Structure-Activity Study of Ghrelin(1-8) Resulting in High Affinity Fluorine-Bearing Ligands for the Ghrelin Receptor

Article abstract of DOI:10.1021/acs.jmedchem.7b00164

The ghrelin receptor, also known as the growth hormone secretagogue receptor 1a (GHS-R1a), is a G-protein-coupled receptor that is differentially expressed in healthy tissue and several cancers, including prostate, testicular, and ovarian. Selectively targeting the ghrelin receptor using fluorine-18 tagged entities would allow localization and visualization of ghrelin receptor expressing carcinomas using PET imaging. The endogenous ligand ghrelin, a 28 amino acid peptide with 3.1 nM affinity, has poor in vivo stability. Here we report on ghrelin(1-8) analogues bearing modifications at residues 1, 3, 4, and 8. The lead analogue, [Inp¹,Dpr³(6-fluoro-2-naphthoate),1-Nal⁴,Thr⁸]ghrelin(1-8), possessed an IC₅₀ value of 0.11 nM that is a 28-fold improvement compared to the natural ligand. A novel 6-fluoro-2-pentafluorophenyl naphthoate (PFPN) prosthetic group was synthesized to incorporate fluorine-18 for PET imaging. This is not only the highest affinity ghrelin analogue reported but also the shortest ghrelin analogue capable of binding GHS-R1a with better affinity than ghrelin(1-28).

Full text of DOI:10.1021/acs.jmedchem.7b00164

Subscriber access provided by UNIV OF DURHAM
Article
Structure-Activity Study of Ghrelin(1-8) Resulting in High
Affinity Fluorine-bearing Ligands for the Ghrelin Receptor
Carlie L. Charron, Jin-Qiang Hou, Mark S. McFarland, Savita
Dhanvantari, Michael S. Kovacs, and Leonard G. Luyt
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b00164 • Publication Date (Web): 18 Jul 2017
Downloaded from http://pubs.acs.org on July 18, 2017
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 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Structure-Activity Study of Ghrelin(1-8) Resulting
in High Affinity Fluorine-Bearing Ligands for the
Ghrelin Receptor
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
a†
b
a
c,d
Carlie L. Charron , Jinqiang Hou , Mark S. McFarland , Savita Dhanvantari , Michael S.
Kovacs , Leonard G. Luyta,b,e*
c,d
a
Department of Chemistry, University of Western Ontario, 1151 Richmond Street, London,
Ontario N6A 5B7 (Canada)
b
London Regional Cancer Program, Lawson Health Research Institute, 790 Commissioners
Road East, London, Ontario N6A 4L6 (Canada)
c
Imaging Program, Lawson Health Research Institute, 268 Grosvenor Street, London, Ontario
N6A 4V2, Canada
d
Department of Medical Biophysics, University of Western Ontario, 1151 Richmond Street,
London, Ontario N6A 5B7 (Canada)
e
Departments of Oncology and Medical Imaging, University of Western Ontario, 1151
Richmond Street, London, Ontario N6A 5B7 (Canada); E-mail: lluyt@uwo.ca
KEYWORDS: Ghrelin, GHS-R1a, Fluorine-18, Peptide Chemistry, Radiochemistry
ABSTRACT
The ghrelin receptor, also known as the growth hormone secretagogue receptor 1a (GHS-R1a),
is a G-protein coupled receptor that is differentially expressed in healthy tissue and several cancers,
including prostate, testicular, and ovarian. Selectively targeting the ghrelin receptor using fluorine-
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 43
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
8 tagged entities would allow localization and visualization of ghrelin receptor expressing
carcinomas using PET imaging. The endogenous ligand ghrelin, a 28 amino acid peptide with 3.1
nM affinity, has poor in vivo stability. Here we report on ghrelin(1-8) analogues bearing
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
3
modifications at residues 1, 3, 4, and 8. The lead analogue, [Inp ,Dpr (6-fluoro-2-naphthoate),1-
4
8
Nal ,Thr ]ghrelin(1-8), possessed an IC50 value of 0.11 nM that is a 28-fold improvement
compared to the natural ligand. A novel 6-fluoro-2-pentafluorophenyl naphthoate (PFPN)
prosthetic group was synthesized to incorporate fluorine-18 for PET imaging. This is not only the
highest affinity ghrelin analogue reported, it is also the shortest ghrelin analogue capable of
binding GHS-R1a with better affinity than ghrelin(1-28).
INTRODUCTION
The recent advances in the field of molecular imaging have vastly influenced disease
monitoring, staging and diagnosis. Molecular imaging techniques, such as optical imaging (OI),
positron emission tomography (PET) and single-photon emission computed tomography (SPECT),
have been used to develop a better understanding of specific molecular events occurring within a
1
disease state. The field of oncology has greatly benefitted from developments in molecular
imaging. Classical methods of cancer diagnosis, such as histology, are slowly being replaced by
new non-invasive methods developed through molecular imaging. One approach to developing
these methodologies is identifying cell surface receptors involved in the proliferation, migration
and cell invasion processes associated with cancer progression and using them as a marker for the
2
disease. For example, somatostatin receptors are highly expressed on the cell surface of specific
neoplastic tissues. Using structure-activity studies, several somatostatin analogues were evaluated
revealing the optimal candidate as octreotide, a cyclic octapeptide. Using a pendant design
ACS Paragon Plus Environment
2

Page 3 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
approach, chelating moieties were appended to the cyclic peptide to incorporate various
radiometals such as 64Cu, 68Ga and 111In. These analogues are now capable of detecting
neuroendocrine tumours using scintigraphy and have been evaluated in preclinical and clinical
studies.3-5 This approach to designing stable, high affinity peptide ligands for transmembrane
receptors present in tumors has led to many new peptide ligands being developed for imaging
purposes. This includes peptides derived from gastrin-releasing peptide, cholecystokinin and
alpha-melanocyte stimulating hormone and this approach continues to be an area of focus for
designing new receptor-binding imaging agents.6
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
An exciting new receptor holding potential relevance to certain cancers is the ghrelin
receptor, one isoform of the growth hormone secretagogue receptor (GHS-R) family. GHS-R has
been detected in two isoforms. The ghrelin receptor, also referred to as GHS-R1a, is a G-protein-
coupled receptor known to induce growth hormone release, control food intake and facilitate
energy metabolism, mediated by its endogenous ligand ghrelin. GHS-R1b is a truncated 5-
transmembrane receptor found to be functionally inactive and possesses no known endogenous
7
ligand. Initially, it was proposed that the ghrelin receptor was solely expressed in the pituitary and
hypothalamus but more comprehensive studies show that while it is predominately expressed in
the pituitary, it is also present in low levels in peripheral tissues including the thyroid, pancreas,
spleen, myocardium and adrenal gland. More importantly, the ghrelin receptor is differentially
expressed in healthy human tissues and several tumour types such as prostate, breast, testicular
and ovarian carcinomas8-12, and therefore may be a potential biomarker for these cancers.
The endogenous ligand for the ghrelin receptor is ghrelin (2), a 28 amino acid peptide
hormone (H-GSS(octanoyl)FLSPEHQRVQQRKESKKPPAKLQPR-OH) that possesses a unique
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 43
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
octanoyl chain on the serine side chain at position 3. Human ghrelin originates from preproghrelin,
a 177-amino acid protein that undergoes proteolytic cleavage to a 28 amino acid peptide known as
desacyl ghrelin (1). Compound 1 possesses the same amino acid sequence as 2, but lacks the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
3
octanoyl chain present at position 3. Without the octanoyl chain, 1 has virtually no affinity (IC50
> 10000 nM) to the ghrelin receptor as determined by a competitive binding assay. 1 then
undergoes acylation by the ghrelin O-acyl transferase (GOAT) to yield the octanoyl modification.
It is this unique modification that significantly increases the affinity to 3.1 nM (Figure 1).14
Ghrelin provides a high affinity template as a starting point for further development into a
radiopeptide. Like many regulatory peptides, ghrelin lacks in vivo stability and requires
modifications such as truncation and amino acid substitution in order to resist in vivo degradation
and maintain affinity to the ghrelin receptor. Much like somatostatin, a methodical sequence-
activity relationship study of ghrelin will help to better understand the role each amino acid plays
in the interaction with its endogenous receptor. We propose that truncation of ghrelin to the first
eight N-terminal amino acids will result in high affinity ghrelin analogues by selectively modifying
amino acids important for receptor interaction. We will also incorporate a radionuclide into the
ghrelin structure in a region that is integral for receptor interaction. This design required the
ACS Paragon Plus Environment
4

Page 5 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
development of a new fluorine-18 prosthetic group and resulted in a high affinity fluorine-
containing ghrelin(1-8) analogue capable of targeting the ghrelin receptor.
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 Preproghrelin undergoes proteolytic cleavage resulting in desacyl ghrelin (1) which
undergoes acylation by ghrelin O-acyl transferase resulting in ghrelin(1-28) (2).
RESULTS AND DISCUSSION
Design of Ghrelin Analogues for Positron Emission Tomography (PET)
A significant quantity of literature has focused on describing studies where the ghrelin(1-28)
structure was modified in order to determine the amino acids necessary for activity, the role of the
octanoyl chain, and the shortest peptide fragment required to activate the receptor.15, 16 Matsumoto
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 43
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
et al. determined that the C-terminal region of the peptide, ghrelin(16-28), had no affinity to the
ghrelin receptor. The N-terminal region of the peptide was then investigated to determine the
minimum amino acid sequence required to ensure affinity. When truncated to the first 14 N-
terminal amino acids, ghrelin(1-14), binding affinity remained relatively intact with an IC50 of 9.6
nM. As further truncation occurred, there was an increasing loss of affinity for the receptor. For
example, ghrelin(1-10) was reported to have an IC50 of 19.0 nM while ghrelin(1-4) had an IC50 of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
80 nM. When ghrelin was truncated to less than 4 amino acids, all affinity to the receptor was
1
5
lost suggesting the first 4 amino acids play an integral role for receptor binding. The same general
trend was observed by Bednarek et al. where no notable affinity to GHS-R1a was observed for
1
4
ghrelin analogues shorter than 5 amino acid residues. Work previously completed in the Luyt
Group showed that a ghrelin (1-5) analogue bearing a fluorinated alkyl substituent in position 3
17
had no affinity to the ghrelin receptor, while longer analogues did. Compilation of these results
has directed our subsequent work to explore marginally longer ghrelin analogues to ensure the
affinity for the ghrelin receptor is not negatively affected.
When VanCraenenbroeck et al. studied ghrelin(1-14) analogues, they discovered that placing an
alanine or tyrosine in position 8 resulted in the most potent ghrelin(1-14) analogues. These findings
suggest that the negatively charged glutamic acid found in natural ghrelin is not the optimal choice
for this position and replacing glutamic acid with various hydrophobic or polar-uncharged side
chains could result in an optimal peptide capable of targeting the ghrelin receptor with equal or
18
greater affinity than ghrelin(1-28). To further explore this amino acid effect, modifications were
made to positions 3 and 8 while the other amino acids remained the same as the natural ghrelin
sequence. At position 3, serine was replaced with the unnatural amino acid diaminopropionic acid
(Dpr) to replace the ester linkage with a more synthetically facile and stable amide linkage.19, 20
ACS Paragon Plus Environment
6

Page 7 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Dpr, bearing an orthogonal protecting group, was selectively deprotected on resin and coupled to
octanoic acid using standard coupling conditions. Position 8 was modified to contain each of the
hydrophobic aromatic side chain amino acids; phenylalanine (F), tyrosine (Y) and tryptophan (W),
and compared to the natural sequence bearing glutamic acid (E). The binding affinity was
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
0
determined using a previously reported competitive binding assay. As summarized in Table 1, E
as residue 8 resulted in an IC50 of 200 nM but replacing this residue with hydrophobic aromatic
side chain greatly increased the binding affinity. W and Y were less effective substitutions,
resulting in 86.3 nM (3c) and 65.0 nM (3a), respectively, while the binding affinity substantially
improved with F to 6.7 nM (3b). Position 8 was also substituted with amino acids bearing polar-
uncharged side chains: serine (S), asparagine (N), glutamine (Q) and threonine (T). These
substitutions lead to higher affinity analogues than the hydrophobic aromatic side chains. The best
th
analogue 3f (Figure 2A), with T in the 8 position, had a 3.3 nM IC50, while N (3d), S (3g) and Q
(
3e) were less effective with 31.9 nM, 28.8 nM and 21.7 nM respectively.
Table 1 Substitution of residue 8 with various natural and unnatural amino acids and the resulting
IC50 (nM) for the ghrelin receptor.
Amino
Unnatural
Acid X Amino Acid X
Compound
No.
IC50
(nM)
E
Y
F
W
N
Q
3
200
65.0
6.7
86.3
31.9
21.7
3a
3b
3c
3d
3e
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 43
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
T
S
3f
3g
3h
3i
3j
3.3
28.8
66.2
45.0
94.1
t
β-homo-Thr
2
-Nal
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
One drawback to using peptides as targeting entities is their limited in vivo stability. Beck-
68
Sickinger et. al. have studied Ga-labelled ghrelin(1-28) and ghrelin(1-16) analogues as potential
PET agents. Although they exhibit similar efficacy for the ghrelin receptor as per native ghrelin
based upon an inositol phosphate turnover assay, they were found to have rapid blood clearance
and suboptimal metabolic stability, especially for the truncated analogues.21-23 In order to increase
the stability of the developed ghrelin(1-8) analogues, bioisosteres can be added to reduce
degradation by endopeptidases and exopeptidases. Bioisosteres are chemical groups that possess
similar physical and chemical properties to another chemical compound and elicit similar
2
4
biological properties. Bioisosteres of F and T, the two lead modifications to position 8, were
incorporated into the peptide sequence in hopes of maintaining the affinity to the ghrelin receptor
as well as increasing the overall stability of the peptide. As summarized in Table 1, two bioisosteres
of T were introduced, D-threonine (3h) and β-homo-threonine (3i). These modifications had a
negative effect on binding affinity by increasing the IC50 to 66.0 nM and 45.0 nM, respectively. 2-
Naphthylalanine (2-Nal, 3j) was also introduced as a non-classical bioisostere for F and also had
a negative effect on IC50, increasing it to 94.1 nM. Overall, two amino acids were found to most
effectively improve binding affinity, F (3b) and T (3f).
ACS Paragon Plus Environment
8

Page 9 of 43
Journal of Medicinal Chemistry
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 2 IC50 curves of lead analogues 3f (A), 5b (B) and 6a (C) after modifications were made
to positions 8, 1 and 4, respectively.
Bednarek, et al. investigated the importance of the octanoyl chain by evaluating several
analogues where serine-3 was acylated with different aliphatic and aromatic acids. When serine-3
is acylated with large aliphatic groups, such as 11-undecanoic acid, the binding affinity is as
effective as natural human ghrelin. Yet, when small, less hydrophobic acids are present, such as
formic acid, all binding affinity is lost. This investigation demonstrates that ghrelin is able to
tolerate large lipophilic groups at position 3 and possibly provides the opportunity to create a short
1
3
ghrelin analogue with a radioisotope integrated into the side chain at position 3. To test this
hypothesis, Rosita et. al. integrated lipophilic groups bearing fluorine-19 and rhenium-185/187 in
position 3 of ghrelin(1-14)-amide to investigate the potential as PET imaging agents. The most
promising fluorine-19 group contained a 12-carbon chain, with an IC50 of 27.9 nM and the most
promising rhenium-185/187 contained a cyclopentadienyl-rhenium complex after a 3-carbon chain
that resulted in a 35.0 nM affinity. These analogues were not ultimately developed into PET
imaging agents but acted as inspiration for further development of ghrelin(1-8) analogues that
require a radioisotope for association to the native receptor. 17
Since it was demonstrated that large lipophilic groups were tolerated in position 3, we
chose to integrate a well-known fluorine-18 prosthetic group, 4-fluorobenzoic acid (4-FBA), at the
end of the Dpr side chain. This would not only provide a lipophilic mass at position 3 but would
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 43
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
18
also provide a straightforward approach for radiolabeling. 4-[ F]FBA has been used to radiolabel
various peptides such as RGD and bombesin, and has well developed radiochemical procedures.25,
2
6
Shown in Table 2, integration of 4-FBA (4a) onto residue 3 resulted in a less than optimal IC50
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
of 146 nM. In order to increase the lipophilicity of the prosthetic group, a bulkier aromatic ring
was added by introducing 6-fluoronaphthoic acid (6-FNA, 4b) and 4’-fluoro-[1,1’-biphenyl]-4-
carboxylic acid (4’-FBC, 4c) using standard coupling conditions. The increased lipophilicity had
a positive effect on binding affinity to 9.9 nM and 33.7 nM, respectively, and led us to explore 6-
FNA as a new fluorine-18 prosthetic group.
Table 2 Substitution of Dpr side chain with various fluorine-bearing aromatic prosthetic groups
and the resulting IC50 (nM) for the ghrelin receptor.
Prosthetic Compound IC50
Group R
No.
(nM)
4
a
b
146
4
9.9
4c
33.7
Following the choice of a suitable prosthetic group for position 3, further modifications
were made to the N-terminal residue 1 and residue 4 in hopes of further increasing in vivo stability
and GHS-R1a affinity by having additional unnatural amino acids present. The N-terminal glycine
was replaced with one of three unnatural amino acids; aminoisobutyric acid (Aib), isonipecotic
acid (Inp) or sarcosine (Sar). The secondary amine found in Sar (5c) had a large negative effect on
ACS Paragon Plus Environment
1
0

Page 11 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
IC50 causing it to rise to 262 nM, while the secondary amine found in Inp (5b, Figure 2B) had the
opposite effect decreasing it to a desirable IC50 of 9.6 nM. Aib, a primary amine, also had a positive
effect on IC50, decreasing it to 9.3 nM (Table 3). When choosing the most suitable candidate for
residue 1, Inp was chosen in order to reduce the presence of primary amines during the
radiochemical prosthetic group labelling. Van Craenenbroeck performed an alanine scan on
ghrelin(1-14) and determined that the F present at residue 4 was integral for interaction with the
receptor. Replacement of this residue with alanine or tyrosine caused a decrease in binding
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
1
6
affinity. Therefore, we hypothesized that replacing F with a larger, more lipophilic side chain
may create a stronger interaction between the ligand and receptor. Substitution of F with Nal-1
and Nal-2 had varied success (Table 3). Nal-1 was able to increase binding affinity greatly from
1
46 nM (4a) to 4.1 nM (6a, Figure 2C) while Nal-2 had a detrimental effect result in a large loss
of affinity. This further establishes the importance of the aromatic group at position 4 of ghrelin
and suggests that there is minimal freedom for positioning of substituents in this hydrophobic
pocket of the receptor. Interestingly, by adding naphthyl groups into the ghrelin(1-8) structure, the
peptides are increasingly becoming structurally related to the growth hormone secretagogue family
of peptidomimetics, for example ipamorelin and G7039, where multiple hydrophobic pockets are
being accessed by the ligand.27
Table 3 Substitution of residue 1 and 4 with various unnatural amino acids and the resulting IC50
(nM) for the ghrelin receptor.
ACS Paragon Plus Environment
1
1

Journal of Medicinal Chemistry
Page 12 of 43
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Amino
Amino
Compound IC50
Acid X
1
Acid X
2
No.
4b
5a
5b
5c
6a
6b
(nM)
9.9
9.2
9.6
262
4.1
1068
G
F
F
F
F
Nal-1
Nal-2
Aib
Inp
Sar
G
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
G
A thorough investigation into the amino acid sequence of ghrelin(1-8)-amide has led to
many analogues with binding affinities comparable to ghrelin(1-28). Combining the most
successful substitutions has led to a lead peptide analogue (Figure 3) to be further developed into
a fluorine-18 PET imaging agent. The best substitutions were Inp, Nal-1 and Thr in positions 1, 4
and 8, respectively and coupling 4-FNA to the side chain of Dpr at position 3. Compound 7 resulted
in the best IC50 of all ghrelin analogues at 0.11 nM and supported the further development of this
peptide into a PET radiopeptide by incorporating fluorine-18. The radioligand binding assay
1
25
showed that the compound 7 is able to displace the I -natural ghrelin in a concentration dependent
manner, which suggests that this compound specifically binds to GHS-R1a and it is suitable to be
developed into a radiotracer for the imaging of GHS-R1a. The stability of compound 7 was
assessed in human serum resulting in over 60% intact peptide after 24 hours of incubation.
(Supplemental Information).
ACS Paragon Plus Environment
1
2

Page 13 of 43
Journal of Medicinal Chemistry
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 Amino acid structure of 7 (A) and IC50 curve (B) of the ghrelin (1-8) analogue bearing
lead modifications to all positions 1, 3, 4 and 8, with an IC50 = 0.11 nM.
To investigate the recognition between 7 and the ghrelin receptor, a molecular docking
2
7
study was performed using the homology model recently developed by our lab (Figure 4). The
docking result shows that the N-terminal end of the compound is situated at the bottom of binding
1
pocket 1 (Figure 4A). The first residue Inp formed H-bond and salt bridge interactions with Ser123
and Glu124 respectively, which is in agreement with Glu124 being reported to serve as an anchor
point for binding and to play a key role in activation of the receptor.28, 29 Interestingly, the fluorine
4
at residue 6-FNA establishes H-bond interaction with Val 205, providing an additional source for
the enhanced binding affinity. Our previous model structure of G-7039 and the ghrelin receptor
complex showed that the hydrophobic sub-pockets in the orthosteric site of the ghrelin receptor
2
7
were occupied by three hydrophobic residues in G-7039. Similarly, the hydrophobic surface in
1
3
4
Figure 4B shows that three residues Inp , Nal , 6-FNA in 7 are bound to those hydrophobic sub-
pockets. This demonstrates that in a step-wise fashion, we have taken the native ghrelin peptide
and morphed the structure to create a peptide-derived entity that interacts with the ghrelin receptor
in an identical manner to a member of the growth hormone secretagogue family.
ACS Paragon Plus Environment
1
3

Journal of Medicinal Chemistry
Page 14 of 43
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4: Model structure of 7 and GHSR-1a complex from molecular docking study. (A) The
1
first residue of 7, Inp forms H-bond and salt bridge interactions with Ser123 and Glu124
4
respectively. The fluorine at the third residue of 7, 6-FNA establishes H-bond interaction with
Val 205. (B) Three hydrophobic pockets in the binding site of GHS-R1a are occupied by Inp1,
3
4
Nal , 6-FNA of 7. (C) The structure of 7 is overlaid with G-7039, obtained from our previous
computational study. The homology model of GHS-R1a generated from our previous
computational study based on the templates (PDB ID: 4BWB, 3ZEV, 4BUO, 4BV0, 4GRV,
EA3) was used. 27
4
Synthesis of Prosthetic Groups and Standards
The synthesis of the new prosthetic group precursor 11 is outlined in scheme 1. Prosthetic
group precursor 11 was obtained in good yield following a 3-step synthesis. The commercially
available starting material, 6-amino-2-naphthoic acid (8), was dissolved in ethanol and acidified
with sulfuric acid (pH 4) to produce 6-amino-2-ethyl naphthoate (9). Following purification by
flash column chromatography, 9 was dimethylated by sodium cyanoborohydride and
formaldehyde to produce 10. The final synthetic step was done by treating the dimethylamine with
ACS Paragon Plus Environment
1
4

Page 15 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the strong methylating agent, methyl trifluoromethanesulfonate (MeOTf), resulting in the
trimethylamine salt. The resulting 6-trimethylamino-2-ethylnaphthoate trifluoromethanesulfonate
(
11) was obtained as a pure white powder after precipitation and recrystallization. This prosthetic
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
group was designed to undergo nucleophilic aromatic substitution in the presence of nucleophilic
fluorine, with the electron withdrawing nature of the ethyl ester activating the ring. Employment
of the standard radiofluorination protocols using potassium carbonate and kryptofix 2.2.2 was
expected to be suitable for radiolabelling of this prosthetic group.
Scheme 1 Synthesis of 6-trimethylamino-2-ethyl-naphthoate triflate (11)
Most fluorine-18-labelled prosthetic groups reported in the literature are designed as
activated esters for facile acylation to peptide targeting entities. The most common activated ester
1
8
reported is the N-hydroxysuccinimide ester used for 4-[ F]-fluorobenzoic acid to result in N-
1
8
succinimidyl-4-[ F]-fluorobenzoate. Unfortunately, acylation of our lead peptide analogue with
an N-hydroxysuccinimide ester of 6-fluoro-2-napthoic acid resulted in degradation of the active
ester due to harsh labeling conditions. The sluggish and low yielding reactions therefore led us to
seek out a different active ester. Pentafluorophenyl (PFP) esters are recognized as having increased
stability and reaction rate when compared to NHS esters.30, 31
ACS Paragon Plus Environment
1
5

Journal of Medicinal Chemistry
Page 16 of 43
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
19
A non-radioactive 6-fluoro-2-pentafluorophenylnaphthoate ([ F]-PFPN) standard was
synthesized using 6-fluoro-2-napthoic acid, pentafluorophenol and DMAP. The desired activated
ester was purified by flash column chromatography in an 86% yield. A peptide precursor of 7 was
synthesized to possess a free amine at the Dpr residue in position 3. This peptide was synthesized
on Rink amide MBHA resin with all amino acids bearing standard protecting groups. Upon
cleavage from the resin using strong acid, all the protecting groups were removed leaving one
primary amine at position 3 and a secondary amine on the N-terminus. This peptide precursor (17)
is capable to coupling to PFPN in the presence of a weak base producing the desired radiolabelled
ghrelin(1-8) peptide.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Radiochemistry
1
8
Full synthesis of [ F]-7 was carried out in three synthetic steps as summarized in Scheme 2. The
first two steps were carried out on an automated Tracer Lab FXFN synthesis box (Figure 5) with
the configuration of the synthesis box being slightly modified to accommodate these steps. Valve
14 (V14) was directly attached to a product output line to bypass the rest of the synthesis system
and transfer the crude reaction material from the reaction vial to a product vial outside the synthesis
box for further manipulation.
The TracerLab FXFN was prepared for synthesis by filling vial 1 with a solution of
potassium carbonate and kryptofix 2.2.2., vials 2 and 3 with dry ACN (acetonitrile) to
azeotropically dry the nucleophilic fluoride-18, vial 4 with the precursor (11) in dry
dimethylsulfoxide (DMSO) and vial 5 with 1M tetramethylammonium hydroxide (TMAH) for
deprotection of the ethyl ester.
ACS Paragon Plus Environment
1
6

Page 17 of 43
Journal of Medicinal Chemistry
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
1
8
Scheme 2 The radiosynthesis of [ F]-PFPN by automated and manual synthesis.
1
8
Figure 5 The modified schematic of Tracer Lab FXFN for the synthesis of 6-[ F]-fluoro-2-
naphthoic acid (15).
1
8
Fluorine-18 was delivered to the synthesis box in H [ O] and trapped on a Waters QMA
2
Carbonate Sep Pak. The activity was then eluted into the reaction vial by the solution in vial 1.
Following azeotropic drying, precursor 11 was introduced to the reaction vial from vial 4 and
1
8
allowed to react at 120 °C for 10 minutes resulting in 6-[ F]-2-ethylnaphoate (14). 1 M TMAH
(
20 μL) in 1 mL of ACN was then added to the reaction mixture from vial 5 and heated to 90 °C
ACS Paragon Plus Environment
1
7

Journal of Medicinal Chemistry
Page 18 of 43
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
for 10 minutes. After cooling the reaction mixture to 40 °C, the reaction crude was then transferred
from the reaction vial to a product vial containing 7 mL of milliQ water and 0.1% TFA in the side
chamber via valve 14. The following steps were carried out manually. The diluted reaction crude
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
8
was loaded onto a tC18 Sep Pak to remove any unreacted fluorine-18. 6-[ F]-4-naphthoic acid
(15) was eluted from the tC18 Sep Pak with 1 mL of acetonitrile into a reaction vial containing 20
mg of pentafluorophenol, 10 mg 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) and 1 mg
4
-dimethylaminopyridine (DMAP). The reaction mixture was allowed to sit at room temperature
for 5 minutes at which time 1 mL of water was added to the crude reaction mixture. The diluted
mixture is then purified by semi-preparative HPLC (55 – 95% acetonitrile in water) to collect the
1
8
pure 6-[ F]-2-pentafluorophenylnaphthoate. Fractions were combined and solvents removed
using speed evaporation resulting in a 6.5% decay corrected yield. The purified 16 was redissolved
in 1 mL of acetonitrile containing 1.5 mg of peptide precursor (17) (Scheme 3). After
administration of 30 μL of diisopropylethylamine (DIPEA), the reaction mixture was warmed at
4
0 °C for 20 minutes before being diluted with 500 μL water and purified by semi-preparative
1
8
HPLC (flow rate: 5ml/min; gradient: 20% to 80% acetonitrile in water with 0.1% TFA), with [ F]-
1
8
7 being collected at 5 minutes in a 3.1% decay corrected yield. The identity of [ F]-7 was
confirmed by co-injection with the non-radioactive standard 7 (Figure 6).
ACS Paragon Plus Environment
1
8

Page 19 of 43
Journal of Medicinal Chemistry
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
1
8
18
Scheme 3 Coupling of [ F]-PFPN to the lead ghrelin(1-8) analogue resulting in [ F]-7.
19
Figure 6 C18 reverse-phase HPLC chromatogram of non-radioactive [ F]-7 (AU) overlaid with
1
8
the radioactive [ F]-7 (mV) product to confirm the identity and purity of the final product.
CONCLUSIONS
ACS Paragon Plus Environment
1
9

Journal of Medicinal Chemistry
Page 20 of 43
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
An increasing body of literature indicates a link between cancer and ghrelin, with ghrelin being
involved in the regulation of a number of processes related to cancer such as migration,
proliferation and inflammation.32, 33 The development of a ghrelin based PET imaging agent could
be a useful tool for the non-invasive imaging of the ghrelin receptor and understanding its
relationship to disease and therapeutic intervention. An important step towards this goal is to
discover more stable and lower molecular weight ghrelin analogues that incorporate a PET isotope.
Through exploring sequence-activity relationships, we report the optimization of positions 1, 3, 4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
and 8 of ghrelin(1-8), resulting in the best reported activity for a ghrelin peptide, with an IC50
=
0
.11 nM. Importantly, the peptide integrates a novel fluorine-containing prosthetic group into
position 3, which is a critical element for receptor recognition. Noteworthy is that this step-wise
peptide discovery approach has resulted in an octapeptide with remarkable structural similarity to
a family of peptidomimetics, the growth hormone secretagogues, which were discovered prior to
there being knowledge of the endogenous ghrelin receptor ligand.
In order to incorporate fluorine-18 into the fluorine-containing ghrelin(1-8) peptide, a new
1
8
aromatic prosthetic group, [ F]-PFPN, was developed using a combination of automated and
18
manual synthesis. Coupling of [ F]-PFPN to the lead peptide analogue with a free amine at residue
3
resulted in the first fluorine-18 labelled, high affinity ghrelin analogue in an overall 3.1%
radiochemical yield (d.c.). Preclinical evaluation of this PET probe is currently underway in cancer
1
8
models, although it is anticipated that [ F]-7 may be a valuable tool for visualizing various tissues
and disease states that are known for ghrelin receptor expression beyond the field of oncology,
such as heart disease.34
EXPERIMENTAL
ACS Paragon Plus Environment
2
0

Page 21 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Materials and Methods
All common solvents were purchased from Fisher Scientific. All protected amino acids, coupling
reagents and resins were purchased from Novabiochem, Peptides International and Chem-Impex
and were used without further purification unless otherwise stated. All reagents were obtained
from Sigma-Aldrich. RP-tC18 Sep-Pak SPE and QMA carbonate SPE cartridges were purchased
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
125
18
18O was
from Waters. [ I]-Ghrelin was purchased from Perkin Elmer. Cyclotron produced [ F]H
2
obtained from Dr. Mike Kovacs at the Nordal Cyclotron & PET Radiochemistry Facility. Lawson
Health Research Institute, London, Ontario, Canada. For analytical HPLC-MS, an Agilent RP-C18
4
1
.6 x 250 mm, 5 μm column was used. For semi-preparative HPLC-MS work, an Agilent RP-C18
9 x 150 mm, 5 μm column was used. A gradient solvent system was used containing 0.1% TFA
in acetonitrile (solvent A) and 0.1% TFA in water (solvent B). For analytical UHPLC-MS, studies
were performed on a Waters, Inc. Acquity UHPLC H-Class system, combined with a Xevo QTof
mass spectrometer (ES+, cone voltage = 30 V). For analytical UHPLC-MS studies, a Waters
Acquity UHPLC BEH C18 2.1 x 50 mm, 1.7 μm column was used with a gradient solvent system
consisting of 0.1% formic acid in acetonitrile (solvent C) and 0.1% formic acid in water (solvent
D). Analytical radio-RP-HPLC (AgilentTM RP-C18 column 4.6 x 150 mm, 5 μm) was performed
on a Waters 1525 Binary HPLC pump containing a Waters 2487 dual λ absorbance detector,
Waters In-Line degasser, a gamma detector and Breeze software (version 3.30).
General Fmoc Synthesis of Ghrelin(1-8) Peptides
Peptide synthesis was carried out manually using Fmoc-based solid-phase peptide chemistry.
Peptides were synthesized at a 0.1 mmol scale on Rink Amide MBHA resin (0.52 mmol/g) using
a glass peptide reaction vessel. The resin was initially swelled with N,N-dimethylformamide
(
DMF), followed by Fmoc deprotection using 2 mL of 20% piperidine in N,N-dimethylformamide
ACS Paragon Plus Environment
2
1

Journal of Medicinal Chemistry
Page 22 of 43
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
DMF) for two cycles (10 minutes, 5 minutes). Amino acids were preactivated by combining 3 eq.
of Fmoc-protected amino acid,
tetramethyluronium hexafluorophosphate (HCTU) and 6 eq. of N,N-diisopropylethylamine
DIPEA) in 2 mL of DMF. The mixture was added to the resin and vortexed for 60 minutes. These
3
eq. of O-(6-chlorobenzotriazol-1-yl)-N,N,N’,N’-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
cycles were repeated until all 8 N-terminal amino acids were coupled to the resin.
Allyloxycarbonyl deprotection of diaminopropionic acid was performed under inert
atmospheric N conditions. DCM was dried over sieves for 24 hours before adding 1 mL to the
2
resin. Two equivalent of phenylsilane in 1 mL dry DCM was then added to the peptide resin
followed by 0.045 eq. of tetrakis(triphenylphosphine) palladium (0) in 1 mL dry DCM. The peptide
vessel was removed from inert conditions and allowed to react for 30 minutes. The resulting free
amine was acylated using 2 eq. of the corresponding acid (octanoic acid, 4-fluorobenzoic acid, 6-
fluoronaphthoic acid or 4’-fluoro-[1,1’-biphenyl]-4-carboxylic acid), 2 eq. of HCTU and 6 eq. of
DIPEA in DMF. The reaction mixture was left to couple overnight.
Full deprotection of the synthesized peptide was performed by adding a 2 mL mixture of 95%
trifluoroacetic acid (TFA), 2.5% triisopropylsilane (TIS) and 2.5% water to the resin and allowed
it to mix for 5 hours. The cleaved peptide was precipitated from solution using ice-cold tert-butyl
methyl ether (TBME) and centrifuged (3000 rpm, 10 minutes) resulting in a crude peptide pellet.
The supernatant was decanted and the resulting peptide pellet was dissolved in 20% acetonitrile in
water, frozen at -78 °C and lyophilized to a white crude powder. Purification was performed using
preparative HPLC-MS and purity of the resulting peptides were analyzed using analytical UHPLC.
These results are summarized in Table 4, with all compounds determined to have ≥95% purity.
Table 4 Analytical data of ghrelin(1-8) analogues 3-3j, 4a-4b, 5a-5c, 6a-6b and 7.
Compound
Number
3
Compound
[M+H]+
Calculated
947.5202
[M+H]+
Found
947.8200
Purity
(%)
96%
Yield
(%)
22%
GS-Dpr(octanoyl)-FLSPE
ACS Paragon Plus Environment
2
2

Page 23 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
3
3
3
3
a
b
c
d
e
GS-Dpr(octanoyl)-FLSPY
981.5409
965.5460
1004.5569 1004.4659 95%
932.5205
946.5362
919.5253
905.5096
919.5253
933.5409
981.4415
965.4459
98%
99%
20%
23%
19%
25%
14%
24%
19%
15%
11%
27%
10%
25%
22%
25%
18%
21%
23%
20%
26%
GS-Dpr(octanoyl)-FLSPF
GS-Dpr(octanoyl)-FLSPW
GS-Dpr(octanoyl)-FLSPN
GS-Dpr(octanoyl)-FLSPQ
GS-Dpr(octanoyl)-FLSPT
GS-Dpr(octanoyl)-FLSPS
GS-Dpr(octanoyl)-FLSPt
GS-Dpr(octanoyl)-FLSP-β-homoT
GS-Dpr(octanoyl)-FLSP-2-Nal
GS-Dpr(4-FB)-FLSPY
932.8500
946.8000
919.8689
905.3419
919.4769
933.4015
99%
98%
99%
99%
95%
99%
3
f
3
3
g
h
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
3
3
i
j
1015.5617 1015.4352 99%
977.4533 977.4948 96%
4
4
a
b
GS-Dpr(6-FN)-FLSPY
1027.4689 1027.5400 99%
1053.4846 1053.1082 99%
1055.5002 1055.8700 98%
1081.5159 1081.4673 97%
1041.4846 1041.8100 99%
1077.4846 1077.3226 96%
1077.4846 1077.3178 98%
1069.5159 1069.5098 99%
4c
5a
5b
5c
6a
6b
7
GS-Dpr(4’-FBF)-FLSPY
Aib-S-Dpr(6-FN)-FLSPY
Inp-S-Dpr(6-FN)-FLSPY
Sar-S-Dpr(6-FN)-FLSPY
GS-Dpr(6-FN)-Nal-1-LSPY
GS-Dpr(6-FN)-Nal-2-LSPY
Inp-S-Dpr(6-FN)-Nal-1-LSPT
Competitive Binding Assay (IC50
)
The affinity for the ghrelin receptor was determined using a radioligand competitive binding assay.
Assays were performed using GHS-R1a transfected HEK293 cells as receptor source and human
[125
I]-ghrelin(1-28) (PerkinElmer Inc.) as the radioligand. Human ghrelin(1-28) was used as
reference to ensure the validity of the results. A suspension of membrane from HEK293/GHS-R1a
cells (50,000 cells per assay tube) were incubated with ghrelin(1-8) peptide analogues (at
-
5
-6
-7
-8
-9
-10
-11
125
concentrations of 10 M, 10 M, 10 M, 10 M, 10 M, 10 M and 10 M) and [ I]-ghrelin
15 pM per assay tube) in binding buffer (25 mM HEPES, 5 mM magnesium chloride, 1 mM
calcium chloride, 2.5 mM EDTA, and 0.4% BSA, pH 7.4). The resulting suspension was incubated
(
1
25
for 20 minutes under shaking (550 rpm) at 37 °C. Unbound [ I]-ghrelin was removed and the
125
amount of [ I]-ghrelin bound to the membranes was measured by gamma counter. IC50 values
were determined by nonlinear regression analysis to fit a 4 parameter dose response curve using
Prism 6 (Version 6.0c). All binding assays were performed in triplicate.
ACS Paragon Plus Environment
2
3

Journal of Medicinal Chemistry
Page 24 of 43
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Molecular docking studies: The homology model of GHS-R1a developed by our lab was used to
study the interactions between the ghrelin analogs and GHS-R1a.27 Docking studies were
3
5
performed using the AUTODOCK 4.2 program. The dimensions of the active site box were
chosen to be large enough (60 60 70) to encompass the entire orthosteric site in the receptor.
Docking calculations were carried out using the Lamarckian genetic algorithm (LGA). A
maximum number of 250,000 energy evaluation was used. The docking experiment was consisted
of 200 independent runs. Default values were used for other parameters. All the calculations were
performed on high-performance computing (HPC) facilities within the Compute Canada
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
36
network. The graphics were generated using DS Visualizer (BIOVIA).
Synthesis of 6-amino-2-ethylnaphthoate (9)
6
8
-Amino-2-naphthoic acid (300 mg, 1.60 mmol) was dissolved in 50 mL of ethanol and heated to
0 °C while stirring in a 100 mL round bottom. Sulfuric acid was added dropwise until pH 6 was
reached. The reaction was left to reflux overnight. Ethanol was removed by rotary evaporation and
the crude oil was redissolved in dichloromethane and transferred to a separatory funnel. Compound
9
was extracted with 0.1 M sodium bicarbonate (50 mL) and dichloromethane (3 x 50 mL). The
organic fractions were combined and dried with MgSO . MgSO was removed by gravity filtration
4
4
and the resulting solution was removed by rotary evaporation. The crude solid was dissolved in 1
mL of dichloromethane with 0.1 % triethylamine. The crude solution was purified by flash column
chromatography (dichloromethane, 0.1% triethylamine) and tracked by TLC. The spot at R
f
= 0.4
was collected into a 100 mL round bottom and rotary evaporated to dryness. Pure 9 was isolated
as a brown solid in 86% yield (295 mg, 1.37 mmol). HRMS (ESI+): m/z calculated for C13
H
13
2
O N,
1
2
15.0946; observed 215.0942. H-NMR Spectrum (400 MHz, CDCl
3
, δ): 8.46 (1H, s, Harom), 7.96
ACS Paragon Plus Environment
2
4

Page 25 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
(
1H, d, Harom, J = 8 Hz), 7.77 (1H, d, Harom, J = 12 Hz), 7.60 (1H, d, Harom, J = 12 Hz), 7.00-6.97
2H, m, Harom), 4.42 (2H, q, CH
2
CH
3
, J = 8 Hz), 1.43 (3H, t, CH
2
CH ).
3
Synthesis of 6-dimethylamino-2-ethylnaphthoate (10)
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
Compound 9 (295 mg, 1.37 mmol) was dissolved in 60 mL of methanol in a 250 mL round bottom
and stirred at room temperature. While stirring, 37 %wt formaldehyde (2.0 mL, 27.4 mmol) was
added dropwise followed by sodium cyanoborohydride (0.86 g, 13.7 mmol) and was left to stir at
room temperature for 12 hours. After 12 hours, starting material was still present by thin layer
chromatography (TLC) (100% dichloromethane), therefore the reaction was heated to 40 °C for 4
hours. The reaction mixture was diluted with 50 mL of fresh dichloromethane and transferred to a
separatory funnel. The crude product was extracted with distilled water (50 mL) and
dichloromethane (3 x 50 mL). The organic fractions were combined and dried with MgSO
4
.
MgSO was removed by gravity filtration and the resulting solution was removed by rotary
4
evaporation to a yellow solid and dried under high vacuum. The crude solid was dissolved in 1 mL
of 10% hexanes in dichloromethane and purified by flash column chromatography (10% hexanes
in dichloromethane) and tracked by TLC. Fractions containing the spot at R = 0.55 were combined
f
into a 100 mL round bottom and rotary evaporated. Pure 10 was obtained in 64% yield (213 mg,
1
0
.87 mmol). HRMS: m/z calculated for C15
H
17
O
2
N, 243.1259; observed 243.1255. H-NMR
Spectrum (400 MHz, CDCl3, δ): 8.40 (1H, s, Carom), 7.92 (1H, d, J= 8 Hz, Carom), 7.81 (1H, d, J=
8 Hz, Carom), 7.70 (1H, d, J= 8 Hz, Carom), 7.28 (1H, d, J= 8 Hz, Carom) 6.96 (1H, s, Carom), 4.33
(
2H, q, J = 8 Hz, CH
Synthesis of 6-trimethylamino-2-ethylnaphthoate triflate (11)
Compound 10 (200 mg, 0.82 mmol) was dissolved in 30 mL of anhydrous dichloromethane dried
2
CH
3
). 3.06 (6H, s, (CH
3
)
2
), 1.35 (3H, t, J = 8 Hz, CH
2
CH
3
).
over sieves. The round bottom was evacuated and charged with N . The reaction mixture was then
2
ACS Paragon Plus Environment
2
5

Journal of Medicinal Chemistry
Page 26 of 43
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
cooled to 0 °C in an ice bath and methyl trifluoromethansulfonate (370 μL, 3.28 mmol) was added
dropwise using an oven dried needle and syringe. The reaction was left to stir. Pure 11 precipitated
out of solution after 1 hour. The precipitate was collected with a Hirsch funnel and rinsed
thoroughly with cold anhydrous dicholoromethane. The resulting white solid was dried under high
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
vacuum, with 11 being isolated in a 78% yield (260 mg, 0.64 mmol). H-NMR Spectrum (400
MHz, DMSO-d6, δ): 8.77 (1H, s, Carom), 8.62 (1H, s, Carom), 8.47 (1H, d, J= 8 Hz, Carom), 8.17-
8
.26 (3H, m, Carom), 4.40 (2H, q, J= 8 Hz, CH
2
CH
3
) 3.72 (9H, s, (CH
3
)
3
), 1.38 (3H, t, J = 8 Hz,
), 165.9 (C=O),
1
3
CH
2
CH
3
). C-NMR Spectrum (100 MHz, DMSO-d6, δ): 223.0 (Carom-N(CH
)
3 3
1
1
34.7 (Carom), 132.3 (Carom), 132.1 (Carom), 130.0 (Carom), 130.0 (Carom), 127.0 (Carom), 120.0 (Carom),
20.0 (Carom), 110.0 (Carom), 62.9 (N-(CH
3
)
3
), 61.6 (CH
2
CH
3
), 21.4 (CH
2
CH ).
3
1
9
Synthesis of [ F]-PFPN
6
-Fluoronaphthoic acid (200 mg, 1.05 mmol) and pentafluorophenol (211 mg, 1.15 mmol) were
dissolved in 100 mL of dichloromethane and cooled to 0 °C in an ice bath. N-(3-
dimethylaminopropyl)-N’-ethylcarbodiimide hydrochloride (EDC) (30.2 mg, 1.58 mmol) and 4-
(
dimethylamino)pyridine (DMAP) (25 mg, 0.21 mmol) were added to the reaction and stirred. The
reaction was tracked by thin-layer chromatography (5% ethyl acetate in hexanes, R = 0.69). After
hour, no starting material was visible by TLC. The reaction was quenched with 30 mL of
f
1
1
9
saturated ammonium chloride (NH Cl) and transferred to a 250 mL separatory funnel. [ F]-PFPN
4
was extracted with 50 mL of ethyl acetate and repeated 3 times. The organic layers were combined
and dried with magnesium sulphate (MgSO ). MgSO was removed by gravity filtration and the
solution was dried using rotary evaporation. The crude solid was dissolved in 1 mL of 5% ethyl
4
4
1
9
acetate in hexanes and loaded onto a silica plug and eluted with 5% ethyl acetate in hexanes. [ F]-
PFPN was isolated as a white solid in 86% yield (322 mg, 0.90 mmol). HRMS: m/z calculated for
ACS Paragon Plus Environment
2
6

Page 27 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
C
H
8
17
H
6
O
2
F
6
, 356.0272; observed 356.0279. H-NMR Spectrum (400 MHz, CDCl
3
, δ): 8.81 (1H, s,
arom), 8.19 (1H, d, Harom, J = 8 Hz), 8.04 (1H, dd, Harom, J = 8 Hz, 2 Hz), 7.93 (1H, d, Harom, J =
Hz), 7.57 (1H, d, Harom, J = 8 Hz), 7.41 (1H, dd, Harom, J = 8 Hz, 2 Hz) 13C-NMR Spectrum (100
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, CDCl
3
, δ): 163.9 (C=O), 162.7 (F-Cnaphthyl), 161.4 ((C=O)O-C
6
F
5
), 137.5 (C
6
F ), 137.4
5
(C ), 132.9 (C
6
F
5
6
F
5
), 132.4 (C ), 132.3 (C
6
F
5
6
F
5
), 129.4 (Cnaphthyl), 128.2 (Cnaphthyl), 128.1 (Cnaphthyl),
1
26.4 (Cnaphthyl), 123.5 (Cnaphthyl), 118.0 (Cnaphthyl), 117.7 (Cnaphthyl), 111.5 (Cnaphthyl), 111.2
(Cnaphthyl).
18
Radiolabelling Procedure of [ F]-PFPN
To prepare 11, Tracerlab FXFN (GE Healthcare) was configured as shown in figure 5. Vials were
loaded as follows: Vial 1: potassium carbonate (2.0 mg), krytofix 2.2.2 (7.0 mg) in 200 μL H O
2
and 800 μL ACN; Vial 2: 1000 μL anhydrous ACN; Vial 3: 1000 μL anhydrous ACN; Vial 4: 2
mg of 11 in 750 μL DMSO; Vial 5: 20 μL TMAH (1M) in 1000 μL ACN.
1
8
Fluorine-18 was produced on a GE Healthcare PETtrace 880 cyclotron (16.5 MeV) via a O(p,n)
nuclear reaction on a high yield Nb25 fluorine-18 target. Fluoride-18 was delivered in a 2 mL
1
8
aliquot to a QMA carbonate Sep-Pak to isolate the [ O]H O from fluoride-18. The fluoride was
2
then eluted into the reaction vial using the potassium carbonate/kryptofix solution in Vial 1. The
fluoride-18 was azeotropically dried by heating the reaction vial to 75 °C under full vacuum for 5
minutes. This repeated 2 more times using the anhydrous ACN in Vials 2 and 3. Once dried, the
radiolabelling precursor 11 was added to the reaction vial from Vial 4, heated to 120 °C and
allowed to react for 10 minutes. Following this reaction, Vial 5 was added to the reaction vial to
hydrolyze the ester, heated to 90 °C for 5 minutes before being cooled to 40 °C and completing
the automated portion of the synthesis. The reactor needle was injected down into the reaction
mixture and helium was pumped into the reaction vial for 3 minutes to transfer the crude mixture
ACS Paragon Plus Environment
2
7

Journal of Medicinal Chemistry
Page 28 of 43
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
from the reaction vial to a product vial via V14.The crude mixture was collected into 20 mL
scintillation vial containing 7 mL of 0.1% TFA in MilliQ water in the side chamber and removed
for manual synthesis.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The diluted mixture was loaded onto a conditioned tC18 Sep Pak to remove unreacted fluoride-
18. Radiolabelled 15 was eluted from the Sep-Pak with 1mL of ACN into a new reaction vial
containing 20 mg pentafluorophenol, 10 mg EDC and 1 mg of DMAP. The mixture was allowed
to react at room temperature for 5 minutes before being diluted with 500 μL of 0.1% TFA in H O
2
and injected onto the HPLC for semi-preparative purification (5 mL/min, 55% to 95% acetonitrile
in water with 0.1% TFA, 10 minutes). Purified 16 was collected at 9 minutes into a scintillation
vial and subsequently evaporated to dryness using a Biotage Speed Rotary Evaporator for 5
minutes. After evaporation, 17 in 750 μL of ACN was added to the scintillation vial followed by
3
0 μL of DIPEA. The reaction mixture was left at 40 °C for 20 minutes, diluted to 1500 μL with
0
.1% TFA in H
2
O and injected on to the HPLC for semi-preparative purification (5 mL/min, 20%
18
to 80% acetonitrile in water with 0.1% TFA). Purified [ F]-7 was collected at 5 minutes for 1.5
minutes into a 10 mL sterile vial resulting in approximately 8 mL of solvent. Solvent was removed
using a Biotage Speed Rotary Evaporator for 5 minutes resulting in a colourless residue.
18
Quality Control of [ F]-PFPN
Using an analytical radio-RP-HPLC equipped with a gamma detector, the radiochemical purity
1
8
and identity of [ F]-7 was confirmed by co-injection with a non-radioactive standard 7. Retention
18
times were consistent at 5.0 minutes. [ F]-7 was obtained in 3.1% decay correct radiochemical
yield and 95% radiochemical purity with a complete synthesis time of 1.5 hours.
ACS Paragon Plus Environment
2
8

Page 29 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
AUTHOR INFORMATION
Corresponding Author
*Email: lluyt@uwo.ca
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
Present Addresses
†School of Chemistry and Bio21 Institute, University of Melbourne, Parkville, Victoria,
Australia
Funding Sources
Canadian Institutes of Health Research; Prostate Cancer Canada; Canada Foundation for
Innovation; Ministry of Research, Innovation and Science (Ontario); Molecular Imaging
Collaborative Program, University of Western Ontario.
ACKNOWLEDGMENT
Thank you to the Nordal Cyclotron & PET Radiochemistry Facility at St. Joseph’s Hospital in
18
[18O] for these procedures. We also
London, Ontario, Canada for generously donating [ F]H
2
thank Ms. Becky McGirr for assistance with cell preparation.
ABBREVIATIONS
4
-FBA, 4-fluorobenzoic acid; 4’-FBC, 4’-fluoro-[1,1’-biphenyl]-4-carboxylic acid; 6-FNA, 6-
fluoronaphthoic acid; ACN, acetonitrile; Aib, aminoisobutyric acid; d.c., decay corrected;
DIPEA, diisopropylethylamine; Dpr, diaminopropionic acid; EDC, 1-ethyl-3-(3-
dimethylaminopropyl)carbodiimide; GHS-R, growth hormone secretagogue receptor; GOAT,
ACS Paragon Plus Environment
2
9