Chemoenzymatic synthesis and evaluation of 3-azabicyclo[3.2.0]heptane derivatives as dopaminergic ligands

Article abstract of DOI:10.1016/j.ejmech.2012.07.025

New 3-azabicyclo[3.2.0]heptane derivatives were synthesized using a multicomponent reaction. Racemic compounds were efficiently resolved by kinetic resolution with immobilized lipase B of Candida antarctica (Novozym 435). The obtained compounds demonstrated greater binding affinity at D_2L and D₃ dopamine receptors compared to D₁ binding sites, and individual enantiomers of the same compound possessed distinct affinities.

Full text of DOI:10.1016/j.ejmech.2012.07.025

European Journal of Medicinal Chemistry 55 (2012) 255e261
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Chemoenzymatic synthesis and evaluation of 3-azabicyclo[3.2.0]heptane
derivatives as dopaminergic ligands
,
b
,
,
,b,
Reet Reinart-Okugbeni a 1, Kerti Ausmees c 1, Kadri Kriis c, Franz Werner c, Ago Rinken a
**
,
,
Tõnis Kangerc
*
a Institute of Chemistry, University of Tartu, Ravila 14, 50411 Tartu, Estonia
b Competence Centre on Reproductive Medicine and Biology, Tiigi 61b, 50410 Tartu, Estonia
c Department of Chemistry, Tallinn University of Technology, Akadeemia tee 15, 12618 Tallinn, Estonia
a r t i c l e i n f o
a b s t r a c t
Article history:
New 3-azabicyclo[3.2.0]heptane derivatives were synthesized using a multicomponent reaction. Racemic
compounds were efficiently resolved by kinetic resolution with immobilized lipase B of Candida ant-
arctica (Novozym 435). The obtained compounds demonstrated greater binding affinity at D2L and D3
dopamine receptors compared to D1 binding sites, and individual enantiomers of the same compound
possessed distinct affinities.
Received 7 October 2011
Received in revised form
9 July 2012
Accepted 17 July 2012
Available online 24 July 2012
Ó 2012 Elsevier Masson SAS. All rights reserved.
Keywords:
Kinetic resolution
CAL-B
3-Azabicyclo[3.2.0]heptane derivatives
Dopamine receptor activity
1. Introduction
heterocycle is a crucial feature. 3-Azabicyclo[3.1.0]hexane 1 deriv-
atives with sulfonamide [3] or triazol [4] are selective dopamine D3
Dopamine is a monoamine neurotransmitter and its receptors
are distributed throughout the body, both in CNS and in the
periphery. According to their structural differences and signaling
properties, dopamine receptors are divided roughly into D1-like
and D2-like receptors, of which D1-like receptors are comprised of
D1 and D5 subtypes and D2-like receptors include D2, D3 and D4
subtypes. Disturbances in brain dopamine receptor signaling lead
to such medical conditions as Parkinson’s disease, schizophrenia
and many other diseases resulting from genetic or environmental
factors [1].
The structureeactivity relationship of efficient dopamine
receptor modulators indicates that a wide variety of structural units
may be needed for the activity. However, a rigid heteroatom con-
taining a bicyclic scaffold is often present in compounds with
a therapeutic potential (Fig. 1) [2].
antagonists. 3-Azabicyclo[3.2.1]octane 2 benzamide derivative
SSR181507 [5] and indolyl-substituted bicyclic compounds [6] are
D2 antagonists. Benzamides of 3-azabicyclo[3.3.1]nonane 3 were
found to be non-selective and showed nearly identical binding to
dopamine D2 and D3 receptors [7]. Compounds with a 3-azabicyclo
[3.2.0]heptane structure 4 have been shown to bind dopamine D2-
like receptors [8]. Our recent paper describes the synthesis of tet-
rasubstituted 3-azabicyclo[3.2.0]heptane derivatives 9 (Scheme 1)
[9]. The method is characterized by a high diastereoselectivity,
simplicity and a wide substrate scope. The starting unsaturated
aldehyde 5 can be either aromatic, heteroaromatic or aliphatic, and
secondary amine 7 can be either acyclic or cyclic. In addition, the
primary hydroxyl group is a possible site for the further derivati-
zation of the compound. This multicomponent cascade reaction is
highly diastereoselective. The cascade involved an aza-Michael
Substituents at the azabicyclo template are usually different and
unique for every ring system. The ring size of the bicyclic
addition of benzyl aminocrotonate 6 to an iminium-activated a,b-
unsaturated aldehyde, followed by the intramolecular Michael
addition and an intramolecular Mannich-type ester enolate attack
to iminium ion. The initially formed bicyclic ester 8 was unstable
under reaction conditions and it was reduced in a one-pot proce-
dure with LiAlH4 to form the target 9.
From the synthetic aspect, the access to 3-azabicyclo[3.2.0]
heptanes is limited mainly to [2 þ 2] photocycloaddition methods
* Corresponding author. Tel.: þ372 6204371; fax: þ372 6202828.
** Corresponding author. Institute of Chemistry, University of Tartu, Ravila 14,
50411 Tartu, Estonia. Tel.: þ372 7375249; fax: þ372 7375264.
E-mail addresses: ago.rinken@ut.ee (A. Rinken), kanger@chemnet.ee (T. Kanger).
These authors have contributed equally.
1
0223-5234/$ e see front matter Ó 2012 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2012.07.025

256
R. Reinart-Okugbeni et al. / European Journal of Medicinal Chemistry 55 (2012) 255e261
resolved enzymatically (the structures are depicted in Table 2). The
method allowed us to obtain both enantiomers needed for the
following pharmacological studies. In most cases, the enantiose-
lectivity of the resolution was not high enough and a repeated
resolution strategy was used. It afforded both enantiomers with
sufficient enantiomeric excess. An acylated (B)-enantiomer (typi-
cally (ꢀ)-enantiomer) and nonacylated (A)-enantiomer (typically
(þ)-enantiomer) were separated by column chromatography and
acylated (B)-isomer was hydrolyzed under alkaline conditions.
Exceptions were the compounds 9e and 9f, with p-bromophenyl
and pyridyl substituents, respectively. They behaved to the
contrary, affording an (ꢀ)-enantiomer as nonacylated and
(þ)-enatiomer as acylated compounds.
Fig. 1. Azabicyclo scaffolds found among dopamine receptor modulators.
[10,11] or Au- [12,13] or Pt-mediated [14] cycloisomerization. Their
asymmetric synthesis has been even less explored. To the best of
our knowledge, crystallization via ditoluyl tartaric acid salt [15] and
Ir-catalyzed asymmetric allylic amination followed by Pt-catalyzed
cyclization [16] are the only examples described in the literature
so far.
For the determination of the absolute configuration the N-tosyl
derivative 11 of 3-azabicyclo[3.2.0]heptane was synthesized and
resolved (Scheme 3).
Kinetic resolution and hydrolysis of the obtained acylated B-
enantiomer afforded a highly crystalline compound. Its absolute
configuration was determined by using X-ray diffraction as (1S, 2R,
5R, 6R, 7R)-configuration (Fig. 2). It is assumed that all B-enantio-
mers of azabicycles that have been acylated by Novozym 435
possess the same configuration as shown.
Herein, we report the synthesis of various new derivatives of
3-azabicyclo[3.2.0]heptanes, a general method for the kinetic
resolution of the obtained racemic products by lipase-catalyzed
reaction, and assess the binding of obtained individual enantiomers
at their potential biological target, dopamine receptors. 3-
azabicyclo[3.2.0]heptanes are more likely to block D2-like recep-
tors [8], which is beneficial for antipsychotic drugs used i.e. for
treating schizophrenia or related disorders [1]. Therefore the aim of
the study was to determine selectivity profile of the synthesized
3-azabicyclo[3.2.0]heptane derivatives and find moieties that
would substantially affect ligand binding to three dopamine
receptor subtypes, D1, D2L and D3.
2.2. Ligand binding to dopamine receptors
Some azabicycloheptyl compounds, similar to the title
compounds, have been shown to bind to D2 and D3 subtypes of
dopamine receptors with a slight preference for the D3 subtype [8].
Therefore, we saw dopamine receptors as a potential target for the
present 3-azabicyclo[3.2.0]heptane derivatives and decided to test
the compounds binding affinities for three different dopamine
receptor subtypes, namely D1, D2L and D3 receptors. In our earlier
studies with dopaminergic compounds various host cell lines for
different receptor subtypes, both human and rat, had been used,
mainly for what was available [19,20]. However, the membrane
environment [21,22] as well as the origin of the receptor gene [23]
might have a substantial impact on ligand binding and signal
transduction properties, so we created a single host (HEK293)
based cell lines stably expressing individual subtypes of human
dopamine receptors (D1, D2L and D3), which is described in detail in
the Experimental Section (Sections 4.3 and 4.4).
All tested 3-azabicyclo[3.2.0]heptane derivatives showed
a preference for D2-like receptors (see Table 2 for results and for the
structures of tested compounds). None of them reached the affinity
of apomorphine, but were comparable with dopamine. On the
other hand, several B-enantiomers had better D2L over D1 selec-
tivity compared to the reference compounds, but none achieved the
level of D3/D2L selectivity of dopamine.
2. Results and discussion
2.1. Lipase-catalyzed kinetic resolution of 3-azabicyclo[3.2.0]
heptane derivatives
The stereodetermining step of the cascade in the synthesis of
heterocycle 9 is the first aza-Michael addition. Our several attempts
to use asymmetric organocatalytic approaches, such as iminium
catalysis with chiral secondary amines (such as diaryl prolinol
derivatives) or hydrogen bonding catalysts (quinidine, cinchonine,
(R)-TRIP and chiral thiourea derivatives) afforded
a racemic
product. The same result was obtained with the auxiliary-based
methods, where enantiomerically pure aminocrotonate deriva-
tives were used in the multicomponent reaction.
As chemical methods of the enantioselective synthesis failed,
we addressed this problem in the enzymatic reactions. It is well
known that lipases are efficient enzymes for the acylation of alco-
hols [17]. The lipase-catalyzed kinetic resolution of the racemic
secondary or even primary alcohols [18] is a widely used method
for the synthesis of enantiomeric alcohols. We used immobilized
lipase B of Candida antarctica (Novozym 435) as an acylating agent
for the kinetic resolution of enantiomers of 3-azabicyclo[3.2.0]
heptane 9 derivatives in EtOAc at room temperature (Scheme 2 and
Table 1).
(B)-Enantiomers had substantially higher affinities (4- to 40-
fold) toward D2-like receptors compared to (A)-enantiomers,
while the influence of the compounds absolute configuration on
the affinities for the D1 receptor was much less pronounced.
The addition of bromine to the phenyl moiety at position 2 of
the core structure slightly improved the affinity for D1 receptors
(compare the compounds 9e and 9a). The highest affinity
In addition to the previously known racemic compounds 9aec
[9], the new derivatives 9d, 9e, 9f and 9g were synthesized and
Scheme 1. A multicomponent synthesis of racemic tetrasubstituted 3-azabicyclo[3.2.0]heptane derivatives.

R. Reinart-Okugbeni et al. / European Journal of Medicinal Chemistry 55 (2012) 255e261
257
Scheme 2. Kinetic resolution of 3-azabicyclo[3.2.0]heptane 9 derivatives.
compound of the current study was compound (B)-9b,which
3. Conclusions
contained a pyrrolidinyl moiety at position 7 and a phenyl
substituent at position 2, of which the pyrrolidinyl group seemed
to improve the binding to D2L/D3 receptors, while the binding
to D1 receptors remains the same (compare (B)-9b and
(B)-9a). The substitution of the pyrrolidine ring by a piperidine
decreased both the affinity and selectivity (compounds (B)-9b and
(B)-9g).
Enantiomers of 3-azabicyclo[3.2.0]heptane 9 derivatives were
efficiently resolved by kinetic enzymatic resolution with immobi-
lized lipase B of Candida antarctica (Novozym 435). The pharmaco-
logical evaluation of the obtained compounds revealed that all
Comparing methyl and phenyl groups, the latter improved the
affinity both for D1 and D2L/D3 receptors, 10- and 50-fold, respec-
tively (compare the compounds (B)-9d and (B)-9b). Generally,
phenyl at position 2 increased the affinity for D1 receptors irre-
spective of the substituents at position 7 (see compounds 9a, 9b, 9e
and 9f). Significant selectivity was detected for only one compound,
(B)-9b, having 80- to 90-fold higher affinity for D2-like receptors
compared to the D1 receptor subtype.
Table 2
Binding affinities and binding selectivity of 3-azabicyclo[3.2.0]heptane derivatives
9aeg ((A)- and (B)-enantiomers) to selected subtypes of the human dopamine
receptors stably expressed in human embryonic kidney (HEK293) cells.
For all the other higher affinity enantiomers, the preference for
D2L and D3 receptors (between 10- and 30-fold) compared to D1
was low. The exception was compound (B)-9c with a methyl at
position 2 which bound very weakly to D1 receptors while
maintaining a considerable affinity for D2L and D3 receptors,
increasing thereby the selectivity for D2-like receptors to more
than 40. The lack of selectivity between D2 and D3 receptor
subtypes, as we encountered here, is a common feature for
dopaminergic compounds and finding ligands that are specific for
the D3 rather than the D2 receptor is one of the major challenges in
dopaminergic receptor pharmacology [24]. In the current study
however the goal was to identify the compounds preference for
selected dopamine receptor subtypes. The binding studies
confirmed our hypothesis that the newly synthesized 3-azabicyclo
[3.2.0]heptane derivatives preferentially bind to D2-like dopa-
mine receptors.
a
Compounds
Ki
(
m
M)
Selectivityb
D1
D2L
D3
D2L/D1 D3/D1 D3/D2L
(A)-9a
(B)-9a
(A)-9b
(B)-9b
(A)-9c
(B)-9c
(A)-9d
(B)-9d
(A)-9e
(B)-9e
(A)-9f
(B)-9f
42 ꢁ 7
22 ꢁ 8
40.8 ꢁ 0.7
22 ꢁ 3
>100
>100
>100
>100
12 ꢁ 2
13.2 ꢁ 0.4
3.4
13
3.2
17
0.9
1.3
0.6
0.9
1.7 ꢁ 0.3
7.4 ꢁ 0.3
1.3 ꢁ 0.2
11 ꢁ 2
5.5
3.6
80
>10 1.8
>85 2.2
0.25 ꢁ 0.02 0.28 ꢁ 0.04 90
38 ꢁ 2
3.9 ꢁ 0.4
66 ꢁ 7
16 ꢁ 3
5 ꢁ 1
21 ꢁ 9
>7
1.77 ꢁ 0.01 >40
36 ꢁ 8
11 ꢁ 1
>2
>4
1.8
>10
>15 1.5
7 ꢁ 1
7.93 ꢁ 0.02 1.4
2.14 ꢁ 0.01 3.9
0.9
3.6
>3
>15 1.4
8.3
34
0.7
0.9
1.2
8 ꢁ 1
2.0 ꢁ 0.1
42 ꢁ 7
12 ꢁ 1
7 ꢁ 1
Table 1
>100
36 ꢁ 10
8.8 ꢁ 0.3
5.3 ꢁ 0.3
1.5 ꢁ 0.1
>3
>10
6.4
25
Kinetic resolution of 3-azabicyclo[3.2.0]heptane 9 derivatives.
>100
(A)-9g
(B)-9g
Dopamine
44 ꢁ 2
49 ꢁ 0.1
12 ꢁ 3
1.3
1.4
140
4.8
0
Entry
Compound
R
NR2
ee of (A)-
ee of (B)-
2.0 ꢁ 0.3
enantiomer
enantiomer
2.1 ꢁ 0.2 0.015 ꢁ 0.004 5.7
800
65
1
2
3
4
5
6
7
9a
9b
9c
9d
9e
9f
Ph
Ph
Me
Me
p-BrPh
2-pyridyl
Ph
NEt2
Pyrrolidinyl
NEt2
Pyrrolidinyl
NEt2
NEt2
99
96
98
99
97
97
95
97
99
94
99
96
99
99
Apomorphine 0.66 ꢁ 0.02 0.05 ꢁ 0.01 0.010 ꢁ 0.004 13
a
Ki value in
m
M represents mean ꢁ SEM from at least two independent experi-
ments carried out in duplicates.
b
Selectivities for D2L over D1 (D2L/D1), for D3 over D1 (D3/D1) and for D3 over D2L
(D3/D2L) receptors are calculated as a ratio of Ki(D1)/Ki(D2L), Ki(D1)/Ki(D3) and
Ki(D2L)/Ki(D3), respectively. Selectivities marked as >are rough estimations due to
very low affinity for the particular subtype.
9g
Piperidinyl

258
R. Reinart-Okugbeni et al. / European Journal of Medicinal Chemistry 55 (2012) 255e261
Scheme 3. Synthesis of N-tosyl-3-azabicyclo[3.2.0]heptane derivative 11.
compounds studied showed a moderate preference for D2-like
receptors and, not surprisingly, the binding affinity depended on the
enantiomeric form of the tested 3-azabicyclo[3.2.0]heptane
derivative.
Reactions sensitive to oxygen or moisture were conducted
under argon atmosphere in flame-dried glassware. Anhydrous
dichloromethane was freshly distilled with CaH2 and anhydrous
tetrahydrofuran with LiAlH4. Commercial reagents were generally
used as received. Petroleum ether used had bp 40e60 ꢂC.
4. Experimental section
4.1. General procedure for the synthesis of racemic 9
Full assignment of 1H and 13C chemical shifts is based on the 1D
and 2D FT NMR spectra on a Bruker AvanceIII 400 instruments.
Deuterosolvent peaks (CHCl3 d ¼ 7.27, CDCl3 d ¼ 77.00) or TMS peak
was used as chemical shift references. Mass spectra were obtained
on a Shimadzu GCMS-QP2010 spectrometer in GC/MS mode (EI,
70 eV). High resolution mass spectra were recorded on LTQ Orbi-
trap (Thermo Electron). IR spectra were recorded on PerkineElmer
Spectrum BX FTIR spectrometer. X-ray diffraction data was
collected on a Bruker SMART X2S at 200 K.
To a solution of the corresponding aldehyde 5 (0.4 mmol) in
anhydrous CH2Cl2 (1.0 mL) with molecular sieves (4Ǻ) dialkyl
amine 7 (0.4 mmol) and N-benzylaminocrotonate 6 (0.2 mmol)
were added. The mixture was stirred at room temperature for
17e42 h. The mixture was concentrated in vacuum and the crude
bicyclic ester 8 was reduced with LiAlH4 (0.8 mmol) in anhydrous
THF (1.0 mL). After 3 h the reaction mixture was cooled to 0 ꢂC and
the reaction was quenched by the addition of water and an aqueous
solution of 4 M aq NaOH. The mixture was dried over K2CO3. The
crude product was purified by chromatography on silica gel
affording bicyclic alcohol 9.
4.1.1. (3-Benzyl-2-exo-methyl-7-exo-pyrrolidin-1-yl-3-azabicyclo
[3.2.0]hept-6-endo-yl)methanol 9d
Yield: 51%, off-white solid mp 102e104 ꢂC. IR (KBr): 3191, 3062,
3028, 2958, 2909, 2800, 1748, 1632, 1494, 1454, 1331, 1240, 1174,
1152, 1128, 1075, 1029, 899, 786, 752, 699 cmꢀ1. 1H NMR (400 MHz,
CDCl3)
d
7.37e7.31 (m, 4H), 7.30e7.24 (m, 1H), 3.85 (d, J ¼ 13.1 Hz,
1H), 3.78 (dd, J ¼ 11.6, 2.0 Hz, 1H), 3.63 (d, J ¼ 13.1 Hz, 1H), 3.62 (dd,
J ¼ 11.5, 5.1 Hz,1H), 3.07 (q, J ¼ 6.6 Hz,1H), 3.00 (dd, J ¼ 10.6 Hz,1H),
2.97 (ddd, J ¼ 9.8, 7.9, 6.2 Hz, 1H), 2.87 (dd, J ¼ 5.7, 3.9 Hz, 1H), 2.63
(dd, J ¼ 10.5, 6.2 Hz, 1H), 2.45 (m, 5H), 2.31 (dd, J ¼ 8.1, 3.7 Hz, 1H),
1.83e1.73 (m, 4H), 0.84 (d, J ¼ 6.7 Hz, 3H). 13C NMR (101 MHz,
CDCl3)
d 138.23, 128.78, 128.48, 127.19, 63.77, 61.85, 60.12, 53.77,
51.20, 50.50, 47.57, 40.50, 34.31, 23.32, 11.70. HRMS (m/z): [M þ Hþ]
calcd for (C19H29N2O)þ, 301.22744; found, 301.22748.
4.1.2. (3-Benzyl-7-exo-diethylamino-2-exo-p-bromophenyl-3-
azabicyclo[3.2.0]hept-6-endo-yl) methanol 9e
Yield: 67%, pale yellow solid mp 43e45 ꢂC. IR (KBr): 3131, 2969,
2906, 2859, 2801, 1945, 1905, 1809, 1639, 1586, 1487, 1468, 1450,
1368, 1343, 1325, 1294, 1195, 1130, 1010, 981, 814, 735, 701,
523 cmꢀ1. 1H NMR (400 MHz, CDCl3)
d 7.46e7.40 (m, 2H), 7.33 (m,
2H), 7.27 (m, 1H), 7.25e7.20 (m, 2H), 6.81 (m, 2H), 3.87 (s, 1H), 3.81
(dd, J ¼ 11.4, 2.9 Hz, 1H), 3.70 (dd, J ¼ 11.4, 6.1 Hz, 1H), 3.46 (d,
J ¼ 13.4 Hz, 1H), 3.28 (d, J ¼ 13.5 Hz, 1H), 3.25 (dd, J ¼ 6.3, 4.8 Hz,
1H), 3.08 (dd, J ¼ 16.3, 8.3 Hz,1H), 2.97 (dd, J ¼ 10.8, 1.4 Hz,1H), 2.74
(dd, J ¼ 10.9, 6.8 Hz, 1H), 2.70 (dd, J ¼ 8.4, 4.7 Hz, 1H), 2.55 (q,
J ¼ 7.2 Hz, 5H), 0.98 (t, J ¼ 7.2 Hz, 6H). 13C NMR (101 MHz, CDCl3)
d
138.34, 131.19, 130.26, 128.59, 128.43, 127.16, 121.39, 70.04, 62.43,
61.79, 54.74, 51.17, 47.63, 41.95, 40.09, 34.36, 10.59. HRMS (m/z):
[M þ Hþ] calcd for (C24H31BrN2O)þ, 443.16925; found, 443.16862.
4.1.3. (3-Benzyl-7-exo-diethylamino-2-exo-pyridyl-3-azabicyclo
[3.2.0]hept-6-endo-yl)methanol 9f
Fig. 2. Molecular moiety in the crystal structure of N-tosyl-3-azabicyclo[3.2.0]heptane
derivative 11. For clarity one formula unit is shown (Z0 ¼ 2). Displacement ellipsoids are
drawn at the 50% probability level.
Yield: 53%, off-white solid mp 92e93 ꢂC. IR (KBr): 3136, 2974,
2939, 2913, 2829, 1589, 1568, 1490, 1474, 1448, 1432, 1368, 1338,

R. Reinart-Okugbeni et al. / European Journal of Medicinal Chemistry 55 (2012) 255e261
259
1288,1233,1190,1174,1151,1127,1079,1019, 992, 852, 779, 764, 747,
707, 650, 604 cmꢀ1. 1H NMR (400 MHz, CDCl3)
4.2.4. (3-Benzyl-2-exo-methyl-7-exo-pyrrolidin-1-yl-3-azabicyclo
[3.2.0]hept-6-endo-yl)methanol 9d
The enantiomeric excess was determined by HPLC, Lux
Amylose-2, Hex:iPrOH 97:3 with 0.1% Et2NH, 1.5 mL/min, UV
d
8.57 (ddd, J ¼ 0.7,
1.7, 4.8, 1H), 7.59 (td, J ¼ 1.8, 7.7, 1H), 7.34e7.14 (m, 6H), 6.90 (d,
J ¼ 7.8, 1H), 4.01 (s, 1H), 3.80 (dd, J ¼ 3.0, 11.4, 1H), 3.66 (dd, J ¼ 6.0,
11.5, 1H), 3.51 (d, J ¼ 13.2, 1H), 3.41 (d, J ¼ 13.2, 1H), 3.29 (dd, J ¼ 4.7,
6.2, 1H), 3.14 (m, 1H), 3.13 (d, J ¼ 8.6, 1H), 2.99 (d, J ¼ 8.5, 1H), 2.76
(dd, J ¼ 4.6, 7.7, 1H), 2.57 (q, J ¼ 7.2, 4H), 2.53e2.49 (m, 1H), 0.99 (t,
254 nm, nonacylated (A)-enantiomer tR ¼ 16.6, ee 99%, [
(c 0.31, CH2Cl2) and acylated (B)-enantiomer tR ¼ 13.9, ee 99%,
] ¼ ꢀ29 (c 0.48, CH2Cl2).
a] ¼ þ30
[a
J ¼ 7.2, 6H). 13C NMR (101 MHz, CDCl3)
d 160.57, 149.16, 138.63,
135.97, 128.78, 128.30, 127.11, 122.87, 122.11, 72.44, 62.45, 61.91,
55.10, 52.03, 47.04, 42.00, 39.88, 34.78, 10.55. HRMS (m/z):
[M þ Hþ] calcd for (C23H31N3O)þ, 366.25399; found, 366.25322.
4.2.5. (3-Benzyl-7-exo-diethylamino-2-exo-p-bromophenyl-3-
azabicyclo[3.2.0]hept-6-endo-yl)methanol 9e
The enantiomeric excess was determined by HPLC, Chiralcel AS,
Hex:iPrOH 98:2 with 0.1% Et2NH, 1.0 mL/min, UV 254 nm, non-
acylated (A)-enantiomer tR ¼ 12.2, ee 96% and acylated (B)-enan-
4.1.4. (3-Benzyl-2-exo-phenyl-7-exo-piperidin-1-yl-3-azabicyclo
[3.2.0]hept-6-endo-yl)methanol 9g
tiomer tR ¼ 17.4, ee 99%, [ ] ¼ þ4.9 (c 2.93, CH2Cl2).
a
Yield: 30%, off-white solid mp 148e151 ꢂC. IR (KBr): 3183, 3065,
3029, 2924, 2782, 1601, 1490, 1452, 1368, 1339, 1261, 1236, 1122,
4.2.6. (3-Benzyl-7-exo-diethylamino-2-exo-pyridyl-3-azabicyclo
[3.2.0]hept-6-endo-yl)methanol 9f
1027, 861, 777, 750, 711, 704 cmꢀ1 1H NMR (400 MHz, CDCl3)
.
d
7.36e7.27 (m, 6H), 7.23 (m, 2H), 6.92 (m, 2H), 3.91 (s,1H), 3.83 (dd,
The enantiomeric excess was determined by HPLC, Chiralcel AS,
Hex:iPrOH 98:2 with 0.1% Et2NH, 1.0 mL/min, UV 254 nm, non-
J ¼ 11.4, 2.6 Hz, 1H), 3.69 (dd, J ¼ 11.4, 5.9 Hz, 1H), 3.42 (d,
J ¼ 13.6 Hz, 1H), 3.31 (d, J ¼ 13.6 Hz, 1H), 3.14 (m, 1H), 3.00 (dd,
J ¼ 10.7, 1.1 Hz, 1H), 2.82e2.75 (m, 3H), 2.56 (m, 1H), 2.41e2.20 (m,
4H), 1.61e1.50 (m, 4H), 1.45 (m, 2H). 13C NMR (101 MHz, CDCl3)
acylated (A)-enantiomer tR ¼ 23.1, ee 97%, [ ] ¼ ꢀ30 (c 1.97, CH2Cl2)
a
and acylated (B)-enantiomer tR ¼ 29.4, ee 99%, [ ] ¼ þ29.5 (c 1.53,
a
CH2Cl2).
d
139.07, 138.61, 128.71, 128.65, 128.35,128.03, 127.44, 127.03, 70.50,
66.11, 62.64, 54.60, 51.17, 51.09, 46.91, 39.83, 34.74, 25.46, 24.39.
HRMS (m/z): [M þ Hþ] calcd for (C25H33N2O)þ, 377.25874; found,
377.25854.
4.2.7. (3-Benzyl-2-exo-phenyl-7-exo-piperidin-1-yl-3-azabicyclo
[3.2.0]hept-6-endo-yl) methanol 9g
The enantiomeric excess was determined by HPLC, Chiralcel AS,
Hex:iPrOH 97:3 with 0.1% Et2NH, 1.0 mL/min, UV 254 nm, non-
4.2. General procedure of enzymatic kinetic resolution of 9
acylated (A)-enantiomer tR ¼ 10.2, ee 95%, [ ] ¼ þ19 (c 0.37, CH2Cl2)
a
and acylated (B)-enantiomer tR ¼ 14.5, ee 99%, [ ] ¼ ꢀ20 (c 0.37,
a
Lipase B of C. antarctica (Novozym 435) (50 mg) was added to
a solution of the racemic compound 9 (50 mg) in EtOAc (1.0 mL).
The resulting mixture was stirred occasionally at room temperature
and the reaction was monitored by TLC (about a conversion 50%).
The typical time of the reaction varied from 4.5 to 7 h, then the
lipase was filtered off and the filtrate concentrated under reduced
pressure. This mixture was purified by chromatography on silica gel
affording alcohol (A)-9 and ester 12. As the enantiomeric excess of
ester 12 could not be determined by chiral HPLC, it was hydrolyzed
to (B)-9, with 4 M NaOH in MeOH by stirring 3 h at room temper-
ature. The enantiomeric excess of (A)-9 and (B)-9 was determined
by HPLC (Chiralcel AS or Lux Amylose-2).
CH2Cl2).
4.2.8. 7-(Exo-(diethylamino)-2-exo-phenyl-3-tosyl-3-azabicyclo
[3.2.0]heptan-6-endo-yl)methanol 11
To a solution of trans-cinnamaldehyde (0.4 mmol) in anhydrous
CH2Cl2 (1.0 mL) diethyl amine (1.0 mmol) and N-tosylaminocroto-
nate (0.2 mmol) were added in the presence of molecular sieves
(4Ǻ). The mixture was stirred at room temperature for 20 h. The
mixture was concentrated in vacuum and the crude bicyclic ester
was reduced with LiAlH4 (0.8 mmol) in anhydrous THF (1.0 mL).
After 3 h the reaction mixture was cooled to 0 ꢂC and was quenched
by the addition of water and an aqueous solution of 4 M aq NaOH.
The mixture was dried over K2CO3. The crude product was purified
by chromatography on silica gel affording bicyclic alcohol 11 (smp
4.2.1. (3-Benzyl-7-exo-diethylamino-2-exo-phenyl-3-azabicyclo
[3.2.0]hept-6-endo-yl)methanol 9a
The enantiomeric excess was determined by HPLC, Chiralcel AS,
Hex:iPrOH 98:2 with 0.1% Et2NH, 1.0 mL/min, UV 254 nm, non-
115e121 ꢂC). 1H NMR (400 MHz, CDCl3)
d
0.95 (t, J ¼ 7.1 Hz, 6H),
2.32 (s, 4H), 2.58e2.48 (m, 5H), 2.68e2.63 (m, 1H), 2.73e2.68 (m,
1H), 3.16e3.06 (m, 1H), 3.36 (dd, J ¼ 11.3, 8.3 Hz, 1H), 3.75e3.62 (m,
2H), 3.87 (dd, J ¼ 11.3, 1.9 Hz, 1H), 4.83 (s, 1H), 7.08e6.98 (m, 4H),
7.23e7.14 (m, 3H), 7.36e7.32 (m, 2H). 13C NMR (101 MHz, CDCl3)
acylated (A)-enantiomer tR ¼ 11.6, ee 99%, [ ] ¼ þ10 (c 0.82, CH2Cl2)
a
and acylated (B)-enantiomer tR ¼ 16.3, ee 95%, [ ] ¼ ꢀ10 (c 0.47,
a
CH2Cl2).
d 10.24, 21.41, 34.53, 40.03, 41.71, 47.17, 49.27, 61.58, 62.54, 68.97,
126.70, 127.02, 127.53, 128.50, 129.11, 136.13, 140.43, 142.77. HRMS
(ESIþ): calculated for (C24H32N2O3S)þ 248.2134 [Mþ], found
248.2134.
4.2.2. (3-Benzyl-2-exo-phenyl-7-exo-pyrrolidin-1-yl-3-azabicyclo
[3.2.0]hept-6-endo-yl) methanol 9b
The enantiomeric excess was determined by HPLC, Chiralcel AS,
Hex:iPrOH 97:3 with 0.1% Et2NH, 1.0 mL/min, UV 254 nm, non-
4.3. Production of cell lines expressing human D1, D2L and D3
dopamine receptors
acylated (A)-enantiomer tR ¼ 11.5, ee 96%, [ ] ¼ þ10 (c 1.22, CH2Cl2)
a
and acylated (B)-enantiomer tR ¼ 15.5, ee 99%, [ ] ¼ ꢀ11 (c 1.46,
a
CH2Cl2).
We created a human embryonic kidney cells (HEK293) based
stable lines expressing individual subtypes of human dopamine
receptors (D1, D2L and D3). Shortly, HEK293 cells (American Type
Culture Collection, Rockville, MD) were maintained in Dulbecco’s
Modified Eagle’s Medium (DMEM) (PAA Laboratories) supple-
mented with 10% fetal bovine serum (GibcoÒ), 100 U/mL penicillin
4.2.3. (3-Benzyl-7-exo-diethylamino-2-exo-methyl-3-azabicyclo
[3.2.0]hept-6-endo-yl)methanol 9c
The enantiomeric excess was determined by HPLC, Lux
Amylose-2, Hex:iPrOH 98:2 with 0.1% Et2NH, 1.0 mL/min, UV
254 nm, nonacylated (A)-enantiomer tR ¼ 22.6, ee 99%, [
0.16, CH2Cl2) and acylated (B)-enantiomer tR ¼ 18.2, ee 94%,
] ¼ ꢀ28 (c 0.61, CH2Cl2).
a
] ¼ þ31 (c
and 100 mg/mL streptomycin (PAA Laboratories). Cells were grown
at 37 ꢂC in a humidified incubator with 5% CO2. The pcDNA3.1þ
[
a
expression vectors (Invitrogen) containing the desired gene of

260
R. Reinart-Okugbeni et al. / European Journal of Medicinal Chemistry 55 (2012) 255e261
human wild type dopamine receptor (DRD1, DRD2L and DRD3) were
purchased from the Missouri S&T cDNA Resource Center. For
transfection, cells were seeded on 6-well plates, cultured 24 h to
reach w90% confluence and transfected with 4 mg of DNA per well
using LipofectamineÔ 2000 (Invitrogen) according to manufac-
turer’s instructions. To obtain stable lines cells were maintained
were initiated by addition of membrane suspension (5 ꢃ 104 to
5 ꢃ 105 cells/assay for different receptor subtypes).
All the reactions were stopped by filtration through thick GF/B
glass fiber filter mats (Whatham) using a FilterMate Harvester
(Model D961962, Perkin Elmer). After five washes with ice-cold
phosphate buffer (20 mM K-phosphate, 100 mM NaCl, pH 7.4),
filter mats were dried in a microwave oven and impregnated with
a MeltiLexÔ B/HS scintillant (Wallac) using a MeltiLexÒ Heatsealer
(Wallac). Filter-bound radioactivity was counted using a Wallac
MicroBeta TriLux 1450 LSC Luminescence Counter (Perkin Elmer).
All pharmacological data were analyzed by means of non-linear
least squares regression analysis using the commercial program
GRAPHPAD PRISMÔ 4.03 (GraphPad Software Inc.). Data were fit to
one-site binding curve and inhibition constant values (Ki) calcu-
lated according to ChengePrusoff equation [29] (in Table 2 repre-
sented as means ꢁ SEM from at least two independent experiments
carried out in duplicates).
and passed for 2 weeks in the presence of 800 mg/mL geneticin
(G418). Four to six G418 resistant colonies per each receptor
subtype were selected and after five further passages the clonal
cultures were tested for receptor expression by radioligand
binding. Henceforth the cells were maintained in the presence of
400
mg/mL of G418. The clonal cell lines with similar receptor
densities were further tested for their binding of known dopami-
nergic ligands and then used in experiments with the test
compounds.
4.4. Radioligand binding and competition binding of 3-azabicyclo
[3.2.0]heptane derivatives
Acknowledgments
The binding characteristics of the new cell lines were assessed
by radioligand binding and competition binding of known dopa-
minergic ligands (dopamine and apomorphine). The radioligands
[3H]SCH23390 and [3H]raclopride have high affinity for D1-like and
D2-like dopamine receptors, respectively, which was also evident
from our saturation binding experiments (data shown below). All
ligand binding experiments were done on membrane suspensions,
prepared as follows. The cells were centrifuged at 800 ꢃ g at room
temperature and the pellet stored at ꢀ80 ꢂC. The frozen pellets
were melted on ice and washed by homogenization with a tissue
homogenizer (Coleparmer Labgen 125) for 30 s in ice-cold PBS and
centrifugation at 800 ꢃ g for 5 min at 4 ꢂC. The pellet was re-
homogenized in 50 mM TriseHCl buffer, (pH ¼ 7.4) and centri-
fuged at 30,000 ꢃ g for 20 min followed by a second resuspension
and homogenization step. The latter homogenization and centri-
fugation steps were repeated once and the final pellet was
homogenized in incubation buffer (IB: 50 mM TriseHCl, 120 mM
NaCl, 5 mM KCl, 5 mM MgCl2, 1 mM EDTA, pH 7.4) typically at
a concentration of 1 ꢃ 107 cells/mL (corresponding to w0.5 mg
protein/mL quantified using Bio-Rad protein assay Kit (Bio-Rad
Laboratories) with BSA as a standard). The membrane preparations
were stored at ꢀ80 ꢂC until further testing.
The authors thank the Estonian Science Foundation (Grants Nos.
8289 and 8314), the Ministry of Education and Research (Grant No.
0140060s12 and 0180032s12) and EU European Regional Devel-
opment Fund (3.2.0101.08-0017, 3.2.0101.11-0029, 30020) for
financial support.
Appendix A. Supplementary material
Supplementary material associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.ejmech.
2012.07.025.
References
[1] J.-M. Beaulieu, R.R. Gainetdinov, The physiology, signaling, and pharmacology
of dopamine receptors, Pharm. Rev. 63 (2011) 182e217.
[2] C.M. Marson, New and unusual scaffolds in medicinal chemistry, Chem. Soc.
Rev. 40 (2011) 5514e5533.
[3] F. Micheli, D. Hamprecht, G. Bonanomi, R. Di Fabio, D. Donati, G. Gentile, et al.,
[3-Azabicyclo[3.1.0]hex-1-yl]phenyl-benzenesulfonamides as selective dopa-
mine D3 antagonists, Bioorg. Med. Chem. Lett. 20 (2010) 5491e5494.
[4] F. Micheli, L. Arista, G. Bonanomi, F.E. Blaney, S. Braggio, A.M. Capelli, et al.,
1,2,4-Triazolyl azabicyclo[3.1.0]hexanes: a new series of potent and selective
dopamine D(3) receptor antagonists, J. Med. Chem. 53 (2010) 374e391.
[5] R. Depoortere, D. Boulay, G. Perrault, O. Bergis, M. Decobert, D. Françon, et al.,
SSR181507, a dopamine D2 receptor antagonist and 5-HT1A receptor agonist.
II: behavioral profile predictive of an atypical antipsychotic activity, Neuro-
psychopharmacology 28 (2003) 1889e1902.
Receptor expression and affinity of radioligands was determined
as described earlier [25,26] with some modifications. Reactions
were carried out in U-bottom 96-well plates in a final volume of
250
150
m
m
L in IB and incubated at 25 ꢂC for 60e90 min. In all assays
L of membrane suspension was added to wells containing
50 mL of radioligand and 50 mL of competing unlabeled ligand (or
[6] N.M. Paul, M. Taylor, R. Kumar, J.R. Deschamps, R.R. Luedtke, A.H. Newman,
buffer) at their five times final concentration. Receptor-positive
clones were identified by one-point assay at a saturating concen-
tration of [3H]SCH23390 or [3H]raclopride (Perkin Elmer) with or
Structureeactivity relationships for
receptor ligands based on N-substituted 3-aryl-8-azabicyclo[3.2.1]octan-3-
ol, J. Med. Chem. 51 (2008) 6095e6109.
a novel series of dopamine D2-like
[7] R.H. Mach, P.S. Hammond, Y.S. Huang, B.O. Yang, Y.P. Xu, J.T. Cheney, et al.,
Structureeactivity relationship studies of N-(9-benzyl)-9-azabicyclo[3.3.1]
nonan-3-beta-yl benzamide analogues for dopamine D-2 and D-3 receptors,
Med. Chem. Res. 9 (1999) 355e373.
[8] K. Drescher, A. Haupt, L. Unger, S.C. Turner, W. Braje, R. Grandel, Azabicy-
cloheptyl Compounds Suitable for Treating Disorders that Respond to
Modulation of the Dopamine D3 Receptor, U.S. Patent WO 2006040176, 2006.
[9] K. Kriis, K. Ausmees, T. Pehk, M. Lopp, T. Kanger, A novel diastereoselective
multicomponent cascade reaction, Org. Lett. 12 (2010) 2230e2233.
[10] T. Bach, C. Pelkmann, K. Harms, Facial diastereoselectivity in the [2 þ 2]-
photocycloaddition of chiral vinylglycine-derived N, N-diallyl amines, Tetra-
hedron Lett. 40 (1999) 2103e2104.
without an unlabeled dopaminergic antagonist, 1
m
M (þ)-butacla-
mol (SigmaeAldrich). Receptor bound radioactivity (“specific
binding”) was calculated as a difference between radioactivity
bound in the absence (“total binding”) and in the presence (“non-
specific binding”) of an unlabeled antagonist. In saturation binding
experiments the concentration of radioligand was varied from 0.2
to 10 nM and “non-specific binding” was determined in the pres-
ence of 1
m
M (þ)-butaclamol. Dissociation constants obtained,
0.5 nM for [3H]SCH23390 binding to D1 receptor and 0.7 nM for [3H]
raclopride binding to both D2L and D3 receptors, were in good
agreement with the data reported earlier for these radioligands and
receptors [27,28]. For 8-point competition assay the concentration
of radioligand was fixed to near Kd value (w1 nM) and concentra-
tion of the test compounds was varied from 0.3 mM to 0.1 nM, the
incubation buffer was supplemented with 1 mM DTT. Reactions
[11] T. Bach, C. Kruger, K. Harms, The stereoselective synthesis of 2-substituted 3-
azabicyclo[3.2.0]heptanes by intramolecular [2
reactions, Synthesis (2000) 305e320.
þ
2]-photocycloaddition
[12] M.R. Luzung, P. Mauleo, F.D. Toste, Gold(I)-catalyzed [2 þ 2]-cycloaddition of
allenenes, J. Am. Chem. Soc. 129 (2007) 12402e12403.
[13] Y.T. Lee, Y.K. Kang, Y.K. Chung, Au(I)-catalyzed cycloisomerization reaction of
amide- or ester-tethered 1,6-enynes to bicyclo[3.2.0]hept-6-en-2-ones, J. Org.
Chem. 74 (2009) 7922e7934.

R. Reinart-Okugbeni et al. / European Journal of Medicinal Chemistry 55 (2012) 255e261
261
[14] T. Matsuda, S. Kadowaki, M. Murakami, Synthesis of 3-acyl-4-
alkenylpyrrolidines by platinum-catalyzed hydrative cyclization of alle-
nynes, Helv. Chim. Acta 89 (2006) 1672e1680.
[15] N. Bischof, E. Buschmann, M. Henningsen, G.A. Krei, R. Munschauer, T. Pfeiffer,
et al., Method for Producing (þ)-Exo-6-phenyl-3-azabicyclo-[3.2.0]heptanes,
U.S. Patent WO2000023423 A1, 2000.
[23] J.E. Leysen, W. Gommeren, J. Mertens, W.H. Luyten, P.J. Pauwels,
M. Ewert, et al., Comparison of in vitro binding properties of a series of
dopamine antagonists and agonists for cloned human dopamine D2S
and D2L receptors and for D2 receptors in rat striatal and mesolimbic
tissues, using [125I] 20-iodospiperone, Psychopharmacology 110 (1993)
27e36.
[16] J.B. Xia, W.B. Liu, T.M. Wang, S.L. You, Enantioselective synthesis of 3-
azabicyclo[4.1.0]heptenes and 3-azabicyclo[3.2.0]heptenes by Ir-catalyzed
asymmetric allylic amination of N-tosyl propynylamine and Pt-catalyzed
cycloisomerization, Chem. Eur. J. 16 (2010) 6442e6446.
[17] K. Faber, Biotransformations in Organic Chemistry, Springer, Heidelberg, 2004.
[18] C. Oger, Z. Marton, Y. Brinkmann, V. Bultel-Poncé, T. Durand, M. Graber, et al.,
Lipase-catalyzed regioselective monoacetylation of unsymmetrical 1,5-
primary diols, J. Org. Chem. 75 (2010) 1892e1897.
[19] R. Reinart, Z. Gyulai, S. Berényi, S. Antus, A. Vonk, A. Rinken, et al., New 2-
thioether-substituted apomorphines as potent and selective dopamine D2
receptor agonists, Eur. J. Med. Chem. 46 (2011) 2992e2999.
[20] A. Sipos, C. Csutorás, S. Berényi, A. Uustare, A. Rinken, Synthesis and neuro-
pharmacological characterization of 2-O-substituted apomorphines, Bioorg.
Med. Chem. 16 (2008) 4563e4568.
[21] A. Rinken, J. Harro, L. Engström, L. Oreland, Role of fluidity of membranes on
the guanyl nucleotide-dependent binding of cholecystokinin-8S to rat brain
cortical membranes, Biochem. Pharmacol. 55 (1998) 423e431.
[22] A. Rinken, Subtype-specific changes in ligand binding properties after solu-
bilization of muscarinic receptors from baculovirus-infected Sf9 insect cell
membranes, J. Pharm. Exp. Ther. 272 (1995) 8e14.
[24] J. Carlsson, R.G. Coleman, V. Setola, J.J. Irwin, H. Fan, A. Schlessinger, et al.,
Ligand discovery from a dopamine D(3) receptor homology model and crystal
structure, Nat. Chem. Biol. (2011) 1e10.
[25] A. Uustare, R. Reinart, A. Rinken, Modulation of dopamine D1 receptor
signaling by adenosine A1 receptors in Sf9 cells requires expression of Gi
proteins, Neurosci. Lett. 406 (2006) 169e173.
[26] J. Harro, A. Terasmaa, M. Eller, A. Rinken, Effect of denervation of the locus
coeruleus projections by DSP-4 treatment on [3H]-raclopride binding to
dopamine D2 receptors and D2 receptoreG protein interaction in the rat
striatum, Brain Res. 976 (2003) 209e216.
[27] A. Rinken, S. Ferré, A. Terasmaa, C. Owman, K. Fuxe, Serotonergic agonists
behave as partial agonists at the dopamine D2 receptor, NeuroReport 10
(1999) 493e495.
[28] Y. Cao, W.C. Sun, L. Jin, K.Q. Xie, X.Z. Zhu, Activation of adenosine A1 receptor
modulates dopamine D1 receptor activity in stably cotransfected human
embryonic kidney 293 cells, Eur. J. Pharmacol. 548 (2006) 29e35.
[29] Y.-C. Cheng, W.H. Prusoff, C. Yung-Chi, Relationship between the inhibition
constant (K1) and the concentration of inhibitor which causes 50 per cent
inhibition (I50) of an enzymatic reaction, Biochem. Pharmacol. 22 (1973)
3099e3108.