Nonpeptide cholecystokinin-2 receptor agonists

Article abstract of DOI:10.1021/jm0010668

In the course of structural explorations around a series of potent CCK₂ receptor antagonists, it was noted that simple N-methylation of the indolic N-H in the parent molecule gave rise to behavior in vivo that was consistent with the compound acting as an agonist. Exploration in vitro confirmed this property, and it was shown that the agonist action could be blocked by the reference CCK₂ receptor antagonist, L-365,260. Further examples of this type of modification were explored, and a common theme with regard to agonist behavior was uncovered. Some molecular modeling is also presented in an attempt to throw light on the nature of the ligand receptor interactions that may be giving rise to the differing properties of these, apparently, structurally similar molecules.

Full text of DOI:10.1021/jm0010668

© Copyright 2001 by the American Chemical Society
Volume 44, Number 8
April 12, 2001
Articles
Non p ep tid e Ch olecystok in in -2 Recep tor Agon ists
S. Barret Kalindjian,* David J . Dunstone, Caroline M. R. Low, Michael J . Pether, Sonia P. Roberts,
Matthew J . Tozer, Gillian F. Watt, and Nigel P. Shankley
J ames Black Foundation, 68 Half Moon Lane, London SE24 9J E, U.K.
Received August 28, 2000
In the course of structural explorations around a series of potent CCK2 receptor antagonists,
it was noted that simple N-methylation of the indolic N-H in the parent molecule gave rise to
behavior in vivo that was consistent with the compound acting as an agonist. Exploration in
vitro confirmed this property, and it was shown that the agonist action could be blocked by
the reference CCK2 receptor antagonist, L-365,260. Further examples of this type of modification
were explored, and a common theme with regard to agonist behavior was uncovered. Some
molecular modeling is also presented in an attempt to throw light on the nature of the ligand
receptor interactions that may be giving rise to the differing properties of these, apparently,
structurally similar molecules.
In tr od u ction
compound is a close analogue of a number of potent AT2
antagonists such as L-158,809, which was derived from
SAR studies on losartan.5 More recently, a number of
papers6-8 from workers at Glaxo have disclosed a series
of benzodiazepine-based CCK1 receptor agonists and
antagonists which also bear a close structural homology
to each other. This area has recently been the subject
of a review by Sugg.9 Prompted by this literature we
wish to report on our observations regarding the dis-
covery of nonpeptide agonists at CCK2 receptors. These
compounds arose as part of a program in which the SAR
of potent antagonists of these receptors were being
explored.10 The compounds differ from most other
described CCK111-15 and CCK2 receptor agonists16 which
have been derived by a more direct consideration of the
structure of tetragastrin, the minimum fully active
fragment of the hormones, cholecystokinin and gastrin,
at CCK2 receptors.17 In addition, we suggest, from a
computer-generated molecular model, that steric rather
than electrostatic fields around the molecules are an
important determinant of the expression of efficacy in
this receptor system.
Agonists of peptide hormone receptors have, in gen-
eral, been derived by minor structural modifications of
the parent hormone.1 In addition, a number of com-
pounds, shown to be antagonists at peptide hormone
receptors in particular tissues, have been shown to
behave as agonists when examined in related systems.
An example of this may be found in a series of peptoids,
devised by Hughes et al.2 as antagonists at cholecysto-
kinin-2 (CCK2) receptors, which were found to exhibit
efficacy.3 Peptides and close analogues usually suffer
from a number of deficiencies which make their utility
as drugs limited. Therefore it is surprising that there
are relatively very few reports in the literature of
nonpeptide-based small molecules that are agonists at
peptide hormone receptors as these would make more
credible drug candidates for any medical conditions
requiring an enhancement of the endogenous effect of
the peptide hormone. Kivlighn et al.4 have described the
discovery of L-162,313, a nonpeptide that mimics the
actions of angiotensin II at the AT2 receptor. This
* Author for correspondence. Tel: 44 (0)20 7737 8282. Fax: 44
(0)20 7274 9687. E-mail: barret.kalindjian@kcl.ac.uk.
10.1021/jm0010668 CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/10/2001

1126 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 8
Sch em e 1. Synthesis of Anhydrides 14, 18, and 20
Kalindjian et al.
Sch em e 2. Synthesis of Regioisomeric Mixtures 3, 8, and 9
Ch em istr y
the diacid 19 followed by ring closure in hot acetic
anhydride gave rise to the N-acylated anhydride 20.
Preparation of the mixtures of regioisomers from the
anhydrides described above is shown in Scheme 2. In
each case the anhydride was opened with 1-adaman-
tanemethylamine in a mixture of triethylamine and
THF to yield an approximately 1:1 mixture of regioiso-
meric amide acids. Treatment of the amide acids with
compound 24 and a variety of coupling reagents gave
rise to 1:1 mixtures of the benzyl ester-protected tri-
amides. The target compounds were isolated as regio-
isomers following deprotection using hydrogen gas over
a 10% palladium on charcoal catalyst in a solvent
mixture of methanol and THF.
The anhydrides described for the first time in this
paper were prepared according to the methodology out-
lined in Scheme 1. The N-methylbenzimidazole anhy-
dride 14 was prepared from dimethyl 4,5-diamino-
phthalic acid 10 by treatment with hot formic acid to
afford the benzimidazole derivative 11, which was
N-methylated using sodium methoxide and methyl io-
dide. The dimethyl ester 12 was hydrolyzed under basic
conditions and the resulting diacid 13 converted to the
target anhydride 14 using hot acetic anhydride. The
modified indole anhydrides were both prepared from
dimethyl indole-5,6-dicarboxylate 15. N-Methylation of
this intermediate attained using cesium carbonate and
methyl iodide followed by hydrolysis and ring closure
gave the requisite anhydride 18, whereas hydrolysis to
As the mixtures shown in Scheme 2 could not be
separated at any stage by column chromatography, the
regioisomerically pure indole derivatives required for

Nonpeptide CCK2 Receptor Agonists
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 8 1127
Sch em e 3. Synthesis of Compounds 4-7a
a
(a) NaH, MeI, THF; (b) NaH, nBuI; (c) H2, 10% Pd-C, MeOH/THF.
Ta ble 1. Comparison of Receptor Activity Values for CCK2
Ligands in Various Assays
testing had to be prepared from separate intermediates
that have already been described.10 These compounds
were prepared as shown in Scheme 3, initially by
alkylation of compounds 28 and 29 using sodium
hydride and the appropriate alkyl iodide, followed by
hydrogenation using a catalyst of 10% palladium on
charcoal in a solvent mixture of methanol and THF.
Resu lts a n d Discu ssion
rat
mouse
cortexd
mouse
We have recently described a series of nonpeptide
CCK2 receptor antagonists based on a disubstituted
indole framework.10 These particular compounds, ex-
emplified by 1 and 2, had nanomolar affinities in the
two principal CCK2 receptor assays as shown in Table
1: namely the immature, isolated, lumen-perfused,
pentagastrin-stimulated rat stomach18 and a radioli-
gand binding assay in mouse cortical membranes using
no.a
Y
X
stomachc
stomache
1
NH
CH
CH
NH
CH
CH
9.90 ( 0.18 8.96 ( 0.24
9.26 ( 0.28 7.87 ( 0.16
9.80 ( 0.34 8.42 ( 0.11
8.99 ( 0.18 9.43 ( 0.18
nof
nog
yes
yes
yes
yes
yes
yes
yes
2
3b
4
NMe
NMe
CH
NBu
CH
5
6
7
NMe 8.57 ( 0.20 8.36 ( 0.02
CH
NBu
7.35 ( 0.24 8.44 ( 0.30
6.86 ( 0.23 7.52 ( 0.21
8.72 ( 0.44 8.74 ( 0.09
8.95 ( 0.45 7.60 ( 0.03
8
NCOCH3 CH
NMe
[
125I]BH-CCK-8S as radiolabel.19 Roberts et al.20 and
9b
N
Harper et al.19 suggest the existence of at least two
CCK2 receptor subtypes with apparently single and
differing populations being present in these two par-
ticular tissues. From the data obtained for compound
1, it may be suggested that it is slightly more selective
for the single receptor site found in the rat stomach
relative to the site found in the mouse cortex. In
contrast, the isolated, lumen-perfused, pentagastrin-
stimulated mouse stomach assay has been shown20 to
contain a mixed and variable population of at least two
CCK2 receptors, and this leads to complex behavior for
selective antagonists. The affinity estimate obtained in
the mouse stomach (pKB 8.81) was similar to the value
found in the cortex, consistent with the lower-affinity
site being predominant in this particular experiment.
Due to the potential complexity in the behavior of
compounds in the mouse stomach bioassay, compounds
were not routinely assessed in this system. However,
with respect to prototype drug discovery in the CCK2
receptor field, the most important and useful property
of this mouse stomach assay is that of all the in vitro
a
Compounds were tested as the bis(N-methyl-D-glucamine)
salts. b Compounds indicated were an approximately 50:50 mixture
of regioisomers between the compound shown and that with X and
Y reversed. c pKB ( SEM values were estimated from single shifts
of pentagastrin concentration-effect curves in the isolated, lumen-
perfused immature rat stomach, calculated assuming an underly-
ing Schild slope of unity and fitted using the Gaddum-Schild
equation. pKB ( SEM competition with 20 pM [125I]BH-CCK-8S
d
for CCK2 binding sites in mouse cortical homogenates from at least
three separate experiments. e Observation of acid secretion due
to compound alone in the isolated, lumen-perfused mouse stomach.
f pKB value of compound estimated from the lowest concentration
which produced a significant shift of pentagastrin concentration-
effect curves (8.81 ( 0.21) with assumptions and calculations as
g
in footnote c. pKB value of compound estimated from the lowest
concentration which produced a significant shift of pentagastrin
concentration-effect curves (7.16 ( 0.43) with assumptions and
calculations as in footnote c.
assays available, it is the most sensitive for detecting
residual efficacy in molecules. This may be rationalized
in terms of a number of tissue-dependent factors such
as high receptor density or particularly efficient receptor
G protein coupling.21 Therefore, compounds such as PD-

1128 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 8
Kalindjian et al.
134,308, which appear as antagonists in assays such as
the isolated, immature, lumen-perfused rat stomach,
can behave as low-efficacy agonists when examined in
the analogous mouse stomach assay.22 It should be
noted that when PD-134,308 was examined in a chronic
gastric fistula dog model in vivo, it behaved as a full
agonist (data not shown) indicating that the relatively
low level of agonism manifest in the mouse stomach was
capable of substantial amplification. Thus examining
selected compounds in the mouse stomach in order to
minimize the chance of encountering agonist-type be-
havior in vivo is an important part of our screening
strategy. Compounds 1 and 2 were found to be devoid
of efficacy when examined in the mouse stomach assay,
a finding that was further confirmed when compound
1 was examined in the chronic gastric fistula dog
model.10
F igu r e 1. Agonist dose-response curves in isolated, lumen-
perfused mouse stomach for compound 3 in the presence and
absence of L-365,260 and pentagastrin (n ) 6).
As part of the exploration of the salient features
controlling the activity of compound 1, it was decided
to examine whether the hydrogen attached to the indole
nitrogen atom had any direct interaction with the
receptor via a hydrogen bond. For this reason, the indole
N-methyl compound was specified and synthesized,
initially as a mixture of regioisomers 3. The affinity of
this compound at CCK2 receptors was similar to that
of compound 1 in both the rat stomach assay and the
mouse cortex radioligand binding assay. The first
surprise with this compound came when it was exam-
ined in vivo in the chronic gastric fistula dog,10 where
it was ineffective at inhibiting pentagastrin-stimulated
acid secretion at a dose of 1 µmol/kg by intravenous
bolus (data not shown). This dose was 10-fold greater
than that of compound 1, which was able to cause a
substantial inhibition in the same model. Intrigued by
this result, we speculated that this apparent lack of
antagonist activity might be because compound 3 was
behaving as an agonist.
Ta ble 2. Characterization of Agonist Response of Ligands in
Isolated Mouse Stomach Bioassay
agonist curve
locationa
affinity of
slope of
Schild plot
no.
L-365,260b
3
4
5
8.40 ( 0.40c
8.69 ( 0.15
7.94 ( 0.17
7.73 ( 0.27
7.04 ( 0.35 (pKB)d NDe
7.02 ( 0.14 (pKB)
7.90 ( 0.24 (pA2)
8.01 ( 0.30 (pA2)
1.02 ( 0.22
0.39 ( 0.24
0.61 ( 0.15
pentagastrin
a
Mean pA50 based on the results of 7-10 different control
b
concentration-response curves. pKB of L-365,260 obtained from
Schild analysis with slope not significantly different from unity.
A pA2 was calculated when the Schild plot slope parameter was
significantly different from unity and represents the shift of the
agonist curve produced by the lowest concentration of the antago-
nist that produces a significant shift. c The standard errors as-
sociated with this data point are large due to the steepness of the
d
dose-response curve. pKB of L-365,260 against agonism of the
compound based on at least six individual data points at only a
single concentration of the antagonist, calculated assuming an
underlying Schild slope of unity and fitted using the Gaddum-
Schild equation. e Schild analysis not performed.
To address this possibility in vitro, we decided to
examine the behavior of the compound in the isolated
mouse stomach bioassay. In this assay, compound 3 was
found to stimulate basal acid secretion, and the con-
centration-response curve obtained for the compound
suggested that it was behaving as a partial agonist with
a potency approximately equal to that of pentagastrin.
We confirmed that this acid secretion was mediated by
CCK2 receptors by looking at whether the agonism could
be blocked by the presence of a receptor antagonist and
used L-365,26023 for this experiment. It was found that
inclusion of this antagonist shifted the dose response
of compound 3, and the pKB obtained for the antagonist
was approximately 7.0 (Figure 1 and Table 2). While
this figure is rather lower than the commonly reported
affinity of this compound at CCK2 receptors (for example
ref 23 quotes a pKi of 8.7), both we20 and others24 have
shown that in some tissues there are sites for which the
compound has an affinity of this order. Given this low
value, it was possible to infer that the agonism observed
in this experiment was being mediated only through the
CCK2 site for which L-365, 260 has the lower affinity,
and this is the one that is commonly found in the rat
stomach. The dose-response curve obtained for penta-
gastrin in the same experiment is shown for compara-
tive purposes. The agonism was confirmed in vivo in
the chronic gastric fistula dog, as acid secretion could
F igu r e 2. Typical acid secretory response due to the admin-
istration of an intravenous bolus dose of compound 3 and a
subcutaneous bolus dose of pentagastrin to conscious, gastric
fistula dogs. Each data block represents the total acid collected
in a 15-min time period.
be seen when the compound was administered intra-
venously. Figure 2 shows the magnitude of this response
relative to that induced by a 4 µg/kg subcutaneous bolus
dose of pentagastrin.

Nonpeptide CCK2 Receptor Agonists
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 8 1129
F igu r e 3. Typical low-energy fields of compounds 1 (top left), 2 (top right), 4 (bottom left), and 5 (bottom right) showing positive
(red), negative (green), and steric (yellow) field points.
The mixture of regioisomers present in 3 were sepa-
rated in order to see whether there was any difference
in activity between them. As shown in Table 1, it was
found that both the individual isomers 4 and 5 had
broadly similar receptor affinities to their corresponding
nonmethylated analogues and that they behaved as
partial agonists in the mouse stomach assay, with com-
pound 4, the regioisomer corresponding to compound 1,
appearing slightly more potent. L-365,260 (30 nM-1
µM) behaved as an antagonist toward the acid secretion
expressed by compound 4, a finding that was consistent
with this effect being CCK2 receptor-mediated agonism.
The Schild slope parameter was not significantly dif-
ferent from unity, indicating that the agonism produced
by compound 4 was via a homogeneous receptor popula-
tion for which L-365,260 expressed its lower affinity
(Table 2, pKB 7.02 ( 0.14). Interestingly, a similar
experiment performed with compound 5 as the agonist
had a flat Schild plot slope (b ) 0.39 ( 0.24), and the
pA2 estimated from the lowest concentration of L-365,-
260 which produced a significant shift of the agonist
curve was approximately 7.90 suggesting that the acid
secretion of this compound might be occurring through
both the putative receptor sites present in this tissue.
As discussed above, given the variable expression of the
two receptors in the mouse stomach and relatively small
amount of agonism detected, such effects are difficult
to quantify accurately.
version of compound 3, and an N-methylbenzimidazole
derivative 9. While maintaining respectable affinity at
CCK2 receptors, all these molecules, to a greater or
lesser extent, caused an acid secretory effect in the
mouse stomach. The amounts observed were generally
small, so that it was not feasible to test whether the
acid secretion was sensitive to L-365,260. However,
given the structural similarities to compound 3 and the
consistency of the observation of the phenomenon with
this type of modification, we felt it was reasonable to
conclude that the efficacy expressed by all these com-
pounds was through CCK2 receptors.
To try to rationalize the emergence of residual efficacy
in the N-substituted indoles, the multiconformational
composite molecular potential field method of Vinter
and Trollope25,26 was used to compare compounds 1, 2,
4, and 5. The first two compounds were antagonists10
and the latter two were agonists when examined in the
mouse stomach bioassay. Electrostatic and van der
Waals field points10 were calculated for all low-energy
conformations of each molecule, and all possible com-
binations of these conformational sets were compared
on the basis of these parameters. The results are
illustrated in Figure 3, which shows typical low-energy
representations of the four molecules, with the corre-
sponding positive (red) and negative (green) electrostatic
field points surrounding them. When represented in this
way, while there are evident differences between mol-
ecules, we could see no consistent pattern of field points
which distinguishes the antagonists from the agonists
in the mouse stomach bioassay. Instead one must look
to the steric differences around the indolic nitrogen atom
for an explanation of the observations. Clearly, where
A number of further compounds were made in which
the indolic N-H was substituted, and all those listed
in Table 1 behaved qualitatively as partial agonists in
the mouse stomach. The molecules made included the
nbutyl homologues of compounds 4 and 5, the acetyl

1130 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 8
Kalindjian et al.
even modest bulk has been introduced, such as in
compounds 4 and 5 with the presence of a methyl group,
residual efficacy has become a property of these mol-
ecules. Intriguingly, substitution of both regioisomers
gives rise to the same phenomenon, and there are two
possible ways of rationalizing this finding. First, if one
assumes that both compounds 4 and 5 bind to the
receptor with the same relative orientation of the two
side chains, then the methyl groups will vary in their
position within the receptor binding pocket. This sug-
gests that the feature of importance with regard to
triggering efficacy at the receptor has to be equally
accessible to methyl groups in both positions. To allow
equivalent access, this putative residue on the receptor
would probably have to be located in the plane perpen-
dicular to that occupied by the bicyclic ring. The second
possibility is that both N-substituted indoles are some-
what wider than their unsubstituted counterparts. Thus
sterically, both would occupy a greater area of space and
so may be impinging on residues in the receptor protein
that might trigger efficacy.
molecules, both of these compounds bind at the same
site on the unactivated CCK2 receptor when they are
behaving as antagonists as is the case in the rat
stomach assay. However, compound 4, which possesses
residual efficacy, is able to induce, or stabilize, an
activated form of the receptor, and in certain tissues,
such as the mouse stomach, this is evident as agonism.
Thus, unless compound 4 and other analogues possess-
ing residual efficacy are able to migrate substantially
within the receptor, or there are a multitude of path-
ways and binding positions possible from which to
activate this receptor, a possibility that cannot be
discounted based on a recent report from Blacker et
al.,30 it is likely that both molecules, the agonist and
the antagonist, are sharing the same principal recogni-
tion site as the parent hormone. Thus, logically in this
case, in contrast to the situation pertaining in the point
mutation studies for substance P receptors referred to
above, there is a substantial congruence between the
site of interaction of the parent hormone with the
receptor and the location of the binding pocket for the
antagonist, compound 1.
The contrast in behavior of the CCK2 and substance
P systems illustrates the dynamic nature of ligand-re-
ceptor interactions and indicates that there may be sev-
eral possible sites from which an antagonist can exert
an inhibitory influence on the actions of an agonist.
When an agonist binds to the receptor it is able to induce
conformational changes in that receptor, via a number
of possible interactions, which lead to a cascade of
events resulting in a signal. Antagonists, it would
appear, fall into two distinct groups. There are those
which compete directly for the agonist site but which
either do not interact with or indeed, stabilize, the
agonist ‘trigger’. Alternatively, there are those which
bind at other sites in the receptor protein either blocking
conformational changes within the receptor that lead
to signal transduction or promoting alternative confor-
mational changes in a such a way as to distort the
binding site for the agonist. This latter mechanism im-
plies an allosteric interaction between the two sites.
In conclusion, during a program aimed at searching
for CCK2 receptor antagonists, we have found that a
particular structural change gives rise to compounds
that show agonist behavior in some of our assays. We
have interpreted this observation in terms of subtle
differences in the steric, rather than electrostatic, fields
around the molecules. It is evident that seemingly minor
changes in the structure of the ligand can lead to
fundamental changes in the nature of the ligand-
receptor interaction not only with respect to affinity but
also in the expression of efficacy.
It should be noted that apparently a similar steric
requirement pertains when one considers the agonist
properties of the CCK1 ligands reported by Aquino,6 in
that the NH compound 34 is an antagonist, whereas
the more sterically demanding, N-ethyl compound 35
behaves as an agonist. This suggests that the activation
mechanism of the CCK1 and CCK2 receptors might
share some common features.
Our findings, and the other recent reports of nonpep-
tide agonists at peptide hormone receptors, also start
to challenge a view that has been gaining credence
about these classes of receptors: that peptide agonists
and nonpeptide antagonists at the same receptor always
bind at different sites on peptide hormone receptors.
Certainly there is good evidence from point mutation
studies that peptide hormones such as substance P bind
to the extracellular domain of their GPCR receptors,27,28
while smaller nonpeptide antagonists interact with
residues in the transmembrane helices. However, there
is physical evidence to suggest that this is not always
the case: For example, in the CCK1 receptor it is
suggested that the hormone penetrates the transmem-
brane region.29 The data described herein supports the
more traditional view that agonist and antagonist
binding sites on the protein can be congruent. Com-
pounds 1 and 4 have broadly similar affinities in the
rat stomach, whereas only compound 4 behaves as an
agonist at the receptors in the mouse stomach. The
nature of the receptor present in the two assays when
our experiments were performed could not be distin-
guished on the basis of their affinity for L-365,260.20
The simplest explanation of these data is that, given
the similarities in the affinity and structure of the
Exp er im en ta l Section
Gen er a l. NMR spectra were recorded on either a Nicolet
GE300 or a Bruker DRX 300 machine. Elemental analyses
were carried out at the London School of Pharmacy, and all
compounds gave analytical results of (0.4% of theoretical
values. Flash column chromatography was performed using
Merck Kieselgel 60 silica grade 9385.
Dim et h yl Ben zim id a zole-5,6-d ica r b oxyla t e (11). Di-
methyl 4,5-diaminophthalate (224 mg, 1.0 mmol), prepared in
several standard steps from 4-nitrophthalic acid, was dissolved
in 97% formic acid (1 mL) and stirred at a gentle reflux for 2
h. On cooling, the solution was carefully poured onto a
saturated aqueous solution of sodium hydrogencarbonate (20

Nonpeptide CCK2 Receptor Agonists
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 8 1131
mL) and then extracted with dichloromethane (2 × 10 mL).
The combined organic layers were dried (magnesium sulfate),
filtered and evaporated resulting in a solid that was recrystal-
lized from ethanol to leave the pale pink colored title compound
(200 mg, 85%): 1H NMR (CDCl3) δ 8.2 (2H, s), 8.0 (2H, br s),
3.9 (6H, s).
Dim et h yl 1-Met h ylb en zim id a zole-5,6-d ica r b oxyla t e
(12). Sodium (25 mg, 1.1 mmol) was dissolved in dry methanol
(3 mL) under an atmosphere of dry nitrogen. Compound 11
(234 mg 1.0 mmol) was added and the solution stirred for 30
min. Methyl iodide (100 µL, 1.6 mmol) was added and the
resulting solution stirred at room temperature overnight. The
solution was then warmed briefly to reflux, cooled, evaporated
and purified by column chromatography (silica, 95% dichlo-
romethane-5% methanol) to leave the title compound (170 mg,
69%) as a white solid: 1H NMR (CDCl3) δ 8.2 (1H, s), 8.0 (1H,
br s), 7.8 (1H, s), 3.94 (6H, s), 3.91 (3H, s).
1-Meth ylben zim id a zole-5,6-d ica r boxylic Acid An h y-
d r id e (14). Step a . To compound 12 (2.27 g, 9.2 mmol)
dissolved in 9:1 methanol:water (30 mL) was added potassium
hydroxide (1.12 g, 20.0 mmol). The solution was stirred
overnight at room temperature. The solution was evaporated
and the residue taken up in a minimum volume of water and
acidified with hydrochloric acid. The precipitated white solid
was filtered, washed with water and dried in vacuo, to leave
compound 13 (1.50 g) which was used without further puri-
fication.
Step b. The crude diacid was suspended in acetic anhydride
(7 mL) and stirred and heated at a reflux for 2 h. The solution
was cooled and diluted with diethyl ether, whereupon a yellow
solid was precipitated, filtered, washed with diethyl ether and
dried in vacuo. This left the title compound (0.95 g, 51%): 1H
NMR (DMSO-d6) δ 8.7 (1H, s), 8.4 (1H, s), 8.3 (1H, s), 3.94
(6H, s), 4.0 (3H, s).
Dim et h yl 1-Met h ylin d ole-5,6-d ica r b oxyla t e (16). Di-
methyl indole-5,6-dicarboxylate10 (15; 1.95 g, 8.3 mmol) was
dissolved in dry dimethylformamide (10 mL); cesium carbonate
(2.72 g, 8.3 mmol) and methyl iodide (2.6 mL, 42.0 mmol) were
added. The reaction mixture was heated to a reflux for 15 min
and then cooled and evaporated. The residue was taken up in
ethyl acetate (50 mL) and washed successively with 2 M
hydrochloric acid (25 mL), water (25 mL), saturated aqueous
sodium hydrogencarbonate (25 mL) and brine (2 × 25 mL).
The organic layer was dried (magnesium sulfate) filtered and
evaporated resulting in a pale yellow gum. This was resus-
pended in dichloromethane and re-evaporated to leave the title
compound as a beige solid (1.94 g, 94%): 1H NMR (CDCl3) δ
8.0 (1H, s), 7.7 (1H, s), 7.2 (1H, d), 6.6 (1H, d), 3.93 (3H, s),
3.91 (3H, s), 3.8 (3H, s).
1-Meth ylin d ole-5,6-d ica r boxylic Acid (17). 16 (1.93 g,
7.8 mmol) was dissolved in 9:2 ethanol:water (20 mL) and
sodium hydroxide (630 mg, 15.7 mmol) was added. The
solution was stirred and heated to a reflux for 15 min. After
cooling and evaporation to a volume of about 2 mL, 2 M
hydrochloric acid was used to precipitate the title compound
which was isolated by filtration, washed with water and dried
in vacuo (1.69 g, 98%): 1H NMR (DMSO-d6) δ 12.6 (2H, Br s),
7.9 (1H, s), 7.7 (1H, s), 7.5 (1H, d), 6.6 (1H, d), 3.8 (3H, s).
1-Meth ylin d ole-5,6-d ica r boxylic Acid An h yd r id e (18).
17 (1.69 g, 7.7 mmol) was suspended in acetic anhydride (14
mL) and stirred and heated at a reflux for 1 h. The solution
was cooled and the yellow crystals were filtered, washed with
diethyl ether and dried in vacuo. This left the title compound
(1.42 g, 92%): 1H NMR (DMSO-d6) δ 8.3 (2H, br s), 7.8 (1H,
br s), 6.8 (1H, br s), 4.0 (3H, br s).
and then extracted with hot acetone (5 × 20 mL). The
combined acetone extracts were evaporated to leave the diacid
as a yellow solid (2.07 g, 77%): 1H NMR (DMSO-d6) δ 11.6
(1H, s), 7.9 (1H, s), 7.7 (1H, s), 7.6 (1H, d), 6.6 (1H, d).
1-Acetylin d ole-5,6-d ica r boxylic Acid An h yd r id e (20).
19 (1.01 g, 4.9 mmol) was dissolved in acetic anhydride (10
mL) and heated at reflux under an atmosphere of argon for
1h. On cooling a yellow, crystalline solid formed which was
collected by filtration, washed in diethyl ether and dried in
vacuo at 50 °C, to leave the title compound (504 mg, 44%): 1H
NMR (DMSO-d6) δ 8.8 (1H, s), 8.33 (1H, s), 8.29 (1H, d), 7.0
(1H, d), 2.7 (3H, s).
1-Meth yl-5(6)-[[(1-a d a m a n tylm eth yl)a m in o]ca r bon yl]-
b en zim id a zole-6(5)-ca r b oxylic Acid (21). Compound 14
(950 mg, 4.7 mmol) was added to a stirred solution of
1-adamantylmethylamine (860 mg, 5.2 mmol) in dry THF (50
mL). The solution was stirred at room temperature for 1.5 h,
whereupon the pale, yellow precipitate was isolated by filtra-
tion, washed with THF and dried in vacuo at 50 °C, to leave
the title compound (1.65 g, 96%): 1H NMR (DMSO-d6) δ 9.5
(1H, bs), 8.3 (1H, s), 7.84 (1H, s), 7.75 (1H, 2 × s), 3.9 (3H, 2
× s), 2.9 (2H, m), 1.9 (3H, s), 1.73 (6H, q), 1.69 (6H, s). The
spectrum being of regioisomers was more complex than would
be expected for a single compound. Integration suggested a
ca. 1:1 mixture.
1-Meth yl-5(6)-[[(1-a d a m a n tylm eth yl)a m in o]ca r bon yl]-
in d ole-6(5)-ca r boxylic Acid (22). Compound 18 (1.42 g, 7.04
mmol) was added to a stirred solution of 1-adamantylmethy-
lamine (1.16 g, 7.0 mmol) in dry THF (40 mL) and triethy-
lamine (1.08 mL, 7.7 mmol). The solution was heated to reflux
for 5 min and then stirred at room temperature for 2 h,
whereupon 2 M hydrochloric acid was added and the precipi-
tate that formed was isolated by filtration, washed with water
and dried in vacuo, to leave the title compound (2.57 g, 99%):
1H NMR (DMSO-d6) δ 8.1-7. 4 (3H, m), 6.5 (1H, m),3.8 (3H,
2 × s), 2.9 (2H, m), 1.9 (3H, s), 1.6 (6H, q), 1.5 (6H, s). The
spectrum being of regioisomers was more complex than would
be expected for a single compound. Integration suggested a
ca. 2:3 mixture, but the predominant regioisomer could not
be identified.
1-Acetyl-5(6)-[[(1-a d a m a n tylm eth yl)a m in o]ca r bon yl]-
in d ole-6(5)-ca r boxylic Acid (23). 20 (578 mg, 2.5 mmol) was
added to a stirred solution of 1-adamantylmethylamine (458
mg, 2.8 mmol) and DMAP (10 mg, catalytic) in dry THF (9
mL) and triethylamine (386 µL, 2.8 mmol). The solution was
stirred at room temperature overnight, whereupon 2 M
hydrochloric acid (20 mL) was added and the mainly aqueous
solution was extracted with dichloromethane (3 × 20 mL). The
organic layer was washed with brine, dried (magnesium
sulfate) filtered and evaporated resulting in the title compound
(1.03 g, 99%): 1H NMR (DMSO-d6) δ 8.7 and 8.4 (1H, 2 × s),
8.2 (1H, t), 8.0 (2H, m), 6.8 (1H, s), 2.9 (2H, m), 2.6 (3H, s), 1.9
(3H, s), 1.6 (6H, q), 1.5 (6H, s). The spectrum being of
regioisomers was more complex than would be expected for a
single compound. Integration suggested a ca. 1:1 mixture.
1-Me t h yl-5(6)-[[[1(S)-[[[3,5-b is(b e n zyloxyca r b on yl)-
p h e n y l]a m in o ]c a r b o n y l]-2-p h e n y le t h y l]a m in o ]c a r -
b on yl]-6(5)-[[(1-a d a m a n t ylm e t h yl)a m in o]ca r b on yl]-
ben zim id a zole (25). 21 (367 mg, 1.0 mmol), 1(S)-[[[3,5-bis-
(benzyloxycarbonyl)phenyl]amino]carbonyl]-2-phenyleth-
ylamine (24;31 510 mg, 1.0 mmol), hydroxybenzotriazole (135
mg, 1.0 mmol), EDC (192 mg, 1.0 mmol) and DMAP (5 mg,
catalytic) were dissolved in dry DMF (3 mL) and stirred at
room temperature overnight. The solution was poured into
stirred water (30 mL) and the resulting buff precipitate was
isolated by filtration, washed with water and dried in vacuo.
Column chromatography (silica, 5% methanol-95% dichlo-
romethane) left the title compound (250 mg, 29%): 1H NMR
(DMSO-d6) δ 10.3 (1H, s), 8.8 (3H, m), 8.3 (3H, m), 7.9 (1H, 2
× s), 7.4-7.0 (16H, m), 5.4 (4H, s), 4.8 (1H, m), 3.9 (3H, 2 ×
s), 3.6-2.5 (4H, m), 1.8 (3H, s), 1.6-1.4 (12H, m). Integration
suggested a ca. 1:1 mixture.
In d ole-5,6-d ica r boxylic Acid (19). Dimethyl indole-5,6-
dicarboxylate10 (15; 3.05 g, 13.1 mmol) was dissolved in a
mixture of ethanol (28 mL) and water (6 mL) and sodium
hydroxide (1.31 g, 32.8 mmol) was introduced. The solution
was stirred overnight at room temperature and concentrated
hydrochloric acid was added so that a pH of 2 was achieved.
The solvent was evaporated to give a white solid which was
re-suspended in ethanol and evaporated followed by treatment
with toluene and evaporation. The residue was dried in vacuo
1-Me t h yl-5(6)-[[[1(S)-[[[3,5-b is(b e n zyloxyca r b on yl)-
ph en yl]am in o]car bon yl]-2-ph en yleth yl]am in o]car bon yl]-

1132 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 8
Kalindjian et al.
6(5)-[[(1-a d a m a n tylm eth yl)a m in o]ca r bon yl]in d ole (26).
22 (500 mg, 1.4 mmol) was dissolved in dry dichloromethane
(40 mL) and PyBOP (710 mg, 1.4 mmol) and diisopropyleth-
ylamine (720 µL, 4.1 mmol) were added. After stirring at room
temperature for 1 h, 2431 (694 mg, 1.4 mmol) was added and
the mixture stirred overnight. The organic layer was washed
successively with 5% potassium hydrogen sulfate (5 mL),
saturated sodium hydrogen carbonate (5 mL) and saturated
brine (5 mL). The organic layer was then dried, filtered and
evaporated before being purified by column chromatography
(silica, 20% ethyl acetate-80% dichloromethane) to leave the
title compound (585 mg, 50%): 1H NMR (DMSO-d6) δ 10.3 (1H,
s), 8.8 (3H, m), 8.4 (1H, m), 8.3 (1H, s), 7.9 and 7.8 (1H, 2 ×
s), 7.4 (16H, m), 7.1 and 6.8 (1H, 2 × s), 6.6 and 6.5 (1H, 2 ×
d), 5.4 (4H, s), 4.8 (1H, m), 3.8 (3H, m), 3.4 and 2.9 (4H, m),
1.8 (3H, s), 1.5 (6H, m), 1.4 (6H, s). Integration of key peaks
suggested a ca. 1:1 mixture.
1-Ace t yl-5(6)-[[[1(S )-[[[3,5-b is(b e n zyloxyca r b on yl)-
ph en yl]am in o]car bon yl]-2-ph en yleth yl]am in o]car bon yl]-
6(5)-[[(1-a d a m a n tylm eth yl)a m in o]ca r bon yl]in d ole (27).
23 (1.18 g, 2.5 mmol), 2431 (958 mg, 1.9 mmol), 1-hydroxyben-
zotriazole (340 mg, 2.5 mmol) and DCCI (519 mg, 2.5 mmol)
were dissolved in dry DMF (13 mL) and stirred at room
temperature overnight. A precipitate was formed which was
removed by filtration, the residue being washed with a small
quantity of dichloromethane. Further dichloromethane (50 mL)
was added to the filtrate and the solution washed successively
with 2 M hydrochloric acid (10 mL), water (10 mL) and
saturated sodium hydrogencarbonate (10 mL). After drying
and filtering, the product was purified by column chromatog-
raphy (silica, 20% ethyl acetate-80% dichloromethane) to
leave the title compound (1.12 g, 67%): 1H NMR (DMSO-d6)
δ 10.3 (1H, s), 9.9 and 9.0 (1H, 2 × d), 8.8 (2H, m), 8.6 (1H, s),
8.5 (1H, m), 8.3 (1H, d), 8.2 (1H, s), 8.0 and 7.9(1H, 2 × m),
7.4 (15H, m), 6.9 and 6.8 (1H, 2 × d), 5.4 (4H, s), 4.7 (1H, m),
3.4 and 2.9 (4H, m), 2.7 (3H, 2 × s), 1.8 (3H, s), 1.6 (6H, m),
1.3 (6H, s). Integration suggested a ca. 1:1 mixture.
Gen er a l Meth od for Hyd r ogen a tion . 1-Meth yl-5(6)-
[[[1(S )-[[(3,5-d ic a r b o x y p h e n y l)a m in o ]c a r b o n y l]-2-
p h en yleth yl]a m in o]ca r bon yl]-6(5)-[[(1-a d a m a n tylm eth -
yl)a m in o]ca r bon yl]in d ole (3). 26 (560 mg, 0.7 mmol) was
dissolved in a 1:1 mixture of THF and methanol (30 mL) and
a catalytic quantity of 10% palladium on charcoal was added.
The suspension was stirred at room temperature under an
atmosphere of hydrogen overnight, filtered through Celite and
evaporated to leave the title compound (442 mg, 99%): 1H
NMR (DMSO-d6) δ 13.2 (2H, br s), 10.2 (1H, 2 × s), 8.7 (3H,
m), 8.4 (1H, m), 8.2 (1H, s), 7.9 and 7.8 (1H, 2 × s), 7.5 (1H,
m), 7.4 (5H, m), 7.1 and 6.8 (1H, 2 × s), 6.6 and 6.5 (1H, 2 ×
d), 4.7 (1H, m), 3.8 and 3.7 (3H, 2 × s), 3.4 and 2.9 (4H, m),
1.8 (3H, s), 1.5 (6H, m), 1.4 (6H, s). Integration suggested a
ca. 1:1 mixture; further characterized as the bis(N-methyl-D-
glucamine) salt. Anal. (C39H40N4O7‚2C7H17NO5‚2.3H2O) C, H,
N.
1-Acetyl-5(6)-[[[1(S)-[[(3,5-d ica r boxyp h en yl)a m in o]ca r -
bon yl]-2-p h en yleth yl]a m in o]ca r bon yl]-6(5)-[[(1-a d a m a n -
tylm eth yl)a m in o]ca r bon yl]in d ole (8). This was prepared
from 27 using the general hydrogenation methodology de-
scribed above for 3: 1H NMR (DMSO-d6) δ 13.3 (2H, br s),
10.2 (1H, 2 × s), 9.0 and 8.9 (1H, 2 × d), 8.6 (3H, m), 8.14 (1H,
s), 8.10 and 7.9 (1H, 2 × s), 8.0 (1H, s), 7.4 (5H, m), 6.9 and
6.8 (1H, 2 × d), 4.7 (1H, m), 3.4 and 2.9 (4H, m), 2.7 (3H, 2 ×
s), 1.8 (3H, s), 1.6 (6H, m), 1.3 (6H, s). Integration suggested
a ca. 1:1 mixture; further characterized as the bis(N-methyl-
D-glucamine) salt. Anal. (C40H40N4O8‚2C7H17NO5‚2.6H2O) C,
H, N.
Integration suggested a ca. 1:1 mixture; further characterized
as the bis(N-methyl-D-glucamine) salt. Anal. (C38H39N5O7‚
2C7H17NO5‚3.0H2O) C, H, N.
1-Meth yl-5-[[[1(S)-[[[3,5-bis(ben zyloxyca r bon yl)p h en -
yl]a m in o]ca r b on yl]-2-p h en ylet h yl]a m in o]ca r b on yl]-6-
[[(1-a d a m a n tylm eth yl)a m in o]ca r bon yl]in d ole (30). Com-
pound 2810 (211 mg, 0.3 mmol) was dissolved in a mixture of
THF (1 mL) and DMF (0.5 mL). The solution was stirred under
an atmosphere of dry nitrogen and sodium hydride (15 mg,
0.3 mmol) was added. Hydrogen gas was evolved for about 5
min whereupon methyl iodide (40 µL, 0.6 mmol)) was added.
The solution was stirred at room temperature for 1 h, diluted
with brine (20 mL) and extracted with dichloromethane (20
mL). The organic layer was washed with brine (2 × 20 mL),
dried (magnesium sulfate), filtered, evaporated and purified
by column chromatography (silica, 85% dichloromethane-15%
ethyl acetate) to leave the title compound (90 mg, 42%): 1H
NMR (DMSO-d6) δ 10.3 (1H, s), 8.8 (3H, m), 8.4 (1H, m), 8.3
(1H, s), 7.8 (1H, s), 7.4 (16H, m), 7.1 (1H, s), 6.5 (1H, d), 5.4
(4H, s), 4.7 (1H, m), 3.8 (3H, s), 3.4 and 2.9 (4H, m), 1.8 (3H,
s), 1.5 (6H, m), 1.4 (6H, s).
1-Meth yl-5-[[(1-a d a m a n tylm eth yl)a m in o]ca r bon yl]-6-
[[[1(S )-[[[3,5-b is(b e n zyloxyca r b on yl)p h e n yl]a m in o]-
car bon yl]-2-ph en yleth yl]am in o]car bon yl]in dole (32). Com-
pound 2910 was converted to the title compound using the same
methodology as for the preparation of 30: 1H NMR (DMSO-
d6) δ 10.3 (1H, s), 8.8 (3H, m), 8.4 (1H, m), 8.3 (1H, s), 7.9
(1H, s), 7.4 (16H, m), 6.8 (1H, s), 6.6 (1H, d), 5.4 (4H, s), 4.7
(1H, m), 3.8 (3H, s), 3.4 and 2.9 (4H, m), 1.8 (3H, s), 1.5 (6H,
m), 1.4 (6H, s).
1-Bu tyl-5-[[[1(S)-[[[3,5-bis(ben zyloxyca r bon yl)p h en yl]-
a m in o]ca r bon yl]-2-p h en yleth yl]a m in o]ca r bon yl]-6-[[(1-
a d a m a n tylm eth yl)a m in o]ca r bon yl]in d ole (31). 2810 was
converted to the title compound using the same methodology
as for the preparation of 30 except that an excess of butyl
iodide was used as alkylating agent instead of methyl iodide:
1H NMR (CDCl3) δ 9.7 (1H, s), 8.8 (2H, s), 8.5 (1H, s), 7.4 (17H,
m), 7.2 (1H, s), 6.5 (1H, d), 6.4 (1H, t), 6.3 (1H, t), 5.4 (4H, s),
5.1 (1H, m), 4.1 (2H, m), 3.3 (2H, m), 3.1 (2H, m), 1.9 (3H, s),
1.8 (2H, m), 1.5 (6H, m), 1.4 (6H, s), 1.3 (2H, m), 0.9 (3H, m).
1-Bu t yl-5-[[(1-a d a m a n t ylm et h yl)a m in o]ca r b on yl]-6-
[[[1(S )-[[[3,5-b is(b e n zyloxyca r b on yl)p h e n yl]a m in o]-
car bon yl]-2-ph en yleth yl]am in o]car bon yl]in dole (33). Com-
pound 2910 was converted to the title compound using the same
methodology as for the preparation of 30 except that an excess
of butyl iodide was used as alkylating agent instead of methyl
iodide: 1H NMR (CDCl3) δ 10.0 (1H, s), 8.9 (2H, s), 8.5 (1H,
s), 7.7 (1H, s), 7.4 (16H, m), 6.9 (1H, s), 6.5 (1H, d), 6.2 (2H,
m), 5.4 (4H, s), 5.1 (1H, m), 4.1 (2H, m), 3.6 (1H, m), 3.3 (1H,
m), 3.1 (2H, d), 1.9 (3H, s), 1.8 (2H, m), 1.5 (6H, m), 1.4 (6H,
s), 1.3 (2H, m), 0.9 (3H, m).
1-Meth yl-5-[[[1(S)-[[(3,5-d ica r boxyp h en yl)a m in o]ca r -
bon yl]-2-p h en yleth yl]a m in o]ca r bon yl]-6-[[(1-a d a m a n tyl-
m eth yl)a m in o]ca r bon yl]in d ole (4). Compound 30 was con-
verted to the title compound using the same methodology as
for the preparation of 3: 1H NMR (DMSO-d6) δ 13.2 (2H, br
s), 10.2 (1H, s), 8.7 (3H, m), 8.4 (1H, m), 8.2 (1H, s), 7.8 (1H,
s), 7.5 (1H, m), 7.4 (5H, m), 7.1 (1H, s), 6.5 (1H, d), 4.7 (1H,
m), 3.7 (3H, s), 3.4 and 2.9 (4H, m), 1.8 (3H, s), 1.5 (6H, m),
1.4 (6H, s); further characterized as the bis(N-methyl-D-
glucamine) salt. Anal. (C39H40N4O7‚2C7H17NO5) C, H, N.
1-Meth yl-5-[[(1-a d a m a n tylm eth yl)a m in o]ca r bon yl]-6-
[[[1(S )-[[(3,5-d ic a r b o x y p h e n y l)a m in o ]c a r b o n y l]-2-
p h en yleth yl]a m in o]ca r bon yl]in d ole (5). Compound 32 was
converted to the title compound using the same methodology
as for the preparation of 3: 1H NMR (DMSO-d6) δ 10.2 (1H,
s), 8.7 (3H, m), 8.4 (1H, t), 8.2 (1H, s), 7.9 (1H, s), 7.5 (1H, d),
7.4 (5H, m), 6.8 (1H, s), 6.6 (1H, d), 4.7 (1H, m), 3.8 (3H, s),
3.4 and 2.9 (4H, m), 1.8 (3H, s), 1.5 (6H, m), 1.4 (6H, s); further
characterized as the bis(N-methyl-D-glucamine) salt. Anal.
(C39H40N4O7‚2C7H17NO5‚3H2O) C, H, N.
1-Met h yl-5(6)-[[[1(S)-[[(3,5-d ica r b oxyp h en yl)a m in o]-
ca r bon yl]-2-p h en yleth yl]a m in o]ca r bon yl]-6(5)-[[(1-a d a -
m a n tylm eth yl)a m in o]ca r bon yl]in d ole (9). This was pre-
pared from 25 using the general hydrogenation methodology
described above for 3: 1H NMR (DMSO-d6) δ 13.0 (2H, br s),
10.2 (1H, s), 8.8 (1H, m), 8.7 (2H, s), 8.6 (2H, m), 8.2 (1H, s),
8.0 and 7.9 (1H, 2 × s), 7.4-7.0 (6H, m), 4.8 (1H, m), 3.9 (3H,
2 × s), 3.6-2.5 (4H, m), 1.8 (3H, s), 1.6-1.4 (12H, m).
1-Bu t yl-5-[[[1(S)-[[(3,5-d ica r b oxyp h en yl)a m in o]ca r -
bon yl]-2-p h en yleth yl]a m in o]ca r bon yl]-6-[[(1-a d a m a n tyl-
m eth yl)a m in o]ca r bon yl]in d ole (6). Compound 31 was con-

Nonpeptide CCK2 Receptor Agonists
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 8 1133
(12) Shiosaki, K.; Lin, C. W.; Kopeka, H.; Craig, R.; Wagenaar, F.;
Bianchi, B. R.; Miller, T. R.; Witte, D. G.; Nadzan, A. M.
Development of CCK Tetrapeptide Analogues as Potent and
Selective CCK-A Receptor Agonists. J . Med. Chem. 1990, 33,
2950-2952.
(13) Shiosaki, K.; Lin, C. W.; Kopeka, H.; Tufano, M. D.; Bianchi, B.
R.; Miller, T. R.; Witte, D. G.; Nadzan, A. M. Boc-CCK-4
Derivatives Containing Side-Chain Ureas as Potent and Selec-
tive CCK-A Receptor Agonists. J . Med. Chem. 1991, 34, 2837-
2842.
verted to the title compound using the same methodology as
for the preparation of 3: 1H NMR (DMSO-d6) δ 13.2 (2H, br
s), 10.2 (1H, s), 8.7 (3H, m), 8.4 (1H, t), 8.2 (1H, s), 7.8 (1H, s),
7.6 (1H, d), 7.4 (5H, m), 7.2 (1H, s), 6.5 (1H, d), 4.8 (1H, m),
4.2 (2H, m), 3.4 and 2.9 (4H, m), 1.8 (5H, m), 1.5 (6H, m), 1.4
(6H, s), 1.2 (2H, m), 0.9 (3H, m); further characterized as the
bis(N-methyl-D-glucamine) salt. Anal. (C42H46N4O7‚2C7H17NO5‚
H2O) C, H, N.
1-Bu t yl-5-[[(1-a d a m a n t ylm et h yl)a m in o]ca r b on yl]-6-
[[[1(S )-[[(3,5-d ic a r b o x y p h e n y l)a m in o ]c a r b o n y l]-2-
p h en yleth yl]a m in o]ca r bon yl]in d ole (7). Compound 33 was
converted to the title compound using the same methodology
as for the preparation of 3: 1H NMR (DMSO-d6) δ 13.3 (2H,
br s), 10.2 (1H, s), 8.8 (3H, m), 8.4 (1H, t), 8.2 (1H, s), 7.9 (1H,
s), 7.5 (1H, d), 7.4 (5H, m), 6.8 (1H, s), 6.6 (1H, d), 4.8 (1H, m),
4.2 (2H, m), 3.4 and 2.9 (4H, m), 1.8 (5H, m), 1.5 (6H, m), 1.4
(6H, s), 1.2 (2H, m), 0.9 (3H, m); further characterized as the
bis(N-methyl-D-glucamine) salt. Anal. (C42H46N4O7‚2C7H17NO5‚
2H2O) C, H, N.
(14) Holladay, M. W.; Kopeka, H.; Miller, T. R.; Bednarz, L.; Nikkel,
A. L.; Bianchi, B. R.; Witte, D. G.; Shiosaki, K.; Lin, C. W.; Asin,
K. E.; Nadzan, A. M. Tetrapeptide CCK-A Agonists: Effect of
Backbone N-Methylations on in Vitro and in Vivo CCK Activity.
J . Med. Chem. 1994, 37, 630-635.
(15) Bernad, N.; Burgaud, B. G. M.; Horwell, D. C.; Lewthwaite, R.
A.; Martinez, J .; Pritchard, M. C. The Design and Synthesis of
the High Efficacy, Non-peptide CCK1 Receptor Agonist PD
170292. Bioorg. Med. Chem. Lett. 2000, 10, 1245-1248.
(16) Semple, G.; Ryder, H.; Kendrick, D. A.; Batt, A. R.; Mathews,
E.; Rooker, D. P.; Szelke, M.; Nishida, A.; Miyata, K. Identifica-
tion and Biological Activity of Novel Peptidomimetic Gastrin/
CCK-B Receptor Agonists. Bioorg. Med. Chem. Lett. 1996, 6,
2971-2976 and references contained therein.
(17) Gregory, R. A.; Tracy, H. J . The Constitution and Properties of
Two Gastrins Extracted from Hog Antral Mucosa. Gut 1964, 5,
103-117.
(18) Welsh, N. J .; Shankley, N. P.; Black, J . W. Comparative Analysis
of the Vagal Stimulation of Gastric Acid Secretion in Rodent
Isolated Stomach Preparations. Br. J . Pharmacol. 1994, 112,
93-96.
(19) Harper, E. A.; Roberts, S. P.; Shankley, N. P.; Black, J . W.
Analysis of Variation in L-365,260 Competition Curves in
Radioligand Binding Assays. Br. J . Pharmacol. 1996, 118, 1717-
1726.
(20) Roberts, S. P.; Harper, E. A.; Watt, G. F.; Gerskowitch, V. P.;
Hull, R. A. D.; Shankley, N. P.; Black, J . W. Analysis of the
Variation in the Action of L-365,260 at CCKB/Gastrin Receptors
in Rat, Guinea-pig and Mouse Isolated Gastric Tissue Assays.
Br. J . Pharmacol. 1996, 118, 1779-1789.
(21) Black, J . W.; Leff, P. Operational Models of Pharmacological
Agonism. Proc. R. Soc. London B 1983, 220, 141-162.
(22) PD-134,308 was found to express partial agonism at CCK-2
receptors in the isolated mouse stomach assay: N. P. Shankley,
personal communication; data to be published separately.
(23) Bock, M. G.; DiPardo, R. M.; Evans, B. E.; Rittle, K. E.; Whitter,
W. L.; Veber, D. F.; Anderson, P. S.; Freidinger, R. M. Benzo-
diazepine Gastrin and Brain Cholecystokinin Receptor Lig-
ands: L-365260. J . Med. Chem. 1989, 32, 13-16.
(24) Patel, M.; Spraggs, C. F. Functional Comparisons of Gastrin/
Cholecystokinin Receptors in Isolated Preparations of Gastric
Mucosa and Ileum. Br. J . Pharmacol. 1992, 106, 275-282.
(25) Vinter, J . G.; Trollope, K. I. Multi-conformational Composite
Molecular Potential Fields in the Analysis of Drug Action. (I)
Methodology and First Evaluation using 5-HT and Histamine
Action as examples. J . Comput.-Aided Drug Des. 1995, 9, 297-
307.
Ack n ow led gm en t. We thank Dr. E. A. Harper for
providing the radioligand binding data presented in
Table 1.
Refer en ces
(1) Small Peptides Chemistry, Biology and Clinical Studies; Dutta,
A. S., Eds.; Elsevier: London, 1993.
(2) Hughes, J .; Boden, P.; Costall, B.; Domeny, A.; Kelly, E.; Horwell,
D.; Hunter, J .; Pinnock, R. D.; Woodruff, G. Development of a
Class of Selective Cholecystokinin Type B Receptor Antagonists
Having Potent Anxiolytic Activity. Proc. Natl. Acad. Sci. U.S.A.
1990, 87, 6728-6732.
(3) Schmassmann, A.; Garner, A.; Flogerzi, B.; Hasan, M. Y.;
Sanner, M.; Varga, L.; Halter, F. Cholecystokinin type
B
Receptor Antagonist PD-136,450 is a Partial Secretory Agonist
in the Stomach and a Full Agonist in the Pancreas of the Rat.
Gut 1994, 35, 270-274.
(4) Kivlighn, S. D.; Huckle, W. R.; Zingaro G. J .; Rivero, R. A.; Lotti,
V. J .; Chang, R. S. L.; Schorn, T. W.; Kevin, N.; J ohnson J r, R.
G.; Greenlee, W. J .; Siegl, P. K. S. Discovery of L-162,313:
A
Nonpeptide that Mimics the Biological Actions of Angiotensin
II. Am. J . Physiol. 1995, 268, R820-3.
(5) Siegl, P. K. S.; Chang, R. S. L.; Mantlo, M. B.; Chakravarty, P.
K.; Ondeyka, D. L.; Greenlee. W. J .; Patchett, A. A.; Sweet, C.
S.; Lotti, V. J . In vivo Pharmacology of L-158,809, a New Highly
Potent and Selective nonpeptide Angiotensin II Receptor An-
tagonist. J . Pharmacol. Exp. Ther. 1992, 262, 139-144.
(6) Aquino, C. J .; Armour, D. R.; Berman, J . M.; Birkemo, L. S.;
Carr, R. A. E.; Croom, D. K.; Dezube, M.; Dougherty J r, R. W.;
Ervin, G. N.; Grizzle, M. K.; Head, J . E. Hirst, G. C.; J ames, M.
K.; J ohnson, M. F.; Miller, L. J .; Queen, K. L.; Rimele, T. J .;
Smith, D. N.; Sugg, E. E. Discovery of 1,5-Benzodiazepines with
Peripheral Cholecystokinin (CCK-A) Receptor Agonist Activity.
1. Optimization of the Agonist “Trigger”. J . Med. Chem. 1996,
39, 562-569.
(26) Vinter, J . G. Extended Electron Distributions Applied to the
Molecular Mechanics of Intermolecular Interactions. J . Comput.-
Aided Drug Des. 1994, 8, 653-668.
(27) Rosenkilde, M. M.; Cahir, M.; Gether, U.; Hjorth, S. A.; Schwartz,
T. W. Mutations along Transmembrane Segment II of the NK-1
Receptor Affect Substance P Competition with Non-peptide
Antagonists but not Substance P Binding. J . Biol. Chem. 1994,
269, 28160-28164.
(28) Gether, U.; J ohansen, T. E.; Snider, R. M.; Lowe III, J . A.;
Nakanishi, S.; Schwartz, T. W. Different Binding Epitopes on
the NK1 Receptor for Substance P and a Non-peptide Antagonist.
Nature 1993, 362, 345-348.
(29) Romano, R.; Bayerl. T. M.; Moroder, L. Lipophilic Derivatisation
and its Effect on the Interaction of Cholecystokinin (CCK)
Nonapeptide with Phospholipids. Biochem. Biophys. Acta 1993,
1151, 111-119.
(7) Henke, B. R.; Willson, T. M.; Sugg, E. E.; Croom, D. K.;
Dougherty J r, R. W.; Queen, K. L.; Birkemo, L. S.; Ervin, G. N.;
Grizzle, M. K.; J ohnson, M. F.; J ames, M. K. 3-(1H-Indazol-3-
ylmethyl)-1,5-benzodiazepines: CCK-A Agonists That Demon-
strate Oral Activity as Satiety Agents. J . Med. Chem. 1996, 39,
2655-2658.
(8) Henke, B. R.; Aquino, C. J .; Birkemo, L. S.; Croom, D. K.;
Dougherty J r, R. W.; Ervin, G. N.; Grizzle, M. K.; Hirst, G. C.;
J ames, M. K.; J ohnson, M. F.; Queen, K. L.; Sherrill, R. G.; Sugg,
E. E.; Suh, E. M. Szewczyk, J . W.; Unwalla, R. J .; Yingling, J .;
Willson, T. M. Optimization of 3-(1H-Indazol-3-ylmethyl)-1,5-
benzodiazepines as Potent, Orally Active CCK-A Agonists. J .
Med. Chem. 1997, 40, 2706-2725.
(30) Blacker, M.; Ren, Y.; Seshadri, L.; McBride, E. W.; Bienborn,
M.; Kopin A. S. CCK-B/Gastrin Receptor Transmembrane
Domain Mutations Selectively Alter Synthetic Agonist Efficacy
without Affecting the Activity of Endogenous Peptides. Mol.
Pharmacol. 2000, 58, 399-406.
(31) Kalindjian, S. B.; Buck, I. M.; Cushnir, J . R.; Dunstone, D. J .,
Hudson, M. L.; Low, C. M. R.; McDonald, I. M.; Pether, M. J .;
Steel, K. I. M.; Tozer, M. J . Improving the Affinity and Selectivity
of a Nonpeptide Series of Cholecystokinin-B/Gastrin Receptor
Antagonists Based Receptor Antagonists Based on the Dibenzo-
bicyclo[2.2.2]octane Skeleton. J . Med. Chem. 1995, 38, 4294-4302.
(9) Sugg, E. E. Nonpeptide agonists for peptide receptors: lessons
from ligands. Annu. Rep. Med. Chem. 1997, 32, 277-283.
(10) Kalindjian, S. B.; Buck, I. M.; Davies, J . M. R.; Dunstone, D. J .;
Hudson, M. L.; Low, C. M. R.; McDonald, I. M.; Pether, M. J .;
Steel, K. I. M.; Tozer, M. J .; Vinter, J . G. Non-Peptide Chole-
cystokinin-B/Gastrin Receptor Antagonists Based on Bicyclic
Heteroaromatic Skeletons. J . Med. Chem. 1996, 39, 1806-1815.
(11) Lin, C. W.; Shiosaki, K.; Miller, T. R.; Witte, D. G.; Bianchi, B.
R.; Wolfram, C. A. W.; Kopeka, H.; Craig, R.; Wagenaar, F.;
Nadzan, A. M. Characterisation of Two Novel Cholecystokinin
Tetrapeptide(30-33) Analogues A-71623 and A-70874, that
Exhibit High Potency and Selectivity for Cholecystokinin-A
Receptors. Mol. Pharmacol. 1990, 39, 346-351.
J M0010668