A labeled neutral endopeptidase inhibitor as a potential tool for tumor diagnosis and prognosis

Article abstract of DOI:10.1002/anie.200500700

(Chemical Equation Presented) Specific targeting of tumors with radioactive isotopes is the key for high-sensitivity scintigraphy detection. A labeled neutral endopeptidase (NEP) inhibitor is described that accumulates in vivo in tumor cells expressing th

Full text of DOI:10.1002/anie.200500700

Communications
Tumor Imaging Agents
A Labeled Neutral Endopeptidase Inhibitor as a
Potential Tool for Tumor Diagnosis and
Prognosis**
Olivier Raguin, Marie-Claude Fourniꢀ-Zaluski,
Anthony Romieu, Andrꢀ Pꢁlegrin, Franꢂois Chatelet,
Didier Pꢀlaprat, Jacques Barbet, Bernard P. Roques,*
and Anne Gruaz-Guyon*
Scintigraphy with labeled somatostatin analogues and posi-
tron-emission tomography (PET) with 18F-labeled 2-deoxy-2-
fluoro-d-glucose (18F-FDG) have greatly improved the diag-
nosis and staging of numerous tumors with somatostatin-
receptor-subtype overexpression or active glucose metabo-
lism, respectively. This entailed the development of numerous
radiolabeled peptides that target other receptors overex-
pressed by tumors that do not accumulate somatostatin
analogues or 18F-FDG.[1,2] Proteolytic enzymes expressed on
many cancer cells are also potential targets for radiopharma-
ceuticals. Recently 18F- or 11C-labeled inhibitors of metal-
loproteinase-2 (MMP-2), which is expressed in a variety of
malignant tumors, have been proposed for use in PET.[3,4]
Neutral endopeptidase (NEP, CD10, CALLA, enkepha-
linase, neprylisin, EC 3.4.24.11) is a membrane-bound zinc
metallopeptidase that interrupts cell signaling by degradation
[*] Dr. A. Romieu, Prof. B. P. Roques
INSERM U266
Universitꢀ Paris V Renꢀ Descartes
4 avenue de l’Observatoire, 75006 Paris (France)
Fax: (+33)1-4326-6918
E-mail: roques@pharmacie.univ-paris5.fr
Dr. O. Raguin,+ Dr. D. Pꢀlaprat,+ Dr. A. Gruaz-Guyon+
EA 3512, Facultꢀ de Mꢀdecine Xavier Bichat
BP416, 16 rue Henri Huchard, 75870 Paris cedex 18 (France)
Fax: (+33)1-4485-6306
E-mail: gruaz@bichat.inserm.fr
Prof. M.-C. Fourniꢀ-Zaluski
Pharmaleads
4 avenue de l’Observatoire, 75270 Paris cedex 06 (France)
Dr. A. Pꢁlegrin+
INSERM EMI 0227, Universitꢀ Montpellier 1
CRLCC Val d’Aurelle
Centre de Recherche en Cancꢀrologie
34298 Montpellier cedex 5 (France)
Prof. F. Chatelet
UFR Saint-Antoine, Universitꢀ Paris VI Pierre et Marie Curie
27 rue Chaligny, 75012 Paris (France)
Dr. J. Barbet+
INSERM U601, Universitꢀ de Nantes
Institut de Biologie
9 quai Moncousu, 44093 Nantes cedex 01 (France)
[+] These authors are members of the CNRS GDR 2352, Immuno-
ciblage des tumeurs, Marseille, France.
[**] We thank the French Ligue Nationale Contre le Cancer for awarding
a research fellowship to O.R.
Supporting Information for this article is available on the WWW
under http://www.angewandte.org or from the author.
4058
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200500700
Angew. Chem. Int. Ed. 2005, 44, 4058 –4061

Angewandte
Chemie
of numerous physiologically active peptides.[5,6] NEP is
expressed in a large variety of tissues including the brain,
kidney, and lung,[7] but it is also overexpressed in various types
of hematological tumors, particularly acute lymphoblastic
leukemia[8] and some lymphomas, as well as in solid tumors
such as melanoma,[9] colonic, pancreatic, and prostatic
adenocarcinomas.[10] Presence, or absence, of NEP expression
has been proposed as a prognostic indicator, depending on the
tumor type.[11–17] In addition, lack of NEP expression in a
prostatic carcinoma may predict decreased androgen sensi-
tivity and a poorer response to hormonal deprivation
therapy.[18,19] A noninvasive scintigraphic technique imaging
tumor-cell accumulation of a labeled NEP inhibitor could be
an interesting diagnosis and/or prognosis tool.
phosphinic moiety, prepared independently by liquid-phase
synthesis, was incorporated onto the peptidyl resin by a
classical coupling step (Scheme 2, see the Supporting Infor-
mation).
Structure–activity relationship studies of NEP inhibi-
tors,[6] site-directed mutagenesis experiments,[6,20,21] and crys-
tallographic studies of enzyme–inhibitor complexes further
elucidated the interactions that stabilize substrates or inhib-
itors in the NEP active site. Potent inhibition of NEP is
obtained with compounds containing a strong zinc ligand as a
phosphinic group, a hydrophobic moiety interacting with the
S’1 subsite of the enzyme,[22] hydrogen-bonding donor and
acceptor groups for binding with Asn542 and Arg717, and a
carboxylate function for Arg102 recognition (Scheme 1).
Scheme 2. Synthesis of the NEP inhibitor AR58. a) Bis(trimethylsilyl)-
acetamide, 708C; b) BOP, DIEA, NMP, solid-phase peptide synthesis
then cleavage from the MBHA resin and final deprotection of the side
chains of the amino acid residues (tert-butyl ether: Tyr, tert-butyl ester:
Asp, N-tert-butoxycarbonyl: Lys) by trifluoroacetic acid to form R1 and
the desired product. BOP=1-benzotriazolyloxytris(dimethylamino)-
phosphonium hexafluorophosphate, DIEA=diisopropylethylamine,
NMP=N-methyl-pyrrolidone, MBHA=4-methyl-benzhydrylamine.
As the phosphinic moiety was synthesized as a mixture of
enantiomers, the inhibitor was obtained as a mixture of two
stereoisomers, which were separated by C18 reversed-phase
HPLC. The first eluted isomer exhibited the best inhibitory
potency: AR58-1 had an inhibition constant of Ki = (3.6 Æ
0.1) ꢀ 10À8 m, versus Ki = (6.7 Æ 0.4) ꢀ 10À7 m for the second
eluted isomer (AR58-2).
High specific activities were obtained by labeling the
DTPA-substituted derivative of AR58-1, referred to as AR60,
with 111In ((191 Æ 20) MBqnmolÀ1). The AR60 equilibrium
binding-affinity constant ((1.1 Æ 0.1) ꢀ 108 mÀ1) was deter-
mined by competition binding experiments (Figure 1) on a
human Burkitt lymphoma cell line (Ramos cells), in which
NEP expression was confirmed by fluorescence-activated cell
sorter (FACS) analysis (see the Supporting Information). This
result suggests that coupling to DTPA does not significantly
lower affinity. The number of NEP binding sites per Ramos
cell (2.6 ꢀ 104–5.7 ꢀ 104) was in the same range as that already
determined from equilibrium binding experiments of 125I-
labeled anti-CD10 antibody (2.1 ꢀ 104–7.5 ꢀ 104).[23]
Scheme 1. Model of NEP active-site recognition by the phosphinic
inhibitors AR58 and AR60 proposed from the crystallographic data
reported by Oefner et al.[24]
Based on these findings, we designed novel compounds
with a biphenylmethyl group and an aspartic acid unit to
interact with the S’1 and S’2 subsites, respectively, and a benzyl
moiety bearing a phosphinyl group as zinc ligand (Scheme 1).
A 6-aminohexanoic acid was introduced as a spacer and its
carboxylate moiety was coupled to a dipeptide amide Tyr-Lys-
NH2, with the tyrosine residue allowing labeling with iodine
isotopes. The chelating agent diethylenetriaminepentaacetic
acid (DTPA) was introduced on the e NH2 group of the lysine
to allow labeling with metal radioisotopes. The peptide part of
the molecule was synthesized by solid-phase methods and the
A significant specific binding of 111In-labeled AR60 to
tissues of the kidneys (mainly the cortical zone), lungs, liver,
and small intestine, but not of muscle, heart, stomach (not
shown), and colon, was demonstrated by in vitro autoradio-
graphy studies (Figure 2), in agreement with NEP expression
already described in these organs.[7] This labeled inhibitor
Angew. Chem. Int. Ed. 2005, 44, 4058 –4061
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4059

Communications
Figure 3. Biodistribution studies of 111In-AR60 in healthy BALB/c mice.
Mice were sacrificed, after an intravenous injection of only 111In-AR60
(10 pmol in 150 mL of phosphate-buffered saline (PBS)) at 3 h (white
bars) or 6 h (gray bars) or at 6 h in the presence of excess thiorphan
(790 nmol, black bars). Tissue radioactivity is expressed as the per-
centage of injected dose per g of tissue (D, mean valueÆstandard
error of the mean (s.e.m.), number of animals: 5 at 3 h, 4 at 6 h).
Injected doses were corrected by subtraction of noninjected and sub-
cutaneously injected material. Additional abbreviations to Figure 2:
bl=blood, sp=spleen, git=gastrointestinal tract, ad=adrenal
tissues, bo=bone, gb=gall bladder, st=stomach.
Figure 1. Typical competition binding curve between 111In-labeled and
unlabeled AR60. Trace amounts of 111In-labeled AR60 were incubated
with Ramos cells in the presence of increasing concentrations of AR60
saturated with nonradioactive indium. (The curve was generated by
computer simulation with the Equilibrium Expert software.[25]) The frac-
tion of radioactivity associated to the cells (bound/total (B/T) in %)
was counted. The mean values from triplicate determinations are plot-
ted. Bars give standard deviation values except those that were smaller
than the points as plotted.
well as the p values: lungs: 4.60 Æ0.93 versus 2.17Æ0.31, p<
0.05; adrenal tissues: 0.60Æ0.08 versus 0.21Æ0.03, p<0.01;
kidneys: 234Æ56 versus 56 Æ33, p<0.01;). Liver labeling was
low and no significant difference was observed in the presence
of thiorphan. These results are consistent with the rapid
decrease of liver activity and the absence of gut activity
through uptake of a tritiated NEP inhibitor, facts which suggest
that several tissues are inaccessible to intravenously injected
inhibitors thanks to the presence of functional barriers.[7]
111In-AR60 uptake in tumors evaluated in SCID mice
intravenously grafted with Ramos human Burkitt lymphoma
cells was high ((10.3 Æ2.4) D, with a range of 21.8–3.5 D, 4.5 h
after injection). Since in this model disseminated tumor-cell
grafts occur in different organs, normal tissue uptakes were
evaluated in healthy SCID mice. Tumor to normal tissue uptake
ratios were high for the brain (144Æ30, not shown), heart,
stomach, blood, adrenal tissues, liver, spleen, and small intestine.
Lower targeting contrasts were obtained for other organs, with
values of around 3 for the colon, muscle, and bone, 1.4 Æ0.3 for
the lungs, and 0.06Æ0.01 for the kidneys (Figure 4).
Figure 2. Autoradiogram of 111In-AR60 binding to tissue sections
(lu=lungs, li=liver, sp=spleen, ki=kidneys, si=small intestine,
co=colon, he=heart, mu=muscle) from healthy BALB/c mice and
intravenous grafts of Ramos cells to SCID mice (tu=tumor). Top:
total binding. Bottom: nonspecific binding evaluated on adjacent sec-
tions in the presence of an excess (10À6 m) of AR58-1. Binding quantifi-
cation, expressed in arbitrary activity units per surface unit, is indi-
cated on each image. The Student t test was used to evaluate the
significance of the results ( : statistical probability value p<0.05;
*
: p<0.01; n.s.: not significant).
**
bound specifically to sections of Ramos cell grafts to SCID
mice.
Since the tumor to normal tissue uptake ratio is the key
for isotopic tumor imaging, 111In-AR60 should provide good
contrast images except for tumors located on or near organs
with high NEP expression, such as the kidneys and lungs. In
We then investigated the in vivo biodistribution of 111In-
AR60 three and six hours after intravenous injection to
healthy BALB/c mice (Figure 3). A fast blood clearance of
111In-AR60 was observed, as assessed by the low residual
blood radioactivity six hours after injection. Activity uptake
was observed primarily in the kidneys and to a lesser extent in
the lungs, liver, and colon. Since about 92% of the colon
activity was associated with the stools, activity uptake in this
organ was related to biliary excretion of 111In-AR60. A low
activity accumulation of < 1% of the injected dose per gram
was observed in other organs six hours after injection. 111In-
AR60 stability in vivo was high since more than 70% was
recovered intact in the serum 1 h after injection.
Figure 4. Tumor to normal tissue uptake ratios UT/N of 111In-AR60 in
SCID mice intravenously grafted with Ramos cells. Four animals were
given 111In-AR60 (10 pmol in 150 mL of PBS) by intravenous injection.
Blood, organs, and tumor masses were collected and counted for
radioactivity 4.5 h later. Tumor involvement of macroscopically suspect
specimens was confirmed by histological analysis. Injected doses were
corrected for losses as described in the legend to Figure 3. Since the
graft is disseminated, results are expressed as the ratio between tumor
uptake (10 samples) and mean organ uptake in nongrafted mice
(mean value of UT/N Æ(s.e.m.)).
Coinjection of 111In-AR60 (10 pmol) and a large excess of
unlabeled Thiorphan (790 nmol), a specific NEP inhibitor,
considerably reduced organ activity in the lungs, adrenal
tissues, and kidneys, a result indicating uptake specificity in
these organs (mean D values Æ (standard error of the mean;
s.e.m.) in the absence versus in the presence of thiorphan as
4060 ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 4058 –4061

Angewandte
Chemie
[19] C. N. Papandreou, B. Usmani, Y. Geng, T. Bogenrieder, R.
some organs of tumor-bearing mice, with no macroscopic
tumor involvement, activity uptake was greater than the mean
uptake of corresponding organs of nongrafted mice. Histo-
logical analysis revealed tumor involvement in four of these
samples, two brain samples with values of 0.78 and 0.57 D
versus (0.07 Æ 0.03) D in ungrafted mice and two adrenal
tissue samples with values of 1.91 and 9.33 D versus (1.07 Æ
0.39) D in ungrafted mice. These results further demonstrated
the specificity of tumor uptake of 111In-AR60.
In conclusion, the fast plasma clearance, the elevated
tumor labeling, and the low nonspecific uptake of 111In-
labeled NEP ligand AR60 in normal organs provide high
uptake ratios between tumor and normal organs except for
the kidneys and lungs, which express high NEP levels.
Noninvasive scintigraphic imaging of NEP expression with
this compound may be a potential tool for the diagnosis and/
or prognosis of different tumor types.
Freeman, S. Wilk, C. L. Finstad, V. E. Reuter, C. T. Powell, D.
Scheinberg, C. Magill, H. I. Scher, A. P. Albino, D. M. Nanus,
Nat. Med. 1998, 4, 50.
[20] N. Dion, P. Cohen, P. Crine, G. Boileau, FEBS Lett. 1997, 411,
140.
[21] A. Devault, V. Sales, C. Nault, A. Beaumont, B. Roques, P. Crine,
G. Boileau, FEBS Lett. 1988, 231, 54.
[22] I. Schechter, A. Berger, Biochem. Biophys. Res. Commun. 1967,
27, 157.
[23] M. Hillairet de Boisferon, O. Raguin, M. Dussaillant, W. Ros-
tene, J. Barbet, A. Gruaz-Guyon, Bioconjugate Chem. 2000, 11,
452.
[24] C. Oefner, B. P. Roques, M. C. Fournie-Zaluski, G. E. Dale, Acta
Crystallogr. D Biol. Crystallogr. 2004, 60, 392.
[25] O. Raguin, A. Gruaz-Guyon, J. Barbet, Anal. Biochem. 2002,
310, 1.
Received: February 24, 2005
Published online: May 31, 2005
Keywords: imaging agents · metalloenzymes ·
.
peptidomimetics · radiopharmaceuticals · tumor targeting
[1] D. Kwekkeboom, E. P. Krenning, M. de Jong, J. Nucl. Med. 2000,
41, 1704.
[2] M. De Jong, D. Kwekkeboom, R. Valkema, E. P. Krenning, Eur.
J. Nucl. Med. Mol. Imaging 2003, 30, 463.
[3] S. Furumoto, K. Takashima, K. Kubota, T. Ido, R. Iwata, H.
Fukuda, Nucl. Med. Biol. 2003, 30, 119.
[4] Q. H. Zheng, X. Fei, T. R. DeGrado, J. Q. Wang, K. L. Stone,
T. D. Martinez, D. J. Gay, W. L. Baity, B. H. Mock, B. E. Glick-
Wilson, M. L. Sullivan, K. D. Miller, G. W. Sledge, G. D.
Hutchins, Nucl. Med. Biol. 2003, 30, 753.
[5] B. Malfroy, J. P. Swerts, A. Guyon, B. P. Roques, J. C. Schwartz,
Nature 1978, 276, 523.
[6] B. P. Roques, F. Noble, V. Dauge, M. C. Fournie-Zaluski, A.
Beaumont, Pharmacol. Rev. 1993, 45, 87.
[7] N. Sales, I. Dutriez, B. Maziere, M. Ottaviani, B. P. Roques,
Regul. Pept. 1991, 33, 209.
[8] M. A. Shipp, J. Vijayaraghavan, E. V. Schmidt, E. L. Masteller,
L. DꢁAdamio, L. B. Hersh, E. L. Reinherz, Proc. Natl. Acad. Sci.
USA 1989, 86, 297.
[9] S. Carrel, L. Zografos, M. Schreyer, D. Rimoldi, Melanoma Res.
1993, 3, 319.
[10] P. Chu, D. A. Arber, Am. J. Clin. Pathol. 2000, 113, 374.
[11] M. Ponzoni, A. J. Ferreri, G. Pruneri, B. Pozzi, S. DellꢁOro, A.
Pigni, G. Pinotti, E. Villa, M. Freschi, G. Viale, C. Capella, Int. J.
Cancer 2003, 106, 288.
[12] T. Tokuhara, M. Adachi, H. Hashida, H. Ishida, T. Taki, M.
Higashiyama, K. Kodama, S. Tachibana, S. Sasaki, M. Miyake,
Jpn. J. Thorac. Cardiovasc. Surg. 2001, 49, 489.
[13] K. Iwaya, H. Ogawa, M. Izumi, M. Kuroda, K. Mukai, Virchows
Arch. 2002, 440, 589.
[14] S. N. Gaidukova, L. A. Sivak, Lik. Sprava 2002, 47.
[15] J. H. Go, W. I. Yang, H. J. Ree, Arch. Pathol. Lab. Med. 2002,
126, 956.
[16] P. Uherova, C. W. Ross, B. Schnitzer, T. P. Singleton, W. G. Finn,
Am. J. Clin. Pathol. 2001, 115, 582.
[17] J. P. Vannier, M. C. Bene, G. C. Faure, C. Bastard, R. Garand, A.
Bernard, Br. J. Haematol. 1989, 72, 156.
[18] S. J. Freedland, D. B. Seligson, A. Y. Liu, A. J. Pantuck, S. H.
Paik, S. Horvath, J. A. Wieder, A. Zisman, D. Nguyen, C. L. Tso,
A. V. Palotie, A. S. Belldegrun, Prostate 2003, 55, 71.
Angew. Chem. Int. Ed. 2005, 44, 4058 –4061
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4061