COMPOSITIONS AND METHODS RELATED TO PEPTIDES THAT SELECTIVELY BIND LEUKEMIA CELLS

BACKGROUND OF THE INVENTION This application claims the priority of U.S. Provisional Application Ser. No. 60/586,814 filed July 10, 2004, the entire disclosure of which is specifically incorporated herein by reference.

The government owns rights in the current invention pursuant to the National Institutes of Health grant number CAl 00632.

I. FIELD OF THE INVENTION The present invention concerns the fields of molecular medicine and targeted delivery of therapeutic agents. More specifically, the present invention relates to compositions and methods for identification and use of peptides that selectively target leukemia cells.

II. DESCRIPTION OF RELATED ART In the past few decades, advancements have been made to the conventional systemic therapy of leukemia resulting in improved patient prognosis and survival. However, due to drug resistance and non-specific cytotoxicity, treatment of leukemia is often associated with adverse, long-term, side effects, and significant relapse and mortality rates. Much attention has been focused on the development of targeted therapies for leukemia. One such approach is to target cell adhesion in leukemia based on the rationale that clinical progression of the disease can be attenuated by controlling adhesion to the extracellular matrix. Normal leukocytes use cell adhesion, and interactions with the extracellular matrix, through the vascular and lymphatic systems to home to selective lymphoid organs. β2 integrins are important receptors mediating the homing of leukocytes. A peptide has been identified that blocks β2-mediated adhesion without significant toxicity. Koivunen et al. (2001) reported the screening of a phage display random peptide libraries for ligands that would inhibit the function of the β2 integrin receptor as it related to leukocyte adhesion. A peptide ligand with a compact disulfide-restrained structure (ADGACPCFLLGCCGA (SEQ ID NO: 8) termed LLG-C4) was identified as an inhibitor of leukocyte cell adhesion and migration.

The LLG motif supported cell adhesion when immobilized in vitro and bound only to cells expressing β2 integrins. The striking specificity of the peptide is explained by its ability to interact with the I-domain (the active site in leukocyte integrins) and with ICAM-I (a functional ligand for β2 integrins). It is well recognized that β2 integrins only become active after selective stimulation by inflammatory cytokines and contact with antigen-presenting cells. Selective inhibition of β2 function would likely prevent or attenuate the progression of leukemia. This peptide binds strongly to acute myelogenous leukemia (AML)-derived cell lines and to the AML cell-enriched leukocyte population isolated from blood of AML patients with no significant binding to leukocytes of healthy donors. In an in vivo leukemia model, the LLG-C4 peptide significantly improved the survival of mice xenografted with human OCI-AML3 cells. Additionally, the peptide prevented the attachment and proliferation of the leukemia cells on growth-supporting mesenchymal cell layer. Despite the promising preliminary data LLG-C4 peptide has properties that are difficult to work with and is poorly soluble in aqueous environments.

SUMMARY OF THE INVENTION The present invention provides additional methods and compositions for the preparation and use of targeting peptides that are selective and/or specific for leukemia cells. In some embodiments, the invention concerns particular targeting peptides selective or specific for leukemia cells, including but not limited to AYHRLRR (SEQ ID NO: 5), GFYWLRS (SEQ ID NO: 6), or SFFYLRS (SEQ ID NO: 7). Other embodiments concern such targeting peptides attached to therapeutic agents. In other embodiments, leukemia or other targeting peptides may be used to selectively or specifically deliver therapeutic agents to target cells, such as leukemia cells. In certain embodiments, the subject methods concern the identification, preparation, and use of targeting peptides selective or specific for a given target cell, tissue or organ, such as leukemia cells.

One embodiment of the invention concerns isolated peptides of 100 amino acids or less in size, comprising at least 3, 4, 5, 6, or 7 contiguous amino acids of a leukemia targeting peptide sequence. In certain aspect of the invention the contiguous amino acids are selected from any of SEQ ID NO: 5, SEQ ID NO: 6, or SEQ ID NO: 7. In a preferred embodiment, the isolated peptide is 50 amino acids or less, more preferably 30 amino acids or less, more preferably 20 amino acids or less, more preferably 10 amino acids or less, still more preferably 7 amino acids or less in size, or even still more preferably 5 amino acids or less in size. In other preferred embodiments, the isolated peptide may comprise at least 4, 5, 6, or 7 contiguous amino acids of a targeting peptide sequence, selected from a leukemia cell targeting peptide including, but not limited to SEQ ID NO: 5, SEQ ID NO: 6, or SEQ ID NO: 7.

In certain embodiments, the isolated peptide may be operatively coupled to an agent. In preferred embodiments, the isolated peptide is covalent coupled to an agent. In various embodiments, the agent is a drug, a chemo therapeutic agent, a radioisotope, a pro- apoptosis agent, an anti-angiogenic agent, a hormone, a cytokine, a growth factor, a cytotoxic agent, a peptide, a protein, an antibiotic, an antibody, a Fab fragment of an antibody, a survival factor, an anti-apoptotic factor, a hormone antagonist, an imaging agent, a nucleic acid or an antigen. These agents are representative only and virtually any agent may be attached to a leukemia targeting peptide and/or administered to a subject within the scope of the invention. In preferred embodiments, the pro-apoptosis agent is gramicidin, magainin, mellitin, defensin, cecropin, (KLAKLAK)2 (SEQ ID NO:1), (KLAKKLA)2 (SEQ ID NO:2), (KAAKKAA)2 (SEQ ID NO:3) or (KLGKKLG)3 (SEQ ID NO:4). In other preferred embodiments, the anti-angiogenic agent is angiostatin 5, pigment epithelium-derived factor, angiotensin, laminin peptides, fϊbronectin peptides, plasminogen activator inhibitors, tissue metalloproteinase inhibitors, interferons, interleukin 12, platelet factor 4, IP-10, Gro-β, thrombospondin, 2-methoxyoestradiol, proliferin-related protein, carboxiamidotriazole, CMlOl, Marimastat, pentosan polysulphate, angiopoietin 2 (Regeneron), interferon- alpha, herbimycin A, PNU145156E, 16K prolactin fragment, Linomide, thalidomide, pentoxifylline, genistein, TNP-470, endostatin, paclitaxel, docetaxel, polyamines, a proteasome inhibitor, a kinase inhibitor, a signaling inhibitor (SU5416, SU6668, Sugen, South San Francisco, CA), accutin, cidofovir, vincristine, bleomycin, AGM- 1470, platelet factor 4 or minocycline. In further preferred embodiments, the cytokine is interleukin 1 (IL-I), IL-2, IL-5, IL-10, IL-I l, IL-12, IL-18, interferon-γ (IF-γ), IF-α, IF-β, tumor necrosis factor-α (TNF-α), or GM-CSF (granulocyte macrophage colony stimulating factor). Such examples are representative only and are not intended to exclude other pro-apoptosis agents, anti-angiogenic agents or cytokines known in the art.

In still other embodiments, targeting peptides attached to one or more therapeutic agents may be administered to a subject, such as an animal, mammal, cat, dog, cow, pig, horse, sheep or human subject. Such administration may be of use for the treatment of various disease states. In certain embodiments, leukemia targeting peptides attached to a cytocidal, pro-apoptotic, anti-angiogenic or other therapeutic agent may be of use in methods to treat hyperproliferative diseases, such as cancer and preferably leukemia.

In other aspects of the invention, the isolated peptide may be attached to an agent that is a macromolecular complex. A "macromolecular complex" refers to a collection of molecules that may be random, ordered or partially ordered in their arrangement. The term encompasses biological organisms such as bacteriophage, viruses, bacteria, unicellular pathogenic organisms, multicellular pathogenic organisms and prokaryotic or eukaryotic cells. The term also encompasses non-living assemblages of molecules, such as liposomes, microcapsules, microparticles, magnetic beads and microdevices. In certain aspects a virus may be a papovaviruses, a simian virus 40, a bovine papilloma virus, a polyoma virus, adenovirus, vaccinia virus, adeno-associated virus (AAV), herpes virus or any of a variety of viruses known in the art. The only requirement is that the complex contains more than one molecule. The molecules may be identical, or may differ from each other. The macromolecular complex may be a virus, a bacteriophage, a bacterium, a liposome, a microparticle, a magnetic bead, a yeast cell, a mammalian cell, a cell, or a microdevice. These are representative examples only and macromolecular complexes within the scope of the present invention may include virtually any complex that may be attached, directly or indirectly, to a targeting peptide and administered to a subject. In certain embodiments, the isolated peptide may be attached to a eukaryotic expression vector, more preferably a gene therapy vector.

In another embodiment, the isolated peptide may be attached to a solid support, preferably magnetic beads, Sepharose beads, agarose beads, a nitrocellulose membrane, a nylon membrane, a column chromatography matrix, a high performance liquid chromatography (HPLC) matrix or a fast performance liquid chromatography (FPLC) matrix.

Additional embodiments of the present invention concern fusion proteins comprising a leukemia targeting peptide. In a preferred embodiment, a fusion protein may comprise at least 3, 4, 5, 6, or 7 contiguous amino acids of a sequence selected from any of SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, or variants and/or mimetics thereof. In some embodiments, larger contiguous sequences, up to a full-length sequence selected from any of SEQ ID NO: 5, SEQ ID NO: 6, or SEQ ID NO: 7 may be used. Certain other embodiments concern compositions comprising isolated peptides or fusion proteins in a pharmaceutically acceptable carrier. Further embodiments concern kits comprising the claimed isolated peptides or fusion proteins in one or more containers.

Other embodiments concern methods of targeted delivery comprising selecting a leukemia targeting peptide for a desired organ, tissue or cell type, attaching the targeting peptide to a molecule, macromolecular complex or gene therapy vector, and providing the peptide attached to the molecule, complex or vector to a subject. Preferably, the targeting peptide is selected to include at least 3, 4, 5, 6, or 7 contiguous amino acids from any of selected from any of SEQ ID NO: 5, SEQ ID NO: 6, or SEQ ID NO: 7. In certain preferred embodiments, the cell is a leukemia cell. In other preferred embodiments, the molecule attached to the targeting peptide is a chemotherapeutic agent, an antigen, an imaging or a diagnostic agent.

Other embodiments of the present invention concern isolated nucleic acids of 300 nucleotides or less in size, encoding a targeting peptide. In preferred embodiments, the isolated nucleic acid is 255, 250, 240, 225, 210, 200, 180, 175, 150, 126, 125, 102, 100, 75, 51, 50, 42, 40, 30, 21, 20, 12, 10 or even 9 nucleotides or less in size. In other preferred embodiments, the isolated nucleic acid is incorporated into a eukaryotic or a prokaryotic expression vector. In even more preferred embodiments, the vector is a plasmid, a cosmid, a yeast artificial chromosome (YAC), a bacterial artificial chromosome (BAC), a virus or a bacteriophage. In other preferred embodiments, the isolated nucleic acid is operatively linked to a leader sequence or a nucleic acid encoding a leader sequence that localizes the nucleic acid or expresses a peptide that is localized to the extracellular surface of a host cell, respectively. The terms "operatively linked" or "operatively coupled" mean connected either directly or indirectly with other intervening moieties in a way that typically forms an operational or a functional connection between entities, e.g., a peptide and therapeutic agent. An alternative example of operatively coupled includes a peptide coupled to a liposome, which contains a therapeutic agent. In the example the peptide is operatively coupled to the therapeutic agent due to formation of an operational connection between the peptide and therapeutic agent mediated by the intervening liposome.

Additional embodiments of the present invention concern methods of treating a disease state, such as leukemia, comprising selecting a targeting peptide that targets cells associated with the disease state, attaching one or more molecules effective to treat the disease state to the peptide, and administering the peptide to a subject with the disease state. Preferably, the targeting peptide includes at least 3, 4, 5, 6, or 7 contiguous amino acids selected from any of selected from any of SEQ ID NO: 5, SEQ ID NO: 6, or SEQ ID NO: 7.

Biopanning and Rapid Analysis of Selective Interactive Ligands (BRASIL), a method for phage display, may results in decreased background of non-specific phage binding, while retaining selective binding of phage to cell receptors (U.S. Patent Application 20040048243, which is incorporated in its entirety herein by reference). In preferred embodiments, targeting peptides are identified by exposing a subject or cells to a phage display library, collecting samples of one or more organs, tissues or cell types, separating the samples into isolated cells or small clumps of cells suspended in an aqueous phase, layering the aqueous phase over an organic phase, centrifuging the two phases so that the cells are pelleted at the bottom of a centrifuge tube and collecting phage from the pellet. In an even more preferred embodiment, the organic phase is dibutylphtalate. In another embodiment, a phage display library displaying the antigen binding portions of antibodies from a subject is prepared, the library is screened against one or more antigens and phage that bind to the antigens are collected. In more preferred embodiments, the antigen is a targeting peptide.

In still other embodiments, methods include selecting a leukemia targeting peptide by obtaining at least one sample comprising leukemia cells; exposing the sample to a peptide library; and recovering one or more peptides that bind to the leukemia cells. The peptide library may be a phage display library. In certain aspects the phage may be recovered by infecting pilus positive bacteria. In other aspects, the phage may be recovered by amplifying phage inserts; ligating the amplified inserts to phage DNA; and producing phage from the ligated DNA. In a preferred embodiment, phage are recovered by BRASIL (Biopanning and Rapid Analysis of Selective Interactive Ligands). The method may further comprise obtaining one or more types of non-leukemic cells and exposing said cells to said peptide library and recovering one or more peptides that do not bind to said one or more types of non-leukemic cells. In additional aspects The method may further comprise preselecting the phage library against non-leukemia cell type; removing phage that bind to the non-leukemia cell type; and selecting the remaining phage against leukemia cells, non- leukemic cells may be lymphocytes, leukocytes, stem cells, myeloid cells or the like.

In particular embodiments, the methods and compositions may be used to identify one or more receptors for a targeting peptide. In alternative embodiments, the compositions and methods may be used to identify naturally occurring ligands for known or newly identified receptors. In some embodiments, the methods may comprise contacting a targeting peptide to an organ, tissue or cell containing a receptor of interest, allowing the peptide to bind to the receptor, and identifying the receptor by its binding to the peptide. In preferred embodiments, the targeting peptide contains at least 3, 4, 5, 6, or 7 contiguous amino acids selected from SEQ ID NO: 5, SEQ ID NO: 6, or SEQ ID NO: 7. In other preferred embodiments, the targeting peptide may comprise a portion of an antibody against the receptor.

In certain embodiments, methods of the invention may utilize intact organs, tissues or cells, or may alternatively utilize homogenates or detergent extracts of the organs, tissues or cells for identifying additional targeting peptide or protein that interact, directly or indirectly with a targeting peptide. In certain embodiments, cells to be contacted may be genetically engineered to express a suspected receptor for the targeting peptide. In a preferred embodiment, the targeting peptide is modified with a reactive moiety that allows its covalent attachment to the receptor. In a more preferred embodiment, the reactive moiety is a photoreactive group that becomes covalently attached to the interacting protein or receptor when activated by light. In another preferred embodiment, the peptide is attached to a solid support and the interacting protein or receptor is purified by affinity chromatography. In other preferred embodiments, the solid support comprises magnetic beads, Sepharose beads, agarose beads, a nitrocellulose membrane, a nylon membrane, a column chromatography matrix, a high performance liquid chromatography (HPLC) matrix or a fast performance liquid chromatography (FPLC) matrix.

In certain aspects of the invention the receptor of interest or a targeted receptor of the invention is a cell surface receptor, more preferably a member of the Plexin or GP-130 like family of receptors, and more preferably GP 130R, plexin bl, GLM-R, plexin Al, plexin A2, neuropilin-1 (NRP-I), NRP-2, or most preferably NRP-I .

In other embodiments, the targeting peptide may inhibit the activity of a receptor upon binding to the receptor. The receptor activity can be assayed by a variety of methods known in the art, including but not limited to catalytic activity and binding activity. In other embodiments, binding of a targeting peptide to a receptor may inhibit a transport activity of the receptor.

In alternative embodiments, one or more ligands for a receptor of interest may be identified by the disclosed methods and compositions. One or more targeting peptides that mimic all or part of a naturally occurring ligand may be identified by phage display and biopanning in vivo or in vitro. A naturally occurring ligand may be identified by homology with a single targeting peptide that binds to the receptor, or a consensus motif of sequences that bind to the receptor. In other alternative embodiments, an antibody may be prepared against one or more targeting peptides that bind to a receptor of interest. Such antibodies may be used for identification or immunoaffinity purification of native ligands.

Other embodiments include methods of treating leukemia in a subject by obtaining a leukemia cell targeting peptide or peptide conjugate as described, selected or made herein, wherein the peptide i) is operatively coupled to and delivers a therapeutic agent to the leukemia cell, ii) inhibits the adhesion of the leukemia cell to a tissue or organ, or iii) is operatively coupled to and delivers a therapeutic agent to the leukemia cell and inhibits the adhesion of the leukemia cell to a tissue or organ; and administering the peptide to the subject.

In certain embodiments, the targeting peptides of the present invention are of use for the selective delivery of therapeutic agents, including but not limited to gene therapy vectors and fusion proteins, to specific organs, tissues or cell types. The skilled artisan will realize that the scope of the claimed methods of use include any disease state that can be treated by targeted delivery of a therapeutic agent to a desired organ, tissue or cell type. Although such disease states include those where the diseased cells are confined to a specific organ, tissue or cell type, other disease states may be treated by an organ, tissue or cell type- targeting approach. In particular embodiments, the organ, tissue or cell type may comprise leukemia.

In yet further embodiments, methods of targeting the delivery of an agent to a leukemia cell in a subject include obtaining a peptide described, selected or made herein operatively coupling the peptide to the agent; and administering the peptide-coupled agent to the subject. The subject is preferably a human, but may include a mouse, a dog, a cat, a rat, a sheep, a horse, a cow, a goat or a pig.

A method of identifying a leukemia cell includes contacting a sample suspected of comprising a leukemia cell with an isolated peptide of the invention; and detecting binding of the peptide to the sample, thereby identifying sample as comprising leukemia cells.

A method of identifying a receptor or protein that interacts with a leukemia targeting peptide, comprising the steps of obtaining a composition suspected of comprising a receptor or protein that interacts with a leukemia cell targeting peptide; contacting the composition with a peptide in accordance with the invention under conditions that permit binding of the peptide to any such receptor or protein present in the composition; and identifying a receptor or protein that binds to the peptide. The composition may comprise leukemia cells. In certain aspects, the methods may further comprise isolating the receptor or protein. In yet other aspects, the methods may further comprise preparing an antibody or antibody fragment that recognizes and binds to the receptor or protein. The agent that one desires to have delivered to leukemia cells may be operatively coupled to the antibody or antibody fragment.

In other embodiments of the invention includes an antibody or antibody fragment that recognizes and binds to a receptor or protein identified by the methods of the invention. The antibody or antibody fragment may further comprise an agent or macromolecular complex that one desires to have delivered to leukemia cells attached to said antibody or antibody fragment.

In yet another aspect of the invention includes methods of selectively targeting a leukemia cell in a patient, comprising the steps of obtaining an antibody or antibody fragment in accordance with the invention; and administering the antibody or fragment to said patient to thereby target the leukemia cells.

In other embodiments, the profile of phage interaction with various leukemia cells may be utilized to characterize known leukemia or to classify a leukemia cell that not been classified or its classification is unknown. A set of standards or profile parameters will be established so that the phage binding profile of various unknown leukemias may then be compared to the profile of known or characterized leukemia and a designation made regarding the classification of the unknown leukemia.

As used herein in the specification, "a" or "an" may mean one or more. As used herein in the claim(s), in conjunction with the word "comprising," the words "a" or "an" may mean one or more than one. As used herein "another" may mean at least a second or more of an item.

Certain embodiments concern methods of obtaining antibodies against an antigen. In preferred embodiments, the antigen comprises one or more targeting peptides. The targeting peptides are prepared and immobilized on a solid support, serum-containing antibodies is added and antibodies that bind to the targeting peptides are collected. BRIEF DESCRIPTION OF THE DRAWINGS

The following drawings form part of the present specification and are included to further demonstrate certain aspects of the present invention. The invention may be better understood by reference to one or more of these drawings in combination with the detailed description of specific embodiments presented herein.

FIG. 1 illustrates exemplary results of BRASIL (n=5) on RPMI-8226 cells validating binding of Molt-4 phage clones.

FIG. 2 illustrates exemplary results of BRASIL (n=3) on Molt-4 cells to validate binding of Molt-4 phage clones.

FIG. 3 illustrates exemplary results of BRASIL (n=3) on CCRF-CEM cells to validate binding of Molt-4 phage clones.

FIG. 4 illustrates exemplary results of BRASIL (n=3) on K562 cells to validate binding of Molt-4 phage clones.

FIG. 5 illustrates exemplary results of BRASIL (n=3) on SR-786 cells to validate binding of Molt-4 phage clones.

FIG. 6 illustrates exemplary results of BRASIL (n=3) on TUR cells to validate binding of Molt-4 phage clones.

FIG. 7 illustrates exemplary results of BRASIL on a panel of cell lines to validate binding of Molt-4 phage clones.

FIG. 8 illustrates exemplary results of BRASIL on AML clinical samples to validate binding of Molt-4 phage clones.

FIG. 9 illustrates exemplary results of BRASIL on AML clinical samples to validate binding of Molt-4 phage clones.

FIG. 10 illustrates exemplary results of BRASIL on AML clinical samples to validate binding of Molt-4 phage clones.

FIG. 11 illustrates exemplary results of BRASIL on AML clinical samples to validate binding of Molt-4 phage clones.

FIGs. 12A - 12B illustrates a comparison of phage binding in AML patient sample, HBMMNC, and HPBMNC. FIG. 13 illustrates exemplary results of BRASIL on AML clinical samples to validate binding of Molt-4 phage clones.

FIG. 14 illustrates the results of studies investigating the peptide binding to Molt-4 cells.

FIG. 15 illustrates an exemplary internalization assay of Molt-4 phage clones in K-562 CML cells. The results suggest that the receptors corresponding to clones 1 and 2 are capable of internalizing the phage with clone 1 exhibiting a higher potential for receptor- mediated internalization in K-562 cells. By contrast, the receptor for clone 3 does not seem to mediate efficient internalization as indicated from the negative FITC staining of anti-phage relative to the IgG isotype control.

FIG. 16. illustrates the validation of neuropilin-1 (NRP-I), a putative receptor for the GFYWLRS peptide insert in Molt-4 phage clone 2 as suggested from BLAST analysis. Phage binding to recombinant rat neuropilin-1 /Fc chimera (rrNRP-1/Fc) was assessed. Results shown are average of three independent assays.

FIG. 17. illustrates the assessment of binding of Molt-4 phage clone 2 to neuropilin-1 -related proteins such as neuropilin-2 and members of the semaphorin family.

FIGs. 18A-18C. illustrates FACS profiles of cell surface expression of NRP-I in Molt-4 (FIG. 18A), OCI- AML3 (FIG. 18B), and K562 (FIG. 18C) leukemia cell lines. Cells were stained as such: cells only, + FITC-goat anti-rabbit IgG, + control rabbit IgG and FITC- goat anti-rabbit IgG, + rabbit anti-human NRP-I and FITC-goat anti -rabbit IgG. M2 indicates % of positive gated cells per sample, Mn stands for the mean fluorescence intensity of the positively gated population.

FIGs. 19A-19C. illustrates assessment of the cytotoxic effect of C-GFYWLRS- C-GG-D(KLAKLAK)2 peptide on leukemia cells in vitro. Increasing doses of either peptide alone, peptides mixed in solution, or synthetically conjugated peptides were added to cultured leukemia cells in proliferating phase and incubated for 24hr at 370C. The WST-I reagent was then added for 4hr and the OD was measured at 450nm. Assuming viability of untreated cells at 100%, the corresponding OD measurement was used as baseline from which % viability and proliferation of all treated samples was extrapolated based on the various OD readouts. FIG. 19A shows the results for Molt-4 cells whereby the C-GFYWLRS-C-GG- D(KLAKLAK)2 peptide resulted in maximal cytotoxicity at concentrations as low as 20 μM. At this dose, significant cytotoxicity was also obtained in treated OCI-AML3 cells (FIG. 19B). At higher doses though, either D(KLAKLAK)2 peptide alone or in mix with the ligand peptide resulted in significant loss of cell viability as well, hence reflecting non-specific toxicity at these doses in OCI- AML3 cells. The CML K562 cell line (FIG. 19C) was also sensitive to the conjugated form of the peptide in that only 37.26 ± 14.25 % (average ± SEM) of the cells were viable upon treatment with 20 μM of the C-GFYWLRS-C-GG- D(KLAKLAK)2 peptide. Results shown represent the average of 3 independent experiments.

DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS

Additional therapeutic compositions and methods for the treatment and therapy of leukemia are still needed, in particular methods and compositions related to leukemia targeting peptides. In one aspect of the invention, the inventors have employed Biopanning and Rapid Analysis of Selective Interactive Ligands (BRASIL) to identify peptides that selectively bind to leukemia cells from a variety of sources. As used herein "selective binding" in no way precludes binding to other cells or material, but connotes the preferential binding of leukemia cells. Selective binding may include a 2, 3, 4, 5, 6, 7, 8, 9, 10 or more fold preference for leukemia cells as compared to non-leukemia cells including normal leukocytes and mesenchymal stem cells (MSC). In one example, human-derived leukemia cell lines were profiled, including those from the NCI-60 cell panel. Screening of the cell lines with a CX7C random phage library, for example, yielded several peptide motifs that bound leukemia cells, of which at least three clones (SEQ ID NO: 5, SEQ ID NO: 6 and SEQ ID NO: 7) exhibited high frequency binding to various leukemia cell lines as compared to the control insert-less phage. Comparison of the selected motifs with available sequences in on¬ line protein databases suggests that a number of candidate proteins, in particular known viral proteins, share homologous sequences with these peptides. These peptides are being use in further studies to identify and purify protein(s) that interact, directly or indirectly, with an identified peptide, including identifying and purifying corresponding receptor(s). In the clinics the newly identified peptides and peptide motifs may serve as targeting moieties, drugs and/or drug leads. Also, the identified peptides can be optimized as delivery vehicles or enhancers for targeted therapy of leukemia.

A "targeting peptide" as used herein is a peptide comprising a contiguous sequence of amino acids, which is characterized by selective localization to an organ, tissue or cell type. Selective localization may be determined, for example, by methods disclosed below, wherein the putative targeting peptide sequence is incorporated into a protein that is displayed on the outer surface of a phage. Administration to a subject of a library of such phage that have been genetically engineered to express a multitude of such targeting peptides of different amino acid sequence is followed by collection of one or more organs, tissues or cell types that are typically derived from a subject and identification of phage found in or associated with that organ, tissue or cell type. A phage expressing a targeting peptide sequence is considered to be selectively localized to a tissue, organ or cell if it exhibits greater binding in that tissue, organ, or cell compared to a control tissue, organ, or cell. Preferably, selective localization of a targeting peptide should result in a two-fold or higher enrichment of the phage or peptide in the target organ, tissue or cell type, compared to a control organ, tissue or cell type. Selective localization resulting in at least a three- fold, four¬ fold, five-fold, six-fold, seven-fold, eight-fold, nine-fold, ten-fold or higher enrichment in the target organ compared to a control organ, tissue or cell type is more preferred.

Alternatively, a phage expressing a targeting peptide sequence that exhibits selective localization preferably shows an increased enrichment in the target organ or cell compared to a control organ or cell when phage recovered from the target organ or cell are injected into or put in contact with a second host or cell population for another round of screening. Further enrichment may be exhibited following a third round of screening.

Another alternative means to determine selective localization is that phage expressing the putative target peptide preferably exhibit a two-fold, more preferably a three- fold or higher enrichment in the target organ or cell type compared to control phage that express a non-specific peptide or that have not been genetically engineered to express any putative target peptides. Yet another means to determine selective localization is that localization to the target organ of phage expressing the target peptide is at least partially blocked by the co-administration of a synthetic peptide containing the target peptide sequence. "Targeting peptide" and "homing peptide" are used synonymously herein.

I. LEUKEMIA

Leukemia is classified by how quickly it progresses. Acute leukemia is fast- growing and can overrun the body within a few weeks or months. By contrast, chronic leukemia is slow-growing and progressively worsens over years.

The blood-forming (hematopoietic) cells of acute leukemia remain in an immature state, so they reproduce and accumulate very rapidly. Therefore, acute leukemia needs to be treated immediately, otherwise the disease may be fatal within a few months. Fortunately, some subtypes of acute leukemia respond very well to available therapies and they are curable. Children often develop acute forms of leukemia, which are managed differently from leukemia in adults. In chronic leukemia, the blood-forming cells eventually mature, or differentiate, but they are not "normal." They remain in the bloodstream much longer than normal white blood cells, and they are unable to combat infection well.

Leukemia also is classified according to the type of white blood cell that is multiplying - that is, lymphocytes (immune system cells), granulocytes (bacteria-destroying cells), or monocytes (macrophage-forming cells). If the abnormal white blood cells are primarily granulocytes or monocytes, the leukemia is categorized as myelogenous, or myeloid, leukemia. On the other hand, if the abnormal blood cells arise from bone marrow lymphocytes, the cancer is called lymphocytic leukemia. Other cancers, known as lymphomas, develop from lymphocytes within the lymph nodes, spleen, and other organs. Such cancers do not originate in the bone marrow and have a biological behavior that is different from lymphocytic leukemia.

There are over a dozen different types of leukemia, but four types occur most frequently. These classifications are based upon whether the leukemia is acute versus chronic and myelogenous versus lymphocytic, that is: Acute Myelogenous (granulocytic) Leukemia (AML), Chronic Myelogenous (granulocytic) Leukemia (CML), Acute Lymphocytic (lymphoblastic) Leukemia (ALL), Chronic Lymphocytic Leukemia (CLL), and Acute Myelogenous Leukemia (AML)

Acute myelogenous leukemia (AML) - also known as acute nonlymphocytic leukemia (ANLL) - is the most common form of adult leukemia. Initial response rates are approximately 65-75% but the overall cure rates are more on the order of 40-50%. Acute leukemia, such as AML, is typically categorized according to a system known as French- American-British (FAB) classification. FAB divides AML into eight subtypes: undifferentiated AML (MO), myeloblasts leukemia (Ml), myeloblasts leukemia (M2), promyelocyte leukemia (M3 or M3), myelomonocytic leukemia (M4), monocytic leukemia (M5), erythroleukemia (M6), or megakaryoblastic leukemia (M7). In addition, patients sometimes develop isolated tumors of the myeloblasts (early granulocytes), e.g., granulocytic sarcoma. II. IDENTIFICATION OF A TARGETING PEPTIDES The invention comprises methods for the identification of one or more targeting peptides or molecular targets that could be utilized for the development of novel therapies in leukemia. Employing Biopanning and Rapid Analysis of Selective Interactive Ligands (BRASIL), human-derived leukemia cell lines are profiled, including, but not limited to the NCI-60 cell panel. Screening of the leukemia cells, including Acute Myelogenous Leukemia (AML) and Acute Lymphoblastic Leukemia (ALL) cell line with CXnC, wherein in can be 4, 5, 6, 7, or more residues, random phage library that yield several peptide motifs. In one example, three clones (encoding SEQ ID NO: 5, SEQ ID NO: 6, and SEQ ID NO: 7) exhibited high frequency binding to various leukemia cell lines as compared to the control insert-less phage. Comparison of the selected motifs with available sequences in on-line protein databases suggests that a number of candidate proteins, in particular known viral proteins, that share homologous sequences with these peptides. Work is underway to validate the lead peptides by assessing their binding specificity in clinical samples from ALL and AML patients. Subsequently, identification and purification of the corresponding receptor(s) and investigation of the interaction mechanism of the ligand-receptor pair candidates are being conducted. Thus, mechanistic studies surrounding these targets are being pursued to provide a rich platform for the identification of signaling pathways relevant to leukemia. The findings will also have important clinical implications in that newly identified motifs may serve as a peptidomimetic drug leads and can be optimized as delivery vehicles for targeted therapy of leukemia.

BRASIL has been successfully used to isolate phage in various cell systems such as activated endothelial cells and tumor cells. BRASIL has also been used to isolate bone marrow homing phage using in vivolex-vivo based strategies. One method includes injecting the phage libraries intravenously and recover the bone marrow after a few minutes. To identify leukemia cells that can bind to peptide from patients cells are incubated with peptide encoding phage or control phage. Phage bound to the cells are recovered, and quantified. Typically, an enrichment of the peptide-displaying phage in comparison to control phage is seen in cells with active β2 integrins. To assess the generality of the response of leukemia cells to these peptides, the effects are typically determined in a panel of cell lines administered intravenously to a model of advanced stage disease. Selected or candidate peptides and control peptides are administered subcutaneously and intraperitoneally. Survival and leukemic cell burden is monitored. Initial in vivo experiments have shown that leukemia develops in immuno- incompetent (SCID) mice injected with human leukemia cells (OCI- AML3; 107 cells/mouse intravenously) and are sensitive to ex-vivo treatment with LLG-C4. Animals treated with LLG-C4 had a significantly longer overall survival after a single treatment. Comparison of the selected motifs with available sequences in on-line protein databases suggests that a number of candidate proteins share homologous sequences with these peptides; some of which may or may not have a high bio-functional relevance to the setting of leukemia.

Certain aspects of the invention include the identification of peptide motifs (ligands) that bind preferentially to leukemia cells from clinical samples. Other aspects include utilizing the leukemia targeted peptide to identify and purify their corresponding receρtor(s) present on the surface of leukemia cells. Still further aspects of the invention include the study of the interaction mechanism of the ligand-receptor pair candidates in the context of leukemia. The inventor describe herein various methods and compositions for the identification of ligands, receptors, and ligand-receptor pairs useful for the development of leukemia-targeted therapy.

A. Phage Display

Recently, an in vivo selection system was developed using phage display libraries to identify organ, tissue or cell type-targeting peptides in a mouse model system. Phage display libraries expressing transgenic peptides on the surface of bacteriophage were initially developed to map epitope binding sites of immunoglobulins (Smith and Scott, 1985 and 1993). Such libraries can be generated by inserting random oligonucleotides into cDNAs encoding a phage surface protein, generating collections of phage particles displaying unique peptides in as many as 109 permutations (Pasqualini and Ruoslahti, 1996; Arap et al, 1998a and 1998b).

A "phage display library" is a collection of phage that have been genetically engineered to express a set of putative targeting peptides on their outer surface. In preferred embodiments, DNA sequences encoding the putative targeting peptides are inserted in frame into a gene encoding a phage capsule protein. In other preferred embodiments, the putative targeting peptide sequences are in part random mixtures of all twenty amino acids and in part non-random. In certain preferred embodiments the putative targeting peptides of the phage display library exhibit one or more cysteine residues at fixed locations within the targeting peptide sequence. Cysteines may be used, for example, to create a cyclic peptide. B. Biopanning and Rapid Analysis of Selective Interactive Ligands (BRASIL)

In preferred embodiments, separation of phage bound to the cells of a target organ, tissue or cell type from unbound phage is achieved using the BRASIL (Biopanning and Rapid Analysis of Soluble Interactive Ligands) technique (PCT Application PCT/USOl/28124 entitled, "Biopanning and Rapid Analysis of Selective Interactive Ligands (BRASIL)" by Arap et al, filed September 7, 2001, incorporated herein by reference in its entirety). In BRASIL, an organ, tissue or cell type is gently separated into cells or small clumps of cells that are suspended in an aqueous phase. The aqueous phase is layered over an organic phase of appropriate density and centrifuged. Cells attached to bound phage are pelleted at the bottom of the centrifuge tube, while unbound phage remain in the aqueous phase. This allows a more efficient separation of bound from unbound phage, while maintaining the binding interaction between phage and cell. BRASIL may be performed in an in vivo protocol, in which organs, tissues or cell types are exposed to a phage display library by intravenous administration, or by an ex vivo protocol, where the cells are exposed to the phage library in the aqueous phase before centrifugation.

C. Preparation of Large Scale Primary Libraries In certain embodiments, primary phage libraries are amplified before injection into a human subject. A phage library is prepared by ligating targeting peptide-encoding sequences into a phage vector, such as fUSE5. The vector is transformed into pilus negative host E. coli such as strain MC 1061. The bacteria are grown overnight and then aliquots are frozen to provide stock for library production. Use of pilus negative bacteria avoids the bias in libraries that arises from differential infection of pilus positive bacteria by different targeting peptide sequences.

To freeze, bacteria are pelleted from two thirds of a primary library culture (5 liters) at 4000 x g for 10 min. Bacteria are resuspended and washed twice with 500 ml of 10% glycerol in water, then frozen in an ethanol/dry ice bath and stored at -80°C.

For amplification, 1.5 ml of frozen bacteria are inoculated into 5 liters of LB medium with 20 μg/ml tetracycline and grown overnight. Thirty minutes after inoculation, a serial dilution is plated on LB/tet plates to verify the viability of the culture. If the number of viable bacteria is less than 5-10 times the number of individual clones in the library (1-2 x 108) the culture is discarded. After growing the bacterial culture overnight, phage are precipitated. About 1/4 to 1/3 of the bacterial culture is kept growing overnight in 5 liters of fresh medium and the cycle is repeated up to 5 times. Phage are pooled from all cycles and used for injection into human subjects.

Intravenous administration of phage display libraries to mice was followed by the recovery of phage from individual organs (Pasqualini and Ruoslahti, 1996). Phage were recovered that were capable of selective homing to the vascular beds of different mouse organs, tissues or cell types, based on the specific targeting peptide sequences expressed on the outer surface of the phage (Pasqualini and Ruoslahti, 1996). A variety of organ and tumor-homing peptides have been identified by this method (Rajotte et al, 1998; Rajotte et al, 1999; Koivunen et al, 1999a; Burg et al, 1999; Pasqualini, 1999). Each of those targeting peptides bound to different receptors that were selectively expressed on the vasculature of the mouse target tissue (Pasqualini, 1999; Pasqualini et al, 2000; Folkman, 1997; Folkman, 1995). Tumor-homing peptides bound to receptors that were upregulated in the tumor angiogenic vasculature of mice (Brooks et al, 1994b; Pasqualini et al, 2000). In addition to identifying individual targeting peptides selective for an organ, tissue or cell type (Pasqualini and Ruoslahti, 1996; Arap et al, 1998a; Koivunen et al, 1999b), this system has been used to identify endothelial cell surface markers that are expressed in mice in vivo (Rajotte and Ruoslahti, 1999).

Attachment of therapeutic agents to targeting peptides resulted in the selective delivery of the agent to a desired organ, tissue or cell type in the mouse model system. Targeted delivery of chemotherapeutic agents and proapoptotic peptides to receptors located in tumor angiogenic vasculature resulted in a marked increase in therapeutic efficacy and a decrease in systemic toxicity in tumor bearing mouse models (Arap et al, 1998a, 1998b; Ellerby e/ fl/., 1999).

The methods described herein for identification of targeting peptides involve the in vivo administration of phage display libraries. Various methods of phage display and methods for producing diverse populations of peptides are well known in the art. For example, U.S. Patents 5,223,409; 5,622,699 and 6,068,829, each of which is incorporated herein by reference in its entirety, disclose methods for preparing a phage library. The phage display technique involves genetically manipulating bacteriophage so that small peptides can be expressed on their surface (Smith and Scott, 1985 and 1993). The potential range of applications for this technique is quite broad, and the past decade has seen considerable progress in the construction of phage-displayed peptide libraries and in the development of screening methods in which the libraries are used to isolate peptide ligands. For example, the use of peptide libraries has made it possible to characterize interacting sites and receptor- ligand binding motifs within many proteins, such as antibodies involved in inflammatory reactions or integrins that mediate cellular adherence. This method has also been used to identify novel peptide ligands that serve as leads to the development of peptidomimetic drugs or imaging agents (Arap et al, 1998a). In addition to peptides, larger protein domains such as single-chain antibodies can also be displayed on the surface of phage particles (Arap et al., 1998a).

Targeting peptides selective for a given organ, tissue or cell type can be isolated by "biopanning" (Pasqualini and Ruoslahti, 1996; Pasqualini, 1999). In brief, a library of phage containing putative targeting peptides is administered to an animal or human and samples of organs, tissues or cell types containing phage are collected. In preferred embodiments utilizing filamentous phage, the phage may be propagated in vitro between rounds of biopanning in pilus-positive bacteria. The bacteria are not lysed by the phage but rather secrete multiple copies of phage that display a particular insert. Phage that bind to a target molecule can be eluted from the target organ, tissue or cell type and then amplified by growing them in host bacteria. If desired, the amplified phage can be administered to a host and samples of organs, tissues or cell types again collected. Multiple rounds of biopanning can be performed until a population of selective binders is obtained. The amino acid sequence of the peptides is determined by sequencing the DNA corresponding to the targeting peptide insert in the phage genome. The identified targeting peptide can then be produced as a synthetic peptide by standard protein chemistry techniques (Arap et al., 1998a, Smith and Scott, 1985). This approach allows circulating targeting peptides to be detected in an unbiased functional assay, without any preconceived notions about the nature of their target. Once a candidate target is identified as the receptor of a targeting peptide, it can be isolated, purified and cloned by using standard biochemical methods (Pasqualini, 1999; Rajotte and Ruoslahti, 1999).

In certain embodiments, a subtraction protocol is used may be used to further reduce background phage binding. The purpose of subtraction is to remove phage from the library that bind to cells other than the cell of interest, or that bind to inactivated cells. In alternative embodiments, the phage library may be prescreened against a subject who does not possess the targeted cell, tissue or organ. For example, placenta-binding peptides may be identified after prescreening a library against a male or non-pregnant female subject. After subtraction the library may be screened against the cell, tissue or organ of interest. In another alternative embodiment, an unstimulated, quiescent cell type, tissue or organ may be screened against the library and binding phage removed. The cell line, tissue or organ is then activated, for example by administration of a hormone, growth factor, cytokine or chemokine and the activated cell type, tissue or organ screened against the subtracted phage library. Other subtraction protocols are known and may be used in the practice of the present invention, for example as disclosed in U.S. Patents 5,840,841, 5,705,610, 5,670,312 and 5,492,807, which are incorporated herein by reference in their entirety.

D. Choice of Phage Display System Previous in vivo selection studies performed in mice preferentially employed libraries of random peptides expressed as fusion proteins with the gene III capsule protein in the fUSE5 vector (Pasqualini and Ruoslahti, 1996). The number and diversity of individual clones present in a given library is a significant factor for the success of in vivo selection. It is preferred to use primary libraries, which are less likely to have an over-representation of defective phage clones (Koivunen et al, 1999b). The preparation of a library should be optimized to between 10 -10 transducing units (T.U.)/ml. In certain embodiments, a bulk amplification strategy is applied between each round of selection.

Phage libraries displaying linear, cyclic, or double cyclic peptides may be used within the scope of the present invention. However, phage libraries displaying 3 to 10 random residues in a cyclic insert (CX3-i0C) are preferred, since single cyclic peptides tend to have a higher affinity for the target organ than linear peptides. Libraries displaying double-cyclic peptides (such as CX3C X3CX3C; Rojotte et al, 1998) have been successfully used. However, the production of the cognate synthetic peptides, although possible, can be complex due to the multiple conformers with different disulfide bridge arrangements.

E. Polyorgan targeting In the standard protocol for phage display biopanning, phage from a single organ are collected, amplified and injected into a new host, where tissue from the same organ is collected for phage rescue and a new round of biopanning. This protocol is feasible in animal subjects. However, the limited availability and expense of processing samples from humans requires an improvement in the protocol. It is possible to pool phage collected from multiple organs after a first round of biopanning and inject the pooled sample into a new subject, where each of the multiple organs may be collected again for phage rescue. The polyorgan targeting protocol may be repeated for as many rounds of biopanning as desired. In this manner, it is possible to significantly reduce the number of subjects required for isolation of targeting peptides for multiple organs, while still achieving substantial enrichment of the organ-homing phage.

In preferred embodiments, phage are recovered from human organs, tissues or cell types after injection of a phage display library into a human subject. In certain embodiments, phage may be recovered by exposing a sample of the organ, tissue or cell type to a pilus positive bacterium, such as E. coli K91. In alternative embodiments, phage may be recovered by amplifying the phage inserts, ligating the inserts to phage DNA and producing new phage from the ligated DNA.

III. TARGETED DELIVERY

Peptides that home to vasculature have been coupled to cytotoxic drugs or proapoptotic peptides to yield compounds that were more effective and less toxic than the parental compounds in experimental models of mice bearing tumor xenografts (Arap et al, 1998a; Ellerby et al, 1999). The insertion of the RGD-4C peptide into a surface protein of an adenovirus has produced an adenoviral vector that may be used for tumor targeted gene therapy (Arap et al, 1998b). The present invention describes methods and compositions for the selective targeting of leukemia cells.

A "receptor" for a targeting peptide includes but is not limited to any molecule or macromolecular complex that binds to a targeting peptide. Non-limiting examples of receptors include peptides, proteins, glycoproteins, lipoproteins, epitopes, lipids, carbohydrates, multi-molecular structures, and a specific conformation of one or more molecules. In preferred embodiments, a "receptor" is a naturally occurring molecule or complex of molecules that is present on the surface of cells within a target organ, tissue or cell type. More preferrably, a "receptor" is a naturally occurring molecule or complex of molecules that is present on the surface of leukemia cells.

The methods used for phage display biopanning in the mouse model system require substantial improvements for use with humans. Techniques for biopanning in human subjects are disclosed in PCT Patent Application PCT/US01/28044, filed September 7, 2001, the entire text of which is incorporated herein by reference. A "subject" refers generally to a mammal. In certain preferred embodiments, the subject is a mouse or rabbit. In more preferred embodiments, the subject is a human. In general, humans suitable for use with phage display are either brain dead or terminal wean patients. The amount of phage library (preferably primary library) required for administration must be significantly increased, preferably to 101 TU or higher, preferably administered intravenously in approximately 200 ml of Ringer lactate solution over about a 10 minute period.

The amount of phage required for use in humans has required substantial improvement of the mouse protocol, increasing the amount of phage available for injection by five orders of magnitude. To produce such large phage libraries, the transformed bacterial pellets recovered from up to 500 to 1000 transformations are amplified up to 10 times in the bacterial host, recovering the phage from each round of amplification and adding LB Tet medium to the bacterial pellet for collection of additional phage. The phage inserts remain stable under these conditions and phage may be pooled to form the large phage display library required for humans.

Samples of various organs and tissues are collected starting approximately 15 minutes after injection of the phage library. Samples are processed as described below and phage collected from each organ, tissue or cell type of interest for DNA sequencing to determine the amino acid sequences of targeting peptides.

With humans, the opportunities for enrichment by multiple rounds of biopanning are severely restricted, compared to the mouse model system. A substantial improvement in the biopanning technique involves polyorgan targeting.

IV. PROTEINS AND PEPTIDES

In certain embodiments, the present invention concerns novel compositions comprising at least one protein or peptide. As used herein, a protein or peptide generally refers, but is not limited to, a protein of greater than about 200 amino acids, up to a full length sequence translated from a gene; a polypeptide of greater than about 100 amino acids; and/or a peptide of from about 3 to about 100 amino acids. For convenience, the terms "protein," "polypeptide" and "peptide are used interchangeably herein.

In certain embodiments the size of at least one protein or peptide may comprise, but is not limited to, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, about 110, about 120, about 130, about 140, about 150, about 160, about 170, about 180, about 190, about 200, about 210, about 220, about 230, about 240, about 250, about 275, about 300, about 325, about 350, about 375, about 400, about 425, about 450, about 475, about 500, about 525, about 550, about 575, about 600, about 625, about 650, about 675, about 700, about 725, about 750, about 775, about 800, about 825, about 850, about 875, about 900, about 925, about 950, about 975, about 1000, about 1100, about 1200, about 1300, about 1400, about 1500, about 1750, about 2000, about 2250, about 2500 or greater amino acid residues.

As used herein, an "amino acid residue" refers to any naturally occurring amino acid, any amino acid derivative or any amino acid mimic known in the art. In certain embodiments, the residues of the protein or peptide are sequential, without any non-amino acid interrupting the sequence of amino acid residues. In other embodiments, the sequence may comprise one or more non-amino acid moieties. In particular embodiments, the sequence of residues of the protein or peptide may be interrupted by one or more non-amino acid moieties. Accordingly, the term protein or peptide encompasses amino acid sequences comprising at least one of the 20 common amino acids found in naturally occurring proteins, or at least one modified or unusual amino acid, including, but not limited to, 2 Aminoadipic acid (Aad), N Ethylasparagine (EtAsn), 3 Aminoadipic acid (Baad), Hydroxylysine (HyI), β alanine, β Amino propionic acid (BaIa), allo Hydroxylysine (AHyI), 2 Aminobutyric acid (Abu), 3 Hydroxyproline (3Hyp), 4 Aminobutyric acid (4Abu), 4 Hydroxyproline (4Hyp), 6 Aminocaproic acid (Acp), Isodesmosine (Ide), 2 Aminoheptanoic acid (Ahe), allo Isoleucine (AIIe), 2 Aminoisobutyric acid (Aib), N Methylglycine (MeGIy), 3 Aminoisobutyric acid (Baib), N Methylisoleucine (MeIIe), 2 Aminopimelic acid (Apm), 6 N Methyllysine (MeLys), 2,4 Diaminobutyric acid (Dbu), N Methylvaline (MeVaI), Desmosine (Des), Norvaline (N va), 2,2' Diaminopimelic acid (Dpm), Norleucine (NIe), 2,3 Diaminopropionic acid (Dpr), Ornithine (Orn), or N Ethylglycine (EtGIy).

Proteins or peptides may be made by any technique known to those of skill in the art, including the expression of proteins, polypeptides or peptides through standard molecular biological techniques, the isolation of proteins or peptides from natural sources, or the chemical synthesis of proteins or peptides. Coding regions for known genes may be amplified and/or expressed using the techniques disclosed herein or as would be know to those of ordinary skill in the art. Alternatively, various commercial preparations of proteins, polypeptides and peptides are known to those of skill in the art.

A. Peptide mimetics

Another embodiment for the preparation of molecule or compound according to the invention is the use of peptide mimetics that mimic characteristics of all or part of the peptides identified herein. Mimetics are molecules that mimic elements of protein secondary structure (see, for example, Johnson et al, 1993, incorporated herein by reference). The underlying rationale behind the use of peptide mimetics is that the peptide backbone of proteins exists chiefly to orient amino acid side chains in such a way as to facilitate molecular interactions, such as those of antibody and antigen. A peptide mimetic is expected to permit molecular interactions similar to the natural molecule. These principles may be used to engineer second generation molecules having many of the natural properties of the targeting peptides disclosed herein, but with altered and even improved characteristics.

B. Fusion proteins

Other embodiments of the present invention concern fusion proteins. These molecules generally have all or a substantial portion of a targeting peptide, linked at the N- or C-terminus, to all or a portion of a second polypeptide or protein. For example, fusions may employ leader sequences from other species to permit the recombinant expression of a protein in a heterologous host. Another useful fusion includes the addition of an immunologically active domain, such as an antibody epitope, to facilitate purification of the fusion protein. Inclusion of a cleavage site at or near the fusion junction will facilitate removal of the extraneous polypeptide after purification. Other useful fusions include linking of functional domains, such as active sites from enzymes, glycosylation domains, cellular targeting signals or transmembrane regions. In preferred embodiments, the fusion proteins of the instant invention comprise a targeting peptide linked to a therapeutic protein or peptide. Examples of proteins or peptides that may be incorporated into a fusion protein include cytostatic proteins, cytocidal proteins, pro-apoptosis agents, anti-angiogenic agents, hormones, cytokines, growth factors, peptide drugs, antibodies, Fab fragments antibodies, antigens, receptor proteins, enzymes, lectins, MHC proteins, cell adhesion proteins and binding proteins. These examples are not meant to be limiting and it is contemplated that within the scope of the present invention virtually and protein or peptide could be incorporated into a fusion protein comprising a targeting peptide. Methods of generating fusion proteins are well known to those of skill in the art. Such proteins can be produced, for example, by chemical attachment using bifunctional cross-linking reagents, by de novo synthesis of the complete fusion protein, or by attachment of a DNA sequence encoding the targeting peptide to a DNA sequence encoding the second peptide or protein, followed by expression of the intact fusion protein.

C. Protein purification

In certain embodiments a protein or peptide may be isolated or purified. Protein purification techniques are well known to those of skill in the art. These techniques involve, at one level, the homogenization and crude fractionation of the cells, tissue or organ to polypeptide and non-polypeptide fractions. The protein or peptide of interest may be further purified using chromatographic and electrophoretic techniques to achieve partial or complete purification (or purification to homogeneity). Analytical methods particularly suited to the preparation of a pure peptide are ion-exchange chromatography, gel exclusion chromatography, polyacrylamide gel electrophoresis, affinity chromatography, immunoaffinity chromatography and isoelectric focusing. An example of receptor protein purification by affinity chromatography is disclosed in U.S. Patent 5,206,347, the entire text of which is incorporated herein by reference. A particularly efficient method of purifying peptides is fast performance liquid chromatography (FPLC) or even high performance liquid chromatography (HPLC).

A purified protein or peptide is intended to refer to a composition, isolatable from other components, wherein the protein or peptide is purified to any degree relative to its naturally-obtainable state. An isolated or purified protein or peptide, therefore, also refers to a protein or peptide free from the environment in which it may naturally occur. Generally, "purified" will refer to a protein or peptide composition that has been subjected to fractionation to remove various other components, and which composition substantially retains its expressed biological activity. Where the term "substantially purified" is used, this designation will refer to a composition in which the protein or peptide forms the major component of the composition, such as constituting about 50%, about 60%, about 70%, about 80%, about 90%, about 95%, or more of the protein or peptide in the composition.

Various methods for quantifying the degree of purification of the protein or peptide are known to those of skill in the art in light of the present disclosure. These include, for example, determining the specific activity of an active fraction, or assessing the amount of protein or peptide within a fraction by SDS/PAGE analysis. A preferred method for assessing the purity of a fraction is to calculate the specific activity of the fraction, to compare it to the specific activity of the initial extract, and to thus calculate the degree of purity therein, assessed by a "-fold purification number." The actual units used to represent the amount of activity will, of course, be dependent upon the particular assay technique chosen to follow the purification, and whether or not the expressed protein or peptide exhibits a detectable activity.

Various techniques suitable for use in protein purification are well known to those of skill in the art. These include, for example, precipitation with ammonium sulphate, PEG, antibodies and the like, or by heat denaturation, followed by: centrifugation; chromatography steps such as ion exchange, gel filtration, reverse phase, hydroxylapatite and affinity chromatography; isoelectric focusing; gel electrophoresis; and combinations of these and other techniques. As is generally known in the art, it is believed that the order of conducting the various purification steps may be changed, or that certain steps may be omitted, and still result in a suitable method for the preparation of a substantially purified protein or peptide.

There is no general requirement that the protein or peptide always be provided in their most purified state. Indeed, it is contemplated that less substantially purified products will have utility in certain embodiments. Partial purification may be accomplished by using fewer purification steps in combination, or by utilizing different forms of the same general purification scheme. For example, it is appreciated that a cation-exchange column chromatography performed utilizing an HPLC apparatus will generally result in a greater "- fold" purification than the same technique utilizing some other chromatography systems. Methods exhibiting a lower degree of relative purification may have advantages in total recovery of protein product, or in maintaining the activity of an expressed protein.

Affinity chromatography is a chromatographic procedure that relies on the specific affinity between a substance to be isolated and a molecule to which it can specifically bind. This is a receptor-ligand type of interaction. The column material is synthesized by covalently coupling one of the binding partners to an insoluble matrix. The column material is then able to specifically adsorb the substance from the solution. Elution occurs by changing the conditions to those in which binding will not occur (e.g., altered pH, ionic strength, and temperature). The matrix should be a substance that itself does not adsorb molecules to any significant extent and that has a broad range of chemical, physical and thermal stability. The ligand should be coupled in such a way as to not affect its binding properties. The ligand should also provide relatively tight binding. And it should be possible to elute the substance without destroying the sample or the ligand.

D. Synthetic Peptides Because of their relatively small size, the targeting peptides of the invention can be synthesized in solution or on a solid support in accordance with conventional techniques. Various automatic synthesizers are commercially available and can be used in accordance with known protocols (see, for example, Stewart and Young, 1984; Tarn et al, 1983; Merrifield, 1986; or Barany and Merrifield, 1979, each incorporated herein by reference). Short peptide sequences, usually from about 6 up to about 35 to 50 amino acids, can be readily synthesized by such methods. Alternatively, recombinant DNA technology may be employed wherein a nucleotide sequence which encodes a peptide of the invention is inserted into an expression vector, transformed or transfected into an appropriate host cell, and cultivated under conditions suitable for expression.

E. Antibodies

In certain embodiments, it may be desirable to make antibodies against the identified targeting peptides or their receptors. The appropriate targeting peptide or receptor, or portions thereof, may be coupled, bonded, bound, conjugated, or chemically-linked to one or more agents via linkers, polylinkers, or derivatized amino acids. This may be performed such that a bispecific or multivalent composition or vaccine is produced. It is further envisioned that the methods used in the preparation of these compositions are familiar to those of skill in the art and should be suitable for administration to humans, i.e., pharmaceutically acceptable. Preferred agents are the carriers are keyhole limpet hemocyanin (KLH) or bovine serum albumin (BSA).

The term "antibody" is used to refer to any antibody like molecule that has an antigen binding region, and includes antibody fragments such as Fab', Fab, F(ab')2, single domain antibodies (DABs), Fv, scFv (single chain Fv), and the like. Techniques for preparing and using various antibody based constructs and fragments are well known in the art. Means for preparing and characterizing antibodies are also well known in the art (See, e.g., Harlow and Lane, 1988; incorporated herein by reference). F. Cytokines and chemokines

In certain embodiments, it may be desirable to couple specific bioactive agents to one or more targeting peptides for targeted delivery to an organ, tissue or cell type. Such agents include, but are not limited to, cytokines, chemokines, pro-apoptosis factors and anti- angiogenic factors. The term "cytokine" is a generic term for proteins released by one cell population that act on another cell as intercellular mediators.

Examples of such cytokines are lymphokines, monokines, growth factors and traditional polypeptide hormones. Included among the cytokines are growth hormones such as human growth hormone, N-methionyl human growth hormone, and bovine growth hormone; parathyroid hormone; thyroxine; insulin; proinsulin; relaxin; prorelaxin; glycoprotein hormones such as follicle stimulating hormone (FSH), thyroid stimulating hormone (TSH), and luteinizing hormone (LH); hepatic growth factor; prostaglandin, fibroblast growth factor; prolactin; placental lactogen, OB protein; tumor necrosis factor- . alpha, and -.beta.; mullerian-inhibiting substance; mouse gonadotropin-associated peptide; inhibin; activin; vascular endothelial growth factor; integrin; thrombopoietin (TPO); nerve growth factors such as NGF-.beta.; platelet-growth factor; transforming growth factors (TGFs) such as TGF-. alpha, and TGF-.beta.; insulin-like growth factor-I and -II; erythropoietin (EPO); osteoinductive factors; interferons such as interferon-α, -β, and -γ; colony stimulating factors (CSFs) such as macrophage-CSF (M-CSF); granulocyte- macrophage-CSF (GM-CSF); and granulocyte-CSF (G-CSF); interleukins (ILs) such as IL-I, IL-l .alpha., IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-IO, IL-I l, IL-12; IL-13, IL-14, IL-15, IL-16, IL-17, IL-18, LIF, G-CSF, GM-CSF, M- CSF, EPO, kit-ligand or FLT-3, angiostatin, thrombospondin, endostatin, tumor necrosis factor and LT. As used herein, the term cytokine includes proteins from natural sources or from recombinant cell culture and biologically active equivalents of the native sequence cytokines.

Chemokines generally act as chemoattractants to recruit immune effector cells to the site of chemokine expression. It may be advantageous to express a particular chemokine gene in combination with, for example, a cytokine gene, to enhance the recruitment of other immune system components to the site of treatment. Chemokines include, but are not limited to, RANTES, MCAF, MIPl -alpha, MIPl -Beta, and IP-IO. The skilled artisan will recognize that certain cytokines are also known to have chemoattractant effects and could also be classified under the term chemokines. G. Imaging agents and radioisotopes

In certain embodiments, the claimed peptides or proteins of the present invention may be attached to imaging agents of use for imaging and diagnosis of various diseased organs, tissues or cell types. Many appropriate imaging agents are known in the art, as are methods for their attachment to proteins or peptides (see, e.g., U.S. Patents 5,021,236 and 4,472,509, both incorporated herein by reference). Certain attachment methods involve the use of a metal chelate complex employing, for example, an organic chelating agent such a DTPA attached to the protein or peptide (U.S. Patent 4,472,509). Proteins or peptides also may be reacted with an enzyme in the presence of a coupling agent such as glutaraldehyde or periodate. Conjugates with fluorescein markers are prepared in the presence of these coupling agents or by reaction with an isothiocyanate.

Non-limiting examples of paramagnetic ions of potential use as imaging agents include chromium (III), manganese (II), iron (III), iron (II), cobalt (II), nickel (II), copper (II), neodymium (III), samarium (III), ytterbium (III), gadolinium (III), vanadium (II), terbium (III), dysprosium (III), holmium (III) and erbium (III), with gadolinium being particularly preferred. Ions useful in other contexts, such as X ray imaging, include but are not limited to lanthanum (III), gold (III), lead (II), and especially bismuth (III).

Radioisotopes of potential use as imaging or therapeutic agents include astatine2", l4carbon, 5 lchromium, 36chlorine, "cobalt, 58cobalt, copper67, 152Eu, gallium67, 3hydrogen, iodine123, iodine125, iodine131, indium1", 59iron, 32phosphorus, rhenium186, rhenium188, 75selenium, 35sulphur, technicium"1" and yttrium90. 125I is often being preferred for use in certain embodiments, and technecium99"1 and indium1" are also often preferred due to their low energy and suitability for long range detection.

Radioactively labeled proteins or peptides of the present invention may be produced according to well known methods in the art. For instance, they can be iodinated by contact with sodium or potassium iodide and a chemical oxidizing agent such as sodium hypochlorite, or an enzymatic oxidizing agent, such as lactoperoxidase. Proteins or peptides according to the invention may be labeled with technetium99'" by ligand exchange process, for example, by reducing pertechnate with stannous solution, chelating the reduced technetium onto a Sephadex column and applying the peptide to this column or by direct labeling techniques, e.g., by incubating pertechnate, a reducing agent such as SNCl2, a buffer solution such as sodium potassium phthalate solution, and the peptide. Intermediary functional groups that are often used to bind radioisotopes that exist as metallic ions to peptides are diethylenetriaminepenta-acetic acid (DTPA) and ethylene diaminetetra-acetic acid (EDTA). Also contemplated for use are fluorescent labels, including rhodamine, fluorescein isothiocyanate and renographin.

In certain embodiments, the claimed proteins or peptides may be linked to a secondary binding ligand or to an enzyme (an enzyme tag) that will generate a colored product upon contact with a chromo genie substrate. Examples of suitable enzymes include urease, alkaline phosphatase, (horseradish) hydrogen peroxidase and glucose oxidase. Preferred secondary binding ligands are biotin and avidin or streptavidin compounds. The use of such labels is well known to those of skill in the art in light and is described, for example, in U.S. Patents 3,817,837; 3,850,752; 3,939,350; 3,996,345; 4,277,437; 4,275,149 and 4,366,241 ; each incorporated herein by reference.

H. Cross-linkers Bifunctional cross-linking reagents have been extensively used for a variety of purposes including preparation of affinity matrices, modification and stabilization of diverse structures, identification of ligand and receptor binding sites, and structural studies. Homobifunctional reagents that carry two identical functional groups proved to be highly efficient in inducing cross-linking between identical and different macromolecules or subunits of a macromolecule, and linking of polypeptide ligands to their specific binding sites. Heterobifunctional reagents contain two different functional groups. By taking advantage of the differential reactivities of the two different functional groups, cross-linking can be controlled both selectively and sequentially. The bifunctional cross-linking reagents can be divided according to the specificity of their functional groups, e.g., amino, sulfhydryl, guanidino, indole, carboxyl specific groups. Of these, reagents directed to free amino groups have become especially popular because of their commercial availability, ease of synthesis and the mild reaction conditions under which they can be applied. A majority of heterobifunctional cross-linking reagents contains a primary amine-reactive group and a thiol-reactive group.

Exemplary methods for cross-linking ligands to liposomes are described in U.S. Patents 5,603,872 and 5,401,511 , each specifically incorporated herein by reference in its entirety. Various ligands can be covalently bound to liposomal surfaces through the cross- linking of amine residues. Liposomes, in particular, multilamellar vesicles (MLV) or unilamellar vesicles such as microemulsified liposomes (MEL) and large unilamellar liposomes (LUVET), each containing phosphatidylethanolamine (PE), have been prepared by established procedures. The inclusion of PE in the liposome provides an active functional residue, a primary amine, on the liposomal surface for cross-linking purposes. Ligands such as epidermal growth factor (EGF) have been successfully linked with PE-liposomes. Ligands are bound covalently to discrete sites on the liposome surfaces. The number and surface density of these sites are dictated by the liposome formulation and the liposome type. The liposomal surfaces may also have sites for non-covalent association. To form covalent conjugates of ligands and liposomes, cross-linking reagents have been studied for effectiveness and biocompatibility. Cross-linking reagents include glutaraldehyde (GAD), bifunctional oxirane (OXR), ethylene glycol diglycidyl ether (EGDE), and a water soluble carbodiimide, preferably l-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC). Through the complex chemistry of cross-linking, linkage of the amine residues of the recognizing substance and liposomes is established.

In another example, heterobifunctional cross-linking reagents and methods of using the cross-linking reagents are described (U.S. Patent 5,889,155, specifically incorporated herein by reference in its entirety). The cross-linking reagents combine a nucleophilic hydrazide residue with an electrophilic maleimide residue, allowing coupling in one example, of aldehydes to free thiols. The cross-linking reagent can be modified to cross- link various functional groups.

V. NUCLEIC ACIDS Nucleic acids according to the present invention may encode a targeting peptide, a receptor protein, a fusion protein, or other protein or peptide. The nucleic acid may be derived from genomic DNA, complementary DNA (cDNA) or synthetic DNA. Where incorporation into an expression vector is desired, the nucleic acid may also comprise a natural intron or an intron derived from another gene. Such engineered molecules are sometime referred to as "mini-genes."

A "nucleic acid" as used herein includes single-stranded and double-stranded molecules, as well as DNA, RNA, chemically modified nucleic acids and nucleic acid analogs. It is contemplated that a nucleic acid within the scope of the present invention may be of almost any size, determined in part by the length of the encoded protein or peptide. It is contemplated that targeting peptides, fusion proteins and receptors may be encoded by any nucleic acid sequence that encodes the appropriate amino acid sequence. The design and production of nucleic acids encoding a desired amino acid sequence is well known to those of skill in the art, using standardized codon tables. In preferred embodiments, the codons selected for encoding each amino acid may be modified to optimize expression of the nucleic acid in the host cell of interest. Codon preferences for various species of host cell are well known in the art.

In addition to nucleic acids encoding the desired peptide or protein, the present invention encompasses complementary nucleic acids that hybridize under high stringency conditions with such coding nucleic acid sequences. High stringency conditions for nucleic acid hybridization are well known in the art. For example, conditions may comprise low salt and/or high temperature conditions, such as provided by about 0.02 M to about 0.15 M NaCl at temperatures of about 500C to about 7O0C. It is understood that the temperature and ionic strength of a desired stringency are determined in part by the length of the particular nucleic acid(s), the length and nucleotide content of the target sequence(s), the charge composition of the nucleic acid(s), and to the presence or concentration of formamide, tetramethyl ammonium chloride or other solvent(s) in a hybridization mixture.

A. Vectors for Cloning, Gene Transfer and Expression In certain embodiments expression vectors are employed to express the targeting peptide or fusion protein, which can then be purified and used. In other embodiments, the expression vectors are used in gene therapy. Expression requires that appropriate signals be provided in the vectors, and which include various regulatory elements, such as enhancers/promoters from both viral and mammalian sources that drive expression of the genes of interest in host cells. Elements designed to optimize messenger RNA stability and translatability in host cells also are known.

1. Regulatory Elements The terms "expression construct" or "expression vector" are meant to include any type of genetic construct containing a nucleic acid coding for a gene product in which part or all of the nucleic acid coding sequence is capable of being transcribed. In preferred embodiments, the nucleic acid encoding a gene product is under transcriptional control of a promoter. A "promoter" refers to a DNA sequence recognized by the synthetic machinery of the cell, or introduced synthetic machinery, required to initiate the specific transcription of a gene. The phrase "under transcriptional control" means that the promoter is in the correct location and orientation in relation to the nucleic acid to control RNA polymerase initiation and expression of the gene.

The particular promoter employed to control the expression of a nucleic acid sequence of interest is not believed to be important, so long as it is capable of directing the expression of the nucleic acid in the targeted cell. Thus, where a human cell is targeted, it is preferable to position the nucleic acid coding region adjacent and under the control of a promoter that transcriptionally active in human cells. Generally speaking, such a promoter might include either a human or viral promoter.

In various embodiments, the human cytomegalovirus (CMV) immediate early gene promoter, the SV40 early promoter, the Rouse sarcoma virus long terminal repeat, rat insulin promoter, and glyceraldehyde-3 -phosphate dehydrogenase promoter can be used to obtain high-level expression of the coding sequence of interest. The use of other viral or mammalian cellular or bacterial phage promoters that are well-known in the art to achieve expression of a coding sequence of interest is contemplated as well, provided that the levels of expression are sufficient for a given purpose.

Where a cDNA insert is employed, one will typically include a polyadenylation signal to effect proper polyadenylation of the gene transcript. The nature of the polyadenylation signal is not believed to be crucial to the successful practice of the invention, and any such sequence may be employed, such as human growth hormone and SV40 polyadenylation signals. Also contemplated as an element of the expression construct is a terminator. These elements can serve to enhance message levels and to minimize read through from the construct into other sequences.

2. Selectable Markers In certain embodiments of the invention, the cells containing nucleic acid constructs of the present invention may be identified in vitro or in vivo by including a marker in the expression construct. Such markers would confer an identifiable change to the cell permitting easy identification of cells containing the expression construct. Usually the inclusion of a drug selection marker aids in cloning and in the selection of transformants. For example, genes that confer resistance to neomycin, puromycin, hygromycin, DHFR, GPT, zeocin, and histidinol are useful selectable markers. Alternatively, enzymes such as herpes simplex virus thymidine kinase (tk) or chloramphenicol acetyltransferase (CAT) may be employed. Immunologic markers also can be employed. The selectable marker employed is not believed to be important, so long as it is capable of being expressed simultaneously with the nucleic acid encoding a gene product. Further examples of selectable markers are well known to one of skill in the art.

3. Delivery of Expression Vectors There are a number of ways in which expression vectors may introduced into cells. In certain embodiments of the invention, the expression construct comprises a virus or engineered construct derived from a viral genome. The ability of certain viruses to enter cells via receptor-mediated endocytosis, to integrate into host cell genome, and express viral genes stably and efficiently have made them attractive candidates for the transfer of foreign genes into mammalian cells (Ridgeway, 1988; Nicolas and Rubinstein, 1988; Baichwal and Sugden, 1986; Temin, 1986). Preferred gene therapy vectors are generally viral vectors.

In using viral delivery systems, one will desire to purify the virion sufficiently to render it essentially free of undesirable contaminants, such as defective interfering viral particles or endotoxins and other pyrogens such that it will not cause any untoward reactions in the cell, animal or individual receiving the vector construct. A preferred means of purifying the vector involves the use of buoyant density gradients, such as cesium chloride gradient centrifugation.

DNA viruses used as gene vectors include the papovaviruses (e.g., simian virus 40, bovine papilloma virus, and polyoma) (Ridgeway, 1988; Baichwal and Sugden, 1986) and adenoviruses (Ridgeway, 1988; Baichwal and Sugden, 1986).

One of the preferred methods for in vivo delivery involves the use of an adenovirus expression vector. Although adenovirus vectors are known to have a low capacity for integration into genomic DNA, this feature is counterbalanced by the high efficiency of gene transfer afforded by these vectors.

Generation and propagation of adenovirus vectors that are replication deficient depend on a unique helper cell line, designated 293, which is transformed from human embryonic kidney cells by Ad5 DNA fragments and constitutively expresses El proteins (Graham et al, 1977.). Since the E3 region is dispensable from the adenovirus genome (Jones and Shenk, 1978), the current adenovirus vectors, with the help of 293 cells, carry foreign DNA in either the El , the E3, or both regions (Graham and Prevec, 1991.). Helper cell lines may be derived from human cells such as human embryonic kidney cells, muscle cells, hematopoietic cells or human embryonic mesenchymal or epithelial cells. Alternatively, the helper cells may be derived from the cells of other mammalian species that are permissive for human adenovirus. Such cells include, for example, Vero cells or other monkey embryonic mesenchymal or epithelial cells. As discussed, the preferred helper cell line is 293. Racher et al (1995) disclose improved methods for culturing 293 cells and propagating adenovirus.

Adenovirus vectors have been used in eukaryotic gene expression (Levrero et al, 1991; Gomez-Foix et al, 1992) and vaccine development (Grunhaus and Horwitz, 1992; Graham and Prevec, 1991). Animal studies have suggested that recombinant adenovirus could be used for gene therapy (Stratford-Perricaudet and Perricaudet, 1991; Stratford- Perricaudet et al, 1990; Rich et al, 1993). Studies in administering recombinant adenovirus to different tissues include tracheal instillation (Rosenfeld et al, 1991; Rosenfeld et al, 1992), muscle injection (Ragot et al, 1993), peripheral intravenous injections (Herz and Gerard, 1993) and stereotactic innoculation into the brain (Le Gal La Salle et al, 1993).

Other gene transfer vectors may be constructed from retroviruses (Coffin, 1990). The retroviral genome contains three genes, gag, pol, and env. that code for capsid proteins, polymerase enzyme, and envelope components, respectively. A sequence found upstream from the gag gene contains a signal for packaging of the genome into virions. Two long terminal repeat (LTR) sequences are present at the 5' and 3' ends of the viral genome. These contain strong promoter and enhancer sequences, and also are required for integration in the host cell genome (Coffin, 1990).

In order to construct a retroviral vector, a nucleic acid encoding a protein of interest is inserted into the viral genome in the place of certain viral sequences to produce a virus that is replication-defective. In order to produce virions, a packaging cell line containing the gag, pol, and env genes, but without the LTR and packaging components, is constructed (Mann et al, 1983). When a recombinant plasmid containing a cDNA, together with the retroviral LTR and packaging sequences is introduced into this cell line (by calcium phosphate precipitation for example), the packaging sequence allows the RNA transcript of the recombinant plasmid to be packaged into viral particles, which are then secreted into the culture media (Nicolas and Rubenstein, 1988; Temin, 1986; Mann et al, 1983). The media containing the recombinant retroviruses is then collected, optionally concentrated, and used for gene transfer. Retroviral vectors are capable of infecting a broad variety of cell types. However, integration and stable expression require the division of host cells (Paskind et al, 1975).

Other viral vectors may be employed as expression constructs. Vectors derived from viruses such as vaccinia virus (Ridgeway, 1988; Baichwal and Sugden, 1986; Coupar et al, 1988), adeno-associated virus (AAV) (Ridgeway, 1988; Baichwal and Sugden, 1986; Hermonat and Muzycska, 1984), and herpes viruses may be employed. They offer several attractive features for various mammalian cells (Friedmann, 1989; Ridgeway, 1988; Baichwal and Sugden, 1986; Coupar et al, 1988; Horwich et al, 1990).

Several non-viral methods for the transfer of expression constructs into cultured mammalian cells also are contemplated by the present invention. These include calcium phosphate precipitation (Graham and van der Eb, 1973.; Chen and Okayama, 1987.; Rippe et al, 1990; DEAE dextran (Gopal, et al, 1985), electroporation (Tur-Kaspa et al, 1986; Potter et al, 1984), direct microinjection, DNA-loaded liposomes and lipofectamine-DNA complexes, cell sonication, gene bombardment using high velocity microprojectiles, and receptor-mediated transfection (Wu and Wu, 1987; Wu and Wu, 1988). Some of these techniques may be successfully adapted for in vivo or ex vivo use.

In a further embodiment of the invention, the expression construct may be entrapped in a liposome. Liposome-mediated nucleic acid delivery and expression of foreign DNA in vitro has been very successful. Wong et al. (1980) demonstrates the feasibility of liposome-mediated delivery and expression of foreign DNA in cultured chick embryo, HeLa, and hepatoma cells. Nicolau et al. (1987) accomplished successful liposome-mediated gene transfer in rats after intravenous injection.

VI. PHARMACEUTICAL COMPOSITIONS Where clinical applications are contemplated, it may be necessary to prepare pharmaceutical compositions - expression vectors, virus stocks, proteins, antibodies and drugs - in a form appropriate for the intended application. Generally, this will entail preparing compositions that are essentially free of impurities that could be harmful to humans or animals.

One generally will desire to employ appropriate salts and buffers to render delivery vectors stable and allow for uptake by target cells. Buffers also are employed when recombinant cells are introduced into a patient. Aqueous compositions of the present invention may comprise an effective amount of a protein, peptide, antibody, fusion protein, recombinant phage and/or expression vector, dissolved or dispersed in a pharmaceutically acceptable carrier or aqueous medium. The phrase "pharmaceutically or pharmacologically acceptable" refers to molecular entities and compositions that do not produce adverse, allergic, or other untoward reactions when administered to an animal or a human. As used herein, "pharmaceutically acceptable carrier" includes any and all solvents, dispersion media, coatings, antibacterial and antifungal agents, isotonic and absorption delaying agents and the like. The use of such media and agents for pharmaceutically active substances is well known in the art. Except insofar as any conventional media or agent is incompatible with the proteins or peptides of the present invention, its use in therapeutic compositions is contemplated. Supplementary active ingredients also can be incorporated into the compositions.

The active compositions of the present invention may include classic pharmaceutical preparations. Administration of these compositions according to the present invention are via any common route so long as the target tissue is available via that route. This includes oral, nasal, buccal, rectal, vaginal or topical. Alternatively, administration may be by orthotopic, intradermal, subcutaneous, intramuscular, intraperitoneal, intraarterial or intravenous injection. Such compositions normally would be administered as pharmaceutically acceptable compositions, described supra.

The pharmaceutical forms suitable for injectable use include sterile aqueous solutions or dispersions and sterile powders for the extemporaneous preparation of sterile injectable solutions or dispersions. In all cases the form must be sterile and must be fluid to the extent that easy syringability exists. It must be stable under the conditions of manufacture and storage and must be preserved against the contaminating action of microorganisms, such as bacteria and fungi. The carrier can be a solvent or dispersion medium containing, for example, water, ethanol, polyol (for example, glycerol, propylene glycol, and liquid polyethylene glycol, and the like), suitable mixtures thereof, and vegetable oils. The proper fluidity can be maintained, for example, by the use of a coating, such as lecithin, by the maintenance of the required particle size in the case of dispersion and by the use of surfactants. The prevention of the action of microorganisms can be brought about by various antibacterial and antifungal agents, for example, parabens, chlorobutanol, phenol, sorbic acid, thimerosal, and the like. In many cases, it is preferable to include isotonic agents, for example, sugars or sodium chloride. Prolonged absorption of the injectable compositions can be brought about by the use in the compositions of agents delaying absorption, for example, aluminum monostearate and gelatin.

Sterile injectable solutions are prepared by incorporating the active compounds in the required amount in the appropriate solvent with various other ingredients enumerated above, as required, followed by filtered sterilization. Generally, dispersions are prepared by incorporating the various sterilized active ingredients into a sterile vehicle which contains the basic dispersion medium and the required other ingredients from those enumerated above. In the case of sterile powders for the preparation of sterile injectable solutions, the preferred methods of preparation are vacuum-drying and freeze-drying techniques which yield a powder of the active ingredient plus any additional desired ingredient from a previously sterile-filtered solution thereof.

VII. THERAPEUTIC AGENTS In certain embodiments, therapeutic agents may be attached to a targeting peptide or fusion protein for selective delivery to, for example, leukemic cells or derivatives thereof. Agents or factors suitable for use may include any chemical compound that induces apoptosis, cell death, cell stasis and/or anti-angiogenesis.

A. Regulators of Programmed Cell Death Apoptosis, or programmed cell death, is an essential process for normal embryonic development, maintaining homeostasis in adult tissues, and suppressing carcinogenesis (Kerr et al, 1972). The Bcl-2 family of proteins and ICE-like proteases have been demonstrated to be important regulators and effectors of apoptosis in other systems. The BcI 2 protein, discovered in association with follicular lymphoma, plays a prominent role in controlling apoptosis and enhancing cell survival in response to diverse apoptotic stimuli (Bakhshi et al, 1985; Cleary and Sklar, 1985; Cleary et al, 1986; Tsujimoto et al, 1985; Tsujimoto and Croce, 1986). The evolutionarily conserved BcI 2 protein now is recognized to be a member of a family of related proteins, which can be categorized as death agonists or death antagonists.

Subsequent to its discovery, it was shown that BcI 2 acts to suppress cell death triggered by a variety of stimuli. Also, it now is apparent that there is a family of BcI 2 cell death regulatory proteins that share in common structural and sequence homologies. These different family members have been shown to either possess similar functions to BcI 2 {e.g., BcIXL, BcIW, BcIS, McI-I, Al, BfI-I) or counteract BcI 2 function and promote cell death (e.g., Bax, Bak, Bik, Bim, Bid, Bad, Harakiri).

Non-limiting examples of pro-apoptosis agents contemplated within the scope of the present invention include gramicidin, magainin, mellitin, defensin, cecropin, (KLAKLAK)2 (SEQ ID NO: 1), (KLAKKLA)2 (SEQ ID NO: 2), (KAAKKAA)2 (SEQ ID NO: 3) or (KLGKKLG)3 (SEQ ID NO: 4).

6. Angiogenic inhibitors In certain embodiments the present invention may concern administration of targeting peptides attached to anti-angiogenic agents, such as angiotensin, laminin peptides, fibronectin peptides, plasminogen activator inhibitors, tissue metalloproteinase inhibitors, interferons, interleukin 12, platelet factor 4, IP-10, Gro-β, thrombospondin, 2- methoxyoestradiol, proliferin-related protein, carboxiamidotriazole, CMlOl, Marimastat, pentosan polysulphate, angiopoietin 2 (Regeneron), interferon-alpha, herbimycin A, PNU145156E, 16K prolactin fragment, Linomide, thalidomide, pentoxifylline, genistein, TNP-470, endostatin, paclitaxel, accutin, angiostatin, cidofovir, vincristine, bleomycin, AGM- 1470, platelet factor 4 or minocycline.

Proliferation of some tumor or cancer cells rely heavily on extensive tumor vascularization, which accompanies cancer progression. Thus, inhibition of new blood vessel formation with anti-angiogenic agents and targeted destruction of existing blood vessels have been introduced as an effective and relatively non-toxic approach to tumor treatment. (Arap et al, 1998; Arap et al., 1998; Ellerby et al, 1999). A variety of anti-angiogenic agents and/or blood vessel inhibitors are known {e.g., Folkman, 1997; Eliceiri and Cheresh, 2001).

C. Cytotoxic Agents

Chemotherapeutic (cytotoxic) agents of potential use include, but are not limited to, 5-fluorouracil, bleomycin, busulfan, camptothecin, carboplatin, chlorambucil, cisplatin (CDDP), cyclophosphamide, dactinomycin, daunorubicin, doxorubicin, estrogen receptor binding agents, etoposide (VP 16), farnesyl-protein transferase inhibitors, gemcitabine, ifosfamide, mechlorethamine, melphalan, mitomycin, navelbine, nitrosurea, plicomycin, procarbazine, raloxifene, tamoxifen, taxol, temazolomide (an aqueous form of DTIC), transplatinum, vinblastine and methotrexate, vincristine, or any analog or derivative variant of the foregoing. Most chemotherapeutic agents fall into the categories of alkylating agents, antimetabolites, antitumor antibiotics, corticosteroid hormones, mitotic inhibitors, and nitrosoureas, hormone agents, miscellaneous agents, and any analog or derivative variant thereof.

Chemotherapeutic agents and methods of administration, dosages, etc. are well known to those of skill in the art (see for example, the "Physicians Desk Reference", Goodman & Gilman's "The Pharmacological Basis of Therapeutics" and in "Remington's Pharmaceutical Sciences" 15th ed., pp 1035-1038 and 1570-1580, each incorporated herein by reference in relevant parts), and may be combined with the invention in light of the disclosures herein. Some variation in dosage will necessarily occur depending on the condition of the subject being treated. The person responsible for administration will, in any event, determine the appropriate dose for the individual subject. Examples of specific chemotherapeutic agents and dose regimes are also described herein. Of course, all of these dosages and agents described herein are exemplary rather than limiting, and other doses or agents may be used by a skilled artisan for a specific patient or application. Any dosage within these points, or range derivable therein is also expected to be of use in the invention.

D. Alkylating agents Alkylating agents are drugs that directly interact with genomic DNA to prevent cells from proliferating. This category of chemotherapeutic drugs represents agents that affect all phases of the cell cycle, that is, they are not phase-specific. An alkylating agent, may include, but is not limited to, a nitrogen mustard, an ethylenimene, a methylmelamine, an alkyl sulfonate, a nitrosourea or a triazines. They include but are not limited to: busulfan, chlorambucil, cisplatin, cyclophosphamide (Cytoxan), dacarbazine, ifosfamide, mechlorethamine (mustargen), and melphalan.

E. Antimetabolites

Antimetabolites disrupt DNA and RNA synthesis. Unlike alkylating agents, they specifically influence the cell cycle during S phase. Antimetabolites can be differentiated into various categories, such as folic acid analogs, pyrimidine analogs and purine analogs and related inhibitory compounds. Antimetabolites include but are not limited to, 5-fluorouracil (5-FU), cytarabine (Ara-C), fludarabine, gemcitabine, and methotrexate. F. Natural Products Natural products generally refer to compounds originally isolated from a natural source, and identified as having a pharmacological activity. Such compounds, analogs and derivatives thereof may be, isolated from a natural source, chemically synthesized or recombinantly produced by any technique known to those of skill in the art. Natural products include such categories as mitotic inhibitors, antitumor antibiotics, enzymes and biological response modifiers.

Mitotic inhibitors include plant alkaloids and other natural agents that can inhibit either protein synthesis required for cell division or mitosis. They operate during a specific phase during the cell cycle. Mitotic inhibitors include, for example, docetaxel, etoposide (VP 16), teniposide, paclitaxel, taxol, vinblastine, vincristine, and vinorelbine.

Taxoids are a class of related compounds isolated from the bark of the ash tree, Taxus brevifolia. Taxoids include but are not limited to compounds such as docetaxel and paclitaxel. Paclitaxel binds to tubulin (at a site distinct from that used by the vinca alkaloids) and promotes the assembly of microtubules.

Vinca alkaloids are a type of plant alkaloid identified to have pharmaceutical activity. They include such compounds as vinblastine (VLB) and vincristine.

G. Antibiotics Certain antibiotics have both antimicrobial and cytotoxic activity. These drugs also interfere with DNA by chemically inhibiting enzymes and mitosis or altering cellular membranes. These agents are not phase specific so they work in all phases of the cell cycle. Examples of cytotoxic antibiotics include, but are not limited to, bleomycin, dactinomycin, daunorubicin, doxorubicin (Adriamycin), plicamycin (mithramycin) and idarubicin.

H. Miscellaneous Agents

Miscellaneous cytotoxic agents that do not fall into the previous categories include, but are not limited to, platinum coordination complexes, anthracenediones, substituted ureas, methyl hydrazine derivatives, amsacrine, L-asparaginase, and tretinoin. Platinum coordination complexes include such compounds as carboplatin and cisplatin (cis- DDP). An exemplary anthracenedione is mitoxantrone. An exemplary substituted urea is hydroxyurea. An exemplary methyl hydrazine derivative is procarbazine (N- methylhydrazine, MIH). These examples are not limiting and it is contemplated that any known cytotoxic, cytostatic or cytocidal agent may be attached to targeting peptides and administered to a targeted organ, tissue or cell type within the scope of the invention.

I. Dosages

The skilled artisan is directed to "Remington's Pharmaceutical Sciences" 15th Edition, chapter 33, and in particular to pages 624-652. Some variation in dosage will necessarily occur depending on the condition of the subject being treated. The person responsible for administration will, in any event, determine the appropriate dose for the individual subject. Moreover, for human administration, preparations should meet sterility, pyrogenicity, and general safety and purity standards as required by the FDA Office of Biologies standards.

EXAMPLES

The following examples are included to demonstrate preferred embodiments of the invention. It should be appreciated by those of skill in the art that the techniques disclosed in the examples which follow represent techniques discovered by the inventors to function well in the practice of the invention, and thus can be considered to constitute preferred modes for its practice. However, those of skill in the art should, in light of the present disclosure, appreciate that many changes can be made in the specific embodiments which are disclosed and still obtain a like or similar result without departing from the spirit and scope of the invention.

EXAMPLE 1

IDENTIFICATION AND CHARACTERIZATION OF MOLT-4 PHAGE CLONES

A. Materials and Methods

Prior to performing binding assays in clinical samples from leukemia patients, BRASIL was used to profile several human-derived leukemia cell lines from the NCI-60 cell panel (dtpws4.ncifcrf.gov) including MOLT-4, CCRF-CEM, SR, RPMI-8226, K-562, and HL-60. The cells were incubated with either the CX5C or CX7C random phage libraries, or with the insert-less control phage (Fd-tet). Phage bound to the cells was then recovered, and quantified in 4 consecutive rounds of panning. In comparison to control phage binding to cells, enrichment in particular for phage displaying peptides that bind specifically to the targeted cell type was obtained. Titer Determination for Selected Molt-4 Phage Clones and Fd-tet Amplified by Aki. Four microliters (μl) each of 10~3 - 10"8 phage serial dilutions in PBS was used to infect 400μl stationary - phase K91 bacteria at RT for 30 min. lOOμl of each infection were cultured in triplicates on LB + Tet + Kana plates at 370C overnight.

PCR and Sequencing to Confirm Identity of Amplified Molt-4 Phage Clones. A PCR master mix was prepared in that 20μl reaction per sample included lμl of 1 :1000 dilution in PBS of each phage clone, lμl Fuse-5 reverse primer (8 pmol/μl), lμl Fuse-5 forward primer (8 pmol/μl), lμl dNTP mix (2.5mM, Promega), 2μl Taq polymerase buffer (10X, Promega), 0.5μl Taq DNA polymerase (5U/μl, Promega), and 13.5 μl dH2O. PCR was performed in polypropylene V-bottom microplates (MJ Research Inc.) using 94°C annealing temperature (T) for 3 min, then 30 cycles of 94°C 10 sec, 60°C 30 sec, 72°C 30 sec. Dilutions (1 : 10) of the PCR products were prepared and stored at 4°C until sent for sequencing while the rest were stored at -2O0C. Seven samples were sequenced at DNA CORE facility at M.D. Anderson Cancer Center, Houston, Tx.

Validation of Selected Molt-4 Phage Clones by Assessing their Binding to a Panel of Leukemia Cell Lines Using BRASIL. Cells grown in suspension were washed once with PBS and cell count was determined using trypan blue vital stain. Cells were suspended at 1x106 cell per lμl in 1 % BSA/RPMI. Cell suspension was then transferred in 200 μl aliquots into micro-centrifuge tubes. Each phage clone (1 x 109 TU) were added to 200 μl cell sample and incubated for 5 hr on ice with occasional slight vortexing. Meanwhile, K91 Kana bacteria were inoculated in TB medium containing supplements and 100 μg/ml kanamycin. Bacteria were used when 1 :10 dilution reached 0.180 to 0.200 OD (600 nm). BRASIL separation. The oil (45 ml dibutyl phtalate + 5 ml cyclohexane in a 50 ml Falcon) was stored at RT and protected from light. Two hundred μl of oil were added to a 400 μl microcentrifuge tube (VWR Scientific Products, cat. No. 20170-326). The phage-cell mix (similar volume) on top of the oil was transferred and spun at 10,000 rpm for 10 min at 4°C. If the pellet remained in the interface, centrifugation was repeated at room temperature. The tubes were placed at - 800C for approximately 10 min (until top layer was frozen or until bacteria were ready). The bottom of the BRASIL tube was cut close to the pellet and then put into a 1.5 ml eppendorf tube. Any residual oil was then removed. Two-hundred μl of the K91-Kana growth medium were added and the cell pellet suspended using a pipette tip. Cell- bound phage and bacteria were incubated for 1- 1.5 hrs at room temperature. Infected bacteria (200 μl) was transferred to 10 ml of pre-warmed LB liquid medium supplemented with 20 μg/ml Tetracyclin and 100 μg/ml Kanamycin and incubated at room temperature for approximately 30 minutes. Dilutions (1:1, 1:10, 1:100) of each sample were plated in triplicate on LB+Tet+Kana plates and incubated at 37°C overmight for colony counts the next day.

Validation of Selected Molt-4 Phage Clones by Assessing their Binding to Acute Myelogenous Leukemia (AML) Clinical Samples Using BRASIL. Ten clinical samples from AML patients were provided by the Leukemia Cell Bank at the M. D. Anderson Cancer Center (Table 2). The samples were covered by an IRB protocol pre-submitted by Dr. Frank Marini, M.D. Anderson Cancer Center. The cryo-preserved samples were stored at -80°C until they were processed. Samples were quickly thawed in a 37°C water bath, counted using trypan blue stain, and resuspended in fresh complete RPMI media at approximately 1x10 cells/ml, then placed in culture at 37°C, 5% CO2 for BRASIL the next day. For BRASIL, cultured cells were washed once with PBS and vital cell count was determined using trypan blue stain. Re-suspended cells at 1x106 per lμl in 1 % BSA/RPMI where applicable then transferred 200 μl aliquots into micro-centrifuge tubes. For some clinical samples with low recovery, a lower cell number was used for BRASIL since a significant percentage of the recovered cells were dead on the day BRASIL was performed. Added 1x109 TU of each phage clone to 200,000 cells per 200 μl sample or scaled down amount of phage accordingly whenever lower number of cells was used and incubated for 2.5 hrs on ice with occasional slight vortexing. Meanwhile, K91 Kana bacteria were inoculated in TB medium containing supplements and 100 μg/ml kanamycin. Bacteria were used when 1 :10 dilution reached 0.180 to 0.200 OD (600 nm). BRASIL separation: The oil (45 ml dibutyl phtalate + 5 ml cyclohexane in a 50 ml Falcon) was stored at room temperature, protected from light. Added 200 μl of oil on a 400 μl - volume microcentrifuge tube (VWR Scientific Products, cat. No. 20170-326). Transferred the phage-cell mix (similar volume) on top of the oil and spun at 10,000 rpm for 10 min at room temperature. Placed the tubes at -800C for approximately 10 min (until top layer was frozen or until bacteria were ready). The bottom of the BRASIL tube was cut close to the pellet and then put into a 1.5 ml eppendorf tube. As much residual oil as possible was removed. Added 200 μl of the K91-Kana growth and suspended the cell pellet well using pipette tips. Incubated the cell-bound phage and bacteria for 1.5 hrs at room temperature. Transferred each 200 μl of infected bacteria to 10 ml of pre-warmed LB liquid medium supplemented with 20 μg/ml Tetracyclin and 100 μg/ml Kanamycin and incubated at room temperature for approximately 30 min. Plated 20, and 50 μl dilutions of each sample in duplicates on LB + Tet + Kana plates and incubated at 37°C overnight for colony counts the next day.

Bone Marrow Stromal Cell Studies. Adherent bone marrow stromal cells (MSCs) were isolated from portions of patient samples and from BMMNCs of healthy donor. MSCs from patients and from healthy marrow will be expanded in culture and BRASIL will be done on them to evaluate if any of the Molt-4 phage clones shows differential binding to MSCs from AML patient as compared to MSCs from healthy marrow. Phage binding assays performed so far on all 9 AML patient samples were performed using all the mononuclear cells in the sample and not MSCs in particular. If any difference is observed upon assaying MSCs only, additional AML samples may be requested from the Leukemia Cell Bank to perform BRASIL on MSCs exclusively.

B. Results

Titers were obtained for the amplified Molt-4 phage clones (Table 2). The titers of all the clones were relatively comparable with Fd-tet phage prep. Panning with the CX5C phage library on MOLT-4, a cell line derived from a patient with Acute Lymphoblastic Leukemia (ALL), the inventors obtained enrichment for several peptides of which SVWFG was expressed at large scale and was further purified using Ni-NTA column. Furthermore, screening MOLT-4 cells with CX7C phage library yielded several peptide motifs with high frequency (Table 1). The binding capacity of these phage-di splayed peptides was assessed by performing BRASIL panning on MOLT-4 cells (Figure 14). In brief, the results show that GFYWLRS (SEQ ID NO: 6) peptide had 10-fold higher phage-binding to MOLT-4 cells relative to the Fd-tet negative control phage. This reflects a high level of expression of the corresponding receptor on the surface of MOLT-4 leukemia cells. In addition, the peptides SFFYLRS (SEQ ID NO: 7) and AYHRLRR (SEQ ID NO: 5) showed 5-fold and 4-fold enhanced binding capacity respectively as compared to the control phage suggesting that their receptors may also be preferentially expressed on the surface of MOLT-4 cells. Table 1. Enriched sequences from CX7C screening on Molt-4 cell line

Table 2. Titer Determination and Sequence Confirmation of Selected Molt-4 Phage Clones

Table 3. Candidate ligands mimicked by Molt-4 phage clones and their putative receptors or interacting proteins.

EXAMPLE 2 IDENTIFICATION OF CORRESPONDING RECEPTORS AND ASSESSMENT OF THE LIGAND-RECEPTOR PAIR INTERACTION MECHANISMS.

A. Material and Methods

The selection of the peptides for which the receptor will be pursued is based on three criteria: (i) the binding avidity of the peptide, (ii) databank searches for identification of the ligand for a given peptide, and (iii) various biochemical and genetic approaches for identification and/or validation of a ligand/receptor pair.

Identifying proteins mimicked by the peptide motifs and their putative receptors using BLAST search. BLAST is an approach that is used to elucidate the identity of the molecules to which the peptide inserts of the selected phage bind. There are successful examples in the literature for identifying several ligand/receptor pairs using the standard protein-protein BLAST [blastp] algorithm. In previous work, peptide motifs were identified with selective affinities for many organs and tumors. The receptors to which the organ and tumor specific phage bind to have been characterized in the context of different vascular beds. It is important to identify the receptors so that a better understanding of what type of molecules the phage peptides recognize in leukemia and what these molecules reveal in terms of their distribution and biological function. The identification of novel receptors in leukemia may shed light on the complex cellular and molecular diversity of this disease.

Validation of candidate receptors by performing phage binding assays on available recombinant proteins. These assays are done with minor modifications to the standard phage binding assays. Commercially available recombinant proteins are immobilized in wells of 96-well plates (lug per well) at 4°C overnight. Excess unbound proteins are removed by a wash step. A blocking step with 1% BSA in PBS follows. Single phage clones are then added to their respective wells using the same TU number and are incubated at RT for 2hr to allow for binding. Unbound phage are removed by several wash steps, followed by K91 infection, transfer to LB media, and plating. Colonies are counted the next day and results are interpreted in terms of fold difference in phage binding to a particular recombinant protein relative to insertless phage which is used as baseline. Synthetic peptides will be used that correspond to the sequence displayed by the phage to perform inhibitory studies. This assay determines whether phage binding is entirely mediated by the peptide displayed by the phage. It is contemplated that the synthetic peptides will inhibit the binding of the corresponding phage in a dose-dependent manner. A control peptide containing unrelated amino acids will be tested at identical concentrations.

GST-fusion proteins can be designed for receptor identification and/or validation by biochemical approaches. Peptide-coding DNA sequences will be amplified by colony PCR using forward and reverse primers that contained BamHI and EcoRI sites, respectively. The amplified sequences will be cloned into the BamHI-EcoRI sites of the GST vector, pGEX-2TK (Amersham/Pharmacia), and the presence of the inserted sequences will be verified by sequence analysis. Positive clones will be transformed into a bacterial expression host strain, BL21(DE3)pLysS (Stratagene) and expression of the GST-fusion proteins will be induced with 200μM IPTG. The GST-fusion proteins will be affinity purified from bacterial lysates by affinity chromatography to immobilized glutathione. The GST-fusion proteins will batch-bound to glutathione Sepharose 4B beads, and the resin rinsed to remove non-specific proteins. GST-fusion proteins will be eluted by incubating the resin with an excess of reduced glutathione, followed by extensive dialysis of the eluted protein against phosphate buffered saline, pH 7.4 (PBS) to remove the glutathione.

Affinity chromatography. If leukemia cell lines can be found to express a particular receptor (this can be evaluated using phage binding assays), it will be possible to use positive cell lines as the source of protein. Once we have identified a gel band in the material that binds to the peptide column and elutes specifically with the corresponding synthetic peptide, the options are to microsequence the protein or to prepare antibodies to it. If the sequence is new, it will be used to design oligonucleotide probes for the isolation of cDNA clones. Antibodies may also be used for the isolation of clones from bacterial expression libraries. A possible, but as yet untested, alternative way of isolating clones for the receptor molecules is to screen a bacterially expressed library with the phage carrying the appropriate insert.

Mass Spectrometry. Mass spectrometric peptide mapping can be used to identify novel target receptors. Polyclonal and monoclonal antibodies raised against the candidate receptors will also be used to purify target proteins. These proteins will be resolved by SDS- PAGE, will be cut out from the SDS gels, and digested in-gel with trypsin. After extraction of the peptides, MALDI-TOF mass spectrometry analysis will be performed to produce a list of peptide masses. This list of peptide masses, in combination with protease specificity, produces a relatively specific "signature" that can be used to search sequence databases. If the protein sequence is present in a database, the protein can be identified with high confidence by this method. The lower detection limit for this approach is currently 1 pmol, at least 10-20- fold better than N-terminal Edman sequencing methods.

Biacore and Protein Arrays. Prospective ligand-receptor pairs will be validated using a Biacore system. The surface plasmon resonance (SPR) measurements will monitor the specificity of interaction, in real time, between two species without the need for labels. As the sample is passed over the immobilized interactant, the progress is monitored through changes in the SPR signal. This system will allow us to measure binding affinities of candidate receptor-ligand pairs through either the rate constant measurements or through the analysis of binding as a function of sample concentration. As a complementary approach, protein arrays can be probed with the most promising ligands (phage or labeled peptides). This may accelerate the identification of receptor-ligand pairs as multiple ligands can be probed simultaneously in the context of a protein array.

Receptor identification and validation by genetic approaches. Yeast two hybrid cloning can be used for receptor identification. DNA encoding the peptide motif with homing properties is fused to a binding domain to screen a cDNA library fused to a transactivation domain. It is contemplated that clones will be found that are capable of binding to the peptide used as the bait; those clones may code for the relevant receptor. However, the encoded protein may also be an irrelevant binder of the peptide, but even in that case it may be useful because antibodies against it may recognize the relevant receptor. Antibodies will be developed to an identified receptor and used to identify cDNA clones, as well as to study the expression of the receptor. Conversely, if cDNA cloning is successful first, recombinant protein will be used to prepare antibodies. An alternative to the conventional protein isolation strategy outlined above is direct expression cloning of receptors in mammalian cells. The COS-cell system will be employed. These cells can be transfected with an expression library from cell or marrow tissue mRNA in the pcDNA3 vector (Invitrogen).

Evaluation of the expression profile of target receptors by cellular immunostaining and immunohistochemistry. Identification of receptors makes it possible to use antibodies against them to accurately map their tissue localization and expression profile in the context of leukemia progression. If the identified receptors are known proteins with commercially available antibodies, these antibodies can be used for immunostaining of cells and for immunohistochemical studies. However, if commercial antibodies are not available, polyclonal antibodies against the receptor can be developed, using for example recombinant proteins developed from cDNA isolation to generate antibodies. Antibody generation is well known in the art. Briefly, for example, after collection of a pre-immune blood sample, two New Zealand White female rabbits will be immunized with 100-500μg of purified recombinant receptor protein in Freund's complete adjuvant. Rabbits will be boosted every 3-4 weeks with lOOμg of purified recombinant protein in Freund's incomplete adjuvant. One week after every booster immunization, a sample of the rabbit sera, along with a sample of the pre-immune sera, will be analyzed for reactivity against the purified recombinant receptor protein by a standard ELISA assay. When the anti-receptor serum titers will be adequately high against the receptor, the final bleed will be obtained from the rabbits. IgGs will be isolated from the sera by standard affinity purification techniques using Protein G resin.

Immunohistochemical staining. Immunohistochemistry will be performed as described previously (Zurita, 2004). Briefly, 4μm sections will undergo antigen retrieval as well as blocking for endogenous biotin. After protein blocking, tissue sections will be incubated with the anti-receptor antibodies followed by secondary detection with the LSAB kit (DAKO). The staining will be scored and evaluated statistically. Additionally, co- localization analyses will be performed using fluorescence with known leukemia cell markers.

Cellular localization of target receptors by confocal microscopy. Laser scanning confocal microscopy will be used to detail the cellular localization of target receptors from tissue specimens. Standard immunohistochemical staining of tissue sections, generally 5μm in thickness, provides information on distribution within the tissue but it is often difficult to determine the cellular localization. Methods have been developed to examine the cellular distribution of receptors in 60μm thick tissue sections in human specimens by combining fluorescence immunohistochemistry and confocal microscopy (Ozawa et al, submitted for publication). Briefly, fresh specimens are fixed in 4% paraformaldehyde in PBS then infiltrated with a 30% sucrose solution in PBS overnight. The sample is then frozen in O. CT. and sectioned at 60μm. Tissue sections are rinsed, blocked with 5% normal serum for 60 minutes and then incubated with primary antibodies overnight. The following day, sections are rinsed with PBS and incubated with secondary antibodies for 4 hr. Samples are rinsed, fixed with 4% paraformaldehyde, rinsed and then mounted with VECTASHIELD (Vector Laboratories, CA). Slides are observed using a laser scanning confocal microscope (Carl Zeiss Inc, Germany) to detail the cellular context of putative target receptors. Assessment of the expression patterns of identified receptors in human leukemia by phage overlay assays. Phage overlay assays will be carried out on human leukemia tissue samples to study the ligand-receptor binding of the identified peptide ligands. Phage clones isolated from the biopanning on leukemia cells will be evaluated for binding to tissue samples derived from leukemia patients. A well-established phage overlay protocol will be used (Arap et al, 2002; Pasqualini and Ruoslahti, 1996). Briefly, selected phage clones or the negative control fd-tet phage (1-5 x 109 TU/tissue section) will be incubated for two hours at room temperature on 4μm tissue sections. If necessary, the tissue sections will be treated for antigen retrieval prior to incubation. The specifically bound phage will be detected with an anti-bacteriophage antibody (1 :500 dilution, Sigma) incubated for one hour at room temperature followed by a peroxidase-conjugated anti-rabbit secondary antibody. Tissue sections will then be washed, developed with DAB, counterstained with hematoxylin, dehydrated and mounted. Additionally, data on ligand-receptor interaction will be obtained using phage/antibody inhibition assays. To evaluate whether the receptor binding phage can inhibit the anti-receptor antibody (either a commercial or a generated polyclonal source) staining, receptor binding phage and negative control phage will be overlaid on serial human tissue sections prior to adding the anti-receptor antibody or negative control antibody. After washes to remove non-specific antibody binding, specific antibody binding will be detected by incubation with the appropriate secondary antibody. In a reversed experimental setting, a receptor binding phage or negative control phage (1-5 x 109 TU/tissue section) will be overlaid on serial human tissue sections pre-incubated with the anti-receptor antibody. After extensive washing to remove non-specific binding, phage binding will be detected with an anti-bacteriophage antibody. The reduced antibody/phage staining with the corresponding receptor binding phage/anti-receptor antibody will be a strong indication of a specific ligand receptor interaction in situ.

Characterization of the synthetic peptides. Once identified Peptides are characterized as to their ability to bind to and interfere with receptor function. In certain aspects it is important that the synthetic peptide exhibits capacity to deliver a toxic entity such as a pro-apoptotic moiety into target cells. For this purpose, several in vitro assays can be performed on leukemia cells with the targeting synthetic peptides. These include cell viability and cytotoxicity assays to determine the proportion of live and dead cells in the tested populations, cell proliferation assays to assess the growth rate and density of a tested cell population based on monitoring changes in total nucleic acid content (L3224, V231 1 1, and C7026, Molecular Probes). For these fluorometric - based assays, the analysis can be done using a flow cytometer, a fluorescence microplate reader, and/or a fluorescence microscope. Alternatively, colorimetric-based assays could be used such as the MTT cell proliferation assay (ATCC) or the WST-I cell proliferation and cell viability assay (Roche) for which the sample measurements can be acquired and analyzed using an absorbance microplate reader. Since the WST-I assay is primarily based on the cleavage of the WST-I tetrazolium salt by mitochondrial dehydrogenases in viable cells, not only it allows for the measurement of cell proliferation in response to growth factors or mitogens, but it also permits the assessment of growth inhibition by physiological mediators or inhibitory antibodies. Moreover, unlike MTT, the WST-I cleavage product is water-soluble so it is a ready-to-use solution. To detect nuclear breakdown and DNA fragmentation in early stage apoptosis, the APO-BrdU TUNEL Assay (Molecular Probes) can be used coupled to flow cytometry and/or fluorescence imaging. To assess phosphatidylserine externalization in intermediate stage apoptosis, fluorescent conjugates to annexin V, a phospholipid - binding protein, can be used whereas late stage in apoptosis can be evaluated based on cytoskeletal collapse and cell membrane permeability to propidium iodide using flow cytometry and/or fluorescence microscopy.

Cell invasion assays. These are based on the ability of invasive cells to degrade matrix proteins coating an 8 micron-pore size polycarbonate membrane and their subsequent migration into the lower compartment of the culture chamber (Calbiochem). The cells can then be labeled, dissociated, and transferred to a fluorescence microplate reader for quantitative measurements. Trans-endothelial migration assays can be performed in 6.5mm- diameter Transwell Plates (Beckton Dickinson). A monolayer of human vascular endothelial cells is grown on gelatin - coated filters (3 micron-pore size for lymphocytic leukemia cell lines and 8 micron-pore size for monocytic leukemia cell lines). A cocktail of chemotactic agents is added to the lower chamber to mediate migration of cells into the lower chamber. Migrating cells are then collected, stained with 0.1% crystal violet, and counted microscopically.

B. Results

Phage internalization into leukemia cells. Phage internalization assays were performed on K-562 cells (FIG. 15) to determine if the sequence insert in either one of the 3 Molt-4 phage clones is capable of receptor-mediated internalization. This would be indicative of whether the corresponding peptide motif has the potential to deliver drugs or apoptotic moieties into leukemia cells. K562 cells mixed with either phage clone or Fd-tet control phage at 2 x 105 cells per 1 x 109 TU were incubated at 37°C overnight to permit for any potential receptor-mediated internalization to occur. Following multiple wash steps with PBS and glycine-based buffer to remove unbound or surface-bound un-internalized phage, the cells were adhered to glass chamber slides pre-coated with poly-D lysine. Subsequently, the samples were fixed and permeabilized with pre-cooled methanol, blocked with 5% normal goat serum and stained with either an anti-fd phage antibody (SIGMA) or non¬ immune rabbit IgG (DAKO). FITC-conjugated goat anti-rabbit antibody (Jackson Labs) was used for secondary staining after which the samples were mounted with VECTASHIELD/DAPI. As shown in FIG. 15, the C-AYHRLRR-C motif insert in clone 1 mediates very strong internalization into K562 cells. To a lesser extent, the C-GFYWLRS-C insert in clone 2 is also capable of internalizing. On the other hand, even though the C- SFFYLRS-C sequence insert in clone 3 showed the strongest binding to leukemia cells in cell lines and patient samples, it did not exhibit any cytoplasmic staining, hence suggesting that it is incapable of internalizing.

Proteins mimicked by the peptide motifs and putative receptors approach. Preliminary analysis of MOLT-4 binding peptides done by comparison of the selected motif sequences with available sequences in on-line protein databases suggests that a number of candidate proteins in particular known viral proteins share homologous sequences with these peptides (Table 3). Peptide motifs have been identified that home to tumors in mice (Pasqualini and Ruoslahti, 1996; Pasqulaini et al, 1997 and 2000). Subsequently, specific receptors were identified in which the tumor-specific motifs bind in the vascular beds (Pasqualini et al, 2000; Burg et al, 1999; Esler and Ruoslahti, 2002; Joyce et al, 2003). Based on these results from mouse tumor models, it is contemplated that some of the selected motifs that bind to leukemia cells may mimic proteins that interact with differentially expressed cell surface markers on leukemia cells. For example, a peptide may mimic a ligand of a receptor via a motif sufficient for receptor recognition. Applying the same approach, peptide motifs corresponding to Molt-4 phage clone inserts were matched for similarity to known human proteins by searching in available online NCBI databases (ncbi.nlm.nih.gov/BLAST/). A minimum of tripeptide homology was used in the matching searches. Examples of candidate proteins (ligands) potentially mimicked by the leukemia- binding peptides and their corresponding putative receptors are listed in Table 3. Some of the proteins mimicked by the peptide motifs either play a role in cell differentiation and adhesion, or have growth factor, or signaling properties with known bio- functional relevance to cancer and leukemia. For instance, the AYHRLRR motif insert in Molt-4 clone 1 mimics galectin -9, a lectin that binds to beta-galactoside and a potent eosinophil chemo-attractant derived from antigen-stimulated T-cells that also plays a role in myeloid differentiation and in cell adhesion. Recently, a study of a panel of 61 human tumor cell lines indicated that Gal-9 over-expression was restricted to hematological, colorectal, and ovarian malignancies (Lahm et al, 2001).

Several interesting proteins with sequence homology to the GFYWLRS motif in clone 2 were identified. Of these, the T-cell lymphoma invasion and metastasis 1 protein (TIAMl), a guanine nucleotide exchange factor that plays a role in Racl activation and src- induced transformation, can directly bind to c-myc and can interfere with apoptosis of cells including leukemia cells (Van Leeuwen et al, 2002). TIAMl can interact with the G-protein coupled lysophosphatidic acid receptor (LPAl receptor). Another candidate ligand that is mimicked by the same motif is the latent transforming growth factor-beta-binding protein (LTBP-2), a structural extracellular matrix protein for targeting TGF-beta action. It was shown that LTBP-2 mediates melanoma cell adhesion via binding to betal - integrin (Vehvilainen et al, 2003). The protease specific for small ubiquitin related modifier -1 (SUMO-I) is an interesting ligand that also shares homology with the clone 2 insert motif. SUMO-I can covalently modify certain proteins including the tumor-suppressive promyelocytic leukemia protein (PML). PML is delocalized from nuclear bodies in acute promyelocytic leukemia (APL) and is degraded in cells infected by several viruses. SUMO-I can also modify E-26 transforming specific ETS-related gene product (ETV6), a transcription repressor that is rearranged in leukemias and congenital fibrosarcoma. It is known that some of the protein modifications mediated by SUMO-I could lead to abnormal cellular localization resulting in neoplastic transformation (Chakraborty et al, 2001).

Focus was placed on candidate ligands that interact with cell surface receptors namely the GP130-like monocyte receptor (GLM-R) and Plexin Bl . Not only GLM-R is related to and can associate with GPl 30 Receptor (GP 130R) and the leukemia inhibitory factor receptor (LIFR), but it is also a soluble type I cytokine receptor similar to receptors for IL-6, IL-I l, IL- 12, and GMCSF 31. On the other hand, Plexin Bl belongs to a novel family of trans-membrane receptors for semaphorins that play a role in axonal guidance, in immune regulation, and in tumori genesis. The closely related members, Plexin- Al and Plexin- A2, form complexes with neuropilin- 1 (NRP-I) as well as with neuropilin-2 (NRP-2) and act as receptors for semaphorins 3A and 3F respectively. Originally identified in the developing nervous system as modulators of axonal guidance, recent reports showed that neuropilins play a major role in angiogenesis and that their overexpression correlates with rumor metastasis and aggressiveness. In the cardiovascular system, both neuropilins were found to function as co-receptors for VEGFi65 and that as homo- or heterodimers, they can modulate its binding to VEGFR2 (KDR) and VEGFRl (Fltl). Additionally, they can recognize other VEGF family members such as the heparin binding from of placental growth factor (both), VEFG-B (only binds NRP-I), and VEGF-C (only binds NRP-2) (Zachary and Gliki, 2001; Neufeld et al, 2002a and 2002b; Nakarmura and Goshima, 2002; Miao et al, 2000; Kawakami et al, 2002; Bachelder et al, 2003). Moreover, the role of leukocyte -associated VEGF in the progression of acute myeloid leukemia has been documented in that higher VEGF levels in vivo correlated with lower survival as well as shorter-disease free survival in patients (Aguayo et al, 1999; Schuch et al, 2002).

Since the GFYWLRS insert in Molt-4 clone 2 mimics Plexin-Bl, membrane bound Sema4D or the closely related Sema3A could be putative receptors recognized by this motif. Additionally, due to the interplay between all these proteins in multi -receptor complexes, it is likely that the peptide insert in clone 2 may bind to either one of the neuropilin receptors. Alignment and a reverse BLAST search were run to check if the peptide motif shares any sequence homology with any known neuropilin-interacting proteins. No 100% linear homology between peptide corresponding to Molt-4 phage clone 2 and VEGF 165 were found. However, two weak matches along the protein sequence were found. Additionally, there were several non-linear weak matches to neuropilin-2 and plexin-Al (Table 4). These findings taken together indicate that neuropilin- 1 may likely be the receptor for the GFYWLRS ligand. Table 4. Non-linear matches were found between GFYWLRS peptide and proteins that interact with neuropilin-1 via receptor dimerization. Protein Length Matched Part Peptide Match Region Matched Residues E-value Protein Description 925 664 667 GFYWLRS 4 7 WL+S 0 51 neuropιlιn-2 [Rattus πorvegicus] 925 686 691 GFYWLRS 2 7 F L+S 1 5 neuropιlιn-2 [Rattus norvegicus] 925 356 360 GFYWLRS 3 7 Y+++S 3 3 neuropιlιn-2 [Rattus norvegicus] 925 656 662 GFYWLRS 1 7 G+ + R+ 16 neuropιlιn-2 [Rattus norvegicus] 1754 981 986 GFYWLRS 2 7 F W S 1 9 NOV/PLEXIN-A1 protein [Homo sapien 1754 511 517 GFYWLRS 1 7 G+ L S 1 9 NOV/PLEXIN-A1 protein [Homo sapien 1754 1296 1300 GFYWLRS 1 5 G +L 21 NOV/PLEXIN-A1 protein [Homo sapien 1754 1137 1142 GFYWLRS 1 6 G L+ 21 NOV/PLEXIN-A1 protein [Homo sapien 1754 1278 1283 GFYWLRS 2 7 F L++ 28 NOV/PLEXIN-A1 protein [Homo sapien 921 564 568 GFYWLRS 3 7 Y R+ 62 NEUROPILIN [Rattus norvegicus]

Receptor identification and validation. Biochemical methods were used to perform binding assays on a number of candidate receptors which were commercially available in recombinant form. Interestingly, relative to control insertless phage, clone 2 with the insert C-GFYWLRS-C showed a 52.27 ± 6.80 (average ± SEM) - fold higher binding to recombinant neuropilin-1 as compared to 1.90 ± 0.67 - fold binding to GP130/Fc, an unrelated recombinant receptor that also was suggested by the BLAST search (P=0.011). The binding was specific since Molt-4 phage clone 1 (C-AYHRLRR-C) and clone 3 (C- SFFYLRS-C) showed no significant binding to either recombinant receptor (FIG. 16). These results indicate that NRP-I is likely the receptor recognized by clone 2 on the surface of leukemia cells.

Additionally, the binding of clone 2 to NRP-I related proteins such as neuropillin-2 (NRP-2) and members of the semaphorin family were assessed. Relative to Fd- tet, clone 2 showed a 6.05 ± 3.69 (average ± SEM) - fold and 8.94 ± 2.70 - fold stronger binding to recombinant NRP-2 and to recombinant semaphorin 3A (Sema3A) respectively. The binding was specific since clone 2 showed no significant binding to SemaόA and to other unrelated recombinant proteins. Taking into consideration that NRP-I and NRP-2 can form heterodimers and Sema3A is a natural ligand to NRP-I, the results indicate that clone 2 likely mimics a dimerization domain in NRP-I itself.

Cell surface expression oj neuropilin-1 in leukemia. The majority of the cell population in all 9 leukemia cell lines studied had positive cell surface expression of neuropilin-1 as determined from fluorescence activated cell sorting (FACS) analysis of PFA- fixed, non-permeabilized cells, immunostained with a rabbit anti -human neuropilin-1 antibody (Santa Cruz). There was slight variability though in the level of expression between the various cell lines. Representative FACS profiles of 3 leukemia cell lines namely, Molt-4 (T-ALL), OCI- AML3 (AML), and K562 (CML) are shown in FIG. 18.

Cytotoxicity of the C-GFYWLRS-C peptide motif coupled to a pro-apoptotic moiety in leukemia cells. Once the motif insert in Molt-4 phage clone 2 has been demonstrated to be capable of receptor mediated internalization and that neuropilin-1, the receptor recognized by this peptide, is expressed on leukemia cell surface, the peptide was assessed for the ability to successfully deliver a pro-apoptotic moiety into leukemia cells. Upon internalization into target cells, the D(KLAKLAK)2 moiety causes apoptosis by physically interfering with and disrupting the mitochondrial membranes resulting in cell death. Using the WST-I cell proliferation and viability assay (Roche), the effect of the pure cyclic form of clone2 peptide on proliferation and viability of leukemia cells in vitro was assessed. A C-GFYWLRS-C-GG-KLAKLAKKLAKLAK-NH2 conjugated form of the peptide was also used to assess its anti-leukemia effect via delivery of the pro-apoptotic moiety. The uncoupled peptide added in doses ranging from 20 μM up to 100 μM had no significant effect on viability of leukemia cells cultured at 8 x 104 cells/100 μl/ well in 96- well plates for 24hr. The peptide slightly enhanced cell proliferation in some of the cell lines tested. By contrast, the KLAKLAK - conjugated peptide resulted in a significant decrease in cell viability in all 9 leukemia cell lines. Molt-4 cells were the most sensitive (FIG. 19) for which the maximal effect was attainable at 20 μM resulting in 19.19 ± 2.34 (average ± SEM) - percentage cell viability as compared to untreated cells (P=O.0004). The cytotoxic effect of the NRP-I targeting peptide coupled to KLAKLAK is specific since treatment of the cells with the D(KLAKLAK)2 peptide alone at similar doses had no significant effect on cell viability. FIG. 19 represents the results from cytotoxicity assays performed on Molt-4, OCI- AML3, and K562 cell lines.

The large-scale screening approach allowed identification of leukemia binding ligands and extraction of useful biological information. The C-GFYWLRS-C, a leukemia- binding motif that utilizes neuropilin-1 as receptor, was shown to likely mimic a dimerization domain. Moreover, a pro-apoptotic peptide was targeted and internalized through this functional ligand-receptor pair. Treatment of leukemia cells in vitro with a pro-apoptotic peptide guided by the C-GFYWLRS-C motif resulted in dose-dependent apoptosis in all 9 leukemia cell lines assessed. Together, these data illustrate the ability of phage-based screening systems for identification of relevant targets in the context of leukemia. Inhibition assays will be performed to confirm the specificity of peptide binding. Additionally, alternative biochemical and genetic based approaches will be used to confirm that neuropilin-1 is the receptor for the GFYWLRS motif and to better map the interaction domain. For that purpose, GST-fusion constructs of Molt-4 clone 2 will be produced. Phage clones isolated from bio-panning on Molt-4 cells will be further evaluated for binding to human tissues in an overlay assay. Paraffin-embedded tissue sections from spleen and lymph node of human leukemia will be overlaid with leukemia-binding CGFYWLRSC-displaying phage. Phage will be detected by using an anti-M13 phage antibody. Neuropilin-1 expression in similar tissue sections will be evaluated by conventional immunostaining with an anti-NRP-1 antibody. Neuropilin-1 will be studied in regard to its elevated in leukemia patients through pathological and functional analyses to confirm that it is a potential target for intervention in leukemia. Moreover, work is underway to assess the cytotoxicity of C- GFYWLRS-C-GG-D(KLAKLAK)2 on imatinib (Gleevec) - resistant, Philadelphia chromosome positive, chronic myelogenous leukemia (CML) cell lines, namely KBM5-R and KBM7-R 41. Furthermore, the efficacy of C-GFYWLRS-C-GG-D(KLAKLAK)2 peptide in mouse models of leukemia including a gleevec - resistant CML model will assessed.

EXAMPLE 3:

EVALUATION OF LEAD PAIRS IN LEUKEMIA MOUSE MODELS FOR THE DEVELOPMENT OF TARGETED THERAPY IN LEUKEMIA.

The choice of the leukemia model will be based on several criteria: (i) the efficiency of engraftment of the leukemia cell line into Severe Combined Immune Deficient (SCID) mice, (ii) the data generated from in vitro cytotoxicity assays considering the most sensitive cell line, and (iii) the aggressiveness of the leukemia model. Several leukemia cell lines may initially be evaluated in pilot in vivo studies to determine optimal experimental conditions such as required number of cells to be injected for successful disease dissemination as well as optimal dose and number of injections of the control and test peptides. Briefly, cultured leukemia cells will be maintained in RPMI-1640 medium (GIBCO) supplemented with 25 mmol/L HEPES buffer, 10% fetal bovine serum (FBS), 100 U/mL penicillin, 100 μg/mL streptomycin, 2 mmol/L 1-glutamine. Cells will be suspended in PBS. Prior to IV injection of the leukemia cells, mice will be first irradiated using 200-250 Rad (cesium source) in order to enhance engraftment. A receptor targeting peptide conjugated to an apoptotic moiety will be evaluated for its anti-leukemia efficacy in combination with chemotherapy (AraC, Gleevec). Peptide-based therapy and/or chemotherapy will be administered IP into tumor-bearing mice. The number of treatment days per cycle as well as the number of successive cycles if necessary is as yet to be determined. Experiments to monitor engraftment and disease progression will include measuring human CD45+ cells in blood samples collected from tail veins of control and test mice using flow cytometry. RBCs will be cleared by lysis. Additionally, the expression of human HLAs in various organs (bone marrow, spleen, liver, lungs, kidneys, and brain) will be evaluated by performing PCR on DNA extracted from these samples and by doing immunohistochemistry on human CD45+ expression in fixed tissue sections. During and following completion of therapeutic treatment, the animals will be monitored daily for survival and for any potential non-specific toxicity.

All of the compositions, methods and apparatus disclosed and claimed herein can be made and executed without undue experimentation in light of the present disclosure. While the compositions and methods of this invention have been described in terms of preferred embodiments, it are apparent to those of skill in the art that variations maybe applied to the compositions, methods and apparatus and in the steps or in the sequence of steps of the methods described herein without departing from the concept, spirit and scope of the invention. More specifically, it are apparent that certain agents that are both chemically and physiologically related may be substituted for the agents described herein while the same or similar results would be achieved. All such similar substitutes and modifications apparent to those skilled in the art are deemed to be within the spirit, scope and concept of the invention as defined by the appended claims. REFERENCES

The following references, to the extent that they provide exemplary procedural or other details supplementary to those set forth herein, are specifically incorporated herein by reference. U.S. Patent 3,817,837 U.S. Patent 3,850,752 U.S. Patent 3,939,350 U.S. Patent 3,996,345 U.S. Patent 4,275,149 U.S. Patent 4,277,437 U.S. Patent 4,366,241 U.S. Patent 4,472,509 U.S. Patent 5,021,236 U.S. Patent 5,206,347 U.S. Patent 5,223,409 U.S. Patent 5,401,511 U.S. Patent 5,622,699 U.S. Patent 5,889,155 U.S. Patent 6,068,829 U.S. Patent 5,492,807 U.S. Patent 5,603,872 U.S. Patent 5,670,312 U.S. Patent 5,705,610 U.S. Patent 5,840,841 U.S. Patent Appln. 20040048243

Aguayo et al, Blood, 94:3717-21, 1999. Arap et al., Curr. Opin. Oncol, 560-565, 1998b. Arap et al, Nature Med. 8, 121-127, 2002. Arap et al, Science, 279:377-380, 1998a. Bachelder et al, Cancer Research, 63:5230-5233, 2003. Baichwal and Sugden, In: Gene Transfer, Kucherlapati (Ed.), NY, Plenum Press, 117-148, 1986. Bakhshi et al., 1985 Barany and Meπϊfield, In: The Peptides, Gross and Meienhofer (Eds.), Academic Press, NY, 1-284, 1979. Brooks et al, Cell, 79: 1157-1 164, 1994b Brooks et al, Science, 264:569-571, 1994a. Burg et al, Cancer Res., 59:2869-2874, 1999. Chakraborty, et al, Cell. Biochem., 82:310-325, 2001. Chen and Okayama, MoI. Cell Biol. , 7(8):2745-2752, 1987. Cleary and Sklar, Proc. Natl. Acad. ScL USA, 82(21):7439-7443, 1985. C\eaτy et al, J. Exp. Med., 164(l):315-320, 1986. Coffin, In: Virology, Fields et al. (Eds.), Raven Press, NY, 1437-1500, 1990. Coupar et al, Gene, 68:1-10, 1988. Eliceiri and Cheresh, Curr. Opin. Cell Biol, 13(5):563-568, 2001.. Ellerby et al, Nature Med., 9:1032-1038, 1999. Essler and Ruoslahti, Proc. Natl Acad. ScL USA, 99:2252-2257, 2002. Folkman, Nature Biotechnol. 15:510, 1997. Folkman, Nature Med., 1 :27-31, 1995 Friedmann, Sconce, 244:1275-1281, 1989. Gomez Foix et al, J. Biol. Chem., 267:25129 25134, 1992. Gopal, MoI Cell Biol, 5:1188-1190, 1985. Graham and Prevec, In: Methods in Molecular Biology: Gene Transfer and Expression Protocol, Murray (Ed.), Humana Press, Clifton, NJ, 7:109-128, 1991. Graham and van der Eb, Virology, 52:456-467, 1973. Graham et al, J. Gen. Virol, 36:59-72, 1977. Grunhaus and Horwitz, Seminar in Virology, 3:237-252, 1992. Harlow and Lane, Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory Press, NY), 1988. Hermonat and Muzycska, Proc. Natl Acad. ScL USA, 81 :6466-6470, 1984. Herz and Gerard, Proc. Natl Acad. ScL USA, 90:2812 2816, 1993. Horwich et al, J. Virol, 64:642-650, 1990. Johnson et al., In: Peptide Turn Mimetics, Pezzuto et al., Eds., Chapman and Hall, NY, 1993. Jones and Shenk, Cell, 13:181 188, 1978. Joyce et al, Cancer Cell, 4, 393-403, 2003. Kawakami et al., Cancer, 95:2196-201, 2002. Koivunen et al., J Cell Biol, 153(5):905-16, 2001. Koivunen et al, Methods MoI Biol, 129:3-17, 1999b. Koivunen et al, Nature Biotechnol, \1:16^,-11Λ, 1999a. Lahm et al, J. Cancer Res. Clin. Oncol, 127:375-386, 2001. Le Gal La Salle et al, Science, 259:988 990, 1993. Levrero et al, Gene, 101 :195 202, 1991. Mann et al, Cell, 33:153-159, 1983. Merrifield, Science, 232: 341-347, 1986 Miao et al.,FASEB J., 14:2532-2539, 2000. Nakamura and Goshima, Advances Exp. Med. and Biology, 515:55-69, 2002. Neufeld et al, Advances Exp. Med. and Biology, 515, 81-90, 2002. Neufeld et al, Trends in Cardiovascular Medicine, 12:13-19, 2002. Nicolas and Rubinstein, In: Vectors: A survey of molecular cloning vectors and their uses, Rodriguez and Denhardt, eds., Stoneham: Butterworth, pp. 494-513, 1988. Nicolau et al, Methods Enzymol, 149:157-176, 1987. Paskind et al, Virology, 67:242-248, 1975. Pasqualini and Ruoslahti, Nature, 380:364-366, 1996. Pasqualini et al, Cancer Res., 60: 722-727, 2000. Pasqualini et al, Nature Biotechnol, 15:542-546, 1997 Pasqualini, J. Nucl. Med., 43:159-162, 1999. PCT Application PCTYUSOl/28124 PCT Appln. PCT/USOl/28044 Physician's Desk Reference, 56th Ed. Medical Economics Co., Inc., NJ, 1862-1866, 2002. Potter et al, Proc. Nat. Acad. ScL USA, 81 :7161-7165, 1984. Racher et al, Biotechnology Techniques, 9:169 174, 1995. Ragot et al, Nature, 361 :647 650, 1993. Rajotte and Ruoslahti, J. Biol Chem., 274:11593-1 1598, 1999 Rajotte et al, J. Clin. Invest., 102:430-437, 1998 Remington's Pharmaceutical Sciences, 15th ed., 33:624-652, Mack Publishing Company, Easton, PA, 1980.. Remington's Pharmaceutical Sciences, 15th ed., pp. 1035-1038 and 1570-1580. Rich et al, Hum. Gene Ther., 4:461 476, 1993. Ridgeway, In: Vectors: A Survey of Molecular Cloning Vectors and Their Uses, Rodriguez et al. (Eds.), Stoneham: Butterworth, pp. 467-492, 1988. Rippe et al, MoI. Cell Biol, 10:689-695, 1990. Rosenfeld et al., Cell, 68:143-155, 1992. Rosenfeld et al., Science, 252:431 434, 1991. Schuch et al., Blood., 100:4622-4628, 2002. Smith and Scott, Meth. Enzymol, 21 :228-257, 1993. Smith and Scott, Science, 228:1315-1317, 1985. Stewart and Young, In: Solid Phase Peptide Synthesis, 2d. ed., Pierce Chemical Co., 1984. Stratford-Perricaudet and Perricaudet, In: Human Gene Transfer, Cohen-Haguenauer et al. (Eds.), John Libbey Eurotext, France, pp. 51-61, 1991. Stratford-Perricaudet et al, Hum. Gene. Then, 1 :241-256, 1990. Tarn et al, J. Am. Chem. Soc, 105:6442, 1983. Temin, In: Gene Transfer, Kucherlapati R, ed., New York, Plenum Press, 149-188, 1986. Tsujimoto and Croce, Proc. Natl Acad. ScL USA, 83(14):5214-5218, 1986. Tsujimoto et al, Nature, 315:340-343, 1985.

Tur-Kaspa et al, MoI. Cell Biol, 6:716-718, 1986. Van Leeuwen et al, J. Biol Chem., 278:400-406, 2002. Vehvilainen et al, J. Biol. Chem., 278:24705-24713, 2003. Ψong et al, Gene, 10:87-94, 1980. Wu and Wu, Biochemistry, 27: 887-892, 1988. Wu and Wu, J. Biol Chem., 262: 4429-4432, 1987. Zachary and Gliki, Cardiovascular Research, 49:568-581, 2001. Zurita, Cancer Res., 64, 435-439, 2004