Identification of pyrazolo-pyrimidinones as GHS-R1a antagonists and inverse agonists for the treatment of obesity

Article abstract of DOI:10.1039/c2md20340e

A pyrazolo-pyrimidinone based series of growth hormone secretagogue receptor type 1a (GHS-R1a) antagonists and inverse agonists were identified using a scaffold hop from known quinazolinone GHS-R1a modulators. Lipophilicity was reduced to decrease hERG activity while maintaining GHS-R1a affinity. SAR exploration of a piperidine substituent was used to identify small cyclic groups as a functional switch from partial agonists to neutral antagonists and inverse agonists. A tool compound was identified which had good overall properties and sufficient oral plasma and CNS exposure to demonstrate reduced food intake in mice through a mechanism involving GHS-R1a. The Royal Society of Chemistry.

Full text of DOI:10.1039/c2md20340e

MedChemComm
View Article Online
View Journal | View Issue
CONCISE ARTICLE
Identification of pyrazolo-pyrimidinones as GHS-R1a
antagonists and inverse agonists for the treatment of
obesity†
Cite this: Med. Chem. Commun., 2013,
4, 456
a
William McCoull, Peter Barton,a Anders Broo,b Alastair J. H. Brown,a
*
David S. Clarke,a Gareth Coope,a Robert D. M. Davies,a Alexander G. Dossetter,a
Elizabeth E. Kelly,a Laurent Knerr,b Philip MacFaul,a Jane L. Holmes,a
Nathaniel Martin,a Jane E. Moore,a David Morgan,a Claire Newton,a
b
a
a
b
¨
Krister Osterlund, Graeme R. Robb, Eleanor Rosevere, Nidhal Selmi,
Stephen Stokes,a Tor S. Svensson,b Victoria B. K. Ullahb and Emma J. Williamsa
A pyrazolo-pyrimidinone based series of growth hormone secretagogue receptor type 1a (GHS-R1a)
antagonists and inverse agonists were identified using a scaffold hop from known quinazolinone GHS-
R1a modulators. Lipophilicity was reduced to decrease hERG activity while maintaining GHS-R1a affinity.
SAR exploration of a piperidine substituent was used to identify small cyclic groups as a functional
switch from partial agonists to neutral antagonists and inverse agonists. A tool compound was
identified which had good overall properties and sufficient oral plasma and CNS exposure to
demonstrate reduced food intake in mice through a mechanism involving GHS-R1a.
Received 9th November 2012
Accepted 15th December 2012
DOI: 10.1039/c2md20340e
www.rsc.org/medchemcomm
have additional benet over ‘neutral’ antagonists in lowering the
set-point for hunger between meals.12
Introduction
Ghrelin, a 28 amino acid acylated peptide hormone is the
endogenous ligand of the growth hormone secretagogue
receptor type 1a (GHS-R1a).1 The major physiological role of
ghrelin appears to be in the control of food intake and energy
homeostasis.2 Energy intake and body weight are tightly regu-
lated at a remarkably consistent set-point by control systems in
the hypothalamus.3 While the role of ghrelin in this system is still
not fully understood, it is clear that acylated ghrelin is released
from specialized mucosal cells in response to hunger cues. This
results in a peak of plasma ghrelin levels before meal initiation4
that is consistent with a role in meal initiation. In addition,
ghrelin infusion in both rodents and humans increases appetite
and food intake.5 Consequently, peripheral and CNS penetrant
ghrelin receptor antagonists have been proposed as potential
therapeutic agents for the treatment of obesity6–10 and type II
diabetes.11 Furthermore, due to the apparent constitutive activity
of the ghrelin receptor demonstrated in recombinant cell
systems and in rodents, we believe that inverse agonists may
We are interested in identifying GHS-R1a antagonists and
inverse agonists for the treatment of obesity. Several structurally
diverse antagonists have been reported9,10,13–22 and some have
been shown to act as inverse agonists.19,22 We were attracted to a
quinazolinone series reported to have demonstrated CNS
exposure (Fig. 1).13,14 Quinazolinone 1 typied this structural
class showing high affinity (IC50 ¼ 0.0001 mM) in a 125I-ghrelin
displacement assay, but with numerous undesirable properties.
Activity at the hERG encoded potassium channel23 (IC50
¼
8.7 mM), modest aqueous solubility at pH 7.4 (45 mM), and
activity against cytochrome P450 enzymes (CYP3A4 inhibition
IC50 ¼ 0.68 mM) are all consistent with high measured lip-
ophilicity (log D7.4 > 4.3). Another quinazolinone 2 (YIL-870)
with lower affinity (IC50 ¼ 0.052 mM) was reported to signi-
cantly reduce body weight in DIO mice.14 However, we have
determined that the effects of this specic compound on body
weight are independent of GHS-R1a (see ESI†). Both 1 and 2 do
not demonstrate pharmacological properties consistent with
inverse agonism as measured in a Tangoꢀ functional assay
thus are likely to represent neutral antagonists or weak partial
agonists. We anticipated that many of these undesirable prop-
erties could be improved upon by reducing lipophilicity and
that the functional activity could be tuned to give an inverse
agonist through SAR exploration. Consequently we planned to
use a scaffold hopping approach, switching the bicyclic core,
then modify the two aryl substituents to improve ligand
aCardiovascular and Gastrointestinal Innovative Medicines Unit, AstraZeneca,
Alderley Park, Maccleseld, Cheshire, SK10 4TG, UK. E-mail: william.mccoull@
astrazeneca.com; Tel: +44 (0) 1625 519444
bCardiovascular
& Gastrointestinal Innovative Medicines Unit, AstraZeneca,
¨
Pepparedsleden 1, 431 83, Molndal, Sweden
† Electronic supplementary information (ESI) available: Synthetic details for the
preparation of compounds and protocols for biological and physicochemical
measurements. See DOI: 10.1039/c2md20340e
456 | Med. Chem. Commun., 2013, 4, 456–462
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Concise Article
MedChemComm
Scheme 2 Reagents and conditions: (a) MeOH, H2SO4, 81%; (b) ArB(OH)2,
Cu(OAc)2, py, air, DCM, 22–81%; (c) Pd/C, H2 or Fe, HCl, 23–87%.
Fig. 1 GHS-R1a modulators from quinazolinone series.13,14
Acylation of the requisite aminopyrazole 11 with 2-methyl-
nicotinic acid was carried out under standard amide coupling
conditions and the resulting amide 12 chlorinated with phos-
phorous pentachloride. Amidine formation with (R)-tert-butyl 3-
(aminomethyl)piperidine-1-carboxylate and cyclisation under
basic conditions gave the pyrazolo-pyrimidinone core in 13. The
intermediate imidoyl chloride was unstable and partly rear-
ranged to an acyl chloride leading to the formation of some bis-
amide side product 15. Removal of the Boc protecting group
and either alkylation or reductive amination allowed for pyr-
azolo-pyrimidinones 14 to be synthesized with a variety of
substituents on the piperidine nitrogen.
lipophilicity efficiency (LLE).24 Since the chiral piperidine is
important to receptor binding we planned to keep this ring but
target the alkyl sidechain on the piperidine nitrogen to probe
SAR for inverse agonism while also reducing basicity to mini-
mise likelihood of off-target pharmacology for a CNS agent.24
We reasoned that the combination of these gross structural
changes would increase the likelihood of avoiding the off-target
effects associated with 2. In this paper we describe the reduc-
tion to practice of this strategy to afford potent pyrazolo-pyr-
imidinone based GHS-R1a antagonists and inverse agonists,
providing a tool compound with oral plasma and CNS exposure
suitable for evaluation in preclinical efficacy models.
Results and discussion
Thieno-pyrimidinone and thiazolo-pyrimidinone synthesis is
exemplied in Scheme 1. Aminoester 5 was obtained from 3
using Liebeskind–Srogl coupling.25 Subsequent condensation
with the imidoyl chloride generated from benzamide 6 afforded
amidines that were directly ring closed using TiCl4 (ref. 26) and
microwave heating, to obtain 7 and 8.
Intermediate 2-substituted 4-aminopyrazoles 11 were
prepared in three steps from nitro pyrazole acid 9 (Scheme 2).
Esterication using continuous ow chemistry minimised
handling large amounts of potentially explosive nitro pyrazole.
Then Chan–Lam couplings27 followed by reduction of the nitro
group efficiently generated a diverse set of 2-aryl pyrazoles 11.
Synthetic routes to allow diversity at three positions of the
pyrazolo-pyrimidinone were developed as shown in Scheme 3.
Scheme 3 Reagents and conditions: (a) 2-Me-nicotinic acid, HATU or HBTU,
DIPEA, DCM or DMF, r.t, 48 h, 71–79%; (b) PCl5, toluene, 110 ꢀC, 4–6 h then (R)-
tert-butyl 3-(aminomethyl)piperidine-1-carboxylate, DIPEA, DCE; (c) K2CO3, DMF,
130 ꢀC, 45 min, 19–45% (2 steps); (d) 4 M HCl in 1,4-dioxane, r.t, 16 h, 80–100%;
(e) Na(AcO)3BH, DCE, AcOH, ketone or aldehyde, 34–65%; or R6–I/Br, K2CO3,
DMF, or dioxane, 80–100 ꢀC, 23–66%; or R6–OTs, MeCN or DMF, 90–130 ꢀC, 8–
44%; (f) R5–COCl, NEt3, DCM, 81%; (g) PS-PPh3, CCl4, NEt3 then aminopyrazole
11, DIPEA or POCl3, DCE, 16 h, 85 ꢀC; (h) K2CO3, DMF, 100 ꢀC, 16 h or TiCl4, DCE,
150 ꢀC, 10 min, 29–73% (2 steps); (i) PS-PPh3, CCl4, NEt3 then 4-amino-5-ethox-
ycarbonyl-2-tetrahydropyranyl-pyrazole; (j) K2CO3, DMF, 100 ꢀC, 16 h, 54%
(2 steps); (k) 2 M HCl, r.t, 16 h, 92%; (l) Cu(OAc)2, py, DCM, air, 22–30%.
Scheme 1 Reagents and conditions: (a) 4-Cl-phenylboronic acid, copper(I)
salicylate, Pd(PPh3)4, THF, 50 ꢀC, 70 h, 69%; (b) (i) 6, POCl3, DCE, 16 h, 85 ꢀC, (ii)
TiCl4, DCE, 170 ꢀC, mW, 10 min, 60% (7) or 6, TiCl4, DCE, 170 ꢀC, mW, 40 min,
16% (8).
This journal is ª The Royal Society of Chemistry 2013
Med. Chem. Commun., 2013, 4, 456–462 | 457

View Article Online
MedChemComm
Concise Article
Fig. 2 Box and whisker plot showing predicted log D of bicyclic cores considered for scaffold hop. Boxes indicate 25th, 50th (median) and 75th percentiles. Whiskers
show full data range, excluding outliers. Blue line connects the mean log D for each core. A yellow background indicates cores that were synthesised in this work (AG,
AJ, BD). A pink background indicates quinazolinone and related compounds, previously reported (CC, CE, CG, CI). 5,6-Bicycles start A or B, 6,6-bicycles start with C. Full
core names and data detail in ESI.†
Alternatively, acylation of an 3-aminomethyl piperidine 16 and measured log D components.29 This is a more reliable
and chlorination28 of the resulting amide 17 under mild method of identifying the contribution of the core in a fully
conditions gave an intermediate imidoyl chloride which could elaborated molecule rather than simply using calculated log D
be smoothly condensed with a 2-substituted 4-aminopyrazoles of the unsubstituted core. Potential cores could then be vetted
11. Cyclisation of the resulting amidine gave the pyrazolo-pyr- for likelihood of lowering log D and since we restricted
imidinone core in 18, avoiding the bis-amide side product.
ourselves to AstraZeneca compounds, synthetic routes were also
To allow for late-stage variation of the aryl substituent on the likely to exist. Fig. 2 shows that the quinazolinone core (CC) is
pyrazole a tetrahydropyranyl (THP)-protected pyrazole was one of the more lipophilic cores and several more hydrophilic
employed. The pyrazolo-pyrimidinone core was assembled with opportunities existed e.g. thiazolo-pyrimidinones (AJ) and pyr-
THP at R2 (20). Then THP-deprotection gave 21 followed by azolo-pyrimidinones (BD) – the core with predicted lowest log D.
Chan–Lam coupling to give the desired arylated compounds 22 Thieno-pyrimidinones (AG) have similar predicted log D to
(along with their N-1 regioisomers).
quinazolinones and several aza-modied quinazolinone cores
Initial chemistry efforts were focused on replacing the qui- (CE, CG and CI) predict lower log D as has been reported by
nazolinone with a less lipophilic bicyclic core. Our design others.18
strategy involved identifying potential bicyclic systems that
Three cores were prepared and compared with quinazoli-
existed in the AstraZeneca compound collection with measured none 1 (Table 1). Thienopyrimidinone 23 was initially selected
log D7.4 (mlog D). Substructure log D values were calculated to conrm changing to a ve-membered ring was tolerated, with
using ACDlog D (pH 7.4) for each substituent, the sum of which some reduction in potency in light of enantiopure 1 being
was subtracted from the mlog D to obtain a predicted log D for compared with racemates. Pyrazolo-pyrimidinone 24 was
the core. Averaging over several compounds is a method to shown to maintain affinity. It is difficult to obtain accurate log
minimise the effect of noise due to errors in both the calculated D7.4 data on high lipophilicity compounds using a shake-ask
Table 1 GHS-R1a binding affinity, lipophilicity, LLE, hERG and pKa for selected bicyclic core variations
Cpd
Core (identier)
R6
GHS-R1a IC50a (mM)
HPLC log Dc
log D7.4
LLEd
hERG IC50 (mM)
pKa
1b
23
24
7
8
25
Quinazolinone (CC)
i-Pr
i-Pr
i-Pr
CH2CH2OMe
CH2CH2OMe
CH2CH2OMe
0.00012
0.0015
0.00028
0.016
0.059
0.0011
5.0
4.8
3.2
5.6
5.1
3.7
>4.3
—
—
>4.3
3.5
3.3
4.9
4.0
6.4
2.2
2.1
5.3
8.7
2.1
3.5
2.1
3.7
2.5
—
—
10.0
8.5
9.2
8.5
Thieno-pyrimidinone (AG)
Pyrazolo-pyrimidinone (BD)
Thieno-pyrimidinone (AG)
Thiazolo-pyrimidinone (AJ)
Pyrazolo-pyrimidinone (BD)
a
d
Mean values of at least two experiments. b (3S)-Piperidine rather than racemate. c Measured using chromatographic method.30 Using HPLC log D.
458 | Med. Chem. Commun., 2013, 4, 456–462
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Concise Article
MedChemComm
method, thus an alternative measure of log D based on chro- reduction of potency. For the next phase of optimisation, R5 ¼
matography30 was used. This HPLC log D data indicated that 2-methyl-3-pyridyl and R2 ¼ 3-methoxyphenyl was selected as
pyrazolo-pyrimidinone 24 demonstrated signicant lip- the optimal substituents based on ease of synthesis and overall
ophilicity decrease over both quinazolinone 1 and thieno-pyr- properties. For example, 34 exhibited aqueous solubility of
imidinone 23 (3.1 for 24, compared to 5.0 for 1 and 4.8 for 23). >2700 mM, low plasma protein binding (63% free and 57% free
High basicity is expected from the N-iso-propyl piperidine in human and mouse respectively), no signicant activity
which, coupled with high lipophilicity is associated with an
increased risk of promiscuity.24 Thus we commonly used ether
containing groups such as methoxyethyl (compounds 7, 8, 25)
to lower pKa. Some reduction in potency resulted but pyrazolo-
Table 3 GHS-R1a binding affinity, and functional effect for selected compounds
varying R6
pyrimidinone 25 again showed lower lipophilicity than thieno-
pyrimidinone 7 or thiazolo-pyrimidinone 8. Use of HPLC log D
to calculate LLE identies pyrazolo-pyrimidinones as superior
to the other cores. Activity against the hERG encoded potassium
channel persisted in the low micromolar range for all these
compounds.
We decided to focus on the pyrazolo-pyrimidinone core but
realised that further lipophilicity reduction was required to
improve overall properties such as hERG activity (Table 2). The
GHS-R1a
IC50
(mM)
Agonist
effecta,b
(%)
Inverse
a
4-Cl-phenyl group could be varied to reduce log D7.4 while still
maintaining nanomolar binding affinity in many cases
(compounds 26–36). Incorporating pyridine into the R5-ortho
tolyl group gave a further log D7.4 reduction while maintaining
potency as in 34 compared to 30. Activity against hERG was now
consistently greater than 10 mM, irrespective of the basicity. As
found previously, reducing the piperidine pKa from 10 to 8.5
was well tolerated but ablation of basicity was not tolerated as in
oxetane 33. Removal of the R2 aryl group or conversion to
saturated substituents such as c-hexyl 36 resulted in signicant
agonist
Cpd
R6
effecta,b (%)
35
37
38
0.0061
0.023
78
21
<10
<10
23
0.0076
<10
39
40
41
42
0.12
<10
11
16
<10
55
0.015
0.0051
0.005
Table 2 GHS-R1a binding potency, hERG activity, log D7.4, LLE and pKa for
<10
70
selected compounds
<10
43
44
0.04
55
74
<10
<10
0.0017
45
46
0.12
53
64
<10
<10
GHS-R1a
IC50
(mM)
hERG
IC50
a
0.0027
Cpd A R6
R2
log D7.4 LLE (mM) pKa
24 CH i-Pr
26 CH i-Pr
27 CH i-Pr
28 CH i-Pr
29 CH i-Pr
30 CH i-Pr
31 CH i-Pr
4-Cl-Ph
4-F-Ph
4-CN-Ph 0.012
4-MeO-Ph 0.0073
3-Cl-Ph
3-MeO-Ph 0.0018
3-CN-Ph 0.032
0.00028
0.0011
—
2.3
—
6.7 >10
3.5
10.0
10.0
47
48
0.0057
0.0016
<10
11
<10
10
>8.3 10.0
2.4
3
5.7 11
10.0
9.8
10.0
9.9
8.5
6.4
9.7
8.4
—
0.0016
5.8 10
6.3 15
5.7 22
6.3 21
4.2
7.5 47
6.9 43
5
49
50
0.012
0.2
<10
14
59
2.5
1.8
1.1
1.2
1.3
1.3
1.4
<10
32
N
N
N
N
N
CH2CH2OMe 4-CN-Ph 0.044
Oxetan-3-yl 4-CN-Ph
i-Pr 3-MeO-Ph 0.0017
CH2CH2OMe 3-MeO-Ph 0.0061
CH2CH2OMe c-Hex 0.37
33
4
34b
35b
36b
51
0.06
45
<10
>100
a
b
a
Mean values of at least two experiments. b (3S)-Piperidine rather than
Mean values of at least two experiments. Maximum% activation,
maximum concentration ¼ 10 mM.
racemate.
This journal is ª The Royal Society of Chemistry 2013
Med. Chem. Commun., 2013, 4, 456–462 | 459

View Article Online
MedChemComm
Concise Article
against cytochrome P450 enzymes CYP1A2, CYP2C9, CYP2C19, antagonism or inverse agonism but with a signicant reduction
CYP2D6 and CYP3A4 (all >30 mM), in addition to high LLE of in affinity thus were not progressed. Variations at R2-aryl
7.5. However, 34 exhibited full agonism in a Tango functional substitution maintained inverse agonism with some variation
assay with EC50 ¼ 0.0013 mM (92% maximal agonist effect) and in maximal response but also a marked SAR for binding affinity
this undesired functional effect was addressed next.
depending on the ether substituent. 4-Methoxy 55 has 4-fold
Our attempts to modulate functional activity focused on the greater affinity than unsubstituted 54. Moving methoxy from 4-
piperidine substituent, based on functional modulation repor- to 3-position as in 41 increased affinity 10-fold further and the
ted in the quinazolinone work.13 In our pyrazolo-pyrimidinones larger 3-ethoxy 56 gave 5-fold more affinity increase. Di-F-methyl
we also found this substituent to inuence functional activity ether 59 exhibited the largest inverse agonist response. The
but with very specic SAR as shown in Table 3. Introduction of larger ethers have higher lipophilicity resulting in increased
cyclic substituents reduced the agonist response as in ethers hERG activity, higher metabolic clearance and only 56 has
37–39 with the 5-membered ring exhibiting potent binding improved LLE over 41. Since we were interested in a CNS
affinity and partial inverse agonism. The activity was found to penetrant compound, a hybrid ligand efficiency parameter (LE
be dependent on stereochemistry with the (S)-tetrahydrofuran fub) incorporating potency and free fraction in brain,32 as an
40 exhibiting neutral antagonism but (R)-tetrahydrofuran 41 indicator of receptor occupancy33 was monitored throughout
exhibiting inverse agonism. Branched acyclic analogues 42–44 the optimisation process. Our aim was to at least match the
are all partial agonists, as are hetaryl analogues exemplied by value of antagonist 1 (LE fub ¼ 6.8) with our pyrazolo-pyr-
45 and 46. Alkyl analogues 47–50 demonstrated similar trend in imidinone inverse agonist tool. Consequently 41 was judged to
functional activity with the 4 and 6-membered rings being have the most attractive prole and conrmed to have suitable
neutral antagonists and the c-pentyl analogue 49 being an plasma and CNS exposure following oral dosing (Table 5).
inverse agonist. Interestingly, the regioisomeric tetrahydro-
In a free feeding test, compound 41 (20 mg per kg po) was
furan 51 is a partial agonist, again highlighting the very specic administered to ad libitum fed GHS-R1a null (KO) and wild-type
nature of functional SAR in this series. While we do not have an (WT) mice and food intake measured at two-hourly intervals
full understanding of the interaction of these compounds with (Fig. 3a). Over the rst 2 h time interval food intake in WT mice
the receptor, a binding pocket has been proposed which can is reduced by 68% but with no signicant change in KO mice.
lead to the opposite efficacy and our results are consistent with The difference in food intake at 2 h between WT and KO
nding.31 Having identied the (R)-tetrahydrofuran as a func- mice treated with vehicle is consistent with what we observe in
tional switch, we performed a further SAR iteration to identify other studies using this model.
an inverse agonist with CNS exposure suitable as a tool for
Similar exposure levels of 41 were observed in KO and WT
preclinical efficacy experiments. Gratifyingly, the (R)-tetrahy- mice and free exposure levels in brain were sustained above the
drofuran maintained inverse agonism when the R2-aryl or R5- binding affinity over the rst 2 h time interval (Fig. 3b). This
pyridyl groups were modied (Table 4). Smaller groups at R5, result is consistent with the feeding being blocked by inverse
for example methyl (52) and iso-butyl (53) did maintain neutral agonist 41 through a mechanism involving GHS-R1a. The effect
Table 4 GHS-R1a binding affinity, functional effect, hERG activity, log D7.4, LLE, human liver microsome (HLM) metabolism and LE fub for selected compounds
Inverse
agonist
a
GHS-R1a IC50
(mM)
Agonist
hERG IC50
HLM Clint
Cpd R5
R2
effecta,b (%) effecta,b (%) (mM)
log D7.4 LLE (ml minꢁ1 mgꢁ1
)
LE fubc
7.2
41
52
53
54
55
56
57
58
59
2-Me-3-pyridyl 3-MeO-Ph
Me
i-Bu
0.0051
0.98
0.5
<10
28
55
<10
36
10
58
53
11
20
81
44
2.9
3.2
12
4.1
14
7.3
6.8
8.4
2
6.3
4.1
3.2
5
17
16
60
15
21
30
67
51
33
3-MeO-Ph
3-MeO-Ph
1.9
3.1
1.7
1.9
2.5
2.9
2.7
2.2
<10
<10
<10
<10
<10
<10
<10
4.7
6
2-Me-3-pyridyl Ph
0.19
2-Me-3-pyridyl 4-MeO-Ph
2-Me-3-pyridyl 3-EtO-Ph
2-Me-3-pyridyl 3-n-PrO-Ph
2-Me-3-pyridyl 3-i-PrO-Ph
0.052
0.00091
0.0048
0.011
5.4
6.5
5.4
5.2
5.4
6.3
7.9
6.6
6.5
6.5
2-Me-3-pyridyl 3-HCF2O-Ph 0.027
a
b
c
Mean value of at least two experiments. Maximum% activation, maximum concentration ¼ 10 mM. LE fub ¼ pIC50 + log10[fub] where fub
measured using brain slice method.34
460 | Med. Chem. Commun., 2013, 4, 456–462
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Concise Article
MedChemComm
Table 5 Selected properties for 41
Conclusions
In summary, a scaffold hop from known quinazolinone GHS-
R1a antagonists to less lipophilic pyrazolo-pyrimidinones was
conducted, followed by further SAR optimization to decrease
lipophilicity and reduce hERG activity. Modication of a
piperidine substituent was used to convert partial agonists into
neutral antagonists and inverse agonists, resulting in 41 which
had good overall properties and sufficient oral plasma and CNS
exposure for use in preclinical efficacy studies. In vivo efficacy of
41 was conrmed in a free feeding study and this effect was
shown to be dependent on the expression of GHS-R1a.
Plasma concna (mM)
1.2
0.24
8.6
Brain concna (mM)
Rat brainb(% free)
Aq. solubility pH 7.4 (mM)
h/rat/mu plasma protein binding (% free)
>1500
30/42/24
a
Compound was dosed 20 mg kgꢁ1 (mouse), in 1% Pluronic F127 and
Acknowledgements
b
compound levels measured at 1 h. Measured using brain slice
method.34
Teresa Collins, Mark Denn and John Swales are acknowledged
for expert technical assistance in generating DMPK data. Dave
Masters, Jenny Morrell, Elizabeth Strutynskyj, Steve Bloor, Rob
Garcia, Sally Johnson, Susan Aiston and Julie Bartlett's expert
technical assistance in generating affinity and functional data is
also acknowledged.
Notes and references
1 M. Kojima, H. Hosoda, Y. Date, M. Nakazato, H. Matsuo and
K. Kangawa, Nature, 1999, 402, 656–660.
2 M. Kojima, H. Hosoda, H. Matsuo and K. Kangawa, Trends
Endocrinol. Metab., 2001, 12, 118–122.
3 M. W. Schwartz, S. C. Woods, D. Porte Jr, R. J. Seeley and
D. G. Baskin, Nature, 2000, 404, 661–671.
4 D. E. Cummings, J. Q. Purnell, R. S. Frayo, K. Schmidova,
B. E. Wisse and D. S. Weigle, Diabetes, 2001, 50, 1714–1719.
5 A. M. Wren, L. J. Seal, M. A. Cohen, A. E. Brynes, G. S. Frost,
K. G. Murphy, W. S. Dhillo, M. A. Ghatei and S. R. Bloom,
J. Clin. Endocrinol. Metab., 2001, 86, 5992–5995.
6 I. Depoortere, Regul. Pept., 2009, 156, 13–23.
7 P. A. Carpino and G. Ho, Expert Opin. Ther. Pat., 2008, 18,
1253–1263.
8 J. Soares, R. Roncon-Albuquerque Jr and A. Leite-Moreira,
Expert Opin. Ther. Targets, 2008, 12, 1177–1189.
9 H. Zhao and G. Liu, Curr. Opin. Drug Discovery Dev., 2006, 9,
509–515.
10 C. Chollet, K. Meyer and A. G. Beck-Sickinger, J. Pept. Sci.,
2009, 15, 711–730.
11 R. S. Ahima, Endocrinology, 2007, 148, 5173–5174.
12 B. Holst and T. W. Schwartz, Trends Pharmacol. Sci., 2004, 25,
113–117.
Fig. 3 (a) Pharmacodynamic effect of 41 in GHS-R1a KO and WT mice: food
intake (g). Mean values ꢂ SEM (**p < 0.01). (b) Pharmacokinetic levels of 41 in
GHS-R1a KO and WT mice expressed as free cover multiples of binding affinity. 1
and 2 refer to plasma and brain samples taken from the same animal at the same
time point of 1 h and 2 h respectively. The 2 h WT_plasma datapoint (blue dia-
mond) is obscured and lies under the 2 h KO_2_plasma time point.
13 J. Rudolph, W. P. Esler, S. O'Connor, P. D. G. Coish,
P. L. Wickens, M. Brands, D. E. Bierer, B. T. Bloomquist,
G. Bondar, L. Chen, C. Chuang, T. H. Claus, Z. Fathi,
W. Fu, U. R. Khire, J. A. Kristie, X. Liu, D. B. Lowe,
A. C. McClure, M. Michels, A. A. Ortiz, P. D. Ramsden,
R. W. Schoenleber, T. E. Shelekhin, A. Vakalopoulos,
W. Tang, L. Wang, L. Yi, S. J. Gardell, J. N. Livingston,
L. J. Sweet and W. H. Bullock, J. Med. Chem., 2007, 50,
5202–5216.
is short lived as during the next time interval (2–4 h), no
signicant effects on food intake were observed, which is
consistent with free exposure levels in brain falling below the
binding affinity.
This journal is ª The Royal Society of Chemistry 2013
Med. Chem. Commun., 2013, 4, 456–462 | 461

View Article Online
MedChemComm
Concise Article
14 W. P. Esler, J. Rudolph, T. H. Claus, W. Tang, N. Barucci,
S. Brown, W. Bullock, M. Daly, L. DeCarr, Y. Li, L. Milardo,
D. Molstad, J. Zhu, S. J. Gardell, J. N. Livingston and
L. J. Sweet, Endocrinology, 2007, 148, 5175–5185.
15 J. T. Mihalic, Y. Kim, M. Lizarzaburu, X. Chen, J. Deignan,
M. Wanska, M. Yu, J. Fu, X. Chen, A. Zhang, R. Connors,
L. Liang, M. Lindstrom, J. Ma, L. Tang, K. Dai and L. Li,
Bioorg. Med. Chem. Lett., 2012, 22, 2046–2051.
S. Varma, M. Campbell, D. DeBartola, R. Kosa-Maines,
S. J. Steyn and K. F. McClure, Bioorg. Med. Chem. Lett.,
2012, 22, 4281–4287.
23 M. H. Bridgland-Taylor, A. C. Hargreaves, A. Easter, A. Orme,
D. C. Henthorn, M. Ding, A. M. Davis, B. G. Small,
C. G. Heapy, N. Abi-Gerges, F. Persson, I. Jacobson,
M. Sullivan, N. Albertson, T. G. Hammond, E. Sullivan,
J.-P. Valentin and C. E. Pollard, J. Pharmacol. Toxicol.
Methods, 2006, 54, 189–199.
16 M. Yu, M. Lizarzaburu, H. Beckmann, R. Connors, K. Dai,
K. Haller, C. Li, L. Liang, M. Lindstrom, J. Ma, A. Motani, 24 P. D. Leeson and B. Springthorpe, Nat. Rev. Drug Discovery,
M. Wanska, A. Zhang, L. Li and J. C. Medina, Bioorg. Med.
Chem. Lett., 2010, 20, 1758–1762.
17 L. Puleo, P. Marini, R. Avallone, M. Zanchet, S. Bandiera,
M. Baroni and T. Croci, Bioorg. Med. Chem., 2012, 20,
5623–5636.
2007, 6, 881–890.
25 C. Kusturin, L. S. Liebeskind, H. Rahman, K. Sample,
B. Schweitzer, J. Srogl and W. L. Neumann, Org. Lett., 2003,
5, 4349–4352.
26 K. Yang, X. He, H. Choi, Z. Wang, D. H. Woodmansee and
H. Liu, Tetrahedron Lett., 2008, 49, 1725–1728.
18 P. Hanrahan, J. Bell, G. Bottomley, S. Bradley, P. Clarke,
E. Curtis, S. Davis, G. Dawson, J. Horswill, J. Keily, 27 J. X. Qiao and P. Y. S. Lam, Synthesis, 2011, 829–
G. Moore, C. Rasamison and J. Bloxham, Bioorg. Med.
Chem. Lett., 2012, 22, 2271–2278.
19 A. Pasternak, S. D. Goble, R. K. deJesus, D. L. Hreniuk,
856.
28 K. Tamura, H. Mizukami, K. Maeda, H. Watanabe and
K. Uneyama, J. Org. Chem., 1993, 58, 32–35.
C. C. Chung, M. R. Tota, P. Mazur, S. D. Feighner, 29 E. Griffen, A. G. Leach, G. R. Robb and D. J. Warner, J. Med.
A. D. Howard, S. G. Mills and L. Yang, Bioorg. Med. Chem.
Lett., 2009, 19, 6237–6240.
Chem., 2011, 54, 7739–7750.
30 K. Valko, J. Chromatogr., A, 2004, 1037, 299–310.
20 F. M. Sabbatini, S. Melotto, G. Bernasconi, S. M. Bromidge, 31 B. Holst, J. Mokrosinski, M. Lang, E. Brandt, R. Nygaard,
L. D'Adamo, M. Rinaldi, C. Savoia, C. Mundi, C. Di
Francesco, L. Zonzini, V. J. A. Costantini, B. Perini,
T. M. Frimurer, A. G. Beck-Sickinger and T. W. Schwartz,
J. Biol. Chem., 2007, 282, 15799–15811.
`
E. Valerio, A. Pozzan, E. Perdona, F. Visentini, M. Corsi 32 M. Pilla, M. Andreoli, M. Tessari, S. Delle-Fratte, A. Roth,
and R. Di Fabio, ChemMedChem, 2011, 6, 1981–1985.
21 M. D. Serby, H. Zhao, B. G. Szczepankiewicz, C. Kosogof,
Z. Xin, B. Liu, M. Liu, L. T. J. Nelson, W. Kaszubska,
H. D. Falls, V. Schaefer, E. N. Bush, R. Shapiro, B. A. Droz,
V. E. Knourek-Segel, T. A. Fey, M. E. Brune, D. W. A. Beno,
S. Butler, F. Brown, P. Shah, E. Bettini, P. Cavallini,
R. Benedetti, D. Minick, P. Smith, B. Tehan,
P. D'Alessandro, O. Lorthioir, C. Ball, V. Garzya,
C. Goodacre and S. Watson, Bioorg. Med. Chem. Lett., 2010,
20, 7521–7524.
T. M. Turner, C. A. Collins, P. B. Jacobson, H. L. Sham and 33 J. Watson, S. Wright, A. Lucas, K. L. Clarke, J. Viggers,
G. Liu, J. Med. Chem., 2006, 49, 2568–2578.
22 D. W. Kung, S. B. Coffey, R. M. Jones, S. Cabral, W. Jiao,
S. Cheetham, P. Jeffrey, R. Porter and K. D. Read, Drug
Metab. Dispos., 2009, 37, 753–760.
´
M. Fichtner, P. A. Carpino, C. R. Rose, R. F. Hank, 34 M. Friden, F. Ducrozet, B. Middleton, M. Antonsson,
M. G. Lopaze, R. Swartz, H. (Tommy) Chen, Z. Hendsch,
B. Posner, C. F. Wielis, B. Manning, J. Dubins, I. A. Stock,
U. Bredberg and M. Hammarlund-Udenaes, Drug Metab.
Dispos., 2009, 37, 1226–1233.
462 | Med. Chem. Commun., 2013, 4, 456–462
This journal is ª The Royal Society of Chemistry 2013