Identification, optimization, and pharmacology of acylurea GHS-R1a inverse agonists

Article abstract of DOI:10.1021/jm500610n

Ghrelin plays a major physiological role in the control of food intake, and inverse agonists of the ghrelin receptor (GHS-R1a) are widely considered to offer utility as antiobesity agents by lowering the set-point for hunger between meals. We identified an acylurea series of ghrelin modulators from high throughput screening and optimized binding affinity through structure-activity relationship studies. Furthermore, we identified specific substructural changes, which switched partial agonist activity to inverse agonist activity, and optimized physicochemical and DMPK properties to afford the non-CNS penetrant inverse agonist 22 (AZ-GHS-22) and the CNS penetrant inverse agonist 38 (AZ-GHS-38). Free feeding efficacy experiments showed that CNS exposure was necessary to obtain reduced food intake in mice, and it was demonstrated using GHS-R1a null and wild-type mice that this effect operates through a mechanism involving GHS-R1a.

Full text of DOI:10.1021/jm500610n

Subscriber access provided by Old Dominion University Libraries & VIVA (Virtual Library of Virginia)
Article
Identification, optimization and pharmacology
of acylurea GHS-R1a inverse agonists
William McCoull, Peter Barton, Alastair Brown, Susanne Bowker, Jennifer Cameron, David
Clarke, Robert Davies, Alastair Dossetter, Anne Ertan, Mark Fenwick, Clive Green, Jane
Holmes, Nathaniel Martin, David Masters, Jane Moore, Nicholas Newcombe, Claire Newton,
Helen Pointon, Graeme R. Robb, Christopher Sheldon, Stephen Stokes, and David Morgan
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm500610n • Publication Date (Web): 26 Jun 2014
Downloaded from http://pubs.acs.org on July 6, 2014
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 32
Journal of Medicinal Chemistry
Identification, optimization and pharmacology of acylurea GHS-R1a inverse agonists
1
2
3
4
5
6
7
8
9
William McCoull,* Peter Barton, Alastair J.H. Brown, Suzanne S. Bowker, Jennifer Cameron, David S.
Clarke, Robert D.M. Davies, Alexander G. Dossetter, Anne Ertan, Mark Fenwick, Clive Green, Jane L.
Holmes, Nathaniel Martin, David Masters, Jane E. Moore, Nicholas J. Newcombe, Claire Newton,
Helen Pointon, Graeme R. Robb, Christopher Sheldon, Stephen Stokes and David Morgan
AstraZeneca, Mereside, Alderley Park, Macclesfield, SK10 4TG, UK
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
KEYWORDS
Ghrelin; Growth hormone secretagogue receptor type 1a; inverse agonist; antagonist; GHSꢀR1a; CNS
penetration; Food intake; Pharmacokinetics; Synthesis.
ABSTRACT
O
S
O
S
N
O
S
Cl
O
O
N
Cl
O
O
N
O
O
Cl
O
O
N
S
S
N
N
N
N
N
O
S
N
N
N
N
N
O
N
O
N
Compound
14
22
38
(AZꢀGHSꢀ14)
GHSꢀR1a binding IC50 1.3 nM
(AZꢀGHSꢀ22)
0.77 nM
(AZꢀGHSꢀ38)
6.7 nM
GHSꢀR1a function
CNS exposure
partial agonist
inverse agonist
nonꢀCNS penetrant
inverse agonist
CNS penetrant
ꢀ
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 32
Ghrelin plays a major physiological role in the control of food intake and inverse agonists of the ghrelin
receptor (GHSꢀR1a) are widely considered to offer utility as antiꢀobesity agents by lowering the setꢀ
point for hunger between meals. We identified an acylurea series of ghrelin modulators from high
throughput screening and optimized binding affinity through SAR studies. Furthermore, we identified
specific substructural changes which switched partial agonist activity to inverse agonist activity, and
optimized physicochemical and DMPK properties to afford the nonꢀCNS penetrant inverse agonist 22
(AZꢀGHSꢀ22) and the CNS penetrant inverse agonist 38 (AZꢀGHSꢀ38). Free feeding efficacy
experiments showed that CNS exposure was necessary to obtain reduced food intake in mice and it was
demonstrated using GHSꢀR1a null and wildꢀtype mice that this effect operates through a mechanism
involving GHSꢀR1a.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
2

Page 3 of 32
Journal of Medicinal Chemistry
Introduction.
1
2
3
4
5
6
7
8
9
Ghrelin, a 28 amino acid acylated peptide hormone is the endogenous ligand of the growth
hormone secretagogue receptor type 1a (GHSꢀR1a).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
regulated at a remarkably consistent setꢀpoint 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 obesity6ꢀ11 and type II diabetes.12 Furthermore, due to the apparent high constitutive
activity13ꢀ15 of the ghrelin receptor demonstrated in recombinant cell systems and in rodents, inverse
agonists may have additional benefit over ‘neutral’ antagonists in lowering the setꢀpoint for hunger
between meals.16
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
While several structurally diverse antagonists have been reported,17ꢀ30 only a smaller number of
these reports have identified inverse agonists.25ꢀ30 Our aim was to identify GHSꢀR1a inverse agonists
for the treatment of obesity and this manuscript details our efforts to obtain a new GHSꢀR1a chemotype
which showed onꢀtarget efficacy in preclinical models.
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 32
Results and Discussion
1
2
3
4
5
6
7
8
9
High throughput screening (HTS) of the AstraZeneca compound collection was conducted and
identified acylurea benzothiazole 1 as a suitable start point for optimization, having moderate binding
affinity (IC50 = 0.21 ꢁM) for GHSꢀR1a but high lipophilicity and low aqueous solubility (Table 1).
Removal of one chlorine as in compound 2 was tolerated for affinity, as was replacement of the
methoxy group with methanesulfonyl which reduced lipophilicity (3). This resulted in an acceptable
ligand lipophilicity efficiency (LLE),31 which we aimed to maintain above 5 throughout optimisation.
Further structural modifications were conducted to understand structureꢀactivity relationships (SAR).
Moving the sulfone group from the 6 to 5 position of the benzothiazole lost more than 10ꢀfold affinity
(4), while the corresponding benzoxazole 5 had no measurable affinity. To improve physicochemical
properties, such as aqueous solubility, basic sidechains were appended to the 6ꢀsulfonyl group in
benzothiazoles 6ꢀ8, resulting in retained affinity and solubilities exceeding 50 ꢁM. At the other end of
the molecule, chemistry allowed us to explore SAR patterns around the phenyl ring. A notable 5ꢀfold
improvement in affinity was obtained by incorporation of a pyrrole group para to the Cl on the phenyl
ring (9). Functional activity was then tested using a receptorꢀspecific βꢀarrestin recruitment assay,
Tango™, in which the GHSꢀR1a was expressed from a TꢀREx™ inducible plasmid with high
constitutive activity. This assay could differentiate agonists and inverse agonists,32 and where
measured, all compounds were shown to exhibit varying levels of partial agonism. Initially, this was not
of concern as we wished to obtain high affinity first, and then explore SAR to convert functional activity
to inverse agonism. This strategy was used successfully by us in a previously reported chemotype for
the same receptor.28
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. Initial SAR exploration optimisation.
R1
R2
Cl
O
O
N
A
N
N
H
H
R6
R5
h GHSꢀR1a Agonist
binding IC50 effect
Aq.
Cpd R6
R5
A
R2
R1
LLEd logD7.4
solubility
(nM)a,b
210
(%)a,c
ꢀ
(ꢁM)e
Cl
H
H
H
H
H
H
H
S
S
S
S
OMe
OMe
SO2Me
H
H
H
ꢀ
> 3.9
> 4
< 0.71
0.78
1
2
3
4
76
16
ꢀ
H
120
37
4.4
2.5
2.5
< 0.47
3.0
SO2Me
4700
10
2.8
ACS Paragon Plus Environment
4

Page 5 of 32
Journal of Medicinal Chemistry
H
H
H
H
H
H
H
H
H
O
S
S
SO2Me
H
H
H
H
H
> 10000
44
ꢀ
3.5
4.7
5.2
5.6
4.8
1.1
2.7
2.3
1.5
2.8
72
64
71
240
ꢀ
5
6
7
8
9
1
2
3
4
5
6
7
SO2(CH2)2ꢀpyrrolidꢀ1ꢀyl
SO2(CH2)2ꢀNHiPr
65
82
51
ꢀ
27
83f
S SO2(CH2)2ꢀ4ꢀMeꢀpiperazinꢀ1ꢀyl
pyrrolꢀ1ꢀyl S
SO2Me
23
aMean of at least 2 independent measurements, unless stated. pIC50 SEM <0.1 for cpds 1ꢀ5, pIC50 SEM <0.5 for cpds 6,7
b
8
9
c
d
and 9. Maximum % activation, maximum concentration = 10 ꢁM. Ligand lipophilicity efficiency, defined as GHSꢀR1a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
e
binding pIC50 ꢀ logD7.4. Aqueous solubility measured at pH 7.4, performed under thermodynamic conditions from solid
samples. fn=1.
Further modification of the basic sulfonyl sidechain was conducted with the aim of increasing
GHSꢀR1a affinity. A key SAR finding was that extension of the alkyl linker to three carbon atoms gave
a step increase in potency (Table 2). An additional benefit of this homologation was to improve
chemical stability by removing the potential for βꢀamino sulfones to eliminate the amino group. High
LLE and good solubility was also achieved across the compound set with piperazine 14 exhibiting the
highest affinity. With the potential for CNS exposure in mind, MDCKꢀMDR1 (MadinꢀDarbyꢀcanineꢀ
kidney cells transfected with MDR1, the human version of Pꢀglycoprotein) permeability and efflux
ratios were obtained.33 High MDCKꢀMDR1 permeability (Papp > 5) and low efflux (ratio < 5) were
obtained except for diazepine 15, which was a notable outlier exhibiting low permeability and high
efflux. Again, all analogues were partial agonists, thus a couple of potent basic sidechains were selected
for a more extensive assessment of SAR in combination with phenyl group modifications (Table 3).
Table 2. Optimising basic side chain for potency.
O
O
S
N
O
Cl
O
R
S
N
N
H
H
h GHSꢀR1a
binding IC50
(nM)a,b
MDCKꢀMDR1 Papp
AꢀB (x10ꢀ6 cm/s)
/ efflux ratiof
Agonist
Aq.
Cpd
R
LLEd logD7.4
effect (%)a,c
solubility (ꢁM)e
ꢀ(CH2)3NMe2
ꢀ(CH2)3NEt2
8.7
13
71
64
6.4
6.2
1.7
1.7
240
280
13 / 1.4
5.4 / 2.7
10
11
N
65
8.6
1.3
71
51
77
5
2.2
1.3
1.8
69
87
17 / 0.9
ꢀ
12
13
14
ꢀ(CH2)3NH(CH2)2OMe
6.8
7.1
(CH2)3
N
N
280
7.1 / 4.2
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 32
(CH2)3
N
N
1
4.2
77
7.4
1.0
340
0.7 / 41
15
2
aMean of at least 2 independent measurements. pIC50 SEM <0.27 for all examples. Maximum % activation, maximum
3
4
5
b
c
d
e
concentration = 10 ꢁM. Ligand lipophilicity efficiency, defined as GHSꢀR1a binding pIC50 ꢀ logD7.4. Aqueous solubility
f
6
7
8
measured at pH 7.4, performed under thermodynamic conditions from solid samples. Compounds were incubated at 10 ꢁM
in cultured MDCKꢀMDR1 cells and permeability was measured in both the A to B and B to A directions
9
The improvement in potency shown by the pyrrole group at R5 (9) indicated that this position
was worth further exploration with different groups and additional aryl substituents. Small R5
substituents (compounds 16ꢀ18) maintained partial agonism and ethyl analogue 16 showed a 10ꢀfold
affinity increase concomitant with a lipophilicity increase that maintained LLE similar to the
unsubstituted compound 14. The larger ethoxy analogue 19 exhibited reduced partial agonism while
still maintaining subꢀnanomolar affinity. As expected, the combination of R5 pyrrole and a basic side
chain yielded a good compound (20) which exhibited partial inverse agonism and subꢀnanomolar
affinity. Encouraged by this finding, we prepared other 5ꢀ and 6ꢀmembered rings, with pyrrolidine 21,
morpholine 22, pyridine 23 and pyrazole 24 all showing high levels of inverse agonism and subꢀ
nanomolar affinity. The SAR change from partial agonism to inverse agonism was specific and subtle.
For example, moving the pyrazole to the R3 position (compound 25), a large structural change, oblated
the inverse agonism with a 10ꢀfold reduction in affinity. However, the small addition of a 3ꢀmethyl
substituent to the pyrazole to give 26 also lost the inverse agonist activity but with maintained affinity.
Moving the 3ꢀmethylꢀpyrazole to the R4 position (compound 27) resulted in a partial agonist with
maintained affinity and further substitution to give the 3,5ꢀdiꢀmethylꢀpyrazole 28 resulted in high
affinity and weak partial agonism. Consequently, to continue optimisation we focused on unsubstituted
5/6ꢀmembered rings at R5 to assess the scope for maintaining inverse agonism with an additional group
at R4. Gratifyingly, we were able to add diverse small substituents at R4 while maintaining high
inverse agonism and subꢀnanomolar affinity for compounds 29ꢀ31. Combining R4 and R5 substitutents
through cyclisation in methylꢀindole 32, did not result in inverse agonism although subꢀnanomolar
affininty was maintained, which was consistent with a R5 substituent of appropriate size being the
dominant determinant of inverse agonism. Three of the R5 groups giving highest inverse agonists were
synthesized with a diꢀethylamine at the opposite end of the molecule rather than a methylꢀpiperazine
(compounds 33ꢀ35). Affinity was reduced with diꢀethylamine compared to methylꢀpiperazine as
expected but surprisingly the pyrrole 33 no longer showed any inverse agonism while both pyrazole 34
and morpholine 35 maintained high inverse agonism. The subtle functional SAR we have uncovered for
this chemical series appears to be heavily dependent on the R5 substituent but is also influenced by
distal functional groups on the molecule. This makes predicting the functional activity of this series
difficult and encouraged us to not settle on any one particular group but rather make a number of
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
6

Page 7 of 32
Journal of Medicinal Chemistry
combinations of groups at both ends of the molecule. In general, our findings are entirely consistent
with reported mutagenesis studies where minor changes in peptide sequence had a major impact on
biological activity at the ghrelin receptor.34
1
2
3
4
5
6
7
8
Table 3. Optimising for inverse agonism.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
R= A
R= B
N
N
N
S
O
O
N
Cl
O
O
R3
R4
S
N
N
H
H
R5
h GHSꢀ
R1a
MDCKꢀ
Inverse
agonist
effect
Agonist
MDR1 Papp Rat po
LLEd logD7.4
AꢀB (x10ꢀ6 Cmax
Rat
free
B:Ph
Rat
Cpd R
R3
R4
R5
Et
binding
IC50
effect
B:Pg
(%)a,c
cm/s) /
efflux ratioe
1.7 / 13
(ꢁM)f
(%)a,c
(nM)a,b
0.078
A
A
A
A
H
H
H
H
H
H
H
H
0
0
0
1
58
84
57
17
7.2
6
2.9
2.3
2.9
2.4
1.5
1.4
1.4
4.2
0.036 0.04
16
17
18
19
HC≡C
cPr
5.1
3.3 / 3.3
2.0 / 9.1
2.8 / 5.0
< 0.03
0.18
ꢀ
0.33
0.31
6.6
7.1
1.2
EtO
0.052 0.52
N
N
A
A
H
H
H
H
0.15
0.29
17
81
0
0
7.2
6.5
2.6
3.1
0.37 / 38
4.5 / 0.55
0.58
ꢀ
ꢀ
ꢀ
20
21
0.64 < 0.03
N
O
A
H
H
H
0.77
84
0
7.4
1.7
0.53 / 69
1.2
< 0.03
ꢀ
22
A
A
A
H
H
H
N
0.69
0.67
8.8
73
76
0
0
ꢀ
6.6
7.1
6.5
2.5
2
0.52 / 19
2.9 / 7.6
0.91 / 25
1
ꢀ
< 0.03
ꢀ
ꢀ
ꢀ
ꢀ
23
24
25
N
N
H
1.2
N
H
ꢀ
1.6
0.012
N
N
N
A
H
H
0.53
0
ꢀ
6.7
2.6
0.68 / 14
0.41
ꢀ
ꢀ
26
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 32
1
2
N
A
H
N
H
0.48
0
42
6.7
2.6
3.5 / 0.86
0.22
ꢀ
ꢀ
27
3
4
5
6
7
8
N
N
A
A
A
H
H
H
H
1.3
0
21
0
6.4
7
2.5
2.2
2.3
0.44 / 76
0.86 / 46
0.40 / 55
0.078
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
28
29
30
9
N
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
MeO
Cl
0.62
0.31
76
75
N
N
ꢀ
7.2
0.03
N
N
A
A
H
H
N≡C
0.24
0.57
73
0
ꢀ
ꢀ
6.8
6.4
2.8
2.8
NV
ꢀ
ꢀ
ꢀ
31
32
Cl
acyl urea
3.9 / 0.71
0.3
0.049 0.04
N
N
N
B
B
H
H
H
H
0.43
3.9
0
ꢀ
ꢀ
6.4
6.3
2.9
2.1
0.10 / 42
0.71 / 12
3.6
1.6
ꢀ
ꢀ
ꢀ
ꢀ
33
34
75
N
N
O
B
H
H
2.9
75
ꢀ
6.7
1.8
3.3 / 5.5
9.4
< 0.01
ꢀ
35
aMean of at least 2 independent measurements. pIC50 SEM <0.24 for all examples. Maximum % activation, maximum
concentration = 10 ꢁM. Ligand lipophilicity efficiency, defined as GHSꢀR1a binding pIC50 ꢀ logD7.4. eCompounds were
b
c
d
incubated at 10 ꢁM in cultured MDCKꢀMDR1 cells and permeability was measured in both the A to B and B to A directions.
fCompound was dosed PO at 2 mg/kg in 1% Pluronic F127. gRatio of the drug concentrations in homogenized brain (B)
h
tissue and plasma (P), 1 h after IV bolus then IV infusion. Calculated using free fraction measured in plasma and brain
(using either brain slice method35 or brain homogenate)
For a number of compounds oral exposure in rat was assessed and modest to good levels were
obtained for most compounds, including both agonists and inverse agonists (Table 3). Assessment of
CNS exposure was conducted using rat infusion studies and it became apparent that generally very low
levels were achieved in brain although compounds 18 and 19 did exhibit a free brain:plasma (B:P) ratios
of 1.2 and 0.52 respectively. This generally low CNS exposure could be related to MDR1 mediated
efflux in some cases and in all cases to low MDCKꢀMDR1 permeability (Papp < 5). Notably morpholine
22 which was viewed as the lead inverse agonist at this time, had highest LLE and showed high oral
exposure, but no detectable level in brain.
ACS Paragon Plus Environment
8

Page 9 of 32
Journal of Medicinal Chemistry
Evaluation of morpholine 22 against chemical properties generally associated with CNS
1
2
medicinal agents revealed that efflux, polar surface area (PSA)36 and rotatable bonds were parameters
all outside the normal range.37 Consequently, our strategy to improve brain exposure involved
increasing MDCKꢀMDR1 permeability (Papp > 5) and reducing MDCKꢀMDR1 efflux (ratio < 3)
through modification of the basic sidechain. We rationalized that reduced basicity could improve
permeability and structural diversity may avoid efflux transporter recognition. The PSA of morpholine
22 was calculated as 124 Å but we reasoned that an intramolecular hydrogen bonding network involving
the acylurea would exist. This effectively hides significant polarity making the actual PSA much less
than calculated, and also effectively reduces hydrogen bond donor/acceptor and rotatable bond counts,
so we were less concerned with reducing these three parameters. The flexible basic sidechain of 22
contributes to a high rotatable bond count of 12 and we targeted to reduce this. With these design
considerations in mind, piperidinylꢀsulfone 12 stood out from earlier compounds (Table 2) as containing
the base most likely to afford CNS exposure going forward, with low efflux and lower rotatable bond
count compared to other basic sidechains explored. Although affinity of this basic group was not as
great as others, we rationalized that combination with optimal Clꢀaryl substituents could yield a
compound with overall balanced properties. Consequently, a small set of piperidinylꢀsulfone
compounds were prepared and profiled (Table 4). Substitution at R5 did improve affinity in all
compounds and indeed this subset of compounds (36ꢀ40) displayed higher inverse agonism (80ꢀ96%)
than previously seen, with the exception of methylꢀindoline 41 which showed 69% inverse agonism.
We had changed PK species from rat to mouse with a view to conducting efficacy studies in mouse and
good oral exposure was obtained for this series. Some improvement in MDCKꢀMDR1 permeability and
efflux was achieved, notably with morpholine 38, which met our CNS design criteria. Gratifyingly, this
did translate into significant brain exposure as measured in mouse (free B:P = 0.42). In general mouse
brain data was consistent with rat infusion brain studies, for example compound 38 gave a free B:P =
1.3 in rat. A single crystal of 38 was obtained and Xꢀray crystallographic analysis38 confirmed that the
expected internally hydrogen bonded conformation did exist, where the hydrogen on the N connected to
the benzothiazole interacts with the oxygen of the carbonyl connected to the aryl ring (Figure 1).
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Xꢀray structure of 38.
Table 4. Optimising for brain exposure.
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 32
O
S
1
2
3
4
5
6
7
N
Cl
O
O
N
O
R3
R4
S
N
N
H
H
R5
8
9
h GHSꢀ
R1a
MDCKꢀ
Inverse
agonist
effect
MDR1 Papp Mouse po
Mouse
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Mouse
AꢀB (x10ꢀ6
cm/s) / efflux
ratioe
Cmax
free
Cpd R3
R4
R5
LLEd logD7.4
binding
IC50
B:P (dose)g
(ꢁM)f
B:Ph
(%)a,c
(nM)a,b
N
H
H
H
H
15
33
96
81
5.3
4.9
2.5
2.6
3.9 / 2.5
1.5
ꢀ
36
37
ꢀ
N
0.90
(2 mg/kg)
N
1.1 / 15
0.64
0.61
N
N
O
0.21
(7 mg/kg)
H
H
6.7
92
5.9
2.3
5.4 / 2.6
6.5
0.42
38
N
H
H
H
0.92
0.4
88
81
5.9
6.4
3.1
3
11 / 0.05
3 / 0.47
5.8
ꢀ
39
40
ꢀ
0.072
(0.9 mg/kg)
N
N≡C
0.84
0.38
Cl
acyl urea
0.15
(2.7 mg/kg)
5.3
0.072
H
7.2
69
2.8
5.1 / 1.2
3
41
N
aMean of at least 2 independent measurements. pIC50 SEM <0.20 for all examples. Maximum % activation, maximum
concentration = 10 ꢁM. Ligand lipophilicity efficiency, defined as GHSꢀR1a binding pIC50 ꢀ logD7.4. eCompounds were
b
c
d
incubated at 10 ꢁM in cultured MDCKꢀMDR1 cells and permeability was measured in both the A to B and B to A directions.
fCompound was dosed PO at 2 mg/kg in 1% Pluronic F127. gRatio of the drug concentrations in homogenized brain (B)
h
tissue and plasma (P), 6 h after PO dosing. Calculated using free fraction measured in mouse plasma and rat brain (using
either brain slice method35 or brain homogenate).
In an attempt to understand the binding mode and subtle changes in SAR for functional activity, we
created a homology model of GHSꢀR1a and docked our compounds into this model. The binding mode
of these compounds is illustrated by compound 38 in Figure 2. The model predicts that the acylurea
subunit makes several hydrogenꢀbonding interactions with Asp99, Arg102 (both TM2) and Asn305
(TM7). The sulfone group makes an interaction with Arg199 while the basic piperidine is pushed into a
subpocket between TM4, TM5 and TM6. At the other end of the molecule, the bulk of the chlorine
ACS Paragon Plus Environment
10

Page 11 of 32
Journal of Medicinal Chemistry
atom causes the phenyl ring to twist out of the plane of the rest of the molecule, placing the morpholine
group into a subpocket located between TM1, TM2 and TM7.
1
2
3
4
5
6
7
8
Figure 2. 38 bound into the ligand binding domain of the human GHSꢀR1a homology model. (a) 3D
structure of the pocket, orientated down the transmembrane axis, viewed from the extracellular side, key
residues and helices are labelled; (b) 3D structure of the pocket, orientated across the transmembrane
axis; (c) 2D depiction of the binding mode and key interactions; (d) schematic image of the receptor
showing the ligand braced between TM1 (red) and TM 5 (lightꢀblue).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(c)
(a)
(d)
(b)
In terms of functional understanding of what favors agonism or inverse agonism, the proposed binding
mode and mechanism of activation appears to be consistent with that proposed for the β2 adrenoceptor
where TM5 moves towards the centre of the pocket upon activation.39 In our model that region is
occupied by the piperidine group, preventing this helix movement. In this context, the effect of the
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 32
morpholine group must also be invoked in order to fully explain inverse agonism SAR. It is inserted
into a subpocket adjacent to TM1 which locks the binding mode, preventing lateral movement of the
ligand that might allow TM5 to shift into the agonist conformation. Effectively, the molecule is a rigid
unit, braced between TM1 and TM5. When the length of the brace is correct an inactive conformation
of the receptor is induced and we observe inverse agonism. It is clear from the data presented that
relatively bulky substituents at both ends of the molecule are required. However, inverse agonism is
easily lost and subtle changes at either end of the molecule may result in subꢀoptimal interactions and
allow partial or even full agonism. This is in agreement with our observed structureꢀfunction
relationships in this series.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis
Several synthesis routes were utilized in the preparation of acylureas (Scheme 1). In general,
substituted benzamides 42 were reacted with oxalyl chloride in THF at 120 °C under microwave
irradiation. Addition of the appropriate aminobenzothiazole 44 to the intermediate acyl isocyanate 43
and again heating to 120 °C gave the desired acylureas 46. For more highly substituted substrates or
those with acid sensitive functional groups present, an alternative method was employed. This involved
synthesis of a carbamate intermediate such as 45 followed by microwave irradiation with the
corresponding benzamide 44 under basic conditions. Alternatively benzoyl carbamates 47 were
prepared, followed by reaction with the aminobenzothiazole 44 under neutral thermal conditions.40
Scheme 1. General synthesis of acylureasa
Cl
O
Cl
O
O
(i)
NH2
(ii)
N
R
R
O
44
42
43
O
O
Cl
(iv)
O
N
S
(iii)
N
S
N
H2N
N
N
H
R'
N
R'
H
42
S
R'
S
H
S
S
O
R
O
O
O
O
O
44
45
46
(vi)
Cl
O
O
O
Cl
Ph
(v)
NH2
O
N
44
H
R
R
42
47
aReagents and conditions: (i) oxalyl chloride, THF, microwave, 120 °C, 1 h; (ii) DIPEA, rt, 1 h or microwave, 120 °C, 5 min;
(iii) DIPEA, THF, isoprenylchloroformate, 0 °C; (iv) tBuOK, THF, reflux, 48 h; (v) NaH, (PhO)2C=O, THF, 1 h; (vi) 100
°C, 30 min.
ACS Paragon Plus Environment
12

Page 13 of 32
Journal of Medicinal Chemistry
To access the piperazine derivatives 14 and 22 we first needed to prepare the iodopropyl substituted
aminobenzothiazole derivative 51 (Scheme 2). This was achieved in three steps from 48 which was
converted to the thioether by reaction with 1ꢀchloroꢀ3ꢀiodopropane. Subsequent oxidation to the sulfone
with mCPBA and conversion to the iodide via a Finkelstein reaction gave 51. The iodopropyl
substituted acylurea 55 was prepared as above, and then without isolation this was converted to the
methylpiperazine 22. A similar procedure was followed for derivative 14 via isolation of intermediate
54.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Synthesis of 14 and 22a
N
N
S
N
S
Cl
I
Cl
N
S
(i)
(ii)
(iii)
H2N
H2N
H2N
H2N
S
S
N
S
O
S
O
SH
O
O
48
49
50
51
O
S
O
S
Cl
R
O
Cl
R
O
O
N
Cl
R
O
O
O
(vi)
(iv), (v)
O
NH2
S
N
N
N
S
N
N
I
H
H
H
H
N
52 R = H
53 R = morpholine
54 R = H
55 R = morpholine
14 R = H
22 R = morpholine
aReagents and conditions: (i) K2CO3, MeCN, 1ꢀchloroꢀ3ꢀiodopropane; (ii) mCPBA, DCM, 38% (2 steps); (iii) NaI, acetone,
reflux, 90%; (iv) oxalyl chloride (1 eq.); THF, microwave 120 °C, 5 min; (v) 51, microwave 120 °C, 5 min, 35% (R=H, 2
steps); (vi) Nꢀmethylpiperazine, rt, 16 h, 58% (R=H), 16% (R=morpholine, 3 steps).
The methyl piperidine derivative 38 was accessed by first preparing the required aminobenzothiazole
derivative 59 (Scheme 3). Reaction of the thiol 56 with mesylate 57 under reducing conditions to
prevent formation of the disulfide gave the thioether 58. This was oxidised to sulfone 59 and then
converted to the acylurea as described previously which was subjected to EschweilerꢀClarke conditions
that removed the protecting group and methylated the piperidine to give 38.
Scheme 3. Synthesis of 38a
O
N
S
OMs
N
Cl
O
O
N
O
O
S
N
S
(iii), (iv)
N
N
(i)
N
S
N
O
H
H
H2N
+
H2N
SH
A
N
O
38
O
O
58 A = S
59 A = SO2
56
57
(ii)
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 32
aReagents and conditions: (i) K2CO3, NaBH4, MeCN/EtOH (9:1), 80 °C, 16 h, 92%; (ii) mCPBA, DCM, rt, 45 min, 68%;
1
2
(iii) 53, oxalyl chloride, THF, 120 °C, 15 min, then 59, 120 °C; (iv) HCOOH, HCHO, 100 °C, 3 h, 15% (2 steps).
3
4
5
6
7
Pharmacokinetic properties and Pharmacology of 14, 22 and 38.
8
9
Compounds 14, 22 and 38, which covered the range of in vitro function from partial agonism through to
near full inverse agonism were profiled in vivo and demonstrated suitable PK properties in rat, mouse
and dog (Table 5). Clearance was generally low, with 14 exhibiting the highest clearance of the three
compounds, but relatively high plasma protein binding (PPB) free levels for 14 translated into low
unbound clearance. Volume was consistent across the three species for each of the compounds and all
showed sufficient oral exposure and bioavailability to be considered as useful in vivo tool compounds.
Where measured, the affinity against rat and mouse isoforms of GHSꢀR1a was consistent with the data
previously obtained for the human isoform (Table 5).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 5. Pharmacokinetic data for key in vivo compounds.
po
PPB %
free
iv/po dose
(mg/Kg)a
iv CL
iv Vss
GHSꢀR1a binding
IC50 (nM)b,c
Cpd
Species
Cmax F (%)
(ꢁM)
(mL/min/Kg) (L/Kg)
Rat
Mouse
dog
8
2/2
2/5
1/1
2/2
2/1.3
2/2
2/2
2/7
2/2
24
35
4.9
4.5
4
0.34
0.71
0.2
0.58
1.3
11
125
3.6
4.9
ꢀ
13
40
14
9.6
0.35
0.6
0.78
0.2
0.9
1.6
8.1
2.6
2.9
0.69
1.9
1.8
10
51
Rat
0.9
0.8
0.3
0.7
0.4
0.4
30
3.4
1.1
ꢀ
Mouse
dog
50
22
38
87
Rat
4.5
8
171
25
ꢀ
Mouse
39
dog
1.0
31
ꢀ
aCompound was dosed IV in 5% DMSO:95% hydroxypropyl beta cyclodextrin; po in 1% Pluronic F127. bMean of at least 2
independent measurements. cpIC50 SEM <0.28 for all examples.
The three tool compounds were tested in mice to investigate the effects on normal food intake.
Ad libitum fed mice were dosed to achieve maximum oral exposure at a time coinciding with peak food
intake, at the beginning of the dark phase, and food intake measured in twoꢀhourly intervals (Figure 3aꢀ
c). In this model, GHSꢀR1a agonist compounds including ghrelin and hexarelin induced a robust
increase in food intake (data not shown) and this effect was replicated by the smallꢀmolecule partial
agonist 14. The effects of inverse agonists in this model appeared dependent on CNS penetration, as
demonstrated by comparison of nonꢀCNS penetrant inverse agonist 22 with CNSꢀpenetrant inverse
agonist 38. At doses achieving significant free plasma cover over the receptor IC50, 22 achieved very
ACS Paragon Plus Environment
14

Page 15 of 32
Journal of Medicinal Chemistry
low CNS exposure, close to the levels from residual blood contamination of brain tissue, while 38
achieved free brain levels that gave free brain cover close to receptor IC50 (Table 6). Compound 38
exhibited significant reduction in free feeding at each of the three 2 h time bins during the first 6 h of the
study, while 22 exhibited a significant reduction in the earliest time bin but no effect during the 2ꢀ6 h
time interval. There is a clear difference between the behavior of these two inverse agonists and the 0ꢀ2
h efficacy of 22 may be explained by a small amount of compound in the brain due to very high plasma
exposure overcoming the high efflux observed for this compound. It is clear that free brain cover, rather
than plasma cover, is driving the more significant 6 h effect shown by 38. Neither inverse agonist has a
significant effect in reducing food intake over a 24 h period, although partial agonist 14 did increase
cumulative food intake over a 24 h period by 19% (Figure 3d).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Effect of compounds 14, 22 and 38 on food intake in mouse free feeding modela
(a) partial agonist 14 (30 mg/Kg)
(b) nonꢀCNS penetrant inverse agonist 22 (100 mg/Kg)
1.5
1.5
Vehicle
Vehicle
p=0.07
11%
14 (30 mg/kg)
22 (100 mg/kg)
1.0
**
30%
1.0
0.5
0.0
**
97%
0.5
**
54%
0.0
0-2h
2-4h
4-6h
0-2h
2-4h
4-6h
Time after dosing
Time after dosing
(c) CNSꢀpenetrant inverse agonist 38 (50 mg/Kg)
(d) 0ꢀ24 h food intake
6
**
19%
1.5
Vehicle
Treated
Vehicle
5
38 (50 mg/kg)
4
3
2
1
0
1.0
0.5
0.0
p=0.08
13%
**
20%
*
37%
14 (30 mg/kg) 38 (50 mg/kg) 22 (100 mg/kg)
Drug treatment
0-2h
2-4h
Time after dosing
4-6h
aMean values ± SEM (*p<0.05, **p<0.01, ***p<0.001)
Table 6. PK multiples for compounds 14, 22 and 38 in mouse free feeding modela
free plasma multiplea
1 h 2 h 6 h
ACS Paragon Plus Environment
free brain multipleb
Cpd
po dose (mg/Kg)
1 h
2 h
6 h
15

Journal of Medicinal Chemistry
Page 16 of 32
30
100
50
310
870
14
220
860 120 B:P < 0.01c B:P < 0.01c B:P < 0.01c
13 3.5 0.40 0.43
20
ꢀ
ꢀ
ꢀ
14
22
38
1
2
3
4
5
6
7
8
ꢀ
aOral exposure measured in plasma corrected for mouse PPB divided by mouse binding affinity. Brain exposure corrected
for rat brain %free divided by mouse binding affinity. cfree B:P not determined because CNS exposure below a level that can
be detected.
b
9
The specificity of the foodꢀintake effect of 38 was investigated using genetically altered mice
lacking the GHSꢀR1a receptor (Figure 4). In the wildꢀtype (WT) mice, similar reductions in food intake
over a 6 h period were observed compared to the previous experiments. In the null (KO) mice, no
significant effects on food intake were observed, confirming that the effect of 38 is mediated via GHSꢀ
R1a. Furthermore, in vitro pharmacological profiling41 of 38 against a panel of over 80 diverse targets
was conducted and exhibited excellent selectivity (see supplementary information). Two targets
showed activity below 1 ꢁM (dopamine transporter, Ki = 0.29 ꢁM; nicotinic acetylcholine receptor α4,
Ki = 0.052 ꢁM). A recent reported ghrelin inverse agonist was designed to avoid activity at the
muscarinic receptor M2.28 Although 38 is an antagonist at M2, it has 550ꢀfold selectivity. In addition,
38 was inactive against hERG measured up to a maximum concentration of 33 ꢁM. Taking all this data
together, 38 is clearly an excellent in vivo tool compound for exploring the inverse agonist
pharmacology of the ghrelin receptor, and further studies will be reported in due course.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Effect of 38 (100 mg/Kg) on food intake in KO/WT mouse free feeding modela
60
WT 38
KO 38
40
20
0
*
-6 -4 -2
0
2
4
6
8 10 12 14 16 18 20 22 24
-20
-40
-60
-80
*
**
Time after dosing (h)
aMean values ± SEM (*p<0.05, **p<0.01, ***p<0.001)
Conclusion
Starting from HTS hit 1, with moderate GHSꢀR1a affinity, we increased affinity and improved
physicochemical properties by incorporation of a basic sulfone sidechain to afford partial agonist 14.
Further SAR optimization led to identification of inverse agonist 22 through discovery that adding a 5ꢀ
or 6ꢀmembered ring at an appropriate position converted partial agonists into inverse agonists.
Reduction of rotatable bond count and optimization of MDCKꢀMDR1 efflux led to increased brain
ACS Paragon Plus Environment
16

Page 17 of 32
Journal of Medicinal Chemistry
exposure in inverse agonist 38. These three tool compounds showed good murine affinity for GHSꢀR1a
and oral exposure and were dosed to free feeding mice. Partial agonist 14 caused increased acute food
intake, while significant reduction of food intake over 6 h was observed with CNS penetrant inverse
agonist 38 but not with nonꢀCNS penetrant inverse agonist 22. 38 also showed acute reduction of food
intake in WT mice but not in KO mice. These results are consistent with free brain cover over GHSꢀ
R1a IC50 being necessary for reduction of food intake, and that this effect occurs through a mechanism
involving GHSꢀR1a.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experimental Section
General. All solvents and chemicals used were reagent grade. Anhydrous solvents THF,
benzene, DCE, DCM and DME were purchased from Aldrich. Flash column chromatography was
carried out using prepacked silica cartridges (from 4 g up to 330 g) from RedisepTM or Silicycle and
eluted using an Isco Companion system. Purity and characterization of compounds were established by
a combination of liquid chromatographyꢀmass spectroscopy (LCꢀMS), gas chromatographyꢀmass
spectroscopy (GCꢀMS) and NMR analytical techniques and was >95% for all test compounds. 1H NMR
were recorded on a Varian INOVA (600 MHz), Varian Gemini 2000 (300 MHz) or Bruker Avance
DPX400 (400 MHz) and were determined in CDCl3 or DMSOꢀd6. Chemical shifts are reported in ppm
relative to TMS (0.00 ppm) or solvent peaks as the internal reference. Splitting patterns are indicated as
follows: s, singlet; d, doublet; t, triplet; m, multiplet; br, broad peak. Elevated temperatures were used
where necessary to sharpen broad NMR peaks due to rotamers and the temperature used is noted for
such compounds. Merck precoated TLC plates (silica gel 60 F254, 0.25 mm, art. 5715) were used for
TLC analysis. Solutions were dried over anhydrous magnesium sulfate and the solvent was removed by
rotary evaporation under reduced pressure.
Synthesis of representative key examples (14, 22 and 38)
6-(3-Chloropropylsulfanyl)benzothiazol-2-amine (49).
To a stirred solution of 2ꢀ
aminobenzothiazoleꢀ6ꢀthiol (10.0 g, 54.9 mmol) in MeCN (250 mL) was added potassium carbonate
(11.4 g, 82.3 mmol) and 1ꢀchloroꢀ3ꢀiodopropane (6.19 mL, 57.6 mmol) and the reaction mixture was
heated at 85 °C for 1 h under nitrogen. The reaction mixture was cooled, filtered and the solid was
washed with MeCN (2 x 100 mL), then the filtrate was concentrated in vacuo and the resulting solid
was triturated with isohexane to give the title compound as a white solid (16.3 g) which was used
1
without further purification. H NMR (400 MHz, DMSOꢀd6) 1.89ꢀ1.97 (2H, m), 2.98 (2H, t), 3.72 (2H,
t), 7.23ꢀ7.30 (2H, m), 7.55 (2H, s), 7.75 (1H, s); m/z MH+ = 259.
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 32
6-(3-Chloropropylsulfonyl)benzothiazol-2-amine (50). To a stirred solution of 49 (16.3 g,
89.9 mmol) in DCM (250 mL) was added mCPBA (32.6 g, 189 mmol) and the reaction mixture was
stirred for 1 h, then washed with aqueous sodium metabisulfite (10% w/v, 100 mL) and saturated
aqueous sodium bicarbonate solution (200 mL). The organic phase was dried (MgSO4), filtered, and
concentrated in vacuo and then the resulting solid was triturated with isohexane to give the title
1
2
3
4
5
6
7
8
1
9
compound as an orange solid (9.9 g, 38%) that was used without further purification. H NMR (400
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
MHz, DMSOꢀd6) 1.95ꢀ2.02 (2H, m), 3.37ꢀ3.39 (2H, m), 3.68 (2H, t), 7.49 (1H, d), 7.67ꢀ7.70 (1H, m),
8.05 (2H, s), 8.25 (1H, d); m/z MH+ = 291.
6-(3-Iodopropylsulfonyl)benzothiazol-2-amine (51). To a stirred solution of 50 (6.0 g, 20.6
mmol) in acetone (100 mL) was added sodium iodide (30.9 mL, 206 mmol) and the reaction mixture
was heated at 55 °C for 16 h. The reaction mixture was cooled, filtered and concentrated in vacuo and
the residue was diluted with DCM (200 mL) and then washed with water (200 mL) and saturated brine
(100 mL). The organic phase was dried, filtered and concentrated in vacuo and then the resulting solid
was triturated with isohexane to give the title compound as an orange solid (7.1 g, 90%) that was used
1
without further purification. H NMR (400 MHz, DMSOꢀd6) 2.04 (2H, q), 3.27 (2H, t), 3.33 (2H, t),
7.48ꢀ7.51 (1H, m), 7.66ꢀ7.70 (1H, m), 8.06ꢀ8.07 (2H, s), 8.24 (1H, d); m/z MH+ = 381.
2-chloro-N-(6-(3-iodopropylsulfonyl)benzo[d]thiazol-2-ylcarbamoyl)benzamide (54). Oxalyl
chloride (2.71 mL, 31.6 mmol) was added to a stirred solution of 2ꢀchlorobenzamide (52) (4.97 g, 31.9
mmol), in THF (270 mL) warmed to 60 ºC, over a period of 5 min under nitrogen. The resulting
solution was stirred at 60 °C for 90 min then evaporated. The residue in THF (20 mL) was added
dropwise to a stirred solution of 51 (11.2 g, 29.4 mmol), and NꢀethylꢀNꢀisopropylpropanꢀ2ꢀamine (15.1
mL, 88.3 mmol) in THF (250 mL) over a period of 10 min under nitrogen. The resulting solution was
stirred at rt overnight, diluted with EtOAc (250 mL), and washed sequentially with water (2 x 100 mL),
and saturated brine (150 mL). The organic layer was dried (MgSO4), filtered, evaporated and purified by
flash silica chromatography, elution gradient (0ꢀ40% EtOAc in DCM) to afford the title compound as a
1
colorless solid (5.8 g, 35%). H NMR (400 MHz, DMSOꢀd6) 2.10ꢀ2.26 (2H, m), 3.30ꢀ3.48 (2H, m),
3.48ꢀ3.52 (2H, m), 7.49ꢀ7.75 (4H, m), 7.98ꢀ8.16 (2H, m), 8.24 (1H, br), 11.70ꢀ12.10 (2H, br); m/z MH+
= 563.
2-chloro-N-((6-((3-(4-methylpiperazin-1-yl)propyl)sulfonyl)benzo[d]thiazol-2-
yl)carbamoyl)benzamide (14). 1ꢀmethylpiperazine (1.71 mL, 15.4 mmol) was added 54 (5.8 g, 10.3
mmol) in THF (40 mL). The resulting solution was stirred at rt overnight then heated to 50 °C for 5 h.
The reaction mixture was diluted with EtOAc (150 mL), and washed sequentially with water (50 mL)
and saturated brine (20 mL). The organic layer was dried (MgSO4), filtered and evaporated to afford
ACS Paragon Plus Environment
18

Page 19 of 32
Journal of Medicinal Chemistry
crude product. The crude product was purified by flash silica chromatography, elution gradient 0 to 20%
MeOH in DCM. The crude product was triturated with Et2O to afford the title compound as a colorless
solid (3.2 g, 58%). 1H NMR (400 MHz, DMSOꢀd6) 1.69ꢀ1.73 (1H, m), 2.21 (4H, s), 2.30 (5H, q), 2.38
1
2
3
4
5
6
13
(4H, s), 3.16 (2H, s), 3.33 (1H, s), 7.43ꢀ7.59 (4H, m), 7.83ꢀ7.88 (2H, m), 8.52 (1H, d); C NMR (700
7
8
9
MHz, DMSOꢀd6) δ 44.7, 51.4, 53.1, 54.0, 54.9, 119.9, 122.2, 125.1, 127.0, 128.8, 129.5, 129.6, 129.6,
131.5, 132.4, 132.6, 135.3, 152.4, 152.7, 165.0, 168.4; HRMS (ES+) for C23H27ClN5O4S2 (MH+): calcd
536.1188; found, 536.1193.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2-Chloro-5-morpholin-4-yl-benzamide (53). 2ꢀChloroꢀ5ꢀfluorobenzamide (1.0 g, 5.76 mmol)
and morpholine (2.51 mL, 28.8 mmol) were dissolved in NMP (10 mL) and the reaction mixture was
heated at 180 °C for 16 h then cooled, diluted with H2O (100 mL) and extracted with EtOAc (3 x 75
mL). The combined organic phases were dried, filtered and concentrated to provide the crude product.
Purification by flash silica chromatography elution gradient (0ꢀ100% EtOAc in isohexane) and
1
trituration with Et2O afforded the title compound as a white solid (0.22 g, 15%): H NMR (300 MHz,
DMSOꢀd6) 3.11 (4H, t), 3.72 (4H, t), 6.93ꢀ6.99 (2H, m), 7.26 (1H, d), 7.47 (1H, s), 7.74 (1H, s); m/z
MH+ = 241.
N-(6-(3-(4-methylpiperazin-1-yl)propylsulfonyl)benzo[d]thiazol-2-ylcarbamoyl)-5-
morpholinobenzamide (22). 53 (530 mg, 2.20 mmol) and oxalyl chloride (0.19 mL, 2.20 mmol) were
suspended in THF (15 mL) and sealed into a microwave tube. The reaction was heated to 120 °C for 5
min in the microwave reactor, then cooled to rt. The vial was cautiously degassed and opened. 51 (758
mg, 1.98 mmol) was added portionwise and the suspension was stirred at 120 °C in the microwave for a
further 5 min. The vial was cooled to rt and cautiously degassed. 1ꢀmethylpiperazine (0.733 mL, 6.61
mmol) was added portionwise and the resulting suspension stirred at rt for 16 h. The reaction mixture
was evaporated to dryness and the residue was purified by flash silica chromatography elution gradient
1
(1ꢀ15% MeOH in DCM) to afford the title compound as a beige solid (220 mg, 16%). H NMR (400
MHz, DMSOꢀd6) 1.71ꢀ1.78 (2H, m), 2.38ꢀ2.47 (9H, m), 2.60 (4H, s), 3.18ꢀ3.20 (2H, m), 3.37 (4H, t),
3.75 (4H, t), 7.11ꢀ7.14 (1H, m), 7.22 (1H, d), 7.40 (1H, d), 7.91ꢀ7.99 (2H, m), 8.65 (1H, d), 11.14 (2H,
13
br s); C NMR (700 MHz, DMSOꢀd6) δ 19.9, 44.6, 47.8, 51.3, 53.1, 54.0, 54.9, 65.8, 114.9, 117.8,
118.8, 120.2, 122.4, 125.3, 130.0, 132.4, 132.9, 134.8, 149.4, 152.0, 152.3, 163.9, 168.4; HRMS (ES+)
for C27H34ClN6O5S2 (MH+): calcd 621.1715; found, 621.1719.
tert-Butyl 4-(2-aminobenzothiazol-6-ylthio)piperidine-1-carboxylate (58). To a solution of
tertꢀbutyl 4ꢀ(methylsulfonyloxy)piperidineꢀ1ꢀcarboxylate42 (57) (20.9 g, 74.8 mmol) in MeCN (900 mL)
was added EtOH (100 mL), 2ꢀaminobenzothiazoleꢀ6ꢀthiol (56) (13.6 g, 74.8 mmol), potassium
carbonate (13.4 g, 97.3 mmol) and sodium borohydride (8.50 g, 224 mmol). The suspension was heated
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 32
at 80 °C for 16 h under nitrogen. The reaction mixture was cooled, concentrated in vacuo and the
residue was partitioned between H2O (900 mL) and DCM (900 mL). The aqueous phase was extracted
with DCM (500 mL) and the combined organic phases were dried and concentrated to give the title
1
2
3
4
5
6
1
compound as a yellow solid (25.1 g, 92%). H NMR (400 MHz, DMSOꢀd6) 1.29ꢀ1.34 (2H, m), 1.38
7
8
9
(9H, s), 1.81ꢀ1.85 (2H, m), 2.85ꢀ2.87 (2H, m), 3.20 (1H, m), 3.81 (2H, m), 7.28 (2H, d), 7.54 (2H, s),
7.77 (1H, t); m/z (MꢀtBu+H)+ = 310.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
tert-Butyl 4-(2-aminobenzothiazol-6-ylsulfonyl)piperidine-1-carboxylate (59). To a solution
of 58 (31.8 g, 87.0 mmol) in DCM (900 mL) was added mCPBA (43.1 g, 183 mmol) portionwise. The
solution was stirred for 45 min before addition of aqueous sodium metabisulphite (20% w/v, 500 mL).
The organic phase was separated, washed with saturated aq. NaHCO3 solution, dried (MgSO4) and
filtered. A precipitate formed on standing, which was filtered to give the title compound (23.4 g, 39%).
The filtrate was concentrated to 100 mL. Another precipitate formed on standing which was filtered to
1
give a second crop of the title compound (9.9 g, 29%). H NMR (400 MHz, DMSOꢀd6) 1.31ꢀ1.39 (11H,
m), 1.83ꢀ1.86 (2H, m), 2.68ꢀ2.71 (2H, m), 3.38 (1H, t), 3.98ꢀ4.01 (2H, m), 7.49 (1H, d), 7.61ꢀ7.64 (1H,
m), 8.01 (2H, s), 8.17 (1H, d); m/z (MꢀH)ꢀ = 396.
2-Chloro-N-(6-(1-methylpiperidin-4-ylsulfonyl)benzothiazol-2-ylcarbamoyl)-5-
morpholinobenzamide (38). 53 (0.72 g, 3.0 mmol) and oxalyl chloride (0.28 mL, 3.15 mmol) were
suspended in THF (15 mL) and heated at 120 °C in a microwave for 5 min. The reaction mixture was
cooled and 59 (1.07 g, 2.70 mmol) was added. The reaction mixture was heated at 120 °C in a
microwave for 5 min. The reaction mixture was cooled and concentrated. The residue was added to a
mixture of acetyl chloride (10 mL) and MeOH (50 mL) (CAUTION, exotherm), stirred for 2 h and
concentrated in vacuo. The residue was diluted with DCM (50 mL) and saturated aq. NaHCO3
solution. The suspension was filtered to give crude 2ꢀchloroꢀ5ꢀmorpholinoꢀNꢀ(6ꢀ(piperidinꢀ4ꢀ
ylsulfonyl)benzothiazolꢀ2ꢀylcarbamoyl)benzamide (1.02 g, 1.81 mmol). This was dissolved in formic
acid (20 mL) and formaldehyde (37% w/w in H2O, 2.0 mL) added. The solution was heated at 100 °C
for 3 h. The reaction mixture was concentrated in vacuo and the residue was dissolved in water. The
solution was neutralized with saturated aq. NaHCO3 solution and extracted with DCM (3 x 150 mL).
The combined organic phases were concentrated in vacuo and purified by flash silica chromatography
1
elution gradient (0ꢀ10% MeOH in DCM) to give the title compound as a white solid (0.23 g, 15%): H
NMR (400 MHz, DMSOꢀd6) 1.56ꢀ1.63 (2H, m), 1.83ꢀ1.97 (4H, m), 2.18 (3H, s), 2.88 (2H, d), 3.18 (4H,
t), 3.75 (4H, t), 7.09ꢀ7.12 (1H, m), 7.20 (1H, d), 7.38 (1H, d), 7.83ꢀ7.86 (1H, m), 7.94 (1H, d), 8.57 (1H,
13
d), 11.65 (2H, br s); C NMR (176 MHz, DMSOꢀd6) 24.6, 45.0, 47.8, 53.2, 59.4, 65.8, 115.1, 118.0,
ACS Paragon Plus Environment
20

Page 21 of 32
Journal of Medicinal Chemistry
118.8, 120.4, 123.6, 126.3, 130.1, 130.8, 132.4, 134.4, 149.4, 151.3, 152.4, 162.9, 168.5; m/z MH+ =
578; HRMS (ES+) for C25H29ClN5O5S2 (MH+): calcd, 578.1293; found, 578.1297.
1
2
3
4
5
6
7
Homology modelling:
8
9
Sequence alignment and structure building. Three different GPCR structures were used as
templates for the h GHSꢀR1a protein model. The closest available structure by sequence similarity is
the rat Neurotensin Receptor Type 1, NTSR1 (pdb code = 4GRV43). This, alongside structures of the
human Kappa Opioid Receptor, OPRK1 (pdb code = 4DJH44) and the human SubstanceꢀP Receptor,
TACR1 (pdb code = 2KS945) was aligned to the h GHSꢀR1a sequence automatically, using constraints
based on principles previously described.46 4DJH was used as the template for TM helices 1 and 2;
2KS9 was used as the template for intracellularꢀloop 3; while 4GRV was used as template for the
majority of the structure. Sequence alignment of these structures is included in the supplementary
information. Choice of template was dictated by overall sequence similarity to h GHSꢀR1a and either
Proline residue placement (in the case of helices) or loop length (in the case of loops). The sequence
alignment and model construction was conducted in MOE (Chemical Computing Group) using the
AMBER12 forcefield and default homology modelling parameters. The top model was accepted.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ligand placement. MOE SiteꢀFinder protocol was used to identify all enclosed pockets within
the structure. The largest of these pockets was located within the transmembrane bundle, close to the
extracellular side and bounded by the extracellular loops. These loops do not significantly descend into
the pocket, leaving it relatively open. This pocket is most likely the binding site for the ghrelin hormone
and was selected for docking studies. Docking was performed for the ligand molecules described herein
using the MOE rigidꢀreceptor docking protocol. Ligand conformations were constrained so that the
acylurea unit forms a hydrogenꢀbonded pseudoꢀring, as shown in the singleꢀcrystal Xꢀray structure of
compound 38 (see Figure 2). The topꢀscoring ligand poses were manually inspected to select the most
likely. Inducedꢀfit docking (where pocket residues are mobile, but constrained) was subsequently
performed on selected poses. Examination of the preferred binding modes for a range of ligands
revealed a series of common interactions with the protein enabling selection of a single binding mode
that was judged to be the best prediction of how these compounds bind to h GHSꢀR1a.
Biological Protocols:
125I-ghrelin displacement assay for GHS-R1a binding IC50 determination (human, mouse
or rat): Isolated plasma membranes from HEK cells stably overexpressing the GHSꢀR1a receptor
(4ꢀg/well) were incubated for 60 min at rt with 20 pM [125I]human ghrelin (Perkin Elmer NEX388) in
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 32
the presence or absence of relevant concentrations of cold competing compounds. Incubations were
performed in 100 ꢀL total volume of assay buffer (50 mM TrisꢀHCl, 5 mM MgCl2.6H2O, 0.01% BSA,
pH7.4) containing 10 ꢀM methyl arachidonyl fluorophosphonate (Sigma M2939). The binding reaction
was stopped by rapid filtration over Whatman GF/C filters preꢀsoaked with 0.5% polyethyeneimine and
rinsed three times with iceꢀcold wash buffer (50 mM TrisꢀHCl, 5 mM MgCl2.6H2O, 50 mM NaCl,
pH7.4) and the radioactivity bound to the membranes was measured using γꢀcounter.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Invitrogen Tango functional assay: TangoTM U2OSꢀGHSꢀR1a assay cells were plated at
10,000 cells per well in 384ꢀwell plate in assay medium (100% Freestyle Medium containing 10 ng/mL
doxycycline) in a final volume of 40 ꢀL and incubated overnight at 37 ºC/5% CO2. Appropriate
concentrations of compounds were also included at the time of plating. Following overnight incubation
cells were loaded with LiveBLAzer FRET B/G substrate for 2 h at rt in the dark and the fluorescence
emission values at 460 nm and 530 nm obtained using a FLIPRTETRA at 409 nm excitation. Maximal
agonist response was measured relative to MK0677.
Mouse Free Feeding: All animal experiments were conducted with strict adherence to licenses
issued under the UK Animals (Scientific Procedures) Act 1986 and after local ethical review and
approval. Studies described were performed under Home Office project license (PPL) number 40/3015.
Male C57Bl6j, GHSr1ꢀ/ꢀ, and wildtype littermate controls were bred at AstraZeneca, Alderley Park.
GHSr1ꢀ/ꢀ mice were generated by Deltagen (San Carlos, CA, USA) as previously described.47 Male mice
(6ꢀ8 weeks on arrival) were acclimatized for 3 weeks to a reversed lightꢀdark cycle (9amꢀ9pm DARK).
Animals were group housed (3ꢀ5/cage) for 14 days then individually housed (1/cage) for 7 days.
Baseline body weight & food intake were measured for the final 5 days before study. Animals were
assigned to treatment groups (n=10ꢀ12) based on the average food intake and body weight recorded over
4 days prior to study, so as to achieve similar means and variance in food intake and body weight in
each group. Compounds were formulated the day before study as suspensions in 1% Pluronic F127
(Sigma, St Louis, USA) in water. Compound or vehicle (10 mL/kg) was administered orally via gavage
approximately 1 h prior to the onset of dark. After administration, animals were given access to a preꢀ
weighed amount of food, which was subsequently reꢀweighed 2, 4, 6 and 24 h after administration.
Pharmacokinetics: Pharmacokinetic studies were performed in male HanꢀWistar (HW) rats,
beagle dogs and CDꢀ1 mice. Intravenous (iv) administration was via bolus administration in a vehicle
containing 5% DMSO 95% HydroxyꢀpropylꢀBꢀCyclodextrin 30% (W/V). Oral dosing was administered
by gavage in a vehicle containing 0.1% F127 in water. Blood samples were collected over a 24 h period
post dose into tubes containing EDTAꢀK2. Plasma was isolated by centrifugation, and the concentration
ACS Paragon Plus Environment
22

Page 23 of 32
Journal of Medicinal Chemistry
of the test compound in plasma was determined by LC/MS/MS after protein precipitation with
acetonitrile. Nonꢀcompartmental analysis was performed to estimate pharmacokinetic parameters using
WinNonLin (version 5.0.1). All animal experiments were conducted with strict adherence to licenses
issued under the UK Animals (Scientific Procedures) Act 1986 and after local ethical review and
approval.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 32
1
2
3
4
5
Supporting Information. Preparation and additional characterization for final compounds, enzymatic
assay procedures, in vivo protocols for efficacy studies and homology model alignment. This material is
available free of charge via the Internet at http://pubs.acs.org.
6
7
8
9
Corresponding Author
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Phone: +44(0)1625519444. Eꢀmail: william.mccoull@astrazeneca.com
Notes
The authors declare no competing financial interest.
Acknowledgements
Teresa Collins, Mark Denn and John Swales are acknowledged for expert technical assistance in
generating DMPK data. 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. Duncan Armstrong, Philip Macfaul, Garry Pairaudeau and Matt Wood are thanked
for useful discussions on this manuscript.
Abbreviations Used
DIPEA, diisopropylethylamine; FA, formic acid; GHSꢀR1a , growth hormone secretagogue receptor
type 1a; h GHSꢀR1a, human isoform; m GHSꢀR1a, murine isoform; ; r GHSꢀR1a, rat isoform; LLE,
ligand lipophilicity efficiency; MDCK, MadinꢀDarby canine kidney; MeCN, acetonitrile; TM, transꢀ
membrane helix.
References
1. Kojima, M.; Hosoda, H.; Date, Y.; Nakazato, M.; Matsuo, H.; Kangawa, K. Ghrelin is a growthꢀ
hormoneꢀreleasing acylated peptide from stomach. Nature (London) 1999, 402, 656ꢀ660.
2. Kojima, M.; Hosoda, H.; Matsuo, H.; Kangawa, K. Ghrelin: discovery of the natural endogenous
ligand for the growth hormone secretagogue receptor. Trends Endocrinol. Metab. 2001, 12, 118ꢀ122.
ACS Paragon Plus Environment
24

Page 25 of 32
Journal of Medicinal Chemistry
3. Schwartz, M. W.; Woods, S. C.; Porte, D., Jr.; Seeley, R. J.; Baskin, D. G. Central nervous system
1
2
3
control of food intake. Nature (London) 2000, 404, 661ꢀ671.
4
5
6
7
8
9
4. Cummings, D. E.; Purnell, J. Q.; Frayo, R. S.; Schmidova, K.; Wisse, B. E.; Weigle, D. S. A
preprandial rise in plasma ghrelin levels suggests a role in meal initiation in humans. Diabetes 2001, 50,
1714ꢀ1719.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5. Wren, A. M.; Seal, L. J.; Cohen, M. A.; Brynes, A. E.; Frost, G. S.; Murphy, K. G.; Dhillo, W. S.;
Ghatei, M. A.; Bloom, S. R. Ghrelin enhances appetite and increases food intake in humans. J. Clin.
Endocrinol. Metab. 2001, 86, 5992ꢀ5992.
6. AlvarezꢀCastro, P.; Pena, L.; Cordido, F. Ghrelin in obesity, physiological and pharmacological
considerations. MiniꢀRev. Med. Chem. 2013, 13, 541ꢀ552.
7. Depoortere, I. Targeting the ghrelin receptor to regulate food intake. Regul. Pept. 2009, 156, 13ꢀ23.
8. Carpino, P. A.; Ho, G. Modulators of the ghrelin system as potential treatments for obesity and
diabetes. Expert Opin. Ther. Pat. 2008, 18, 1253ꢀ1263.
9. Soares, J.; RonconꢀAlbuquerque, R., Jr; LeiteꢀMoreira, A. Ghrelin and ghrelin receptor inhibitors:
agents in the treatment of obesity. Expert Opin. Ther. Targets 2008, 12, 1177ꢀ1189.
10. Zhao, H.; Liu, G. Growth hormone secretagogue receptor antagonists as antiꢀobesity therapies? Still
an open question. Curr. Opin. Drug Discovery Dev. 2006, 9, 509ꢀ515.
11. Chollet, C.; Meyer, K.; BeckꢀSickinger, A. G. Ghrelinꢀa novel generation of antiꢀobesity drug:
design, pharmacomodulation and biological activity of ghrelin analogues. J. Pept. Sci. 2009, 15, 711ꢀ
730.
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 32
12. Ahima, R. S. Antagonism of ghrelin for glycemic control in type 2 diabetes mellitus? Endocrinology
2007, 148, 5173ꢀ5174.
1
2
3
4
5
6
7
8
9
13. Damian, M.; Marie, J.; Leyris, J.; Fehrentz, J.; Verdie, P.; Martinez, J.; Baneres, J.; Mary, S. High
constitutive activity is an intrinsic feature of ghrelin receptor protein: a study with a functional
monomeric GHSꢀR1a receptor reconstituted in lipid discs. J. Biol. Chem. 2012, 287, 3630ꢀ3641.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
14. Holst, B.; Cygankiewicz, A.; Jensen, T. H.; Ankersen, M.; Schwartz, T. W. High constitutive
signaling of the ghrelin receptorꢀidentification of a potent inverse agonist. Mol. Endocrinol. 2003, 17,
2201ꢀ2210.
15. Mokrosinski, J.; Frimurer, T. M.; Sivertsen, B.; Schwartz, T. W.; Holst, B. Modulation of
constitutive activity and signaling bias of the ghrelin receptor by conformational constraint in the second
extracellular loop. J. Biol. Chem. 2012, 287, 33488ꢀ33502.
16. Holst, B.; Schwartz, T. W. Constitutive ghrelin receptor activity as a signaling setꢀpoint in appetite
regulation. Trends Pharmacol. Sci. 2004, 25, 113ꢀ117.
17. Rudolph, J.; Esler, W. P.; O'Connor, S.; Coish, P. D. G.; Wickens, P. L.; Brands, M.; Bierer, D. E.;
Bloomquist, B. T.; Bondar, G.; Chen, L.; Chuang, C.; Claus, T. H.; Fathi, Z.; Fu, W.; Khire, U. R.;
Kristie, J. A.; Liu, X.; Lowe, D. B.; McClure, A. C.; Michels, M.; Ortiz, A. A.; Ramsden, P. D.;
Schoenleber, R. W.; Shelekhin, T. E.; Vakalopoulos, A.; Tang, W.; Wang, L.; Yi, L.; Gardell, S. J.;
Livingston, J. N.; Sweet, L. J.; Bullock, W. H. Quinazolinone derivatives as orally available ghrelin
receptor antagonists for the treatment of diabetes and obesity. J. Med. Chem. 2007, 50, 5202ꢀ5216.
18. Esler, W. P.; Rudolph, J.; Claus, T. H.; Tang, W.; Barucci, N.; Brown, S.; Bullock, W.; Daly, M.;
DeCarr, L.; Li, Y.; Milardo, L.; Molstad, D.; Zhu, J.; Gardell, S. J.; Livingston, J. N.; Sweet, L. J.
Smallꢀmolecule ghrelin receptor antagonists improve glucose tolerance, suppress appetite, and promote
weight loss. Endocrinology 2007, 148, 5175ꢀ5185.
ACS Paragon Plus Environment
26

Page 27 of 32
Journal of Medicinal Chemistry
19. Mihalic, J. T.; Kim, Y.; Lizarzaburu, M.; Chen, X.; Deignan, J.; Wanska, M.; Yu, M.; Fu, J.; Chen,
X.; Zhang, A.; Connors, R.; Liang, L.; Lindstrom, M.; Ma, J.; Tang, L.; Dai, K.; Li, L. Discovery of a
new class of ghrelin receptor antagonists. Bioorg. Med. Chem. Lett. 2012, 22, 2046ꢀ2051.
1
2
3
4
5
6
7
8
9
20. Puleo, L.; Marini, P.; Avallone, R.; Zanchet, M.; Bandiera, S.; Baroni, M.; Croci, T. Synthesis and
pharmacological evaluation of indolinone derivatives as novel ghrelin receptor antagonists. Bioorg.
Med. Chem. 2012, 20, 5623ꢀ5636.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
21. Hanrahan, P.; Bell, J.; Bottomley, G.; Bradley, S.; Clarke, P.; Curtis, E.; Davis, S.; Dawson, G.;
Horswill, J.; Keily, J.; Moore, G.; Rasamison, C.; Bloxham, J. Substituted azaquinazolinones as
modulators of GHSrꢀ1a for the treatment of type II diabetes and obesity. Bioorg. Med. Chem. Lett. 2012,
22, 2271ꢀ2278.
22. Sabbatini, F. M.; Melotto, S.; Bernasconi, G.; Bromidge, S. M.; D'Adamo, L.; Rinaldi, M.; Savoia,
C.; Mundi, C.; Di Francesco, C.; Zonzini, L.; Costantini, V. J. A.; Perini, B.; Valerio, E.; Pozzan, A.;
Perdonà, E.; Visentini, F.; Corsi, M.; Di Fabio, R. Azabicyclo[3.1.0]hexaneꢀ1ꢀcarbohydrazides as potent
and selective GHSR1a ligands presenting a specific in vivo behavior. ChemMedChem 2011, 6, 1981ꢀ
1985.
23. Serby, M. D.; Zhao, H.; Szczepankiewicz, B. G.; Kosogof, C.; Xin, Z.; Liu, B.; Liu, M.; Nelson, L.
T. J.; Kaszubska, W.; Falls, H. D.; Schaefer, V.; Bush, E. N.; Shapiro, R.; Droz, B. A.; KnourekꢀSegel,
V. E.; Fey, T. A.; Brune, M. E.; Beno, D. W. A.; Turner, T. M.; Collins, C. A.; Jacobson, P. B.; Sham,
H. L.; Liu, G. 2,4ꢀDiaminopyrimidine derivatives as potent growth hormone secretagogue receptor
antagonists. J. Med. Chem. 2006, 49, 2568ꢀ2578.
24. Moulin, A.; Brunel, L.; Boeglin, D.; Demange, L.; Ryan, J.; M'Kadmi, C.; Denoyelle, S.; Martinez,
J.; Fehrentz, J. The 1,2,4ꢀtriazole as a scaffold for the design of ghrelin receptor ligands: development of
JMV 2959, a potent antagonist. Amino Acids 2013, 44, 301ꢀ314
.
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 32
25. Holst, B.; Lang, M.; Brandt, E.; Bach, A.; Howard, A.; Frimurer, T. M.; BeckꢀSickinger, A.;
Schwartz, T. W. Ghrelin receptor inverse agonists: identification of an active peptide core and its
interaction epitopes on the receptor. Mol. Pharmacol. 2006, 70, 936ꢀ946.
1
2
3
4
5
6
7
8
9
26. Pasternak, A.; Goble, S. D.; de, J., Reynalda K.; Hreniuk, D. L.; Chung, C. C.; Tota, M. R.; Mazur,
P.; Feighner, S. D.; Howard, A. D.; Mills, S. G.; Yang, L. Discovery and optimization of novel 4ꢀ
[(aminocarbonyl)amino]ꢀNꢀ[4ꢀ(2ꢀaminoethyl)phenyl]benzenesulfonamide ghrelin receptor antagonists.
Bioorg. Med. Chem. Lett. 2009, 19, 6237ꢀ6240
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
27. Kung, D. W.; Coffey, S. B.; Jones, R. M.; Cabral, S.; Jiao, W.; Fichtner, M.; Carpino, P. A.; Rose,
C. R.; Hank, R. F.; Lopaze, M. G.; Swartz, R.; Chen, H.; Hendsch, Z.; Posner, B.; Wielis, C. F.;
Manning, B.; Dubins, J.; Stock, I. A.; Varma, S.; Campbell, M.; DeBartola, D.; KosaꢀMaines, R.; Steyn,
S. J.; McClure, K. F. Identification of spirocyclic piperidineꢀazetidine inverse agonists of the ghrelin
receptor. Bioorg. Med. Chem. Lett. 2012, 22, 4281ꢀ4287.
28. McCoull, W.; Barton, P.; Broo, A.; Brown, A. J. H.; Clarke, D. S.; Coope, G.; Davies, R. D. M.;
Dossetter, A. G.; Kelly, E. E.; Knerr, L.; MacFaul, P.; Holmes, J. L.; Martin, N.; Moore, J. E.; Morgan,
D.; Newton, C.; Osterlund, K.; Robb, G. R.; Rosevere, E.; Selmi, N.; Stokes, S.; Svensson, T. S.; Ullah,
V. B. K.; Williams, E. J. Identification of pyrazoloꢀpyrimidinones as GHSꢀR1a antagonists and inverse
agonists for the treatment of obesity. Med. Chem. Commun. 2013, 4, 456ꢀ462.
29. McClure, K. F.; Jackson, M.; Cameron, K. O.; Kung, D. W.; Perry, D. A.; Orr, S. T. M.; Zhang, Y.;
Kohrt, J.; Tu, M.; Gao, H.; Fernando, D.; Jones, R.; Erasga, N.; Wang, G.; Polivkova, J.; Jiao, W.;
Swartz, R.; Ueno, H.; Bhattacharya, S. K.; Stock, I. A.; Varma, S.; Bagdasarian, V.; Perez, S.; Kellyꢀ
Sullivan, D.; Wang, R.; Kong, J.; Cornelius, P.; Michael, L.; Lee, E.; Janssen, A.; Steyn, S. J.; Lapham,
K.; Goosen, T. Identification of potent, selective, CNSꢀtargeted inverse agonists of the ghrelin receptor.
Bioorg. Med. Chem. Lett. 2013, 23, 5410ꢀ5414.
ACS Paragon Plus Environment
28

Page 29 of 32
Journal of Medicinal Chemistry
30. Bhattacharya, S. K.; Andrews, K.; Beveridge, R.; Cameron, K. O.; Chen, C.; Dunn, M.; Fernando,
D.; Gao, H.; Hepworth, D.; Jackson, V. M.; Khot, V.; Kong, J.; Kosa, R. E.; Lapham, K.; Loria, P. M.;
Londregan, A. T.; McClure, K. F.; Orr, S. T. M.; Patel, J.; Rose, C.; Saenz, J.; Stock, I. A.; Storer, G.;
VanVolkenburg, M.; Vrieze, D.; Wang, G.; Xiao, J.; Zhang, Y. Discovery of PFꢀ5190457, a potent,
selective, and orally bioavailable ghrelin receptor inverse agonist clinical candidate. ACS Med. Chem.
Lett. 2014, 5, 474ꢀ479.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
31. Leeson, P. D.; Springthorpe, B. The influence of drugꢀlike concepts on decisionꢀmaking in
medicinal chemistry. Nat. Rev. Drug Discovery 2007, 6, 881ꢀ890.
32. Hanson, B. J.; Wetter, J.; Bercher, M. R.; Kopp, L.; FuerstenauꢀSharp, M.; Vedvik, K. L.; Zielinski,
T.; Doucette, C.; Whitney, P. J.; Revankar, C. A homogeneous fluorescent liveꢀcell assay for measuring
7ꢀtransmembrane receptor activity and agonist functional selectivity through betaꢀarrestin recruitment.
J. Biomol. Screening 2009, 14, 798ꢀ810.
33. Di, L.; Rong, H.; Feng, B. Demystifying brain penetration in central nervous system drug discovery.
J. Med. Chem. 2013, 56, 2ꢀ12.
34. Els, S.; Schild, E.; Petersen, P. S.; Kilian, T.; Mokrosinski, J.; Frimurer, T. M.; Chollet, C.;
Schwartz, T. W.; Holst, B.; BeckꢀSickinger, A. G. An aromatic region to induce a switch between
agonism and inverse agonism at the ghrelin receptor. J. Med. Chem. 2012, 55, 7437ꢀ7449.
35. Friden, M.; Ducrozet, F.; Middleton, B.; Antonsson, M.; Bredberg, U.; HammarlundꢀUdenaes, M.
Development of a highꢀthroughput brain slice method for studying drug distribution in the central
nervous system. Drug Metab. Dispos. 2009, 37, 1226ꢀ1233.
36. PSA was calculated using inꢀhouse software according to a published methodology: Ertl, P.; Rohde,
B.; Selzer, P. Fast calculation of molecular Polar Surface Area as a sum of fragmentꢀbased contributions
and its application to the prediction of drug transport properties. J. Med. Chem. 2000, 43, 3714ꢀ3717.
ACS Paragon Plus Environment
29