A new series of highly potent growth hormone-releasing peptides derived from ipamorelin

Article abstract of DOI:10.1021/jm9801962

A new series of GH secretagogues derived from ipamorelin is described. In an attempt to obtain ora bioavailability, by reducing the size and the number of potential hydrogen-bonding sites of the compounds, a strategy using the peptidomimetic fragment 3-(a

Full text of DOI:10.1021/jm9801962

J . Med. Chem. 1998, 41, 3699-3704
3699
A New Ser ies of High ly P oten t Gr ow th Hor m on e-Relea sin g P ep tid es Der ived
fr om Ip a m or elin
Michael Ankersen,*,§ Nils L. J ohansen,§ Kjeld Madsen,§ Birgit S. Hansen,† Kirsten Raun,‡
Karin Kramer Nielsen,∇ Henning Thøgersen,§ Thomas K. Hansen,§ Bernd Peschke,§ J esper Lau,§
Behrend F. Lundt,§ and Peter H. Andersen§
Departments of Medicinal Chemistry Research, Assay and Cell Technology, Growth Hormone Pharmacology, and
Pharmacokinetics, Health Care Discovery and Preclinical Development, Novo Nordisk A/ S, Novo Nordisk Park, DK-2760
Måløv, Denmark
Received March 30, 1998
A new series of GH secretagogues derived from ipamorelin is described. In an attempt to obtain
oral bioavailability, by reducing the size and the number of potential hydrogen-bonding sites
of the compounds, a strategy using the peptidomimetic fragment 3-(aminomethyl)benzoic acid
and sequential backbone N-methylations was applied. Several compounds from this series
release GH with high in vitro potency and efficacy in a rat pituitary cell assay and high in vivo
potency and efficacy in anesthetized rats. The tetrapeptide NNC 26-0235 (3-(aminomethyl)-
benzoyl-D-2Nal-N-Me-D-Phe-Lys-NH2) shows, following iv administration, comparable in vivo
potency to ipamorelin, GHRP-2, and GHRP-6 with an ED50 in swine at 2 nmol/kg. NNC 26-
0235 demonstrated a 10% oral bioavailability in dogs, and NNC 26-0235 and ipamorelin were
able to increase basal GH level by more than 10-fold after oral administration of a dose of 1.8
and 2.7 mg/kg, respectively. The tripeptide NNC 26-0323 (3-(aminomethyl)benzoic acid-N-
Me-D-2Nal-N-Me-D-Phe-ol) which showed moderate in vitro potency but lacked in vivo potency
demonstrated a 20% oral bioavailability in rats.
In tr od u ction
The most extensively studied GHRPs are the hexapep-
tides GHRP-2 and GHRP-6, which demonstrated GH-
releasing properties in vivo4 and were efficacious for GH
release in humans.5 The Merck group has recently
disclosed a number of nonpeptidyl mimics of GHRP-6,
among those the benzolactam L-692,429 and the de-
rivatized tripeptide MK677, which was the first potent
GH secretagogue in vivo with good oral bioavailability
in dogs.6
Human growth hormone (hGH), a 191-amino acid
hormone, is a pleiothrophic hormone interacting with
most tissues in the human body. The most prominent
effects of hGH are the growth-promoting effect, the
protein anabolic effect and the lipolytic effect. Synthesis
and release of GH from somatotrophs in the anterior
pituitary are under tight control by two hypothalamic
hormones: GH-releasing hormone (GHRH, 44 amino
acids) which stimulates synthesis and release and
somatostatin (SRIF, 14 amino acids) which inhibits GH
release.1
In addition to the effect of GHRH, Bowers discovered
in 1977 that release of GH can be induced by GH-
releasing peptides (GHRPs).2 It is widely believed that
these compounds elicit their effect on both the hypo-
thalamic and pituitary level and work synergistically
with GHRH.3 GH, GHRH, and SRIF are all polypep-
tides and incapable of being efficacious via the oral
route.
The development of orally active drugs, with the
capability of inducing GH secretion, has been speculated
for some time. Such drugs could mediate their effect
through the GHRP, GHRH, or SRIF pathway. An
advantage of drugs mediating their effect through such
pathways, compared to recombinant hGH, could be
maintenance of a physiological pulsatile pattern of GH
secretion.
The medicinal importance of peptides as oral drugs
is limited because peptides in general have two principle
disadvantages compared with most other drugs: namely,
high susceptibility to hydrolysis of the peptide bonds
and poor intestinal absorption. To improve the oral
bioavailability of peptides, a number of strategies have
been applied, such as coating of peptides, use of pen-
etration enhancers, or administration of peptidase
inhibitors.7 An alternative approach may be to chemi-
cally alter the peptidic nature by reducing the size of
the molecule and by reducing the number of potential
hydrogen-bonding sites, defined as heteroatoms with the
ability to form hydrogen bonds.8 We have previously
reported the pentapeptide ipamorelin as a highly potent
and selective GH releaser.9
Herein we report a structure-activity study of ana-
logues of ipamorelin aiming at increased oral bioavail-
ability and retained potency. Our strategy was to
gradually reduce the size and polarity of the peptide by
methylation of the amides, remove superfluous polar
groups, and yet still retain the high potency by fixing
the two aromatic group and the N-terminal amino
group, previously identified as essential pharmaco-
phores for GH-releasing activity.10
* Corresponding author. Phone: +4544434911. Fax: +4544663450.
E-mail: miak@novo.dk.
§ Department of Medicinal Chemistry Research.
† Department of Assay and Cell Technology.
‡ Department of Growth Hormone Pharmacology.
3 Department of Pharmacokinetics.
S0022-2623(98)00196-4 CCC: $15.00 © 1998 American Chemical Society
Published on Web 08/20/1998

3700 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 19
Ta ble 1. NMR Assignments of Protons of Ipamorelin
Ankersen et al.
NH
HR
Hâ, Hâ3
other
Aib
His
D-2Nal
D-Phe
Lys-NH2
1.5 (Meâ), 1.5 (Meâ3
)
NDa
8.3
8.4
4.6
4.8
4.6
4.1
2.9, 2.8
3.3, 3.0
3.1, 3.1
1.7, 1.5
6.8 (Hδ), 7.7 (Hꢀ)
7.7 (H1), 7.4 (H3), 7.9 (H4), 8.0 (H5), 7.6 (H6, H7), 7.9 (H8)
7.3 (H2, H6), 7.4 (H3, H5), 7.4 (H4)
1.0 (Hγ, Hγ), 1.5 (Hδ, Hδ), 2.9 (Hꢀ, Hꢀ), 7.0, 7.4 (NH2
8.2
a
ND, not detected.
Ta ble 2. NMR Assignments of Protons of the Major and Minor Isomer of NNC 26-0235
NH
HR
Hâ, Hâ3
other
Major Isomer
AMB
D-2Nal
N-Me-D-Phe
Lys-NH2
4.2 (CH2), 7.7 (H2), 7.3-7.6 (H4, H5, H6)
6.6-7.6 (H1-H8)
8.6
7.6
5.0
4.3
4.2
3.1, 3.1
3.2, 2.7
1.7, 1.5
2.6 (Me), 6.6-7.6 (H2-H6)
1.3 (Hγ, Hγ), 1.6 (Hδ, Hδ), 2.9 (Hꢀ, Hꢀ), 7.5 (NH2)
Minor Isomer
AMB
D-2Nal
N-Me-D-Phe
Lys-NH2
4.1 (CH2), 7.6 (H2), 7.3-7.6 (H4, H5, H6)
6.6-7.6 (H1-H8)
8.7
8.0
4.8
5.0
4.2
2.9, 2.3
2.9, 2.3
1.6, 1.5
2.8 (Me), 6.6-7.6 (H2-H6)
1.2 (Hγ, Hγ), 1.5 (Hδ, Hδ), 2.8 (Hꢀ, Hꢀ), 7.5 (NH2)
Ch em istr y
The compounds 1-3 and 5-12 were all prepared
stepwise on solid support (Rink resin) using Boc- or
Fmoc-peptide synthesis strategy based on commercially
available Boc- or Fmoc-protected amino acids. Com-
pounds 4 and 13 were prepared in a similar manner on
a N-methyl-PAL resin.11 Compound 14 was prepared
in solution using traditional amide coupling (1-hydroxy-
7-azabenzotriazole, 1-ethyl-3-(3-(dimethylamino)propyl)-
carbodiimide hydrochloride) conditions, starting from
(R)-(methylamino)-3-phenylpropan-1-ol.12 N-Methyla-
tion of the involved amino acids was obtained using
sodium hydride and methyl iodide in THF.13
NMR An a lysis of Ip a m or elin a n d NNC 26-0235.
The proton resonances of ipamorelin and NNC 26-0235
observed in water solution at 310 K have been assigned
from a series of proton 2D-NMR spectra (DQF-COSY,
TOCSY, and ROESY spectra). Probably due to fast
exchange of amide protons with the water protons, only
weak amide resonances were observable at this tem-
perature. Only one conformer of ipamorelin was ob-
served, while two distinct conformers of NNC 26-0235
were observed in solution as a result of restricted
rotation around the N-methylated D-2Nal-D-Phe amide
bond. The ratio between the two conformers was
approximately 60:40. The presence of a strong NOE
between methyl protons of N-Me-D-Phe and HR of D-2Nal
in the major conformer suggests that the predominant
conformation of the peptide bond is trans. This is
further substantiated by the observed (weak) NOE
between HR of D-2Nal and HR of N-Me-D-Phe in the
minor conformer (cis conformer), which is absent in the
most populated conformation. The assignments of
protons of ipamorelin and for the major and minor
isomer of NNC 26-0235 are provided in Tables 1 and 2.
release induced by GHRP-6). In the rat model either
male or female rats were used. No significant difference
was observed of the potency in the two genders. When
comparing the efficacy, a significantly lower Emax was
observed for the female rat. Since all the Emax results
are listed in percent of GHRP-6 Emax, these as well can
be compared. The lead compound, ipamorelin, stimu-
lates GH release from rat primary pituitary cells in a
dose-dependent manner with an EC50 of 1.4 nM and an
Resu lts a n d Discu ssion
Str u ctu r e-Activity Rela tion sh ip . The GH-releas-
ing properties of the compounds 1-14 are represented
in Table 3 by EC50 as the potency in vitro in a rat
pituitary cell assay and by ED50 representing the in vivo
potency in rats. The efficacy Emax (maximal GH re-
leased) is given as percent of GHRP-6 (the maximal GH

GH-Releasing Peptides Derived from Ipamorelin
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 19 3701
Ta ble 3. In Vitroa Potency and Efficacy in Primary Rat Pituitary Cells, in Vivob Potency and Efficacy after iv Administration in
Rats, and Chemical Characterization of the Prepared Growth Hormone Secretagogues
rat pit cell
HPLC tR (min)
PDMS
in vivo rats
b
EC50
Emax
system system
(m/z)
c
no.
peptide sequencea
(nmol) (%)
ED50
Emax
A1
B1
(M + H+)
GRHP-6 H-His-D-Trp-Ala-Trp-D-Phe-Lys-NH2
GHRP-2 H-D-Ala-D-2Nal-Ala-Trp-D-Phe-Lys-NH2
2.2
1.8
100 115M
60F
100
100
100
90
120
90
85 30M
30F
1
2
H-Aib-His-D-2Nal-D-Phe-Lys-NH2 (NNC 26-0161, ipamorelin)
H-Aib-His-D-2Nal-N-Me-D-Phe-Lys-NH2
1.3
2.0
1.5
3.8
85
7.0
10
110
0.5
5.0
170
85 85M
100 1.8M
85 14F
110
17.42
21.78
21.08
26.37
23.22
22.42
26.72
22.22
26.78
25.25
30.73
28.68
31.20
29.90
20.13
23.53
22.10
29.08
25.05
24.13
27.82
23.50
27.88
26.25
32.47
30.05
36.35
31.52
636.5
726.0
740.1
625.3
622.4
622.0
635.5
621.9
636.5
650.8
510.2
521.6
535.7
508.5
3
4
5
H-Aib-His-N-Me-D-2Nal-N-Me-D-Phe-Lys-NH2
H-Aib-His-N-Me-D-2Nal-N-Me-D-Phe-NH-Me
H-(2-AMB)-D-2Nal-D-Phe-Lys-NH2
85
60
6
H-(3-AMB)-D-2Nal-D-Phe-Lys-NH2
85 >4800M
85 65M
50
7
8
H-(N-Me-3-AMB)-D-2Nal-D-Phe-Lys-NH2
H-(4-AMB)-D-2Nal-D-Phe-Lys-NH2
30
9
H-(3-AMB)-D-2Nal-N-Me-D-Phe-Lys-NH2 (NNC 26-0235)
H-(3-AMB)-N-Me-D-2Nal-N-Me-D-Phe-Lys-NH2
H-(3-AMB)-D-2Nal-N-Me-D-Phe-NH2
H-(3-AMB)-N-Me-D-2Nal-N-Me-D-Phe-NH2
H-(3-AMB)-N-Me-D-2Nal-N-Me-D-Phe-NH-Me
H-(3-AMB)-N-Me-D-2Nal-N-Me-D-Phe-ol (NNC 26-0323)
80 80M
80 146F
85 >1000F
85 >266F
110
90
100
10
11
12
13
14
115
40
265
75 >1000F
a
b
Aib, aminoisobutyric acid; AMB, (aminomethyl)benzoic acid. Rat pituitary cell assay (mean ( SEM of 1-5 separate experiments).
Emax of GHRP-6 is 230 ( 50 ng/mL. Rats (Mmale rat; Ffemale rat), mean ED50 values ((SEM) (n ) 5-35). Emax of GHRP-6 in male and
female rats are 1167 ( 120 and 732 ( 52 ng/mL, respectively.
c
Emax at 80% of GHRP-6 and in rats with an ED50 of 85
nmol/kg and an Emax at 120% after iv administration.
N-Methylation of D-Phe in ipamorelin afforded the
compound 2, which showed equipotency in vitro to
ipamorelin but surprisingly showed a 50-fold increase
in in vivo potency compared to ipamorelin. Further
N-methylation of D-2Nal gave the dimethylated com-
pound 3, which also showed equipotency in vitro and
increased in vivo potency, although to a lesser extent.
In an attempt to evaluate the bioavailability of the
compounds following oral administration, the com-
pounds were injected intravenously (iv) or administered
orally via gavage to rats. In this model less than 1% of
GHRP-2 and ipamorelin, respectively, was absorbed
orally. The oral bioavailability of compounds 2 and 3
in this model was determined to ∼1%, and we hypoth-
esized at this stage that the low bioavailability of these
compounds was caused not only by the number of
secondary amide bonds but also to a much higher extent
by the ionic nature of both the lysine in the C-terminal
and the histidine in the N-terminal.
By replacing the N-terminal dipeptide fragment H-
Aib-His-OH with the peptidomimetic fragments 2-(ami-
nomethyl)benzoic acid (2-AMB), 3-(aminomethyl)benzoic
acid (3-AMB),14 or 4-(aminomethyl)benzoic acid (4-
AMB), we discovered that H-Aib-His-OH could be
replaced by 3-AMB (6) and still retain high in vitro
potency. This compound 6 did not release GH in rats
even at doses higher than 4800 nmol/kg, but driven by
the information that N-methylation of D-Phe in ipamo-
relin increased the in vivo potency dramatically, we
introduced the corresponding N-methyl group into 6,
affording NNC 26-0235 (compound 9); indeed the same
in vivo potency as that of ipamorelin was obtained. The
increased in vivo potency caused by the N-methylation
of ipamorelin (leading to 2) and 6 (leading to NNC 26-
0235), respectively, may be due to an increased phar-
macokinetic stability, or it may be a result of an
increased population of the cis conformer around the
D-2Nal and D-Phe amide bond, as suggested by proton
resonances of NNC 26-0235 (Table 2). The cis con-
former may in particular be important for receptor
recognition in compounds with the semirigid 3-AMB as
the N-terminal amino acid, as exemplified by more than
10-fold increase of in vitro potency going from 6 to NNC
26-0235.
A further N-methylation of D-2Nal leading to the
dimethylated compound 10 caused a slight decrease of
in vitro and in vivo potency in rats. An investigation
of the influence of the C-terminal lysine on the potency
was carried out, and by removing the lysine from 9,
resulting in 11, a 350-fold loss of in vitro potency and a
total loss of in vivo activity were observed.
Assuming the C-terminal lysine amide was unfavor-
able for oral bioavailability and the fact that we did not
totally lose the in vitro activity encouraged us to
elaborate on 11 as the lead compound. An N-methyla-
tion at D-2Nal (compound 12) slightly increased the
activity, and a further N-methylation in the C-terminal
(compound 13) increased the in vitro potency 4-5-fold.
The hypothesis that reduced size and reduced number
of hydrogen-bonding sites eventually would lead to oral
bioavailability encouraged us to further reduce the
polarity by replacing the C-terminal amide with an
alcohol. The alcohol NNC 26-0323 (14), which has the
“optimal” structure in this series, with only four hydro-
gen-bonding sites, released GH in vitro but with a lower
potency. Although NNC 26-0323 was inactive in vivo,
we were encouraged to test this compound for oral
bioavailability in rats to prove our hypothesis. In rats
the oral bioavailability (fpo) of NNC 26-0323 was deter-
mined to be 20%, indicating the importance of reduced
size (MW 509.7) and reduced number of hydrogen-
bonding sites.
In Vivo Ch a r a cter iza tion of NNC 26-0235 in
Sw in e. The fact that NNC 26-0235 (9) showed good in
vivo potency in rats and contained a reduced number
of hydrogen-bonding sites encouraged us to determine
the potency of NNC 26-0235 in species other than rats.
Swine seems to be an interesting animal model because
of their close similarity to humans regarding several
physiological aspects, including GH endocrinology.9

3702 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 19
Ankersen et al.
Ta ble 4. Effect of NNC 26-0235, Ipamorelin, GHRP-2, and
to be isosteric to the much more hydrophilic dipeptide
fragment Aib-His leading to NNC 26-0235. NNC 26-
0235 demonstrated in vivo potencies in rats and swine
comparable to those of ipamorelin and GHRP-6. NNC
26-0235 resulted in an increased basal GH level by more
than 10-fold after oral administration in dogs in a dose
of 1.8 mg/kg. Despite the reduced number of hydrogen-
bonding sites of NNC 26-0235 compared to ipamorelin,
only a slightly higher oral bioavailability was obtained.
With an oral bioavailability of 20% in rats, NNC 26-
0323 proved the hypothesis that reduced size and
reduced number of hydrogen-bonding sites would result
in improved oral bioavailability. Although NNC 26-
0323 lacked the in vivo potency in rats, it may be
considered as a lead compound for future discovery.
Our main objective of this study was to design and
prepare new GH secretagogues with high potency and
high oral bioavailability. We obtained orally active
compounds, and although the oral bioavailability was
dissatisfactory, we created the basis for a new series of
compounds with high oral bioavailability derived from
ipamorelin. These compounds are described in the
following paper.15
GHRP-6 in 40-kg Female Danish Slaughter Swinea
compd
ED50 (nmol/kg)
Emax (% of GHRP-6)
NNC 26-0235
ipamorelin
GHRP-2
2.0 ( 0.3
2.3 ( 0.03
0.6 ( 0.2*
3.9 ( 1.4
70 ( 3*
88 ( 0.3
75 ( 7
GHRP-6
100 ( 9
a
The values were calculated from dose-response curves based
on mean plasma GH concentrations obtained 10 min after
administration of test compounds. Each value is given as mean (
SEM (n ) 6-10). *Significantly different from GHRP-6 value (p
< 0.05). Emax of GHRP-6 is 74 ( 7 ng/mL.
F igu r e 1. Dose-response curves for ipamorelin, GHRP-6,
GHRP-2, and NNC 26-0235 in swine. Plasma pGH concentra-
tions obtained 10 min after iv administration of test compound.
The data are means ( SEM from six swine.
Exp er im en ta l Section
Ch em istr y. Gen er a l Meth od s. The compounds 1-13
were synthesized stepwise on a solid support using the Boc
peptide synthesis strategy on an Applied Biosystems model
430A peptide synthesizer or using the Fmoc strategy on an
Applied Biosystems model 431A peptide synthesizer. The
standard protocols supplied from the manufacturer were used.
In all cases the crude peptides were purified using semi-
preparative RP-HPLC to a purity of >95% (based on HPLC
with detection at 214 nm). The compounds were all character-
ized by analytical HPLC and plasma desorption mass spec-
trometry (see Table 3). The molecular mass found was in all
cases in agreement with the expected structure.
The RP-HPLC analysis was performed using UV detection
at 214, 254, 276, and 301 nm on a Vydac 218TP54 4.6-mm ×
250-mm 5-µm C-18 silica column, which was eluted at 1 mL/
min at 42 °C. Two different elution conditions were used.
Method A1: The column was equilibrated with 5% acetonitrile
in a buffer consisting of 0.1 M ammonium sulfate, which was
adjusted to pH 2.5 with 4 M sulfuric acid. After injection the
sample was eluted by a gradient of 5-60% acetonitrile in the
same buffer during 50 min. Method B1: The column was
equilibrated with 5% acetonitrile/0.1% TFA/water and eluted
by a gradient of 5% acetonitrile/0.1% TFA/water to 60%
acetonitrile/0.1% TFA/water during 50 min.
Consequently, NNC 26-0235, ipamorelin, GHRP-2, and
GHRP-6 were tested intravenously in swine in doses
ranging from 0.03 to 500 nmol/kg, and the results are
presented in Table 4.
As shown in Table 4, NNC 26-0235 releases GH in
swine in a dose-dependent manner (see Figure 1) with
the same ED50 as ipamorelin and GHRP-6 although a
slight but significant decrease in Emax compared to
GHRP-6 was observed. The high potency of NNC 26-
0235 together with its chemical structure, i.e., reduced
number of hydrogen-bonding sites, tempted us to de-
termine the oral bioavailability of this compound in
dogs.
Or a l Bioa va ila bility of NNC 26-0235 a n d Ip a m o-
r elin in Dogs. To get a crude impression of the oral
bioavailability of ipamorelin and NNC 26-0235, two dogs
were dosed po and iv with ipamorelin and one dog was
dosed po and iv with NNC 26-0235. Blood samples were
drawn at intervals up to 6 h after dosing and assayed
for growth hormone, ipamorelin, and NNC 26-0235,
respectively. Table 5 shows doses, Cmax, AUC(0-6 h),
and oral bioavailability (fpo) after iv and oral adminis-
tration of ipamorelin and NNC 26-0235 to beagle dogs.
Recognizing the sparse data (only two and one dog used,
respectively) and despite the fact that ipamorelin and
NNC 26-0235 were able to significantly increase basal
GH level by more than 10-fold after po administration,
the low oral bioavailability discouraged us to further
determine the GH release after po administration,
although a significant increase of basal GH level may
be expected with even lower doses.
PDMS analysis was performed on a Bio-ion (Applied Bio-
systems) system using a Californium 252 (Cf252) source on a
nitrocellulose matrix. NMR analysis was carried out in a 90%
mixture of water and deuterium oxide. Proton spectra were
recorded at 310 K on a BRUKER AMK-2 400 spectrometer.
Mixing times of 68 ms (TOCSY) and 200 ms (ROESY) were
applied. All chemical shifts are reported using the ppm scale
and with the water resonance as chemical shift reference (δ
) 4.7 ppm).
H-Aib-His-D-2Na l-D-P h e-Lys-NH2 (1, Ip a m or elin ).
A
total of 4.53 g of the peptide resin Boc-Aib-His(Trt)-D-2Nal-D-
Phe-Lys(Boc)-NH resin was prepared according to the Fmoc
strategy on an Applied Biosystems 431A peptide synthesizer
in two identical runs (to obtain sufficient material) in 1-mmol
scale using the manufacturer-supplied FastMoc UV protocols
which employ HBTU-mediated couplings in NMP and UV
monitoring of the deprotection of the Fmoc protection group.
The starting resin used for the synthesis was 2 × 1.75 g of
4-((2′,4′-dimethoxyphenyl)(Fmoc-amino)methyl)phenoxy resin
(Rink resin) (Novabiochem, Bad Soden, Germany, cat. #01-
64-0013) with a substitution capacity of 0.55 mmol/g. The
Con clu sion
A new series of GH secretagogues derived from
ipamorelin has been described. The peptidomimetic
fragment 3-(aminomethyl)benzoic acid has been shown

GH-Releasing Peptides Derived from Ipamorelin
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 19 3703
Ta ble 5. Doses, Cmax, AUC(0-6 h), and Oral Bioavailability for Ipamorelin and NNC 26-0235 after iv and po Administration to
Beagle Dogs
dose (mg/kg)
Cmax (ng/mL)
AUC (h‚ng/mL)
compd
iv
po
basal GH (ng/mL)
iv
po
iv
po
f
po (%)
ipamorelin (1)a
0.5
0.5
2.7
1.8
1.6/5.4
1.2
33/79
39
12/64
18
42/153
137
17/94
48
4/6
∼10
NNC 26-0235 (9)b
a
b
Ipamorelin was tested in two dogs; both values are listed. NNC 26-0235 was only tested in one dog.
protected amino acid derivatives used were Fmoc-Lys(Boc)-
OH, Fmoc-D-Phe-OH, Boc-D-2Nal-OH, Boc-His(Trt)-OH, and
Boc-Aib-OH. The peptide was cleaved from the 4.53-g peptide
resin by treatment with 54 mL of TFA/phenol/ethanedithiol/
thioanisole/water (40:3:1:2:2) for 3 h at room temperature.
Purification was done by semipreparative RP-HPLC, and
the final product was characterized by amino acid analysis,
analytical RP-HPLC, and plasma desorption mass spectros-
copy. Amino acid analysis and mass spectrometry were in
agreement with the expected structure within the experimen-
tal error of the method (mass spectrometry ( 0.9 amu, amino
acid analysis ( 10%). In addition 1H NMR analysis was
performed, and the spectra were assigned (see Table 1). The
NMR spectra fully supported the proposed structure.
Com p ou n d s 2, 3, a n d 5-12. Compounds 2, 3, and 5-12
were prepared in analogy to ipamorelin (1).
H-(3-AMB)-D-2Nal-N-Me-D-Phe-Lys-NH2 (9, NNC 26-0235).
1H NMR analysis was performed, and the spectra were
assigned (see Table 2). The NMR spectra fully supported the
proposed structure.
Com p ou n d s 4 a n d 13. Compounds 4 and 13 were pre-
pared in analogy to ipamorelin (1) but using N-methyl-PAL-
resin11 instead of Rink resin.
H-(3-AMB)-N-Me-D-2Na l-N-Me-D-P h e-ol (14, NNC 26-
0323). To a solution of Boc-N-Me-D-2Nal-OH (165.7 mg, 0.5
mmol), (2R)-(methylamino)-3-phenylpropan-1-ol12 (165.2 mg,
1.0 mmol), and 1-hydroxy-7-azabenzotriazole (68.1 mg, 0.5
mmol) in a mixture of dimethylformamide (2 mL) and meth-
ylene chloride (4 mL) at 0 °C was added 1-ethyl-3-(3-(di-
methylamino)propyl)carbodiimide hydrochloride (115 mg, 0.6
mmol), and the mixture was stirred for 1 h at 0 °C and for 18
h at room temperature.
The methylene chloride was removed from the mixture by
a stream of nitrogen, ethyl acetate (50 mL) was added, and
the resulting mixture was extracted sequentially with 5%
aqueous NaHCO3 (100 mL), water (100 mL), 5% aqueous
KHSO4 (100 mL), and water (100 mL). The resulting organic
phase was dried (Na2SO4) and concentrated in vacuo to an oil.
The oil was dissolved in methylene chloride/trifluoroacetic acid
(1:1) (4 mL) and stirred. After 10 min the mixture was
concentrated by a stream of nitrogen, the resulting oil was
redissolved in 70% acetonitrile/0.03 M hydrogen chloride (20
mL), water (480 mL) was added, and the mixture was
lyophilized.
acetonitrile in 0.05 M (NH4)2SO4, which was adjusted to pH
2.5 with 4 M H2SO4. The column was eluted with a gradient
of 28-38% acetonitrile in 0.05 M (NH4)2SO4, pH 2.5, at 10 mL/
min during 47 min at 40 °C, and the compound-containing
fractions were collected, diluted with 3 volumes of water, and
applied to a Sep-Pak C18 cartridge (Waters part #51910) which
was equilibrated with 0.1% trifluoroacetic acid. The compound
was eluted from the Sep-Pak cartridge with 70% acetonitrile/
0.1% trifluoroacetic acid and isolated from the eluate by
lyophilization after dilution with water: crude yield, 31.5%;
purified yield, 20.8%. HPLC: tR ) 29.90 min (A1); tR ) 31.52
min (B1). PDMS: m/z ) 508.5 (M + H)+
In Vitr o Ch a r a cter iza tion in Ra t P itu ita r y Cell Assa y.
The Sprague-Dawley male albino rats (250 ( 25 g) were
purchased from Møllegaard, Lille Skensved, Denmark. The
rats were housed in group cages (4-8 animals/cage) and placed
in rooms with 12-h light cycle. The room temperature varied
from 19 to 24 °C and the humidity from 30% to 60%.
The effect of different GH secretagogues on the GH release
from primary rat pituitary cells was measured. The method
is a modification of that of Heiman16 and is described in Raun
et al.9 Briefly, the cells were isolated from rat pituitaries, and
dispersed into single cells using trypsin, and cultured for 3
days. The cells were then washed and stimulated for 15 min
with different GH secretagogues. The supernatant was de-
canted and assayed for GH content in a rat GH SPA assay.
In Vivo Ch a r a cter iza tion in An esth etized Ra ts. The
animals (male or female) were purchased and housed under
the same conditions as described above.The method is de-
scribed in Raun et al.9 Briefly, pentobarbital dissolved in 0.9%
NaCl (final concentration 50 mg/mL) was administered ip (dose
volume 1.2 mL/kg) to male or female rats (250 ( 25 g). After
15 min, a blood sample of 0.4 mL was obtained from the orbital
vein (sample “0”). Immediately after blood sampling, 2 mL/
kg of a test compound solution was administered iv as a bolus
injection through the tail vein. After 10 min, a new sample
of 0.4 mL was obtained from the orbital vein (sample “10”).
The samples were collected in heparin-coated vials and as-
sayed for rat GH. At least five different doses of each
compound were tested using six rats per dose level.
Ca lcu la tion s. Dose-response relations were constructed
using the Hill equation (Prism, GraphPad). The efficacy
(maximal GH released, Emax) was expressed in percent of
maximal GH released by GHRP-6. The potency (EC50/ED50
)
3-Boc-(aminomethyl)benzoic acid (502.6 mg, 0.5 mmol) was
dissolved in methylene chloride (10 mL) by addition of 2 drops
of dimethylformamide and then converted to the symmetrical
anhydride by stirring with 1-ethyl-3-(3-(dimethylamino)pro-
pyl)carbodiimide hydrochloride (191.6 mg, 1.0 mmol) for 10
min. A solution of the above lyophilized 2(R)-(H-N-Me-D-2Nal-
N-Me)-3-phenylpropanol and diisopropylethylamine (342 mL)
in methylene chloride (5 mL) was added to this mixture and
reacted for 20 h at room temperature. The reaction mixture
was concentrated to an oil and redissolved in ethyl acetate (50
mL). The solution was extracted sequentially with 5% aqueous
NaHCO3 (100 mL), water (100 mL), 5% aqueous KHSO4 (100
mL), and H2O (100 mL). The resulting organic phase was
dried with Na2SO4 and concentrated in vacuo to an oil. The
oil was then dissolved in methylene chloride/trifluoroacetic acid
(1:1) (4 mL) and stirred. After 10 min the mixture was
concentrated by a stream of nitrogen, and the resulting oil was
redissolved in water (480 mL) and lyophilized.
was determined as the concentration inducing half-maximal
stimulation of the GH release as measured from the basal
level. For statistical comparisons Student’s t-test was used.
All parameters were tested for normality using the Gaussian
distribution. If the variances (F-test) were significantly dif-
ferent, then Student’s t-test with Welch’s correction was used
(Prism 2.0, GraphPad, Intuitive Software for Science).
P h a r m a cok in etics, Or a l Bioa va ila bility. Ra ts: The oral
bioavailability of NNC 26-0323 was determined in 40 fasted
male rats. For po administration the rats received 40 mg/kg
via gavage and for iv administration 1.0 mg/kg via a tail vein.
Blood was drawn by heart puncture in two animals per time
point at intervals up to 6 h after dosing.
Dogs: Oral bioavailability studies were conducted in male
and female Beagle dogs. The dogs were fasted overnight prior
to dosing. Diet was withheld for at least 3 h postdosing. A
1-week washout period separated po and iv dosing. The
compounds were administered in a vehicle of citrate/phosphate
buffer, pH 5.0. For po administration the dogs received a dose
of 2.7 mg/kg of body weight of ipamorelin and 1.8 mg/kg of
body weight of NNC 26-0235 via gavage. For iv administration
The crude product was then purified by semipreparative
HPLC in seven runs on a 25-mm × 250-mm column packed
with 7-µm C-18 silica which was preequilibrated with 28%

3704 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 19
Ankersen et al.
the dogs received a dose of 0.5 mg/kg of body weight as a bolus
in a hind leg vein. EDTA blood samples were drawn from a
front leg vein at intervals up to 6 h after dosing.
Refer en ces
(1) Merimee, T. J .; Laron, Z. Growth Hormone, IGF-I and Growth
- New Views of Old Concepts; Freund Publishing House Ltd.:
London-Tel Aviv, 1996; Vol. 4, Chapter 1.
(2) Bowers, C. Y.; Chang, J .; Momany, F.; Folkers, K. In Molecular
Endocrinology; MacIntyre, I., Ed.; Elsevier/North-Holland: Am-
sterdam, 1977; p 287.
(3) Smith, R.; Van Der Ploeg, L. H. T.; Howard, A. D.; Feighner, S.
D.; Cheng, K.; Hickey, G. J .; Wyvratt, M. J .; Fisher, M. H.;
Nargund, R. P.; Patchett, A. A. Peptidomimetic regulation of
growth hormone secretion. Endocrinol. Rev. 1997, 5, 621-645.
(4) Bowers, C. Y. GH releasing peptides - structure and kinetics.
J . Pediatr. Endocrinol. 1993, 6, 21-31.
(5) Bowers, C. Y.; Alster, D. K.; Frentz, J . M. The GH-releasing
activity of synthetic hexapeptide in normal men and short
statured children after oral administration. J . Clin. End. Metab.
1992, 74, 292-298.
Blood samples were placed on an ice-water bath im-
mediately after sampling. Plasma was separated by centrifu-
gation and stored frozen until analysis. Plasma was analyzed
for GH using a porcine GH ELISA. An HPLC assay with UV
detection and solid-phase extraction was developed for each
compound. Analytical C8 columns and disposable C3 extrac-
tion columns were used. The oral bioavailability was calcu-
lated as the total area under the plasma concentration versus
time curve following po administration divided by the area
following iv administration, appropriately corrected for dose
(eq 1):
(6) Nargund, R. P.; Van der Ploeg, L. H. T. Growth hormone
secretagogues. In Annual Reports in Medicinal Chemistry;
Bristol, J . A., Ed.; Academic Press: San Diego, CA, 1997; Vol.
32, Chapter 22.
(7) Kramer, W.; Girbig, F.; Gutjahr, U.; Kowalewski, S. In The
Intestinal Oligopeptide Transporter, in Peptide-Based Drug
Design; Taylor, M. D., Amidon, G. L., Eds.; American Chemical
Society: Washington, DC, 1995.
AUCpo‚doseiv
AUCiv‚dosepo
f )
(1)
In Vivo Ch a r a cter iza tion in Con sciou s Sw in e. For in
vivo test of the GH secretagogues, six female 30-40-kg Danish
slaughter swine of the breed Landrace Yorkshire cross (Lars
Holmenlund, DK) were used. The model is described in Raun
et al.9 Briefly, jugular catheters were inserted and fixed under
general halothane anesthesia at day 0. The test compound
was dissolved in phosphate/citrate buffer diluted in saline
containing 0.5% porcine serum albumin. Blood samples were
drawn from the jugularis catheter at frequent intervals from
1 h prior to stimulation until 3 h poststimulation. The
compounds were administered as single iv injections in
increasing doses with 24-h intervals between the doses, five
doses per experiment. Plasma was analyzed for porcine GH
(pGH) by ELISA.
(8) J ack, D. B. Drug absorption: Getting it across. Drug News
Perspect. 1997, 10 (6), 370-373.
(9) Raun, K.; Hansen, B. S.; J ohansen, N. L.; Thøgersen, H.;
Madsen, K.; Ankersen, M.; Andersen, P. H. Ipamorelin:
A
potent, selective growth hormone secretagogue. Eur. J . Endo-
crinol., submitted.
(10) Ankersen, M.; Hansen, B. S. Novel simple thioureas with growth
hormone releasing properties. Eur. J . Med. Chem. 1997, 32,
631-635.
(11) Albericio, F.; Kneib-Cordonier, N.; Biancalna, S.; Gera, L.;
Masada, R. I.; Hudson, D.; Barany, G. Preparation and applica-
tion of the 5-(4-(9-fluoronylmethyloxycarbonyl)aminomethyl-3,5-
dimethoxyphenoxy)-valeric acid (PAL) handle of the solid-phase
synthesis of C-terminal peptide amides under mild conditions.
J . Org. Chem. 1990, 55, 3730-3743.
(12) McKennon, M. J .; Meyers, A. I. J . A convenient reduction of
amino acids and their derivatives. J . Org. Chem. 1993, 58, 3568-
3571.
Ca lcu la tion s. The basal GH level, for the individual swine,
was calculated as the average of the three GH values obtained
prior to stimulation. Peak hormone levels (Cmax) adjusted for
basal level, obtained following administration of test com-
pound, were used to characterize the hormone response of
individual swine. Dose-response curve was constructed using
Cmax GH plasma concentrations. Fitting to the Hill equation
or hyperbolic Michaelis-Menten equation was performed by
nonlinear regression using Prism software (GraphPad). Using
the obtained Emax values of the compound, the potency (ED50
values) was calculated as the dose inducing half-maximal
stimulation, i.e., the increase in plasma GH from basal. All
parameters were tested for normality using the Gaussian
distribution. If the variances (F-test) were significantly dif-
ferent, then Student’s t-test with Welch’s correction was used
(Prism 2.0, GraphPad, Intuitive Software for Science).
(13) Cheung, S. T.; Benoiton, N. L. N-Methylamino acids in peptide
synthesis. V. The synthesis of N-tert-buthoxycarbonyl, N-meth-
ylamino acids by N-methylation. Can. J . Chem. 1977, 906-910.
(14) (a) Stewart, F. H. C. Synthesis of m-(aminomethyl)benzoic acid
peptides. Aust. J . Chem. 1983, 36, 2511-2516. (b) Nigam, M.;
Seong, C.-M.; Qian, Y.; Hamilton, A. D.; Sebti, S. M. Potent
inhibition of human tumor p21 farnesyltransferase by A1A2-
lacking p21 CA1A2X peptidomimetics J . Biol. Chem. 1993, 268,
20695-20698. (c) Wityak, J .; Fevig, J . M.; J ackson, S. A.;
J ohnson, A. L.; Mousa, S. A.; Parthasarathy, A.; Wells, G. J .;
DeGrado, W. F.; Wexler, R. R. Synthesis and antiplatelet activity
of DMP 757 analogues. Bioorg. Med. Chem. Lett. 1995, 5, 2097-
2100.
(15) Hansen, T. K.; Ankersen, M.; Hansen, B. S.; Raun, K.; Nielsen,
K. K.; Lau, J .; Peschke, B.; Lundt, B. F.; Thøgersen, H.;
J ohansen, N. L.; Madsen, K.; Andersen, P. H. Novel orally active
growth hormone secretagogues. J . Med. Chem. 1998, 41, 3705-
3714.
(16) Heiman, M. L.; Nekola, M. V.; Murphy, W. A.; Lance, V. A.; Coy,
D. H. An extremely sensitive in vitro model for elucidating
structure-activity relationships of growth hormone releasing
factor analogues. Endocrinology 1985, 116, 410-415.
Ack n ow led gm en t. We are thankful to Vibeke Rode,
Henriette Christensen, Annette Heewagen, Lotte G.
Sørensen, Anne G. Christiansen, Edward Kristensen,
and Lene von Voss for technical assistance.
J M9801962