Benzo- and cyclohexanomazindol analogues as potential inhibitors of the cocaine binding site at the dopamine transporter

Article abstract of DOI:10.1021/jm010301z

A series of mazindol (1), homomazindol (2), and bishomomazindol (3) derivatives with a benzo or cyclohexano ring fused at various sites were prepared as part of an SAR study to determine the effect of increased aliphatic and aromatic lipophilicity on sele

Full text of DOI:10.1021/jm010301z

4110
J . Med. Chem. 2002, 45, 4110-4118
Ben zo- a n d Cycloh exa n om a zin d ol An a logu es a s P oten tia l In h ibitor s of th e
Coca in e Bin d in g Site a t th e Dop a m in e Tr a n sp or ter
William J . Houlihan,*,† Umer F. Ahmad,† J udith Koletar,† Lawrence Kelly,† Leonard Brand,† and
Theresa A. Kopajtic‡
Charles A. Dana Research Institute, Drew University, Hall of Sciences, Madison, New J ersey 07940, and
National Institute on Drug Abuse, Intramural Research Program, P.O. Box 5180, Baltimore, Maryland 21224
Received J une 29, 2001
A series of mazindol (1), homomazindol (2), and bishomomazindol (3) derivatives with a benzo
or cyclohexano ring fused at various sites were prepared as part of an SAR study to determine
the effect of increased aliphatic and aromatic lipophilicity on selected in vitro assays used to
identify potential cocaine-like and cocaine antagonism activity. Very good (IC50 ) 2-3 nM)
inhibition of [3H] WIN 35,428 and [125I] RTI-55 binding on rat or guinea pig striatal membranes
and HEK cells expressing cDNA for the human dopamine transporter (HEK-hDAT) was shown
by the 8,9-benzomazindol 25 and 9,10-benzohomomazindol 28. All new compounds were weaker
inhibitors of [3H] DA uptake in HEK-hDAT cells than 1 and 2. No improvement in the binding
selectivity ratio (SERT/DAT and NET/DAT) was found when compared to 2. Compounds 25
and 28 showed a considerable increase versus 1 in uptake/binding discrimination ratios at the
DAT (311.0 and 182.1 vs 0.9), SERT (33.6 and 127.3 vs 1.9), and NET (7.3 and 10.0 vs 0.3).
In tr od u ction
The search for safe and effective medications in the
treatment of cocaine abuse has focused mainly on
substances that can displace its binding at the dopamine
transporter (DAT) in the striatum.1 The interaction of
cocaine with the DAT results in the blockage of the
uptake of DA into the presynaptic axon terminal and
leads to an increase of dopamine in the synapse. This
results in increased interaction of DA with its receptors
on postsynaptic neurons that gives rise to reinforcing
and stimulatory effects.2
The preceding3 and earlier papers4,5 from our labo-
ratories have reported on structure-activity relation-
ships (SAR) studies on mazindol (1) and its ring A
homologues , homomazindol (2), and bishomomazindol
(3) as potential inhibitors of cocaine (4a ) binding at the
DAT. That work focused on the effect of substituents
on the pendant aryl group, rings C and C-5 or C-6,
enlargement of ring A, and modification of the carbonyl
group in the keto tautomer of mazindol (M-k eto in
Figure 1). In addition, an interaction model of how 1 to
3 could interact at the DAT was proposed.3
F igu r e 1. Tautomeric forms of aryl substituted mazindol (M)
and homomazindol (HM) analogues in neutral and acidic
media.
16, 19, 22, 24, and 28), two (25 and 32), or three (33)
benzene rings, one cyclohexane ring (8, 12, and 13), and
one benzene and cyclohexane ring (30) on rings A, C,
and D (Figure 1).
Ch em istr y
In this paper, we investigate the effects of enhancing
the lipophilicity of 1 to 3 by adding one (6, 7, 11, 15,
The synthesis of the compounds used in this study is
given in Schemes 1-5.
Treatment of the known6 N,o-dilithio derivative of
2-phenylimidazoline (5) with the methyl esters of 1- and
2-naphthoic and 5,6,7,8-tetrahydronaphthoic acids gave
6, 7, and 8 (Scheme 1). Compound 6 was determined to
* To whom correspondence should be addressed. E-Mail Address:
whouliha@drew.edu.
† Charles A. Dana Research Institute.
‡ National Institute on Drug Abuse.
10.1021/jm010301z CCC: $22.00 © 2002 American Chemical Society
Published on Web 09/05/2002

Benzo- and Cyclohexanomazindol Analogues
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19 4111
Sch em e 1a
Sch em e 3a
a
Reagents/conditions. (a) (i) n-BuLi, TMEDA, THF, N2, rt; (ii)
methyl-1-naphthoate, THF, rt, overnight. (b) (i) n-BuLi, THF, N2,
rt; (ii) methyl-2-naphthoate, THF, rt, overnite. (c) (i) Same as those
in panel b; (ii) methyl-5,6,7,8-tetrahydronaphthoate, THF, rt,
overnight.
a
Reagents/conditions. (a) (i) MeI, N2, 6 days, rt; (ii) 17-MeI,
EDA, reflux, 10 h. (b) (i) n-BuLi, THF, N2, rt; (ii) methyl
4-chlorobenzoate, THF, overnight, rt. (c) EDA, xylene reflux, 7 h.
(d) LiAlH4, THF, N2, 18°, 0.25 h.
Sch em e 2a
Sch em e 4a
a
Reagents/conditions. (a) (i) n-BuLi, THF, N2; (ii) methyl-4-
chlorobenzoate, THF, overnight, rt. (b) (i) Same as those in panel
a; (ii) methyl-2-naphthoate, THF, overnight, rt. (c) 1,3-Diamino-
propane, p-tolSO3H, 220°, 2.5 h. (d) trans-1,2-Diaminocyclohexane,
p-tolSO3H, 225°, 3 h.
Sch em e 5a
a
Reagents/conditions. (a) 1,2-Diaminobenzene, HOAc, concd.
HCl, 80°, 0.5 h. (b) (i) 10, NaH, DMF, N2; (ii) air, 6 h, rt. (c) cis- or
trans-1,2-diaminocyclohexane, xylenes, reflux; (ii) MeOH, air, 96
h, rt. (d) (i) 2-Aminobenzylamine, xylenes, reflux; (ii) i-PrOH, air,
56 h, rt. (e) (i) 1,2-Benzenedimethanamine, xylenes, reflux; (ii)
i-PrOH, air, 56 h, rt.
a
Reagents/conditions. (a) (i) n-BuLi, THF, argon, rt; (ii) methyl
4-chlorobenzoate, THF, rt, overnight. (b) (i) Same as thsoe in panel
a; (ii) methyl 2-naphthoate, THF, overnight.
exist as a mixture of ol and keto tautomers (ca. 5:95)
by the presence of characteristic 13C NMR signals at
88.3 and 197.44 ppm due to C-5 and CdO, respectively,
while 7 and 8 are exclusively in the ol form. Inspection
of Dreiding models suggests that the keto form pre-
dominates because of steric interaction of the naphthyl
group with positions 3 and 6 when 6 is in the ol form.
Condensation of 97 with 1, 2-diaminobenzene in a
mixture of warm HOAc/HCl gave 11H-isoindolo-[2,1-a]-
benzimidazole 10. Air oxidation8 of the C-11 sodium salt
of 10 in DMF gave the known9 11-ol derivative 11.
Treatment of 9 with racemic cis- or trans-1, 2-diami-
nocyclohexane in refluxing xylenes followed by exposure
to air of the resulting 11-H compounds gave the 11-ols
12 and 13. The 13C NMR of 12 gave two C-11 signals
(87.85 and 88.34) and two CdN signals (165.81 and
167.86), and 13 showed C-11 (87.82 and 89.04) and
CdN signals (166.83 and 168.02), indicating that both
existed as diastereomeric mixtures of at least two and
1
possibly four compounds each. H NMR analysis of the
hydrogens at 5a and 9a in 12 disclosed that each
integrated for 0.5 H, indicating a 50:50 diastereomeric
mixture. A similar analysis of 13 showed that several
atoms integrated for 0.4 or 0.6 H corresponding to a 40:
60 diastereoismeric mixture. No attempt was made to

4112 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19
Houlihan et al.
Ta ble 1. Inhibition of [3H] WIN 35,428 Binding at the
separate the mixtures because of the known3 ability of
a C-OH in mazindol analogues containing an imidazo
ring A to form the corresponding CdO tautomer under
mild conditions and then revert to the C-OH form
(Figure 1).
Dopamine Transporter on Guinea Pig Striatal Membranesa
compd
no.
IC50, nMb
IC50 ratio
mazindol/analog
(% inhib. at 10-6 M)
1
2
6
7
8
42.6 ( 2.6
4.2 ( 0.6
1841 ( 147
4.1 ( 0.3
(35)
1.0
10.1
0.02
Reaction of 9 with 2-aminobenzylamine or 1,2-ben-
zenedimethanamine in refluxing xylenes, followed by
10.4
1
exposure to air, gave compounds with H and 13 C NMR
<0.02
<0.02
5.1
11
12
13
14
16
19
22
24
25
28
30
32
33
(30)
consistent with a single entity in the ol form that could
be either 14 or 15 and the expected 16, respectively.
The large separation (4.13 and 4.62 ppm) of the meth-
ylene hydrogen signals in the 1H NMR of 14/15 suggests
that one of the hydrogen atoms is in the vicinity of a
polar environment. This is possible in 14, where one
hydrogen is shielded by the p-chlorophenyl group.
Additional evidence favoring 14 is found by comparing
its UV spectrum with 16. Compound 16 has a UV
spectrum similar to that of bishomomazindol (3), with
maxima at 201 and 272 nm, whereas the UV of 14/15
gave multiple maxima between 203 and 356 nm that
are in agreement with extended conjugation as found
in 14, but not 15.
8.4 ( 0.4
2.7 ( 0.2
(45)
15.8
<0.02
<0.02
0.2
1.3
4.0
13.0
20.3
<0.02
<0.02
<0.02
(12)
218 ( 13
36.1 ( 2.7
10.6 ( 1.1
3.3 ( 0.2
2.1 ( 0.2
(5)
(5)
(20)
a
b
See Experimental Section for details. Data represent the
mean (( SEM) of three assays at five concentrations for each
compound.
The known10 oxazole 17 was converted to the imida-
zoline 18 by treatment with iodomethane and then
reacting the resultant methiodide salt with refluxing 1,
2-diaminoethane. Lithiation of 18 with n-butyllithium
followed by treatment with methyl 4-chlorobenzoate
gave a mixture of two compounds with very close Rf
values (0.24 and 0.29) and NMR spectra, suggesting
that lithiation had occurred at both C-1 and C-3 to give
19 and the known11 22. Separation by plate chroma-
tography resulted in the isolation of the Rf ) 0.24
Lithiation of 2-(9-phenanthryl)-imidazoline (31) at
room temperature and then reaction with methyl 4-chlo-
robenzoate or naphthoate gave 32 and 33 as the ol
tautomers (Scheme 5).
Biology
The compounds listed in Table 1 were tested for their
ability to displace [3H] WIN 35,428 (WIN) binding from
guinea pig striatal membranes. Those more active (7,
13, 24, 25, and 28) than mazindol were evaluated in
displacing bound WIN from rat striatal membranes,
displacing [125I] RTI-55 (RTI) and blocking biogenic
amines (DA, 5-HT, NE) uptake in HEK cells expressing
cDNA for the human dopamine (hDAT), serotonin
(hSERT), and norepinephrine (hNET) transporters.
1
substance as a single entity. Comparison of the H and
13C NMR spectrum of 22, prepared by condensation of
keto-acid 2012 with 1,2-diaminoethane to 21 followed by
lithium aluminum hydride reduction, with the Rf ) 0.24
material showed the two were different compounds.
Additional evidence for assigning the Rf ) 0.24 sub-
1
stance as 19 comes from the H NMR spectrum which
Resu lts a n d Discu ssion
has a doublet at 8.10 ppm that can be assigned to the
H on C-11. The H on C-11 in 22 gave the expected
singlet at 8.33 ppm. It is interesting to note that no
lithiation is reported to occur at the C-3 position of N,N-
dimethyl-2-naphthamide.9
Lithiation of naphthylimidazoline13 (23), followed by
treatment with methyl esters of 4-chlorobenzoic and
2-naphthoic acid, gave the expected benzo-[g]-imidazo-
[2,1-a]-isoindol-5-ols 24 and 25, respectively (Scheme 4).
The inhibition of WIN binding at the DAT of guinea
pig striatal membranes for all compounds evaluated in
this work is given in Table 1.
Placement of a benzene ring at the 2′,3′-positions (6)
of the pendant aryl group in mazindol resulted in a 43-
fold loss of activity, while placement at the 3′,4′-position
(7) gave a 10-fold increase. The finding with 6 is in
contrast to the work of Davies,14 who reported that
replacement of a 3â-phenyl by a 1- or 2-naphthyl group
in the 2â-acyl-3â-aryl-8-azabicyclo-[3.2.1]-octane series
resulted in increased activity at inhibiting the binding
of RTI on rat striatal membranes. The results with 7 is
consistent with the increase in inhibition of binding at
the DAT when a phenyl was replaced by a 2-naphthyl
group in (-)-3â-substituted ecgonine methyl esters,15
piperidine-based analogues of cocaine,16 and the work
of Davies cited above.14 The marked difference in
activity between 6 and 7 in this series is most likely
due to 6 existing almost exclusively in the keto tau-
tomer, which has been shown3 to reduce activity in
substituted-aryl mazindol analogues. When the 2-naph-
thyl group in 7 was modified to the tetrahydronaphthyl
8, a greater than 250-fold loss of activity occurred. This
indicates that the Ring D binding of mazindol (Figure
1) favors only aromatic groups.
Condensation of 1-cyanonaphthalene (26) with p-
toluenesulfonic acid salts of 1,3-diaminopropane and
racemic trans-1,2-diaminocyclohexane at ca. 220 °C gave
27 and 29 respectively. Treatment of 27 with n-butyl-
lithium at room temperature and then methyl-4-chlo-
robenzoate gave 28 as the ol tautomer. Lithiation of 29
in an icebath, followed by reaction with methyl-4-chloro-
benzoate, gave 30 as a mixture of up to four compounds
(Scheme 4). The 13C NMR spectrum of 30 gave two C-13
(88.9 and 90.1 ppm) and CdN signals (168.5 and 170.4
1
ppm), while the H NMR spectrum gave several hydro-
gen atoms integrating for 0.5 H, indicating a 50:50
diastereoisomer mixture at C-13. No attempt was made
to separate these because, like 12 and 13, the presence
of an imidazo ring A makes the C-OH at C-13 subject
to ol-keto tautomerism (Figure 1).

Benzo- and Cyclohexanomazindol Analogues
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19 4113
Ta ble 2. Comparison of the Inhibition of Binding at the DAT on Guinea Pig and Rat Striatal Membranes and HEK HDAT Cells
[3H] WIN 35,428
IC50, nM
[
125I] RTI-55
ratios
HEK
g. pig
Ki, nM
rat
HEK
rat
compd
g. piga
rat
HEK hDATb
g. pig
1
7
13
24
25
28
42.6 ( 2.6
4.1 ( 0.3
2.7 ( 0.2
10.6 ( 1.1
3.3 ( 0.2
2.1 ( 0.2
12.94 ( 1.15
23.26 ( 3.26
34.20 ( 1.42
15.97 ( 2.19
38.11 ( 6.29
37.27 ( 1.53
45 ( 1
0.3
5.7
12.7
1.5
11.6
17.8
1.1
11.0
56.3
2.6
0.9
1.0
3.5
1.9
4.4
1.7
0.07
0.05
45 ( 20
152 ( 40
27.2 ( 6.5
2.83 ( 0.60
2.01 ( 0.85
a
b
Data from Table 1. Data from Table 3.
Addition of a benzene ring at the 2,3-position (11 and
14) of mazindol and homomazindol and the 3,4-position
(16) of bishomomazindol all resulted in a large loss
(100-500-fold) in binding activity, whereas addition of
a cyclohexane ring as a cis (12) or trans (13) isomer gave
a 5- and 16-fold increase. The loss of activity in 11 and
14 is most likely due to the conversion of the strongly
basic imine nitrogen in mazindol and homomazindol to
the weakly basic anilino nitrogen. Although 16 contains
a basic imine group, the benzene group at the 3,4-
position likely extends into a binding site pocket that
appears to tolerate only aliphatic interactions. Evidence
for this is demonstrated by the enhanced activity shown
by the cyclohexano analogues 12 and 13.
In contrast to the guinea pig, these compounds were
less potent than mazindol in displacing WIN binding
on the rat membranes. In addition, the order of activity
and the difference between the most and least active
compound was smaller. On the guinea pig membranes,
the activity is 28 > 13 > 25 > 7 > 24, and the rat
membrane pattern of activity is 24 > 7 > 13 > 28 )
25. Only compound 24 gave similar IC50 values, differing
by 1.5-fold. The greater variation of the present com-
pounds to fit in the guinea pig and rat DAT binding
pockets in the same sequence may be due to the larger
change in the molecular shape, size, or lipophilicity than
that of 1-3, where only the size of ring A was changed.
The selectivity of the five compounds to bind to the
DAT, SERT, and NET and their discrimination for
uptake inhibition at these transporters were measured
by displacement of [125I] RTI-55 binding and blocking
the uptake of [3H] DA, [3H] 5-HT, and [3H] NE on HEK
cells expressing cDNA for the human transporters
(Tables 3 and 4).
The 5-fold loss of activity with 6,7-benzo analogue 19
of mazindol is possibly due to some interference with
the ability of the OH group to interact at its binding
site. Compound 22, the 7,8-benzo analogue of mazindol
showed a slight (1.3-fold) increase in activity, while 24,
the 8,9-benzo analogue, gave a 4-fold increase. Replace-
ment of the p-chlorophenyl in 24 by a 2-naphthyl group
(25) enhanced the activity of mazindol by 13-fold.
However, when compared with 7, the 2-naphthyl ana-
logue of mazindol, the addition of the 8,9-benzo ring
increased activity by 1.3-fold. When the benzo ring was
added to the 9,10-position of homomazindol, there was
a 20-fold increase in binding relative to mazindol, but
only a 2-fold increase compared to homomazindol. In
summary, adding a benzo ring at the 6,7-, 7,8-, or 8,9-
positions in mazindol, or the 9,10-position in homo-
mazindol, gave a modest increase (1.3-4-fold) or de-
crease (5-fold) in binding.
All of the new compounds, except 13, were more
effective in inhibition of RTI binding than blocking DA
uptake at the DAT. Compounds 25 and 28 were potent
inhibitors of RTI binding with Ki values in the same
range as those of 2 and ca. 20-fold greater than those
of mazindol (1). In addition, 25 and 28 gave the best
discrimination ratios of 311 and 182, respectively, a
considerable improvement over the 0.9 and 2.2 ratios
for mazindol (1) and homomazindol (2).
Displacement of RTI binding on the HEK-hDAT cells
also showed a pattern of activity different from that of
WIN on guinea pig and rat striatal membranes with 28
> 25 > 24 > 7 > 13. Good correlation between HEK
and guinea pig activity was found with 1, 24, 25, and
28, whereas only 7 and 24 correlated with HEK and
rat activity.
Addition of a trans-cyclohexano ring to the 2,3-
position of 24 to form 30 resulted in a 500-fold decrease
in activity, while it gave a 15-fold increase (13) when
added to mazindol. The likely cause of this loss in
activity is that the addition of a second ring to the
heterocyclic nucleus of mazindol or homomazindol re-
sults in a molecule too large to fit in the same binding
pocket. This is further demonstrated in compounds 32
and 33, where the addition of a second benzo ring in
the 6,7-position of 24 and 25 resulted in a >500-fold
loss of activity.
Previous studies4 from our laboratories have shown
that mazindol, homomazindol, and bishomomazindol
displace WIN binding on guinea pig and rat membranes
in the same order (3 > 2 > 1). To determine if there is
any correlation between the activity found on the guinea
pig versus rat striatal membranes, we evaluated five
of the most active compounds (7, 13, 24, 25, and 28) for
their ability to displace WIN binding on rat striatal
membranes (Table 2).
At both the SERT and NET, 7, 25, and 28 were more
effective than mazindol (1) or homomazindol (2) in
inhibiting the binding of RTI. However, none of the
compounds showed a better selectivity ratio than 2.
Compounds 25 and 28 did show better discrimination
ratios of 33.6 and 127.3 compared to 1.9 and 1.4 for 1
and 2 at the SERT and 7.3 and 10.0 compared to 0.3
for both 1 and 2 at the NET. It is interesting to note
that 7, where the p-chlorophenyl group in mazindol is
replaced by a 2-naphthyl group, is the most potent
inhibitor of RTI binding and 5-HT uptake at the SERT.
Similar findings have also been found in tropane14-16
and piperidine17 series, where the replacement of an
aryl group by a 2-naphthyl has resulted in an increase
of selectivity for the cocaine binding site on the SERT.

4114 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19
Houlihan et al.
Ta ble 3. Inhibition of [125I] RTI-55 Binding at the DA, 5-HT, and NE Transporter Sites on HEK-hDAT, HEK-hSERT, and
HEK-hNET Cellsa
binding,b Ki, nM
selectivity ratio
compd
hDAT
hSERT
hNET
SERT/DAT
NET/DAT
1
2
7
13
24
25
28
45 ( 1
50 ( 15
39 ( 11
18 ( 12
18.2 ( 2
1.1
22.9
0.05
3.9
8.2
3.9
2.8
0.4
0.7
0.4
0.4
10.7
0.24
4.1
1.2
1.9
2.6
1.6
1.5
2.4
1.7 ( 0.8
45 ( 20
2.3 ( 1
11 ( 1
152 ( 40
27.2 ( 6.5
2.83 ( 0.60
2.01 ( 0.85
960 ( 40
419 ( 36
640 ( 30
600 ( 230
224 ( 84
11.0 ( 4.1
5.5 ( 2.0
384 ( 196
297 ( 32
270 ( 59
690 ( 170
33.4 ( 72
5.2 ( 2.1
5.3 ( 2.1
1530 ( 300
645 ( 87
1525 ( 280
cocainec
cocained
cocainee
a
b
See Experimental Section for details. Values are mean ( standard error of two or three independent experiments, each conducted
with triplicate determinations. c Cocaine as reference for 1 and 7. dCocaine as the reference for 2, 13, and 24. e Cocaine as the reference
for 25 and 28.
noted. Proton (1H) and carbon (13C) nuclear magnetic reso-
nance (NMR) spectra were recorded at 300 or 500 and 75.5
Ta ble 4. Inhibition of DA, 5-HT, and NE Uptake to
HEK-hDAT, HEK-hSERT, andHEK-hNET Cellsa
MHz respectively, on a Bruker DPX-300 or DRX-500 spec-
trometer. Chemical shifts (δ) are reported in parts per million
discrimination
ratio
uptake,b IC50, nM
hSERT
uptake/bindingc
(ppm) downfield from tetramethylsilane (TMS) as the internal
standard. The ultraviolet spectra (UV) were obtained in 95%
EtOH on a Shimadzu Model UV-2101 PC ultraviolet spec-
trometer. Mass spectra (MS) were determined on a Micromass
Quattro II mass spectrometer using APCI or ES ionization
modes (positive or negative) and 0.2% ammonium formate in
50% aqueous acetonitrile as solvent. Thin-layer chromatogra-
phy (TLC) was carried out on all compounds using glass plates
coated with silica gel HF-254 (E. Merck, AG). If not otherwise
specified, chemicals were obtained from the Aldrich Chemical
Co.
compd
hDAT
hNET
DAT SERT NET
1
2
7
13
24
25
28
43 ( 20
94 ( 32
3.7 ( 0.4 53 ( 7
4.9 ( 0.5 0.9
4.9 ( 0.5 2.2
4.5 ( 1.5 1.5
323 ( 91 0.4
1.9
1.4
0.8
>17
14.3
0.3
0.3
0.4
0.5
0.4
7
66 ( 10
63 ( 25
85 ( 34
1.8 ( 1.3
>10 µM
3,210 ( 870 13.0 ( 1.7 3.2
880 ( 210 370 ( 40
366 ( 82 700 ( 160
38 ( 12
53 ( 14
311 34
182 127
10
cocained 168 ( 32 300 ( 101
cocainee 315 ( 78 420 ( 130
cocainef 330 ( 58 325 ( 171
583 ( 178 0.2
220 ( 76 0.8
192 ( 61 0.5
0.8
1.4
1.2
0.4
0.3
0.1
5-(1-Na p h th yl)-2,3-d ih yd r o-5H-im id a zo-[2, 1-a ]-isoin -
d ol-5-ol (6) a n d 1-Na p h th yl-[2-(4,5-d ih yd r o-1H-im id a zol-
2yl)-p h en yl]-m eth a n on e (6a ). To a stirred solution of 2.5M
n-BuLi in hexane (4.11 mL, 0.01 mol), N,N,N′,N′-tetrameth-
ylethylenediamine (1.19 g, 1.54 mL, 0.01 mol) and THF (10
mL) under a N2 atmosphere was added dropwise at room
temperature a solution of 2-phenylimidazoline (0.68 g, 0.005
mol) in THF (10 mL). After stirring at room temperature for
2 h, the solid was treated dropwise with a solution of methyl
1-naphthoate (1.30 g, 0.007 mol) in THF (10 mL) and then
stirred overnight. The mixture was cooled in an icebath and
treated dropwise with saturated NH4Cl solution (2.0 mL),
followed by anhydrous MgSO4 (ca. 10 g). The resultant solid
was filtered and washed with THF (10 mL), and the combined
filtrates were concentrated in vacuo to give 0.27 g (20%) of 6
and 6a , mp 209-210 °C dec (CH3OH/DMF); 1H NMR (DMSO-
d6) δ 3.35 (s, 4H), 4.09 (m, 0.05H), 4.23 (m, 0.05H), 6.83 (brs,
1H), 7.13 (d, 1H), 7.39 (t, 1H), 7.62 (m, 2H), 7.68 (m, 5H), 7.98
(d, 1H), 8.06 (d, 1H), 8.72 (d, 1H); 13C NMR (DMSO-d6) δ 60.00,
122.37, 124.13, 124.95, 125.54, 126.16, 126.63, 126.98, 127.35,
127.39, 127.85, 128.74, 129.36, 129.76, 130.19, 130.25, 131.02,
131.31, 131.61, 133.20, 136.24, 141.11, 163.07 (CdN), 197.44
(CdO); MS m/z 301 (MH+). Anal. (C20H16N2O) C, H, N.
a
b
See Experimental Section for details. Values are mean (
standard error of two or three independent experiments, each
conducted with triplicate determinations. c See Table 3 for binding
values. Cocaine as the reference for 1 and 7. e Cocaine as the
d
reference for 2, 13, and 24. f Cocaine as the reference for 25 and
28.
F igu r e 2. Sites (bold lines) where addition of a benzo or
cyclohexano ring enhanced binding at the DAT or DA uptake
inhibition of mazindol (1) or homomazindol (2).
Con clu sion s
The addition of a benzo ring at certain sites on rings
C and D and a cyclohexano ring on ring A of mazindol
(1) or homomazindol (2) produced several compounds
(7, 13, 24, 25, and 28) with potent DAT binding and
DA uptake inhibition (Figure 2). Some of these, in
particular 25 and 28, showed a considerable increase
in uptake/binding discrimination ratios over 1 and 2 at
the DAT, SERT, and NET. In vivo studies used to define
potential cocaine treatment agents are in progress or
completed on 7, 13, 24, 25 and 28 and will be published
elsewhere.
5-(2-Naph th yl)-2,3-dih ydr o-5H-im idazo-[2,1-a]-isoin dol-
5-ol (7). Following the procedure given for 6 and 6a , but using
methyl-2-naphthoate, gave 7 (29%), mp 185-186 °C dec (CH3-
1
OH/DMF); H NMR (DMSO-d6) δ 2.93 (m, 2H), 4.18 (m, 2H),
6.95 (s, 1H), 7.29 (d, 2H), 7.46-8.01 (m, 7H), 8.11 (s, 1H); 13
C
NMR (DMSO-d6) δ 42.3, 60.9, 88.9 (C-5), 122.9, 124.8, 125.1,
125.2, 125.7, 127.1, 128.0, 128.3, 128.8, 129.1, 129.7, 130.9,
133.3, 133.6, 139.3, 155.3, 167.9 (CdN); MS m/z 301 (MH+).
Anal. (C20H16N2O) C, H, N.
5-(5,6,7,8-Tetr a h yd r o-2-n a p h th yl)-2,3-d ih yd r o-5H-im i-
d a zo-[2,1-a ]-isoin d ol-5-ol (8). Following the procedure for 6
and 6a , but using methyl 5,6,7,8-tetrahydro-2-naphththoate,
gave 8, (33%), mp 183-184 °C, dec (CH3OH/DMF); 1H (DMSO-
d6) δ 1.73 (m, 4H), 2.74 (m, 4H), 2.90 (q, 1H), 3.30 (q, 1H),
4.13 (m, 2H), 6.90 (t, 1H), 7.08 (s, 1H), 7.23 (t, 2H), 7.44 (m,
2H), 7.57 (m, 1H), 7.67 (m, 1H); 13C NMR (DMSO-d6) δ 22.45,
22.58, 28.72, 28.81, 41.36, 59.88, 87.99 (C-5), 125.80, 127.07,
Exp er im en ta l Section
Melting points were determined in a Thomas-Hoover capil-
lary melting point apparatus and are not corrected. Analyses
were performed by the Robertson Microlit Laboratories, Inc.,
Madison, NJ , and are within (0.4% of theory, unless otherwise

Benzo- and Cyclohexanomazindol Analogues
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19 4115
127.47, 129.08, 129.52, 130.20, 131.23, 136.12, 136.40, 136.47,
137.93, 154.79, 167.08 (CdN); MS m/z 30.5 (MH+). Anal.
(C20H20N2O) C, H, N.
(4,060), 292 (5,954), 305 (7,037), 318 (6,225) 332 (5,572), 356
(2,479); MS m/z 347 (MH+). Anal. (C21H15ClN2O) C, H, N.
13-(4-Ch lor op h en yl)-6,13-d ih yd r o-11H-isoin d olo-[2,1-
b]-[2,4]-ben zod ia zep in -13-ol (16). Following the procedure
used to prepare 14, but using 1, 2-benzenedimethanamine,
gave 16 (46%), mp > 225 °C (i-PrOH/DMF); 1H NMR (CD3-
OD) δ 4.38 (d, 1H), 4.49 (d, 1H), 4.70 (d, 1H), 4.85 (d, 1H),
6.93 (t, 2H), 7.12 (t, 1H), 7.19 (t, 1H), 7.27 (m, 2H), 7.35 (m,
5H), 7.58 (t, 1H); 13C NMR (DMSO-d6) δ 45.23, 50.51, 93.81
(C-13), 122.06, 123.23, 128.02, 128.54, 128.61, 128.98, 129.19,
129.46, 131.26, 133.37, 135.19, 137.78, 141.32, 141.76, 146.70,
153.93 (CdN); UV λ max 201 nm (ꢀ 21,500), 272 (1,820); MS
m/z 361 (MH+). Anal. (C22H17ClN2O) C, H, N.
2-(2-Na p h th yl)-4,5-d ih yd r o-1H-im id a zole (18). A solu-
tion of iodomethane (31.7 mL, 72.4 g, 0.51 mol), and 4,4-
dimethyl-2-(2-naphthyl)-4,5-dihydro-oxazole10 (17: 11.5 g, 0.051
mol) under a N2 atmosphere was stirred at room temperature
for 6 days. The resultant solid was filtered off to give 10.3 g
(55%) of 17‚MeI; MS m/z 241 (MH+).
A solution of 17‚MeI (10.0 g, 0.027 mol) and 1,2-diamino-
ethane (2.0 mL, 1.80 g, 0.03 mol) was stirred and refluxed for
10h, cooled in an icebath, and treated dropwise with 2N KOH
(50 mL) and toluene (50 mL). The organic layer was separated,
dried (MgSO4), and filtered, and the filtrate was concentrated
in vacuo to give 1.73 g (33%) of 18 mp 105-106 °C (acetone);
1H NMR (CDCl3) δ 3.91 (bs, 4H), 7.47 (m, 2H), 9.30 (m, 4H),
8.20 (s, 1H); MS m/z 197 (MH+). Anal. (C13H12N2) C, H, N.
5-(4-Ch lor op h en yl)-2,3-d ih yd r o-5H -b en z-[e]-im id a zo-
[2,1-a ]-isoin d ol-5-ol (19). To a stirred solution of 1.6M n-BuLi
in hexanes (8 mL, 0.013 mol) under a N2 atmosphere was
added dropwise a solution of 18 (1.00 g, 0.005 mol) in THF
(15 mL). After an additional 15 min, a solution of methyl
4-chlorobenzoate (1.75 g, 0.01 mol) in THF (15 mL) was added
dropwise and allowed to stir overnight. The mixture was
treated dropwise with saturated NH4Cl solution (5 mL), and
H2O (10 mL), and after 15 min, the resultant solid was filtered
off to give 0.41 g (25%) of a mixture of 19 and 22, mp 193-
196 °C; Rf 0.24 and 0.29 (5% MeOH/CH2Cl2); 1H NMR (DMSO-
d6) δ 2.63 (q, 0.5H), 3.01 (q, 0.5H), 3.38 (m, 1.0H), 4.14 (q,
0.4H), 4.23 (m, 0.6H), 6.91 (s), 7.02 (s), 7.18 (s), 7.23 (d), 7.38
(d), 7.42-7.48 (m), 7.58 (m), 7.68 (m), 7.75 (s), 7.81 (d), 7.87
(d), 7.97 (m), 8.02 (s), 8.05 (d), 8.10 (m), 8.32 (s), 8.33 (s); 13C
NMR (DMSO-d6) δ 41.1, 42.0, 60.8, 60.9, 88.9, 89.3, 119.5,
122.8, 123.6, 125.1, 126.1, 127.6, 128.0, 128.3, 128.5, 128.7,
128.9, 129.2, 129.3, 129.8, 129.9, 130.9, 131.2, 133.7, 135.2,
135.2, 135.9, 140.1, 141.5, 150.8, 151.3, 167.1, 167.7; MS m/z
335 (MH+).
A 30 mg sample of 19 and 22 was dissolved in CH2Cl2 (15
mL) and spotted on 25 plates of 20 × 20 cm Fisherbland plates
500 µM thickness and developed with 7% MeOH/CH2Cl2. The
upper spot was isolated to give 17 mg of 19, mp 194-196 °C
dec, Rf ) 0.24 (5% MeOH/CH2Cl2); 1H NMR (DMSO-d6) δ 2.68
(d, 1H), 3.38 (bs, 1H), 4.15 (bs, 2H), 7.01 (bs, 1H: OH), 7.28
(bs, 2H), 7.38 (d, 2H), 7.48 (t, 1H), 7.57 (t, 1H), 7.82 (d, 1H),
7.88 (bs, 1H), 8.05 (d, 1H), 8.10 (d, 1H); 13C NMR (DMSO-d6)
δ 42.00, 60.52, 88.94 (C-5), 119.08, 124.76, 126.02, 127.26,
127.48, 128.34, 128.81, 129.27, 130.63, 133.20, 135.51, 139.92,
150.52, 167.75 (CdN); MS m/z 335 (MH+). Anal. (C20H15ClN2O)
C, H, N.
11-(4-Ch lor op h en yl)-11H-isoin d olo-[2,1-a ]-ben zoim id a -
zole (10). To a stirred solution of 1,2-diaminobenzene (2.12
g, 0.02 mol) in concd. HCl (5 mL) was added slowly a solution
of 2-(4-chlorobenzoyl)-benzaldehyde (4.89 g, 0.02 mol) in acetic
acid (100 mL). The mixture was heated to 80 °C for 0.5h and
then concentrated in vacuo. The residue was treated with 2N
NaOH until basic to litmus paper, and the resultant solid was
filtered off and crystallized from 2-propanol to give 2.15 g (34%)
of 10, mp 199-202 °C; 1H NMR (DMSO-d6) δ 6.00 (s, 1H:
H-11), 7.00-7.62 (m, 10H), 7.8-8.2 (d, 2H); MS m/z 317 (MH+).
Anal. (C20H13ClN2) C, H, N.
11-(4-Ch lor op h en yl)-11H-isoin d olo-[2,1-a ]-ben zim id a -
zol-11-ol (11). To a stirred solution of 10 (1.00 g, 0.032 mol)
in dry DMF (15 mL) under a N2 atmosphere was added
portionwise sodium hydride (0.17 g, 0.07mol as 60% in mineral
oil). After ca. 0.5h stirring at room temperature, the N2 was
stopped and a slow stream of dry air was bubbled into the
solution for ca. 6h. The mixture was treated with CH3OH (0.5
mL) and then H2O (75 mL). The resultant solid was filtered
to give 0.64 g (61%) of 11, mp 220-221 °C (lit,8 mp 222-223
°C).
(()-cis-11-(4-Ch lor op h en yl)-5a ,6,7,8,9,9a -h exa h yd r o-
11H-isoin d olo-[2,1-a ]-ben zim id a zol-11-ol (12). A mixture
of 2-(4-chlorobenzoyl) benzaldehyde (1.44 g, 0.006 mol), cis-1,
2-diaminocyclohexane (1.0 g, 0.009 mol), and xylene (50 mL)
was stirred and refluxed in a flask equipped with a Dean-
Stark water separator until the H2O layer in the sidearm
remained constant (ca. 5 h). The mixture was concentrated in
vacuo and the residue dissolved in MeOH (20 mL) and stirred
at room temperature in the presence of air for 96h. The solid
was filtered off to give 0.57 g (28%) of 12, mp 183-184 °C (CH3-
OH/DMF); 1H NMR (DMSO-d6) δ 1.01-1.65 (m, 6H), 1.88 (m,
2H), 2.98 (m, 0.5H), 3.78 (m, 0.5H), 3.95 (quint., 0.5H), 4.08
(quint., 0.5H), 6.78 (s, 0.5H; OH), 7.00 (s, 0.5H; OH), 7.05 (t,
0.5H), 7.09 (d, 1H), 7.29 (d, 0.5H), 7.42 (m, 3H), 7.52 (m, 1H),
7.63 (d, 1H), 7.67 (t, 0.5H), 7.73 (m, 0.5H); 13C NMR (DMSO-
d6) δ 19.88, 20.35, 21.83, 22.01, 24.46,25.18, 30.47, 54.60, 56.02,
68.70, 68.87, 87.58 and 88.34 (C-11), 121.72, 121.95, 123.59,
123.79, 127.47, 127.75, 127.82, 127.87, 127.92, 128.27, 128.38,
128.85, 129.00, 129.94, 131.29, 131.60, 132.22, 136.88, 138.26,
141.96, 154.33, 154.36, 165.81 and 167.86 (CdN); MS m/z 339
(MH+). Anal. (C20H19ClN2O) C, H, N.
(()-tr a n s-11-(4-Ch lor op h en yl)-5a ,6,7,8,9,9a -h exa h yd r o-
11H-isoin d olo-[2,1-a ]-ben zim id a zol-11-ol (13). Prepared by
1
the procedure of Aeberli,9 et al. H NMR (DMSO-d6) δ 1.13-
1.67 (m, 5H), 1.78 (m, 2H), 2.04 (t, 0.4H), 2.27 (t, 0.6H), 3.07
(t, 0.6H), 3.43 (t, 0.4H), 3.58 (m, 1H), 6.82 (s, 0.5H; OH), 7.02
(d, 0.5H), 7.09 (s, 0.5H; OH), 7.18 (t, 0.5H), 7.32 (d, 0.5H), 7.44
(m, 3H), 7.55 (quint, 1H), 7.68 (m, 2.5H); 13C NMR (DMSO-
d6) δ 23.86, 23.95, 24.76, 25.08, 29.38, 29.68, 30.97, 31.18,
63.55, 65.74, 87.82 and 89.04 (C-11), 121.69, 122.07, 123.77,
124.07, 127.25, 127.73, 127.93, 127.95, 128.18, 128.96, 129.19,
129.84, 131.41, 131.92, 132.29, 132.38, 137.79, 141.93, 153.44,
153.82, 166.85 and 168.62 (CdN); MS m/z 339 (MH+).
12-(4-Ch lor op h en yl)-10,12-d ih yd r o-isoin d olo-[1, 2-b]-
qu in a zolin -12-ol (14). A mixture of 2-(4-chlorobenzoyl)-
benzaldehyde (4.89 g, 0.02 mol), 2-aminobenzylamine (3.67 g,
0.03 mol) and xylene (150 mL) was stirred and refluxed in a
flask equipped with a Dean-Stark water separator until the
H2O layer in the sidearm remained constant. The solution was
concentrated in vacuo and the semisolid residue treated with
i-PrOH (75 mL) and stirred at room temperature in the
presence of air for ca. 56h. The resultant light yellow solid was
filtered off to give 2.68 g (39%) of 14, mp > 225 °C (i-PrOH/
DMF); 1H NMR (CD3OD) δ 4.13 (d, 1H), 4.62 (d, 1H), 7.02 (m,
2H), 7.12 (s, 1H; OH), 7.17 (d, 1H), 7.28 (t, 1H), 7.45 (m, 4H),
7.53 (t, 1H), 7.87 (t, 1H); 13C NMR (DMSO-d6) δ 92.86 (C-12),
121.57, 122.80, 123.72, 125.55, 127.75, 128.86, 128.93, 129.12,
129.51, 130.05, 132.50, 132.73, 133.74, 138.83, 143.74, 148.49,
155.53 (CdN); UV λ max 203 nm ( ° 29,500) 224 (22,463), 275
11b-(4-Ch lor op h en yl)-1,2,3,11b-tetr a h yd r o-5H-ben z-[f]-
im id a zo-[2,1-a ]-isoin d ol-5-on e (21). A mixture of 3-(4-
chlorobenzoyl)-2-naphthoic acid (4.00 g, 0.013 mol), 1,2-
diaminoethane (1.55 g, 1.72 mL, 0.026 mol), and xylene (100
mL) was stirred and refluxed in a flask equipped with a Dean-
Stark water separator until the H2O layer in the sidearm
remained constant (ca. 7h). The mixture was stirred overnight
at room temperature, and the resultant solid was filtered off
to give 3.85 g (90%) of 21, mp 213-215 °C (i-PrOH) (lit.11 mp
1
217-219 °C); H NMR (DMSO-d6) δ 2.18 (brs, 1H; NH), 3.21
(m, 2H), 3.68 (m, 1H), 3.91 (m, 1H), 7.35 (d, 2H), 7.53 (m, 2H);
7.68 (m, 3H), 7.79 (t, 1H), 7.95 (t, 1H), 8.35 (s, 1H); MS m/z
335 (MH+).
5-(4-Ch lor op h en yl)-2,3-d ih yd r o-5H -b en z-[f]-im id a zo-
[2,1-a ]-isoin d ol-5-ol (22). To a stirred mixture of LiAlH4

4116 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19
Houlihan et al.
(0.144 g, 0.004 mol) and THF (25 mL) under a N2 atmosphere
and cooled in an icebath was added dropwise a solution of 21
(1.005 g, 0.003 mol) in THF (15 mL) at such a rate that the
internal temperature did not excede 18 °C. After an additional
15 min, it was treated with 2N KOH (0.45 mL), H2O (0.6 mL),
and anhydr. Na2SO4 (4 g) and stirred overnight at room
temperature. The solid was filtered off to give 0.257 g (25%)
of 22, Rf ) 0.29 (5% MeOH/CH2Cl2) mp 194-195 °C (lit.11 mp
224-226 °C); 1H NMR (DMSO-d6) δ 3.01 (q, 1H), 3.38 (m, 1H),
4.21 (m, 2H), 6.79 (s, 1H; OH), 7.43 (A2B2, 4H), 7.58 (m, 2H),
7.75 (s, 1H), 7.93 (m, 1H), 8.11 (m, 1H), 8.33 (s, 1H); 13C NMR
(DMSO-d6) δ 41.26, 60.08, 87.58 (C-5), 122.02, 122.85, 125.26,
126.86, 127.58, 128.18, 128.45, 128.58, 129.16, 132.57, 132.95,
134.47, 140.71, 150.53, 166.42 (CdN); MS m/z 335 (MH+).
Anal. (C20H15ClN2O) C, H, N.
tr a n s-2-(1-Na p h th yl)-3a ,4,5,6,7,7a -h exa h yd r o-1H-ben -
zim id a zole (29). A solution of trans-1, 2-diaminocyclohexane
(28.55 g, 30.70 mL, 0.25 mol), p-toluenesulfonic acid monohy-
drate (47.60 g, 0.25 mol) in H2O (50 mL) was stirred at room
temperature for ca. 23h, concentrated in vacuo to give 49.2 g
(69%) of trans-1, 2-diaminocyclohexane‚p-tol SO3H acid salt,
mp > 225 °C (i-PrOH); 1H NMR (DMSO-d6) δ 1.10 (d, 4H),
1.35 (d, 2H), 1.75 (d, 2H), 2.21 (s, 3H), 2.40 (t, 2H), 7.05 (d,
2H), 7.43d (2H). Anal. (C13H22N2O3S) C, H, N.
A stirred mixture of 1-cyanonaphthalene (7.66 g, 0.05 mol),
and the above salt (15.75 g, 0.055 mol) was heated in an oilbath
to 225 °C for 3h, allowed to cool to room temperature, dissolved
in DMF (35 mL), poured into H2O (100 mL), and stirred for
ca. 12h. The solid was filtered off to give 7.75 g (37%) of 29‚
p-TolSO3H, mp 180-181 °C (DMF/i-PrOH); MS m/z 251
(MH+). Anal. (C24H26N2O3S) C, H, N.
A mixture of 29‚p-TolSO3H (6.32 g, 0.015 mol), 5N NaOH
(20 mL) and CH2Cl2 (50 mL) was stirred at room temperature
for ca. 2h. The CH2Cl2 layer was separated, dried (MgSO4),
filtered, and concentrated in vacuo to give 3.52 g (94%) of 29,
mp 184-185 °C (i-PrOH); 1H NMR (DMSO-d6) δ 1.35 (q, 2H),
1.52 (d, 2H), 1.82 (d, 2H), 2.21 (d, 2H), 3.05 (d, 2H), 7.00 (s,
1H: NH), 7.53 (m, 3H), 7.73 (d, 1H), 7.96 (m, 2H), 8.83 (q,
1H); MS m/z 251 (MH+). Anal. (C17H18N2) C, H, N.
(()-tr a n s-13-(4-Ch lor op h en yl)-7a ,8,9,10,11,11a -h exa h y-
d r o-13H -b en zo-[e]-isoin d olo-[2,1-a ]-b en zoim id a zol-13-
ol (30). To a stirred solution of 1.6M n-BuLi in hexanes
(4.8 mL, 0.008 mol) under N2 and cooled in an icebath was
added dropwise a solution of 29 (1.00 g, 0.0035 mol) in THF
(25 mL). After stirring an additional 15 min, the mixture was
5-(4-Ch lor op h en yl)-2,3-d ih yd r o-5H-b en z-[g]-im id a zo-
[2,1-a ]-isoin d ol-5-ol (24). To a stirred solution of 2-(1-
naphthyl)-4, 5-dihydro-1H-imidazole13 (23: 1.26 g, 0.0064 mol)
in THF under a N2 atmosphere was added at room tempera-
ture a solution of 1.6 M n-BuLi in hexanes (10 mL, 0.016 mol).
After an additional 2h stirring, the solution was treated
dropwise with a solution of methyl-4-chlorobenzoate (2.18 g,
0.013 mol). The mixture was cooled in an icebath and treated
dropwise with a saturated NH4Cl solution (3 mL), H2O (4 mL)
and stirred overnight at room temperature. The resultant solid
was filtered off to give 0.462 g (22%) of 24, mp 200-201 °C
1
dec (i-PrOH/DMF); H NMR (DMSO-d6) δ 2.93 (q, 1H), 3.36
(q, 1H), 4.28 (m, 2H), 6.98 (s, 1H; OH), 7.38 (d, 1H), 7.44 (A2B2,
4H), 7.63 (t, 1H), 7.72 (t, 1H), 8.05 (t, 2H), 8.27 (d, 1H); 13C
NMR (DMSO-d6) δ 41.93, 61.48, 88.68 (C-5), 121.88, 123.04,
125.43, 127.74, 128.24, 128.87, 129.02, 129.24, 129.31, 132.88,
133.42, 133.78, 140.46, 154.15, 168.56 (CdN); MS m/z 335
(MH+). Anal. (C20H15ClN2O) C, H, N.
treated dropwise with
a solution of methyl p-chloroben-
zoate (1.19 g, 0.007 mol) in THF (10 mL) and then stirred
overnight at room temperature. The mixture was treated
dropwise with saturated NH4Cl solution (5 mL), H2O (5 mL),
and MgSO4 (5 g). The solids were filtered off and the fil-
trate concentrated in vacuo to give 0.44 g (32%) of 30, mp
5-(2-Na p h th yl)-2,3-d ih yd r o-5H-ben zo-[g]-im id a zo-[2,1-
a ]-isoin d ol-5-ol (25). Following the procedure for 24, but
using methyl-2-naphthoate, gave 25 (33%), mp 218-219 °C
1
216-217 °C, dec (CH3OH); H NMR (DMSO-d6) δ 1.28-1.51
1
dec (CH2Cl2/MeOH); H NMR (DMSO-d6) δ 2.93 (q, 1H), 4.32
(m, 2.5H), 1.64-1.75 (m, 3H), 1.88 (t, 2H), 2.42 (d, 1.0H), 3.18
(t, 0.5H), 3.65 (t, 0.5H), 3.76 (t, 0.5H), 6.96 (s, 0.5H: OH), 7.12
(d, 0.5H), 7.22 (s, 0.5H), 7.36 (d, 0.5H), 7.48 (m, 3H), 7.62-
7.81 (m, 4H), 8.08 (d, 1H), 8.15 (d, 0.5H), 8.18 (d, 0.5H); 13C
NMR (DMSO-d6) δ 24.8, 24.9, 25.7, 26.0, 30.3, 30.6, 31.9,
32.1, 64.6, 66.4, 78.2, 78.3, 88.9 and 90.1 (C-13), 121.8, 122.1,
123.2, 123.7, 125.5, 127.8, 127.9, 128.3, 128.9, 129.0, 129.1,
129.3, 129.4, 132.9, 133.4, 133.8, 138.1, 142.5, 153.5, 153.8,
(m, 2H), 7.01 (s, 1H; OH), 7.32 (d, 1H), 7.42 (d, 1H), 7.54 (m,
2H), 7.67 (t, 1H), 7.77 (t, 1H), 7.90 (m, 1H), 8.01 (t, 1H), 8.08
(d, 2H), 8.17 (s, 1H), 8.82 (d, 1H); 13C NMR (DMSO-d6) δ 41.21,
60.57, 88.21 (C-5), 121.18, 122.24, 124.22, 124.55, 125.01,
126.23, 126.78, 127.37, 127.89, 127.94, 128.21, 128.41, 131.91,
132.46, 132.74, 132.85, 137.87, 153.62, 167.87 (CdN); MS m/z
351 (MH+). Anal. (C24H18N2O) C, H, N.
168.5 and 170.4 (CdN); MS m/z 389 (MH+). Anal. (C24H21
-
2-(1-Na p h t h yl)-1,4,5,6-t et r a h yd r op yr im id in e (27). A
stirred mixture of 1, 3-diaminopropane (14.80 g, 16.8 mL, 0.20
mol), 1-cyanonaphthalene (14.05 g, 0.09 mol), and p-toluene-
sulfonic acid monohydrate (19.1 g, 0.10 mol) was heated in an
oil bath at 210-220 °C for 2.5 h. The viscous mixture was
cooled to ca. 95 °C, treated with H2O (100 mL), allowed to cool
to room temperature, and then treated with 5N NaOH (75
mL). After stirring overnight, the solid was filtered off and
crystallized from toluene to give 10.80 g (57%) of 27, mp 152-
154 °C; 1H NMR (CDCl3) δ 1.95 (quin, 2H), 3.55 (t, 4H), 7.40-
7.58 (m, 4H), 7.84 (d, 2H), 8.26 (dd, 1H); MS m/z 211 (MH+).
Anal. (C14H14N2) C, H, N.
ClN2O) C, H, N.
2-(9-P h en a n th r yl)-4,5-d ih yd r o-1H-im id a zole (31). A so-
lution of 9-cyanophenanthrene (4.07 g, 0.02 mol), 1,2-diami-
noethane (1.04 g, 1.6 mL, 0.024 mol), p-toluene sulfonic acid‚
H2O (2.28 g, 0.012 mol), and ethylene glycol (15 mL) under N2
was stirred and refluxed for ca. 48 h. The mixture was treated
at room temperature with 3% NaOH (100 mL) and stirred for
3 h, and the resultant solid filtered off to give 2.50 g (50%) of
31, mp 190-191 °C; 1H NMR (DMSO-d6) δ 3.76 (bs, 4H), 6.98
(s, NH), 7.70 (m, 4H), 8.05 (d, 1H), 8.14 (s, H-10), 8.88 (m,
3H); 13C NMR (DMSO-d6) δ 122.76, 122.92, 126.69, 126.83,
127.04, 127.69, 127.82, 127.85, 128.93, 129.09, 129.87, 130.10,
130.21, 164.24 (CdN); MS m/z 247 (MH+). Anal. (C17H14N2),
C, H, N.
5-(4-Ch lor oph en yl)-2,3-dih ydr o-5H-diben z-[e.g]-im idazo-
[1, 2-a ]-isoin d ol-5-ol (32). To a stirred solution of 31 (1.23 g,
0.005 mol) and THF (20 mL) under argon was added dropwise
at room temperature 1.6M n-BuLi in hexanes (9 mL, 0.014
mol). After an additional 15 min, the mixture was treated
dropwise with a solution of methyl 4-chlorobenzoate (1.70 g,
0.01 mol) in THF (5 mL), stirred overnight, and then treated
with saturated NH4Cl solution (10 mL) and H2O (5 mL). After
ca. 15 min, the resultant solid was filtered off to give 0.76 g
(39%) of 32, mp 226 °C dec (CH3OH/CH2Cl2); 1H NMR (DMSO-
d6) δ 2.60 (q, 1H), 3.35 (q, 1H), 4.23 (q, 2H), 7.30 (s, 1H, OH),
7.35 (s, 1H), 7.39 (m, 3H), 7.60 (t, 1H), 7.75 (t, 1H), 7.84 (m,
2H), 8.02 (d, 1H), 8.95 (d, 1H), 9.05 (t, 1H); 13C NMR (DMSO-
d6) δ 42.04, 60.75, 88.76 (C-5), 122.50, 123.79, 124.22, 125.88,
6-(4-Ch lor op h en yl)-2,3,4,6-tetr a h yd r o-ben z-[g]-p yr im -
id o-[2,1-a ]-isoin d ol-6-ol (28). To a stirred solution of 1.6 M
n-BuLi in hexanes (15.6 mL, 0.025 mol) under N2 at room
temperature, there was added dropwise a solution of 27 (2.10
g, 0.01 mol) in THF (40 mL). After an additional 2h, the
mixture was treated dropwise with a solution of methyl
4-chlorobenzoate (4.26 g, 0.025 mol) in THF (25 mL), stirred
overnight, and then treated with saturated NH4Cl solution (7
mL). After stirring 2h, the resultant solid was filtered off to
give 0.70 g (21%) of 28, mp 220-221 °C; 1H NMR (DMSO-d6)
δ 1.82 (m, 2H), 2.91 (m, 1H), 3.58 (m, 2H), 6.78 (s, 1H; OH),
7.28 (d, 1H), 7.42 (s, 4H), 7.57 (t, 1H), 7.62 (t, 1H), 7.94 (t,
2H), 9.40 (d, 1H); 13C NMR (DMSO-d6) δ 20.50, 36.55, 44.30,
90.73 (C-6), 119.99, 124.90, 126.16, 126.33, 127.36, 128.04,
128.25, 128.45, 131.03, 132.45, 133.22, 139.85, 146.57,
155.17, 162.30 (CdN), MS m/z 349 (MH+). Anal. (C21H17ClN2O)
C, H, N.

Benzo- and Cyclohexanomazindol Analogues
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19 4117
126.02, 126.07, 126.57, 127.48, 128.14, 128.22, 128.53, 128.91,
130.88, 132.25, 132.81, 139.13, 149.91, 167.79 (CdN); MS m/z
385 (MH+). Anal. (C24H17ClN2O) C, H, N.
2.5 mM CaCl2, 1.2 mM MgSO4, 10µM pargyline, 100 µM
tropolone, 0.2% glucose and 0.02% ascorbic acid, buffered with
25 mM HEPES), 25 µL of [125I] RTI-55 (40-80 pM final
concentration), and additional buffer sufficient to bring the
final volume to 250 µL. The membranes are preincubated with
unknowns for 10 min prior to the addition of [125I] RTI-55. The
assay tubes were incubated at 25 °C for 90 min, and the
binding was terminated by filtration over GF/C filters using
a Tomtec 96-well cell harvester. Filters are washed for six
seconds with ice-cold saline. Scintillation fluid was added to
each square and radioactivity remaining on the filter was
determined using a Wallac µ- or â-plate reader. Specific
binding was defined as the difference in binding observed in
the presence and absence of 5 µM mazindol (HEK-hDAT and
HEK-hNET) or 5 µM imipramine (HEK-hSERT). Two or three
independent competition experiments were conducted with
duplicate determinations. GraphPAD Prism was used to
analyze the ensuing data, with IC50 values converted to Ki
values using the Cheng-Prusoff equation.
In h ibition of [3H] Neu r otr an sm itter Uptake in HEK293
Cells Exp r essin g Recom bin a n t Biogen ic Am in e Tr a n s-
p or ter s. HEK293 cells expressing hDAT, hSERT, or hNET
were grown to confluence as described above. The medium was
removed, and the cells were washed twice with phosphate
buffered saline (PBS) at room temperature. Following the
addition of Krebs Hepes buffer (3 mL), the plates were warmed
in a 25 °C water bath for 5 min. The cells were gently scraped
and then triturated with a pipet. Cells from multiple plates
were combined. One plate provides enough cells for 48 wells,
which is required to generate data on two complete curves for
the test compounds. Krebs-HEPES (350 µL) and test com-
pounds (50 µL) were added to 1-mL vials and placed in a 25
°C water bath. Specific uptake is defined as the difference in
uptake observed in the presence and absence of 5 µm imi-
pramine (HEK-hSERT). Cells (50 µL) are added and preincu-
bated with the unknowns for 10 min. The assay is initiated
by the addition of [3H] dopamine, [3H] serotonin, or [3H]
norepinephrine (50 µL, 20 nM final concentration). Filtration
through Whatman GF/C filters presoaked in 0.05% polyeth-
ylenimine is used to terminate uptake after 10 min. The IC50s
are calculated applying the GraphPAD Prism program to
triplicate curves made up of 6 drug concentrations each. Two
or three independent determinations of each curve are made.
Additional details about this assay have been published.
5-(2-Na p h th yl)-2,3-d ih yd r o-5H-d iben zo-[e.g]-im id a zo-
[1, 2-a ]-isoin d ol-5-ol (33). Following the procedure for 32,
but using methyl 2-naphthoate, gave 33 (28%), mp 212-214
°C dec (CH2Cl2/MeOH); 1H NMR (DMSO-d6) δ 2.59 (q, 1H),
3.38 (q, 1H), 4.23 (m, 2H), 7.07 (d, 1H), 7.34 (s, 1H: OH), 7.48
(d, 3H), 7.66 (t, 1H), 7.78 (d, 1H), 7.83 (m, 4H), 7.95 (d, 1H),
8.08 (d, 1H), 8.18 (s, 1H), 8.88 (m, 2H), 9.07 (d, 1H); 13C NMR
(DMSO-d6) δ 41.77, 60.35, 88.79 (C-5), 122.21, 123.40, 123.47,
123.75, 123.82, 125.00, 125.45, 125.75, 126.31, 126.38, 127.02,
127.33, 127.48, 127.68, 127.86, 128.16, 130.51, 131.84, 132.40,
132.78, 137.24, 149.85, 167.67 (CdN); MS m/z 351 (MH+).
Anal. (C28H20N2O) C, H, N.
[3H]-WIN 35,428 Bin d in g Assa ys. Rat. Brains from male
Sprague-Dawley rats weighing 200-225 g (Taconic Labs)
were removed, striatum dissected, and placed on ice. Mem-
branes were prepared by homogenizing tissues in 20 volumes
(w/v) of ice cold modified Krebs-HEPES buffer (15 mM HEPES,
127 mM NaCl, 5 mM KCl, 1.2 mM MgSO4, 2.5 mM CaCl2, 1.3
mM NaH2PO4, 10 mM D-glucose, pH adjusted to 7.4) using a
Brinkman Polytron (setting 6 for 20 s.) and centrifuged at
20 000×g for 10 min at 4 °C. The resulting pellet was
suspended in buffer, recentrifuged, and resuspended in buffer
to a concentration of 10 mg/mL. Ligand binding experiments
were conducted in assay tubes containing 0.5 mL modified
Krebs-HEPES buffer for 60 min on ice. Each tube contained
1.5 nM [3H] WIN 35,428 (specific activity 84 Ci/mmol) and 2.5
mg striatal tissue (original wet weight). Nonspecific binding
was determined using 0.1 mM cocaine HCl. For determination
of binding affinity, triplicate samples of membranes were
preincubated for 5 min in the presence or absence of the
compound being tested. Incubations were terminated by rapid
filtration through Whatman GF/B filters, presoaked in 0.1%
BSA, using a Brandel R48 filtering manifold (Brandel Instru-
ments Gaithersburg, Maryland). The filters were washed twice
with 5 mL cold buffer and transferred to scintillation vials.
Beckman Ready Safe (3.0 mL) was added, and the vials were
counted the next day using a Beckman 6000 liquid scintillation
counter (Beckman Coulter Instruments, Fullerton, CA). Data
were analyzed by using GraphPad Prism software (San Diego,
CA).
B.Gu in ea P ig. Compounds listed in Table 1 were tested
at NOVASCREEN, a division of Oceanix Biosciences Corpora-
tion, Hanover, MD. In brief, guinea pig striatal membranes,
[3H] WIN 35,428, and the test compound at concentrations of
10-10 to 10-6 M in DMSO/H2O), were incubated with 50 mM
TRIS-HCl (pH 7.4) containing 100 mM NaCl at 25 °C for 2 h.
The reaction assay was terminated by rapid vacuum filtration
onto glass fiber filters. Radioactivity trapped onto the fibers
is determined and compared to control values in order to
ascertain any interaction of test compound with the uptake
site.
Ack n ow led gm en t. Part of this work was supported
by grants (RO3 DA08516-02, RO1 DA10533-01) to
W.J .H. from the National Institute on Drug Abuse
(NIDA). The authors acknowledge the service of Dr. A.
J anowsky of the VA Medical Center, Portland, OR, for
the HEK cell studies (N01DA-3-8303). Special thanks
to Dr. J amie Biswas of NIDA for coordinating the
testing at the above NIDA centers. Analytical services
were provided by Drs. Emil Fu and Mark Hayward,
Leonard Hargiss, J im Munson, Richard Beveridge, and
their associates at Novartis Pharmaceuticals, Summit,
NJ .
[
125I] RTI-55 Bin d in g Assa y. All test compounds were
prepared as 10 mM stock solution in DMSO. Subsequent
dilutions were made in assay buffer, achieving a final concen-
tration of 0.1%. Pipetting was conducted using a Biomek 2000
robotic work station. HEK293 cells expressing hDAT, hSERT,
or hNET inserts were grown to 80% confluence on 150 mm
diameter tissue culture dishes and serve as the tissue source.
The medium was poured off the plate, the plate was washed
with 10 mL of calcium- and magnesium-free phosphate-
buffered saline, and lysis buffer (10 mL; 2 mM HEPES with 1
mM EDTA) was added. After 10 min, cells were scraped from
the plates, poured into centrifuge tubes, and centrifuged
30 000g for 20 min. The supernatant fluid was removed, and
the pellet was resuspended in 12-32 mL of 0.32 M sucrose
using a Polytron at setting 7 for 10 s. The resuspension volume
depends on the density of binding sites within a cell line and
is chosen to reflect binding of 10% or less of the total
radioactivity. Each assay tube contained 50 µL of membrane
preparation (about 10-15 µg of protein), 25 µL of test
compound or buffer (Krebs-HEPES, pH 7.4; 122 mM NaCl,
Refer en ces
(1) Carrol, F. I.; Lewin, A. H.; Kuhar, M. J . Dopamine Transporter
Uptake Blockers: Structure Activity Relationships. In Neu-
rotransmitter Transporters: Structure and Function; Reith, M.
E. A., Ed.; Humana Press: Totawa, NJ , 1997; pp 263-295.
(2) (a) Kuhar, M. J .; Ritz, M. C.; Boja, J . W. The Dopamine
Hypothesis of the Reinforcing Properties of Cocaine. Trends
Neurosci. 1991, 14, 299-302. (b) Kuhar, M. J .; Ritz, M. C.;
Sharkey, J . Cocaine Receptors on Dopamine Transporters Medi-
ate Cocaine-Reinforced Behavior. In Mechanisms of Cocaine
Abuse and Toxicity; Clovet, D., Agshar, K. and Brown, R., Eds.;
U.S. Government Printing Office: Washington, D,C, 1998; pp
14-22. (c) Giros, B.; J aber, M.; J ones, S. R.; Wrightman, R. M.;
Caron, M. G. Hyperlocomotion and Indifference to Cocaine and
Amphetamine in Mice Lacking the Dopamine Transporter.
Nature 1996, 379, 606-612.

4118 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 19
Houlihan et al.
(3) Houlihan, W. J .; Kelly, L.; Pankuch, J .; Koletar, J .; Brand, L.;
J anowsky, A.; Kopajtic, T. A. Mazindol Analogues as Potential
Inhibitors of the Cocaine Binding Site at the Dopamine Trans-
porter. J . Med. Chem. 2002, 45, 4097.
(4) Houlihan, W. J .; Boja, J . W.; Parrino, V. A.; Kopajtic, T. A.;
Kuhar, M. J . Halogenated Mazindol Analogues as Potential
Inhibitors of the Cocaine Binding Site at the Dopamine Trans-
porter. J . Med. Chem. 1996, 39, 4935-4941.
(5) Houlihan, W. J .; Boja, J . W.; Kopajtic, T. A.; Kuhar, M. J .;
Degrado, S. J .; Toledo, L. Positional Isomers and Analogues of
Mazindol as Potential Inhibitors of the Cocaine Binding Site on
the Dopamine Transporter Site. Med. Chem. Res. 1998, 8, 77-
90.
(6) Houlihan, W. J .; Parrino, V. A.: Directed Lithiation of 2-Phenyl-
and 2-(o-Methylphenyl)imidazoline. J . Org. Chem. 1982, 47,
5177-5180.
(7) Metlesics, W.; Sternbach, L. H. U.S. Patent 3,905,994, 1975.
(8) Aberli, P.; Eden, P.; Gogerty, J . H.; Houlihan, W. J .; Penberthy,
C. Anorectic Agents 2. Structural Analogues of 5-(p-Chlorophe-
nyl)-2,3-dihydro-5H-imidazo [2, 1-a] isoindol-5-ol. J . Med. Chem.
1975, 18, 182-185.
(9) Eberle, M. K.; Houlihan, W. J .; Schirm, P. The Preparation of
11-Aryl-11H-isoindolo [2,1-b] benzimidazols. J . Org. Chem. 1973,
38, 3872-3874.
(10) Meyers, A. I.; Lutomski, K. A.; Lacher, D. Regio- and Stereo
Selective Addition of Organolithiums to Naphthalenes. An
Efficient Synthesis of 1,1,2-Trisubstituted and trans-2-disubsti-
tuted Dihydronaphthalenes. Tetrahedron 1988, 3107-3118.
(11) Sulkowski, T. S.; Mascitti, A. A. U.S. Patent 3,770,766, 1973.
(12) Waldmann, H.; Mathiowetz, H. U¨ bereinen neuen Aufbau in der
lin. Benzanthrachinon-Reihe. Chem. Ber. 1931, 64, 1713-1722.
(13) Hill, A. J .; J ohnston, J . V. Amidines Derived from Ethylenedi-
amine. II. Imidazolines. J . Am. Chem. Soc. 1954, 76, 922-923.
(14) (a) Davies, H. M. L.; Gilliatt, V.; Kuhn, L. A.; Saikali, E.; Ren,
P.; Hammond, P. S.; Sexton, T.; Childers, S. R. Synthesis of 2â-
Acyl-3â-(substituted naphthyl)-8-azabicyclo [3.2.1] octanes and
Their Binding Affinities at Dopamine and Serotonin Transporter
Sites. J . Med. Chem. 2001, 44, 1509-1515. (b) Davies, H. M. L.;
Kuhn, L. A.; Thornley, C.; Matasi, J . J .; Sexton, T.; Childress,
S. R. Synthesis of 3â-Aryl-8-azabicyclo [3.2.1] octanes with High
Binding Affinities and Selectivities for the Serotonin Transporter
Site. J . Med. Chem. 1996, 39, 2554-2558. (c) Davies, H. M. L.;
Saikali, E.; Huby, N. J .; Gilliatt, V. J .; Matsai, J . J .; Sexton, T.;
Childers, S. R. Synthesis of 2â-Acyl-3â-aryl-8-azabicyclo [3.2.1]
octanes and Their Binding Affinities at Dopamine and Serotonin
Transporter Sites in Rat Striatum and Frontal Cortex. J . Med.
Chem. 1994, 37, 1262-1268.
(15) Lieske, S. F.; Yang, B.; Eldefrawi, M. E.; MacKerell, J r., A. D.;
Wright, J . (-)-3â-Substituted Ecgonine Methyl Esters as Inhibi-
tors for Cocaine Binding and Dopamine Uptake. J . Med. Chem.
1998, 41, 864-876.
(16) 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 Monoamine
Transporters. J . Med. Chem. 2000, 43, 2982-2991.
(17) Tamiz, A. P.; Zhang, J .; Flippen-Anderson, J . L.; Zhang, M.;
J ohnson, K. M.; Deschaux, O.; Tella, S.; Kozikowski, A. P.
Further SAR Studies of Piperdine-Based Analogues of Cocaine.
2. Potent Dopamine and Serotonin Reuptake Inhibitors. J . Med.
Chem. 2000, 43, 1215-1222.
(18) Katz, J . L.; Izenwasser, S.; Kline, R. H.; Allen, A. C.; Newman,
A. H. Novel 3 -Diphenylmethoxytropane Analogues: Selective
Dopamine Uptake Inhibitors with Behavioral Effects Distinct
from Those of Cocaine. J . Pharmacol. Exp. Ther. 1999, 288, 302-
315.
(19) Avor, K. S.; Singh, S.; Seale, T. W.; Pouw, B.; Basmadjian, G. P.
2â-Substituted Analogues of 4′-Iodocaine: Synthesis and Dopam-
ine Transporter Binding Potencies. J . Med. Chem. 1998, 41,
2380-2389.
J M010301Z