WO2003033675A2 - Identification of binding partners for specific proteins - Google Patents

Abstract

This invention relates to newly characterized protein DGI-3 and peptides ('DGI-3-binders') and polypeptides ('DGI-3-partners') that bind to this protein. Also related are polypeptide complexes comprising DGI-3 and DGI-3-binders or DGI-3 and DGI-3-partners, as well as antibodies that bind to these polypeptide complexes. Additionally, this invention relates to nucleic acids encoding the DGI-3, DGI-3-binder, or DGI-3-partner peptides or polypeptides. The invention also relates to diagnostics and therapeutics employing the disclosed DGI-3, DGI-3-binder, or DGI-3-partner polypeptides, peptides, or polynucletides. In particular, this invention relates to diagnostic and therapeutic applications directed to cancers and similar disorders.

Description

IDENTIFICATION OF BINDING PARTNERS FOR SPECIFIC PROTEINS RELATED APPLICATION

This application is a continuation-in-part of U.S. Application Serial No. 60/342,892, filed October 19, 2001 , which is hereby incorporated herein by reference in its entirety. FIELD OF THE INVENTION

This invention relates to isolated peptides ("DGI-3-binders") and polypeptides ("DGI-3-partners") that bind to a previously uncharacterized gene product DGI-3. The invention also relates to polypeptide complexes comprising DGI-3 and DGI-3-binders or DGI-3 and DGI-3-partners, as well as antibodies that bind to these polypeptide complexes. The invention further relates to nucleic acids encoding the DGI-3, DGI-3-binder, or DGI-3-partner peptides or polypeptides. Also encompassed are diagnostic reagents and methods, and therapeutic compositions and methods employing the DGI-3, DGI-3-binder, or DGI-3-partner polypeptides, peptides, and polynucleotides of the invention. Specifically related are diagnostic and therapeutic applications directed to diseases associated with defects in cell growth, proliferation, and/or attachment (e.g., neoplastic growth and cancers). BACKGROUND OF THE INVENTION

The human genome is estimated to contain over 30,000 genes. Despite rapid progress in identifying genes, progress in identifying the activity and function of gene products lags far behind. A number of methods have been reported for determining the functions and/or medical conditions associated with specific genes. One general method called genomic 'knockout' abolishes the function of a given gene by deleting the gene or by making mutations in the gene, e.g., small deletions or insertions. Genomic knockouts, performed in cell lines and animals, have revealed the functions of several genes. However, genomic knockouts have limited applications. Genomic knockouts can only be used to identify conditions that occur from a null mutation at a very early stage in development. Yet, many diseases occur from graded alterations in gene activity. A disease state may occur from changes in gene expression, for example, through increased or decreased expression levels, timing, and/or localization. Furthermore, genomic knockouts provide little information on finding an optimal drug target. Similarly, genomic knockouts do not provide an investigator with a simple tool for obtaining small organic molecules that are useful for animal phenotyping and for use as drug leads.

A second approach to elucidating gene function is the use of antisense nucleic acids to prevent gene expression by inhibiting transcription or translation. In this approach, antisense molecules are administered via gene therapy or conventional therapy (e.g., intravenous administration). Advantageously, these methods allow investigators to alter timing and levels of gene expression. However, many antisense methods do not allow precise control of expression levels. In addition, antisense molecules may be difficult to localize to the tissue of interest.

Recent information has made it clear that there are large networks of genes encoding interrelated products. One gene can influence the expression level of one or several other genes. Gene networks have been elucidated by DNA chip technology, which allows for the quantitation of mRNA from a large number of genes (see, e.g., U.S. Patent No. 5,800,992 and WO 95/35505, incorporated herein by reference). Although DNA chip technology provides useful maps of gene networks, such technology cannot easily be used to distinguish initial interactions from secondary interactions. Further, DNA chip technology is not useful for characterizing genes of unknown function.

Many gene products function by binding to one or more other peptides or proteins. Presently, there are few approaches for identifying a protein's partner, i.e., the protein with which the target gene product directly interacts. This information is critical as protein:ligand interactions are involved in important cell processes, such as signaling (e.g., information transfer in signaling cascades) and molecular processing. The identification of protein partners can be used to determine, for example, the ligand for a receptor, the substrate for an enzyme, and the regulatory protein for an enzyme complex.

In the classical approach to partner identification, the protein target and its partner are isolated in a complex. In a modern approach, the target and partner are constructed as two fusion proteins that generate a signal upon interaction. As examples of the latter, yeast two-hybrid systems have been developed. Yet, two-hybrid systems have several disadvantages, including high levels of false positives, incompatibility with certain targets (e.g., RNAs and membrane bound proteins cannot be used), and problems with postranslational modifications. Moreover, approaches based on two- hybrid systems are not easily applied to a large number of genes of unknown function.

Partner information is critical to developing a target-binding assay for the identification of drug leads. A large number of binding assays exists, including in vitro and in vivo formats. Most in vitro formats require input of both target and ligand, and only a few formats require only target input. Further, formats that require only a protein target generate a high frequency of false positives, i.e., compounds that bind but do not cause a change in target activity. Such formats would require extensive, laborious screening protocols to identify ligands for previously uncharacterized protein targets. Previous studies screened for partners by panning uncharacterized proteins against phage displayed peptide libraries. However, these experiments failed to identify naturally occurring ligands. For example, panning of the EPO (erythropoietin) and TPO (thrombopoietin) receptors identified peptides that were active but contained no natural TPO or EPO sequence motifs. Other studies have shown similar results (see, e.g., U.S. Patent No. 5,877,007, U.S. Patent No. 5,770,377, and U.S. Patent No. 6,010,861 , incorporated herein by reference). In contrast, this invention employs novel methods for identifying naturally-occurring (i.e., endogenous) binding partners of uncharacterized gene products, and establishing the functions of these gene products. These methods were first described in co-owned patent applications, U.S. Patent Application Serial No. 09/852,455 and International Patent Application No. PCT/US01/15092, which are incorporated herein by reference in their entirety. SUMMARY OF THE INVENTION

In accordance with the methods of this invention, uncharacterized gene product DGI-3 (KIAA0186; GenBank Ace. No. NM_021067; Nagase et al., 1996, DNA Res. 3:17-24) was used to identify DGI-3-binding peptides. These peptides ("DGI-3-binders") were used to identify endogenous DGI-3- binding partners including cytohesins 1-4 (GenPept Ace. Nos. Q15438, Q99418, 043739, Q9UIA0); Ras-GAP (Ras GTPase activating protein; Q15283), GNRP (guanine-nucleotide releasing protein; GenPept Ace. No. Q13972), Trio (GenPept Ace. No. 075962), and Rhophilin (GenPept Ace. No. Q61085).

Cytohesins facilitate the exchange of GDP for GTP on ARFs (ADP- ribosylation factors). Ras-GAP (Ras GTPase activating protein) enhances the GTPase activity of the Ras oncogene. GNRP (guanine nucleotide release protein) promotes the exchange of GDP for GTP on Ras. Trio facilitates the exchange of GDP by GTP on Ras-like proteins Rho/Rac. Rhophilin specifically binds to GTP-Rho. Notably, these DGI-3-binding partners ("DGI-3-partners") have been implicated in various cellular processes, including cell growth, proliferation, and attachment. Accordingly, DGI-3, DGI-3-binders, DGI-3-partners, and modulators of DGI-3/DGI-3- binder or DGI-3/DGI-3-partner interactions find use as diagnostics and therapeutics for diseases relating to defects in these cellular processes (e.g., neoplastic growth and cancers). The invention encompasses isolated DGI-3, DGI-3-binder, and DGI-

3-partner polypeptides, peptide fragments, or protein fusions thereof. Specifically encompassed are polypeptide-complexes comprising DGI-3 and DGI-3-binders or DGI-3 and DGI-3-partners, as well as fragments of these complexes.

The invention further encompasses isolated nucleic acids comprising nucleotide sequences of DGI-3, DGI-3-binders, or DGI-3-partners. In one aspect, the invention provides nucleic acids encoding polypeptides that bind to DGI-3. This invention also encompasses to probes, primers, antisense sequences, and vectors comprising DGI-3, DGI-3-binder, or DGI-3-partner nucleotide sequences, as well as host cells comprising these vectors. Also encompassed are antibodies and antibody fragments that bind to DGI-3, DGI-3-binder, DGI-3-partner polypeptides, variants, or fragments thereof. In particular, the invention encompasses polyclonal and monoclonal antibodies and antibody fragments that bind to polypeptide- complexes comprising DGI-3 and DGI-3-binders or DGI-3 and DGI-3- partners. In accordance with this invention, antibodies are preferably complex-specific, i.e., the antibodies specifically bind to complexes of DGI- 3, DGI-3-binders, or DGI-3-partners. In various aspects, the antibodies of the invention can be used as protein purification reagents.

Further encompassed are ligands (e.g., agonists, antagonists, inhibitors, or other modulators) that bind to DGI-3, DGI-3-binders, or DGI-3- partner polypeptides, variants, or fragments thereof. In particular, this invention encompasses ligands that alter (e.g., inhibit or enhance) the formation or stability of polypeptide-complexes comprising DGI-3 and DGI- 3-binders or DGI-3 and DGI-3-partners. The invention also encompasses diagnostic reagents comprising the disclosed DGI-3, DGI-3-binder, and DGI-3-partner polynucleotides, polypeptides, peptides, antibodies, or ligands. Also included are compositions (e.g., pharmaceutical compositions) comprising one or more DGI-3, DGI-3-binder, or DGI-3-partner polynucleotides, polypeptides, peptides, antibodies, or ligands. This invention further encompasses methods of obtaining the disclosed DGI-3 and DGI-3-partner nucleic acids, polypeptides, peptides, antibodies, and ligands. In addition, this invention encompasses diagnostic methods for identifying or monitoring diseases relating to DGI-3 or DGI-3- partners (e.g., neoplasms, including cancers). Also encompassed are methods of treatment using the pharmaceutical compositions of the invention. Specifically included are cancer treatments utilizing the disclosed DGI-3, DGI-3-binder, DGI-3-partner polynucleotides, polypeptides, peptides, antibodies, fragments, and/or modulators. Additional aspects and embodiments encompassed by this invention will be apparent from the detailed description and exemplification herein below. BRIEF DESCRIPTION OF THE DRAWINGS

The appended drawings of the figures are presented to further describe the invention and to assist in its understanding through clarification of its various aspects. In the figures of this invention, the nucleotide and amino acid sequences are represented by their one-letter abbreviations.

Figures 1A-1C show the nucleotide and amino acid sequence information for DGI-3. Figure 1A: The nucleotide sequence of DGI-3 obtained from cDNAs (GenBank Ace. No. NMJ321067; Nagase et al., 1996, DNA Res. 3:17-24; SEQ ID NO:18). Figure 1B: The encoded amino acid sequence for DGI-3 (GenPept Ace. No. NP_066545; SEQ ID NO:19). Figure 1C: The nucleotide sequence of the DGI-3 probe used in Northern blot analysis (SEQ ID NO:184). Figures 2A-2D show the amino acid sequences for the cytohesins.

Figure 2A: the amino acid sequence for cytohesin-1 (GenPept Ace. Nos. Q15438; SEQ ID NO:20). Figure 2B: The amino acid sequence for cytohesin-2 (GenPept Ace. No. Q99418; SEQ ID NO:21 ). Figure 2C: The amino acid sequence for cytohesin-3 (GenPept Ace. No. 043739; SEQ ID NO:22). Figure 2D: The amino acid sequence for cytohesin-4 (GenPept Ace. No. Q9UIA0; SEQ ID NO:23). Figures 3A-3B show the amino acid sequences for GNRP and Ras- GAP. Figure 3A: The amino acid sequence for GNRP (guanine-nucleotide releasing protein; GenPept Ace. No. Q13972; SEQ ID NO:24). Figure 3B: The amino acid sequence for Ras-GAP (Ras GTPase activating protein; GenPept Ace. No. Q15283; SEQ ID NO:25).

Figures 4A-4B show the amino acid sequences for Trio and Rhophilin. Figure 4A: The amino acid sequence for Trio (GenPept Ace. No. 075962; SEQ ID NO:26). Figure 4B: The amino acid sequence for Rhophilin (GenPept Ace. No. Q61085; SEQ ID NO:27). Figure 5 shows the possible mechanism of action for DGI-3 through the interaction with cytohesins 1-4 via their pleckstrin homology (PH) domains, and the interaction with Trio via its tyrosine kinase domain. DGI-3 may act to alter (e.g., enhance, inhibit or regulate) the actions of these proteins. Figure 6A shows the possible mechanism of action for DGI-3 through the interaction with Ras-GAP via its pleckstrin homology (PH) domain, and the interaction with GNRP via its guanine nucleotide exchange factor (GEF) domain. The Ras-GAP and GNRP proteins act as positive or negative regulators of Ras, and DGI-3 may act to alter (e.g., enhance, inhibit, or regulate) the interactions of these proteins. Figure 6B shows the possible combined mechanisms of action of DGI-3 in cell adhesion, cytoskeletal reorganization, and cell growth.

Figure 7 shows the position of the DGI-3 gene (KIAA0186) on human chromosome 20. Distances are expressed as kilobases (kb). Image was obtained from the euGenes website (University of Indiana, Bloomington IN).

Figure 8 shows results of Northern Blot analysis of DGI-3 expression in normal and breast cancer cell lines. HMEC: human mammary epithelial cells; HBL-100: transformed breast epithelial cells; T47-D, MCF-7, MDA- MB-157, MDA-MB-231 , MDA-MB-453: breast carcinoma cells. Figure 9 shows results of Northern Blot analysis of DGI-3 expression in normal human tissues.

Figure 10 shows results of non-radioactive Northern blot analysis of DGI-3 expression in various cancer cell lines. Lane 1 : MDA-MB-453 breast carcinoma cells; Lane 2: HT29 colorectal carcinoma; Lane 3: WiDr colorectal carcinoma Lane 4: T47D breast carcinoma cells; Lane 5: Caco-2 colorectal carcinoma; Lane 6: HBL-100 transformed breast epithelial cells; Lane 7: PANC-1 pancreatic carcinoma; Lane 8: DU-145 prostate carcinoma ; Lane 9: HL-60 myeloid leukemia cells; Lane 10: SW480 colorectal carcinoma cells; Lane 11 : MDA-MB-231 breast carcinoma cells; Lane 12: MCF-7 breast carcinoma cells; Lane 13: A549 lung carcinoma cells; Lane 14: MiaPaCa pancreatic carcinoma cells.

Figures 11A-11C show DGI-3 sequence information obtained from AceView from NCBI (National Center for Biotechnology Information, Bethesda, MD). Figure 11 A: Nucleotide sequence of the 3' untranslated region (3' UTR; SEQ ID NO: 132) and the 5' untranslated region (5' UTR; SEQ ID NO:166) of DGI-3; Figures 11B-11C: Coding sequence of DGI-3 (SEQ ID NO:133). Figure 11 D: Exons 1-7 (SEQ ID NO 167-SEQ ID NO: 173) as determined by Ensembl Human GeneView (Wellcome Trust Sanger Institute, Hinxton, Cambridge, United Kingdom). DGI-3 is Ensembl Gene ID ENSG00000101003 located between 25376420-25416477 bp on human chromosome 20.

Figures 12A-12B show an amino acid sequence alignment of DGI-3 and DGI-3 orthologs generated by the PRETTY program (Wisconsin Package, Accelrys). Plurality: 2.00 Threshold: 3 AveWeight 1.00 AveMatch 2.78 AvMisMatch -2.25. YVS6_CAEEL: Caenorhabditis elegans ortholog (SEQ ID NO:134; GenBank Ace. No. Q22019); Y181 : human DGI- 3 (SEQ ID NO: 135); Q9CZ15: Mus musculus ortholog (SEQ ID NO: 136) Q9W0I7: Drosophilia melangogaster ortholog (SEQ ID NO:137); Q9SSC0 Arabidopis thaliana ortholog (SEQ ID NO:138); YD13_YEAST Sacharomyces cerevisiae ortholog (SEQ ID NO:139; GenBank Ace. No.

Q12488); Q9P7X6: Sacharomyces pombe ortholog (SEQ ID NO:140).

Figures 13A-13C show the nucleotide sequence of the DGI-3 promoter region. Figures 13A-13B: The 4.3 kb fragment of the DGI-3 promoter region obtained from BAC clone RP11-384-D7 (SEQ ID NO:141 ).

The promoter region starts at 30 bp upstream from the translation start site of DGI-3. Figure 13B: The adjacent sequence and translation start site of

DGI-3 (SEQ ID N0.142).

Figures 14A-14B illustrate vectors used for analysis of the DGI-3 promoter region. Figure 14A: Vector map of pGL3-Enhancer Vector

(Promega) used to test the DGI-3 promoter. Figure 14B: Vector map of the pGL3-Enhancer Vector-DGI-3-promoter construct.

Figures 15A-15C show sequence analysis of peptides identified by panning against DGI-3. Figure 15B: Results from BLASTP analysis of DGI-3 binders using the PAM30 matrix and the swissprot sequence database. The results were used to assign endogenous protein partners for

DGI-3. Figure 15B: Sequence alignments of the A1 , G3, and C5 peptides and their endogenous counterparts. Figure 15C: The A1 , G3, and C5 peptides correspond to endogenous proteins known to interact with the pleckstrin homology domain (PHD), guanine nucleotide exchange factor

(GEF) domain, or protein kinase domain (PKD), respectively.

Figures 16A-16B show peptides identified from an A1 secondary peptide library panned against DGI-3. The amino acids are represented by their one-letter abbreviation. The ratios over background are determined by dividing the signal at 405 nm (E-Tag, DGI-3, or LDH) by the signal at 405 nm for non-fat milk. Sp/lrr = the ratio of specific binding over non-specific binding; LDH = lactate dehydrogenase (negative control).

Figure 17 shows preferred peptides constructed based on sequence data provided by the A1 secondary peptide library. The amino acids are represented by their one-letter abbreviation. The ratios over background are determined by dividing the signal at 405 nm (E-Tag, DGI-3, or LDH) by the signal at 405 nm for non-fat milk. Sp/lrr = the ratio of specific binding over non-specific binding; LDH = lactate dehydrogenase (negative control).

Figures 18A-18C show morphological changes in NIH3T3 cells following transfection with DGI-3. Figure 18A: NIH3T3 normal mouse fibroblasts; Figure 18B: NIH3T3 cells plus cytohesin; Figure 18C: NIH3T3 cells plus DGI-3.

Figures 19A-19H show DGI-3 and control antisense, peptide, and polypeptide sequences, and their effect on cell growth. Figure 19A: DGI-3 antisense sequence (SEQ ID NO:143); Figure 19B: Cytohesin-1 PHD antisense sequence (SEQ ID NO:144); Figure 19C: Cytohesin-1 nucleotide and polypeptide sequence (GenBank Ace. No. NM 004762; SEQ ID NO:207-208); Figure 19D: Nucleotide and amino acid sequences of peptides tested: VEGF ligand (SEQ ID NO:145-146); A1 (SEQ ID NO:147- 148); and G3 (SEQ ID NO:149-150); Figure 19E: MCF-7 cells. Lane 1 : cells alone; Lane 2: cells plus TransFast™; Lane 3: cells plus vector; Lane 4: cells plus DGI-3 antisense; Lane 5: cells plus A1 peptide; Lane 6: cells plus G3 peptide; Lane 7: cells plus cytohesin; Lane 8: cells plus VEGF ligand. Cell Number (x 10⁵); Statistically significant using Student's T test and ANOVA (Sigma Stat). Figure 19F: HBL-100 cells; Lane 1 : cells alone; Lane 2: cells plus TransFast™; Lane 3: cells plus vector; Lane 4: cells plus DGI-3 antisense; Lane 5: cells plus A1 peptide; Lane 6: cells plus G3 peptide; Lane 7: cells plus cytohesin; Lane 8: cells plus VEGF ligand. Cell Number (x 10⁵); No statistical significance observed between the groups seen with ANOVA. Figure 19G: MCF-7 cells; Lane 1 : cells alone; Lane 2: cells plus empty vector; Lane 3: cells plus DGI-3 antisense; Lane 4: cells plus cytohesin-1 PHD antisense. Figure 19H: HBL-100 cells; Lane 1 : cells alone; Lane 2: cells plus empty vector; Lane 3: cells plus DGI-3 antisense; Lane 4: cells plus cytohesin-1 PHD antisense.

Figures 20A-20B show the effect of adeno-associated virus expression of DGI-3 antisense in human cell lines. Cells were incubated with Giemsa stain 96-hours post infection. AAV empty vector or AAV-DGI was added at a concentration of 1 :1 , 1 :10, or 1 :100 with complete medium. Figure 20A: Cells plus vector. Figure 20B: Cells plus AAV:DGI-3 antisense. Figure 21 shows DGI-3 specific peptides modified with penetrating peptide (penetratin) for use as inhibitors of cancer cell growth.

Figure 22 shows silencer RNA (siRNA) construction kit oligonucleotides for use as inhibitors of cancer cell growth.

Figures 23A-23B demonstrate that DGI-3 binds to the pleckstrin homology domain. Figure 23A: Results from a competitive ELISA used to evaluate DGI-3 binding. Figure 23B: Format of the competitive ELISA.

Figure 24 shows amino acid sequence analysis of the DGI-3 polypeptide. Data was generated using the website COILS at worldwide web.ch.embnet.org/software/COILS_form.html (Swiss Institute of Bioinformatics, Lausanne, Switzerland; A. Lupas et al., 1991 , Science 252: 1162-1164). Underlined and shaded residues (amino acids 25-47) are predicted to have a coiled-coil structure. Tyrosines (Y) and serines (S) predicted as targets for phosphorylation are shown with bold font and shading. One of the tyrosine phosphorylation targets is included within the predicted coil-coiled domain (tyr40).

Figure 25 shows Biacore analysis of binding of DGI-3 with peptide A1.

Figure 26 shows DGI-3 phosphorothioate antisense primers for use as inhibitors of cancer cell growth. Primers were designed to cluster near the translation start (ATG) codon.

Figure 27 shows results of RT-PCR analysis of DGI-3 expression in human cell lines. PCR was performed with DGI-3 outside primers (see below). Lane 1 : 1 kb Plus Marker (ResGen, Invitrogen Corp., Carlsbad, CA); Lane 2: T47D cells; Lane 3: NIH3T3 cells expressing cytohesin; Lane 4: NIH3T3 cells; Lane 5: MCF-7 cells; Lane 6: NIH3T3 cells expressing DGI-3; Lane 7: HBL-100 cells; Lane 8: 293 cells; Lane 9: human lung RNA (CLONTECH) 0.5 μg. The 1 kb Plus marker included the following molecular weight species (bottom to top): 100, 200, 300, 400, 500, 650, 850, 1000, 1650, 2000, up to 12000 in 1000 base increments.

Figure 28 shows results of RT-PCR analysis of DGI-3 expression in NIH3T3 cells and transfectants. Lane 1 : 1 kb Plus Marker (ResGen, Invitrogen); Lane 2: NIH3T3 cells; primers pcDNA 3.1 Forward and pcDNA 3.1 Reverse (non-specific bands); Lane 3: NIH3T3 cells; primers pcDNA 3.1 Forward and DGI-3 Internal Reverse (no band); Lane 4: NIH3T3 expressing DGI-3; primers pcDNA 3.1 Forward and pcDNA 3.1 Reverse (non-specific bands); Lane 5: NIH3T3 cells expressing DGI-3; primers pcDNA 3.1 Forward and DGI 3 Internal Reverse (expected band of 400 bp). The 1 kb Plus marker included the following molecular weight species (bottom to top): 100, 200, 300, 400, 500, 650, 850, 1000, 1650, 2000, up to 12000 in 1000 base increments. Figure 29 shows the crystal structure of the pleckstrin homology domain of murine Grp1. Image was obtained from the NCBI website, as deposited by M. Feruson et al., Structure Of The Pleckstrin Homology Domain From Grp1 In Complex With Inositol 1 ,3,4,5-Tetrakisphosphate; (MMDB (Molecular Modeling Database) Accession No. 14112; PDB (Protein Data Bank) Accession No. 1 FHX). The PHD residues highlighted in light gray represent the predicted binding site of DGI-3, as identified by the A1 peptide.

Figures 30A-30B show the sequence information for hRas and activated hRas. Figure 30A: The nucleotide and amino acid sequence information (SEQ ID NO:185-186) for Homo sapiens v-Ha-Ras Harvey rat sarcoma viral oncogene homolog (hRas; GenBank Accession No. NM_005343). The nucleotide and amino acid sequence information (SEQ ID NO: 187-188) for activated hRas. DETAILED DESCRIPTION OF THE INVENTION To aid in the understanding of the specification and claims, the following definitions are provided. DEFINITIONS

Use of the terms "SEQ ID NO:3-SEQ ID NO:11" etc., is intended, for convenience, to refer to each SEQ ID NO. individually, and is not intended to refer to the sequences collectively. The invention encompasses each sequence individually, as well as any combination thereof.

"Nucleic acid or "polynucleotide" as used herein refers to purine- and pyrimidine-containing polymers of any length, either polyribonucleotides or polydeoxyribonucleotide or mixed polyribo-polydeoxyribonucleotides. This includes single-and double-stranded molecules, i.e., DNA-DNA, DNA-RNA and RNA-RNA hybrids, as well as "protein nucleic acids" (PNA) formed by conjugating bases to an amino acid backbone. This also includes nucleic acids containing modified bases. Polynucleotides, e.g., oligonucleotides, include naturally-occurring species or synthetic species formed from naturally-occurring subunits or their close homologs. The term may also refer to moieties that function similarly to polynucleotides, but have non- naturally-occurring portions. Thus, polynucleotides may have altered sugar moieties or inter-sugar linkages. Exemplary among these are phosphorothioate and other sulfur containing species which are known in the art. A "coding sequence" or a "protein-coding sequence" is a polynucleotide sequence capable of being transcribed into mRNA and/or capable of being translated into a polypeptide. The boundaries of the coding sequence are typically determined by a translation start codon at the 5'-terminus and a translation stop codon at the 3'-terminus. A "complement" or "complementary sequence" of a nucleic acid sequence as used herein refers to the antisense sequence that participates in Watson-Crick base-pairing with the original sequence.

A "probe" or "primer" refers to a nucleic acid or oligonucleotide that forms a hybrid structure with a sequence in a target region due to complementarily of at least one sequence in the probe or primer with a sequence in the target region. The term "vector" as used herein refers to a nucleic acid molecule capable of replicating another nucleic acid to which it has been linked. A vector, for example, can be a plasmid.

"Host" includes prokaryotes and eukaryotes. The term includes an organism or cell that is the recipient of a vector.

"Isolated", as used herein, refers to nucleic or amino acid sequences that are at least 60% free, preferably 75% free, and most preferably 90% free from other components with which they are associated with in in vivo or in vitro systems. As used herein the terms "protein" and "polypeptide" are synonymous. The term "peptide" refers to fragments or portions of proteins, and relatively short amino acid sequences, e.g., less than 50 amino acids in length. Peptides which are fragments or portions of proteins or other peptides preferably have at least one functional activity (e.g., binding, signaling, anti-cancer, or antigenic activity) as the complete protein or peptide sequence.

The term "antigenic" refers to the ability of a molecule (e.g., a polypeptide or peptide) to bind to its specific antibody, or an antibody fragment, with sufficiently high affinity to form a detectable antigen-antibody complex.

The term "ligand" as used herein describes any molecule, protein, peptide, or compound with the capability of directly or indirectly altering the physiological function, stability, or levels of a polypeptide or polynucleotide.

In specific aspects, ligands may directly bind to a polypeptide or polynucleotide.

A "sample" as used herein refers to a biological sample, such as, for example, tissue or fluid isolated from an individual (including, without limitation, plasma, serum, cerebrospinal fluid, lymph, tears, saliva, milk, pus, and tissue exudates and secretions) or from in vitro cell culture constituents, as well as samples obtained from, for example, a laboratory procedure. General descriptions of the foregoing terms and others are known in the art. See, e.g., Roitt et al., 1989, Immunology, 2^nd Edition, CN. Mosby Company, New York; Male et al., 1991 , Advanced Immunology, 2^nd Edition, Grower Medical Publishing, New York. DGI-3-PARTNERS

As described in detail below, uncharacterized gene product DGI-3 (Nagase et al., 1996, DNA Res. 3:17-24) was used to identify DGI-3- binding peptides. These peptides ("DGI-3-binders", also called hot-spot ligands) were used to identify DGI-3-binding partners ("DGI-3-partners") including cytohesins 1-4, Ras-GAP, GNRP, Trio, and Rhophilin.

Cytohesin-1 : Plasma membrane phospholipids are active in cell signaling (Divecha and Irvine, 1995, Cell 80:269-278; Spiegel et al., 1996, Curr. Opin. Cell Biol. 8:159-167). In particular, phosphatidylinositol 4,5- bisphosphate (PI(4,5)P2) is a precursor for the production of second messengers in two major cell-signaling pathways. This signaling is initiated by activation of either tyrosine kinase- or G-protein-linked cell surface receptors (Divecha and Irvine, 1995, Cell 80:269-278). The first pathway involves hydrolysis of PI(4,5)P2 by agonist-stimulated phospholipase C (PLC). This produces second messengers, inositol 1 ,4,5-trisphosphate (IP3) and diacylglycerol. IP3 triggers Ca 2+ release from intracellular stores (Clapham, 1995, Cell 80:259-268; Thomas et al., 1996, FAESEB J. 10:1505-1517), while diacylglycerol enhances the activity of various isomers of protein kinase C (Mellor and Parker, 1998, Biochem. J. 332:281-292). The second pathway involves phosphatidylinositol 3-kinase (PI 3-kinase)- catalyzed production of phosphatidylinositol 3,4,5-trisphosphate (PIP3). This product is dephosphorylated to phosphatidylinositol 3,4- bisphosphate (PI(3,4)P2) by SHIP (Stephens et al., 1993, Biochem. Biophys. Acta 1179:27-75; Toker and Cantley, 1997, Nature 387:673-767; Vanhaesebroek et al., 1997, Trends Biochem. Sci. 22:267-272). PIP3 and IP4 (inositol-1 ,3,4,5-tetrakisphosphate) have been implicated in various cell processes. In particular, PIP3 is involved in the regulation of numerous intracellular mechanisms (Hawkins et al., 1997, Biochem. Soc. Trans. 25:1147-1151 ) including mitogenesis (Fantl et al., 1992, Cell 69:413-423), membrane trafficking (Joly et al., 1995, J. Biol. Chem. 270:13225-13230), chemotaxis (Wennstrόm et al., 1994, Curr. Biol. 4:385-393), glucose transport (Okada et al., 1994, J. Biol. Chem. 269:3568- 3573), actin reorganization (Wymann and Acaro, 1994, Biochem. J. 298:517-520) and neurite outgrowth (Kimura et al., 1994, J. Biol. Chem. 269:18961-18967; Jackson et al., 1996, J. Cell Sci. 109:289-300; Kobayashi et al., 1997, J. Biol. Chem. 272:16089-16092; Kita et al., 1998, J. Cell. Sci. 111 :907-915).

A number of proteins have been identified as potential PIP3 downstream targets, including protein kinase B (Alessi and Cohen, 1998, Curr. Opin. Genet. Dev. 8(1 ):55-62), 3-phosphoinositide-dependent kinases (Alessi et al., 1997, Curr. Biol. 7:261-269; Stephens et al., 1998, Science 279:710-714; Anderson et al., 1998, Curr. Biol. 8:684-691 ), Bruton's tyrosine kinase (Btk) (Salim et al., 1996, EMBO J. 15:6241-6250), synaptotagmins (Schiavo et al., 1996, Proc. Natl. Acad. Sci. USA 93:13327- 13332), protein kinase C (Mellor and Parker, 1998, 1998, Biochem. J. 332:281-292), centaurin-α (Hammonds-Odie et al., 1996, J. Biol. Chem. 271 :18859-18868) and ADP-ribosylation factor guanine-nucleotide exchange factors (Kolanus et al., 1996, Cell 86:233-242; Chardin et al., 1996, Nature 384:481-484; Klarlund et al., 1997, Science 275:1927-1930; Venkateswalu et al., 1998, Curr. Biol. 8:463-466; K. Venkateswarlu et al., 1999, J. Cell Sci. 112:1957-1965). ADP-ribosylation factors (ARFs) are Ras-related GTPases that control membrane trafficking. ARFs are active when bound to GTP, and inactive when bound to GDP (M.S. Boguski and F. McCormick, 1993, Nature 366:643-654). The switch from GDP- to the GTP-bound states is controlled by GNRPs, which contain a central Sec7 domain. ARF-GNRPs can be subdivided into two major classes: large (-100 kDa) and small (45 to 50 kDa; CL. Jackson and J.E. Casanova, 2000, Trends Cell Biol. 10:60- 67; J.G. Donaldson and CL. Jackson, 2000, Curr. Opin. Cell Biol. 12:475- 482). Large GEFs are inhibited specifically by the fungal metabolite brefeldin A (BFA). In contrast, small GEFs are insensitive to BFA (A. Peyroche et al., 1999, Mol. Cell. 3, 275-285; M. Sata et al., 1999, Proc. Natl. Acad. Sci. USA 96:2752-2757; P. Chardin and F. McCormick, 1999, Cell 97:153-155).

Cytohesin-1 acts as a small GNRP for ADP-ribosylation factors (E. Meacci et al., 1997, Proc. Natl. Acad. Sci. USA 94:1745-1748; G. Pacheco- Rodriguez et al., 1998, J. Biol. Chem. 273:26543-26548; T. Knorr et al., 2000, Eur. J. Biochem. 267:3784-3791 ). Cytohesin-1 is a 47 kDa protein comprising an N-terminal coiled-coil domain, a central Sec7 domain, and a C-terminal module consisting of a pleckstrin homology (PH) domain and a polybasic region (W. Kolanus and L. Zeitlmann, 1998, Curr. Top. Microbiol. Immunol. 231 :33-49). Notably, cytohesin-1 has been shown to bind in vivo to PIP3 via the pleckstrin homology domain (K. Vankateswarlu et al., 1999, J. Cell Sci. 112:1957-1965). In addition, the ARF1 -exchange activity of a cytohesin-1 homolog was found to increase in the presence of PIP3 (Klarlund et al., 1998, J. Biol. Chem. 273:1859-1862). These observations suggest that PIP3 recruits cytohesin-1 to the plasma membrane and stimulates its ARF-exchange activity.

Notably, cytohesin-1 also specifically interacts with the cytoplasmic domain of the leukocyte integrin αι.β2 (Kolanus et al., 1996, Cell 86:233- 242). Cytohesin-1 is thereby thought to regulate adhesion of the extracellular domain of the αι_β₂-integrin to intercellular adhesion molecule 1 (ICAM-1 ) (W. Kolanus and L. Zeitlmann, 1998, Curr. Top. Microbiol. Immunol. 231 :33-49). Investigators have demonstrated that overexpression of either full-length cytohesin-1 or the Sec7 domain, results in a constitutive increase in the adhesion of αι.β₂ to ICAM-1. Conversely, expression of the cytohesin-1 pleckstrin homology domain specifically inhibits the activation of α_Lβ₂ in a dominant negative fashion (Kolanus et al., 1996, Cell 86:233-242; Nagel et al., 1998, J. Biol. Chem. 278:3568-3573). Cytohesin-1 may thereby coordinate ARF-exchange activity with regulation of αι_β₂ integrin- mediated cell adhesion. Cytohesin-1 shows a limited expression pattern, but is highly enriched in hematopoietic cells (W. Kolanus et al., 1996, Cell 26:233-242). Cytohesin-2 (also called ARNO (ARF-nucleotide-binding site opener)), shares 84% amino acid sequence identity with cytohesin-1. Both cytohesin- 1 and cytohesin-2 share nearly identical pleckstrin homology domains and similar biological function as guanine nucleotide exchange factors. However, cytohesin-2 shows a more ubiquitous expression pattern (P. Chardin et al., 1996, Nature 384:481-484; S. Frank et al., 1998, J. Biol. Chem. 273:23-27).

One set of DGI-3-partners was identified as cytohesins 1-4 (described in detail, below). DGI-3:cytohesin binding was mediated via the cytohesin PHD. Without wishing to be bound by theory, the following mechanisms of action has been suggested for DGI-3. Since PHDs also bind phospholipid second messages, it is possible that these molecules are involved in the DGI-3/cytohesin interaction. In particular, phospholipid second messages may act as allosteric activators of DGI-3/cytohesin-1. Because cytohesin-1 mediates ARF-GEF activity and αι_β2 integrin-mediated cell adhesion, DGI-3 may be involved in both processes. In addition, the specific expression patterns for cytohesin-1 and cytohesin-1 suggests that DGI-3 may mediate different effects in different cell types via its interactions with the cytohesin PHDs.

Rhophilin/Trio: Rho/Rac proteins are Ras-related GTPases that regulate a diverse set of cellular processes, including cytoskeleton reorganization, cell growth, gene transcription, cell-cycle progression, membrane trafficking, and cell-cell adhesion. Rhophilin was identified as a protein that specifically bound to GTP-Rho. The Rho-binding domain of Rhophilin shares 40% identity with the regulatory domain of a protein kinase, PKN. PKN also binds to GTP-Rho and is activated by this binding both in vitro and in vivo. This suggests that serine/threonine protein kinases may act as Rho effectors and contain amino acid sequence motifs for binding to GTP-Rho shared by other Rho target proteins (Watanabe, 1996, Science 271:645-648).

The functions of Rho/Rac GTPases are positively regulated by guanine nucleotide releasing proteins (GNRPs), which promote the exchange of GDP for GTP. Trio is a complex, multidomain Rho/Rac GNRP containing three enzyme domains: a serine/threonine kinase (PSK) domain and two guanine nucleotide exchange (GEF) domains. In Trio, one GEF domain has rad GEF activity and the other GEF domain has rhoA GEF activity. Trio also contains four spectrin-like domains, two pleckstrin-like domains, and an Ig-like domain (A. Debant et al., 1996, Proc. Natl. Acad. Sci. USA 93:5466-5471 ; T.B. Brady, The Interactive Fly, 28^th Edition, posted Monday, July 16, 2001 , at hypertext transfer protocol://sdb. bio.purdue.edu/fly/aimain/1aahome.htm).

To assess the functional role of the Trio GEF domains, NIH3T3 cell lines were stably transfected with constructs expressing the individual GNRP domains (K. Seipel et al., 1999, J. Cell Sci. 112 (Pt 12):1825-34). Cells expressing the N-terminal GEF domain showed prominent membrane ruffling, whereas cells expressing the C-terminal GEF domain produced lamellae with miniruffles. Cells expressing the N-terminal GEF domain also showed more rapid cell spreading, haptotactic cell migration, and anchorage-independent growth than normal cells (Seipel et al., 1999). In addition, expression of full-length Trio in COS cells altered actin cytoskeleton organization, and the distribution of focal contact sites (Seipel et al., 1999, J. Cell Sci. 112(Pt 12):1825-34).

The Trio PSK domain shows all the features of a functional PSK (S.K. Hanks and A.M. Quinn, 1991 , Methods Enzymol. 200:38-62), suggesting that Trio has kinase activity. The Trio PSK domain has the highest degree of sequence similarity with calcium/calmodulin-dependent PSKs, although it is not yet known whether Trio PSK activity requires calmodulin. Trio purified by immunoprecipitation did not show kinase activity toward artificial substrates, suggesting that Trio has strict substrate specificity (A. Debant et al., 1996, Proc. Natl. Acad. Sci. USA 93:5466- 5471 ).

In Drosophila, Trio exhibited dosage-sensitive, reciprocal genetic interactions with Abl tyrosine kinase, suggesting a role in formation of embryonic axon pathways (E.C Liebl et al., 2000, Neuron 25:107-118). Trio protein was localized to axons in the central nervous system (CNS) of embryos and was found to be highly expressed in regions of the brain, including the mushroom body (MB; T. Awasaki et al., 2000, Neuron 26(1 ):119-31 ). In Drosophila embryos, Trio null mutations caused inhibition or misdirection of axonal growth and malformation of the MB (Awasaki et al., 2000, Neuron 26(1 ):119-31 ). The sum of these findings suggest that Trio is a multifunctional protein that is necessary for cell migration and growth (Seipel et al., 1999, J. Cell Sci. 112(Pt 12):1825-34) and the directional extension of neurites (Awasaki et al., 2000, Neuron 26(1 ):119-31 ; J. Bateman et al., 2000, Neuron 26:93-106; Liebl et al., 2000, Neuron 25:107- 118). Another DGI-3-partner was identified as Trio (described in the

Examples, below). DGI-3:Trio binding was mediated via the Trio PSK domain. The substrate for the Trio kinase domain is presently unknown. Without wishing to be bound by theory, a possible model for DGI-3 function that involves cytohesin-1 and Trio is shown in Figure 5. In accordance with this model, DGI-3 connects the cytohesin pathway (cellular attachment) with the Rho/Rac pathway (cytoskeleton reorganization). In addition, DGI-3 may also function by inhibiting cytohesin-mediated effects on the integrin receptor; e.g., by blocking attachment to the extracellular matrix, thereby increasing metastasis. Ras-GAP/GNRP: Mammalian Ras genes encode closely related, small proteins with molecular weights of 21 ,000 Daltons (p21). Ras proteins have been shown to bind GDP and GTP and possess an intrinsic GTPase activity (see, e.g., H. Bourne et al., 1990, Nature 348:125-132; H.R. Bourne et al., 1991 , Nature 349(6305): 117-27). In normal cells, Ras is predominantly in an inactive GDP-bound form. The release of GDP and binding of GTP is triggered by an extracellular stimulus. The bound GTP causes a conformational change that enables Ras to interact with effector molecules, and initiates signaling. Subsequently, the GTP-bound form is inactivated by hydrolysis of GTP to GDP (see, e.g., F. Watzinger, T. Lion, 1999, Ras Family Atlas: Genetics, Cytogenetics, Oncology, and Haematology, available online at world wide web.infobiogen.fr/services/chromcancer/Deep/ras.html).

The intrinsic GTPase activity of Ras is very low, i.e., about 10^"2 min^"1. To accelerate the low rate of hydrolysis, regulatory proteins such as GAPs (GTPase activating proteins) bind to the GTP-bound form of Ras and stimulate the GTPase activity (see, e.g., M.R. Ahmadian et al., 1996, J. Biol. Chem. 271 :16409). Other regulatory proteins, termed GNRPs or GEFs (guanine nucleotide releasing proteins; guanine nucleotide exchange factors) facilitate the exchange of GDP for GTP (T. Satoh et al., 1992, Cancer Biol. 3:169). Specifically, GNRPs catalyze the dissociation of GDP from Ras. Ras proteins have been shown to be active in cell proliferation, differentiation, transformation, and apoptosis by relaying mitogenic and growth signals (R. Khosravi et al., 1994, Cancer Metastasis Rev. 13:67). In particular, uncontrolled activation of Ras can lead to oncogenesis. Normally, GAPs keep Ras predominantly an inactive GDP-bound state. However, the abundance of GAPs appears to be limited, as overexpression of non-mutated Ras results in deregulated activation of Ras and oncogenic transformation. Interestingly, the interaction of Ras with Ras-GAP has been reported to be regulated through the pleckstrin homology (PH) domain (K. Jonelle et. al., 2000, J. Biol. Chem. 275:35021-35027) identified by DGI-3- binder A1 (see below). Also, Ras-GAP may also be involved in cell motility via an interaction with p190 Rho-GAP (V. Sarang et al., 2000, J. Cell Biol. 149: 457-470).

Notably, several members in the Ras protein family have been linked to oncogenesis, including H-, K-, and N-Ras (see, e.g., M. Malumbres and A. Pellicer, 1998, Front. Biosci. 3:d887-912). A high incidence of Ras gene mutations has been reported in certain cancers. For example, mutations have been identified in malignant tumors of the pancreas (80-90%; C. Almoguera et al., 1988, Cell 53:549; V.T. Smit et al., 1988, Nucleic Acids Res. 1988; 16:7773), colorectal carcinomas (30-60%; J. Breivik et al., 1994, Br. J. Cancer 69:367; D.A. Spandidos et al., 1996, Tumori. 81 :7), non- melanoma skin cancer (30-50%; M. Barbacid, 1990, Eur. J. Clin. Invest. 20:225; S. Rodenhuis, 1992, Semin. Cancer Biol. 3:241 ), hematopoietic neoplasia of myeloid origin (18-30%; J. Breivik et al., 1994, Br. J. Cancer 69:367; T. Nakagawa et al., 1992, Oncology 49: 114; A. Neubauer et al., 1994, Blood 83:1603; T. Satoh et al., 1992, Semin. Cancer Biol. 3:169), and in seminoma (25-40%; M.P. Mulder et al., 1989, Oncogene 4:1345; M. Ridanpaa et al., 1993, Environ. Health Perspect. 101(Suppl 3):185-187). In other cancers, mutant Ras genes are found at lower frequencies. For example, mutations are found in breast carcinoma (0-12%; CF. Rochlitz et al., 1989, Cancer Res. 49:357; D.A. Spandidos, 1987, Anticancer Res. 7:991 ), glioblastoma, and neuroblastoma (0-10%; K. Ballas et al., 1988, Eur. J. Pediatr. 147:313; O. Brustle et al., 1996, Cancer 69:2385; CM. Ireland, 1989, Cancer Res. 49:5530). Overall, activating Ras mutations can be found in human malignancies with a frequency of 15-20%. Another set of DGI-3-partners was identified as GNRP and Ras-GAP

(described in the Examples, below). DGI-3:GNRP binding was mediated by the GNRP GEF domain. GNRP facilitates the exchange of Ras-bound GDP for GTP via the GEF domain. DGI-3:Ras-GAP binding was mediated by the Ras-GAP pleckstrin homology domain. Ras-GAP activity appears to be regulated by its pleckstrin homology domain (K. Jonelle et. al., 2000, J. Biol. Chem. 275:35021-35027). In particular, the pleckstrin homology domain has been shown to bind intramolecularly to the Ras catalytic domain on Ras-GAP and specifically inhibit Ras-mediated signaling and transformation, but not normal cellular growth. A non-limiting, theoretical model for DGI-3 protein function in the Ras pathway is shown in Figure 6. In this model, the DGI-3 interaction with GNRP activates Ras, while DGI-3 blocks the inhibitory role of Ras-GAP, leading to uncontrolled cell growth and oncogenesis. NUCLEIC ACIDS This invention relates to isolated nucleic acids comprising nucleotide sequences of DGI-3 (e.g., SEQ ID NO:18, SEQ ID NO:132-SEQ ID NO:133, SEQ ID NO:141-SEQ ID NO:142, and SEQ ID NO:167-173), DGI-3-binders (e.g., SEQ ID NO:28-SEQ ID NO:36), or DGI-3-partners. Also related are isolated nucleic acids encoding the amino acid sequences of DGI-3 (e.g., SEQ ID NO:19), DGI-3-binders (e.g., SEQ ID NO:3-SEQ ID NO:11 , SEQ ID NO:58-SEQ ID NO:109, SEQ ID NO:110-SEQ ID NO:112, and SEQ ID NO:114-SEQ ID NO:117), and DGI-3-partners (e.g., SEQ ID NO:20-SEQ ID NO:27). Such nucleic acids may be single stranded or double stranded, and may include DNA or RNA molecules (e.g., DNA, DNA/DNA, RNA, or RNA/DNA) comprising at least 15, 20, 40, 50, 60, 100, 200, 500 or more contiguous nucleotides, or the complements thereof. Closely related variants are also included as part of this invention, as well as recombinant nucleic acids sharing at least 50%, 60%, 70%, 80%, 85%, 90%, 95%, 98%, or 99% sequence identity with the nucleic acids described above which would be identical to nucleic acids from DGI-3 or DGI-3-partner genes except for one or more substitutions, deletions, or additions. Preferably, nucleic acids comprising altered sequences maintain their desired function (e.g., hybridization, expression, etc.). Further, included are the regulatory regions of DGI-3 and DGI-3-partner genes required for accurate expression of the respective gene. In a preferred embodiment, this invention is directed to at least 15 contiguous nucleotides of the nucleic acid sequences of DGI- 3, DGI-3-binders, or DGI-3-partners.

This invention further relates to recombinant DGI-3, DGI-3-binder, or DGI-3-partner nucleic acids (DNA or RNA) that are characterized by their ability to hybridize to (a) a nucleic acid encoding a protein or polypeptide, such as a nucleic acid having any of the sequences of DGI-3 or a DGI-3- partner or (b) a fragment of the foregoing. For example, a fragment can comprise the minimum nucleotides of a DGI-3 or DGI-3-partner gene required to encode a functional protein, or to encode a functional equivalent thereof. A functional equivalent can include a polypeptide, which, when incorporated into a cell, has all or part of the activity of a DGI-3 or DGI-3- partner protein (e.g., binding, signaling, anti-cancer, or antigenic activity). A nucleic acid that hybridizes to a nucleic acid encoding a DGI-3, DGI-3- binder, or DGI-3-partner protein or peptide can be double- or single- stranded. Hybridization to DNA includes hybridization to the sense strand, or to the antisense strand.

This invention also relates to nucleic acids that encode a polypeptide having the amino acid sequence of a DGI-3, DGI-3-binder, or DGI-3-partner polypeptides, or a functional equivalent thereof. A functional equivalent of a DGI-3 protein and DGI-3-partner includes fragments or variants that perform at least one characteristic function of the protein (e.g., antigenic, binding, or signaling activity). Preferably, a functional equivalent will share at least 65% sequence identity with the DGI-3 protein or DGI-3-partner.

Sequence identity calculations can be performed using computer programs, hybridization methods, or calculations. Preferred computer program methods to determine identity and similarity between two sequences include, but are not limited to, the GCG program package, BLASTN, BLASTX, TBLASTX, and FASTA (J. Devereux et al., 1984, Nucleic Acids Research 12(1 ):387; S.F. Altschul et al., 1990, J. Molec. Biol. 215:403-410; W. Gish and D.J. States, 1994, Nature Genet. 3:266-272; W.R. Pearson and D.J. Lipman, 1988, Proc Natl. Acad. Sci. USA 85(8):2444-8). The BLAST programs are publicly available from NCBI (Bethesda, MD) and other sources. The well-known Smith Waterman algorithm may also be used to determine identity.

For example, nucleotide sequence identity can be determined by comparing a query sequences to sequences in publicly available sequence databases (NCBI) using the BLASTN2 algorithm (S.F. Altschul et al., 1997, Nucl. Acids Res., 25:3389-3402). The parameters for a typical search are: E = 0.05, v = 50, B = 50, wherein E is the expected probability score cutoff, V is the number of database entries returned in the reporting of the results, and B is the number of sequence alignments returned in the reporting of the results (S.F. Altschul et al., 1990, J. Mol. Biol., 215:403-410). In another approach, nucleotide sequence identity can be calculated using the following equation: % identity = ((number of identical nucleotides) / (alignment length in nucleotides)) x 100. For this calculation, alignment length includes internal gaps but not includes terminal gaps. Alternatively, nucleotide sequence identity can be determined experimentally using specific hybridization conditions.

In accordance with this invention, polynucleotide alterations are selected from the group consisting of at least one nucleotide deletion, substitution, including transition and transversion, insertion, or modification (e.g., via RNA or DNA analogs). Alterations may occur at the 5' or 3' terminal positions of the reference nucleotide sequence or anywhere between those terminal positions, interspersed either individually among the nucleotides in the reference sequence or in one or more contiguous groups within the reference sequence. Alterations of a polynucleotide sequence of a DGI-3, DGI-3-binder, and DGI-3-partner may create nonsense, missense, or frameshift mutations in these coding sequences, and thereby alter the polypeptides encoded by the polynucleotides following such alterations.

Such altered nucleic acids, including DNA or RNA, can be detected and isolated by hybridization under high stringency conditions or moderate stringency conditions, for example, which are chosen to prevent hybridization of nucleic acids having non-complementary sequences. "Stringency conditions" for hybridizations is a term of art which refers to the conditions of temperature and buffer concentration which permit hybridization of a particular nucleic acid to another nucleic acid in which the first nucleic acid may be perfectly complementary to the second, or the first and second may share some degree of complementarity which is less than perfect.

For example, certain high stringency conditions can be used which distinguish perfectly complementary nucleic acids from those of less complementarity. "High stringency conditions" and "moderate stringency conditions" for nucleic acid hybridizations are explained in F.M. Ausubel et al. (eds), 1995, Current Protocols in Molecular Biology, John Wiley and Sons, Inc., New York, NY, the teachings of which are hereby incorporated by reference. In particular, see pages 2.10.1-2.10.16 (especially pages 2.10.8-2.10.11 ) and pages 6.3.1-6.3.6. The exact conditions which determine the stringency of hybridization depend not only on ionic strength, temperature and the concentration of destabilizing agents such as formamide, but also on factors such as the length of the nucleic acid sequence, base composition, percent mismatch between hybridizing sequences and the frequency of occurrence of subsets of that sequence within other non-identical sequences. Thus, high or moderate stringency conditions can be determined empirically.

High stringency hybridization conditions are typically carried out at 65 to 68°C in 0.1 X SSC and 0.1% SDS. Highly stringent conditions allow hybridization of nucleic acid molecules having about 95 to 100% sequence identity. Moderate stringency hybridization conditions are typically carried out at 50 to 65°C in 1 X SSC and 0.1% SDS. Moderate stringency conditions allow hybridization of sequences having at least 80 to 95% nucleotide sequence identity. Low stringency hybridization conditions are typically carried out at 40 to 50°C in 6 X SSC and 0.1% SDS. Low stringency hybridization conditions allow detection of specific hybridization of nucleic acid molecules having at least 50 to 80% nucleotide sequence identity. Examples of high, medium, and low stringency conditions can be found in Sambrook et al., 1989. Exemplary conditions are also described in M.H. Krause and S.A. Aaronson, 1991 , Methods in Enzymology, 200:546- 556; Ausubel et al., 1995. It is to be understood that the low, moderate and high stringency hybridization/washing conditions may be varied using a variety of ingredients, buffers, and temperatures well known to and practiced by the skilled practitioner.

It is understood that, as a result of the degeneracy of the genetic code, many nucleic acid sequences are possible which encode a DGI-3 protein or DGI-3-partner. Some of these will have little homology to the nucleotide sequences of any known or naturally-occurring DGI-3 or DGI-3- partner genes, but can be used to produce the proteins and polypeptides of this invention by selection of combinations of nucleotide triplets based on codon choices. Such variants and their complements, while not hybridizable to a naturally-occurring DGI-3 or DGI-3-partner genes, are contemplated within this invention.

The nucleic acid sequences of this invention may be derived from a variety of sources including DNA, cDNA, synthetic DNA, synthetic RNA, or combinations thereof. Such sequences may comprise genomic DNA, which may or may not include naturally-occurring introns. Moreover, such genomic DNA may be obtained in association with promoter regions or poly(A)+ sequences. The sequences, genomic DNA, or cDNA may be obtained in any of several ways. Genomic DNA can be extracted and purified from suitable cells by means well known in the art. Alternatively, mRNA can be isolated from a cell and used to produce cDNA by reverse transcription or other means.

The nucleic acids described herein are used in the methods of this invention for production of proteins or polypeptides, through incorporation into cells, tissues, or organisms. In one embodiment, DNA containing all or part of the coding sequence for a DGI-3 protein or DGI-3-partner, or a fragment thereof, is incorporated into a vector for expression of the encoded polypeptide or peptide in suitable host cells. The encoded polypeptide or peptide is capable of normal activity, such as antigenic, binding, or signaling activity. The nucleic acids of this invention find use as primers and probes for chromosome and gene mapping, antisense sequences, tissue distribution studies, identification and cloning full length genes, identification and cloning of homologous or orthologous sequences (wild-type and mutants), and in diagnostic applications. The primers of this invention may comprise all or a portion of the nucleotide sequence of DGI-3 or DGI-3-partner genes, or a complementary sequence thereof.

Probes may also be used for the detection DGI-3, DGI-3-binder, or DGI-3-partner nucleic acids, and should preferably contain at least 50%, preferably at least 80%, identity to a DGI-3, DGI-3-binder, or DGI-3-partner polynucleotide, or a complementary sequence, or fragment thereof. The probes of this invention may be DNA or RNA, the probes may comprise all or a portion of the nucleotide sequence of DGI-3, DGI-3-binder, or DGI-3- partner nucleic acids, or a complementary sequence thereof, and may include promoter, enhancer elements, and introns of the naturally-occurring DGI-3 or DGI-3-partner gene. VECTORS AND HOST CELLS

The nucleic acids described herein are used in the methods of this invention for production of polypeptides or peptides, through incorporation into cells, tissues, or organisms. In one embodiment, DNA containing all or part of the coding sequence for DGI-3, DGI-3-partner, or fragment thereof is incorporated into a vector for expression of the encoded polypeptide in suitable host cells. The encoded polypeptides consisting of DGI-3, DGI-3- binders, DGI-3-partners, or their functional equivalents are capable of activity (i.e., binding, signaling, or antigenic activity). A large number of vectors, including bacterial, yeast, and mammalian vectors, have been described for replication and/or expression in various host cells or cell-free systems, and may be used for gene therapy as well as for simple cloning or protein expression.

In one aspect, an expression vector comprises a nucleic acid encoding a DGI-3, DGI-3-binder, or DGI-3-partner polypeptide or peptide, as described herein, operably linked to at least one regulatory sequence. Regulatory sequences are known in the art and are selected to direct expression of the desired protein in an appropriate host cell. Accordingly, the term regulatory sequence includes promoters, enhancers and other expression control elements (see DN. Goeddel, 1990, Methods Enzymol. 185:3-7). Enhancer and other expression control sequences are described in Enhancers and Eukaryotic Gene Expression, 1983, Cold Spring Harbor Press, Cold Spring Harbor, ΝY. It should be understood that the design of the expression vector may depend on such factors as the choice of the host cell to be transfected and/or the type of polypeptide to be expressed. Several regulatory elements (e.g., promoters) have been isolated and shown to be effective in the transcription and translation of heterologous proteins in the various hosts. Such regulatory regions, methods of isolation, manner of manipulation, etc. are known in the art. Νon-limiting examples of bacterial promoters include the β-lactamase (penicillinase) promoter; lactose promoter; tryptophan (trp) promoter; araBAD (arabinose) operon promoter; lambda-derived Pi promoter and Ν gene ribosome binding site; and the hybrid tac promoter derived from sequences of the trp and lac UV5 promoters. Νon-limiting examples of yeast promoters include the 3- phosphoglycerate kinase promoter, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) promoter, galactokinase (GAL1 ) promoter, galactoepimerase promoter, and alcohol dehydrogenase (ADH1 ) promoter. Suitable promoters for mammalian cells include, without limitation, viral promoters, such as those from Simian Virus 40 (SV40), Rous sarcoma virus (RSV), adenovirus (ADV), and bovine papilloma virus (BPV). Preferred replication and inheritance systems include M13, ColE1 , SV40, baculovirus, lambda, adenovirus, CEΝ ARS, 2μm ARS and the like. While expression vectors may replicate autonomously, they may also replicate by being inserted into the genome of the host cell, by methods well known in the art.

DNA sequences can be optimized, if desired, for more efficient expression in a given host organism or expression system. For example, codons can be altered to conform to the preferred codon usage in a given host cell or cell-free translation system using well-established techniques. Codon usage data can be obtained from publicly-available sources, for example, the Codon Usage Database at world wide web.kazusa.or.jp/codon/. In addition, computer programs that translate amino acid sequence information into nucleotide sequence information in accordance with codon preferences (i.e., backtranslation programs) are widely available. See, for example, Backtranslate program from Genetics Computer Group (GCG), Accelrys, Inc., Madison, WI; and Backtranslation Applet from Entelechon GmbH, Regensburg, Germany. Thus, using the polypeptide and peptide sequences disclosed herein, one of ordinary skill in the art can design nucleic acids to yield optimal expression levels in the translation system or host cell of choice.

To obtain expression in eukaryotic cells, terminator sequences, polyadenylation sequences, and enhancer sequences that modulate gene expression may be required. Sequences that cause amplification of the gene may also be desirable. These sequences are well known in the art. Furthermore, sequences that facilitate secretion of the recombinant product from cells, including, but not limited to, bacteria, yeast, and animal cells, such as secretory signal sequences and/or preprotein or proprotein sequences, may also be included. Such sequences are well described in the art.

Expression and cloning vectors will likely contain a selectable marker, a gene encoding a protein necessary for survival or growth of a host cell transformed with the vector. The presence of this gene ensures growth of only those host cells that express the inserts. Typical selection genes encode proteins that 1 ) confer resistance to antibiotics or other toxic substances, e.g., ampicillin, neomycin, methotrexate, etc.; 2) complement auxotrophic deficiencies, or 3) supply critical nutrients not available from complex media, e.g., the gene encoding D-alanine racemase for Bacilli. Markers may be an inducible or non-inducible gene and will generally allow for positive selection. Non-limiting examples of markers include the ampicillin resistance marker (i.e., beta-lactamase), tetracycline resistance marker, neomycin/kanamycin resistance marker (i.e., neomycin phosphotransferase), dihydrofolate reductase, glutamine synthetase, and the like. The choice of the proper selectable marker will depend on the host cell, and appropriate markers for different hosts as understood by those of skill in the art.

Suitable expression vectors for use with this invention include, but are not limited to, pUC, pBluescript (Stratagene), pET (Novagen, Inc., Madison, WI), and pREP (Invitrogen) plasmids. Vectors can contain one or more replication and inheritance systems for cloning or expression, one or more markers for selection in the host, e.g., antibiotic resistance, and one or more expression cassettes. The inserted coding sequences can be synthesized by standard methods, isolated from natural sources, or prepared as hybrids. Ligation of the coding sequences to transcriptional regulatory elements (e.g., promoters, enhancers, and/or insulators) and/or to other amino acid encoding sequences can be carried out using established methods.

Suitable cell-free expression systems for use with this invention include, without limitation, rabbit reticulocyte lysate, wheat germ extract, canine pancreatic microsomal membranes, E. coli S30 extract, and coupled transcription/translation systems (Promega Corp., Madison, WI). These systems allow the expression of recombinant polypeptides or peptides upon the addition of cloning vectors, DNA fragments, or RNA sequences containing protein-coding regions and appropriate promoter elements. Non-limiting examples of suitable host cells include bacteria, archea, insect, fungi (e.g., yeast), plant, and animal cells (e.g., mammalian, especially human). Of particular interest are Escherichia coli, Bacillus subtilis, Saccharomyces cerevisiae, SF9 cells, C129 cells, 293 cells, Neurospora, and immortalized mammalian myeloid and lymphoid cell lines. Techniques for the propagation of mammalian cells in culture are well- known (see, Jakoby and Pastan (eds), 1979, Cell Culture. Methods in Enzymology, volume 58, Academic Press, Inc., Harcourt Brace Jovanovich, NY). Examples of commonly used mammalian host cell lines are VERO and HeLa cells, CHO cells, and WI38, BHK, and COS cell lines, although it will be appreciated by the skilled practitioner that other cell lines may be used, e.g., to provide higher expression desirable glycosylation patterns, or other features.

Host cells can be transformed, transfected, or infected as appropriate by any suitable method including electroporation, calcium chloride-, lithium chloride-, lithium acetate/polyethylene glycol-, calcium phosphate-, DEAE- dextran-, liposome-mediated DNA uptake, spheroplasting, injection, microinjection, microprojectile bombardment, phage infection, viral infection, or other established methods. Alternatively, vectors containing the nucleic acids of interest can be transcribed in vitro, and the resulting RNA introduced into the host cell by well-known methods, e.g., by injection (see, Kubo et al., 1988, FEBS Letts. 241 :119). The cells into which have been introduced nucleic acids described above are meant to also include the progeny of such cells.

The nucleic acids of the invention may be isolated directly from cells. Alternatively, the polymerase chain reaction (PCR) method can be used to produce the nucleic acids of the invention, using either RNA (e.g., mRNA) or DNA (e.g., genomic DNA) as templates. Primers used for PCR can be synthesized using the sequence information provided herein and can further be designed to introduce appropriate new restriction sites, if desirable, to facilitate incorporation into a given vector for recombinant expression. Using the information provided in the nucleotide sequences of DGI-3,

DGI-3-binders, and DGI-3-partners, one skilled in the art will be able to clone and sequence all representative nucleic acids of interest, including nucleic acids encoding complete protein-coding sequences. It is to be understood that non-protein-coding sequences contained within DGI-3 and DGI-3-partner genes are also within the scope of the invention. Such sequences include, without limitation, sequences important for replication, recombination, transcription, and translation. Non-limiting examples include promoters and regulatory binding sites involved in regulation of gene expression, and 5'- and 3'-untranslated sequences (e.g., ribosome-binding sites) that form part of mRNA molecules. The nucleic acids of this invention can be produced in large quantities by replication in a suitable host cell. Natural or synthetic nucleic acid fragments, comprising at least 15 contiguous bases coding for a desired peptide or polypeptide can be incorporated into recombinant nucleic acid constructs, usually DNA constructs, capable of introduction into and replication in a prokaryotic or eukaryotic cell. Usually the nucleic acid constructs will be suitable for replication in a unicellular host, such as yeast or bacteria, but may also be intended for introduction to (with and without integration within the genome) cultured mammalian or plant or other eukaryotic cells, cell lines, tissues, or organisms. The purification of nucleic acids produced by the methods of this invention is described, for example, in Sambrook et al., 1989; F.M. Ausubel et al., 1992, Current Protocols in Molecular Biology, J. Wiley and Sons, New York, NY.

The nucleic acids of this invention can also be produced by chemical synthesis, e.g., by the phosphoramidite method described by Beaucage et al., 1981 , Tetra. Letts. 22:1859-1862, or the triester method according to Matteucci et al., 1981 , J. Am. Chem. Soc, 103:3185, and can performed on commercial, automated oligonucleotide synthesizers. A double-stranded fragment may be obtained from the single-stranded product of chemical synthesis either by synthesizing the complementary strand and annealing the strands together under appropriate conditions or by adding the complementary strand using DNA polymerase with an appropriate primer sequence.

Large quantities of the nucleic acids of this invention may be prepared by cloning the DGI-3, DGI-3-binder, or DGI-3-partner nucleic acids or portions thereof into expression vectors in compatible prokaryotic or eukaryotic host cells. The most commonly used prokaryotic hosts are strains of Escherichia coli, although other prokaryotes, such as Bacillus subtilis or Pseudomonas may also be used. Mammalian or other eukaryotic host cells, such as those of yeast, filamentous fungi, plant, insect, or amphibian or avian species, may also be useful for isolation of the nucleic acids of the invention.

Prokaryotic or eukaryotic cells comprising the nucleic acids of this invention will be useful not only for the production of the nucleic acids and proteins of this invention, but also, for example, in studying the characteristics of DGI-3 and DGI-3-partner proteins. Cells and animals that carry a DGI-3 or DGI-3-partner gene can be used as model systems to study and test for substances that have potential as therapeutic agents. The cells are typically cultured mesenchymal stem cells. These may be isolated from individuals with a somatic or germline DGI-3 or DGI-3-partner gene. Alternatively, the cell line can be engineered to carry a DGI-3 or DGI- 3-partner gene, in accordance with established methods. After a test substance is applied to the cells, the phenotype of the cell is determined. Any trait of the genetically modified cells can be assessed, including proliferation, attachment, and response to putative therapeutic agents. PEPTIDES AND POLYPEPTIDES

A further aspect of this invention pertains to isolated polypeptides and peptides such as DGI-3 (e.g., SEQ ID NO:19), DGI-3-binders (e.g., SEQ ID NO:3-SEQ ID NO:11 , SEQ ID NO:58-SEQ ID NO:109, SEQ ID NO:110-SEQ ID NO:112, and SEQ ID NO:114-SEQ ID NO:117), and DGI- 3-partners (e.g., SEQ ID NO:20-SEQ ID NO:27). The polypeptides and peptides of this invention can be isolated, synthetic, or recombinant. The DGI-3, DGI-3-binders, and DGI-3-partners may be obtained as individual polypeptides or part of a polypeptide-complex. In various aspects, a polypeptide-complex may comprise one or more molecules of DGI-3 in an association with one or more DGI-3-binders (e.g., multiple copies of the same peptide binder or single copies of different peptide binders). Similarly, a polypeptide-complex may comprise one or more molecules of DGI-3 in an association with one or more DGI-3-partners (e.g., multiple copies of the same binding partner or single copies of different binding partners).

This invention encompasses the DGI-3, DGI-3-binders, and DGI-3- partners, and fragments and functional equivalents thereof. The term "functional equivalent" is intended to include proteins which differ in amino acid sequence from the DGI-3, DGI-3-binder, or DGI-3-partner polypeptides or peptides but where such differences result in a modified protein which performs at least one characteristic function of the polypeptide (e.g., binding, signaling, anti-cancer, or antigenic activity). For example, a functional equivalent of a polypeptide may have a modification such as a substitution, addition, or deletion of one or more amino acid residues that are not directly involved in the function of this polypeptide.

It is also possible to vary the structure of a DGI-3, DGI-3-binder, or DGI-3-partner polypeptide or peptide for such purposes as increasing solubility, enhancing activity, antigenicity, or stability (e.g., shelf life ex vivo and resistance to proteolytic degradation in vivo). Such variants are considered functional equivalents of the polypeptides and peptides as defined herein. Preferably, polypeptides and peptides are modified so that they retain activity. Those residues shown to be essential for activity can be modified by replacing the essential amino acid with another, preferably similar amino acid residue (i.e., a conservative substitution) whose presence is shown to enhance, diminish, but not eliminate, or not effect receptor interaction. In addition, those amino acid residues that are not essential for binding or other activity can be modified by being replaced by another amino acid whose incorporation may enhance, diminish, or not effect reactivity.

Polypeptide and peptide variants include mutants differing by the addition, deletion, or substitution of one or more amino acid residues. Also included are modified polypeptides and peptides in which one or more residues are modified, and mutants comprising one or more modified residues. Useful modifications may include phosphorylation, sulfation, reduction/alkylation (Tarr, 1986, Methods of Protein Microcharacterization, J. E. Silver, Ed., Humana Press, Clifton, NJ, pp. 155-194); acylation (Tarr, supra); chemical coupling (Mishell and Shiigi (Eds), 1980, Selected Methods in Cellular Immunology, W H Freeman, San Francisco, CA; U.S. Patent No. 4,939,239); and mild formalin treatment (Marsh, 1971 , Int. Arch, of Allergy and Appl. Immunol. 41 :199-215). Additionally, D-amino acids, non-natural amino acids, or non-amino acid analogs can be substituted or added to produce a modified polypeptide. Furthermore, the polypeptides disclosed herein can be modified using polyethylene glycol (PEG) according to known methods (S.I. Wie et al., 1981 , Int. Arch. Allergy Appl. Immunol. 64(1 ):84- 99) to produce a protein conjugated with PEG. In addition, PEG can be added during chemical synthesis of the protein. Modifications or sequence variations may occur at the amino- or carboxy-terminal positions of the reference polypeptide sequence or anywhere between those terminal positions, interspersed either individually among the amino acids in the reference sequence or in one or more contiguous groups within the reference sequence.

Polypeptides or peptides may also be modified with a label capable of providing a detectable signal, either directly or indirectly, including, but not limited to, radioisotope, fluorescent, and enzyme labels. Fluorescent labels include, for example, Cy™3, Cy™5, Alexa, BODIPY, fluorescein (e.g., FluorX, DTAF, and FITC), rhodamine (e.g., TRITC), auramine, Texas Red, AMCA blue, and Lucifer Yellow. Preferred isotope labels include ³ H, ¹⁴ C, 32 P, ³⁵ S, ³⁶ Cl, ⁵¹ Cr, ⁵⁷ Co, ⁵⁸ Co, ⁵⁹ Fe, ⁹⁰ Y, ¹²⁵ I, ¹³¹ I, and ¹⁸⁶ Re. Preferred enzyme labels include peroxidase, β-glucuronidase, β-D- glucosidase, β-D-galactosidase, urease, glucose oxidase plus peroxidase, and alkaline phosphatase (see, e.g., U.S. Pat. Nos. 3,654,090; 3,850,752 and 4,016,043). Enzymes can be conjugated by reaction with bridging molecules such as carbodiimides, diisocyanates, glutaraldehyde, and the like. Enzyme labels can be detected visually, or measured by calorimetric, spectrophotometric, fluorospectrophotometric, amperometric, or gasometric techniques. Other labeling systems, such as avidin/biotin, Tyramide Signal Amplification (TSA™), are known in the art, and are commercially available (see, e.g., ABC kit, Vector Laboratories, Inc., Burlingame, CA; NEN® Life Science Products, Inc., Boston, MA).

In one aspect of this invention, DGI-3-binders (e.g., SEQ ID NO:3- SEQ ID NO:11 , SEQ ID NO:58-SEQ ID NO:109, SEQ ID NO:110-SEQ ID NO:112, and SEQ ID NO:114-SEQ ID NO:117) may be used as reagents to isolate, identify, quantify, or localize DGI-3 or DGI-3-related proteins (e.g., mouse or Drosophila DGI-3 orthologs, described below). For example, DGI- 3-binders can be used to form DGI-3:DGI-3-binder complexes, and thereby isolate or purify the DGI-3 protein. In particular, the DGI-3 protein can be precipitated by labeled DGI-3-binders, or the DGI-3 protein can be isolated by DGI-3 binders affixed to a solid surface (e.g., resin, beads, microtiter plates, etc.). As another example, labeled DGI-3-binders can be used in protein blotting experiments to determine the presence, absence, or expression levels of DGI-3 protein. In addition, labeled DGI-3 binders can be used in histological experiments, to localize DGI-3 protein in cells or tissues. Polypeptide and peptide variants may have "conservative" changes, wherein a substituted amino acid has similar structural or chemical properties, e.g., replacement of leucine with isoleucine. More infrequently, a variant may have "nonconservative" changes, e.g., replacement of a glycine with a tryptophan. Guidance in determining which amino acid residues can be substituted, inserted, or deleted without abolishing biological or immunological activity can be determined using computer programs well known in the art, for example, DNASTAR software (DNASTAR, Inc., Madison, WI). As non-limiting examples, conservative substitutions in a DGI-3 or DGI-3-partner amino acid sequence can be made in accordance with the following table:

TABLE 1

Substantial changes in function or immunogenicity can be made by selecting substitutions that are less conservative than those shown in the table, above. For example, non-conservative substitutions can be made which more significantly affect the structure of the polypeptide in the area of the alteration, for example, the alpha-helical, or beta-sheet structure; the charge or hydrophobicity of the molecule at the target site; or the bulk of the side chain. The substitutions which generally are expected to produce the greatest changes in the polypeptide's properties are those where 1 ) a hydrophilic residue, e.g., seryl or threonyl, is substituted for (or by) a hydrophobic residue, e.g., leucyl, isoleucyl, phenylalanyl, valyl, or alanyl; 2) a cysteine or proline is substituted for (or by) any other residue; 3) a residue having an electropositive side chain, e.g., lysyl, arginyl, or histidyl, is substituted for (or by) an electronegative residue, e.g., glutamyl or aspartyl; or 4) a residue having a bulky side chain, e.g., phenylalanine, is substituted for (or by) a residue that does not have a side chain, e.g., glycine.

Sequence tags (e.g., FLAG® tags) or amino acids, such as one or more lysines, can be added to the peptide sequences of the invention (e.g., at the N-terminal or C-terminal ends) as described in detail herein. Sequence tags can be used for peptide purification or localization. Lysines can be used to increase peptide solubility or to allow for biotinylation. Alternatively, amino acid residues located at the carboxy and amino terminal regions of the consensus motifs described below, which comprise sequence tags (e.g., FLAG® tags), or which contain amino acid residues that are not associated with a strong preference for a particular amino acid, may optionally be deleted providing for truncated sequences. Certain amino acids (e.g., C-terminal or N-terminal residues) such as lysine which promote the stability or biotinylation of the amino acids sequences may be deleted depending on the use of the sequence, as for example, expression of the sequence as part of a larger sequence which is soluble, or linked to a solid support.

Polypeptide or protein fragments (i.e., peptides) can range in size from 5 amino acid residues to all but one residue of the entire amino acid sequence. Thus, a peptide can be at least 5, 15, 20, 25, 30, 50, 100, 200, 236, 250, 300, 500, 800, or more contiguous amino acid residues of a DGI- 3 or DGI-3-partner protein or polypeptide. In one embodiment, the percent amino acid sequence identity between a DGI-3, DGI-3-binder, or DGI-3- partner polypeptide or peptide, and a functional equivalent thereof is at least 50%. In a preferred embodiment, the percent amino acid sequence identity is at least 65%. More preferably, the percent amino acid sequence identity is at least 75%, still more preferably, at least 80%, and even more preferably, at least 90%. Percent sequence identity can be calculated using computer programs or direct sequence comparison. Preferred computer program methods to determine identity between two sequences include, but are not limited to, the GCG program package, FASTA, BLASTP, and TBLASTN (see, e.g., D.W. Mount, 2001 , Bioinformatics: Sequence and Genome Analysis, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY). The BLASTP and TBLASTN programs are publicly available from NCBI and other sources. The well-known Smith Waterman algorithm may also be used to determine identity.

Exemplary parameters for amino acid sequence comparison include the following: 1 ) algorithm from Needleman and Wunsch, 1970, J Mol. Biol. 48:443-453; 2) BLOSSUM62 comparison matrix from Hentikoff and Hentikoff, 1992, Proc. Natl. Acad. Sci. USA 89:10915-10919; 3) gap penalty = 12; and 4) gap length penalty = 4. A program useful with these parameters is publicly available as the "gap" program (Genetics Computer Group, Madison, WI). The aforementioned parameters are the default parameters for polypeptide comparisons (with no penalty for end gaps). Alternatively, polypeptide sequence identity can be calculated using the following equation: % identity = ((the number of identical residues) / (alignment length in amino acid residues)) x 100. For this calculation, alignment length includes internal gaps but does not include terminal gaps. In accordance with this invention, polypeptide or peptide sequences may be identical to DGI-3, DGI-3-binder, or DGI-3-partner polypeptides or peptides, or may include up to a certain integer number of amino acid alterations. Polypeptide alterations are selected from the group consisting of at least one amino acid deletion, substitution, including conservative and non-conservative substitution, or insertion.

The invention also relates to isolated, synthesized and/or recombinant portions or fragments of a DGI-3, DGI-3-binder, or DGI-3- partner polypeptide as described herein. Polypeptide fragments (i.e., peptides) can be made which have full or partial function on their own, or which when mixed together (though fully, partially, or nonfunctional alone), spontaneously assemble with one or more other polypeptides to reconstitute a functional protein having at least one functional characteristic of a DGI-3 or DGI-3-partner protein of this invention. In addition, peptides may comprise, for example, one or more domains of a DGI-3-partner, disclosed herein (e.g., PH, GEF, or protein kinase domains). Nucleic acids comprising protein-coding sequences can be used to direct the expression polypeptides or peptides in intact cells or in cell-free translation systems. The coding sequence can be tailored, if desired, for more efficient expression in a given host organism, and can be used to synthesize oligonucleotides encoding the desired amino acid sequences. The resulting oligonucleotides can be inserted into an appropriate vector and expressed in a compatible host organism or translation system.

The polypeptides and peptides of this invention, including functional equivalents, may be isolated from wild-type or mutant cells (e.g., human cells or cell lines), from heterologous organisms or cells (e.g., bacteria, yeast, insect, plant, and mammalian cells), or from cell-free translation systems (e.g., wheat germ, microsomal membrane, or bacterial extracts) in which a protein-coding sequence has been introduced and expressed. Furthermore, the polypeptides and peptides may be part of recombinant fusion proteins. The polypeptides and peptides can also, advantageously, be made by synthetic chemistry. Polypeptides and peptides may be chemically synthesized by commercially available automated procedures, including, without limitation, exclusive solid phase synthesis, partial solid phase methods, fragment condensation or classical solution synthesis, described in detail below.

Methods for polypeptide and peptide purification are well-known in the art, including, without limitation, preparative disc-gel electrophoresis, isoelectric focusing, HPLC, reversed-phase HPLC, gel filtration, ion exchange and partition chromatography, and countercurrent distribution. For some purposes, it is preferable to produce the polypeptide or peptide in a recombinant system in which the protein contains an additional sequence (e.g., epitope or protein) tag that facilitates purification. Non-limiting examples of epitope tags include c-myc, haemagglutinin (HA), polyhistidine (6X-HIS) (SEQ ID NO:1 ), GLU-GLU, and DYKDDDDK (SEQ ID NO:2) (FLAG®) epitope tags. Non-limiting examples of protein tags include glutathione-S-transferase (GST), green fluorescent protein (GFP), and maltose binding protein (MBP).

In one approach, the coding sequence of a polypeptide or peptide can be cloned into a vector that creates a fusion with a sequence tag of interest. Suitable vectors include, without limitation, pRSET (Invitrogen Corp., San Diego, CA), pGEX (Amersham-Pharmacia Biotech, Inc., Piscataway, NJ), pEGFP (CLONTECH Laboratories, Inc., Palo Alto, CA), and pMAL™ (New England BioLabs (NEB), Inc., Beverly, MA) plasmids. Following expression, the epitope, or protein tagged polypeptide or peptide can be purified from a crude lysate of the translation system or host cell by chromatography on an appropriate solid-phase matrix. In some cases, it may be preferable to remove the epitope or protein tag (i.e., via protease cleavage) following purification. As an alternative approach, antibodies produced against a protein or peptide can be used as purification reagents. Other purification methods are also possible.

Both the naturally occurring and recombinant forms of the polypeptides of the invention can advantageously be used to screen compounds for binding activity. Many methods of screening for binding activity are known by those skilled in the art and may be used to practice the invention. Several methods of automated assays have been developed in recent years so as to permit screening of tens of thousands of compounds in a short period of time. Such high-throughput screening methods are particularly preferred. The use of high-throughput screening assays to test for inhibitors is greatly facilitated by the availability of large amounts of purified polypeptides and peptides, as provided by the invention. The polypeptides and peptides of the invention also find use as therapeutic agents as well as antigenic components to prepare antibodies.

Many conventional techniques in molecular biology, protein biochemistry, and immunology may be used to produce the amino acid sequences for use with this invention. To obtain recombinant polypeptides and peptides, coding sequences may be cloned into any suitable vectors for expression in intact host cells or in cell-free translation systems by methods well known in the art (see Sambrook et al., 1989) and described herein. The particular choice of the vector, host, or translation system is not critical to the practice of the invention. According to methods known in the art, polypeptides and peptides can be chemically synthesized by commercially available automated procedures, including, without limitation, exclusive solid phase synthesis, partial solid phase methods, fragment condensation, classical solution synthesis. In addition, recombinant and synthetic methods of peptide production can be combined to produce semi-synthetic polypeptides and peptides. The polypeptides and peptides of the invention are preferably prepared by solid phase peptide synthesis methods and techniques as described by Merrifield, 1963, J. Am. Chem. Soc. 85:2149; J.M. Stewart and J.D. Young, 1984, Solid Phase Peptide Synthesis, 2nd edition, Pierce Chemical Co., Rockford, IL; W.C Chan and P.D. White (Eds.), 2000, Fmoc Solid Phase Peptide Synthesis: A Practical Approach, Oxford University Press; Kaiser et al., 1970, Anal. Biochem. 34:595.

This invention provides specific amino acid sequences that bind DGI- 3. However, additional sequences may be obtained in accordance with the procedures described herein below. Advantageously, DGI-3-binders can be used to modulate the binding of DGI-3 to its partners. For example, DGI-3- binders can be used to alter the downstream actions of DGI-3, and thereby alter the cellular processes resulting from these actions (e.g., oncogenic transformation, metastasis, motility, and adhesion). In addition, DGI-3- binders can be used to identify the corresponding binding sites on the DGI-3 protein. This information, in turn, can be used to design peptides or polypeptides that mimic the binding action of the DGI-3 protein ("DGI-3- surrogates"). DGI-3-surrogates can be used to alter the downstream actions of DGI-3-partners, such as Ras-GAP, GNRP, Trio, Rhophilin, and the cytohesins, and thereby alter the cellular processes resulting from these actions (e.g., cell growth, attachment, motility, and proliferation). SEQUENCE ANALYSIS

Publicly available sequence databases, e.g., GenBank, GenPept, SWISS-PROT, Protein Data Bank (PDB), Protein Information Resource (PIR), Human UniGene (National Center for Biotechnology Information), can be used to determine sequences that share homology with DGI-3, DGI-3- binder, or DGI-3-partner nucleotide or amino acid sequences, or fragments thereof. Alternatively, privately owned sequence databases, e.g., the Incyte Genomics sequence database (Incyte Genomics), can be used. Databases with relatively few redundant sequences, e.g., PIR or SWISS-PROT databases, may be used to improve the statistical significance of a sequence match. However, databases which are more comprehensive and up-to-date, e.g., GenBank, GenPept, and Incyte Genomics sequence databases (Incyte Genomics, Inc., St. Louis, MO), are preferred.

Any method known in the art can be used to align and compare a DGI-3, DGI-3-binder, or DGI-3-partner sequence with the sequences present in a sequence database. Preferably, the BLAST program is used (S.F. Altschul et al., 1990, J. Mol. Biol. 215:403-410; S. Karlin et al., 1990, Proc. Natl. Acad. Sci. USA 87:2264-68; S. Karlin et al., 1993, Proc. Natl. Acad. Sci. USA 90:5873-7). BLAST identifies local alignments between the sequence of the previously identified protein and the protein sequences in the database, and predicts the probability of the local alignment occurring by chance. Although the original BLAST programs utilized ungapped local alignments, more recently developed BLAST programs such as WU- BLAST2/BLAST v2.0 (S. F. Altschul et al., 1996, Methods Enzymol. 266:460-480) have been modified to incorporate gapped local alignments similar to SSEARCH (T.F. Smith et al., 1981 , J. Mol. Biol. 147:195-197) and FASTA programs (W.R. Pearson, 1990, Methods Enzymol. 183:63-98). In addition, position-specific-iterated BLAST (PSI-BLAST) programs have been developed to identify weak but biologically relevant sequence similarities (S.F. Altschul et al., 1997, Nucleic Acids Res. 25:3389-3402). Furthermore, pattern-hit-initiated BLAST (PHI-BLAST) programs have been designed to identify specific patterns or sequence motifs shared by distantly-related proteins (Z. Zhang et al., 1998, Nucleic Acids Res. 26:3986-3990). Specialized BLAST programs are also available for performing searches of human, microbial, and malaria genome sequences, as well as searches for vector, immunoglobulin, and predicted human consensus sequences (NCBI).

Both FASTA and BLAST programs identify very short exact sequence matches between the query sequence and the databases sequences, analyze the best short sequence matches ("hits") to determine if longer stretches of sequence similarity are present, and then optimize the best hits by dynamic programming (S.F. Altschul et al., 1990, J. Mol. Biol. 215:403-410; W.R. Pearson, 1990, Methods Enzymol. 183:63-98). In contrast, the SSEARCH program compares the query sequence to all the sequences in the database via pair-wise sequence comparisons (T.F. Smith et al., 1981 , J. Mol. Biol. 147:195-197). Thus, the SSEARCH program is considered more sensitive than the BLAST and FASTA programs, but it is also significantly slower. The BLAST and FASTA programs utilize several approximations to increase their searching speed, and utilize statistical parameters (see below) to increase sensitivity and selectivity to approximate the performance of the SSEARCH program.

It is understood in the art that BLAST comparison of amino acid sequences requires a substitution matrix program (PAM30, PAM70, PAM120, PAM250, BLOSUM45, BLOSUM62, BLOSUM80, etc.; see, e.g., S.F. Altschul, 1991 , J. Mol. Biol., 219:555-565; S.F. Altschul, 1993, J. Mol. Evol. 36:290-300; M.O. Dayhoff et al., 1978, Atlas of Protein Sequence and Structure, 5:345-352; R.M. Schwartz and M.O. Dayhoff, 1978, Atlas of Protein Sequence and Structure, 5:353-358; G.H. Gonnet et al., 1992, Science, 256:1443-1445; S. Henikoff and J.G. Henikoff, 1992, Proc. Natl. Acad. Sci. USA, 89:10915-10919; S. Henikoff and J.G. Henikoff, 1993, Proteins, 17:49-61 ). A substitution matrix program scores each possible amino acid residue substitution, reflecting the probability that the residue is related to the corresponding residue in the query. Although different substitution matrices use different scoring systems for quantifying the relationship between the compared residues, the results from different substitution matrices can be readily compared and evaluated by a person of average skill in the art. In fact, it is common practice to run several parallel BLAST searches, using a different substitution matrix for each search (S.F. Altschul, 1991 , J. Mol. Biol., 219:555-565; S.F. Altschul, 1993, J. Mol. Evol., 36:290-300; S. Henikoff and J.G. Henikoff, 1993, Proteins, 17(1 ):49-61 ).

In some instances, BLAST analysis of amino acid sequences may require a filtering program (e.g., SEG or XNU). A filtering program removes repetitive regions from the query sequence (e.g., proline-rich regions). Although certain filtering programs determine regions of low compositional complexity (e.g., SEG), while other filtering programs determine regions with short, periodic, internal repeats (e.g., XNU), the results from different filtering programs are comparable and can be evaluated by a person of average skill in the relevant art. Moreover, it is common practice to combine, alternate, or omit the filtering programs as required for a specific query sequence (see, e.g., B. Birren et al. (Eds), 1997, Genome Analysis: A Laboratory Manual, Volume 1: Analyzing DNA, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY).

In addition, BLAST analysis may utilize different search parameters (e.g., scoring matrices and filters) depending on the specific features of the query sequence. For example, it is understood by the skilled practitioner that the BLOSUM matrices are not suitable for the shortest queries. Thus, the PAM matrices, such as PAM30 or PAM70, may be used for short query sequences (R.M. Schwartz and M.O. Dayhoff, supra; M.O. Dayhoff et al., supra). In general, as understood by the skilled person in the art, the following scoring matrices are recommended:

>85 BLOSUM62

(see the National Center for Biotechnology Information website at world wide web.ncbi.nlm.nih.gov/BLAST/matrix_info.html).

Additionally, PAM120 and BLOSUM62 matrices are designed to identify moderately diverged sequences, and these matrices may miss short/strong sequence similarities, or long/weak sequence similarities (S. Altschul et al., 1994, Nat. Genet. 6:119-129). For this reason, it has been generally recommended that researchers employ at least 3 separate substitution matrices for BLAST analysis (S. Altschul et al., 1993, J. Mol. Evol. 36:290-300; S. Altschul et al., 1994, Nat. Genet. 6:119-129). The use of different BLAST parameters with a particular query sequence is routinely practiced in the art and is well within the capability of an average artisan (see, e.g., Altschul et al., 1994, Nature Genet, 6; 119-129; Altschul et al., 1996, Methods Enzymol., 266:460-480; S. Henikoff and J.G. Henikoff, 1993, Proteins, 17(1 ):49-61 ).

BLAST analysis of nucleic acid sequences generally require a unitary matrix program, which scores each position as "+" if the nucleotides match, or "-" if the nucleotides do not match. In addition, BLAST comparisons of nucleic acid sequences may involve a nucleic acid-specific filtering program, such as DUST (J.M. Hancock and J.S. Armstrong, 1994, Comput. Appl. Biosci. 10:67-70). Notably, a particular sequence alignment program (e.g., BLAST, FASTA, or SSEARCH) and appropriate parameters (e.g., scoring matrices and filters) can be chosen based on the requirements of a sequence search, or individual preferences. In some cases, it may be preferable to use more than one program or set of parameters in order to confirm or evaluate search alignment results. PEPTIDE LIBRARIES

Peptide libraries produced and screened according to this invention are useful in providing new ligands for DGI-3 (e.g., SEQ ID NO:19) or DGI- 3-partners (e.g., SEQ ID NO:20-SEQ ID NO:27). Peptide libraries can be designed and panned according to methods described in detail herein, and methods generally available to those in the art (see, e.g., U.S. Patent No. 5,723,286 issued March 3, 1998 to Dower et al.). In one aspect, commercially available phage display libraries can be used (e.g., RAPIDLIB^® or GRABLIB^®, DGI BioTechnologies, Inc., Edison, NJ; Ph.D. C7C Disulfide Constrained Peptide Library, New England Biolabs). In another aspect, an oligonucleotide library can be prepared according to methods known in the art, and inserted into an appropriate vector for peptide expression. For example, vectors encoding a bacteriophage structural protein, preferably an accessible phage protein, such as a bacteriophage coat protein, can be used. Although one skilled in the art will appreciate that a variety of bacteriophage may be employed in this invention, in preferred embodiments the vector is, or is derived from, a filamentous bacteriophage, such as, for example, f1 , fd, Pfl , M13, etc. In particular, the fd-tet vector has been extensively described in the literature (see, e.g., Zacher et al., 1980, Gene 9:127-140; Smith et al., 1985, Science 228:1315-1317; Parmley and Smith, 1988, Gene 73:305-318).

The phage vector is chosen to contain or is constructed to contain a cloning site located in the 5' region of the gene encoding the bacteriophage structural protein, so that the peptide is accessible to receptors in an affinity enrichment procedure as described herein below. The structural phage protein is preferably a coat protein. An example of an appropriate coat protein is pill. A suitable vector may allow oriented cloning of the oligonucleotide sequences that encode the peptide so that the peptide is expressed at or within a distance of about 100 amino acid residues of the N- terminus of the mature coat protein. The coat protein is typically expressed as a preprotein, having a leader sequence.

Thus, desirably the oligonucleotide library is inserted so that the N- terminus of the processed bacteriophage outer protein is the first residue of the peptide, i.e., between the 3'-terminus of the sequence encoding the leader protein and the δ'-terminus of the sequence encoding the mature protein or a portion of the 5' terminus. The library is constructed by cloning an oligonucleotide which contains the variable region of library members (and any spacers, as discussed below) into the selected cloning site. Using known recombinant DNA techniques (see generally, Sambrook et al., 1989, Molecular Cloning, A Laboratory Manual, 2d ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y., 1989), an oligonucleotide may be constructed which, inter alia; 1 ) removes unwanted restriction sites and adds desired ones; 2) reconstructs the correct portions of any sequences which have been removed (such as a correct signal peptidase site, for example); 3) inserts the spacer residues, if any; and/or 4) corrects the translation frame (if necessary) to produce active, infective phage.

The central portion of the oligonucleotide will generally contain one or more DGI-3-binding sequences and, optionally, spacer sequences. The sequences are ultimately expressed as peptides (with or without spacers) fused to or in the N-terminus of the mature coat protein on the outer, accessible surface of the assembled bacteriophage particles. The size of the library will vary according to the number of variable codons, and hence the size of the peptides, which are desired. Generally the library will be at least 10⁶ members, usually at least 10⁷, and typically at least 10⁸ members. To generate the collection of oligonucleotides which forms a series of codons encoding a random collection of amino acids and which is ultimately cloned into the vector, a codon motif is used, such as (NNK)_X, where N may be A, C, G, or T (nominally equimolar), K is G or T (nominally equimolar), and x is typically up to about 5, 6, 7, 8, or more, thereby producing libraries of penta-, hexa-, hepta-, and octa-peptides or larger. The third position may also be G or C, designated "S". Thus, NNK or NNS 1 ) code for all the amino acids; 2) code for only one stop codon; and 3) reduce the range of codon bias from 6:1 to 3:1.

It should be understood that, with longer peptides, the size of the library that is generated may become a constraint in the cloning process. The expression of peptides from randomly generated mixtures of oligonucleotides in appropriate recombinant vectors is known in the art (see, e.g., Oliphant et al., Gene 44:177-183). For example, the codon motif (NNK)₆ produces 32 codons, one for each of 12 amino acids, two for each of five amino acids, three for each of three amino acids and one (amber) stop codon. Although this motif produces a codon distribution as equitable as available with standard methods of oligonucleotide synthesis, it results in a bias against peptides containing one-codon residues. In particular, a complete collection of hexacodons contains one sequence encoding each peptide made up of only one-codon amino acids, but contains 729 (3⁶) sequences encoding each peptide with only three-codon amino acids.

An alternative approach to minimize the bias against one-codon residues involves the synthesis of 20 activated trinucleotides, each representing the codon for one of the 20 genetically encoded amino acids. These are synthesized by conventional means, removed from the support while maintaining the base and 5-OH-protecting groups, and activated by the addition of 3'O-phosphoramidite (and phosphate protection with b- cyanoethyl groups) by the method used for the activation of mononucleosides (see, generally, McBride and Caruthers, 1983, Tetrahedron Letters 22:245). Degenerate oligocodons are prepared using these trimers as building blocks. The trimers are mixed at the desired molar ratios and installed in the synthesizer. The ratios will usually be approximately equimolar, but may be a controlled unequal ratio to obtain the over- to under-representation of certain amino acids coded for by the degenerate oligonucleotide collection. The condensation of the trimers to form the oligocodons is done essentially as described for conventional synthesis employing activated mononucleosides as building blocks (see, e.g., Atkinson and Smith, 1984, Oligonucleotide Synthesis, M.J. Gait, Ed., p. 35-82). This procedure generates a population of oligonucleotides for cloning that is capable of encoding an equal distribution (or a controlled unequal distribution) of the possible peptide sequences. Advantageously, this approach may be employed in generating longer peptide sequences, since the range of bias produced by the (NNK)ε motif increases by threefold with each additional amino acid residue.

When the codon motif is (NNK)_X, as defined above, and when x

10 equals 8, there are 2.6. x 10 possible octa-peptides. A library containing most of the octa-peptides may be difficult to produce. Thus, a sampling of the octa-peptides may be accomplished by constructing a subset library using up to about 10% of the possible sequences, which subset of recombinant bacteriophage particles is then screened. If desired, to extend the diversity of a subset library, the recovered phage subset may be subjected to mutagenesis and then subjected to subsequent rounds of screening. This mutagenesis step may be accomplished in two general ways: the variable region of the recovered phage may be mutagenized, or additional variable amino acids may be added to the regions adjoining the initial variable sequences. To diversify around active peptides (i.e., binders) found in early rounds of panning, the positive phage can be sequenced to determine the identity of the active peptides. Oligonucleotides can then be synthesized based on these peptide sequences. The syntheses are done with a low level of all bases incorporated at each step to produce slight variations of the primary oligonucleotide sequences. This mixture of (slightly) degenerate oligonucleotides can then be cloned into the affinity phage by methods known to those in the art. This method produces systematic, controlled variations of the starting peptide sequences as part of a secondary library. It requires, however, that individual positive phage be sequenced before mutagenesis, and thus is useful for expanding the diversity of small numbers of recovered phage.

An alternate approach to diversify the selected phage allows the mutagenesis of a pool, or subset, of recovered phage. In accordance with this approach, phage recovered from panning are pooled and single stranded DNA is isolated. The DNA is mutagenized by treatment with, e.g., nitrous acid, formic acid, or hydrazine. These treatments produce a variety of damage to the DNA. The damaged DNA is then copied with reverse transcriptase, which misincorporates bases when it encounters a site of damage. The segment containing the sequence encoding the receptor- binding peptide is then isolated by cutting with restriction nuclease(s) specific for sites flanking the peptide coding sequence. This mutagenized segment is then recloned into undamaged vector DNA, the DNA is transformed into cells, and a secondary library according to known methods. General mutagenesis methods are known in the art (see Myers et al., 1985, Nucl. Acids Res. 13:3131-3145; Myers et al., 1985, Science 229:242-246; Myers, 1989, Current Protocols in Molecular Biology Vol. I, 8.3.1-8.3.6, F. Ausubel et al., eds, J. Wiley and Sons, New York).

In another general approach, the addition of amino acids to a peptide or peptides found to be active, can be carried out using various methods. In one, the sequences of peptides selected in early panning are determined individually and new oligonucleotides, incorporating the determined sequence and an adjoining degenerate sequence, are synthesized. These are then cloned to produce a secondary library. Alternatively, methods can be used to add a second DGI-3-binding sequence to a pool of peptide- bearing phage. In accordance with one method, a restriction site is installed next to the first DGI-3-binding sequence. Preferably, the enzyme should cut outside of its recognition sequence. The recognition site may be placed several bases from the first binding sequence. To insert a second DGI-3- binding sequence, the pool of phage DNA is digested and blunt-ended by filling in the overhang with Klenow fragment. Double-stranded, blunt-ended, degenerately synthesized oligonucleotides are then ligated into this site to produce a second binding sequence juxtaposed to the first binding sequence. This secondary library is then amplified and screened as before.

While in some instances it may be appropriate to synthesize longer peptides to bind certain receptors, in other cases it may be desirable to provide peptides having two or more DGI-3-binding sequences separated by spacer (e.g., linker) residues. For example, the binding sequences may be separated by spacers that allow the regions of the peptides to be presented to the receptor in different ways. The distance between binding regions may be as little as 1 residue, or at least 2-20 residues, or up to at least 100 residues. Preferred spacers are 3, 6, 9, 12, 15, or 18 residues in length. For probing large binding sites or tandem binding sites, the binding regions may be separated by a spacer of residues of up to 20 to 30 amino acids. The number of spacer residues when present will typically be at least 2 residues, and often will be less than 20 residues.

The oligonucleotide library may have binding sequences which are separated by spacers (e.g., linkers), and thus may be represented by the formula: (NNK)y - (abc)_n - (NNK)_Z where N and K are as defined previously (note that S as defined previously may be substituted for K), and y+z is equal to about 5, 6, 7, 8, or more, a, b and c represent the same or different nucleotides comprising a codon encoding spacer amino acids, n is up to about 3, 6, 9, or 12 amino acids, or more. The spacer residues may be somewhat flexible, comprising oligo-glycine, or oligo-glycine-glycine-serine, for example, to provide the diversity domains of the library with the ability to interact with sites in a large binding site relatively unconstrained by attachment to the phage protein. Rigid spacers, such as, e.g., oligo-proline, may also be inserted separately or in combination with other spacers, including glycine spacers. It may be desired to have the DGI-3-binding sequences close to one another and use a spacer to orient the binding sequences with respect to each other, such as by employing a turn between the two sequences, as might be provided by a spacer of the sequence glycine-proline-glycine, for example. To add stability to such a turn, it may be desirable or necessary to add cysteine residues at either or both ends of each variable region. The cysteine residues would then form disulfide bridges to hold the variable regions together in a loop, and in this fashion may also serve to mimic a cyclic peptide. Of course, those skilled in the art will appreciate that various other types of covalent linkages for cyclization may also be used. Spacer residues as described above may also be situated on either or both ends of the DGI-3-binding sequences. For instance, a cyclic peptide may be designed without an intervening spacer, by having a cysteine residue on both ends of the peptide. As described above, flexible spacers, e.g., oligo-glycine, may facilitate interaction of the peptide with the selected receptors. Alternatively, rigid spacers may allow the peptide to be presented as if on the end of a rigid arm, where the number of residues, e.g., proline residues, determines not only the length of the arm but also the direction for the arm in which the peptide is oriented. Hydrophilic spacers, made up of charged and/or uncharged hydrophilic amino acids, (e.g., Thr, His, Asn, Gin, Arg, Glu, Asp, Met, Lys, etc.), or hydrophobic spacers of hydrophobic amino acids (e.g., Phe, Leu, lie, Gly, Val, Ala, etc.) may be used to present the peptides to receptor binding sites with a variety of local environments. Notably, some peptides, because of their size and/or sequence, may cause severe defects in the infectivity of their carrier phage. This causes a loss of phage from the population during reinfection and amplification following each cycle of panning. To minimize problems associated with defective infectivity, DNA prepared from the eluted phage can be transformed into appropriate host cells, such as, e.g., E. coli, preferably by electroporation (see, e.g., Dower et al., Nucl. Acids Res. 16:6127-6145), or well known chemical means. The cells are cultivated for a period of time sufficient for marker expression, and selection is applied as typically done for DNA transformation. The colonies are amplified, and phage harvested for affinity enrichment in accordance with established methods. Phage identified in the affinity enrichment may be re-amplified by infection into the host cells. The successful transformants are selected by growth in an appropriate antibiotic(s), e.g., tetracycline or ampicillin. This may be done on solid or in liquid growth medium. For growth on solid medium, the cells are grown at a high density

(about 10⁸ to 10⁹ transformants per m²) on a large surface of, for example, L-agar containing the selective antibiotic to form essentially a confluent lawn. The cells and extruded phage are scraped from the surface and phage are prepared for the first round of panning (see, e.g., Parmley and Smith, 1988, Gene 73:305-318). For growth in liquid culture, cells may be grown in L-broth and antibiotic through about 10 or more doublings. The phage are harvested by standard procedures (see Sambrook et al., 1989, Molecular Cloning, 2^nd ed.). Growth in liquid culture may be more convenient because of the size of the libraries, while growth on solid media likely provides less chance of bias during the amplification process. For affinity enrichment of desired clones, generally about 10³ to 10⁴ library equivalents (a library equivalent is one of each recombinant; 10⁴ equivalents of a library of 10⁹ members is 10⁹ x 10⁴ = 10¹³ phage), but typically at least 10² library equivalents, up to about 10⁵ to 10⁶, are incubated with a receptor (or portion thereof) to which the desired peptide is sought. The receptor is in one of several forms appropriate for affinity enrichment schemes. In one example the receptor is immobilized on a surface or particle, and the library of phage bearing peptides is then panned on the immobilized receptor generally according to procedures known in the art. In an alternate scheme, a receptor is attached to a recognizable ligand (which may be attached via a tether). A specific example of such a ligand is biotin. The receptor, so modified, is incubated with the library of phage and binding occurs with both reactants in solution. The resulting complexes are then bound to streptavidin (or avidin) through the biotin moiety. The streptavidin may be immobilized on a surface such as a plastic plate or on particles, in which case the complexes (phage/peptide/receptor/biotin/ streptavidin) are physically retained; or the streptavidin may be labeled, with a fluorophor, for example, to tag the active phage/peptide for detection and/or isolation by sorting procedures, e.g., on a fluorescence-activated cell sorter. Phage that associate with DGI-3 via non-specific interactions are removed by washing. The degree and stringency of washing required will be determined for each receptor/peptide of interest. A certain degree of control can be exerted over the binding characteristics of the peptides recovered by adjusting the conditions of the binding incubation and the subsequent washing. The temperature, pH, ionic strength, divalent cation concentration, and the volume and duration of the washing will select for peptides within particular ranges of affinity for the receptor. Selection based on slow dissociation rate, which is usually predictive of high affinity, is the most practical route. This may be done either by continued incubation in the presence of a saturating amount of free ligand, or by increasing the volume, number, and length of the washes. In each case, the rebinding of dissociated peptide-phage is prevented, and with increasing time, peptide- phage of higher and higher affinity are recovered. Additional modifications of the binding and washing procedures may be applied to find peptides that bind receptors under special conditions. Once a peptide sequence that imparts some affinity and specificity for the receptor molecule is known, the diversity around this binding motif may be embellished. For instance, variable peptide regions may be placed on one or both ends of the identified sequence. The known sequence may be identified from the literature, or may be derived from early rounds of panning in the context of this invention. Notably, sequences identified from secondary peptide libraries can be used to identify amino acid residues that are important for binding or other activity. For example, as described herein below, a secondary peptide library was constructed from the DGI-3-binder A1. Peptides from the A1 secondary library were panned against the DGI-3 polypeptide to identify A1 variants that bound to DGI-3 (Figures 16A-16B). The sequences of the A1 variants were analyzed to determine the amino acid residues that were important for DGI-3 binding activity. Conserved and variable residues were identified from the A1 variants. This analysis revealed the following consensus sequence: XιX₂X₃RWFCLX₉HXnGX₁₃Xι₄Xi₅CXι₇X₁₈Xi₉X₂o (SEQ ID NO:209), wherein Xi is selected from the group consisting of T, N, S, D, and H, and is preferably N, more preferably T; X₂ is selected from the group consisting of I, V, and F, and is preferably I; X₃ is selected from the group consisting of R, L, S, P, G, Y, S, Q, H, and F, and is preferably H, L, or S; X₉ is selected from the group consisting of A, V, L, and I, and is preferably V; X₁₁ is selected from the group consisting of W and Y, and is preferably W; Xι₃ is selected from the group consisting of T, P, V, I, N, Q, and S, and is preferably P; X-ι₄ is selected from the group consisting of E, T, D, and A, and is preferably E; X-ι₅ is selected from the group consisting of G, F, A, S, V, R, D, M, and L, and is preferably D; Xι₇ is selected from the group consisting of L, R, V, Q, P, M, E, and K, and is preferably L; Xι₈ is selected from the group consisting of A, G, S, V, T, P, R, G, and E, and is preferably G, more preferably A; X₁₉ is selected from the group consisting of R, G, D, S, P, and L, and is preferably R; and X₂₀ is selected from the group consisting of T, P, A, S, N, and R, and is preferably T. STRUCTURAL STUDIES Purified DGI-3 (e.g., SEQ ID NO:19), DGI-3-binders (e.g., SEQ ID

NO:3-SEQ ID NO:11 , SEQ ID NO:58-SEQ ID NO:109, SEQ ID NO:110- SEQ ID NO:112, and SEQ ID NO:114-SEQ ID NO:117), or DGI-3-partners (e.g., SEQ ID NO: 20-SEQ ID NO:27), or fragments or complexes thereof, can be analyzed by well-established methods (e.g., X-ray crystallography, NMR, CD, etc.) to determine the three-dimensional structures of these molecules. The three-dimensional structures, in turn, can be used to model intermolecular interactions. Exemplary methods for crystallization and X-ray crystallography are found in P.G. Jones, 1981 , Chemistry in Britain, 17:222- 225; C. Jones et al. (eds), Crystallographic Methods and Protocols, Humana Press, Totowa, NJ; A. McPherson, 1982, Preparation and Analysis of Protein Crystals, John Wiley & Sons, New York, NY; T.L. Blundell and L.N. Johnson, 1976, Protein Crystallography, Academic Press, Inc., New York, NY; A. Holden and P. Singer, 1960, Crystals and Crystal Growing, Anchor Books-Doubleday, New York, NY; R.A. Laudise, 1970, The Growth of Single Crystals, Solid State Physical Electronics Series, N. Holonyak, Jr., (ed), Prentice-Hall, Inc.; G.H. Stout and L.H. Jensen, 1989, X-ray Structure Determination: A Practical Guide, 2nd edition, John Wiliey & Sons, New York, NY; Fundamentals of Analytical Chemistry, 3rd. edition, Saunders Golden Sunburst Series, Holt, Rinehart and Winston, Philadelphia, PA, 1976; P.D. Boyle of the Department of Chemistry of North Carolina State University at hypertext transfer protocol://laue.chem. ncsu.edu/web /GrowXtal.html; M.B. Berry, 1995, Protein Crystalization: Theory and Practice, Structure and Dynamics of E. coli Adenylate Kinase, Doctoral Thesis, Rice University, Houston TX; available online at world wide web.bioc.rice.edu/~berry/papers/ crystalization/crystalization.html. For X-ray diffraction studies, single crystals can be grown to suitable size. Preferably, a crystal has a size of 0.2 to 0.4 mm in at least two of the three dimensions. Crystals can be formed in a solution comprising a DGI-3, DGI-3-binder, or DGI-3-partner polypeptide (e.g., 1.5-200 mg/ml) and reagents that reduce the solubility to conditions close to spontaneous precipitation. Factors that affect the formation of polypeptide crystals include: 1 ) purity; 2) substrates or co-factors; 3) pH; 4) temperature; 5) polypeptide concentration; and 6) characteristics of the precipitant. Preferably, the DGI-3, DGI-3-binder, or DGI-3-partner polypeptides are pure, i.e., free from contaminating components (at least 95% pure), and free from denatured polypeptides. In particular, polypeptides can be purified by FPLC and HPLC techniques to assure homogeneity (see, Lin et al., 1992, J. Crystal. Growth. 122:242-245). Optionally, substrates or co-factors for DGI- 3, DGI-3-binders, or DGI-3-partners can be added to stabilize the quaternary structure of the proteins and promote lattice packing. Suitable precipitants for crystallization include, but are not limited to, salts (e.g., ammonium sulphate, potassium phosphate); polymers (e.g., polyethylene glycol (PEG) 6000); alcohols (e.g., ethanol); polyalcohols (e.g., 1-methyl-2,4 pentane diol (MPD)); organic solvents; sulfonic dyes; and deionized water. The ability of a salt to precipitate polypeptides can be generally described by the Hofmeister series: P0₄ ^3" > HP0 ^2" = S0₄ ^2" > citrate > CH₃C0₂ ^" > Cl^" > Br^" > N0₃ ^" > CI0₄ ^" > SCN^"; and NH₄ ⁺ > K⁺ > Na⁺ > Li⁺. Non-limiting examples of salt precipitants are shown below (see Berry, 1995).

TABLE 2

High molecular weight polymers useful as precipitating agents include polyethylene glycol (PEG), dextran, polyvinyl alcohol, and polyvinyl pyrrolidone (A. Poison et al., 1964, Biochem. Biophys. Acta. 82:463-475). In general, polyethylene glycol (PEG) is the most effective for forming crystals. PEG compounds with molecular weights less than 1000 can be used at concentrations above 40% v/v. PEGs with molecular weights above 1000 can be used at concentration 5-50% w/v. Typically, PEG solutions are mixed with -0.I % sodium azide to prevent bacterial growth.

Typically, crystallization requires the addition of buffers and a specific salt content to maintain the proper pH and ionic strength for a protein's stability. Suitable additives include, but are not limited to sodium chloride (e.g., 50-500 mM as additive to PEG and MPD; 0.15-2 M as additive to PEG); potassium chloride (e.g., 0.05-2 M); lithium chloride (e.g., 0.05-2 M); sodium fluoride (e.g., 20-300 mM); ammonium sulfate (e.g., 20-300 mM); lithium sulfate (e.g., 0.05-2 M); sodium or ammonium thiocyanate (e.g., 50- 500 mM); MPD (e.g., 0.5-50%); 1 ,6 hexane diol (e.g., 0.5-10%); 1 ,2,3 heptane triol (e.g., 0.5-15%); and benzamidine (e.g., 0.5-15%).

Detergents may be used to maintain protein solubility and prevent aggregation. Suitable detergents include, but are not limited to non-ionic detergents such as sugar derivatives, oligoethyleneglycol derivatives, dimethylamine-N-oxides, cholate derivatives, N-octyl hydroxyalkylsulphoxides, sulphobetains, and lipid-like detergents. Sugar- derived detergents include alkyl glucopyranosides (e.g., C8-GP, C9-GP), alkyl thio-glucopyranosides (e.g., C8-tGP), alkyl maltopyranosides (e.g., C10-M, C12-M; CYMAL-3, CYMAL-5, CYMAL-6), alkyl thio- maltopyranosides, alkyl galactopyranosides, alkyl sucroses (e.g., N- octanoylsucrose), and glucamides (e.g., HECAMEG, C-HEGA-10; MEGA- 8). Oligoethyleneglycol-derived detergents include alkyl polyoxyethylenes (e.g., C8-E5, C8-En; C12-E8; C12-E9) and phenyl polyoxyethylenes (e.g., Triton X-100). Dimethylamine-N-oxide detergents include, e.g., C10-DAO; DDAO; LDAO. Cholate-derived detergents include, e.g., Deoxy-Big CHAP, digitonin. Lipid-like detergents include phosphocholine compounds. Suitable detergents further include zwitter-ionic detergents (e.g., ZWITTERGENT 3-10; ZWITTERGENT 3-12); and ionic detergents (e.g., SDS).

Crystallization of macromolecules has been performed at temperatures ranging from 60°C to less than 0°C However, most molecules can be crystallized at 4°C or 22°C Lower temperatures promote stabilization of polypeptides and inhibit bacterial growth. In general, polypeptides are more soluble in salt solutions at lower temperatures (e.g., 4°C), but less soluble in PEG and MPD solutions at lower temperatures. To allow crystallization at 4°C or 22°C, the precipitant or protein concentration can be increased or decreased as required. Heating, melting, and cooling of crystals or aggregates can be used to enlarge crystals. In addition, crystallization at both 4°C and 22°C can be assessed (A. McPherson, 1992, J. Cryst. Growth. 122:161-167; C.W. Carter, Jr. and C.W. Carter, 1979, J. Biol. Chem. 254:12219-12223; T. Bergfors, 1993, Crystalization Lab Manual).

A crystallization protocol can be adapted to a particular polypeptide or peptide. In particular, the physical and chemical properties of the polypeptide can be considered (e.g., aggregation, stability, adherence to membranes or tubing, internal disulfide linkages, surface cysteines, chelating ions, etc.). For initial experiments, the standard set of crystalization reagents can be used (Hampton Research, Laguna Niguel, CA). In addition, the CRYSTOOL program can provide guidance in determining optimal crystallization conditions (Brent Segelke, 1995, Efficiency analysis of sampling protocols used in protein crystallization screening and crystal structure from two novel crystal forms of PLA2, Ph.D. Thesis, University of California, San Diego; available online at world wide web.ccp14.ac.uk/ccp/web-mirrors/llnlrupp/crystool/crystool.htm). Exemplary crystallization conditions are shown below (see Berry, 1995).

TABLE 3

Robots can be used for automatic screening and optimization of crystallization conditions. For example, the IMPAX and Oryx systems can be used (Douglas Instruments, Ltd., East Garston, United Kingdom). The CRYSTOOL program (Segelke, supra) can be integrated with the robotics programming. In addition, the Xact program can be used to construct, maintain, and record the results of various crystallization experiments (see, e.g., D.E. Brodersen et al., 1999, J. Appl. Cryst. 32: 1012-1016; G.R. Andersen and J. Nyborg, 1996, J. Appl. Cryst. 29:236-240). The Xact program supports multiple users and organizes the results of crystallization experiments into hierarchies. Advantageously, Xact is compatible with both CRYSTOOL and Microsoft® Excel programs.

Four methods are commonly employed to crystallize macromolecules: vapor diffusion, free interface diffusion, batch, and dialysis. The vapor diffusion technique is typically performed by formulating a 1 :1 mixture of a solution comprising the polypeptide of interest and a solution containing the precipitant at the final concentration that is to be achieved after vapor equilibration. The drop containing the 1 :1 mixture of protein and precipitant is then suspended and sealed over the well solution, which contains the precipitant at the target concentration, as either a hanging or sitting drop. Vapor diffusion can be used to screen a large number of crystallization conditions or when small amounts of polypeptide are available. For screening, drop sizes of 1 to 2 μl can be used. Once preliminary crystallization conditions have been determined, drop sizes such as 10 μl can be used. Notably, results from hanging drops may be improved with agarose gels (see K. Provost and M.-C Robert, 1991 , J. Cryst. Growth. 110:258-264). Free interface diffusion is performed by layering of a low density solution onto one of higher density, usually in the form of concentrated protein onto concentrated salt. Since the solute to be crystallized must be concentrated, this method typically requires relatively large amounts of protein. However, the method can be adapted to work with small amounts of protein. In a representative experiment, 2 to 5 μl of sample is pipetted into one end of a 20 μl microcapillary pipette. Next, 2 to 5 μl of precipitant is pipetted into the capillary without introducing an air bubble, and the ends of the pipette are sealed. With sufficient amounts of protein, this method can be used to obtain relatively large crystals (see, e.g., S.M. Althoff et al., 1988, J. Mol. Biol. 199:665-666).

The batch technique is performed by mixing concentrated polypeptide with concentrated precipitant to produce a final concentration that is supersaturated for the solute macromolecule. Notably, this method can employ relatively large amounts of solution (e.g., milliliter quantities), and can produce large crystals. For that reason, the batch technique is not recommended for screening initial crystallization conditions.

The dialysis technique is performed by diffusing precipitant molecules through a semipermeable membrane to slowly increase the concentration of the solute inside the membrane. Dialysis tubing can be used to dialyze milliliter quantities of sample, whereas dialysis buttons can be used to dialyze microliter quantities (e.g., 7-200 μl). Dialysis buttons may be constructed out of glass, perspex, or Teflon™ (see, e.g., Cambridge Repetition Engineers Ltd., Greens Road, Cambridge CB4 3EQ, UK; Hampton Research). Using this method, the precipitating solution can be varied by moving the entire dialysis button or sack into a different solution. In this way, polypeptides can be "reused" until the correct conditions for crystallization are found (see, e.g., C.W. Carter, Jr. et al., 1988, J. Cryst. Growth. 90:60-73). However, this method is not recommended for precipitants comprising concentrated PEG solutions. Various strategies have been designed to screen crystallization conditions, including 1 ) pl screening; 2) grid screening; 3) factorials; 4) solubility assays; 5) perturbation; and 6) sparse matrices. In accordance with the pl screening method, the pl of a polypeptide is presumed to be its crystallization point. Screening at the pl can be performed by dialysis against low concentrations of buffer (less than 20 mM) at the appropriate pH, or by use of conventional precipitants.

The grid screening method can be performed on two-dimensional matrices. Typically, the precipitant concentration is plotted against pH. The optimal conditions can be determined for each axis, and then combined. At that point, additional factors can be tested (e.g., temperature, additives). This method works best with fast-forming crystals, and can be readily automated (see M.J. Cox and P.C Weber, 1988, J. Cryst. Growth. 90:318- 324). Grid screens are commercially available for popular precipitants such as ammonium sulphate, PEG 6000, MPD, PEG/LiCI, and NaCl (see, e.g., Hamilton Research).

The incomplete factorial method can be performed by 1 ) selecting a set of -20 conditions; 2) randomly assigning combinations of these conditions; 3) grading the success of the results of each experiment using an objective scale; and 4) statistically evaluating the effects of each of the conditions on crystal formation (see, e.g., C.W. Carter, Jr. et al., 1988, J. Cryst. Growth. 90:60-73). In particular, conditions such as pH, temperature, precipitating agent, and cations can be tested. Dialysis buttons are preferably used with this method. Typically, optimal conditions/combinations can be determined within 35 tests. Similar approaches, such as "footprinting" conditions, may also be employed (see, e.g., E.A. Stura et al., 1991 , J Cryst. Growth. 110:1-2).

The perturbation approach can be performed by altering crystallization conditions by introducing a series of additives designed to test the effects of altering the structure of bulk solvent and the solvent dielectric on crystal formation (see, e.g., Whitaker et al., 1995, Biochem. 34:8221- 8226). Additives for increasing the solvent dialectric include, but are not limited to, NaCl, KCI, or LiCI (e.g., 200 mM); Na formate (e.g., 200 mM); Na₂HP0 or K₂HP0 (e.g., 200 mM); urea, triachloroacetate, guanidium HCl, or KSCN (e.g., 20-50 mM). A non-limiting list of additives for decreasing the solvent dialectric include methanol, ethanol, isopropanol, or tert-butanol (e.g., 1-5%); MPD (e.g., 1 %); PEG 400, PEG 600, or PEG 1000 (e.g., 1- 4%); PEG MME (monomethylether) 550, PEG MME 750, PEG MME 2000 (e.g., 1-4%).

As an alternative to the above-screening methods, the sparse matrix approach can be used (see, e.g., J. Jancarik and S.-H.J. Kim, 1991 , Appl. Cryst. 24:409-411 ; A. McPherson, 1992, J. Cryst. Growth. 122:161-167; B. Cudney et al., 1994, Acta. Cryst. D50:414-423). Sparse matrix screens are commercially available (see, e.g., Hampton Research; Molecular Dimensions, Inc., Apopka, FL; Emerald Biostructures, Inc., Lemont, IL). Notably, data from Hampton Research sparse matrix screens can be stored and analyzed using ASPRUN software (Douglas Instruments).

Exemplary conditions for an initial screen are shown below (see Berry, 1995). TABLE 4

Tray 2:

The initial screen can be used with hanging or sitting drops. To conserve the sample, tray 2 can be set up several weeks following tray 1. Wells 31-48 of tray 2 can comprise a random set of solutions. Alternatively, solutions can be formulated using sparse methods. Preferably, test solutions cover a broad range of precipitants, additives, and pH (especially pH 5.0-9.0).

Seeding can be used to trigger nucleation and crystal growth (Stura and Wilson, 1990, J. Cryst. Growth. 110:270-282; C. Thaller et al., 1981 , J. Mol. Biol. 147:465-469; A. McPherson and P. Schlichta, 1988, J. Cryst. Growth. 90:47-50). In general, seeding can be performed by transferring crystal seeds into a polypeptide solution to allow polypeptide molecules to deposit on the surface of the seeds and produce crystals. Two seeding methods can be used: microseeding and macroseeding. For microseeding, a crystal can be ground into tiny pieces and transferred into the protein solution. Alternatively, seeds can be transferred by adding 1-2 μl of the seed solution directly to the equilibrated protein solution. In another approach, seeds can be transferred by dipping a hair in the seed solution and then streaking the hair across the surface of the drop (streak seeding; see Stura and Wilson, supra). For macroseeding, an intact crystal can be transferred into the protein solution (see, e.g., C. Thaller et al., 1981 , J. Mol. Biol. 147:465-469). Preferably, the surface of the crystal seed is washed to regenerate the growing surface prior to being transferred. Optimally, the protein solution for crystallization is close to saturation and the crystal seed is not completely dissolved upon transfer.

In certain cases, publicly-available structural information can be used for the purposes of this invention. For example, the structure of the pleckstrin homology domain is available from the NCBI website, as part of the crystal structure of murine GRP1 (Figure 29). See Mus musculus MMDB: 14112; PDB: 1 FHX; Structure Of The Pleckstrin Homology Domain From Grp1 In Complex With Inositol 1 ,3,4,5-Tetrakisphosphate, deposited by M. Ferguson et al., 2-Aug-00. To identify drug candidates, the PHD can be used as a pharmacophore for modeling small molecule drug candidates that inhibit DGI-3 binding to its partner. In particular, small molecules that bind to the PHD hot-spot (highlighted in Figure 29) can be identified by virtual (i.e., in silico) screening (see, e.g., S. Putta et al., 2002, J Chem. Inf. Comput. Sci. 42(5): 1230-1240; D. Eros et al., 2002, Curr. Med. Chem. 9(20): 1819-29; R.D. Taylor et al., 2002, J. Comput. Aided Mol. Des. 16(3):151-66). Based on the identification of peptide A1 , the region within the PHD to which DGI-3 binds has been identified. This region includes amino acids 21-31 of the PHD (RRWFILTDNCL; SEQ ID NO:189), which corresponds to amino acids 280-290 of cytohesin-1. Various in silico approaches may be used to identify small molecules which bind to this region and thereby block the interaction between the PHD and DGI-3, and between cytohesin-1 and DGI-3. ANTIBODIES

Another aspect of the invention pertains to antibodies directed to DGI-3 (e.g., SEQ ID NO:19), DGI-3-binders (e.g., SEQ ID NO:3-SEQ ID NO:11 , SEQ ID NO:58-SEQ ID NO:109, SEQ ID NO:110-SEQ ID NO:112, and SEQ ID NO:114-SEQ ID NO:117), or DGI-3-partners (e.g., SEQ ID NO:20-SEQ ID NO:27) or fragments or variants thereof. Specifically included are antibodies directed to polypeptide-complexes of DGI-3 and DGI-3-partners, polypeptide-complexes of DGI-3 and DGI-3-binders, or fragments of these complexes. In particular embodiments, antibodies directed to polypeptide-complexes of DGI-3/DGI-3-partners or DGI-3/DGI-3- binders are complex-specific, i.e., the antibodies do not bind to the components separately. The invention provides polyclonal and monoclonal antibodies that bind to DGI-3, DGI-3-binders, or DGI-3-partners, fragments, or complexes thereof. The antibodies may be elicited in an animal host (e.g., rabbit, goat, mouse, or other non-human mammal) by immunization with disorder-associated immunogenic components. Antibodies may also be elicited by in vitro immunization (sensitization) of immune cells. The immunogenic components used to elicit the production of antibodies may be isolated from cells or chemically synthesized. The antibodies may also be produced in recombinant systems programmed with appropriate antibody- encoding DNA. Alternatively, the antibodies may be constructed by biochemical reconstitution of purified heavy and light chains. The antibodies include hybrid antibodies, chimeric antibodies, and univalent antibodies. Also included are Fab fragments, including Fabi and F(ab)₂ fragments of antibodies.

An isolated DGI-3, DGI-3-binder, or DGI-3-partner polypeptide, peptide, or complex thereof, can be used as an immunogen to generate antibodies using standard techniques for polyclonal and monoclonal antibody preparation. Full-length polypeptides can be used or, alternatively, the invention provides antigenic peptide portions of these polypeptides for use as immunogens. The antigenic peptide of DGI-3 or a DGI-3-partner comprises a sufficient number of contiguous amino acid residues of the amino acid sequence, or a variant thereof, to encompass an epitope of a DGI-3, DGI-3-binder, or DGI-3-partner polypeptide such that an antibody raised against the peptide forms a specific immune complex with a DGI-3, DGI-3-binder, or DGI-3-partner amino acid sequence. Typically, about 5 contiguous amino acids are sufficient to define an epitope.

An appropriate immunogenic preparation can contain, for example, 1 ) recombinantly produced DGI-3, DGI-3-binder, or DGI-3-partner polypeptides; 2) chemically synthesized DGI-3, DGI-3-binder, or DGI-3- partner polypeptides, 3) fragments of these polypeptides (i.e., peptides); or 4) complexes comprising these polypeptides. The preparation can further include an adjuvant, such as Freund's complete or incomplete adjuvant, or similar immunostimulatory agent. A number of adjuvants are known and used by those skilled in the art. Non-limiting examples of suitable adjuvants include incomplete Freund's adjuvant, mineral gels such as alum, aluminum phosphate, aluminum hydroxide, aluminum silica, and surface-active substances such as lysolecithin, pluronic polyols, polyanions, peptides, oil emulsions, keyhole limpet hemocyanin, and dinitrophenol. Further examples of adjuvants include stearyl tyrosine (A. Nixon-George et al., 1990, J. Immunol. 144:4798-4802; Paoletti, et al., 1997, J. Infect. Diseases 175:1237-9; U.S. Pat. No. 4,258,029 to Moloney et al.; U.S. Pat. No. 5,683,699 to Jennings, et al.), N-acetyl-muramyl-L-threonyl-D-isoglutamine (thr-MDP), N-acetyl-nor-muramyl-L-alanyl-D-isoglutamine (CGP 11637, referred to as nor-MDP), N-acetylmuramyl-Lalanyl-D-isoglutaminyl-L- alanine-2-(1'-2'-dipalmitoyl-sn-glycero-3 hydroxyphosphoryloxy)-ethylamine (CGP 19835A, referred to as MTP-PE), and RIBI adjuvant (e.g., Detox®, Corixa Corp., Seattle WA) which contains three components extracted from bacteria, monophosphoryl lipid A, trehalose dimycolate and cell wall skeleton (TDM+CWS+MPL®; Corixa Corp.) in a 2% squalene/Tween®80 emulsion. A particularly useful adjuvant comprises 5% (wt/vol) squalene, 2.5% Pluronic L121 polymer and 0.2% polysorbate in phosphate buffered saline (Kwak et al., 1992, New Eng. J. Med. 327:1209-1215). Preferred adjuvants include complete BCG, Detox, (RIBI, Immunochem Research Inc.), ISCOMS, and aluminum hydroxide adjuvant (Superphos, Biosector). The effectiveness of an adjuvant may be determined by measuring the amount of antibodies directed against the immunogenic peptide.

Polyclonal antibodies to DGI-3, DGI-3-binder, or DGI-3-partner polypeptides, peptides, or polypeptide-complexes thereof, can be prepared as described above by immunizing a suitable subject with a DGI-3, DGI-3- binder, or DGI-3-partner immunogen. The antibody titer in the immunized subject can be monitored over time by standard techniques, such as with an enzyme linked immunosorbent assay (ELISA) using immobilized DGI-3, DGI-3-binder, or DGI-3-partner polypeptide, peptide, or polypeptide- complex. If desired, the antibody molecules can be isolated from the mammal (e.g., from the blood) and further purified by well-known techniques, such as protein A chromatography to obtain the IgG fraction.

At an appropriate time after immunization, e.g., when the antibody titers are highest, antibody-producing cells can be obtained from the subject and used to prepare monoclonal antibodies by standard techniques, such as the hybridoma technique (see Kohler and Milstein, 1975, Nature

256:495-497; Brown et al., 1981 , J. Immunol. 127:539-46; Brown et al.,

1980, J. Biol. Chem. 255:4980-83; Yeh et al., 1976, PNAS 76:2927-31 ; and Yeh et al., 1982, Int. J. Cancer 29:269-75), the human B cell hybridoma technique (Kozbor et al., 1983, Immunol. Today 4:72), the EBV-hybridoma technique (Cole et al., 1985, Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, Inc., pp. 77-96) or trioma techniques.

The technology for producing hybridomas is well-known (see generally R. H. Kenneth, 1980, Monoclonal Antibodies: A New Dimension In Biological Analyses, Plenum Publishing Corp., New York, NY; E.A. Lerner,

1981 , Yale J. Biol. Med., 54:387-402; M.L. Gefter et al., 1977, Somatic Cell Genet. 3:231-36). In general, an immortal cell line (typically a myeloma) is fused to lymphocytes (typically splenocytes) from a mammal immunized with a DGI-3, DGI-3-binder, or DGI-3-partner immunogen as described above, and the culture supernatants of the resulting hybridoma cells are screened to identify a hybridoma producing a monoclonal antibody that binds DGI-3, DGI-3-binder, or DGI-3-partner polypeptides, peptides, or polypeptide-complexes.

Any of the many well known protocols used for fusing lymphocytes and immortalized cell lines can be applied for the purpose of generating an monoclonal antibody to a DGI-3, DGI-3-binder, or DGI-3-partner, peptide, or polypeptide-complex (see, e.g., G. Galfre et al., 1977, Nature 266:55052; Gefter et al., 1977; Lerner, 1981 ; Kenneth, 1980). Moreover, the ordinarily skilled worker will appreciate that there are many variations of such methods. Typically, the immortal cell line (e.g., a myeloma cell line) is derived from the same mammalian species as the lymphocytes. For example, murine hybridomas can be made by fusing lymphocytes from a mouse immunized with an immunogenic preparation of this invention with an immortalized mouse cell line. Preferred immortal cell lines are mouse myeloma cell lines that are sensitive to culture medium containing hypoxanthine, aminopterin, and thymidine (HAT medium). Any of a number of myeloma cell lines can be used as a fusion partner according to standard techniques, e.g., the P3-NS1/1-Ag4-1 , P3-x63-Ag8.653, or Sp2/0-Ag14 myeloma lines. These myeloma lines are available from ATCC (American Type Culture Collection, Manassas, VA). Typically, HAT-sensitive mouse myeloma cells are fused to mouse splenocytes using polyethylene glycol (PEG). Hybridoma cells resulting from the fusion arc then selected using HAT medium, which kills unfused and unproductively fused myeloma cells (unfused splenocytes die after several days because they are not transformed). Hybridoma cells producing a monoclonal antibody of the invention are detected by screening the hybridoma culture supernatants for antibodies that bind DGI-3, DGI-3-binder, or DGI-3-partners, peptides, or polypeptide-complexes, e.g., using a standard ELISA assay.

Alternative to preparing monoclonal antibody-secreting hybridomas, a monoclonal antibody can be identified and isolated by screening a recombinant combinatorial immunoglobulin library (e.g., an antibody phage display library) with the corresponding DGI-3, DGI-3-binder, or DGI-3- partner, peptide, or polypeptide-complex to thereby isolate immunoglobulin library members that bind these molecules. Kits for generating and screening phage display libraries are commercially available (e.g., the Pharmacia Recombinant Phage Antibody System, Catalog No. 27-9400-01 ; and the Stratagene SurfZAP™ Phage Display Kit, Catalog No. 240612).

Additionally, examples of methods and reagents particularly amenable for use in generating and screening antibody display library can be found in, for example, A. Blume U.S. Patent No. 6,010,861 , Ladner et al. U.S. Pat. No. 5,223,409; Kang et al. International Publication No. WO 92/18619; Dower et al. International Publication No. WO 91/17271 ; Winter et al. International Publication WO 92/20791 ; Markland et al. International Publication No. WO 92/15679; Breitling et al. International Publication WO 93/01288; McCafferty et al. International Publication No. WO 92/01047; Garrard et al. International Publication No. WO 92/09690; Ladner et al. International Publication No. WO 90/02809; Fuchs et al., 1991 , Bio/Technology 9:1370-1372; Hay et al., 1992, Hum. Antibod. Hybridomas 3:81-85; Huse et al., 1989, Science 246:1275-1281 ; Griffiths et al., 1993, EMBO J 12:725-734; Hawkins et al., 1992, J. Mol. Biol. 226:889-896; Clarkson et al., 1991 , Nature 352:624-628; Gram et al., 1992, PNAS 89:3576-3580; Garrad et al., 1991 , Bio/Technology 9:1373-1377;

Hoogenboom et al., 1991 , Nuc. Acid Res. 19:4133-4137; Barbas et al.,

1991 , PNAS 88:7978-7982; and McCafferty et al., 1990, Nature 348:552-55.

Additionally, recombinant antibodies to a DGI-3, DGI-3-binder, or

DGI-3-partner, peptide, or polypeptide-complex, such as chimeric and humanized monoclonal antibodies, comprising both human and non-human portions, can be made using standard recombinant DNA techniques. Such chimeric and humanized monoclonal antibodies can be produced by recombinant DNA techniques known in the art, for example using methods described in Robinson et al. International Application No. PCT/US86/02269; Akira, et al. European Patent Application 184,187; Taniguchi, M., European Patent Application 171 ,496; Morrison et al. European Patent Application 173,494; Neuberger et al. PCT International Publication No. WO 86/01533; Cabilly et al. U.S. Pat. No. 4,816,567; Cabilly et al. European Patent Application 125,023; Better et al., 1988, Science 240:1041-1043; Liu et al.,

1987, PNAS 84:3439-3443; Liu et al., 1987, J. Immunol. 139:3521-3526; Sun et al., 1987, PNAS 84:214-218; Nishimura et al., 1987, Cane. Res.

47:999-1005; Wood et al., 1985, Nature 314:446-449; and Shaw et al.,

1988, J. Natl. Cancer Inst. 80:1553-1559; S.L. Morrison, 1985, Science 229:1202-1207; Oi et al., 1986, BioTechniques 4:214; Winter U.S. Pat. No. 5,225,539; Jones et al., 1986, Nature 321 :552-525; Verhoeyan et al., 1988, Science 239:1534; and Bcidler et al., 1988, J. Immunol. 141 :4053-4060.

An antibody against a DGI-3, DGI-3-binder, or DGI-3-partner, peptide, or polypeptide-complex (e.g., monoclonal antibody) can be used to isolate the corresponding molecules by standard techniques, such as affinity chromatography or immunoprecipitation. For example, antibodies can facilitate the purification of natural polypeptides or polypeptide-complexes from cells and of recombinantly produced polypeptides or complexes produced in host cells. In addition, an antibody that binds to a DGI-3, DGI- 3-binder, or DGI-3-partner, peptide, or complex can be used to detect the corresponding molecule (e.g., in a cellular lysate or cell supernatant) in order to evaluate patterns or levels of protein expression or complexation. Such antibodies can also be used as diagnostics to monitor polypeptide or polypeptide-complex levels in tissue as part of a clinical testing procedure, e.g., to, for example, determine the efficacy of a given treatment regimen as described in detail herein. In addition, antibodies to a DGI-3, DGI-3-binder, or DGI-3-partner, peptide, or polypeptide-complex can be used as therapeutics for the treatment of diseases related to abnormal gene expression or function, e.g., as relating to the development of neoplasms and cancers. LIGANDS DGI-3, DGI-3-binder, or DGI-3-partner polynucleotides (e.g., SEQ ID

NO:18; SEQ ID NO:28-SEQ ID NO:36), polypeptides or peptides (e.g., SEQ ID NO:3-SEQ ID NO:11 ; SEQ ID NO:19-SEQ ID NO:27; SEQ ID NO:58- SEQ ID NO:109; SEQ ID NO:110-SEQ ID N0:112; and SEQ ID N0:114- SEQ ID N0:117), variants, or fragments thereof, can be used to screen for ligands that alter the levels or activity of the DGI-3 or DGI-3-partner polypeptides. In accordance with this invention, ligands are agonists, antagonists, inhibitors, or other modulators. DGI-3, DGI-3-binder, or DGI-3- partner polynucleotides, polypeptides, peptides, variants, or fragments thereof can used to identify synthetic or recombinant ligands, or endogenous ligands, that bind to DGI-3 or DGI-3-partner polypeptides or polynucleotides. Of particular interest, are ligands that alter the interaction of DGI-3 and its partners. For example, useful ligands may increase, decrease, or otherwise alter the rate of formation or the stability of DGI- 3/DGI-3-partner complexes. Notably, the invention encompasses methods of identifying DGI-3 ligands (e.g., agonists, antagonists, inhibitors, or other modulators), as well as the ligands identified by these methods, compositions (e.g., pharmaceutical compositions) comprising these ligands, and methods of using these ligands in diagnostic and therapeutic applications, as described in detail herein.

In one aspect of this invention, the full-length DGI-3 (e.g., SEQ ID NO:19) or DGI-3-partner (e.g., SEQ ID NO: 20-SEQ ID NO: 27) polypeptide, or a complex thereof, is used to identify ligands. Alternatively, variants or portions of DGI-3 or DGI-3-partner polypeptides or polypeptide-complexes are used. Such portions may comprise, for example, a DGI-3-binder (e.g., SEQ ID NO:3-SEQ ID NO:11 , SEQ ID NO:58-SEQ ID NO:109, SEQ ID NO:110-SEQ ID NO:112, and SEQ ID NO:114-SEQ ID NO:117), or one or more domains of the polypeptides (e.g., PH, GEF, or protein kinase domains) disclosed herein. Particularly useful are screening assays that identify agents that have relatively low levels of toxicity in human cells. A wide variety of assays may be used for this purpose, including in vitro protein-protein binding assays, electrophoretic mobility shift assays, immunoassays, and the like. Ligands that bind to the DGI-3, DGI-3-binder, or DGI-3-partner polypeptides or polynucleotides of the invention have use in diagnostic applications and/or pharmaceutical compositions relating to neoplastic growth and cancer, as described in detail herein. Ligands may be identified from test agents encompassing numerous chemical classes, though typically they are organic molecules, e.g., small molecules. Preferably, test agents have a molecular weight of less than 5000 daltons, more preferably, test agents have a molecular weight of more than 50 and less than 2,500 daltons. Such test agents can comprise functional groups necessary for structural interaction with proteins, particularly hydrogen bonding, and typically include at least an amine, carbonyl, hydroxyl or carboxyl group, preferably at least two of the functional chemical groups. Useful test agents often comprise cyclical carbon or heterocyclic structures and/or aromatic or polyaromatic structures substituted with one or more of the above functional groups. Such test agents can also comprise biomolecules including peptides, saccharides, fatty acids, steroids, purines, pyrimidines, derivatives, structural analogs, or combinations thereof.

Test agents may include, for example, 1 ) peptides such as soluble peptides, including Ig-tailed fusion peptides and members of random peptide libraries (see, e.g., Lam et al., 1991 , Nature 354:82-84; Houghten et al., 1991 , Nature 354:84-86) and combinatorial chemistry-derived molecular libraries made of D- and/or L-configuration amino acids; 2) phosphopeptides (e.g., members of random and partially degenerate, directed phosphopeptide libraries, see, e.g., Songyang et al, 1993, Cell 72:767-778); 3) antibodies (e.g., polyclonal, monoclonal, humanized, anti-idiotypic, chimeric, and single chain antibodies as well as Fab, F(ab')₂, Fab expression library fragments, and epitope-binding fragments of antibodies); and 4) small organic and inorganic molecules. Peptide libraries for use with the invention can include as well as in vitro display libraries (see, e.g., R. Roberts and J. Szostak, 1997, Proc. Natl. Acad. Sci. USA 94:12297-12302), as well as non-display libraries (e.g., R. Frank, 2002, J. Immunol. Methods 267(1 ):13-26; R. Liu et al., 2002, Am. Chem. Soc. 124(26):7678-80; C. Pinilla et al., 2001 , Cancer Res. 61 (13):5153-60).

Test agents can be obtained from a wide variety of sources including libraries of synthetic or natural compounds. Synthetic compound libraries are commercially available from, for example, Maybridge Chemical Co. (Trevillet, Cornwall, UK), Comgenex (Princeton, NJ), Brandon Associates (Merrimack, NH), and Microsource (New Milford, CT). A rare chemical library is available from Aldrich Chemical Company, Inc. (Milwaukee, WI). Natural compound libraries comprising bacterial, fungal, plant or animal extracts are available from, for example, Pan Laboratories (Bothell, WA). In addition, numerous means are available for random and directed synthesis of a wide variety of organic compounds and biomolecules, including expression of randomized oligonucleotides. The combinatorial chemistry libraries of the invention include display and non-display libraries. Alternatively, libraries of natural compounds in the form of bacterial, fungal, plant and animal extracts can be readily produced. Methods for the synthesis of molecular libraries are readily available (see, e.g., DeWitt et al., 1993, Proc. Natl. Acad. Sci. USA 90:6909; Erb et al., 1994, Proc. Natl. Acad. Sci. USA 91 :11422; Zuckermann et al., 1994, J. Med. Chem. 37:2678; Cho et al., 1993, Science 261 :1303; Carell et al., 1994, Angew. Chem. Int. Ed. Engl. 33:2059; Carell et al., 1994, Angew. Chem. Int. Ed. Engl. 33:2061 ; and in Gallop et al., 1994, J. Med. Chem. 37:1233). In addition, natural or synthetic compound libraries and compounds can be readily modified through conventional chemical, physical and biochemical means (see, e.g., Blondelle et al., 1996, Trends in Biotech. 14:60), and may be used to produce combinatorial libraries. In another approach, previously identified pharmacological agents can be subjected to directed or random chemical modifications, such as acylation, alkylation, esterification, amidification, and the analogs can be screened for DGI-3- and/or DGI-3- partner-modulating activity. Numerous methods for producing combinatorial libraries are known in the art, including those involving biological libraries; spatially addressable parallel solid phase or solution phase libraries; synthetic library methods requiring deconvolution; the 'one-bead one-compound' library method; and synthetic library methods using affinity chromatography selection. The biological library approach is limited to polypeptide libraries, while the other four approaches are applicable to polypeptide, non-peptide oligomer, or small molecule libraries of compounds (K. S. Lam, 1997, Anticancer Drug Des. 12:145). Non-limiting examples of small molecules, small molecule libraries, combinatorial libraries, and screening methods are described in B. Seligmann, 1995, "Synthesis, Screening, Identification of Positive Compounds and Optimization of Leads from Combinatorial Libraries: Validation of Success" p. 69-70. Symposium: Exploiting Molecular Diversity: Small Molecule Libraries for Drug Discovery, La Jolla, CA, Jan. 23-25, 1995 (conference summary available from Wendy Warr & Associates, 6 Berwick Court, Cheshire, UK CW4 7HZ); E. Martin et al., 1995, J. Med. Chem. 38:1431-1436; E. Martin et al., 1995, "Measuring diversity: Experimental design of combinatorial libraries for drug discovery" Abstract, ACS Meeting, Anaheim, CA, COMP 32; and E. Martin, 1995, "Measuring Chemical Diversity: Random Screening or Rationale Library Design" p. 27-30, Symposium: Exploiting Molecular Diversity: Small Molecule Libraries for Drug Discovery, La Jolla, Calif. Jan. 23-25, 1995 (conference summary available from Wendy Warr & Associates, 6 Berwick Court, Cheshire, UK CW4 7HZ).

Libraries may be screened in solution (e.g., Houghten, 1992, Biotechniques 13:412-421 ), or on beads (Lam, 1991 , Nature 354:82-84), chips (Fodor, 1993, Nature 364:555-556), bacteria or spores (Ladner U.S. Pat. No. 5,223,409), plasmids (Cull et al., 1992, Proc. Natl. Acad. Sci. USA 89:1865-1869), or on phage (Scott and Smith, 1990, Science 249:386-390; Devlin, 1990, Science 249:404-406; Cwirla et al., 1990, Proc. Natl. Acad. Sci. USA 97:6378-6382; Felici, 1991 , J. Mol. Biol. 222:301-310; Ladner, supra).

Where the screening assay is a binding assay, a DGI-3, DGI-3- binder, or DGI-3-partner polypeptide, polynucleotide, variant, polypeptide- complex, or fragment thereof, may be joined to a label, where the label can directly or indirectly provide a detectable signal. Various labels include radioisotopes, fluorescers, chemiluminescers, enzymes, specific binding molecules, particles, e.g., magnetic particles, and the like. Specific binding molecules include pairs, such as biotin and streptavidin, digoxin and antidigoxin, etc. For the specific binding members, the complementary member would normally be labeled with a molecule that provides for detection, in accordance with known procedures.

A variety of other reagents may be included in the screening assay. These include reagents like salts, neutral proteins, e.g., albumin, detergents, etc., that are used to facilitate optimal protein-protein binding and/or reduce non-specific or background interactions. Reagents that improve the efficiency of the assay, such as protease inhibitors, nuclease inhibitors, anti-microbial agents, etc., may be used. The components are added in any order that produces the requisite binding. Incubations are performed at any temperature that facilitates optimal activity, typically between 4° and 40°C Incubation periods are selected for optimum activity, but may also be optimized to facilitate rapid high-throughput screening. Normally, between 0.1 and 1 h will be sufficient. In general, a plurality of assay mixtures is run in parallel with different agent concentrations to obtain a differential response to these concentrations. Typically, one of these concentrations serves as a negative control, i.e., at zero concentration or below the level of detection.

To perform cell-free ligand screening assays, it may be desirable to immobilize either a DGI-3, DGI-3-binder, or DGI-3-partner polypeptide, polynucleotide, fragment, or polypeptide-complex, to a surface to facilitate identification of ligands that bind to these molecules, as well as to accommodate automation of the assay. For example, a fusion protein comprising a DGI-3 polypeptide and an affinity tag can be produced. In one embodiment, a glutathione-S-transferase/phosphodiesterase fusion protein comprising a DGI-3 polypeptide is adsorbed onto glutathione sepharose beads (Sigma Chemical, St. Louis, MO) or glutathione-derivatized microtiter plates. Cell lysates (e.g., containing ³⁵S-labeled polypeptides) are added to the coated beads under conditions to allow complex formation (e.g., at physiological conditions for salt and pH). Following incubation, the coated beads are washed to remove any unbound polypeptides, and the amount of immobilized radiolabel is determined. Alternatively, the complex is dissociated and the radiolabel present in the supernatant is determined. In another approach, the beads are analyzed by SDS-PAGE to identify the bound polypeptides.

Ligand-binding assays can be used to identify agonist or antagonists that alter the function or levels of a DGI-3 or DGI-3-partner polypeptide. Such assays are designed to detect the interaction of test agents (e.g., small molecules) with DGI-3, DGI-3-binder, or DGI-3-partner polypeptides, polynucleotides, polypeptide-complexes, or fragments or portions thereof. Interactions may be detected by direct measurement of binding. Alternatively, interactions may be detected by indirect indicators of binding, such as stabilization/destabilization of protein structure, or activation/inhibition of biological function. Non-limiting examples of useful ligand-binding assays are detailed below.

Ligands that bind to DGI-3, DGI-3-binder, or DGI-3-partner polypeptides, polynucleotides, polypeptide-complexes, or fragments or portions thereof, can be identified using real-time Bimolecular Interaction Analysis (BIA; Sjolander et al., 1991 , Anal. Chem. 63:2338-2345; Szabo et al., 1995, Curr. Opin. Struct. Biol. 5:699-705). BIA-based technology (e.g., BIAcore™; LKB Pharmacia, Sweden) allows study of biospecific interactions in real time, without labeling. In BIA, changes in the optical phenomenon surface plasmon resonance (SPR) is used determine real-time interactions of biological molecules.

Ligands can also be identified by scintillation proximity assays (SPA, described in U.S. Patent No. 4,568,649). In a modification of this assay that is currently undergoing development, chaperonins are used to distinguish folded and unfolded proteins. A tagged protein is attached to SPA beads, and test agents are added. The bead is then subjected to mild denaturing conditions (such as, e.g., heat, exposure to SDS, etc.) and a purified labeled chaperonin is added. If a test agent binds to a target, the labeled chaperonin will not bind; conversely, if no test agent binds, the protein will undergo some degree of denaturation and the chaperonin will bind.

Ligands can also be identified using a binding assay based on mitochondrial targeting signals (Hurt et al., 1985, EMBO J. 4:2061-2068; Eilers and Schatz, 1986, Nature 322:228-231 ). In a mitochondrial import assay, expression vectors are constructed in which nucleic acids encoding particular target proteins are inserted downstream of sequences encoding mitochondrial import signals. The chimeric proteins are synthesized and tested for their ability to be imported into isolated mitochondria in the absence and presence of test compounds. A test compound that binds to the target protein should inhibit its uptake into isolated mitochondria in vitro. The ligand-binding assay described in Fodor et al., 1991 , Science 251 :767-773, which involves testing the binding affinity of test compounds for a plurality of defined polymers synthesized on a solid substrate, can also be used. Ligands that bind to DGI-3, DGI-3-binder, or DGI-3-partner polypeptides or peptides can be identified using two-hybrid assays (see, e.g., U.S. Pat. No. 5,283,317; Zervos et al., 1993, Cell 72:223-232; Madura et al., 1993, J. Biol. Chem. 268:12046-12054; Bartel et al., 1993, Biotechniques 14:920-924; Iwabuchi et al., 1993, Oncogene 8:1693-1696; and Brent WO 94/10300). The two-hybrid system relies on the reconstitution of transcription activation activity by association of the DNA- binding and transcription activation domains of a transcriptional activator through protein-protein interaction. The yeast GAL4 transcriptional activator may be used in this way, although other transcription factors have been used and are well known in the art. To carry out the two-hybrid assay, the GAL4 DNA-binding domain, and the GAL4 transcription activation domain are expressed, separately, as fusions to potential interacting polypeptides.

In one embodiment, the "bait" protein comprises a DGI-3 polypeptide fused to the GAL4 DNA-binding domain. The "fish" protein comprises, for example, a human cDNA library encoded polypeptide fused to the GAL4 transcription activation domain. If the two, coexpressed fusion proteins interact in the nucleus of a host cell, a reporter gene (e.g., LacZ) is activated to produce a detectable phenotype. The host cells that show two-hybrid interactions can be used to isolate the containing plasmids containing the cDNA library sequences. These plasmids can be analyzed to determine the nucleic acid sequence and predicted polypeptide sequence of the candidate ligand. Alternatively, methods such as the three-hybrid (Licitra et al., 1996, Proc. Natl. Acad. Sci. USA 93:12817-12821 ), and reverse two-hybrid (Vidal et al., 1996, Proc. Natl. Acad. Sci. USA 93:10315-10320) systems may be used. Commercially available two-hybrid systems such as the CLONTECH Matchmaker™ systems and protocols (CLONTECH Laboratories, Inc., Palo Alto, CA) may be also be used (see also, A.R. Mendelsohn et al., 1994, Curr. Op. Biotech. 5:482; E.M. Phizicky et al., 1995, Microbiological Rev. 59:94; M. Yang et al., 1995, Nucleic Acids Res. 23:1152; S. Fields et al., 1994, Trends Genet. 10:286; and U.S. Patent No. 6,283,173 and 5,468,614).

Several methods of automated assays have been developed in recent years so as to permit screening of tens of thousands of test agents in a short period of time. High-throughput screening methods are particularly preferred for use with this invention. The ligand-binding assays described herein can be adapted for high-throughput screens, or alternative screens may be employed. For example, continuous format high throughput screens (CF-HTS) using at least one porous matrix allows the researcher to test large numbers of test agents for a wide range of biological or biochemical activity (see United States Patent No. 5,976,813 to Beutel et al.). Moreover, CF-HTS can be used to perform multi-step assays. Other useful screening assays are based on those disclosed in

International application WO 96/04557, which is incorporated herein in its entirety. Briefly, WO 96/04557 discloses the use of reporter peptides that bind to active sites on targets and possess agonist or antagonist activity at the target. These reporters are identified from recombinant libraries and are either peptides with random amino acid sequences or variable antibody regions with at least one CDR region that has been randomized (rVab). The reporter peptides may be expressed in cell recombinant expression systems, such as for example in E. coli, or by phage display (see WO 96/04557 and Kay et al. 1996, Mol. Divers. 1(2): 139-40, both of which are incorporated herein by reference). The reporters identified from the libraries may then be used in accordance with this invention either as therapeutics themselves, or in competition binding assays to screen for other molecules, preferably small, active molecules, which possess similar properties to the reporters and may be developed as drug candidates to provide agonist or antagonist activity. Preferably, these small organic molecules are orally active.

Additionally, an in vitro competitive receptor binding assay can be used as the basis of a heterogeneous screen for small organic molecular replacements for DGI-3-binders. Occupation of the active site of DGI-3 can be quantified by time-resolved fluorometric detection (TRFD) with streptavidin-labeled europium (saEu) complexed to biotinylated peptides (bP). In this assay, saEu can form a ternary complex with bP and DGI-3 (i.e., DGI-3:bP:saEu complex). The TRFD assay format is well-established, sensitive, and quantitative (Tompkins et al., 1993, J. Immunol. Methods 163:209-216). The assay can use a single-chain antibody or a biotinylated peptide. In such assays, soluble DGI-3 is coated on the surface of microtiter wells, blocked by a solution of 0.5% bovine serum albumin (BSA) and 2% non-fat milk in PBS, and then incubated with biotinylated peptide or rVab. Unbound bP is then washed away and saEu is added to complex with receptor-bound bP. Upon addition of the acidic enhancement solution, the bound europium is released as free Eu³⁺ which rapidly forms a highly fluorescent and stable complex with components of the enhancement solution. The DGI-3:bP bound saEu is then converted into its highly fluorescent state and detected by a detector such as Wallac Victor II (EG&G Wallac, Inc.)

Phage display libraries can also be screened for ligands that bind to DGI-3, as described above. Details of the construction and analyses of these libraries, as well as the basic procedures for biopanning and selection of binders, have been published (see, e.g., WO 96/04557; Mandecki et al., 1997, Display Technologies - Novel Targets and Strategies, P. Guttry (ed), International Business Communications, Inc. Southborogh, MA, pp. 231- 254; Ravera et al., 1998, Oncogene 16:1993-1999; Scott and Smith, 1990, Science 249:386-390); Grihalde et al., 1995, Gene 166:187-195; Chen et al., 1996, Proc. Natl. Acad. Sci. USA 93:1997-2001 ; Kay et al., 1993, Gene 128:59-65; Carcamo et al., 1998, Proc. Natl. Acad. Sci. USA 95:11146- 11151 ; Hoogenboom, 1997, Trends Biotechnol. 15:62-70; Rader and Barbas, 1997, Curr. Opin. Biotechnol. 8:503-508; all of which are incorporated herein by reference).

The designing of mimetics to a known pharmaceutically active compound is a known approach to the development of pharmaceuticals based on a "lead" compound. This might be desirable where the active compound is difficult or expensive to synthesize or where it is unsuitable for a particular method of administration, e.g., peptides are generally unsuitable active agents for oral compositions as they tend to be quickly degraded by proteases in the alimentary canal. Mimetic design, synthesis, and testing are generally used to avoid large-scale screening of molecules for a target property.

There are several steps commonly taken in the design of a mimetic from a compound having a given target property. First, the particular parts of the compound that are critical and/or important in determining the target property are determined. In the case of a peptide, this can be done by systematically varying the amino acid residues in the peptide (e.g., by substituting each residue in turn). These parts or residues constituting the active region of the compound are known as its "pharmacophore". Once the pharmacophore has been found, its structure is modeled according to its physical properties (e.g., stereochemistry, bonding, size, and/or charge), using data from a range of sources (e.g., spectroscopic techniques, X-ray diffraction data, and NMR). Computational analysis, similarity mapping (which models the charge and/or volume of a pharmacophore, rather than the bonding between atoms), and other techniques can be used in this modeling process.

In a variant of this approach, the three dimensional structure of the ligand and its binding partner are modeled. This can be especially useful where the ligand and/or binding partner change conformation on binding, allowing the model to take account of this in the design of the mimetic.

A template molecule is then selected, and chemical groups that mimic the pharmacophore can be grafted onto the template. The template molecule and the chemical groups grafted on to it can conveniently be selected so that the mimetic is easy to synthesize, is likely to be pharmacologically acceptable, does not degrade in vivo, and retains the biological activity of the lead compound. The mimetics found are then screened to ascertain the extent they exhibit the target property, or to what extent they inhibit it. Further optimization or modification can then be carried out to arrive at one or more final mimetics for in vivo or clinical testing. DIAGNOSTICS

As discussed herein below, DGI-3 and DGI-3-partners are associated with various cellular processes, including cell growth, proliferation, and attachment. Defects in these processes can lead to the development of neoplasms, including metastatic cancers. Non-limiting examples of cancers include bone cancer, brain cancer, breast cancer, endocrine system cancers, gastrointestinal cancers (e.g., rectal, colorectal and pancreatic cancers) male genituorinary cancers (e.g., prostate cancer), germ cell cancers, gynecologic cancers (e.g., ovarian, cervical, endometrial, and vulvar cancers), head and neck cancers, leukemia, lung cancer, lymphomas (e.g., Hodgkin's and non-Hodgkin's lymphomas), skin cancers, and urinary cancers (e.g., bladder and kidney cancers). Metastatic cancers include, but are not limited to, cancers affecting bone, breast, lung, brain, spinal cord, skin, ovaries, bladder, and gastrointestinal tissues. Other specific cancers are described below. This invention therefore provides compositions (e.g., diagnostic reagents) comprising DGI-3, DGI-3-binder, or DGI-3-partner polynucleotides (e.g., SEQ ID NO:18; SEQ ID NO:28-SEQ ID NO:36), polypeptides or peptides (e.g., SEQ ID NO:3-SEQ ID NO:11 ; SEQ ID NO:19-SEQ ID NO:27; SEQ ID NO:58-SEQ ID NO:109; SEQ ID NO:110- SEQ ID NO:112; and SEQ ID NO:114-SEQ ID NO:117), antibodies, and fragments thereof that can be useful in diagnosing and monitoring the treatment of these conditions.

Antibody-based diagnostics: In one embodiment of this invention, antibodies which specifically bind to a DGI-3, DGI-3-binder, or DGI-3-partner polypeptide or polypeptide complex may be used for the diagnosis of conditions or diseases characterized by altered levels of DGI-3 or DGI-3- partner polypeptides or polypeptide complexes. Alternatively, such antibodies may be used in assays to monitor patients being treated with a DGI-3, DGI-3-binder, or DGI-3-partner polypeptide, polynucleotide, antibody, or modulator. The antibodies useful for diagnostic purposes may be prepared in the same manner as those for use in therapeutic methods, described herein. Antibodies may be raised to a full-length DGI-3 or DGI-3-partner polypeptide. Alternatively, the antibodies may be raised to portions or variants of these polypeptides. In one aspect of the invention, antibodies are prepared to bind to a DGI-3-partner polypeptide fragment comprising one or more domains of the polypeptide (e.g., PH, GEF, or kinase domains), as described in detail herein.

Diagnostic assays for a DGI-3 or DGI-3-partner polypeptide include methods that utilize the antibody and a label to detect the protein in biological samples (e.g., human body fluids, cells, tissues, or extracts of cells or tissues). The antibodies may be used with or without modification, and may be labeled by joining them, either covalently or non-covalently, with a reporter molecule. A wide variety of reporter molecules that are known in the art may be used, several of which are described herein.

Many immunoassay formats are known in the art, and the particular format used is determined by the desired application. An immunoassay can use, for example, a monoclonal antibody directed against a single disease- associated epitope, a combination of monoclonal antibodies directed against different epitopes of a single disease-associated antigenic component, monoclonal antibodies directed towards epitopes of different disease-associated antigens, polyclonal antibodies directed towards the same disease-associated antigen, or polyclonal antibodies directed towards different disease-associated antigens. Protocols can also, for example, use solid supports, or may involve immunoprecipitation. Typically, immunoassays use either a labeled antibody or a labeled antigenic component (i.e., to compete with the antigen in the sample for binding to the antibody). Exemplary labels are described in the sections shown above.

In accordance with this invention, "competitive" (U.S. Pat. Nos. 3,654,090 and 3,850,752), "sandwich" (U.S. Pat. No. 4,016,043), and "double antibody," or "DASP" assays may be used. Several procedures for measuring the amount of a DGI-3 or DGI-3-partner polypeptide in a sample (e.g., ELISA, RIA, and FACS) are known in the art and provide a basis for diagnosing altered or abnormal levels of polypeptides or polypeptide complexes. Normal or standard values for a polypeptide or polypeptide complex are established by incubating biological samples taken from normal subjects, preferably human, with antibody to a polypeptide or polypeptide complex under conditions suitable for association. The amount of standard antibody-antigen association may be quantified by various methods; photometric means are preferred. Levels of the polypeptide or polypeptide complex in the subject sample, negative control (normal) sample, and positive control (disease) sample are compared with the standard values. Deviation between standard and subject values establishes the parameters for diagnosing disease.

Kits suitable for antibody-based diagnostic applications typically include one or more of the following components:

(1 ) Antibodies: The antibodies may be pre-labeled. Alternatively, the antibody may be unlabeled and the ingredients for labeling may be included in the kit in separate containers, or a secondary, labeled antibody is provided. Antibodies may be monoclonal or polyclonal, and may be directed to DGI-3, DGI-3-binders, DGI-3-partners, or polypeptide complexes thereof; and

(2) Reaction components: The kit may also contain other suitably packaged reagents and materials needed for the particular immunoassay protocol, including solid-phase matrices, if applicable, and standards. The kits referred to above may include instructions for conducting the test. Furthermore, in preferred embodiments, the diagnostic kits are adaptable to high-throughput and/or automated operation.

Nucleic acid-based diagnostics: The invention also provides methods for detecting altered levels or sequences of DGI-3 or DGI-3-partner nucleic acids in a sample, such as in a biological sample (e.g., human body fluids, cells, tissues, or extracts of cells or tissues). The presence of DGI-3 or DGI-3-partner polynucleotide sequences can be detected by DNA-DNA or DNA-RNA hybridization, or by amplification, using probes or primers comprising at least a portion of a DGI-3 or DGI-3-partner polynucleotide, or a sequence complementary thereto. In particular, nucleic acid amplification- based assays can use DGI-3 or DGI-3-partner oligonucleotides or oligomers (i.e., primers) to detect cells (e.g., host cells or biological sample cells) containing DGI-3 or DGI-3-partner DNA or RNA. Preferably, DGI-3 and DGI-3-partner nucleic acids useful as probes or primers in diagnostic methods include oligonucleotides at least 8 contiguous nucleotides in length, more preferably at least 10, 15, 20, 25, 30, 40, or 50 contiguous nucleotides in length, that hybridize specifically with DGI-3 or DGI-3-partner nucleic acids. Probes can be DNA or RNA and preferably contain at least 50%, preferably at least 80%, identity to a DGI-3, DGI-3-binder, or DGI-3- partner polynucleotide, or a complementary sequence, or a fragment thereof.

Several methods can be used to produce specific probes for DGI-3 or DGI-3-partner polynucleotides. For example, labeled probes can be produced by oligo-labeling, nick translation, end-labeling, or PCR amplification using a labeled nucleotide. Alternatively, DGI-3 or DGI-3- partner polynucleotide sequences, or any portions or fragments thereof, may be cloned into a vector for the production of an mRNA probe. Such vectors are known in the art, are commercially available, and may be used to synthesize RNA probes in vitro by addition of an appropriate RNA polymerase, such as T7, T3, or SP(6) and labeled nucleotides. These procedures may be conducted using a variety of commercially available kits (e.g., from Amersham-Pharmacia; Promega Corp.; and U.S. Biochemical Corp., Cleveland, OH). Suitable reporter molecules or labels which may be used include radionucleotides (e.g., ³²P, ³H, and ³⁵S), enzymes, fluorescent (e.g., rhodamine, fluorescein, and Cy™3, Cy™5), chemiluminescent, or chromogenic agents, and other labels (e.g., DNP, digoxigenin, and biotin) such as substrates, cofactors, inhibitors, magnetic particles, and the like. A sample to be analyzed, such as, for example, a tissue sample (e.g., hair or buccal cavity) or body fluid sample (e.g., blood, saliva, or urine), may be contacted directly with the nucleic acid probes. Alternatively, the sample may be treated to extract the nucleic acids contained therein. It will be understood that the particular method used to extract DNA will depend on the nature of the biological sample. The resulting nucleic acid from the sample may be subjected to gel electrophoresis or other size separation techniques, or, the nucleic acid sample may be immobilized on an appropriate solid matrix without size separation. In accordance with this invention, diagnostic assays may be used to distinguish between the absence, presence, increase, and decrease of DGI- 3 or DGI-3-partner mRNA levels, and to monitor regulation of polynucleotide levels during therapeutic treatment or intervention. For example, DGI-3 or DGI-3-partner polynucleotide sequences, or fragments, or complementary sequences thereof, can be used in Southern or Northern analysis, dot blot, or other membrane-based technologies; in PCR technologies; or in dip stick, pin, ELISA or biochip assays utilizing fluids or tissues from patient biopsies to detect the status of, e.g., levels or overexpression of DGI-3 or DGI-3- partner genes, or to detect altered gene expression. Such qualitative or quantitative methods are well known in the art (G.H. Keller and M.M. Manak, 1993, DNA Probes, 2^nd Ed, Macmillan Publishers Ltd., England; D.W. Dieffenbach and G. S. Dveksler, 1995, PCR Primer: A Laboratory Manual, Cold Spring Harbor Press, Plainview, NY; B.D. Hames and S.J. Higgins, 1985, Gene Probes 1, 2, IRL Press at Oxford University Press, Oxford, England).

Methods suitable for quantifying the expression of DGI-3 or DGI-3- partner genes include radiolabeling or biotinylating nucleotides, co- amplification of a control nucleic acid, and standard curves onto which the experimental results are interpolated (P.C Melby et al., 1993, J. Immunol. Methods 159:235-244; and C. Duplaa et al., 1993, Anal. Biochem. 212(1 ):229-36). The speed of quantifying multiple samples may be accelerated by running the assay in an ELISA format where the oligomer of interest is presented in various dilutions and a spectrophotometric or colorimetric response gives rapid quantification. In accordance with established methods, microarray formats (i.e., DNA chips or biochips) may also be used (see, e.g., D.A. Rew DA, 2001 , Eur. J. Surg. Oncol. 27(5):504- 8; P.J. Planet et al., 2001 , Genome Res. 11 (7): 1149-55; J. Quackenbush, 2001 , Nat. Rev. Genet. 2(6):418-27; O.P. Kallioniemi et al., 2001 , Hum. Mol. Genet. 10(7):657-62; U.S. Patent No. 6,015,702 to P. Lai et al.; M. Schena (Ed.), 2000, Microarray Biochip Technology, Eaton Publishing). Kits suitable for nucleic acid-based diagnostic applications typically include the following components:

(1 ) Probe DNA: The probe DNA may be prelabeled. Alternatively, the probe DNA may be unlabeled and the ingredients for labeling may be included in the kit in separate containers. Probes may hybridize to DGI-3 or DGI-3-partner nucleic acids; and

(2) Hybridization reagents: The kit may also contain other suitably packaged reagents and materials needed for the particular hybridization protocol, including solid-phase matrices, if applicable, and standards. THERAPEUTICS As indicated herein below, DGI-3, DGI-3 binders, and DGI-3-partners are associated with various cell processes, including cell growth, proliferation, and adhesion. Uncontrolled activation of cell growth and proliferation can lead to neoplastic growth and/or oncogenesis, whereas inhibition of cell attachment can lead to metastasis. Accordingly, the method of this invention encompasses the use of DGI-3, DGI-3-binders, DGI-3-partners, or combinations thereof to alter cell functions so that one or more of the processes associated with neoplastic growth or oncogenesis, including metastatic growth, is reduced. Non-limiting examples of cancers for which the methods and compositions of this invention may be used to provide a therapeutic benefit, include bone cancer, brain cancer, breast cancer, endocrine system cancers, gastrointestinal cancer (e.g., rectal, colorectal and pancreatic cancer) male genituorinary cancer (e.g., prostate cancer), germ cell cancer, gynecologic cancer (e.g., ovarian, cervical, endometrial, and vulvar cancer), head and neck cancer, leukemia, lung cancer, lymphoma (e.g., Hodgkin's and non-Hodgkin's lymphoma), skin cancer, and urinary cancers (e.g., bladder and kidney cancers). Metastatic cancers include, but are not limited to, cancers affecting bone, breast, lung, brain, spinal cord, skin, ovaries, bladder, and gastrointestinal tissues. Cancers may include carcinomas, sarcomas, melanomas, lymphomas, leukemias, myelomas, nerve cell tumors, and germ cell tumors. Specific breast cancers include, but are not limited to, in situ carcinomas, ductal carcinomas in situ (DCIS), lobular carcinomas in situ (LCIS), or lobular neoplasias, invasive ductal and lobular tumors, Paget's Disease, and cystosarcoma phyllodes. Colon cancers and related conditions include, but are not limited to, adenocarcinomas, tubular adenomas, tubulovillous adenomas (villoglandular polyps), villous (papillary) adenomas (with or without adenocarcinoma), hamartomas, polypoid carcinomas, pseudopolyps, lipomas, leiomyomas, familial polyposis, Gardner's syndrome, and Peutz-Jeghers syndrome. Pancreatic cancers include, but are not limited to, ductal adenocarcinomas, cystadenocarcinomas, intraductal papillary-mucinous tumors, insulinomas, Zollinger-Ellison syndrome (Z-E Syndrome; gastrinoma), vipomas, and glucagonomas. Specific skin cancers include, but are not limited to, basal cell carcinomas, squamous cell carcinomas, and melanomas. Non-limiting endometrial cancers include adenocarcinomas, sarcomas, mesodermal tumors, leiomyosarcomas, and endometrial stromal sarcomas. Specific prostate cancers include, but are not limited to, adenocarcinomas and sarcomas, or pre-cancerous conditions, such as prostate intraepithelial neoplasia (PIN). Specific lung cancers include, but are not limited to, cancers relating to tumors such as bronchial carcinoid (bronchial adenoma), chondromatous hamartoma (benign), solitary lymphoma, and sarcoma (malignant) tumors, as well as cancers relating to multifocal lymphomas. In addition, bronchogenic carcinomas may present as squamous cell carcinomas, small cell carcinomas, non-small cell carcinomas, or adenocarcinomas.

This invention therefore provides compositions (e.g., pharmaceutical compositions, see infra) comprising DGI-3, DGI-3-binder, or DGI-3-partner polynucleotides (e.g., SEQ ID NO:18; SEQ ID NO:28-SEQ ID NO:36), polypeptides or peptides (e.g., SEQ ID NO:3-SEQ ID NO:11 ; SEQ ID NO:19-SEQ ID NO:27; SEQ ID NO:58-SEQ ID NO:109; SEQ ID NO:110- 112; and SEQ ID NO:114-SEQ ID NO:117), antibodies, ligands, variants, or fragments thereof that can be useful in treating individuals with these conditions. Also provided are methods employing DGI-3, DGI-3-binder, or DGI partner polynucleotides, polypeptides, peptides, antibodies, ligands, or variants, portions, or fragments to identify drug candidates that can be used to prevent, treat, or ameliorate such disorders. Drug screening and design: This invention provides methods of screening for drugs using DGI-3, DGI-3-binder, or DGI-3-partner polypeptides, peptides, or polypeptide-complexes thereof, in competitive binding assays, according to methods well-known in the art. For example, competitive drug screening assays can be employed using a complex comprising DGI-3 and a DGI-3-binder, and screening for a test compound that disrupts, enhances, or otherwise alters the polypeptide-complex.

This invention further provides methods of rational drug design employing a DGI-3, DGI-3-binder, or DGI-3-partner polynucleotide, polypeptide, antibody, polypeptide-complex, or portion or functional equivalent thereof. The goal of rational drug design is to produce structural analogs of biologically active polypeptides of interest or of small molecules with which they interact (e.g., agonists, antagonists, or inhibitors). In turn, these analogs can be used to fashion drugs which are, for example, more active or stable forms of the polypeptide, or which, e.g., enhance or interfere with the function of the polypeptide in vivo (see, e.g., Hodgson, 1991 , Bio/Technology, 9:19-21 ). An example of rational drug design is the development of HIV protease inhibitors (Erickson et al., 1990, Science, 249:527-533).

In one approach, one first determines the three-dimensional structure of a DGI-3, DGI-3-binder, or DGI-3-partner polypeptide, peptide, or polypeptide-complex, by x-ray crystallography, computer modeling, or a combination thereof. Useful information regarding the structure of a polypeptide can also be gained by computer modeling based on the structure of homologous or orthologous proteins. In the case of cytohesin-1 , the three-dimensional structure has been determined for the Sec7 domain (S.F. Betz et al., 1998, Proc. Natl. Acad. Sci. USA 95(14):7909-14). The determined structures of DGI-3, DGI-binders, or DGI-3-partners can be used to devise modulators (i.e., see Szardenings et al, 1998).

In addition, DGI-3, DGI-3-binder, or DGI-3-partner polypeptides, or portions thereof, can be analyzed by alanine scans (Wells, 1991 , Methods in Enzymol., 202:390-411 ). In this technique, each amino acid residue in a DGI-3, DGI-3-binder, or DGI-3-partner polypeptide is replaced by alanine, and its effect on the activity of the polypeptide is determined. In another approach, an antibody specific to a DGI-3, DGI-3-binder, or DGI-3-partner polypeptide, peptide, or polypeptide-complex can be isolated, selected by a functional assay, and then analyzed to solve its crystal structure. In principle, this approach can yield a pharmacore upon which subsequent drug design can be based.

Alternatively, it is possible to bypass protein crystallography altogether by generating anti-idiotypic antibodies (anti-ids) to a functional, pharmacologically active antibody. As a mirror image of a mirror image, the binding site of the anti-ids is predicted to be an analog of the corresponding DGI-3, DGI-3-binder, or DGI-3-partner polypeptide, peptide, or polypeptide- complex. The anti-id can then be used to identify and isolate peptides from banks of chemically or biologically produced banks of peptides. Selected peptides can subsequently be used as pharmacores. Non-limiting examples of methods and computer tools for drug design are described in R. Cramer et al., 1974, J. Med. Chem. 17:533; H. Kubinyi (ed) 1993, 3D QSAR in Drug Design, Theory, Methods, and Applications, ESCOM, Leiden, Holland; P. Dean (ed) 1995, Molecular Similarity in Drug Design, K. Kim "Comparative molecular field analysis (ComFA)" p. 291-324, Chapman & Hill, London, UK; Y. et al., 1993, J. Comp.-Aid. Mol. Des. 7:83-102; G. Lauri and P.A. Bartlett, 1994, J. Comp.- Aid. Mol. Des. 8:51-66; P.J. Gane and P.M. Dean, 2000, Curr. Opin. Struct. Biol. 10(4):401-4; H.O. Kim and M. Kahn, 2000, Comb. Chem. High Throughput Screen. 3(3): 167-83; G.K. Farber, 1999, Pharmacol Ther. 84(3):327-32; and H. van de Waterbeemd (Ed.) 1996, Structure-Property Correlations in Drug Research, Academic Press, San Diego, CA.

In another aspect of this invention, cells and animals that carry a human DGI-3 or DGI-3-partner gene or an variant thereof can be used as model systems to study and test for substances that have potential as therapeutic agents. After a test agent is administered to animals or applied to the cells, the phenotype of the animals/cells can be determined.

In accordance with these methods, one may design drugs that result in, for example, altered DGI-3 or DGI-3-partner activity or stability. Such drugs may act as inhibitors, agonists, or antagonists of these polypeptides, or the complexes formed by these peptides. By virtue of the availability of DGI-3, DGI-3-binder, and DGI-3-partner nucleotide sequences, sufficient amounts of these polypeptides may be produced to perform such analytical studies as x-ray crystallography. In addition, the knowledge of the DGI-3 and DGI-3-partner polypeptide sequences will guide those employing computer-modeling techniques in place of, or in addition to x-ray crystallography.

Pharmaceutical compositions: This invention contemplates compositions comprising one or more DGI-3, DGI-3-binder, or DGI-3-partner polynucleotide, polypeptide, peptide, antibody, ligand (e.g., agonist, antagonist, or inhibitor), polypeptide-complex, or fragments or variants thereof, and a physiologically acceptable carrier, excipient, or diluent. This invention further contemplates pharmaceutical compositions useful in practicing the therapeutic methods of this invention. Preferably, a pharmaceutical composition includes, in admixture, a pharmaceutically acceptable excipient (carrier) and one or more of a DGI-3, DGI-3-binder, or DGI-3-partner polynucleotide, polypeptide, peptide, ligand, antibody, polypeptide-complex, or fragment, portion, or variant thereof, as described herein, as an active ingredient. Notably, the rapidity and ease of peptide synthesis and the avoidance of gene interference has made peptide-based therapeutics (e.g., cancer therapeutics) advantageous in many cases (see, e.g., P.W. Latham, 1999, Nat. Biotechnol. 17:755-757).

The preparation of pharmaceutical compositions that contain DGI-3, DGI-3-binder, or DGI-3-partner molecules as active ingredients is well understood in the art. Typically, such compositions are prepared as injectables, either as liquid solutions or suspensions, however, solid forms suitable for solution in, or suspension in, liquid prior to injection can also be prepared. The preparation can also be emulsified. The active therapeutic ingredient is often mixed with excipients that are pharmaceutically acceptable and compatible with the active ingredient. Suitable excipients are, for example, water, saline, dextrose, glycerol, ethanol, or the like and combinations thereof. In addition, if desired, the composition can contain minor amounts of auxiliary substances such as wetting or emulsifying agents, pH-buffering agents, which enhance the effectiveness of the active ingredient. A DGI-3, DGI-3-binder, or DGI-3-partner polynucleotide, polypeptide, peptide, ligand, antibody, polypeptide-complex, or fragment, portion, or variant thereof can be formulated into the pharmaceutical composition as neutralized physiologically acceptable salt forms. Suitable salts include the acid addition salts (i.e., formed with the free amino groups of the polypeptide or antibody molecule) and which are formed with inorganic acids such as, for example, hydrochloric or phosphoric acids, or such organic acids as acetic, oxalic, tartaric, mandelic, and the like. Salts formed from the free carboxyl groups can also be derived from inorganic bases such as, for example, sodium, potassium, ammonium, calcium, or ferric hydroxides, and such organic bases as isopropylamine, trimethylamine, 2- ethylamino ethanol, histidine, procaine, and the like.

The pharmaceutical compositions can be administered systemically by oral or parenteral routes. Non-limiting parenteral routes of administration include subcutaneous, intramuscular, intraperitoneal, intravenous, transdermal, inhalation, intranasal, intra-arterial, intrathecal, enteral, sublingual, or rectal. In one particular embodiment of this invention, the disclosed pharmaceutical compositions are administered via mucoactive aerosol therapy (see, e.g., M. Fuloria and B.K. Rubin, 2000, Respir. Care 45:868-873; I. Gonda, 2000, J. Pharm. Sci. 89:940-945; R. Dhand, 2000, Curr. Opin. Pulm. Med. 6(1 ):59-70; B.K. Rubin, 2000, Respir. Care 45(6):684-94; S. Suarez and A.J. Hickey, 2000, Respir. Care. 45(6):652-66). Intravenous administration, for example, can be performed by injection of a unit dose. The term "unit dose" when used in reference to a pharmaceutical composition of this invention refers to physically discrete units suitable as unitary dosage for humans, each unit containing a predetermined quantity of active material calculated to produce the desired therapeutic effect in association with the required diluent; i.e., carrier, or vehicle.

Pharmaceutical compositions are administered in a manner compatible with the dosage formulation, and in a therapeutically effective amount. The quantity to be administered depends on the subject to be treated, capacity of the subject's immune system to utilize the active ingredient, and degree of modulation of DGI-3 or DGI-3-partner activity desired. Precise amounts of active ingredient required to be administered depend on the judgment of the practitioner and are specific for each individual. However, suitable dosages may range from about 0.1 to 20 mg, preferably about 0.5 to about 10 mg, and more preferably one to several, milligrams of active ingredient per kilogram body weight of individual per day and depend on the route of administration. Suitable regimes for initial administration and booster shots are also variable, but are typified by an initial administration followed by repeated doses at one or more hour intervals by a subsequent injection or other administration. Alternatively, continuous intravenous infusions sufficient to maintain concentrations of 10 nM to 10 μM in the blood are contemplated. An exemplary pharmaceutical formulation comprises: DGI-3 antagonist or inhibitor (5.0 mg/ml); sodium bisulfite USP (3.2 mg/ml); disodium edetate USP (0.1 mg/ml); and water for injection q.s.a.d. (1.0 ml). For further guidance in preparing pharmaceutical formulations, see, e.g., Gilman et al. (eds), 1990, Goodman and Gilman's: The Pharmacological Basis of Therapeutics, 8th ed., Pergamon Press; and Remington's Pharmaceutical Sciences, 17th ed., 1990, Mack Publishing Co., Easton, PA; Avis et al. (eds), 1993, Pharmaceutical Dosage Forms: Parenteral Medications, Dekker, New York; Lieberman et al. (eds), 1990, Pharmaceutical Dosage Forms: Disperse Systems, Dekker, New York.

In yet another aspect of this invention, antibodies that specifically react with a DGI-3, DGI-3-binder, or DGI-3-partner polypeptide, peptides, or polypeptide-complexes comprised thereof can be used as therapeutics. In particular, such antibodies can be used to block the activity of a DGI-3, DGI- 3-binder, or DGI-3-partner polypeptide or DGI-3/DGI-3-partner complex. Antibodies or fragments thereof can be formulated as pharmaceutical compositions and administered to a subject. It is noted that antibody-based therapeutics produced from non-human sources can cause an undesired immune response in human subjects. To minimize this problem, chimeric antibody derivatives can be produced. Chimeric antibodies combine a non- human animal variable region with a human constant region. Chimeric antibodies can be constructed according to methods known in the art (see Morrison et al., 1985, Proc. Natl. Acad. Sci. USA 81 :6851 ; Takeda et al., 1985, Nature 314:452; U.S. Patent No. 4,816,567 of Cabilly et al.; U.S. Patent No. 4,816,397 of Boss et al.; European Patent Publication EP 171496; EP 0173494; United Kingdom Patent GB 2177096B).

In addition, antibodies can be further "humanized" by any of the techniques known in the art, (e.g., Teng et al., 1983, Proc. Natl. Acad. Sci. USA 80:7308-7312; Kozbor et al., 1983, Immunology Today 4: 7279; Olsson et al., 1982, Meth. Enzymol. 92:3-16; International Patent Application WO92/06193; EP 0239400; U.S. Patent No. 5,585,089 to Queen et al.). Humanized antibodies can also be obtained from commercial sources (e.g., Scotgen Limited, Middlesex, England). Immunotherapy with a humanized antibody may result in increased long-term effectiveness for the treatment of chronic disease situations or situations requiring repeated antibody treatments.

It should be noted that the DGI-3, DGI-3-binder, or DGI-3-partner polynucleotides, polypeptides, peptides, ligands, antibodies, polypeptide- complexes, or fragments, portions, or variants thereof, may be administered individually, or in combination with other DGI-3 antagonists or inhibitors. Alternatively, the disclosed DGI-related therapeutics can be used in combination with other cancer therapies, e.g., surgery, radiation, biological response modification, immunotherapy, hormone therapy, and/or chemotherapy. In particular, one or more anti-cancer agent can be used in combination with one or more DGI-3-related therapeutic.

Non-limiting examples of anti-cancer agents include aldesleukin (e.g., Proleukin®, IL-2, lnterleukin-2); altretamine (e.g., Hexalen®); asparaginase (e.g., Elspar®, Oncaspar®, Pegaspargase), bleomycin (e.g., Blenoxane®); capecitabine (e.g., Xeloda™); carmustine (e.g., BCNU, BiCNU®, Gliadel®); cladribine (e.g., Leustatin®); platinum compounds such as cisplatin (e.g., Platinol®) and carboplatin (e.g., Paraplatin®, CBDCA); cyclophosphamide (e.g., Cytoxan®, Cytoxan®IV, Neosar®); cytarabine (e.g., Cytosar-U®, Ara- C, Cytosine arabinoside, DepoCyt™);dacarbazine (e.g., DTIC, DTIC- Dome®); actinomycin (e.g., Cosmegen®, actinomycin-D); taxines such as docetaxel (e.g., Taxotere®) and paclitaxel (e.g., Taxol®, Paxene®); doxorubicin (e.g., Adriamycin®, Rubex®, Doxorubicin HCL); liposomal doxorubicin (e.g., Doxil®, Evacet™); estramustine, etoposide (e.g.,. Etopophos®, Toposar®, VP-16, VePesid®); fludarabine (e.g., Fludara®); fluorouracil (e.g., Adrucil®, Efudex®, Fluoroplex®, Fluorouracil IV; 5-FU); gemcitabine (e.g., Gemzar®); hydroxyurea (e.g., Hydrea®); idarubicin (e.g., Idamycin®); ifosfamide (e.g., Ifex®); interferons such as interferon-α (e.g., Alferon N®, Intron A®, Roferon-A®, Wellferon®); lymphocyte-activated killer cells, tumor necrosis factors, topoisomerase inhibitors; irinotecan (e.g., Camptosar®, Cpt-11 HCL); megestrol (e.g., Megace®, Pallace®, Megestrol Acetate); methotrexate (MTX); mitomycin (e.g., Mutamycin®, Mitomycin®- C); mitotane (e.g., Lysodren®); mitoxantrone (e.g., Novantrone®); navelbine, pilocarpine (e.g., Isopto Carpine®, Pilocar®, Salagen®); rituximab (e.g., Rituxan®); tamoxifen (e.g., Nolvadex®); topotecan (e.g., Hycamtin® Topotecan HCl); monoclonal antibodies such as trastuzumab (e.g., Herceptin®); and vinca alkaloids such as vinblastine (e.g., Velban®), vincristine (e.g., Oncovin®, Vincasar®), and vinorelbine tartrate (e.g., Navelbine®).

Antisense Nucleic Acids: A further embodiment of the invention is antisense nucleic acids (e.g., SEQ ID NO:143) or oligonucleotides (e.g., SEQ ID NO:126-SEQ ID NO:131 ) that are complementary, in whole or in part, to a target molecule comprising a sense strand of DGI-3. The DGI-3 target can be DNA, or its RNA counterpart (i.e., wherein thymine (T) is present in DNA and uracil (U) is present in RNA). When introduced into a cell, antisense nucleic acids or oligonucleotides can hybridize to all or a part of the sense strand of DGI-3, thereby inhibiting gene expression or replication.

In a particular embodiment of the invention, an antisense nucleic acid or oligonucleotide is wholly or partially complementary to, and can hybridize with, a target nucleic acid (either DNA or RNA) having the sequence of DGI- 3 (e.g., SEQ ID NO:18, SEQ ID NO:132-SEQ ID NO:133, SEQ ID NO:141- SEQ ID NO:142, SEQ ID NO:166-SEQ ID NO:173, or fragments thereof). For example, an antisense nucleic acid or oligonucleotide comprising 16 contiguous nucleotides can be sufficient to inhibit expression of the DGI-3 protein. Alternatively, an antisense nucleic acid or oligonucleotide can be complementary to 5' or 3' untranslated regions, or can overlap the translation initiation codon (5' untranslated and translated regions) of DGI-3, or its functional equivalent. In another embodiment, the antisense nucleic acid is wholly or partially complementary to, and can hybridize with, a target nucleic acid that encodes a DGI-3 polypeptide.

In addition, oligonucleotides can be constructed which will bind to duplex nucleic acid (i.e., DNA:DNA or DNA:RNA), to form a stable triple helix-containing or triplex nucleic acid. Such triplex oligonucleotides can inhibit transcription and/or expression of a gene encoding DGI-3, or its functional equivalent (M.D. Frank-Kamenetskii and S.M. Mirkin, 1995, Ann. Rev. Biochem. 64:65-95). Triplex oligonucleotides are constructed using the base-pairing rules of triple helix formation and the nucleotide sequence of the gene or mRNA for DGI-3.

The present invention encompasses methods of using oligonucleotides in antisense inhibition of the function of DGI-3. In the context of this invention, the term "antisense oligonucleotide" refers to naturally-occurring species or synthetic species formed from naturally- occurring subunits or their close homologs. The term may also refer to moieties that function similarly to oligonucleotides, but have non-naturally- occurring portions. Thus, oligonucleotides may have altered sugar moieties or inter-sugar linkages. Exemplary among these are phosphorothioate and other sulfur containing species which are known in the art.

In preferred embodiments, at least one of the phosphodiester bonds of the oligonucleotide has been substituted with a structure that functions to enhance the ability of the compositions to penetrate into the region of cells where the RNA whose activity is to be modulated is located. It is preferred that such substitutions comprise phosphorothioate bonds, methyl phosphonate bonds, or short chain alkyl or cycloalkyl structures. In accordance with other preferred embodiments, the phosphodiester bonds are substituted with structures which are, at once, substantially non-ionic and non-chiral, or with structures which are chiral and enantiomerically specific. Persons of ordinary skill in the art will be able to select other linkages for use in the practice of the invention.

Antisense oligonucleotides may also include species that include at least some modified base forms. Thus, purines and pyrimidines other than those normally found in nature may be so employed. Similarly, modifications on the furanosyl portions of the nucleotide subunits may also be effected, as long as the essential tenets of this invention are adhered to. Examples of such modifications are 2'-0-alkyl- and 2'-halogen-substituted nucleotides. Some non-limiting examples of modifications at the 2' position of sugar moieties which are useful in the present invention include OH, SH, SCH₃, F, OCH₃, OCN, 0(CH₂)_n NH₂ and 0(CH₂)_π CH₃, where n is from 1 to about 10. Such oligonucleotides are functionally interchangeable with natural oligonucleotides or synthesized oligonucleotides, which have one or more differences from the natural structure. All such analogs are comprehended by this invention so long as they function effectively to hybridize with DGI-3 DNA or RNA to inhibit the function thereof. The antisense oligonucleotides in accordance with this invention preferably comprise from about 3 to about 50 subunits. It is more preferred that such oligonucleotides and analogs comprise from about 8 to about 25 subunits and still more preferred to have from about 12 to about 20 subunits. As defined herein, a "subunit" is a base and sugar combination suitably bound to adjacent subunits through phosphodiester or other bonds.

Antisense nucleic acids or oligonulcleotides can be produced by standard techniques (see, e.g., Shewmaker et al., U.S. Patent No.

5,107,065. The oligonucleotides used in accordance with this invention may be conveniently and routinely made through the well-known technique of solid phase synthesis. Equipment for such synthesis is available from several vendors, including PE Applied Biosystems (Foster City, CA). Any other means for such synthesis may also be employed, however, the actual synthesis of the oligonucleotides is well within the abilities of the practitioner. It is also will known to prepare other oligonucleotide such as phosphorothioates and alkylated derivatives. The oligonucleotides of this invention are designed to be hybridizable with DGI-3 RNA (e.g., mRNA) or DNA. For example, an oligonucleotide (e.g., DNA oligonucleotide) that hybridizes to DGI-3 mRNA can be used to target the mRNA for RnaseH digestion. Alternatively, an oligonucleotide that hybridizes to the translation initiation site of DGI-3 mRNA can be used to prevent translation of the mRNA. In another approach, oligonucleotides that bind to the double-stranded DNA of DGI-3 can be administered. Such oligonucleotides can form a triplex construct and inhibit the transcription of the DNA encoding DGI-3 polypeptides. Triple helix pairing prevents the double helix from opening sufficiently to allow the binding of polymerases, transcription factors, or regulatory molecules. Recent therapeutic advances using triplex DNA have been described (see, e.g., J.E. Gee et al., 1994, Molecular and Immunologic Approaches, Futura Publishing Co., Mt. Kisco, NY).

As non-limiting examples, antisense oligonucleotides may be targeted to hybridize to the following regions: mRNA cap region; translation initiation site; translational termination site; transcription initiation site; transcription termination site; polyadenylation signal; 3' untranslated region; 5' untranslated region; 5' coding region; mid coding region; and 3' coding region. Preferably, the complementary oligonucleotide is designed to hybridize to the most unique 5' sequence of DGI-3 , including any of about 15-35 nucleotides spanning the 5' coding sequence. Appropriate oligonucleotides can be designed using OLIGO software (Molecular Biology Insights, Inc., Cascade, CO; available online at hyperlink transfer protocol on the world wide web at oligo.net). Non-limiting examples of DGI-3 antisense oligonucleotides include those shown in Figure 26 (SEQ ID NO:126-SEQ ID NO:131 ). In accordance with the present invention, the antisense oligonucleotides can be synthesized, formulated as a pharmaceutical composition, and administered to a subject. The synthesis and utilization of antisense and triplex oligonucleotides have been previously described (e.g., H. Simon et al., 1999, Antisense Nucleic Acid Drug Dev. 9:527-31 ; F.X. Barre et al., 2000, Proc. Natl. Acad. Sci. USA 97:3084-3088; R. Elez et al., 2000, Biochem. Biophys. Res. Commun. 269:352-6; E.R. Sauter et al., 2000, Clin. Cancer Res. 6:654-60). Alternatively, expression vectors derived from retroviruses, polio viruses, adenoviruses, alphaviruses, herpes or vaccinia viruses, or from various bacterial plasmids may be used for delivery of nucleotide sequences to the targeted organ, tissue or cell population. Methods which are well known to those skilled in the art can be used to construct recombinant vectors which will express nucleic acid sequence that is complementary to the nucleic acid sequence encoding a DGI-3 polypeptide. These techniques are described both in Sambrook et al., 1989 and in Ausubel et al., 1992. For example, DGI-3 expression can be inhibited by transforming a cell or tissue with an expression vector that expresses high levels of untranslatable sense or antisense DGI-3 sequences. Even in the absence of integration into the DNA, such vectors may continue to transcribe RNA molecules until they are disabled by endogenous nucleases. Transient expression may last for a month or more with a non-replicating vector, and even longer if appropriate replication elements included in the vector system.

Various assays may be used to test the ability of DGI-3-specific antisense oligonucleotides to inhibit DGI-3 expression. For example, DGI-3 mRNA levels can be assessed northern blot analysis (Sambrook et al., 1989; Ausubel et al., 1992; J.C. Alwine et al. 1977, Proc. Natl. Acad. Sci. USA 74:5350-5354; I.M. Bird, 1998, Methods Mol. Biol. 105:325-36), quantitative or semi-quantitative RT-PCR analysis (see, e.g., W.M. Freeman et al., 1999, Biotechniques 26:112-122; Ren et al., 1998, Mol. Brain Res. 59:256-63; J.M. Cale et al., 1998, Methods Mol. Biol. 105:351-71 ), or in situ hybridization (reviewed by A.K. Raap, 1998, Mutat. Res. 400:287-298). Alternatively, antisense oligonucleotides may be assessed by measuring levels of DGI-3 polypeptide, e.g., by western blot analysis, indirect immunofluorescence, immunoprecipitation techniques (see, e.g., J.M. Walker, 1998, Protein Protocols on CD-ROM, Humana Press, Totowa, NJ). Gene Therapy: DGI-3, DGI-3-binder, and DGI-3-partner polynucleotides of the invention also find use as gene therapy reagents. In recent years, significant technological advances have been made in the area of gene therapy for both genetic and acquired diseases (Kay et al., 1997, Proc. Natl. Acad. Sci. USA, 94:12744-12746). Gene therapy can be defined as the transfer of DNA for therapeutic purposes. Improvement in gene transfer methods has allowed for development of gene therapy protocols for the treatment of diverse types of diseases. Gene therapy has also taken advantage of recent advances in the identification of new therapeutic genes, improvement in both viral and non-viral gene delivery systems, better understanding of gene regulation, and improvement in cell isolation and transplantation. Gene therapy would be carried out according to generally accepted methods as described by, for example, Friedman, 1991 , Therapy for Genetic Diseases, Friedman, Ed., Oxford University Press, pages 105-121.

Gene transfer systems known in the art may be useful in the practice of the gene therapy methods of this invention. These include viral and non- viral transfer methods. A number of viruses have been used as gene transfer vectors, including alphavirus, e.g., sindbis virus (W.F. Cheng et al., 2002, Hum. Gene Ther. 13(4):553-68), polio virus (reviewed in C.J. Bostock et al., 1990, Vet. Microbiol. 23(1-4):55-71 ), polyoma, i.e., SV40 (Madzak et al., 1992, J. Gen. Virol., 73:1533-1536), adenovirus (Berkner, 1992, Curr. Top. Microbiol. Immunol., 158:39-6; Berkner et al., 1988, Bio Techniques, 6:616-629; Gorziglia et al., 1992, J. Virol., 66:4407-4412; Quantin et al., 1992, Proc. Natl. Acad. Sci. USA, 89:2581-2584; Rosenfeld et al., 1992, Cell, 68:143-155; Wilkinson et al., 1992, Nucl. Acids Res., 20:2233-2239; Stratford-Perricaudet et al., 1990, Hum. Gene Ther., 1 :241-256), vaccinia virus (Mackett et al., 1992, Biotechnology, 24:495- 499), adeno-associated virus (Muzyczka, 1992, Curr. Top. Microbiol. Immunol., 158:91- 123; Ohi et al., 1990, Gene, 89:279-282), herpes viruses including HSV and EBV (Margolskee, 1992, Curr. Top. Microbiol. Immunol., 158:67-90; Johnson et al., 1992, J. Virol., 66:2952-2965; Fink et al., 1992, Hum. Gene Ther, 3:11- 19; Breakfield et al., 1987, Mol. Neurobioi, 1 :337-371 ; Fresse et al., 1990, Biochem. Pharmacol., 40:2189-2199), and retroviruses of avian (Brandyopadhyay et al., 1984, Mol. Cell Biol., 4:749-754; Petropouplos et al., 1992, J. Virol., 66:3391-3397), murine (Miller, 1992, Curr. Top. Microbiol. Immunol., 158:1-24; Miller et al., 1985, Mol. Cell Biol., 5:431-437; Sorge et al., 1984, Mol. Cell Biol., 4:1730-1737; Mann et al., 1985, J. Virol., 54:401- 407), and human origin (Page et al., 1990, J. Virol., 64:5370-5276; Buchschalcher et al., 1992, J. Virol., 66:2731-2739). Most human gene therapy protocols have been based on disabled murine retroviruses.

Non-viral gene transfer methods known in the art include chemical techniques such as calcium phosphate coprecipitation (Graham et al., 1973, Virology, 52:456-467; Pellicer et al., 1980, Science, 209:1414-1422), mechanical techniques, for example microinjection (Anderson et al., 1980, Proc. Natl. Acad. Sci. USA, 77:5399-5403; Gordon et al., 1980, Proc. Natl. Acad. Sci. USA, 77:7380-7384; Brinster et al., 1981 , Cell, 27:223-231 ; Constantini et al., 1981 , Nature, 294:92-94), membrane fusion-mediated transfer via liposomes (U.S. Patent No. 5,908,777 to Lee et al.; Feigner et al., 1987, Proc. Natl. Acad. Sci. USA, 84:7413-7417; Wang et al., 1989, Biochemistry, 28:9508-9514; Kaneda et al., 1989, J. Biol. Chem., 264:12126-12129; Stewart et al., 1992, Hum. Gene Ther., 3:267-275; Nabel et al., 1990, Science, 249:1285-1288; Lim et al., 1992, Circulation, 83:2007- 2011 ), and direct DNA uptake and receptor-mediated DNA transfer (Wolff et al., 1990, Science, 247:1465-1468; Wu et al., 1991 , BioTechniques, 11:474- 485; Zenke et al., 1990, Proc. Natl. Acad. Sci. USA, 87:3655-3659; Wu et al., 1989, J. Biol. Chem., 264:16985-16987; Wolff et al., 1991 , BioTechniques, 11 :474-485; Wagner et al., 1991 , Proc. Natl. Acad. Sci. USA, 88:4255-4259; Cotten et al., 1990, Proc. Natl. Acad. Sci. USA, 87:4033-4037; Curiel et al., 1991 , Proc. Natl. Acad. Sci. USA, 88:8850- 8854; Curiel et al., 1991 , Hum. Gene Ther, 3:147-154).

In one approach, plasmid DNA is complexed with a polylysine- conjugated antibody specific to the adenovirus hexon protein, and the resulting complex is bound to an adenovirus vector. The trimolecular complex is then used to infect cells. The adenovirus vector permits efficient binding, intemalization, and degradation of the endosome before the coupled DNA is damaged. In another approach, liposome/DNA is used to mediate direct in vivo gene transfer. While in standard liposome preparations the gene transfer process is non-specific, localized in vivo uptake and expression have been reported in tumor deposits, for example, following direct in situ administration (Nabel, 1992, Hum. Gene Ther. 3:399- 410).

Suitable gene transfer vectors possess a promoter sequence, preferably a promoter that is cell-specific and placed upstream of the sequence to be expressed. The vectors may also contain, optionally, one or more expressible marker genes for expression as an indication of successful transfection and expression of the nucleic acid sequences contained in the vector. In addition, vectors can be optimized to minimize undesired immunogenicity and maximize long-term expression of the desired gene product(s) (see Nabe, 1999, Proc. Natl. Acad. Sci. USA 96:324-326). Moreover, vectors can be chosen based on cell-type that is targeted for treatment. Notably, gene transfer therapies have been initiated for the treatment of various pulmonary diseases (see, e.g., M.J. Welsh, 1999, J. Clin. Invest. 104(9): 1165-6; D.L. Ennist, 1999, Trends Pharmacol. Sci. 20:260-266; S.M. Albelda et al., 2000, Ann. Intern. Med. 132:649-660; E. Alton and C. Kitson C, 2000, Expert Opin. Investig. Drugs. 9(7):1523- 35). Illustrative examples of vehicles or vector constructs for transfection or infection of the host cells include replication-defective viral vectors, DNA virus or RNA virus (retrovirus) vectors, such as adenovirus, herpes simplex virus and adeno-associated viral vectors. Adeno-associated virus vectors are single stranded and allow the efficient delivery of multiple copies of nucleic acid to the cell's nucleus. Preferred are adenovirus vectors. The vectors will normally be substantially free of any prokaryotic DNA and may comprise a number of different functional nucleic acid sequences. An example of such functional sequences may be a DNA region comprising transcriptional and translational initiation and termination regulatory sequences, including promoters (e.g., strong promoters, inducible promoters, and the like) and enhancers which are active in the host cells. Also included as part of the functional sequences is an open reading frame (polynucleotide sequence) encoding a protein of interest. Flanking sequences may also be included for site-directed integration. In some situations, the 5'-flanking sequence will allow homologous recombination, thus changing the nature of the transcriptional initiation region, so as to provide for inducible or non-inducible transcription to increase or decrease the level of transcription, as an example. In general, the encoded and expressed DGI-3, DGI-3-binder, or DGI-

3-partner polypeptide may be intracellular, i.e., retained in the cytoplasm, nucleus, or in an organelle, or may be secreted by the cell. For secretion, the natural signal sequence present in the polypeptide may be retained. When the polypeptide or peptide is a fragment of a full-length protein, a signal sequence may be provided so that, upon secretion and processing at the processing site, the desired protein will have the natural sequence. Specific examples of coding sequences of interest for use in accordance with this invention include the DGI-3 and DGI-3-partner polypeptide-coding sequences shown in the GenBank and GenPept entries described herein. As previously mentioned, a marker may be present for selection of cells containing a vector construct. The marker may be an inducible or non- inducible gene and will generally allow for positive selection under induction, or without induction, respectively. Examples of marker genes include neomycin, dihydrofolate reductase, glutamine synthetase, and the like. The vector employed will generally also include an origin of replication and other genes that are necessary for replication in the host cells, as routinely employed by those having skill in the art. As an example, the replication system comprising the origin of replication and any proteins associated with replication encoded by a particular virus may be included as part of the construct. The replication system must be selected so that the genes encoding products necessary for replication do not ultimately transform the cells. Such replication systems are represented by replication-defective adenovirus (see G. Acsadi et al., 1994, Hum. Mol. Genet. 3:579-584) and by Epstein-Barr virus. Examples of replication defective vectors, particularly, retroviral vectors that are replication defective, are BAG, (see Price et al., 1987, Proc. Natl. Acad. Sci. USA, 84:156; Sanes et al., 1986, EMBO J., 5:3133). It will be understood that the final gene construct may contain one or more genes of interest, for example, a gene encoding a bioactive metabolic molecule. In addition, cDNA, synthetically produced DNA or chromosomal DNA may be employed utilizing methods and protocols known and practiced by those having skill in the art.

According to one approach for gene therapy, a vector encoding a DGI-3 and/or DGI-3-partner polypeptide is directly injected into the recipient cells (in vivo gene therapy). Alternatively, cells from the intended recipients are explanted, genetically modified to encode a DGI-3 and/or DGI-3-partner polypeptide, and reimplanted into the donor (ex vivo gene therapy). An ex vivo approach provides the advantage of efficient viral gene transfer, which is superior to in vivo gene transfer approaches. In accordance with ex wVo gene therapy, the host cells are first transfected with engineered vectors containing at least one gene encoding a DGI-3, DGI-3-binder, or DGI-3- partner polypeptide, suspended in a physiologically acceptable carrier or excipient such as saline or phosphate buffered saline, and the like, and then administered to the host. The desired gene product is expressed by the injected cells, which thus introduce the gene product into the host. The introduced gene products can thereby be utilized to treat or ameliorate a disorder (e.g., neoplastic growth or cancer) that is related to altered levels, activities, and/or interactions of the DGI-3 or DGI-3-partner polypeptides.

Lipid Vehicles: In certain cases, viral or non-viral gene transfer or other therapeutics (e.g., peptides, polypeptides, or small molecules) may be enhanced by the use of lipid vehicles, e.g., liposomes. The principal lipid of the vehicle is, preferably, phosphatidylcholine, but can include various natural (e.g., tissue derived L-α-phosphatidyl: egg yolk, heart, brain, liver, soybean) and/or synthetic (e.g., saturated and unsaturated 1 ,2-diacyl-S/V- glycero-3-phosphocholines, 1-acyl-2-acyl-SΛ/-glycero-3-phosphocholines, 1 ,2-diheptanoyl-SΛ/-glycero-3-phosphocholine) derivatives of the same. Such lipids can be used alone, or in combination with a helper lipid. Preferred helper lipids are non-ionic or uncharged at physiological pH. Particularly preferred non-ionic lipids include, but are not limited to, cholesterol and DOPE (1 ,2-dioleolylglyceryl phosphatidylethanolamine), with cholesterol being most preferred. The molar ratio of a phospholipid to helper lipid can range from about 3:1 to about 1 :1 , more preferably from about 1.5:1 to about 1 :1 , and most preferably, the molar ratio is about 1 :1.

A liposome used for the preparation of a vehicle of the invention is, in simplest form, composed of two lipid layers. The lipid layer may be a monolayer, or may be multilamellar and include multiple layers. Constituents of the liposome may include, for example, phosphatidylcholine, cholesterol, phosphatidylethanolamine, etc. Phosphatidic acid, which imparts an electric charge, may also be added. Exemplary amounts of these constituents used for the production of the liposome include, for instance, 0.3 to 1 mol, preferably 0.4 to 0.6 mol of cholesterol; 0.01 to 0.2 mol, preferably 0.02 to 0.1 mol of phosphatidylethanolamine; 0.0-0.4 mol, preferably 0-0.15 mol of phosphatidic acid per 1 mol of phosphatidylcholine. Liposomes for use with the present invention can be constructed by well-known techniques (see, e.g., G. Gregoriadis (ed.), 1993, Liposome Technology Vols. 1-3, CRC Press, Boca Raton, FL). Lipids are typically dissolved in chloroform and spread in a thin film over the surface of a tube or flask by rotary evaporation. If liposomes comprised of a mixture of lipids are desired, the individual components are mixed in the original chloroform solution. After the organic solvent has been eliminated, a phase consisting of water optionally containing buffer and/or electrolyte is added and the vessel agitated to suspend the lipid. Optionally, the suspension is then subjected to ultrasound, either in an ultrasonic bath or with a probe sonicator, until the particles are reduced in size and the suspension is of the desired clarity. For transfection, the aqueous phase is typically distilled water and the suspension is sonicated until nearly clear, which requires several minutes depending upon conditions, kind, and quality of the sonicator. Commonly, lipid concentrations are 1 mg/ml of aqueous phase, but could be higher or lower by about a factor of ten.

Liposomes can be produced in accordance with established methods. For example, a mixture of the above-mentioned lipids, from which the solvents have been removed, can be emulsified by the use of a homogenizer, lyophilized, and melted to obtain multilamellar liposomes. Alternatively, unilamellar liposomes can be produced by the reverse phase evaporation method (Szoka and Papahadjopoulos, 1978, Proc. Natl. Acad. Sci. USA 75:4194-4198). Unilamellar vesicles can also be prepared by sonication or extrusion. Sonication is generally performed with a bath-type sonifier, such as a Branson tip sonifier (G. Heinemann Ultrashall und Labortechnik, Schwabisch Gmund, Germany) at a controlled temperature as determined by the melting point of the lipid. Extrusion may be carried out by biomembrane extruders, such as the Lipex Biomembrane Extruder (Northern Lipids Inc, Vancouver, British Columbia, Canada). Defined pore size in the extrusion filters may generate unilamellar liposomal vesicles of specific sizes. The liposomes can also be formed by extrusion through an asymmetric ceramic filter, such as a Ceraflow Microfilter (commercially available from the Norton Company, Worcester, MA).

Following liposome preparation, the liposomes that have not been sized during formation may be sized by extrusion to achieve a desired size range and relatively narrow distribution of liposome sizes. A size range of about 0.2-0.4 microns will allow the liposome suspension to be sterilized by filtration through a conventional filter (e.g., a 0.22 micron filter). The filter sterilization method can be carried out on a high throughput basis.

Several techniques are available for sizing liposomes to a desired size, including, ultrasonication, high-speed homogenization, and pressure filtration (M.J. Hope et al., 1985, Biochimica et Biophysica Acta 812:55; U.S. Patent Nos. 4,529,561 and 4,737,323). Sonicating a liposome suspension either by bath or probe sonication produces a progressive size reduction down to small unilamellar vesicles less than about 0.05 microns in size. Multilamellar vesicles can be recirculated through a standard emulsion homogenizer until selected liposome sizes, typically between about 0.1 and 0.5 microns. The size of the liposomal vesicles may be determined by quasi-elastic light scattering (QELS) (see Bloomfield, 1981 , Ann. Rev. Biophys. Bioeng. 10:421-450). Average liposome diameter may be reduced by sonication of formed liposomes. Intermittent sonication cycles may be alternated with QELS assessment to guide efficient liposome synthesis.

Liposomes can be extruded through a small-pore polycarbonate membrane or an asymmetric ceramic membrane to yield a well-defined size distribution. Typically, a suspension is cycled through the membrane one or more times until the desired liposome size distribution is achieved. The liposomes may be extruded through successively smaller-pore membranes, to achieve a gradual reduction in liposome size. For use in the present invention, liposomes have a size of about 0.05 microns to about 0.5 microns. More preferred are liposomes having a size of about 0.05 to about 0.2 microns. Various conditions can be used to trigger the liposome to release its payload or active agent, including pH, ionic strength, controlled release and antibody attachment. Research related to pH-sensitive liposomes has focused principally on anionic liposomes comprised largely of phosphatidylethanolamine (PE) bilayers (see, Huang et al., 1989, Biochemistry 28:9508-9514; Duzgunes et al., 1990, "pH-Sensitive Liposomes" Membrane Fusion J. Wilschut and D. Hoekstra (eds.), Marcel- Decker Inc., New York, NY pp. 713-730; Yatvin et al., 1980, Science, 210, 1253-1255). More recently, pH-sensitive cationic liposomes have been developed to mediate transfer of DNA into cells. For instance, researchers have described a series of amphiphiles with headgroups containing imidazole, methylimidazole, or aminopyridine moieties (see, Budker et al., 1996, Nature Biotech. 14:760-764). Also described are lipid molecules within liposome assemblies that are capable of structural reorganization upon a change in pH (see, e.g., U.S. Patent No. 6,200,599 to Nantz et al.).

From the detailed description herein, it will be clear to those skilled in the art that the lipid vehicles are useful for both in vitro and in vivo applications. The vehicles of the present invention will find use for nearly any in vitro or in vivo application requiring delivery of bioactive agents (e.g., nucleic acids, peptides, polypeptides, or antibodies) and/or drugs (e.g., anticancer treatments) into cells. In certain embodiments of the invention, lipid vehicles such as liposomes can be combined with penetrating peptides (e.g., penetratin, described below, or TAT) to increase efficiency of drug/therapeutic delivery (see, e.g., Y.L. Tseng et al., 2002, Mol. Pharmacol. 62:864-872).

EXAMPLES The examples as set forth herein are meant to exemplify the various aspects of this invention and are not intended to limit the invention in any way. EXAMPLE 1 : IDENTIFICATION AND ISOLATION OF DGI-3

DGI-3 was originally designated as unknown gene KIAA0186 (GenBank Ace. No. NM_021067; Nagase et al., 1996, DNA Res. 3:17-24). BLASTP 2.2.1 analysis (S.F. Altschul et al., 1997, Nucleic Acids Res. 25:3389-3402) indicated that the human DGI-3 protein showed significant homology to DGI-3 orthologs identified from Caenorhabditis elegans ortholog (GenBank Ace. No. Q22019); Mus musculus (GenBank Ace. No. Q9W017); Drosophilia melangogaster (GenBank Ace. No. QPW017); Arabidopis thaliana (GenBank Ace. No. Q9SSC0); Sacharomyces cerevisiae (GenBank Ace. No. Q12488); and Sacharomyces pombe (GenBank Ace. No. Q9P7X6) cells (Figures 12A-12B). The nucleotide and amino acid sequences for DGI-3 are shown in Figures 1A-1B.

An IMAGE clone containing the DGI-3 sequence (IMAGE: 4333095; accession number BF692084) was obtained from the American Type Culture Collection (ATCC, Manassas VA). The clone included 813 bp and encompassed the entire DGI-3 open reading frame (ORF). The 588 bp DGI ORF was amplified using PCR, the amplified sequence was digested with the appropriate enzymes, the digested DNA was ligated into the appropriate expression vectors (see below), and the cloned DNA was sequenced. DGI- 3 was first expressed in E. coli. The DGI-3 ORF was cloned into expression vector pTrcHis A (Invitrogen, Carlsbad, CA) using restriction sites Kpn I and EcoRI, which were added to the DGI-3 by PCR. Protein yields were satisfactory, but most of protein was found in inclusion bodies. Next, DGI-3 was expressed using an in vitro translation system (Rapid Translation System RTS 5000, Roche Diagnostics GmbH, Mannheim Germany). For this system, the DGI-3 ORF was cloned into the plVEX2.3-MCS vector (Roche Diagnostics) using restriction sites Xho I and Kpn I, which were added to DGI-3 by PCR. In this case, the protein was found to be soluble, and was used for panning and ELISA analysis. The primers used for PCR amplification are shown as follows:

DGI 3 Kpn I forward primer for the TrcHis A vector: 5'-gat eta ggt ace atg ttc tgc gaa aaa gcc atg-3' (SEQ ID NO:38);

DGI 3 Eco Rl reverse primer for the TrcHis A vector: 5'-gat eta gaa ttc tga cag gat tgt etc cag gac-3' (SEQ ID NO:39); DGI 3 Xho I reverse primer for the I VEX vector: 5'-gat egg etc gag tga cag gat gtg etc cag gac tcc-3' (SEQ ID NO:40); DGI 3 Nde I forward primer for the I VEX vector: 5'-gat eta cat atg ttc tgc gaa aaa gcc atg gaa c-3' (SEQ ID NO:41 ). EXAMPLE 2: BIOPANNING WITH DGI-3

For these studies, a 20mer random library designated RAPIDLIB20 was used (R.C. Pillutla et al., 2001 , BMC Biotechnology 1 :6-14). One hundred microliters of protein solution diluted to 5 ng/μl was added to an appropriate number of wells in a 96-well microtiter plate (NUNC Maxisorp) and incubated at 4°C overnight. Unbound solution was discarded and 100 μl phage was added to each well. For the first round, the input phage titer was 1 x 10¹³ cfu/ml. For rounds 2, 3, and 4 the input phage titer was approximately 10¹¹ cfu/ml. Phage were allowed to bind for 2-3 h at room temperature. The wells were then quickly washed 13 times with 400 μl/well of PBS. Bound phage were eluted by incubation with 150 μl/well of 50 mM glycine-HCI, pH 2.2 with 0.1 % BSA for 5 min. The resulting solution was then neutralized with Tris-HCl, pH 8.0.

Log phase TG1 cells were infected with the eluted phage by incubation at 37°C for 1 h. Helper phage (M13K07) were added (multiplicity of infection (MOI)=15) and cells were incubated in the presence of 50 μg/ml ampicillin and 2% glucose for 1 h at 37°C with shaking at 250 rpm. Following infection, cells were pelleted, and resuspended in the initial culture volume of 2xYT containing 50 μg/ml ampicillin and 50 μg/ml kanamycin. Cells were grown overnight at 37°C with shaking at 225 rpm. Cells from the overnight culture were pelleted and phage were recovered in the supernatant. Phage were precipitated with 6% PEG 8000, 300 mM NaCl and chilled on ice for 1 h. Precipitated phage was pelleted by centrifugation at 10,000 x g for 30 min, and resuspended in PBS (1/100 of the initial volume).

EXAMPLE 3: ELISAANALYSIS OF PHAGE

For analysis of individual clones, colonies were picked and phage prepared as described above using helper phage, M13KO7. Ninety-six-well microtiter plates were coated with 100 ng/well of target protein at 4°C overnight. Plates were then blocked with PBS containing 2% non-fat milk for 1 h at room temperature. One hundred microliters per well of phage from isolated clones were added and plates were incubated at room temperature for 2 h. The phage solution was then removed, and the wells were washed three times with PBS at room temperature. Anti-M13 antibody conjugated to horseradish peroxidase (Pharmacia) was diluted 1 :3000 in PBS and added to each well (100 μl/well). Incubation was for 1 h at room temperature, followed by PBS washes as described. Color was developed by addition of ABTS solution (100 μl/well; Boehringer, Indianapolis, IN). Plates were analyzed at 405 nm. A clone was considered "positive" if the A_tos of the well was > 2-fold over background. Results are shown in Table 5, below. EXAMPLE 4: DGI-3 PANNING RESULTS Overall, 92 clones from the DGI-3 panning experiments were sequenced. Nine unique peptides were identified, including four high- specificity binders (DGI-3-binders designated A1 , G3, D9, and C5), one potential binder (DGI-3-binder designated D1 ) and four non-binders. Each of the high-specificity DGI-3-binders were isolated multiple times. Table 5 shows the encoded amino acid sequences and results for each of the clones panned against DGI-3. Table 6 shows the nucleotide sequences of the clones identified.

In Table 5, the number shown in subscript following each peptide name represents the number of times the peptide was isolated from the 92 DGI-3 clones; For example, DGI-3-binder A1 was isolated 58 times from a total of 92 peptides sequenced; E-Tag = peptide binding to E-tag antibody (AAAGAPVPYPDPLEPRP; SEQ ID NO:37) fused to peptide; DGI-3 = binding to DGI-3 polypeptide (specific binding); IGF1 R = peptide binding to IGF-1 R polypeptide (background binding); SP/IRR = ratio of specific binding to irrelevant (background) binding; . The Q residues represent translation read-through at TAG stop codons. Values shown represent ELISA units read at 405 nm (described below); Higher values indicated higher binding specificity for target; Binders = peptides that showed specific (HIT or POTENTIAL) binding to DGI-3; Cysteines shown in bold are sites of potential loops or other constrained conformations; Both A1 and G3 binders contain cysteines separated by 8 amino acid residues.

TABLE 5

TABLE 6

EXAMPLE 5: DATABASE ANALYSIS OF DGI-3-BINDERS

DGI-3-binders that were identified and shown to bind specifically to DGI-3 were used to identify DGI-3-binding proteins (DGI-3-partners). A Phenogenix® approach was used as described in co-owned patent applications, U.S. Patent Application Serial No. 09/852,455 and International Patent Application No. PCT/US01/15092, which are incorporated herein by reference. These methods employed several different database search programs. Initially, the entire peptide sequence and consensus motifs of the DGI-3-binders were entered into an Advanced BLAST search (available from the NCBI website at world wide web.ncbi.nlm.nih.gov/blast/blast.cgi?Jform=1 ) using the following parameters:

1. Programs: BLASTP, TBLASTN;

2. Databases: protein and nucleotide databases including dbest (ESTs), dsts (STSs), and htgs (unfinished high throughput genomic sequences);

3. Expect value: 1000 to 20000;

4. Matrix: PAM30 or PAM70; and

5. Query: consensus motif alone and varying combinations of sequence at the N- and C-terminal ends.

In subsequent steps, motifs identified by sequence alignment programs like MEME (Multiple EM for Motif Elicitation), (hypertext transfer protocol://meme. sdsc.edu/meme/website/intro. html) were also used to search the available databases using MAST (Motif Alignment and Search Tool, hypertext transfer protocol://meme.sdsc.edu/mem/website/mast- intro.html). Motifs and consensus domains were further used as query patterns to search the protein databases using Patternfind (available online at world wide web.isrec.isb-sib.ch/software/PATFND_ form.html). For Patternfind, the following parameters were used: 1. Databases: non-redundant, Swissprot, TREATS, and

TROGON; 2. Limit: between 10 and 5000; and

3. Query: consensus motif alone and varying combinations of sequence at N- and C-terminal ends.

Data obtained from the various searches were analyzed using the following approach:

1. Results of different searches were analyzed independently and then combined to search for similar classes of proteins (e.g., nucleic acid binding proteins, kinases) that emerged;

2. The best matches identified in more than one kind of search (e.g. same protein/ORF picked up by BLAST searches using different parameters, or by both BLAST and Patternfind) were picked. The amino acid sequence of the protein in the identified region was compared with other peptide surrogates containing this amino acid sequence; and

3. The protein interaction was evaluated in view of the cellular function of the target, DGI-3.

The criteria for identification of a DGI-3-partner (i.e., partner hit) included the following:

1. Search produced an exact match of > 5-7 amino acids;

2. Same partner appeared in at least 50% of the top hits of any one search, and/or the same or related hits occurred in multiple searches;

3. Search produced expected class of protein partners based on function, cellular location, or tissue/disease distribution; and

4. Partner candidate produced a phenotype change when added into the appropriate model system. Unless there was an exact sequence match (criterion 1 ), the candidate partner was required to satisfy at least two other criteria to be considered a partner hit. EXAMPLE 6: RESULTS OF DATABASE ANALYSIS

A BLASTP search and database analysis was performed using the Phenogenix® approach described. The results are described as follows. 1. A1 : Database analysis of DGI-3-binder A1 identified the pleckstrin domain (PHD) of cytohesins 1-4 as a DGI-3-partner hit. Cytohesins promote the exchange of GDP by GTP on ARFs (ADP- ribosylation factors). The PHDs bind the second messengers PIP3 and IP4. Database analysis of A1 also identified RAS-GAP as a DGI-3-partner hit. RAS-GAP facilitates GTPase activity of the Ras oncogene. Results for A1 are shown in Table 7, below.

2. G3: Database analysis of DGI-3-binder G3 identified the Ras GEF (GTP Exchange Factor) domain on GNRP as a DGI-3-partner hit. GNRP promotes the exchange of GDP for GTP on Ras.

3. D9: Database analysis identified no partner hits;

4. C5: Database analysis of the DGI-3-binder C5 identified the protein kinase domain of Trio as a partner hit. The Trio protein contains 3 catalytic sites (two GEF domains and one serine/threonine kinase domain) and promotes the exchange of GDP by GTP on Rho/Rac.

5. D1 : Database analysis of DGI-3-binder D1 peptide identified Rhophilin as a DGI-3-partner hit. Rhophilin is a GTP-Rho binding protein. Results of database analysis are summarized in Table 8, below.

TABLE 7

TABLE 8

The amino acid sequences for cytohesins 1-4 are shown in Figures 2A-2D. The amino acid sequences for GNRP and Ras-GAP are shown in Figures 3A-3B. The amino acid sequences for Trio and Rhophilin are shown in Figures 4A-4B. GenBank (GB) and GenPept (GP) Accession Numbers for DGI-3 and each of the DGI-3-partner hits are shown in Table 9, below.

TABLE 9

BLASTP searches were performed for each of the partner hits. In addition, amino acid sequence alignments were constructed for each of the peptides and their partner hits. The sequence alignments are shown in Table 10, below.

TABLE 10

PEPTIDE SEQUENCE SEQ ID NO:

Peptide C5 E ARCWRW- - LVEGWGQL 6 Trio LDCWRWGSLTEGKIRAH 12

Trio LDCWRWGSL TEGKIRAH 12

Peptide Al T I RRWFCLAHWGTEGCLART 3 Cytohesins WKRRWFIL TDNCLYYF 13 RAS -GAP KNFKKRWFCL TSRELTYH 14 GNRP KWQTKWFAL LQNLL 15

Peptide G3 GLGWRDPVCVP FRGMRLCLV 4

GNRP TLRNCDP - CVP YLGMYLTDLAF 16

Trio QGRLLDCWRWGSLTEGKIRA 17

Table 10: Amino acid residues in bold show identical residues shared by at least two sequences. Amino acid residues underlined show conservative substitutions. Notably, Trio protein, which was identified by DGI-3-binder C5, also showed amino acid sequence homology to DGI-3-binder A1. DGI-3-binder

A1 also showed regions of amino acid homology to GNRP, which was identified by DGI-3-binder G3. BLASTP analysis of Trio also identified regions of homology to GNRP.

EXAMPLE 7: DGI-3 AMINO ACID AND NUCLEOTIDE SEQUENCE ANALYSIS

PSORT II analysis (K. Nakai; hypertext transfer protocol://psort.nibb.ac.jp) predicted that the subcellular location of DGI-3 was cytoplasmic (47% probability) or nuclear (39% probability). Less likely locations include the mitochondria (8% probability) and the cytoskeleton (4% probability). In addition, PSORT II analysis has identified a coil-coil domain in the DGI-3 polypeptide that includes residues 25 to 53 (NEDGLRQVLEEMKALYEQNQSDVNEAKS; SEQ ID NO:151 ).

AceView analysis of DGI-3 (KIAA0186) obtained from the NCBI website (AceView provided by humangenes.org; D.Thierry-Mieg et al., Construction and automatic annotation of cDA-supported genes using ACEmbly) indicated that the gene was delineated by 26 sequences from 25 cDNA clones. It has been localized to human chromosome 20 on the direct strand, including base 25376375 to base 25421392. According to RefSeq annotation (NCBI Reference Sequence Project) the cytogenetic location of DGI-3 is 20p11.1. The gene includes 45017 bp of genomic DNA. Further analysis indicated that an in-frame stop in the DGI-3 5'UTR is positioned 60 bp before the initiating methionine. The DGI-3 3'UTR contains about 2564 bp followed by a polyA tail. The standard AAT AAA polyadenylation signal is observed about 21 bp before the DGI-3 polyA tail. The length of the 3'UTR points to a regulatory function.

According to one set of AceView data, the DGI-3 gene is predicted to include 8 introns that follow the consensus gt-ag rule for splicing. Details regarding the DGI-3 introns are presented in Table 11 , below. For Table 11 , the reference sequence for calculating the start, end, and length values is Ensembl number ENSG00000101003 (Wellcome Trust Sanger Institute).

TABLE 11

According to another set of AceView data, the DGI-3 mRNA is predicted to include 1 1 confirmed introns, 9 of which are alternative. Comparison to the genomic sequence revealed 1 1 introns that follow the consensual gt-ag rule. DGI-3 is also predicted to produce, by alternative splicing, 3 different transcript variants A, B, C, which encode 3 proteins. DGI-3 splice variant A mRNA is 3.29 kb long, and includes a very long 3' UTR. DGI-3 splice variant B mRNA is 2.09 kb long, and includes a very long 3' UTR. DGI-3 splice variant B mRNA is 0.47 kb long. Additional data regarding the DGI-3 splice variants is shown in Tables 12-14, below. For Table 12 and 14, the reference sequences for calculating the start, end, and length values is Ensembl number ENSG00000101003 (Wellcome Trust Sanger Institute) located between 25376375-25421392 bp on human chromosome 20. For Table 13, the reference sequence is shown in Figures 11 B-11C.

TABLE 12

TABLE 13

Polypeptide Starts Ends Position in mRNA Length

Variant A Met Stop 141 to 730 196 aa

Variant B Met Stop 290 to 727 145 aa

Variant C Met Stop 198 to 404 68 aa

TABLE 14

It should be noted that the analysis described herein identified only 7 exons in DGI-3 (Figure 11D) and only one DGI-3 splice product (Figures 8- 10). However, it is possible that the other DGI-3 variants and splice products predicted by AceView may be observed in specific cells, tissue types, and/or developmental stages. EXAMPLE 8: RNA ISOLATION AND NORTHERN BLOT ANALYSIS

Radioactive methods: HMEC (human mammary epithelial cells) were obtained from Cambrex Corporation (East Rutherford, NJ). HBL-100 (tranformed breast epithelial cells), and T47-D, MCF-7, MDA-MB-157, MDA- MB-231 , MDA-MB-453 (breast carcinoma cells) were obtained from ATCC (Manassas, VA). Total cellular RNA was isolated using the QIAGEN RNeasy Mini Kit (QIAGEN) according to the manufacturer's protocol. Northern blotting was performed as described (Z.Z. Su et al., 1999, Proc. Natl. Acad. Sci. USA 96:15115-15120; D.C. Kang et al., 1998, Proc. Natl. Acad. Sci. USA 95:13788-13793). Fifteen micrograms of RNA were denatured and electrophoresed in 1.2% agarose gels with 3% formaldehyde. The RNA was transferred to nylon membranes and hybridized sequentially with ³²P-labeled cDNA probes as described (Z.Z. Su et al., 1999, Proc. Natl. Acad. Sci. USA 96:15115-15120; D.C. Kang et al., 1998, Proc. Natl. Acad. Sci. USA 95:13788-13793). Following hybridization, the filters were washed and visualized by autoradiography. The cDNA probes included full-length human actin, full-length human GAPDH, and full length DGI-3. The nucleotide sequence of the DGI-3 probe is show in Figure 1C. Results are shown in Figure 8 and Table 15. The results indicated that DGI-3 transcript was overexpressed in transformed breast epithelial cells (HBL-100) and breast cancer cells (MDA-MB-157, MDA-MB- 231 , MDA-MB-453). In addition, DGI-3 transcript was highly overexpressed in certain breast cancer cells (T47-D and MCF-7).

TABLE 15

In separate experiments, DGI-3 expression was analyzed in normal human tissues using a multiple tissue northern blot (MTN; CLONTECH) according to the manufacturer's instructions. Results are shown in Figure 9. The results indicated that DGI-3 transcript was restricted to heart, liver, kidney, and adrenal gland tissue. Additional information from the GeneCard™ website (hypertext transfer protocol://bioinformatics. weizmann.ac.il/cards/; Weizmann Institute of Science, Rehovot, Israel) indicated that the DGI-3 transcript was also observed brain tissue.

Non-radioactive methods: HBL-100 (transformed breast epithelial cells); PANC-1 and MiaPaCa (pancreatic carcinoma cells); DU-145 (prostate carcinoma cells); HL-60 (myeloid leukemia cells); HT29, Caco-2, WiDr, and SW480 (colorectal carcinoma cells); T47D, MDA-MB-231 , MDA- MB-453, and MCF-7 (breast carcinoma cells); A549 (lung carcinoma cells) were all obtained from ATCC. PolyA+ RNA was purified from cancer cell lines using the FastTrack 2.0 Kit (Invitrogen). The Buffer Puffer Gel System (Owl Separation Systems, Portsmouth, NH) combined with the Northern Max-Gly System (Ambion, Austin, TX) was used for gel resolution of polyA+ RNA. RNA samples (5 μg per lane) were prepared in glyoxal loading dye and were electrophoresed on 2% agarose gel. The gel was run in 1 X Gel Prep/Gel Running Buffer (Ambion) at a voltage of 5 v/cm for 1.5 h. Following this, the RNA was transferred to membrane using the Panther Semidry Electroblotter (Owl Separation Systems). For the transfer, 1 X TBE buffer was used and a constant current (220 mM) was applied for 2.5 h. The RNA was UV-cross-l inked to the membrane using the Optimal Crosslink Mode (Spectronics Corporation, Westbury, NY). Two identical gels were run as described above, one for hybridization with the DGI-3 probe and one for hybridization with the beta-actin control probe. The probe for detection of DGI-3 and beta-actin (control) RNA transcripts was generated by using the PCR DIG Probe Synthesis Kit (Roche). Primers for synthesis of the DGI-3 probe included Forward primer: 5'-cgattcgttctagaatgttctgcgaaaaagccatg-3' (SEQ ID NO:152); Reverse primer: 5'- cga ttc gtt eta gaa tga aag ctt tgt atg aac aaa ac-3' (SEQ ID NO: 153). Primers for synthesis of the actin probe included Forward primer: 5'-caa gag atg gcc acg get gct-3' (SEQ ID NO:154); Reverse primer: 5'-tcc ttc tgc ate ctg teg gca-3' (SEQ ID NO:155). For the PCR reaction, 5 μl 10 X Buffer, 5 μl 50 X dNTP plus dUTP, 0.75 μl enzyme (High Fidelity Enzyme Mix; Roche), 1 μl Forward primer (100 μM), 1 μl Reverse primer (100 μM), 1 μl (100 ng) DNA (DGI-3 cloned into plVEX2.3d (Roche Applied Science)), and 36.25 μl H₂0 were combined in a final volume of 50 μl.

The reaction was first incubated at 95°C for 2 min, then cycled 30 times at 95°C for 30 sec, 62°C for 30 sec, and 72°C for 1 min, and then incubated at 72°C for 7 min. The PCR primers for the DGI-3 template generated a 700 bp fragment covering the entire open reading frame of the gene. Hybridization was carried out using DIG Easy Hyb (Roche). For the reaction, 3 μl of the DIG-labeled probes were added to 5 ml of the DIG Esy Hyb solution. This mixture was added to the membrane, and the membrane was incubated overnight at 42°C. Two stringent washes were performed with 2 X SSC and 0.1 % SDS for 5 min at room template. This was followed by two washes with in 0.1 X SSC and 0.1 % for 15 min at 68°C under constant agitation. Additional washing and development was carried out using the DIG Wash and Block Buffer Set (Roche). The nucleotide sequence of the DGI-3 probe is shown in Figure 1C. Results are shown in Figure 10. The results indicated that DGI-3 transcript was overexpressed in breast cancer cell lines T47-D and MCF-7, and also in colorectal cancer cell line Caco-2 and pancreatic cancer cell line PANC-1.

Additional data was obtained from GeneExpress® (GeneLogic, Gaithersburg MD; Tables 16-17, below). Table 16 shows DGI-3 Gene expression in normal tissues. In Table 16, the column marked "Present" shows the number of samples that were DGI-3-positive out of the total number of samples tested. Table 17 shows differential DGI-3 gene expression corresponding to disease state/morphology. In Table 17, NOS indicates not otherwise specified.

TABLE 16

In addition, data showing DGI-3 gene profiling in the NCI-60 cell set was obtained from the SOURCE website (Stanford online universal resource for clones and ESTs; original data from D.T. Ross et al., 2000, Nature Genetics 24:227-235). For gene profiling, mRNA from normal breast samples (CLONTECH) was used as a control, and cell lines included HL-60 (acute myeloid leukemia), K562 (chronic myeloid leukemia), NCI-H226 (non-small cell lung carcinoma), COLO 205 (colorectal carcinoma), LOX-IMVI (melanoma), SNB-19 (central nervous system cancer), OVCAR-3 and OVCAR-4 (ovarian carcinoma), CAKI-1 (renal carcinoma), PC-3 (prostate carcinoma) and MCF-7 and Hs578T (breast carcinoma), Table 18. Gene profiling data was combined with Northern data showing DGI-3 expression (Figure 8; Table 18). The combined data indicated that DGI-3 transcript was observed in breast cancer, melanoma, colorectal cancer, pancreatic cancer, ovarian cancer, and leukemic cells. TABLE 18

EXAMPLE 9: PROMOTER ISOLATION AND ANALYSIS

Cloning methods: Experiments were performed to investigate the regulatory elements that control DGI-3 gene expression. A 4.3 kb Xho I- Hind III fragment (Figures 13A-13B) located upstream from the translational start site of DGI-3 was cloned into the pGL3-Enhancer Vector (Promega). The pGL3-Enhancer Vector contains the luciferase reporter gene but does not contain a promoter sequence (Figure 14A). BAC clone RP11-384-D7 (ResGen, Invitrogen Corporation, Carlsbad, CA) comprising the DGI-3 genomic sequence, was used in a PCR reaction with forward and reverse primers incorporating the Xho I and Hind III restriction sites. BAC DNA was first purified (QIAGEN) and the concentration was determined. For the PCR reaction, 0.5 μl (100 ng) genomic BAC DNA, 5 μl primers (10 μM; Reverse primer: 5'-tta get aag ctt aac cac cag etc tea cgc caa aat g-3'; SEQ ID NO:156; Forward primer: 5'-acg tea etc gag ggg tta ttt tct ccc tgt tea ggt ctg-3'; SEQ ID NO:157), 1 μl dNTPs (50 X), ThermalAce Buffer (1X; Invitrogen), and 2 U ThermalAce enzyme (Invitrogen) were incubated in a total volume of 50 μl. The PCR reaction tube was first incubated at 95°C for 10 min, cycled for 28 cycles at 95°C for 30 sec, 68°C for 30 sec, 74°C for 4 min, and then incubated at 74°C for 10 min. The PCR product which corresponded to the predicted 4.4 kb fragment was gel purified, digested with Xho I and Hind III and ligated to the pGL3-Enhancer Vector (Figure 14B). The sequence of the clone was confirmed by sequencing analysis.

Luciferase reporter gene assay: The luciferase reporter gene assay was used to allow precise quantification of firefly luciferase activity in transfected MCF-7 and NIH3T3 cells. MCF-7and NIH3T3 cells were grown stably in culture media RPMI supplemented with 10% FBS (Gibco). Cells were plated at 1 x 10⁵ cells/well in 12-well tissue culture plates, and grown overnight at 37°C in 5% CO₂. After approximately 18 h, the cells were transfected with 3 μg of pGL3-Enhancer Vector-DGI-3-promoter, 3 μg of pGL-Enhancer and 3 μg of pGL3-control (Promega). Each construct was tested in duplicate assays as follows.

Pre-warmed RPMI plus serum free medium was added to a sterile 2 ml tube with 3 μg of DNA, and vortexed immediately. TransFast™ reagent (Promega) was added to yield a 1 :1 charge ratio of TransFast™ to DNA, and vortexed immediately. The TransFast™/DNA mixture was incubated at room temperature for 10-15 min. Growth medium was removed from the cells, and 2 ml TransFast™/DNA mixture was added to the cells. Cells were returned to the 37°C incubator for 1 h. At the end of the incubation period, cells were gently overlaid with 3 ml complete medium. Cells were returned to the incubator without removing the transfection medium containing the Trans Fast Reagent. Incubation of the transfected cells was continued for 48 h in the 37°C incubator. Following this, the cells were starved in RPMI plus 2% FBS for 8 h. Cells were then were stimulated by adding, to every other well, IGF (20 nM; PeproTech, Inc., Rocky Hill, NJ) in 3 ml RPMI plus 2% FBS. Cells with or without IGF were incubated overnight. The tranfected cells were then evaluated using the Luciferase Reporter Gene Assay Kit (Roche). For transient transfection, cells were typically harvested 48-72 h after transfection. All reagents were stored at or below -15-25°C. For the assay, lysis buffer (supplied with Roche kit) was prepared by diluting 1 part lysis buffer (5 X) with 4 parts distilled water. The culture supernatant was aspirated, and cells were carefully rinsed with 1 X PBS. The remaining PBS was aspirated completely with a fine-tipped pipette. The minimum volume of lysis buffer was added to cover the cells (250 μl for 60 mm dishes or 100 μl for 35 mm dishes). Solubilized cells were transferred to microcentrifuge tubes and incubated at room temperature for a total of 15 min (the time for incubation was calculated from the addition of lysis buffer to the cells). To remove cellular debris, the tubes were spun for 5-10 second in a microcentrifuge at maximum speed. The supernatant was transferred from the tubes to black 96-well microtiter plates. To start the reaction immediately, 20-50 μl cell extract was used. The luciferase assay reagent was prepared by adding 10 ml reaction buffer (Bottle 1 ) to the lyophilized luciferase substrate (Bottle 2; both supplied with Roche kit) and mixing well. Next, 100 μl luciferase assay reagent was added to the cell extracts. The plate was immediately analyzed in a Wallac 1420 multilabel counter (VICTOR 2). Results from the assay are shown in Table 19, below. The results indicated that the cloned DGI-3 promoter region was active in different cell types.

TABLE 19

* The pGL3 vector containing the DGI-3 promoter 4.3 kb fragment.

**The pGL3-control vector (Promega) has SV40 promoter and gives high constitutive luciferase activity when transfected into cells. EXAMPLE 10: SECONDARY LIBRARY PEPTIDES

Construction of A1 secondary libraries: All libraries were designed with the short FLAG® Epitope DYKD (SEQ ID NO:158; Hopp et al., 1988, Bio/Technology 6:1205-1210) at the N-terminus of the listed sequence and an E-tag epitope (GAPVPYPDPLEPR; SEQ ID NO: 159) at the C-terminus. Amino acid mutations were introduced into the A1 sequence via oligonucleotide synthesis. The mutated oligonucleotide was used as the template in a PCR reaction with two shorter 5' biotinylated oligonucleotide primers contributing the restriction sites. The primers included the following sequences: Forward 5'-Biotin-tgttcctttctatgcggcccagccggccatggcg-3' (SEQ ID NO: 190) and Reverse 5'-Biotin-cgaaatcttttggactcacactgcg-3'; SEQ ID NO: 191 ). The library was then produced essentially as described (Ravera et al., 1998, Oncogene 16:1993-1999). The resulting PCR products were inserted into the pCANTAB5E vector (Pharmacia) digested with Sfi\ and Λ/ofl. The ligation reaction was carried out overnight at 15 °C. The ligation product was purified and electroporations were performed with Escherichia coli strain TG1 electrocompetent cells ((F'_fraD36 lacP Δ(/acZ)M15 proAB)/supE A(hsdM-mcrB)s r_k ^"m_k ^" McrB^') thi L (lac-proAB). Cells were pooled and an aliquot was plated to determine the total number of transformants. The diversity of the secondary cell library was determined to

10 be 1 x 10 clones.

Rescue of secondary phage libraries: The secondary cell library was rescued according to the standard phage preparation protocol (Ravera et al., 1998, Oncogene 16:1993-1999). The phage were titered by infection of TG1 cells. The phage titer for the library was determined to be 1 x 10¹³ cfu/ml. The secondary phage library was subsequently used in panning experiments.

Panning A1 secondary libraries: A standard method was used to coat and block all microtiter plates. DGI-3 protein was obtained from the in vitro expression system described above (Rapid Translation System RTS

5000, Roche Diagnostics GmbH, Mannheim Germany). Plates were coated with DGI-3 protein in PBS. One hundred microliters of solution containing 500 ng DGI-3 protein was added to wells in a 96-well microtiter plate (MaxiSorp plates, Nunc) and incubated overnight at 4°C. Wells were then blocked with a solution of 2% non-fat milk in PBS (MPBS) at room temperature (RT) for at least 1 h. Four to eight wells coated with DGI-3 were used for each round of panning. One hundred microliters of phage were added to each well. For the first round, the input phage titer was ~10¹³ cfu/ml. For subsequent rounds, the input phage titer was approximately 10¹² cfu/ml. Phage were allowed to bind for 2-3 h at RT. The wells were then quickly washed 13 times with 300 μl/well of PBS.

Bound phage were eluted by incubation with 150 μl/well of 50 mM glycine-HCI, pH 2.0 for 5 min. The resulting solution was pooled and then neutralized with Tris-HCl, pH 8.0. Log phase TG1 cells were infected with the eluted phage, in 2xYT medium for 1 h at 37°C prior to the addition of helper phage, ampicillin, and glucose (2% final concentration). After incubation for another hour at 37°C, the cells were spun down and resuspended in 2xYT-AK medium. The cells were then returned to the shaker and incubated overnight at 37°C. Phage amplified overnight were then precipitated and subjected to the next round of panning. A total of 96 clones were picked at random from rounds 3 and 4 and screened for binding activity.

ELISA analyses of phage from A1 secondary libraries: For phage pools, cells from frozen stocks were grown and phage were prepared as described above. For analysis of individual clones, colonies were picked and phage prepared as described above. Subsequent steps were the same for pooled and clonal phage. Microtiter wells were coated and blocked as described above. Wells were coated with either DGI-3 protein or LDH (lactate dehydrogenase) as a negative control. Phage resuspended in MPBS (PBS containing 2% non-fat milk) were added to wells (100 μl/well) and incubated at room temperature for 1 h. Following this, the phage solution was removed, and the wells were washed three times with PBS at room temperature. Anti-M13 antibody conjugated to horseradish peroxidase (Pharmacia Biotech) was diluted 1 :3000 in MPBS and added to each well (100 μl/well). Incubation was carried out for another hour at room temperature, followed by PBS washes as described. Color was developed by addition of ABTS solution (100 μl/well; Boehringer). Color development was stopped by adjusting each well to 0.5% SDS. Plates were analyzed at 405 nm using a SpectraMax 340 plate reader (Molecular Devices) and SoftMax Pro software (Molecular Devices Corporation, Sunnyvale CA). Data points were averaged after subtraction of appropriate blanks. A clone was considered "positive" if the A os of the well was > 2-fold over background. Peptides identified by panning the A1 secondary library are shown in Figures 16A-16B.

Determination of amino acid preferences: Amino acid preferences for each peptide were determined as follows. The expected frequency of each of the 20 amino acids at that position was calculated based on codon usage and % doping for that library. This was compared to the actual frequency of occurrence of each amino acid at every position after four rounds of biopanning. Any amino acid that occurred at a frequency >2-fold was considered preferred. Most preferred amino acid(s) were considered to be those that showed the greatest fold enrichment after panning. Preferred peptides were synthesized based on the information provided by the A1 peptide secondary library peptides (Figure 17). EXAMPLE 11 : STABLE TRANSFECTANTS

Cells and reagents: NIH3T3 cells were obtained from ATCC (Manassas, VA). Plasmids were prepared using endotoxin-free Mega-Prep Kit (QIAGEN). NIH3T3-DGI-3 and NIH3T3-cytohesin-1 (negative control) transfectants were constructed using the TransFast™ (Promega) lipofectamine reagent in the protocol described below. For the transfection, 2 μg DNA was used (pcDNA3.1 Hygro+ Vector with DGI-3 or cytohesin-1 inserts). For the cytohesin-1 construct, the cytohesin-1 sequence (GenBank Ace. No. accession NM_004762.1 ) was cloned by PCR amplification using Human Universal Quick-Clone™ cDNA (CLONTECH) and ThermalAce enzyme (Invitrogen). The PCR reaction mixture included 100 ng Quick- Clone™ cDNA, Forward primer (1 μM; 5'-tcc cgc ace atg gag gag gac gac age tac-3'; SEQ ID NO:174), Reverse primer (1 μM; 5'-ttg get gca cgc tea gtg teg ctt c-3'; SEQ ID NO:175), dNTPs (200 μM), ThermalAce Buffer (1X; Invitrogen), and 2 U ThermalAce enzyme (Invitrogen) in a total volume of 50 μl. The PCR reaction tube was incubated at 95°C for 5 min, then incubated for 35 cycles at 95°C for 45 sec, 68°C for 45 sec, 72°C for 2 min, and then incubated at 72°C for 3 min. The PCR product corresponding to the predicted 1.1 kb fragment was gel purified and blunt end ligated into the pZErO™-2 vector (Invitrogen) digested with EcoRV. Construction was confirmed by sequencing analysis. The cytohesin-1 sequence was subcloned into pcDNA3.1/Hygro vector by PCR amplification. The PCR reaction included 100 ng pZerO-2- cytohesin-1 , Forward primer (1 μM; 5'-cga ttc gtt eta gaa tgg agg agg acg aca get ac-3'; SEQ ID NO:176), Reverse primer (1 μM; 5'-cga ttc gtc gta cct tag tgt cgc ttc gtg gag gag ac-3'; SEQ ID NO:177), dNTPs (200 μM), ThermalAce Buffer (1X; Invitrogen), and 2 U ThermalAce enzyme (Invitrogen) in a total volume of 50 μl. The PCR reaction tube was incubated at 95°C for 5 min, then incubated for 30 cycles at 95°C for 45 sec, 68°C for 45 sec, 72°C for 2 min, and then incubated at 72°C for 3 min. The PCR product corresponding to the predicted 1.1 kb fragment was gel purified, digested with Kpn I and Xba I, and ligated into the pcDNA3.1/Hygro vector (Invitrogen) digested with Kpn I and Xba I. Construction was confirmed by sequencing analysis.

The DGI-3 sequence was cloned into the pcDNA3.1/Hygro vector (Invitrogen) by PCR amplification. The PCR reaction mixture included 100 ng DGI-3 template, Forward primer (1 μM; 5'-cga ttc gtt eta gaa tgt tct gcg aaa aag cca tg-3'; SEQ ID NO:178), Reverse primer (1 μM; 5'-cga ttc gtg gta cct tat gac agg atg tgc tec agg ac-3'; SEQ ID NO:179), dNTPs (200 μM), ThermalAce Buffer (1X; Invitrogen), and 2 U ThermalAce enzyme (Invitrogen) in a total volume of 50 μl. The PCR reaction tube was incubated at 95°C for 5 min, then incubated for 30 cycles at 95°C for 45 sec, 68°C for 45 sec, 72°C for 2 min, and then incubated at 72°C for 3 min. The PCR product corresponding to the predicted 0.7 kb fragment was gel purified, digested with Kpn I and Xba I, and ligated into the pcDNA3.1/Hygro vector (Invitrogen) digested with Kpn I and Xba I. Construction was confirmed by sequencing analysis.

TransFast™ transfection protocol: Cells were grown in RPMI supplemented with 10% fetal bovine serum (Gibco), and then plated at 1 x 10⁵ cells/well in a 6-well culture dish. Cells were incubated in 5% C0₂ at 37°C overnight. For each well, 1 ml serum-free medium was added to a sterile 15 ml conical, pre-mixed with 2 μg DNA (pcDNA3.1 -cytohesin-1 or pcDNA3.1-DGI-3), and vortexed. A 1 :1 charge ratio of TransFast™ Reagent to DNA was used (i.e., 3 μl reagent per 1 μg DNA). Accordingly, 6 μl of TransFast™ was added to tube containing DNA and serum-free media, and the mixture was immediately vortexed. The TransFast™/DNA/media mixture was incubated for 10-15 min at room temperature. The growth media was removed from the cells. Prior to addition, the mixture was vortexed. Then, 1 ml TransFast™/DNA/media mixture was added to each well, and cells were promptly returned to the incubator for 1 h.

At the end of the incubation period, cells were overlaid with 2 ml pre- warmed complete growth media. Transfection medium containing the TransFast™ Reagent was not removed. Cells were returned to the incubator for 24 h. At 24 h post transfection, media was aspirated and replaced with RPMI with 10% FBS plus 25 μg/ ml hygromycin B (Roche). Media was changed daily to remove necrosis factors. When foci formed, and the cells looked healthy, cells were transferred to a T-25 flask. Concentration of hygromycin was increased to 50 μg/ml in complete media after 1 week. Stable transfectants were allowed to grow for 4 weeks and passaged 4 times before used for further analysis. Results are shown in Figures 18A-18C. The results indicate that stable transfection of DGI-3 causes a transformed morphology in NIH3T3 cells.

EXAMPLE 12: TRANSIENT TRANFECTIONS USING DGI-3 ANTISENSE AND DGI-3-BINDERS

Antisense transient transfections: For antisense and peptide experiments, both MCF-7 (breast carcinoma cells) and HBL-100 (transformed breast epithelial cells) cell lines were used. The DGI-3 antisense sequence (Figure 19A) was cloned into pcDNA 3.1 vector digested with BamH\ and Xho I. Using PCR amplification, BamH\ and Xho I restriction sites were added to the DGI-3 antisense sequence. The PCR primers included DGI-3 antisense forward primer: 5'-gcatttccc attgggatcctgacaggatgtgctccaggactccttgtc-3' (SEQ ID NO:160); and DGI-3 antisense reverse primer: 5'-gcatgcttcaagtctcgagatgttctgc gaaaaagccatggaac-3' (SEQ ID NO:161 ). The PHD antisense sequence (Figure 19B) was obtained from the cytohesin-1 antisense strand using the following primers. The PHD insert was cloned into the pcDNA 3.1 + vector between the Hind III and Not I sites. The PCR primers included: PHD antisense forward primer: 5' gcg atg etc aag ctt get gat ggc tgc ttt aat gca c 3' (SEQ ID NO:192); and PHD antisense reverse primer: 5' gca ttt aat ctg egg ccg cc gaga agg ctg get att gaa act c 3' (SEQ ID NO:193).

For transfections, cells were plated at 1 x 10⁵ per well in a 12-well tissue culture plate and incubated overnight. For each well, 3 μg of DNA (pcDNA3.1 -cytohesin-1 PHD antisense or pcDNA3.1 -DGI-3 antisense) was transfected using the TransFast™ protocol from Promega. Each construct was tested in triplicate. At 24 h, the media was refreshed. At 72 h, the cells were counted following trypsinization in a 37°C incubator. For trypsinization, cells were incubated for 5 min with 900 μl 0.05% trypsin/EDTA (Gibco). Cells were removed from the incubator, and 100 μl fetal bovine serum was added to neutralize action of trypsin and prevent cell death. Following this, 100 μl cells in trypsin/FBS were stained with 100 μl Trypan Blue. A 15 μl aliquot of the mixture was added to a 0.1 mm deep hemacytometer (Bright Line, Reichert Ophthalmic Instruments, Depew, NY).

Peptide/Protein transient transfections: Peptides and proteins including A1 , G3, cytohesin-1 and VEGF ligand were cloned into pcDNA3.1 (Figures 19C-19D). For each peptide or protein, the coding sequence was cloned into the pcDNA 3.1 Hygro+ vector between the Hind III and Not I sites. The Kozak initiation sequence (5'-ccaccatgg-3') was added to the constructs to promote translation. A histidine tag was added to the C- terminus of each peptide (HHHHHH-stop; SEQ ID NO:1 ).

The cytohesin-1 sequence was cloned by PCR amplification using Human Universal QuickClone™ cDNA (CLONTECH) and ThermalAce enzyme (Invitrogen). The PCR reaction mixture included 100 ng QuickClone™ cDNA, Forward primer (1 μM; 5'-tcc cgc ace atg gag gag gac gac age tac-3'; SEQ ID NO:194), Reverse primer (1 μM; 5'-ttg get gca cgc tea gtg teg ctt c-3'; SEQ ID NO:195), dNTPs (200 μM), ThermalAce Buffer (1X; Invitrogen), and 2 U ThermalAce enzyme in a total volume of 50 μl. The PCR reaction tube was incubated at 95°C for 5 min, then incubated for 35 cycles at 95°C for 45 sec, 68°C for 45 sec, 72°C for 2 min, and then incubated at 72°C for 3 min. The PCR product which corresponded to the predicted 1.1 kb fragment was gel purified and blunt end ligated into the pZErO™-2 vector (Invitrogen) digested with Eco/ . Construction was confirmed by sequencing analysis.

Next, the cytohesin-1 sequence was cloned into the pcDNA 3.1 Hygro vector between the Xba I and Kpn I sites. PCR amplification was used to create the DNA insert. The PCR reaction mixture included 100 ng pZErO™-2-cytohesin-1 , Forward primer (1 μM; 5'-cga ttc gtt eta gaa tgg agg agg acg aca get ac-3'; SEQ ID NO:196), Reverse primer (1 μM; 5'-cga ttc gtg gta cct tag tgt cgc ttc gtg gag gag ac-3'; SEQ ID NO:197), dNTPs (200 μM), ThermalAce Buffer (1X; Invitrogen), and 2 U ThermalAce enzyme in a total volume of 50 μl. The PCR reaction tube was incubated at 95°C for 5 min, then incubated for 35 cycles at 95°C for 45 sec, 68°C for 45 sec, 72°C for 2 min, and then incubated at 72°C for 3 min. The PCR product which corresponded to the predicted 1.1 kb fragment was gel purified, digested with Xba I and Kpn I, and ligated into pcDNA 3.1 Hygro vector digested with Xba I and Kpn I. Construction was confirmed by sequencing analysis.

The VEGF ligand was cloned by PCR amplification using Human Universal QuickClone™ cDNA (CLONTECH) and ThermalAce enzyme (Invitrogen). The PCR reaction mixture included 100 ng QuickClone™ cDNA, Forward primer (1μM; 5'-ccg ate agg gag aga gag att gga aac atg-3'; SEQ ID NO:198), Reverse primer (1 μM; 5'-ttg get gca cgc tea gtg teg ctt c- 3'; SEQ ID NO:199), dNTPs (200 μM), ThermalAce Buffer (1X; Invitrogen), and 2 U ThermalAce enzyme in a total volume of 50 μl. The PCR reaction tube was incubated at 95°C for 5 min, then incubated for 35 cycles at 95°C for 45 sec, 68°C for 45 sec, 72°C for 2 min, and then incubated at 72°C for 3 min. The PCR product which corresponded to the predicted 0.4 kb fragment was gel purified and blunt end ligated into the pZErO™-2 vector (Invitrogen) digested with EcoRV. Construction was confirmed by sequencing analysis. Following this, the VEGF ligand was subcloned into the pcDNA 3.1 Hygro vector between the Hind III and Not I sites. PCR amplification was used to create the DNA insert. The PCR reaction included Forward primer: 5'-gcg gcc cag ccg gcc atg gcg tec egg cga aga gaa gag aca cat tg-3' (SEQ ID NO:200); and Reverse primer: 5'-gat tec ttg egg ccg caa taa gta ccg tat ata aaa cac ttt c-3' (SEQ ID NO:201 ).

For transfections, cells were plated at 1 x 10⁵ per well in a 12-well tissue culture plate and incubated overnight. For each well, 3 μg of DNA (pcDNA3.1-VEGF ligand; pcDNA3.1-A1 ; or pcDNA3.1-G3) was transfected using the TransFast™ protocol from Promega. Each construct was tested in triplicate. At 24 h, the media was refreshed. At 72 h, the cells were counted following trypsinization in a 37°C incubator. For trypsinization, cells were incubated for 5 min with 900 μl 0.05% trypsin/EDTA (Gibco). Cells were removed from the incubator, and 100 μl fetal bovine serum was added to neutralize action of trypsin and prevent cell death. Following this, 100 μl cells in trypsin/FBS were stained with 100 μl Trypan Blue. A 15 μl aliquot of the mixture was added to a 0.1 mm deep hemacytometer (Bright Line by Reichert). Result of the antisense and peptide experiments are shown in Figures 19E-19H. The results indicated that DGI-3 antisense and A1 and G3 peptides substantially inhibited the growth of MCF-7 cells. In contrast, control antisense, peptide, and protein sequences did not affect cell growth. Further, no effect on growth was observed in HBL-100 cells. EXAMPLE 13: ADENO-ASSOCIATED VIRUS EXPRESSING DGI-3

Cells and vectors: To study the effects of DGI-3 antisense constructs, the AAV Helper-Free System was used (Stratagene; Cat #240071 ). HEK-293 (human embryonic kidney cells) and HT1080 (human fibrosarcoma cells) cell lines were obtained from ATCC (Manassas, VA). HEK-293 cells were grown in DMEM with 10% FBS and Glutamax (concentrated glucose; Gibco). HT1080 cells were grown in DMEM supplemented with 10% FBS. Sequences were individually cloned into the pCMV-MCS vector supplied with the kit (Stratagene). The DGI 3 antisense sequence was cloned into the pCMV-MCS vector. PCR amplification was used to create the DNA insert. The PCR reaction mixture included 100 ng of DGI3 in pcDNA 3.1 Hygro; Forward primer (1 μM; 5'- gcg gcc cag ccg gcc tga cag gat tgt etc cag gac tec ttg tc-3'; SEQ ID NO: 180); Reverse primer (1 μM; 5'-gca att tea tgc ggc cgc atg ttc tgc gaa aaa gcc atg g-3'; SEQ ID NO:181 ); dNTPs (200 μM); ThermalAce Buffer (1 X; Invitrogen); and 2 U ThermalAce enzyme in a total volume of 50 μl. The PCR reaction tube was incubated at 95°C for 5 min, then incubated for 30 cycles at 95°C for 45 sec, 68°C for 45 sec, 72°C for 2 min, and then incubated at 72°C for 3 min. The PCR product which corresponded to the predicted 1.1 kb fragment was gel purified, digested with Sfi I and Not I, and ligated into the pCMV-MCS vector digested with Sfi I and Not I. Correct construction was confirmed by sequencing analysis. The Not I fragment was cut out of pCMV-MCS-DGI-3 and this expression cassette was subcloned into the ITR-containing vector pAAV. The pAAV-DGI-3, pHELPER, and pAAV-RC (control) constructs were then used to transfect HEK-293 cells.

Preparation of cells: HEK-293 cells were seeded onto 100 mm tissue culture plates in 10 ml of DMEM growth medium (3 x 10⁶ cells/plate). Cells were incubated for 48 h and inspected to determine confluency. Cells at 80% confluency were used for transfection. Plasmids (pAAV-DGI-3, pAAV- RC, and pHELPER) to were removed from storage at -20°C. The concentration of each plasmid was adjusted to 1 mg/ml in TE Buffer, pH 7.5. For each plasmid solution, 10 μl (10 μg) was pipetted (30 μg total) into a 15 ml conical tube containing 6 ml room temperature DMEM, and the mixture was vortexed. Next, 90 μl TransFast™ reagent was added, vortexed, and incubated at temperature for 10-15 min. Growth media was aspirated from the cells. The TransFast™/DMEM/DNA mixture was vortexed briefly and 6 ml was overlaid on each plate. The plates were returned to the 37°C incubator for 1 h. While the cells were incubating, DMEM complete growth medium was pre-warmed. To each plate, 12 ml DMEM was added. The plates were returned to the incubator for an additional 72 h. At the 24 h time point, cells lifted from the plate, and the pH of the media changed sufficiently to turn the media yellow. At the 48 h time point, the control lacZ transfected plate was stained using a β-Gal Kit (Invitrogen β-Gal Staining Kit Cat # 45-0450). Transfection efficiency was calculated by counting blue cells and total cells in 5 random views.

Preparation of viral stocks: A dry-ice ethanol bath and a 37°C water bath was prepared. Transfected cells in DMEM growth medium were transferred to a 15 ml conical tube. To transfer the cells, each plate was held at an angle cells while scraping with a cell lifter. The cell suspension was subjected to four rounds of freeze/thaw by alternating the tubes between the dry-ice ethanol bath and the 37°C water bath, vortexing briefly after each thaw. Cellular debris was collected by centrifugation at 10,000 X g for 10 min at room temperature. The supernatant (primary virus stock) was transferred to a fresh tube. Viral stocks were stored at -80°C.

Viral titer measurement: HT1080 cells were plated into 24-well plates at 60% confluence in 0.5 ml DMEM growth medium. Cells were evenly spread in the wells to ensure accurate titer determination, and incubated at 37°C overnight.

Viral stock dilution and application: AAV viral stocks were diluted in 2 ml volumes of DMEM growth medium. For each viral stock, 6 sets of dilutions were made, ranging from 10-2 to 10-7 cells/ml. The volume of each dilution was sufficient to titer the viral stocks in triplicate. Wild-type adenovirus was added to the AAV stock dilutions for a final MOI of 50 adenovirus per cell. This was equivalent to ~2 x 10⁷ adenovirus particles for every 2 ml AAV stock dilution. The amount of virus was sufficient to infect ~3 x 10⁵ cells in a total of three wells. Growth medium was aspirated from wells to be infected. Next, 0.5 ml of each dilution was added to separate wells of the 24-well plates. The titer was performed in triplicate, adding 0.5 ml of each dilution to each of the three wells. The plates were incubated at 37°C for 20 h.

Detection of infected cells: The growth medium was aspirated. Cells were fixed and stained using the β-Gal Staining Kit (Invitrogen) according to the manufacturer's instructions. Blue-stained cells were counted in wells with appropriate cell densities. The number of viral particles (number of stained cells) per milliliters of stock was calculated. Intensity of cell staining was variable among infected cells. Both faintly- as well as intensely-stained cells were counted. Testing for viral induced cell death: HBL100 and MCF7 cells were plated according to the procedure for viral titer measurement (above). HBL100 cells were grown in RPMI and MCF7 cells were grown in EMEM growth medium. Cells were incubated overnight. Viral stock was added to the cells at 100%, and dilutions of 1 :10, 1 :100. and 1 :1000, and cells were returned to the incubator for 72 h. At the end of the incubation, cells were stained or counted using the trypsinization protocol (above). Results are shown in Figures 20A-20B and summarized in Table 20, below. The results indicate that the adeno-associated virus expressing DGI-3 antisense inhibited the growth of various cancer cells, including MCF-7 (breast carcinoma cells), CaCo-2 (colorectal carcinoma cells), PANC-1 (pancreatic carcinoma cells).

TABLE 20

EXAMPLE 14: PENETRATING PEPTIDES. SMALL INTERFERING RNAs, AND ANTISENSE OLIGONUCLEOTIDES Penetrating peptides: In one approach, penetrating peptides are used to inhibit the growth of DGI-3-expressing tumor cells. Penetratin was originally identified as a 16mer peptide derived from the homeodomain of Antennapedia (D. Derossi et al., 1996, J. Biol. Chem. 271 :18188-18193). Penetratin has been reported to efficiently gain entry into cells via pathway(s) independent of endocytosis, receptors, or transporters (D. Derossi et al., 1996, J. Biol. Chem. 271 :18188-18193). The penetratin peptide has recently been used to enhance efficacy of anti-cancer treatments using Smac/DIABLO apoptosis protein (C.R. Arnt et al., 2002, J. Biol. Chem. (e-published manuscript M207578200, JBC Papers in Press, September 5, 2002) and polo-box peptide (J. Yuan et al., 2002, Cancer Res. 62:4186-4190).

In accordance with the present invention, the penetratin sequence is added to the C-terminus of DGI-3 binders such as A1 , G3, and C5 (Figure 21). To test the penetrating peptides, cells are plated at 1 x 10⁵ cells/well in a 12-well tissue culture plate, and incubated overnight. Various concentrations of the peptides are added to each well in serum free medium (approximate range 1 μM to 30 μM). Each peptide is tested in triplicate. At 24 h, the media is refreshed. At 72 h, the cells are counted by trypsinization in the 37°C incubator. For trypsinization, media was aspirated and 900 μl 0.05 % Trypsin/EDTA. (Gibco) was added directly to cells. Plates were returned to the 37°C incubator for 5 min. After removal from the incubator, 100 μl fetal bovine serum is added to neutralize action of trypsin and prevent cell death. Next, 100 μl cells in trypsin/FBS is stained with 100 μl Trypan Blue. From this mixture, 15 μl is used to count cells on a 0.1 mm deep hemacytometer (Bright Line by Reichert).

Small interfering RNAs: In a alternate approach, short silencer RNAs (siRNAs) are used to inhibit expression of DGI-3, and thereby inhibit the growth of DGI-3-expressing cancer cells. RNA interference involves the suppression of specific genes using complementary double stranded RNAs (P.A. Sharp, 1999, Genes Devel. 13:139). Toxicity from RNA interference can be avoided by using synthetic, short (21-23 nucleotide) interfering RNAs, i.e., silencer RNAs (S.M. Elbashir et al., 2001 , Nature 411 :494). To allow long-term suppression of specific genes, siRNA systems (e.g., pSUPER) have been developed (T. Brummelkamp et al., 2002, Science 296:550-553; J.D. Thompson, 2002, Drug Discovery Today 7:912-917) and certain vectors have been made commercially available.

In accordance with the present invention, siRNAs are produced according to the instructions provided with the Silencer siRNA Construction Kit (Catalog # 1620; Ambion). The Ambion in vitro siRNA transcription system generates siRNAs in large quantities using in vitro synthesis followed by a column purification. Oligonucleotides for use this kit (Figure 22) were designed using the siRNA Target Finder and Design Tool (available online at hypertext transfer protocol://ambion.com/techlib/misc/ siRNA_finder.html). The program automatically added the sequence CCTGTCTC to the 3' end of each oligonucleotide. This sequence is complementary to the T7 Promoter Primer supplied in the kit. Generally, oligonucleotides for use as siRNAs are synthesized 29mers. In this case, the oligonucleotides were designed to include 21 bases of the 5' end of DGI-3 or hRas (Figure 22). The siRNAs are amplified following manufacturer's protocols as described in the kit (instructions available online at world wide web. ambion.com/techlib/prot/bp 1620.pdf). The upper and lower oligonucleotides (Figure 22) are then annealed following the manufacturer's instructions.

To test the siRNAs, RiboJuice™ siRNA Transfection Reagent (Novagen; Cat # 71115-3) is used to transfect siRNA complexes into MCF-7 cells. Cells are plated at 5 x 10⁴ cells/well in a 24-well cell culture plate. The plates include 250 μl complete growth media (EMEM with 10% FBS plus 10 μg/ml insulin). Insulin may be obtained from SIGMA (Cat. # I-9278). Plates are incubated overnight, to allow cells to reach between 50%-80% confluency prior to transfection. siRNA stocks (1 μM) are prepared. Pre- warmed 37°C serum-free medium (47 μl) is added to a small sterile Eppendorf microtube. RiboJuice™ (3 μl) is added to the tube, and the combination is mixed thoroughly by gentle vortexing. The mixture is incubated at room temperature for 5 min. siRNA (0.3 to 7.5 μl of 1 μM stock solution) is added to the RiboJuice™/medium mixture, and gently combined to yield a final concentration of 1-25 nM. The siRNA/RiboJuice™/medium mixture is incubated at room temperature for 5-15 min. To each well, 50 μl of the mixture is added in a drop-wise fashion. The drops are distributed evenly along the wells, and the plates are gently rocked to ensure complete distribution. The final volume in the well is 300 μl (250 μl complete medium plus 50 μl transfection mixture). Cells are incubated for 72 h in 37°C in a 5% C0₂ incubator. At 24 h, an additional 300 μl complete medium is added without aspiration. After 72 h, the plates are stained or cells are counted with trypsinization. For trypsinization, 250 μl trypsin is added for 5 min followed by the addition of 50 μl FBS.

DGI-3 antisense oligonucleotides: DGI-3 antisense oligonucleotides are used to inhibit the growth of DGI-3 expressing tumor cells. DGI-3 antisense oligonucleotides (Figure 26) were designed to cluster around the translation start site. Oligonucleotides were produced by Invitrogen using phosphorothioate chemistry, and all were HPLC purified. To test the oligonucleotides, cells are plated at 1 x 10⁵ cells/well in a 12-well tissue culture plate, and incubated overnight. Next, 5 μM of the antisense oligonucleotides is added to each well. Transfection is carried out using the TransFast™ protocol from Promega. Each oligonucleotide is tested in triplicate. At 24 h, the media is refreshed. At 72 h, the cells are counted by trypsinization in the 37°C incubator for 5 min with 900 μl 0.05% trypsin/EDTA (Gibco). After removal from the incubator, 100 μl fetal bovine serum is added to neutralize action of trypsin and prevent cell death. Following this, 100 μl cells in trypsin/FBS are stained with 100 μl Trypan Blue. From this, 15 μl is removed and used to count cells on a 0.1 mm deep hemacytometer (Bright Line by Reichert). EXAMPLE 15: BINDING ANALYSIS USING COMPETITION ELISAs

Cloning and in vitro expression of DGI-3 and PHD: DNA coding sequences for DGI-3 and PHD were cloned into the plVEX 2.3d circular expression vector (Roche) using the Rapid Translation System RTS 500 (RTS 500 E. coli HY kit, Roche Diagnostics GmbH). In each case, the coding sequence was inserted between the 5' Nde I and 3' Xho I sites, in frame with the His-tag provided in the vector. Expression reactions were performed by pipetting 10 ml Feeding Solution and 1 ml Reaction Solution into their respective compartments in the reaction device (provided with the kit). Following this, 10 μg of plasmid DNA was added to each reaction mixture and the reaction was allowed to continue for 24 h at 30°C in the RTS 500 Instrument prior to harvesting the expressed protein.

Solubilizing, refolding and purification of insoluble proteins: Analysis of protein products obtained from the in vitro system indicated that adequate expression levels were obtained. However, high expression levels caused segregation of the protein products to the insoluble fraction of the expression reaction. The expressed proteins were therefore recovered as insoluble pellets, denatured, and subjected to a standard refolding regimen prior to use in panning experiments or other assays. After overnight expression at 4°C in the Roche RTS in vitro expression system, the reaction mixtures were removed from their disposable reaction vessels. The insoluble fractions were pelleted by a 10 min spin at maximum speed in a microfuge. The pellet from each reaction mixture was dissolved by gentle vortexing in 1.0 ml Denaturation Buffer (50 mM TRIS, 8 M urea, 20 mM - mercaptoethanol). Insoluble material was removed by another 10 min spin in the microfuge. The protein concentration of the soluble fraction was determined by Bradley assay (Bio-Rad Protein Assay). For refolding, aliquots of solubilized proteins were diluted to a protein concentration of 25 μg/ml in Denaturation Buffer and the diluted mixtures were loaded into 10,000 MWCO dialysis cassettes (Slide-A-Lyzer®, Pierce). Each 10 ml dilution mixture was dialyzed against three changes of 4L Dialysis/Refold Buffer (50 mM ammonium bicarbonate, 100 mM NaCl, pH 9.0) at 4°C. The first two changes were performed for a minimum of 2 h, each. The final dialysis was carried out overnight. The His-tagged proteins were purified by loading them onto a 1.6 x 1.2 cm Ni-Superflo column (2.4 ml, QIAGEN) equilibrated with Equilibration Buffer (50 mM NaH₂P0₄, 300 mM NaCl, 10 mM imidazole, pH 8.0) followed by elution with Equilibration Buffer containing 500 mM imidazole.

ELISA competition between phage and PHD: A competition assay (Figure 23B) was used to show physical interaction between DGI-3 and the pleckstrin homology domain. Phage were rescued from E. coli TG1 cells containing phagemid DNA for expression of A1 , G3, or C5. On day one, 1 ml fresh overnight culture of cells containing A1 , G3, or C5 phagemid was used to inoculate 20 ml 2xYT medium with 2% glucose plus ampicillin (50 μg/ml). Cells were grown at 37°C with shaking at 250 rpm to OD₆oo= 0.6- 1.0. Next, 0.75 ml M13K07 (helper phage) was added to each tube. Cells were incubated at 37°C with shaking at 250 rpm for 1 h. Cells were spun down at 3000 rpm for 15 min at 4°C. The supemantant was discarded, and cells were resuspended in 20 ml 2xYT medium with ampicillin plus kanamycin (no glucose). Cells were incubated overnight at 37°C, with shaking at 250 rpm. Plates were coated with target (500 ng/well of DGI-3), E-tag antibody (100 ng/well), LDH (500 ng/well) in PBS, or non-fat milk (NFM) control for background binding. Plates were incubated at 4°C overnight.

On day three, the plates were inverted to remove the coating solutions. P1 tubes were spun at 3500 rpm for 15 min at 4°C. The supernatant was poured into fresh tubes. Plates were blocked for 1 h with 2% NFM-PBS. The blocking solution was removed, and 50 μl of PHD protein (800nM final concentration) was added to rows A and B of the plates. To rows B through H, 50 μl dilution buffer (PBS) was added. Using a pipette, 50 μl PHD was mixed and transferred from row B to row C. Pipette tips were discarded. This process was repeated until row G was completed. From row G, 50 μl was discarded, leaving row H as the no competitor control. Next, 50 μl of phage was added per well, and the phage were allowed to incubate at RT for 2 h. The plates were inverted to remove the phage, and washed three times w/ PBS-Tween®20 (0.5% Tween®20). Following this, 100 μl of HRP-conjugated anti-M13 (1 :3000 dilution) was added, and the antibody solution was allowed to incubate at RT for 1 h. The plates were inverted to remove the antibody solution, and washed three times with PBS-Tween®20 (0.5% Tween®20). Next, 100 μl ABTS (BioFx Laboratories, Owings Mills, MD) was added to each well, incubated to allow color development (5-15 min), and plates were read at OD₄os. Results are shown in Figure 23A. The results indicate that the recombinant PHD domain was able to compete with A1 peptide for binding to DGI-3, as predicted. In contrast, G3 and C5 peptides, which bind to different and non- overlapping hot-spots on DGI-3, were not able to compete with recombinant PHD for binding to DGI-3. EXAMPLE 16: BINDING ANALYSIS USING BIACORE

Purification of DGI-3: DGI-3 was cloned into plVEX2.3d vector using PCR amplification. The PCR reaction mixture included 100 ng pcDNA 3.1- DGI-3, Forward primer (1 μM; 5'-cga ttc gtt eta gaa tgt tct gcg aaa aag cca tg-3'; SEQ ID NO:182), Reverse primer (1 μM; 5'-cga ttc gtg gta cct tat gac agg atg tgc tec agg ac-3'; SEQ ID NO:183), dNTPs (200 μM), ThermalAce Buffer (1X; Invitrogen), and 2 U ThermalAce enzyme (Invitrogen) in a total volume of 50 μl. The PCR reaction tube was first incubated at 95°C for 5 min, then incubated for 30 cycles at 95°C for 45 sec, 68°C for 45 sec, 72°C for 1 min, and then incubated at 72°C for 3 min. The PCR product corresponding to the predicted 0.6 kb fragment was gel purified, digested with Kpn\ and Xbal, and ligated to the plVEX2.3d vector (Roche). Construction was confirmed by sequencing analysis. The plVEX2.3d-DGI-3 construct was used for in vitro translation using the Rapid Translation System RTS 500 E. coli HY kit (Roche). The standard kit protocol was followed using 10 μg of plVEX2.3d-DGI-3 DNA (purified by QIAGEN midi column) run on the RTS 500 instrument for 22 h at 30°C with stirring. The insoluble DGI-3 translation product (1 ml) was collected by centrifugation with a tabletop microcentrifuge at top speed for 10 min. The pellet was washed three times with PBS buffer and dissolved in 1 ml of denaturing buffer containing 8 M urea, 20 mM β-mercaptoenthanol, 50 mM Tris-HCl, pH 8. After incubation at room temperature for 2 h, 39 ml column buffer A (500 mM Na₂HP0₄(pH 8), 0.3 M NaCl, 6 M urea and 10 mM imidazole) was added and passed through the Ni-NTA column (Poros® 4.6x100, Perseptive Biosystems). The column was washed with column buffer B (500 mM Na₂HP0₄ (pH 8), 0.3 M NaCl, 6 M urea and 20 mM imidazole) and eluted with Buffer C (500 mM Na₂HP0₄ (pH 8), 0.3 M NaCl, 6 M urea and 300 mM imidazole). The DGI-3 protein peaks were pooled and diluted to 25 μg/ml with refolding buffer and incubated at 4°C overnight. The refolded protein was dialyzed in fresh refolding buffer and lyophilized.

Biacore analysis: DGI-3 protein was diluted to 25 μg/ml in 10 mM sodium acetate (pH 4). A CM5 chip (Biacore, Piscataway, NJ) was activated by 1 :1 NHS/EDC mixture (Biacore) for 4 min. DGI-3 was immobilized on the chip at about 200 RU. The surfaces were then deactivated by passage of 1 M ethanolamine. Synthetic peptide A1 was diluted in HBS-P buffer (10 mM HEPES, pH 7.4, 150 mM NaCl, 0.005% surfactant P20) and injected at 15 μl/min with 4 min association phase and 3 min dissociation phase. Kinetic data was analyzed using the BIAevaluation softer, version 3.0 (Biacore). The binding curve was corrected for background and bulk refractive index contribution by subtraction of blank injection. The model used for was the 1 :1 Langmuir binding interaction describing 1 :1 binding between DGI-3 and A1. Local kinetic data was obtained with chi=1 below the common accepted chi=2. Results are shown in Figure 25. The results indicated that the A1 peptide binds to DGI-3 with an affinity of 80 nM. EXAMPLE 17: RNA ISOLATION AND RT-PCR ANALYSIS

RT-PCR methods: RNA was isolated from NIH3T3, NIH3T3-DGI-3, 293, T47D, HBL100, MCF7, and MCF7-DGI-3 cells using the Absolutely RNA Miniprep Kit (Stratagene; Cat. # 400800) according to the manufacturer's instructions. Cells were collected and spun down to obtain approximately 1 x 10⁷ cells. Cells were lysed using a guanidine thiocyanate buffer (provided by the Stratagene kit), and RNA was bound to a spin column. DNA was removed using DNase I and the RNA washed and eluted. Eluted RNA was used directly in RT-PCR reactions. For the amplification reaction, the TITANIUM one-step RT-PCR Kit

(CLONTECH) was used with the following primers:

DGI 3 outside forward 1 :

5'-tac cat ttt ggc gtg aga get ggt ggt tgg-3' (SEQ ID NO: 162) DGI 3 outside reverse 1 :

5'-agg gag gtg aag gga gtg gtg gag aga gtg-3' (SEQ ID NO:163)

DGI 3 inside forward 1 :

5'-cat ttt ggc gtg aga get ggt ggt tgg-3' (SEQ ID NO: 164)

DGI 3 inside reverse 1 : 5'-cat gag ttg agt gaa gcc tgg aag tgc ctc-3' (SEQ ID NO: 165)

In a 50 μl reaction volume, 0.5 μg RNA was added. The PCR samples were incubated at 50°C for 1 h, 94°C for 5 min, then for 30 cycles at 94°C for 30 sec, 65°C for 30 sec, 68°C for 60 sec, and then at 68°C for 3 min. Secondary PCR was performed with internal primers using Thermozyme Polymerase (Invitrogen) and 1 μl of primary RT-PCR reaction mixture. The PCR samples were incubated at 94°C for 5 min, then for 30 cycles at 94°C for 30 sec 65°C for 30 sec 72°C for 60 sec, and then at 72°C for 3 min. Results are shown in Figures 27-28. The results indicated that

DGI-3 was expressed in the breast cancer cell lines MCF-7 and T47D, as well as MCF-7-DGI-3 transfectants. DGI-3 was also expressed in non- tumorigenic breast cell line HBL-100 and in NIH3T3-DGI-3 transfectants.

EXAMPLE 18: DEVELOPMENT OF A HIGH-THROUGHPUT ALPHAScreen™ ASSAY Target production: The target, DGI-3(His)₆, is produced using a cell- free, in vitro protein expression system (Rapid Translation Systems, RTS 500; Roche Applied Sciences of Indianapolis, IN) as described above. This system employs plVEx Rapid Translation System vectors with 3' His-tags. The system uses instrumentation that incorporates two chambers separated by a semi-permeable membrane. Each chamber contains 1) a reaction compartment for coupled transcription/translation from an enhanced Escherichia coli lysates; and 2) a feeding compartment for substrates and energy components. This allows a continuous-exchange for removing the inhibitory products that may accumulate during protein synthesis, and continuously replacing required substrates and energy sources. A single reaction can continue in this system for up to 24 h and is expected to produce than 200 to 500 μg of product. Proteins greater than 120 kDa in size have been translated with this system, and it is especially suited for cytotoxic proteins that cannot be expressed in traditional cell-based expression systems.

DGI-3(His)₆ product is then tested using Biacore analysis to determine that it has been properly refolded. The target is covalently attached to the chip and A1 peptide will be used at several concentrations to determine the on- and off-rates of binding, which will be used to calculate the Kd for the peptide Interaction with the target. This will be compared to the phage ELISA competition data. Peptide synthesis: Peptides A1 and A1 -biotin are synthesized by automated solid phase synthesis, using either FMOC or tBOC chemistries by a commercial vendor (Peptide Technologies Corporation, Gaithersburg, MD). Peptide purification is accomplished by reverse phase HPLC. The final peptide is analyzed by mass spectrometry and analytical HPLC. Compound library: A library of compounds synthesized by combinatorial chemistry techniques will be obtained from a supplier of in- house designed and synthesized novel drug-like organic molecules for high throughput screening.

Assay development: Recently, bead-based technology based on the principle of luminescent oxygen channeling (Ullman et al., 1996, Clin. Chem. 42:1518-1526; Ullman et al:, 1994, Proc. Natl. Acad. Sci. USA 91 :5426- 5430) has become commercially available (e.g., ALPHAScreen™; Perkin Elmer) This assay format offers the advantages of being homogeneous, fluorescence-based, easy to miniaturize for robotics. In addition, the format does not suffer from the distance limitations of TR-FRET (time-resolved fluorescence resonance energy transfer) assays. The detection limit of ALPHAScreen™ assays is 200 nm, whereas the detection limit of TR-FRET assays is 9 nm.

The DGI-3(His)₆ target is first tested for binding to nickel-conjugated acceptor beads. To 384-well low volume polystyrene microplates (e.g., ProxiPlate™-384; PerkinElmer Life Sciences, Boston, MA), biotin-(His)6- peptide is added at a final concentration of 10^"8 M. Next, DGI-3(His)₆ is added at concentrations ranging from 10^"10 to 10^"5 M (final volume 20 μl/well). Nickel conjugated acceptor beads are added at a final concentration of 20 μg/ml. Plates are incubated for 30 min in the dark at room temperature. After this, streptavidin conjugated donor beads are added at a final concentration of 20 μg/ml. Plates are then incubated for 60 min in the dark at room temperature. At the end of the incubation period, the fluorescence signal at 520 nm is read on a plate reader (e.g., Fusion-α HT; Packard BioScience Co., Meriden, CT). The values are expressed as AlphaScreen™ activity in counts per second (cps) and plotted against Log [DGI-3(His)₆] (M). Assay optimization may include variation of detector reagents (acceptor and donor bead concentrations), competition analysis of A1 versus PHD, order of reagent addition, kinetics of complex formation, and DMSO sensitivity. High-throughput methods: Relative potencies of combinatorial chemistry compounds as compared to DGI-3-binders (e.g., A1 or G3 peptide) are analyzed in a competition system. The system utilizes biotinylated-A1 (b-A1 ) and recombinant DGI-3(His)₆. Detection of the target ligand interaction will be measured in an amplified luminescent proximity homogeneous assay (ALPHAScreen™; BioSignal-Packard, Montreal, Canada). The assay will be performed in 384-well low volume polystyrene microplates (e.g., ProxiPlate™-384; PerkinElmer Life Sciences) with a final volume of 20 μl. Final incubation conditions are 10^'9 to 10^"7 M b-A158, 10^"10 to 10^"8 M DGI-3(His)₆, combinatorial chemistry compounds at 10^~5 M, 0.025 M HEPES (pH 7.4 at 25°C), 0.100 M NaCl, 0.1 % BSA (Cohn Fraction V; Sigma Chemical Co., St. Louis, MO), 5-20 μg/ml nickel-conjugated acceptor beads, and 5-20 μg/ml streptavidin-conjugated donor beads.

For the first step of the assay, b-A1 , DGI-3(His)6, and combi-chem compound are incubated for 2 h at room temperature. Non-specific binding (background) is determined in the presence of 10^"5 M A1 peptide. For the second step, acceptor beads are added and the incubation is continued for 0.5 h. For the final step, donor beads are added and the incubation is continued for an additional 1 h. At the end of the incubation period, the fluorescence signal at 520 nm is read on a plate reader (e.g., Fusion-α HT; Packard BioScience Co.). Primary data are analyzed, e.g., by importation into ActivityBase™ (ID Business Solutions Ltd, Guildford, UK), background corrected, normalized to buffer controls and then expressed as % specific binding. Data is be validated if the Z'-factor (Zhang et al., 1999, J. Biomol. Screen. 4:67-73) for this assay is expected to be greater than 0.7 (Σ = 1- (3σ++3σ.)/|μ+-μ.|) and the signal-to-background (S/B) ratio is expected to be between 30 and 300.

ICgn analysis of "HITS": HITS are evaluated in a competitive binding assay to determine their ability to displace A1 -biotin in a dose-dependent manner. This allows an accurate estimation of HIT potency, IC₅o, relative to A1 displacement curves. The data is fit to a four-parameter non-linear regression analysis ( y = min + (max-min)/(1 +10^Λ((loglC₅o-x)*Hillslope)) ), and is used to determine IC ₀ values. The Z'-factor (Zhang et al., 1999, J. Biomol. Screen. 4:67-73) for this assay is expected to be greater than 0.7 (Z = 1-(3σ++3σ.)/|μ+-μ.|) and the signal-to-background (S/B) ratio is expected to be between 30 and 300.

EXAMPLE 19: SCREENING FOR INHIBITORS THAT BIND AT THE DGI-3 PROMOTER

A cancer cell line overexpressing DGI-3 (e.g., MCF-7-DGI-3) is grown stably in culture media RPMI supplemented with 10% FBS (Gibco). Cells are plated at 1 x 10⁵ cells/well in 12-well tissue culture plates, and grown overnight at 37°C in 5% CO₂. After approximately 18 h, the cells are transfected with 3 μg of pGL3-Enhancer Vector-DGI-3-promoter construct. Cells are grown for an additional 48 h to allow insertion of the plasmid into the genome and selected in one of several selecting agents including but not limited to G418, hygromycin or zeomycin. Stable clones expressing the promoter are isolated, expanded, and used in the assay described below.

A library of compounds synthesized by combinatorial chemistry techniques is obtained from a supplier of in-house designed and synthesized novel drug-like organic molecules for high throughput screening. Cells expressing the DGI-3 promoter are plated at 1 x10³ cells/well in complete medium and allowed to attach overnight at 37°C. The individual small molecules from the combinatorial library are dissolved in DMSO and added to each well so that the final concentration of the molecules is 20 μM. Cells are incubated for an additional 24 h at 37°C.

For the assay, lysis buffer (supplied with the Luciferase Reporter Gene Assay Kit (Roche)) is prepared by diluting 1 part lysis buffer (5 X) with 4 parts distilled water. The culture supernatant is aspirated, and cells are carefully rinsed with 1 X PBS. The remaining PBS is aspirated completely with a fine-tipped pipette. The minimum volume of lysis buffer is added to cover the cells (250 μl for 60 mm dishes or 100 μl for 35 mm dishes). Solubilized cells are transferred to microcentrifuge tubes and incubated at room temperature for a total of 15 min. Note: the time for incubation is calculated from the addition of lysis buffer to the cells. To remove cellular debris, the tubes are spun for 5-10 second in a microcentrifuge at maximum speed. The supernatant is transferred from the tubes to black 96-well microtiter plates.

To immediately start the reaction, 20-50 μl of the cell extract are used. The luciferase assay reagent is prepared by adding 10 ml reaction buffer (Bottle 1 ) to the lyophilized luciferase substrate (Bottle 2; both supplied with Roche kit) and mixing well. Next, 100 μl luciferase assay reagent is added to the cell extracts. The plate is immediately analyzed in a Wallac 1420 multilabel counter (VICTOR 2). Inhibition of the luciferase signal (< 50% of non-treated cells) is considered a hit (i.e., inhibitory agent).

EXAMPLE 20: TRANSFORMATION OF CELLS BY DGI-3 IN COOPERATION WITH RAS Based on DGI-3-partner identification (above), DGI-3 is predicted to be part of the Ras signaling pathway. To test this hypothesis, established methods are used (see, e.g., K. Jacobsen et al., 2002, Oncogene 21: 3058- 3067). DGI-3 and the Ras oncogene are cloned into the pcDNA 3.1 Hygro+ vector between the Hind III and Not I sites. The Kozak initiation sequence (5'-ccaccatgg-3') is added to the constructs to promote translation. The human Ras sequence is obtained from GenBank accession number NM_005343 (Figure 30A). This corresponds to the Homo sapiens v-Ha-ras coding sequence (Harvey rat sarcoma viral oncogene homolog; hRas). The GenBank sequence is used to design cloning primers. The primers extend into the 5' and 3' untranslated regions and generate a 570 bp open reading frame. The hRas sequences is then cloned by PCR amplification using Human Universal QuickClone™ cDNA (CLONTECH), ThermalAce high fidelity polymerase (Invitrogen), Forward primer (5'- ccctgaggagcgatgacggaatataagctg-3'; SEQ ID NO:202) and Reverse primer (5'-gtccccctcacctgcgtcaggagagcacac-3'; SEQ ID NO:203). The resulting PCR product is cloned into a pZErO™ cloning vector (Invitrogen) using blunt end ligation, and cloning is confirmed by sequence analysis.

The activated hRas sequence is then created by changing G to T at base 29 as found in Ras T24 (Figure 30B). PCR amplification is performed using Activated Forward primer (5'-gga ata taa get ggt ggt ggt ggt cgc egg egg tgt ggg caa gag-3'; SEQ ID NO:204) and Reverse cloning primer (5'- gtccccctcacctgcgtcaggagagcacac-3'; SEQ ID NO:203). The resulting PCR product is cloned into the pZErO™-2 vector (Invitrogen) using blunt end ligation. Both hRas and activated hRas sequences are subcloned (separately) into pcDNA 3.1 Hygro vector between the Xba I and Kpn I sites. PCR amplification is used to create the DNA inserts. The PCR reaction includes Forward primer (5'-ggc caa ttc gtc tag aat ggc gac gga ata taa get ggt ggt g-3'; SEQ ID NO:205) and Reverse Primer (5'-gca ttt cag gta cct cag gag age aca cac ttg cag ctc-3'; SEQ ID NO:206).

For transfections, normal rodent fibroblasts, including but not limited to NIH3T3 and Rat 1A (JHU-25; ATCC), are plated at 1-2 x 10⁵ cell per well in a 6-well plate and allowed to adhere overnight at 37°C. The following experimental groups are set up in triplicate: 1 ) cells alone; 2) cells plus DGI- 3; 3) cells plus hRas; 4) cells plus activated hRas; 5) cells plus DGI-3 and hRas; 6) cells plus DGI-3 and activated Ras. To each well, 3 μg of DNA (pcDNA3.1 -DGI-3 and pcDNA3.1-Ras) is transfected using the TransFast™ protocol from Promega. At 24 h, the media is refreshed. Cells are incubated for 10-14 days to allow development of foci. After this, cells are washed, fixed with 3.7% formaldehyde, stained with Giemsa (5% in PBS, GibcoBRL) and the foci are counted. It is expected that activated hRas and DGI-3 will be transforming and the combination will have a significant increase in the number of foci due to cooperation between these factors.

EXAMPLE 21 : SCREENING DGI-3 TRANSFECTANTS WITH A COMBINATORIAL LIBRARY

Normal rodent fibroblasts (e.g., NIH3T3) or other mammalian cells are obtained from ATCC. Cells are grown stably in culture media, such as RPMI supplemented with 10% fetal bovine serum (Gibco). Cells are plated at 1 x 10⁵ cells/well in a 6-well culture dish. DNAs of interest (e.g., pcDNA3.1 Hygro+-DGI-3) are prepared using an endotoxin free Mega-Prep Kit (QIAGEN). After 24 h, transfectants are selected by treatment with 25 μg/ml hygromycin in complete media (Roche; Cat. # 843 555). Concentration of hygromycin is increased to 50 μg/ml in complete media after 1 week. Stable transfectants are allowed to grow for 4 weeks and passaged 4 times before further use.

A combinatorial chemical library is obtained and the compounds are diluted in DMSO to a final concentration of 2 mM or greater. Tranfectants and controls (e.g., NIH3T3-DGI-3 and NIH3T3 cells) are plated in 96- or

384- well plates at concentrations of 3 x 10³ or 3 x 10² cells/well, respectively, in media such as RPMI-1640 medium with 10% fetal bovine serum. After the cells have adhered, the chemical compounds are added at a final concentration of 20 μM/well. After 24 h, cell viability is determined using the ATP assay as described below. The CellTiter-Glo™ Luminescent Cell Viability Assay (Promega

Corporation, Madison, WI) is a homogeneous method of determining the number of viable cells in culture based on quantitation of the ATP present. As ATP is an indicator of metabolically active cells, the CellTiter-Glo™ Assay is designed for use with multi-well formats, making it ideal for automated high-throughput screening (HTS), cell proliferation and cytotoxicity assays. The homogeneous assay procedure involves addition of a single reagent (CellTiter-Glo™ Reagent) directly to cells cultured in serum-supplemented medium. Cell washing, removal of medium or multiple pipetting steps are not required. The system detects as few as 15 cells/well in a 384-well format in 10 minutes after reagent addition and mixing.

For the assay, opaque-walled multi-well plates (e.g., Costar® sterile black clear bottom tissue culture treated polystyrene, Corning Inc., Corning, NY) are used. Mammalian cells in culture medium are added to the plates: 100 μl/well for 96-well plates or 25 μl/well for 384 well plates. Control wells containing only medium are used to establish background luminescence. Test compounds are added to the experimental wells, and incubated with the cell culture. Plates are equilibrated at room temperature for approximately 30 min. An equal volume of CellTiter-Glo™ Reagent is added to each well with cell culture. As examples, 100 μl of Reagent is added to 100 μl of cell culture in the 96-well plates, whereas 25 μl of Reagent is added to 25 μl of cell culture in the 384-well plates. Contents are mixed for 2 min on an orbital shaker to induce cell lysis. The plate is incubated at room temperature for 10 min to stabilize the luminescence signal. The luminescence is recorded using an integration time of 0.25-1 sec/well. A hit (i.e., antagonist agent) is identified as having a 50% or greater inhibitory effect on the transfectants (e.g., NIH3T3-DGI-3) with no effect on control cells (e.g., NIH3T3). SUMMARY

DGI-3 was originally identified as a transcript identified from a subtraction library designated as KIAA0186 (GenBank Ace. No. NMJD21067; Nagase et al., 1996, DNA Res. 3:17-24). The DGI-3 gene has been located to chromosome 20p11.1 (Figure 7) and is predicted to contain 7 exons (Figure 11D). Several structural and functional motifs are predicted from the sequence of DGI-3 including a coil-coiled domain and several phosphorylation sites (Figure 24). As described herein, Northern blot analysis showed that DGI-3 expression was restricted in normal tissues to heart, liver, lung, and brain tissues (Figure 9). In addition, DGI-3 was over-expressed in several human tumor cell lines, with the highest levels observed in breast carcinoma cell lines, MCF-7 and T47D (Figure 8). DGI- 3 was also expressed in the non-tumorigenic breast cells (HBL-100), but it was not expressed in primary human mammary epithelial cells (HMEC; Figure 8). Similar results were seen using RT-PCR analysis, as described herein (Figures 27-28). Additional information about DGI-3 expression in human tumor cells was obtained from GeneExpress™ (GeneLogic; Tables 16-17) and the SOURCE website (Ross et al., 2000, Nature Genetics 24:227-234).

In accordance with this invention, an IMAGE clone was obtained and the protein expressed by in vitro translation (Roche RTS500). The His- tagged protein was purified by affinity chromatography and refolded using sequential dialysis in ammonium biocarbonate. Two highly diverse (10¹⁰ to 10¹¹ diversity) random peptide phage display libraries and one 25mer library (>10¹⁰ diversity) with a constrained cysteine in position 5 were used for 4 rounds of panning. Four peptides, A1 , G3, C5, and D1 , and were found to be specific DGI-3-binders by ELISA. The peptide sequences of A1 , G3, C5, and D1 were determined and analyzed by BLAST searching (Figure 15A). Alignment of the peptide sequences revealed several putative natural partners for DGI-3. Interestingly, the identified partners were members of the GTP exchange factor family of signaling proteins. In addition, three of the peptides were used to identify hot-spot domains on the partner proteins that bound to DGI-3 (Figure 15C). For peptide A1 , this was shown by a competition ELISA using a recombinant version of the pleckstrin homology domain (PHD) of cytohesin-1. Results indicated that recombinant PHD competed with A1 , but not the other peptides, for binding to DGI-3 (Figure 23A). Based on this data, a functional network has been predicted for DGI- 3 (Figure 6B). As described herein, DGI-3 was extensively characterized using the methods of the invention. In one experiment, NIH3T3 cells were transfected with constructs comprising the complete open reading frames of DGI-3 (NIH3T3-DGI-3) and cytohesin-1 (NIH3T3-cytohesin-1 ) inserted into pcDNA3.1 vector. A striking change in morphology was observed for the cells expresssing DGI-3 (Figures 18A-18C). The NIH3T3-DGI-3 cells became round and loosely attached to the substratum. In contrast, the NIH3T3-cytohesin-1 cells retained the morphological characteristics of parental NIH3T3 cells.

To examine the role of DGI-3 in oncogenesis, a full-length version of the gene in the antisense orientation was delivered by various means to cancer cell lines. In one set of experiments, the antisense sequence was delivered by adeno-associated (AAV) virus to MCF-7 (breast carcinoma cells), HBL-100 (non-tumorigenic breast cells), CaCo-2 (colorectal carcinoma cells), and PANC-1 (pancreatic carcinoma cells) cell lines. The AAV:DGI-3 antisense construct was cytotoxic to the three cancer cell lines, but had no effect on the growth of HBL-100 cells (Figure 20B). In a separate series of experiments, the DGI-3 antisense sequence was transiently transfected into MCF-7 and HBL-100 cells along with controls. The pcDNA3.1 -DGI-3 antisense construct reduced the growth of MCF-7 but not HBL-100 cells (Figures 19E-19H). At the same time, pcDNA3.1-A1 and pcDNA3.1-G3 constructs were transiently transfected into MCF-7 and HBL- 100 cells. In both cases, the constructs were found to be cytotoxic in MCF- 7 but not HBL-100 cells (Figures 19E-19F). These results suggested that the DGI-3 may play a different functional role in cancer and non-cancer cells. For example, without wishing to be bound by theory, DGI-3 may function as a maintainance or survival factor for tumor cells via interactions with signaling pathways that respond to external stimuli.

In conclusion, the methods of the invention were used to identify peptides (also called hot-spot ligands or DGI-3-binders), which were used to validate DGI-3 as a cancer target. DGI-3 was initially identified from a subtraction library as a gene of unknown function (Nagase et al., 1996, DNA Res. 3:17-24). Using the Phenogenix® approach of the invention, the identified peptides were used to characterize DGI-3 as a component of several important cellular pathways. The results described herein have shown that DGI-3 plays a critical role in tumor cell growth and survival. The peptides of the invention therefore find use as tools to define endogenous partners and pathways for DGI-3, and as therapeutics for treating cancers responsive to DGI-3.

Advantageously, this invention provides various means to alter (i.e., control) the cellular pathway(s) involving DGI-3. For example, inhibition of DGI-3 activity or expression is expected to inhibit the activity or expression of other components in the DGI-3 pathway. In particular, the activity or expression of DGI-3-partners, such as cytohesin-1 , cytohesin-2, cytohesin- 3, cytohesin-4, Ras-GAP, GNRP, Trio, and/or Rhophilin may be decreased by factors that inhibit DGI-3 expression or activity (e.g., anti-DGI-3 antibodies, DGI-3 antisense sequences, DGI-3 siRNAs, DGI-3 binders, or small molecule DGI-3 antagonists or inhibitors). Related components such as Ras may also be inhibited or deactivated by these factors.

As various changes can be made in the above compositions and methods without departing from the scope and spirit of the invention, it is intended that all subject matter contained in the above description, shown in the accompanying drawings, or defined in the appended claims be interpreted as illustrative, and not in a limiting sense.

The contents of all patents, patent applications, published articles, books, reference manuals, texts and abstracts cited herein are hereby incorporated by reference in their entirety to more fully describe the state of the art to which this invention pertains.

Claims

WHAT IS CLAIMED IS:

1. An isolated nucleic acid comprising a nucleotide sequence encoding an amino acid sequence selected from the group consisting of A1 (SEQ ID NO:3), G3 (SEQ ID NO:4), D9 (SEQ ID NO:5), C5 (SEQ ID NO:6), D1 (SEQ ID NO:7), H2 (SEQ ID NO:8), H3 (SEQ ID NO:9), G12 (SEQ ID NO:10), E7 (SEQ ID NO:11 ), A1 variants (SEQ ID NO:58-SEQ ID NO:109 and SEQ ID NO:110-SEQ ID NO:112), and penetrating variants (SEQ ID NO:114-SEQ ID NO:117).

2. An isolated nucleic acid comprising a nucleotide sequence selected from the group consisting of A1 (SEQ ID NO:28), G3 (SEQ ID NO:29), D9 (SEQ ID NO:30), C5 (SEQ ID NO:31 ), D1 (SEQ ID NO:32), H2 (SEQ ID NO:33), H3 (SEQ ID NO:34), G12 (SEQ ID NO:35), and E7 (SEQ ID NO:36).

3. An isolated nucleic acid comprising at least 15 contiguous nucleotides of a nucleotide sequence selected from the group consisting of A1 (SEQ ID NO:28), G3 (SEQ ID NO:29), D9 (SEQ ID NO:30), C5 (SEQ ID NO:31 ), D1 (SEQ ID NO:32), H2 (SEQ ID NO:33), H3 (SEQ ID NO:34), G12 (SEQ ID NO:35), and E7 (SEQ ID NO:36).

4. An isolated nucleic acid comprising a nucleotide sequence that is at least 80% identical to the nucleotide sequence of claim 2.

5. An isolated nucleic acid comprising a nucleotide sequence that is complementary to the nucleotide sequence of claim 3.

6. A vector comprising the isolated nucleic acid of claim 1.

7. A vector comprising the isolated nucleic acid of claim 2.

8. A vector comprising the isolated nucleic acid of claim 3.

9. A vector comprising the isolated nucleic acid of claim 4.

10. A virus comprising the isolated nucleic acid of claim 1 , wherein the virus is selected from the group consisting of retrovirus, adenovirus, adeno- associated virus, herpes virus, vaccinia virus, polio virus, and sindbis virus.

11. A virus comprising the isolated nucleic acid of claim 2, wherein the virus is selected from the group consisting of retrovirus, adenovirus, adeno- associated virus, herpes virus, vaccinia virus, polio virus, and sindbis virus.

12. A virus comprising the isolated nucleic acid of claim 3, wherein the virus is selected from the group consisting of retrovirus, adenovirus, adeno- associated virus, herpes virus, vaccinia virus, polio virus, and sindbis virus.

13. A virus comprising the isolated nucleic acid of claim 4, wherein the virus is selected from the group consisting of retrovirus, adenovirus, adeno- associated virus, herpes virus, vaccinia virus, polio virus, and sindbis virus.

14. A host cell comprising the vector of claim 6, wherein the host cell is selected from the group consisting of bacterial, yeast, insect, mammalian, and plant cells.

15. A host cell comprising the vector of claim 7, wherein the host cell is selected from the group consisting of bacterial, yeast, insect, mammalian, and plant cells.

16. A host cell comprising the vector of claim 8, wherein the host cell is selected from the group consisting of bacterial, yeast, insect, mammalian, and plant cells.

17. A host cell comprising the vector of claim 9, wherein the host cell is selected from the group consisting of bacterial, yeast, insect, mammalian, and plant cells.

18. A primer comprising the isolated nucleic acid of any one of claims 3 and 5.

19. A probe comprising the isolated nucleic acid of any one of claims 3 and 5.

20. An isolated peptide comprising an amino acid sequence selected from the group consisting of A1 (SEQ ID NO:3), G3 (SEQ ID NO:4), D9 (SEQ ID NO:5), C5 (SEQ ID NO:6), D1 (SEQ ID NO:7), H2 (SEQ ID NO:8), H3 (SEQ ID NO:9), G12 (SEQ ID NO:10), E7 (SEQ ID NO:11 ), A1 variants (SEQ ID NO:58-SEQ ID NO:109 and SEQ ID NO:110-SEQ ID NO:112), and penetrating variants (SEQ ID NO:114-SEQ ID NO:117).

21. An isolated peptide comprising at least 8 contiguous amino acids of an amino acid sequence selected from the group consisting of A1 (SEQ ID NO:3), G3 (SEQ ID NO:4), D9 (SEQ ID NO:5), C5 (SEQ ID NO:6), D1 (SEQ ID NO:7), H2 (SEQ ID NO:8), H3 (SEQ ID NO:9), G12 (SEQ ID NO:10), E7 (SEQ ID NO:11 ), and A1 variants (SEQ ID NO:58-SEQ ID NO:109 and SEQ ID NO:110-SEQ ID NO:112).

22. An isolated peptide which binds to a DGI-3 polypeptide, wherein the peptide comprises an amino acid sequence that is at least 65% identical to the amino acid sequence of claim 20.

23. An antibody that binds to the isolated peptide of claim 20.

24. An antibody that binds to the isolated peptide of claim 21.

25. An antibody that binds to the isolated peptide of claim 22.

26. The antibody of claim 23 which is monoclonal.

27. The antibody of claim 24 which is monoclonal.

28. The antibody of claim 25 which is monoclonal.

29. An isolated polypeptide complex comprising a DGI-3 polypeptide and a peptide comprising an amino acid sequence selected from the group consisting of A1 (SEQ ID NO:3), G3 (SEQ ID NO:4), D9 (SEQ ID NO:5), C5 (SEQ ID NO:6), D1 (SEQ ID NO:7), H2 (SEQ ID NO:8), H3 (SEQ ID NO:9), G12 (SEQ ID NO:10), E7 (SEQ ID NO:11 ), A1 variants (SEQ ID NO:58- SEQ ID NO:109 and SEQ ID NO:110-SEQ ID NO:112), and penetrating variants (SEQ ID NO:114-SEQ ID NO:117).

30. An isolated polypeptide complex comprising a DGI-3 polypeptide and a partner polypeptide selected from the group consisting of cytohesin-1 , cytohesin-2, cytohesin-3, cytohesin-4, Ras-GAP, GNRP, Trio, and Rhophilin.

31. An antibody that binds to the isolated polypeptide complex of claim 29.

32. An antibody that binds to the isolated polypeptide complex of claim 30.

33. The antibody of claim 31 which is monoclonal.

34. The antibody of claim 32 which is monoclonal.

35. A peptide library comprising amino acid sequence variants of a peptide selected from the group consisting of A1 (SEQ ID NO:3), G3 (SEQ ID NO:4), D9 (SEQ ID NO:5), C5 (SEQ ID NO:6), D1 (SEQ ID NO:7), H2 (SEQ ID NO:8), H3 (SEQ ID NO:9), G12 (SEQ ID NO:10), E7 (SEQ ID NO:11 ), A1 variants (SEQ ID NO:58-SEQ ID NO:109 and SEQ ID NO:110- SEQ ID NO:112), and penetrating variants (SEQ ID NO:114-SEQ ID NO:117).

36. The peptide library of claim 35 which is selected from the group consisting of a display peptide library and a non-display peptide library.

37. A method of identifying a DGI-3-binder comprising: screening the peptide library of claim 35 for an amino acid sequence that binds to DGI-3, wherein binding indicates identification of a DGI-binder.

38. A method of identifying a DGI-3-binder comprising: screening the peptide library of claim 36 for an amino acid sequence that binds to DGI-3, wherein binding indicates identification of a DGI-binder.

39. A method of identifying a DGI-3 antagonist agent comprising:

1 ) contacting the isolated polypeptide complex of claim 29 with test agents to identify a test agent that disrupts the complex;

2) contacting cancer cells with the test agent identified in (1 ) to determine effect on cell growth, wherein inhibition of cell growth indicates identification of a DGI-3 antagonist agent.

40. A method of identifying a DGI-3 antagonist agent comprising:

1 ) contacting the isolated polypeptide complex of claim 30 with test agents to identify a test agent that disrupts the complex;

2) contacting cancer cells with the test agent identified in (1 ) to determine effect on cell growth, wherein inhibition of cell growth indicates identification of a DGI-3 antagonist agent.

41. A method of identifying a DGI-3 antagonist agent comprising:

1 ) contacting a DGI-3 polypeptide with the isolated peptide of claim 20 to form a complex;

2) contacting the complex of (1) with test agents to identify a test agent that disrupts the complex; and

3) contacting cancer cells with the test agent identified in (2) to determine effect on cell growth, wherein inhibition of cell growth indicates identification of a DGI-3 antagonist agent.

42. A method of identifying a DGI-3 antagonist agent comprising:

1) contacting a DGI-3 polypeptide with the isolated peptide of claim 21 to form a complex;

2) contacting the complex of (1 ) with test agents to identify a test agent that disrupts the complex; and

3) contacting cancer cells with the test agent identified in (2) to determine effect on cell growth, wherein inhibition of cell growth indicates identification of a DGI-3 antagonist agent.

43. A method of identifying a DGI-3 antagonist agent comprising:

1 ) contacting a DGI-3 polypeptide with a partner polypeptide selected from the group consisting of cytohesin-1 , cytohesin-2, cytohesin-3, cytohesin-4, Ras-GAP, GNRP, Trio, and Rhophilin to form a complex;

2) contacting the complex of (1 ) with test agents to identify a test agent that disrupts the complex; and

3) contacting cancer cells with the test agent identified in (2) to determine effect on cell growth, wherein inhibition of cell growth indicates identification of a DGI-3 antagonist agent.

44. A method of diagnosing cancer comprising:

1 ) incubating the antibody of claim 33 with a biological sample under conditions to allow the antibody to associate with a DGI-3 polypeptide complex in the sample; and

2) measuring levels of the antibody-complex association formed in (1 ), wherein an increase in these levels compared to standard levels indicates diagnosis of cancer.

45. A method of diagnosing cancer comprising:

1 ) incubating the antibody of claim 34 with a biological sample under conditions to allow the antibody to associate with a DGI-3 polypeptide complex in the sample; and

2) measuring levels of the antibody-complex association formed in (1 ), wherein an increase in these levels compared to standard levels indicates diagnosis of cancer.

46. A method of diagnosing cancer comprising:

1 ) incubating a primer comprising at least 15 contiguous nucleotides of a nucleotide sequence selected from the group consisting of SEQ ID NO:38-SEQ ID NO:41 , SEQ ID NO:152-SEQ ID NO:153, SEQ ID NO:156-SEQ ID NO:157, SEQ ID NO:160-SEQ ID NO:165, SEQ ID NO:174-175, and SEQ ID NO:178-183 with a biological sample under conditions to allow the primer to amplify a DGI-3 nucleic acid in the sample; and

2) measuring levels of the amplification product formed in (1 ), wherein an increase in these levels compared to standard levels indicates diagnosis of cancer.

47. A method of diagnosing cancer comprising:

1 ) incubating an anti-DGI-3 antibody with a biological sample under conditions to allow the antibody to bind to a DGI-3 polypeptide in the sample and form a complex;

2) measuring levels of the DGI-3-antibody complex formed in (1 ), wherein an increase in these levels compared to standard levels indicates diagnosis of cancer.

48. A method of diagnosing cancer comprising:

1 ) incubating a probe comprising at least 15 contiguous nucleotides of a nucleotide sequence SEQ ID NO: 184 with a biological sample under conditions to allow the probe to hybridize a DGI-3 nucleic acid in the sample; and

2) measuring levels of the nucleic acid hybrid formed in (1 ), wherein an increase in these levels compared to standard levels indicates diagnosis of cancer.

49. The method of any one of claims 44-48, wherein the cancer is selected from the group consisting of brain cancer, breast cancer, lung cancer, pancreatic cancer, bladder cancer, and kidney cancer, rectal cancer, colorectal cancer, prostate cancer, ovarian cancer, cervical cancer, endometrial cancer, vulvar cancer, leukemia, lymphoma, head and neck cancer, bone cancer, and skin cancer.

50. A pharmaceutical composition comprising the antibody of claim 27 and a physiologically acceptable carrier, excipient, or diluent.

51. A pharmaceutical composition comprising the antibody of claim 33 and a physiologically acceptable carrier, excipient, or diluent.

52. A pharmaceutical composition comprising the antibody of claim 34 and a physiologically acceptable carrier, excipient, or diluent.

53. A pharmaceutical composition comprising the peptide of claim 21 and a physiologically acceptable carrier, excipient, or diluent.

54. A pharmaceutical composition comprising the vector of claim 8 and a physiologically acceptable carrier, excipient, or diluent.

55. A pharmaceutical composition comprising the virus of claim 12 and a physiologically acceptable carrier, excipient, or diluent.

56. A pharmaceutical composition comprising the host cell of claim 16 and a physiologically acceptable carrier, excipient, or diluent.

57. A method of treating cancer comprising: administering the pharmaceutical composition of any one of claims 50-56 in an amount sufficient to treat the cancer.

58. The method of claim 57, wherein the cancer is selected from the group consisting of brain cancer, breast cancer, lung cancer, pancreatic cancer, bladder cancer, and kidney cancer, rectal cancer, colorectal cancer, prostate cancer, ovarian cancer, cervical cancer, endometrial cancer, vulvar cancer, leukemia, lymphoma, head and neck cancer, bone cancer, and skin cancer.

59. The method of claim 57, wherein the pharmaceutical composition is formulated as a liposome.

60. A method of treating cancer comprising: administering a pharmaceutical composition comprising a DGI-3 antisense sequence and a physiologically acceptable carrier, excipient, or diluent, in an amount sufficient to treat the cancer, wherein the antisense sequence comprises at least 15 contiguous nucleotides of a nucleotide sequence selected from the group consisting of SEQ ID NO:126-SEQ ID NO:131.

61. A method of treating cancer comprising: administering a pharmaceutical composition comprising a DGI-3 antisense sequence and a physiologically acceptable carrier, excipient, or diluent, in an amount sufficient to treat the cancer, wherein the antisense sequence comprises a nucleotide sequence that shares 80% sequence identity with the nucleotide sequence of claim 60.

62. A method of treating cancer comprising: administering a pharmaceutical composition comprising a DGI-3 antisense sequence and a physiologically acceptable carrier, excipient, or diluent, in an amount sufficient to treat the cancer, wherein the antisense sequence comprises at least 50 contiguous nucleotides of the nucleotide sequence of SEQ ID NO:143.

63. A method of treating cancer comprising: administering a pharmaceutical composition comprising a DGI-3 antisense sequence and a physiologically acceptable carrier, excipient, or diluent, in an amount sufficient to treat the cancer, wherein the antisense sequence comprises a nucleotide sequence that shares 80% sequence identity with the nucleotide sequence of claim 62.

64. A method of treating cancer comprising: administering a pharmaceutical composition comprising a DGI-3 silencer sequence and a physiologically acceptable carrier, excipient, or diluent, in an amount sufficient to treat the cancer, wherein the silencer sequence is a transcription product of a nucleotide sequence comprising at least 15 contiguous nucleotides of any one of SEQ ID NO:118-SEQ ID NO:125.

65. A method of treating cancer comprising: administering a pharmaceutical composition comprising a DGI-3 silencer sequence and a physiologically acceptable carrier, excipient, or diluent, in an amount sufficient to treat the cancer, wherein the silencer sequence is a transcription product of nucleotide sequence that shares 80% sequence identity with the nucleotide sequence of claim 64.

66. A method of treating cancer comprising: administering a pharmaceutical composition comprising an anti-DGI-3 antibody, and a physiologically acceptable carrier, excipient, or diluent, in an amount sufficient to treat the cancer.

67. The method of any one of claims 60-66, wherein the cancer is selected from the group consisting of brain cancer, breast cancer, lung cancer, pancreatic cancer, bladder cancer, and kidney cancer, rectal cancer, colorectal cancer, prostate cancer, ovarian cancer, cervical cancer, endometrial cancer, vulvar cancer, leukemia, lymphoma, head and neck cancer, bone cancer, and skin cancer.

68. The method of any one of claims 60-66, wherein the pharmaceutical composition is formulated as a liposome.

69. The method of any one of claims 60-66, wherein the sequence is expressed by a vector or a virus.

70. The method of claim 69, wherein the virus is selected from the group consisting of retrovirus, adenovirus, adeno-associated virus, herpes virus, vaccinia virus, polio virus, and sindbis virus.

71. A method of identifying a DGI-3 inhibitory agent

1 ) contacting a DGI-3 nucleic acid comprising at least 50 contiguous nucleotides of a nucleotide sequence selected from the group consisting of SEQ ID NO: 141 , 142, and 161 with test agents to identify a test agent that binds to the nucleic acid;

2) contacting cancer cells with the test agent identified in (1 ) to determine effect on cell growth, wherein inhibition of cell growth indicates identification of a DGI-3 inhibitory agent.

72. A method of identifying a DGI-3 antagonist agent comprising:

1 ) contacting a DGI-3 polypeptide with test agents to identify a test agent that binds to the DGI-3 polypeptide; and

2) contacting cancer cells with the test agent identified in (1 ) to determine effect on cell growth, wherein inhibition of cell growth indicates identification of a DGI-3 antagonist agent.

73. A method of identifying a DGI-3 antagonist agent comprising:

1 ) contacting cells expressing DGI-3 polypeptide with test agents; and

2) identifying a test agent which inhibits the growth of the cells, wherein inhibition of cell growth indicates identification of a DGI-3 antagonist agent.

74. A method of identifying a DGI-3 antagonist agent comprising:

1 ) contacting a pleckstrin homology domain polypeptide comprising an amino acid sequence RRWFILTDNCL (SEQ ID NO:189) with test agents to identify a test agent that binds to the polypeptide and forms a complex; and2) contacting the polypeptide-test agent complex formed in (1 ) with a DGI-3 polypeptide to determine the effect on polypeptide-DGI-3 complex formation, wherein inhibition of polypeptide-DGI-3 complex formation indicates identification of a DGI-3 antagonist agent.

75. A method of identifying a DGI-3 antagonist agent comprising:

1 ) screening a computer library of test agents to identify a test agent that is predicted to bind to a pleckstrin homology domain polypeptide comprising an amino acid sequence RRWFILTDNCL (SEQ ID NO:189);

2) contacting the test agent identified in (1 ) to confirm that the test agent binds to the polypeptide and forms a complex; and

3) contacting the polypeptide-test agent complex formed in (2) with a DGI-3 polypeptide to determine the effect on polypeptide-DGI-3 complex formation, wherein inhibition of polypeptide-DGI-3 complex formation indicates identification of a DGI-3 antagonist agent.

76. A method of identifying a DGI-3 antagonist agent comprising:

1 ) contacting a DGI-3 polypeptide with test agents to identify a test agent that binds to the polypeptide and forms a complex;

2) contacting the complex of (1 ) with the isolated peptide of claim 20 to determine the effect on peptide-polypeptide complex formation, wherein inhibition of peptide-polypeptide complex formation indicates identification of a DGI-3 antagonist agent.

77. A method of identifying a DGI-3 antagonist agent comprising:

1 ) contacting a DGI-3 polypeptide with test agents to identify a test agent that binds to the polypeptide and forms a complex;

2) contacting the complex of (1 ) with the isolated peptide of claim 21 to determine the effect on peptide-polypeptide complex formation, wherein inhibition of peptide-polypeptide complex formation indicates identification of a DGI-3 antagonist agent.

78. A method for inhibiting activity in a mammalian cell of a gene whose activity is regulated by DGI-3, comprising: decreasing DGI-3 activity in the cell, and thereby inhibiting activity of the gene.