Discovery of potent cyclic pseudopeptide human tachykinin NK-2 receptor antagonists

Full text of DOI:10.1021/jm0010217

© Copyright 2000 by the American Chemical Society
Volume 43, Number 22
November 2, 2000
Communications to the Editor
Ch a r t 1
Discover y of P oten t Cyclic
P seu d op ep tid e Hu m a n Ta ch yk in in NK-2
Recep tor An ta gon ists
Danilo Giannotti,* Enzo Perrotta, Cristina Di Bugno,
Rossano Nannicini, Nicholas J . S. Harmat,
Alessandro Giolitti, Riccardo Patacchini,
Anna Rita Renzetti, Luigi Rotondaro,
Sandro Giuliani, Maria Altamura, and
Carlo A. Maggi
Menarini Ricerche S.p.A., Via Sette Santi 3,
I-50131 Firenze, Italy
Received J uly 3, 2000
In tr od u ction . The tachykinins substance P (SP),
neurokinin A (NKA), and neurokinin B (NKB) are a
family of neuropeptides that are widely distributed in
the mammalian peripheral and central nervous systems
and produce their biological actions by activating three
distinct receptor types, termed NK-1, NK-2, and NK-3.
NKA exerts its biological effects mainly by activation
of the tachykinin NK-2 receptor. The human NK-2
receptor has been identified and validated as a suitable
target for development of novel drugs to be used for
treatment of a number of diseases in the respiratory,
gastrointestinal, and genitourinary tracts.1
Since the rational design of the bicyclic hexapeptide
MEN 10627 or cyclo(Met-Asp-Trp-Phe-Dap-Leu)cyclo-
(2â-5â)2,3 (Chart 1), a potent and selective tachykinin
NK-2 receptor antagonist,4,5 the glycosylated analogue
MEN 11420 (Nepadutant) or cyclo{[Asn(â-D-GlcNAc)-
Asp-Trp-Phe-Dap-Leu]cyclo(2â-5â)}6 (Chart 1) was se-
lected for further development and is currently in phase
II clinical trials. Structural studies on MEN 10627 have
shown the presence of a type I and type II â-turn, with
Trp-Phe and Leu-Met as corner residues, respec-
tively.2,7,8 In addition site-directed mutagenesis studies
on labeled MEN 11420 suggested a primary role of the
Trp-Phe moiety in the binding interactions with the
tachykinin NK-2 receptor.9,10 On this background a
chemical program was undertaken to search for lower
molecular weight compounds which maintained the
same potency and selectivity of our reference com-
pounds.
Initially we prepared the monocyclic compound 5a
(Scheme 1) that possesses only micromolar binding
affinity; however, we considered this value sufficient to
use 5a as a lead compound for a new family of NK-2
antagonists. In fact Figure 1 shows the comparison of
the type I â-turns for compounds 5a and MEN 10627.11
The smaller monocyclic compound maintains the struc-
ture of the larger bicyclic peptide, with exact overlap-
ping of the Trp-Phe motif. This comparison suggests
that the monocyclic compound 5a should be the small-
molecule model of the Trp-Phe â-turn for which we were
looking.
It has been observed for other pseudopeptide NK-2
antagonists that an increase in lipophilicity was related
to the increment of NK-2 receptor antagonist activity.12
For this reason we introduced a benzyl group at posi-
tions 8, 9, 12, and 13 (Table 1), as a probe in order to
improve the binding affinity of our lead compound 5a .
We now report on the discovery of this novel class of
potent monocyclic pseudopeptides as human NK-2
receptor antagonists and part of our work aimed at
optimizing the binding affinity.
Ch em istr y. The chemical synthesis of 5a and of
compounds containing the benzyl group at positions 9,
* Corresponding author. Phone: +39 0555680746. Fax: +39
0555680419. E-mail: fbenincasa@menarini-ricerche.it.
10.1021/jm0010217 CCC: $19.00 © 2000 American Chemical Society
Published on Web 10/13/2000

4042 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 22
Sch em e 1. Synthesis of Compounds 5a -c, 7, and 9a
Communications to the Editor
a
Reagents: (a) Z-Phe-OH, EDC, HOBT, DMF (80%); (b) H2, Pd/C, MeOH (90-97%); (c) Z-Trp-OH, EDC, HOBT, DMF (95%); (d) succinic
anhydride, DMF (95%); (e) TFA, CH2Cl2 (75-85%); (f) EDC, HOBT, DMF (75-85%); (g) 2-benzylbutanedioic acid 4-tert-butyl ester, EDC,
HOBT, DMF (70-80%).
Ta ble 1. Chemical Structure and Affinities for the Tachykinin
NK-2 Receptor Evaluated in Binding Experiments (pKi values)
and Functional in Vitro Experiments (pKB values)
a
a
compd
R
R1
R2
R3
config
pKi
pKB
5a
5b
5c
H
H
H
H
5.9 ( 0.1 6.3 ( 0.15
5.9 ( 0.1 5.4 ( 0.19
5.9 ( 0.1 5.5 ( 0.18
H
Bzl
Bzl
H
H
H
(R)
(S)
H
H
Bzl
H
H
H
H
H
Bzl
H
H
F igu r e 1. Superposition of MEN 10627 (light) and 5a (dark).
7
H
(R,S) 6.2 ( 0.1 5.7 ( 0.2
9
14b
14c
H
H
(R,S) 6.1 ( 0.1 6.1 ( 0.09
12, and 13 was performed as described in Scheme 1. The
synthesis of compounds containing the substituent at
position 8 is described in Scheme 2.
The diamines 1 were prepared starting from the
corresponding Boc-amino acids using known pro-
cedures13-15 and were reacted with commercially avail-
able protected amino acids under standard peptide
coupling conditions.
As described in Scheme 1 the diamines 1a -c were
condensed with Z-phenylalanine, and after removal of
the Z-protecting group, the resulting intermediates 2
were condensed with Z-tryptophan to give products 3.
For the 9-substituted benzyl derivatives, compounds
3a -c were converted to tetrapeptides 4a -c by Z-
deprotection and reaction with succinic anhydride. After
Boc-deprotection of the diamine moiety of the molecule,
the corresponding zwitterions were finally cyclized to
give the products 5a -c. For the 13-substituted benzyl
derivatives, the compound 3a was reacted after Z-
deprotection with 2-benzylbutanedioic acid 4-tert-butyl
ester16 to give 6 where subsequent Boc-deprotection and
cyclization furnished 7 as a diastereoisomeric mixture.
For the 12-substituted benzyl derivatives, the interme-
Bzl
Bzl
H
(S)
(R)
5.9 ( 0.1 5.8 ( 0.05
8.7 ( 0.3 7.6 ( 0.09
H
a
pKi and pKB are -log of Ki and KB, respectively; see refs 17
and 18.
diate resulting from Boc-deprotection of 3a was reacted
with 2-benzylbutanedioic acid 4-tert-butyl ester to give
8. This was submitted to Z- and tert-butyl ester depro-
tection and cyclized to give the cyclopeptide 9, as a
diastereoisomeric mixture.
As described in Scheme 2 the diamines 1b-j were
transformed into Fmoc-protected diamines 10 and con-
densed with Boc-phenylalanine. Boc-Deprotection fur-
nished compounds 11 which were reacted with Boc-
tryptophan and submitted to Fmoc-deprotection to give
12. Reaction of intermediates 12 with succinic anhy-
dride and subsequent Boc-deprotection furnished inter-
mediates 13 that were cyclized to products 14.
The cyclization step on these compounds was per-
formed in DMF at 0.01 M concentration in order to
minimize intermolecular condensation reactions.
Resu lts a n d Discu ssion . The binding17 data and the

Communications to the Editor
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 22 4043
Sch em e 2. Synthesis of Compounds 14b-ja
F igu r e 2. Superposition of 5a (light) and 14c (dark).
pseudopeptide 14c that possesses the same nanomolar
potency (pKi ) 8.7 ( 0.3) as MEN 11420 (pKi ) 8.6 (
0.08) and is slightly less potent than MEN 10627 (Table
2).
The NK-2 binding affinity and functional activity are
enantioselective properties of this chemical series, the
C-8 (R)-stereoisomer 14c being about 3 orders of mag-
nitude more potent than the C-8 (S)-compound 14b in
binding and 2 orders of magnitude more potent in
functional activity. The selective antagonist activity of
14c for the tachykinin receptors was tested in guinea-
pig isolated ileum bioassay,20 and the resulting values:
NK-1 pKB ) 5.7 ( 0.1, NK-2 pKB ) 7.8 ( 0.06, NK-3
pKB ) 6.1 ( 0.08, convinced us to consider compound
14c as the structural lead (designated MEN 11558) for
a new class of compounds with potent NK-2 receptor
antagonist activities.
a
Reagents: (a) Fmoc-OSu, THF (70-75%); (b) TFA, CH2Cl2
(85-95%); (c) Boc-Phe-OH, EDC, HOBT, DIPEA, DMF (82-94%);
(d) Boc-Trp-OH, EDC, HOBT, DIPEA, DMF (87-97%); (e) Et2NH,
DMF (66-75%); (f) succinic anhydride, DMF (90-95%); (g) EDC,
HOBT, DMF (75-85%).
Ta ble 2. Affinities of the Test Compounds for the Tachykinin
NK-2 Receptor Evaluated in Binding Experiments (pKi values)
and Functional in Vitro Experiments (pKB values)
Figure 2 shows the comparison between 14c and 5a .
The good overlap indicates that the introduction of a
benzyl group in position 8 with the (R)-configuration
does not change the â-turn conformation of the mono-
cyclic structure. This observation suggests that the
increase in activity observed in 14c is probably due to
favorable interaction of the benzyl group with the NK-2
receptor rather than to a conformational constraint
effect.
To understand the role of the lipophilic substituent
in position 8 and to optimize the binding affinity, we
prepared other C-8 derivatives with the (R)-configura-
tion (Table 2). Initial replacement of the benzyl moiety
by the cyclohexylmethyl group dramatically decreased
the pKi and pKB values of the resulting compound 14d .
For this reason we used aromatic residues such as
phenylethyl by which 14e was obtained. However it also
showed a reduction in affinity (pKi) and a poor pKB
value. A slighty better result was observed for compound
14f where the benzyl moiety was substituted with a
2-naphthylmethyl residue.
a
a
compd
14c
14d
14e
14f
14g
14h
14i
R
pKi
pKB
-CH2(Ph)
-CH2(cyclohexyl)
-CH2CH2(Ph)
-CH2(2-naphthyl)
-CH2[(4-OMe)Ph]
-CH2[(4-CF3)Ph]
-CH2[(3,4-diF)Ph]
-CH2[(3,4-diCl)Ph]
8.7 ( 0.3
6.5 ( 0.1
7.6 ( 0.09
6.5 ( 0.01
6.5 ( 0.09
7.0 ( 0.01
7.1 ( 0.01
7.1 ( 0.01
7.8 ( 0.02
8.1 ( 0.03
8.1 ( 0.1
8.0 ( 0.07
8.1 ( 0.25
8.0 ( 0.15
7.7 ( 0.1
8.8 ( 0.1
14j
MEN 10627
MEN 11420
10.0 ( 0.17
9.2 ( 0.09
8.6 ( 0.08
8.7 ( 0.07
a
See corresponding footnote in Table 1.
in vitro functional test18 of the novel NK-2 receptor
antagonists are presented in Tables 1 and 2.19
Assuming that the aromatic ring and the interatomic
distance between its centroid and the C-8 carbon of the
core structure can play a key role on the activity of our
compounds, we used substituted benzyl derivatives to
obtain compounds 14g-j. The benzyl group substituted
at the para position with electron-withdrawing groups,
e.g. trifluoromethyl (14h ), or the electron-donating
methoxy group (14g) did not show any improvement in
Table 1 shows that compound 14c is the best NK-2
antagonist of this series, while no significant difference
exists between our lead compound 5a and the other
benzyl derivatives. It is noteworthy that introduction
of the benzyl group at the C-8 position with (R)-
configuration increased binding affinity about 1000-fold
with respect to 5a and led to the discovery of the cyclic

4044 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 22
Communications to the Editor
(5) Quartara, L.; Pavone, V.; Pedone, C.; Lombardi, A.; Renzetti,
A. R.; Maggi, C. A.; A Review of the Design, Synthesis and
Biological Activity of the Bicyclic Hexapeptide Tachykinin NK2
Antagonist MEN 10627. Regul. Pept. 1996, 65, 55-59.
(6) Catalioto, R.-M.; Criscuoli, M.; Cucchi, P.; Giachetti, A.; Gian-
notti, D.; Giuliani, S.; Lecci, A.; Lippi, A.; Patacchini, R.;
Quartara, L.; Renzetti, A. R.; Tramontana, M.; Arcamone, F.;
Maggi, C. A. MEN 11420 (Nepadutant), a Novel Glycosylated
Bicyclic Peptide Tachykinin NK2 Receptor Antagonist. Br. J .
Pharmacol. 1998, 123, 81-91.
(7) Lombardi, A.; D’Auria, G.; Saviano M.; Maglio, O.; Nastri, F.;
Quartara, L.; Pedone, C.; Pavone, V. Bicyclic Peptides as Type
I/Type II â-Turn Scaffolds. Biopolymers 1997, 40, 505-518.
(8) Lombardi, A.; D’Auria, G.; Maglio, O.; Nastri, F.; Quartara, L.;
Pedone, C.; Pavone, V. A Novel Rigid â-Turn Molecular Scaffold.
J . Am. Chem. Soc. 1998, 120, 5879-5886.
(9) Renzetti, A. R.; Catalioto, R.-M.; Criscuoli, M.; Cucchi, P.; Ferrer,
C.; Giolitti A.; Guelfi, M.; Rotondaro, L.; Warner, F. J .; Maggi,
C. A. Relevance of Aromatic Residues in Transmembrane
Segments V to VII for Binding of Peptide and Nonpeptide
Antagonists to the Human Tachykinin NK2 Receptor. J . Phar-
macol. Exp. Ther. 1999, 290, 487-495.
(10) Giolitti, A.; Cucchi, P.; Renzetti, A. R.; Rotondaro, L.; Zappitelli,
S.; Maggi, C. A. Molecular Determinants of Peptide and Non-
peptide NK-2 Receptor Antagonists Binding Sites of the Human
Tachykinin NK-2 Receptor by Site-directed Mutagenesis. Neu-
ropharmacology 2000, 39, 1422-1429.
(11) Conformational studies were performed through “simulated
annealing” techniques, in the Sybyl molecular modeling software
package (Tripos Inc., St. Louis, MO).
(12) Quartara, L.; Fabbri, G.; Ricci, R.; Patacchini, R.; Pestellini, V.;
Maggi, C. A.; Pavone, V.; Giachetti, A.; Arcamone, F. Influence
of Lipophilicity on the Biological Activity of Cyclic Pseudopeptide
NK-2 Receptor Antagonists. J . Med. Chem. 1994, 37, 3630-3638.
(13) Stanfield, F. C.; Parker, J . E.; Kanellis, P. Synthesis of Protected
Amino Alcohols: A Comparative Study. J . Org. Chem. 1981, 46,
4799-4800.
(14) Kokotos, G.; Costantinou-Kokotou, V. Modified Amino Acids and
Peptides. Part 2. A Convenient Conversion of Amino and Peptide
Alcohols into Amines. J . Chem. Res. (S) 1992, 391.
(15) O’Brien, P. M.; Sliskovic, D. R.; Blankley, J . C.; Roth, B. D.;
Wilson, M. W.; Hamelehle, K. L.; Krause, B. R.; Stanfield, R. L.
Inhibitors of Acyl-CoA: Cholesterol O-Acyl Transferase (ACAT)
as Hypocholesterolemic Agents. 8. Incorporation of Amide or
Amine Functionalities into a Series of Disubstituted Ureas and
Carbamates. Effects on ACAT Inhibition in Vitro and Efficacy
in Vivo. J . Med. Chem. 1994, 37, 1810-1822.
Ta ble 3. Kd Values for the Two Antagonists MEN 11420 and
14c Against [125I]NKA (NKA) and [3H]SR 48968 (SR) on the
Wild-Type and Four Point-Mutated Human NK-2 Receptors
(see text)
wild-type Tyr206Ala Tyr206Phe Tyr266Phe Phe270Ala
NKA SR NKA SR
NKA SR NKA SR NKA SR
MEN
11420
14c
2.1
4.4
>1000 3.1 1.9
26
34
48
48
291
208
2.7 10
1960 3.6 1.3
activity if compared to 14c. The substitution of positions
3 and 4 of the benzyl group by halogen atoms was more
satisfactory: in fact the 3,4-difluoro derivative 14i
retains the activity of 14c and the 3,4-dichloro deriva-
tive 14j showed an improvement in the activity values:
pKi ) 10 ( 0.17 and pKB ) 8.1 ( 0.03.
On the basis of these results, we hypothesized that
while the benzyl moieties in compounds 14c,i,j produce
a new interaction with a previously unexplored hydro-
phobic pocket of the NK-2 receptor, the Trp-Phe moiety
of our new class of cyclic compounds interacts with the
NK-2 receptor similarly as in MEN 11420.10 To confirm
this, we tested the binding affinity of 14c toward a series
of point-mutated human NK-2 receptors transfected in
CHO cells that were previously identified as relevant
for the interaction with Trp-Phe in MEN 11420.10 The
results, reported in Table 3, agree with this hypothesis
adding experimental support to our initial assumption.
These compounds possess pKi and pKB values compa-
rable to those of MEN 10627 and MEN 11420 with the
advantage of lower molecular weight and fewer stereo-
genic centers, two factors that confer to this type of
molecule an interesting opportunity for further develop-
ment.
Ongoing studies are evaluating the potential thera-
peutic utility of compounds in this series, and new
derivatives related to MEN 11558 will be reported in
due time.
(16) Conrow, R.; Portoghese, P. S. Efficient Preparation of Polyfunc-
tional R-Diketones from Carboxylic Acids. J . Org. Chem. 1986,
51, 938-940.
Ack n ow led gm en t. We thank Dr. Antonio Triolo
and Dr. Giuseppe Balacco (Menarini Ricerche S.p.A.)
for mass spectra and NMR spectroscopic determina-
tions. This work was supported in part by MURST (IMI
Grant No. 63217).
(17) Binding affinity for the human NK-2 receptor transfected in
CHO cells was determined in competition experiments using
[
125I]NKA as radioligand (see ref 6). The pKi values, calculated
using EBDA and LIGAND programs (Munson, P. J .; Rodband.
Anal. Biochem. 1980, 107, 220-239) in sequence, represent the
mean value determined from 2-6 experiments, each performed
in duplicate.
Su p p or tin g In for m a tion Ava ila ble: MS and 1H NMR
spectra for compounds 14c,i,j. This material is available free
of charge via the Internet at http://pubs.acs.org.
(18) Functional studies were performed in the rabbit isolated pul-
monary artery, set up as described previously: Patacchini, R.;
et al. Br. J . Pharmacol. 1991, 104, 91. The pKB values represent
the average of 3-8 independent determinations.
(19) The purity of final compounds was found to be g95% by HPLC.
(20) The selectivity of compound 14c was assayed on the guinea-pig
isolated ileum, set up for recording NK-1 or NK-3 receptor-
mediated responses (Patacchini, R.; Maggi, C. A. Tachykinin
Receptor Assays. In Current Protocols in Pharmacology; Enna,
S., Ferkany, M., Williams, J ., Kenakin, T., Porsolt, R., Sullivan,
J ., Eds.; J . Wiley and Sons: New York, 1998; Chapter 4, units
4.10, 4.10.1-4.10.26). The data obtained in the former bioassay
were compared to the affinity of compound 14c shown at the
NK-2 receptor of the guinea-pig isolated bronchus set up as
described previously: Maggi, C. A.; et al. Eur. J . Pharmacol.
1991, 197, 167-174. The pKB values represent the average of
3-8 independent determinations.
Refer en ces
(1) Maggi, C. A.; Patacchini, R.; Rovero, P.; Giachetti, A. Tachykinin
Receptors and Tachykinin Receptor Antagonists. J . Auton.
Pharmacol. 1993, 13, 23-93.
(2) Pavone, V.; Lombardi, A.; Nastri, F.; Saviano, M.; Maglio, O.;
D’Auria, G.; Quartara, L.; Maggi, C. A.; Pedone, C. Design and
Structure of a Novel Neurokinin A Receptor Antagonist Cyclo-
(-Met1-Asp2-Trp3-Phe4-Dap5-Leu6)cyclo(2â-5â). J . Chem. Soc.
Perkin Trans. 2 1995, 987-993.
(3) Pavone, V.; Lombardi, A.; Maggi, C. A.; Quartara, L.; Pedone,
C. Conformational Rigidity versus Flexibility in a Novel Peptidic
Neurokinin A Receptor Antagonist. J . Pept. Sci. 1995, 1, 236-
240.
(4) Maggi, C. A.; Astolfi, M.; Giuliani, S.; Goso, C.; Manzini, S.;
Meini, S.; Patacchini, R.; Pavone, V.; Pedone, C.; Quartara, L.;
Renzetti, A. R.; Giachetti, A. MEN 10,627 a Novel Polycyclic
Peptide Antagonist of Tachykinin NK2 Receptors. J . Pharmacol.
Exp. Ther. 1994, 271, 1489-1500.
J M0010217