Synthesis and biological activity of NK-1 selective, N-backbone cyclic analogs of the C-terminal hexapeptide of substance P

Article abstract of DOI:10.1021/jm960154i

The application of the concept of backbone cyclization to linear substance P (SP) analogs is presented. We describe the synthesis, characterization, and biological activity of a series of backbone-to-amino- terminus cyclic analogs of the C-terminal hexapeptide of SP. These analogs were designed on the basis of NMR data and molecular modeling of the selective NK-1 analog WS-septide (Ac[Arg⁶,Pro⁹]SP_6-11). A series of peptides with the general formula: cyclo[-(CH₂)(m)-NH-CO-(CH₂)(n)-CO-Arg- Phe-Phe-N-]-CH₂-CO-Leu-Met-NH₂ (n = 2, 3, 6 and m = 2, 3, 4) was synthesized by solid phase methodology using Fmoc chemistry for the main chain and Boc chemistry for the building units [N(α)-(ω-aminoalkyl)Gly] side chains. Cyclization was performed on the resin after removal of the Boc protecting group from the ω-aminoalkyl chain. Cyclic and precyclic analogs were compared. They were purified by HPLC and characterized by mass spectroscopy and NMR. Biological activity and selectivity to the NK-1 neurokinin receptor were found to depend on cyclization and the ring size: The most active and selective analog had a ring of 20 atoms. This analog was found to have enhanced metabolic stability in various tissue preparation compared to WS-septide.

Full text of DOI:10.1021/jm960154i

3174
J . Med. Chem. 1996, 39, 3174-3178
Syn th esis a n d Biologica l Activity of NK-1 Selective, N-Ba ck bon e Cyclic An a logs
of th e C-Ter m in a l Hexa p ep tid e of Su bsta n ce P
Gerardo Byk,†,§ David Halle,‡ Irena Zeltser,† Gal Bitan,† Zvi Selinger,‡ and Chaim Gilon*,†
Departments of Organic Chemistry and Biological Chemistry, The Hebrew University of J erusalem, J erusalem 91904, Israel
Received February 26, 1996X
The application of the concept of backbone cyclization to linear substance P (SP) analogs is
presented. We describe the synthesis, characterization, and biological activity of a series of
backbone-to-amino-terminus cyclic analogs of the C-terminal hexapeptide of SP. These analogs
were designed on the basis of NMR data and molecular modeling of the selective NK-1 analog
WS-septide (Ac[Arg6,Pro9]SP6-11). A series of peptides with the general formula: cyclo[-(CH2)m-
NH-CO-(CH2)n-CO-Arg-Phe-Phe-N-]-CH2-CO-Leu-Met-NH2 (n ) 2, 3, 6 and m ) 2, 3, 4) was
synthesized by solid phase methodology using Fmoc chemistry for the main chain and Boc
chemistry for the building units [NR-(ω-aminoalkyl)Gly] side chains. Cyclization was performed
on the resin after removal of the Boc protecting group from the ω-aminoalkyl chain. Cyclic
and precyclic analogs were compared. They were purified by HPLC and characterized by mass
spectroscopy and NMR. Biological activity and selectivity to the NK-1 neurokinin receptor
were found to depend on cyclization and the ring size: The most active and selective analog
had a ring of 20 atoms. This analog was found to have enhanced metabolic stability in various
tissue preparation compared to WS-septide.
In tr od u ction
Tachykinins are a large family of peptides which
share the common sequence Phe-Xaa-Gly-Leu-Met-NH2
at their C-terminus and show a great variation in their
N-terminal sequence. While most of the tachykinins are
F igu r e 1. Amino acid sequences of the mammalian tachyki-
nins (neurokinins).
from nonmammalian origin, the three peptides called
two out of the three receptor subtypes, while activity
neurokinins: substance P (SP), neurokinin A (NKA),
on the third receptor remained unchanged. Thus,
and neurokinin B (NKB), were identified in mammals
N-methylation of the Phe7-Phe8 peptide bone inferred
(Figure 1).
selectivity to the NK-3 receptor whereas N-methylation
of the Phe8-Gly9 peptide bone caused selectivity to the
Extensive studies implicated the neurokinins in a
NK-1 receptors.3 On the basis of structure-activity
relationship studies, we prepared the NK-1 selective
agonist WS-septide (Ac[Arg6,Pro9]SP6-11 and the highly
variety of physiological functions such as transmission
of pain stimuli, exocrine gland secretion, intestinal
motility, vasodilation, neuronally mediated inflamma-
tory skin reaction, and behavioral responses.1 The large
variety of biological activities exerted by the neurokinins
has indicated the existence of several receptors (or
receptor subtypes). The three neurokinin receptors are
NK-1, NK-2, and NK-3. Even though the neurokinin
receptors are activated preferentially by each of the
neurokinins (NK-1 by SP, NK-2 by NKA, and NK-3 by
NKB), the selectivity of each neurokinin to its receptor
is rather poor.1,2 Thus, for example, NKB activates both
the NK-1 and NK-3 receptors with almost the same
EC50 (4.2 and 1.3 nM, respectively). Since apparently
each of the receptor subtypes mediates distinct physi-
ological functions, it is essential to have receptor selec-
tive agonists and antagonists in order to relate the exact
physiological function to each receptor. In our labora-
tory the design of selective neurokinin receptor agonists
came out of a systematic study in which each of the
peptide bonds in the C-terminal hexapeptide of SP was
replaced by an N-methylated peptide bond. These
studies revealed that N-methylation of specific peptide
bonds resulted in selective loss of biological activity on
selective NK-3 agonist senktide (succ[Asp6,N-MePhe8]-
4,5
SP6-11
;
(see Table 1). Senktide was found to be
metabolically stable in many tissues,6 whereas WS-
septide was found to be metabolically unstable (Figure
3). This fact prompted us to design novel metabolically
stable NK-1 agonists.
Design of Ba ck bon e Cyclic An a logs
Theoretical energy calculations7 indicated that the
C-terminal hexapeptide of SP had a folded conformation
stabilized by hydrogen bonding between the C-terminal
carboxamide and the carboxamide of Gln6. This model,
which was later corroborated by NMR studied,8,9
prompted researchers to prepare cyclic analogs of SP
in order to stabilize the C-terminal hexapeptide in the
predicted bent conformation. However, although some
success was made in preparation of NK-2 antago-
nists,10,11 all the cyclic analogs in which cyclization
involved in the C-terminal hexapeptide were either
inactive or 2-3 orders of magnitude less active as
agonists than SP itself.12-18
There were two possible explanations for the lack of
activity of these cyclic analogs: Structure-activity
relationship studies indicated that only minor modifica-
tions were permissible in the structure or chirality of
the side chains constituting the active region of SP-
(-Phe-Phe-Gly-Leu-Met-NH2).1 It was, hence, plausible
† Department of Organic Chemistry.
‡ Department of Biological Chemistry.
§ Present address: UMR-133 CNRS/RPR, Rhoˆne Poulenc Rorer,
Centre de Recherche de Vitry-Alfortville, 13 Quai J ules Guesde, B.P.
14, 94403 Vitry-sur-Seine, France.
X Abstract published in Advance ACS Abstracts, J uly 1, 1996.
S0022-2623(96)00154-9 CCC: $12.00 © 1996 American Chemical Society

N-Backbone Cyclic Analogs of SP
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 16 3175
Ta ble 1. Structure, Activity, and Selectivity of Linear,
Backbone Cyclic and Precyclic SP Analogs
F igu r e 2. Various modes of N-backbone cyclization.
that the loss of activity of the cyclic peptides stemmed
from alterations in the side chains or the C-terminal
carboxamide necessary for receptor activation. It was
also possible that the cyclization conferred constraints
that prevented the peptides from attaining the active
conformation. We attributed the loss of biological
activity of the cyclic peptides to structural rather than
conformational causes.
Conformational analysis of senktide and WS-septide
19
using NMR studies in DMSO-d6 revealed that each
of these selective linear agonists had an exclusive
predominant conformation in DMSO-d6. The neuroki-
nin NK-1 receptor selective analog WS-septide adopted
mainly a type I â conformation, stabilized by hydrogen
bonds between Phe8 CdO and Met11 N-H and between
Arg6 N-HG and Met11 carboxamide. The existence of
a predominant conformation in this analog was at-
tributed to the incorporation of Pro9 which stabilized
the conformation. We then faced the question: Was the
predominant conformation of WS-septide found by NMR
studies relevant to its biologically active conformation?
This question could be answered by solving the 3D
structure of the WS-septide:NK-1 complex. Since this
information has not been available, the problem was
approached by the design and synthesis of constrained
WS-septide cyclic analogs based on the findings of the
NMR studies.
In order to overcome the limitations of the classical
cyclization methods, which led mostly to inactive pep-
tides, we suggested a new general concept of backbone
cyclization (BC) which would not necessarily alter either
side chains or the carboxyl and amino termini. Accord-
ing to this method, cyclization can be accomplished by
adjoining atoms in the peptide backbone rather than
side chains or terminal groups (Figure 2.) NR and/or
CR hydrogens are replaced by ω-functional alkyl chains,
which can then be interconnected to form the desired
ring.21,22
peptides containing a lactam ring, which differed from
each other by the number of atoms in the ringsranging
from 17 atoms (peptide 1, Table 1) to 22 atoms (peptide
6, Table 1).
P ep tid e Syn th esis. Backbone cyclic peptides were
synthesized by the solid phase peptide synthesis meth-
odology using Boc chemistry for the first two amino
acids (Leu and Met) and then Fmoc chemistry for the
rest of the peptide. Boc was used for the protection of
the ω-amino group of the NR-(ω-aminoalkyl)Gly building
units.23,24 No special procedure was required for the
coupling of the Fmoc-NR-[ω-(Boc-amino)alkyl]Gly-OH
building units to the growing peptide. After the removal
of the Fmoc protecting group, repeated couplings with
BOP were used to ensure coupling of Fmoc-Phe to the
secondary R-amine of the Gly building units. Monitor-
ing the coupling by Kaiser test was impossible. This
difficultly was overcome using amino acid analysis of
peptide-resin hydrolysates after every coupling to the
secondary amine and checking its completion. Gener-
ally three repetitions were necessary to complete the
condensation.
After the coupling of Fmoc-Arg and removal of the
Fmoc protecting group, the peptides were reacted with
dicarboxylic acid spacers (as their anhydrides) with a
catalytic amount of DMAP. Then the Boc protecting
group on the NR-(ω-aminoalkyl)Gly unit was removed,
and the peptide was cyclized on the resin. In the case
of peptide 4 (n ) 3, m ) 4), the peptide-resin obtained
was divided into two parts after the appropriate wash-
Resu lts a n d Discu ssion
Molecular models of the predominant conformation
of WS-septide based on the NMR data indicated that
conformational restriction could be imposed on WS-
septide by cyclization in which the nitrogen of the Phe8-
Pro9 peptide bond would be attached to the amino
terminal Arg6 by a ring. Since the molecular models
could not define the exact distance between the nitrogen
of the terminal Arg and the nitrogen of the Phe-Pro
bond, we prepared a series of six cyclic homologous

3176 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 16
Byk et al.
F igu r e 3. Time course of degradation of WS-septide (b) and cycloseptide (9) by slices of (left) liver and (right) parotid gland.
ing and drying. In one part the free amine was
acetylated and the free carboxyl group was amidated
with N-methylamine to yield the counterpart precyclic
peptide 9 (Table 1). The second part of the peptide-resin
was subjected to cyclization. Another precyclic peptide,
peptide 8, Ac[Arg6,NR-(ω-aminopropyl)Gly9]SP6-11, was
obtained by acetylation of the N-terminus of the linear
peptide before removal of the Boc protecting group. The
cyclizations were carried out on the resin using BOP
reagent. The time required for cyclization was ring size
dependent. For example for the cyclization of peptide
1 (ring size ) 17 atoms), the cyclization step was
repeated six times to get almost negative Kaiser test (6
days). Cyclization times were significantly shorters3
days for peptides 2-4scoming to a few hours in the
cases of peptides 5 and 6. The quality of the crude
products (after cleavage from the resin) depended
inversely on the cyclization time. Shorter cyclization
times gave crude peptides of higher quality.
It was previously suggested25 that conformational
factors dictated the speed and extent of cyclization. The
difficulty in the cyclization step encountered in peptide
1 was probably due to intensified steric constraint
exerted on the peptide upon cyclization. Moreover, the
extension of cyclization time enhanced side reactions
such as dimerization, partial cleavage from the resin,
partial racemization, etc. (no data). After cyclization,
the peptides were cleaved from the resin by HF and
purified by RP-HPLC. Characterization of the purified
cyclic and precyclic peptides was performed by fast atom
bombardment (FAB) MS, amino acid analysis, and, in
some cases, NMR. FAB MS/MS sequencing was used
to unequivocally establish the structure of the cyclic
peptides. Peptides 3-5 and 7 were fully characterized
by NMR.26,27
Biologica l Activity a n d Meta bolic Sta bility. The
biological activity of the cyclic and precyclic peptides was
assayed on three different smooth muscle prepara-
tions: guinea pig ileum (GPI), rat vas deferens (RVD),
and rat portal vein (RPV). Assay of the activity of the
different analogs in these systems provided quantitative
determination of their selectivity toward the three
neurokinin receptors (NK-1 and NK-3 receptors in GPI,
NK-2 in RVD, and NK-3 in RPV). Blocking the action
of acetylcholine, which is released by the neuronal NK-3
receptor, by atropine provided a selective GPI assay for
the NK-1 receptor. The resistance of the peptide
analogs to digestion by proteases was studied by incu-
bating the peptides with slices or homogenates of liver
and parotid gland and measuring the residual activity
after incubation by the GPI assay.
full selectivity to the NK-1 receptor were exhibited by
the 20-membered ring analog 4 [EC50 (nM): NK-1, 5;
NK-2, >200 000; NK-3, >10 000]. This particular cyclic
analog, which we have termed cycloseptide, had com-
parable activity and selectivity to the linear parent
peptide WS-septide. Peptide 1, which had a smaller
ring, was 3 orders of magnitude less active compared
to cycloseptide. We attributed this loss of activity to
enhanced conformational constraint which distorted the
Arg6-Phe7-Phe8 region. The decrease in selectivity of
the larger rings containing analogs 5 and 6 (21 and 22
atoms, respectively) was attributed to ring flexibility
which allowed activation of the NK-3 receptor.
In order to demonstrate that the high biological
activity and selectivity of cycloseptide indeed resulted
from cyclization, the activity and selectivity of two
precyclic analogs of cycloseptide (analogs 8 and 9) were
compared to that of cycloseptide (see Table 1). The low
activity and lack of selectivity of the precyclic linear
analogs proved the importance of cyclization in achiev-
ing the conformational restriction required for activity
and selectivity.
Cycloseptide also showed protracted activity in vari-
ous tissues. In Figure 3 the time course of the degrada-
tion of cycloseptide is compared to that of WS-septide
in liver (Figure 3, left) and parotid gland slices (Figure
3, right). The linear WS-septide was metabolically
unstable, and its activity was lost already after a few
minutes (half-life in liver slices was 4 min and in parotid
slices 6 min). Cycloseptide retained about 80% of its
original activity even after 120 min of incubation with
parotid gland slices. A similar pattern of behavior could
be seen with liver slices: 50% of the biological activity
of cycloseptide was retained after 30 min of incubation
with liver slices, whereas WS-septide was deactivated
almost immediately.
The protracted activity of cycloseptide in parotid and
liver slices was attributed solely to cyclization. Both
cycloseptide and WS-septide contain N-alkylated amino
acids in position 9 [NR-(ω-aminoalkyl)Gly unit and Pro,
respectively]. In both analogs the amino and carboxy
terminals are blocked, and both have an Arg-Phe bond
susceptible to degradation by trypsin or trypsin-like
enzymes. Nevertheless, cycloseptide was found much
more stable to degradation the WS-septide.
Str u ctu r e-Activity Rela tion sh ip s Stu d ies. Tak-
ing into account the biological activity and selectivity
of the resulting backbone cyclic peptides 1-6, we
selected the highly active and selective NK-1 agonist
cycloseptide and synthesized a closely related cyclic
peptide in which the chain -(CH2)3-NH-CO-(CH2)4-CO-
Arg- in cycloseptide was replaced by the chain -(CH2)2-
NH-CO-(CH2)2-CO-NH-CH2-CO-Arg- in peptide 7 (see
The activity and selectivity of peptides 1-6 (Table 1)
depended greatly on the ring size. Optimal activity and

N-Backbone Cyclic Analogs of SP
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 16 3177
Ta ble 2. Physical Data of Cyclic and Precyclic Analogs 1-9
amino acid analysis
peptide
m
n
ring size
FAB MS (calcd MW)
Arg
Phe
Leu
Met
Gly unit
Gly
1.1
1
2
3
4
5
6
2
2
3
4
2
3
2
2
3
3
3
6
6
2
3
3
17
18
19
20
21
22
20
894 (894.1)
908 (908.1)
923 (922.2)
937 (936.2)
951 (950.2)
964 (964.2)
951 (951.2)
856 (855.5)
955 (955.3)
1.0
0.9
1.0
1.0
0.9
0.9
0.7
0.8
0.9
2.0
2.0
2.0
2.0
2.0
2.0
1.8
2.0
1.8
1.0
1.0
1.0
1.0
1.0
1.0
1.0
2.0
1.0
0.9
1.0
1.0
0.9
1.0
1.0
0.9
0.9
0.9
1.0
0.9
1.1
1.0
1.0
1.0
1.1
1.1
1.0
7
8a
9a
4
a
Precyclic analog.
The Fmco-NR-[ω-(Boc-amino)alkyl]Gly-OH building units were
Table 1). Both analogs had the same sequence, the
same location of the ring, and the same ring size (20
atoms). The two analogs differed in the number of
methylene groups vs amide bonds in the ring: Analog
7 had five methylene groups and six amide bonds,
whereas cycloseptide had seven methylene groups and
five amide bonds. Peptide 7 was found to be completely
inactive toward all of the neurokinin receptors (Table
1).
We assumed that the loss of activity in peptide 7 was
due to excessive conformational constraint: The back-
bone of the ring of peptide 7 adopted a conformation
which precluded the bioactive conformation essential for
the binding and activation of any of the neurokinin
receptors. This assumption was validated by detailed
NMR studies and MD calculations of analogs 3-5 and
7.27 Comparison of the conformations of the active
analogs 4 and 5 to that of the inactive analog 7 revealed
that while the ring conformations of the active analogs
were very similar, they differed considerably from that
of the inactive analog 7.27 These results prompted us
to propose the bioactive conformation for the backbone
of the sequence -Arg-Phe-Phe- in cycloseptide.
synthesized according to procedures previously described.23,24
FAB MS was performed on a ZAB-3HF FAB/tandem mass
spectrometer or on an API-III LC/MS/MS instrument. Amino
acid analysis was performed on LKB-4400 apparatus equipped
with an SP-4100 computing integrator and a double channel
detector at 440 and 570 nm. Amino acids were detected by
ninhydrin after postcolumn derivatization. Hydrolysis of the
peptides: 1 mg of the peptide was heated at 110 °C in the
presence of 6 N HCl in a sealed tube for 24 h. After the
hydrolysis the samples were dried over KOH and dissolved in
sodium citrate buffer, 3.3 M (pH ) 2.2).
Peptides were synthesized by the manual solid phase
method using MBHA resin (1% cross-linked 80-180 mesh, 0.9
mequiv/g substitution level) obtained from Peninsula, U.K.
They were cleaved from the resin with dry HF in a Peptide
Institute, Inc. (J apan) apparatus Type I. All peptides were
purified by semipreparative reverse phase HPLC on a Merck-
Hitachi 655A apparatus equipped with an LC-5000 gradient
pump and a UV-vis detector set at 220 nm. The columns for
the analytical HPLC were Lichrospher RP-8 and RP-18 from
Merck (5 × 250 mm) and for the semipreparative purifications
HIBAR RP-8 and RP-18 (9 × 250 mm). The purity of the
peptides was checked by analytical HPLC using MeCN (0.085%
TFA)/TDW (0.1% TFA) gradients. Final products were ob-
tained as lyophilisates. Based on the amount of amine on the
resin, the yields for the cyclic analogs ranged between 10%
and 40%.
Con clu sion s
Gen er a l P r oced u r es. 4-Methylbenzhydrylamine resin
was used for synthesis of all peptides. Coupling of Boc- or
Fmoc-amino acids was performed using 6-fold excess of the
protected amino acid, 6-fold excess of BOP, and 12-fold excess
of diisopropylethylamine (DIEA) in DMF for 2 h, after preac-
tivation of the protected amino acid and the coupling reagent
for 5 min. After coupling of the first amino acid to the resin,
the residual free amines were capped using 10-fold excess of
acetic anhydride, 10-fold excess of DIEA, and 1 equiv of
4-(dimethylamino)pyridine (DMAP). Fmoc was deprotected
with 20% piperidine in DMF for 30 min. Boc was deprotected
with 55% TFA in DCM for 2 min and then for an additional
30 min followed by neutralization with 5% DIEA in DMF.
Acylation with succinic and glutaric anhydrides was performed
in DMF using 6-fold excess of anhydride and 1 equiv of DMAP.
Adipic acid anhydride was prepared by preactivation of 6 equiv
of the dicarboxylic acid for 30 min with 6 equiv of diisopropyl-
carbodiimide (DIC) in DMF and then addition of this mixture
to the peptide-resin together with 6 equiv of DIEA and 1 equiv
of DMAP.
Cyclization was performed by repeated reaction cycles with
6-fold excess of BOP and DIEA in DMF. Peptides were
cleaved from the resin and completely deprotected by treat-
ment with HF. The reaction was carried out for 90 min at -5
°C with 20 mL of HF/g of resin in the presence of 1 mL of
anisole/g of resin. After evaporation of the HF, the resin was
washed six times with hexane and six times with ether, and
the dried resin-peptide was extracted four times with 10 mL
of 30% acetic acid. Lyophilization of the combined acetic acid
extracts yielded the crude products which were purified as
described above. Physical data for peptides 1-9 are given in
Table 2.
In this article we demonstrated the feasibility of
backbone cyclization as a tool for generation of selective
and metabolically stable peptidomimetic analogs of
linear peptides, by the synthesis of potent cyclic analogs
of the C-terminal hexapeptide of SP. One of these
analogs, cycloseptide (peptide 4, Table 1), showed
compared activity and selectivity to the most active and
NK-1 selective linear analog, WS-septide. Moreover,
cycloseptide was metabolically stable, as indicated by
incubation with different tissue preparations, whereas
WS-septide was not.
Precyclic linear analogs of cycloseptide (peptides 8 and
9) showed low biological activity and selectivity, thus
suggesting that the high biological activity and selectiv-
ity of cycloseptide resulted directly from the cyclization
of the peptide. Comparative studies of cycloseptide and
its inactive analog 7 demonstrated that the ring con-
formations of the active and nonactive analogs were
different and that the backbone in the ring portion of
cycloseptide was stabilized in the bioactive conformation
essential for the binding and activation of the NK-1
receptor, or very close to it.
Exp er im en ta l Section
Equ ip m en t a n d Ma ter ia ls. Fmoc-amino acids were pur-
chased from Chemical Dynamics Corp. Boc-amino acids were
purchased from Nova Biochem, Switzerland. BOP reagent
was purchased from Richelieu Biotechnologies Inc., Canada.

3178 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 16
Byk et al.
(12) Neubert, K.; Bergmann, J .; Mansfield, J .; Hartrodt, B.; Barth,
A.; J akubke, H. D. Synthesis of Cyclic and Cyclically Branched
Tachykinin Partial Sequences. 1. Synthesis of the Homodet-
icCyclic Eledoisin6-11 Hexapeptide. Pharmazie 1985, 40, 456-
459.
(13) Neubert, K.; Hartrodt, B.; Mehlis, B.; Ruger, M.; Bergmann, J .;
J akubke, H. D.; Barth, A. Synthesis of Cyclic and Cyclically
Branched Tachykinin Partial Sequences. 2. Synthesis of the
HomodeticCyclic Substance P6-11 Hexapeptide. Pharmazie 1985,
40, 532-535.
(14) Chassaing, G.; Lavielle, S.; Ploux, O.; J ulien S.; Convert, O.;
Marquet, A.; Beaujouan, J . C.; Torrens, Y.; Glowinsky, J .
Biological and Conformational Studies of Cyclic Analogs of
Substance P. In Peptides 1984; Ragnarsson, U., Ed.; Almquist
& Wiksell: Stockholm, 1985; pp 345-348.
(15) Sandberg, B. E. B.; Bishai, W. R.; Hanna, P. Synthesis of Cyclic
Analogs Based on Substance P. In Peptides 1984; Ragnarson,
U., Ed.; Almquist & Wiksell: Stockholm, 1985; pp 369-372.
(16) Theodoropoulos, D.; Poulos, C.; Gatos, D.; Cordopatis, D.; Escher,
E.; Mizrahi, J .; Regoli, D.; Dalietos, D.; Furst, A.; Lee, T. D.
Conformationally Restricted C-Terminal Peptides of Substance
P. Synthesis, Mass Spectral Analysis and Pharmacological
Properties. J . Med. Chem. 1985, 28, 1536-1539.
(17) Darman, P. S.; Landis, G. C.; Smits, J . R.; Hirning, L. D.; Gulya,
K.; Yamamura, H. I.; Burks, T. F.; Hruby, V. J . Conformationally
Restricted Cyclic Analogues of Substance P: Insight into the
Receptor Binding Process. Biochem. Biophys. Res. Commun.
1985, 127, 656-662.
(18) Mutulis, F. K.; Mutule, I. E.; Maurops, G. H.; Sekacis, I. P.;
Grigoyeva, V. D.; Kukaine, E. M.; Golubeva, V. V.; Myshliakova,
N. V.; Klusha, V. E.; Chipens, G. I. Substance P - a New Type of
Cyclic Analogs. Bioorg. Khim. 1985, 11, 1276-1278.
(19) Levian-Teitelbaum, D.; Kolodny, N.; Chorev, M.; Selinger, Z.;
Gilon, C. 1H-NMR studies on receptor-selective Substance P
analogs reveal distinct predominant conformations in DMSO-
d6. Biopolymers 1989, 28, 51-64.
(20) Gilon, C.; Halle, D.; Chorev, M.; Selinger, Z.; Byk, G. Backbone
Cyclization: A New Method for Conferring Conformational
Constraint on Peptides. Biopolymers 1991, 31, 745-750.
(21) Gilon, C.; Halle, D.; Chorev, M.; Selinger, Z.; Byk, G. SAR Studies
of Cycloseptide: Effects of Cyclization and Charge at Position
6. In Peptides 1992; Smith, J . A., Rivier, J ., Eds.; ESCOM
Science BV: The Netherlands, 1993; pp 476-477.
(22) Gilon, C.; Zeltser, I.; Rashti-Behar, V.; Muller, D.; Bitan, G.;
Halle, D.; Bar-Akiva, G.; Selinger, Z.; Byk, G. Backbone cycliza-
tion as a tool for imposing conformational constraint on peptides.
In Peptide Chemistry 1992; Yanaihara, N., Ed.; ESCOM Science
BV: The Netherlands, 1993; pp 482-485.
Biologica l Assa ys. Guinea pig ileum (GPI) assay was
performed according to the procedures described by Wormser
et al.3 Rat vas deferens (RVD) assay was performed according
to the procedures described by Chorev et al.28 Rat portal vein
(RPV): Rat was sacrificed by decapitation. The abdomen was
cut open, and all the internal organs were moved to the side.
The portal vein was tied at both ends within the animal and
cleaned from all surrounding tissues. The cleaned tissue was
immersed in a bath containing Tirode solution, aerated with
a mixture of CO2:O2 (5:95). One of the tied ends was attached
to a glass hook and the other end to a transducer lever, in
order to measure the contraction. Tension was about 0.5 g.
The tissue was left in the bath at 37 °C for 1 h; then the
peptides were added at 20 min intervals, to prevent desensi-
tization of the receptor. Resistance to digestion by proteases
assay was performed according to the procedures described
by Chorev et al.28
Abbr evia tion s: Fmoc, 9-fluorenylmethoycarbonyl; BOP,
benzotriazol-1-yloxytris(dimethylamino)phosphonium hexaflu-
orophosphate; DIEA, diisopropylethylamine; DMAP, 4-(di-
methylamino)pyridine; DMSO, dimethyl sulfoxide; TFA, tri-
fluoroacetic acids; SPPS, solid phase peptide synthesis.
Ack n ow led gm en t. The authors are grateful to
Peptor Co. Rehovot, the GIF, and the Center for Pain,
The Hebrew University, J erusalem, for their financial
support. The authors are also grateful to Dr. J . W.
Metzger and Prof. G. J ung from the Eberhard-Karls
Universita¨t, Tu¨bingen, Germany, for performing the
API LC/MS/MS determinations and to Dr. K. Eckart
and Prof. H. Schwartz from the Technical University,
Berlin, Germany, for performing the FAB MS measure-
ments.
Refer en ces
(1) Regoli, D.; Boudon, A.; Fouchere, J .-L. Receptors and Antagonists
for Substance P and Related Peptides. Pharmacological Rev.
1994, 46, 551-599.
(2) Laufer, R.; Wormser, U.; Friedman, Z. Y.; Gilon, C.; Chorev, M.;
Selinger, Z. Neurokinin B is a preferred agonist for neuronal
Substance P receptor and its action is antagonized by enkepha-
lin. Proc. Natl. Acad. Sci. U.S.A. 1985, 82, 7444-7448.
(3) Wormser, U.; Laufer, R.; Hart, Y.; Chorev, M.; Gilon, C.; Selinger,
Z. Highly selective agonists for Substance P receptor subtypes.
EMBO J . 1986, 5, 2805-2808.
(4) Laufer, R.; Gilon, C.; Chorev, M.; Selinger, Z. Characterization
of a neurokinin B receptor site in rat brain using a highly
selective radioligand. J . Biol. Chem. 1986, 261, 10257-10263.
(5) Laufer, R.; Gilon, C.; Chorev, M.; Selinger, Z. Desensitization
with a Selective Agonist Discriminates between Multiple Ta-
chykinin Receptores. Pharmacol. Exp. Ther. 1986, 245, 639-
643.
(23) Byk, G.; Gilon, C. Building Units for N-Backbone Cyclic Peptides.
1. Synthesis of Protected N-(ω-Aminoalkylene)amino Acids and
Their Incorporation into Dipeptide Units. J . Org. Chem. 1992,
57, 5687-5691.
(24) Muller, D.; Zeltser, I.; Bitan, G.; Gilon, C. Building Units for
N-Backbone Cyclic Peptides. 3. Synthesis of Protected N-(ω-
Aminoalkylene)amino Acids and N-(ω-Carboxylkylene)amino
Acids. Tetrahedron, submitted for publication.
(25) Schiller, P. W.; Nguyen, T. M.-D.; Lamieur, C. Synthesis and
OpioidActivity of Cyclic Dynorphin Analogs. Tetrahedron 1988,
44, 733-743.
(26) Saulitis, J .; Mierke, D. F.; Byk, G.; Gilon, C.; Kessler, H.
Conformation of Cyclic Analogues of Substance P: NMR and
Molecular Dynamics in Dimethyl Sulfoxide. J . Am. Chem. Soc.
1992, 114, 4818-4827.
(6) Halle, D. Ph.D. Thesis, submitted to the Senate of the Hebrew
University, October 1993.
(7) Manavalan, P.; Momany, F. A. Conformational Energy Calcula-
tions on Substance P. Int. J . Pept. Protein Res. 1982, 20, 351-
356.
(27) Golic-Grdadolnik, S.; Mierke, D.; Byk, G.; Zeltser, I.; Gilon, C.;
Kessler, H. Comparison of the Conformation of Active and
Nonactive Backbone Cyclic Analogs of Substance P as a Tool to
Elucidate Features of the Bioactive Conformatio: NMR and
Molecular Dynamics in DMSO and Water. J . Med. Chem. 1994,
37, 2145-2152.
(28) Chorev, M.; Rubini, M.; Hart, Y.; Gilon, C.; Wormser, U.;
Selinger, Z. Metabolically Stable Analogues of Substance P:
Persistant Action of Partially Modified Retro-Inverso Analogues
of Substance P on Rat Parotid and Hypothalamic Slices. Eur.
J . Pharmacol. 1986, 127, 187-195.
(8) Chassaing, G.; Convert, O.; Laville, S. Preferencial Conformation
of Substance P in Solution. Eur. J . Biochem. 1986, 154, 77-
85.
(9) Young, J . K.; Anklin, C.; Hicks, R. P. NMR and Molecular
Modeling Investigations of the Neuropeptide Substance P in the
Presence of 15 mM Sodium Dodecyl Sulfate Micelles. Biopoly-
mers 1994, 34, 1449-1462.
(10) Williams, B. J .; Curtis, N. R.; McKnight, A. T.; Maguire, J .;
Foster, A.; Tridgett, R. Development of NK-2 Selective Antago-
nists. Regul. Pept. 1988, 21, 189P.
(11) Maggi, C. A.; Quartara, L.; Giuliani, S.; Patacchini, R. Tachy-
kinin NK-2 Receptor Antagonists. Drugs Future 1993, 18, 155-
158.
J M960154I