Design, synthesis and pharmacological characterization of a potent radioiodinated and photoactivatable peptidic oxytocin antagonist

Article abstract of DOI:10.1021/jm010125u

Using a segment strategy, we have synthesized four iodinated photoactivatable cyclic peptidic ligands of oxytocin, bearing a β-mercapto-ββ-cyclopentamethylene propionic group (Pmp) on their N-terminus. All the syntheses were RP-HPLC monitored, and the com

Full text of DOI:10.1021/jm010125u

3022
J . Med. Chem. 2001, 44, 3022-3030
Design , Syn th esis a n d P h a r m a cologica l Ch a r a cter iza tion of a P oten t
Ra d ioiod in a ted a n d P h otoa ctiva ta ble P ep tid ic Oxytocin An ta gon ist∞
Eric Carnazzi,*,†,‡ Andre´ Aumelas,§ Bernard Mouillac,|| Christophe Breton,||, Laurent Guillou,†
Claude Barberis,|| and Rene´ Seyer†
CNRS UPR 9023 and INSERM U 469, CCIPE, 141, rue de la Cardonille, 34094 Montpellier Cedex 5, France, and
CNRS UMR C 995 and INSERM U 414, CBS, Faculte´ de Pharmacie, Montpellier, France
Received March 19, 2001
Using a segment strategy, we have synthesized four iodinated photoactivatable cyclic peptidic
ligands of oxytocin, bearing a â-mercapto-ââ-cyclopentamethylene propionic group (Pmp) on
their N-terminus. All the syntheses were RP-HPLC monitored, and the compounds were HPLC
purified. They were characterized by 1H NMR, MALDI-TOF, or FAB mass spectrometries. The
affinities of Pmp-Tyr (Me)-Ile-Thr-Asn-Cys-Gly-Orn-Phe(3I,4N3)-NH2 (20), Pmp-Tyr -Ile-Thr-
Asn-Cys-Gly-Orn-Phe(3I,4N3)-NH2 (21), Pmp-Tyr (Me)-Ile-Thr-Asn-Cys-P r o-Orn-Phe(3I,4N3)-
NH2 (22), and Pmp-Tyr -Ile-Thr-Asn-Cys-P r o-Orn-Phe(3I,4N3)-NH2 (23) were evaluated as
inhibition constants (Ki, in nM) for the human oxytocin receptor expressed in Chinese hamster
ovary cells by displacement of a radioiodinated disulfide-cyclized antagonist (Elands et al. Eur.
J . Pharmacol. 1987, 147, 197-207). The most potent of them, compound 22, was synthesized
by another method in order to allow its radiolabeling by 125I. Its dissociation constant (Kd) for
the human oxytocin receptor, directly measured in saturation studies, was 0.25 ( 0.04 nM,
and its antagonist properties were determined by inactivation of phospholipase C, thus obtaining
an inactivation constant (Kinact) of 0.18 ( 0.02 nM, evaluated by inositol phosphate accumulation.
This compound is a very good tool for the mapping of peptidic antagonist binding sites in the
human oxytocin receptor.
In tr od u ction
OT acts through receptors that belong to the G-protein-
coupled receptor, characterized by seven putative trans-
membrane domains.2 In the previous decade, several OT
receptors have been cloned from various mammalian
species.3-5 Functionally coupled to phospholipase C,
oxytocin receptors6 share a high degree of peptide
sequence homology with the different vasopressin re-
ceptor subtypes. As well, the peptide sequence of OT is
structurally very close to that of vasopressin, and
numerous analogues have been described to date for
both neuropeptides.7 On careful and meticulous survey
of the literature,8 we considered that the V1a vasopressin
receptor subtype could serve as a good model for the
study of structure-activity relationships. Furthermore,
because OT and vasopressin receptors belong to the
same family, we thought that the structural information
obtained for the V1a vasopressin receptor subtype could
be very useful for the study and understanding of
ligand-receptor interactions of the OT receptor itself.
Therefore, we were interested in designing and synthe-
sizing a series of OT antagonists because of their
potential therapeutic interest, for instance, in the treat-
ment of premature labor.9-11
Site-directed mutagenesis studies12 and construction
of chimeric OT receptor have been performed;13 how-
ever, limited information was available regarding the
OT receptor binding domains. A complementary ap-
proach is the use of photoactivatable analogues. On the
basis of the formation of a covalent bond between the
ligand and its receptor upon UV irradiation, this method
allows the mapping of membrane proteins, particularly
under denaturing conditions.14 To date, several ligands
for the photolabeling of OT receptor have been re-
Oxytocin (OT) is a neurohypophyseal nonapeptide
that regulates several physiological functions. It stimu-
lates the contraction of uterine and mammary gland
muscles during parturition and lactation, respectively.1
∞ Abbreviations: Abbreviations used are in accordance to the
IUPAC-IUB Commission (Eur. J . Biochem. 1984, 138, 9-37). The
following abbreviations were also used: BHA resin, benzhydrylamine
resin; Boc, tert-butyloxycarbonyl; BOP, (1H-1,2,3-benzotriazol-1-yl-oxy)-
tris(dimethylamino)phosphonium hexafluorophosphate; BSA, bovine
serum albumine; CHO, Chinese hamster ovary cells; COSY, correlated
spectroscopy; Dcb, dichlorobenzyl; DFQ-COSY, double-quantum-
filtered correlated spectroscopy; DMF, N,N-dimethylformamide; EDTA,
ethylenediamine tetraacetate; FAB, fast atom bombardment; HPLC,
high-performance liquid chromatography; (i-Pr)2EtN, diisopropylethyl-
amine; [125I]-OTA, Pmp-Tyr(Me)-Ile-Thr-Asn-Cys-Pro-Orn-Tyr(3125I)-
NH2; Kact, activation constant; Kinact, inactivation constant; MALDI-
TOF, matrix-assisted laser desorption ionization time-of-flight; MBHA
resin, p-methylbenzhydrylamine resin; mol equiv, molar equivalent;
MS, mass spectroscopy; NMR, nuclear magnetic resonance spectros-
copy; NOESY, nuclear Overhauser effect spectroscopy; Orn, ornithine;
OT, oxytocin, Cys-Tyr-Ile-Gln-Asn-Cys-Pro-Leu-Gly-NH2; OTA, oxy-
tocin antagonist, Pmp-Tyr(Me)-Ile-Thr-Asn-Cys-Pro-Orn-Tyr-NH2; PBS,
phosphate buffered saline; PAGE, polyacrylamide gel electrophoresis;
Pmp, â-mercapto-ââ-cyclopentamethylene propionic acid; PyBOP, (ben-
zotriazolyloxy)tripyrrolidinophosphonium hexafluorophosphate; SDS,
sodium dodecyl sulfate; TOCSY, total correlation spectroscopy; Tris,
tris(hydroxymethyl)aminomethane; Xan, xanthyl; Z, benzyloxycar-
bonyl; ZOTA, azido oxytocin antagonist (18).
* To whom correspondence should be addressed. Phone:
(33) 4 67 54 86 59. Fax: (33) 4 67 54 86 35. E-mail: carnazzi@
colombes.pharma.univ-montp1.fr.
† CNRS UPR 9023.
‡ Present address: Laboratoire des Aminoacides Peptides et Pro-
te´ines, UMR5810-CNRS, Universite´ Montpellier I et II, Faculte´ de
Pharmacie, 15, av. Ch. Flahaut, 34060 Montpellier Cedex, France.
§ CNRS UMR C 995 and INSERM U 414.
|| INSERM U 469.
Present address: Laboratoire d’Endocrinologie des Anne´lides,
UPRESA CNRS 8017, SN3, Universite´ des Sciences et Technologies
de Lille, 59655 Villeneuve d’Ascq Cedex, France.
10.1021/jm010125u CCC: $20.00 © 2001 American Chemical Society
Published on Web 08/02/2001

A Peptidic Oxytocin Antagonist
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 18 3023
ported.15-19 However, they were mainly synthesized by
direct coupling of bulky aromatic groups on functional
amino acid side chains, and no direct identification of
receptor residues in close interaction with the probe has
been reported. Thus, we exploited OT analogues specif-
ically designed according to a different approach, con-
sisting of the minimal structural modification of the
ligand peptide sequence.20 We synthesized them using
a combination of solid phase and solution synthesis and
characterized them pharmacologically by measuring
their affinities on CHO cells, stably expressing the
human OT receptor. We selected the analogue from the
series displaying the highest affinity and subsequently
developed a synthesis strategy for radiolabeling. The
structure of the nonradioactive iodinated analogue was
Sch em e 1. Synthesis of Analogues 20-23 for
Preliminary Affinity Determinationa
1
ascertained by H NMR spectrometry, and its pharma-
cological properties were determined using saturation
binding experiments and also measure of inositol phos-
phate accumulation.
Resu lts
P r obe Design . It was generally observed that struc-
tural modifications of the cyclic part of the OT analogue
family often resulted in a loss of both affinity and
activity of the ligands while chemical modifications of
the C-terminal section were better tolerated.21 Accord-
ingly, we designed photoactivatable OT analogues on
the basis of the following: (i) the replacement of the
N-terminal Cys residue by the Pmp group22 increases
both the affinity and the antagonist potency of the
analogues, (ii) the substitution of the Gln4 and Pro7
residues for Thr4 and Gly7 residues, respectively, results
in an increased selectivity of the ligand for the OT
receptor,23 and (iii) O-alkylation of the Tyr2 residue also
increases the antagonist potency of the analogues.24,25
Thus, we selected, as lead compounds, antagonist Pmp-
Tyr(Me)-Ile-Thr-Asn-Cys-Pro-Orn-Tyr(3125I)-NH2 ([125I]-
OTA), on the basis of its maximum affinity, and
analogue Cys-Tyr-Ile-Thr-Asn-Cys-Gly-Leu-Gly-NH2, on
the basis of its maximum selectivity for the OT recep-
tor.23,26,27 To maximally conserve the structure of [I]-
OTA, we selected the azido group to replace the p-hy-
droxyl group of the C-terminal Tyr9 residue,28 resulting
in residue Phe(4N3)9. We also explored two positions by
introducing either a Tyr or a Tyr(Me) residue in position
2 and a Gly or a Pro residue in position 7 (Scheme 1).
We finally chose to introduce the iodine atom on the
same aromatic ring as the azido group (Phe(3I,4N3)9,
keeping in view that the close proximity of the 125I tracer
to the covalent bond established on photoactivation
would lower the risk of losing signal detection owing to
enzymatic cleavage of the ligand. Though it has been
suggested that partial deiodination might occur during
the irradiation step due to intramolecular rearrange-
ments of the highly reactive nitrene intermediate,29,30
many results were reported with such a configura-
tion,31-33 and several reagents offering such structural
characteristic are currently available commercially (see
Pierce catalog).
a
Right part of the scheme: solution synthesis of dipeptide 7.
Left part of the scheme: solid-phase synthesis of heptapeptides
8-11 and disulfide-cyclization giving heptapeptides 12-15. Lower
part of the scheme: coupling of heptapeptides 12-15 to dipeptide
7 giving compounds 16-19, followed by their Boc deprotection
giving analogues 20-23.
during the synthesis of the azido group and enable the
reaction with the Tyr2 residue, and (iii) perform the
selective iodination of the Phe9 residue at the Phe(4NH2)9
intermediate stage without any unexpected reaction at
position 2 (Tyr2-containing analogues). Accordingly, we
selected a segment-coupling strategy taking advantage
of the presence of the nonracemizable Gly or Pro
residues at position 7. We combined solid-phase and
solution peptide synthesis to obtain, respectively, the
N-terminal heptapeptide segments (residues 1-7) and
the common C-terminal dipeptide 7 (residues 8-9) for
the final coupling.
P r obes Syn th esis. All heptapeptide fragments (com-
pounds 12-15) were synthesized on a chloromethylated
resin, using Boc as protecting group and PyBOP as
coupling reagent;34 HF deprotected; cyclized by air-
oxidation (Scheme 1); and subsequently purified by RP-
HPLC. Dipeptide 7 was synthesized in solution accord-
ing to the synthetic route described in Scheme 1. The
reactions of cyclization, coupling, hydrogenation, iodina-
tion, and azidation were monitored by HPLC, using a
dual wavelength detection (214 and 254 nm).35 The
second UV detector was set at 254 nm to take advantage
of the high absorbance properties of the aryldiazonium
and arylazido groups at that wavelength. Indeed, the
To synthesize analogues 20-23, the general structure
of which is given in Scheme 1, we had to (i) conciliate
the catalytic hydrogenation of a Phe(4NO2) precursor
residue and the presence of a disulfide bridge, (ii) control
the reactivity of the diazonium intermediate produced

3024 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 18
Carnazzi et al.
presence of these chromophores in the different peptide
intermediates always resulted in an important decrease
of the corresponding ꢀ214/ꢀ254 ratio, which was particu-
larly helpful for peptide identification during HPLC
monitoring. In a similar way, the disulfide cyclization
of compounds 12-15 to compounds 16-19 resulted in
a 2-fold decrease of the ꢀ214/ꢀ254 ratio in each case. This
observation, along with the increased polarity of the
cyclized peptides (1-3% in CH3CN elution difference),
facilitated the identification of the synthesized com-
pounds, subsequenly confirmed by Ellman’s reagent and
MALDI-TOF MS. Compounds 20-23 were obtained by
coupling each heptapeptide 12-15 to the dipeptide 7,
followed by deprotection of the Orn(Boc)8 residue and
RP-HPLC purification. The removal of the Boc group
resulted in an important shift in the HPLC elution,
approximately 7% lower CH3CN as compared to the Boc
precursor, but no change of the ꢀ214/ꢀ254 ratios was
observed. The purities of compounds 20-23 were as-
sessed by HPLC and their structures were confirmed
by FAB-MS. It is also noticeable that O-methylation of
the phenol group of Tyr2 resulted in a 6% CH3CN
increase of elution and that all Pro7-containing peptides
and intermediates elute 1-2% higher than the corre-
sponding Gly7-containing peptides. According to this
synthetic route, compound 18 was always obtained as
the major product of a mixture of three compounds and
was particularly difficult to purify because of very close
HPLC elution percent at 55, 55.7, and 56% CH3CN,
respectively. This observation, together with the need
for a convenient radiolabeling of the compounds,
prompted us to design another synthesis strategy.
F igu r e 1. Concentration-dependent binding experiments of
analogues 20-23 to human OT receptor stably expressed in
CHO cells. Membrane preparations (2-3 µg) were incubated
for 1 h at 30 °C in darkness with increasing amounts of each
peptide (10 pM to 10 µM) and 200-220 pM of [125I]-OTA.
Values of specific binding, measured in the presence of
unlabeled ligands, are expressed as the percentage of the total
binding measured in the absence of competitor. The apparent
dissociation constants (Ki) of the nonradioactive ligands were
calculated from the relation Ki ) IC50/(1 + [L]/Kd), where IC50
is the concentration of analogue resulting in 50% inhibition
of total binding and [L] and Kd are the concentration and the
dissociation constant of [125I]-OTA, respectively, with Kd ) 0.18
nM. Results are the mean of triplicate values and are
representative of, at least, three independent experiments.
Ki ≈ 2.3 and 18 nM respectively) and (ii) for each Pro7-
or Gly7-containing analogue, O-alkylation of the Tyr2
residue with a methyl group resulted in a slight increase
in affinity (Ki shifting from 18 to 2.3 nM for analogues
21 and 20 and from 0.6 to 0.3 nM for analogues 23 and
22). Ligand 22, whose structure differs from antagonist
[I]-OTA only by the presence of an azido group in place
on the hydroxyl group, showed the best affinity for the
human OT receptor expressed in CHO cells. On the
basis of these findings, we selected compound 22 for
radioiodination (becoming ligand 29), to determine its
ability to interact directly with specific receptors, em-
ploying saturation experiments.
Syn th esis of th e Ra d ioa ctive Liga n d 29 ([125I]-
ZOTA). To synthesize ligand 29, we developed a dif-
ferent schedule in order to reduce the handling of
radioactive intermediate compounds that would have
been generated at several steps of the segment coupling
strategy. Indeed, since the iodine atom may only be
introduced at the Phe(4NH2) stage, because of the
structural delocalization of the π-electrons on the meta
position of the aromatic ring (Figure 2), the following
reactions of azidation, segment coupling, and Boc depro-
tection would have to be performed on radioactive
compounds. Therefore, we designed a strategy consist-
ing of the solid-phase synthesis of nonapeptide 25, with
the direct incorporation of a Phe(4NHZ) residue and its
subsequent iodination and complete azidation.37 Peptide
25 has been synthesized on a p-methylbenzhydrylamine
resin, using Boc as temporary protecting group, PyBOP
as coupling reagent, and a p-aminophenylalanine resi-
due with a benzyloxycarbonyle group as side chain
protection. The product was cyclized by air-oxidation
and RP-HPLC purified and its structure checked by
MALDI-TOF MS (Figure 2a).
NMR Stu d y. The chemical structure of peptide 22,
the analogue observed to be the most promising candi-
date for photolabeling experiments, was confirmed by
two-dimensional 1H NMR. All expected spin systems
were identified in the TOCSY experiment, and their
assignment was obtained by using the sequential pro-
cedure described by Wu¨thrich.36 The combination of the
COSY, TOCSY, and NOESY spectra allowed us to
unambiguously assign all expected spins systems as
well as aromatic signals of substituted rings of Tyr(Me)2
and Phe(3I,4N3)9. Indeed, in the case of Phe(3I,4N3)9,
due to the 3-iodo and 4-azido substitutions, the three-
signal system composed of two doublets (6-H and 5-H)
and a singlet (2-H) was clearly observed with the
TOCSY. Of the two doublets, the 6-H assignment was
performed on the basis of NOE with CâH2 protons. In a
similar way, the 2,6-H and 3,5-H aromatic signals of
the Tyr(Me)2 were assigned on the basis of their NOE
with the CâH2 and the methoxy group, respectively. The
observation of a strong NOE between the HR-proton of
Cys6 with the Hδ-Hδ protons of Pro7 is indicative of
the trans conformation for the Cys6/Pro7 amide bond.
Affin it ies of t h e Non r a d ioa ct ive Iod in a t ed Li-
ga n d s 20-23. The affinities of ligands 20-23 were
determined on membrane preparations from CHO cells
stably expressing the human OT receptor. We per-
formed concentration-dependent binding inhibitions us-
ing the specific radiolabeled antagonist [125I]-OTA as
radioligand23 (Figure 1). The data suggest that (i)
analogues containing a Pro residue in position 7 (22 and
23, Ki ≈ 0.3 and 0.6 nM respectively) display a better
affinity than those containing a Gly residue (20 and 21

A Peptidic Oxytocin Antagonist
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 18 3025
F igu r e 3. Scatchard representation of a saturation binding
experiment performed with radioiodinated analogue 29. Mem-
brane preparations of CHO cells stably expressing the human
OT receptor (2-3 µg) were incubated in the presence of
increasing concentrations of peptide 29 (10-900 pM) for 1 h
at 30 °C in darkness. Specific binding was expressed as the
difference between total and nonspecific binding determined
in the presence of 800 nM of the corresponding nonradioactive
iodinated analogue (22). The results are the mean of triplicate
values and are representative of three independent experi-
ments, giving a Kd value of 0.25 ( 0.04 nM.
oped procedures using either ICl or NaI/Iodo-Gen for
nonradioactive iodinations at a nanomolar scale (Figure
2b). Monoiodinated peptide 26 was obtained by an
equimolar reaction of ICl in MeOH with freshly dis-
solved peptide 25 (≈1 nmol), purified by RP-HPLC and
characterized by MALDI-TOF MS (Figure 2c). Com-
pound 26 (approximately 0.5 nmol) was then progres-
sively diazotized with an excess of NaNO2 in HCl at 0
°C, in darkness, to obtain a mixture of two compounds
that could not be identified by MALDI-TOF MS, prob-
ably because of their photosensitivity (Figure 2d, peaks
27, 28). Peak 28 might be the corresponding diazonium
intermediate since it disappeared progressively upon
addition of NaN3 in excess (2 mol equiv toward NaNO2)
at room temperature to give ligand 22 ([I]-ZOTA, Figure
2e), which was characterized by FAB-MS (Figure 2f).
Identical strategy was employed for synthesizing the
radioactive ligand 29 ([125I]-ZOTA), except that com-
pound 25 was iodinated with Na125I (1 mCi) in the pres-
ence of Iodo-Gen and HPLC purified for diazotization
and azidation. Ligand 29 was purified by HPLC and
stored in liquid nitrogen. The identity of radioactive
ligand 29 was determined by HPLC coinjection of
compounds 22 and 29 and measuring the radioactivity
associated with the 0.5-mL HPLC fractions.
P h a r m a cologica l Stu d ies. The Scatchard repre-
sentation of the saturation binding experiments of
ligand 29 is linear (Figure 3), indicating that this
compound interacts with a single class of receptors (Hill
coefficient ) 0.97). The determination of its affinity gave
a Kd value of 0.25 ( 0.04 nM, suggesting that the
replacement of the initial hydroxyl group by an azido
had nearly no influence on the affinity of the parent
ligand (Kd of [125I]-OTA ) 0.18 ( 0.04 nM).
To determine its biological activity, we measured the
ability of ligand 22 (the nonradioactive iodinated form
of 29) to inhibit OT-induced inositol phosphate ac-
cumulation in CHO cells, in a concentration-dependent
F igu r e 2. MS and HPLC monitoring of analogue 22 synthesis
according to the strategy designed for obtaining radioactive
analogue 29. Upper part of the figure: detailed sequence of
precursor 25. Right part of the figure: description of the “Phe8”
moiety at different steps of the synthesis, respectively, as inter-
mediate “Phe(4NH2)” (25), “Phe(3I,4NH2)” (26), “Phe(3I,4N2OH)”
(hypothetic intermediate 27), “Phe(3I,4N2+)” (28), and “Phe-
(3I,4N3)” (22). Lower part of the figure: photoactivated form
of analogue 22 represented as its highly reactive nitrene (short-
life intermediate 30). Panels a and c: MALDI-TOF mass
spectra of compound 25 and 26. Panel f: FAB mass spectra of
azidopeptide 22. The M + 1 molecular main peak (1305) is
always accompanied by an unidentified degradation product
resulting from bombardment. Panels b, d, and e: chromato-
grams illustrating the main steps for the synthesis of analogue
22. RP-HPLC monitoring was performed on a C18-column at
a flow rate of 2 mL/min in the linear CH3CN gradient mode
(dotted line as shown on panel b) using 0.1% CF3CO2H
acidified eluents. The detection was performed at two distinct
wavelengths (214 nm upward and 254 nm downward). Panel
b: The iodination of compound 25 by ICl in MeOH gives the
expected monoiodinated derivative 26 and traces of the more
hydrophobic diiodinated compound (third peak). Panel d: The
diazotization of compound 26 was performed at 0 °C in
darkness using an excess of NaNO2/HCl and results in the
formation of two products (27 and 28) having very different
absorbance properties (ꢀ214/ꢀ254). Panel e: Once compound 26
reacted completely, an excess of NaN3 was added and the
reaction mixture was allowed to reach room temperature to
give the hydrophobic azidopeptide 22.
The radiolabeling procedure of compound 25 was first
optimized using nonradioactive iodine (127I) at the usual
radioiodination concentration scale.38 Owing to 50% of
iodine incorporation, for a reaction with 1 mCi Na125
I
(specific activity of 2200 Ci/mmol), we, therefore, devel-

3026 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 18
Carnazzi et al.
F igu r e 4. Determination of the antagonist properties of
analogue 22. The ability of analogue 22 to inhibit OT-induced
inositol phosphate accumulation was determined in CHO cells
stably expressing the human OT receptor. myo-[2-3H]Inositol-
prelabeled- and PBS-equilibrated CHO cells were incubated
for 10 min in PBS supplemented with 10 mM LiCl, in the
absence or presence of increasing concentrations of ligand 22
(1 pM to 1 µM, all hatched open and gray-colored bars). Cells
were then incubated in the presence of 10 nM OT (all gray-
colored open and hatched bars) for an additional 15 min period,
and the reaction was stopped by addition of ice-cold perchloric
acid followed by neutralization of the samples. The radioactiv-
ity associated to extracted inositol phosphates was measured,
and the results were expressed as disintegrations per min/
well, with no correction for the basal activity (open bar).
Results are from a single experiment representative of three
independent experiments each performed in triplicate. The
apparent inactivation constant Kinact ) 0.18 ( 0.02 nM was
calculated from Kinact ) IC50/(1 + [OT]/Kact), where IC50 is the
concentration of analogue 22 yielding 50% inhibition of specific
total inositol phosphate accumulation, [OT] ) 10 nM, and Kact
is the concentration of OT inducing half-maximum accumula-
tion of inositol phosphates (Kact ) 12.9 nM for CHO cells
expressing the wild-type human OT receptor).
F igu r e 5. Photolabeling of the human oxytocin receptor.
Membrane preparations of CHO cells stably expressing the
OT receptor (500 µg, 4 mL) were incubated with freshly
radioiodinated analogue 29 (≈2 nM) in a Tris:HCl buffer (Tris
50 mM, MgCl2 5 mM, pH 7.4, 0.5 mg/mL BSA) with (left lane)
or without (right lane) an excess of 10-5 M oxytocin, for 1 h at
30 °C. Membranes were then washed twice using cold Tris:
HCl buffer (without BSA, 10 mL) to remove unbound ligand
29 and were resuspended in the same buffer for UV irradiation
(254 nm, 1 min) in a Petri dish, on ice. The membranes were
washed twice again and solubilized in Laemmli buffer, and
the proteins were separated by sodium dodecyl sulfate poly-
acrylamide gel electrophoresis (gel with 12% bisacrylamide)
as were molecular mass standards. Equivalent amounts of
proteins (25 µg) were loaded into each well. The gel was further
fixed, dried, and autoradiographed on Kodak film for ap-
proximately 10 h at -70 °C. Positions of protein standard
masses are indicated on the left part of the figure. The
autoradiogram of the dried gel is representative of at least
three distinct experiments.
manner.12,39 As shown in Figure 4, the ligand 22 was
able to inhibit the action of 10 nM OT, at a concentration
displaying half-maximal stimulation of inositol phos-
phate production, with a calculated Kinact of 0.18 ( 0.02
nM. In addition, the incubation of high concentrations
of ligand 22 on CHO cells (up to 1 µM) resulted in the
production of inositol phosphate levels comparable to
those of the basal activity, suggesting that this ligand
is a potent antagonist for the human OT receptor.
To determine whether analogue 22 ([I]-ZOTA) was
able to covalently and specifically bind to the human
OT receptor, membrane preparations of transfected
CHO cells were incubated with analogue 29 ([125I]-
ZOTA), the radioactive form of ligand 22. Then, mem-
branes were washed, irradiated at 254 nm, and sub-
jected to sodium dodecyl sulfate-polyacrylamide gel
electrophoresis (SDS-PAGE) for autoradiography. As
shown in Figure 5, the labeling (right lane) was specific
since it could be completely displaced when the incuba-
tion was performed with an excess of oxytocin (left lane).
Irradiation of the membranes resulted in the strong
labeling of a major protein band migrating at an
apparent molecular mass of 70-75 kDa, which is in
agreement with the expected molecular mass for the
native glycosylated human OT receptor. An additional
faint band at around 42 kDa was also specifically
labeled and could correspond to the molecular mass of
the proteic core of the human OT receptor, as deduced
from its cDNA sequence. This labeling was 254 nm-
dependent and easily observable, even under denaturing
conditions such as for SDS-PAGE, and no labeling at
all could be observed in the absence of UV irradiation
(data not shown).44
Discu ssion
A survey of the literature reveals that the photo-
activatable OT derivatives previously described bear the
photoreactive group and the 125I-tracer on separate
residues and were obtained by introduction of bulky
arylazido groups to functional amino acids side
chains.17-19,40 Though the affinity criterion is not con-
sidered to be crucial for photolabeling experiments, we
always obtained the best results using high-affinity
ligands, particularly because of the absence of non-
specific labeling.41 Therefore we selected, as lead com-
pounds, analogues having high affinity and high selec-
tivity for the OT receptors, and we designed new
photoreactive analogues in a manner so as to maintain

A Peptidic Oxytocin Antagonist
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 18 3027
as much as possible their close interactions with the
targeted receptor. Thus, we modified an aromatic resi-
due already present in the peptide sequence. We re-
placed the Tyr9 p-hydroxyl group of antagonist [I]-OTA
with an azido group, because it is activated in a nitrene
intermediate (Figure 2, active form 30)42 that, among
the known reactive groups, possesses structural similar-
ity to the hydroxyl function.28 This position of the azido
group should allow a more precise and representative
mapping of the receptor-binding site, as compared to
ligands previously reported,17-19,40 for which the photo-
reactive groups were more distant because of the
additional aromatic structures necessary for their in-
troduction. It has also been reported that the C-terminal
part of OT analogues was barely sensitive to chemical
modifications, suggesting that the last two residues
were probably not important for affinity. Accordingly,
we showed that the replacement of the hydroxyl group
with a hydrophobic azido group did not modify signifi-
cantly the affinity of the molecule (ligand 22, Kd ) 0.25
( 0.02 nM) compared to [I]-OTA (0.18 ( 0.04 nM).
The introduction of the photoreactive group on the
Phe9 residue offers the opportunity to further function-
alize these ligands. Indeed, the remaining free amino
group of the Orn8 side chain could serve as an anchor
point for the introduction of fluorophores, biotin, or other
groups, allowing the study of ligand-receptor interac-
tions by different biochemical approaches. These de-
rivatizations can be easily performed using the segment
coupling strategy, particularly adapted for the prepara-
tion, in high amounts, of various chemically modified
C-terminal dipeptides (8-9) in solution. We also devel-
oped in this work an experimental procedure allowing
the complete azidation of radioiodinated compounds,
suggesting that such syntheses can be performed rou-
tinely with efficiency and reproducibility, though sub-
nanomolar amounts of compounds are usually involved
in these reactions. Among the series of photoactivatable
OT ligands we designed and synthesized, analogue 22
displayed the highest affinity toward the human OT
receptor (stably expressed in CHO cells). It was able to
inhibit the OT-induced inositol phosphate accumulation,
indicating that the introduction of the photoactivatable
azido group did not modify the antagonist properties of
the parent compound. Furthermore, the trans confor-
mation, mainly observed for the Cys6-Pro7 amide bond
by NMR studies, seemed not to alter these antagonistic
properties, unlike what had been reported recently.43
Exp er im en ta l Section
Gen er a l. Protected L-amino acids, Boc-L-Tyr(Me)-OH,
Pmp(Mob), Boc-Phe(p-NHZ)-OH, PyBOP, BHA, and MBHA
resins were from either Propeptide, Bachem, or Novabiochem.
Solid-phase peptide synthesis was performed using a manual
device as previously described,34 and solvents used were of
analytical grade. The pH of the organic solutions during solid-
phase and solution synthesis was determined using moistened
pH indicator paper. Peptides were deprotected using HF (from
Matheson, provided by Interchim) on a Kel-F apparatus (from
the Protein Research Foundation). Couplings of the hepta-
peptidic segments to H-Orn(Boc)-Phe(3I,4N3)-NH2 segment
were performed in UV spectroscopy grade quality Me2SO and
DMF. The ligand Pmp-Tyr(Me)-Ile-Thr-Asn-Cys-Pro-Orn-Tyr-
NH2 (OTA) was a generous gift from Dr Manning and was
radioiodinated as previously described.26 All diazo and azido
peptides were handled in darkness.
HP LC. Analytical HPLC monitorings were performed using
three types of conditions. For system I, a Merck Lichrosorb
C18 column (5-µm particle size, 4 × 250 mm, 100 Å porosity)
was fitted to a Shimadzu LC 9A pump equipped with two
Waters 440 and 441 photometers, operating at 214 and 254
nm, respectively. Elution was performed with linear gradients
of mobile phase B (CF3CO2H:CH3CN, 0.1/100, v/v) in mobile
phase A (CF3CO2H:H2O, 0.1/100, v/v) at a flow rate of 2 mL/
min and gradients of 1% B/min (void volume, v0 ≈ 1.7 mL).
The values reported as percent elution correspond to the
percent of CH3CN of the eluent passing through the detector
cell at the time of UV detection and were corrected for the
void volume (v0). For system II, a Merck Lichrospher C18 end-
capped column (5-µm particle size, 100 Å porosity, 4 × 250
mm) was fitted, instead of the Lichrosorb column described
in system I, and run at a 1 mL/min flow rate. For system III,
the radioiodinated compounds were handled on a specially
devoted installation, including a Lichrocart Purospher C18
column (250 × 4 mm; 5-µm particle size, end-capped) connected
to a high-pressure Beckman gradient system with two 114 M
pumps and a 421 A controller, a 440 Waters detector set at
254 nm, and a LKB fraction collector. Elution was performed
with linear gradients of mobile phase B (CF3CO2H:CH3CN:
H2O, 0.1/60/40, v/v/v) in mobile phase A (CF3CO2H:H2O, 0.1/
100, v/v) at a flow rate of 1 mL/min and gradients of 1% B/min.
Highly radioactive 0.5-mL fractions were measured on a Mini-
Assay type 6-20 γ-counter (Numelec) equipped with an
attenuator.
Semipreparative HPLC was performed at each step of the
synthesis utilizing the high-pressure gradient mode employing
a dual pump system (model 400, Applied Biosystem), driven
by a 738A controller (Applied Biosystem) and fitted with either
a Whatman Partisil ODS 3 Magnum 20 column (10-µm particle
size, 22 × 500 mm, v0 ≈ 110 mL) or a Vydac 218TP1022
column (10-µm particle size, 100 Å porosity, 22 × 250 mm, v0
≈ 50 mL). UV detection was performed using a Waters 440
absorbance detector (254 nm) and an ABI 738A spectropho-
tometric detector (set on 214 nm). Peptide purifications were
done using linear gradients of 0.2%, 0.5%, or 1% CH3CN/min
at a flow rate of 10 mL/min, as previously reported.35 Briefly,
the peak corresponding to the desired peptide was collected
in several fractions using a Gilson model 202 fraction collector,
coupled to the 214-nm detector and operating in the peak
detection mode. Each fraction was further checked by analyti-
cal HPLC and finally selected on the basis of the purity rather
than synthetic yield. The fractions were then flash frozen and
lyophilized. The stock solutions of 10-3 M ligands were
periodically checked by analytical HPLC before use, and their
purity was assessed by integration of the peaks at 214 and
254 nm.
Thus we can conclude that ligand 22 is a promising
candidate for the study of antagonist-receptor interac-
tions, as observed from the results obtained on the
mapping of the human OT receptor-binding site.44
Furthermore, the synthesis procedure can serve for the
design of novel photoactivatable ligands. As an example,
it would be possible to isolate the aryldiazonium inter-
mediate (compound 28), using appropriate stabilizing
counterions (CF3CO2-, BF4-).45 Aryldiazonium groups
are photoreactive groups generating arylcations upon
UV irradiation.46 These are positively charged and are
characterized by a different reactivity compared to
arylnitrenes. Therefore, such compounds, used in com-
bination with arylazido derivatives, could possibly
complete the mapping of the OT receptor antagonist-
binding site.
NMR Sp ectr oscop y. The lyophilized sample (2 mg) was
solubilized in [2H6]Me2SO, and all NMR experiments were
recorded at 27 °C on a Bruker AMX spectrometer, operating
1
at 600 MHz for H nucleus. Chemical shifts are quoted relative
to the [2H5]Me2SO resonance fixed at 2.5 ppm. Typically,
DQF-COSY spectra47 were collected into an 800 × 1024 data
matrix with 32 scans per t1 value and TOCSY spectra48 with

3028 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 18
Carnazzi et al.
a mixing time of 80 ms into a 512 × 1024 data matrix with 16
scans per t1 value. NOESY spectra49 were acquired in the
phase-sensitive mode using time-proportional phase incre-
mentation with a mixing time of 200 ms. All the data were
processed with the UXNMR software. One zero-filling and a
π/4 phase-shifted sine bell window function were applied in
both dimensions before Fourier transformation. Routine char-
acterization of heptapeptide intermediates and of dipeptide 6
were performed on a Bruker AMX 360 spectrometer operating
at 360 Hz.
styrene (1% cross-linked, 200-400 mesh), using a Boc strategy
with PyBOP as coupling reagent, (i-Pr)2EtN as tertiary amine,
and CH2Cl2 or dimethylformamide as solvents. The following
side-chain-protected amino acids were used: Boc-Pmp(Mob)-
OH, Boc-Cys(Mob)-OH, Boc-Tyr(Dcb)-OH, and Boc-Asn(Xan)-
OH. All couplings were monitored by the qualitative ninhydrin
test. Recouplings were necessary for the incorporation of Boc-
Asn(Xan)-OH and Boc-Tyr(Dcb)-OH residues. The peptides
were then deprotected (including peptide-resin bond) by HF
treatment, purified by semipreparative HPLC, and lyophilized.
Ma ss Sp ectr om etr y. MALDI-TOF . The samples were
prepared by mixing an equal volume of a 10-5 M peptide
solution and of a saturated solution of R-cyano-4-hydroxy-
cinnamic acid in CF3CO2H:CH3CN:H2O, 0.1/50/50 (v/v/v). The
mixture was then applied (0.5 µL) on the target of a Bruker
Biflex III mass spectrometer and dried at room temperature.50
Desorption of the samples was performed using a 337-nm
laser, and the positive ion mass spectra were obtained in a
linear and reflector mode with an acceleration of 19 kV.
External mass calibration was performed with low-mass
peptide standards (angiotensine I and II). FAB-MS was
performed at the Universite´ de Montpellier 2, using thio-
glycerol as matrix.
Peptides 8 (HPLC elution %, 43; ꢀ214/ꢀ254, 24), 9 (HPLC
elution %, 38; ꢀ214/ꢀ254, 24), 10 (HPLC elution %, 44; ꢀ214/ꢀ254
,
33), and 11 (HPLC elution %, 39; ꢀ214/ꢀ254, 22) were then
cyclized by air-oxidation in an aqueous solution at low
concentration (1 mg/mL) maintained at pH 8, using (i-Pr)2EtN,
to give the corresponding disulfide cyclic peptides Pmp-
Tyr(Me)-Ile-Thr-Asn-Cys-Gly-OH (12; HPLC elution %, 41; ꢀ214
/
ꢀ254, 13), Pmp-Tyr-Ile-Thr-Asn-Cys-Gly-OH (13; HPLC elution
%, 35; ꢀ214/ꢀ254, 12), Pmp-Tyr(Me)-Ile-Thr-Asn-Cys-Pro-OH (14;
HPLC elution %, 43; ꢀ214/ꢀ254, 16), and Pmp-Tyr-Ile-Thr-Asn-
Cys-Pro-OH (15; HPLC elution %; 36; ꢀ214/ꢀ254, 14; MALDI-
TOF (M + H)+ ) 864.7; (M + Na)+ ) 886.7; (M + K)+ ) 902.7;
calculated exact mass ) 863.4).
Syn t h esis of t h e H -Or n (Boc)-P h e(3I,4N3)-NH2 Seg-
m en t (7). The following reactions were monitored using the
analytical HPLC system I. Boc-Phe(4NO2)-OH (10 mmol;
HPLC elution %: 38; ꢀ214/ꢀ254: 1.3) was amidated in dimethyl-
formamide using N-methylmorpholine (10 mmol), isobutyl-
chlorocarbonate (10 mmol), and an excess of 14 M ammonia
(2 mL) to give Boc-Phe(4NO2)-NH2 (1) as a precipitate after a
10 min reaction (HPLC elution %: 32; ꢀ214/ꢀ254: 1.2). The Boc
protecting group of compound 1 was then removed by a
mixture of CF3CO2H:CH2Cl2 (1/1, v/v) to give H-Phe(4NO2)-
NH2 (2), which was isolated by precipitation in diethyl ether
(HPLC elution %, 13; ꢀ214/ꢀ254, 1.0). Compound 2 (4.5 mmol)
was then coupled to Fmoc-Orn(Boc)-OH (4.5 mmol; HPLC
elution %, 50; ꢀ214/ꢀ254, 1.9) in DMF using PyBOP (4.5 mmol)
and (i-Pr)2EtN (4 mol equiv). The dipeptide Fmoc-Orn(Boc)-
Phe(4NO2)-NH2 (3) (HPLC elution %, 53; ꢀ214/ꢀ254, 1.9) was
obtained, in 89% yield, by precipitation in water and was then
alternately rinsed with diethyl ether and hexane. The Fmoc
protecting group of compound 3 was then removed using
diethylamine in dimethylformamide (10%, v/v) to give
Syn th eses of Com p ou n d s 16-19. The following reactions
were monitored using the analytical HPLC system II. Peptide
segments 12, 13, 14, and 15 were coupled to dipeptide 7, the
reaction mixtures were then acidified using CF3CO2H and
applied on a HPLC column to give Pmp-Tyr(Me)-Ile-Thr-Asn-
Cys-Gly-Orn(Boc)-Phe(3I,4N3)-NH2 (16), Pmp-Tyr-Ile-Thr-Asn-
Cys-Gly-Orn(Boc)-Phe(3I,4N3)-NH2 (17), Pmp-Tyr(Me)-Ile-Thr-
Asn-Cys-Pro-Orn(Boc)-Phe(3I,4N3)-NH2 (18), and Pmp-Tyr-Ile-
Thr-Asn-Cys-Pro-Orn(Boc)-Phe(3I,4N3)-NH2 (19). Compounds
16-19 were then deprotected on their Orn side chain. For 10-
mg scale preparations, the Boc group was removed using
CF3CO2H:CH2Cl2 (1/1, v/v) for 1 h, and the trifluoroacetate salt
was precipitated using cold diethyl ether and recovered by
centrifugation (10 mL conical glass tubes). For small scale
preparations (<1 mg), the Boc compounds were deprotected
in CH3CN:CF3CO2H, (1/1, v/v, 200 µL) for 1-2 h, under HPLC
monitoring, and were directly injected on the preparative
HPLC column to give, respectively, Pmp-Tyr(Me)-Ile-Thr-Asn-
Cys-Gly-Orn-Phe(3I,4N3)-NH2 (20; HPLC elution %, 44; ꢀ214
/
ꢀ254, 2.6), Pmp-Tyr-Ile-Thr-Asn-Cys-Gly-Orn-Phe(3I,4N3)-NH2
(21; HPLC elution %, 38; ꢀ214/ꢀ254, 2.1), Pmp-Tyr(Me)-Ile-Thr-
Asn-Cys-Pro-Orn-Phe(3I,4N3)-NH2 (22; HPLC elution %, 48;
ꢀ214/ꢀ254, 2.3), and Pmp-Tyr-Ile-Thr-Asn-Cys-Pro-Orn-Phe(3I,4N3)-
NH2 (23; HPLC elution %, 39; ꢀ214/ꢀ254, 2.6).
H-Orn(Boc)-Phe(4NO2)-NH2 (4) (HPLC elution %, 27; ꢀ214/ꢀ254
,
1.4), which was further precipitated in diethyl ether.
Compound 4 was then reduced by hydrogen bubbling
over catalyst (10% Pd on charcoal) in a solution of MeOH:
dimethylformamide:12 M HCl (10/90/1, v/v/v). The resulting
peptide H-Orn(Boc)-Phe(4NH2)-NH2 (5) (HPLC elution %, 15;
ꢀ214/ꢀ254, 37.5) was obtained in 90% yield following removal of
the catalyst by filtration of the reaction mixture on Celite.
Peptide 5 (0.45 mmol) was then iodinated using 0.01 M ICl (1
mol equiv) in MeOH to give a mixture of 5 and mono- and
diiodinated compounds. The reaction mixture was applied on
the HPLC column to isolate the desired monoiodinated peptide
Syn t h esis of t h e R a d ioiod in a t ed P ep t id e 29 ([125I]-
ZOTA). The synthesis was first performed using nonradio-
active iodine (127I) according to a modified synthesis protocol,
and all reactions were monitored under the conditions of HPLC
system I (Figure 2). Pmp-Tyr(Me)-Ile-Thr-Asn-Cys-Pro-Orn-
Phe(4NH2)-NH2 (24) was synthesized on a MBHA resin using
a Boc/PyBOP/HF strategy, followed by air-oxidation cycliza-
tion, to give Pmp-Tyr(Me)-Ile-Thr-Asn-Cys-Pro-Orn-Phe(4NH2)-
NH2 (25), purified by HPLC and characterized by MALDI-TOF
mass spectrometry (found M + 1 ) 1153.8, calculated exact
mass ) 1152.6) (Figure 2a). Iodination was then performed
by reacting with an equimolar amount of freshly dissolved 25
in MeOH with ICl in MeOH, to give a mixture of compound
25, Pmp-Tyr(Me)-Ile-Thr-Asn-Cys-Pro-Orn-Phe(3I,4NH2)-NH2
(26), and diiodinated compound (Figure 2b). Compound 26 was
purified using HPLC and characterized by MALDI-TOF MS
(found M + 1 ) 1279.7, calculated exact mass ) 1278.4)
(Figure 2c). Diazotization and azidation of 26 were then done
using the same amounts as for the further radiolabeling
procedure. Briefly, compound 26 (0.1 nmol) was diazotized with
an excess of 10-3 M NaNO2 (10 µL approximately) at 0 °C in
darkness, in a solution of 12 M HCl (10 µL) for 1 h before
addition of NaN3 (2 mol equiv over NaNO2, allowed to reach
room temperature), yielding compound 22 (Figure 2e), which
was characterized by FAB MS (M + 1 ) 1305, calculated exact
mass ) 1304.4) (Figure 2f). The synthesis of Pmp-Tyr(Me)-
H-Orn(Boc)-Phe(3I,4NH2)-NH2 (6) (HPLC elution %, 25; ꢀ214
/
,
ꢀ254, 7.5). 1H NMR data (ppm): Orn residue, HR, 3.74; Hââ
1.67; Hγγ, 1.43; Hδδ, 2.92; side chain NH2, 6.83; Boc protecting
group, 1.39; Phe residue, NH, 8.5; HR, 4.4; Hââ, 2.83-2.65;
H6, H5, and H2 (aromatic), 6.99, 6.79, and 7.49, respectively;
NH2 (aromatic amine), 8.02; NH2-C-terminus amide, 7.51-
7.07. Dipeptide 6 was finally diazotized and azidated in
darkness in 2 M HCl using solutions of 1 M NaNO2 (1.1 mol
equiv, 10 min at 0 °C) and of 1 M NaN3 (1.05 mol equiv), with
HPLC-monitoring at room temperature. The azidopeptide
H-Orn(Boc)-Phe(3I,4N3)-NH2 (7) (HPLC elution %, 37; ꢀ214/ꢀ254
,
1.6) was then purified by semipreparative HPLC.
Syn th eses of Segm en ts 12-15. The following reactions
were monitored using the analytical HPLC system II. Pmp-
Tyr(Me)-Ile-Thr-Asn-Cys-Gly-OH (8), Pmp-Tyr-Ile-Thr-Asn-
Cys-Gly-OH (9), Pmp-Tyr(Me)-Ile-Thr-Asn-Cys-Pro-OH (10),
and Pmp-Tyr-Ile-Thr-Asn-Cys-Pro-OH (11) were synthesized
according to a general protocol previously described.34 Briefly,
the peptides were synthesized on a chloromethylated poly-
Ile-Thr-Asn-Cys-Pro-Orn-Phe(3125I,4N3)-NH2
(29,
[
125I]-

A Peptidic Oxytocin Antagonist
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 18 3029
ZOTA) was performed according to the same procedure, except
that the iodination of peptide 25 (10-3 M, 40 µL) was done
with Na125I (Amersham, 1 mCi) in a phosphate buffer (pH 6.8,
55 µL) using an Iodo-Gen-coated Eppendorf vessel (300 µL)
for 5 min. Radiolabeled peptide 29 ([125I]-ZOTA) was purified
under the conditions of HPLC system III, concentrated under
vacuum (Speed-Vac apparatus), and stored in liquid nitrogen.
The identity of 29 was determined by measure of the radio-
activity of 0.5-mL HPLC-eluted fractions of coinjected com-
pounds 22 and 29.
Mem br a n e P r ep a r a tion s. The human OT receptor cDNA
was a generous gift from Dr T. Kimura.3 OT receptor was
stably expressed in CHO cells as previously described.12 Cells
were washed twice, in PBS without Ca2+ and Mg2+, followed
by the addition of a lysis buffer (15 mM Tris:HCl, 2 mM MgCl2,
0.3 mM EDTA, pH 7.4). The cells were then scrapped,
homogenized (Polytron), and centrifuged at 100g for 5 min at
4 °C. Supernatants were collected and centrifuged at 44000g
for 20 min at 4 °C. Pellets were suspended in a Tris:HCl buffer
(15 mM Tris:HCl, 5 mM MgCl2, pH 7.4) and centrifuged again
at 44000g for 20 min at 4 °C. The pellets were resuspended in
a convenient volume of the same buffer and proteins were
determined. Membranes were stored in liquid nitrogen or used
immediately.
ards (Kaleidoscope, Bio-Rad). The gel was fixed in MeCO2H/
MeOH/Me2SO/H2O, 16:40:2:42 (v/v), dried under vacuum, and
autoradiographed using Kodak XAR-5 films at -70 °C.
Ack n ow led gm en t. This work was supported by
INSERM, CNRS, and MRT. Thanks are due to Dr N.
Gale´otti for help in MALDI-TOF experiments, Dr P.
J ouin for constant interest and advice, M. Passama and
L. Charvet for illustrations, and Dr Tuhinadri Sen for
reading the manuscript.
Su p p or tin g In for m a tion Ava ila ble: A table of the
chemical shifts of 22, a table of the structures and properties
of 20-23, and parts of the TOCSY and NOESY spectra of 22.
This material is available free of charge via the Internet at
http/::pubs.acs.org.
Refer en ces
(1) Gainer, H.; Wray, S. In The Physiology of Reproduction; 2nd ed.;
Knobil, E., Neill, J ., Eds.; Raven Press: New York, 1994; pp
1099-1129.
(2) Bockaert, J .; Pin, J . P. Molecular tinkering of G protein-coupled
receptors: an evolutionary success. EMBO J 1999, 18, 1723-
1729.
Bin d in g Exp er im en ts. Affinities of analogues 20-23 were
determined from competition experiments using [125I]-OTA
(200-220 pM) as radioligand with 2-3 µg of membrane
proteins.8,26 The ligands were tested at concentrations ranging
from 10 pM to 10 µM, and the experiments were performed in
triplicate. The Kd of compound 29 ([125I]-ZOTA) was determined
in saturation experiments with 3 µg of membrane proteins
using concentrations ranging from 10 to 900 pM. Incubations
were performed in triplicate at 30 °C for 1 h in a Tris:HCl
buffer (15 mM Tris:HCl, 5 mM MgCl2, pH 7.4) containing 1
mg/mL BSA.8 Bound and free ligand were separated by
filtration (Brandel apparatus) over Whatman GF/C filters
previously soaked with a 10 mg/mL BSA solution for 3-4 h.
The binding data were analyzed by nonlinear least-squares
regression using the computer program Ligand.51
In ositol P h osp h a te Assa ys. The antagonist potency of
compound 22 ([I]-ZOTA) was determined by measuring the
accumulation of inositol phosphates as previously described.12
Briefly, CHO cells expressing the OT receptor were cultured
in 6-well dishes for 48 h in Dulbecco’s modified Eagle’s
medium-supplemented medium and labeled for 24 h with myo-
[2-3H]inositol (Dupont New England Nuclear) at 1 µCi/mL, in
a serum- and inositol-free medium. Cells were then washed
twice with PBS and incubated in PBS for 1 h. They were then
incubated in PBS containing 10 mM LiCl with or without
analogue 22, at various concentrations for 10 min. The cells
were then stimulated using 10 nM OT for an additional period
of 15 min. Incubation was stopped with perchloric acid. Inositol
phosphate extraction was performed using anion-exchange
chromatography (Dowex AG-1 × 8, formate form, 200-400
mesh, Bio-Rad) and the radioactivity was measured. A Kinact
value was calculated as Kinact ) IC50/(1 + [OT]/Kact), where IC50
corresponds to the concentration of analogue leading to the
half-inhibition of the stimulation induced by OT, [OT] ) 10
nM and Kact ) 12.9 nM.12
P h otola belin g of th e Hu m a n Oxytocin Recep tor . CHO
cells membranes expressing the human OT receptor (500 µg,
4 mL) were suspended in buffer A (50 mM Tris:HCl, 5 mM
Mg2+, pH 7.4) containing BSA (0.5 mg/mL) in low-adsorption,
borosilicated tubes (Corex). Incubations were performed using
≈2 nM of freshly radioiodinated ligand 29 for 1 h at 30 °C, in
darkness, in the presence or absence of 10-5 M OT for specific
labeling determination. To remove unbound ligand, mem-
branes were washed with cold buffer A without BSA (10 mL)
and centrifuged (44000g, 20 min) twice. The remaining pellet
was resuspended in cold buffer A (1 mL) and UV-irradiated
for 1 min at 254 nm, in ice-cold siliconized Petri dish.
Membranes were then washed twice again using buffer A (1
mL) and finally solubilized in Laemmli buffer for SDS-PAGE
(12% acrylamide gel, 1 mm thickness) with prestained stand-
(3) Kimura, T.; Tanizawa, O.; Mori, K.; Brownstein, M. J .; Okayama,
H. Structure and expression of a human oxytocin receptor
[published erratum appears in Nature 1992, 357, 176]. Nature
1992, 356, 526-529.
(4) Gorbulev, V.; Buchner, H.; Akhundova, A.; Fahrenholz, F. Mo-
lecular cloning and functional characterization of V2 [8-lysine]
vasopressin and oxytocin receptors from a pig kidney cell line.
Eur. J . Biochem. 1993, 215, 1-7.
(5) Rozen, F.; Russo, C.; Banville, D.; Zingg, H. H. Structure,
characterization, and expression of the rat oxytocin receptor gene
[published erratum appears in Proc. Natl. Acad. Sci. U.S.A.
1996, 93, 12051]. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 200-
204.
(6) J ard, S.; Elands, J .; Schmidt, A.; Barberis, C. In Progress in
Endocrinology; Imura, H., Shizume, K., Eds.; Elsevier: Amster-
dam, 1988; pp 1183-1189.
(7) Barberis, C.; Morin, D.; Durroux, T.; Mouillac, B.; Guillon, G.;
Seyer, R.; Hibert, M.; Tribollet, E.; Manning, M. Molecular
pharmacology of AVP and OT receptors and therapeutic poten-
tial. Drug News Persp. 1999, 12, 279-292.
(8) Phalipou, S.; Seyer, R.; Cotte, N.; Breton, C.; Barberis, C.; Hibert,
M.; Mouillac, B. Docking of linear peptide antagonists into the
human V1a vasopressin receptor. Identification of binding
domains by photoaffinity labeling. J . Biol. Chem. 1999, 274,
23316-23327.
(9) Williams, P. D.; Bock, M. G.; Evans, B. E.; Freidinger, R. M.;
Pettibone, D. J . Progress in the development of oxytocin antago-
nists for use in preterm labor. Adv. Exp. Med. Biol. 1998, 449,
473-479.
(10) Goodwin, T. M.; Zograbyan, A. Oxytocin receptor antagonists.
Update. Clin. Perinatol. 1998, 25, 859-871.
(11) Chan, W. Y.; Wo, N. C.; Stoev, S. T.; Cheng, L. L.; Manning, M.
Discovery and design of novel and selective vasopressin and
oxytocin agonists and antagonists: the role of bioassays. Exp.
Physiol. 2000, 85 Spec No, 7S-18S.
(12) Chini, B.; Mouillac, B.; Balestre, M. N.; Trumpp-Kallmeyer, S.;
Hoflack, J .; Hibert, M.; Andriolo, M.; Pupier, S.; J ard, S.;
Barberis, C. Two aromatic residues regulate the response of the
human oxytocin receptor to the partial agonist arginine vaso-
pressin. FEBS Lett. 1996, 397, 201-206.
(13) Postina, R.; Kojro, E.; Fahrenholz, F. Separate agonist and
peptide antagonist binding sites of the oxytocin receptor defined
by their transfer into the V2 vasopressin receptor. J . Biol. Chem.
1996, 271, 31593-31601.
(14) Dorman, G.; Prestwich, G. D. Using photolabile ligands in drug
discovery and development. Trends Biotechnol. 2000, 18, 64-
77.
(15) Stadel, J . M.; Goodman, D. B.; Galardy, R. E.; Rasmussen, H.
Synthesis and characterization of 2-nitro-5-azidobenzoylgly-
cyloxytocin, an oxytocin photoaffinity label. Biochemistry 1978,
17, 1403-1408.
(16) Estrada, E. F.; Barra, V.; Caorsi, C. E.; Troncoso, S.; Ruiz-Opazo,
N.; Gonzalez, C. B. Identification of the V1 vasopressin receptor
by chemical cross-linking and ligand affinity blotting. Biochem-
istry 1991, 30, 8611-8616.
(17) Kojro, E.; Hackenberg, M.; Zsigo, J .; Fahrenholz, F. Identification
and enzymatic deglycosylation of the myometrial oxytocin recep-
tor using a radioiodinated photoreactive antagonist. J . Biol.
Chem. 1991, 266, 21416-21421.

3030 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 18
Carnazzi et al.
(18) Hinko, A.; Soloff, M. S. Characterization of oxytocin receptors
in rabbit amnion involved in the production of prostaglandin E2.
Endocrinology 1992, 130, 3547-3553.
(37) Ruat, M.; Korner, M.; Garbarg, M.; Gros, C.; Schwartz, J . C.;
Tertiuk, W.; Ganellin, C. R. Characterization of histamine H1-
receptor binding peptides in guinea pig brain using [125I]iodo-
azidophenpyramine, an irreversible specific photoaffinity probe.
Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 2743-2747.
(38) Barberis, C.; Balestre, M. N.; J ard, S.; Tribollet, E.; Arsenijenic,
Y.; Dreifuss, J . J .; Bankowski, K.; Manning, M.; Chan, W. Y.;
Schlosser, S. S.; Holsboer, F.; Elands, J . Characterization of a
novel, linear radioiodinated vasopressin antagonist: an excellent
radioligand for vasopressin V1a receptors. Neuroendocrinology
1995, 62, 135-146.
(39) Mouillac, B.; Chini, B.; Balestre, M. N.; Elands, J .; Trumpp-
Kallmeyer, S.; Hoflack, J .; Hibert, M.; J ard, S.; Barberis, C. The
binding site of neuropeptide vasopressin V1a receptor. Evidence
for a major localization within transmembrane regions. J . Biol.
Chem. 1995, 270, 25771-25777.
(40) Kojro, E.; Hackenberg, M.; Klein, U.; Fahrenholz, F. In Peptides
1990; Giralt, E., Andreu, D., Eds.; Escom Science Publishers B.
V.: New York, 1991; pp 674-675.
(19) Hinko, A.; Soloff, M. S.; Potier, M. Molecular size characteriza-
tion of oxytocin receptors in rabbit amnion. Endocrinology 1992,
130, 3554-3559.
(20) Carnazzi, E.; Aumelas, A.; Seyer, R. In Peptides 1990; Giralt,
E., Andreu, D., Eds.; Escom Science Publishers B. V., 1991; pp
219-220.
(21) Manning, M.; Olma, A.; Klis, W.; Kolodziejczyk, A.; Nawrocka,
E.; Misicka, A.; Seto, J .; Sawyer, W. H. Carboxy terminus of
vasopressin required for activity but not binding. Nature 1984,
308, 652-653.
(22) Nestor, J . J ., J r.; Ferger, M. F.; du Vigneaud, V. [1-Beta-
mercapto-beta,beta-pentamethylenepropionic acid]oxytocin, a
potent inhibitor of oxytocin. J . Med. Chem. 1975, 18, 284-287.
(23) Lowbridge, J .; Manning, M.; Haldar, J .; Sawyer, W. H. Synthesis
and some pharmacological properties of [4-threonine, 7-glycine]-
oxytocin, [1-(L-2-hydroxy-3-mercaptopropanoic acid), 4-threo-
nine, 7-glycine]oxytocin (hydroxy[Thr4, Gly7]oxytocin), and
[7-glycine]oxytocin, peptides with high oxytocic-antidiuretic
selectivity. J . Med. Chem. 1977, 20, 120-123.
(41) Carnazzi, E.; Aumelas, A.; Barberis, C.; Guillon, G.; Seyer, R. A
new series of photoactivatable and iodinatable linear vasopressin
antagonists. J . Med. Chem. 1994, 37, 1841-1849.
(24) Bankowski, K.; Manning, M.; Haldar, J .; Sawyer, W. H. Design
of potent antagonists of the vasopressor response to arginine-
vasopressin. J . Med. Chem. 1978, 21, 850-853.
(25) Bankowski, K.; Manning, M.; Seto, J .; Haldar, J .; Sawyer, W.
H. Design and synthesis of potent in vivo antagonists of oxytocin.
Int. J . Pept. Protein Res. 1980, 16, 382-391.
(42) Possmayer, F. E.; Hartog, A. F.; Berden, J . A.; Graber, P.
Covalent modification of the catalytic sites of the H+-ATPase
from chloroplasts and 2-nitreno-ADP. Modification of the cata-
lytic site 1 (tight) and catalytic sites 1 and 2 together impairs
both uni-site and multisite catalysis of ATP synthesis and ATP
hydrolysis. Biochim. Biophys. Acta 2000, 1459, 202-217.
(43) Belec, L.; Slaninova, J .; Lubell, W. D. A study of the relationship
between biological activity and prolyl amide isomer geometry
in oxytocin using 5-tert-butylproline to augment the Cys6-Pro7
amide cis-isomer population. J . Med. Chem. 2000, 43, 1448-
1455.
(44) Breton, C.; Chellil, H.; Kabbaj-Benmansour, M., Carnazzi, E.;
Seyer, R.; Phalipou, S.; Morin, D.; Durroux, T.; Zingg, H.;
Barberis, C.; Mouillac, B. Direct identification of human oxytocin
receptor-binding domains using a photoactivatable cyclic peptide
antagonist: comparison with the human V1a vasopressin recep-
tor. J . Biol. Chem. 2001, 276, 26931-26941.
(45) Colas, C.; Goeldner, M. An efficient procedure for the synthesis
of cristalline aryldiazonium trifluoroacetates. Synthetic applica-
tions. Eur. J . Org. Chem. 1999, 1357, 7-1366.
(46) Kotzyba-Hibert, F.; Kapfer, I.; Goeldner, M. Recent trends in
photoaffinity labeling. Angew. Chem., Int. Ed. Engl. 1995, 34,
1296-1312.
(47) Rance, M.; Sorensen, O. W.; Bodenhausen, G.; Wagner, G.; Ernst,
R. R.; Wuthrich, K. Improved spectral resolution in cosy 1H NMR
spectra of proteins via double quantum filtering. Biochem.
Biophys. Res. Commun. 1983, 117, 479-485.
(48) Rance, M. Improved techniques for homonuclear rotating-frame
and isotropic mixing experiments. J . Magn. Reson. 1987, 74,
557-564.
(26) Elands, J .; Barberis, C.; J ard, S.; Tribollet, E.; Dreifuss, J . J .;
Bankowski, K.; Manning, M.; Sawyer, W. H. 125I-labeled d(CH2)5-
[Tyr(Me)2, Thr4, Tyr-NH29]OVT: a selective oxytocin receptor
ligand. Eur. J . Pharmacol. 1987, 147, 197-207.
(27) Elands, J .; Barberis, C.; J ard, S. [3H]-[Thr4, Gly7]OT: a highly
selective ligand for central and peripheral OT receptors. Am. J .
Physiol. 1988, 254, E31-38.
(28) Weber, P. J .; Beck-Sickinger, A. G. Comparison of the photo-
chemical behavior of four different photoactivatable probes. J .
Pept. Res. 1997, 49, 375-383.
(29) Watt, D. S.; Kawada, K.; Leyva, E.; Platz, M. S. Exploratory
photochemistry of iodinated aromatic azides. Tetrahedron Lett.
1989, 30, 899-902.
(30) Henriksen, U.; Buchardt, O. Aryl azides as photolabels. Reten-
tion of iodine during photochemical ring expansion of an iodi-
nated tyrosine derivative. Tetrahedron Lett. 1990, 31, 2443-
2444.
(31) Resek, J . F.; Ruoho, A. E. Photoaffinity labeling the beta-
adrenergic receptor with an iodoazido derivative of norepineph-
rine. J . Biol. Chem. 1988, 263, 14410-14416.
(32) Balczon, R. D.; Brinkley, B. R. Synthesis of azidotubulin:
a
photoaffinity label for tubulin-binding proteins. Biochemistry
1989, 28, 8490-8496.
(33) Kinnunen, P. M.; Klopf, F. H.; Bastiani, C. A.; Gelfman, C. M.;
Lange, L. G. 12-[(5-iodo-4-azido-2-hydroxybenzoyl)amino]do-
decanoic acid: biological recognition by cholesterol esterase and
acyl-CoA: cholesterol O-acyltransferase. Biochemistry 1990, 29,
1648-1654.
(49) Macura, S.; Huang, Y.; Sutter, D.; Ernst, R. Two-dimensional
chemical chemical exchange and cross relaxation spectroscopy
of coupled nuclear spins. J . Magn. Reson. 1981, 43, 259-281.
(50) Cohen, S. L. Influence of matrix solution conditions on the
MALDI-MS analysis of peptides and proteins. Anal. Chem. 1996,
68, 31-37.
(51) Munson, P. J .; Rodbard, D. Ligand: a versatile computerized
approach for characterization of ligand-binding systems. Anal.
Biochem. 1980, 107, 220-239.
(34) Coste, J .; Le-Nguyen, D.; Castro, B. PyBOP: a new peptide
reagent devoid of toxic byproduct. Tetrahedron Lett. 1990, 31,
205-208.
(35) Seyer, R.; Aumelas, A. Synthesis of biotinylated and photo-
reactive probes for angiotensin receptors. J . Chem. Soc., Perkin
Trans. 1 1990, 3289-3299.
(36) Wu¨thrich, K. NMR of Proteins & Nucleic Acids; J ohn Wiley &
Sons: New York, 1986.
J M010125U