Syntheses and structure-activity relationship (SAR) studies of 2,5-diazabicyclo[2.2.1]heptanes as novel α7 neuronal nicotinic receptor (NNR) ligands

Article abstract of DOI:10.1016/j.bmcl.2010.04.105

Biaryl substituted 2,5-diazabicyclo[2.2.1]heptanes have been synthesized and tested for their affinity toward α7 neuronal nicotinic receptors (NNRs). SAR studies established that 5-N-methyl substituent, heteroaryl linker and the nature of terminal aryl gr

Full text of DOI:10.1016/j.bmcl.2010.04.105

Bioorganic & Medicinal Chemistry Letters 20 (2010) 3636–3639
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Syntheses and structure–activity relationship (SAR) studies
of 2,5-diazabicyclo[2.2.1]heptanes as novel
receptor (NNR) ligands
a7 neuronal nicotinic
a
a
a
a
Tao Li a, , William H. Bunnelle , Keith B. Ryther , David J. Anderson , John Malysz ,
Rosalind Helfrich a, Jens H. Grønlien a, Monika Håkerud a, Dan Peters b,
Michael R. Schrimpf a, Murali Gopalakrishnan a, Jianguo Ji a,
*
a Neuroscience Research, GPRD, Abbott, Dept. R4DE, 100 Abbott Park Road, Abbott Park, IL 60064-6117, USA
b NeuroSearch A/S, Pederstrupvej 93, 2750 Ballerup, Denmark
a r t i c l e i n f o
a b s t r a c t
Article history:
Biaryl substituted 2,5-diazabicyclo[2.2.1]heptanes have been synthesized and tested for their affinity
Received 25 February 2010
Revised 22 April 2010
Accepted 23 April 2010
Available online 28 April 2010
toward a7 neuronal nicotinic receptors (NNRs). SAR studies established that 5-N-methyl substituent, het-
eroaryl linker and the nature of terminal aryl group are critical for the ligand to achieve potent
a7 NNR
agonist activity.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
2,5-Diazabicyclo[2.2.1]heptanes
Neuronal nicotinic receptors
NNRs
a7
Cognitive deficits
a
7 Neuronal nicotinic receptors (
a
7 NNRs) belong to the family
tron (7).11 Although several selective
a7 NNR agents have
of ACh-gated cation channels, which includes multiple subtypes
advanced to human clinical trials, to date none has been brought
resulting from pentameric combinations of 12 known subunits
(a2–a
10 and b2–b4).1 These homo-oligomeric channels are con-
sidered to play an important role in modulating biochemical and
neurophysiological events underlying cognitive and attentive pro-
cesses, sensory gating, and neuroprotection.2 The convergence of
evidence from pharmacological, genetic and early clinical studies
N
OMe
N
N
suggests that selective
a7 NNR agonists have potential for the
N
(2)
treatment of a variety of cognitive deficits associated with Alzhei-
(1)
MeO
mer’s disease (AD) and schizophrenia.3 Accordingly, a range of
DMXB-A (GTS-21)
nicotine
structurally diverse
anabaseine and quinuclidine scaffolds, have been identified to-
a7 NNR ligands, largely emerging from the
O
Cl
H
H
O
N
N
wards further preclinical validation, and ultimately, clinical studies
NH
O
in humans.4 As shown in Scheme 1, examples of
a7 NNR agonists
O
O
that have been extensively used for preclinical in vivo studies in-
clude nicotine (1),5 the anabaseine derivative DMXB-A (GTS-21)
(2),6 spiro-oxazolidinone AR-R17779 (3),7 quinuclidine amides
PNU-282987 (4)8 and TC-5619 (6),9 SSR-180711 (5),10 and tropise-
N
N
N
(4)
(5)
(3)
N
AR-R17779
PNU-282987
TC-5619
NH
O
O
O
Br
N
O
N
N
(6)
(7)
* Corresponding author. Tel.: +1 847 937 3299; fax: +1 847 937 9195.
E-mail address: Tao.Li@abbott.com (T. Li).
tropisetron
SSR-180711
Present address: Hutchison Medipharma Limited, 720 Cai Lun Road, Shanghai,
China.
Scheme 1. Structures of a7 NNR agents have been extensively profiled in literature.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.04.105

T. Li et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3636–3639
3637
to market.5–11 The search for novel and selective
continues to receive considerable attention. In this communica-
tion, we describe the synthesis and structure–activity relationship
(SAR) studies of biaryl substituted 2,5-diazabicyclo[2.2.1]heptanes
a7 NNR agents
produce an agonist response at concentrations 6100
expressing the recombinant human
other heteromeric NNR subtypes (
that good selectivity for 7 over other NNR subtypes can be
achieved with this scaffold.
l
M in cells
a
4b2 NNR, nor did it activate
3b4 and 4b4). This suggests
a
a
a
as novel a7 NNR ligands.
Initially discovered by Barlocco and co-workers12,13 2-(6-chlo-
Based on the preliminary SAR results shown in Table 1, aryl sub-
stitution on pyridazine and 5-N-methylation on 2,5-diazabicy-
ropyridazin-3-yl)-2,5-diazabicyclo[2.2.1]heptane (9a) was found
to be a potent
a
4b2 NNR ligand (Ki = 6 nM in the [3H]cytisine bind-
clo[2.2.1]heptane led to an increase in
decrease in 4b2 NNR activity. Between enantiomers, (1S,4S) com-
pounds are preferable because they are not only more selective
a7 NNR affinity, but
ing assay). Barlocco also reported that the 5-N-methyl analogue
(9b) exhibits diminished binding affinity (Ki = 65 nM), while a sec-
ond methylation to the 5-N,N-dimethylated quaternary ammo-
a
with respect to 7 activity than
a
a4b2, but also 20–30-fold more
nium salt (9c) restores
a
4b2 NNR binding affinity (Ki = 4.3 nM).11
potent in the a7 binding assay. Based on these observations,
Consistent with our studies on substituent control of NNR subtype
selectivity,14 we found that replacing the chloro substituent in 9a
with a phenyl group, as in (1S,4S)-10a (Scheme 2) caused a dra-
5-N-methylated (1S,4S) compounds were chosen for further
exploration.
To further explore the potential of 2,5-diazabicyclo[2.2.1]hep-
matic attenuation of
affinity at the 7 NNR subtype (Table 1). Likewise, the 5-N-methyl
analog (1S,4S)-10b had weak 4b2 NNR affinity. In this case the
N-methyl group does not have a detrimental effect on 4b2 NNR
binding, but appears to enhance the 7 NNR affinity (Table 1, entry
7). Even more dramatically, the quaternary ammonium salt
(1S,4S)-10c (A-585539). exhibited 7 NNR binding affinity in the
a
4b2 NNR binding affinity, but enhanced
tane a7 NNR ligands, our efforts focused on the SAR of biaryl
a
substituted (1S,4S)-5-methyl-2,5-diazabicyclo[2.2.1]heptanes. As
shown in Scheme 3, the condensation of 5-N-Boc-2,5-diazabicy-
clo[2.2.1]heptane (1S,4S)-8 with a heteroaryl halide followed by
reductive animation with formalin in formic acid provided an array
of halo-heteroaryl intermediates 11, 13, 15, and 17. Aryl halides
that are activated toward nucleophilic aromatic substitution, such
as 3,6-dichloropyridazine, 2-chloropyrazine and 2,5-dibromopyr-
azine, reacted efficiently with (1S,4S)-8 in the presence of Hünig’s
base in DMSO at 100 °C. Regioselective amination of 5-bromo-2-
chloropyridine was accomplished by using the palladium mediated
procedures initially developed by Buchwald and Hartwig, then by
Ji.16 Under standard Suzuki coupling reaction conditions,17 halide
intermediates (11, 13, 15, and 17) reacted smoothly with a variety
of boronic acids in the presence of a palladium catalyst, giving biar-
yl substituted (1S,4S)-5-methyl-2,5-diaza-bicyclo[2.2.1]heptanes
12, 14, 16, and 18 in moderate to excellent chemical yield.
a
a
a
a
picomolar range (Table 1, entry 8). Although (1S,4S)-10c has lim-
ited therapeutic potential due to poor CNS access, its favorable
binding properties have made it our preferred radioligand for
in vitro binding at
ing is enhanced in the (1R,4R)-diamine series (Ref. 12 and Table
1), the opposite trend pertains to 7 NNR affinity, where (1S,4S)-
a a4b2 bind-
7 NNRs.15 Interestingly, while the
a
10b and (1S,4S)-10c are significantly more potent than their
(1R,4R) counterparts (Table 1, entries 7–10). Consistent with its
low affinity for the native rat
a4b2 receptor, (1S,4S)-10b did not
Biaryl substituted 2,5-diazabicyclo[2.2.1]heptane ligands were
evaluated for their ability to inhibit radioligands [3H]cytisine and
[3H]A-585539 binding to rat brain membranes. The functional
activity of these ligands was investigated using recombinant hu-
Cl
c
Ph
N
N
a
b
NH
N
N
N
N
N
A
A
man
show low affinity for
sine binding (Ki >10
a
7 NNR expressed in Xenopus oocytes. All of these compounds
4b2 NNR based on displacement of [3H]cyti-
M), which is consistent with our early finding
Boc
a
l
8
9a, A=NH
9b, A=NMe
10a, A=NH
d
e
for 10b. On the other hand, many of them also exhibit potent
a7
9c, A=Me2N+I-
10b, A=NMe
agonist activities. These results are summarized in Table 2.
As shown in Table 2, several SAR trends immediately emerged.
First, the nature of the terminal Ar group plays a very important
10c, A=Me2N+I-
role on
a7 NNR agonist activities. For instance, benzofuran-5-yl
Scheme 2. Synthesis of 2-(6-phenylpyridazin-3-yl)-2,5-diazabicyclo-[2.2.1]hep-
tanes. Reagents and conditions: (a) 2,6-dichloropyridazine, i-Pr2NEt, DMF, 100 °C,
16 h (78%); (b) HCl; (c) PhB(OH)2, Pd(PPh3)4, K2CO3 (aq), dioxane, 70 °C, 10 h, >70%;
(d) HCHO, NaBH(OAc)3, MeCN, rt, 10 h, >90%; (e) MeI, CH2Cl2, >80%.
and benzothiophen-5-yl analogues (12b and 12c) show compara-
ble binding affinity relative to phenyl analog 12a, but have approx-
imately a 4–6-fold increase in
1–3). N-Containing bicyclic aryl groups, such as indazol-5-yl and
indol-5-yl, led to a significant boost on both 7 NNR binding affin-
ity and functional potency (entries 4–5). Meanwhile, analog 12e
showed very weak [3H]-dofetilide binding affinity (Ki >20
M)
a7 NNR functional potency (entries
a
Table 1
Rat
a7 NNR binding affinity for 2-(6-phenylpyridazin-3-yl)-2,5-diazabicyclo
l
[2.2.1]heptanes
and excellent human liver microsome stability (>97% remaining
after 30 min incubation) thanks to the indol-5-yl substituent. An
electron-donating 2-methyl substituent on the indole (12i) shows
a slight increase in binding affinity and functional potency, but an
Entry
Compound
a
4b2
a7
Kia (nM)
Kib (nM)
1
2
3
4
5
6
7
8
9
10
(1S,4S)-9a
(1R,4R)-9a
(1S,4S)-9b
(1R,4R)-9b
(1S,4S)-9c
(1S,4S)-10a
(1S,4S)-10b
(1S,4S)-10c
(1R,4R)-10b
(1R,4R)-10c
6
0.2
139
26
303
486
1480
748
21.5
10.1
9.3
0.066
212
2.2
electron-withdrawing 2-CF3 group (12f) decreases
a7 binding
affinity by 5-fold relative to 12e (entries 5–7). Second, the substi-
tution site at the terminal aryl ring is important for activity. Indol-
5-yl is more favorable than indol-4-yl and indol-6-yl. For example,
12e is 35-fold more potent than 12g. Third, the linking heteroaryl
4.3
31,200
12,700
700
26,700
5500
ring has an effect on a7 NNR activity. In terms of the a7 NNR bind-
ing affinities, there is a clear preference for pyridine and pyridazine
over pyrimidine and pyrazine. Compounds 18 and 12 are at least
several fold more potent than their 14 and 16 counterparts. The
pyridin-3-yl linker appears to be optimal; ligands 18a–18h show
Displacement studies with [3H]cytisine using rat brain homogenates. The Ki
represents mean values obtained from independent experiments (n P 3,
SEM 6 10%.
a
Displacement of [3H]A-585539.
b
very potent
a7 NNR activity. For instance, indol-5-yl compound

3638
T. Li et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3636–3639
a, b
c
Boc
N
NH
Me
N
N
Cl
Me
N
N
Ar
N N
11
N N
12
(1S,4S)-8
(1S,4S)-8
N
N
d, b
e, b
f, b
c
c
Me
Me
N
N
N
N
Br
Br
Me
Me
N
N
N
N
Ar
Ar
N
N
13
14
N
N
(1S,4S)-8
(1S,4S)-8
N
15
N
16
c
Me
N
N
Cl
Me
N
N
Ar
N
N
17
18
Scheme 3. Synthesis of biaryl substituted 2,5-diazabicyclo[2.2.1]heptanes. Reagents and conditions: (a) 3,6-dichloropyridazine (1.2 equiv), i-Pr2NEt/DMF (v. 1/1, 0.5 M),
100 °C, 16 h, >70%; (b) HCHO (aq), HCOOH (aq), 100 °C, 1 h; (c) Ar-B(OH)2 (1.2 equiv), Pd(PPh3)4 (0.05 equiv), K2CO3 (aq), dioxane, 70 °C, 10 h, >70%; (d) (1) 2-chloropyrazine
(1.2 equiv) i-Pr2NEt/DMSO (v. 1/1, 0.5 M), 100 °C, 16 h; (2) NBS (1.0 equiv), MeCN, rt, 3 h; (e) 5-bromo-2-chloropyrimidine, i-Pr2NEt/DMSO (v. 1/1, 0.5 M), 100 °C, 16 h; (f)
5-bromo-2-chloropyridine, Pd2(dba)3 (0.02 equiv), Xantphos (0.06 equiv), t-BuONa (1.5 equiv), toluene (0.1 M), 100 °C, 3 h.
Table 2
Biological properties of biaryl substituted 2,5-diazabicyclo[2.2.1]heptanes
Entry
Compounds
Ar
a7 Binding assay
a7 Functional assay
[3H]-Dofetilide binding assay
Kia (nM)
EC50
(
lM)
Maxc (%)
Kid
(lM)
b
1
2
3
4
5
6
7
8
9
12a
12b
12c
12d
12e
12f
12g
12h
12i
14a
14b
14c
14d
14e
14f
14g
14h
16a
16b
16c
16d
16e
16f
Phenyl
10
14
9.5
2.7
1.5
1.1
0.9
1.2
2.7
1.7
0.4
5.6
11
55
75
78
70
79
69
85
65
70
75
41
71
77
90
71
84
59
45
20
31
83
77
37
69
51
6
Benzofuran-5-yl
Benzo[b]thiophen-5-yl
Indazol-5-yl
Indol-5-yl
2-CF3-indol-5-yl
Indol-4-yl
Indol-6-yl
2-Methyl-indol-5-yl
Phenyl
Benzofuran-5-yl
Benzo[b]thiophen-5-yl
Indazol-5-yl
Indol-5-yl
2-CF3-indol-5-yl
Indol-4-yl
8.8
1.8
> 55
21
2.2
> 57
12
3.1
0.9
0.4
2.1
13
7.0
0.3
18
61
35
6.3
1.0
17
12
26
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
5.3
1.2
> 65
14
1.9
> 51
9.0
22
1.3
0.56
>35
4.3
1.85
13
12
1.5
0.54
12
2.1
3.2
7.4
27
26
50
61
95
Indol-6-yl
Phenyl
Benzofuran-5-yl
Benzo[b]thiophen-5-yl
Indazol-5-yl
Indol-5-yl
2-CF3-indol-5-yl
Indol-4-yl
33
33
7.6
1.1
15
0.8
0.7
18
16g
18a
18b
18c
18d
18e
18f
73
4.6
1.8
4.7
5.3
0.3
1.0
2.8
3.9
1.6
Phenyl
3.1
4.4
2.9
0.3
0.2
3.0
10
10
Benzofuran-5-yl
Benzo[b]thiophen-5-yl
Indazol-5-yl
Indol-5-yl
2-CF3-indol-5-yl
Indol-4-yl
2.7
0.55
46
11
2.0
32
56
72
56
28
59
58
18g
18h
Indol-6-yl
5.4
2.9
a
Displacement experiments of [3H]A-585539 in rat brain (n P 3, SEM 6 10%).
b
Current activation measured by two-electrode voltage clamp using the automated parallel oocyte electrophysiology test station (POETs)18 and EC50 values determined by
analysis of concentration–response parameters using the nonlinear curve-fitting in Graphpad Prism.
c
Responses evoked by test compounds were measured as the evoked peak (maximal) inward current relative to the baseline and normalized to responses evoked by
10 mM ACh in the same oocyte.
d
[3H]-Dofetilide-isolated membrane binding assay.19
18e shows a Ki value of 0.2 nM and indazol-5-yl ligand 18d had an
EC50 value of 0.3 M.
Virtually all of the compounds listed in Table 2 are partial ago-
nists at 7, although a few (entries 7, 14, and 17) do approach full
efficacy. Nevertheless, our previous work20 has indicated that pro-
cognitive effects of a7 agonists require very little channel opening,
l
and occur at exposures well below those predicted by EC50’s mea-
sured in vitro. Therefore, the compounds in Table 2 exhibit phar-
macological profiles consistent with precognitive efficacy in
animal models.
a

T. Li et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3636–3639
3639
11. Papke, R. L.; Schiff, H. C.; Jack, B. A.; Horenstein, N. A. Neurosci. Lett. 2005, 378,
140.
In this study, we have discovered that biaryl substituted
2,5-diazabicyclo[2.2.1]heptanes are potent and selective 7 NNR
agents. SAR studies established that 5-N-methyl, heteroaryl linker
and the nature of terminal aryl group are critical for the ligands to
achieve potent a7 NNR agonist activity.
a
12. Toma, L.; Quadrelli, P.; Bunnelle, W. H.; Anderson, D. J.; Meyer, M. D.;
Cignarella, G.; Gelain, A.; Barlocco, D. J. Med. Chem. 2002, 45, 4011.
13. Bunnelle, W. H.; Daanen, J. F.; Ryther, K. B.; Schrimpf, M. R.; Dart, M. J.; Gelain,
A.; Meyer, M. D.; Frost, J. M.; Anderson, D. J.; Buckley, M.; Curzon, P.; Cao, Y.;
Puttfarcken, P.; Searle, X.; Ji, J.; Putman, C. B.; Surowy, C.; Toma, L.; Barlocco, D.
J. Med. Chem. 2007, 50, 3627.
14. Bunnelle, W. H.; Tietje, K. R.; Frost, J. M.; Peters, D.; Ji, J.; Li, T.; Scanio, M. J. C.;
Shi, L.; Anderson, D. J.; Dyhring, T.; Grønlien, J. H.; Ween, H.; Thorin-hagene, K.;
Meyer, M. D. J. Med. Chem. 2009, 52, 4126.
15. Anderson, D. J.; Bunnelle, W. H.; Surber, B.; Du, J.; Surowy, C.; Tribollet, E.;
Marguerat, A.; Bertrand, D.; Gopalakrishnan, M. J. Pharmacol. Exp. Ther. 2008,
324, 179.
16. (a) Buchwald, S. L. Top. Curr. Chem. 2002, 219, 131; (b) Hartwig, J. F. Acc. Chem.
Res. 1998, 31, 825–860; (c) Ji, J.; Li, T.; Bunnelle, W. H. Organic Lett. 2003, 5,
4611.
References and notes
1. (a) Dani, J. A.; Bertrand, D. Annu. Rev. Pharmacol. Toxicol. 2007, 47, 699; (b) Gotti,
C.; Zoli, M.; Clementi, F. Trends Pharmacol. Sci. 2006, 27, 482.
2. (a) Maskos, U. J. Neurochem. 2007, 100, 596–602; (b) Gotti, C.; Riganti, L.;
Clementi, F. Curr. Pharm. Design 2006, 12, 407.
3. (a) Martin, L. F.; Freedman, R. Int. Rev. Neurobiol. 2007, 78, 225–246; (b)
Berman, J. A.; Talmage, D. A.; Role, L. W. Int. Rev. Neurobiol. 2007, 78, 193–223;
(c) Mazurov, A.; Hauser, T.; Miller, C. H. Curr. Med. Chem. 2006, 13, 1567; (d)
Hashimoto, K.; Koike, K.; Shimizu, E.; Iyo, M. Curr. Med. Chem. 2005, 3, 171.
4. (a) Broad, L. M.; Sher, E.; Astles, P. C.; Zwart, R.; O’Neill, M. J. Drugs Fut. 2007, 32,
161–170; (b) Mazurov, A.; Hauser, T.; Miller, C. Curr. Med. Chem. 2006, 13, 1567.
5. Dani, J. A. Biol. Psychiatry 2001, 49, 166–174.
17. Suzuki, A. Pure Appl. Chem. 1991, 63, 419–422.
18. Trumbull, J. D.; Maslana, E. S.; McKenna, D. G.; Nemcek, T. A.; Niforatos, W.;
Pan, J. Y.; Parihar, A. S.; Shieh, C. C.; Wilkins, J. A.; Briggs, C. A.; Bertrand, D.
Receptors Channels 2003, 9, 19.
19. Diaz, G. J.; Daniell, K.; Leitza, S. T.; Martin, R. L.; Su, Z.; McDermott, J. S.; Cox, B.
F.; Gintant, G. A. J. Pjarmacol. Toxicol. Methods 2004, 50, 187.
6. Kem, W.; Soti, F.; Wildeboer, K.; LeFrancois, S.; MacDougall, K.; Wei, D.; Chou,
K.; Arias, H. R. Marine Drugs 2006, 3, 255.
20. (a) Briggs, C. A.; Grønlien, J. H.; Curzon, P.; Timmermann, D. B.; Ween, H.;
Thorin-Hagene, K.; Kerr, P.; Anderson, D. J.; Malysz, J.; Dyhring, T.; Olsen, G. M.;
Peters, D.; Bunnelle, W. H.; Gopalakrishnan, M. British J. Pharmacol. 2009, 158,
1486; (b) Tietje, K. R.; Anderson, D. J.; Bitner, R. S.; Blomme, E. A.; Brackemeyer,
P. J.; Briggs, C. A.; Browman, K. E.; Bury, D.; Curzon, P.; Drescher, K. U.; Frost, J.
M.; Fryer, R. M.; Fox, G. B.; Grønlien, J. H.; Håkerud, M.; Gubbins, E. J.; Halm, S.;
Harris, R.; Helfrich, R. J.; Kohlhaas, K. L.; Law, D.; Malysz, J.; Marsh, K. C.; Martin,
R. L.; Meyer, M. D.; Molesky, A. L.; Nikkel, A. L.; Otte, S.; Pan, L.; Puttfarcken, P.
S.; Radek, R. J.; Robb, H. M.; Spies, E.; Thorin-Hagene, K.; Waring, J. F.; Ween, H.;
Xu, H.; Gopalakrishnan, M.; Bunnelle, W. H. C. N. S. Neurosci. Ther. 2008, 14, 65.
7. van Kampen, M.; Selbach, K.; Schneider, R.; Schiegel, E.; Boess, F.; Schreiber, R.
Psychopharmacology 2004, 172, 375.
8. Bodnar, A. L.; Cortes-Burgos, L. A.; Cook, K. K.; Dinh, D. M.; Groppi, V. E.; Hajos,
M.; Higdon, N. R.; Hoffmann, W. E.; Hurst, R. S.; Myers, J. K.; Rogers, B. N.; Wall,
T. M.; Wolfe, M. L.; Wong, E. J. Med. Chem. 2005, 48, 905.
9. Bencherif, M.; Hauser, T. A.; Jordan, K. G.; Gatto, G. J. J. Mol. Neurosci. 2006, 30,
17.
10. Pichat, P.; Bergis, O. E.; Terranova, J.-P.; Urani, A.; Duarte, C.; Santucci, V.;
Gueudet, C.; Voltz, C.; Steinberg, R.; Stemmelin, J.; Oury-Donat, F.; Avenet, P.;
Griebel, G.; Scatton, B. Neuropsychopharmacology 2007, 32, 17–34.