Synthesis and pharmacology of site specific cocaine abuse treatment agents: 8-substituted isotropane (3-azabicyclo[3.2.1]octane) dopamine uptake inhibitors

Article abstract of DOI:10.1021/jm020410t

A series of 8-substituted-3-azabicyclo[3.2.1] octanes (isotropanes) were synthesized and tested for inhibitor potency using [³H]WIN 35,428 binding at the dopamine (DA) transporter, [³H]citalopram binding at the serotonin (5-HT) transporter, and [³H]DA uptake assays. The synthesis started with a Mannich condensation of cyclopentanone, benzylamine, and fomaldehyde to afford N-benzyl-3-azabicyclo [3.2.1]octan-8-one (6). The 8-phenyl group was introduced by Grignard addition to ketone 6 or nucleophilic displacement via a triflate of the corresponding alcohol 7a. The 8β-phenyl-8α-alcohols from Grignard addition generally have low affinity for the two transporters and do not effectively inhibit the uptake of [³H]DA. The 8β-phenyl compound (14) without the hydroxyl group at C-8 was much more potent (22-fold) for [³H]WIN 35,428 binding inhibition than the corresponding 8β-phenyl-8α-hydroxy compound (7a). The 8α-phenyl compound 8a was almost as potent as cocaine in binding to the DA transporter (IC₅₀ = 234 nM vs 159 nM for cocaine), whereas the C-8 epimer, compound 14, was somewhat less potent (IC₅₀ = 785 nM). The lower potency of 14 (β-orientation of 8-phenyl group) as compared to 8a (α-orientation) was unexpected, based on modeling studies comparing the new compounds to WIN 35,065-2, an analogue of cocaine. The benzhydryl ethers at C-8 (17), analogous to the benztropines, had better selectivity than the corresponding phenyl compounds, 8a and 14, for the DA transporter as compared to the 5-HT transporter. The isotropane and benzisotropine analogues seem to bind in a manner that is more similar to that of the benztropine compounds 5 rather than those of cocaine and WIN 35,065-2.

Full text of DOI:10.1021/jm020410t

1456
J . Med. Chem. 2003, 46, 1456-1464
Syn th esis a n d P h a r m a cology of Site Sp ecific Coca in e Abu se Tr ea tm en t Agen ts:
8-Su bstitu ted Isotr op a n e (3-Aza bicyclo[3.2.1]octa n e) Dop a m in e Up ta k e
In h ibitor s
Deog-Il Kim,† Margaret M. Schweri,‡ and Howard M. Deutsch*,†
School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, Georgia 30332-0400, and School of Medicine,
Mercer University, Macon, GA 31201-1554
Received September 18, 2002
A series of 8-substituted-3-azabicyclo[3.2.1]octanes (isotropanes) were synthesized and tested
for inhibitor potency using [3H]WIN 35,428 binding at the dopamine (DA) transporter, [3H]-
citalopram binding at the serotonin (5-HT) transporter, and [3H]DA uptake assays. The
synthesis started with a Mannich condensation of cyclopentanone, benzylamine, and fomal-
dehyde to afford N-benzyl-3-azabicyclo[3.2.1]octan-8-one (6). The 8-phenyl group was introduced
by Grignard addition to ketone 6 or nucleophilic displacement via a triflate of the corresponding
alcohol 7a . The 8â-phenyl-8R-alcohols from Grignard addition generally have low affinity for
the two transporters and do not effectively inhibit the uptake of [3H]DA. The 8â-phenyl
compound (14) without the hydroxyl group at C-8 was much more potent (22-fold) for [3H]WIN
35,428 binding inhibition than the corresponding 8â-phenyl-8R-hydroxy compound (7a ). The
8R-phenyl compound 8a was almost as potent as cocaine in binding to the DA transporter
(IC50 ) 234 nM vs 159 nM for cocaine), whereas the C-8 epimer, compound 14, was somewhat
less potent (IC50 ) 785 nM). The lower potency of 14 (â-orientation of 8-phenyl group) as
compared to 8a (R-orientation) was unexpected, based on modeling studies comparing the new
compounds to WIN 35,065-2, an analogue of cocaine. The benzhydryl ethers at C-8 (17),
analogous to the benztropines, had better selectivity than the corresponding phenyl compounds,
8a and 14, for the DA transporter as compared to the 5-HT transporter. The isotropane and
benzisotropine analogues seem to bind in a manner that is more similar to that of the
benztropine compounds 5 rather than those of cocaine and WIN 35,065-2.
In tr od u ction
agents that will fall along a continuum ranging from
full agonist to full antagonist at the cocaine binding
sites. Those compounds that prove to be full agonists
will be developed as a substitution therapy for cocaine
abusers, analogous to the use of methadone8 (a long-
acting and orally effective opiod agonist) for the treat-
ment of heroin addiction. To be effective in this appli-
cation, these agents will need to have a slow onset and
long duration of action. At the opposite end of the
continuum are agents that might be able to inhibit the
binding of cocaine to the transporter yet not interfere
with DA transport. While this is theoretically possible,
to our knowledge, no such compounds have yet been
identified. Work in this area (see Figure 1) has focused
on the tropane ring system (8-azabicyclo[3.2.1]octane).
For example, WIN 35,065-2 (2), in which the â-benzoate
ester of cocaine is replaced by â-phenyl, results in a
more potent compound.9 Although the nitrogen atom
and phenyl group are commonly regarded as important
elements of the pharmacophore, there are some ex-
amples of bicyclo[3.2.1]octanes without nitrogen, com-
pounds 3, which show potent binding to the DAT when
the phenyl ring is 3,4-dichloro-substituted.10 This may
mean that in part, the complete three-dimensional
topology of the ligand may be more important for
binding to the biological macromolecule than the indi-
vidual pharmacophore elements. Interestingly, piperi-
dine analogues11 such as 4, which lack cocaine’s two
carbon bridge, are much less potent than WIN 35,065-2
Cocaine (1) abuse and addiction continues to be a
problem that plagues our society.1 Once a person tries
cocaine, an individual sometimes cannot predict or
control the extent to which he or she will continue to
use the drug.2 According to Monitoring the Future
National Survey Results on Drug Use, 1975-2001,3 in
1998, the proportion of teenagers who have used cocaine
at least once in their lifetime increased slightly in the
early 1990s, although it is still lower than its peak in
1985.
Despite a large amount of research in this area,4 there
are currently no safe and effective pharmacotherapies
available to help habitual users of cocaine who want to
stop their usage of the drug. Its reinforcing and stimu-
lant properties are, in part, a consequence of its
propensity to inhibit monoamine transport systems, in
particular the dopamine transporter (DAT).5 This inhi-
bition results in an increase in synaptic dopamine levels,
which then leads to overstimulation of postsynaptic
dopamine receptors. However, the importance of sero-
toneric mechanisms is being increasingly recognized.6
The work described here is part of an ongoing project7
aimed at the development of drugs that are targeted at
the DAT to treat cocaine abuse. The goal is to synthesize
* To whom correspondence should be addressed. Fax: 404-894-7452.
E-mail: howard.deutsch@chemistry.gatech.edu.
† Georgia Institute of Technology.
‡ Mercer University.
10.1021/jm020410t CCC: $25.00 © 2003 American Chemical Society
Published on Web 03/15/2003

Site Specific Cocaine Abuse Treatment Agents
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 8 1457
reduced by treatment with Bu3SnH and a catalytic
amount of AIBN. Reduction of the S-methyl xanthate
group also occurred only from the â-face to afford
R-phenyl compounds 8a ,b. Selective reduction of the
xanthate of 7b, without the reduction of the aromatic
chlorine, occurred when 1 equiv of Bu3SnH was used.
The stereochemistry of these products, 8a ,b, was
determined as the R-orientation in part by 1H NMR
spectroscopy. The nuclear Overhauser enhancement
spectroscopy (NOESY) spectrum of 8a showed an in-
teraction between the benzylic proton at C-8 and the
bridge protons at C-6 and C-7 and no interaction
between the aromatic protons of the phenyl ring at C-8
and the bridge protons at C-6 and C-7. In addition, there
was an interaction between the protons at C-2 and C-4
and the aromatic protons of the phenyl ring at C-8. The
structure of 8a was further confirmed by single-crystal
X-ray crystallography (Figure 3).
F igu r e 1. Dopamine uptake inhibitors.
Debenzylation of 8a with hydrogen and Pd afforded
9a . Chloroethyl chloroformate16 was used for debenzyl-
ation of 8b to obtain 9b and avoid the hydrogenolysis
of the aromatic chlorine. Reductive alkylation of amines
9a ,b with formaldehyde afforded the N-methylamines,
10a ,b (Scheme 1).
F igu r e 2. Design of isotropane.
(33-fold). That means that the [3.2.1]bridge structure
is important for potent binding at the DAT.
Benztropine (5), 3R-diphenylmethoxytropane,12 is a
dopamine uptake inhibitor, equipotent to cocaine, that
inhibits cocaine-induced central nervous system (CNS)
stimulant activity in animal modes. Generally, 4-â-
substituted isomers are more potent than the corre-
sponding 4-R-substituted compounds for the inhibition
of WIN binding in the tropane series. However, all
benztropine derivatives have shown reversed potency;
that is, the R-isomers are more potent than the â-iso-
mers.
Our approach to the synthesis of the â-phenyl isomer
14 was envisioned to be by a nucleophilic substitution
reaction using a good R-oriented leaving group at the
C-8 position, based on the preference for â-approach of
reagents (Scheme 2). As expected, the NaBH4 reduction
of ketone 6 afforded only the R-alcohol 13a , which was
converted to the triflate using triflic anhydride. When
the crude triflate was treated with phenyllithium at -78
°C, thin-layer chromatography and 1H NMR analysis
did not show any recognizable major product. In addi-
tion, the 8â-bromo compound 15 was obtained as the
only major product when the triflate was treated with
PhMgBr at -78, 0, or 25 °C. However, the â-phenyl
isomer 14 could be obtained in low yield when a mixture
of PhMgBr and CuBr‚Me2S was added to the triflate of
R-alcohol in hexamethylphosphoramide (HMPA) solvent
at 90 °C. This procedure also gave a low yield of the
deoxygenated product 16.
We report here a series of novel tropane analogues,
which we have named “isotropane”, 3-azabicyclo[3.2.1]-
octane. Tropane and isotropane are structurally differ-
ent only in the locations of the two carbon bridge as
shown in Figure 2. In the isotropanes, the two carbon
bridge in part takes the place of the methyl ester of
cocaine. These analogues, which do not have an ester
group, have a significantly different topology as com-
pared to compounds that have been previously synthe-
sized.
The stereochemistry at C-8 of the product 14 was
determined by 1H NMR spectroscopy. The benzylic
proton at C-8 of â-isomer 14 was a broadened singlet,
while that of the R-isomer 8a was a clear triplet. The
coupling constants between H-C8 and the adjacent
protons on C1 and C5 can be explained by the dihedral
angles between these protons in 14 and 8. Using the
global minimized structure of 8a and 14 as calculated
by MM2,17 the calculated dihedral angle between the
protons at C-8 and the protons at C-1 and C-5 is 55° in
8a while it is 80° in 14. According to Karplus’s equa-
tion,18 the vicinal coupling constant would be expected
to be between 2.5 and 4.5 Hz when the dihedral angle
is 55° and between 0.0 and 0.2 Hz when the dihedral
angle is 80°. The proton NMR of 8a shows a triplet (J
) 4.2 Hz) at 3.00 ppm while 14 shows a singlet at 2.74
ppm. Therefore, 8a and 14 could be assigned as the R-
and â-stereoisomers, respectively.
Resu lts a n d Discu ssion
Ch em istr y. N-Benzyl-3-azabicyclo[3.2.1]octan-8-one
(6) was prepared by a Mannich condensation of cyclo-
pentanone, benzylamine, and formaldehyde (Scheme
1).13 In our hands, the N-methyl analogue of 6 could not
be isolated when methylamine was used instead of
benzylamine in the Mannich condensation.14 Addition
of Grignard reagents such as phenylmagesium bromide
to the ketone 6 introduced the aromatic group at C-8.
Nucleophilic reagents are reported14 to approach with
only â-directional preference for this compound. This
afforded only the R-alcohols such as 7a ,b; single-crystal
X-ray crystallography of 7a supported the structural
assignments and the preferential â-approach of phenyl-
magnesium bromide (Figure 3).
Ionic reduction of the hydroxyl group of 7a ,b failed
under a variety of conditions, such as NaI/CH3CN/TMSI,
NaBH4/CF3CO2H, and ZnI2/NaBH4. However, reduction
under free radical conditions15 was successful to afford
benzylamines 8a ,b. In this procedure, the hydroxyl
group was converted to an S-methyl xanthate and
The benzisotropine analogues, 8-diphenylmethoxy-
isotropanes 17a ,b, were synthesized to compare their
pharmacophores with benztropine (5). To get the â-al-
cohol 13b, reactions of R-triflate of 13a were investi-

1458 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 8
Kim et al.
Sch em e 1a
a
Reagents and conditions: (a) Benzylamine/37% HCHO. (b) Grignard reagent/THF. (c) (i) NaH/imidazole/THF, CS2, MeI; (ii) Bu3SnH/
AIBN/benzene. (d) H2, Pd/C. (e) Chloroethyl chloroformate/dichloroethane. (f) (CH2O)n, NaBH4, molecular sieve. (g) 37% HCHO/HCO2H.
â-alcohol 13b involve heating to reflux a mixture of the
triflate of 13a , 1.5 equiv of p-toluenesulfonic acid in
cosolvents of water, benzene, and dimethyl sulfoxide
1
(DMSO) (1:10:4). The H NMR splitting patterns of the
protons at C-8 in these isomers 13a ,b are very similar
to compounds 8a and 14 allowing their stereochemistry
to be assigned. The â-isomer 13b shows a singlet while
the R-isomer 13a is triplet (J ) 4.8 Hz). Using the
method developed by Meltzer et al.,19 the coupling
between isotropan-8-ols 13a or b and benzhydrol af-
forded ethers 17a ,b in good yields, respectively (Scheme
2).
P h a r m a cology. The compounds were tested for their
ability to inhibit the binding of [3H]WIN 35,428 to the
cocaine binding site on the DAT in synaptosomal
membrane preparations from rat striatal tissue,7a [3H]-
DA uptake into rat striatal synaptosomes,7a,20 and [3H]-
citalopram binding to the 5-HTT in synaptosomal
membrane preparations from rat cortical tissue.21 The
results are summarized in Table 1.
In h ibition of [3H]WIN 35,428 Bin d in g. Substitu-
ents such as methyl or benzyl on the N-atom of isotro-
panes have little effect on binding potencies. For ex-
F igu r e 3. ORTEP of 7a (up) and 8a (down).
ample (N-substituent, IC50), compounds 8a (N-benzyl;
Sch em e 2a
234 nM), 10a (N-methyl; 360 nM), and 9a (N-hydrogen;
258 nM) show similar potencies. The same trend has
been reported for many tropane compounds, for ex-
ample, WIN 35,065-2 (N-methyl; 23.0 nM) and WIN 35,-
981 (N-hydrogen; 30.8 nM).22
The aromatic group at C-3 in tropanes such as WIN
35,065-2 is regarded as a critical pharmacophore for the
recognition site. Binding affinities for the DAT are also
highly dependent upon substituents on the phenyl
ring.4,23 Likewise, a phenyl group at C-8 in isotropanes
greatly increases their potency. For example, 8a (C-8R-
phenyl) is 104-fold more potent than compound 16
without a phenyl group. This effect is greatly dimin-
ished, however, when the other substituent at C-8 is a
hydroxyl group (compare 13a and 7a ). The orientation
of the phenyl group at C-8 has a significant effect on
potency. Compound 8a (R-orientation) is more potent
than the corresponding â-epimer (14) by 3.4-fold. The
same trend is also observed for the benzhydryl ethers
(benzisotropines); for example, 17a (R-orientation; IC50
) 363 nM) is slightly more potent than 17b (â-orienta-
tion; IC50 ) 536 nM). These results contrast to phenyl-
tropanes such as WIN 35,065-2 and its C-3 epimer,
a
Reagents and conditions: (a) NaBH4. (b) Tf2O, pyridine. (c)
TsOH‚H2O/water/DMSO/benzene. (d) Benzhydrol/TsOH‚H2O/
benzene. (e) PhMgBr, CuBr‚Me2S, HMPA, 90 °C.
gated. With aqueous NaOH or KOH solutions, only the
R-alcohol 13a was obtained, presumably by hydrolysis.
However, 13b could be obtained by reaction under acidic
conditions. The best conditions (60% yield) to obtain the

Site Specific Cocaine Abuse Treatment Agents
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 8 1459
Ta ble 1. Binding and Uptake Data for 3-Azabicyclo[3.2.1]octane Isotropane Derivatives and Reference Compound (Value ( SEM)a
inhibition assays IC50 (nM)
% inhibition
[3H]CIT at
10 µM
selectivity 5-HHT/DAT
[3H]CIT/[3H]WIN
35,428
compdb
R2
DAT
DAT
5-HTT
no.
R1
R3
OH
OH
Ph
[3H]WIN 35,428
[3H]DA
[3H]CIT
1
7a
7b
8a
8b
9a
cocaine
Ph
4-ClPh
H
H
H
160 ( 15
17500 ( 4800
1510 ( 260
234 ( 10
404 ( 26
27 100 ( 560
2630 ( 79
309 ( 5.5
389
2.43
0.17
0.66
14.8
4.2
16
4.4
5.1
0.81
∼1.0
>1.0
>0.45
4.9
>2.2
>0.41
∼27
>18
Bn
Bn
Bn
Bn
H
3060 ( 450
997 ( 110
3460 ( 640
402 ( 13
4000 ( 850
363 ( 3.0
1850 ( 310
220 ( 24
∼10 000
4-ClPh
Ph
95.5 ( 20
601 ( 13
258 ( 70
893 ( 6
478 ( 59
1040 ( 110
926 ( 88
9b
H
H
H
H
Ph
4-ClPh
Ph
4-ClPh
OH
81.5 ( 2.5
360 ( 43
10a
10b
11
CH3
CH3
H
273 ( 59
9830 ( 420
10 100 ( 710
22 100 ( 1000
785 ( 67
8480 ( 370
9770 ( 440
45 000 ( 8600
4450 ( 480
8640 ( 900
35 700 ( 3800
735 ( 78
46.2 ( 2.8
27.8 ( 2.6
23.8 ( 4.8
12
13a
14
15
16
CH3
Bn
Bn
Bn
Bn
Bn
Bn
Ph
H
Ph
Br
H
H
OH
OH
H
H
>10 000
>10 000
3880 ( 440
>10 000
4630 ( 830
24 300 ( 660
363 ( 37
30.4 ( 9.5
24.4 ( 4.2
61.0 ( 4.6
29.8 ( 0.5
H
>10 000
17a
17b
OCHPh2
H
∼10 000
OCHPh2
536 ( 61
1980 ( 220
>10 000
a
b
Abbreviations: Bn, benzyl; Ph, phenyl; CIT, citalopram. Compounds tested as HCl salts.
which have IC50 values of 101 (R-orientation) and 23
nM (â-orientation), repectively.23a However, benztropines
show the same trend as isotropanes; that is, R-benz-
tropine (Ki ) 118 nM) is 7.2-fold more potent than
â-benzisotropine.12
35, 428 binding. Compounds with high DR values (g36)
have theoretical promise as cocaine antagonists7a be-
cause they should be able to prevent cocaine from
binding at a dose that does not affect DA uptake. The
range of values obtained for the present series was 1.3
(8a ) to 6.3 (8b), suggesting that the compounds in this
series will likely be more agonist- than antagonist-like.
Compounds 8a ,b are unique in that this pair represents
the only instance in which a chemical modification (4-
Cl substitution of the R-phenyl group) caused a con-
comitant increase in potency against binding and a
decrease in potency against uptake. In all other cases,
the direction of the change in potency is the same in
both assays but to differing degrees.
The usefulness of the DR has been questioned because
it has been shown to vary greatly depending on assay
conditions.25 Data collected under our closely controlled
experimental conditions as part of a previous study of
another series of compounds revealed an unexpected
correlation between the DR values and the slopes of
dose-response curves measuring their generalization
to cocaine,21a suggesting that even if this parameter
proves unsatisfactory as a predictor of antagonist ef-
ficacy, it may have other useful applications.
In h ibition of [3H]Cita lop r a m Bin d in g. These com-
pounds generally showed low potency against [3H]-
citalopram binding at the 5-HTT with IC50 values
ranging from 220 (10b) to well over 10 000 nM. As with
[3H]WIN 35,428 binding, phenyl substitution increased
potency; for example, the phenyl-containing compounds
8a (R-phenyl; 3460 nM) and 14 (â-phenyl; 3880 nM)
were more potent than the desphenyl compound 16,
whose IC50 was well over 10 000 nM. Unlike the [3H]-
WIN 35,428 binding results, however, there is no
orientation effect of the phenyl group at C-8 between
isomers 8a and 14. A slight preference for the R-orien-
tation is retained for the benzhydryl ethers (17a vs 17b),
however, despite the fact that their affinity for the [3H]-
citalopram binding site is markedly reduced.
A 4-chloro substituent on the phenyl ring of 8b and
9b increases the potency about 3-fold as compared to
the corresponding unsubstituted phenyl derivatives 8a
and 9a . The most potent compound in this study is the
4-chlorophenyl-substituted derivative, 9b (IC50 ) 81.5
nM), which is 2-fold more potent than (-)-cocaine.
Interestingly, the N-Me derivative (10a ) showed little
or no increase in potency upon 4-chloro substitution of
the phenyl ring (10b). In contrast, the 4-chloro substitu-
tion of the phenyl group of the tropane WIN 35,065-2
was reported to elicit a 20-fold increase in inhibitory
potency at the stimulant binding site.9,23 The compound
with an 8R-hydroxyl group (7a ) has a much lower
affinity (22-fold) as compared to the corresponding
compound (14) without the hydroxyl group.
The Hill coefficients (nH) for most of the compounds
in the binding assay were close to unity (data not shown
here). The two compounds with the highest nH values
were 17a ,b with values of 1.66 ( 0.16 and 1.38 ( 0.08,
respectively. These compounds required the maximal
amount of DMSO tolerated by the assay to solubilize
them. It is possible, therefore, that these higher nH
values associated with these compounds were artifacts
resulting from their extreme hydrophobicity.
In h ibition of [3H]DA Up ta k e. Similar conclusions
to those discussed about [3H]WIN 35,428 binding can
be reached when considering the inhibition of [3H]DA
uptake since there is a high correlation between them.
The correlation is not perfect, and a term called the
“discrimination ratio” (DR) is utilized by us to detect
any marked departure from linearity that occurs be-
tween these two measures.7a,24 The DR value is defined
as a ratio of the IC50 for a given compound against [3H]-
DA uptake to its corresponding IC50 against [3H]WIN

1460 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 8
Kim et al.
Whereas the hydroxyl group at C-8 greatly reduces
the potency for [3H]WIN 35,428 binding inhibition, it
does not change the potency for [3H]citalopram inhibi-
tion. For example, R-oriented hydroxy compound 7a
(3060 nM) and the corresponding compound without
hydroxyl group 14 (3880 nM) show similar potencies,
as do 13a and 16. The chlorine substituent (8b, 9b, and
10b) on the phenyl ring at C-8 increases the potencies
from 8.4- to 11.0-fold when they are compared to the
corresponding unsubstituted compounds (8a , 9a , and
10a ). These increases in potency are somewhat larger
than that for [3H]WIN 35,428 binding inhibition (about
3-fold increase for these compounds). As with [3H]WIN
35,428 binding, the substituents on the N-atom, benzyl
(8b; 402 nM), hydrogen (9b; 363 nM), or methyl (10b;
220 nM), do not show a significant effect on the
potencies. The N-methylisotropanes, 10a and b, show
small increases in potency as compared to the corre-
sponding norisotropane compounds, 9a ,b. In compari-
son, norcocaine and WIN 35,981 (N-hydrogen) have been
reported to be 8.3- and 12.6-fold more potent, respec-
tively, than cocaine and WIN 35,065-2 (N-methyl).4,23b
Tr a n sp or ter Selectivity. Selectivity of all of the
compounds screened for the two transporters was
calculated by comparing the ratio of IC50 values for [3H]-
citalopram to [3H]WIN 35,428 (5-HTT/DAT) as shown
in Table 1. The compound that shows the highest
selectivity (ratio ∼27) for the DA transporter is R-ben-
zisotropine 17a . Thus, none of the compounds studied
were highly selective.
F igu r e 4. (A) Overlay of WIN 35,065-2 (dark) and N-methyl-
8â-phenyl-3-azabicyclo[3.2.1]octane (gray). (B) Overlay of WIN
35,065-2 (dark) and 10a (gray).
Generally, the selectivity for the DAT against 5-HTT
is decreased in these derivatives when chlorine is
substituted on the phenyl ring at C-8. For example,
compounds 8a , 9a , and 10a , which have an unsubsti-
tuted phenyl ring at C-8, are 15-, 16-, and 5.1-fold
selective, respectively, for the DA transporter. However,
the corresponding compounds 8b, 9b, and 10b, which
have a 4-chlorophenyl ring at C-8, are only 4.2-, 4.4-,
and 0.8-fold selective, respectively, for the DAT.
Gen er a l Con sid er a t ion . [3H]WIN 35,428 binding
data for tropane compounds, such as WIN 35,065-2,
show that â-orientation of the phenyl group at C-3 is
more potent (5-fold) than for the R-orientation.4,20 As
shown in Figure 4A,B, the isotropane N-methyl-3-aza-
8â-phenylbicyclo[3.2.1]octane (â-orientation of a phenyl
group) is more closely superimposed topologically with
WIN 35,065-2 (Figure 4A) than N-methyl-3-aza-8R-
phenylbicyclo[3.2.1]octane 10a (R-orientation) when the
N atom and phenyl group of each structure are overlaid,
respectively.17 This leads to the expectation that the 8â-
phenylisotropane would be a more potent inhibitor of
[3H]WIN 35,428 binding at the DAT. However, the
R-orientation 8a (234 nM) is more potent by 3.4-fold
than the â-isomer 14 (785 nM). This trend is similar to
that of the benztropines12 (5), which show that the
R-orientation is 7-fold more potent. This may imply that
the isotropane analogues bind in a manner that is more
similar to that of benztropine compounds 5 rather than
those of cocaine and WIN 35,065-2.
6-fold increase, respectively, for mono- and dichloro
substitution. Interestingly, 4-chlorophenyl derivatives
of isotropanes C-8 (8b, 9b, and 10b) also show a small
increase in potency (less than 3-fold) when they are
compared, respectively, to the corresponding unsubsti-
tuted compounds (8a , 9a , and 10a ). Again, this implies
that the isotropane analogues may be more similar in
binding characteristics to the benztropines than to the
tropane analogues.
The fact that the isotropanes resemble the benz-
tropines more than the tropanes in biochemical activity
could have far-reaching implications for the develop-
ment of pharmacotherapies for the treatment of cocaine
abuse. Several benztropines have been found to be
potent inhibitors of both [3H]WIN 35,428 binding and
DA uptake, yet have behavioral profiles that indicate
diminished psychomotor stimulant effects and self-
administration potential relative to cocaine.26
Con clu sion s
Derivatives of a novel structure, 8-phenyl-3-azabicyclo-
[3.2.1]octane, which we have named isotropane, were
designed and synthesized as dopamine uptake inhibi-
tors, which might be used as lead compounds for the
development of treatment agents for cocaine abuse. The
isomers at C-8 of the isotropane structure were synthe-
sized selectively, and these derivatives were tested for
their binding to the DAT and 5-HTT and their ability
to inhibit [3H]DA uptake. Several alternate explanations
of the structure-activity relationships of these com-
pounds are offered in terms of known pharmacophores.
When chlorine is substituted at the 4-position of the
phenyl ring of WIN 35,065-2, the potency for the
inhibition of WIN binding is increased by 20-fold. This
4-chlorine substitution effect on the phenyl ring is
reduced for benztropine analogues, which show a 4- and
Compound 8b shows a relatively high affinity (IC50
)

Site Specific Cocaine Abuse Treatment Agents
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 8 1461
95.5 nM) for the [3H]WIN 35,428 binding site and has
little selectivity (4.2-fold) for the DAT over the 5-HTT.
The benzisotropine derivative (17a ) shows better selec-
tivity (27-fold) but somewhat less potency (IC50 ) 363
nM). The isotropane analogues seem to bind in a
manner that is more similar to that of benztropines
rather than those of cocaine and WIN 35,065-2. Further
testing and synthesis of these and other related com-
pounds are in progress.
S-methyl xanthate (468 mg, 1.22 mmol), Bu3SnH (495 µL, 1.83
mmol), and AIBN (61 mg) in benzene (10 mL) was degassed
with argon at room temperature and refluxed for 12 h. After
it was cooled and concentrated, the residue was recrystallized
from hexane to afford colorless crystals of 8a (276 mg, 82%);
mp 108-109 °C. 1H NMR (CDCl3, 300 MHz): δ 7.34-7.24 (m,
8H), 7.21-7.14 (m, 2H), 3.28 (s, 2H), 3.00 (t, J ) 4.2 Hz, 1H),
2.63 (br s, 2H), 2.45-2.36 (m, 4H), 2.02-1.95 (m, 2H), 1.83-
1.79 (m, 2H). 13C NMR (CDCl3, 100 MHz): δ 140.7, 139.9,
129.4, 128.5, 128.3, 128.0, 126.5, 125.2, 62.3, 53.5, 47.4, 36.7,
29.2. HRMS (EI) calcd for C20H23N m/z, 277.1830; found,
277.1814.
Exp er im en ta l Section
(8a )-HCl. mp 168-169 °C. 1H NMR (D2O, 300 MHz): δ
7.26-7.12 (m, 10H), 3.88 (s, 2H), 3.00 (br s, 5H), 2.81 (s, 2H),
Gen er a l Meth od s. Starting materials were purchased from
Aldrich Chemical Co. and used without further purification.
Tetrahydrofuran (THF) was dried over sodium benzophenon
ketyl prior to distillation under argon. Benzene was dried over
sodium before it was used. Compound 6 was synthesized
according to the literature.13 Flash chromatography was run
using 230-400 mesh silica gel. Melting points were deter-
mined on a Mel-Temp apparatus and are uncorrected. 1H (300
or 500 MHz), 13C (75 or 100 MHz) NMR, and COSY/NOESY
spectra were obtained on either a Varian Gemini-300 or a
Bruker DSX-500 spectrometer. High-resolution mass spectra
(electron impact (EI) or fast atom bombardment (FAB)) were
recorded on a VG Analytical 70-SE mass spectrometer equipped
with a 11-250J data system. Elemental analyses were obtained
from Atlantic Microlabs, Atlanta, GA. Free bases were dis-
solved in CH2Cl2 and converted to HCl salts by the addition
of 1 M HCl (1.5 equiv) in diethyl ether. An excess amount of
HCl was removed under reduced pressure, and the solid was
recrystallized from MeOH/EtOAc/diethyl ether/CH2Cl2.
N-Ben zyl-3-azabicyclo[3.2.1]-8â-ph en yloctan -8r-ol (7a).
A 1 M solution of phenylmagnesium bromide (6.3 mL) was
added to a solution of 6 (900 mg, 4.18 mmol) in THF (20 mL)
at 0 °C. After it was stirred for 2 h, the mixture was quenched
with water and extracted with petroleum ether. The organic
layer was dried over MgSO4, filtered, and concentrated. The
residue was purified by silica gel chromatography eluting with
1:6 EtOAc:hexane to afford a white solid, 7a (1.04 g, 85%);
1.99-1.96 (m, 2H), 1.70 (d, J ) 8.7 Hz, 2H). Anal. (C20H24
NCl‚0.3H2O): C, H, N, Cl.
-
N-Ben zyl-3-a za b icyclo[3.2.1]-8r-(4-ch lor o)p h en yloc-
ta n e (8b). Compound 8b was made using a similar procedure
as that used to prepare 8a except that 1 equiv of Bu3SnH was
employed, and the colorless crystals of 12b (mp 134-135 °C)
1
were obtained with 76% yield. H NMR (CDCl3, 500 MHz): δ
7.34-7.22 (m, 9H), 3.32 (s, 2H), 2.99 (t, J ) 4.0 Hz, 1H), 2.63
(br s, 2H), 2.45 (dd, J ) 3.8, 10.9 Hz, 2H), 2.37 (d, J ) 10.8
Hz, 2H), 2.03 (m, 2H), 1.86 (m, 2H). 13C NMR (CDCl3, 100
MHz): δ 139.8, 139.3, 130.9, 130.7, 128.5, 128.4, 128.0, 126.6,
62.3, 53.4, 46.9, 36.8, 29.2. HRMS (EI) calcd for C20H22NCl
m/z, 311.1441; found, 311.1435.
(8b)-HCl. mp 189-190 °C. 1H NMR (D2O, 500 MHz): δ
7.38 (t, J ) 7.4 Hz, 1H), 7.28 (t, J ) 7.7 Hz, 2H), 7.25 (d, J )
8.6 Hz, 2H), 7.19 (d, J ) 7.6 Hz, 2H), 7.11 (d, J ) 8.2 Hz, 2H),
3.98 (s, 2H), 3.06 (m, 5H), 2.86 (br s, 2H), 2.07 (m, 2H), 1.80
(d, J ) 8.7 Hz, 2H). Anal. (C20H23NCl2‚1.7H2O): C, H, N, Cl.
3-Aza bicyclo[3.2.1]-8r-p h en ylocta n e (9a ). A mixture of
8a (250 mg, 0.903 mmol) and Pd/C (10%, 40 mg) in AcOH/
EtOAc (2:5) was shaken overnight under hydrogen atmosphere
at 40 psi. The mixture was filtered on Celite and concentrated
under reduced pressure. The residue was mixed with water,
treated with K2CO3 (pH 10), and extracted with CH2Cl2 to
afford 9a (154 mg, 91%). 1H NMR (CDCl3, 300 MHz): δ 7.33-
7.27 (m, 4H), 7.19-7.14 (m, 1H), 3.10 (s, 1H), 3.08 (d, J ) 12.0
Hz, 2H), 2.56 (br s, 2H), 2.43 (dd, J ) 3.0, 12.6 Hz, 2H), 1.96
(m, 2H), 1.91 (m, 1H), 1.84 (m, 2H). 13C NMR (CDCl3, 75
MHz): δ 140.4, 129.2, 128.4, 125.3, 47.7, 46.2, 36.9, 28.3.
HRMS (EI) calcd for C13H17N m/z, 187.1361; found, 187.1354.
1
mp 90-91 °C. H NMR (CDCl3, 300 MHz): δ 7.49 (d, J ) 8.4
Hz, 2H), 7.38-7.22 (m, 8H), 3.60 (s, 2H) 2.87 (d, J ) 10.5 Hz,
2H), 2.62 (dd, J ) 3.3, 10.5 Hz, 2H), 2.41 (s, 2H), 1.79 (d, J )
7.8 Hz, 2H), 1.58 (br. s, 1H), 1.40 (m, 2H). HRMS (EI) calcd
for C20H23NO m/z, 293.1780; found, 293.1779.
(7a )-HCl. mp 239-241 °C. 1H NMR (D2O, 300 MHz): δ
7.54-7.32 (m, 10H), 4.26 (s, 2H), 3.63 (d, J ) 12.0 Hz, 2H),
3.23 (d, J ) 12.0 Hz, 2H), 2.72 (s, 2H), 1.57 (s, 4H). Anal.
(C20H24NOCl): C, H, N, Cl.
(9a )-HCl. mp 291-294 °C. 1H NMR (D2O, 300 MHz): δ
7.38 (t, J ) 6.9 Hz, 2H), 7.22 (d, J ) 6.9 Hz, 2H), 7.14 (t, J )
7.2 Hz, 1H), 3.06 (br s, 2H), 3.02 (s, 1H), 2.85 (br s, 1H), 2.79
(m, 3H), 2.01 (m, 2H), 1.70 (d, J ) 8.4 Hz, 2H). Anal. (C13H18
NCl‚0.2H2O): C, H, N, Cl.
-
N-Ben zyl-3-azabicyclo[3.2.1]-8â-(4-ch lor o)-ph en yloctan -
8r-ol (7b). Using a similar procedure as that used to prepare
7a , a white solid (mp 118-119 °C) was obtained with 86%
3-Azabicyclo[3.2.1]-8r-(4-ch lor o)ph en yloctan e‚HCl (9b‚
HCl). A mixture of 8b (84 mg, 0.270 mmol) in dichloroethane
(3 mL) and chloroethyl chloroformate (31 µL, 0.287 mmol) was
cooled to 0 °C. After 10 min, the solution was refluxed
overnight and then concentrated. The residue was dissolved
in methanol (3 mL) and heated for 1 h at 50 °C. After it was
cooled, the mixture was concentrated and recrystallized to give
1
yield. H NMR (CDCl3, 300 MHz): δ 7.42 (d, J ) 8.7 Hz, 2H),
7.31 (d, J ) 8.1 Hz, 2H), 7.43-7.22 (m, 5H), 3.59 (s, 2H), 2.85
(d, J ) 10.8 Hz, 2H), 2.61 (dd, J ) 3.6, 10.8 Hz, 2H), 2.36 (d,
J ) 1.5 Hz, 2H), 1.80 (d, J ) 7.5 Hz, 2H), 1.54 (s, 1H), 1.37
(m, 2H). HRMS (EI) calcd for C20H22NOCl m/z, 327.1390;
found, 327.1379.
1
9b (55 mg, 79%); mp 272-273 °C. H NMR (D2O, 300 MHz):
(7b)-HCl. mp 238-241 °C. 1H NMR (D2O, 300 MHz): δ
7.42 (d, J ) 4.2 Hz, 2H), 7.35 (d, J ) 4.8 Hz, 2H), 7.46-7.31
(m, 5H), 4.26 (s, 2H), 3.62 (d, J ) 12.0 Hz, 2H), 3.23 (d, J )
δ 7.32 (d, J ) 8.4 Hz, 2H), 7.21 (d, J ) 8.4 Hz, 2H), 3.07 (d, J
) 12.6 Hz, 2H), 3.05 (m, 1H), 2.88 (dd, J ) 2.7, 12.6 Hz, 2H),
2.79 (br s, 2H), 2.05 (m, 2H), 1.74 (d, J ) 8.4 Hz, 2H). 13C NMR
(D3O, 100 MHz): δ 136.3, 131.2, 130.1, 128.7, 44.6, 43.7, 33.5,
25.7. HRMS (CI) calcd for C13H17NCl m/z, 222.1050; found,
222.1037 (M + 1). Anal. (C13H17NCl2): C, H, N, Cl.
11.1 Hz, 2H), 2.67 (s, 2H), 1.57 (br s, 4H). Anal. (C20H23
NOCl2): C, H, N, Cl.
-
N-Ben zyl-3-a za bicyclo[3.2.1]-8r-p h en ylocta n e (8a ). A
mixture of alcohol 7a (406 mg, 1.38 mmol), 60% NaH in
mineral oil (166 mg, 4.16 mmol), and a catalytic amount of
imidazole (7 mg) in THF (6 mL) was refluxed overnight under
argon, and carbondisulfide (2 mL) was added to the mixture.
After 1 h, methyl iodide (1 mL) was added and refluxed for 1
h. The mixture was cooled, quenched with water, and extracted
with petroleum ether. The organic layer was dried over MgSO4,
filtered, and concentrated. The residue was purified by silica
gel chromatography eluting 1:30 EtOAc:hexane to afford
yellow solid (468 mg, 89%). S-Methyl xanthate formation was
N-Meth yl-3-a za bicyclo[3.2.1]-8r-p h en ylocta n e (10a ). To
a solution of 9a (330 mg, 1.76 mmol) in methanol (3 mL) was
added paraformaldehyde (265 mg, 8.82 mmol), the molecular
sieves (5 Å, 500 mg), and NaBH4 (201 mg, 5.31 mmol), and it
was stirred for 3 days. The mixture was filtered on Celite and
1
concentrated to give compound 10a (324 mg, 92%). H NMR
(CDCl3, 300 MHz): δ 7.34-7.24 (m, 4H), 7.18-7.13 (m, 1H),
3.00 (t, J ) 3.6 Hz, 1H), 2.66 (br s, 2H), 2.46 (dd, J ) 3.6, 11.1
Hz, 2H), 2.35 (d, J ) 10.5 Hz, 2H), 2.07 (s, 3H), 1.95-1.85 (m,
4H). 13C NMR (CDCl3, 75 MHz): δ 140.6, 129.3, 128.3, 125.2,
1
confirmed by H NMR (δ 2.42, singlet, S-CH3). A solution of

1462 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 8
Kim et al.
56.0, 47.1, 46.0, 36.5, 29.1. HRMS (EI) calcd for C14H19N m/z,
Hz, 2H), 2.06 (m, 4H), 1.74 (d, J ) 6.6 Hz, 2H), 1.56 (m, 2H),
1.44 (m, 1H), 1.34 (d, J ) 11.1 Hz, 1H). HRMS (EI) calcd for
C14H19N m/z, 201.1517; found, 201.1520.
(16)-HCl. mp 215-216 °C. 1H NMR (D2O, 300 MHz): δ
7.62 (m, 2H), 7.43 (m, 3H), 4.17 (d, J ) 4.5 Hz, 2H), 3.21 (d, J
) 11.1, 2H), 3.75 (t, J ) 10.8 Hz, 2H), 2.48 (dd, J ) 2.1, 9.3
Hz, 2H), 2.39 (br s, 2H), 1.82 (m, 2H), 1.58 (m, 2H), 1.39 (d, J
) 11.7 Hz, 1H). Anal. (C14H20NCl): C, H, N, Cl.
201.1518; found, 201.1502.
(10a )-HCl. mp 244-246 °C. H NMR (D2O, 300 MHz): δ
1
7.38-7.20 (m, 5H), 3.15-3.00 (m, 5H), 2.90 (br s, 2H), 2.48 (s,
3H), 2.08 (m, 2H), 1.78 (d, J ) 8.7 Hz, 2H). Anal. (C14H20NCl‚
0.1H2O): C, H, N, Cl.
N-Met h yl-3-a za b icyclo[3.2.1]-8r-(4-ch lor o)p h en yloc-
ta n e (10b). Compound 9b (148 mg, 0.668 mmol) was dissolved
in 2 mL of 37% HCHO and 1 mL of formic acid at room
temperature, and the mixture was refluxed overnight. After
it was cooled, the mixture was concentrated, and the residue
was dissolved in water and treated with K2CO3. The mixture
was extracted with CH2Cl2 (three times) to afford a crystalline
10b (128 mg, 81%); mp 74-75 °C. 1H NMR (CDCl3, 300
MHz): δ 7.28 (d, J ) 8.4 Hz, 2H), 7.18 (d, J ) 8.4 Hz, 2H),
2.92 (t, J ) 3.9 Hz, 1H), 2.60 (br s, 2H), 2.43 (dd, J ) 3.3, 11.4
Hz, 2H), 2.27 (d, J ) 11.1 Hz, 2H), 2.05 (s, 3H), 1.92-1.83 (m,
4H). 13C NMR (CDCl3, 75 MHz): δ 139.1, 130.9, 130.7, 128.5,
55.8, 46.4, 45.9, 36.6, 29.1. HRMS (EI) calcd for C14H18NCl
m/z, 235.1125; found, 235.1129.
3-Aza bicyclo[3.2.1]-8â-p h en ylocta n -8r-ol (11). Using a
similar procedure with the synthesis of 9a , compound 7a was
hydrogenated in acetic acid to afford a white solid 11 (91%);
1
mp 209-211 °C. H NMR (CDCl3, 300 MHz): δ 7.49 (d, J )
6.9 Hz, 2H), 7.39-7.26 (m, 3H), 3.54 (d, J ) 12.6 Hz, 2H), 2.63
(dd, J ) 2.7, 12.6 Hz, 2H), 2.37 (br s, 2H), 1.83 (br s, 2H), 1.60-
1.50 (m, 4H).
1
(11)-HCl. mp >320 °C. H NMR (D2O, 300 MHz): δ 7.41
(m, 2H), 7.32-7.23 (m, 3H), 3.63 (d, J ) 12.3 Hz, 2H), 3.10 (d,
J ) 12.0 Hz, 2H), 2.70 (br s, 2H), 1.59 (br s, 4H). Anal. (C13H18
NOCl‚0.05H2O): C, H, N, Cl.
-
N-Meth yl-3-azabicyclo[3.2.1]-8â-ph en yloctan -8r-ol (12).
Using a similar procedure with the synthesis of 10a , compound
11a was treated to afford 12 (99%). 1H NMR (CDCl3, 300
MHz): δ 7.44 (d, J ) 7.2 Hz, 2H), 7.35-7.22 (m, 3H), 2.78 (d,
J ) 10.5 Hz, 2H), 2.56 (d, J ) 10.5 Hz, 2H), 2.36 (s, 2H), 2.26
(s, 3H), 2.30-1.90 (br s, 1H), 1.66 (d, J ) 7.8 Hz, 2H), 1.37
(m, 2H). HRMS (self-CI) calcd for C14H20NO m/z, 218.1545;
found, 218.1526 (M + 1).
1
(10b)-HCl. mp 282-284 °C. H NMR (D2O, 300 MHz): δ
7.29 (d, J ) 8.4 Hz, 2H), 7.17 (d, J ) 8.4 Hz, 2H), 3.06 (dd, J
) 2.7, 13.2 Hz, 2H), 2.99 (br s, 1H), 2.92 (d, J ) 12.6 Hz, 2H),
2.79 (br s, 2H), 2.42 (s, 3H), 2.01 (m, 2H), 1.70 (d, J ) 12.0
Hz, 2H). Anal. (C14H19NCl2‚0.2H2O): C, H, N, Cl.
N-Ben zyl-3-a za bicyclo[3.2.1]-8â-p h en ylocta n e (14). To
a solution of 6 (1.27 g, 5.91 mmol) in CH3OH (15 mL) was
added NaBH4 (300 mg, 7.93 mmol) at 0 °C. After 2 h, the
mixture was quenched with water and extracted with CH3Cl
to afford N-benzyl-3-azabicyclo[3.2.1]octan-8R-ol 13a (1.2 g,
95%). 1H NMR (D2O, 300 MHz): δ 7.37-7.23 (m, 5H), 3.97 (t,
J ) 4.8 Hz, 1H), 3.52 (s, 2H), 2.60 (d, J ) 10.2 Hz, 2H), 2.45
(dd, J ) 3.3, 11.1 Hz, 2H), 1.96 (br s, 1H), 1.82 (m, 2H), 1.65
(m, 2H).
Compound 13a (90 mg, 0.41 mmol) was treated with Tf2O
(150 µL, 0.892 mmol) and pyridine (300 µL, 3.71 mmol) in CH2-
Cl2 (3 mL) for 30 min at 0 °C. The mixture was poured into
saturated NaHCO3 solution and extracted with CH2Cl2. The
organic layer was concentrated after drying, and the residue
was dissolved in HMPA (1.5 mL). To the solution was added
a mixture of phenylmagnesium bromide (1 M, 1.5 mL) and
CuBr‚Me2S (5 mol %) under argon at 90 °C. After 2 h, the
mixture was quenched with water and extracted with diethyl
ether. The combined organic layer was dried over MgSO4,
concentrated, and purified by flash column chromatography
with 1:20 EtOAc:hexane to afford N-benzyl-3-azabicyclo[3.2.1]-
8â-bromooctane 15 (40 mg, 35%) and the mixture of 14 and
N-benzyl-3-azabicyclo[3.2.1]octane 16 (37 mg). This mixture
was once more purified by flash column chromatography with
1:15:15 diethyl ether:petroleum ether:CH2Cl2 to afford 14 (9.1
mg, 8%) and 16 (26.4 mg, 32%).
1
(12)-HCl. mp >300 °C. H NMR (D2O, 300 MHz): δ 7.41
(m, 2H), 7.39-7.30 (m, 3H), 3.59 (d, J ) 12.6 Hz, 2H), 3.31 (d,
J ) 12.3 Hz, 2H), 2.79 (s, 3H), 2.73 (s, 2H), 1.60 (s, 4H). Anal.
(C14H20NOCl‚0.25H2O): C, H, N, Cl.
N-Ben zyl-3-a za bicyclo[3.2.1]-octa n -8â-ol (13b). Com-
pound 13a (90 mg, 0.41 mmol) was treated with Tf2O (150 µL,
0.892 mmol) and pyridine (300 µL, 3.71 mmol) in the CH2Cl2
(3 mL) for 30 min at 0 °C. The mixture was poured into
saturated NaHCO3 solution and extracted with CH2Cl2. After
it was dried with anhydrous MgSO4, the organic layer was
concentrated to give the triflate as a quantitative yield.
Without further purification, it was refluxed with 1.5 equiv of
p-TsOH‚H2O in a mixture of water (1 mL), DMSO (4 mL), and
benzene (10 mL) under reflux for 3 days. After it was quenched
with saturated K2CO3, the mixture was extracted with CH2-
Cl2 and the organic layer was concentrated to give the crude
product, which was purified by flash column chromatography
with 1:2 ethyl acetate:hexane to afford the â-isomer 13b (54
mg, 60%); mp 103-104 °C. 1H NMR (CDCl3, 300 MHz): δ
7.35-7.20 (m, 5H), 3.75 (s, 1H, C-8), 3.45 (s, 2H), 2.65 (ddd, J
) 1.5, 4.8, 9.6 Hz, 2H), 2.11 (d, J ) 10.8 Hz, 2H), 2.05 (br. s,
2H), 1.82-1.72 (m, 4H). 13C NMR (CDCl3, 75 MHz): δ 139.5,
128.6, 128.1, 126.7, 81.0, 61.5, 58.2, 41.1, 26.1. MS (CI) for
C
14H20NO m/z, 218.3 (M + 1).
N-Ben zyl-3-a za bicyclo[3.2.1]-8â-p h en ylocta n e (14). mp
70-71 °C. 1H NMR (CDCl3, 300 MHz): δ 7.35-7.14 (m, 10H),
3.53 (s, 2H), 2.83 (dd, J ) 3.0, 9.9 Hz, 2H), 2.74 (s, 1H), 2.48
(br s, 2H), 2.28 (d, J ) 10.2 Hz, 2H), 1.74 (d, J ) 7.5 Hz, 2H),
1.57 (m, 2H). HRMS (EI) calcd for C20H23N m/z, 277.1830;
found, 277.1834.
N-Ben zyl-8â-d ip h en ylm eth oxy-3-a za bicyclo[3.2.1]-oc-
ta n e (17b). A mixture of 8â-alcohol 13b (75 mg, 0.346 mmol),
benzhydrol (127 mg, 0.691 mmol), p-toluenesulfonic acid
monohydrate (131 mg, 0.691 mmol), and benzene (20 mL) was
heated at reflux for 6 h. Benzhydrol (127 mg, 0.691 mmol) and
p-toluenesulfonic acid monohydrate (65 mg, 0.346 mmol) were
added, and the reaction mixture was further refluxed over-
night. Benzene was removed, and the residue was dissolved
in water, basified with K2CO3, and extracted with CH2Cl2. The
extract was concentrated after drying with anhydrous MgSO4,
and the residue was chromatographed over silica gel (1:30
ethyl acetate:hexane) to afford 17b (109 mg, 82%). 1H NMR
(CDCl3, 300 MHz): δ 7.31-7.20 (m, 15H), 5.45 (s, 1H), 3.47
(s, 1H), 3.41 (s, 2H), 2.64 (dd, J ) 4.2, 11.1 Hz, 2H), 2.23 (br.
s, 2H), 1.99 (d, J ) 10.8 Hz, 2H), 1.87 (m, 2H), 1.73 (m, 2H).
13C NMR (CDCl3, 75 MHz): δ 142.9, 139.5, 128.6, 128.2, 127.1,
126.7, 85.8, 79.5, 61.5, 58.3, 39.5, 26.8. HRMS (EI) calcd for
(14)-HCl. mp 214-215 °C. 1H NMR (D2O, 300 MHz): δ
7.38 (s, 5H), 7.30-7.14 (m, 5H), 4.20 (s, 2H), 3.34 (dd, J ) 3.3,
12.9 Hz, 2H), 3.22 (d, J ) 12.6 Hz, 2H), 2.97 (s, 1H), 2.72 (br
s, 2H), 1.67 (m, 2H), 1.54 (m, 2H). Anal. (C20H24NCl‚0.3H2O):
C, H, N, Cl.
1
N-Ben zyl-3-a za bicyclo[3.2.1]-8â-br om oocta n e (15). H
NMR (CDCl3, 300 MHz): δ 7.28 (m, 5H), 4.14 (s, 1H), 3.43 (s,
2H), 2.71 (ddd, J ) 1.2, 4.2, 9.9 Hz, 2H), 3.36 (m, 2H), 2.19 (d,
J ) 10.8 Hz, 2H), 2.04 (m, 2H), 1.81 (d, J ) 9.0 Hz, 2H). MS
(CI) calcd for C14H18NBr m/z, 279; found, 280.1 (M + 1).
(15)-HCl. mp 218-219 °C. 1H NMR (D2O, 300 MHz): δ
7.35 (m, 5H), 4.27 (s, 1H), 4.13 (s, 2H), 3.26 (dd, J ) 3.0, 13.2
Hz, 2H), 3.16 (d, J ) 12.6 Hz, 2H), 2.51 (br s, 2H), 2.15 (m,
2H), 1.62 (d, J ) 8.7 Hz, 2H). Anal. (C14H19NBrCl): C, H, N,
Br, Cl.
C
27H29NO m/z, 383.2249; found, 383.2252 (M+).
(17b)-HCl. mp >172.5-174 °C. 1H NMR (D2O, 300 MHz):
δ 7.33-7.20 (m, 15H), 5.55 (s, 1H), 4.05 (s, 2H), 3.71 (s, 1H),
3.20 (d, J ) 10 Hz, 2H), 2.90 (d, J ) 13.2 Hz, 2H), 2.44 (br s,
2H), 1.92 (m, 2H), 1.55 (m, 2H). Anal. (C27H30NOCl‚0.06HCl):
C, H, N, Cl.
N-Ben zyl-3-azabicyclo[3.2.1]octan e (16). 1H NMR (CDCl3,
300 MHz): δ 7.31-7.06 (m, 5H), 3.46 (s, 2H), 2.62 (d, J ) 9.3

Site Specific Cocaine Abuse Treatment Agents
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 8 1463
(7) (a) Deutsch, H. M.; Collard, D. M.; Zhang, L.; Burnham, K. S.;
Deshpande, A. K.; Holtzman, S. G.; Schweri, M. M. Synthesis
and Pharmacology of Site-Specific Cocaine Abuse Treatment
Agents: 2-(Aminomethyl)-3-phenylbicyclo[2.2.2]alkane and -[2.2.1]-
alkane Dopamine Inhibitors. J . Med. Chem. 1999, 42, 882-895.
(b) J avanmard, S.; Deutsch, H. M.; Collard, D. M.; Kuhar, M.
J .; Schweri, M. M. Synthesis and Pharmacology of Site-Specific
Cocaine Abuse Treatment Agents: 2-Substituted-6-amino-5-
phenylbicyclo[2.2.2]octanes. J . Med. Chem. 1999, 42, 4836-4843.
(8) Kleber, H. D.; Gawin, F. H. Cocaine: Pharmacology, Effects, and
Treatment of Abuse. NIDA Res. Monogr. 1984, 50, 111-129.
(9) Carroll, F. I.; Gao, Y.; Rahman, M. A.; Abraham, P.; Parham,
K.; Lewin, A. H.; Boja, J . W.; Kuhar, M. J . Synthesis, Ligand
Binding, QSAR, and CoMFA Study of 3â-(p-Substituedphenyl)-
tropane-2â-carboxylic Acid Methyl Esters. J . Med. Chem. 1991,
34, 2719-2725.
(10) (a) Meltzer, P. C.; Blundell, P.; Yong, Y. F.; Chen, Z.; George,
C.; Gonzalez, M. D.; Madras, B. K. 2-Carbomethoxy-3-aryl-8-
bicyclo[3.2.1]octanes: Potent Non-Nitrogen Inhibitors of Monoam-
ine Transporters. J . Med. Chem. 2000, 43, 2982-2991. (b)
Meltzer, P. C.; Liang, A. Y.; Blundell, P.; Gonzalez, M. D.; Chen,
Z.; George, C.; Madras, B. K. 2-Carbomethoxy-3-aryl-8-oxabicyclo-
[3.2.1]octanes: Potent non-nitrogen inhibitors of monoamine
transporters. J . Med. Chem. 1997, 40, 2661-2673.
(11) Kozikowski, A. P.; Araldi, G. L.; Boja, J .; Meil, W. M.; J ohn, K.
M.; Flippen-Anderson, J . L.; George, C.; Saiah, E. Chemistry and
Pharmacology of the Piperidine-Based Analogues of Cocaine.
Identification of Potent DAT Inhibitors Lacking the Tropane
Skeletone. J . Med. Chem. 1998, 41, 1962-1969.
(12) Newman, A. H.; Allen, A. C.; Izenwasser, S.; Katz, J . L. Novel
3R-(Diphenylmethoxy)tropane Analogues: Potent Dopamine
Uptake Inhibitors without Cocaine-like Behavioral Profiles. J .
Med. Chem. 1994, 37, 2258-2261.
(13) Lowe, J . A., III; Drozda, S. E.; Mclean, S.; Bryce, K. D.; Crawford,
R. T.; Snider, R. M.; Longo, K. P.; Nagahisa, A.; Tuschiya, M.
Aza-Tricyclic Substance P Antagonists. J . Med. Chem. 1994, 37,
2831-2840.
(14) House, H. O.; Bryant, W. M., III. The Addition of Organometallic
Reagents to Azabicyclic Ketones. J . Org. Chem. 1965, 30, 3634-
3642.
(15) Lee, H.-Y.; Lee, S.; Kim, D.-I.; Kim, B. G.; Bahn, J . S.; Kim, S.
Total Synthesis of R-Cedrene; A New Strategy Utilizing N-
Aziridine Radical Chemistry. Tetrahedron Lett. 1998, 39, 7713-
7716.
(16) Olofson, R. A.; Martz, J . T.; Senet, J .-P.; Piteau, M.; Malfroot,
T. A New Reagent for the Selective, High Yield N-Dealkylation
of tertiary Amines; Improved Syntheses of Naltrexone and
Nalbuphine. J . Org. Chem. 1984, 49, 2081-2082.
(17) CS Chem 3D Pro 4.0; Cambridge Soft Corp.: Cambridge, MA.
(18) Karplus, M. Vicinal Proton Coupling in Nuclear Magnetic
Resonance. J . Am. Chem. Soc. 1963, 85, 2870-2871.
(19) 19. Meltzer, P. C.; Liang, A. Y.; Madras, B. K. The Discovery of
an Unusually Selective and Novel Cocaine Analog: Difluoropine.
Synthesis and Inhibition of Binding at Cocaine Recognition Sites.
J . Med. Chem. 1994, 37, 2001-2010.
(20) 20. (a) Reith, M. E. A.; Selmeci, G. Radiolabeling of dopamine
uptake sites for cocaine, mazindaol, and GBR 12935. Naunyn-
Schmiedberg’s Arch. Pharmacol. 1992, 345, 309-318. (b) Schw-
eri, M. M.; J acobson, A. E.; Lessor, R. A.; Rice, K. C. Metaphit
inhibits dopamine transport and binding of [3H]methylpheni-
date, a proposed marker for the dopamine transport complex.
Life Sci. 1989, 45, 1689-1698.
(21) Schweri, M. M.; Deutsch, H. M.; Massey, A. T.; Holtzman, S. G.
Biochemical and behavioral characterization of novel methyl
phenidate analogues. J . Pharmcol. Exp. Ther. 2002, 301, 527-
535.
(22) Boja, J . W.; Kuhar, M. J .; Kopajtic, T.; Yang, E.; Abraham, P.;
Lewin, A. H.; Carroll, F. I. Secondary Amine Analogues of 3â-
(4′-Substituted phenyl) tropane-2â-carboxylic Acid Esters and
N-norcocaine Exhibit Enhanced Affinity for Serotonin and
Norepinepherin Transporters. J . Med. Chem. 1994, 37, 1220-
1223.
N -Be n zyl-8r-d ip h e n ylm e t h oxy-3-a za b icyclo[3.2.1]-
octa n e (17a ). Using a similar coupling procedure to prepare
ether 17b, compound 13a was coupled with benzhydrol to
afford 17a (79%). 1H NMR (CDCl3, 300 MHz): δ 7.41-7.20
(m, 15H), 5.45 (s, 1H), 3.56 (t, J ) 4.5 Hz, 1H), 3.54 (s, 2H),
2.77 (d, J ) 10.2 Hz, 2H), 2.41 (dd, J ) 3.3, 10.2 Hz, 2H), 2.03
(br. s, 2H), 1.77 (m, 2H), 1.56 (m, 2H).
1
(17a )-HCl. mp >172.5-174 °C. H NMR (D2O, 300 MHz):
δ 7.37-6.96 (m, 15H), 5.35 (s, 1H), 4.07 (s, 2H), 3.55 (t, J )
4.8 Hz, 1H), 3.31 (d, J ) 12.0 Hz, 2H), 2.93 (dd, J ) 2.7, 12.6
Hz, 2H), 2.09 (br s, 2H), 1.64-1.48 (m, 4H). Anal. (C27H30NOCl‚
0.14H2O): C, H, N, Cl.
P h a r m a cology. [3H]WIN 35,428 Bin d in g. The synthe-
sized compounds were screened for activity in a striatal
synaptosomal membrane preparation by a previously described
method.7a,20
[3H]DA Up ta k e. Accumulation of [3H]DA was determined
as previously described.20
[3H]Citalopr am Bin din g to th e Tr an spor ter . The method
used here was as described previously.21 Briefly, a P2 fraction
prepared as described above from selected rat cortical tissue
was homogenized in 16 volumes of ic-cold assay buffer (25 mM
sodium phosphate buffer, pH 7.7, at 0 °C). Binding was
initiated by addition of 150 µL of the P2 suspension to samples
containing 750 µL of assay buffer, 50 µL of the test compound,
25 µL of water or clomipramine (to define nonspecific binding;
final concentration, 1 µM), and 25 µL of [3H]CIT (final
concentration, 2 nM). Test compounds were dissolved in water,
dilute DMSO, or dilute methanol, with the concentration of
organic solvent in any assay tube limited to 0.3 vol %. Usually,
a range of seven drug concentrations was used to generate the
dose-response curve; triplicate samples were run at each
concentration. The samples were incubated for 3 h at 0 °C.
The reaction was terminated by addition of 5 mL of ice-cold
assay buffer to the assay tubes, followed by rapid filtration
through glass microfiber filter strips (presoaked in 0.05%
polyethyleneimine) under reduced pressure using a Brandel
30 port cell harvester (Brandel Inc., Gaithersburg, MD). Assay
tubes were washed with an additional 5 mL aliquot of ice-
cold assay buffer. The filters were extracted and counted for
radioactivity as described above. After correction of the data
for displacement of filter binding by clomipramine and/or the
test drug, IC50 values and Hill coefficients were determined
as described above. The dose-response curves usually con-
tained a range of seven concentrations of the test drug, with
triplicate determinations made at each concentration.
Ack n ow led gm en t. This work was supported in part
by grants from the National Institute of Drug Abuse,
DA 06305 and DA 11541. We are particularly grateful
to Dr. Amy Hauck Newman, NIDA, for her helpful
suggestion. We thank Ms. Monica Stafford, Ms. Le-
keisha Nelson, and Mr. Eric Shields for their assistance
in the biochemical assays.
Refer en ces
(1) J ohnston, L. D.; O’Malley, P. M.; Bachman, J . G. The Monitoring
the Future National Survey Results on Adolescent Drug Use:
Overview of Key Findings, 2001; NIH Publication No. 02-5105;
National Institute on Drug Abuse: Bethesda, MD, 2002; pp 61.
(2) NIDA Research ReportsCocaine Abuse and Addiction; NIH
Publication No. 99-4342; National Institute on Drug Abuse:
Bethesda, MD, May 1999.
(23) (a) Holmquist, C. R.; Keverline-Frantz, K. I.; Abram, P.; Boja,
J . W.; Kuhar, M. J .; Carroll, F. I. 3R-(4′-Substituted phenyl)-
tropane-2â-carboxylic Acid Methyl Esters: Novel Ligands with
High Affinity and Selectivity at the Dopamine Transporter. J .
Med. Chem. 1996, 39, 4139-4141. (b) Carroll, F. I.; Lewin, A.
H.; Boja, J . W.; Kuhar, M. J . Cocaine Receptor: Biochemical
Characterization and Structure-Activity Relationships of Co-
caine and Analogues at the Dopamine Transporter. J . Med.
Chem. 1992, 35, 969-981.
(24) Deutsch, H. M.; Shi, Q.; Gruszecka-Kowalik, E.; Schweri, M. M.
Synthesis and Pharmacology of Potential Cocaine Antagonists.
2. Structure-Activity relationship Studies of Aromatic Ring-
Substituted Methylphenidate Analogues. J . Med. Chem. 1996,
39, 1201-1209.
(3) J ohnston, L. D.; O’Malley, P. M.; Bachman, J . G. Monitoring
the Future National Survey Results on Drug Use, 1975-2001;
NIH Publication No. 02-5106; National Institute on Drug
Abuse: Bethesda, MD, 2002; Vol. I: Secondary School Students,
503 pp.
(4) Singh, S. Chemistry, Design, and Structure-Activity Relation-
ship of Cocaine Antagonists. Chem. Rev. 2000, 100, 925-1024.
(5) Ritz, M. C.; Lamb, R. J .; Goldberg, S. R.; Kuhar, M. J . Cocaine
receptors on dopamine transporters are related to self-admin-
istration of cocaine. Science 1987, 237, 1219-1223.
(6) Rocha, B. A.; Scearce-Levie, K.; Lucas, J . J .; Hiroi, N.; Castanon,
N.; Crabbe, J . C.; Nestler, E. J .; Hen, R. Increased vulnerability
to cocaine in mice lacking the serotonin-1B receptor. Nature
1998, 393, 175-178.

1464 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 8
Kim et al.
(25) (a) Rothman, R. B.; Becketts, K. M.; Radesca, L. R.; De Costa,
B. R.; Rice, K. C.; Carroll, F. I.; Dersch, C. M. Studies of the
biogenic amine transporters. II. A brief study on the use of [3H]-
DA-uptake-inhibition to transporter-binding-inhibition ratios for
the in vitro evaluation of putative cocaine antagonists. Life Sci.
1993, 53, PL267-PL272. (b) Xu, C.; Coffey, L. L.; Reith, M. E.
A. Translocation of dopamine and binding of 2â-carbomethoxy-
3â-(4-fluorophenyl)tropane (WIN35,428) measured under identi-
cal conditions in rat striatal synaptosomal preparations. Inhi-
bition by various blockers. Biochem. Pharmacol. 1995, 49, 339-
350.
(26) (a) Woolverton, W. L.; Hecht, G. S.; Agoston, G. E.; Katz, J . L.;
Newman, A. H.; et al. Further studies of the reinforcing effects
of benzotropine analogs in rhesus monkeys. Psychopharmacology
(Berlin) 2001, 154, 375-382. (b) Katz, J . L.; Agoston, G. E.;
Alling, K. L.; Kline, R. H.; Forster, M. J .; Woolverton, W. L.;
Kopajtic, T. A.; Newman, A. H. Dopamine transporter binding
without cocaine-like behavioral effects: synthesis and evaluation
of benztropine analogues alone and in combination with cocaine
in rodents. Psychopharmacology (Berlin) 2001, 154, 362-374.
J M020410T