Optimisation of imidazole compounds as selective TAAR1 agonists: Discovery of RO5073012

Article abstract of DOI:10.1016/j.bmcl.2012.06.060

A series of imidazole compounds has been identified which affords potent and selective partial and full agonists of the TAAR1 receptor. Starting from 2-benzyl-imidazoline screening hits, a series of structurally related 2-benzyl- and 4-benzyl-imidazoles was investigated first, but it proved highly challenging to obtain compounds having sufficient selectivity against the adrenergic alpha 2 receptor. This issue could be successfully addressed by modification of the linker region and SAR exploration led to the discovery of highly selective isopropyl-substituted 4-aminomethyl-imidazole compounds. The work culminated in the identification of the selective TAAR1 partial agonist RO5073012 (4-chlorophenyl)-(1H-imidazol-4-ylmethyl)-isopropyl-amine, 24), which has a good pharmacokinetic profile after oral administration in rodents. RO5073012 has been found to be active in a behavioural rat model which is considered indicative for schizophrenia.

Full text of DOI:10.1016/j.bmcl.2012.06.060

Bioorganic & Medicinal Chemistry Letters 22 (2012) 5244–5248
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Optimisation of imidazole compounds as selective TAAR1 agonists: Discovery
of RO5073012
⇑
Guido Galley , Henri Stalder, Annick Goergler, Marius C. Hoener, Roger D. Norcross
F. Hoffmann-La Roche Ltd, Pharmaceuticals Division, Preclinical Research, CH-4070 Basel, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of imidazole compounds has been identified which affords potent and selective partial and full
agonists of the TAAR1 receptor. Starting from 2-benzyl-imidazoline screening hits, a series of structurally
related 2-benzyl- and 4-benzyl-imidazoles was investigated first, but it proved highly challenging to
obtain compounds having sufficient selectivity against the adrenergic alpha 2 receptor. This issue could
be successfully addressed by modification of the linker region and SAR exploration led to the discovery of
highly selective isopropyl-substituted 4-aminomethyl-imidazole compounds. The work culminated in
the identification of the selective TAAR1 partial agonist RO5073012 (4-chlorophenyl)-(1H-imidazol-4-
ylmethyl)-isopropyl-amine, 24), which has a good pharmacokinetic profile after oral administration in
rodents. RO5073012 has been found to be active in a behavioural rat model which is considered indica-
tive for schizophrenia.
Received 3 May 2012
Revised 15 June 2012
Accepted 18 June 2012
Available online 23 June 2012
Keywords:
TAAR1 agonist
Trace amines
Imidazole
SAR
Ó 2012 Elsevier Ltd. All rights reserved.
It has been known for over four decades that ‘trace amines’ such
as b-phenethylamine (PEA) and p-tyramine are present in the
mammalian CNS, but the precise function of these putative neuro-
transmitters could thus far only be hypothesised. The trace amines
have been postulated to play a role in modulation of the classical
neurotransmitters such as dopamine and serotonin, and dysregula-
tion of trace amine levels has been linked to highly prevalent
psychiatric diseases such as depression, schizophrenia and bipolar
disorder.1 The first member of the GPCR family for which the trace
amines are the endogenous ligands was not identified until 2001,
and since then a total of nine trace amine-associated receptors
(TAARs) have been classified.2 Within this family, the trace amines
PEA and p-tyramine have been shown to be selective agonists for
the TAAR1 receptor. However, the primary amines PEA and p-tyra-
mine are rapidly catabolised by monoamine oxidase and the ensu-
ing low metabolic stability precludes their use as drugs and
severely limits their suitability to serve as tool compounds for
in vivo experiments to further investigate the pharmacology of
TAAR1. Accordingly, there is much interest in the identification
of safe, selective and drug-like TAAR1 agonists which may offer
the potential to deliver innovative therapeutics for psychiatric
indications.3
GPCRs. Therefore, on the one hand it might be challenging to ob-
tain TAAR1 agonists having high selectivity versus other biogenic
amine GPCRs, but on the other hand there should be a high likeli-
hood to find TAAR1 agonists by screening a corporate compound
collection containing compounds from historic biogenic amine-
based drug discovery programs.
Anticipating a high hit-rate, we thus decided to screen a sub-set
of about 55,000 compounds from the Roche compound library using
a functional assay employing the chimeric human/rat TAAR1 recep-
tor.4 300 hit compounds were identified having an EC50 <10
lM, 15
of which showed an EC50 <1
lM. Structural evaluation of the hits
revealed that many actives clustered around a ‘2-benzyl-imidazo-
line’ motif and most were derivatives of the known adrenergic com-
pounds tolazoline 1, naphazoline 2 and tetryzoline 3 (Fig. 1).5
For SAR exploration, around one hundred 2-benzyl-imidazolines
5 were either purchased or synthesised from the corresponding
benzyl cyanide 4 by reaction with ethylenediamine and sulphur
N
N
N
NH
N
H
N
H
The structures of the natural agonists PEA and p-tyramine sug-
gest that the TAAR1 agonist pharmacophore is intrinsically small
and shares a common motif with several other biogenic amine
1
2
3
⇑
Corresponding author. Tel.: +41 61 6880437; fax: +41 61 6888714.
E-mail address: guido.galley@roche.com (G. Galley).
Figure 1. Derivatives of tolazoline 1, naphazoline 2 and tetryzoline 3 were found as
screening hits.
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.06.060

G. Galley et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5244–5248
5245
N
N
CN
O
O
a)
N
N
H
b)
H
a)
b), c)
N
Et
Et
MeO
OMe
R
R
R
MeO
OMe
8
5
6
4
7
9
Scheme 1. Synthesis of 2-benzylimidazolines and 2-benzylimidazoles. Reagents
and conditions: (a) ethylenediamine, S8, 200 °C, W, 20 min, 17–80%; (b) DMSO,
l
Br
N
N
(COCl)2, CH2Cl2, ꢀ78 °C, then addition of imidazoline in CH2Cl2, ꢀ78 °C, 50 min,
d)
f)
then addition of NEt3, warm up to rt., 43–84%.
e)
N
N
Br
Et
H
H
Et
Et
Et
Et
Et
under microwave irradiation (Scheme 1).6 Because the related
2-benzyl-imidazoles 6 were available in a single further step from
the imidazolines by oxidation,7 we were also interested to include
these in our studies.
10
6d
5d
Scheme 2. Synthesis of compound 5d and 6d. Reagents and conditions: (a) 2,4-
dimethyl-pent-3-ylamine, p-TsOH, toluene, reflux, 40 min, 99%; (b) 2.6 equiv EtLi,
THF, ꢀ10 °C, 2 h (c) 4 M aq HCl in THF, reflux, 3 h, 93%; (d) Ph3P, CBr4, CH2Cl2, <5 °C
then rt, 30 min, 60%; (e) ethylenediamine, CH2Cl2, rt, 18 h, 48%; (f) DMSO, (COCl)2,
NEt3, CH2Cl2, ꢀ78 °C to rt., 2.5 h, 75%.
Since affinity for the adrenergic
a
2 receptors is known for many
representatives of this class8 we established an
a2 binding assay to
measure selectivity. For TAAR1, a binding assay using the human
TAAR1 receptor (hTAAR1) was used as the primary screen.9 To
avoid undesirable cardiovascular or CNS side effects arising from
Further in vitro characterisation of RO4992479 (6d) in a Cerep
Custom Profile (comprising over 120 receptors, ion channels and
enzymes)11 showed significant additional activities only at the
a2 activity, a selectivity ratio of 100-fold or better was set as the
threshold for choosing compounds for in vivo characterisation.
Testing the compounds revealed that both the benzylimidazo-
line class 5 and the benzylimidazole class 6 delivered potent li-
gands for the TAAR1 receptor, with sub-micromolar affinity
observed for compounds bearing small lipophilic substituents on
the phenyl ring.
imidazoline receptor I1 (Ki = 0.66 lM) and at 5HT-1A (1.3 lM).
Unfortunately, rodent PK studies with RO4992479 revealed low
bioavailability after oral administration,12 which we attributed to
first-pass hepatic oxidative metabolism at the benzylic positions,
so a search for alternative compounds was necessary.
Unfortunately, most of the active compounds were more potent
To extend the scope of our studies we decided to investigate the
regioisomeric 4-imidazole compounds and to vary the linker be-
tween this basic moiety and the phenyl ring. Accordingly, 4-ben-
zyl- and 4-phenethyl-imidazole 11 and 12 were found to be very
potent hTAAR1 ligands (Table 2).
at the adrenergic
a2 receptor than at the hTAAR1 receptor (e.g.,
5a–d in Table 1, many further examples tested are not shown).
In the imidazole series a few compounds exhibited a degree of
selectivity against the adrenergic
a2 receptor: all of these carry
two ortho substituents with o,o0-diethyl being optimal for potency
and selectivity (6d = RO4992479, which is 100-fold selective for
2-Aminomethyl-imidazoles13 13 and 14 were found to be
potent as well, and we anticipated these compounds might show
higher metabolic stability compared to carbon linked compounds
(N vs C at benzylic position). We therefore decided to investigate
hTAAR1 vs a2).
Since most of the o,o0-disubstituted compounds were difficult to
obtain by the benzyl cyanide route (Scheme 1), an improved
synthesis was developed as shown for compounds 5d and 6d in
Scheme 2. The corresponding benzaldehyde 9, which is available
from 7 according to the methodology of Flippin et al,10 is trans-
formed into the dibromo-olefin 10, which is further cyclised with
ethylenediamine at room temperature to yield benzylimidazoline
5d. Compound 6d can then be obtained by Swern oxidation as
before.
the effect on selectivity for TAAR1 versus a2 of varying the N-alkyl
substituent in a series of 2-aminomethyl-imidazoles (Table 3).
Whereas methyl- or ethyl-substituted compounds (14, 15) are
unselective, isopropyl substitution has a dramatic effect, with
compound 16 displaying a 24-fold selectivity for hTAAR1 versus
adrenergic
a2 receptor. In contrast, substitution with benzyl or
phenyl leads to much less selective compounds (17, 18). Interest-
ingly, although the N-alkyl substituents are similar in size, the
N-cyclopropyl compound 19 is less potent and less selective than
the N-isopropyl compound 16.
Bicyclic examples where the N-alkyl substituent is tethered to
the aromatic o-position (20, 21) were found to be in the same
Table 1
TAAR1 binding and selectivity versus
a
2 for o,o0-disubstituted 2-benzylimidazolines
5a–f and benzylimidazoles 6a–f
potency range at hTAAR1 but were not selective versus a2 (Table 4).
N
N
N
H
N
H
R
R
R
R
Table 2
TAAR1 binding and selectivity versus
a2 for 4-imidazole compounds 11–12
N
5a-f
6a-f
NH
R
Compd
R
Imidazoline
Imidazoline
Imidazole
Imidazole
affinity Ki
hTAAR1
(nM)
selectivity Ki
affinity Ki
selectivity Ki
a2/Ki
hTAAR1
Compd
R
Affinity Ki hTAAR1
(nM)
Selectivity Ki a2/Ki
hTAAR1
a2/Ki hTAAR1 hTAAR1
(nM)
5a/6a
5b/6b
5c/6c
5d/6d
5e/6e
5f/6f
H
Cl
1640
500
82
300
825
0.06
0.05
0.77
0.68
4.0
400
1390
36
24
630
4.7
0.7
4.5
100
14
11
12
20
2
2.5
5.1
Me
Et
iPr
CF3 43000
n.d.
13300
n.d.

5246
G. Galley et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5244–5248
Table 3
Table 6
TAAR1 binding and selectivity versus
a
2 for 4-imidazole compounds 13–19
TAAR1 binding and selectivity versus
a2 for 4-imidazole compounds with heterocy-
clic substitution in place of phenyl 36–39
N
N
NH
N
R
R1
N
NH
R2
Compd
R
Affinity Ki hTAAR1 (nM)
Selectivity Ki a2/Ki hTAAR1
Compd R1
R2 Affinity Ki hTAAR1
Selectivity Ki a2/Ki
hTAAR1
13
14
15
16
17
18
19
H
65
11
48
22
32
5.3
1.6
1.6
23.7
4.0
0.9
(nM)
Me
Et
iPr
Bn
Ph
cPr
36
37
iPr 195
12
62
N
69
100
3.3
iPr
12
N
MeO
N
Table 4
TAAR1 binding and selectivity versus
indoline 20 and 1,2,3,4-tetrahydroquinoline 21
a
2 for 4-imidazole compounds derived from
38
39
iPr 100
33
3
N
Cl
N
N
N
NH
Et
84
N
MeO
()n
Compd
n
Affinity Ki hTAAR1 (nM)
Selectivity Ki a2/Ki hTAAR1
20
21
1
2
35
12
0.14
4
The fact that lipophilic substituents on the phenyl ring are pre-
ferred, and polarity is apparently less tolerated by the receptor in
this region of the molecule, is also reflected by the weaker affinity
of compounds bearing heterocycles in place of phenyl. When we
tested a small set of pyridine and pyrimidine derivatives (Table
6), only the methoxypyridine analogue 37 stands out with regard
to potency and selectivity. Based on the observed lower functional
activity (vide infra) of many of these heterocyclic compounds, we
did not investigate this approach further.
A general route for the synthesis of 4-aminomethyl-imidazole
compounds 43 and 44, employing two consecutive reductive ami-
nation procedures without requiring protection of the imidazole
moiety, is shown in Scheme 3. For the preparation of the isopropyl
compounds 44, 2-methoxypropene was used in the presence of
sodium acetoxy-borohydride and TFA.14 A small amount of hemi-
aminal ether 45 could be detected in the reaction mixture in some
cases, but this could be transformed to the desired product 44 by
acidic work-up.
We decided to investigate the SAR of the 2-aminomethyl-imid-
azole class further, with a special focus on the isopropyl series.
Table 5 shows examples prepared to determine the influence of
substitution on the phenyl ring. Small lipophilic substituents are
generally preferred. Introduction of chlorine in meta position leads
to a gain in potency and selectivity for TAAR1 versus a2 (23); para-
chloro substitution is beneficial as well and affords the most
selective isopropyl compound (24). Dichloro or meta-methoxy sub-
stitution is also favourable (25, 26, 30), whereas introduction of
polarity (e.g., by attaching a sulfone substituent, 32) leads to less
active TAAR1 ligands. In terms of the substituent on the nitrogen,
the larger tert-butyl group affords the most selective compound,
although at the expense of affinity at hTAAR1 (35).
Table 5
TAAR1 binding and selectivity versus
a
2 for 4-imidazole compounds 22–35
R1
NH2
H
N
N
R1
N
a)
N
NH
R1
+
N
H
NH
N
42
CHO
R2
40
41
Compd R1
R2
Affinity Ki hTAAR1
Selectivity Ki a2/Ki
hTAAR1
H
N
MeO
H
N
(nM)
N
N
R2
N
iPr
iPr
iPr
iPr
iPr
iPr
iPr
iPr
iPr
iPr
iPr
Me
Et
40
4
6
4
4
18
12
13
5
76
72
140
71
37
16
34
104
67
33
8
N
22
23
24
25
26
27
28
29
30
31
32
33
34
35
o-Cl
m-Cl
p-Cl
3,4-DiCl
3,5-DiCl
o-F
m-F
p-F
m-MeO
p-MeO
m-MeSO2
m-Cl
m-Cl
m-Cl
c)
b)
N
42
+
N
N
R1
R1
R1
43
44
45
d)
80
128
11
2
Scheme 3. Synthesis of aminomethyl-imidazole compounds 42–44 by reductive
amination. Reagents and conditions: (a) NaBH4, MeOH, rt, 4 h, 60–99%; (b) R2-CHO,
NaCNBH3, ZnCl2, MeOH, 40 °C, 18 h, 25–88%; (c) 2-methoxypropene, NaBH(OAc)3,
CF3COOH, CH2ClCH2Cl, 60 °C, 18 h, 32–64%; (d) 1 N aq HCl, H2O, stir or reflux
mixture of 45 and 44 during work-up, 1 h, quantitative.
1.3
21
200
tBu
33

G. Galley et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5244–5248
5247
dimethylaminosulfonyl group15 is thermally stable even under
microwave heating for longer periods, longer reaction times being
necessary when using the isopropyl-substituted amines. At the end
of the sequence the dimethylaminosulfonyl group can be cleaved
off by either strongly acidic or strongly basic conditions to yield
the pyrimidine derivatives 38, 39.
N
N
O
O
S
S
H
N
O
O
a)
b)
N
N
N
N
N
O
R1
CHO
CHO
N
The synthetic route shown in Scheme 3 was also used to pro-
duce a set of regioisomeric 2-aminomethyl-imidazoles, by starting
from imidazole-2-carbaldehyde instead of imidazole-4-carbalde-
hyde. These compounds were tested for binding and selectivity
(Table 7). As was the case for the 4-imidazoles, meta- or para-
chloro substitution on the phenyl ring improves binding affinity
and selectivity (53, 54) but the compounds are in general at least
10-fold weaker in binding at hTAAR1. Again, isopropyl substitution
on the nitrogen leads to the most selective compound (53).
Next we studied the functional activity of aminomethyl-imidaz-
ole compounds at the mouse, rat and human TAAR1 receptors;9
data for selected compounds is shown in Table 8. Most of the ami-
nomethyl-imidazole compounds are partial agonists (relative to
the endogenous agonist PEA defined as 100% agonism) having sim-
ilar activity at the mouse, rat and human TAAR1 receptors; the
minimal inter-species differences should simplify the development
of potential drug candidates from this series.
46
H
41
47
N
Cl
c)
S
H
N
O
N
N
N
N
+
N
R2
R1
R1
N
d) or e)
N
N
N
N
N
38, 39
R2
R2
48
Scheme 4. Microwave-assisted synthesis of pyrimidine compounds 38, 39 using
the dimethylaminosulfonyl protecting group. Reagents and conditions: (a) Me2N-
SO2Cl, NEt3, CH2Cl2, rt, 18 h, 94%; (b) R1NH2, if R1 = Et: Pd/C, H2, MeOH, rt, 18 h, 91%;
if R1 = iPr: NaBH4, MeOH, 4 h, 82%; (c) W, 160–180 °C, iPrOH, 30–90 min, if R2 = Cl:
l
chromatographic separation of regioisomers, 6–90%; (d) 5 M HCl in EtOH, 80 °C, 2 h,
35–80%; (e) NaOMe in MeOH or DME, 80 °C, 18 h, 40–60%.
Based on the potency, selectivity and balanced functional activ-
ity across species, we decided to investigate compound 24
(=RO5073012) in more detail. The experimentally determined
physicochemical properties (Table 9) are in a favourable range
for a CNS drug;16 in this regard the low molecular weight is also
advantageous (MW = 249.7 g/mol). A Cerep Custom Profile cover-
ing over 60 targets11 showed weak activities at the imidazoline
Table 7
TAAR1 binding and selectivity versus
a
2 for 2-imidazole compounds 49–54
H
N
R1
N
N
receptor I1 (Ki = 0.77
l
M),17 as had been seen with the first lead
compound 6d.18
R2
Because of its low to moderate in vitro clearance in microsomes,
RO5073012 was tested in pharmacokinetic (PK) studies in mice
and rats and was found to have moderate volume of distribution,
low to moderate clearance and excellent oral bioavailability in
both species (Table 10). The measured brain/plasma ratios indi-
cated good CNS penetration. Binding to plasma proteins in vitro
was only moderate, with concomitantly high unbound fraction
(8% in mouse plasma; 6% in rat plasma).
Compd R1
R2
H
Affinity Ki hTAAR1 (nM) Selectivity Ki a2/Ki hTAAR1
49
50
51
52
53
H
H
H
H
m-
Cl
4250
0.2
0.2
0.7
8
Me 1270
Et
710
470
68
iPr
iPr
64
54
p-Cl
iPr
138
22
Following the demonstration of favourable PK properties in
rodents, RO5073012 was tested in a rat behavioural model: the
reversal of cocaine-induced hyperlocomotion paradigm was
chosen since it should reflect the modulatory nature of TAAR1 on
the dopaminergic signalling pathway and is considered relevant
for psychiatric diseases.20–22 The dose-dependent effect seen after
p.o. administration of RO5070312 (see Fig. 2) provides evidence
When we tried to obtain the heterocyclic analogues by using
this procedure we were only able to synthesise the pyridine ana-
logues 36 and 37. For the pyrimidine derivatives we developed a
thermal halogen displacement route using a dimethylaminosulfo-
nyl protected imidazole derivative 46 according to Scheme 4. The
Table 8
Functional activity at mouse, rat and human TAAR1 receptor for selected compounds
Compd
EC50/Emax mouse TAAR1 (nM/%)
EC50/Emax rat TAAR1
EC50/Emax human TAAR1 (nM/%)
16
23
24
30
37
53
28/45
10/50
23/26
72/55
98/35
500/54
7.3/34
27/22
25/24
61/25
25/18
400/60
6.8/69
4.6/61
23/35
37/29
40/59
320/92
Table 9
Measured physicochemical properties of RO5073012 (24)
Solubilitya
97
(l
g/ml)
logD (pH 7.4)
Permeation coefficientb (10ꢀ6 cm sꢀ1
2.7
)
pKa
3.3
3.5/6.7
a
Thermodynamic measurement with solid material at final pH 6.5.
Membrane permeability as measured in the PAMPA assay.19
b

5248
G. Galley et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5244–5248
Table 10
assistance in the synthesis of the compounds. We would also like
In vitro and in vivo PK parameters of RO50373012 (24)
to thank Veit Metzler, Danièle Buchy and Sylvie Chaboz for per-
forming in vitro pharmacology experiments, Roland Mory and Sean
Durkin for performing in vivo behavioural experiments, Gerhard
Hoffmann for coordinating & interpreting the rodent PK studies
and Jean-Luc Moreau for coordinating and interpreting behavioural
studies in rat.
Parameter
Mouse
Rat
Microsomal clearancea
Cl (ml/min/mg protein)
i.v. dose
Clearance (ml/min/kg)
Vss (l/kg)
17.5
5 mg/kg
24
70
4.8 mg/kg
40
2.0
4.1
Brain/plasma ratio
p.o. dose
cmax (ng/ml)
0.6
6.6
3.8 mg/kg
519
Supplementary data
10 mg/kg
1740
1.8
t1/2 (h)
1.0
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2012.06.
060. These data include MOL files and InChiKeys of the most
important compounds described in this article.
Bioavailability F (%)
86
66
a
Intrinsic clearance measured from in vitro incubations with mouse or rat liver
microsomes; for human liver microsomes a clearance value of 18 ml/min/mg pro-
tein was measured.
References and notes
1. Berry, M. D. J. Neurochem. 2004, 90, 257.
HORIZONTAL
ACTIVITY
2. Lindemann, L.; Hoener, M. Trends Pharmacol. Sci. 2005, 26, 274.
3. Branchek, T. A.; Blackburn, T. P. Curr. Opin. Pharmacol. 2003, 3, 90.
4. Reese, E. A.; Bunzow, J. R.; Arttamangkul, S.; Sonders, M. S.; Grandy, D. K. J.
Pharmacol. Exp. Ther. 2007, 321, 178.
5. Petrusewicz, J.; Kaliszan, R. Pharmacology 1986, 33, 149.
6. Mohammadpoor-Baltork, I.; Abdollahi-Alibeik, M. Bull. Korean Chem. Soc. 2003,
24, 1354.
7. Huh, D. H.; Ryu, H.; Kim, Y. G. Tetrahedron 2004, 60, 9857.
8. Petrusewicz, J.; Kaliszan, R. Gen. Pharmacol. 1991, 22, 819.
9. For assay descriptions see Supplementary data.
10. Flippin, L. A.; Carter David, S.; Dubree, N. J. P. Tetrahedron Lett. 1993, 34, 3255.
11. Cerep, Le bois l’Evêque, 86600 Celle l’Evescault, France (www.cerep.fr); see also
Supplementary data.
12. The following PK parameters were determined after oral administration in
rodents: cmax = 300 ng/ml, t1/2 = 0.95 h, F = 21% following a dose of 7.1 mg/kg
p.o. in mice; cmax = 100 ng/ml, t1/2 = 4.9 h, F = 17% following a dose of 10 mg/kg
p.o. in rats.
13. Various aminomethyl-imidazoles have been described as
a2- adrenergic
ligands Bagley, J. R.; Brockunier, L. L.; Kudzma, L. V.; Pandit, C. R.; Pratt, D.
V.; Galdes, A.; Jerussi, T. P.; Del Vecchio, R. A.; Harris, S.; Bylund, D. B. Med.
Chem. Res. 1994, 4, 346.
Figure 2. In vivo activity of RO5073012 (24) in the cocaine-induced hyperlocomo-
tion test in rat: the hyperlocomotion observed after administration of cocaine
(20 mg/kg i.p.) is reduced when the rats are pre-treated with different doses of 24
p.o.24
14. Reddy, T. J.; Leclair, M.; Proulx, M. Synlett 2005, 583.
15. Chadwick, D. J.; Ngochindo, R. I. J. Chem. Soc., Perkin Trans. 1 1984, 481.
16. Wager, T. T.; Chandrasekaran, R. Y.; Hou, X.; Troutman, M. D.; Verhoest, P. R.;
Villalobos, A.; Will, Y. ACS Chem. Neurosci. 2010, 1, 420.
17. Revel, F. G.; Meyer, C. A.; Bradaia, A.; Jeanneau, K.; Calcagno, E.; André, C. B.;
Haenggi, M.; Miss, M. T.; Galley, G.; Norcross, R. D.; Invernizzi, R. W.; Wettstein,
J. G.; Moreau, J. L.; Hoener, M. C. Neuropsychopharmacology, in press. For Cerep
profile see Supplementary data.
18. TAAR1 agonistic activity of imidazoline receptor ligands has recently been
published: Hu, L. A.; Zhou, T.; Ahn, J.; Wang, S.; Zhou, J.; Hu, Y.; Liu, Q. Biochem.
J. 2009, 424, 39.
19. Kansy, M.; Senner, F.; Gubernator, K. J. Med. Chem. 1998, 1007, 41.
20. Revel, F. G.; Moreau, J.-L.; Gainetdinov, R. R.; Bradaia, A.; Sotnikova, T. D.; Mory,
R.; Durkin, S.; Groebkezbinden, K.; Norcross, R. D.; Meyer, C. A.; Metzler, V.;
Chaboz, S.; Ozmen, L.; Trube, G.; Pouzet, B.; Bettler, B.; Caron, M. G.; Wettstein,
J.; Hoener, M. C. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 8485.
that TAAR1 agonists have a CNS activity profile which may be of
therapeutic benefit.23
In summary, we have identified low molecular weight imidazole
compounds which are potent, selective, and drug-like TAAR1
agonists. Starting from 2-benzylimidazoline screening hits, we
investigated related 2- and 4-benzyl imidazoles but selectivity
versus the adrenergic a2 receptor was difficult to achieve. Modifica-
tion of the linker region led to discovery of 4-aminomethyl-imidaz-
ole compounds having strong affinity at TAAR1 and good selectivity
21. Revel, F. G.; Moreau, J. L.; Pouzet, B.; Mory, R.; Bradaia, A.; Buchy, D.; Metzler,
V.; Chaboz, S.; Groebke Zbinden, K.; Galley, G.; Norcross, R. D.; Tuerck, D.;
Bruns, A.; Morairty, S. R.; Kilduff, T. S.; Wallace, T. L.; Risterucci, C.; Wettstein, J.
G.; Hoener, M. C. Mol. Psychiatry 2012. http://dx.doi.org/10.1038/mp.2012.57.
Epub ahead of print.
22. Revel, F. G.; Moreau, J. L.; Gainetdinov, R. R.; Ferragud, A.; Velázquez-Sánchez,
C.; Sotnikova, T. D.; Morairty, S. R.; Harmeier, A.; Groebke Zbinden, K.;
Norcross, R. D.; Bradaia, A.; Kilduff, T. S.; Biemans, B.; Pouzet, B.; Caron, M. G.;
Canales, J. J.; Wallace, T. L.; Wettstein, J. G.; Hoener, M. C. Biol. Psychiatry 2012.
http://dx.doi.org/10.1016/j.biopsych.2012.05.014. Epub ahead of print.
versus a2 could be obtained for compounds carrying an N-isopropyl
substituent. This work culminated in the identification of the
selective TAAR1 partial agonist RO5073012 (4-chlorophenyl)-(1H-
imidazol-4-ylmethyl)-isopropyl-amine, 24), which has
a good
pharmacokinetic profile after oral administration in rodents and is
active in a CNS behavioural model in rat.
Acknowledgments
23. All in vivo experiments were conducted in compliance with Swiss Federal and
Cantonal laws on animal research and AAALAC regulations and received prior
approval by the Cantonal Veterinary Office.
The authors are grateful to Roman Hutter, Phillip Schmid,
Daniela Kruesi and Sandy Desrat for their excellent technical
24. For details on the method see Supplementary Material.