2-carbomethoxy-3-(diarylmethoxy)-1αH,5αH-tropane analogs: Synthesis and inhibition of binding at the dopamine transporter and comparison with piperazines of the GBR series

Article abstract of DOI:10.1021/jm950463t

We recently reported a new class of tropanes, based on benztropine, that bind uniquely, in the S-configuration, to the dopamine transporter. We have now extended this series to evaluate the effects of substituents on the nitrogen and the diarylmethoxy group. Herein we have described the synthesis and biological evaluation of a series of 2-carbomethoxy-3-(diaryl-methoxy)- 1αH,5αH-tropane (2-carbomethoxybenztropine) analogs. Examination of the binding data obtained for these compounds shows that while the 4,4'-difluoro compound is potent and selective for the dopamine transporter, introduction of larger groups such as 4,4'-dichloro, 4,4'-dibromo, 4,4'-diiodo, or 4,4'- dimethyl on the 3-diphenylmethoxy moiety reduces this potency. However, although introduction of only one group (e.g., 4-chloro, 4-bromo, 4-iodo, or 4-methyl) leads to a similar reduction of binding affinity, these monosubstituted 2-carbomethoxybenztropines are significantly more potent than the related disubstituted compounds. Finally, from the data for the N- substituted 2-carbomethoxybenztropine analogs, it is evident that steric bulk can be tolerated at the nitrogen site. A comparison of structure-activity relationship data for the tropanes, GBR analogs, and these benztropines indicates that the 2-carbomethoxy-benztropine analogs may be more like the GBR analogs in their mode of binding to the dopamine transporter than like the tropanes. This conclusion supports the notion that the binding site for (-)-cocaine [and the (1R)-tropanes] may differ from that of the 2- carbomethoxybenztropine analogs.

Full text of DOI:10.1021/jm950463t

J . Med. Chem. 1996, 39, 371-379
371
2-Ca r bom eth oxy-3-(d ia r ylm eth oxy)-1rH,5rH-tr op a n e An a logs: Syn th esis a n d
In h ibition of Bin d in g a t th e Dop a m in e Tr a n sp or ter a n d Com p a r ison w ith
P ip er a zin es of th e GBR Ser ies
Peter C. Meltzer,*,† Anna Y. Liang,† and Bertha K. Madras‡
Organix Inc., 65 Cummings Park, Woburn, Massachusetts 01801, and Department of Psychiatry, Harvard Medical School and
New England Regional Primate Center, Southborough, Massachusetts 01772
Received J une 23, 1995X
We recently reported a new class of tropanes, based on benztropine, that bind uniquely, in the
S-configuration, to the dopamine transporter. We have now extended this series to evaluate
the effects of substituents on the nitrogen and the diarylmethoxy group. Herein we have
described the synthesis and biological evaluation of a series of 2-carbomethoxy-3-(diaryl-
methoxy)-1RH,5RH-tropane (2-carbomethoxybenztropine) analogs. Examination of the binding
data obtained for these compounds shows that while the 4,4′-difluoro compound is potent and
selective for the dopamine transporter, introduction of larger groups such as 4,4′-dichloro, 4,4′-
dibromo, 4,4′-diiodo, or 4,4′-dimethyl on the 3-diphenylmethoxy moiety reduces this potency.
However, although introduction of only one group (e.g., 4-chloro, 4-bromo, 4-iodo, or 4-methyl)
leads to a similar reduction of binding affinity, these monosubstituted 2-carbomethoxybenz-
tropines are significantly more potent than the related disubstituted compounds. Finally, from
the data for the N-substituted 2-carbomethoxybenztropine analogs, it is evident that steric
bulk can be tolerated at the nitrogen site. A comparison of structure-activity relationship
data for the tropanes, GBR analogs, and these benztropines indicates that the 2-carbomethoxy-
benztropine analogs may be more like the GBR analogs in their mode of binding to the dopamine
transporter than like the tropanes. This conclusion supports the notion that the binding site
for (-)-cocaine [and the (1R)-tropanes] may differ from that of the 2-carbomethoxybenztropine
analogs.
In tr od u ction
sulfones, sufficient to inhibit protonation of the amine
under physiological conditions does not reduce binding
potency. With respect to nitrogen substitution, it had
already been reported11 that introduction of an alkyl or
allyl group did not eliminate binding potency and, more
recently, we23 have shown that introduction of an
iodoallyl group on the tropane nitrogen can lead to
potent and selective compounds for the DAT. Altropane
(Figure 1), a product of that work, is currently undergo-
ing development as a potential SPECT imaging agent.23
This result has been corroborated by the work of
Goodman24 in his synthesis of a chloro analog of
altropane.
The reinforcing properties and stimulant effects of
cocaine have been associated with its propensity to bind
to monoamine transporters, particularly the dopamine
transporter (DAT).1-11 Although structure-activity
relationship (SAR) studies of cocaine and its analogs
have offered insight into the possible mode of cocaine
binding to the DAT, a comprehensive picture of the
binding interaction to the DAT at the molecular level
has yet to emerge. Indeed, even though SAR studies
on the classical tropane analogs of cocaine3,5,6,12-14
appeared to provide a consistent model for this interac-
tion, subsequent studies have revealed inconsistencies
with the initial reports. Thus, Carroll proposed12,15-17
that the molecular requirements for binding of cocaine
and its tropane analogs at the DAT include four factors,
namely, a 2â-carboxy ester, a basic nitrogen capable of
protonation at physiological pH, the R-configuration of
The active (-)-cocaine series, including all the tropane
analogs of WIN 35,428, are of the R-configuration.
Guided by our desire to seek topologically different
tropanes, we recently synthesized and evaluated the
receptor-binding properties of all eight isomers of 2-car-
bomethoxy-3-[bis(4-fluorophenyl)methoxy]tropane and
reported25 that the S-enantiomer, (S)-(+)-2â-carbo-
methoxy-3R-[bis(4-fluorophenyl)methoxy]tropane, 2a (di-
fluoropine or O-620), was considerably more potent (IC50
) 10.9 nM) and selective (DAT over the 5HT transporter
) 324:1) than any of the other seven isomers, including
the R-enantiomers. Until the discovery of difluoropine
(2a ; Figure 1) in 1994, no potent tropanes had been
found in the S-configuration. This compound, a 2â-
carbomethoxy analog of benztropine, is uniquely of the
S-configuration. In further contrast to the classical 3â-
substituted cocaine and tropane analogs, it has a 3R-
diphenylmethoxy substituent. As compound 2a is in the
S-configuration, it has no substituent in the 2â-position
of (1R)-cocaine [or the (1R)-tropanes]; rather, one of the
phenyl rings may topologically occupy that position.
the tropane, and a 3â-aromatic ring at C-3. Later,
18
Davies
reported that introduction of 2â-ketones did
not reduce potency. Kozikowski has since demonstrated
that hydrogen bonding at the C-2 site is not required
since introduction of unsaturated and saturated alkyl
groups19,20 does not diminish binding. Further, doubt
has been cast on the notion of an ionic bond between a
protonated amine (at physiological pH) and the pre-
sumed aspartate residue on the DAT,21 since reduction
of nitrogen nucleophilicity,22 by introduction of N-
* Mail correspondence to Dr. Peter Meltzer, Organix Inc., 65
Cummings Park, Woburn, MA 01801.
† Organix Inc.
‡ Harvard Medical School and New England Regional Primate
Center.
X Abstract published in Advance ACS Abstracts, December 15, 1995.
0022-2623/96/1839-0371$12.00/0 © 1996 American Chemical Society

372 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 2
Meltzer et al.
Sch em e 1
Ch em istr y
The (S)-2-carbomethoxy-3-(diarylmethoxy)-1RH,5RH-
tropanes (2-carbomethoxybenztropines) of this study
were prepared from the alcohol 125 which was reacted
with a series of benzhydrols under azeotropic distillation
in the presence of catalytic p-toluenesulfonic acid to
obtain compounds 2 in moderate yields. In order to
obtain the N-substituted compounds 4, the difluoro
compound 2a was N-demethylated with R-chloroethyl
chloroformate as solvent to provide the nor compound
3 in 58% yield. N-Alkylation of 3 was then achieved
with an appropriate alkyl bromide with KF/Celite in
acetonitrile. Thus, 3 was reacted with benzyl bromide,
1-bromo-3-phenylpropane, 1-bromo-3-(4-bromophenyl)-
propane, and 1-chloro-5-phenylpentane, to yield 4a -c,e,
respectively. Yields of 80-90% were obtained.
The 4-iodo compounds were obtained by conversion
of the corresponding bromo compounds to tributylstan-
nyl derivatives. Reaction with N-iodosuccinimide (NIS)
then gave the iodo compounds. Thus, 2e,k ,l and 4d
were derived from their stannyl precursors upon treat-
ment with NIS in tetrahydrofuran. Stannylation to
provide these precursors was achieved by reaction of
2c,i and 4c with bis(tributyltin) in the presence of
tetrakis(triphenylphosphine)palladium.
F igu r e 1. Three classes of compounds that inhibit binding
of [3H]WIN 35,428 to the dopamine transporter (DAT).
Within the phenyltropane series, halogen and other
substituents on the single aromatic ring have dramatic
effects, not only on the affinity of these compounds for
the DAT but also on their DAT vs serotonin transporter
selectivity.3,5,6,12-14,18-20 As this series has a diphenyl-
methoxy in the 3-position, it is of considerable interest
to determine whether halogen and other substituents
have analogous effects. Equally significant is whether
these effects are dependent upon substitution in each,
or both, of the aromatic rings.
The enantiomeric purity of the end products was
critical. We therefore developed a supplementary NMR
method for assessing enantiomeric purity of 2a . Thus,
The diphenylmethoxy substituent is also present in
another family of potent and selective DAT inhibitors,
namely, the 1,4-dialkylpiperazines, the so-called GBR
series, first reported in 1980.26 Van der Zee had noted
the use of benztropine as an anti-Parkinsonian agent
and prepared this new series in an attempt to enhance
activity by replacement of the tropine moiety in benz-
tropine with a piperazine. This work was a precursor
to our own work in this area.
1
an H-NMR spectrum measured in the presence of 80%
by weight of the chiral shift reagent europium D-3-
(heptafluorobutyryl)camphorate showed two base-line-
separated singlet resonances for the diagnostic diben-
zylic proton at δ 5.93 and 6.07 for the mixture of the
two enantiomers (S)- and (R)-2a . These resonance
positions were concentration dependent, and therefore
the shift experiment must be optimized for the com-
pound and the NMR field strength. The pure (S)-2a
showed only a singlet at δ 6.0 under these conditions.
It should be noted that compounds in which each of
the diphenylmethoxy rings were differently substituted,
such as 2g-m , possess a newly introduced benzylic
chiral center. These 1S-compounds were not resolved
again, and their receptor binding was evaluated as
We now report the synthesis and receptor-binding
properties of a series of analogs of 2a . These analogs
have been designed to probe the effect of substitution
on the aromatic rings of the diphenylmethoxy group
present in 2a (Table 1) and also to explore the effects
of introduction of arylalkyl groups on the tropane
nitrogen in order to create analogs of the GBR structure
(Table 2).

2-Carbomethoxy-3-(diarylmethoxy)tropane Analogs
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 2 373
diastereomeric pairs. Their NMR spectra proved in-
structive. Whereas the methoxy resonance for sym-
metric compounds such as 2a -f appeared as a singlet
at δ 3.7, that resonance appeared as two distinct singlets
at about δ 3.7 for the diastereomeric pairs.
The ethylphenyl compound 2m and the fluorenyloxy
compound 2n were synthesized in order to examine the
need for two aromatic rings and the effect of a planar
system at this site, in comparison with the angled
diphenylmethoxy, respectively. Both compounds were
prepared analogously, albeit in much poorer yields, to
those presented above; the appropriate alcohol was
reacted with the intermediate 1. Thus 2m was obtained
in 14% yield and 2n in 29% yield.
Biology
The affinities (IC50) of the difluoropine (2a ) analogs
for the dopamine and serotonin transporters were
determined in competition studies using [3H]-3â-(4-
fluorophenyl)tropane-2â-carboxylic acid methyl ester
([3H]WIN 35,428 or [3H]CFT) to label the DAT5 and [3H]-
citalopram to label the serotonin transporter. The data
are presented in Tables 1 and 2. Studies were con-
ducted in cynomolgus monkey striatum for two reasons.
First, these compounds are part of an ongoing investi-
gation of SARs at the DAT in this tissue,5,14,25 and
meaningful comparisons with an extensive data base
can be made. Equally important, cynomolgus monkeys
are to be used as subjects for PET (positron emission
tomography) imaging with some targeted compounds in
the course of this program.
F igu r e 2. Alternate representation of difluoropine to show
possible overlap with GBR 12909.
and the benztropines have proven to be potent inhibitors
of the DAT.
The structural similarity between compounds in the
GBR series and the benztropine analogs is instructive.26
Most notable is the common diphenylmethoxy function-
ality. It is intriguing to speculate how these two classes
of compounds may bind similarly. Specifically, Dut-
ta28,29 has evaluated the contribution of each of the two
nitrogen atoms in the piperidine ring of the GBR series
for conferring binding potency. The nitrogen distal to
the diphenylmethoxy substituent was found to confer
greater binding potency than did the proximal nitrogen.
If 2a and GBR 12909 are superimposed using the
common diphenylmethoxy substituents of each com-
pound, then the tropane nitrogen of 2a and the distal
nitrogen (shown in boldface) of GBR 12909 (Figure 2)
occupy the same region of space.
Competition studies were conducted with a fixed
concentration of radioligand and a range of concentra-
tions of the test drug. All drugs inhibited [3H]WIN
35,428 and [3H]citalopram binding in a concentration-
dependent manner.
Discu ssion
Many structurally different compounds (including
cocaine, tropanes, GBR compounds, mazindol, and
methylphenidate) have proven to be potent inhibitors
of [3H]cocaine-binding sites in striatum. In this brain
region the majority of the cocaine- or WIN35,428-
binding sites are associated with the DAT. Three
classes of compounds (Figure 1) have furnished par-
ticularly potent ligands and have provided the spring-
board for the work presented here. Specifically, the
(1R)-tropanes (C-3-aryl analogs), the piperazines (GBR),
These observations support the notion that multiple
binding sites exist on the DAT such that binding of the
2-carbomethoxybenztropines and GBR analogs does not
fit the binding domain in exactly the same way as does
cocaine or the (1R)-tropanes. This interpretation lends
Sch em e 2

374 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 2
Meltzer et al.
Ta ble 1. Inhibition of [3H]WIN 35,428 Binding to the DAT and [3H]Citalopram Binding to the Serotonin Transporter by Analogs of
2a in Cynomolgus Monkey Caudate-putamena
IC50 (nM)
selectivity
compd
benztropine
O-672, difluorobenztropine
O-620, 2a
O-651, 2b
O-734, 2c
O-652, 2d
O-743, 2e
O-701, 2f
O-810, 2g
R1
R2
DAT
5HT
(DAT/5HT)
312 ( 1.1
44.6 ( 8.37
10.9 ( 1.2
33.5 ( 4.5
72.0 ( 3.65
91.4 ( 0.85
909 ( 79
240 ( 18.4
13.2 ( 1.9
15.8 ( 0.95
24.0 ( 4.6
49.5 ( 6.0
55.9 ( 10.3
389 ( 29.4
1480 ( 569
6030 ( 815
20.3 ( 3.54
24 100 ( 14 800
5340 ( 2110
3530 ( 1480
10 100 ( 1740
2430 ( 339
3360 ( 1480
8550 ( 442
77
120
324
302
34
37
9.4
41
373
377
240
266
166
11
F
H
Br
Cl
I
F
H
Br
Cl
I
CH3
F
Cl
Br
CH3
I
CH3
H
9800 ( 2680
4930 ( 1200
5960 ( 467
O-819, 2h
O-723, 2i
O-820, 2j
O-739, 2k
H
H
H
H
5770 ( 493
13 200 (n ) 1)
9280 ( 1640
4390 ( 829
O-738, 2l
Br
I
O-784, 2m b
O-660, 2n c
O-657, 3d d
ethyl
fluorenoxy
F
21 000 ( 5300
3440 ( 542
14
0.6
F
a
Tissue (4 mg/mL original wet tissue weight) was incubated with each radioligand and 7-14 concentrations of a cocaine congener.
Nonspecific binding of [3H]WIN 35,428 was measured with 30 µM (-)-cocaine and of [3H]citalopram with 1 µM fluoxetine. IC50 values
were computed by the EBDA computer program and are the means ((SD or SEM) of 2-4 independent experiments, each conducted in
triplicate. One aromatic ring has been replaced by an ethyl group. c The diphenylmethoxy is replaced with a fluorenyloxy; cpd is racemic.
b
d
This is N-demethylated 2a .
hope for the discovery of a cocaine antagonist which
displays reduced influence on the uptake of dopamine
itself.25
tonin transporter (IC50 ) 10 µM) also decreased. The
resulting transporter selectivity was similar for the two
compounds. The dibromo- (2c, IC50 ) 72.0 nM), dichlo-
ro- (2d , IC50 ) 91.4 nM), diiodo- (2e, IC50 ) 909 nM)
and dimethyl- (2f, IC50 ) 240 nM) substituted deriva-
tives were less potent than the difluoro or unsubstituted
compounds. The resulting rank order of potency of the
disubstituted compounds was diF > diH > diBr > diCl
> diCH3 > diI.
The rank order for the monosubstituted compounds
2g-k was F > Cl > Br > H > CH3 > I, and a
comparison of the mono- vs dihalogenated molecules
proved interesting. The mono- (2g) and di- (2a ) fluori-
nated compounds were approximately equipotent (IC50
The goal of this study was therefore 2-fold. First,
to explore the SAR of these new 2-carbomethoxybenz-
tropine analogs, and second, to explore their possible
relationship to compounds in the GBR series. We were
guided by both the SAR in the tropane area14,15 as well
as the SAR in the area of the GBR analogs.26,30,31
Therefore we synthesized the compounds presented in
Tables 1 and 2. All compounds were prepared in the
1S-series since we have shown unequivocally25 that in
the 2-carbomethoxybenztropine series, the R-configured
compounds are relatively inactive.
An examination of the binding data presented in
Table 1 shows that the most potent compound for
binding to the DAT, in this series, was the difluoro
analog 2a . This compound also manifested excellent
selectivity for the DAT over the 5HT receptor (324:1),
similar to the less potent (IC50 ) 33.5 nM) compound
2b. Compound 2a was previously reported to be a
potent (IC50 ) 10.9 nM) and selective (DAT:5HT ) 324:
1) ligand at the dopamine transporter.25 The effects of
other halogen or methyl substituents on the diphenyl-
methoxy rings were now investigated under similar
conditions (Table 1). In the absence of the fluoro groups,
binding affinity (2b, IC50 ) 33.5 nM) for the dopamine
transporter decreased 3-fold and affinity for the sero-
) 10-13 nM). The monochloro compound 2h (IC50 )
15.8 nM) was 5-6 times as potent as the dichloro
compound 2d . Introduction of only one bromine atom
(2i, IC50 ) 24.0 nM) resulted in a compound of retained
substantial potency, while the dibromo compound 2c
was again 3 times less potent. Introduction of an iodine
(2k ) resulted in a further reduction of potency (IC50
55.9 nM), and as expected, the diodo compound 2e was
)
at least 1 order of magnitude less potent. The mono-
methyl (2j, IC50 ) 49.5 nM) and dimethyl (2f, IC50
)
240 nM) compounds displayed a similar preference for
the monosubsituted compound. As may be expected, the
monobromo-monoiodo compound 2l showed potency (IC50
) 389 nM) between that of the diodo and dibromo

2-Carbomethoxy-3-(diarylmethoxy)tropane Analogs
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 2 375
Ta ble 2. Inhibition of [3H]WIN 35,428 Binding to the DAT and
[3H]Citalopram Binding to the Serotonin Transporter by
N-Substituted 2a Analogs in Cynomolgus Monkey
Caudate-putamena
compounds. These results indicate that while a halogen
was preferred, the increase in steric bulk as the halogen
size was increased, and as it was introduced into both
rings, was detrimental to binding to the DAT.
A comparison of selectivity for binding to the DAT
versus binding to the 5HT transporter was interesting.
In the disubstituted series only the difluoro (2a ) and
unsubstituted (2b) compounds displayed selectivity
>300:1, while the dibromo (2c), dichloro (2d ), diiodo
(2e), and dimethyl (2f) compounds showed much poorer
selectivity (<50:1).
In contrast, the monohalogenated analogs all showed
selectivity >150:1, and the monofluoro (2g) and mono-
chloro (2h ) compounds showed the highest selectivity
(>370:1) of all the analogs tested. Thus, it is evident
that, in general, the more active compounds were also
more selective and, furthermore, that introduction of
halogens on both rings caused reduction of selectivity.
Binding affinity was measured on the diastereomeric
mixture for each of the monosubstituted compounds,
and it remains to be seen whether affinity and/or
selectivity for any of these compounds resides predomi-
nantly in one diastereoisomer. It is interesting to note
that the monohalogenated compounds in this series are
significantly more selective for the DAT over the 5HT
transporter than are the classical halogenated tro-
panes.14
IC50 (nM)
selectivity
compd
R
n
DAT
5HT
(DAT/5HT)
O-755, 4a
O-747, 4b
O-786, 4c
O-788, 4d
O-764, 4e
H
1
3
3
3
5
223 ( 53
22.0 ( 11.9
80.2 ( 8.8
119 ( 11
99.0 ( 28
4970 ( 700
19.7 ( 3
234 ( 0.5
2200 ( 1250
550 ( 63
22
0.9
3
19
6
H
Br
I
H
a
Tissue (4 mg/mL original wet tissue weight) was incubated
with each radioligand and 7-14 concentrations of a cocaine
congener. Nonspecific binding of [3H]WIN 35,428 was measured
with 30 µM (-)-cocaine and of [3H]citalopram with 1 µM fluoxetine.
IC50 values were computed by the EBDA computer program and
are the means ((SD or SEM) of 2-4 independent experiments,
each conducted in triplicate.
fore it is likely that the manner in which the aryl rings
of the (1S)-2-carbomethoxybenztropines and the GBR
analogs interact at the receptor site differs from that
interaction of the classical C-3 aryl (1R)-tropanes as well
as from the unsubstituted benztropines. Specifically,
steric bulk is less well tolerated in the diphenylmethoxy
analogs. In contrast, it is favored in the C-3 (1R)-aryl
tropane series.
Substituents on the nitrogen of the GBR analogs and
the 2-carbomethoxybenztropines showed some sililari-
ties. Comparison of the relative binding potencies of
the compounds of different alkyl chain length on the
nitrogen, reported here, with those GBR analogs re-
ported by Van der Zee showed that the optimum chain
had a three-carbon link in both series. Thus in the GBR
series, potency increases as the chain length increases
from one carbon to three or four carbons.26 In the
2-carbomethoxybenztropine series, we prepared com-
pounds with one, three, and five methylene groups in
the chain. Here, the one-carbon linker was the least
potent, while the three-carbon linker was optimum (4b,
IC50 ) 22.0 nM). The five-carbon linker was 5-fold
weaker (4e, IC50 ) 99.0 nM) (Table 2) than the three-
carbon linker.
Two compound are of particular interest. Compound
2m was prepared in order to determine the neccessity
for the two aromatic rings of the diphenylmethoxy
system. Clearly, substitution of one ring by an ethyl
group is deleterious to binding to the DAT. Compound
2n was prepared to investigate the effect of planarity
on the diaromatic system. A planar moiety substan-
tially decreased the molecule’s ability to bind to the
DAT.
Substituents on the nitrogen of the tropane ring had
various effects on dopamine and serotonin transporter
affinity and selectivity (Table 2). In a comparison of
three compounds in which the aromatic ring was un-
substituted but the length of the spacer arm was varied,
the three-spacer carbon chain, 4b, had relatively high
affinity for the dopamine transporter (IC50 ) 22 nM)
compared with the five-carbon (4e, IC50 ) 99 nM) or
one-carbon (4a , IC50 ) 223 nM) spacer chain. Iodo (4d )
or bromo (4c) substituents on the aromatic ring of the
three-spacer compound decreased affinity ca. 4-5-fold.
In all cases, selectivity for the DAT had been reduced
compared with the parent compound 2a .
A comparison of SAR in the tropanes, benztropines,
and GBR series is interesting. While data from [3H]-
WIN 35,428 binding (tropanes and benztropines) and
those from inhibition of dopamine uptake26 (GBR ana-
logs) are not fully comparable because these are two
different biological assays, nevertheless, the relative
potency of the compounds is still instructive.30 The
relative potency of the unsubstituted and chlorine- and
fluorine-substituted compounds revealed that the rank
order of potency for the disubstituted (1S)-2-carbomethox-
ybenztropine analogs 2a ,b,d (F > H > Cl) was similar
to the order for the disubstituted GBR analogs (F = H
> Cl).26 In sharp contrast, the order for the C-2-
unsubstituted benztropines (F > Cl > H)27 and the C-3
(1R)-aryl tropanes (Cl > F > H)14 differed from each
other and from the 2-carbomethoxybenztropines. There-
Substitution on the aryl group attached to the nitro-
gen linker affected binding potency markedly. While
the unsubstituted compound 4b had an IC50 ) 22.0 nM,
introduction of a 4-bromo substituent (4c) reduced this
affinity 4-fold, and the affinity of the 4-iodo derivative
4d was reduced by 5-fold. A similar differentiation is
not as obvious in the GBR series, since the analogous
compounds have similar IC50s in the range of 1.8-2.0
nM.
Con clu sion s
Herein we have described the synthesis and biological
evaluation of a series of 2-carbomethoxybenztropine
analogs. Examination of the binding data obtained for

376 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 2
Meltzer et al.
1H, H-5), 3.59 (d, 1H, H-1), 3.69 (s, 3H, OCH3), 4.00 (d, 1H,
H-3), 5.36 (s, 1H, OCH), 6.90-7.36 (m, 8H, ArH); analytical
HPLC (Chiralcel OC column, eluting with hexane/2-propanol,
these compounds shows that while the 4,4′-difluoro
compound 2a is potent and selective for the dopamine
transporter, introduction of larger groups such as 4,4′-
dichloro, 4,4′-dibromo, 4,4′-diiodo, or 4,4′-dimethyl on
the 3-diphenylmethoxy moiety reduces this potency.
However, although introduction of only one group (e.g.,
4-chloro, 4-bromo, 4-iodo, or 4-methyl) also leads to a
reduction of binding affinity compared with 2a , these
monosubstituted 2-carbomethoxybenztropines, with the
exception of the monofluoro analog, are significantly
more potent and selective than the related disubstituted
compounds.
98:2, + 0.1% DEA) tR ) 7.8 min; [R]21 +21.6° (c ) 1, MeOH).
D
Anal. (C23H25NO3F2) C, H, N.
NMR with europium D-3-(heptafluorobutyryl)camphorate
was conducted at 100 MHz. A sample of (S)-2a (5 mg) and
(R)-2a (5 mg) was prepared in CDCl3 (0.5 mL) containing
europium D-3-(heptafluorobutyryl)camphorate (8.1 mg). Two
base-line-separated singlet resonances at δ 6.07 and 5.93 for
the diagnostic dibenzylic protons were apparent. In a 10%
mixture of (R)-2a in (S)-2a , with 52% by weight of shift reagent
in CDCl3 (0.5 mL), the major peak was at δ 5.98 [(S)-2a ] and
the minor at δ 5.85 [(R)-2a ].
(S)-(+)-2â-Ca r b om et h oxy-3r-(d ip h en ylm et h oxy)t r o-
p a n e (2b). Compound 2b was prepared from (S)-alloecgonine
methyl ester, 1, and benzhydrol as described for 2a . A yellow
viscous oil was obtained (74%): Rf 0.52 (EtOAc + 2% NH4-
OH); 1H-NMR (100 MHz, CDCl3) δ 1.62-2.13 (m, 6H), 2.19 (s,
3H, NCH3), 2.76 (s, 1H, H-2), 3.08 (bs, 1H, H-5), 3.56 (bs, 1H,
H-1), 3.66 (s, 3H, OCH3), 4.02 (d, 1H, H-3), 5.38 (s, 1H, OCH),
Finally, from the data reported for the N-substituted
2-carbomethoxybenztropine analogs, it is clear that
substantial steric bulk can be tolerated at the nitrogen
site. A comparison of SAR data for the tropanes, GBR
analogs, and these benztropines indicates that the
2-carbomethoxybenztropine analogs may be more like
the GBR analogs in their mode of binding to the DAT
than like the tropanes. This conclusion would support
the notion that the binding site for (-)-cocaine [and the
(1R)-tropanes] may be different from that of the 2-car-
bomethoxybenztropine analogs. This difference may
possibly lead to the discovery of a cocaine antagonist
which does not effect dopamine reuptake.
7.30 (m, 10H, ArH); [R]21 +36.4° (c ) 1, MeOH). HCl salt:
D
mp 178-179 °C. Anal. (C23H27NO3‚HCl) C, H, N, Cl.
(S )-(+)-2â-Ca r b om e t h oxy-3r-[b is(4-b r om op h e n yl)-
m eth oxy]tr op a n e (2c). Compound 2c was prepared from
(S)-alloecgonine methyl ester, 1, and 4,4′-dibromobenzhydrol
as described for 2a . A white solid was obtained (68%): mp
119-121°C; Rf 0.55 (EtOAc + 2% NH4OH); 1H-NMR (100 MHz,
CDCl3) δ 1.60-2.15 (m, 6H), 2.20 (s, 3H, NCH3), 2.72 (s, 1H,
H-2), 3.10 (bs, 1H, H-5), 3.58 (bs, 1H, H-1), 3.69 (s, 3H, OCH3),
3.98 (d, 1H, H-3), 5.30 (s, 1H, OCH), 7.05-7.55 (m, 8H, ArH);
Exp er im en ta l Section
[R]21 +18.2° (c ) 1, MeOH). Anal. (C23H25NO3Br2) C, H, N,
D
NMR spectra were recorded on either a Bruker 100, a
Varian XL 400, or a Bruker 300 NMR spectrometer. TMS was
used as internal standard. Melting points are uncorrected and
were measured on a Gallenkamp melting point apparatus.
Optical rotations were measured at the sodium D line at 21°C
using a J ASCO DIP 320 polarimeter (1 dcm cell). Thin layer
chromatography (TLC) was carried out on Baker Si 250F
plates. Visualization was accomplished with either iodine
vapor, UV exposure, or treatment with phosphomolybdic acid
(PMA). Preparative TLC was carried out on Analtech uni-
plates (silica gel GF 2000 µm). Flash chromatography was
carried out on Baker silica gel (40 mM). Elemental analyses
were performed by Atlantic Microlab, Atlanta, GA. A Beck-
man 1801 scintillation counter was used for scintillation
spectrometry; 0.1% bovine serum albumin and (-)-cocaine
were purchased from Sigma Chemicals.
[3H]WIN 35,428 [[3H]CFT, 2â-carbomethoxy-3â-(4-fluoro-
phenyl)-N-[3H]methyltropane, 79.4-87.0 Ci/mmol] and [3H]-
citalopram (86.8 Ci/mmol) were purchased from DuPont-New
England Nuclear (Boston, MA). (R)-(-)-Cocaine hydrochloride
for the pharmacological studies was donated by the National
Institute on Drug Abuse (NIDA). Fluoxetine was donated by
E. Lilly & Co. HPLC analyses were carried out on a Waters
510 system with detection at 254 nm on a Chiralcel OC column
(flow rate, 1 mL/min).
Br.
(S )-(+)-2â-Ca r b om e t h oxy-3r-[b is(4-ch lor op h e n yl)-
m eth oxy]tr op a n e (2d ). Compound 2d was prepared from
(S)-alloecgonine methyl ester, 1, and 4,4′-dichlorobenzhydrol
as described for 2a . An off-white solid was obtained (73%):
mp 110-112 °C; Rf 0.55 (EtOAc + 2% NH4OH); 1H-NMR (100
MHz, CDCl3) δ 1.62-2.13 (m, 6H), 2.18 (s, 3H, NCH3), 2.72
(s, 1H, H-2), 3.09 (bs, 1H, H-5), 3.58 (bs, 1H, H-1), 3.69 (s, 3H,
OCH3), 3.98 (d, 1H, H-3), 5.33 (s, 1H, OCH), 7.27 (m, 8H, ArH);
[R]21 +20.7° (c ) 1, MeOH). Anal. (C23H25NO3Cl2) C, H, N,
D
Cl.
(S )-(+)-2â-Ca r b om e t h oxy-3r-[b is(4-m e t h ylp h e n yl)-
m eth oxy]tr op a n e (2f). Compound 2f was prepared from (S)-
alloecgonine methyl ester, 1, and 4,4′-dimethylbenzhydrol as
described for 2a . A viscous oil was obtained (78%): Rf 0.55
1
(EtOAc + 2% NH4OH); H-NMR (100 MHz, CDCl3) δ 1.65-
2.13 (m, 6H), 2.19 (s, 3H, NCH3), 2.30 (s, 6H, ArCH3), 2.76 (s,
1H, H-2), 3.08 (bs, 1H, H-5), 3.55 (bs, 1H, H-1), 3.69 (s, 3H,
OCH3), 4.01 (d, 1H, H-3), 5.32 (s, 1H, OCH), 7.02-7.32 (m,
8H, ArH). HCl salt: mp 164-166 °C; [R]21 +35.7 ° (c ) 1,
D
MeOH). Anal. (C25H31NO3‚HCl‚0.25H2O) C, H, N, Cl
(S)-(+)-2â-Ca r bom eth oxy-3r-[(4-flu or op h en yl)p h en yl-
m eth oxy]tr op a n e (2g). Compound 2g was prepared from
(S)-alloecgonine methyl ester, 1, and 4-fluorobenzhydrol as
described for 2a . An off-white solid was obtained (78%): mp
115-117 °C; Rf 0.55 (EtOAc + 2% NH4OH); 1H-NMR (100
MHz, CDCl3) δ 1.65-2.15 (m, 6H), 2.18 (s, 3H, NCH3), 2.74
(s, 1H, H-2), 3.08 (bs, 1H, H-5), 3.57 (bs, 1H, H-1), 3.67 (s, 3H,
OCH3), 4.00 (d, 1H, H-3), 5.36 (s, 1H, OCH), 6.97 (t, 2H, ArH),
Gen er a l P r oced u r e for Syn t h esis of t h e Dia r yl-
m eth oxytr op a n es: (S)-(+)-2â-Ca r bom eth oxy-3r-[bis(4-
flu or op h en yl)m et h oxy)]t r op a n e (2a ). (S)-Alloecgonine
methyl ester, 125 (900 mg, 4.5 mmol), 4,4′-difluorobenzhydrol
(1.99 g, 9 mmol), p-toluenesulfonic acid monohydrate (1.3 g,
6.78 mmol), and benzene (50 mL) in a 100 mL round bottom
flask fitted with a Dean-Stark trap and condenser were
heated at reflux for 18 h. More 4,4′-difluorobenzhydrol (1.0
g) and p-toluenesulfonic acid monohydrate (60 mg) were added,
and the reaction mixture was heated at reflux for a further 5
h. The reaction mixture was cooled to room temperature. NH4-
OH (1 mL) was added followed by H2O (20 mL) and EtOAc
(100 mL). The dried (K2CO3) organic layer was concentrated
to dryness. The residue was flash chromatographed over silica
gel (EtOAc, 2% NH4OH, 2% MeOH in EtOAc) to afford (S)-
(+)-2â-carbomethoxy-3R-[bis(4-fluorophenyl)methoxy]tro-
pane, 2a (1.4 g, 78%), as a white solid: mp 130-131°C; Rf 0.53
7.20-7.36 (m, 7H, ArH); [R]21 +21.4° (c ) 1, MeOH). Anal.
D
(C23H26NO3F).
(S)-(+)-2â-Ca r bom eth oxy-3r-[(4-ch lor op h en yl)p h en yl-
m eth oxy]tr op a n e (2h ). Compound 2h was prepared from
(S)-alloecgonine methyl ester, 1, and 4-chlorobenzhydrol as
described for 2a . A yellow viscous oil was obtained (69%): mp
1
(HCl salt) 128-130 °C; Rf 0.60 (EtOAc + NH4OH); H-NMR
(100 MHz, CDCl3) δ 1.65-2.10 (m, 6H), 2.180 (s, 3H, NCH3),
2.74 (s, 1H, H-2), 3.08 (bs, 1H, H-5), 3.56 (bs, 1H, H-1), 3.66
(s, 3H, OCH3), 4.00 (d, 1H, H-3), 5.35 (s, 1H, OCH), 7.27 (bs,
9H, ArH); [R]21 +36.0° (c ) 1, MeOH). Anal. (C23H26NO3-
D
Cl‚HCl‚3/4H2O).
(S )-2â-Ca r b om e t h oxy-3r-[(4-b r om op h e n yl)p h e n yl-
m eth oxy]tr op a n e (2i). Compound 2i was prepared from (S)-
alloecgonine methyl ester, 1, and 4-bromobenzhydrol as de-
1
(EtOAc + 2% NH4OH); H-NMR (100 MHz, CDCl3) δ 1.80-
2.15 (m, 6H), 2.20 (s, 3H, NCH3), 2.73 (s, 1H, H-2), 3.11 (bs,

2-Carbomethoxy-3-(diarylmethoxy)tropane Analogs
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 2 377
scribed for 2a . A pale yellow oil was obtained (77%): Rf 0.65
(EtOAc + 2% NH4OH); H-NMR (100 MHz, CDCl3) δ 1.67-
2.14 (m, 6H), 2.18 (s, 3H, NCH3), 2.74 (s, 1H, H-2), 3.09 (bs,
1H, H-5), 3.58 (bs, 1H, H-1), 3.66, 3.67 (2s, 3H, OCH3), 3.98
(d, 1H, H-3), 5.33 (s, 1H, OCH), 7.15-7.53 (m, 9H, ArH). HCl
salt: mp 120-122 °C. Anal. (C23H26NO3Br‚HCl‚1.5H2O) C,
H, N, Br, Cl.
H-1), 3.68 (s, 3H, OCH3), 3.99 (d, 1H, H-3), 5.26 (s, 1H, OCH),
6.93-7.76 (m, 8H, ArH); [R]21D +16.0° (c ) 0.5, MeOH). Anal.
(C23H25NO3I2) C, H, N, I.
1
(S)-(+)-2â-Ca r b om et h oxy-3r-[(4-iod op h en yl)p h en yl]-
m eth oxy]tr op a n e (2k ). Compound 2k was prepared from
(S)-2â-carbomethoxy-3R-[[4-(tributylstannyl)phenyl]phenyl-
methoxy]tropane as described for 2e. A yellow oil was
obtained (95%): Rf 0.53 (EtOAc + 2% NH4OH); 1H-NMR (100
MHz, CDCl3) δ 1.67-2.18 (m, 6H), 2.22 (s, 3H, NCH3), 2.74
(s, 1H, H-2), 3.12 (bs, 1H, H-5), 3.63 (bs, 1H, H-1), 3.67, 3.69
(2s, 3H, OCH3), 4.02 (d, 1H, H-3), 5.32 (s, 1H, OCH), 7.01-
7.66 (m, 9H, ArH). HCl salt: mp 130-132°C; [R]21D +30.3° (c
) 1, MeOH). Anal. (C23H26NO3I‚HCl‚0.5 H2O) C, H, N, Cl, I.
(S)-(+)-2â-Ca r bom eth oxy-3R-[(4-br om op h en yl)(4-iod o-
p h en yl)m eth oxy]tr op a n e (2l). Compound 2l was prepared
from (S)-2â-carbomethoxy-3R-[(4-bromophenyl)[4-(tributyl-
stannyl)phenyl]methoxy]tropane as described for 2e. A white
solid was obtained (76%): mp 126-128 °C; Rf 0.53 (EtOAc +
2% NH4OH); 1H-NMR (100 MHz, CDCl3) δ 1.68-2.13 (m, 6H),
2.35 (s, 3H, NCH3), 2.78 (s, 1H, H-2), 3.28 (bs, 1H, H-5), 3.73
(bs, 4H, H-1, OCH3), 4.08 (d, 1H, H-3), 5.29 (s, 1H, OCH), 6.94-
(S)-(+)-2â-Ca r bom eth oxy-3r-[(4-m eth ylp h en yl)p h en yl-
m eth oxy]tr op a n e (2j). Compound 2j was prepared from (S)-
alloecgonine methyl ester, 1, and 4-methylbenzhydrol as
described for 2a . A yellow viscous oil was obtained (78%): mp
(HCl salt) 148-150°C; Rf 0.60 (EtOAc + 2% NH4OH); 1H-NMR
(100 MHz, CDCl3) δ 1.70-2.13 (m, 6H), 2.30 (s, 3H, NCH3),
2.76 (s, 1H, H-2), 3.07 (bs, 1H, H-5), 3.56 (bs, 1H, H-1), 3.66
(s, 3H, OCH3), 3.98 (d, 1H, H-3), 5.29 (s, 1H, OCH), 7.04-7.35
(m, 9H, ArH); [R]21 +34.1° (c ) 1, MeOH). Anal. (C24H29
-
D
NO3‚HCl‚1/2H2O).
Gen er a l P r oced u r e for P r ep a r a tion of th e Tr ibu tyl-
sta n n yl In ter m ed ia tes: (S)-2â-Ca r bom eth oxy-3r-[bis[4-
(t r ib u t ylst a n n yl)p h en yl]m et h oxy]t r op a n e. (S)-(+)-2â-
Carbomethoxy-3R-[bis(4-bromophenyl)methoxy]tropane, 2c (0.38
g, 0.73 mmol), in toluene (10 mL) was degassed by bubbling
N2 into the mixture. Bis(tributyltin) (0.88 mL, 1.74 mmol) was
added followed by tetrakis(triphenylphosphine)palladium (13.4
mg, 0.12 mmol). Degassing was continued for another 15 min.
The reaction mixture was heated at reflux for 2 h, filtered
through Celite, and then concentrated to dryness. The residue
was purified by flash chromatography (hexanes, 10% Et3N/
hexanes) to afford 81 mg (12%) of (S)-2â-carbomethoxy-3R-[bis-
[4-(tributylstannyl)phenyl]methoxy]tropane as a light brown
oil: Rf 0.5 (10% Et3N/hexanes, two elutions); 1H-NMR (100
MHz, CDCl3) δ 0.70-2.13 (m, 60H), 2.18 (s, 3H, NCH3), 2.73
(s, 1H, H-2), 3.08 (bs, 1H, H-5), 3.55 (bs, 1H, H-1), 3.65 (s, 3H,
OCH3), 4.01 (d, 1H, H-3), 5.30 (s, 1H, OCH), 7.18-7.46 (m,
8H, ArH). Also eluted was 197 mg of (S)-2â-carbomethoxy-
3R-[[4-(tributylstannyl)phenyl](4-bromophenyl)methoxy]tro-
pane as a light brown oil: Rf 0.27 (10% Et3N/hexanes, two
7.72 (m, 8H, ArH); [R]21D +20.5° (c ) 1, MeOH). Anal. (C23H25
-
NO3BrI‚1/2H2O) C, H, N, Br, I.
(S )-(+)-2â-Ca r b om e t h oxy-3r-[b is(4-flu or op h e n yl)-
m eth oxy]-N-[3-(4-iodoph en yl)pr opyl]n or tr opan e (4d). (S)-
2â-Carbomethoxy-3R-[bis(4-fluorophenyl)methoxy]-N-[3-[4-
(tributylstannyl)phenyl]propyl]nortropane (83 mg, 0.1 mmol)
in anhydrous THF (5 mL) was degassed by bubbling N2 for 15
min. N-Iodosuccinimide (24 mg, 0.107 mmol) was added, and
the reaction mixture was stirred at room temperature for 15
min. THF was removed under reduced pressure. The residue
was dissolved in 10 mL of CH2Cl2 and washed with saturated
NaHCO3 (5 mL) and brine (5 mL). The dried (K2CO3) CH2Cl2
layer was concentrated to dryness. The residue was purified
by flash chromatography (25% EtOAc/hexanes) to afford 66
mg (100%) of (S)-2â-carbomethoxy-3R-[bis(4-fluorophenyl)-
methoxy]-N-[3-(4-iodophenyl)propyl]nortropane, 4d , as a light
brown oil: Rf 0.55 (EtOAc/hexane, 1:1); 1H-NMR (100 MHz,
CDCl3) δ 1.50-2.22 (m, 10H), 2.55 (t, 2H, NCH2), 2.73 (bs, 1H,
H-2), 3.08 (bs, 1H, H-5), 3.65 (s, 3H, OCH3), 3.68 (bs, 1H, H-1),
4.01 (d, 1H, H-3), 5.34 (s, 1H, OCH), 6.87-7.62 (m, 12H, ArH).
1
elutions); H-NMR (100 MHz, CDCl3) δ 0.70-2.18 (m, 33H),
2.19 (s, 3H, NCH3), 2.74 (bs, 1H, H-2), 3.09 (bs, 1H, H-5), 3.58
(bs, 1H, H-1), 3.66, 3.68 (2s, 3H, OCH3), 4.00 (d, 1H, H-3), 5.31
(s, 1H, OCH), 7.15-7.55 (m, 8H, ArH). In addition, 180 mg
(47%) of recovered starting material 2c was obtained.
(S)-2â-Ca r bom eth oxy-3r-[[4-(tr ibu tylsta n n yl)p h en yl]-
p h en ylm eth oxy]tr op a n e. (S)-2â-Carbomethoxy-3R-[(4-bro-
mophenyl)phenylmethoxy]tropane, 2i, was treated as de-
scribed above to provide (S)-2â-carbomethoxy-3R-[[4-(tri-
butylstannyl)phenyl]phenylmethoxy]tropane as a light brown
oil (57%): Rf 0.37 (10% Et3N/hexanes, two elutions); 1H-NMR
(100 MHz, CDCl3) δ 0.71-2.14 (m, 33H), 2.20 (s, 3H, NCH3),
2.76 (bs, 1H, H-2), 3.09 (bs, 1H, H-5), 3.58 (bs, 1H, H-1), 3.68
(s, 3H, OCH3), 4.03 (d, 1H, H-3), 5.35 (s, 1H, OCH), 7.15-7.63
(m, 9H, ArH).
HCl salt: mp 113-115 °C; [R]21 +20.2° (c ) 0.55, MeOH).
D
Anal. (C31H32NO3F2I‚HCl‚1.5H2O) C, H, N, Cl, I.
(S)-2â-Ca r b om e t h oxy-3R-(e t h ylp h e n ylm e t h oxy)t r o-
p a n e (2m ). (S)-Alloecgonine methyl ester, 1 (200 mg, 1
mmol), 1-phenylpropanol (273 mg, 2 mmol), p-toluenesulfonic
acid monohydrate (288 mg, 1.5 mmol), and benzene (50 mL)
in a 100 mL round bottom flask fitted with a Dean-Stark trap
and condenser were heated at reflux for 18 h. 1-Phenylpro-
panol (180 mg) and p-toluenesulfonic acid monohydrate (50
mg) were added, and the reaction mixture was heated at reflux
for another 5 h. The reaction mixture was cooled to room
temperature and concentrated to dryness. The residue was
dissolved in ether (25 mL) and washed with water (20 mL).
The aqueous layer was basified with NH4OH and extracted
with CH2Cl2 (2 × 25 mL). The dried (K2CO3) organic layer
was concentrated to dryness. The residue was flash chro-
matographed over silica gel (EtOAc, 2% NH4OH, 5% MeOH
in EtOAc) to afford 45 mg (14%) of (S)-2â-carbomethoxy-3R-
(ethylphenylmethoxy)tropane, 2m , as a pale yellow oil: Rf 0.53
(S)-2â-Car bom eth oxy-3r-[bis(4-flu or oph en yl)m eth oxy]-
N-[3-[4-(tr ibu tylsta n n yl)p h en yl]p r op yl]n or tr op a n e. (S)-
2â-Carbomethoxy-3R-[bis(4-fluorophenyl)methoxy]-N-[3-(4-
bromophenyl)propyl]nortropane, 4c, was treated as described
above to provide (S)-2â-carbomethoxy-3R-[bis(4-fluorophenyl)-
methoxy]-N-[3-[4-(tributylstannyl)phenyl]propyl]nortropane
(49%) as a light brown oil: Rf 0.73 (25% EtOAc/hexane).
Gen er a l P r oced u r e for Iod in a tion of th e Tr ibu tyl-
sta n n yl Com p ou n d s: (S)-(+)-2â-Ca r bom eth oxy-3R-[bis-
(4-iod op h en yl)m et h oxy]t r op a n e (2e). (S)-(+)-2â-Carbo-
methoxy-3R-[bis[4-(tributylstannyl)phenyl]methoxy]tropane (80
mg, 0.086 mmol) in anhydrous THF (3 mL) was degassed by
bubbling N2 for 15 min. N-Iodosuccinimide (39 mg, 0.17 mmol)
was added, and the reaction mixture was stirred at room
temperature for 15 min. THF was removed under reduced
pressure. The residue was dissolved in 10 mL of EtOAc and
washed with saturated Na2CO3 (5 mL) and brine (5 mL). The
dried (K2CO3) organic layer was concentrated to dryness. The
residue was purified by flash chromatography (5% Et3N, 10%
Et2O in hexanes) to afford 38 mg (72%) of (S)-(+)-2â-car-
bomethoxy-3R-[bis(4-iodophenyl)methoxy]tropane, 2e, as an
off-white solid: mp 156-158 °C; Rf 0.53 (EtOAc + 2% NH4-
OH); 1H-NMR (100 MHz, CDCl3) δ 1.45-2.13 (m, 6H), 2.20 (s,
3H, NCH3), 2.71 (s, 1H, H-2), 3.10 (bs, 1H, H-5), 3.60 (bs, 1H,
1
(EtOAc + 2% NH4OH); H-NMR (100 MHz, CDCl3) δ 0.86 (t,
3H, CH3), 1.50-2.10 (m, 8H), 2.17 (s, 3H, NCH3), 2.60, 2.73
(2bs, 1H, H-2), 3.05 (bs, 1H, H-5), 3.52 (bs, 1H, H-1), 3.61, 3.68
(2s, 3H, OCH3), 3.82 (d, 1H, H-3), 4.20 (t, 1H, OCH), 7.26 (m,
5H, ArH); [R]21D +36.7° (c ) 0.39, MeOH). HCl salt: mp 150-
152 °C. Anal. (C19H27NO3‚HCl‚1/3H2O) C, H, N, Cl.
(RS)-2â-Ca r bom eth oxy-3r-(flu or en yloxy)tr op a n e (2n ).
(RS)-Alloecgonine methyl ester, 1 (200 mg, 1 mmol), 9-hy-
droxyfluorene (366 mg, 2 mmol), p-toluenesulfonic acid mono-
hydrate (290 mg, 1.5 mmol), and benzene (50 mL) in a 100
mL round bottom flask fitted with a Dean-Stark trap and
condenser were heated at reflux for 18 h. 9-Hydroxyfluorene
(366 mg) and p-toluenesulfonic acid monohydrate (80 mg) were
added, and the reaction mixture was heated at reflux for
another 5 h. The reaction mixture was cooled to room
temperature and concentrated to dryness. The residue was

378 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 2
Meltzer et al.
dissolved in CH2Cl2 (30 mL) and basified with NH4OH. The
dried (K2CO3) organic layer was concentrated to dryness. The
residue was chromatographed over silica gel (EtOAc, 3% NH4-
OH, 5% MeOH in EtOAc) to afford 104 mg (29%) of (RS)-2â-
carbomethoxy-3R-(fluorenyloxy)tropane, 2n , as a light brown
oil: Rf 0.48 (EtOAc + 2% NH4OH); 1H-NMR (100 MHz, CDCl3)
δ 1.80-2.15 (m, 6H), 2.20 (s, 3H, NCH3), 2.86 (s, 1H, H-2),
3.12 (bs, 1H, H-5), 3.61 (bs, 1H, H-1), 3.70 (s, 3H, OCH3), 4.40
Tissu e Sou r ces a n d P r ep a r a tion . Brain tissue from
adult male and female cynomolgus monkeys (Macaca fascicu-
laris) was stored at -85 °C in the primate brain bank at the
New England Regional Primate Research Center. The cau-
date-putamen was dissected from coronal slices and yielded
1.4 ( 0.4 g of tissue. Membranes were prepared as described
previously. Briefly, the caudate-putamen was homogenized
in 10 vol (w/v) of ice-cold Tris‚HCl buffer (50 mM, pH 7.4 at
4°C) and centrifuged at 3800g for 20 min in the cold. The
resulting pellet was suspended in 40 vol of buffer, and the
entire procedure was repeated twice. The membrane suspen-
sion (25 mg original wet weight of tissue/mL) was diluted to
12 mL/mL for [3H]WIN 35,428 or [3H]citalopram assay in
buffer just before assay and was dispersed with a Brinkmann
Polytron homogenizer (setting no. 5) for 15 s. All experiments
were conducted in triplicate, and each experiment was re-
peated in each of two to three preparations from individual
brains.
Dop a m in e Tr a n sp or ter Assa y. The DAT was labeled
with [3H]WIN 35,428 (81-84 Ci/mmol; DuPont-NEN). The
affinity of [3H]WIN 35,428 for the DAT was determined in
experiments by incubating tissue with a fixed concentration
of [3H]WIN 35,428 and a range of concentrations of unlabeled
WIN 35,428. The assay tubes received, in Tris‚HCl buffer (50
mM, pH 7.4 at 0-4 °C; NaCl, 100 mM), the following
constituents at a final assay concentration: CFT, 0.2 mL (1
pM-100 or 300 nM), [3H]WIN 35,428 (0.3 nM), and membrane
preparation, 0.2 mL (4 mg original wet weight of tissue/mL).
The 2 h incubation (0-4 °C) was initiated by addition of
membranes and terminated by rapid filtration over Whatman
GF/B glass fiber filters presoaked in 0.1% bovine serum
albumin (Sigma Chemical Co.). The filters were washed twice
with 5 mL of Tris‚HCl buffer (50 mM) and incubated overnight
at 0-4 °C in scintillation fluor (Beckman Ready-Value, 5 mL),
and radioactivity was measured by liquid scintillation spec-
trometry (Beckman 1801; cpm were converted to dpm following
determination of counting efficiency (>45%) of each vial by
external standardization. Total binding was defined as [3H]-
CFT bound in the presence of ineffective concentrations of
unlabeled CFT (1 or 10 pM). Nonspecific binding was defined
as [3H]CFT bound in the presence of an excess (30 µM) of (-)-
cocaine. Specific binding was the difference between the two
values. Competition experiments to determine the affinities
of other drugs at [3H]CFT-binding sites were conducted using
procedures similar to those outlined above. Stock solutions
of water-soluble drugs were dissolved in water or buffer, and
stock solutions of other drugs were made in a range of ethanol/
HCl solutions. Several of the drugs were sonicated to promote
solubility. The stock solutions were diluted serially in the
assay buffer and added (0.2 mL) to the assay medium as
described above.
(d, 1H, H-3), 5.48 (s, 1H, OCH), 7.20-7.70 (m, 8H, ArH); [R]21
D
-4.41° (c ) 2.45, MeOH). HCl salt: mp 155 °C. Anal. (C23H25
NO3‚HCl‚0.5H2O) C, H, N, Cl.
-
(S )-(+)-2â-Ca r b om e t h oxy-3r-[b is(4-flu or op h e n yl)-
m eth oxy]n or tr op a n e (3). (S)-(+)-2â-Carbomethoxy-3R-[bis-
(4-fluorophenyl)methoxy]tropane, 2a (0.5 g, 1.24 mmol), and
R-chloroethyl chloroformate (ACE-Cl) (2 mL) were combined
and heated at reflux for 1 h. Excess ACE-Cl was then removed
under reduced pressure, and methanol (10 mL) was added to
the residue. The mixture was heated at reflux for 1 h and
then concentrated to dryness. The residue was purified by
flash chromatography to afford 280 mg (58%) of (S)-(+)-2â-
carbomethoxy-3R-[bis(4-fluorophenyl)methoxy]nortropane, 3,
as a light brown oil: Rf 0.26 (EtOAc + 2% NH4OH); 1H-NMR
(100 MHz, CDCl3) δ 1.74-2.25 (m, 6H), 2.73 (bs, 1H, H-2), 3.50
(bs, 1H, H-5), 3.68 (s, 3H, OCH3), 3.84 (d, 2H, H-3, H-1), 5.35
(s, 1H, OCH), 6.92-7.35 (m, 8H, ArH). Anal. (C22H23
NO3F2‚0.25H2O) C, H, N.
-
Gen er a l P r oced u r e for Alk yla tion of th e Nor tr o-
p a n es: (S)-(+)-2â-Ca r bom eth oxy-3r-[bis(4-flu or op h en -
yl)m et h oxy]-N-b en zyln or t r op a n e
(4a ).
(S)-2â-Car-
bomethoxy-3R-[bis(4-fluorophenyl)methoxy]nortropane, 3 (100
mg, 0.26 mmol), benzyl bromide (51 mg, 0.29 mmol), 50% KF/
Celite (150 mg, 1.29 mmol), and anhydrous acetonitrile (5 mL)
were combined and heated at reflux for 18 h. Ether (15 mL)
was added after the reaction mixture was cooled to room
temperature, and the mixture was filtered over Celite. The
filtrate was concentrated to dryness. The residue was purified
by flash chromatography (5% Et3N/hexanes) to afford 111 mg
(90%) of (S)-2â-carbomethoxy-3R-[bis(4-fluorophenyl)methoxy]-
N-benzylnortropane, 4a , as a pale yellow oil: Rf 0.24 (10%
Et3N/hexane); 1H-NMR (100 MHz, CDCl3) δ 1.70-2.20 (m, 6H),
2.68 (bs, 1H, H-2), 3.15-3.58 (m, 4H, H-5, NCH2, H-1), 3.47
(s, 3H, OCH3), 3.99 (d, 1H, H-3), 5.34 (s, 1H, OCH), 6.90-7.34
(m, 13H, ArH); [R]21 +37.8° (c ) 1, MeOH). HCl salt: mp
D
130-132 °C. Anal. (C29H29NO3F2‚HCl) C, H, N, Cl.
(S )-(+)-2â-Ca r b om e t h oxy-3r-[b is(4-flu or op h e n yl)-
m et h oxy]-N-(3-p h en ylp r op yl)n or t r op a n e (4b ). (S)-2â-
Carbomethoxy-3R-[bis(4-fluorophenyl)methoxy]nortropane, 3,
and 1-bromo-3-phenylpropane were treated as for 4a to provide
1
a pale yellow oil (81%): Rf 0.33 (10% Et3N/hexane); H-NMR
(100 MHz, CDCl3) δ 1.62-2.28 (m, 10H), 2.61 (t, 2H, NCH2),
2.70 (s, 1H, H-2), 3.10 (bs, 1H, H-5), 3.63 (s, 3H, OCH3), 3.69
(d, 1H, H-1), 4.00 (d, 1H, H-3), 5.34 (s, 1H, OCH), 6.90-7.36
Ser oton in Tr a n sp or ter Assa y. The serotonin transporter
was assayed in caudate-putamen membranes using conditions
similar to those for the DAT. The affinity of [3H]citalopram
(specific activity, 82 Ci/mmol; DuPont-NEN) for the serotonin
transporter was determined in experiments by incubating
tissue with a fixed concentration of [3H]citalopram and a range
of concentrations of unlabeled citalopram. The assay tubes
received, in Tris‚HCl buffer (50 mM, pH 7.4 at 0-4 °C; NaCl,
(m, 13H, ArH). HCl salt: mp 98-100 °C; [R]21 30.2° (c ) 1,
D
MeOH). Anal. (C31H33NO3F2‚HCl‚0.5H2O) C, H, N, Cl.
(S )-(+)-2â-Ca r b om e t h oxy-3r-[b is(4-flu or op h e n yl)-
m eth oxy]-N-[3-(4-br om op h en yl)p r op yl]n or tr op a n e (4c).
(S)-2â-Carbomethoxy-3R-[bis(4-fluorophenyl)methoxy]nortro-
pane, 3, and 1-bromo-3-(4-bromophenyl)propane were treated
as for 4a to provide a light brown oil (77%): Rf 0.55 (EtOAc/
hexane, 1:1); 1H-NMR (100 MHz, CDCl3) δ 1.50-2.30 (m, 10H),
2.55 (t, 2H, NCH2), 2.73 (bs, 1H, H-2), 3.08 (bs, 1H, H-5), 3.62
(s, 3H, OCH3), 3.68 (bs, 1H, H-1), 4.00 (d, 1H, H-3), 5.34 (s,
1H, OCH), 6.87-7.45 (m, 12H, ArH). HCl salt: mp 91-94
100 mM), the following constituents at
a final assay
concentration: citalopram, 0.2 mL (1 pM-100 or 300 nM), [3H]-
citalopram (1 nM), and membrane preparation, 0.2 mL (4 mg
original wet weight of tissue/mL). The 2 h incubation (0-4
°C) was initiated by addition of membranes and terminated
by rapid filtration over Whatman GF/B glass fiber filters
presoaked in 0.1% poly(ethyleneimine). The filters were
washed twice with 5 mL of Tris‚HCl buffer (50 mM) and
incubated overnight at 0-4 °C in scintillation fluor (Beckman
Ready-Value, 5 mL), and radioactivity was measured by liquid
scintillation spectrometry (Beckman 1801); cpm were con-
verted to dpm following determination of counting efficiency
(>45%) of each vial by external standardization. Total binding
was defined as [3H]citalopram bound in the presence of
ineffective concentrations of unlabeled citalopram (1 or 10 pM).
Nonspecific binding was defined as [3H]citalopram bound in
the presence of an excess (10 µM) of fluoxetine. Specific
°C; [R]21 +22.9° (c ) 0.81, MeOH). Anal. (C31H32NO3F2Br‚
D
HCl‚0.5H2O) C, H, N, Br, Cl.
(S )-(+)-2â-Ca r b om e t h oxy-3r-[b is(4-flu or op h e n yl)-
m et h oxy]-N-(3-p h en ylp en t yl)n or t r op a n e (4e). (S)-2â-
Carbomethoxy-3R-[bis(4-fluorophenyl)methoxy]nortropane, 3,
and 1-chloro-3-phenylpentane were treated as for 4a to provide
1
a pale yellow oil (76%): Rf 0.28 (10% Et3N/hexane); H-NMR
(100 MHz, CDCl3) δ 1.22-2.22 (m, 14H), 2.59 (t, 2H, NCH2),
2.70 (bs, 1H, H-2), 3.10 (bs, 1H, H-5), 3.62 (s, 3H, OCH3), 3.68
(bs, 1H, H-1), 3.99 (d, 1H, H-3), 5.33 (s, 1H, OCH), 6.90-7.33
(m, 13H, ArH); [R]21D +23.5° (c ) 0.86, MeOH). Anal. (C33H37
NO3F2) C, H, N.
-

2-Carbomethoxy-3-(diarylmethoxy)tropane Analogs
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 2 379
(16) Carroll, F. I.; Abraham, P.; Lewin, A.; Parham, K. A.; Boja, J .
W.; Kuhar, M. J . Isopropyl and phenyl esters of 3â-(4-substituted
phenyl) tropane-2â-carboxylic acids. Potent and selective com-
pounds for the dopamine transporter. J . Med. Chem. 1992, 35,
2497-2500.
binding was the difference between the two values. Competi-
tion experiments to determine the affinities of other drugs at
[3H]citalopram-binding sites were conducted using procedures
similar to those outlined above.
(17) Boja, J . W.; McNeill, R. M.; Lewin, A.; Abraham, P.; Carroll, F.
I.; Kuhar, M. J . Selective dopamine transporter inhibition by
cocaine analogs. Mol. Neurosci. 1992, 3, 984-986.
(18) Davies, H. M. L.; Saikali, E.; Sexton, T.; Childers, S. R. Novel
2-substituted cocaine analogs: binding properties at dopamine
transport sites in rat striatum. Eur. J . Pharmacol. Mol. Phar-
macol. 1993, 244, 93-97.
(19) Kozikowski, A. P.; Roberti, M.; Xiang, L.; Bergmann, J . S.;
Callahan, P. M.; Cunningham, K. A.; J ohnson, K. M. Structure-
activity relationship studies of cocaine: replacement of the C-2
ester group by vinyl argues against H-bonding and provides an
esterase-resistant, high-affinity cocaine analogue. J . Med. Chem.
1992, 35, 4764-4766.
(20) Kozikowski, A. P.; Roberti, M.; J ohnson, K. M.; Bergmann, J .
S.; Ball, R. G. SAR of Cocaine: further exploration of structural
variations at the C-2 center provides compounds of subnano-
molar binding potency. Bioorg. Med. Chem. Lett. 1993, 3, 1327-
1332.
(21) Kitayama, S.; Shimada, S.; Xu, H.; Markham, L.; Donovan, D.
H.; Uhl, G. R. Dopamine transporter site-directed mutations
differentially alter substrate transport and cocaine binding. Proc.
Natl. Acad. Sci. U.S.A. 1993, 89, 7782-7785.
(22) Kozikowski, A. P.; Saiah, M. K. E.; Bergmann, J . S.; J ohnson,
K. M. Structure-activity relationship studies of N-sulfonyl
analogs of cocaine: role of ionic interaction in cocaine binding.
J . Med. Chem. 1994, 37, 3440-3442.
(23) Elmaleh, D. R.; Madras, B. K.; Shoup, T. M.; Byon, C.; Hanson,
R. N.; Liang, A. Y.; Meltzer, P. C.; Fischman, A. J . Radiosyn-
thesis and evaluation of E and Z-[125I]-2â-carbomethoxy-3â-(4-
fluorophenyl)-N-(iodoprop-1-en-3-yl)nortropane (Altropane): A
selective SPECT agent for imaging DA reuptake sites. J . Nucl.
Chem. 1995, accepted for publication.
(24) Goodman, M. M.; Kung, M.-P.; Kabalka, G. W.; Kung, H. F.;
Switzer, R. Synthesis and characterization of radioiodinated
N-(3-iodopropen-1-yl)2â-carbomethoxy-3â-(4-chlorophenyl)tro-
panes: Potential dopamine reuptake site imaging agents. J . Med
Chem. 1994, 37, 1535-1542.
(25) 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.
(26) van der Zee, P.; Koger, H. S.; Gootjes, J .; Hespe, W. Aryl 1,4-
dialk(en)ylpiperazines as selective and very potent inhibitors of
dopamine uptake. Eur. J . Med. Chem. 1980, 15, 363-370.
(27) Newman, A. H.; Allen, A. C.; Izenwasser, S.; Katz, J . L. Novel
3R-(diphenylmethoxy)tropane analogs: potent dopamine uptake
inhibitors without cocaine-like behavioral profiles. J . Med Chem.
1994, 37, 2258-2261.
(28) Dutta, A. K.; Meltzer, P. C.; Madras, B. K. Positional importance
of the nitrogen atom in novel piperidine analogs of GBR 12909:
Affinity and selectivity for the dopamine transporter. Med.
Chem. Res. 1993, 3, 209-222.
(29) Madras, B. K.; Reith, M. E. A.; Meltzer, P. C.; Dutta, A. K. O-526,
a piperidine analog of GBR 12909, retains high affinity for the
dopamine transporter in monkey caudate putamen. Eur. J .
Pharmacol. 1994, 267, 167-173.
Ack n ow led gm en t. The authors acknowledge Bos-
ton Life Sciences for funding a part of this work (B.K.M.)
and NIDA Grant No. DA06303 and Division of Research
Resources RR00168 (B.K.M.) for support.
Refer en ces
(1) Kennedy, L. T.; Hanbauer, I. Sodium sensitive cocaine binding
to rat striatal membrane: possible relationship to dopamine
uptake sites. J . Neurochem. 1983, 34, 1137-1144.
(2) Schoemaker, H.; Pimoule, C.; Arbilla, S.; Scatton, B.; J avoy-Agid,
F.; Langer, S. Z. Sodium dependent [3H]cocaine binding associ-
ated with dopamine uptake sites in the rat striatum and human
putamen decrease after dopaminergic denervation and in Par-
kinson’s disease. Naunyn-Schmiedeberg’s Arch. Pharmacol.
1985, 329, 227-235.
(3) Reith, M. E. A.; Meisler, B. E.; Sershen, H.; Lajtha, A. Structural
requirements for cocaine congeners to interact with dopamine
and serotonin uptake sites in mouse brain and to induce
stereotyped behavior. Biochem. Pharmacol. 1986, 35, 1123-
1129.
(4) 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.
(5) Madras, B. K.; Fahey, M. A.; Bergman, J .; Canfield, D. R.;
Spealman, R. D. Effects of cocaine and related drugs in nonhu-
man primates: I. [3H]Cocaine binding sites in caudate-putamen.
J . Pharmacol. Exp. Ther. 1989, 251, 131-141.
(6) Madras, B. K.; Spealman, R. D.; Fahey, M. A.; Neumeyer, J . L.;
Saha, J . K.; Milius, R. A. Cocaine receptors labeled by [3H]2â-
Carbomethoxy-3â(4-fluorophenyl)tropane. Mol. Pharmacol. 1989,
36, 518-524.
(7) Bergman, J .; Madras, B. K.; J ohnson, S. E.; Spealman, R. D.
Effects of cocaine and related drugs in nonhuman primates. III.
Self-administration by squirrel monkeys. J . Pharmacol. Exp.
Ther. 1989, 251, 150-155.
(8) Canfield, D. R.; Spealman, R. D.; Kaufman, M. J .; Madras, B.
K. Autoradiographic localization of cocaine binding sites by [3H]-
CFT ([3H]WIN 35,428) in the monkey brain. Synapse 1990, 6,
189-195.
(9) Kaufman, M. J .; Madras, B. K. Severe depletion of cocaine
recognition sites associated with the dopamine transporter in
Parkinson’s diseased striatum. Synapse 1991, 9, 43-49.
(10) Kuhar, M. J .; Ritz, M. C.; Boja, J . W. The dopamine hypothesis
of the reinforcing properties of cocaine. Trends Neurosci. 1991,
14, 299-302.
(11) Madras, B. K.; Kamien, J . B.; Fahey, M.; Canfield, D.; Milius,
R. A.; Saha, J . K.; Neumeyer, J . L. N-Modified fluorophenyltro-
pane analogs of cocaine with high affinity for [3H]cocaine
receptors. Pharmacol. Biochem. Behav. 1990, 35, 949-953.
(12) 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-substituted phenyl)-
tropane-2â-carboxylic acid methyl esters. J . Med. Chem. 1991,
34, 2719-2725.
(13) Carroll, F. I.; Mascarella, S. W.; Kuzemko, M. A.; Gao, Y.;
Abraham, P.; Lewin, A. H.; Boja, J . W.; Kuhar, M. J . Synthesis,
Ligand Binding, and QSAR (CoMFA and Classical) Study of 3â-
(3′-Substituted phenyl)-, 3â-(4′-Substituted phenyl)-, 3â-(3′,4′-
Disubstituted phenyl)tropane-2â-carboxylic Acid Methyl Esters.
J . Med. Chem. 1994, 37, 2865-2873.
(14) Meltzer, P. C.; Liang, A. Y.; Brownell, A.-L.; Elmaleh, D. R.;
Madras, B. K. Substituted 3-phenyltropane analogs of cocaine:
synthesis, inhibition of binding at cocaine recognition sites, and
Positron Emission Tomography imaging. J . Med. Chem. 1993,
36, 855-862.
(15) Carroll, F. I.; Lewin, A. H.; Boja, J . W.; Kuhar, M. J . Cocaine
Receptor: Biochemical Characterization and Structure Activity
Relationships of Cocaine Analogues at the Dopamine Trans-
porter. J . Med. Chem. 1992, 35, 969-981.
(30) Andersen, P. H. Biochemical and pharmacological characteriza-
tion of [3H]GBR 12935 binding in vitro to rat striatal mem-
branes: Labeling of the dopamine uptake complex. J . Neuro-
chem. 1987, 48, 1887-1896.
(31) Andersen, P. H. The dopamine uptake inhibitor GBR 12909:
selectivity and molecular mechanism of action. Eur. J . Phar-
macol. 1989, 166, 493-504.
J M950463T