A Non-Peptide NK1 Receptor Agonist Showing Subpicomolar Affinity

Article abstract of DOI:10.1021/jm034219a

3-Quinolinecarboxamides have been synthesized and evaluated for their binding to the human NK₁ receptor. Several secondary amide derivatives show NK₁ receptor affinity in the picomolar range. The most active compound, hydroxymethylca

Full text of DOI:10.1021/jm034219a

J . Med. Chem. 2004, 47, 1315-1318
Letters
1315
Ch a r t 1
A Non -P ep tid e NK1 Recep tor Agon ist
Sh ow in g Su bp icom ola r Affin ity
Andrea Cappelli,*,‡ Germano Giuliani,‡
Gal.la Pericot Mohr,‡ Andrea Gallelli,‡
Maurizio Anzini,‡ Salvatore Vomero,‡
Aroldo Cupello,† Simona Scarrone,†
Mario Matarrese,§ Rosa Maria Moresco,§
Ferruccio Fazio,§ Federica Finetti,#
Lucia Morbidelli,# and Marina Ziche#
research effort in this area within the framework of the
pharmaceutical industry, and a large number of potent
NK1 antagonists have been developed in the past
decade.3 In 1995, researchers from Takeda discovered
the potent antagonist properties of 4-phenylisoquinoli-
none and naphthyridinone derivatives 2,4 which are
apparently unrelated to the most known NK1 receptor
antagonists. Despite the huge amount of work per-
formed in this field and the large number of non-peptide
NK1 receptor antagonists described so far, the agonism
for this receptors still appears to be confined to peptide
compounds (since potent non-peptide agonists are, as
far as we know, still lacking).
On the other hand, similar research efforts on the
development of non-peptide angiotensin II (AT1) or
cholecystokinin-2 (CCK2) receptor antagonists (AT1 and
CCK2 receptors are GPCRs as NK1) showed that slight
structural variations of certain antagonists caused the
intrinsic efficacy to shift toward agonism.5 Our experi-
ence in the study of the interaction of arylpiperazine,
tropane, and quinuclidine ligands with the 5-HT3 recep-
tor (a ligand-gated ion channel) suggests that the
transformation of an antagonist into an agonist by
means of slight structural modifications is confined to
certain compounds only.6 Thus, we started a program
of structural manipulation of the original Takeda an-
tagonists 2 leading to the synthesis (see Supporting
Information) and the pharmacological characterization
of 3-quinolinecarboxamides 3 related to our quinolin-
ecarboxamide peripheral benzodiazepine receptor
ligands.7 These studies led to the discovery of the first
non-peptide NK1 receptor agonist endowed with sub-
picomolar affinity.
Dipartimento Farmaco Chimico Tecnologico and
European Research Centre for Drug Discovery and
Development, Universita` degli Studi di Siena, Via A. Moro,
53100 Siena, Italy, Istituto di Bioimmagini e Fisiologia
Molecolare, CNR, Sezione di Genova, Via de Toni 5, 16132
Genova, Italy, Istituto di Bioimmagini e Fisiologia
Molecolare, CNR, Istituto H. S. Raffaele, Universita` di
Milano/ Bicocca, Via Olgettina 60, 20132 Milano, Italy, and
Sezione di Farmacologia, Dipartimento di Biologia
Molecolare, Universita` degli Studi di Siena,
Via A. Moro, 53100 Siena, Italy.
Received October 13, 2003
Abst r a ct : 3-Quinolinecarboxamides have been synthesized
and evaluated for their binding to the human NK1 receptor.
Several secondary amide derivatives show NK1 receptor af-
finity in the picomolar range. The most active compound,
hydroxymethylcarboxamide 3h showing an IC50 value in the
subpicomolar range, behaved as an agonist of NK1 receptor in
endothelial cell proliferation, inositol phosphate turnover, and
NO-mediated cyclic GMP accumulation, thus proving it to be
the first non-peptide NK1 receptor agonist showing very high
potency.
In tr od u ction . Substance P (SP) is an undecapeptide
member of the tachykinin family. The comparative
analysis of the pharmacological properties of various
tachykinins has provided the evidence for the existence
of three receptor subtypes (termed NK1, NK2, and NK3)
showing different preferences in the interaction with
their endogenous ligands. All the tachykinin receptors
identified up to now belong to the family of the G-
protein-coupled receptors (GPCRs) and are linked to the
inositol phosphate signal transduction pathway.1 By
interacting with the NK1 receptor, SP elicits a variety
of biological responses including the transmission of
pain and stress signals, smooth muscle contraction, the
induction of neurogenic inflammation, endothelium-
dependent vasodilation, and angiogenesis. Since 1991,
the discovery2 of the first selective non-peptide NK1
antagonist, CP-96,345 (1, Chart 1), has catalyzed much
Resu lts a n d Discu ssion . Compounds 3a -r were
tested for their activity in inhibiting the specific binding
of [125I]BH-SP (100 pM) to native human NK1 receptor
expressed in astrocytoma UC11MG cells compared with
unlabeled SP and reference non-peptide antagonist
8
L-703,606 . The binding inhibition could be described
by a single site for most of the tested compounds, while
a second lower affinity component was necessary to fit
the data of compounds 3c,n more satisfactorily. The
binding of [125I] SP was confirmed to be due to NK1
receptors because it was completely displaced by the
non-peptide antagonist to the human NK1 receptors
L-703,606, with an IC50 of 1.6 nM. The results of the
binding studies (Table 1) reveal that some secondary
* To whom correspondence should be addressed. Phone: +39 0577
234320. Fax: +39 0577 234333. E-mail: cappelli@unisi.it.
‡ Dipartimento Farmaco Chimico Tecnologico and European Re-
search Centre for Drug Discovery and Development, Universita` degli
Studi di Siena.
† Istituto di Bioimmagini e Fisiologia Molecolare.
§ Universita` di Milano/Bicocca.
#
Dipartimento di Biologia Molecolare, Universita` degli Studi di
Siena.
10.1021/jm034219a CCC: $27.50 © 2004 American Chemical Society
Published on Web 02/11/2004

1316 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 6
Letters
Ta ble 1. Binding Affinities of 3a -r to Human Endogenous
NK1 Receptor in UC11MG Cells
binding
IC50, nMa
(SEM, %)
binding
%
b
compd
3a
3b
3c
3d
3e
3f
3g
3h
3i
3j
3k
3l
3m
3n
3o
R1
R2
H
CH3
H
CH3 CH3
H
H
H
H
CH3
>1000
0.030(22)
0.075(12)
3.1(6)
83c
78d
73c
57c
71c
100
100
75c
100
100
100
82c
40e
71c
CH2Cl
CH2Br
26(5)
0.010(17)
52(17)
0.00010(13)
16(8)
11(21)
0.0017(25)
9.0(13)
0.011(20)
0.22(7)
CH3 CH2Br
CH2OH
H
CH3 CH2OH
CH3 CH2OCH3
H
CH2NH2
CH3 CH2NH2
H
H
H
H
H
H
CH2N(H)CH3
CH2N(CH3)2
CH2N(CH3)CH2C6H5
180(36)
3p
3q
3r
SP
CH2N(CH2CH2)2NCH3 >1000
CH2-phthalimido
>1000
CH2OSi(CH3)2C(CH3)3 17(13)
100
100
100
0.12(8)
1.6(9)
L-703,606
a
F igu r e 1. Curves of the inhibition of [125I]SP (100 pM) binding
to NK1 receptors in UC11MG cells by various concentrations
of 3h (top) and 3k (bottom). The values on the ordinates are
relative bindings (referenced to the [125I]SP binding in the
absence of the compound, taken as 1.0), and those on the
abscissas are the logarithms of the compound concentrations
(log C). The single inhibition points are the averages of at least
three different experiments, each run in duplicate. The
continuous curves represent the best fits to the data.
Each value was calculated from at least three independent
experiments performed in duplicate and represents the concentra-
tion giving half the maximum inhibition of [125I]SP specific binding
b
to human endogenous NK1 receptor in UC 11 MG cells. Percent
of labeled SP total binding that is inhibited by the compound under
scrutiny with the reported IC50 value. c These compounds failed
to completely inhibit the specific [125I]SP binding even at the
d
maximal concentration tested. An IC50 value of 3900 nM was
calculated for the second binding component (22% of the [125I]SP
specific binding) of 3d . e An IC50 value of 11 nM was calculated
for the second binding component (60%) of 3n .
Ch a r t 2
amide derivatives (3c,f,h ,k ,m ) show NK1 receptor af-
finity in the picomolar range. In particular, hydroxym-
ethyl derivative 3h shows a subpicomolar IC50 value and
is, to our knowledge, the most potent NK1 receptor
ligand so far described (Figure 1).
On the other hand, tertiary amides 3d ,g,i,l were
much less potent than their secondary counterparts
3c,f,h ,k . One exception is represented by the simplest
N-methyl-3-quinolinecarboxamide derivative 3b, which
is more than 30 000 times more potent than the corre-
sponding secondary derivative 3a (Chart 2). The appar-
ent discrepancy can be easily explained in light of the
different conformational preferences shown by 3a with
respect to those of the remaining compounds. The
presence of the hydrogen substituents in position 2 of
the quinoline nucleus and on the amide nitrogen could
allow 3a to populate conformations in which the amide
hydrogen interacts with the π electron cloud of the
pendent phenyl (N-H‚‚‚π hydrogen bond) and the Cd
O group interacts with the hydrogen atom in position 2
(C-H‚‚‚O hydrogen bond).9 Conversely, such conforma-
tions are not populated in 3b because of the steric
repulsion between N-CH3 and the pendent phenyl
group. This hypothesis is strongly supported by the
chemical shift of the proton in position 2 (9.32 ppm in
3a and 8.87 ppm in 3b in CDCl3). Therefore, the
bioactive conformation of these 3-quinolinecarboxamide
derivatives 3 appears to be characterized by the out-of-
plane orientation of the carbonyl dipole guaranteed by
the presence of substituents in position 2 of the quino-
line nucleus.
between secondary and tertiary amides. In the second-
ary amide series, the replacement of a hydrogen atom
of the methyl group in position 2 of 3c with a bromine
atom (3f), an amino group (3k ), or a hydroxy group (3h )
enhances the NK1 receptor affinity from 1 to almost 3
orders of magnitude. The most notable enhancing effects
involve compounds 3h ,k showing H-bond donor sub-
stituents, which could play a major role in the stabiliza-
tion of the ligand-receptor complex.10 Additional sub-
stituents on the amine nitrogen of 3k progressively
decrease the affinity as a consequence of the steric bulk
increase (compare 3k with 3m -p ), suggesting a limited
dimension of the pocket accommodating the substitu-
ents in position 2. In the tertiary amide series, the
2-unsubstituted derivative 3b (R2 ) H) is the most
The structure-affinity relationship (SAFIR) analysis
of these ligands reveals a dualism in the SAFIR trends

Letters
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 6 1317
potent compound and even the introduction of small
groups [CH3 (3d ), CH2OH (3i), CH2OCH3 (3j), CH2NH2
(3l), and CH2Br (3g)] produces dramatic affinity de-
creases. Taken together, the SAFIR data suggest that
secondary amide derivatives 3c,e,f,h ,k ,m -o,r bind NK1
receptor with different modalities with respect to ter-
tiary amides 3b,d ,g,i,j,l. Probably, the N-H group of
the secondary amides plays a key role in the interaction
with the NK1 receptor and affects the orientation of
these amides in the binding crevice, while the binding
of tertiary amides is governed by the strong directional
interaction of the CdO group. Secondary amides could
perhaps also stabilize a different conformation of the
NK1 receptor.
P h a r m a cologica l Ch a r a cter iza tion of 3h . Be-
cause it showed the highest affinity for NK1 receptor,
3h was selected for pharmacological characterization in
cultured endothelial cells previously demonstrated to
possess the NK1 receptor.11 Postcapillary endothelial
cells isolated from bovine heart (CVEC) proliferate and
migrate when stimulated with NK1 selective agonists,
thus leading to angiogenesis. NK1 receptor mediated
angiogenesis has been demonstrated to be linked to the
nitric oxide synthase/cyclic GMP pathway.11,12 We se-
lected the microvascular endothelium as a target tissue
to evaluate the properties of the compound for the
potential therapeutic implications NK1 ligands show on
angiogenesis-dependent diseases. Compound 3h was
tested on CVEC proliferation, inositol phosphate turn-
over, and cyclic GMP accumulation. Thus, sparse and
synchronized CVEC were exposed to increasing concen-
trations (0.01-100 nM) of 3h .
F igu r e 2. Effect of 3h on CVEC proliferation: (A) comparison
with SP, with cells that were incubated with increasing
concentrations (0.01-100 nM) of 3h and SP; (B) effect of NK1
antagonist, with cells that were incubated with 3h (0.01-100
nM) either in the absence (black bars) or in the presence of
L-703,606 (1 nM) (white bars). Cellular proliferation was
evaluated after 48 h of exposure to the test substances. Fixed
and stained cells were microscopically counted, and the data
were expressed as total number of cells counted per well. The
values are the mean ( SEM of five experiments run in
triplicate; P < 0.05 for 3h alone vs basal response (Student’s
t test).
Cell proliferation, measured after 48 h of incubation,
was induced by the compound at all the concentrations
tested. A bell-shaped dose-response curve was ob-
tained, as previously reported for SP peptides,13 without
any cytotoxic effect at higher concentrations. The maxi-
mal effect was reached at 0.1 nM of 3h and 1 nM SP
(40% increase over basal response) (Figure 2A), while
Sar9-SP-sulfone induced the maximal effect at 10 nM
(162 ( 3 counted cells per well vs 120 ( 4 in basal
condition, P < 0.01). To verify whether 3h behaves as
an NK1 agonist, cell proliferation was evaluated
in CVEC treated with the NK1 receptor antagonist
L-703,60614 (1 nM). The presence of L-703,606 com-
pletely impaired the proliferative effect of 3h , thus
supporting the fact that 3h is a NK1 agonist (Figure
2B).
Tritiated inositol monophosphate (IP1) accumulation
was measured by ion exchange chromatography in
CVEC exposed for 15-60 min to increasing concentra-
tions of substance P or 3h (1-1000 nM).
Compound 3h produced a concentration-dependent
increase in IP1 levels, producing maximal stimulation
at 1000 nM (Figure 3). IP1 accumulation was detected
with SP only at 10 µM and after 1 h of incubation (38 (
2% increase over basal response).
F igu r e 3. Effect of 3h on IP1 levels in postcapillary endo-
thelial cells. IP1 accumulation was measured by ion exchange
chromatography in [3H]myoinositol labeled CVEC treated with
3h for 15 min. Data are expressed as % increase over basal
activity. The values are the mean ( SEM of at least three
determinations; P < 0.05 for 3h alone vs basal control
conditions (Student’s t test).
while no increase was observed when CVEC was pre-
w
treated with L-703,606 or L-NMMA (L -nitromono-
methyl-L-arginine, a NOS inhibitor)15 (Figure 4B), sup-
porting the involvement of NK1 and NOS pathway
activation. These data thus demonstrate that 3h be-
haves like an NK1 receptor agonist in endothelial cells,
stimulating their proliferation and producing an intra-
cellular accumulation of IP1 and NO-mediated cyclic
GMP.
Con clu sion s. 3-Quinolinecarboxamides bearing the
3,5-bis(trifluoromethyl)benzyl group as the main sub-
stituent of the amide nitrogen and different substituents
in position 2 of the quinoline have been synthesized and
evaluated for their potential ability to inhibit the specific
binding of [125I]BH-SP to native human NK1 receptor.
This structure-affinity relationship study has led to the
discovery of several potent NK1 receptor ligands en-
Compound 3h and SP were then tested on cGMP
production. As shown in Figure 4A, both compounds
induced a concentration-dependent activation of guan-
ylate cyclase. SP produced a characteristic bell-shaped
response curve, while 3h continued to increase cGMP
accumulation at doses higher than 100 nM. Cyclic GMP
levels increased 4 times in the presence of 100 nM 3h ,

1318 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 6
Letters
Refer en ces
(1) Hokfelt, T.; Pernow, B.; Wahren, J . Substance P: A Pioneer
amongst Neuropeptides. J . Intern. Med. 2001, 249, 27-40.
(2) Snider, R. M.; Constantine, J . W.; Lowe, J . A., III; Longo, K. P.;
Lebel, W. S.; Woody, H. A.; Drozda, S. E.; Desai, M. C.; Vinick,
F. J .; Spencer, R. W.; Hess, H.-J . A Potent Nonpeptide Antago-
nist of the Substance P (NK1) Receptor. Science 1991, 251, 435-
437.
(3) Sorbera, L. A.; Silvestre, J .; Castaner, J . GR-205171. Drugs
Future 1999, 24, 254-260.
(4) Natsugari, H.; Ikeura, Y.; Kiyota, Y.; Ishichi, Y.; Ishimaru, T.;
Saga, O.; Shirafuji, H.; Tanaka, T.; Kamo, I.; Doi, T.; Otsuka,
M. Novel, Potent, and Orally Active Substance P Antagonists:
Synthesis and Antagonist Activity of N-Benzylcarboxamide
Derivatives of Pyrido[3,4-b]pyridine. J . Med. Chem. 1995, 38,
3106-3120.
(5) (a) Kalindjian, S. B.; Dunstone, D. J .; Low, C. M. R.; Pether, M.
J .; Roberts, S. P.; Tozer, M. J .; Watt, G. F.; Shankley, N. P.
Nonpeptide Cholecystokinin-2 Agonists. J . Med. Chem. 2001,
44, 1125-1133. (b) Perlman, S.; Costa-Neto, C. M.; Miyakawa,
A. A.; Schambye, H. T.; Hjorth, S. A.; Paiva, A. C.; Rivero, R.
A.; Greenlee, W, J .; Schwartz, T. W. Dual Agonistic and
Antagonistic Property of Nonpeptide Angiotensin AT1 Ligands:
Susceptibility to Receptor Mutations. Mol. Pharmacol. 1997, 51,
301-311 and references therein.
(6) Cappelli, A.; Anzini, M.; Vomero, S.; Mennuni, L.; Makovec, F.;
Hamon, M.; De Benedetti, P. G.; Menziani, M. C. The Interaction
of the 5-HT3 Receptor with Arylpiperazine, Tropane, and Qui-
nuclidine Ligands. Curr. Top. Med. Chem. 2002, 2, 599-624.
(7) Anzini, M.; Cappelli, A.; Vomero, S.; Seeber, M.; Menziani, M.
C.; Langer, T.; Hagen, B.; Manzoni, C. Bourguignon, J .-J .
Mapping and Fitting the Peripheral Benzodiazepine Receptor
Binding Site by Carboxamide Derivatives. Comparison of Dif-
ferent Approaches to Quantitative Ligand-Receptor Interaction
Modeling. J . Med. Chem. 2001, 44, 1134-1150.
F igu r e 4. Effect of 3h on cyclic GMP accumulation in
CVEC: (A) effect of increasing concentrations of SP and 3h ,
where cyclic GMP levels were measured by enzyme immu-
noassay in CVEC monolayers treated with stimuli for 10 min;
(B) involvement of NK1 and NOS activation in which L-703,-
606 (1 nM) or L-NMMA (200 µM) was given 45 min before the
addition of 3h (100 nM). The data are expressed as pmol of
cyclic GMP/mg of protein. The values are the mean ( SEM of
at least three determinations; P < 0.05 for 3h alone vs basal
control conditions (Student’s t test).
(8) The binding studies were performed according to the procedure
recommended by the manufacturer (Amersham Pharmacia
Biotech, Amersham, U.K.).
(9) Cheng, J .; Kang, C.; Zhu, W.; Luo, X.; Mok Puah, C.; Chen, K.;
Shen, J .; J iang, H. N-Methylformamide-Benzene Complex as a
Prototypical Peptide N-H‚‚‚π Hydrogen-Bonded System: Den-
sity Functional Theory and MP2 Studies. J . Org. Chem. 2003,
68, 7490-7495 and references therein.
(10) The contribution of a neutral-neutral hydrogen bond has been
estimated to be around 0.5-1.5 kcal/mol in binding (equivalent
to a 2- to 15-fold increase in affinity), while a hydrogen bond
between a charged and a neutral component can contribute up
to 4.7 kcal/mol (equivalent to a 3000-fold increase in affinity).
Davis A. M.; Teague, S. J . Hydrogen Bonding, Hydrophobic
Interactions, and Failure of the Rigid Receptor Hypothesis.
Angew. Chem., Int. Ed. 1999, 38, 736-749.
(11) (a) Ziche, M.; Morbidelli, L.; Parenti, A.; Amerini, S.; Granger,
H. J .; Maggi, C. A. Substance P Increases Cyclic GMP Levels
on Coronary Postcapillary Venular Endothelial Cells. Life Sci.
1993, 53, PL229-234. (b) Ziche, M.; Parenti, A.; Ledda, F.;
Dell’Era, P.; Granger, H. J .; Maggi, C. A.; Presta, M. Nitric Oxide
Promotes Proliferation and Plasminogen Activator Production
by Coronary Venular Endothelium through Endogenous bFGF.
Circ. Res. 1997, 80, 845-852
dowed with picomolar affinity. In particular, the hy-
droxymethylcarboxamide 3h showed an IC50 value in
the subpicomolar range and behaved as an agonist of
NK1 receptor in the endothelial cell proliferation, inosi-
tol phosphate turnover, and NO-mediated cyclic GMP
accumulation. The compound displayed the highest
potency (subnanomolar) in the functional assay on
microvascular endothelium (cell proliferation), while it
was considerably weaker in the intracellular signal
assays. This finding, also observed in the case of SP,
may reflect the poor association between these signals
and cell growth. In conclusion, 3h is the first non-
peptide NK1 receptor agonist showing very high potency
and theoretically useful as a tool in the study of NK1
receptor function. Moreover, the discovery of non-
peptide tachykinin agonists seems to be of therapeutical
relevance in angiogenesis-dependent diseases where
endothelial cell proliferation is required to promote
vascularization and healing of ischemic or damaged
tissues.
(12) Ziche, M.; Morbidelli, L.; Masini, E.; Amerini, S.; Granger, H.
J .; Maggi, C. A.; Geppetti, P.; Ledda, F. Nitric Oxide Mediates
Angiogenesis in Vivo and Endothelial Cell Growth and Migration
in Vitro Promoted by Substance P. J . Clin. Invest. 1994, 94,
2036-2044.
Ack n ow led gm en t. Thanks are due to Italian MIUR
for financial support. Prof. Stefania D’Agata D’Ottavi’s
(Universita` di Macerata, Italy) and Prof. Antonio
Giachetti’s (Lifetech s.r.l., Firenze, Italy) careful reading
and discussion of the manuscript are also acknowledged.
Dr. F. Finetti is a recipient of a fellowship from
Fondazione Callerio, Trieste, Italy.
(13) Ziche, M.; Morbidelli, L.; Pacini, M.; Geppetti, P.; Alessandri,
G.; Maggi, C. A. Substance P Stimulates Neovascularization in
Vivo and Proliferation of Cultured Endothelial Cells. Microvasc.
Res. 1990, 40, 264-278.
(14) Cascieri, M. A.; Ber, E.; Fong, T. M.; Sadowski, S.; Bansal, A.;
Swain, C.; Seward, E.; Frances, B.; Burns, D.; Strader, C. D.
Characterization of the Binding of a Potent, Selective, Radioio-
dinated Antagonist to the Human Neurokinin-1 Receptor. Mol.
Pharmacol. 1992, 42, 458-463.
(15) Moncada, M.; Higgs, A.; Furghgott, R. XIV. International Union
of Pharmacology Nomenclature in Nitric Oxide Research. Phar-
macol. Rev. 1997, 49, 137-142.
Su p p or tin g In for m a tion Ava ila ble: Details of the syn-
thesis and characterization of 3 (chemistry, NMR, MS, and
pharmacology). This material is available free of charge via
the Internet at http://pubs.acs.org.
J M034219A