Synthesis and monoamine transporter affinity of 3-aryl substituted trop-2-enes

Article abstract of DOI:10.1016/j.bmcl.2005.08.082

A series of new 3-aryl-tropanes have been synthesized, and their affinities and selectivities were evaluated for monoamine transporters. (1RS)-3-(Fluoren-2-yl)-8-methyl-8-azabicyclo[3.2.1]oct-2-ene exhibited the highest affinity for the human serotonin tr

Full text of DOI:10.1016/j.bmcl.2005.08.082

Bioorganic & Medicinal Chemistry Letters 15 (2005) 5488–5493
Synthesis and monoamine transporter affinity of 3-aryl
substituted trop-2-enes
Aleksej Krunic,a S. V. Santhana Mariappan,b,* Marteen E. A. Reithc and
William J. Dunn, IIId
aDepartment of Biopharmaceutical Sciences, College of Pharmacy MC 865, University of Illinois at Chicago,
833 S. Wood Street, Rm 335, Chicago, IL 60612-7231, USA
bDepartment of Chemistry, College of Liberal Arts and Sciences, CB-181, University of Iowa, Iowa City, IA 52242-1294, USA
cDepartment of Psychiatry, New York University School of Medicine, RM HN 604A, 550 1st Avenue, New York, NY 10016, USA
dDepartment of Medicinal Chemistry and Pharmacognosy, College of Pharmacy MC 781, University of Illinois at Chicago,
833 S. Wood St., Rm 539, Chicago, IL 60612-7231, USA
Received 6 July 2005; revised 26 August 2005; accepted 29 August 2005
Available online 3 October 2005
Abstract—A series of new 3-aryl-tropanes have been synthesized, and their affinities and selectivities were evaluated for monoamine
transporters. (1RS)-3-(Fluoren-2-yl)-8-methyl-8-azabicyclo[3.2.1]oct-2-ene exhibited the highest affinity for the human serotonin
transporter (IC50 = 14.5 nM). It is also 52-fold and 230-fold selective over human dopamine and norepinephrine transporters,
respectively.
Ó 2005 Elsevier Ltd. All rights reserved.
The transporter proteins involved in the neuronal
monoamine uptake are attractive targets for the treat-
ment of cocaine abuse and other psychiatric disor-
ders.1–4 Pharmacological treatment for cocaine abuse
and addiction, however, remains elusive. The focus on
the dopamine transporter (DAT) as the primary thera-
peutic target for cocaine addiction has been proposed
on the basis of self-administration studies, in which rein-
forcing properties of cocaine and related drugs correlate
with their potencies at inhibiting [3H]mazindol binding
of dopamine (DA), but not to binding of serotonin (5-
HT) or norepinephrine (NE) to their respective trans-
porters.5 The Ôdopamine hypothesisÕ assumes that
behaviors associated with cocaine addiction result from
the accumulation of dopamine in the synapse and its ac-
tions on one or more dopamine receptor subclasses.6
SERT knockouts developed place preference for co-
caine.8 Altogether, these data suggest that the 5-HT sys-
tem could interact with the DA system and contribute to
the reinforcing actions of cocaine. Conversely, in double
knockout mice that lack both SERT and NET, the
rewarding properties of cocaine were significantly en-
hanced, indicating that the actions of cocaine on the
NET might result in aversive effects.9
Among the many structurally diverse classes of monam-
ine uptake inhibitors, 3-aryl-tropanes have renewed inter-
est due to their strong affinities and selectivities.10–13
Recent conformational and binding studies revealed
definitive roles for the orientations of the aryl substituent
and the electron lone pair of the bridge head nitrogen in
determining their biological activities.10,12 SAR studies
on a number of rigid phenyl tropanes, in which the elec-
tron lone pair was fixed in either orientation (equatorial
or axial, relative to the piperidine ring), revealed that
the SERT favored equatorial orientation, while the
DAT showed a slightly higher preference for the axial
conformation.12 On the other hand, the selectivity of the
NET was not significantly affected.
However, cocaine blocks several monoamine transport-
ers, including those for serotonin (SERT) and norepi-
nephrine (NET). Recent evidence showed that double
DAT–SERT knockout mice failed to develop the condi-
tional place preference.7 In contrast, single DAT or
Keywords: Monoamine transporters; Uptake; Tropane.
*
Herein, we report on the activity of a series of semi-rigid
3-aryl-8-methyl-8-aza-bicyclo-[3.2.1]oct-2-enes (3-aryl-
Corresponding author. Tel.: +1 319 335 1336; fax: +1 319 335
1270; e-mail: santhana-velupillai@uiowa.edu
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.08.082

A. Krunic et al. / Bioorg. Med. Chem. Lett. 15 (2005) 5488–5493
5489
trop-2-enes) with the aim to probe the spatial require-
ments of the aryl binding pocket of the human mono-
amine transporters and to correlate the orientation of
the aryl moieties to the biological activity. As a refer-
Me
Me
Me
Me
Me
Me
Me
Me
ence,
a
few selected 3-aryl-8-methyl-8-azabicy-
clo[3.2.1]octane-3-ols, lacking the 2,3-double bond, are
also evaluated for their activity. The solution structures
of the ligand molecules have been determined by NMR
spectroscopy to establish orientational preferences of
the aryl and the N-methyl groups in relation to the tro-
pene skeleton (published elsewhere). These ligands can
be viewed either as cocaine derivatives lacking the 2-sub-
stituent, with the a,b-vinylene group incorporated into
the tropane skeleton (Fig. 1), or as the semi-rigid ana-
logs of styrene with spectroscopically detectable rota-
tional isomers.14,15 The nature of the selected aryl
groups confers various rotational degrees of freedom
resulting in distinctive conformations and overall molec-
ular shapes as a result of the hindered rotation about the
single bond (Fig. 2).
1
2
Figure 2. Structures of a,b-naphthylethylenes.15
and no nucleophilic attack occurs. The two generated
carbanions can also lead to two possible addition
products.
The compound 10a was formed during the isolation of
10 following the dehydration reaction as a result of
quaternization of N8 by dichloromethane (Scheme 1).
Tertiary amines were known to be quaternized readily
by dihalomethanes.19 However, the formation of 10a
as a result of carbene addition to the tertiary amine can-
not be completely ruled out, since among the six aryl-
tropenes synthesized, an alternative product was
formed only in the case of 10.20 In addition, high pres-
sure conditions are generally required for the quatern-
ization of tertiary amines by dichloromethane. The
preferential alkylation of N8 over the double bond
may be due to the steric hindrance introduced by the
orthogonal benzene ring.
Commercially available tropine-3-one, 14 (Scheme 1),
was used as a starting material in all the reactions.
The intermediate alcohols (3, 5, 7, 9, and 11) result from
the exo attack of the carbanions generated from the
organometallic reagent on 14, under an inert nitrogen
atmosphere and at different temperatures and solvents.
The mixtures were allowed to warm up to room temper-
ature and then quenched with a 1 M solution of the cold
hydrochloric acid. The alcohols 3, 5, 7, 9, and 11 were
then refluxed in neat trifluoroacetic acid for variable
reaction times, depending on the steric hindrance of
the tertiary alcohol, to give the alkenes 4, 6, 8, 10, and
12 (Scheme 1) in quantitative yields.16
Indole derivative 13 (Scheme 1), a rigid analog of sero-
tonin, was synthesized by refluxing 14 with indole and
NaOMe in MeOH. The direct route to 13 was chosen
over the two-step nucleophilic addition–dehydration
pathway to eliminate the protection–deprotection step
of the indole nitrogen.21
Aryllithium reagents were used for 5, 7, 9, and 11,
whereas arylmagnesium bromide was used for 3. Arylli-
thium reagents were synthesized starting from either the
appropriate arylbromides or n-butyllithium at ꢀ78 °C.
There are several reported reactions for the deprotona-
tion of the unsubstituted fluorene, among which depro-
tonation with n-butyllithium at the 0 °C produced the
highest yield for 5.17 The aryllithium reagent of 7 was
obtained by the deprotonation of 2-bromofluorene at
ꢀ25 °C, the optimal conditions found for this reaction.
The Grignard reagent was generated for alcohol 3 from
commercially available 2-bromofluorene. Methylene
bridge protons have pKa of 22 and so the deprotonation
becomes a competing reaction with the transmetalation
even at low temperature.18 Self-condensation prevails
In general, the yields in these reactions range from 5% to
43%. The lowest yield (cumulative overall yield) was ob-
served for 4 and is likely either due to the self-alkylation
or the reprotonation of the carbanions generated during
the course of the reaction. The same reasoning is appli-
cable to the transmetalation reaction. Lower yields ob-
served with 9, in comparison to 11, are attributed to
the increased steric hindrance in the 2-lithiumbiphenyl
carbanion that also influences its generation.
8
N
N
N
Me
Me
Me
COOMe
O
1
7
2
6
5
4
Ar
3
Ar
OH
O
I
II
cocaine
Figure 1. The structures of cocaine (1) and the tropane-based ligands (I and II). The numbering scheme of the tropane system is shown for cocaine.

5490
A. Krunic et al. / Bioorg. Med. Chem. Lett. 15 (2005) 5488–5493
Scheme 1. Synthetic route to 3-aryl-8-methyl-8-aza-bicyclo[3.2.1]oct-2-enes 3–13: Reagents and conditions: (a) ArBr, n-BuLi/THF, ꢀ78 °C to rt, (b)
ArBr (Ar = fluorenyl), Mg, MeI, Et2O, hm, reflux, (c) TFA, reflux, (d) indole, NaOMe, MeOH, reflux. The overall isolated yields of the products are
as follows: 4, 8%; 5, 19%; 6, 17%; 7, 5%; 9, 15%; 10, 15%; 10a, 4%; 11, 43%; 12, 42%; 13, 11%. The yields are not listed for 3 and 8; 3 was used without
purification in the dehydration step, while 8 was obtained in minimal quantities. Atom numbering for each 3-aryl substituent is also shown.
The fluorenyl group in C3 position of 3 and its alkene
derivative 4 (Scheme 1) has two structural effects: (a)
projects the extended sp2 functionality into the binding
pocket and (b) fixes the orientation of the methylene
bridge in such a way that the aromatic ring is almost
planar. Compounds 5, 6, 7, and 8 (Scheme 1) are the
positional isomers of 3 and 4. These molecules can
adopt at least two conformations, as dictated by the
sp3 hybridized carbon at C90. The fluorenyl ring can
be oriented either in plane or orthogonal to the double
bond. The latter orientation positions the benzene rings
above and below the plane of the tropane ring. The
biphenyl substituted tropanes 9, 10 and 11, 12 (Scheme
1) can be viewed as ortho and meta substituted biphe-
nyls. The lack of methylene linker makes them more
conformationally flexible than fluorenes 3 and 4. The
smaller pyrrole ring adjacent to the tropane ring, on
the other hand, is expected to decrease the steric barrier
for rotation and increase torsional flexibility of the in-
dole portion in 13.
were observed, in line with the earlier findings that cor-
related the effect of 2-substituents on the affinity toward
rat DAT (IC50 of 5180 nM measured for 3-exo-benzoyl-
oxytropane).23 Steric bulk directly below the tropane
ring, introduced by 3-endo OH, was tolerated to a cer-
tain extent only by the hSERT. Compound 5 is an
exception with submicromolar activity (543 nM for the
hDAT) and is also highly selective for the hDAT (65-
fold over the hSERT). For all other compounds the
selectivity was reversed from 1.6- to 6.5-fold in favor
of the hSERT. The reversal of selectivity observed be-
tween 5 and 7 is attributed to the introduction of a bro-
mine atom at 2-position of the fluorenyl moiety.
Considering the aryl binding pockets on the hDAT
and hSERT as narrow clefts of different sizes, it would
be expected that the two compounds to have different
biological activities.13,24,25 The single bond connecting
C3 and C90 allows rotational flexibility sufficient to pro-
vide a better fit of the aryl group into the binding pock-
et, enabling bromine atom in 5 to engage in electrostatic
and/or dipolar interactions with the amino acid residues
above the plane of the tropane ring. Presumably these
interactions are nonexistent with the hSERT. The steric
bulk below the proximal benzene ring was not tolerated
by any of the three human monoamine transporters, as
observed with the compounds 9 and 11. In the biphenyl
series, the addition of a second benzene in the meta po-
sition, 11, was particularly detrimental to the activity of
the hDAT. In the ortho substituted isomer 9, the rota-
tion of the distal benzene ring is restricted relative to
its counterpart in 11.
The transporter-binding affinities were measured by
using competitive binding assays against standard li-
gands in accordance with the NIDA protocol.22 The tri-
tiated (3H-labeled) WIN-35,428, CitalopramÒ, and
NisoxetineÒ were used as standard ligands for the
DAT, SERT, and NET, respectively.
Table 1 lists the affinities and selectivities of the 3-aryl-
tropan-3-ols 5, 7, 9, and 11 toward the hDAT, hSERT,
and hNET. Generally, affinities in the micromolar levels

A. Krunic et al. / Bioorg. Med. Chem. Lett. 15 (2005) 5488–5493
5491
Table 1. Affinities and selectivities of (1RS)-3-aryl-8-methyl-azabicyclo[3.2.1]octane-3-ols for human DAT, SERT and NET
Compound
IC50 (nM)a,b
Selectivity
hDAT
543 150
hSERT
hNET
hDAT/hSERT
hDAT/hNET
hSERT/hNET
5
7
35,400 7300
2570 1100
2520 400
158,000 5000
62,600 3100
66,300 4700
41,500 11,800
0.015
6.5
0.0034
0.27
0.06
0.22
16,800 1300
3980 430
21,900 4800
0.041
0.038
0.091
9
11
1.58
5.81
3770 1690
0.53
a IC50 is the concentration at which 50% inhibition occurs.
b IC50 values are means SEM from three independent experiments.
Much lower affinities were observed for hNET, as evi-
denced from the IC50 values for the displacement of
[3H]nisoxetine.
et al.24 Second, the double bond hinders rotation around
the single bond limiting the number of conformers.
Third, flattening of the piperidine ring in combination
with the nature of the aryl group influences the orienta-
tion of the N-methyl group. The quantitative nuclear
Overhauser effect analysis of the ligand molecules at
room temperature provided distance estimates of the
various interproton pairs and the subsequent restrained
energy minimization calculations revealed three-dimen-
sional solution structures including the preferential ori-
entations of the N-methyl group (equatorial N-methyl
orientation is favored for 4, 6, 10, and 12, whereas axial
conformation is favored for 13). The preferential orien-
tations of the N-methyl group in these molecules were
also further confirmed by dynamic NMR and inversion
magnetization transfer data. However, the low tempera-
ture dynamic NMR measurements revealed two popula-
tions for at least three sets of protons with the major
conformers being equatorial for 4, 6, 10, and 12, and axi-
al for 13, respectively. In accordance with the results of
Smith et al.12 (the ligands are expected to be in the axial
conformation to have higher activity and selectivity
toward hSERT), it is tempting to speculate that the
equilibrium is shifted to the conformer with the axial
N-methyl orientation in the presence of the transporter
proteins for 4, 6, 10, and 12.
Table 2 lists the affinities and selectivities of the 3-aryl-
substituted trop-2-ene ligands (4, 6, 10, 12, and 13) to-
ward the hDAT, hSERT, and hNET. The ligands
showed much lower activities for the hNET and were
generally selective for the hSERT. Compounds 4, 6,
12, and 13 showed submicromolar affinities for hSERT.
Compounds 4 and 13 were the only ones with submi-
cromolar affinities for the hDAT. DAT activity is sensi-
tive to steric bulk at the para position of the phenyl
ring.26 Compound 4 was found to be the most active
with an IC50 of 14.5 nM at the hSERT (Table 2) and
was also highly selective, namely 52-fold and 230-fold
selective for the hSERT over hDAT and hNET, respec-
tively. The second most active compound, 13, has re-
duced affinity for the hSERT, about 5-fold, with
slightly improved activity at the hDAT compared to 4.
Other ligands largely remained inactive with IC50 for
10a reaching the highest values of 67,000 and
16,100 nM for the hDAT and hSERT, respectively.
A multiplicity of factors can be attributed to the ob-
served high affinity and selectivity of the aryl tropenes
toward hSERT. First, the removal of the hydroxyl
group (4, 6, 10, 12, and 13) through dehydration of
the alcohols resulted in the flattening of the piperidine
ring, making it more rigid and placing the aromatic moi-
ety farther into the binding pocket.13,24,25 The binding
pocket, here, refers to the lipophilic pocket present in
the transporter proteins that is expected to interact with
the aryl group. The size of the aryl group has also been
suggested to play a role in determining the selectivity of
hSERT over hDAT.24 The improved selectivity of 4 in
comparison with the naphthyl analog reported in Appell
et al.10 is in line with the model proposed by Davis
Even though the biological activities can be correlated
satisfactorily to the conformational features of the li-
gands, the relative affinities cannot be discerned in terms
of various contributions. Hence, we cannot make any
definitive statements on whether the increased activity
of 13 compared to 6, 10, and 12 or the decreased activity
of 13 compared to 4 is a result of the differences in the
conformational equilibria.
Compound 10 displayed only slight improvements over
its alcohol precursor, most notably for the hNET
Table 2. Affinities and selectivities of (1RS)-3-aryl-8-methyl-azabicyclo[3.2.1]oct-2-enes for human DAT, SERT, and NET
Compound
IC50 (nM)a,b
Selectivity
hDAT
757 152
hSERT
hNET
3360 560
hDAT/hSERT
hDAT/hNET
hSERT/hNET
4
6
14.5 2.5
690 120
52.2
13.6
5.4
0.23
0.113
1.44
0.0043
0.0083
0.27
9350 170
9240 1370
67,000 17,300
7220 930
603 34
82,700 4800
6430 1810
2020 710
10
10a
12
13
1710 700
16,100 1400
670 142
4.2
33.2
8.0
0.029
0.018
23,000 300
3956 978
10.8
8.3
0.314
0.15
72.6 3.4
a IC50 is the concentration at which 50% inhibition occurs.
b IC50 values are means SEM from three independent experiments.

5492
A. Krunic et al. / Bioorg. Med. Chem. Lett. 15 (2005) 5488–5493
N+
N
Cl
Me
Me
1) 40% NaOH/CH2Cl2
2) H2O
10
10a
Scheme 2. Reaction conditions leading to quaternary salt 10a.
binding (10-fold). The second aromatic ring underneath
the tropane ring disturbs its proper orientation at the
binding sites of both the hDAT and hSERT. Reposi-
tioning of the aromatic ring in 12 led to a 3-fold increase
in the activity at the hSERT.
References and notes
1. Reith, M. E. A.; Sershen, H.; Allen, D. L.; Lajtha, A. Mol.
Pharmacol. 1983, 23, 600.
2. Javitch, J. A.; Blaustein, R. O.; Snyder, S. H. Mol.
Pharmacol. 1984, 26, 35.
3. Singer, H. S.; Hahn, I. H.; Moran, T. H. Ann. Neurol.
1991, 30, 558.
4. Pearce, P. K. B.; Seeman, P.; Jellinger, K.; Tourtelotte, W.
Eur. Neurol. 1989, 30, 9.
5. Ritz, M. C.; Lamb, R. J.; Goldberg, S. R.; Kuhar, M. J.
Science 1987, 237, 1219.
6. Boja, J. W.; Vaughan, R.; Patel, A.; Shaya, E. K.;
Kuhar, M. J. In Dopamine Receptors and Transporter;
Niznik, H. B., Ed.; Marcel Dekker: New York, 1994, pp
611–644.
Structurally, 10a resembles its precursor 10. The
added N-chloromethyl group is positioned over the
ethylene bridge (Scheme 2). In the case of 10a, the
steric bulk of the biphenyl group and the presence
of the positive charge altogether had negative impact
on the biological activity at the hDAT and hSERT.
The presence of the positive charge on the nitrogen
is detrimental to its biological activity; the methyl
iodide salt of the cocaine is completely inactive at
the DAT.27 Alternatively, whether 10a acts as an
irreversible inhibitor of the monoamine transporters
is not known.
7. Sora, I.; Hall, F. S.; Andrews, A. M.; Itokawa, M.;
Li, X. F.; Wei, H. B.; Wichems, C.; Lesch, K. P.;
Murphy, D. L. Proc. Natl. Acad. Sci. U.S.A. 2001, 98,
5300.
8. Sora, I.; Wichems, C.; Takanishi, N.; Li, X. F.; Zeng, Z.;
Revay, R.; Lesch, K. P.; Murphy, D. L.; Uhl, G. R. Proc.
Natl. Acad. Sci. U.S.A. 1998, 95, 7669.
9. Uhl, G. R.; Hall, F. S.; Sora, I. Mol. Psychiatry 2002, 7,
21.
10. Appell, M. D.; Dunn, W. J., III; Reith, M. E. A.; Miller,
L.; Flippen-Anderson, J. L. Bioorg. Med. Chem. 2002, 10,
1197.
11. Appell, M. D.; Berfield, J. L.; Wang, L. C.; Dunn, W. J.,
III; Chen, N.; Reith, M. E. A. Biochem. Pharmacol. 2004,
67, 293.
12. Smith, M. P.; Johnson, K. M.; Zhang, M.; Flippen-
Anderson, J. L.; Kozikowski, A. P. J. Am. Chem. Soc.
1998, 120, 9072.
13. Singh, S. Chem. Rev. 2000, 100, 925.
14. Anderson, J. E.; Hazlehurst, C. J. J. Chem. Soc., Chem.
Commun. 1980, 1188.
15. Anderson, J. E.; Barkel, D. J.; Cooksey, C. J. Tetrahedron
Lett. 1983, 24, 1077.
16. Perregaard, J.; Moltzen, E. K.; Meier, E.; Sanchez, C.
J. Med. Chem. 1995, 38, 1998.
The size and the orientation of the aryl moiety are
important and 13 serves as a good example. Reducing
the size of the aromatic ring from six-membered to
five-membered relaxes the rotation around the single
bond, thus decreasing the C2–C3–C20–C30 torsional an-
gle. At the same time, conjugation between two p sys-
tems becomes feasible. A relatively smaller heterocycle,
compared to the rest of the C3 substituents, has slightly
improved affinity for the hDAT, but the affinity for the
hSERT is five times lower than 4.
In conclusion, 3-aryl-trop-2-enes were found to have
good selectivity for the hSERT, and absence of the 2-
substituent was found not to be critical for activity at
the hSERT and to a certain level at the hDAT. The aryl
binding pocket at the hSERT appears to be an elongated
narrow cleft, tolerating a second additional phenyl ring,
unlike the hDAT. The shifting of the conformational
equilibrium toward the axial N-methyl conformer by
hSERT may contribute to its high selectivity.
17. Streitwieser, A., Jr.; Xie, L.; Wang, P.; Bachrach, S. M.
J. Org. Chem. 1993, 58, 1778.
18. Bors, D. A.; Kaufman, M. J.; Streitwieser, A., Jr. J. Am.
Chem. Soc. 1985, 107, 6975.
19. Almarzogi, B.; Grorge, A. V.; Isaacs, N. S. Tetrahedron
1986, 42, 601.
Acknowledgments
20. Saunders, M.; Murray, R. W. Tetrahedron 1960, 11, 1.
21. Palmer, S. L.; Mabic, S.; Castagnoli, N., Jr. J. Med. Chem.
1997, 40, 1982.
22. Deutsch, H. M.; Schweri, M. M.; Culbertson, C. T.;
Zalkow, L. H. Life Sci. 1994, 55, PL115.
The authors thank Drs. Dejan Nikolic and Benjamin
Johnson for recording of the mass spectra, and Lijuan
C. Wang and Janet L. Berfield of Dr. ReithÕs reaserch
group for the binding studies.

A. Krunic et al. / Bioorg. Med. Chem. Lett. 15 (2005) 5488–5493
5493
23. Reith, M. E. A.; Meisler, B. E.; Sershen, H.; Lajtha, A.
Biochem. Pharmacol. 1986, 35, 1123.
24. Davis, H. M. L.; Gilliatt, V.; Kuhn, L. A.; Saikali, E.;
Ren, P.; Hammond, P. S.; Sexton, T.; Childers, S. R.
J. Med. Chem. 2001, 44, 1509.
26. Carroll, F. I.; Mascarella, S. W.; Kuzemko, M. A.;
Abraham, P.; Parham, K.; Lewin, A. H.; Boja, J. W.
J. Med. Chem. 1994, 37, 2865.
27. Abraham, P.; Pitner, J. B.; Lewin, A. H.; Boja, J. W.;
Kuhar, M. J.; Carroll, F. I. J. Med. Chem. 1992, 35,
141.
25. Carroll, F. I. J. Med. Chem. 2003, 46, 1775.