New fused benzazepine as selective D3 receptor antagonists. Synthesis and biological evaluation. Part one: [h]-fused tricyclic systems

Article abstract of DOI:10.1016/j.bmcl.2007.12.066

The synthesis and SAR of a new series of potent and selective dopamine D₃ receptor antagonists is reported. The introduction of a tricyclic [h]-fused benzazepine moiety on the recently disclosed scaffold of 1,2,4-triazol-3-yl-thiopropyl-tetrahy

Full text of DOI:10.1016/j.bmcl.2007.12.066

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 901–907
New fused benzazepine as selective D3 receptor antagonists.
Synthesis and biological evaluation. Part one:
[h]-fused tricyclic systems
Fabrizio Micheli,a,* Giorgio Bonanomi,a Simone Braggio,a Anna Maria Capelli,b
Paolo Celestini,e Federica Damiani,d Romano Di Fabio,a Daniele Donati,a
Stefania Gagliardi,e Gabriella Gentile,a Dieter Hamprecht,a Marcella Petrone,a
Stefano Radaelli,e, Giovanna Tedesco,b Silvia Terreni,a
Angela Worbyc and Christian Heidbredera,à
aGlaxoSmithKline, Psychiatry Centre of Excellence for Drug Discovery, Via Fleming, 4, 37135 Verona, Italy
bGlaxoSmithKline, Molecular Discovery Research, Via Fleming, 4, 37135 Verona, Italy
cHarlow NFSP, Essex, UK
dEuropean Patent Office, Munich DE, USA
eNiKem Research S.r.l., Via Zambeletti, 25, 20021, Baranzate, Italy
Received 22 November 2007; revised 18 December 2007; accepted 19 December 2007
Available online 3 January 2008
Abstract—The synthesis and SAR of a new series of potent and selective dopamine D3 receptor antagonists is reported. The intro-
duction of a tricyclic [h]-fused benzazepine moiety on the recently disclosed scaffold of 1,2,4-triazol-3-yl-thiopropyl-tetra-
hydrobenzazepines is reported. A full rat pharmacokinetic characterization is also reported.
Ó 2007 Elsevier Ltd. All rights reserved.
Following the isolation and characterization of the
cDNA for the dopamine D3 receptor,1 a number of non-
selective and selective dopamine D3 receptor antagonists
have been reported.2 Growing evidence suggests that
selective antagonists at dopamine D3 receptor can re-
duce the reinforcing efficacy of drugs of abuse, reverse
cognitive deficits, and show efficacy in animal models
of schizophrenia.2a GSK showed a long-standing inter-
est in this field and contributed to the discovery of selec-
tive D3 receptor antagonists.3–5 While SB-2770113 (1,
Fig. 1) represented an excellent tool to the purpose of
target validation, the recently disclosed benzazepine
derivative 25–10 (Fig. 1) represents a more balanced mol-
ecule with improved developability criteria. In the pres-
ent study we investigated the possibility to ‘fuse’ the
pendant heterocycles attached to the benzazepine
(BAZ) template of the series to which 25–10 belongs with
the BAZ moiety itself to develop tricyclic derivatives.
Depending on the bond on which the fusion is achieved,
this task can lead to [h]-fused or [g]-fused derivatives.
This manuscript deals with the former class, while the
latter one will be the topic of the second part.
Each new chemical entity (NCE) prepared was assayed
for its agonistic versus antagonistic properties using a
functional GTPcS assay expressing the human dopa-
mine D3 receptor.5 All the compounds here reported
proved to be antagonists at the D3 receptor.5 The objec-
tive of the screening cascade for this specific series was
to identify molecules having at least 100-fold selectivity
versus dopamine D2 and histamine H1 receptors (func-
tional assays), and being endowed with 100-fold
selectivity versus the hERG ion channel (Dofetilide
binding assay).5 Generic developability screens such as
CYPEX bactosome P450 inhibition and rat and human
in vitro clearance in liver microsomes (Cli) were
included early in the screening cascade. Substituted 3-
[(3-chloropropyl)thio]-4-methyl-5-aryl-4 H-1,2,4-triazoles
Keywords: Rat; D3 antag; SAR; DMPK; In vitro; In vivo; Selective.
*
Corresponding author. Tel.: +39 045 8218515; fax: +39 045
8218196; e-mail: fm20244@gsk.com
Present address: Cambrex, Milan, Italy.
à
Present address: Philip Morris USA, Richmond, VA, USA.
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.12.066

902
F. Micheli et al. / Bioorg. Med. Chem. Lett. 18 (2008) 901–907
N
N
N
N
O
N
N
N
S
N
N
N
N
1
2
Figure 1. GSK selective D3 antagonist.
were reacted5–10 with tricyclic derivatives, according to
the general Scheme 1, at 60 °C in DMF using K2CO3
overnight and purified through column chromatography
to obtain the products described in Tables 1–7.
carbonyl group and acidic NH) were present, showed an
excellent D3 potency and selectivity versus D2 receptor.
Unfortunately, the hERG value did not match the selec-
tivity criteria described in the screening cascade.
For the preparation of the alpha-substituted derivatives,
reductive ammination was used as depicted in Scheme 2.
The introduction of an oxazinone system (4) on the thio-
triazole scaffold (11, 12) led to a marked increase in
In this series it is important to highlight that the het-
erocyclic decoration on the thiotriazole moiety can be
modified with no significant loss in terms of D3 affin-
ity (12 vs. 11, 14 vs. 13). Taking into account the high
PSA values displayed by derivatives 11–17 (ranging
PSA11 (values > 90 A with respect to 65 A for deriva-
tive 2) and was detrimental both to the desired D3 affin-
ity and to the selectivity of the molecule, leading to an
increased unwanted interaction with the hERG channel
(Table 1). This could have been due to either a potential
detrimental effect of a hydrogen bond (HB) donor (mor-
pholinone N–H) or to the presence of an additional HB
acceptor (morpholinone C@O). The first hypothesis
may find some support in the fact that N-methylation
(13, 14) allowed some recovery in terms of D3 affinity
and significantly reduced hERG affinity. On the other
hand, while the sulfonamide (15) further increased the
desired affinity with the removal of both features, the
acetamide (16) had a similar profile to 11 in terms of
D3 affinity having introduced again a carbonyl group
into the system. However, the [1,4]oxazino[2,3-h][3]ben-
zazepine derivative (17), where none of the features (i.e.,
2
2
b
b
2
b
from 90 to 130 A ), the pharmacokinetic (PK) prop-
erties of 15 were assessed in vivo in the rat.12 The
compound showed low blood clearance (Clb = 32 ml/
min/kg), moderate half life (T1/2 = 1.2 h), and low dis-
tribution volume (Vd = 2.4 l/kg). One may assume that
this parameter (Vd), coupled with the high PSA value
mentioned above, led to low brain penetration: the
brain to blood (B/B) ratio, measured through sam-
pling of the blood and brain12 1 h after i.v. adminis-
tration, was actually equal to 0.2.
Consequently, it was decided to move from 6-membered
anellated BAZ to 5-membered anellated BAZ to modu-
late the PSA parameter and the isoxazolo[4,5-h][3]ben-
zazepine scaffold was designed (Table 2).
6
R
N
N
N
N
N
N
n
N
S
8
Ar
N
N
S
Ar
S
N
N
O
84-89
90
O
N
N
N
n
N
S
N
N
N
S
Ar
5
3
N
9
O
N
R
N
N
R
m
n
1
4
Ar
N
11-17
7
6
80-83
N
N
Cl
n
2
S
Ar
R
N
5
N
O
3
N
N
N
N
N
R
n
n
S
Ar
S
Ar
O
N
N
4
R
18-21
22-73
2
R
O
N
S
6
R
N
O
3
R
5
N H
N
N H
N
R
N H
N H
N
O
N
R
O
O
4
1
R
7
6
8
4
5
O
N H
N
F
N
N
CF3
d
N
N
c
F
O
Ar =
a
b
e
f
9
Scheme 1. Reagents and conditions: K2CO3, DMF 60 °C, cat NaI and appropriate tricyclic 4–8, overnight. n = 0, 1, 2. m = 0, 1. R as defined in each
table.

F. Micheli et al. / Bioorg. Med. Chem. Lett. 18 (2008) 901–907
903
Table 1. Affinity results
O
N
N
N
S
Ar
N
O
N
R1
m
2
b
Compound
R1
m
Ar
D3 fpKi
D2 fpKi
H1 fpKi
hERG pIC50
PSA ðA Þ
69
65
ACD logD
1
N.A.
N.A.
H
8.4
8.8
6.6
5.6
7.4
7.2
7.9
6.8
8.1
6.4
6.5
6.2
6.1
6.8
5.7
6.6
5.7
nt
5.7
5.7
8.2
7.2
5.3
4.6
5.1
5.0
6.4
3.8
5.3
3.5
1.3
4.2
2.1
2.9
4.7
4.2
2
11
12
13
14
15
16
17
1
1
1
1
0
0
0
a
b
a
b
a
a
a
<5.6
<5.0
<5.6
<6.7
<5.6
<5.6
<5.9
110
124
102
115
127
102
93
H
Me
Me
MeSO2
Ac
6.0
nt
H
SEM for D3 GTPcS, H1 FLIPR and HERG data sets is 0.1 and for the D2 GTPcS data is 0.2. N.A., not applicable; nt, not tested.
Table 2. Affinity results
2
R
N
N
N
N
S
Ar
O
N
2
b
Compound
R2
Ar
D3 fpKi
D2 fpKi
H1 fpKi
hERG pIC50
PSA ðA Þ
ACD logD
18
19
20
21
Me
Me
Et
a
b
a
d
7.7
6.8
7.6
7.9
<6.2
<5.9
<6.1
<6.9
7.0
nt
6.1
5.2
6.0
6.9
98
86
73
60
5.5
3.4
6.1
6.2
6.9
6.2
Me
SEM for D3 GTPcS, H1 FLIPR and HERG data sets is 0.1 and for the D2 GTPcS data is 0.2. nt, not tested.
The first derivative prepared (18) had actually a lower
2
The first compound prepared, 22, was slightly less po-
tent than the corresponding isoxazolyl derivative 18,
but it was much more selective and devoid of any hERG
activity. Also for this compound, the rat PK profile was
immediately performed, showing a relatively high Clb
(47 ml/min/kg), T1/2 = 1.7 h and F = 34%. One may as-
sume that because of its Vd (4.3 l/kg), its brain penetra-
tion was very good showing a B/B ratio = 5.4. A slight
increase in the clogD13 value shown by compound 26
(5.6 vs. 5.0 for 22) led to a 10-fold increase in the D3
affinity, in agreement with the analysis deriving from
the D3 receptor model (Fig. 3). In order to assess
whether or not the increased lipophilicity was detrimen-
tal to the in vivo PK properties, the profile of this deriv-
ative was studied. The compound was endowed with a
relatively higher Clb and Vd (57 ml/min/kg and 5.1 l/
kg, respectively) with respect to 22, but both F (31%)
and T1/2 (1.6 h) were almost identical. A further increase
in the lipophilicity of the n-Pr derivative (28,
clogD = 6.1) led to an improvement in D3 affinity, but
was also paralleled by an increase in hERG affinity;
moreover the overall PK parameters were much more
affected leading to low F (6%), probably because of high
Clb (62 ml/min/kg).
b
(98 A ) PSA value compared to the majority of the scaf-
fold reported above, but it was still higher than our ref-
2
b
erence compound 2 (65 A ). The in vivo rat PK profiling
was therefore put forward in the screening cascade to as-
sess the brain penetration of the series.
Compound 18 showed acceptable bioavailability (F) and
half life (F = 22%; T1/2 = 1.3 h) with acceptable Vd and
Clb (3.3 l/kg and 40 ml/min/kg, respectively), but more
importantly, it demonstrated high B/B (7.7) with a large
amount of compound in the brain. The introduction of
the oxazolyl moiety on the thiotriazole (19) contributed
b
to a further reduction of PSA (86 A2). This had a posi-
tive impact on hERG, but also the desired D3 affinity
was 10-fold reduced. The introduction of an Et substitu-
ent (20) in the tricyclic portion did not substantially
change the affinity, as well as the presence of a para-
CF3 Ph substituent (21) on the thiotriazole moiety
which, however, was detrimental both for hERG activ-
ity and for some in vitro PK properties.
Based on receptor modeling5 results analysis, the isoxaz-
olyl portion in the tricyclic scaffold was rearranged to an
oxazolyl one, to maintain similar PSA values, reduce the
favorable interactions within the hERG channel (Fig. 2)
because of the position of R3 substituent, and have a
better interaction with key features in the D3 receptor
pocket (Fig. 3).
As predicted by the SAR developed so far, the introduc-
tion of the i-Pr substituent (29, 30 vs. 22, 23) was posi-
tive in terms of desired D3 affinity; while both
molecules demonstrated similar and positive in vitro

904
F. Micheli et al. / Bioorg. Med. Chem. Lett. 18 (2008) 901–907
Table 3. Affinity results
N
O
3
R
N
N
N
S
Ar
N
2
b
Compound
R3
Ar
D3 fpKi
D2 fpKi
H1 fpKi
hERG pIC50
PSA ðA Þ
ACD logD
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
Me
Me
a
b
b
a
a
b
a
a
b
a
b
a
c
7.2
6.8
6.7
7.3
7.9
7.5
8.4
8.5
8.0
7.3
6.7
9.1
6.9
7.6
7.5
7.0
7.0
8.1
7.6
7.3
7.4
8.2
7.7
6.6
7.9
7.4
8.3
9.0
8.3
8.5
8.1
8.4
<5.6
<5.6
<5.6
<5.6
<6.3
<6.3
<6.2
<5.9
<6.1
<6.2
<6.2
<6.0
<6.1
<6.1
<6.1
<6.3
<6.2
<6.2
<6.3
<6.2
<6.2
<6.1
<6.1
<6.2
<6.2
<6.2
<6.2
<6.3
<6.4
<6.2
<6.3
<6.4
<5.7
5.6
<5.0
4.5
5.2
5.4
5.3
5.2
6.2
5.7
5.4
6.0
<5.4
6.1
4.3
5.4
5.0
<4.2
4.8
6.6
5.9
6.0
5.6
5.9
nt
98
111
86
73
73
86
73
73
86
73
86
73
73
60
60
69
73
60
60
73
73
73
86
82
73
86
76
76
73
86
91
104
5.0
2.9
2.6
4.8
5.6
3.4
6.1
5.9
3.8
5.0
2.9
6.2
3.6
5.7
5.1
2.7
4.2
6.2
5.6
6.2
6.4
6.0
3.8
4.9
5.9
3.8
2.1
4.9
8.2
6.1
5.9
3.7
H
<5.3
5.6
H
Et
<5.9
<5.5
<5.6
<5.5
<5.5
<5.8
<5.5
<5.3
<5.4
<5.4
<5.4
<5.5
<5.6
<5.5
<5.6
<5.8
<5.5
<5.7
<5.7
<5.5
<5.5
<5.5
<5.5
<5.5
<5.6
<5.6
<5.6
<5.5
Et
Pr
i-Pr
i-Pr
CyPr
CyPr
t-Bu
Me
Me
d
e
Me
Me
f
Et
Et
c
d
e
Et
CF3
a
a
a
b
a
a
b
a
a
a
b
a
b
CH2CF3
CF2Me
CF2Me
CH2OMe
CH2CyPr
CH2CyPr
Piperidyl
Me2N
CF2CF3
CF2CF3
pir*
<4.8
5.9
6.0
5.3
4.7
7.0
6.7
6.3
6.0
pir*
SEM for D3 GTPcS, H1 FLIPR and HERG data sets is 0.1 and for the D2 GTPcS data is 0.2. nt, not tested; pir*, 1,3-dimethyl-1H-pyrazol-5-yl.
PK parameters (P450 and rat Cli), their in vivo proper-
ties were sub-optimal probably due to high Clb (81 and
152 ml/min/kg, respectively). Actually very low F% and
T1/2 values were achieved. An attempt to slow down the
potential metabolic degradation of the i-Pr substituent
was undertaken with the introduction of the cyclopropyl
derivative (31, 32), but the outcome was not positive in
terms of D3 affinity. The introduction of the t-Bu moiety
(33), on the other hand, led to a very high D3 affinity
with 1000-fold selectivity over the D2, and H1 receptors
and the hERG channel. Unfortunately, the rat Clb of
this derivative was too high (137 ml/min/kg) to make it
progressable down the screening cascade. At the same
time, the exploration of the right-hand side of the mol-
ecule introducing different aromatic or heterocyclic
groups was performed. The introduction of a 3-pyridyl
derivative (34) had a positive effect on the hERG side,
but was detrimental to the main target, while a p-CF3
Ph (35) was well balanced in terms of affinity profile.
Its PK properties showed a Clb equal to 62 ml/min/kg,
with a relatively good T1/2 (1.5 h), a moderate Vd
(5.4 l/kg), while F was 17%. Brain penetration was also
good, with a B/B ratio = 2.9. In accordance with the
SAR developed for this series, the introduction of an
Et group (39) was positive for the primary affinity and
had no major effect on the PK parameters
(Clb = 48 ml/min/kg; Vd = 3.4 l/kg; T1/2 = 1.1 h;
F = 17% ; B/B = 3.1).
The introduction of groups less prone to metabolic deg-
radation, like the –CF3, in the tricyclic scaffold, (41) led
to a much reduced Clb (20 ml/min/kg) and to a relatively
balanced PK profile (Vd = 5.0 l/kg; T1/2 = 3.3 h;
F = 31%; B/B = 2.8), while the introduction of spaced
cyclopropyl derivatives (46, 47) maintained reasonable
affinity and a relatively good PK profile (F = 25%,
Vd = 4.3 l/kg, Clb = 50 ml/min/kg; and F = 32%,
Vd = 1.7 l/kg; Clb = 26 ml/min/kg for 46 and 47, respec-
tively). The introduction of amine moieties was positive
from the affinity point of view and the piperidine deriv-
ative 48 showed a 1000-fold selectivity versus hERG
with an excellent B/B ratio (5.1); its Clb was unfortu-
nately relatively high (76 ml/min/kg) leading to poor F
data (3%). More promising was the dimethyl amino
derivative 49 that not only achieved nanomolar potency
and excellent selectivity, but was also endowed with a

F. Micheli et al. / Bioorg. Med. Chem. Lett. 18 (2008) 901–907
905
Table 4. Affinity results
m
N
O
3
R
N
N
N
S
Ar
n
N
4
R
2
b
Compound
R3
R4
m
n
Ar
D3 fpKi
D2 fpKi
H1 fpKi
hERG pIC50
PSA ðA Þ
ACD logD
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
Me
Et
Br
Br
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
0
0
0
0
0
0
0
1
1
1
1
1
1
a
a
a
a
a
a
b
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
7.6
<7.2
7.9
7.4
7.5
7.4
6.5
7.4
7.4
7.7
8.2
7.6
7.2
7.5
6.5
6.7
6.5
6.5
6.6
6.9
6.6
6.9
7.1
7.3
7.0
7.0
<6.2
<6.2
<6.2
<6.0
<6.0
<6.0
<6.2
<6.1
6.4
<5.9
<6.3
6.0
5.6
5.9
6.4
5.3
6.3
6.6
4.3
5.3
5.8
6.6
5.8
5.8
6.6
4.3
5.2
5.6
6.2
4.9
5.6
6.4
4.9
5.4
6.0
5.0
5.5
5.7
73
73
73
73
73
73
86
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
5.8
6.4
7.0
5.5
6.0
6.6
2.9
5.9
6.4
7.0
5.5
6.0
6.7
5.3
6.1
6.6
7.2
5.8
6.3
6.9
5.8
6.3
6.9
6.1
6.6
7.3
t-Bu
Me
Et
Br
Me
Me
Me
H
<5.7
<5.7
5.8
t-Bu
Me
Me
Et
<5.6
<5.7
5.7
Br
Br
CF3
Me
Et
Br
6.0
6.8
5.9
Me
Me
Me
H
<5.6
<5.6
5.6
6.2
6.2
CF3
Me
Me
Et
<6.1
<6.0
<6.0
<6.1
<6.0
<6.0
<5.8
<6.0
<6.0
<6.0
<5.8
<6.0
<5.8
<5.5
<5.8
<5.6
<5.6
<5.6
<5.6
<5.6
<5.6
<5.6
<5.7
6.0
Br
Br
CF3
Me
Et
Br
Me
Me
Me
Me
Me
Me
Br
CF3
Me
Et
CF3
Me
Et
Br
Br
<5.6
<6.3
CF3
SEM for D3 GTPcS, H1 FLIPR and HERG data sets is 0.1 and for the D2 GTPcS data is 0.2.
Table 5. Affinity results
N
S
5
R
N
N
N
S
Ar
N
2
b
Compound
R5
Ar
D3 fpKi
D2 fpKi
H1 fpKi
hERG pIC50
PSA ðA Þ
ACD logD
80
81
82
83
Me
Me
Et
a
b
a
b
7.7
7.1
8.6
8.4
<6.1
<6.0
<6.3
<6.5
6.0
5.8
5.6
5.5
6.6
5.9
60
73
60
73
5.6
3.4
6.1
3.9
<5.7
<5.6
Et
SEM for D3 GTPcS, H1 FLIPR and HERG data sets is 0.1 and for the D2 GTPcS data is 0.2.
good PK profile (Clb = 29 ml/min/kg; T1/2 = 1.6 h;
Vd = 2.8 l/kg; F = 45%; B/B = 1.1). The compound was
therefore progressed along the screening cascade to
in vivo animal disease models and the exploration was
further continued for better understanding of the SAR
around this series.
spacer length (60–66 vs. 54–59) was relatively well toler-
ated, as well as its reduction in nonsubstituted systems
(67). The introduction of a substitution alpha to the ba-
sic nitrogen in the system led to a general decrease in
hERG activity, with no significant effect on D3 affinity
(74–79).
The tolerance of the D3 receptor versus the introduction
of relatively basic groups in the tricyclic template was
further confirmed by the introduction of a substituted
pyrazole (52, 53) in the system, while the introduction
of a Bromine atom or of a Methyl group on the tricyclic
portions was almost neutral (54–59). The increase in the
The modification of the tricyclic portion with the
replacement of the oxazolyl moiety with a thiazolyl
one was well tolerated from the D3 affinity point of view
(80–83, Table 5) and showed a similar trend for the
hERG profile. From the PK point of view, derivative
80 showed no major issues in terms of P450 or Cli and

906
F. Micheli et al. / Bioorg. Med. Chem. Lett. 18 (2008) 901–907
Table 6. Affinity results
m
6 N
N
N
N
R
S
Ar
n
N
4
O
R
2
b
Compound
R6
m
n
Ar
D3 fpKi
D2 fpKi
H1 fpKi
hERG pIC50
PSA ðA Þ
ACD logD
84
85
86
87
88
89
Me
i-Pr
Me
i-Pr
Me
i-Pr
0
0
0
0
1
1
1
1
2
2
1
1
a
a
a
a
a
a
6.2
7.6
<6.0
<5.8
<6.0
<6.0
<6.0
<6.0
5.5
<5.7
<5.6
<5.6
<5.6
<5.6
nt
67
67
67
67
67
67
3.1
4.0
3.2
4.0
3.4
4.3
5.4
4.5
5.4
4.4
5.6
6.6
7.4
<6.3
6.9
SEM for D3 GTPcS, H1 FLIPR and HERG data sets is 0.1 and for the D2 GTPcS data is 0.2. nt, not tested.
Table 7. Affinity results
N
N
S
N
N
N
N
N
2
b
Compound
D3 fpKi
8.6
D2 fpKi
6.4
H1 fpKi
6.2
hERG pIC50
6.0
PSA ðA Þ
65
ACD logD
90
5.8
SEM for D3 GTPcS, H1 FLIPR and HERG data sets is 0.1 and for the D2 GTPcS data is 0.2.
O
4
4
4
4
R
R
R
R
OH
O
N
O
N
O
N
3
3
N
N
3
R
2
O
N
3
3
N
NH
R
N
R
S
N
N
R
1
10 a-f
Ar
OH
Cl
74-79
O
OH
6
R
N
N
6
R
6
R
2
6
R
3
N
N
NH
N
N
N
N
S
N
N
1
10 a-f
Ar
O
O
O
O
OH
Cl
N
N
88-89
Ar
N
HS
10 a-f
Scheme 2. Reagents and conditions: (1) 4-hydroxy-butan-2-one, NaBH(OAc)3, THF 0 °C, rt; (2) SOCl2, CHCl3, 0 °C, rt; 3 h; (3) Et3N, NaI, DMF
70 °C, 24 h.
its profile was comparable to derivative 22 showing a
Clb = 34 ml/min/kg, a T1/2 = 2.2 h, a Vd = 4.3 l/kg, and
F = 25%. The brain penetration was also excellent, with
a B/B ratio equal to 4.4.
high polarity of the new left-hand side. To counterbal-
ance the reduction in clogD in the new scaffold, a higher
degree of lipophilicity in the substitution pattern was
introduced to improve the D3 affinity (e.g., 85 vs. 84).
2
b
Nonetheless, the much reduced PSA (67 A for 85)
was probably the reason for low B/B (0.2) showed by
the NCE, while the remaining parameters were similar
to previously tested compounds (Vd = 3.5 l/kg;
Clb = 36 ml/min/kg; T1/2 = 1.5 h; F = 10%).
Further modifications, balancing the slight increase in
lipophilicity due to the ethyl substituent on the left-hand
side with the oxazolyl moiety on the right portion of the
molecule (83), led to a very balanced profile also in
terms of hERG affinity.
Finally, using the same chemistry described in Scheme 1,
the pyrazolyl derivative 90 was prepared to complete the
analysis of the pattern of substitution in the h-fused scaf-
fold. From the in vitro point of view this led to excellent
affinity and selectivity profile as reported below in Table 7.
The introduction of a more hydrophilic lactam portion
led to a more significant modification of the tricyclic sys-
tem, and the first derivative prepared (84) showed poor
affinity at the desired receptor, probably due to the very

F. Micheli et al. / Bioorg. Med. Chem. Lett. 18 (2008) 901–907
907
This exploration will be the topic of the following Part 2
of this manuscript.
Acknowledgment
We thank Dr. Carla Marchioro and her analytical group
for the high quality support received.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmcl.2007.12.066.
Figure 2. An energetically favorable docking solution of 21 in the
closed state of the hERG receptor model. This solution suggests that
this ligand can establish hydrophobic interactions with Phe656 and
Tyr652 while its protonated nitrogen can make an HB interaction with
Ser624 in the pore helix.
References and notes
1. Sokoloff, P.; Giros, B.; Martres, M. P.; Bouthenet, M. L.;
Schwartz, J. C. Nature 1990, 347, 146.
2. (a) Micheli, F.; Heidbreder, C. Recent Patents CNS Drug
Discov. 2006, 1, 271; (b) Zhang, A.; Kan, Y.; Li, F. Expert
Opin. Ther. Patent 2006, 16, 587.
asn352
3. Reavill, C.; Taylor, S. G.; Wood, M. D.; Ashmeade, T.;
Austin, N. E.; Avenell, K. Y.; Boyfield, I.; Branch, C. L.;
Cilia, J.; Coldwell, M. C.; Hadley, M. S.; Hunter, A. J.;
Jeffrey, P.; Jewitt, F.; Johnson, C. N.; Jones, D. N.;
Medhurst, A. D.; Middlemiss, D. N.; Nash, D. J.; Riley,
G. J.; Routledge, C.; Stemp, G.; Thewlis, K. M.; Trail, B.;
Vong, A. K.; Hagan, J. J. J. Pharmacol. Exp. Ther. 2000,
294, 1154.
4. Macdonald, G. J.; Branch, C. L.; Hadley, M. S.; Johnson,
C. N.; Nash, D. J.; Smith, A. B.; Stemp, G.; Thewlis, K.
M.; Vong, A. K.; Austin, N. E.; Jeffrey, P.; Winborn, K.
Y.; Boyfield, I.; Hagan, J. J.; Middlemiss, D. N.; Reavill,
C.; Riley, G. J.; Watson, J. M.; Wood, M.; Parker, S. G.;
Ashby, C. R., Jr. J. Med. Chem. 2003, 46, 4952.
5. Micheli, F.; Bonanomi, G.; Blaney, F. E.; Braggio, S.;
Capelli, A. M.; Checchia, A.; Curcuruto, O.; Damiani, F.;
Di-Fabio, R.; Donati, D.; Gentile, G.; Gribble, A.;
Hamprecht, D.; Tedesco, G.; Terreni, S.; Tarsi, L.;
Lightfoot, A.; Pecoraro, M.; Petrone, M.; Perini, O.;
Piner, J.; Rossi, T.; Worby, A.; Pilla, M.; Valerio, E.;
Griffante, C.; Mugnaini, M.; Wood, M.; Scott, C.;
Andreoli, M.; Lacroix, L.; Schwarz, A.; Gozzi, A.; Bifone,
A.; Ashby, C. R., Jr.; Hagan, J. J.; Heidbreder, C. J. Med.
Chem. 2007, 50, 5076.
tyr365
ser182
hys349
ser193
ser192
phe345
thr368
asp110
trp342
phe197
Figure 3. Derivative 26 was probed for its ability to bind a dopamine
D3 receptor model based on the bovine rhodopsin crystal structure.
Using a docking approach based on Macromodel, a possible binding
mode shows a strong interaction of the basic benzazepine nitrogen
with Asp110 on transmembrane helix 3 (TM3). Hydrogen bond
interactions can also be observed between the triazole ring and Asn352
(to a lesser extent, also to Hys349 and Ser182). The fused oxazolo-
benzazepine scaffold extends through the ‘classic’ agonist binding
pocket toward TM5, with the Et group (R3) positively fitting in the
available space. Two aromatic residues, Phe345 and Tyr365, can form
p-stacking interactions, respectively, with the oxazolo-benzazepine
ring and the quinoline mojety.
6. Hadley, M.S.; Lightfoot, A.P.; MacDonald, G.J.; Stemp,
G. WO2002040471, 2002.
In summary, substitution of a simple benzazepine ring
with a more complex tricyclic structure was well toler-
ated by the D3 receptor and some NCEs here described
showed remarkable selectivity versus D2, H1 receptors
and the hERG channel. Among this new class of selec-
tive D3 receptor antagonists, a number of derivatives
with balanced rat PK profiles were also discovered.
These will be very useful in the continuation of the
screening cascade to further validate the concept of
selective D3 receptor antagonism in different therapeutic
areas.
7. Bonanomi, G.; Damiani, F.; Gentile, G.; Hamprecht,
D.W.; Micheli, F.; Tarsi, L.; Tedesco, G.; Terreni, S.
WO2005118549, 2005.
8. Hamprecht, D.; Micheli, F.; Tarsi, L.; Tedesco, G. WO
2005123717, 2005.
9. Bonanomi, G.; Cardullo, F.; Damiani, F.; Gentile, G.;
Hamprecht, D.; Micheli, F.; Tarsi, L. WO 2006002928,
2006.
10. Arista, L.; Bonanomi, G.; Damiani, F.; Hamprecht, D.;
Micheli, F.; Tarsi, L.; Tedesco, G. WO 2005087764, 2005.
11. Ertl, P.; Rohde, B.; Selzer, P. J. Med. Chem. 2000, 43,
3714.
12. The research complied with national legislation and with
company policy on the Care and Use of Animals and with
related codes of practice.
The exploration of the new tricyclic scaffold was to be
completed probing the sensitivity of the D3 receptor to
the [g]-fusion and to the introduction of much more po-
lar systems into the new template.
13. ACD
Labs
www.acdlabs.com.
suite
is
available
from
http://