WO2012045719A2

WO2012045719A2 - New vascular targeting peptides

Info

Publication number: WO2012045719A2
Application number: PCT/EP2011/067276
Authority: WO
Inventors: Gian-Paolo Rizzardi; Catia Traversari
Original assignee: Molmed Spa
Priority date: 2010-10-05
Filing date: 2011-10-04
Publication date: 2012-04-12
Also published as: WO2012045719A3

Abstract

The present invention relates to new vascular targeting peptides containing the retro- inverso peptide rGn, such peptides being characterized by peptide bond reversal and inversion of chirality of the aminoacids of the parent peptide NGR. The retro-inverso rGn peptides are able to target tumor vessels since they are effective ligands of CD13 receptor expressed in angiogenic endothelium. The present invention also relate to a new therapeutic strategy against cancer based on the use of conjugates of peptides containing rGn motif and a therapeutic moiety. The presence of the rGn motif allows a specific targeting to tumor vessels of the therapeutic moiety thus causing an increase in efficacy and a reduction of toxic effects.

Description

New vascular targeting peptides

Field of the invention

The present invention relates to vascular targeting for the treatment of angiogenesis- dependent diseases, such as tumors. More particularly, the present invention relates to new vascular targeting peptides containing the retro-inverso peptide rGn, such peptides being characterized by peptide bond reversal and inversion of chirality of the aminoacids of the parent peptide NGR. The retro-inverso rGn peptides of the present invention are able to bind CD13 isoforms expressed in angiogenic endothelium.

Background Vascular targeting is a very promising approach for the treatment of angiogenesis- dependent diseases particularly for tumor treatment. Targeted drug-delivery to tumor vasculature preferentially localizes the drug at the targeted site, thus increasing efficacy and decreasing systemic side effects. Interestingly, activated endothelial cells and pericytes in tumor neovasculature have been shown to express molecules that are characteristic of angiogenic vessels and are virtually not expressed in normal vessels. Notably, an isoform of aminopeptidase N (CD13) has been shown to be selectively expressed on endothelial cells of tumor-associated vessels [Curnis et al. (2002) Cancer Res] and to bind to peptides containing the Asn-Gly-Arg (NGR)-motif. Such "vascula r addresses" allow the selective ta rgeting of systemically administrated therapies to tumors, including for instance chemotherapeutic agents and cytokines [Pasqualini and Ruoslahti (1996) Nature; Corti A et al. (2008) Blood]. In particular, the NGR-hTNF conjugate has been shown to induce potent antitumor effects and is currently tested in phase ll-lll clinical trials [Gregorc et al. (2009) Br J Cancer; Gregorc et al. (2010) J Clin Oncol; Gregorc et al. (2010) Eur J Cancer; van Laarhoven et al. (2010) Clin Cancer Res; Santoro et al. (2010) Br J Cancer; Santoro et al. (2010) Eur J Cancer]. Therefore, selective targeting of CD13 isoforms expressed in tumor associated vessels appears to be an effective strategy to obtain anticancer drugs. There is a need to obtain new peptides that are able to effectively target CD13 receptor, the present invention addresses this issue. The retro-inverso modification of a peptide consists in peptide bond reversal and inversion of chirality of the parent peptide. Even though in some cases retro-inverso peptides resulted to be potent analogues of parent peptides [Chorev, M. & Goodman, M. (1995). Trends Biotechnol; Fletcher, M.D. & Campbell, M.M. (1998) Chem. Rev.] it is recognised that side-chain topology is not perfectly maintained between a peptide and its retro-enantiomer [Fredinger, R.M. & Veber, D.F. (1979). J. Am. Chem. Soc.]. Therefore retro- inverso modification of biologically active peptides can seriously affect their function.

Particularly the retro-inverso approach resulted to be unsuccessful for RGD peptides, a potent ligand of ανβ3 integrin. Werm uth et al.; (1997) J. Am. Chem . Soc, evaluated the effect of retro-inverso modification on cyclo-RGD peptides particularly on their activity as vitronectin antagonists. All RGD-retro-inverso peptides showed reduced activity compared with their parent compounds, except for one case in which the activity increased.

Dal Pozzo et al.; (2000) J. Peptide Res., obtained six retro-inverso tri- and tetrapeptide a na logues of RGD that showed to have a 2±3-fold decrease in potency or tota l loss of bioactivity. S uch data we re expla i ned by a uthors o n the basis of structu re-activity relationship. Particularly, when the bioactive peptide conformation has critical spatial requirements, retro-inversion causes perturbations incompatible with the affinity to the receptor.

Therefore, it is clear that retro-inverso modification of biologically effective peptides causes a totally unpredictable effect on the ca pability of such peptides of binding to the same receptor and it is unexpected that a retro-inverso peptide be a comparable or even a better ligand than its parent peptide.

Definition

Amino acids will be herein indicated according to conventional single letter code. In addition, in order to distinguish the chirality of alpha-carbon of each amino acid, L-isomers will be indicated using capital letters and D-isomer will be indicated using small letters. Therefore, for example, L-isomer of amino acid arginine will be indicated as "R", while D-isomer will be indicated as "r". Summary of the invention

The present invention relates to new vascular targeting peptides containing the retro- inverso peptide rGn, such peptides being characterized by peptide bond reversal and inversion of chirality of the aminoacids of the parent peptide NGR. Retro-inverso modification of biologically active peptides leads in some cases to effective analogues. Despite this, in many other cases such modification affected the ability of such analogue to bind the same receptor of the parent peptide. The most relevant example in the field of vascular targeting is the case of RGD peptides whose retro-inverso did not maintain the ability to bind ανβ3 receptor. The peptides of the present invention contain the rGn motif, wherein r is the D isomer of arginine, G is glycine and n is the D isomer of asparagine. Such rGn peptide is the retro- inverso of the peptide NGR (asparagine-glycine-arginine) and, therefore, it is characterized by peptide bond reversal and inversion of chirality of the aminoacids of its parent peptide. We surprisingly found that rGn peptides are able to bind CD13 isoforms expressed in angiogenic endothelium and, therefore, that they represent a new effective targeting motif able to target tumor vessels.

The present invention then also relates to a new therapeutic strategy against tumors based on the selective delivery of a therapeutic moiety to tumor vessels. Such selective delivery can be obtained by conjugating the rGn containing peptide to a therapeutic moiety such as an anticancer drug or a cytokine or by co-administering rGn containing peptides able to internalize tumor tissue with such moiety. Thanks to the effectiveness of the rGn motif, the therapeutic moiety will be localized at the targeted site, thus increasing efficacy and decreasing systemic side effects.

Statements of the invention According to a first aspect of the invention there is provided a peptide comprising the rGn motif wherein said peptide is a ligand of the CD13 receptor. Preferably the peptide is up to 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, or 100 aminoacids in length.

Preferably the peptide comprises the sequence XrGnX' wherein X is selected from the group consisting of L, V, A, C, G, Y, P, H, K, Q, I, I, v, a, c, g, y, p, h, k, q and i and X' is selected from the group consisting of C, G, H, L, E, T, Q, R, S, P, c, g, h, I, e, t, q, r, s, and p.

In one embodiment the peptide comprises a sequence selected from CrGnCVSGCAGRC, rGnAHA, GrGnG, CVLrGnMEC, CrGnC, CGrGnG, CrGnG, CrGnGG, CrGnCGVRSSSRTPSDKY, LrGnE, YrGnT, LQCICTGrGnGEWKCE LQCISTGrGnGEWKCE CICTGrGnGEWKC, CISTGrGnGEWKC, M RCTCVG rG nG EWTCY, M RCTSVG rG nG EWTCY CTCVGrGnGEWTC CTSVGrGnGEWTC, crGncvsGcagrc, rGnaha, GrGnG, cvlrGnmec, crGnc, cGrGnG, crGnG, crGnGG, crGncGVRSSSRTPSDKY, IrGne, yrGnt, IqcictGrGnGewkce, IqcistGrGnGewkce cictGrGnGewkc, cistGrGnGewkc, mrctcvGrGnGewtcy, mrctsvGrGnGewtcy ctcvGrGnGewtc and ctsvGrGnGewtc.

Preferably the peptide comprises a sequence selected from cycloCVLrGnMEC, linear CrGnC, cyclic CrGnC, cyclic CGrGnG, cyclic CrGnG, cyclic CrGnGG, cyclocvlrGnmec, linear crGnc, cyclic crGnc, cyclic cGrGnG, cyclic crGnG, cyclic crGnGG, linear crGncGVRSSSRTPSDKY or cyclic crGncGVRSSSRTPSDKY.

In another embodiment the peptide have a sequence selected from CrGnCVSGCAGRC, rGnAHA, GrGnG, CVLrGnMEC, CrGnC, CGrGnG, CrGnG, CrGnGG, CrGnCGVRSSSRTPSDKY LrGnE, YrGnT, LQCICTGrGnGEWKCE LQCISTGrGnGEWKCE CICTGrGnGEWKC, CISTGrGnGEWKC, M RCTCVG rGnG EWTCY, M RCTSVG rGnG EWTCY CTCVGrGnGEWTC CTSVGrGnGEWTC, crGncvsGcaGrc, rGnaha, GrGnG, cvlrGnmec, crGnc, cGrGnG, crGnG, crGnGG, IrGne, yrGnt, IqcictGrGnGewkce IqcistGrGnGewkce cictGrGnGewkc, cistGrGnGewkc, mrctcvGrGnGewtcy, mrctsvGrGnGewtcy ctcvGrGnGewtc ctsvGrGnGewtc and crGncGVRSSSRTPSDKY.

Preferably the peptide have a sequence selected from cycloCVLrGnMEC, linear CrGnC, cyclic CrGnC, cyclic CGrGnG, cyclic CrGnG, cyclic CrGnGG, linear CrGnCGVRSSSRTPSDKY, cyclic CrGnCGVRSSSRTPSDKY, cyclocvlrGnmec, linear crGnc, cyclic crGnc cyclic cGrGnG, cyclic crGnG, cyclic crGnGG, linear crGncGVRSSSRTPSDKY or cyclic crGncGVRSSSRTPSDKY.

In a further embodiment the peptides are CendR peptides which are capable to penetrate into tumor tissues. CendR peptides have been disclosed to be able to bind to neuropilin-1 and to have penetration and transportation activity through tissues; such activity can be useful for drug delivery application [Teesalu et al., (2009) PNAS]. In one embodiment the CendR peptides of the invention are peptides comprising the sequence rGnXGPX' wherein X is selected from K a nd R a nd X' is selected from D a nd E. Preferably the peptide CendR peptide have sequence selected from crGnRGPDc or CrGnRGPDC. According to another aspect of the invention there is provided a pharmaceutical composition comprising a peptide selected from crGnRGPDc or CrGnRGPDC and an antitumor agent. Preferably the antitumor agent is selected from a drug, cytokine, cytokine fragment toxin, apoptotic peptide, biological response modifier radionuclide, viral particle, gene or an imaging compound. More preferably the antitumor agent is an anticancer agent such as doxorubicin, melphalan, cis-platin, gemcitabine, taxol or a kinase inhibitor such as sunitinib, sorafenib, dasatinib, erlotinib, axitinib or lapatinib or a cytokine selected from TNFa, ΤΝΡβ, or IFNy.

In one embodiment the peptide of the present invention does not comprise an additional therapeutic agent, such as an anticancer agent or cytokine.

According to another aspect of the present invention there is provided a conjugation product comprising a peptide of the present invention. Preferably the peptide is a peptide comprising the rGn motif, including but not limited to CrGnCVSGCAGRC, rGnAHA, GrGnG, CVLrGnMEC, CrGnC, CrGnCG, CGrGnG, CrGnG, CrGnGG, CrGnRGPDC, LrGnE, YrGnT LQCICTGrGnGEWKCE LQC I STG rGnGEWKCE, CICTGrGnGEWKC, CISTGrGnGEWKC, M RCTCVG rG nG EWTCY, M RCTSVG rG nG EWTCY CTCVGrGnGEWTC, CTSVGrGnGEWTC, CrGnCGVRSSSRTPSDKY, crGncvsGcaGrc, rGnaha, GrGnG, cvlrGnmec, crGnc, crGncG, cGrGnG, crGnG, crGnGG, crGnRGPDc, IrGne, yrGnt IqcictGrGnGewkce IqcistGrGnGewkce, cictGrGnGewkc, cistGrGnGewkc, mrctcvGrGnGewtcy, mrctsvGrGnGewtcy ctcvGrGnGewtc or ctsvGrGnGewtc. More preferably the peptide have a sequence selected from cycloCVLrGnMEC, linear CrGnC, cyclic CrGnC, cyclic CGrGnG, cyclic CrGnG, cyclic CrGnGG, CrGnRGPDC, linear CrGnCGVRSSSRTPSDKY, cyclic CrGnCGVRSSSRTPSDKY, cyclocvlrGnmec, linear crGnc or cyclic crGnc, cyclic cGrGnG, cyclic crGnG, cyclic crGnGG, crGnRGPDc, linear CrGnCGVRSSSRTPSDKY or cyclic CrGnCGVRSSSRTPSDKY.

Preferably the conjugation product is between the peptide and a drug, cytokine, cytokine fragment toxin, apoptotic peptide, biological response modifier radionuclide, viral particle, gene or an imaging compound. More preferably the drug is an anticancer agent such as doxorubicin, melphalan, cis-platin, gemcitabine, taxol or a kinase inhibitor such as sunitinib, sorafenib, dasatinib, erlotinib, axitinib, or lapatinib.

More preferably the cytokine is TNF, preferably TNFa or TNF , or IFNy. The peptide can be coupled directly to the anticancer drug or the cytokine. I n one embodiment the peptide is coupled indirectly to the anticancer drug or the cytokine through a spacer, which can be a single amino acid, an amino acid sequence or an organic residue, such as 6-aminocapryl-N-hydroxysuccinimide. Preferably the spacer is a single amino acid such as glycine (G). In another embodiment the peptide is selected from cyclic peptides CGrGnG, CrGnG, CrGnGG, cGrGnG, crGnG or crGnGG, that are directly coupled to the anticancer agent through the free thiol group of the cysteine side chain.

According to another aspect of the present invention there is provided a pharmaceutical composition comprising a pharmaceutically effective amount of a conjugation product of the present invention, preferably comprising a pharmaceutically acceptable carrier, diluent or excipient.

The composition of the present invention may be in the form of an injectable solution or suspension or a liquid for infusions.

The composition of the present invention may be in the form of liposomes.

The composition of the present invention may further comprise another antitumor agent, such as, but not limited to doxorubicin or melphalan, or cis-platin or gemcitabine or taxol or a diagnostic tumor-imaging compound.

According to another aspect of the present invention, there is provided a pharmaceutica l composition comprising the peptides of the present invention and a further drug, cytokine, cytokine fragment toxin, apoptotic peptide, biological response modifier radionuclide, viral particle, gene or an imaging compound. In one embodiment the drug is an anticancer agent such as doxorubicin, melphalan, cis-platin, gemcitabine or taxol. In another embodiment the cytokine is TNF, preferably TNFa or TNF , or IFNy.

According to another aspect of the present invention there is provided use of the conjugation product or the pharmaceutical compositions of the present invention for treatment or diagnosis of a patient suffering from a cancer, such as but not limited to lung, pancreas, breast, colon, larynx or ovary cancer. Detailed description of the invention

A detailed description of preferred features and embodiments of the invention will be described by way of non-limiting example.

The invention can be put into practice by a person of ordinary skill in the art who will em ploy, unless otherwise indicated, conventiona l techniques of chemistry, molecu la r biology, microbiology, recombinant DNA and immunology. All such techniques are disclosed and explained in published literature. See, for example, J. Sambrook, E. F. Fritsch, and T. Maniatis, 1989, Molecular Cloning: A Laboratory Manual, Second Edition, Books 1-3, Cold Spring Harbor Laboratory Press; Ausubel, F.M. et al. (1995 and periodic supplements; Current Protocols in Molecula r Biology, ch. 9, 13, a nd 16, John Wiley & Sons, New York, N.Y.); Current Protocols in Immunology, ch. 12, John Wiley & Sons, New York, N.Y.); B. Roe, J. Crabtree, and A. Kahn, 1996, DNA Isolation and Sequencing: Essential Techniques, John Wiley & Sons; J . M. Polak and James O'D. McGee, 1990, In Situ Hybridization: Principles and Practice; Oxford University Press; M. J . Gait (Editor), 1984, Oligonucleotide Synthesis: A Practical Approach, Irl Press; and, D. M. J. Lil ley a nd J . E. Da hlberg, 1992, Methods of Enzymology: DNA Structure Part A: Synthesis and Physical. Analysis of DNA Methods in Enzymology, Academic Press. All these publications are incorporated by reference.

Retro-inverso concept

The present invention relates to a new targeting motif able to target angiogenic vessels, particularly tumor angiogenic vessels. The inventors found that peptides containing the retro-inverso peptide rGn, are able to specifically target CD13 isoforms expressed on angiogenic endothelium.

The targeting motif rGn is a retro-inverso of NGR peptide. Retro-inverso modification of biologically functional peptides involves the synthetic assemblage of amino acids with [alpha]-carbon stereochemistry opposite to that of the corresponding L-amino acids, i.e. D- amino acids, in reverse order with respect to the native peptide sequence. A retro-inverso analogue thus has reversed termini and reversed direction of peptide bonds. The biologic activity of a retro-inverso peptide is unpredictable. Peptide bond reversion and inversion of chirality can affect the topology of a peptide. A bioactive peptide usually has critical spatial requirements necessary to bind to its receptor. Retro-inversion can cause perturbation that can affect peptide affinity to its receptor. This effect is even more important for those peptides in which peptide bonds are involved in binding where reversal results in mismatched interaction. In the field of vascular targeting, retro-inversion of RGD containing peptides causes reduction or, in some cases, loss of activity.

With the aim of finding new ligands of CD13 receptors, the inventors surprisingly found that retro-inversion of the NGR sequence results in peptides containing the rGn sequence which maintain ability to bind CD13 isoforms expressed in angiogenic endothelium, a nd which have an higher binding affinity with such receptor with respect to the parent peptide. These effects, together with higher stability of the retro-inverso peptides, characterize the peptides of the present invention, which are new and efficient vascular targeting peptides. rGn peptides The present invention relates to peptides containing the "rGn" motif, wherein r is the D- isomer of arginine, G is glycine and n is the D-isomer of asparagine. The rGn peptide is the retro-inverso of the NGR peptide. Both peptides are ligands of CD13.

Particularly peptides containing the rGn motif are new and more efficient ligands of CD13. In fact, the rGn motif: 1. has the same ability of the NGR motif to bind CD13

2. binds the target cells with a 10-fold higher affinity than NGR

3. has higher stability than NGR because it does not undergo Asn deamidation thus preventing Asn->Asp/isoAsp transformation

Therefore, in this case, retro-inversion causes an important improvements in the activity of NGR peptides leading to new and more potent agents able to bind tumor vessels. The term "peptide" as used herein includes polypeptides and proteins. The term "polypeptide" includes single-chain polypeptide molecules as well as multiple- polypeptide complexes where individual constituent polypeptides are linked by covalent or non-covalent means. The term "polypeptide" includes peptides of two or more amino acids in length, typically having more than 5, 10 ,20, 30, 40 , 50 or 100, amino acids. rGn peptides will be used to target a therapeutic moiety to angiogenic vasculature particularly tumor angiogenic vasculature. The peptides are preferably linear or cyclic peptides comprising the rGn motif, such as but not limited to CrGnCVSGCAGRC, rGnAHA, GrGnG, CVLrGnMEC, CrGnC, CrGnCG, CGrGnG, CrGnG, CrGnGG, CrGnCGVRSSSRTPSDKY LrGnE, YrGnT LQCICTGrGnGEWKCE LQC I STG rG n G E W KC E, C I CTG rG n G E W KC, CISTGrGnGEWKC, M RCTCVG rG nG EWTCY, M RCTSVG rG nG EWTCY, CTCVGrGnGEWTC, CTSVGrGnGEWTC, crGncvsGcaGrc, rGnaha, GrGnG, cvlrGnmec, crGnc, crGncG, cGrGnG, crGnG, crGnGG, CrGnCGVRSSSRTPSDKY IrGne, yrGnt IqcictGrGnGewkce IqcistGrGnGewkce, cictGrGnGewkc, cistGrGnGewkc, mrctcvGrGnGewtcy, mrctsvGrGnGewtcy ctcvGrGnGewtc or ctsvGrGnGewtc. More preferably cycloCVLrGnMEC, linear CrGnC, cyclic CrGnC, cyclic CGrGnG, cyclic CrGnG, cyclic CrGnGG, linear CrGnCGVRSSSRTPSDKY, cyclic CrGnCGVRSSSRTPSDKY, cyclocvlrGnmec, linear crGnc cyclic crGnc, cyclic cGrGnG, cyclic crGnG, cyclic crGnGG, linear CrGnCGVRSSSRTPSDKY or cyclic CrGnCGVRSSSRTPSDKY.

Apart from the rGn motif, the other amino acids of these peptides can be both in D- or L- conformation.

It may be desirable to use derivatives of the peptides of the invention which are conformationally constrained. Such constrains maintain the peptides in the active conformation. The active conformation of the peptide may be stabilised by a covalent modification, such as cyclization or by incorporation of gamma-lactam or other types of bridges. Cyclization of CrGnC, for example, is achieved through a disulphide bridge between the two cysteines. The N-terminal domain of the peptides can optionally be acetylated.

In other cases peptides may be head to tail cyclised through the formation of a peptide bonds between their N and C terminus. Therefore, for example, cyclization of peptides CGrGnG, CrGnG, CrGnGG, cGrGnG, crGnG or crGnGG, is achieved through a peptidic bridge between N and C terminus, thus obtaining, a cyclic peptide having a free thiol group on the side chain of the cysteine. Such free thiol group can be exploited for the coupling of rGn peptides to a diagnostic or therapeutic moiety containing, for example, a maleimide functional group [Nallamothu et al. (2006) AAPS PharmSciTech]. Such cou pling ca n be obtained by a Michael addition type reaction resulting in covalent attachment of the rGn containing peptide to the therapeutic or diagnostic moiety.

In a further embodiment the rGn containing peptides are CendR peptides [Teesalu et al.;(2009) PNAS]. CendR peptides contains a C-terminal arginine (or rarely lysine) residue in a consensus context R/KXXR/K. C-terminal exposure of such consensus is an essential feature to assure the activity of such peptides that interact with Neuropilin-1 and penetrate tumor tissue. Strategies that combine ability to home and further penetrate tumors have been previously disclosed (Sugahara et al. (2009) Cancer Cells) in which the RGD motif, that homes tumor vasculature, has been incorporated into the R/KXXR/K consensus that allows tumor penetration via Neuropilin-1. A si m i la r strategy ca n be used with rG n peptides of the invention that are able to home tumor vasculature by binding CD13 receptor. In one embodiment the CendR peptides of the invention are peptides comprising the sequence rGnXGPX' wherein X is selected from K and R and X' is selected from D and E. Preferably the Cend peptide have sequence crGnRGPDc or CrGnRGPDC which are referred to herein as irGn. It has been shown that the peptides irGn are able to target tumor vessels and that cause tumor specific accumulation of a co-administered agent in the tumor.

The irGn peptides of the invention, can be used in combination with anticancer agent in order to allow tumor vessel targeting and subsequent penetration of such agent into the tumor. In one embodiment there is provided a pharmaceutical composition comprising a irGn peptide and an anticancer agent. The anticancer agent may be selected from but not limited to a drug, a cytokine, a toxin, an apoptotic peptide, a radionuclide, a viral particle, a gene or an imaging compound.

Preferably the anticancer agent is a drug is selected from doxorubicin, melphalan, cis- platin, gemcitabine taxol or a kinase inhibitor such as sunitinib, sorafenib, dasatinib, erlotinib, axitinib, or lapatinib. More preferably the anticancer agent is a cytokine selected fromTNFa, TNF , IFNy. In another embodiment the irGn peptide and the the anticancer agent are covalently linked to form a conjugate. rGn conjugates In another aspect, the present invention relates to a new therapeutic strategy against tumors based on selective delivery of therapeutic moiety to tumor vessels. Such strategy can be achived by administering conjugation products comprising the rGn containing peptides of the present invention to a therapeutic moiety.

The selective targeting obtained by the rGn motif allows localization of the therapeutic moiety at tumor vessels with a consequent improvement in therapeutic efficacy as well as a decrease of side effects.

The therapeutic moiety may be but not limited to, a drug, a cytokine, a toxin, an apoptotic peptide, a radionuclide, a viral particle, a gene or an imaging compound.

In one embodiment the drug is an anticancer agent such as doxorubicin, melphalan, cis- platin, gemcitabine taxol or a kinase inhibitor such as sunitinib, sorafenib, dasatinib, erlotinib, axitinib, or lapatinib.

In another embodiment the cytokine is selected from TNFa, TNF , IFNy.

The peptide can be coupled directly to the cytokine or to the anticancer agent or indirectly through a spacer, which can be a single amino acid, an amino acid sequence or an organic residue, such as 6-aminocapryl-N-hydroxysuccinimide. The coupling procedures are known to those skilled in the art and comprise genetic engineering or chemical synthesis techniques.

In one embodiment the peptide or targeting moiety is linked to the cytokine N- terminus or C-terminus thus minimising any interference in the binding of the modified cytokine to its receptor. Alternatively, the peptide or targeting moiety can be linked to amino acid residues which a re amido- or ca rboxylic-bond acceptors, which may be natura lly occurring on the molecule or artificially inserted using genetic engineering techniques.

The peptides, conjugates and compositions of the invention may be used in therapeutic treatment. It is to be appreciated that all references herein to treatment include curative, palliative and prophylactic treatment.

I n one embodiment the peptides, conjugates or pharmaceutica l compositions may be used to treat or prevent cancer including but not limited to cancer of the lung, pancreas, breast, colon, larynx or ovary. Preferably the cancer comprises a solid tumor or a ny tumor expressing CD13 in angiogenic tumor vasculature. I n an additiona l related embodiment, a tissue to be treated is a tumor tissue of a patient with a solid tumor, a metastases, a ski n cancer, a breast cancer, a hemangioma or angiofibroma a nd the like ca ncer, and the angiogenesis to be inhibited is tumor tissue angiogenesis where there is neovascula rization of a tumor tissue. Typical solid tumor tissues treatable by the present methods include lung, pancreas, breast, colon, laryngeal, ovaria n, and the like tissues. The peptides, conjugates a nd pha rmaceutical compositions of the invention can be used in combined, separated or sequential prepa rations, also with other diagnostic or thera peutic substances such as, but not limited to doxorubicin or melpha lan, or cis-platin or gemcitabine or taxol or a diagnostic tumor-imaging compound. Co-administration of peptides containing rGn motif with a therapeutic moiety can be a further way to obtain a selective delivery. Description of the figures

Figure 1. The rGn and NGR peptides have the same recognition pattern. Adherent GR4 (upper panels) and MR300 (lower pa nels) cells were incubated with biotinylated rGn a nd NGR peptides followed by STV-Qd. Magnification, 400x; red, Qd blue, nuclear staining with DAPI . CD13 expression was evaluated by FACS a nalysis with the CD13-specific mAb WM15 (green lines)

Figure 2. Competitive binding of la beled rGn peptide with rGn, NGR and isoDGR peptides to MR300 cells. Single cell suspension (A) or adherent (B) MR300 cells were stained with rGn- biotin-STV-Qd alone (none) or in the presence of an excess of either rGn, NGR or isoDGR peptides. Cells were then analyzed by (A) FACS and the results reported as mean fluorescence intensity (MFI); or (B) by fluorescence microscopy assays, Magnification, 400x; red, Qd; blue, nuclear staining with DAPI . Figure 3. The rGn and NGR peptides bind to tumor-associated vessels. Frozen sections of murine breast carcinoma (N202) a nd human renal cell carcinoma were incubated with biotinylated rGN and NGR peptides followed by STV-Qd and immunostained with FITC- labeled a nti-CD31 mAb. Nuclear staining with DAPI .

Figure 4. In vivo administered rGn and NGR peptides bind to tumor-associated vessels. Qd- conjugated peptides were administered to mice bearing the ABl mesothelioma. Fresh tumor sections were then stained with anti-CD31 mAb.

Figure 5. In vivo administered rGn and NGR peptides compete for the binding to tumor- associated vessels. (A) Qds conjugated to rGn peptides were administered a lone or in combination with an excess of free NGR or rGn peptides, to mice bearing RMA tumors. Fresh tumor sections were then stained with anti-CD31 mAb. (B) Qds conjugated to NGR peptides were administered alone or in combination with an excess of free rGn, NGR and unrelated SGR peptides to tumor-bearing. Fresh tumor sections were then stained with anti-CD31 mAb.

Figure 6. Binding affinity of the NGR peptide is affected by acetylation of the N-terminal a- amino group. Competitive binding of NGR-Qd with various doses NGR or acetyl-NGR petides. Magnification, 400x; red, Qd; nuclear staining with DAPI .

Figure 7. Binding affinity of the rGn vs. NGR peptide tested using as probe either the rGn-Qd (left panel) or the NGR-Qd (right panel).

Figure 8. rGn binding to the target cells requires CD13 expression. Mesangioblasts were transduced with lentiviral vectors to express CD13-specific and control shRNAs. Transduced cells were seeded on slides and stained with the anti-CD13 WM15 mAb, NGR-biotin and rGn- biotin peptides, followed by fluorescent -streptavidin. The binding was quantified by the CellF program and reported as arbitra ry units. Figure 9. (A) NGR and rGn peptides specifically pull down CD13. Immunoblot for CD13 (left panel) and aV (right panel), of pulled-down proteins from mesangioblasts by either isoDGR, NGR or rGn biotinylated peptides. (B) NGR and rGn peptides directly interact with CD13 on the cell membrane. Immunoblot for CD13 (left panel) and extravidine (right panel), of CD13 molecules immunoprecipitated by anti-CD13 mAb from mesangioblasts previously labeled with either isoDGR, NGR or rGn biotinylated peptides. (C) The 150 kDa band pulled-down by rGn peptide contains only CD13 molecules. Immunoblot for CD13 (left panel) and extravidin (right panel) of total cellular lysate of mesangioblasts previously labeled with the rGn biotinylated peptide (input), CD13 isolated molecules (CD13 IP), and CD13-cleared lysate (unbound).

Figure 10. Differential stability of rGn and NGR peptides. MALDI-TOF mass spectrometry analysis of the reported peptides after incubation at 37°C for 20 hours in PBS.

Figure 11. Binding of untreated and heat-treated rGn and NGR peptides to ανβ3 integrin. The biotinylated peptides were treated as described in figure 6 and then added to microtiter plates coated with ανβ3. After washing, the binding was detected by chromogenic reaction.

Figure 12. Toxic effect of repeated administrations of rGn or NGR peptide (15 mg/kg, i.p.) to CT26 tumor-bearing mice. Animals (six/group) were treated daily and body weights measured.

Figure 13. Tumor-specific entry of Evans Blue into tumor tissue in irGn-injected mice. Mice bearing the AB1 mesothelioma were intravenously injected with 1 mg of Evans Blue, followed 5 minutes later by 4 μιηοΙ/kg of irGn peptide or vehicle alone. Tumors were collected 30 minutes later and the amount Evans Blue was then measured. Evans Blue specifically accumulates in the tumor when co-administered with irGn peptide.

Examples Example I: General methods

Synthesis and folding of rGn containing peptides To perform binding studies the peptides crGnc-G-VRSSSRTPSDYK and CNGRC-G- VRSSSRTPSDYK were synthesized, comprising the crGnc and the CNGRC targeting motives , one glycine(G) and the first 10 aminoacids of the human TNF (VRSSSRTPSD) as spacer, one tyrosine (Y) to allow spettrofotometric quantification of the peptides, and one lysine (K) to allow peptide conjugation to biotin and/or Qds.

All the peptides used in this study, were assembled manually using the stepwise solid phase Fmoc method onto a preloaded Tyrosine-2-Chlorotrityl resin from Novabiochem (Laufelfingen, CH): The Ν-α-9-Fluorenylmethoxycarbonyl amino acid used for the synthesis had the following side-chain protecting groups: trityl for Cysteine and Asparagine, tert-butyl fo r As pa rti c Aci d, Se ri n e, Th re o n i n e a n d Ty rosi n e, a n d 2, 2, 4, 6, 7,-pentamethyl- dihydrobenzofuran-5-sulfonyl for Arginine, t-Boc for Lysine. After completion of the synthesis, Fmoc-deprotected peptides were cleaved f ro m t he resi n a n d sid e ch a i n deprotected, at room temperature by treatment with a mixture of, 5% water, 5% phenol, 5% thioanisole, 2.5% ethandithiol, 2.5% triisopropylsilane 80% and trifluoroacetic acid (reagent K) for 3 h. The resins were filtered and the peptide solutions were added in drops to cold tert-butylmethyether to precipitate the peptides. After centrifugation and washing three times with tert-butylmethyether, the peptides were suspended in 5% aqueous acetic acid and lyophilized.

Peptide purification: Analytical and semipreparative Reversed Phase High Performance Liquid Chromatography (RP-HPLC) were carried out on a Tri Rotar-VI HPLC system equipped with a MD-910 multichannel detector for analytical purposes or with a Uvidec-100-VI variable UV detector for preparative purpose (all from JASCO, Tokyo, Japan). A Jupiter 5u C18 90A column (150 x 4.6 mm) was used for analytical runs and a Jupiter lOu C18 90A (250 x 21.2 mm) for peptide purification. (The chromatographic columns were from Phenomenex, Torrance CA, USA)

Peptide folding: The purified peptide solutions from the semipreparative column, were diluted with water to a final peptide concentration of 100 micromolar, the pH adjusted with 2M Sodium Hydroxide to neutralize the trifluoroacetic acid, Tris.HCI 1M, pH 8 buffer was added to a final concentration of 20 mM and to the solutions were added two equivalents of hydrogen peroxide (rispetto ad un equivalente di peptide) and allowed to stir. The oxidative folding was followed by analytical RP-HPLC and the folded oxidized monomeric peptides eluted earlier from the column respect to the reduced peptides and few polymeric material eluted later. After completion of the folding (30-60 minutes), and no free sulphydryl groups were detectable as assessed by Ellman test (4), the peptide solutions were buffered to pH 2.2 with phosphoric acid and loaded by the chromatographic pump on the semi-preparative column and purified from the polymeric material. Finally the peptides were U.V. quantified, transferred in vials and liophilized.

Cells Peptide binding assay: Conjugation of peptides to QD605 was performed as detailed in Curnis F., J. Biol Chem. 2010. 4-6 x 105 cells were plated on glass chamber slides (BD Falcon). After 48 hours, staining was performed incubating the cells with QD605-labeled peptides for 2 hours at room temperature in binding buffer. The cells were then fixed with paraformaldeide solution and nuclei were stained with DAPI. Images were acquired on fluorescence microscopy (Olympus BX71) using the CellF software.

Competition studies: For competition experiments the NGR-Qd/ rGn-Qd staining was performed as described above in the presence of various amounts of competitor peptides. Images were acquired and the staining was quantified using the CellF software

Binding to integrin ανβ3: biotinylated rGn and NGR peptides were incubated at 37°C for 6 hours in Ammonium Bicarbonate (0.1M) and 0.1% BSA (i.e. heat treated condition) and then added to microtiterplates coated with ανβ3. After washing, the binding was detected by chromogenic reaction.

In vivo studies

Binding to tumor vessels: Conjugation of peptides to QD605 was performed as detailed in Curnis F., J. Biol Chem. 2010. Qds (20 μΙ) alone or conjugated to rGn, NGR or to unrelated SGR peptides, were diluted in 200 μΙ of saline and injected intravenously in tumor-bearing mice. After 10 min of circulation, the mice were perfused (3ml/min) with saline for 5 min, then tumors were collected. Ana lysis of tumor tissues in whole mount was performed after incubation with FITC-anti-CD31 mAb that specifica lly stains blood vessels.

Competition studies: For competition experiments the NGR-Qd/ rGn-Qd infusion in tumor- bea ri ng mice was performed as descri bed a bove, i n the presence of a n excess (800 μg) of free competitor peptides. After 10 min of ci rculation, mice were perfused (3m l/mi n) with saline for 5 min, then tumors were collected. Analysis of tumor tissues in whole mount was performed as described above.

I nternalization experiments: Tumor mice were injected intravenously with 100 μΙ of sa li ne containing 1 mg of Eva ns Blue fol lowed 5 minutes later by 4 μιηοΙ/kg of irGn (CrGnRGPDC) peptide in PBS, or PBS alone. After 30 min of circulation, the mice were perfused with saline containing 1% BSA and heparin (50U/ml), and tissues were collected. For Evans Blue quantification, the dye was extracted from tissues in N,Ndimethylformamide for 24 hours at 37°C and quantified by measuring the absorbance at 600 nm

Example I I : CD13 targeting Evaluation of crGnc binding ability in vitro and in vivo

To evaluate the recognition pattern of the crG nc (hereafter reported as rGn) peptide motif, cells and tissues, previously cha racterized for their NGR binding ability, were stained by incubation with the biotynilated peptides, synthesized as previously described, followed by streptavidin (STV) labeled with different fluorochromes. I n some experiments the peptides were conjugated to qua ntum-dots (Qd). The results of the staining performed on adherent cells from primary cultures (HUVEC and mesangioblasts) and established cell lines showed that the NGR and rGn peptides have the same recognition pattern (table 1).

I n particula r, GR4 cells did not bind the peptides even if they express high level of CD13 (figure 1 plot on the left), in agreement with the concept that the NGR peptide binds only a defined CD13 isoform (Curnis et al Cancer Res 2002), and suggesting that the two peptides recognized the same ligand. To further confirm this hypothesis, competition experiments were performed. The binding of biotinylated rGn peptide to MR300 cells was competed by a 20-fold excess of either rG n or NG R pe ptide, but not of the u nre lated isoDG R peptide (figu re 2A) . Si m i la r results were obtained on adherent MR300

Table 1 Peptide reactivity on human adherent cel

cells (figure 2B).

Cell Lines Histotype NGR rGn

MR300 NSCLC ++ ++

665/2 melanoma

GR4 melanoma - -

ES2 ovarian carcinoma + +

SKOV3 ovarian carcinoma + +

Hela cervix carcinoma - -

PC3 prostate carcinoma + +

JU77 mesothelioma + +

One-58 mesothelioma + 10

Primary Cells Histotype NGR rGn

HUVEC endothelial cells + +

MB mesangioblasts ++ ++

To characterize the pattern of reactivity of the rGn peptide on tumor tissues, sections of a human renal cell carcinoma (figure 3, lower panels) were stained with the rGn peptide (in red) and with an anti-CD31 mAb, recognizing endothelia l cel ls of blood vessels (in green). The reactivity of the NGR peptide was tested on a murine breast carcinoma (i.e. N202; figure 3, upper panels) in the same experimental conditions. As expected, both the peptides preferentia lly recognize the CD31 positive tumor vessels.

The in vivo homing of NG R a nd rG n peptides to tumor vessels was i nvestigated by the administration of Qd-labeled peptides to tumor-bearing mice.

After perfusion, the tumor masses were recovered and analysed by whole mount histology after ex vivo staining with anti-CD31 mAb. In vivo both the rGn and the NGR peptides specifica lly bind tumor associated vessels (figure 4). To further confirm in vivo, that the two peptides recognized the same ligand, rGn-Qds were administered alone (fig 5A right panels) or with an excess (800 μg) of unlabelled NGR (figure 5A left panels) or rGn (figure 5B middle panels) peptides to tumor-bea ring mice. The co-administration of free NGR a nd rGn peptides abolished the binding of rGn-Qds to tu mor vessels, th us demonstrating that the two peptides recognized the same ligand. These results were further confirmed by competing the binding of NGR-Qds with unlabelled rGn and NGR peptides. Indeed, both peptides inhibited the binding of NGR-Qds, whereas the unrelated SGR peptide did not (figure 5B).

Altogether, these data suggest that the NGR and the retro-inverso rGn peptides recognize the same receptor on the cell surface of target cells both in vitro and in vivo.

Affinity studies

Previous studies on the binding of NGR peptides to adherent cells showed that acetylation of the N-terminal a-amino group can significantly modify the binding properties of this peptide.

In particular, we analyzed the capability of NGR and acetyl-NGR to compete the binding of NGR-Qd to MR300 cells. We found that NGR inhibits the binding of NGR-Qd 3-fold more efficiently than acetyl-NGR (figure 6), thus suggesting that the acetyl-NGR motif binds endothelial cells with a lower affinity.

In line with this observation, only the binding of the non-acetylated NGR peptide can be detected by FACS analysis of single cell suspension of MR300 cells (table 2). Interestingly, in this assay, the rGn peptide stained non-adherent cells more intensely than the NGR peptide, thus suggesting that the retro-inverso peptide bind the cell surface with higher affinity than the NGR peptide.

Table 2 FACS analysis of single cell suspension

Cell Lines Mean Intensity Fluorescence (MFI)

Acetil-NGR NGR rGn

MR300 0 21 69

GR4 0 0 0

To formally validate this hypothesis, we performed competition studies showing that the IC₅₀ of the rGn motif is about 10-fold higher than that of the NGR peptide. These results were obtained using as probe either the rGn itself or the NGR peptide (figure 7).

Characterization of the CD13/rGn peptide interaction

To further characterize the binding properties of the rGn peptide, the specificity of the ligand-peptide interaction was validated by the use of CD13-silenced cells obtained by lentiviral vector transduction with appropriate shRNAs. Transduced cells were stained with either anti-CD13 mAb, NGR or rGn peptides, and the binding was quantified by the CellF program. Transduction of primary mesangioblasts with CD13-specific (figure 8, yellow bar), unlike with control shRNAs (figure 8, blue bar), inhibited expression of CD13, as well as, the binding of rGn and NGR peptides. Binding of _acisoDGR peptide, specific for integrins, was not affected (figure 8).

The physical interaction of the rGn motif with the CD13 molecules expressed on the surface of the target cells was demonstrated by pull-down experiments. Adherent MR300 (data not shown) and mesangioblasts (figure 9) were incubated with either biotynilated isoDGR, NGR or rGn peptides. Following protein crosslinking with Bis(Sulfosuccinimidyl) suberate (BS3), cells were lysed, and the peptide-bou nd protei ns we re isolated by streptavidin-conjugated magnetic beads and analyzed by immunobloting for av and CD13 after SDS-PAGE separation.

As shown in figure 9 panel A, the monomeric and the dimeric form of CD13 were pulled- down by both the NGR and the rGn peptide, and not by the isoDGR peptide (right panel), which specifically recognize the av-integrin (left panel). Noteworthy, in agreement with the affinity studies, the amount of CD13 isolated by the rGn peptide was more than that bound by the NGR peptide.

Equal amounts of CD13 were immunoprecipitated (IP) by CD13-specific mAb from cellular lysates of mesangioablasts previously labelled with either isoDGR, NGR or rGn biotinylated peptides (figure 8B left panel). The presence of CD13-linked biotinylated peptides was detected by extravidine hybridization, only in association with the CD13 isolated from NGR and rGn labelled mesangioblasts (figure 9B right panel). These results formally demonstrate that NGR and rGn peptides directly interact with CD13 on the cell membrane.

Of note, when CD13 was completely depleted from lysates of mesangioblasts labelled with the rGn biotinylated peptide (unbound in figure 9C left panel), the biotinylated 150 kDa band was no more detectable (unbound figure 9C right panel). These data demonstrate that the 150 kDa band pulled-down by the rGn peptide contains CD13 molecules.

Example III : Stability Studies It has recently been demonstrated that the Asn residue of NGR can rapidly deamidate and generate Asp and isoAsp residues. This spontaneous reaction occurs by nucleophilic attack of the backbone NH center on the Asn side-chain leading to formation of a succinimide intermediate (Geiger, JBC 1987). Hydrolysis of succinimide leads, in turn, to formation of mixtures of isoDGR and DGR, with an overall gain of 1 Da (Curnis JBC 2006). The transition of NGR to isoDGR/DGR is associated with change of the ligand pattern of the peptide, from CD13 to integrin ανβ3 (Curnis JBC 2006; Spitaleri 2008). To verify whether deamidation also occurs in the rGn peptide, peptide degradation was monitored by MALDI-TOF mass spectrometry. Storage of CNGRCGVRSSSRTPSDKY-biotin peptide at 37 °C for 6 hours in ammonium bicarbonate (0.1M) and 0.1% BSA (the so-ca lled "heat-treated" condition), generated a large amount of a product characterized by a gain of 1 Da (figure 10), likely corresponding to a isoDGR/DGR mixture. Conversely, storage of bioti n-Ahx-Ahx-kyrvgcrG nc- coh2 under the same conditions did not modify the mass-spectrometry profile (figure 11). This observation was then verified at functional level, looking at the ability of the untreated and heat-treated compounds to bind recombinant ανβ3 integrin. In agreement, with the mass spectrometry data, it was found that only the NGR peptide could generate functional molecules in these conditions (figure 11), thus demonstrating that the retro- inverso peptide did not undergo deamidation. The same results were obtained using Qdot labeled CNGRCGVRSSSRTPSDKY peptide, which deamidates as expected, and crGncGVRSSSRTPSDKY peptide that is not prone to deamidation.

Along with the high metabolic stability of retro-inverso peptides, which contain D aminoacid bonds that are stable to enzymatic cleavage, the inability of rGn to deamidate is an additional advantage of the retro-inverso rGn peptide.

Example IV: Evaluation of rGn toxicity in vivo The potential toxicity of the rGn peptide was investigated. CT26 colon carcinoma-bearing mice were treated with NGR and rGn peptides (300 μg, i.p.) daily for five days. Neither loss of weight (figure 12), nor other toxicities were observed, thus demonstrating that the rGn peptide per se is safe.

Example V: Evaluation of the irGn tumor-penetrating activity in vivo To investigate the tumor-penetrating activities of the irGn retroinverso peptide (CrGnRGPDC), a murine mesothelioma model was used. As shown in figure 13, the chemically synthesized irGn peptide, when co-injected with the albumin binding dye Evans blue, caused tumor specific accumulation of the dye in the tumor.

References

Ausubel, F.M. et al., 1995 and periodic supplements; Current Protocols in Molecular Biology, ch. 9, 13, and 16, John Wiley & Sons, New York, N.Y., Current Protocols in Immunology, ch. 12, John Wiley & Sons, New York, N.Y.

Chorev M, Goodman M. Recent developments in retro peptides and proteins-an ongoing topochemical exploration. Trends Biotechnol 1995;13(10):438-45.

Corti A, Curnis F, Arap W, Pasqualini R. The neovasculature homing motif NGR: more than meets the eye. Blood 2008;112(7):2628-35.

Curnis F, Arrigoni G, Sacchi A, et al. Differential binding of drugs containing the NGR motif to CD13 isoforms in tumor vessels, epithelia, and myeloid cells. Cancer Res 2002;62(3):867-74.

Curnis F, Cattaneo A, Longhi R, et al. Critical role of flanking residues in NGR-to- isoDGR transition and CD13/integrin receptor switching. J Biol Chem

2010;285(12):9114-23.

Curnis F, Longhi R, Crippa L, et al. Spontaneous formation of L-isoaspartate and gain of function in fibronectin. J Biol Chem 2006;281(47):36466-76.

Curnis F, Sacchi A, Borgna L, Magni F, Gasparri A, Corti A. Enhancement of tumor necrosis factor alpha antitumor immunotherapeutic properties by targeted delivery to aminopeptidase N (CD13). Nat Biotechnol 2000;18(ll):1185-90.

Dal Pozzo A, Fagnoni M, Bergonzi R, et al. Synthesis and anti-aggregatory activity of linear retro-inverso RGD peptides. J Pept Res 2000;55(6):447-54.

Fletcher MD, Campbell MM. Partially Modified Retro-lnverso Peptides: Development,

Synthesis, and Conformational Behavior. Chem Rev 1998;98(2):763-96.

Freidinger RM, Veber DF. Peptides and their retro enantiomers are topological^ nonidentical. J Am Chem Soc 1979;101(20):6129-31.

Geiger T, Clarke S. Deamidation, isomerization, and racemization at asparaginyl and aspartyl residues in peptides. Succinimide-linked reactions that contribute to protein degradation. J Biol Chem 1987;262(2):785-94.

Gregorc V, Citterio G, Vitali G, et al. Defining the optimal biological dose of NGR- hTNF, a selective vascular targeting agent, in advanced solid tumours. Eur J Cancer 2010;46(l):198-206.

Gregorc V, Santoro A, Bennicelli E, et al. Phase lb study of NGR-hTNF, a selective vascular targeting agent, administered at low doses in combination with doxorubicin to patients with advanced solid tumours. Br J Cancer 2009;101(2):219-24.

Gregorc V, Zucali PA, Santoro A, et al. Phase II Study of Asparagine-Glycine-Arginine- Human Tumor Necrosis Factor {alpha}, a Selective Vascular Targeting Agent, in Previously Treated Patients With Malignant Pleural Mesothelioma. J Clin Oncol 2010. D. M. J. Lilley and J. E. Dahlberg, 1992, Methods of Enzymology: DNA Structure Part A Synthesis and Physical. Analysis of DNA Methods in Enzymology, Academic Press. Pasqualini R, Ruoslahti E. Organ targeting in vivo using phage display peptide libraries. Nature 1996;380(6572):364-6.

J. M. Polak and James O'D. McGee, 1990, In Situ Hybridization: Principles and

Practice; Oxford University Press; M. J . Gait (Editor), 1984, Oligonucleotide Synthesis A Practical Approach, Irl Press. B. Roe, J. Crabtree, and A. Kahn, 1996, DNA Isolation and Sequencing: Essential Techniques, John Wiley & Sons.

J. Sambrook, E. F. Fritsch, and T. Maniatis, 1989, Molecular Cloning: A Laboratory Manual, Second Edition, Books 1-3, Cold Spring Harbor Laboratory Press.

Santoro A, Pressiani T, Citterio G, et al. Activity and safety of NGR-hTNF, a selective vascular-targeting agent, in previously treated patients with advanced hepatocellular carcinoma. Br J Cancer 2010;103(6):837-44.

Santoro A, Rimassa L, Sobrero AF, et al. Phase II study of NGR-hTNF, a selective vascular targeting agent, in patients with metastatic colorectal cancer after failure of standard therapy. Eur J Cancer 2010;46(15):2746-52.

Spitaleri A, Mari S, Curnis F, et al. Structural basis for the interaction of isoDGR with the RGD-binding site of alphavbeta3 integrin. J Biol Chem 2008;283(28):19757-68. Teesalu T, Sugahara KN, Kotamraju VR, Ruoslahti E., et al. C-end rule peptides mediate neuropilin-l-dependent cell, vascular, and tissue penetration. Proc Natl Acad Sci U S A. 2009 Sep 22;106(38):16157-62

van Laarhoven HW, Fiedler W, Desar IM, et al. Phase I clinical and magnetic resonance imaging study of the vascular agent NGR-hTNF in patients with advanced cancers (European Organization for Research and Treatment of Cancer Study 16041). Clin Cancer Res 2010;16(4):1315-23.

Sugahara KN, Teesalu T, KarmaliPP, Kotamraju VR, et al. Tissue-penetreting delivery of compounds and nanoparticles into tumors. Cancell Cell. 2009; 16(6):510-520. Nallamothu R, Wood GC, Pattillo CB, A tumor vasculature targeted liposome delivery system for Combretastatin A4: design characterization and in vitro evaluation

Claims

1. A peptide comprising the rGn motif, wherein such peptide is a ligand of the CD13 receptor.

2. A peptide according to claim 1 wherein such peptide is up to 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, or 100 aminoacids in length.

3. A peptide according to claim 1 or 2 wherein such peptide comprises the sequence XrGnX' wherein X is selected from the group consisting of L, V, A, C, G, Y, P, H, K, Q, I, I, v, a, c, g, y, p, h, k, q and i and X' is selected from the group consisting of C, G, H, L, E, T, Q, R, S, P, c, g, h, I, e, t, q, r, s, and p.

4. A peptide according to anyone of claims 1 to 3 comprising a sequence selected from CrGnCVSGCAGRC, rGn AH A, GrGnG, CVLrGnMEC, CrGnC, CGrGnG, CrGnG, CrGnGG, CrGnRGPDC, CrGnCGVRSSSRTPSDKY L rG n E, Y rG nT, LQC I CTG rG n G E W KC E LQCISTGrGnGEWKCE CICTGrGnGEWKC, CISTGrGnGEWKC, M RCTCVG rG nG EWTCY, M RCTSVG rG nG EWTCY CTCVGrGnGEWTC CTSVGrGnGEWTC, crGncvsGcagrc, rGnaha, GrGnG, cvlrGnmec, crGnc, cGrGnG, crGnG, crGnGG, crGnRGPDc, crGncGVRSSSRTPSDKY, IrGne, yrGnt, IqcictGrGnGewkce, IqcistGrGnGewkce cictGrGnGewkc, cistGrGnGewkc, mrctcvGrGnGewtcy, mrctsvGrGnGewtcy ctcvGrGnGewtc and ctsvGrGnGewtc.

5. A peptide according to claim 4 comprising a sequence selected from cycloCVLrGnMEC, linear CrGnC, cyclic CrGnC, cyclic CGrGnG, cyclic CrGnG, cyclic CrGnGG, CrGnRGPDC, CrGnCGVRSSSRTPSDKY, cyclocvlrGnmec, linear crGnc, cyclic crGnc , cyclic cGrGnG, cyclic crGnG, cyclic crGnGG, crGncGVRSSSRTPSDKY and CrGnRGPDC.

6. A conjugation product comprising a peptide according to anyone of claims 1 to 5 and a therapeutic moiety.

7. A conjugation product according to claim 6 wherein the therapeutic moiety is selected from a drug, a cytokine, cytokine fragment toxin, apoptotic peptide, biological response modifier radionuclide, viral particle, gene, or an imaging compound.

8. A conjugation product according to claim 7 wherein the the drug is selected from doxorubicin, melphalan, cis-platin, gemcitabine, taxol or a kinase inhibitor.

9. A conjugation product according to claim 7 wherein the cytokine is selected from TNFa or TNF , or lFNy.

10. A conjugation product according to anyone of claims 6 to 9 wherein the peptide is linked to the therapeutic moiety through a spacer.

11. A conjugation product according to claim 10 wherein the spacer is the a mino acid glycine.

12. A pharmaceutical composition com prising a pha rmaceutically effective a mount of a conjugation product according to any one of claims 6 to 11 and a pharmaceutically acceptable carrier, diluent or excipient thereof.

13. A pharmaceutical composition according to claim 12 in the form of an injectable solution or suspension or a liquid for infusions.

14. A pharmaceutical composition according to anyone of claims 12 or 13 further comprising another antitumor agent selected from doxorubicin, melphalan, cis-platin, gemcitabine, taxol or a kinase inhibitor a diagnostic tumor-imaging compound.

15. A pharmaceutical composition comprising a peptide comprising rGn motif, wherein such peptide is a ligand of CD13 receptor, and an anticancer agent

16. A pharmaceutical composition according to claim 15 wherein the peptide is selected from crGnRGPDc or CrGnRGPDC

17. A pharmaceutical composition according to anyone of claims 15 or 16 wherein the anticancer agent selected from drug, a cytokine, cytokine fragment toxin, apoptotic peptide, biological response modifier radionuclide, viral particle, gene, or an imaging compound

18. A pharmaceutical composition according to claim 17 wherein the drug is selected from doxorubicin, melphalan, cis-platin, gemcitabine, taxol or a kinase inhibitor.

19. A pharmaceutical composition according to claim 17 wherein the cytokine is selected from TNFa or TNF , or lFNy.

20. Use of the conjugation product according to claims 6 to 11 or of the pharmaceutical composition according to claims 12 to 19 for treating or diagnosing a patient suffering from a cancer.