Rational design, discovery, and synthesis of a novel series of potent growth hormone secretagogues

Article abstract of DOI:10.1021/jm010207i

In the joint experimental and computational efforts reported here to obtain novel chemical entities as growth hormone secretagogues (GHSs), a small database of peptides and non-peptides known to have GHS activity was used to generate and assess a 3D pharmacophore for this activity. This pharmacophore was obtained using a systematic and efficient procedure, "DistComp", developed in our laboratory. The 3D pharmacophore identified was then used to search 3D databases to explore chemical structures that could be novel GHSs. A number of these were chosen for synthesis and assessment of their ability to release growth hormone (GH) from rat pituitary cells. Among the compounds tested, those with a benzothiazepin scaffold were discovered with micromolar activity. To facilitate lead optimization, a second program, a site-dependent fragment QSAR procedure was developed. This program calculates a library of chemical and physical properties of "fragments" or chemical components in a known pharmacophore and determines which, if any, of these properties are important for the observed activity. The combined use of the 3D pharmacophore and the results of the site-dependent fragment QSAR analysis led to the discovery and synthesis of a novel series of potent GHSs, a number of which had nanomolar in vitro activity.

Full text of DOI:10.1021/jm010207i

4082
J . Med. Chem. 2001, 44, 4082-4091
Ra tion a l Design , Discover y, a n d Syn th esis of a Novel Ser ies of P oten t Gr ow th
Hor m on e Secr eta gogu es
Ping Huang,# Gilda H. Loew,# Hidenori Funamizu,† Mitsuo Mimura,† Nobuo Ishiyama,† Mitsuo Hayashida,†
Tadashi Okuno,‡ Osafumi Shimada,‡ Akihiko Okuyama,§ Satoru Ikegami,§ J un Nakano,§ and Kiyoshi Inoguchi*,§
Molecular Research Institute, 2495 Old Middlefield Way, Mountain View, California 94043, and Central Research
Laboratories, Kaken Pharmaceutical Co., Ltd., 14 Shinomiya-minamikawara, Yamashina, Kyoto 607-8042, J apan
Received May 9, 2001
In the joint experimental and computational efforts reported here to obtain novel chemical
entities as growth hormone secretagogues (GHSs), a small database of peptides and non-peptides
known to have GHS activity was used to generate and assess a 3D pharmacophore for this
activity. This pharmacophore was obtained using a systematic and efficient procedure,
“DistComp”, developed in our laboratory. The 3D pharmacophore identified was then used to
search 3D databases to explore chemical structures that could be novel GHSs. A number of
these were chosen for synthesis and assessment of their ability to release growth hormone
(GH) from rat pituitary cells. Among the compounds tested, those with a benzothiazepin scaffold
were discovered with micromolar activity. To facilitate lead optimization, a second program, a
site-dependent fragment QSAR procedure was developed. This program calculates a library of
chemical and physical properties of “fragments” or chemical components in a known pharma-
cophore and determines which, if any, of these properties are important for the observed activity.
The combined use of the 3D pharmacophore and the results of the site-dependent fragment
QSAR analysis led to the discovery and synthesis of a novel series of potent GHSs, a number
of which had nanomolar in vitro activity.
In tr od u ction
Growth hormone (GH), a 191 amino acid peptide that
is synthesized and stored in the pituitary gland, is the
primary anabolic hormone of the body. It is released
from the pituitary by two distinct pathways. The first
is through the action of the hypothalamic hormones,
growth hormone releasing hormone (GHRH) which
stimulates release and by the action of somatostatin,
which is inhibitory. In the second pathway, distinct from
that of GHRH, peptides structurally unrelated to GHRH
such as the hexapeptide His-D-Trp-Ala-Trp-D-Phe-Lys-
NH2 (GHRP-6) were found to stimulate GH release. In
addition to stimulation of growth, GH is known to have
a number of basic effects on the metabolic processes,
e.g. stimulation of protein synthesis and free fatty acid
mobilization and to cause a switch in energy metabolism
from carbohydrate to fatty acid metabolism. Promising
use has been made of recombinant human GH for the
treatment of GH deficient children and adults, and the
potential clinical benefit of GH replacement therapy is
being established.1
Since the discovery of the peptidyl GH secretagogue
GHRP-6 (1, Figure 1) by Bowers and Momany et al.,2,3
GH secretagogues have received considerable attention
in the last several years with the hope for more
physiological GH release by oral administration than
* To whom correspondence should be addressed. Phone: +81-75-
594-0787. Fax: +81-75-594-0790. E-mail: inoguchi_kiyoshi@kaken.co.jp.
Dedicated to Dr. Gilda H. Loew on the occasion of her 70th
birthday.
F igu r e 1. Known GH secretagogues (1) GHRP-6; (2) L-692,-
429; and (3) MK-0677.
#
by subcutaneous injection of GH itself.4,5 Significant
progress has been made toward identifying small mol-
ecules that mimic the mechanism of action and in vivo
Molecular Research Institute.
† Department of Chemistry, Kaken.
‡ Department of Pharmacology, Kaken.
§ Department of Drug Discovery Research, Kaken.
10.1021/jm010207i CCC: $20.00 © 2001 American Chemical Society
Published on Web 10/18/2001

Potent Growth Hormone Secretagogues
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 24 4083
Ta ble 1. Compounds Chosen for 3D Pharmacophore
Development
properties of GHRP-6. Examples are the benzolactam
L-692,4296,7 (2, Figure 1) and the spiropiperidine-based
MK-06778 (3, Figure 1). Experiments in rat primary
pituitary cells indicated that L-692,429 and MK-0677
are mechanistically indistinguishable from the hexapep-
tide GHRP-6 and, hence, are peptidomimetic agonists.6,9
Subsequently, a G-protein coupled receptor in pituitary
and hypothalamus that functions in GH release was
cloned and shown to be the target of the GH secreta-
gogues.10 This receptor is distinct from both the GHRH
and somatostatin receptors. Competitive binding studies
with the GH secretagogue receptor have provided direct
evidence to show that GHRP-6, L-692,429, and MK-0677
are agonist mimetics of one another.11 A recent study
using site-directed mutagenesis and molecular modeling
also classified the binding of peptide and non-peptide
agonists in the context of its receptor environment.12
These results led to the conclusion that all peptide and
non-peptide ligands shared a common binding domain
in transmembrane (TM) region 3 of the GH secretagogue
receptor, although identical disposition of all agonists
at the binding site may not be required.12 In addition
to L-692,429 and MK-0677, some other GH secreta-
gogues reported in the literature include camphor-based
compounds,13 quinazolinones,14 and quite a number of
potent peptidyl analogues of GHRP-6,15-20 including KP-
102 (GHRP-220). Furthermore, the recent discovery of
an endogenous agonist of the GHS receptor, ghrelin,21
provides a basis for elucidating the physiological func-
tions of GHSs including GH secretion. While several
different chemical entities are undergoing preclinical
and clinical studies, it is still important to continue to
design, synthesize, and assess novel nonpeptidyl GH
secretagogues for potential clinical use.
activity
compound
sequence
EC50 (nM)a
A. Peptides
GHRP-6
His-D-Trp-Ala-Trp-D-Phe-Lys-NH2
5.0
100
0.2
1.0
10
10
>1000
[D-Lys6]GHRP-6 His-D-Trp-Ala-Trp-D-Phe-D-Lys-NH2
KP102
D-Ala-D-âNal-Ala-Trp-D-Phe-Lys-NH2
D-Ala-D-âNal-Ala-Trp-D-Phe-Lys-OH
D-Ala-D-âNal-Ala-Trp-D-Phe-OH
D-Ala-D-Phe-Ala-Phe-D-Phe-Lys-NH2
His-D-Trp-Val-Trp-D-Phe-Lys-NH2
KP102-OH
KP102 (N-5)
KP102 (Tu-4)
[Val3]GHRP-6
compound
activity EC50 (nM)
B. Non-peptdes
MK-0677
L-164080
G7134
1.3b
3.0c
0.1d
1.8e
G7220
a
For a direct comparison, activities of listed peptides were
measured at Kaken. Reference 8. c Reference 17. Reference 18.
b
d
e Reference 19.
Com p u ter -Aid ed Ra tion a l Design
The computational design strategy for the develop-
ment of novel GH secretagogues was implemented in
three stages: (1) conceptual stage, 3D pharmacophore
development; (2) exploratory stage, database search and
lead generation; and (3) optimization stage, development
of QSAR for refinement.
3D P h a r m a cop h or e Develop m en t. Although a
specific G-protein coupled GHS receptor has recently
been identified,9,10 it is still difficult to extract precise
structural information for a ligand design other than
the importance of a protonable amine. Thus, a ligand-
based computational design strategy was used. The
objective of these computational studies was to identify
and characterize the molecular determinants of GHS
activity using a set of known potent peptidyl and
nonpeptidyl agonists. Specifically, six peptidyl and four
nonpeptidyl agonists were chosen. They are listed in
Table 1, together with their EC50s for the release of GH
from rat pituitary cells. Also included in this study as
negative control was a peptide, [Val3]GHRP-6, found to
be devoid of such activity, as also shown in Table 1. The
structures of these compounds are shown in Figure 2a,b.
These agonists were used to develop 3D pharmacoph-
ores, which represent the three-dimensional arrange-
ments of specific functional groups essential for activity.
To this end, an in-house program, DistComp,22 was used
to systematically determine whether there is any com-
mon spatial arrangement of key functional groups
shared by the known GH secretagogues that are absent
in the inactive analogue. And then the resulting 3D
pharmacophore, representing a particular spatial ar-
rangement of atoms or functional groups common to the
agonists, can be used to search databases to discover
novel GH secretagogues that satisfy these requirements.
These candidate GH secretagogues can then be acquired
or synthesized and assessed for their ability to release
GH.
To this end, in the joint experimental and computa-
tional efforts reported here, a small database of peptides
and non-peptides known to have GHS activity was used
to develop and assess a 3D pharmacophore for this
activity. This pharmacophore was generated using a
systematic and efficient procedure, DistComp, developed
in our laboratory.22 The 3D pharmacophore identified
was then used to search 3D databases to discover
chemical entities that could be novel GH releasing
agonists. A number of these were chosen for synthesis
and assessment of their ability to release GH from rat
pituitary cells. Several promising lead compounds were
discovered with micromolar activity. To facilitate lead
optimization, a second program, site-dependent frag-
ment QSAR procedure was developed. This program
calculates a library of chemical and physical properties
of “fragments” or chemical components in a known
pharmacophore and determines which, if any, of these
properties are important for the observed activity. The
combined use of the 3D pharmacophore and the results
of the site-dependent fragment QSAR analysis together
with medicinal chemistry efforts and SAR studies led
to the discovery of a novel series of potent GH secreta-
gogues, a number of which had nanomolar in vitro
activity. Further modifications of these potent in vitro
agonists led to the discovery of compounds with en-
hanced bioavailability and in vivo activity. These com-
pounds to be described in later publications are prom-
ising candidates for clinically useful GH replacement
therapy.
The search for the form in which flexible molecules
such as peptides bind to receptors is a challenging task
because many low-energy conformations are accessible
and they coexist in equilibrium. The complexity in-
creases enormously when several diverse families of
fairly flexible molecules are included, and the goal is to

4084 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 24
Huang et al.
Ta ble 2. Conformational Libraries of Peptides and
Non-peptides Used To Develop the 3D Pharmacophore
conformations found for each analogue within 6 kcal/
mol in energy from the lowest energy. We see from this
table that there are many low-energy conformers within
6 kcal/mol from the lowest energy in each molecule, not
only in the peptides but also in the nonpeptidyl agonists.
All these low-energy conformations were used as input
to the DistComp program and considered in the sys-
tematic search for a 3D pharmacophore.
compound
no. of conformers ∆E e 6 kcal/mol
A. Peptides
563
GHRP-6
[D-Lys6]GHRP-6
KP102
505
105
KP102-OH
KP102(N5)
KP102(Tu4)
[Val3]GHRP-6
243
162
165
216
The first use of DistComp was to generate a 3D
pharmacophore common to all six peptidyl agonists but
absent in the inactive analogue using the set of six
pharmacophoric sites (A)-(F) shown in Figure 2a.
Among the sites (A)-(F) in the figure, site (A) represents
the N-terminal C alpha carbon, which is directly linked
to a protonated amine and a small hydrophobic group.
Sites (B) and (E) are proton-accepting groups. Sites (C),
(D), and (F) are aromatic rings. Figure 3 shows an
overlay of the six peptidyl agonists with these six
pharmacophoric sites superimposed. The conformations
used in this superimposition (Figure 3) are the lowest
energy conformer of each analogue that satisfied the
requirements of the pharmacophore. The energy of this
conformer relative to the lowest energy in each agonist
is given in Table 3. There is no low-energy conformer
of the inactive analogue [Val3]GHRP-6 (5, Figure 2a)
that shares this common peptide pharmacophore. This
is because that in the bioactive conformation of GHRP-
6, the methyl moiety of Ala3 is close to the aromatic ring
of Trp4 (∼3 Å). A direct replacement of the Ala with Val
in this bioactive conformation results in significant
steric conflicts. Therefore, [Val3]GHRP-6 is not tolerated
by the bioactive conformation of GHRP-6. This finding
provided support for the validity of the peptide phar-
macophore identified as well as an explanation for why
[Val3]GHRP-6 is inactive.
B. Non-peptides
MK-0677
L164,080
G7134
228
1073
216
G7220
168
identify the common geometric arrangements of moi-
eties that are determinants of receptor recognition or
activation because all low-energy conformations of each
molecule should be included in analysis. A novel com-
puter program, DistComp, has thus been developed22
in our laboratory to perform systematic and automated
comparisons of molecular conformations in different
molecules for the determination of 3D pharmacophores.
DistComp provides a procedure for identifying common
spatial arrangements of selected moieties in a given set
of molecules. No prior assumption of an active confor-
mation is necessary. There is also no need for a rigid
template. To achieve this goal, conformational libraries
consisting of a set of unique conformers for each
compound used are required as input to the DistComp
program. All these low-energy conformations are then
considered in the systematic search for a 3D pharma-
cophore.
Another key input central to this procedure is the
selection of sets of common functional moieties assumed
to be important for recognition or activation. The
validity of these candidate recognition or activation sites
is then assessed by the program. Specifically, for each
hypothetical set of recognition or activation moieties
selected, the program systematically determines whether
any common 3D relationships among them exist in
active analogues but are absent in inactive ones. Each
set of proposed chemical moieties that satisfies this
requirement, together with the common spatial ar-
rangements identified, comprise candidate 3D pharma-
cophores.
The next computational effort was to use the Dist-
Comp program to determine if a common 3D pharma-
cophore exists among the peptidyl and nonpeptidyl
agonists. To this end, the four nonpeptidyl agonists
shown in Figure 2b, together with their conformational
libraries, were added to the input compound list for
DistComp. The results indicated that when a subset of
the peptide pharmacophoric sites (A-C, F), shown in
Figure 2b, were used, a common 3D pharmacophore was
identified. This 3D pharmacophore, common to both
peptidyl and nonpeptidyl GH secretagogues, is shown
in Figure 4. Shown in this figure are the four pharma-
cophoric sites and the common distances found between
them. Figure 5 shows the overlay of the nonpeptidyl
agonists, MK-0677 (3),8 L-164,080 (10),17 G7134 (11),18
and G7220 (12)19 with the peptide agonist GHRP-6. We
see in this figure that there is a good superimposition
of the four pharmacophoric sites despite the structural
diversity of these active compounds. The energies ∆E
of each conformer used in the overlay relative to the
lowest energy of the molecule are shown in Table 3.
As mentioned, conformational libraries consisting of
a set of unique conformers for each compound used are
also required as input to the DistComp program. Thus,
prior to use of DistComp, the conformational libraries
of each of the selected molecules (Table 1) were gener-
ated by a procedure of repeated cycles of high (900 K)
and low (300 K) temperature molecular dynamics
combined with energy minimization of molecular struc-
tures.23 These cycles were continued until no new low-
energy conformers were generated. All these calcula-
tions were performed using QUANTA4.0/CHARMm
(Molecular Simulations, Inc., San Diego, CA). The
conformations obtained for each compound were then
clustered into unique families by a set of torsional
angles in the molecule with a window of 60°. The results
of the conformational search are summarized in Table
2. Shown in this table are the number of unique
Da ta ba se Sea r ch , Discover y, a n d Syn th esis of
Lea d Com p ou n d . The 3D pharmacophore common to
both peptidyl and nonpeptidyl GH secretagogues devel-
oped (Figure 4) was used to search databases to discover
diverse and novel leads that complied with the 3D
requirements of this pharmacophore. Because it is a
“through space” 3D pharmacophore, rather than a
“through bonds” 2D pharmacophore, compounds with

Potent Growth Hormone Secretagogues
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 24 4085
F igu r e 2. Known GH secretagogues selected for 3D pharmacophore studies: (a) six peptidyl agonists and one inactive peptide
used as a negative control. Also shown in this figure are the six pharmacophoric sites (A-F) used for determination of a common
3D pharmacophore in peptidyl agonists and (b) four nonpeptidyl agonists. Also shown in this figure are the four pharmacophoric
sites (A-C,F) used for determination of a common 3D pharmacophore for peptidyl and nonpeptidyl agonists.
entirely different molecular scaffolds may be explored
to comply with it. The 3D databases used for search
included MDDR (Molecular Design Drug Report) and
NCI (National Cancer Institute) Databases using ISIS
(MDL Information Systems, Inc., San Leandro, CA).
Compounds obtained from database search were then
evaluated and modified using the following criteria. (1)
Each candidate compound was computationally as-
sessed to verify that the 3D pharmacophore require-
ments were met by a low-energy conformer. (2) Since
our goal was to discover novel and diverse new leads,
scaffold novelty and chemical diversity as well as
conformational rigidity were important in selecting
compounds.
If the overlay of a candidate compound with the
known agonists, such that all the pharmacophoric sites
were superimposed, demonstrated a significant portion
of the candidate compound occupying a region unoc-
cupied by the known active compounds, this compound
was then modified to eliminate the extra region, if its
scaffold was of particular interest. In this process,
synthetic feasibility was imposed for a more practical
design. The criterion that we used in the compound
modification step included the compliance with the 3D

4086 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 24
Huang et al.
F igu r e 3. Stereoview showing the superimposition of the six peptides using the pharmacophoric sites (purple) shown in Figure
2a: (1) GHRP-6, green; (4) [DLys6]GHRP-6, blue; (6) KP-102, yellow; (7) KP-102-OH, red; (8) KP-102(N-5), white; and (9) KP-
102(Tu-4), lemon .
Ta ble 3. Energy of the “Bioactive Conformation” Relative to
the Lowest Energy in Each Agonist
compound
∆E (kcal/mol)
GHRP-6
1.7
3.6
3.1
5.3
2.6
3.5
3.2
4.0
2.1
2.4
[D-Lys6]GHRP-6
KP102
KP102-OH
KP102(N5)
KP102(Tu4)
G7134
G7220
MK-0677
L164080
F igu r e 4. Common 3D pharmacophore for peptidyl and
nonpeptidyl GH secretagogues. Shown in this figure are the
four pharmacophoric sites and the common distances between
them determined by DistComp. This 3D pharmacophore was
used for rational design of novel GH secretagogues.
F igu r e 5. Overlay of four nonpeptidyl agonists, MK-0677 (3,
blue), L164,080 (10, red), G7134 (11, yellow), and G7220 (12,
white), with the peptide agonist GHRP-6 (1, green) using the
pharmacophoric sites (purple) shown in Figure 4.
the activity and the physical and electronic properties
of each pharmacophoric site that comprises the 3D
pharmacophore. If so, the desired site-dependent prop-
erties could then be used as additional criteria for
optimizing and refining compounds discovered on the
basis of conformity to the 3D pharmacophore. This site-
dependent fragment QSAR method was developed to
further characterize compounds already determined to
conform to a given 3D pharmacophore defined in a
specific geometric arrangement of selected functional
groups. To this end, the most useful capability built into
the site-dependent fragment QSAR analysis was the
identification and calculation of potentially relevant
properties of each pharmacophoric site (i.e. site-depend-
pharmacophore (Figure 4) and ease of synthesis. Using
these strategies, several candidates were selected for
synthesis and biological evaluation. Among these com-
pounds, an initial new lead 16a (Figure 6) with micro-
molar activity in GH release in rat pituitary cells was
successfully discovered.
Develop m en t of a Novel Site-Dep en d en t F r a g-
m en t QSAR P r ogr a m for Lea d Op tim iza tion . The
goal of the computational studies in this stage was to
provide a rational basis for lead optimization to enhance
the activity of the initial lead. To this end, a novel site-
dependent fragment QSAR method was developed to
explore whether there were any correlations between

Potent Growth Hormone Secretagogues
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 24 4087
binding interactions such as amines, phenyl, or carbonyl
groups and their substituents or connecting groups.
The results of the site-dependent fragment QSAR
analysis for the six peptidyl agonists (Figure 2a) indi-
cated that the hydrophobicities of the sites (C), (E), and
(F) in Figure 2a were favored for activity, while the
overall hydrophobicity of the entire molecule was,
however, unfavorable. Therefore, to increase the GH
secretion activity of compounds that agree with the
peptide pharmacophore, hydrophobic (C), (E), and (F)
sites but hydrophilic nature of the whole molecule were
preferred. These results provided a useful guide for the
modification of specific pharmacophoric sites for en-
hancing GHS activities.
Ch em istr y
The lead compound 16a was designed from the result
of database searching using the strategy described
above, and the extra regions, comparing to the known
GH secretagogues in the overlay, were eliminated as
much as possible. The syntheses of 16a and its various
analogues (16b-e) were fairly straightforward as shown
in Scheme 1. Typical intermediates 13a -c were coupled
to several aromatic amino acid derivatives 14a -d to
give 15a -e. The desired products 16a and 16e were
afforded by an acidic deprotection. 16b, 16c, and 16d
were derived through an oxidation step before the
deprotection, as shown in Scheme 1.
F igu r e 6. Chemical structures and GHS activities of novel
benzothiazepin compounds.
ent properties) rather than the properties of the entire
molecule. These properties could then be used in a
regression analysis to identify the ones that modulate
activity. Among the properties calculated for each site
were the following: (1) regional net charge; (2) polar-
izability; (3) free energy of solvation; (4) van der Waals
volume; (5) hydrophobicity; (6) proton donating capabil-
ity, defined as the partial charge on the proton donor;
and (7) proton accepting capability, defined as the
partial charge on the proton acceptor. The definition of
the pharmacophoric sites or fragments already identi-
fied by Distcomp was an important input to the site-
dependent fragment QSAR analysis. These pharma-
cophoric sites are molecular fragments capable of making
Resu lts a n d Discu ssion on th e Novel
Ben zoth ia zep in Com p ou n d s
As discussed above, the modeling and database search
studies led to the discovery and synthesis of the initial
lead 16a (Figure 6). A complete conformational study
of this molecule indicated that it fits well to the 3D
pharmacophore shown in Figure 4. The four pharma-
cophoric sites (A-C, F), shown in Figure 7, conform to
those in Figure 4. An overlay of this molecule with MK-
0677 is shown in Figure 8, in which a low-energy
conformation of 16a (0.6 kcal/mol relative to the lowest
energy) has a good superimposition with MK-0677.
Sch em e 1. Preparation of Novel GH Secretagogues (16a -e)

4088 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 24
Huang et al.
ment QSAR studies which suggested that enhancing the
hydrophilic nature of the whole molecule was favorable
for activity. In addition, several analogues similar to 16e
were synthesized which retained nanomolar potency
(data not shown).
The discovery of the initial lead (16a ) and its ana-
logues (16b-e) provided some new insights for the
development of nonpeptidyl agonists: (1) the N-terminal
moiety in peptidyl agonists (Figure 2a) can be replaced
by a fairly simple aliphatic diamine; (2) a reversed
peptide backbone, as in 16a -e, is well-tolerated sug-
gesting that side chain groups, not backbone amides,
make important interactions with the receptors in these
GHS compounds; and (3) compounds 16a -e are potent
GHSs with a novel benzothiazepin scaffold. The lead
discovery strategy reported in this study is distinct from
the usual approach of lead identification via high-
throughput screens. The present study illustrates the
successful use of 3D pharmacophore models and frag-
ment QSAR analyses in drug design.
F igu r e 7. The four pharmacophoric sites in compound 16a
that are common to the pharmacophore shown in Figure 4.
Con clu sion
The computer-aided drug design strategies used in
the present study have provided a crucial basis for the
discovery of novel nonpeptidyl GH secretagogues incor-
porating a benzothiazepin scaffold. An in-house program
DistComp was used to develop a 3D pharmacophore
common to some known peptidyl and nonpeptidyl GH
secretagogues. Database searches using this pharma-
cophore together with strategies used for compound
evaluation and modification led to the discovery of a
promising lead. A site-dependent fragment QSAR method
was developed for fragmental property refinement.
Compound modifications using these results as a guide
led to the discovery and synthesis of agonists with
nanomolar in vitro activity. The studies presented here
provided an important rational basis for the develop-
ment of novel potent peptidomimetics.
F igu r e 8. Overlay of compound 16a (green) with MK-0677
(blue).
Also shown in Figure 6 are some analogues of 16a
and their GHS activities. 16b differs from 16a in the
chiral configuration and the oxidation state of the sulfur
atom. 16b was synthesized with the considerations that
a R-phenylalanine moiety (16b) would be more resistant
to proteases than the S-configuration (16a ) and that a
sulfone group (16b) could be more stable in vivo than
sulfur (16a ). The modeling study indicated that both R-
and S-chiral centers in this scaffold can conform to the
3D pharmacophore in Figure 4. In fact, 16b showed
similar activity to 16a (EC50 ) 1 µM). A replacement of
the phenyl ring in 16b with a more hydrophobic
naphthalene ring resulted in 16c with a 10-fold increase
in activity. This change was made based on the results
of the site-dependent fragment QSAR analysis described
earlier which suggested that the hydrophobicity of this
ring, which is the site (C) in Figure 7 and in the
pharmacophore (Figure 4), was favored for activity. A
further modification of 16c with the linkage one atom
longer between the benzothiazepin and the phenylala-
nine moieties led to 16d with another 10-fold increase
in activity. A complete conformational study of an
analogue similar to 16d (compound not shown) demon-
strated that with this longer linkage the molecule can
still conform well to the 3D pharmacophore shown in
Figure 4. To further improve potency, 16e was synthe-
sized to introduce a hydroxyl group to the diamine
moiety thereby increasing the solubility of the molecule.
Indeed, 16e showed potent GHS activity (EC50 ) 1 nM),
consistent with the results of the site-dependent frag-
Exp er im en ta l Section
Ch em istr y (Gen er a l). Column chromatography was per-
formed on silica gel 60 (Merck) unless otherwise noted. Mass
spectra were recorded using a J EOL J MS-HX110A mass
spectrometer. Proton magnetic resonance (1H NMR: 270 MHz)
spectra were recorded using a J EOL J NM EX-270 spectrom-
eter and reported in ppm (δ) downfield from the internal
standard tetramethylsilane.
Gen er a l P r oced u r e for th e P r ep a r a tion of 13a -c. 4-(4-
Oxo-2,3-d ih yd r o-1,5-ben zoth ia zep in -5-yl)bu ta n oic Acid
(13c). To a mixture of 125 mg (0.70 mmol) of 4-oxo-2,3-dihydro-
1,5-benzothiazepin and 280 mg (2.0 mmol) of K2CO3 in 3 mL
of DMF was added at room temperature a solution of 195 mg
(1.0 mmol) of 4-bromobutanoic acid ethyl ester in 2 mL of DMF.
After the mixture was stirred overnight, the solvent was
removed under vacuum, and then the resulting residue was
diluted with 2 mL of EtOH. To this solution was added 2 mL
of 1 N NaOH at 0 °C. The mixture was allowed to warm to
room temperature, stirred for 3 h, and then concentrated under
reduced pressure. The resulting residue was partitioned
between 0.1 N HCl (15 mL) and EtOAc (15 mL), while cooling
on ice. The organic layer was washed with brine (2 × 10 mL),
dried over anhydrous sodium sulfate, filtered, and concen-
trated to dryness. The residue was chromatographed (CHCl3:
MeOH ) 100:1) to isolate an ester form of 13c (160 mg, 85%).
Then to a solution of this ester in 5 mL of MeOH was added
1 mL of 1 N NaOH at 0 °C. After being stirred at 0 °C for 3 h,
the reaction mixture was adjusted to pH 2 with 1 N HCl and
then concentrated. The residue was diluted with 10 mL of

Potent Growth Hormone Secretagogues
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 24 4089
water and extracted with EtOAc (2 × 20 mL). The organic
layer was dried (Na2SO4), filtered, and then concentrated to
give 13c (102 mg, 79%): 1H NMR (CDCl3) δ: 1.83 (m, 2H), 2.53
(m, 4H), 3.35 (m, 2H), 3.53 (m, 1H), 4.40 (m, 1H), 7.28 (m,
2H), 7.45 (m, 1H), 7.63 (d, J ) 7.6 Hz, 1H); MS (FAB) m/z 266
(M + H)+.
2-(4-Oxo-2,3,-d ih yd r o-1,5-ben zoth ia zep in -5-yl)eth a n o-
ic Acid (13a ). A procedure similar to that above for 13c
employing 2-bromoethanoic acid methyl ester, instead of
4-bromobutanoic acid ethyl ester, gave 13a (97%):1H NMR
(CDCl3) δ: 2.68 (bs, 2H), 3.40 (bs, 2H), 4.06 (d, J ) 14.5 Hz,
1H), 4.94 (d, J ) 14.8 Hz, 1H), 7.05 (bs, 1H), 7.33 (m, 3H),
7.63 (d, J ) 7.6 Hz, 1H); MS (FAB) m/z 238 (M + H)+.
To a stirring solution of 210 mg (0.54 mmol) of 14d in 3 mL of
DMF were added 79 mg (0.30 mmol) of 13c, 92 mg (0.68 mmol)
of HOBt, and 125 mg (0.67 mmol) of EDC at 0 °C. The mixture
was allowed to warm to room temperature and stirred over-
night. The reaction mixture was poured into 40 mL of
saturated aqueous sodium hydrogen carbonate. The resulting
precipitate was collected by filtration and then dissolved in 5
mL of EtOAc. This solution was dried over anhydrous sodium
sulfate, filtered, and then concentrated to dryness. Chroma-
tography of the residue using CHCl3/MeOH (100:1) gave 15e
(240 mg, 92%): 1H NMR (CDCl3) δ: 1.42 (s, 9H), 1.70 (m, 1H),
1.93 (m, 1H), 2.15 (m, 1H), 2.38 (m, 3H), 2.73 (m, 1H), 3.22
(m, 9H), 3.70 (m, 1H), 4.35 (m, 1H), 4.83 (m, 1H), 5.30 (m,
1H), 6.22 (m, 1H), 7.03 (m, 1H), 7.23 (m, 1H), 7.40 (m, 5H),
7.60 (d, J ) 7.6 Hz, 1H), 7.67 (s, 1H), 7.82 (m, 3H); MS (FAB)
m/z 635 (M + H)+.
N-[1-(S)-(2-ter t-Bu toxycar bon ylam in oeth ylcar bam oyl)-
2-p h en yleth yl]-2-(4-oxo-2,3-d ih yd r o-1,5-ben zoth ia zep in -
5-yl)a ceta m id e (15a ). A coupling procedure similar to that
above for 15e employing 14a and 13a , instead of 14d and 13c,
gave 15a (93%): 1H NMR (CDCl3) δ: 1.43 (s, 9H), 2.63 (m, 2H),
3.20 (m, 8H), 4.05 (d, J ) 16.5 Hz, 1H), 4.55 (d, J ) 16.2 Hz,
1H), 4.75 (m, 1H), 5.70 (m, 1H), 6.99 (m, 5H), 7.28 (m, 1H),
7.45 (m, 2H), 7.65 (d, J ) 7.6 Hz, 1H); MS (FAB) m/z 555 (M
+ H)+.
N-[1-(R)-(2-ter t-Bu toxycar bon ylam in oeth ylcar bam oyl)-
2-p h en ylet h yl]-2-(1,1,4-t r ioxo-2,3-d ih yd r o-1,5-b en zot h i-
a zep in -5-yl)a ceta m id e (15b). A coupling procedure similar
to that above for 15a employing 14b, instead of 14a , gave 15b
(96%): 1H NMR (CDCl3) δ: 1.43 (s, 9H), 2.68 (m, 2H), 3.22 (m,
8H), 4.05 (d, J ) 15.5 Hz, 1H), 4.55 (d, J ) 15.5 Hz, 1H), 4.75
(m, 1H), 5.70 (m, 1H), 6.98 (m, 5H), 7.29 (m, 1H), 7.40 (m,
2H), 7.65 (d, J ) 7.6 Hz, 1H); MS (FAB) m/z 555 (M + H)+.
N-[1-(R)-(2-ter t-Bu toxycar bon ylam in oeth ylcar bam oyl)-
2-(n a p h th a len -2-yl)eth yl]-2-(1,1,4-tr ioxo-2,3-d ih yd r o-1,5-
ben zoth ia zep in -5-yl)a ceta m id e (15c). A coupling procedure
similar to that above for 15b employing 14c, instead of 14b,
gave 15c (98%): 1H NMR (CDCl3) δ: 1.43 (s, 9H), 2.61 (m,
2H), 3.29 (m, 8H), 4.00 (d, J ) 16.8 Hz, 1H), 4.55 (d, J ) 16.2
Hz, 1H), 4.85 (m, 1H), 5.70 (m, 1H), 7.07 (m, 3H), 7.40 (m,
6H), 7.62 (m, 1H), 7.75 (m, 1H); MS (FAB) m/z 577 (M + H)+.
N-[1-(R)-(2-ter t-Bu toxycar bon ylam in oeth ylcar bam oyl)-
2-(n a p h th a len -2-yl)eth yl]-3-(1,1,4-tr ioxo-2,3-d ih yd r o-1,5-
ben zoth ia zep in -5-yl)p r op ion a m id e (15d ). A coupling pro-
cedure similar to that above for 15c employing 13b, instead
of 13a , gave 15d (98%): 1H NMR (CDCl3) δ: 1.38 (s, 9H), 2.45
(m, 4H), 3.13 (m, 8H), 3.60 (m, 1H), 4.55 (m, 2H), 6.65 (m,
1H), 7.50 (m, 13H); MS (FAB) m/z 591 (M + H)+.
Gen er a l Dep r otection P r oced u r e for th e P r ep a r a tion
of 16a a n d e. N-[1-(R)-(3-Am in o-2-h yd r oxyp r op ylca r ba m -
oyl)-2-(n a p h t h a len -2-yl)et h yl]-4-(4-oxo-2,3-d ih yd r o-1,5-
ben zoth ia zep in -5-yl)bu tyr a m id e Hyd r och lor id e (16e). To
a stirring solution of 80 mg (0.13 mmol) of 15e in 2 mL of
EtOAc was added 1 mL of 4 N HCl under cooling on ice-water
and then stirring was continued at room temperature for 3 h.
The reaction mixture was concentrated under reduced pres-
sure. To the oily residue was added 5 mL of ether, and the
precipitated product was collected by filtration and dried under
reduced pressure to give 16e (69 mg, 92%) as a white solid:
1H NMR (DMSO-d6) δ: 1.43 (m, 2H), 2.05 (m, 2H), 2.43 (m,
3H), 2.85 (m, 2H), 3.23 (m, 6H), 3.72 (bs, 1H), 4.00 (bs, 1H),
4.55 (m, 1H), 5.55 (bs, 1H), 7.24 (t, J ) 7.8 Hz, 1H), 7.40 (m,
5H), 7.58 (d, J ) 7.3 Hz, 1H), 7.80 (m, 6H), 8.14 (d, J ) 7.3
Hz, 1H), 8.23 (m, 1H); MS (FAB) m/z 535 (M + H)+; HRFABMS
calcd for C29H35N4O4S (MH+), 535.2379; found, 535.2391.
3-(4-Oxo-2,3-d ih yd r o-1,5-ben zoth ia zep in -5-yl)p r op ion -
ic Acid (13b). Similarly to the N-alkylation procedure for 13c
employing 2-propenoic acid tert-butyl ester, instead of 4-bro-
mobutanoic acid ethyl ester, the ester intermediate was
prepared to give the ester form of 13b. The title compound
1
13b was obtained by acidic (TFA) hydrolysis (85%): H NMR
(CDCl3) δ: 2.58 (m, 3H), 2.88 (m, 1H), 3.35 (m, 2H), 3.70 (m,
1H), 4.53 (m, 1H), 7.30 (m, 2H), 7.48 (m, 1H), 7.61 (d, J ) 7.9
Hz, 1H); MS (FAB) m/z 252 (M + H)+.
Gen er a l P r oced u r e for t h e P r ep a r a t ion of 14a -d .
N-[3-(t er t -Bu t oxyca r b on yla m in o)-2-h yd r oxyp r op yl]-2-
(R)-a m in o-3-(n a p h th a len -2-yl)p r op ion a m id e (14d ). To a
mixture of 1.1 g (3.1 mmol) of Cbz-D-naphthylalanine and 500
mg (2.6 mmol) of 1,1-dimethylethyl-N-(3-amino-2-hydroxypro-
pyl)carbamate in 15 mL of DMF were added 530 mg (3.9 mmol)
of HOBt and 720 mg (3.8 mmol) of EDC successively under
cooling on ice. The mixture was allowed to warm to room
temperature and stirred overnight. The reaction mixture was
poured into 150 mL of saturated NaHCO3. The resulting
precipitate was collected by filtration and dried under reduced
pressure to give 1.43 g of white powder. Then a solution of
1.43 g of the obtained powder in 15 mL of DMF was stirred
under 4 atm of hydrogen in the presence of 400 mg of
palladium-on-charcoal (10%: w/w) at 35 °C for 4 d. The
reaction mixture was filtered to remove the catalyst, and the
filtrate was concentrated by evaporation. The resulting residue
was crystallized with n-hexane/EtOAc (1:1) to give 970 mg (2.5
mmol, 98%) of 14d : 1H NMR (CDCl3) δ: 1.43 (s, 9H), 1.80 (bm,
3H), 3.16 (m, 6H), 3.74 (m, 2H), 5.17 (bs, 1H), 7.35 (d, J ) 8.6
Hz, 1H), 7.46 (m, 2H), 7.65 (s, 1H), 7.80 (m, 4H); MS (FAB)
m/z 388 (M + H)+.
N-[2-(ter t-Bu toxyca r bon yla m in o)eth yl]-2-(S)-a m in o-3-
p h en ylp r op ion a m id e (14a ). A procedure similar to that
above for 14d employing Cbz-L-phenylalanine and 1,1-dim-
ethylethyl-N-(2-aminoethyl)carbamate, instead of Cbz-D-naph-
thylalanine and 1,1-dimethylethyl-N-(3-amino-2-hydroxypropyl)-
carbamate, gave 14a (57%): 1H NMR (CDCl3) δ: 1.43 (s, 9H),
1.56 (bs, 2H), 2.70 (s, 1H), 3.25 (m, 3H), 3.35 (m, 2H), 3.63 (m,
1H), 4.87 (s, 1H), 7.28 (m, 5H), 7.52 (bs, 1H); MS (FAB) m/z
308 (M + H)+.
N-[2-(ter t-Bu toxyca r bon yla m in o)eth yl]-2-(R)-a m in o-3-
p h en ylp r op ion a m id e (14b). A procedure similar to that
above for 14a employing Cbz-D-phenylalanine, instead of Cbz-
L-phenylalanine, gave 14b (67%):1H NMR (CDCl3) δ: 1.43 (s,
9H), 1.69 (bs, 2H), 2.73 (s, 1H), 3.25 (m, 3H), 3.35 (m, 2H),
3.60 (m, 1H), 4.87 (s, 1H), 7.28 (m, 5H), 7.52 (bs, 1H); MS (FAB)
m/z 308 (M + H)+.
N-[2-(ter t-Bu toxyca r bon yla m in o)eth yl]-2-(R)-a m in o-3-
(n a p h th a len -2-yl)p r op ion a m id e (14c). A procedure similar
to that above for 14b employing Cbz-D-naphthylalanine,
instead of Cbz-D-phenylalanine, gave 14c (67%):1H NMR
(CDCl3) δ: 1.43 (s, 9H), 1.56 (bs, 2H), 2.88 (s, 1H), 3.28 (m,
2H), 3.39 (m, 3H), 3.68 (m, 1H), 4.91 (bs, 1H), 7.36 (dd, J )
13.5, 13.5 Hz, 1H), 7.46 (m, 2H), 7.59 (bs, 1H), 7.65 (s, 1H),
7.80 (m, 3H); MS (FAB) m/z 356 (M + H)+.
N-[1-(S)-(2-Am in oeth ylca r ba m oyl)-2-p h en yleth yl]-2-(4-
oxo-2,3-d ih yd r o-1,5-ben zoth ia zep in -5-yl)a ceta m id e Hy-
d r och lor id e (16a ). White solid: 93% from 15a : 1H NMR
(DMSO-d6) δ: 2.43 (m, 2H), 2.82 (m, 3H), 3.09 (m, 1H), 3.32
(m, 4H), 3.84 (m, 2H), 4.57 (bs, 2H), 7.23 (m, 7H), 7.40 (m,
1H), 7.59 (d, J ) 6.3 Hz, 1H), 7.97 (bs, 2H), 8.33 (m, 2H); MS
(FAB) m/z 427 (M + H)+.
Gen er a l Cou p lin g P r oced u r e for t h e P r ep a r a t ion
of 15a -e. N-[1-(R)-(3-ter t-Bu t oxyca r b on yla m in o-2-h y-
d r oxyp r op ylca r b a m oyl)-2-(n a p h t h a len -2-yl)et h yl]-4-(4-
oxo-2,3-dih ydr o-1,5-ben zoth iazepin -5-yl)bu tyr am ide (15e).

4090 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 24
Huang et al.
Gen er a l Oxid a tion a n d Dep r otection for th e P r ep a r a -
tion of 16b, c, a n d d . N-[1-(R)-(2-Am in oeth ylca r ba m oyl)-
2-p h en ylet h yl]-2-(1,1,4-t r ioxo-2,3-d ih yd r o-1,5-b en zot h i-
a zep in -5-yl)a ceta m id e Hyd r och lor id e (16b). To a stirring
solution of 110 mg (0.20 mmol) of 15b in 2 mL of CH2Cl2 was
added a solution of 195 mg (0.0.5 mmol) of MCPBA in 4 mL of
CH2Cl2 under cooling on ice-water, and then stirring was
continued at room temperature for 3 h. The reaction mixture
was poured into 30 mL of saturated NaHCO3 under cooling
on ice-water, and then the organic layer was extracted by
EtOAc (2 × 30 mL), dried over Na2SO4, filtered, and evapo-
rated. Chromatography of the residue using CHCl3/MeOH
(100:1) gave an oil (110 mg, 94%). Then the similar procedure
to this oil for preparing 16e gave 16b (445 mg, 90%) as a white
solid: 1H NMR (DMSO-d6) δ: 2.45 (m, 1H), 2.60 (m, 1H), 2.83
(m, 3H), 3.11 (m, 1H), 3.33 (m, 3H), 3.83 (m, 2H), 4.55 (m,
2H), 7.27 (m, 6H), 7.58 (m, 1H), 7.75 (m, 1H), 7.90 (d, J ) 7.6
Hz, 1H), 7.98 (bs, 2H), 8.26 (bs, 1H), 8.52 (d, J ) 8.3 Hz, 1H);
MS (FAB) m/z 459 (M + H)+.
Refer en ces
(1) De Bore H.; Blok, G.-J .; Van Der Veen, E. Clinical aspects of
growth hormone deficiency in adults. Endocr. Rev. 1995, 16, 63-
86.
(2) Momany, F. A.; Bowers, C. Y.; Reynolds, G. A.; Hong, A.;
Newlander, K. Conformational energy studies and in vitro and
in vivo activity data on growth hormone-releasing peptides.
Endocrinology 1984, 114, 1531-1536.
(3) Bowers, C. Y.; Momany, F. A.; Reynolds, G. A.; Hong, A. On the
in vitro and in vivo activity of a new synthetic hexapeptide that
acts on the pituitary to specifically release growth hormone.
Endocrinology 1984, 114, 1537-1545.
(4) Laron, Z. Growth hormone secretagogues: clinical experience
and therapeutical potential. Drugs 1995, 50, 595-601.
(5) Saenger, P. Editorial: oral growth hormone secertagogues better
than Alice in Wonderland’s growth elixir. J . Clin. Endroclinol.
Metab. 1996, 81, 2773-2775.
(6) Smith, R. G.; Cheng, K.; Schoen, W. R.; Pong, S. S.; Hickey, H.;
J acks, T.; Bulter, B.; Chan, W. W. S.; Chaung, L. Y. P.; J udith,
F.; Taylor, J .; Wyvratt, M. J .; Fisher, M. H. A nonpeptidyl growth
hormone secretagogue. Science 1993, 260, 1640-1643.
(7) Schoen, W. R.; Pisano, J . M.; Prendergast, K.; Wyvratt, M. J .;
Fisher, M. H.; Cheng, K.; Chan, W. W. S.; Bulter, B.; Smith, R.
G.; Ball, R. G. A novel 3-substituted benzazepinone growth
hormone secretagogue (L-692,429). J . Med. Chem. 1994, 37,
897-906.
(8) Patchett, A. A.; Nargund, R. P.; Tata, J . R.; Chen, M. H.;
Barakat, K. J .; J ohnston, D. B. R.; Cheng, K.; Chan, W. W. S.;
Bulter, B.; Hickey, G.; J acks, T.; Schleim, K.; Pong, S. S.;
Chaung, L. Y. P.; Chen, H. Y.; Frazier, E.; Leung, K. H..; Chiu,
S. H. L.; Smith, R. G. Design and biological activity of L-163,-
191 (MK-0677): A potent, orally active growth hormone secre-
tagogue. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 7001-7005.
(9) Pong, S. S.; Chaung, L. Y. P.; Dean, D. C.; Nargund, R. P.;
Patchett, A. A.; Smith, R. G. Identification of a new receptor for
growth hormone secretagogues. Mol. Endocrinol. 1996, 10, 57-
61.
N-[1-(R)-(2-Am in oeth ylca r ba m oyl)-2-(n a p h th a len -2-yl-
)eth yl]-2-(1,1,4-tr ioxo-2,3-d ih yd r o-1,5-ben zoth ia zep in -5-
yl)a ceta m id e Hyd r och lor id e (16c). White solid: 79% from
1
15c: H NMR (DMSO-d6) δ: 2.53 (m, 2H), 3.50 (m, 9H), 4.68
(m, 2H), 7.63 (m, 13H), 8.38 (m, 1H), 8.60 (d, J ) 8.6 Hz, 1H);
MS (FAB) m/z 509 (M + H)+.
N-[1-(R)-(2-Am in oeth ylca r ba m oyl)-2-(n a p h th a len -2-yl-
)eth yl]-3-(1,1,4-tr ioxo-2,3-d ih yd r o-1,5-ben zoth ia zep in -5-
yl)p r op ion a m id e Hyd r och lor id e (16d ). White solid: 85%
from 15d : 1H NMR (DMSO-d6) δ: 2.33 (m, 4H), 2.85 (m, 3H),
3.60 (m, 7H), 4.49 (m, 1H), 7.48 (m, 7H), 7.78 (m, 6H), 8.31
(bs, 2H); MS (FAB) m/z 509 (M + H)+.
In Vitr o Assa y of GH Relea se fr om Ra t P itu ita r y Cells.
Compounds 16a -e were evaluated by in vitro testing for the
efficacy and potency to release growth hormone (GH) in rat
primary anterior pituitary cells. Preparation of rat primary
anterior pituitary cells was essentially the same as described
by Chen et al.24 Namely, rats were sacrificed by decapitation,
and then the pituitary was quickly removed. The anterior
pituitaries were digested with 0.2% collagenase, 0.2% hyaru-
lonidase, and 200 U/mL DNase I in Hank’s balanced salt
solution. The cells were resuspended in Dulbecco’s Modified
Eagle’s medium containing 7.5% horse serum, 5.0% fetal
calf serum, 1.0% nonessential amino acids, 100 U/mL penicil-
lin, and 100 µg/mL streptomycin and then adjusted to 1.0 ×
105 cells/mL. The suspension (0.5 mL) was placed in each
ell of 48-well trays and cultured for 3 days before the
experiments.
(10) Howard, A. D.; Feighner, S. D.; Cully, D. F.; Arena, J . P.;
Liberator, P. A.; Rosenblum, C. I.; Hamelin, M.; Hreniuk, D. L.;
Palyha, O. C.; Anderson, J .; Paress, P. S.; Diaz, C.; Chou, M.;
Liu, K. K.; McKee, K. K.; Pong, S. S.; Chaung, L. Y.; Elbrecht,
A.; Dashkevicz, M.; Heavens, R.; Rigby, M.; Sirinathsinghji, D.
J . S.; Dean, D. C.; Melillo, D. G.; Patchett, A. A.; Nargund, R.
P.; Griffin, P. R.; DeMartino, J . A.; Gupta, S. K.; Schaeffer, J .
M.; Smith, R. G.; Van der Ploeg, L. H. T. A receptor in pituitary
and hypothalamus that functions in growth hormone release.
Science 1996, 273, 974-977.
(11) Dean, D. C.; Nargund, R. P.; Pong, S. S.; Chaung, L. Y. P.;
Griffin, P.; Melillo, D. G.; Ellsworth, R. L.; Van der Ploeg, L. H.
T.; Patchett, A. A.; Smith, R. G. Development of a high specific
activity sulfur-35-labeled sulfonamide radioligand that allowed
the identification of a new growth hormone secretagogue recep-
tor. J . Med. Chem. 1996, 39, 1767-1770.
(12) Feighner, S. D.; Howard, A. D.; Prendergast, K.; Palyha, O. C.;
Hreniuk, D. L.; Nargund, R. P.; Underwood, D.; Tata, J . R.;
Dean, D. C.; Tan, C. P.; McKee, K. K.; Woods, J . W.; Patchett,
A. A.; Smith, R. G.; Van der Ploeg, L. H. T. Structural require-
ments for the activation of the human growth hormone secre-
tagogue receptor by peptide and nonpeptide secretagogues. Mol.
Endocrinol. 1998, 12, 137-145.
(13) Nargund, R. P.; Barakat, K. H.; Cheng, K.; Chan, W. W. S.;
Butler, B. R.; Smith, R. G.; Patchett, A. A. Synthesis and
biological activities of camphor-based non-peptide growth hor-
mone secretagogues. Bioorg. Med. Chem. Lett. 1996, 6, 1265-
1270.
On the experiments, cells were washed twice with the above
medium containing 20 mM HEPES (pH 7.4). GH release was
initiated by addition of the medium containing 20 mM HEPES
and a test compound. Incubation was carried out for 15 min
at 37 °C, and then released GH into the medium was measured
by a standard radioimmunoassay (RIA) procedure. EC50 values
were reported based upon the maximum GH level which
resulted from 10-8 M KP-102.
(14) Ye, Z.; Gao, Y.; Bakshi, R. K.; Chen, M. H.; Rohrer, S. P.;
Feighner, S. D.; Pong, S. S.; Howard, A. D.; Blake, A.; Birzin, E.
T.; Locco, L.; Parmer, R. M.; Chan, W. W. S.; Schaeffer, J . M.;
Smith, R. G.; Patchett, A. A.; Nargund, R. P. Modeling directed
design and biological evaluation of quinazolinones as nonpeptidic
growth hormone secretagogues. Bioorg. Med. Chem. Lett. 2000,
10, 5-8.
(15) Ankersen, M.; J ohansen, N. L.; Madsen, K.; Hansen, B. S.; Raun,
K.; Nielsen, K. K.; Thogersen, H.; Hansen. T. K.; Peschke, B.;
Lau, J .; Lundt, B. F.; Andersen, P. H. A new series of highly
potent growth hormone-releasing peptides derived from ipamo-
relin. J . Med. Chem. 1998, 41, 3699-3704.
Ack n ow led gm en t. The authors thank Drs. Danni
Harris (MRI), Sherry Chang (MRI), Katsuhiro Uchida
(Kaken), Katsuhiko Ase (Kaken), Norio Kajikawa (Kak-
en), and Katsumasa Zama (Kaken) for constructive
discussions. We also thank Ms. Miyako Ohhara (Kaken)
and Ms. Kazumi Sumita (Kaken) for NMR and mass-
spectral analyses.
Su p p or tin g In for m a tion Ava ila ble: Combustion analy-
ses and HPLC data for peptides (4-9), which were prepared
by a conventional solid-phase synthesis with using a peptide
synthesizer (PSSM8, SHIMADZU Co., Kyoto, J apan), except
for GHRP-6 which was purchased from BACHEM AG (Product
No. H-9990.0005) and used as received. This material is
available free of charge via the Internet at http://pubs.acs.org.
(16) Deghenghi, R.; Cananzi, M. M.; Torsello, A.; Battisti, C.; Muller,
E. E.; Locatelli, V. GH-releasing activity of Hexarelin, a new
growth hormone releasing peptide, in infant and adult rats. Life
Sci. 1994, 54, 1321-1328.
(17) Yang, L.; Morriello, G.; Pan, Y.; Nargund, R. P.; Barakat, K.;
Prendergast, K.; Cheng, K.; Chan, W. W.; Smith, R. G.; Patchett,
A. A. Tripeptide growth hormone secretagogues. Bioorg. Med.
Chem. Lett. 1998, 8, 759-764.

Potent Growth Hormone Secretagogues
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 24 4091
(18) Elias, K. A.; Ingle, G. S.; Burnier, J . P.; Hammonds, R. G.;
McDowell, R. S.; Rawson, T. E.; Somers, T. C.; Stanley, M. S.;
Cronin, M. J . In vitro characterization of four novel classes of
growth hormone-releasing peptide. Endocrinology 1995, 12,
5694-5699.
(19) McDowell, R. S.; Elias, K. A.; Stanley, M. S.; Burdick, D. J .;
Burnier, J . P.; Chan, K. S.; Fairbrother, W. J .; Hammonds, R.
G.; Ingle, G. S.; J acobsen, N. E.; Mortensen, D. L.; Rawson, T.
E.; Won, W. B.; Clark, R. G.; Somers, T. C. Growth hormone
secretagogues: characterization, efficacy, and minimal bioactive
conformation. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 11165-
11169.
(20) Bowers, C. Y. GH releasing peptides - - structure and kinetics.
J . Pediatr. Endocrinol. 1993, 6, 21-31.
(21) Kojima, M.; Hosoda, H.; Date, Y.; Nakazato. M.; Matsuo, H.;
Kangawa, K. Ghrelin is a growth-hormone-releasing acylated
peptide from stomach. Nature 1999, 402, 656-660.
(22) Huang, P.; Kim, S.; Loew, G. Development of a common 3D
pharmacophore for µ-opioid recognition from peptides and non-
peptides using a novel computer program. J . Comput.-Aided Mol.
Design 1997, 11, 21-28.
(23) Chew, C.; Villar, H. O.; Loew, G. H. Theoretical study of the
flexibility and solution conformation of the cyclic opioid peptides
[D-Pen2, D.-Pen5]enkephalin and [D-Pen2, L-Pen5]enkephalin.
Mol. Pharmcol. 1991, 39, 502-510.
(24) Wu, D.; Chen, C.; Katoh, K.; Zhang, J .; Clarke, I. J . The effect
of GH-releasing peptide-2 (GHRP-2 or KP 102) on GH secretion
from primary cultured ovine pituitary cell can be abolished by
a specific GH-releasing factor (GRF) receptor antagonist. J .
Endocrinol. 1994, 140, R9-13.
J M010207I