Potent achiral agonists of the growth hormone secretagogue (ghrelin) receptor. Part 2: Lead optimisation

Article abstract of DOI:10.1016/j.bmcl.2007.12.021

A series of small molecule orally bioavailable ghrelin receptor agonists have been identified through systematic optimisation of a high throughput screening hit. Crown Copyright

Full text of DOI:10.1016/j.bmcl.2007.12.021

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2203–2205
Potent achiral agonists of the growth hormone
secretagogue (ghrelin) receptor. Part 2: Lead optimisation
Jason Witherington,a,* Lee Abberley,a Michael A. Briggs,a Katharine Collis,a
David K. Dean,a Alessandra Gaiba,a N. Paul King,a Helmut Kraus,a Nicola Shuker,a
Jon G. A. Steadman,a Andrew K. Takle,a Gareth Sanger,b Graham Wadsworth,b
Sharon Butler,c Fiona McKay,c Alison Muir,c Kim Winbornc and Tom D. Heightmand
aDepartment of Medicinal Chemistry & DMPK, Neurology & GI Centre of Excellence for Drug Discovery, GlaxoSmithKline,
New Frontiers Science Park, Third Avenue, Harlow, Essex CM19 5AW, UK
bDepartment of Biology, Neurology & GI Centre of Excellence for Drug Discovery, GlaxoSmithKline,
New Frontiers Science Park, Third Avenue, Harlow, Essex CM19 5AW, UK
cMolecular Discovery Research, Department of Screening and Compound Profiling, GlaxoSmithKline, New Frontiers Science Park,
Third Avenue, Harlow, Essex CM19 5AW, UK
dMolecular Discovery Research, Department of Medicinal Chemistry, GlaxoSmithKline, New Frontiers Science Park,
Third Avenue, Harlow, Essex CM19 5AW, UK
Received 8 November 2007; revised 10 December 2007; accepted 11 December 2007
Available online 15 December 2007
Abstract—A series of small molecule orally bioavailable ghrelin receptor agonists have been identified through systematic optimi-
sation of a high throughput screening hit.
Crown copyright Ó 2008 Published by Elsevier Ltd. All rights reserved.
Ghrelin, a 28-amino acid gastric hormone containing a
unique post-translational modification on serine 3,
exhibits a wide range of biological activities via its pos-
tulated cognate receptor, the growth hormone secreta-
gogue receptor (GHSR1a). In humans as well as in
rodents, ghrelin stimulates pituitary growth hormone
(GH) secretion1 and in addition increases food intake
and body weight gain and regulates energy balance.2,3
Agents which mimic the actions of ghrelin have poten-
tial not only in growth hormone replacement therapy,
but also in disorders requiring increased nutritional in-
take, such as cancer-induced cachexia and post-opera-
tive ileus, and in motility disorders such as neurogenic
and diabetic gastroparesis.4
(1).5 Whilst this represents an exciting new class of ghre-
lin receptor agonists, this series generally suffers from
poor oral exposure which may be attributed to the rela-
tively high lipophilicity and poor solubility. Herein we
report the subsequent SAR optimisation and in vivo
properties of these compounds. Previous SAR explora-
tion around the arylsulfonamide and amine moieties
had demonstrated that these parts of the molecule were
highly sensitive to modifications; hence our initial efforts
targeted core modifications that would lower the overall
lipophilicity of the compounds (Table 1). Interestingly
although removal of the indoline ring resulted in a con-
siderable reduction of potency, compound 2 did retain
appreciable potency and hence further open chain ana-
logues were prepared. Subsequent removal of the meth-
oxy group was also tolerated (cf. 2 and 3), whereas
replacement with an electron withdrawing led to a sig-
nificant reduction in potency (cf. 2 and 4). A range of
targets whereby there is the potential to form an intra-
molecular hydrogen bond with the highly acidic sulfon-
amide moiety were also targeted with the aim of
increasing the potential for good permeability and hence
oral bioavailability. Gratifyingly the ortho methoxy
In the preceding paper, we highlighted the identification
and subsequent optimisation of a series of small mole-
cule ghrelin receptor agonists, typified by SB-791016
Keywords: Ghrelin; Growth hormone; Agonist.
*
Corresponding author. Tel.: +44 1279 627832; fax: +44 1279
627685; e-mail: Jason_Witherington@GSK.com
0960-894X/$ - see front matter Crown copyright Ó 2008 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.12.021

2204
J. Witherington et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2203–2205
Table 1. Core modifications
7) or alternative hydrogen bonding substituents gener-
ally led to a reduction in potency (e.g., 9, 10, and 11).
NH
With compound 5 in hand we sought to optimise the
(het)aryl sulfonamide moiety (Table 2). The presence
of an (het)aryl substituent in the para position appears
optimal for potency (cf. 13 with 12 and 28) consistent
with our hypothesis that an appropriately placed lipo-
philic group potentially mimics the octanyl group pres-
ent on serine 3 of ghrelin which has been reported to
be critical for potency and efficacy.6 Further support
for this hypothesis comes from the finding that introduc-
tion of polarity also leads to a reduction in potency (cf.
13 with 18, 19 and 20), however functionalisation of the
terminal aryl ring can lead to further improvements in
potency (cf. 13 with 14, 15 and 16). Interestingly, intro-
duction of either a thiophene or furan group is also well
tolerated (22 and 23) and the potency can be further en-
hanced by modulation of the lipophilicity around this
part of the molecule (24, 25, and 26). Introduction of
a linker residue between the two aromatic rings gener-
ally led to a reduction in potency (29 and 30).
H
N
N
S
N
S
O
O
R1
R2
#
R1
R2
hGHSR pEC50 (IA)7
2
H
OMe
H
7.8 (0.9)
7.4 (0.8)
6.2 (0.9)
8.0 (1.1)
7.4 (0.8)
7.0 (1.1)
7.4 (1.0)
7.0 (0.8)
7.4 (0.9)
6.1 (0.9)
3
H
4
H
CN
H
5
MeO
EtO
iPrO
Cl
6
H
7
H
8
H
9
F
H
10
11
MeCO
Me2NCO
H
H
Table 2. Side chain SAR
NH
Having identified several potent ligands we sought to
determine whether they displayed improved in vivo
properties relative to the indoline sulfonamides typified
by SB-791016 (1). Encouragingly, a number of com-
pounds displayed robust performance in a rat feeding
assay8 following oral dosing and were subsequently fur-
ther profiled. Gratifyingly, compounds 22 and 24 re-
tained a good selectivity profile over the 5-HT1B
receptor and displayed encouraging pharmacokinetic
properties with low blood clearance and excellent bio-
availability (Table 3).
H
N
R
N
S
O
O
MeO
#
R1
hGHSR pEC50 (IA)
5
(2-Pyridyl)-4-thienyl
4-Br-Ph
Ph-4-Ph
8.0 (1.1)
6.9 (0.7)
8.2 (1.0)
9.4 (1.2)
8.7 (1.0)
9.3 (1.0)
8.6 (0.9)
6.5 (0.7)
6.5 (0.6)
7.4 (0.8)
7.8 (0.8)
8.4 (1.0)
8.4 (0.9)
9.1 (1.0)
9.1 (0.9)
8.7 (0.9)
7.7 (1.0)
6.8 (1.0)
7.2 (0.9)
7.0 (0.8)
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
(5-OMe,2-F-Ph)-4-Ph
(3-OMe-Ph)-4-Ph
(5-Cl,2-F-Ph)-4-Ph
(1,3-Oxazol-5-yl)-4-thienyl
(4-Pyridyl)-4-Ph
O
NH
S
O
(5-Pyrimidinyl)-4-Ph
(2-Oxo-1-pyrrolidinyl)-4-Ph
1H-Pyrazol-1-yl)-4-Ph
(2-Thienyl)-4-Ph
S
N
N
N
OMe
(2-Furanyl)-4-Ph
(5-Me-2-furanyl)-4-Ph
(2-Thienyl)-4-(2-Cl-Ph)
(2-Thienyl)-4-(2-F-Ph)
(2-Furanyl)-3-Ph
(1) SB-791016, hGHSR pEC50 (IA) 9.8 (0.9)
The furanyl analogue 24 was obtained starting with the
reaction of the commercially available bromide 31 and
piperazine 32 under cross coupling conditions to afford
the N-aryl piperazine 33. Reduction of the nitro moiety
and subsequent selective sulfonylation of the resulting
aniline, employing the commercially available 4-bromo
benzene sulfonyl chloride, afforded the sulfonamide 12.
Finally, treatment of 12 with the commericially available
Ph-3-Ph
PhO-4-Ph
BnO-4-Ph
analogue 5 demonstrated encouraging potency. At-
tempts to replace the potential metabolically labile
methoxy group with either larger alkoxy groups (6 and
Table 3. 5-HT1B selectivity and rat pharmacokinetic profiles of selected ghrelin agonists
#
5-HT1B pKi
Feeding responsea (3 mg/kg, po)
Fpob,c (%)
Cl (iv) (mL/min/kg)
T1/2 (h)
VSS (L/kg)
22
24
<5.5
<5.5
292%
544%
74%
75%
26
24
7
4
1.0
1.9
1.6 0.1
3.3 0.0
a Food intake measured between 0 and 2 h post-dose expressed relative to vehicle (n = 6).
b Male Sprague–Dawley rats (n = 1–3).
c Dose: iv infusion at 1 mg/kg; po at 3 mg/kg (n = 3, SD).

J. Witherington et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2203–2205
2205
NH
HN
NH
NH
32
a
Br
O2N
O
H2N
O
N
O2N
O
N
b
31
33
34
SO2Cl
Br
O
S
NH
B(OH)2
H
O
d
N
Br
N
c
O
O
12
NH
O
S
H
N
N
O
O
O
24
Scheme 1. Reagents and conditions: (a) Cs2CO3, Pd(OAc)2, (RS)-BINAP, dioxane, reflux (68%); (b) H2, Pd/C, EtOH, 25 °C (100%); (c) pyridine,
DCM, 25 °C (85%); (d) Pd(PPh2)2Cl2 (0.1 mol%), Na2CO3, DME, reflux (90%).
Martin, N. M.; Patterson, M.; Taylor, S. A.; Ghatei, M. A.;
Kamm, M. A.; Johnston, C.; Bloom, S. R.; Emmanuel, A.
V. Gut 2005, 54, 1693.
5-methyl furanyl boronic acid under Suzuki cross cou-
pling conditions afforded the desired analogue 24
(GSK894490A) in good overall yield (Scheme 1).
5. Heightman, T. D.; Scott, J. S.; Longley, M.; Dean, D. K.;
Elliott, R.; Hutley, G.; Witherington, J.; Abberley, L.;
Passingham, B.; Berlanga, M.; Frailes, M.; Wise, A.;
Powney, B.; Muir, A.; McKay, F.; Butler, S.; Winborn,
K.; Gardner, C.; Darton, J.; Campbell, C.; Sanger, G. Bio.
Med. Chem. Lett. 2007, 17, 6584.
6. Matsumoto, M.; Hosoda, H.; Kitajima, Y.; Morozumi, N.;
Minamitake, Y.; Tanaka, S.; Matsuo, H.; Kojima, M.;
Hayashi, Y.; Kangawa, K. Biochem. Biophys. Res. Com-
mun. 2001, 287, 142.
7. GHSR Agonist BACMAM FLIPR Assay was used to
determine the potency and efficacy of the test compounds.
Media were aspirated from cell plates using a cell washer
(leaving 10 ll of media). Cells are immediately loaded with
In summary, having previously successfully identified
SB-791016 as a potent agonist of the ghrelin receptor
we have systematically identified the key pharmaco-
phoric features responsible for high potency. In com-
mon with the literature the SAR around the natural
ligand ghrelin, an appropriately placed lipophilic side
chain is essential for optimal potency in our non peptidic
agonists. Furthermore, replacement of the indoline moi-
ety with an ortho methoxy phenyl ring led to the identi-
fication of a series of compounds that display excellent
in vivo efficacy and pharmacokinetic properties.
loading
buffer
(Tyrodes
(Elga
water + 145 mM
NaCl + 5 mM KCl + 20 mM HEPES + 10 mM glu-
cose + 1 mM MgCl2) + 1.5 mM CaCl2 + 0.714 mg/ml pro-
benicid (predissolved in 1 M NaOH) + 0.5 mM brilliant
black + 2.5 lM Fluo 4 dye) and incubated at 37.5 °C for
1 h. Ten microliters from compound plates is then added
immediately to cell plates using a FLIPR 3 calcium
imaging instrument. Fluorescence measurements are then
taken. Intrinsic activity is expressed relative to human
ghrelin.
References and notes
1. Takaya, K.; Ariyasu, H.; Kanamoto, N.; Iwakura, H.;
Yoshimoto, A.; Harada, M.; Mori, K.; Komatsu, Y.; Usui,
T.; Shimatsu, A.; Ogawa, Y.; Hosoda, K.; Akamizu, T.;
Kojima, M.; Kangawa, K.; Nakao, K. J. Clin. Endocrinol.
Metab. 2000, 85, 4908.
2. Tschop, M.; Smiley, D. L.; Heiman, M. L. Nature 2000,
407, 908.
8. Test compounds were assessed for their effects on food
intake in singly housed, freely feeding male Sprague–
Dawley rats. Groups of 6 rats received either vehicle (1%
(w/v) methylcellulose in water, 1 ml kgÀ1), or test com-
pounds (3 mg kgÀ1) po. Total food consumption was
measured up to 2 h post-dose.
3. Nakazato, M.; Murakami, N.; Date, Y.; Kojima, M.; Matsuo,
H.; Kangawa, K.; Matsukura, S. Nature 2001, 409, 194.
4. (a) Binn, M.; Albert, C.; Gougeon, A.; Maerki, H.; Coulie,
B.; Lemoyne, M.; Rabasa Lhoret, R.; Tomasetto, C.;
Poitras, P. Peptides 2006, 27, 1603; (b) Murray, C. D. R.;