Identification of a series of 4-[3-(quinolin-2-yl)-1,2,4-oxadiazol-5-yl] piperazinyl ureas as potent smoothened antagonist hedgehog pathway inhibitors

Article abstract of DOI:10.1016/j.bmcl.2011.07.031

The Hedgehog (Hh-) signalling pathway is a key developmental pathway and there is a growing body of evidence showing that this pathway is aberrantly reactivated in a number of human tumors. Novel agents capable of inhibiting this pathway are sought, and an entirely novel series of smoothened (Smo) antagonists capable of inhibiting the pathway have been identified through uHTS screening. Extensive exploration of the scaffold identified the key functionalities necessary for potency, enabling potent nanomolar Smo antagonists like 91 and 94 to be developed. Optimization resulted in the most advanced compounds displaying low serum shift, clean off-targets profile, and moderate clearance in both rats and dogs. These compounds are valuable tools with which to probe the biology of the Hh-pathway.

Full text of DOI:10.1016/j.bmcl.2011.07.031

Bioorganic & Medicinal Chemistry Letters 21 (2011) 5274–5282
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Identification of a series of 4-[3-(quinolin-2-yl)-1,2,4-oxadiazol-5-yl]piperazinyl
ureas as potent smoothened antagonist hedgehog pathway inhibitors
Jesus M. Ontoria a, Laura Llauger Bufi a, Caterina Torrisi a, Alberto Bresciani a, Claudia Giomini a,
Michael Rowley a, Sergio Serafini a, Hu Bin b, Wu Hao b, Christian Steinkühler a, Philip Jones a,
⇑
a IRBM, Merck Research Laboratories Rome, Via Pontina km 30,600, Pomezia, 00040 Rome, Italy
b WuXi AppTec Co., Ltd, 288 Fute Zhong Road, Waigaoqiao Free Trade Zone, Shanghai 200131, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
The Hedgehog (Hh-) signalling pathway is a key developmental pathway and there is a growing body of
evidence showing that this pathway is aberrantly reactivated in a number of human tumors. Novel agents
capable of inhibiting this pathway are sought, and an entirely novel series of smoothened (Smo) antag-
onists capable of inhibiting the pathway have been identified through uHTS screening. Extensive explo-
ration of the scaffold identified the key functionalities necessary for potency, enabling potent nanomolar
Smo antagonists like 91 and 94 to be developed. Optimization resulted in the most advanced compounds
displaying low serum shift, clean off-targets profile, and moderate clearance in both rats and dogs. These
compounds are valuable tools with which to probe the biology of the Hh-pathway.
Received 18 June 2011
Revised 29 June 2011
Accepted 6 July 2011
Available online 14 July 2011
Keywords:
Hedgehog pathway
Smoothened antagonist
Hedgehog inhibitor
Ó 2011 Elsevier Ltd. All rights reserved.
Several developmental pathways regulate patterning, growth
and cell migration during embryonic development, of which the
hedgehog (Hh-) signalling pathway is a one of these. While this
pathway is critical during development, its role in later life is lim-
ited to a small number of tissues and to a maintenance/repair
role.1–4 However, in the last decade evidence has emerged showing
that this pathway is aberrantly activated in a growing number of
cancers, and strategies to inhibit this pathway are being evaluated
for therapeutic use to target defined tumors dependent on this
pathway.
The Hh-signalling cascade is initiated by binding of Hh-protein to
its cellular membrane receptor Patched (Ptch), and this binding re-
lieves Ptch-mediated repression of the GPCR-like receptor Smooth-
ened (Smo). Alleviation of this repression initiates an intracellular
signalling cascade culminating with the Gli-transcription factors
(Gli1-3) regulating genes involved in differentiation, proliferation
and survival. In tumors, mutations in Ptch and Smo, as well as down-
stream factors SUFU and Gli, have been identified, thereby activat-
ing the pathway.1,2
(2),6 has demonstrated encouraging results in phase I studies in
advanced basal cell carcinoma.7 We have recently disclosed the
identification and optimization of a novel series of [6,5]-bicyclic
tetrahydroimidazo[1,5-a]pyrazine-1,3(2H,5H)-dione Smo antago-
nists,8 culminating in the development of MK-5710 (3).9 An alter-
native method to inhibit the pathway include small molecule that
binds Hedgehog,10 such as robotnikinin (4) as well as binders to Gli
proteins like GANT61 (5),11 and other compounds whose mecha-
nisms of action are being elucidated.12,13
Herein we described the development of an alternative class of
structurally distinct, potent Smo antagonists with acceptable
pharmacokinetics in preclinical species, culminating in the
identification of 91 and 94.
The program began with an high throughput screening cam-
paign against the corporate screening collection to identify small
molecules useful for the treatment of Hh dependent malignancies
where compounds were screened for their ability to inhibit the Hh
pathway in Shh-light2 cells expressing a Gli-dependent reporter
gene, simultaneously monitoring for cytotoxicity.14 Hits were then
profiled in a whole cell Smo binding assay in 2% fetal bovine serum
(FBS), measuring the ability of compounds to displace a Bodipy-
labeled cyclopamine derivative.15 The initial screen identified three
closely related 4-[3-(quinolin-2-yl)-1,2,4-oxadiazol-5-yl]piperidi-
nyl urea (6–8) as Smo antagonists, and inhibitors of the Hh-path-
way ( Fig. 2). All three compounds inhibited the pathway with
several hundred nanomolar activity, and inhibited binding of the
labeled cyclopamine derivative at similar concentrations.
Several agents have been demonstrated to show preclinical
antitumor activity in animal models,3,5 The majority of the re-
search has focused on Smo antagonists, including the natural prod-
uct Cyclopamine (1) ( Fig. 1), and various synthetic molecules.
Recently, the most advanced of these, Vismodegib (GDC-0449)
⇑
Corresponding author. Present Address: Belfer Institute for Applied Cancer
Sciences, Dana Farber Cancer Institute, 450 Brookline Avenue, Boston, MA 02115,
USA. Tel.: +1 617 582 9648.
The ortho-chloro derivative (6) was profiled more extensively
and the compound showed high clearance in rat microsomes
E-mail address: philip_jones@dfci.harvard.edu (P. Jones).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.07.031

J. M. Ontoria et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5274–5282
5275
Table 2
Cl
Exploration of the central saturated heterocyclic ring system
HN
H
N
O
MeO
H
O
N
HN
O
H
N
HN
( )a
H
N
N
N
N
Cl
N
N
Ph
N
O
H
H
O
( )b
O
N
O
HO
SO2Me
Cyclopamine (1)
MK-5710 (3)
R
Compound Light2 IC50 (nM)
SmoBind 2% FBS IC50 (nM)
Vismodegib (2)
N
N
7
410
180
O
Cl
NH
NMe2
NMe2
N
N
11
12
28% inh. at 10
15% inh. at 10
l
M
M
39% inh. at 10
38% inh. at 10
l
l
M
M
O
O
N
N
NH
O
l
Robotnikinin (4)
GANT61 (5)
Figure 1. Representation Hh-pathway inhibitors.
with modest hERG activity (62% inh. at 10
of CYPs 2C9 and 3A4 (62 / 58% inh. at 10
l
M), and also inhibition
lM).
Encouraged by these results, further SAR exploration was
undertaken, extensively exploring the SAR of each of the four com-
ponents of the lead scaffold: the quinoline ring system; the 1,2,4-
oxadiazole; the piperidine ring system, and the right hand side
capping group.
HN R
O
N
O
N
N
N
Initial studies exploring the importance of the 1,2,4-oxadiazole
rapidly established that this isomeric ring system was critical from
binding to Smo, as the alternative isomers (9 and 10) were inferior
to 7 (Table 1). The isomeric 1,2,4-oxadiazole (10) was round 6-fold
less active, while the corresponding 1,3,4-isomer (9) displayed
only weak micromolar affinity.
Cl
MeO
6
7
8
Similar studies were conducted to explore the importance of
the piperidine ring (Table 2), and ring contraction to either the cor-
responding azetidine or pyrrolidine ring systems resulted in abol-
ishment of affinity for Smo, with both 11 and 12 showing only
Light2 IC50 (nM)
510
380
410
180
480
350
Smo Bind 2% FBS IC50 (nM)
Figure 2. Lead structures from uHTS campaign.
modest inhibition at 10 lM.
Knowing that the piperidine moiety of 6 was one of the sites for
oxidative metabolism, some efforts were undertaken to try and
(Clint >200
man microsomes (Clint = 65 and 45
l
L/min/mgP) and moderate clearance in dog and hu-
lL/min/mgP respectively). In
vivo the compound showed low clearance in rats with
Cl = 11 min/ml/kg, which could be attributed to the lipophilic nat-
ure of the compound which displayed log D = 4.25, and high plas-
ma protein binding (fraction unbound (fu) rat 1%, human 0.5%).
The oral bioavailability of 6 was acceptable at 38%. Limited metab-
olite identification studies in microsomes pointed towards the
piperidine and chlorophenyl groups as being prone to oxidative
metabolism. The compound displayed some off-target activity
Table 3
Effect of substituents on piperidine ring system
Cl
HN
O
N
O
N
X
N
N
R
Compound
Light2 IC50 (nM)
510
SmoBind 2% FBS IC50 (nM)
N
6
380
790
Table 1
Exploration of the oxadiazole ring system
13a
14
680
N
N
MeO
HN
O
X
Z
N
900
640
620
N
Y
N
15
2400
R
Compound
Light2 IC50 (nM)a
410
SmoBind 2% FBS IC50(nM)a
N
16
17
2500
1300
6000
2100
N
N
7
180
N
N
O
O
O
HO
F
9
57% inh. at 20
2400
lM
4700
15000
N
N
18
19
1300
140
1000
170
N
N
10
N
N
a
Values are means of at least 3 experiments (std. dev. were within 25% of the IC50
values).
a
Mixture of diastereomers.

5276
J. M. Ontoria et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5274–5282
prevent this metabolism (Table 3), either by addition of additional
substituents, or through changing the electronic of the ring system.
Unfortunately, introduction of additional substituents on this 6-
membered ring proved to be detrimental to affinity, as all substi-
tute piperidines displayed inferior potency to 6, even addition of
a 4-fluoro group in 18 resulted in a 3-fold loss in activity. However,
replacement of the piperidine for a piperazine as in the case of 19,
resulted in a 3-fold improvement in both Smo binding affinity
(IC50 = 170 nM) and also a similar improvement in activity in the
light2 assay, where 19 inhibited the hedgehog pathway with
IC50 = 140 nM. Noticeably, the introduction of the additional nitro-
gen in the piperazine scaffold, resulted in lower log D = 2.95, and
accordingly improved both the CYP inhibition (2D6, 2C9 and 3A4
IC50 >25 lM) and hERG (IC50 = 12 lM) profiles.
Having established that the central portion of these Smo antag-
onists was critical for the affinity, attention focused on the two
extremities of the molecule, focusing first on replacement of the
2-quinoline ring (Table 4). The bicyclic heterocyclic framework is
Table 4
Exploring the SAR of the left-hand side heterocyclic group
R
N
HN R'
N
O
N
O
MeO
Cl
R’
R
Cpd
SmoBind 2% FBS IC50 (nM)
Cpd
SmoBind 2% FBS IC50 (nM)
20
>2.5
>2.5
l
l
M
M
21
>2.5
l
M
22
23
>2.5
l
M
24
26
>2.5
>2.5
l
l
M
M
N
N
25
27
>2.5
>2.5
l
l
M
M
N
N
28
30
51% inh at 2
lM
29
31
63% inh at 2.5 lM
N
>2.5
>2.5
l
l
M
M
>2.5
>2.5
l
l
M
M
N
N
32
34
33
N
2200
N
35
>2.5
l
M
36
6
>2.5
380
l
M
7
180
N
N
N
N
37
38
750
1900
39
40
>2.5
>2.5
l
l
M
M
N
N
41
42
1100
N
N
>2.5
l
M
M
N
N
N
43
44
2200
N
>2.5
l
N
N
45
1000
N

J. M. Ontoria et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5274–5282
5277
important for potent inhibition of the hedgehog pathway, as alter-
ation to a simple phenyl (20–21), or monocyclic heterocycle such
as pyridines (23–27), pyrimidine (30–31) or pyrazine (32–33) re-
sulted in significant losses in activity. Only the addition of a
3-methyl-2-pyridyl group in 28 resulted in some modest inhibition
of the pathway with around 50% inhibition at 2 lM. When the 2-
quinoline was replaced by a 2-naphthyl group approximately a
10-fold loss in potency was observed with 34 displaying
When larger more lipophilic alkyl ureas were explored an inter-
esting trend was observed, and cycloalkyl ureas like cyclopentyl
(68) and cyclohexyl (69) displayed sub-micromolar activity in both
assays. In particular, the cyclohexyl 69 had a very interesting pro-
file, inhibiting the pathway with IC50 = 57 nM in the light2 assay
and showing IC50 = 280 nM in the binding assay, where it competes
for binding against Bodipy-cyclopamine. Despite the increase in
potency of this compound, the cyclohexyl urea suffered from both
IC50 = 2.2
l
M. This position of the nitrogen in the quinoline ring
high turnover in rat microsomes (Clint >300 lL/min/mgP) and high
was critical as when the corresponding 1-isoquinoline (35–36),
or 3-isoquinoline were prepared these too displayed inferior inhi-
plasma protein binding (human fu = 0.5%). The corresponding car-
bamate (70) was round 10-fold less active in the functional assay,
with IC50 = 580 nM.
bition of the Hh-pathway with 37 displaying IC50 = 0.5
Smo binding assay and IC50 = 1.6 M in the light2 assay. Similarly
the 3-quinoline 38 display IC50 = 1.9 M in the binding assay. Addi-
lM in the
l
Having established the initial SAR in this series, attention fo-
cused on optimizing the best compounds from this series, and
attention focused on some specific areas. Most notable was the po-
tency of these compounds, and the majority of these compounds
still displayed activities in the hundreds of nanomolar range, with
69 being one of the most active compounds, light2 IC50 = 57 nM.
Secondly, the high lipophilicity of these compounds needed to be
addressed as these Smo antagonists displayed high plasma protein
binding and serum shift. Accordingly, the binding assay was opti-
mised so that compounds could also be evaluated in the presence
of 20% normal human serum (NHS). For example, when 6 was
tested in the Smo binding assay in the presence of 20% NHS, the
l
tion of a second nitrogen into the bicyclic scaffold also proved to be
detrimental as quinazoline (40), quinoxaline (41), and naphthyri-
dines (42–44) all showed weaker activity than the corresponding
2-quinoline (6), with the quinoxaline (41) being the most potent
of these displaying IC50 = 1.1 lM in the Smo binding assay. In
contrast, one suitable replacement for the 2-quinoline group
proved to be the corresponding N-methylbenzimidazole, and while
45 displayed slightly weaker affinity in the binding assay with
IC50 = 1.0 lM, it appeared to be equipotent with 6 in the light2 as-
say (IC50 = 430 nM). Interestingly this more polar compound, mea-
sured log D = 2.85, proved to have improved hERG profile
IC50 shifted about 10-fold, IC50 = 3.5 lM. Initial data suggested this
(IC50 = 32
inh., 3A4: 35% inh. at 10
l
M), and reduced CYP inhibition liabilities (2C9: 44%
could be improved by modulating the log D of these compounds as
the corresponding N-methylbenzimidazole (45) with measured
log D = 2.85, proved to have a reduced serum shift, with only a 5-
fold loss in activity when the binding assay was run in the presence
lM).
Knowing that the 4-[3-(quinolin-2-yl)-1,2,4-oxadiazol-5-
yl]piperidinyl ring system was one of the best core scaffolds with
which to interrogate the left hand side SAR, attention focused on
the urea fragment (Table 5). It was rapidly established that aro-
matic ureas were one of the best functionalities to introduce at that
position as simple amides were generally inactive, and only when
large lipophilic moieties like the cycloheptane group (46) were
introduced was good inhibition of the Hh-pathway seen. Similarly
alkyl groups such as cyclohexylmethyl (47) or ortho-chlorobenzyl
(48) displayed only micromolar activity. The corresponding carba-
mate analogy 50 was demonstrated to be equipotent to 7, however,
during further study it was shown that the phenolic carbamates
were unstable and rapidly cleaved. The homologated benzyl carba-
mate (51) was displayed weaker micromolar activity. With this
knowledge an in-depth study into the urea group was conducted,
where it was rapidly established that an aromatic urea was bene-
ficial for activity as the ethyl urea (52) was displayed greater than
of 20% NHS (IC50 = 1.0 and 4.5 lM respectively). Similarly, the
piperazine bearing the ortho-chloroaniline (19) displayed
only a 4-fold shift in the Smo binding assay, displaying with
IC50 = 170 and 700 nM in the presence of 2% FBS and 20% NHS
respectively. The third area for improvement was the stability of
these compounds in liver microsomes, as the majority of these
compounds displayed high turnover in rat liver microsomes 6:
Clint >200, benzimidazole (45): = 174, and cyclohexyl (69):
>300 lL/min/mgP.
Attempts to reduce the plasma protein binding of lead 6 focused
on the introduction of polar substituents on either the left- or
right-hand side of the leads structures, as demonstrated from the
previous work that neither the quinoline or phenyl groups could
be altered to more polar heterocycles. Addition of a polar nitrile,
or methylsulfone on the urea fragment generally resulted in a loss
of potency (Table 6), however, the introduction of the para-meth-
ylsulfone in 71 did result in result in a reduction of serum shift to
2.5 lM activity in the Smo binding assay, whereas a simple phenyl
urea (53) displayed IC50 = 480 nM. It was also established that pri-
mary ureas were preferred over secondary ureas, as methylation of
the NH of 7 to give 56 resulted in a 8-fold loss in activity (Smo bind
IC50 = 180 vs. 1400 nM), similarly when the tolyl derivative 57 was
cyclised to the corresponding tetrahydroquinoline urea (58) a sub-
stantial loss in activity was observed in the functional assay (Light2
IC50 = 440 vs. 2300 nM).
only 3-fold, with Smo Bind IC50 = 1.5 and 4.1 lM respectively and
log D = 2.8. Alternatively, addition of a methoxy group at various
positions of the quinoline was also attempted, and resulted in a
substantial loss of activity (similar results were seen with other
substituents, data not shown). The exception to this was the intro-
duction of groups at the 5- or 6-position, which was tolerated to a
limited extent, although frequently resulting in a 5-fold loss in
potency.
Substituents were tolerated on the aniline ring at the ortho- and
meta- positions, although introduction of a para-methoxy (55)
group resulted in a loss of activity (68% inh at 4
l
M). A wide range
Attention then switched to the cyclohexyl urea moiety in an ef-
fort to address primarily potency and stability, and to a lesser ex-
tent the serum shift (Table 7). Initial data revealed that the
introduction of polar functionality in this region of the molecule
was detrimental for affinity, as the introduction of a hydroxyl
group (81), or alteration of the cyclohexyl to either a pyran (82)
or a functionalised piperidine (83–85) resulted in a loss in potency.
Interestingly, both the piperidine (83) and the N-acetyl piperidine
(85) display minimal serum shift. A significant breakthrough came
with the discover of the thiopyran dioxide (87) and 4,4’-dif-
luorocyclohexyl (88) derivatives, for although these two com-
pounds showed inferior activity to 69, they both displayed
of substituents could be accommodated at the ortho-position
without having a significant impact on the affinity of these Smo
antagonists, including methoxy (7), chloro (6), methyl (57), trifluo-
romethyl (60), isopropyl (8) and nitrile (61) groups. However, for-
mation of a ring between the ortho and meta-positions, as in
quinoline (62) was found to cause around a 3-fold loss in affinity.
Similarly introduction of heteroatoms into the aromatic ring
proved to be detrimental as the corresponding pyridine derivatives
of 6, compounds 63–65, all displayed between 3- and 6-fold weak-
er affinity than the chloroaniline, and similarly the isoxazole (66)
displayed only micromolar affinity.

5278
J. M. Ontoria et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5274–5282
Table 5
Exploring the SAR of the right-hand side urea and related groups
N
O
N
N
R
N
R
Compound
Light2 IC50 (nM)
840
SmoBind 2% FBS IC50 (nM)
1000
46
O
47
48
>2.5
>2.5
l
l
M
M
Cl
Cl
49
50
51
>2.5
110
>2.5
l
M
O
O
OMe
O
O
250
>2.5
lM
l
l
M
M
O
O
O
H
N
52
53
>2.5
480
H
N
OMe
H
N
7
410
410
180
220
>2.5
O
O
O
H
N
OMe
OMe
54
55
H
N
l
M
OMe
N
56
57
58
3100
440
1400
510
O
O
H
N
N
2300
670
O
Cl
H
N
6
510
380
470
340
350
O
O
H
N
59
60
8
Cl
CF3
H
N
700
480
O
O
H
N

J. M. Ontoria et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5274–5282
5279
Table 5 (continued)
R
Compound
Light2 IC50 (nM)
190
SmoBind 2% FBS IC50 (nM)
CN
H
N
O
61
490
N
H
N
62
63
1400
1600
1300
1500
O
Cl
H
N
N
O
Cl
H
N
64
2000
1100
O
N
Cl
H
N
65
66
2800
3000
>2.5 lM
O
O
N
H
N
1500
N
O
H
N
67
68
410
630
750
420
O
O
H
N
H
N
69
70
57
280
280
O
O
O
580
Table 6
Exploring the SAR in an effort to affect the serum shift
R
6
5
4
Cl
3
7
HN
R'
N
O
N
8
N
N
O
R
R’
Compound
SmoBind 2% FBS IC50(nM)
SmoBind 20% NHS IC50(nM)
H
H
H
H
H
6
380
3500
4100
4-SO2Me
5-SO2Me
5-CN
4-CN
H
H
H
H
H
71
72
73
74
75
76
77
78
79
80
1500
1200
320
>2.5
>2.5
l
l
M
M
H
1900
3-MeO
4-MeO
5-MeO
6-MeO
7-MeO
8-MeO
>2.5
>2.5
>2.5
1800
>2.5
>2.5
l
l
l
M
M
M
>2.5
>2.5
>2.5
l
l
l
M
M
M
lM
>2.5
l
M
H
lM
minimal serum shift, and 87 showed essentially no turn over in rat
liver microsomes (Clint = 8 L/min/mgP) and in rats the compound
showed modest clearance Cl = 40 mL/min/kg and a terminal half
life of 0.9 h. Analysis of the urine showed that the parent was the
only species excreted.
Introduction of the cyclohexyl or thiopyran dioxide fragments
into the benzimidazole core resulted in compounds with similar
in vitro properties to the corresponding quinolines 69 and 87,
and both benzimidazoles (89 and 90) displayed around one micro-
molar affinity in the Smo binding assay in the presence of 20% NHS,
IC50 = 1.1 and 0.76
ies, the cyclohexyl derivative in the benzimidazole series (89) dis-
played high turnover in rat microsomes (Clint = 192 L/min/mgP)
and relatively high plasma protein binding (rat fu = 2%). In rats
the compound displayed moderate clearance (Cl = 26 mL/min/kg),
T1/2 = 1.3 h and good oral bioavailability (F = 59%).
A substantial leap in potency was seen with the replacement
of the piperidinyl central ring by the corresponding piperazine,
and unlike the quinoline system where a 3-fold gain in potency
was observed going from 6 to 19, at least a 20-fold improvement
lM respectively. Similarly to the quinoline ser-
l
l

5280
J. M. Ontoria et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5274–5282
Table 7
Combining the optimal substituents
A
B
HN
O
R
N
O
N
X
N
N
A-B
X
Compound
Light2 IC50 (nM)
57
Smo Bind 2% FBS IC50 (nM)
Smo Bind 20% NHS IC50 (nM)
1600
CH@CH
CH
69
280
CH@CH
CH
81
2700
2500
65% inh. at 25 lM
OH
CH@CH
CH@CH
CH
82
83
480
1700
2900
8600
3700
O
CHs
2500
NH
N
CH@CH
CH@CH
CH
CH
84
85
3200
500
3400
2300
Me
N
1300
O
O
CH@CH
CH@CH
CH
CH
86
87
850
360
1800
860
4800
1600
S
S
O
O
F
CH@CH
NMe
CH
CH
CH
N
88
89
90
91
92
93
94
480
350
690
3
1500
1100
760
33
F
89
350
5
O
NMe
S
O
NMe
F
NMe
N
7
25
F
F
CH@CH
CH@CH
N
53
5
45
13
180
39
F
N
in potency with the benzimidazoles 89 and 91. The simple cyclo-
hexyl urea improved in the light2 assay which was run in the pres-
ence of minimal serum from 89 nM for 89 to 5 nM for 91. The
addition of the piperazine also had a beneficial effect in reducing
the lipophilicity (log D = 2.57) of that scaffold, which resulted in a
lower log D = 2.76 resulted in only a 4-fold serum shift with
IC50 = 180 nM in 20% NHS, one of the most potent compounds
identified in this series. Room for improvement remained with
the pharmacokinetics as this oxadiazole displayed relatively high
turnover in liver microsomes (Clint: rat = 189, dog = 54 and hu-
leap in activity from 1.1
l
M to 33 nM when this pair of compounds
man = 60 lL/min/mgP), and moderate-high clearance in rats
was tested in the Smo binding assay in 20% NHS. This piperazine
scaffold also showed modestly improved stability in rat microsomes
(Clint = 119 lL/min/mgP), but in vivo in rats no dramatic change was
seen (Cl = 26 mL/min/kg) and due to poor solubility of this particular
compound the oral bioavailability was low (F = 10%).
The combination of the piperazine and the 2-quinoline yielded
some very interesting derivatives suitable for further characterisa-
tion, with excellent levels of potency, low serum shift and favour-
able PK parameters in preclinical species. The simple cyclohexyl
(93) showed 5-fold improved potency in the Smo binding assay
in the presence of 2% FBS (IC50 = 45 nM) compared to 69, and the
(Cl = 54 mL/min/kg) with T1/2 = 1.1 h and modest oral bioavailabil-
ity (F = 24%). In dogs moderate clearance was observed (Cl = 16 mL/
min/kg) and T1/2 = 1.4 h. The corresponding 4,4’-difluorocyclohexyl
(94) displayed improved activity with IC50 = 5 nM in the light 2 as-
say, and similar activity in the Smo binding assay in low serum
IC50 = 13 nM. Similar to the unsubstituted cyclohexyl (93), this
compound showed only a 3-fold serum shift with IC50 = 39 nM in
20% NHS (log D = 2.67). The addition of the two fluorine atoms
on the cyclohexyl ring stabilised the molecule to oxidative metab-
olism and it displayed improved stability in microsomes (Clint
:
rat = 61, and human = 8 L/min/mgP). This corresponded to
l

J. M. Ontoria et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5274–5282
5281
R'
R'
R'
R"
i
ii, iii
R"
R"
NH
OH
HO2C
NBoc
( )n
HN R
O
N
O
N
O
N
N
N
NBoc
( )n
N
HN
N
N
v
( )n
95
96
97
98
HN R
N
HO2C
O
Het
101
iv
NH
HN R
N
Het CN
N
Het
N
O
O
HN OH
99
100
102
ii, iii
N
N
N
N
iv
i, vi
CN
HN
R
NH
N
O
N
N
O
N
N
N
H
Me
H
Me
N
NBoc
HN
N
N
N
OH
O
103
104
105
106
NBoc
x
vii
NH
N
N
O
N
N
O
N
O
N
N
N
CCl3
N
NBoc
HN
N
N
O
N
N
OH
viii
107
108
110
111
112
x
ii, iii
ix
H
N
HN R
O
N
N
N
N
N
N
O
Cl
N
O
N
O
O
109
113
N
N
N
N
x, vi
vii
ii, iii
HN
R
N
NH
N
H
N
O
N
N
O
N
N
H
Me
Me
CCl3
N
NBoc
N
N
HN
N
N
N
104
114
115
116
OH
O
O
Scheme 1. Synthetic procedure for the preparation of Smo antagonists. Reagents and conditions: (i) 1.2 equiv CDI, DMF, RT, then 1.1 equiv CDI, toluene,
DCM; (iii) RCNO, DIPEA, DCE/THF or R-NH2, CDI, MeCN 50 °C or Triphosgene, DIPEA, DCM, À20 °C then R-NH2; (iv) H2NOH, NaHCO3, MeOH, ; (v) TBTU, Et3N, DCM, then
TBAF, THF; (vi) MeI, KOH, acetone; (vii) Cl3CCOCl, Cl3CCO2H, , 12 h; (viii) CDI, MeCN, , 4 h; (ix) Neat POCl3, 190 °C, uW, 1 hr; (x) Boc-piperazine, DIPEA, THF.
D, 12 h; (ii) TFA/
D
D
D
improved clearance in rats (Cl = 32 mL/min/kg) and T1/2 = 1.3 h. In
dogs moderate clearance was observed (Cl = 20 mL/min/kg) and
T1/2 = 1.1 h. The addition of the piperazine moiety also improved
the off-targets profile as neither compound displayed CYP inhibitor
the 5-chlorooxadiazole (110) in modest 30% yield. In this case chlo-
ride displacement proceeded smoothly enabling the required final
compounds (113) to be prepared. Similar chemistry was used to
synthesize the corresponding benzimidazoles (116), although in
this case the displacement of the trichloromethyl group in 114 pro-
ceeded uneventfully.
In summary, a novel series of weak Smo antagonists were iden-
tified through uHTS screening, while these compounds are a novel
class of Smo antagonists other ureido derivatives have been de-
scribed as Smo antagonists.18–20 Thorough SAR exploration of the
scaffold identified the key functionalities necessary for potency,
enabling potent nanomolar Smo antagonists like 91 and 94 to be
identified. Optimisation of lipophilicity resulted in the most ad-
vanced compounds displaying low serum shift and clean off-tar-
gets profile, while reduction of intrinsic clearance in liver
microsomes resulted in 94 showing moderate clearance in both
rats and dogs. These compounds are valuable tools with which to
further probe the biology of the Hh-pathway.
or induction at 30 lM, and IC50 >30 lM in hERG binding assays.
These compounds were prepared as outlined in Scheme 1,16
quinoline-2-yl amidoximes were condensed with various hetero-
cyclic carboxylic acids, and then cyclised to form the required
1,2,4-oxadiazoles 97 using CDI in refluxing toluene. Deprotection
and urea formation then furnished the desired Smo antagonists
(98). Similar chemistry could be applied for the formation of
1,2,4-oxadiazoles (102) bearing various aromatic heterocycles at
the 3-position, in this case coupling of the piperidine carboxylic
acid with the heterocyclic amidoxime was followed up cyclodehy-
dration using TBAF as described by Gangloff.17 Structural modifica-
tion to allow introduction of the central piperazine required
alternative chemistry, and accordingly N-hydroxyquinoline-
2-carboximidamide (107) was reacted with trichloroacetyl chlo-
ride to form the 5-trichloromethyloxadiazole (108), although
displacement with N-Boc piperazine successfully yielded the de-
sired piperazine derivative (111) the reaction was complicated by
formation of piperazine amide 112 resulting from partial hydroly-
sis of 108. Alternatively, the oxadiazol-5(4H)-one (109) could be
prepared and reaction with neat POCl3 in the microwave to give
Acknowledgments
The authors are extremely grateful to F. Fiore, M. Fonsi,
E. Monteagudo, and M. V. Orsale for performing pharmacokinetic
studies.

5282
J. M. Ontoria et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5274–5282
10. Stanton, B. Z.; Peng, L. F.; Maloof, N.; Nakai, K.; Wang, X.; Duffner, J. L.; Taveras,
References and notes
K. M.; Hyman, J. M.; Lee, S. W.; Koehler, A. N.; Chen, J. K.; Fox, J. L.; Mandinova,
A.; Schreiber, S. L. Nature Chem. Biol. 2009, 5, 154.
11. Lauth, M.; Bergström, A.; Shimokawa, T.; Toftgård, R. Proc. Natl. Acad. Sci. U.S.A.
2007, 104, 8455.
12. Hyman, J. M.; Firestone, A. J.; Heine, V. M.; Zhao, Y.; Ocasio, C. A.; Han, K.; Sun,
M.; Rack, P. G.; Sinha, S.; Wu, J. J.; Solow-Cordero, D. E.; Jiang, J.; Rowitch, D. H.;
Chen, J. K. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 14132.
13. Rifai, Y.; Arai, M. A.; Sadhu, S. K.; Ahmed, F.; Ishibashi, M. Bioorg. Med. Chem.
Lett. 2011, 21, 718.
14. Taipale, J.; Chen, J. K.; Cooper, M. K.; Wang, B.; Mann, R. K.; Milenkovic, L.;
Scotts, M. P.; Beachy, P. A. Nature 2000, 406, 1005; Frank-Kamenetsky, M.;
Zhang, X. M.; Bottega, S.; Guicherit, O.; Wichterle, H.; Dudek, H.; Bumcrot, D.;
Wang, F. Y.; Jones, S.; Shulok, J.; Rubin, L. L.; Porter, J. A. J. Biol. 2002, 1, 10.
15. Chen, J. K.; Taipale, J.; Cooper, M. K.; Beachy, P. A. Genes Dev. 2002, 16, 2743.
16. Dessole, G.; Jones, P.; Llauger Bufi, L.; Muraglia, E.; Ontoria Ontoria, J. M.;
Torrisi, C. WO2010/013037, 2010.
17. Gangloff, A. R.; Litvak, J.; Shelton, E. J.; Sperandio, D.; Wang, V. R.; Rice, K. D.
Tetrahedron Lett. 2001, 42, 1441.
18. Brunton, S. A.; Stibbard, J. H. A.; Rubin, L. L.; Kruse, L. I.; Guicherit, O. M.; Boyd,
E. A.; Price, S. J. Med. Chem. 2008, 51, 1108.
1. Rubin, L. L.; de Sauvage, F. J. Nat. Rev. Drug Disc. 2006, 5, 1026.
2. Xie, J. Curr. Oncol. Rep. 2008, 10, 107.
3. Gallinari, P.; Filocamo, G.; Jones, P.; Pazzaglia, S.; Steinkuhler, C. Exp. Opin. Drug
Discov. 2009, 4, 525.
4. Ingham, P. W.; McMahon, A. P. Genes Dev. 2001, 15, 3059.
5. Peukert, S.; Miller-Moslin, K. ChemMedChem 2010, 5, 500.
6. Robarge, K. D.; Brunton, S. A.; Castanedo, G. M.; Cui, Y.; Dina, M. S.; Goldsmith,
R.; Gould, S. E.; Guichert, O.; Gunzner, J. L.; Halladay, J.; Jia, W.; Khojasteh, C.;
Koehler, M. F. T.; Kotkow, K.; La, H.; LaLonde, R. L.; Lau, K.; Lee, L.; Marshall, D.;
Marsters, J. C.; Murray, L. J.; Qian, C.; Rubin, L. L.; Salphati, L.; Stanley, M. S.;
Stibbard, J. H. A.; Sutherlin, D. P.; Ubhayaker, S.; Wang, S.; Wong, S.; Xie, M.
Bioorg. Med. Chem. Lett. 2009, 19, 5576.
7. Von Hoff, D. D.; LoRusso, P. M.; Rudin, C. M.; Reddy, J. C.; Yauch, R. L.; Tibes, R.;
Weiss, G. J.; Borad, M. J.; Hann, C. L.; Brahmer, J. R.; Mackey, H. M.; Lum, B. L.;
Darbonne, W. C.; Marsters, J. C., Jr.; de Sauvage, F. J.; Low, J. A. New Eng. J. Med.
2009, 361, 1164.
8. Malancona, S.; Altamura, S.; Filocamo, G.; Kinzel, O.; Martin Hernando, J. I.;
Rowley, M.; Scarpelli, R.; Steinkühler, C.; Jones, P. Bioorg. Med. Chem. Lett. 2011.
doi:10.1016/j.bmcl.2011.06.024.
19. Manetti, F.; Faure, H.; Roudaut, H.; Gorojankina, T.; Traiffort, E.; Schoenfelder,
A.; Mann, A.; Solinas, A.; Taddei, M.; Ruat, M. Mol. Pharmacol. 2010, 78, 658.
20. Roudaut, H.; Traiffort, E.; Gorojankina, T.; Vincent, L.; Faure, H.; Schoenfelder,
A.; Mann, A.; Manetti, F.; Solinas, A.; Taddei, M.; Ruat, M. Mol. Pharmacol 2011,
79, 453.
9. Kinzel, O.; Alfieri, A.; Altamura, S.; Brunetti, M.; Bufali, S.; Colaceci, F.; Ferrigno,
F.; Filocamo, G.; Fonsi, M.; Gallinari, P.; Malancona, S.; Martin Hernando, J. I.;
Monteagudo, E.; Orsale, M. V.; Palumbi, M. C.; Pucci, V.; Rowley, M.; Sasso, R.;
Scarpelli, R.; Steinkühler, C.; Jones, P. Bioorg. Med. Chem. Lett. 2011.
doi:10.1016/j.bmcl.2011.06.023.