The discovery of biaryl carboxamides as novel small molecule agonists of the motilin receptor

Article abstract of DOI:10.1016/j.bmcl.2008.10.072

Optimisation of urea (5), identified from high throughput screening and subsequent array chemistry, has resulted in the identification of pyridine carboxamide (33) which is a potent motilin receptor agonist possessing favourable physicochemical and ADME profiles. Compound (33) has demonstrated prokinetic-like activity both in vitro and in vivo in the rabbit and therefore represents a promising novel small molecule motilin receptor agonist for further evaluation as a gastroprokinetic agent.

Full text of DOI:10.1016/j.bmcl.2008.10.072

Bioorganic & Medicinal Chemistry Letters 18 (2008) 6429–6436
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
The discovery of biaryl carboxamides as novel small molecule agonists
of the motilin receptor
a
c
b
Susan M. Westaway a, , Samantha L. Brown , Elizabeth Conway , Tom D. Heightman ,
Christopher N. Johnson b, Kate Lapsley b, Gregor J. Macdonald b, David T. MacPherson b, Darren J. Mitchell a,
James W. Myatt b, Jon T. Seal a, Steven J. Stanway b, Geoffrey Stemp a, Mervyn Thompson b, Paolo Celestini d,
Andrea Colombo d, Alessandra Consonni d, Stefania Gagliardi d, Mauro Riccaboni d, Silvano Ronzoni d,
Michael A. Briggs b, Kim L. Matthews b, Alexander J. Stevens a, Victoria J. Bolton a, Izzy Boyfield b,
Emma M. Jarvie b, Sharon C. Stratton b, Gareth J. Sanger a
*
a Immuno-Inflammation Centre of Excellence for Drug Discovery, GlaxoSmithKline, Medicines Research Centre, Gunnels Wood Road, Stevenage, Herts SG1 2NY, UK
b Neurosciences Centre of Excellence for Drug Discovery, GlaxoSmithKline, New Frontiers Science Park, Third Avenue, Harlow, Essex CM19 5AW, UK
c Molecular Discovery Research, GlaxoSmithKline, New Frontiers Science Park, Third Avenue, Harlow, Essex CM19 5AW, UK
d NiKem Research Srl, Via Zambeletti, 25, 20021 Baranzate (MI), Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Optimisation of urea (5), identified from high throughput screening and subsequent array chemistry, has
resulted in the identification of pyridine carboxamide (33) which is a potent motilin receptor agonist pos-
sessing favourable physicochemical and ADME profiles. Compound (33) has demonstrated prokinetic-like
activity both in vitro and in vivo in the rabbit and therefore represents a promising novel small molecule
motilin receptor agonist for further evaluation as a gastroprokinetic agent.
Received 11 September 2008
Revised 15 October 2008
Accepted 16 October 2008
Available online 19 October 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Motilin
Motilin receptor agonist
Motilin is a 22-amino acid peptide secreted by the enterochro-
maffin cells of the small intestine and is the endogenous ligand of
the 7-TM motilin receptor, previously known as GPR38.1 This
receptor is located primarily in the gastrointestinal tract on the en-
teric nerves, smooth muscle and gastric vagal nerve terminals with
the highest levels present in the stomach and duodenum.1–3 Moti-
lin is thought to act as a ‘house-keeper’, promoting migrating mo-
tor complexes which sweep the GI tract in the fasted state.4
However, it has also been shown that motilin5 and other motilin
receptor agonists6 promote gastric emptying and propulsion of
GI tract contents in an anal direction in the fed state. Consequently,
motilin receptor agonists have the potential for the treatment of
conditions where gastric and upper intestinal motility is impaired;
for example, in certain subsets of patients suffering from gastropa-
resis7 and functional dyspepsia8 and also in patients in a critical
care setting where treatment of GI stasis can result in improved
recovery times.9
and potential for cardiac side-effects, it is not suitable for chronic
use. Several groups have made extensive progress in modifying
erythromycin giving a class of motilin receptor agonists known
as the motilides.11 Despite demonstrating improved gastric empty-
ing in healthy volunteers, many of these motilides (e.g. KC11458,12
ABT-22913) have suffered difficulties in the clinic. These have var-
iously been ascribed to tachyphylaxis, dose selection/dosing regi-
men, inappropriate patient selection or potential selectivity
issues.11,14 However, two members of the motilide class, mitemci-
nal15 (GM-611) (2) (Phase II) and PF-04548043/KOS-219714c
(Phase I) are currently progressing in clinical trials. Other groups
have disclosed non-motilide agonists (3, 4)16,17 but these are gen-
erally high molecular weight compounds (>550) (Fig. 1).
At GlaxoSmithKline, we embarked on a programme towards the
discovery of small molecule motilin receptor agonists. Thus, a high
throughput screening campaign and efficient array chemistry led
to the discovery of (5)18 (Fig. 2) which possessed good activity at
the recombinant human motilin receptor (pEC50 7.3)19,20 and also
promoted nerve-mediated contractile response in isolated rabbit
gastric antrum. However, (5) also possessed high clogP and molec-
ular weight and consequently, P450 inhibition21 (3A4 IC50 0.2 and
The antibiotic erythromycin (1) (Fig. 1) is also a motilin receptor
agonist10 and is frequently used in a clinical setting for the treat-
ment of gastroparesis.7 However, due to its antibiotic activity
0.4
lM) and microsomal clearance profiles (CLi: human >50, rat
* Corresponding author. Tel.: +44 01438 763469.
E-mail address: sue.m.westaway@gsk.com (S.M. Westaway).
23 mL minÀ1 gÀ1 liver) were undesirable. In this letter we report
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.10.072

6430
S. M. Westaway et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6429–6436
Me
iPr
Me
Me
Ph
O
N
O
N
O
Me
Me
O
HO
HO
O
O
O
N
H
O
O
N
NH2
Me
HO
Me
Me
Me
NH
N
N
NH
O
O
Me
O
O
Me
H
H
HO
O
O
N
N
HO
Me
MeO
Me
N
Me
Me
Ph
O
O
Me
OH
O
O
Me
OH
O
O
O
Me
Me
Me
Me
O
O
O
N
H
O
Me OMe
Me OMe
1
2
3
4
Figure 1. Erythromycin, mitemcinal and reported non-motilide motilin receptor agonists.
MeO
MeO
H
N
H
N
H
N
H
N
Me
Me
OMe
Me
Me
N
N
N
N
Me
N
H
N
N
N
N
OPh
OPh
OPh
O
O
O
O
5
6
7
8
hMotilinR pEC50 7.3
MW 548, cLogP 7.4
hMotilinR pEC50 6.6
MW 549, cLogP 6.3
hMotilinR pEC50 6.4
MW 487, cLogP 5.0
hMotilinR pEC50 7.1
MW 457, cLogP 5.3
Figure 2. (Piperazinylmethyl)biphenylmethyl urea and amide motilin receptor agonists.
on further optimisation of urea (5) leading to the identification of
pyridine- and imidazole-2-carboxamide agonists of the motilin
receptor.
the side-chain of (9) to give (10) resulted in a slight drop in po-
tency. A variety of other linkers as exemplified by (12)–(14) also
gave good levels of agonist activity but the importance of the
optionally substituted phenyl group at the terminus of the side-
chain was demonstrated by compound (15) where R is isobutyl
such that a drop in potency was observed despite the compound
having a clogP similar to that of compound (8). Introduction of a
higher degree of polarity into the terminal ring was also detrimen-
tal to potency, for example see isoxazole (16).
We were pleased to observe that reduction in clogP also led to
an improvement in the in vitro ADME profiles for some of these
analogues, Table 1. For those compounds possessing the terminal
phenyl ring in the side-chain, phenoxyacetamide (8) showed much
improved P450 3A4 inhibition and microsomal clearance profiles
when compared to urea (5). Replacement of the oxygen linker in
(8) with a sulfone (13) gave increased potency and a further drop
in clogP although this was not accompanied by a significant
Preliminary SAR showed that the cis-2,6-dimethylpiperazine
gave enhanced agonist potency at the motilin receptor when com-
pared to an unsubstituted piperazine head group. We discovered
that the presence of this group allowed us to replace one of the
side-chains in the moderately active amides (6) and (7)18 (Fig. 2)
with a simple methyl group to give compound (8) which possessed
reduced molecular weight and ꢀ3- to 5-fold improvement in po-
tency when compared with (6) and (7). In comparison, replace-
ment of the 3-methoxyphenethyl side-chain of (5) with a methyl
group resulted in a 40-fold drop in agonist potency (data not
shown). Therefore, a variety of other amide analogues of (8) were
prepared (Table 1).
The 4-fluoro-analogue (9) showed a 2.5-fold improvement in
potency compared to (8) but removal of the oxygen linker from
Table 1
Agonist activity (FLIPR) at human motilin receptor,20 microsomal clearance and P450 inhibition profiles21 of amide derivatives (8)–(16).22
H
Me
N
Me
N
Me
N
R
O
Compound
R
MW
clogP
hMotilinR pEC50
CLi (mL minÀ1 gÀ1) human, rat
P450 3A4 IC50 (lM)
5
8
9
10
11
12
13
14
15
16
—
PhO–
4-F-PhO–
4-F-Ph–
3-Cl-Ph–
PhS–
568
457
475
459
476
473
506
534
421
446
7.4
5.3
5.6
5.2
5.7
5.8
4.3
5.0
5.3
3.3
7.3
7.1
7.5
7.3
7.1
8.5
8.0
7.7
6.6
6.0
>50, 23
4.9, 9.5
—
—
20, 17
—
6.8, 4.3
—
—
0.2, 0.4
7.0, 24
6.0, 7.3
—
1.4, 1.5
—
3, 8.9
—
—
PhSO2–
(4-MePh)SO2NH–
iPrCH2–
3-Me-5-isoxazoyl–
2.4, 2.9
40, >100

S. M. Westaway et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6429–6436
6431
Boc
N
H
N
H
N
H
N
Me
Me
Me
Me
Me
Me
Me
Me
N
H
N
N
N
(a)
(b)
(c)
CHO
CHO
Br
17
18
19
(d)
(d)
R
N
H
Me
Me
Me
N
Me
Me
N
N
R=H: (e)
Me
N
N
H
R
R=Boc : (f)
O
8-16
20a R=H
20b R=Boc
Scheme 1. Reagents and conditions: (a) 4-bromobenzaldehyde, NaBH(OAc)3, DCM, 0 °C–rt, 87%; (b) 2-formylbenzene boronic acid, 3 M Na2CO3aq, Pd(PPh3)4, 1,2-DME, 80 °C,
90%; (c) Boc anhydride, Et3N, DCM, rt, 68%; (d) MeNH2 (8 M in THF), MeOH, rt, 16 h then NaBH4, rt, 82% R = H, 69% R = Boc; (e) RCH2CO2H, PS-carbodiimide, HOBt, DCM, rt, 2–
13%; (f) RCH2CO2H, PS-carbodiimide, HOBt, DCM, rt then HCl/Et2O, DCM, rt, 25–78%.
Table 2
Agonist activity (FLIPR) at human motilin receptor,20 microsomal clearance and P450 inhibition profiles21 of amide derivatives (21)–(28).22
H
Me
N
Me
R2
N
N
R3
O
Compound
–NR2R3
clogP
hMotilinR pEC50
8.1
CLi (mL minÀ1 gÀ1) human, rat
40, 26
P450 3A4 IC50
0.9, 1.5
(lM)
Et
N
21
5.9
22
23
24
25
26
27
28
6.5
5.5
4.8
5.8
5.2
4.9
4.3
8.0
7.2
8.3
7.3
8.6
7.2
<5
29, 7
—
0.6, 0.5
8.4, 14
12, 17
5.2, 3.5
—
N
N
F
F
O
H
N
7, 20
8, 33
—
N
N
N
H
N
Cl
F
H
N
O
—
4.8, 3.6
—
N
N
F
F
H
N
—
O

6432
S. M. Westaway et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6429–6436
improvement in the in vitro ADME profile. Replacement of the ter-
minal phenyl ring with a more polar heteroaromatic such as the
isoxazole (16) resulted in a favourable ADME profile with low
microsomal clearance and P450 3A4 inhibition. However, since
the potency of this compound was much reduced, we were faced
with finding a delicate balance between the degree and position
of lipophilicity required for motilin agonist activity and the nature
and position of more polar functionality which would give accept-
able in vitro ADME properties.
The synthesis of amide analogues (8)–(16) was undertaken as
shown in Scheme 1. Reductive amination of 4-bromobenzalde-
hyde with cis-2,6-dimethylpiperazine followed by standard Suzu-
ki–Miyaura coupling with 2-formylbenzeneboronic acid gave
aldehyde (18). Further reductive amination with methylamine
gave key intermediate (20a) which was converted to products
(8), (10) and (16) by coupling with the appropriate acids in
the presence of polymer-supported carbodiimide and hydroxy-
benzotriazole. Due to the low yields obtained for the final step
in the synthesis of these three analogues, a modified route was
used for the remaining analogues. The piperazine in (18) was
Boc-protected before reductive amination with methylamine to
give (20b). Amide coupling and final Boc-deprotection then oc-
curred in improved overall yields to give compounds (9) and
(11)–(15).
Compounds (8)–(16) possessed two benzylamino- groups
which was potentially sub-optimal from a metabolic stability per-
spective. We had previously shown that the LHS benzylpiperazine
moiety was a requirement for motilin agonist activity.18 Therefore,
we proceeded to investigate replacement of the benzylic amide on
the RHS of the molecule through preparation of the reversed
amides. Compound (21),23 Table 2, showed improved agonist po-
tency at the motilin receptor compared with analogues (10) and
H
N
H
N
Me
Me
Me
Me
R2
N
OHC
CO2Me
N
N
CO2Me
(b), (c)
(a)
(d), (e) or (f)
I
O
R3
CO2H
22-28
29
30
Scheme 2. Reagents and conditions: (a) 4-formylbenzeneboronic acid, Na2CO3, Pd(OAc)2, nBu4NBr, 1,2-DME/H2O (1:1), microwave, 150 °C, 10 min, 75%; (b) cis-2,6-
dimethylpiperazine, NaBH(OAc)3, DCM, rt, 98%; (c) 2M NaOHaq, EtOH, rt, 77%; (d) compounds (22), (25) and (28): R1R2NH, EDCI.HCl, HOBt, DMF, 50 °C, 25–42%; (e)
compounds (23), (24) and (27): R1R2NH, PS-carbodiimide, HOBt, DMF, 50 °C, 6–36%; (f) compound (26): R1R2NH, EDCI.HCl, DMAP, DCM/DMF, rt, 19%.
Table 3
Agonist activity (FLIPR) at human motilin receptor,20 microsomal clearance and P450 inhibition profiles21 of amide derivatives (31)–(36).22
H
Me
N
Me
A
X
N
Y
O
N
B
Compound
Aa
Ba
X
Y
clogP
hMotilinR pEC50
<5
CLi (mL minÀ1 gÀ1) human, rat
P450 3A4 IC50
—
(lM)
N
31
Ph
NH
4-F
4.0
—
S
32
33
34
35
Ph
NH
NH
NH
NH
CH2
4-F
4-F
3-F
3-F
4-F
3.8
4.1
3.6
4.3
5.4
6.4
7.7
7.8
5.6
8.0
—
—
N
Ph
Ph
Ph
Ph
2.6, 2.8
3.4, 4.1
—
>100, >100
58, 46
—
N
N
N
N
36
16, 16
13, 5.6
N
a
When A is phenyl the regiochemistry is 1,4. When B is phenyl the regiochemistry is 1,2.

S. M. Westaway et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6429–6436
6433
Table 4
Agonist activity (FLIPR) at human motilin receptor20 and microsomal clearance data for amide derivatives (37)–(47).22
H
Me
N
Me
H
N
2
3
4
N
Y
O
N
N
X
2
3
Compound
X
Y
clogP
hMotilinR pEC50
CLi (h, r) (mL minÀ1 gÀ1
)
33
37
38
39
40
41
42
43
44
45
46
47
H
H
H
H
H
H
H
H
4-F
3-F
2-F
4-CN
3-CN
2-CN
4-CONH2
3,4-diF
4-OMe
4-F
4.1
4.1
4.1
3.8
3.8
3.8
2.5
4.3
3.8
4.3
4.3
4.1
7.7
8.2
7.3
6.2
7.7
6.1
5.6
8.7
6.0
8.4
8.0
8.3
2.6, 2.8
4.4, 6.5
—
—
3.5, 4.6
—
—
7.0, 5.9
—
5.5, 6.5
9.4, 12.0
5.3, 5.9
H
2-F
2-F
2-OMe
3-F
4-F
h, human; r, rat.
Table 5
Further profiling of amides (33) and (34).
Property
33
34
P450 IC50
(l
M)
1A2, 2C9, 2C19 >100; 2D6 15
1A2 74; 2C9, 2C19 >100; 2D6 29
hERG binding pIC50
hGhrelinR pEC50
MW
LogD (pH 7.4)
Aq solubility (HCl salt) (mg/mL)
<4.5
7.0
501
0.72
P1
<4.6
6.3
490
0.82
P0.5
800
600
400
200
0
Erythromycin (1)
Cpd (33)
Cpd (34)
Table 6
Prokinetic-like activity of erythromycin (1), compounds (33) and (34) on isolated
rabbit gastric antrum.
Compound
Concentration (
l
M)
Potentiation (%)
Na
1
10
3
1
0.3
10
3
1
0.3
3
1.5
0.3
368 74
490 117
189 71
69 32
486 151
343 120
299 33
168 63
224 45
302 60
73 12
5
5
5
5
3
3
3
3
4
4
3
33
34
10-8
10-7
10-6
10-5
[Drug] M
a
N, number of strips of gastric antrum tissue used per test conc.
Figure 3. Prokinetic-like activity of erythromycin (1), compounds (33) and (34) on
isolated rabbit gastric antrum.
of the benzylic terminal group with a phenoxy group (23) gave a
10-fold reduction in overall lipophilicity together with a 6-fold
reduction in potency but we were pleased to observe that P450
3A4 inhibition was also reduced by >10-fold. A further reduction
in overall lipophilicity was achieved by replacement of the ether
linker with an amino linker as in (24). This compound showed im-
proved agonist potency and a similar P450 3A4 inhibition profile to
(23) but microsomal clearance was still too high. We were some-
what surprised to observe that the 4-chloro analogue (25) was
10-fold less potent than the unsubstituted analogue (24) despite
being 10-fold more lipophilic and that the microsomal clearance
was not significantly affected by the presence of this substituent.
In the limited set of compounds prepared in this series, 4-fluoro
substitution (26) gave the best potency with pEC50 8.6 and this
(11). However, removal of the RHS benzylic centre offered no
improvement in microsomal clearance levels despite (21) possess-
ing a similar overall level of lipophilicity to (11).
Conformational constraint of the RHS of this series of reversed
amides was also examined through preparation of 4-substituted
piperidine amide derivatives (22)–(28) as shown in Scheme 2.
The methylene linked compound (22) showed an overall profile
similar to that of non-constrained analogue (21) with excellent po-
tency at the motilin receptor (pEC50 8.0), high clogP, high micro-
somal clearance and significant P450 3A4 inhibition, Table 2.
However, it was apparent that the nature of the linking atom be-
tween the piperidine and the terminal phenyl ring had a marked
effect on both potency and in vitro ADME properties. Replacement

6434
S. M. Westaway et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6429–6436
H
N
H
N
O
H
N
O
N
N
(d)
O
N
N
(a), (b)
(c)
F
F
HN
HO
TfO
N
Boc
F
48
49
Me
50
H
N
Me
H
H
N
N
N
OHC
O
N
N
N
N
O
(e)
(f)
F
F
51
33
Scheme 3. Reagents and conditions: (a) (i) 4-fluoroaniline, AcOH, 1,2-DCE, rt, (ii) NaBH(OAc)3, 1,2-DCE, rt, 96%; (b) 2 M HCl, 1,4-dioxane, 60 °C, quant; (c) (i) 3-
hydroxypyridine-2-carboxylic acid, iBuOCOCl, Et3N, DCM, (ii) (48), iii—2 M NaOHaq
94%; (d) N-phenyl bis(trifluoromethylsulfonimide), Et3N, DCM, quant; (e) 4-
formylbenzene boronic acid, Na2CO3, Pd(PPh3)4, 1,2-DME, H2O, 50 °C, quant.; (f)cis-2,6-dimethylpiperazine, NaBH(OAc)3, 1,2-DCE, rt, 63%.
,
F
F
F
H
N
F
Me
Me
O
NH
N
NH
NH
O
OHC
O
HN
(d)
(c)
N
(a), (b)
(e)
N
O
N
N
N
Boc
N
N
N
N
HN
N
H
52
53
54
34
Scheme 4. Reagents and conditions: (a) (i) 3-fluoroaniline, AcOH, 1,2-DCE, rt, (ii) NaBH(OAc)3, 1,2-DCE, rt, 96%; (b) 2 M HCl, 1,4-dioxane, 60 °C, quant; (c) ethyl imidazole-2-
carboxylate, Me3Al, PhMe, rt, 54%; (d) 4-formylbenzene boronic acid [Cu(OH).TMEDA]2Cl2, O2, 1,2-DCE, 50 °C, 87%; (e)cis-2,6-dimethylpiperazine, AcOH, 1,2-DCE, 60 °C, then
NaBH(OAc)3, 1,2-DCE, rt, 52%.
group together with the 3-fluoro analogue were used in further
investigations.
Table 5. In addition, these compounds showed >100-fold selectiv-
ity against a variety of receptor, ion channel and enzyme targets
except for the closely related ghrelin receptor24 where (33) showed
5-fold selectivity (pEC50 7.0) and (34) showed 30-fold selectivity
(pEC50 6.3).
The in vivo pharmacokinetic properties of (33) and (34) were
assessed. In the rat, pyridine (33) was a moderate clearance com-
pound (CLb 53 mL minÀ1 kgÀ1) with 13% oral bioavailability (Fpo).
However, on dosing to the dog a much improved profile was ob-
served, particularly with regard to oral absorption; (33) still
showed moderate clearance (CLb 19 mL minÀ1 kgÀ1) but its oral
bioavailability was 58% in this species. Imidazole (34) showed a
similar profile to (33) in the rat with CLb 61 mL minÀ1 kgÀ1 and
Fpo 17%.
The gastric prokinetic-like activity of (33) and (34) was also as-
sessed whereby the compounds were tested for their effects on
contractions evoked by electrical-field-stimulation (EFS) of iso-
lated rabbit gastric antrum tissue.25 This type of stimulation results
in nerve-mediated contractions of the muscle and both motilin and
erythromycin have demonstrated a robust ability to facilitate the
amplitude of contractions in this assay. Thus, application of a test
compound results in potentiation of the contractions because of
its ability to activate the motilin receptors located on cholinergic
neurons in this tissue. Figure 3 and Table 6 show the effects of
(33) and (34), compared to erythromycin, when applied to the tis-
sue at a range of concentrations.
Modification of the (phenylamino)piperidine amide series tar-
geting further improvement of the in vitro ADME profile was car-
ried out by replacing each of the core phenyl rings with a
selection of heteroaromatic or heterocyclic rings as exemplified
in Table 3. Replacement of the phenyl ‘A-ring’ with 3-pyridyl gave
an inactive compound (31); however, the thiazole analogue (32)
was better tolerated although potency was still lower than we re-
quired. Attention was then turned to analogues where the ‘B-ring’
phenyl was modified. Replacement of the phenyl ‘B-ring’ with pyr-
idine (33) resulted in a 10-fold drop in lipophilicity and we were
pleased to observe that this change was not only tolerated with re-
spect to agonist potency at the motilin receptor (pEC50 7.7), but
also gave a much improved P450 3A4 inhibition and microsomal
clearance profiles. The analogous imidazole (34) showed a similar
level of potency but the pyrrolidine analogue (35) was consider-
ably less potent despite possessing a similar overall level of lipo-
philicity. The beneficial effect of combining the amino linker with
the pyridyl ‘B-ring’ is also illustrated by comparison of (33) with
the benzyl analogue (36). The 20-fold increase in lipophilicity gave
only a marginal increase in potency but was significantly detri-
mental to the in vitro ADME profile.
The SAR around pyridyl compound (33) was further explored as
shown in Table 4. It was found that various alternative substitution
patterns resulted in higher potency at the motilin receptor, for
example compounds (37), (43) and (45)–(47). The clogP values
for these compounds were either the same as, or not significantly
higher than for (33), but the addition of extra substituents in the
2-position of the phenyl ‘A-ring’ or the terminal phenyl ring was
detrimental for in vitro metabolic stability. Therefore, (33) and
the imidazole (34) were selected for further profiling as shown in
These data show that in the range of 0.3–10
possessed a profile (potentiation 486 151% at 10 uM) comparable
to that of erythromycin (Emax 490 117% at 3 M). At a higher con-
centration of 30 M, compound (33) showed a still greater poten-
lM, pyridine (33)
l
l
tiation with Emax 1260 60% (n = 4) (data not shown in graph for
clarity). Whilst the imidazole (34) was somewhat less efficacious,

S. M. Westaway et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6429–6436
11. Peeters, T. L. Neurogastroenterol. Motil. 2006, 18, 1.
6435
its maximal effect (Emax 302 60% at 1.5 lM) was still similar to
12. (a) Sann, H. Functional Disorders of the Gastrointestinal Tract, Krause, G., Ed.; IOS
Press, 2005.; (b) Russo, A.; Stevens, J. E.; Giles, N.; Krause, G.; O’Donovan, D. G.;
Horowitz, M.; Jones, K. L. Aliment. Pharmacol. Ther. 2004, 20, 333.
13. (a) Verhagen, M. A. M. T.; Samson, M.; Maes, B.; Geypens, B. J.; Ghoos, Y. F.;
Smout, A. J. P. M. Aliment. Pharmacol. Ther. 1997, 11, 1077; (b) Talley, N. J.;
Verlinden, M.; Snape, W.; Beker, J. A.; Ducrotte, P.; Dettmer, A.; Brinkhoff, H.;
Eaker, E.; Ohning, G.; Miner, P. B.; Mathias, J. R.; Fumagalli, I.; Staessen, D.;
Mack, R. J. Aliment. Pharmacol. Ther. 2000, 14, 1653; (c) Talley, N. J.; Verlinden,
M.; Geenen, D. J.; Hogan, R. B.; Riff, D.; McCallum, R. W.; Mack, R. J. Gut 2001,
49, 395.
14. (a) Tack, J.; Peeters, T. Gut 2001, 49, 317; (b) Poitras, P.; Peeters, T. L. Curr. Opin.
Endocrinol. Diabetes Obes. 2008, 15, 54; (c) Sanger, G. J.; Alpers, D. H.
Neurogastroenterol. Motil. 2008, 30, 177; (d) Sanger, G. J. Drug Disc. Today
2008, 13, 234.
15. Mccallum, R. W.; Cynshi, O. Aliment. Pharmacol. Ther. 2007, 26, 107. See also
company website http://www.roche.com.
16. Li, J. J.; Chao, H.-G.; Wang, H.; Tino, J. A.; Lawrence, R. M.; Ewing, W. R.; Ma, Z.;
Yan, M.; Slusarchyk, D.; Seethala, R.; Sun, H.; Li, D.; Burford, N. T.; Stoffel, R. H.;
Salyan, M. E.; Li, C. Y.; Witkus, M.; Zhao, N.; Rich, A.; Gordon, D. A. J. Med. Chem.
2004, 47, 1704.
17. Jasserand, D.; Antel, J.; Preuschoff, U.; Brueckner, R.; Sann, H.; Wurl, M.;
Eickelmann, P. PCT Int. Appl. WO 2002092592, 2002; Can 137:384755.
18. Heightman, T. D.; Conway, E.; Corbett, D. F.; Macdonald, G. J.; Stemp, G.;
Westaway, S. M.; Celestini, P.; Gagliardi, S.; Riccaboni, M.; Ronzoni, S.; Vaidya,
K.; Butler, S.; McKay, F.; Muir, A.; Powney, B.; Winborn, K.; Wise, A.; Jarvie, E.
M.; Sanger, G. J. Bioorg. Med. Chem. Lett. 2008, 18, 6423.
the effect of (33) at a similar concentration. Although (33) showed
a sub-optimal level of selectivity at the ghrelin receptor, it has pre-
viously been demonstrated that ghrelin does not show any con-
tractile activity on rabbit gastric antrum under EFS conditions.26
Hence, the prokinetic-like activity observed with (33) in this assay
may be attributed to its motilin receptor agonist activity.
Compounds (33) and (34) were prepared according to Schemes
3 and 4, respectively.27 The RHS 4-arylaminopiperidines were syn-
thesised via standard reductive amination and deprotection meth-
ods. For compound (33), piperidine (48) was coupled with 3-
hydroxypicolinic acid using isobutyl chloroformate to activate
the acid. A basic hydrolysis was also required after the coupling
to hydrolyse the isobutyl ester formed in the initial step back to
the free hydroxy compound (49). Activation of the hydroxyl as
the triflate (50) was followed by standard Suzuki–Miyaura cou-
pling to give the biaryl core (51) in quantitative yield. Finally, a fur-
ther reductive amination step with cis-2,6-dimethylpiperazine
proceeded regioselectively to give compound (33). The imidazole
(34) was prepared using a similar strategy whereby the piperidine
(52) was coupled with ethyl imidazole-2-carboxylate in the pres-
ence of trimethylaluminium to give (53) in good yield. Reaction
of imidazole (53) with 4-formylbenzeneboronic acid utilising a
copper-mediated oxidative coupling28 resulted in the formation
of the biaryl core (54) in excellent yield. As for compound (33), a
final reductive amination step furnished the target compound (34).
In summary, starting from urea lead (5) a strategy involving
modulation of lipophilicity in such a way as to maintain activity
at the motilin receptor resulted in the identification of amides pos-
19. Potency data quoted in this paper was obtained using a FLIPR assay format
with recombinant human motilin receptor stably expressed in a HEK293 cell
line20
whereas data quoted in the previous publication (Ref. 18) used a CHO
cell line. pEC50 values represent the mean from at least three independent
experiments with SEM 6 0.2 (SD 6 0.3) in all cases except compound (25)
where SEM = 0.2 (SD = 0.4, n = 5) and compounds (8), (10), (24), (31), (35), (44)
and (47), where n = 2. Comparison of activities of standard compounds in these
assay formats:
Cell line
Human motilin
Erythromycin
Compound
pEC50 (n)
pEC50 (n)
5 pEC50 (n)
sessing
a key heteroaryl-containing core and 4-(phenylami-
HEK293
CHO
9.3 (77)
10.4 (770)
6.2 (7)
7.3 (4)
7.3 (17)
8.0 (80)
no)piperidine RHS. In particular, compounds (33) and (34)
combined good agonist activity at the motilin receptor with
favourable in vitro DMPK and physicochemical profiles. Pyridyl
analogue (33) also showed a highly promising level of prokinetic-
like activity in isolated rabbit gastric antrum and this, in combina-
tion with its in vivo pharmacokinetic profile, has resulted in further
assessment of the compound in an in vivo rabbit model of whole
gut transit. In this model, (33) showed a statistically significant in-
crease in both the number and weight of faecal pellets produced by
the rabbits at 2 h after a 3 mg/kg iv bolus dose of the compound.29
In conclusion, compound (33), GSK326416,30 represents an
exciting novel small molecule motilin receptor agonist for further
evaluation as a gastric prokinetic agent.
20. FLIPR assay protocol: HEK-293 cells stably expressing the human motilin
receptor were seeded (30,000 cells/100 L growth media/well) into poly-
lysine coated 96-well black-wall, clear-bottom microtitre plates (Corning)
24 h prior to assay. On the day of assay the cells were loaded with 2
l
D-
lV
(final) Fluo-4-AM fluorescent indicator dye (Molecular Probes) and 1 mM
(final) probenicid in assay buffer (145 mM sodium chloride, 2.5 mM
potassium chloride, 10 mM Hepes, 10 mM glucose, 1.2 mM magnesium
chloride, 1.5 mM calcium chloride and 0.1% BSA) (50
added to each well). Plates were incubated for 1 h at 25 °C, before being
washed four times with 100 L assay buffer using the EMBLA cell washer;
150 L residual being left after the final wash. The cells were then incubated
lL loading solution
l
l
at 25 °C for 20 min and the plates were then assayed on a Fluorometric
Imaging Plate Reader (FLIPR, Molecular Devices). Test compounds were
prepared in assay buffer without probenecid. In the FLIPR, 50
compound was added to the cells and changes in fluorescence measured
over 2-min timeframe. Maximum change in fluorescence over baseline
was used to determine agonist response and concentration response curves
were constructed, using 4-parameter logistic equation. All compounds
lL of test
References and notes
a
1. Feighner, S. D.; Tan, C. P.; McKee, K. K.; Palyha, O. C.; Hreniuk, D. L.; Pong, S.-S.;
Austin, C. P.; Figueroa, D.; MacNeil, D.; Cascieri, M. A.; Nargund, R.; Bakshi, R.;
Abramovitz, M.; Stocco, R.; Kargman, S.; O’Neill, G.; Van Der Ploeg, L. H. T.;
Evans, J.; Patchett, A. A.; Smith, R. G.; Howard, A. D. Science 1999, 284, 2184.
2. Xu, L.; Depoortere, I.; Tomasetto, C.; Zandecki, M.; Tang, M.; Timmermans, J.-P.;
Peeters, T. L. Reg. Peptides 2005, 124, 119.
3. Takeshita, E.; Matsuura, B.; Dong, M.; Miller, L. J.; Matsui, H.; Onji, M. J.
Gastroenterol. 2006, 41, 223.
4. (a) Itoh, Z. Peptides 1997, 18, 593; (b) Sanger, G. J.; Lee, K. Nat. Rev. Drug Disc.
2008, 7, 241.
5. (a) Christofides, N. D.; Modlin, I. M.; Fitzpatrick, M. L.; Bloom, S. R.
Gastroenterology 1979, 76, 903; (b) Peeters, T. L.; Muls, E.; Janssens, J.; Urbain,
J. L.; Bex, M.; Van Cutsem, E.; Depoortere, I.; De Roo, M.; Vantrappen, G.;
Bouillon, R. Gastroenterology 1992, 102, 97.
6. Mantides, A.; Xynos, E.; Chrysos, E.; Georgopoulos, N.; Vassilakis, J. S. Am. J.
Gastroenterol. 1993, 88, 198.
7. Abell, T. L.; Bernstein, R. K.; Cutts, T.; Farrugia, G.; Forster, J.; Hasler, W. L.;
McCallum, R. W.; Olden, K. W.; Parkman, H. P.; Parrish, C. R.; Pasricha, P. J.;
Prather, C. M.; Soffer, E. E.; Twillman, R.; Vinik, A. I. Neurogastroenterol. Motil.
2006, 18, 263.
8. Hiyama, T.; Yoshihara, M.; Matsuo, K.; Kusunoki, H.; Kamada, T.; Ito, M.;
Tanaka, S.; Nishi, N.; Chayama, K.; Haruma, K. J. Gastroenterol. Hepatol. 2007, 22,
304.
a
were full agonists in this assay.
21. P450 data determined using Gentest enzyme source/assay format. Two IC50
values are quoted for 3A4 corresponding to inhibition of turnover of two
substrates, DEF (Diethoxyfluorescein) and PPR (Phenylpiperazinylmethy-
lbenzylresofurin), respectively.
22. All new compounds gave satisfactory 1H NMR and mass spectral data.
23. Compound (21) was prepared in five steps as follows: (i) phenethylamine,
acetyl chloride, triethylamine, DCM, rt, quant; (ii) LiAlH4, THF, 50 °C, 98%; (iii)
2-IC6H4 COCl, DIPEA, DCM, rt, 76%; (iv) 4-(OHC)C6H4B(OH)2, Pd(PPh3)4, 3 M
Na2CO3aq, DME, reflux, 46%; (v) cis-2,6-dimethylpiperazine, DCM, rt, 1 h then
NaBH(OAc)3, rt, 72%.
24. Witherington, J.; Abberley, L.; Briggs, M. A.; Collis, K.; Dean, D. K.; Gaiba, A.;
King, N. P.; Kraus, H.; Shuker, N.; Steadman, J. G. A.; Takle, A. K.; Sanger, G.;
Wadsworth, G.; Butler, S.; McKay, F.; Muir, A.; Winborn, K.; Heightman, T. D.
Bioorg. Med. Chem. Lett. 2008, 18, 2203.
25. Jarvie, E. M.; North Laidler, V. J.; Corcoran, S.; Bassil, A.; Sanger, G. J. Brit. J.
Pharmacol. 2007, 150, 455.
26. (a) Dass, N. B.; Hill, J.; Muir, A.; Testa, T.; Wise, A.; Sanger, G. J. Brit. J. Pharmacol.
2003, 140, 948; (b) Depoortere, I.; Thijs, T.; Theelemans, L.; Robberecht, P.;
Peters, T. L. J. Pharmacol. Exp. Ther. 2003, 305(2), 660.
27. MacDonald, G. J.; Stanway, S. J.; Thompson, M.; Westaway, S. M. PCT Int. Appl.
WO 2007007018, 2007; Can 146:163136.
9. Chapman, M. J.; Nguyen, N. Q.; Fraser, R. J. L. Curr. Opin. Crit. Care 2007, 13, 187.
10. Peeters, T.; Matthijs, G.; Depoortere, I.; Cachet, T.; Hoogmartens, J.;
Vantrappen, G. Am. J. Physiol. 1989, 257, G470.
28. Collman, J. P.; Zhong, M. Org. Lett. 2000, 2, 1233.

6436
S. M. Westaway et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6429–6436
29. Unsworth, W.; Smith, J. E.; Campbell, C. A.; Bate, S. T.; Matthews, K. L.; Stevens,
A. J.; McNulty, S.; Lee, K.; Westaway, S. M.; Stratton, S. C.; O’Shaughnessy, C. T.
Gastroenterology 2006, 130(4 Suppl. 2) (Abstr. T2040).
(6H, m), 3.20–3.25 (3H, m), 3.56 (2H, s), 4.51 (1H, m), 6.45 (2H, m), 6.84 (2H, t, J
8.8 Hz), 7.40 (3H, m), 7.46 (2H, d, J 6.5 Hz), 7.77 (1H, dd, J 8.0, 1.5 Hz), 8.62 (1H,
dd, J 4.8, 1.5 Hz). MS: (ES+) 502 (MH+).
30. Characterising data for (33): 1H NMR (250 MHz, CDCl3) d (ppm): 0.72 (1H, m),
1.04 (6H, d, J 6.3 Hz), 1.15 (1H, m), 1.61–1.72 (4H, m), 1.97 (1H, br d), 2.74–3.00