WO1999055869A1 - Novel polypeptide growth factors and materials and methods for making them - Google Patents

Abstract

Polypeptide growth factors, methods of making them, polynucleotides encoding them, and methods of using them are disclosed. These polypeptides include the human and mouse zalpha5 polypeptides shown in SEQ ID NO:2 and SEQ ID NO:15, respectively, as well as homologs and fragments thereof. Multimers of the polypeptides are also disclosed. The polypeptides, multimeric proteins, and polynucleotides can be used in the study and regulation of cell and tissue development, as components of cell culture media, and as diagnostic agents.

Description

Description NOVEL POLYPEPTIDE GROWTH FACTORS

AND MATERIALS AND METHODS FOR MAKING THEM

BACKGROUND OF THE INVENTION

In multicellular animals, cell growth, differentiation, and migration are controlled by polypeptide growth factors. These growth factors play a role in both normal development and pathogenesis, including the development of solid tumors.

Polypeptide growth factors influence cellular events by binding to cell-surface receptors. Binding initiates a chain of signalling events within the cell, which ultimately results in phenotypic changes such as cell division, protease production, cell migration, expression of cell surface proteins, and production of additional growth factors.

Angiogenesis, the sprouting of capillaries from existing blood vessels, is one such growth factor- dependent developmental process. During angiogenesis, vascular endothelial cells re-enter the cell cycle, degrade underlying basement membrane, and migrate to form new capillary sprouts. These cells then differentiate, and mature vessels are formed. This process of growth and differentiation is regulated by a balance of pro- angiogenic and anti-angiogenic factors. Angiogenesis occurs during embryonic development, as well as in the adult organism during pregnancy, the female reproductive cycle, and wound healing. In addition, angiogenesis occurs during a variety of pathological conditions, including diabetic retinopathy, macular degeneration, atherosclerosis, psoriasis, rheumatoid arthritis, and solid tumor growth. For review, see Breier et al . , Thrombosis and Haemostasis 213:678-683, 1997. Angiogenesis is regulated by the vascular endothelial growth factors

(VEGFs) and the angiopoietins . The VEGFs act through at least three cell surface receptors, designated Flt-1, Flk-

1, and Flt-4. The expression of these receptors is limited to certain cell types and/or developmental stages, thereby defining the functions of the ligands. Data obtained from receptor- and growth factor-deficient mice indicate that the VEGFs are essential for vascular development in the embryo. Angiopoietin-1 (Ang-1) , acting through the Tie-2 receptor (also known as Tek) , is believed to regulate a later stage of vascular development

(reviewed by Hanahan, Science 277 :48-50 , 1997), directing the maturation and stabilization of blood vessels through its action on endothelial cells and the surrounding matrix or mesenchyme. The recently discovered angiopoietin-2 (Ang-2) is an antagonist of Tie-2-mediated activity. Ang- 2 causes a loosening of vessel structure and loss of contact between endothelial cells and the matrix, making the endothelial cells more accessible to VEGF . This destabilization is an initial step in angiogenesis, and both VEGF and Ang-2 are up-regulated at sites of ongoing angiogenesis. Ang-2 is also highly expressed during vascular regression in non-productive ovarian follicles.

Cell differentiation and maturation are also under control of growth factors. For example, the hematopoietic factors erythropoietin, thrombopoietin, and G-CSF stimulate the production of erythrocytes, platelets, and neutrophils, respectively, from precursor cells in the bone marrow. Development of mature cells from pluripotent progenitors may require the presence of a plurality of factors .

In addition to their role in angiogenesis, the angiopoietins may be regulators of hematopoiesis . Endothelial cells and hematopoietic stem cells are believed to be derived from a common precursor cell, and Tie receptors are expressed on both cell types. Tie receptors are expressed in several leukemia cell lines with predominantly megakaryoblastic markers (Batard et al . , Blood 82:2212-2220, 1996; Kukk et al . , Bri t . J.

Haema tol . 9_.8:195-203, 1997). Analysis of Tie expression in hematopoietic progenitor cells indicates the presence of Tie-mediated pathways in both early hematopoiesis and differentiation and/or proliferation of B cells (Hashiyama et al., Blood 82:93-101, 1996).

The role of growth factors in controlling cellular processes makes them likely candidates and targets for therapeutic intervention. Platelet-derived growth factor, for example, has been disclosed for the treatment of periodontal disease (U.S. Patent No. 5,124,316) and gastrointestinal ulcers (U.S. Patent No. 5,234,908) . Inhibition of PDGF receptor activity has been shown to reduce intimal hyperplasia in injured baboon arteries (Giese et al . , Restenosis Summit VIII, Poster Session #23, 1996; U.S. Patent No. 5,620,687) . Vascular endothelial growth factors have been shown to promote the growth of blood vessels in ischemic limbs (Isner et al . , The Lancet 3_4_8: 370-374 , 1996), and have been proposed for use as wound-healing agents, for treatment of periodontal disease, for promoting endothelialization in vascular graft surgery, and for promoting collateral circulation following myocardial infarction (WIPO Publication No. WO 95/24473; U.S. Patent No. 5,219,739). VEGFs are also useful for promoting the growth of vascular endothelial cells in culture. A soluble VEGF receptor (soluble flt-1) has been found to block binding of VEGF to cell-surface receptors and to inhibit the growth of vascular tissue in vitro {Biotechnology News 16_.(17) :5-6, 1996). Experimental evidence suggests that inhibition of angiogenesis may be used to block tumor development {Biotechnology News, Nov.

13, 1997) and that angiogenesis is an early indicator of cervical cancer {Br. J. Cancer 1 - 1410-1415 , 1997). The hematopoietic cytokine erythropoietin has been developed for the treatment of anemias (e.g., EP 613,683). More recently, thrombopoietin has been shown to stimulate the production of platelets in vivo (Kaushansky et al . , Na ture

369:568-571, 1994) .

In view of the proven clinical utility of growth factors, there is a need in the art for additional such molecules for use as both therapeutic agents and research tools and reagents. Growth factors are used in the laboratory to study developmental processes, and in laboratory and industry settings as components of cell culture media.

SUMMARY OF THE INVENTION

It is an object of the present invention to provide novel growth factors, polynucleotides encoding them, and methods of making them.

It is another object of the invention to provide compositions and methods for modulating the proliferation, differentiation, migration, and metabolism of responsive cell types and for regulating tissue development. Within one aspect of the invention there is provided an isolated polypeptide of at least 15 amino acid residues comprising an epitope-bearing portion of a polypeptide of SEQ ID NO : 2 or SEQ ID NO: 15. Within one embodiment, the polypeptide comprises at least 15 contiguous residues of SEQ ID NO: 2, such as residues 19- 42, 88-116, or 404-430.

Within a second aspect of the invention there is provided an isolated polypeptide comprising a sequence of amino acids of the formula B_x-L_y-C_z, wherein B is at least 70% identical to residues m to n of SEQ ID NO : 2 , wherein m is from 15 to 23 and n is from 199-207; L is at least 70% identical to residues o to p of SEQ ID NO: 2, wherein o is n+1 and p is from 234 to 242; C is at least 70% identical to residues q to r of SEQ ID NO : 2 , wherein q is from p+1 to 246 and r is from 408 to 460; and each of x, y, and z is individually 0 or 1 , subject to the limitations that at least one of x and z is 1 and if both x and z are 1, y is 1. Within certain embodiments, B is at least 80% identical to residues m to n of SEQ ID NO : 2 , L is at least 80% identical to residues o to p of SEQ ID NO : 2 , and C is at least 80% identical to residues q to r of SEQ ID NO : 2. Within other embodiments, x is 1 and B comprises residues 23-199 of SEQ ID NO : 2 , or z is 1 and C comprises residues 246-408 of SEQ ID NO : 2. Within other embodiments, the polypeptide comprises residues XI -X2 of SEQ ID NO : 2 , wherein XI is selected from the group consisting of 15, 16, 17, 18, 19, 20, 21, 22, 23, 200, 201, 202, 203, 204, 205, 206, 207, 208, 225, 231, 232, 233, 235, 236, 237, 238, 239, 240, 241, 242, 243, and 246, and X2 is from 408 to 460; or XI is selected from the group consisting of 15, 16, 17, 18, 19, 20, 21, 22, and 23, and X2 is selected from the group consisting of 199, 200, 201, 202, 203, 204, 205, 206, 207, 224, 230, 231, 232, 234, 235, 236, 237, 238, 239, 240, 241, 242, and 408 through 460. The polypeptide may comprise cysteine residues at positions corresponding to residues 246, 274, 394, and 408 of SEQ ID NO : 2. Within another embodiment, the polypeptide is glycosylated. Within other embodiments, the isolated polypeptide is covalently linked to a moiety selected from the group consisting of affinity tags, toxins, radionuclides , enzymes, and fluorophores . Preferred affinity tags include polyhistidine, protein A, glutathione S transferase, substance P, and an immunoglobulin heavy chain constant region. Within a related embodiment, the isolated polypeptide further comprises a proteolytic cleavage site between the polypeptide and the affinity tag.

Within a third aspect of the invention there is provided an isolated multimeric protein comprising a first polypeptide as disclosed above non-covalently associated with a second polypeptide, wherein the protein modulates cell proliferation, differentiation, migration, adhesion, or metabolism. Within one embodiment, the first and second polypeptides are the same. Within an alternative embodiment, the first and second polypeptides are different .

Within a fourth aspect of the invention there is provided an expression vector comprising the following operably linked elements: (a) a transcription promoter; (b) a DNA segment encoding a polypeptide as disclosed above; and (c) a transcription terminator. Within one embodiment the expression vector further comprises a secretory signal sequence operably linked to the DNA segment. Within another embodiment, the polypeptide comprises a sequence of amino acid residues selected from the group consisting of residues 23-238 of SEQ ID NO : 2 and residues 17-238 of SEQ ID NO : 2. Within a further embodiment, the polypeptide comprises a sequence of amino acid residues selected from the group consisting of residues 23-460 of SEQ ID NO : 2 and residues 17-460 of SEQ ID NO:2.

Within a fifth aspect of the invention there is provided a protein produced by a method comprising the steps of (a) culturing a cell containing an expression vector as disclosed above, and (b) isolating the protein encoded by the DNA segment and produced by the cell.

Within a sixth aspect of the invention there is provided a cultured cell into which has been introduced an expression vector as disclosed above, wherein the cell expresses the DNA segment and produces a polypeptide encoded by the DNA segment .

Within a seventh aspect of the invention there is provided a method of making a protein comprising the steps of (a) culturing a cell containing an expression vector as disclosed above, and (b) isolating the protein encoded by the DNA segment and produced by the cell .

Within an eighth aspect of the invention there are provided antibodies that bind to the polypeptides and/or multimeric proteins disclosed above. Within one embodiment, the antibodies specifically bind to a 7

polypeptide as shown in SEQ ID NO : 2 from residue 18 through residue 460 or to a fragment thereof.

These and other aspects of the invention will become evident upon reference to the following detailed description of the invention.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 shows an alignment of representative human (SEQ ID NO:2) and mouse (SEQ ID NO:15) zalpha5 amino acid sequences. Amino acid residues are represented by the conventional single-letter codes.

Fig. 2 illustrates the vector pHB12-8.

DETAILED DESCRIPTION OF THE INVENTION Prior to setting forth the invention in detail, it may be helpful to the understanding thereof to define the following terms:

The term "affinity tag" is used herein to denote a polypeptide segment that can be attached to a second polypeptide to provide for purification of the second polypeptide or provide sites for attachment of the second polypeptide to a substrate. In principal, any peptide or protein for which an antibody or other specific binding agent is available can be used as an affinity tag. Affinity tags include a poly-histidine tract, protein A (Nilsson et al . , EMBO J. 4_.: 1075, 1985; Nilsson et al . ,

Methods Enzymol . 198 :3 , 1991), glutathione S transferase (Smith and Johnson, Gene 62:31, 1988), Glu-Glu affinity tag (Grussenmeyer et al . , Proc . Na tl . Acad . Sci . USA 2:7952-4, 1985) (SEQ ID NO:9), substance P, Flag™ peptide

(Hopp et al . , Biotechnology 6_.: 1204-1210 , 1988; reagents available from Eastman Kodak, New Haven, CT) , streptavidin binding peptide, maltose binding protein, cellulose binding protein, thioredoxin, ubiquitin, T7 polymerase, or other antigenic epitope or binding domain. See, in general, Ford et al . , Protein Expression and Purifica tion 2_.: 95-107, 1991. DNAs encoding affinity tags are available from commercial suppliers (e.g., Pharmacia Biotech, Piscataway, NJ) .

The term "allelic variant" is used herein to denote any of two or more alternative forms of a gene occupying the same chromosomal locus. Allelic variation arises naturally through mutation, and may result in phenotypic polymorphism within populations . Gene mutations can be silent (no change in the encoded polypeptide) or may encode polypeptides having altered amino acid sequence. The term allelic variant is also used herein to denote a protein encoded by an allelic variant of a gene .

The terms " amino-terminal " and " carboxyl - terminal" are used herein to denote positions within polypeptides. Where the context allows, these terms are used with reference to a particular sequence or portion of a polypeptide to denote proximity or relative position. For example, a certain sequence positioned carboxyl- terminal to a reference sequence within a polypeptide is located proximal to the carboxyl terminus of the reference sequence, but is not necessarily at the carboxyl terminus of the complete polypeptide.

"Angiogenic" denotes the ability of a compound to stimulate the formation of new blood vessels from existing vessels, acting alone or in concert with one or more additional compounds. Angiogenic activity is measurable as endothelial cell activation, stimulation of protease secretion by endothelial cells, endothelial cell migration, capillary sprout formation, and endothelial cell proliferation. Angiogenesis can also be measured using any of several in vivo assays as disclosed herein.

A "complement" of a polynucleotide molecule is a polynucleotide molecule having a complementary base sequence and reverse orientation as compared to a reference sequence. For example, the sequence 5¹

ATGCACGGG 3 ' is complementary to 5 ' CCCGTGCAT 3 ' . The term "corresponding to", when applied to positions of amino acid residues in sequences, means corresponding positions in a plurality of sequences when the sequences are optimally aligned. The term "degenerate nucleotide sequence" denotes a sequence of nucleotides that includes one or more degenerate codons (as compared to a reference polynucleotide molecule that encodes a polypeptide) . Degenerate codons contain different triplets of nucleotides, but encode the same amino acid residue (i.e., GAU and GAC triplets each encode Asp) .

A "DNA segment" is a portion of a larger DNA molecule having specified attributes. For example, a DNA segment encoding a specified polypeptide is a portion of a longer DNA molecule, such as a plasmid or plasmid fragment, that, when read from the 5' to the 3' direction, encodes the sequence of amino acids of the specified polypeptide .

The term "expression vector" is used to denote a DNA molecule, linear or circular, that comprises a segment encoding a polypeptide of interest operably linked to additional segments that provide for its transcription. Such additional segments include promoter and terminator sequences, and may also include one or more origins of replication, one or more selectable markers, an enhancer, a polyadenylation signal, etc. Expression vectors are generally derived from plasmid or viral DNA, or may contain elements of both.

The term "isolated", when applied to a polynucleotide, denotes that the polynucleotide has been removed from its natural genetic milieu and is thus free of other extraneous or unwanted coding sequences, and is in a form suitable for use within genetically engineered protein production systems. Such isolated molecules are those that are separated from their natural environment and include cDNA and genomic clones. Isolated DNA molecules of the present invention are free of other genes 10

with which they are ordinarily associated, but may include naturally occurring 5 ' and 3 ' untranslated regions such as promoters and terminators. The identi ication of associated regions will be evident to one of ordinary skill in the art (see for example, Dynan and Tijan, Na ture

316:774-78, 1985) .

An "isolated" polypeptide or protein is a polypeptide or protein that is found in a condition other than its native environment, such as apart from blood and animal tissue. In a preferred form, the isolated polypeptide is substantially free of other polypeptides, particularly other polypeptides of animal origin. It is preferred to provide the polypeptides in a highly purified form, i.e. greater than 95% pure, more preferably greater than 99% pure. When used in this context, the term "isolated" does not exclude the presence of the same polypeptide in alternative physical forms, such as dimers or alternatively glycosylated or derivatized forms.

"Operably linked", when referring to DNA segments, indicates that the segments are arranged so that they function in concert for their intended purposes, e.g., transcription initiates in the promoter and proceeds through the coding segment to the terminator.

A "polynucleotide" is a single- or double- stranded polymer of deoxyribonucleotide or ribonucleotide bases read from the 5' to the 3' end. Polynucleotides include RNA and DNA, and may be isolated from natural sources, synthesized in vi tro, or prepared from a combination of natural and synthetic molecules. A "polypeptide" is a polymer of amino acid residues joined by peptide bonds, whether produced naturally or synthetically. Polypeptides of less than about 10 amino acid residues are commonly referred to as "peptides" . The term "promoter" is used herein for its art- recognized meaning to denote a portion of a gene containing DNA sequences that provide for the binding of 11

RNA polymerase and initiation of transcription. Promoter sequences are commonly, but not always, found in the 5' non-coding regions of genes.

A "protein" is a macromolecule comprising one or more polypeptide chains. A protein may also comprise non- peptidic components, such as carbohydrate groups. Carbohydrates and other non-peptidic substituents may be added to a protein by the cell in which the protein is produced, and will vary with the type of cell. Proteins are defined herein in terms of their amino acid backbone structures; substituents such as carbohydrate groups are generally not specified, but may be present nonetheless.

A "secretory signal sequence" is a DNA sequence that encodes a polypeptide (a "secretory peptide") that, as a component of a larger polypeptide, directs the larger polypeptide through a secretory pathway of a cell in which it is synthesized. The larger polypeptide is commonly cleaved to remove the secretory peptide during transit through the secretory pathway. Molecular weights and lengths of polymers determined by imprecise analytical methods (e.g., gel electrophoresis) will be understood to be approximate values. When such a value is expressed as "about" X or "approximately" X, the stated value of X will be understood to be accurate to +10%.

The present invention provides novel growth factor polypeptides and proteins. This novel growth factor, termed "zalpha5", exhibits significant amino acid sequence homology to the previously described angiopoietin-1 (Davis et al . , Cell 12:1161-1169, 1996) and angiopoietin-2 (Maisonpierre et al . , Sci ence 277 : 55-60 ,

1997) . For example, a representative polypeptide of the present invention is approximately 30% identical to Ang-1 throughout its length when the sequences are aligned to produce a 441 amino acid residue overlap (residues 17-455 of SEQ ID NO: 2), accounting for overlaps as disclosed below. The polypeptide is approximately 29% identical to 12

Ang-2 when the sequences are aligned to produce a 406 amino acid residue overlap (residues 50-455 of SEQ ID NO: 2) .

Experimental data indicate that zalpha5 is produced in cells of the liver. Using multiple tissue Northern blots, zAlpha5 transcript is detected primarily in liver. Immunohistochemistry (IHC) using a polyclonal antibody raised against full-length zalpha5 detects protein in normal human and monkey liver. These data suggest that one or more of the cell types that line the sinusoids (sinusoidal endothelial cells, Ito cells, or Kupffer cells) is the source of zalpha5. In normal liver, the space of Disse and Kupffer cells appear to stain strongly for zalpha5 protein. In si tu hybridization of mouse liver slices confirms the presence of transcript in the liver starting as early as embryonic day 13.5 as well as in adult liver. Neither cultured hepatocytic nor cultured sinusoidal endothelial cells contain zalpha5 transcript. While not wishing to be bound by theory, this result suggests that another non-parenchymal cell type in the liver produces zalphaδ . When whole rat livers are perfused with collagenase, and the resulting cell mixture is fractionated, the Kupffer cell fraction contains noticeably more transcript than the other fractions. Experimental data further suggest that sinusoidal lining cells may also be a target for zalpha5. Alterations in the expression level and localization of zalpha5 are observed by IHC in liver cirrhosis, hepatitis, thrombosis and centrilobular necrosis. Zalpha5 may thus be involved in organ-specific liver biology, such as through effects on the sinusoidal endothelial cells and/or hepatocytes . There is thought to be significant paracrine interaction between the cell types lining the liver sinusoids. Given that zalpha5 transcript is detected in adult liver rather than transiently during embryogenesis, it may function in vivo to maintain the liver sinusoidal 13

architecture. As such it may be a protective factor in the event of liver damage.

The proteins zalpha5, angiopoietin-1, angiopoietin-2 , fibrinogen-beta, and fibrinogen-gamma form an evolutionarily related family. Their amino acid sequences include three distinct segments designated, from the amino terminus to the carboxyl terminus, A, B, and C. Segment A comprises a secretory signal peptide. Segment B is predicted to be an extended helical region which, in interaction with analogous extended helical regions in other chains, produces multimeric (e.g., dimeric, trimeric, or tetrameric) coiled coil structures. Coiled coils are bundles of extended α-helices (generally 2-4 helices) wound into a superhelix. The sequence of a single component chain of a coiled coil shows a heptad repeat in the chemical nature of sidechains. This "coiled coil" structure is characterized by a "knobs-into-holes" packing of amino acid sidechains in the core of the bundle. See, Lupas, TIBS 2JL:375-382, 1996. Among the family members there is only limited amino acid sequence conservation in segment B. Segment B has some similarity to the primary structure of myosin, which forms a prototypical dimeric coiled coil. Segment C, as revealed by the crystal structure of the gamma chain of human fibrinogen (Spraggon et al . , Na ture 389 :455-462, 1997), forms a prominent globular region of approximately 250 residues. Segment C is sometimes referred to as a "fibrinogen homology domain". This domain occurs in numerous other proteins, including tenascin-C (Erickson and Bourdon, Ann . Rev. Cell Biol . 5 : 71-92, 1989), restrictin, and ficolin.

Within zalphaδ (SEQ ID NO: 2), segment A, the secretory signal peptide, extends from residue 1 through residue 18. Segment B, the extended helical region, continues through residue 203. The core helices of segment B comprise residues Arg36-Leu50, Phe78-Ile92, 14

Valll3-Leul27, and Ilel75-Leul89 of SEQ ID NO : 2. Segment C, the fibrinogen homology domain, extends from residue 239 to the carboxyl terminus of the protein. In addition, zalpha5 includes an interdomain region or linker (segment L) between segments B and C. Those skilled in the art will recognize that domain boundaries are somewhat imprecise and can be expected to vary by up to ± 4 residues from the specified positions.

There is a potential proteolytic cleavage site at residues 204-205 (Arg-Arg) of SEQ ID NO : 2 , at the approximate boundary between segments B and L. Certain cell types may cleave the zalphaδ polypeptide at this site to produce a segment B polypeptide terminating at residue 203. Experimental data indicate that zalpha5 produced in an insect cell/baculovirus expression system is cleaved in segment L, resulting in a segment C-containing polypeptide having an amino terminus at residue 225 of SEQ ID NO : 2 , with a minor fraction having an amino terminus at residue 231. Full-length recombinant zalpha5 polypeptide having an amino terminus at residue 17 was also produced.

Those skilled in the art will recognize that a variety of zalpha5 polypeptides and multimers thereof can be produced by selecting different expression systems, adding or deleting proteolytic cleavage sites through mutagenesis, synthesizing polypeptides in vi tro, or through the application of other methods known in the art.

In many cases, the structure of the final polypeptide product will result from processing of the nascent polypeptide chain by the host cell, thus the final sequence of a zalpha5 polypeptide produced by a host cell will not always correspond to the full sequence encoded by the expressed polynucleotide. For example, expressing the complete zalpha5 sequence in a cultured mammalian cell is expected to result in removal of at least the secretory peptide, while the same polypeptide produced in a prokaryotic host would not be expected to be cleaved. Differential processing of individual chains may result in 15

heterogeneity of expressed polypeptides and the production of heterodimeric zalpha5 proteins.

In general, zalpha5 polypeptides will contain at least segment B residues 23-199 or segment C residues 246- 408 as shown in SEQ ID NO : 2. These polypeptides can be represented by the formula B_x-L_y-C_z, wherein B is at least 70% identical to residues m to n of SEQ ID NO: 2; L is at least 70% identical to residues o to p of SEQ ID NO : 2 ; and C is at least 70% identical to residues q to r of SEQ ID NO : 2. Within this formula, m is from 15 to 23, n is from 199-207, o is n+1, p is from 234 to 242, q is from p+1 to 246, r is from 408 to 460, and x, y, and z are each 0 (i.e., the indicated domain is absent) or 1 (i.e., the indicated domain is present), subject to the limitations that at least one of x and z is 1 and if both x and z are 1, y is 1. Representative polypeptides (with reference to SEQ ID NO: 2) are shown below in Table 1.

16

Table 1

17-460 23-238

17-408 23-460

17-238 204-408 17-230 204-460

17-224 225-408

17-203 225-460

17-199 231-408

23-203 231-460 23-204 232-408

23-205 232-460

23-235 233-408

23-240 233-460

Sequence homology within protein family provides evidence that zalpha5 will form multimeric complexes . At the junction of segments B and C in the fibrinogen beta, gamma, and alpha chains reside two cysteine residues spaced in a CXXXC motif (SEQ ID NO:3). These cysteines, in interaction with cysteine residues in other monomers, form a "disulfide ring" motif that stabilizes the C- terminal portion of segment B. Angiopoietin-1 , which forms a trimer, contains only one of these cysteines; angiopoietin-2, which forms a dimer, and zalpha5 contain neither of them. When this cysteine is mutated, angiopoietin-1 forms dimers rather than trimers (Davis et al . , WIPO publication WO 96/31598). Zalpha5 multimeric complexes may be heteromultimers, comprising other angiopoietin-like monomers (including dissimilar zalphaδ polypeptides); or homomultimers , comprising two or more identical zalpha5 monomeric units. Experimental data indicate that the protein is produced as a monomer or as a non-covalent multimer, depending upon the expression system. Any of the zalpha5 polypeptides disclosed above any form multimers.

Segment B in human fibrinogen gamma, beta, and alpha contains several cysteines which, with interchain 17

disulfide bonds, form another "disulfide ring" that links the trimeric alpha/beta/gamma multimer to a second multimer, forming a hexameric complex. Corresponding cysteine residues are not present in zalpha5. While not wishing to be bound by theory, these sequence dissimilarities provide evidence that dimers or trimers of zalpha5 will not combine to form tetrameric or hexameric complexes .

In segment C of each of human fibrinogen beta and gamma chains there are four cysteine residues that form intrachain disulfide bonds. Four cysteines are also present, with similar spacing, in segment C of zalpha5. By analogy, disulfide bonds are predicted to occur in zalphaδ between Cys246 and Cys274 of SEQ ID NO : 2 , and between Cys394 and Cys408 of SEQ ID NO: 2. This prediction is confirmed by analysis of recombinant zalpha5.

Zalpha5 has four potential N-linked glycosylation sites, at residues 23, 115, 296, and 357 of SEQ ID NO : 2. Different lots of recombinant zalphaδ showed different amounts of N-linked glycosylation, with two to three oligosaccharides per molecule on the average. The first site was found to be almost never occupied, the second site to be occupied about half the time, and the last two sites to be usually occupied. There were also several (range of 2-5) O-linked oligosaccharides per molecule .

Zalpha5 proteins of the present invention are characterized by their growth factor activity. These proteins modulate the proliferation, differentiation, migration, adhesion, or metabolism of responsive cell types. Biological activity of zalpha5 proteins is assayed using in vi tro or in vivo assays designed to detect cell proliferation, differentiation, migration or adhesion; or changes in cellular metabolism (e.g., production of other growth factors or other macromolecules) . Many suitable assays are known in the art, and representative assays are disclosed herein. Assays using cultured cells are most convenient for screening, such as for determining the effects of amino acid substitutions, deletions, or insertions. However, in view of the complexity of developmental processes (e.g., angiogenesis, wound healing) , in vivo assays will generally be employed to confirm and further characterize biological activity. Certain in vi tro models, such as the three-dimensional collagen gel matrix model of Pepper et al . {Biochem .

Biophys . Res . Comm . 189 : 824-831 , 1992), are sufficiently complex to assay histological effects. Assays can be performed using exogenously produced proteins, or may be carried out in vivo or in vi tro using cells expressing the polypeptide (s) of interest. Assays can be conducted using zalpha5 proteins alone or in combination with other growth factors, such as members of the VEGF family or hematopoietic cytokines (e.g., EPO, TPO, G-CSF, stem cell factor) . Representative assays are disclosed below.

Those skilled in the art will recognize that there is considerable latitude in amino acid sequence, and that equivalent polypeptides can be produced by engineering amino acid changes into the representative human polypeptide sequence shown in SEQ ID NO : 2 or an allelic variant or ortholog thereof. It is preferred that these engineered variant polypeptides are at least 70% identical to the polypeptide of SEQ ID NO : 2. Such polypeptides will preferably be at least 75% identical, more preferably 80% identical, still more preferably at least 90% identical, and most preferably 95% or more identical to SEQ ID NO : 2. Preferred candidate amino acid substitutions within human zalpha5 are suggested by alignment of the human (SEQ ID NO: 2) and mouse (SEQ ID NO: 15) sequences as shown in Fig. 1, which sequences are approximately 76% identical overall.

Percent sequence identity is determined by conventional methods. See, for example, Altschul et al . , Bull . Ma th . Bio . 48_.= 603-616, 1986, and Henikoff and 19

Henikoff, Proc . Na tl . Acad . Sci . USA _.89_. : 10915-10919 , 1992.

Briefly, two amino acid sequences are aligned to optimize the alignment scores using a gap opening penalty of 10, a gap extension penalty of 1, and the "BLOSUM62" scoring matrix of Henikoff and Henikoff { ibid. ) as shown in Table

2 (amino acids are indicated by the standard one-letter codes) . The percent identity is then calculated as:

Total number of identical matches x 100

[length of the longer sequence plus the number of gaps introduced into the longer sequence in order to align the two sequences]

Table 2

A R N D C Q E G H I L K M F P S T W Y V w>

00

A 4

R -1 5

N -2 0 6

D -2 -2 1 6

C 0 -3 -3 -3 9

Q -1 1 0 0 -3 5

10 E -1 0 0 2 -4 2 5

G 0 -2 0 -1 -3 -2 -2 6

H -2 0 1 -1 -3 0 0 -2 8 ro o

I -1 -3 -3 -3 -1 -3 -3 -4 -3 4

L -1 -2 -3 -4 -1 -2 -3 -4 -3 2 4

15 K -1 2 0 -1 -3 1 1 -2 -1 -3 -2 5

M -1 -1 -2 -3 -1 0 -2 -3 -2 1 2 -1 5

F -2 -3 -3 -3 -2 -3 -3 -3 -1 0 0 -3 0 6

P -1 -2 -2 -1 -3 -1 -1 -2 -2 -3 -3 -1 -2 -4 7

S 1 -1 1 0 -1 0 0 0 -1 -2 -2 0 -1 -2 -1 4

20 T 0 -1 0 -1 -1 -1 -1 -2 -2 -1 -1 -1 -1 -2 -1 1 5

W -3 -3 -4 -4 -2 -2 -3 -2 -2 -3 -2 -3 -1 1 -4 -3 -2 11 o

Y -2 -2 -2 -3 -2 -1 -2 -3 2 -1 -1 -2 -1 3 -3 -2 -2 2 7

CΛ

V 0 -3 -3 -3 -1 -2 -2 -3 -3 3 1 -2 1 -1 -2 -2 0 -3 -1

© 00

OS

21

The level of identity between amino acid sequences can be determined using the "FASTA" similarity search algorithm disclosed by Pearson and Lipman { Proc .

Na tl . Acad . Sci . USA 15:2444, 1988) and by Pearson {Meth . Enzymol . 183 :63 , 1990). Briefly, FASTA first characterizes sequence similarity by identifying regions shared by the query sequence (e.g., SEQ ID NO: 2) and a test sequence that have either the highest density of identities (if the ktup variable is 1) or pairs of identities (if ktup=2), without considering conservative amino acid substitutions, insertions, or deletions. The ten regions with the highest density of identities are then rescored by comparing the similarity of all paired amino acids using an amino acid substitution matrix, and the ends of the regions are "trimmed" to include only those residues that contribute to the highest score. If there are several regions with scores greater than the "cutoff" value (calculated by a predetermined formula based upon the length of the sequence and the ktup value) , then the trimmed initial regions are examined to determine whether the regions can be joined to form an approximate alignment with gaps. Finally, the highest scoring regions of the two amino acid sequences are aligned using a modification of the Needleman-Wunsch- Sellers algorithm (Needleman and Wunsch, J^". Mol . Biol .

4_8:444, 1970; Sellers, SIAM J. Appl . Ma th . _.26:787, 1974), which allows for amino acid insertions and deletions. Illustrative parameters for FASTA analysis are: ktup=l, gap opening penalty=10, gap extension penalty=l, and substitution matrix=BLOSUM62. These parameters can be introduced into a FASTA program by modifying the scoring matrix file ("SMATRIX"), as explained in Appendix 2 of Pearson, 1990 { ibid. ) .

FASTA can also be used to determine the sequence identity of nucleic acid molecules using a ratio 22

as disclosed above. For nucleotide sequence comparisons, the ktup value can range between one to six, preferably from four to six.

The present invention includes polypeptides having one or more conservative amino acid changes as compared with the amino acid sequence of SEQ ID NO : 2. The BLOSUM62 matrix (Table 2) is an amino acid substitution matrix derived from about 2,000 local multiple alignments of protein sequence segments, representing highly conserved regions of more than 500 groups of related proteins (Henikoff and Henikoff, ibid . ) . Thus, the BLOSUM62 substitution frequencies can be used to define conservative amino acid substitutions that may be introduced into the amino acid sequences of the present invention. As used herein, the term "conservative amino acid substitution" refers to a substitution represented by a BLOSUM62 value of greater than -1. For example, an amino acid substitution is conservative if the substitution is characterized by a BLOSUM62 value of 0, 1, 2, or 3. Preferred conservative amino acid substitutions are characterized by a BLOSUM62 value of at least one 1 (e.g., 1, 2 or 3) , while more preferred conservative amino acid substitutions are characterized by a BLOSUM62 value of at least 2 (e.g., 2 or 3) .

Engineered variant zalpha5 polypeptides are characterized as having one or more amino acid substitutions, deletions or additions. These changes are preferably of a minor nature, that is conservative amino acid substitutions and other substitutions that do not significantly affect the folding or activity of the polypeptide; small deletions, typically of one to about 30 amino acids; and amino- or carboxyl -terminal extensions, such as an amino-terminal methionine residue, a small linker peptide of up to about 20-25 residues, or an extension that facilitates purification (an affinity 23

tag) . See, in general Ford et al . , Protein Expression and Purification 2 : 95-107, 1991, which is incorporated herein by reference. Two or more affinity tags may be used in combination. DNAs encoding affinity tags are available from commercial suppliers (e.g., Pharmacia Biotech, Piscataway, NJ; New England Biolabs, Beverly, MA) . It is preferred to limit amino acid deletions and substitutions to the globular region (when present) of a zalpha5 polypeptide, although conservative substitutions or deletions of one to a few residues can be made in the extended helical region. Polypeptides comprising affinity tags can further comprise a polypeptide linker and/or a proteolytic cleavage site between the zalphaδ polypeptide and the affinity tag. Preferred cleavage sites include thrombin cleavage sites and factor Xa cleavage sites.

The proteins of the present invention can also comprise non-naturally occuring amino acid residues. Non-naturally occuring amino acids include, without limitation, trans-3-methylproline, 2 , 4-methanoproline, cis-4 -hydroxyproline, trans-4. -hydroxyproline, JV- methylglycine, allo-threonine, methylthreonine, hydroxyethylcysteine , hydroxyethylhomocysteine , nitroglutamine, homoglutamine, pipecolic acid, tert- leucine, norvaline, 2-azaphenylalanine, 3- azaphenylalanine, 4-azaphenylalanine, and 4- fluorophenylalanine . Several methods are known in the art for incorporating non-naturally occuring amino acid residues into proteins. For example, an in vi tro system can be employed wherein nonsense mutations are suppressed using chemically aminoacylated suppressor tRNAs . Methods for synthesizing amino acids and aminoacylating tRNA are known in the art. Transcription and translation of plasmids containing nonsense mutations is carried out in a cell -free system comprising an E. coli S30 extract and 24

commercially available enzymes and other reagents. Proteins are purified by chromatography. See, for example, Robertson et al . , J. Am. Chem. Soc. 113 :2722 , 1991; Ell an et al . , Methods Enzymol . 202 :301, 1991; Chung et al . , Science 2_5_9: 806-809 , 1993; and Chung et al., Proc. Natl. Acad. Sci. USA _.90_.: 10145-10149 , 1993). In a second method, translation is carried out in Xenopus oocytes by microinjection of mutated mRNA and chemically aminoacylated suppressor tRNAs (Turcatti et al . , J. Biol. Chem. 271:19991-19998, 1996). Within a third method, E . coli cells are cultured in the absence of a natural amino acid that is to be replaced (e.g., phenylalanine) and in the presence of the desired non-naturally occuring amino acid(s) (e.g., 2-azaphenylalanine, 3-azaphenylalanine, 4- azaphenylalanine, or 4-fluorophenylalanine) . The non- naturally occuring amino acid is incorporated into the protein in place of its natural counterpart. See, Koide et al., Biochem. 3_3: 7470-7476 , 1994. Naturally occuring amino acid residues can be converted to non-naturally occuring species by in vi tro chemical modification.

Chemical modification can be combined with site-directed mutagenesis to further expand the range of substitutions (Wynn and Richards, Protein Sci. 2_.:395-403, 1993).

Essential amino acids in the polypeptides of the present invention can be identified according to procedures known in the art, such as site-directed mutagenesis or alanine-scanning mutagenesis (Cunningham and Wells, Science 244 , 1081-1085, 1989; Bass et al . ,

Proc . Natl . Acad . Sci . USA _.88_.: 4498-4502 , 1991). In the latter technique, single alanine mutations are introduced at every residue in the molecule, and the resultant mutant molecules are tested for biological activity as disclosed below to identify amino acid residues that are critical to the activity of the molecule. Multiple amino acid substitutions can be made and tested using known methods of mutagenesis and 25

screening, such as those disclosed by Reidhaar-Olson and Sauer { Science 241 :53-57, 1988) or Bowie and Sauer { Proc .

Na tl . Acad . Sci . USA 8_: 2152-2156, 1989). Briefly, these authors disclose methods for simultaneously randomizing two or more positions in a polypeptide, selecting for functional polypeptide, and then sequencing the mutagenized polypeptides to determine the spectrum of allowable substitutions at each position. Other methods that can be used include phage display (e.g., Lowman et al., Biochem . 3_0: 10832-10837 , 1991; Ladner et al . , U.S.

Patent No. 5,223,409; Huse, WIPO Publication WO 92/06204) and region-directed mutagenesis (Derbyshire et al . , Gene

£6:145, 1986; Ner et al . , DNA 1 : 121 , 1988).

Amino acid sequence changes are made in zalpha5 polypeptides so as to minimize disruption of higher order structure essential to biological activity. In this regard, it is generally preferred to retain the coiled- coil-forming propensity in the amino-terminal , extended helical region of zalpha5 and the sequence homology to fibrinogen in the carboxyl-terminal region (when present) , particularly the conserved cysteine residues

(residues 246, 274, 394, and 408 of SEQ ID NO:2) .

Changes in the coiled-coil region can be analyzed using analytical software available on the World Wide Web at http://ulrec3.unil.ch/software/COILS_form.html and http : //ostrich. lcs .mit . edu/cgi-bin/score .

Variants of the disclosed zalpha5 DNA and polypeptide sequences can be generated through DNA shuffling as disclosed by Stemmer, Na ture 370 : 389-391 , 1994 and Stemmer, Proc . Natl . Acad . Sci . USA 91:10747-

10751, 1994. Briefly, variant genes are generated by in vi tro homologous recombination by random fragmentation of a parent gene followed by reassembly using PCR, resulting in randomly introduced point mutations. This technique can be modified by using a family of parent genes, such 26

as allelic variants or genes from different species, to introduce additional variability into the process. Selection or screening for the desired activity, followed by additional iterations of mutagenesis and assay provides for rapid "evolution" of sequences by selecting for desirable mutations while simultaneously selecting against detrimental changes .

Mutagenesis methods as disclosed above can be combined with high volume or high-throughput screening methods to detect biological activity of zalpha5 variant polypeptides. For example, the chick chorioallantoic assay disclosed below can be carried out on large numbers of samples. Assays that can be scaled up for high throughput include mitogenesis assays, which can be run in a 96 -well format. Mutagenized DNA molecules that encode active zalphaδ polypeptides can be recovered from the host cells and rapidly sequenced using modern equipment. These methods allow the rapid determination of the importance of individual amino acid residues in a polypeptide of interest, and can be applied to polypeptides of unknown structure.

Using the methods discussed above, one of ordinary skill in the art can prepare a variety of polypeptide fragments or variants of SEQ ID NO : 2 that retain the activity of wild-type zalphaδ .

The present invention also provides polynucleotide molecules, including DNA and RNA molecules, that encode the zalpha5 polypeptides disclosed above. A representative DNA sequence encoding the amino acid sequence of SEQ ID NO : 2 is shown in SEQ ID NO : 1. A representative mouse zalphaδ DNA sequence is shown in SEQ ID NO: 14, with the encoded protein shown in SEQ ID NO: 15. Those skilled in the art will readily recognize that, in view of the degeneracy of the genetic code, considerable sequence variation is possible among these polynucleotide molecules. SEQ ID NO : 4 is a degenerate DNA sequence that encompasses all DNAs that encode the zalpha5 polypeptide 27

of SEQ ID NO: 2. Those skilled in the art will recognize that the degenerate sequence of SEQ ID NO : 4 also provides all RNA sequences encoding SEQ ID NO : 2 by substituting U for T. Thus, zalpha5 polypeptide-encoding polynucleotides comprising nucleotide 70 to nucleotide 714 of SEQ ID NO : 4 , and their RNA equivalents are contemplated by the present invention, as are segments of SEQ ID NO : 4 encoding other polypeptides shown in Table 1. Table 3 sets forth the one-letter codes used within SEQ ID NO : 4 to denote degenerate nucleotide positions. "Resolutions" are the nucleotides denoted by a code letter. "Complement" indicates the code for the complementary nucleotide (s) . For example, the code Y denotes either C or T, and its complement R denotes A or G, A being complementary to T, and G being complementary to C.

28

TABLE 3

Nucleotide Resolutions Complement Resolutions

^ ^ ! ^

C C G G

G G C C

T T A A

R A|G Y c|τ

Y C|T R A|G

M A|C K G|T

K G|T M A|C

S C|G S C|G w A|T W A|T

H A|C|T D A|G|T

B C|G|T v A|C|G v A|C|G B C|G|T

D A|G|T H A|C|T

N A|C|G|T N A|C|G|T

The degenerate codons used in SEQ ID NO: 4, encompassing all possible codons for a given amino acid, are set forth in Table 4, below.

29

TABLE ■ 4

Amino One- Degenerate

Acid Letter Codons Codon

Code

Cys C TGC TGT TGY

Ser S AGC AGT TCA TCC TCG TCT WSN

Thr T ACA ACC ACG ACT CAN

Pro P CCA CCC CCG CCT CCN

Ala A GCA GCC GCG GCT GCN

Gly G GGA GGC GGG GGT GGN

Asn N AAC AAT AAY

Asp D GAC GAT GAY

Glu E GAA GAG GAR

Gin Q CAA CAG CAR

His H CAC CAT CAY

Arg R AGA AGG CGA CGC CGG CGT MGN

Lys K AAA AAG AAR

Met M ATG ATG

He I ATA ATC ATT ATH

Leu L CTA CTC CTG CTT TTA TTG YTN

Val V GTA GTC GTG GTT GTN

Phe F TTC TTT TTY

Tyr Y TAC TAT TAY

Trp W TGG TGG

Ter - TAA TAG TGA TRR

Asn | Asp B RAY

Glu | Gin Z SAR

Any X NNN

Gap -

One of ordinary skill in the art will appreciate that some ambiguity is introduced in determining a degenerate codon, representative of all possible codons encoding each amino acid. For example, the degenerate codon for serine (WSN) can, in some circumstances, encode arginine (AGR) , and the degenerate 3 0

codon for arginine (MGN) can, in some circumstances, encode serine (AGY) . A similar relationship exists between codons encoding phenylalanine and leucine . Thus, some polynucleotides encompassed by the degenerate sequence may encode variant amino acid sequences, but one of ordinary skill in the art can easily identify such variant sequences by reference to the amino acid sequence of SEQ ID NO: 2. Variant sequences can be readily tested for functionality as described herein. One of ordinary skill in the art will also appreciate that different species can exhibit preferential codon usage. See, in general, Grantham et al . , Nuc . Acids Res . _.8:1893-912, 1980; Haas et al . Curr .

Biol . 1:315-24, 1996; Wain-Hobson et al . , Gene 13 :355-64, 1981; Grosjean and Fiers, Gene 11=199-209, 1982; Holm,

Nuc . Acids Res . JL4: 3075-87, 1986; and Ikemura, J. Mol .

Biol . 151:573-97, 1982. Introduction of preferred codon sequences into recombinant DNA can, for example, enhance production of the protein by making protein translation more efficient within a particular cell type or species. Therefore, the degenerate codon sequence disclosed in SEQ ID NO : 4 serves as a template for optimizing expression of polynucleotides in various cell types and species commonly used in the art and disclosed herein. Within preferred embodiments of the invention the isolated polynucleotides will hybridize to similar sized regions of SEQ ID NO : 1 , or a sequence complementary thereto, under stringent conditions. In general, stringent conditions are selected to be about 5°C lower than the thermal melting point (T_m) for the specific sequence at a defined ionic strength and pH. The T_m is the temperature (under defined ionic strength and pH) at which 50% of the target sequence hybridizes to a perfectly matched probe. Typical stringent conditions are those in which the salt concentration is up to about 0.03 M at pH 7 and the temperature is at least about 60°C. 31

As previously noted, the isolated polynucleotides of the present invention include DNA and RNA. Methods for preparing DNA and RNA are well known in the art. In general, RNA is isolated from a tissue or cell that produces large amounts of zalpha5 RNA. Liver is a preferred tissue. Zalpha5 transcripts have also been detected in liver tumor tissue. Total RNA can be prepared using guanidine HCl extraction followed by isolation by centrifugation in a CsCl gradient (Chirgwin et al . , Biochemistry ϋ:52-94, 1979). Poly (A) ⁺ RNA is prepared from total RNA using the method of Aviv and Leder { Proc . Natl . Acad. Sci . USA 69_.: 1408-1412, 1972).

Complementary DNA (cDNA) is prepared from poly (A) ⁺ RNA using known methods. In the alternative, genomic DNA can be isolated. Polynucleotides encoding zalpha5 polypeptides are then identified and isolated by, for example, hybridization or PCR.

A full-length clone encoding zalpha5 can be obtained by conventional cloning procedures. Complementary DNA (cDNA) clones are preferred, although for some applications (e.g., expression in transgenic animals) it may be preferable to use a genomic clone, or to modify a cDNA clone to include at least one genomic intron. Methods for preparing cDNA and genomic clones are well known and within the level of ordinary skill in the art, and include the use of the sequence disclosed herein, or parts thereof, for probing or priming a library. Expression libraries can be probed with antibodies to zalpha5, receptor fragments, or other specific binding partners.

Zalpha5 polynucleotide sequences disclosed herein can also be used as probes or primers to clone 5 ' non-coding regions of a zalpha5 gene. In view of the tissue-specific expression observed for zalpha5 by Northern blotting, this gene region is expected to provide for liver- and kidney-specific expression. 32

Promoter elements from a zalphaδ gene could thus be used to direct the tissue-specific expression of heterologous genes in, for example, transgenic animals or patients treated with gene therapy. Cloning of 5' flanking sequences also facilitates production of zalpha5 proteins by "gene activation" as disclosed in U.S. Patent No. 5,641,670. Briefly, expression of an endogenous zalpha5 gene in a cell is altered by introducing into the zalpha5 locus a DNA construct comprising at least a targeting sequence, a regulatory sequence, an exon, and an unpaired splice donor site. The targeting sequence is a zalpha5 5' non-coding sequence that permits homologous recombination of the construct with the endogenous zalpha5 locus, whereby the sequences within the construct become operably linked with the endogenous zalpha5 coding sequence. In this way, an endogenous zalpha5 promoter can be replaced or supplemented with other regulatory sequences to provide enhanced, tissue-specific, or otherwise regulated expression. Those skilled in the art will recognize that the sequences disclosed in SEQ ID NOS : 1 and 2 represent a single allele of human zalpha5. Allelic variants of these sequences can be cloned by probing cDNA or genomic libraries from different individuals according to standard procedures.

The present invention further provides counterpart polypeptides and polynucleotides from other species ( "orthologs" ) . Of particular interest are zalpha5 polypeptides from other mammalian species, including murine, porcine, ovine, bovine, canine, feline, equine, and other primate polypeptides. Orthologs of human zalpha5 can be cloned using information and compositions provided by the present invention in combination with conventional cloning techniques. For example, a cDNA can be cloned using mRNA obtained from a tissue or cell type that expresses zalpha5 as disclosed above. A library is then prepared from mRNA of a 33

positive tissue or cell line. A zalpha5 -encoding cDNA can then be isolated by a variety of methods, such as by probing with a complete or partial human or mouse cDNA or with one or more sets of degenerate probes based on the disclosed sequences. A cDNA can also be cloned using the polymerase chain reaction, or PCR (Mullis, U.S. Patent No. 4,683,202), using primers designed from the representative human or mouse zalphaδ sequences disclosed herein. Within an additional method, the cDNA library can be used to transform or transfect host cells, and expression of the cDNA of interest can be detected with an antibody to zalpha5 polypeptide. Similar techniques can also be applied to the isolation of genomic clones.

For any zalpha5 polypeptide, including variants and fusion proteins, one of ordinary skill in the art can readily generate a fully degenerate polynucleotide sequence encoding that variant using the information set forth in Tables 3 and 4, above. Moreover, those of skill in the art can use standard software to devise zsig51 variants based upon the nucleotide and amino acid sequences described herein. The present invention thus provides a computer-readable medium encoded with a data structure that provides at least one of the following sequences: SEQ ID NO:l, SEQ ID NO : 2 , SEQ ID NO : 4 , SEQ ID NO: 14, SEQ ID NO: 15, and portions thereof. Suitable forms of computer-readable media include magnetic media and optically-readable media. Examples of magnetic media include a hard or fixed drive, a random access memory (RAM) chip, a floppy disk, digital linear tape (DLT) , a disk cache, and a ZIP disk. Optically readable media are exemplified by compact discs (e.g., CD-read only memory (ROM) , CD-rewritable (RW) , and CD-recordable) , and digital versatile/video discs (DVD) (e.g., DVD-ROM, DVD- RAM, and DVD+RW) . The zalphaδ polypeptides of the present invention, including full-length polypeptides, biologically active fragments, and fusion polypeptides 34

can be produced in genetically engineered host cells according to conventional techniques. Suitable host cells are those cell types that can be transformed or transfected with exogenous DNA and grown in culture, and include bacteria, fungal cells, and cultured higher eukaryotic cells. Eukaryotic cells, particularly cultured cells of multicellular organisms, are preferred. Techniques for manipulating cloned DNA molecules and introducing exogenous DNA into a variety of host cells are disclosed by Sambrook et al . , Molecular Cloning: A

Laboratory Manual , 2nd ed., Cold Spring Harbor Laboratory

Press, Cold Spring Harbor, NY, 1989, and Ausubel et al . , eds., Current Protocols in Molecular Biology, John Wiley and Sons, Inc., NY, 1987. In general, a DNA sequence encoding a zalpha5 polypeptide is operably linked to other genetic elements required for its expression, generally including a transcription promoter and terminator, within an expression vector. The vector will also commonly contain one or more selectable markers and one or more origins of replication, although those skilled in the art will recognize that within certain systems selectable markers may be provided on separate vectors, and replication of the exogenous DNA may be provided by integration into the host cell genome. Selection of promoters, terminators, selectable markers, vectors and other elements is a matter of routine design within the level of ordinary skill in the art . Many such elements are described in the literature and are available through commercial suppliers.

To direct a zalpha5 polypeptide into the secretory pathway of a host cell, a secretory signal sequence (also known as a leader sequence, prepro sequence or pre sequence) is provided in the expression vector. The secretory signal sequence may be that of zalpha5, or may be derived from another secreted protein 35

(e.g., t-PA; see, U.S. Patent No. 5,641,655) or synthesized de novo . The secretory signal sequence is operably linked to the zalpha5 DNA sequence, i.e., the two sequences are joined in the correct reading frame and positioned to direct the newly sythesized polypeptide into the secretory pathway of the host cell . Secretory signal sequences are commonly positioned 5 ' to the DNA sequence encoding the polypeptide of interest, although certain signal sequences may be positioned elsewhere in the DNA sequence of interest (see, e.g., Welch et al . , U.S. Patent No. 5,037,743; Holland et al . , U.S. Patent No. 5, 143, 830) .

Cultured mammalian cells are preferred hosts for use within the present invention. Methods for introducing exogenous DNA into mammalian host cells include calcium phosphate-mediated transfection (Wigler et al . , Cell _.14:725, 1978; Corsaro and Pearson, Somatic

Cell Genetics 2:603, 1981; Graham and Van der Eb,

Virology 12:456, 1973), electroporation (Neumann et al . , EMBO J. 1:841-845, 1982), DEAE-dextran mediated transfection (Ausubel et al . , ibid.), and liposome- mediated transfection (Hawley-Nelson et al . , Focus 15 : 73 ,

1993; Ciccarone et al . , Focus 15:80, 1993). The production of recombinant polypeptides in cultured mammalian cells is disclosed, for example, by Levinson et al., U.S. Patent No. 4,713,339; Hagen et al . , U.S. Patent

No. 4,784,950; Palmiter et al . , U.S. Patent No.

4,579,821; and Ringold, U.S. Patent No. 4,656,134.

Suitable cultured mammalian cells include the COS-1 (ATCC No. CRL 1650), COS-7 (ATCC No. CRL 1651), BHK (ATCC No.

CRL 1632), BHK 570 (ATCC No. CRL 10314), 293 (ATCC No.

CRL 1573; Graham et al . , J^". Gen . Virol . 3_6: 59-72, 1977) and Chinese hamster ovary (e.g.^' CH0-K1, ATCC No. CCL 61; or CHO DG44, Chasin et al . , So . Cell . Molec . Genet . 12: 555, 1986) cell lines. Additional suitable cell lines 36

are known in the art and available from public depositories such as the American Type Culture Collection, Manassas, VA. In general, strong transcription promoters are preferred, such as promoters from SV-40 or cytomegalovirus . See, e.g., U.S. Patent No. 4,956,288. Other suitable promoters include those from metallothionein genes (U.S. Patent Nos. 4,579,821 and 4,601,978) and the adenovirus major late promoter. Expression vectors for use in mammalian cells include pZP-1 and pZP-9, which have been deposited with the American Type Culture Collection, Manassas, VA USA under accession numbers 98669 and 98668, respectively.

Drug selection is generally used to select for cultured mammalian cells into which foreign DNA has been inserted. Such cells are commonly referred to as " transfectants" . Cells that have been cultured in the presence of the selective agent and are able to pass the gene of interest to their progeny are referred to as "stable transfectants . " A preferred selectable marker is a gene encoding resistance to the antibiotic neomycin. Selection is carried out in the presence of a neomycin- type drug, such as G-418 or the like. Selection systems can also be used to increase the expression level of the gene of interest, a process referred to as "amplification." Amplification is carried out by culturing transfectants in the presence of a low level of the selective agent and then increasing the amount of selective agent to select for cells that produce high levels of the products of the introduced genes. A preferred amplifiable selectable marker is dihydrofolate reductase, which confers resistance to methotrexate . Other drug resistance genes (e.g. hygromycin resistance, multi-drug resistance, puromycin acetyltransferase) can also be used. The adenovirus system (disclosed in more detail below) can also be used for protein production in vi tro . 37

By culturing adenovirus-infected non-293 cells under conditions where the cells are not rapidly dividing, the cells can produce proteins for extended periods of time. For instance, BHK cells are grown to confluence in cell factories, then exposed to the adenoviral vector encoding the secreted protein of interest. The cells are then grown under serum- free conditions, which allows infected cells to survive for several weeks without significant cell division. In an alternative method, adenovirus vector-infected 293 cells can be grown as adherent cells or in suspension culture at relatively high cell density to produce significant amounts of protein (See Gamier et al., Cytotechnol . 15:145-55, 1994). With either protocol, an expressed, secreted heterologous protein can be repeatedly isolated from the cell culture supernatant, lysate, or membrane fractions depending on the disposition of the expressed protein in the cell. Within the infected 293 cell production protocol, non-secreted proteins can also be effectively obtained. Other higher eukaryotic cells can also be used as hosts, including insect cells, plant cells and avian cells. The use of Agroba cterium rhizogenes as a vector for expressing genes in plant cells has been reviewed by Sinkar et al . , J^". Biosci . (Bangalore) 11:47-58, 1987. Insect cells can be infected with recombinant baculovirus, commonly derived from Autographa calif ornica nuclear polyhedrosis virus (AcNPV) according to methods known in the art. Within a preferred method, recombinant baculovirus is produced through the use of a transposon- based system described by Luckow et al . (J. Virol .

12:4566-4579, 1993) . This system, which utilizes transfer vectors, is commercially available in kit form

(Bac-to-Bac™ kit; Life Technologies, Rockville, MD) . The transfer vector (e.g., pFastBacl™; Life Technologies) contains a Tn7 transposon to move the DNA encoding the 3 8

protein of interest into a baculovirus genome maintained in E. coli as a large plasmid called a "bacmid." See,

Hill-Perkins and Possee, J^". Gen . Virol . 21=971-976, 1990;

Bonning et al . , J. Gen . Virol . 21:1551-1556, 1994; and Chazenbalk and Rapoport , J. Biol . Chem . 270 :1543-1549,

1995. In addition, transfer vectors can include an in- frame fusion with DNA encoding a polypeptide extension or affinity tag as disclosed above. Using techniques known in the art, a transfer vector containing a zalpha5- encoding sequence is transformed into E. coli host cells, and the cells are screened for bacmids which contain an interrupted lacZ gene indicative of recombinant baculovirus. The bacmid DNA containing the recombinant baculovirus genome is isolated, using common techniques, and used to transfect Spodoptera frugiperda cells, such as Sf9 cells. Recombinant virus that expresses zalpha5 protein is subsequently produced. Recombinant viral stocks are made by methods commonly used the art .

For protein production, the recombinant virus is used to infect host cells, typically a cell line derived from the fall armyworm, Spodoptera frugiperda

(e.g., Sf9 or Sf21 cells) or Trichoplusia ni (e.g., High

Five™ cells; Invitrogen, Carlsbad, CA) . See, for example, U.S. Patent No. 5,300,435. Serum- free media are used to grow and maintain the cells. Suitable media formulations are known in the art and can be obtained from commercial suppliers. The cells are grown up from an inoculation density of approximately 2-5 x 10⁵ cells to a density of 1-2 x 10⁶ cells, at which time a recombinant viral stock is added at a multiplicity of infection (MOI) of 0.1 to 10, more typically near 3. Procedures used are generally known in the art .

Fungal cells, including yeast cells, can also be used within the present invention. Yeast species of particular interest in this regard include Saccharomyces 39

cerevisiae , Pichia pastoris , and Pichia methanolica .

Methods for transforming S . cerevisiae cells with exogenous DNA and producing recombinant polypeptides therefrom are disclosed by, for example, Kawasaki, U.S. Patent No. 4,599,311; Kawasaki et al . , U.S. Patent No. 4,931,373; Brake, U.S. Patent No. 4,870,008; Welch et al . , U.S. Patent No. 5,037,743; and Murray et al . , U.S. Patent No. 4,845,075. Transformed cells are selected by phenotype determined by the selectable marker, commonly drug resistance or the ability to grow in the absence of a particular nutrient (e.g., leucine) . A preferred vector system for use in Saccharomyces cerevisiae is the

POT1 vector system disclosed by Kawasaki et al . (U.S.

Patent No. 4,931,373), which allows transformed cells to be selected by growth in glucose-containing media.

Suitable promoters and terminators for use in yeast include those from glycolytic enzyme genes (see, e.g.,

Kawasaki, U.S. Patent No. 4,599,311; Kingsman et al . ,

U.S. Patent No. 4,615,974; and Bitter, U.S. Patent No. 4,977,092) and alcohol dehydrogenase genes. See also

U.S. Patents Nos. 4,990,446; 5,063,154; 5,139,936 and

4,661,454. Transformation systems for other yeasts, including Hansenula polymorpha , Schizosa ccharomyces pombe, Kl uyveromyces lactis , Kl uyveromyces fragilis , Ustilago maydis , Pichia pastoris , Pichia methanolica ,

Pichia guillermondii and Candida mal tosa are known in the art. See, for example, Gleeson et al . , J. Gen .

Microbiol . 132 :3459-3465, 1986; Cregg, U.S. Patent No. 4,882,279; and Raymond et al . , Yeast 14_., 11-23, 1998. Aspergillus cells may be utilized according to the methods of McKnight et al . , U.S. Patent No. 4,935,349. Methods for transforming Acremonium chrysogenum are disclosed by Sumino et al . , U.S. Patent No. 5,162,228. Methods for transforming Neurospora are disclosed by Lambowitz, U.S. Patent No. 4,486,533. 40

The use of Pichia methanolica as host for the production of recombinant proteins is disclosed in U.S.

Patents No. 5,716,808 and No. 5,736,383, and WIPO

Publications WO 97/17450 and WO 97/17451. DNA molecules for use in transforming P . methanolica will commonly be prepared as double-stranded, circular plasmids, which are preferably linearized prior to transformation. A preferred method of transformation is electroporation as disclosed in U.S. Patent No. 5,854,039. For polypeptide production in P. methanolica , it is preferred that the promoter and terminator in the plasmid be that of a P . methanolica gene, such as a P . methanoli ca alcohol utilization gene {AUG1 or AUG2) . Other useful promoters include those of the dihydroxyacetone synthase (DHAS) , formate dehydrogenase (FMD) , and catalase (CAT) genes. To facilitate integration of the DNA into the host chromosome, it is preferred to have the entire expression segment of the plasmid flanked at both ends by host DNA sequences. A preferred selectable marker for use in Pichia methanolica is a P. methanolica ADE2 gene, which encodes phosphoribosyl-5-aminoimidazole carboxylase (AIRC; EC 4.1.1.21), which allows ade2 host cells to grow in the absence of adenine . For large-scale, industrial processes where it is desirable to minimize the use of methanol, it is preferred to use host cells in which both methanol utilization genes {AUG1 and AUG2) are deleted.

For production of secreted proteins, host cells deficient in vacuolar protease genes { PEP4 and PRB1 ) are preferred.

Prokaryotic host cells, including strains of the bacteria Escherichia coli , Ba cill us and other genera are also useful host cells within the present invention. Techniques for transforming these hosts and expressing foreign DNA sequences cloned therein are well known in the art (see, e.g., Sambrook et al . , ibid.). When 41

expressing a zalpha5 polypeptide in bacteria such as E . col i , the polypeptide may be retained in the cytoplasm, typically as insoluble granules, or may be directed to the periplasmic space by a bacterial secretion sequence. In the former case, the cells are lysed, and the granules are recovered and denatured using, for example, guanidine isothiocyanate or urea. The denatured polypeptide can then be refolded and dimerized by diluting the denaturant , such as by dialysis against a solution of urea and a combination of reduced and oxidized glutathione, followed by dialysis against a buffered saline solution. In the latter case, the polypeptide can be recovered from the periplasmic space in a soluble and functional form by disrupting the cells (by, for example, sonication or osmotic shock) to release the contents of the periplasmic space and recovering the protein, thereby obviating the need for denaturation and refolding.

Transformed or transfected host cells are cultured according to conventional procedures in a culture medium containing nutrients and other components required for the growth of the chosen host cells. A variety of suitable media, including defined media and complex media, are known in the art and generally include a carbon source, a nitrogen source, essential amino acids, vitamins and minerals. Media may also contain such components as growth factors or serum, as required. The growth medium will generally select for cells containing the exogenously added DNA by, for example, drug selection or deficiency in an essential nutrient which is complemented by the selectable marker carried on the expression vector or co-transfected into the host cell. Liquid cultures are provided with sufficient aeration by conventional means, such as shaking of small flasks or sparging of fermentors . It is preferred to purify the polypeptides and proteins of the present invention to >80% purity, more 42

preferably to >90% purity, even more preferably >95% purity, and particularly preferred is a pharmaceutically pure state, that is greater than 99.9% pure with respect to contaminating macromolecules, particularly other proteins and nucleic acids, and free of infectious and pyrogenic agents. Preferably, a purified polypeptide or protein is substantially free of other polypeptides or proteins, particularly those of animal origin.

Expressed recombinant zalpha5 proteins (including chimeric polypeptides and multimeric proteins) are purified by conventional protein purification methods, typically by a combination of chromatographic techniques. See, in general, Affinity Chromatography : Principles & Methods, Pharmacia LKB Biotechnology, Uppsala, Sweden, 1988; and Scopes, Protein Purification: Principles and Practice, Springer-Verlag, New York, 1994. Proteins comprising a polyhistidine affinity tag (typically about 6 histidine residues) are purified by affinity chromatography on a nickel chelate resin. See, for example, Houchuli et al . , Bio/Technol . 6_ : 1321-1325,

1988. Proteins comprising a glu-glu tag can be purified by immunoaffinity chromatography according to conventional procedures. See, for example, Grussenmeyer et al . , ibid . The proteins of the present invention can be isolated by exploitation of their binding properties. For example, immobilized metal ion adsorption chromatography (IMAC) can be used to purify polyhistidine-tagged polypeptides. Briefly, a gel is first charged with divalent metal ions to form a chelate

(E. Sulkowski, Trends in Biochem. 2:1-7, 1985) .

Histidine-rich polypeptides will be adsorbed to this matrix with differing affinities, depending upon the metal ion used, and will be eluted by competitive elution, lowering the pH, or use of strong chelating agents. For example, a zalphaδ protein comprising a 43

polyhistidine affinity tag (typically about 6 histidine residues) can be purified by affinity chromatography on a nickel chelate resin. See, for example, Houchuli et al . , Bio/Technol . 6_ : 1321-1325, 1988. Other methods of purification include purification of glycosylated proteins by lectin affinity chromatography and ion exchange chromatography (Methods in Enzymol . , Vol. 182, "Guide to Protein Purification", M. Deutscher, (ed.), Academic Press, San Diego, 1990, 529-539) . Fusion polypeptides comprising other affinity tags (e.g., maltose-binding protein, glu-glu tag, or an immunoglobulin domain) can be constructed and purified using the appropriate binding agent (e.g., an antibody), the selection of which will be evident to those skilled in the art .

Zalpha5 polypeptides can also be prepared through chemical synthesis according to methods known in the art, including exclusive solid phase synthesis, partial solid phase methods, fragment condensation or classical solution synthesis. See, for example, Merrifield, J. Am. Chem. Soc. 15:2149, 1963; Stewart et al . , Solid Phase Peptide Synthesis (2nd edition), Pierce Chemical Co., Rockford, IL, 1984; Bayer and Rapp, Chem .

Pept . Prot . 2:3, 1986; and Atherton et al . , Solid Phase Peptide Synthesis: A Practical Approach, IRL Press,

Oxford, 1989.

Using methods known in the art, zalpha5 proteins can be prepared as monomers or multimers; glycosylated or non-glycosylated; pegylated or non- pegylated; and may or may not include an initial methionine amino acid residue.

Target cells for use in zalpha5 activity assays include vascular cells (especially endothelial cells and smooth muscle cells) , hematopoietic (myeloid and lymphoid) cells, liver cells (including hepatocytes, fenestrated endothelial cells, Kupffer cells, and Ito 44

cells) , fibroblasts (including human dermal fibroblasts and lung fibroblasts), fetal lung cells, articular synoviocytes , pericytes, chondrocytes , osteoblasts, and prostate epithelial cells. Endothelial cells and hematopoietic cells are derived from a common ancestral cell, the hemangioblast (Choi et al . , Development

125:725-732, 1998) .

Activity of zalphaδ proteins can be measured in vi tro using cultured cells or in vivo by administering molecules of the claimed invention to an appropriate animal model. For example, mitogenic activity can be measured using known assays, including ³H-thymidine incorporation assays (as disclosed by, e.g., Raines and Ross, Methods Enzymol . 109 :749-773 , 1985 and Wahl et al . , Mol . Cell Biol . 1:5016-5025, 1988), dye incorporation assays (as disclosed by, for example, Mosman, J^". Immunol .

Meth . 15:55-63, 1983 and Raz et al . , Acta Trop . 68 : 139-

147, 1997) or cell counts. Preferred mitogenesis assays measure incorporation of ³H-thymidine into (1) 20% confluent cultures to look for the ability of zalpha5 proteins to further stimulate proliferating cells, and

(2) quiescent cells held at confluence for 48 hours to look for the ability of zalpha5 proteins to overcome contact-induced growth inhibition. See also, Gospodarowicz et al . , J^". Cell . Biol . 21:395-405, 1976;

Ewton and Florini, Endocrinol . 106 :577-583 , 1980; and

Gospodarowicz et al . , Proc . Na tl . Acad. Sci . USA 16:7311-

7315, 1989. Differentiation can be assayed using suitable precursor cells that can be induced to differentiate into a more mature phenotype . For example, mesenchymal stem cells can be used to measure the ability of zalphaδ protein to stimulate differentiation into osteoblasts. Differentiation is indicated by the expression of osteocalcin, the ability of the cells to mineralize, and the expression of alkaline phosphatase, 45

all of which can be measured by routine methods known in the art .

Zalpha5 activity may also be detected using assays designed to measure zalphaδ- induced production of one or more additional growth factors or other macromolecules . Preferred such assays include those for determining the presence of hepatocyte growth factor (HGF) , epidermal growth factor (EGF) , transforming growth factor alpha (TGFα) , interleukin-6 (IL-6) , VEGF, acidic fibroblast growth factor (aFGF) , angiogenin, and other macromolecules produced by the liver. Suitable assays include mitogenesis assays using target cells responsive to the macromolecule of interest, receptor-binding assays, competition binding assays, immunological assays (e.g., ELISA) , and other formats known in the art. Metalloprotease secretion is measured from treated primary human dermal fibroblasts, synoviocytes and chondrocytes . The relative levels of collagenase, gelatinase and stromalysin produced in response to culturing in the presence of a zalpha5 protein is measured using zymogram gels (Loita and Stetler- Stevenson, Cancer Biology 1:96-106, 1990).

Procollagen/collagen synthesis by dermal fibroblasts and chondrocytes in response to a test protein is measured using ³H-proline incorporation into nascent secreted collagen. ³H-labeled collagen is visualized by SDS-PAGE followed by autoradiography (Unemori and Amento, J^". Biol .

Chem . 265 : 10681-10685, 1990) . Glycosaminoglycan (GAG) secretion from dermal fibroblasts and chondrocytes is measured using a 1 , 9-dimethylmethylene blue dye binding assay (Farndale et al . , Biochim . Biophys . Acta 883 :173-

177, 1986) . Collagen and GAG assays are also carried out in the presence of IL-lβ or TGF-β to examine the ability of zalpha5 protein to modify the established responses to these cytokines. 46

Monocyte activation assays are carried out (1) to look for the ability of zalphaδ proteins to further stimulate monocyte activation, and (2) to examine the ability of zalphaδ proteins to modulate attachment - induced or endotoxin-induced monocyte activation

(Fuhlbrigge et al . , J. Immunol . 13j): 3799-3802, 1987).

IL-lβ and TNFα levels produced in response to activation are measured by ELISA (Biosource, Inc. Camarillo, CA) . Monocyte/macrophage cells, by virtue of CD14 (LPS receptor) , are exquisitely sensitive to endotoxin, and proteins with moderate levels of endotoxin-like activity will activate these cells.

Hematopoietic activity of zalpha5 proteins can be assayed on various hematopoietic cells in culture. Preferred assays include primary bone marrow colony assays and later stage lineage-restricted colony assays, which are known in the art (e.g., Holly et al . , WIPO Publication WO 95/21920) . Marrow cells plated on a suitable semi-solid medium (e.g., 50% methylcellulose containing 15% fetal bovine serum, 10% bovine serum albumin, and 0.6% PSN antibiotic mix) are incubated in the presence of test polypeptide, then examined microscopically for colony formation. Known hematopoietic factors are used as controls. Mitogenic activity of zalpha5 polypeptides on hematopoietic cell lines can be measured as disclosed above.

Cell migration is assayed essentially as disclosed by Kahler et al . {Arteriosclerosis, Thrombosis , and Vascular Biology 12:932-939, 1997). A protein is considered to be chemotactic if it induces migration of cells from an area of low protein concentration to an area of high protein concentration. The assay is performed using modified Boyden chambers with a polystryrene membrane separating the two chambers (Transwell; Corning Costar Corp.) . The test sample, diluted in medium containing 1% BSA, is added to the 47

lower chamber of a 24-well plate containing Transwells. Cells are then placed on the Transwell insert that has been pretreated with 0.2% gelatin. Cell migration is measured after 4 hours of incubation at 37°C. Non- migrating cells are wiped off the top of the Transwell membrane, and cells attached to the lower face of the membrane are fixed and stained with 0.1% crystal violet . Stained cells are then extracted with 10% acetic acid and absorbance is measured at 600 nm. Migration is then calculated from a standard calibration curve. Cell migration can also be measured using the matrigel method of Grant et al . ("Angiogenesis as a component of epithelial -mesenchymal interactions" in Goldberg and Rosen, Epi thelial -Mesenchymal Interaction in Cancer, Birkhauser Verlag, 1995, 235-248; Baatout, Anticancer

Research 12:451-456, 1997) .

Cell adhesion activity is assayed essentially as disclosed by LaFleur et al . (J. Biol . Chem . 222:32798-

32803, 1997) . Briefly, microtiter plates are coated with the test protein, non-specific sites are blocked with BSA, and cells (such as smooth muscle cells, leukocytes, or endothelial cells) are plated at a density of approximately 10⁴ - 10^s cells/well. The wells are incubated at 37°C (typically for about 60 minutes) , then non-adherent cells are removed by gentle washing.

Adhered cells are quantitated by conventional methods

(e.g., by staining with crystal violet, lysing the cells, and determining the optical density of the lysate) .

Control wells are coated with a known adhesive protein, such as fibronectin or vitronectin.

Assays for angiogenic activity are also known in the art. For example, the effect of zalphaδ proteins on primordial endothelial cells in angiogenesis can be assayed in the chick chorioallantoic membrane angiogenesis assay (Leung, Science 246 : 1306-1309 , 1989; 48

Ferrara, Ann. NY Acad. Sci . 752 :246-256, 1995). Briefly, a small window is cut into the shell of an eight-day old fertilized egg, and a test substance is applied to the chorioallantoic membrane. After 72 hours, the membrane is examined for neovascularization. Other suitable assays include microinjection of early stage quail { Coturnix coturnix japonica) embryos as disclosed by

Drake et al . { Proc . Natl . Acad. Sci . USA .92_.: 7657-7661 ,

1995) ; the rodent model of corneal neovascularization disclosed by Muthukkaruppan and Auerbach { Science

205 : 1416-1418 , 1979), wherein a test substance is inserted into a pocket in the cornea of an inbred mouse; and the hampster cheek pouch assay (Hόckel et al . , Arch .

Surg. 121:423-429, 1993) . Induction of vascular permeability, which is indicative of angiogenic activity, is measured in assays designed to detect leakage of protein from the vasculature of a test animal (e.g., mouse or guinea pig) after administration of a test compound (Miles and Miles, J^". Physiol . 118:228-257, 1952; Feng et al . , J^". Exp . Med . 183 :1981-1986, 1996). In vi tro assays for angiogenic activity include the tridimensional collagen gel matrix model (Pepper et al . Biochem .

Biophys . Res . Comm . 189:824-831, 1992 and Ferrara et al . ,

Ann . NY Acad . Sci . 732:246-256, 1995), which measures the formation of tube-like structures by microvascular endothelial cells; and matrigel models (Grant et al . , ibid . ) , which are used to determine effects on cell migration and tube formation by endothelial cells seeded in matrigel, a basement membrane extract enriched in laminin. It is preferred to carry out angiogenesis assays in the presence and absence of VEGF to assess possible combinatorial effects. It is also preferred to use VEGF as a control within in vivo assays. 4 9

Angiogenic factors are also expected to find use in the reduction or prevention of restenosis following invasive procedures such as balloon angioplasty and stent placement. For example, VEGF has been shown to promote vessel re-endothelialization and to reduce intimal hyperplasia in animal models of restenosis (Asahara et al . , Circula tion 11:2802-2809, 1995; Callow et al . , Growth Factors JLO.: 223-228 , 1994); efficacy of zalpha5 polypeptides can be tested in these and other known models.

Stimulation of coronary collateral growth can be measured in known animal models, including a rabbit model of peripheral limb ischemia and hind limb ischemia and a pig model of chronic myocardial ischemia (Ferrara et al . , Endocrine Reviews 18:4-25, 1997). Zalpha5 proteins are assayed in the presence and absence of VEGF and basic FGF to test for combinatorial effects. These models can be modified by the use of adenovirus or naked DNA for gene delivery as disclosed in more detail below, resulting in local expression of the test protein (s) .

Efficacy of zalpha5 polypeptides in promoting wound healing can be assayed in animal models. One such model is the linear skin incision model of Mustoe et al .

{ Science 237 : 1333 , 1987). In a typical procedure, a 6- cm incision is made in the dorsal pelt of an adult rat, then closed with wound clips. Test substances and controls (in solution, gel, or powder form) are appied before primary closure. It is preferred to limit administration to a single application, although additional applications can be made on succeeding days by careful injection at several sites under the incision. Wound breaking strength is evaluated between 3 and 21 days post wounding. In a second model, multiple, small, full-thickness excisions are made on the ear of a rabbit. The cartilage in the ear splints the wound, removing the variable of wound contraction from the evaluation of 50

closure. Experimental treatments and controls are applied. The geometry and anatomy of the wound site allow for reliable quantification of cell ingrowth and epithelial migration, as well as quantitative analysis of the biochemistry of the wounds (e.g., collagen content) . See, Mustoe et al . , J. Clin . Invest . 12=694, 1991. The rabbit ear model can be modified to create an ischemic wound environment, which more closely resembles the clinical situation (Ahn et al . , Ann . Plas t . Surg. 24 : 17 , 1990) . Within a third model, healing of partial- thickness skin wounds in pigs or guinea pigs is evaluated (LeGrand et al . , Growth Factors 1:307, 1993).

Experimental treatments are applied daily on or under dressings. Seven days after wounding, granulation tissue thickness is determined. This model is preferred for dose-response studies, as it is more quantitative than other in vivo models of wound healing. A full thickness excision model can also be employed. Within this model, the epidermis and dermis are removed down to the panniculus carnosum in rodents or the subcutaneous fat in pigs. Experimental treatments are applied topically on or under a dressing, and can be applied daily if desired. The wound closes by a combination of contraction and cell ingrowth and proliferation. Measurable endpoints include time to wound closure, histologic score, and biochemical parameters of wound tissue. Impaired wound healing models are also known in the art (e.g., Cromack et al . , Surgery 113 :36, 1993; Pierce et al . , Proc . Natl . Acad .

Sci . USA 16:2229, 1989; Greenhalgh et al . , Amer. J^". Pa thol . 1H:1235, 1990). Delay or prolongation of the wound healing process can be induced pharmacologically by treatment with steroids, irradiation of the wound site, or by concomitant disease states (e.g., diabetes) . Linear incisions or full-thickness excisions are most commonly used as the experimental wound. Endpoints are 51

as disclosed above for each type of wound. Subcutaneous implants can be used to assess compounds acting in the early stages of wound healing (Broadley et al . , Lab .

Invest . 61:571, 1985; Sprugel et al . , Amer . J. Pa thol . 129 : 601, 1987) . Implants are prepared in a porous, relatively non- inflammatory container (e.g., polyethylene sponges or expanded polytetrafluoroethylene implants filled with bovine collagen) and placed subcutaneously in mice or rats. The interior of the implant is empty of cells, producing a "wound space" that is well-defined and separable from the preexisting tissue. This arrangement allows the assessment of cell influx and cell type as well as the measurement of vasculogenesis/angiogenesis and extracellular matrix production. Expression of zalpha5 polynucleotides in animals provides models for further study of the biological effects of overproduction or inhibition of protein activity in vivo . Zalpha5 -encoding polynucleotides and antisense polynucleotides can be introduced into test animals, such as mice, using viral vectors or naked DNA, or transgenic animals can be produced.

One in vivo approach for assaying proteins of the present invention utilizes viral delivery systems. Exemplary viruses for this purpose include adenovirus, herpesvirus, retroviruses, vaccinia virus, and adeno- associated virus (AAV) . Adenovirus, a double-stranded DNA virus, is currently the best studied gene transfer vector for delivery of heterologous nucleic acids. For review, see Becker et al . , Meth . Cell Biol . 43 : 161-89,

1994; and Douglas and Curiel, Science & Medicine 4_.: 44 -53,

1997. The adenovirus system offers several advantages. Adenovirus can (i) accommodate relatively large DNA inserts; (ii) be grown to high-titer; (iii) infect a broad range of mammalian cell types; and (iv) be used 52

with many different promoters including ubiquitous, tissue specific, and regulatable promoters. Because adenoviruses are stable in the bloodstream, they can be administered by intravenous injection. By deleting portions of the adenovirus genome, larger inserts (up to 7 kb) of heterologous DNA can be accommodated. These inserts can be incorporated into the viral DNA by direct ligation or by homologous recombination with a co-transfected plasmid. In an exemplary system, the essential El gene is deleted from the viral vector, and the virus will not replicate unless the El gene is provided by the host cell (e.g., the human 293 cell line) . When intravenously administered to intact animals, adenovirus primarily targets the liver. If the adenoviral delivery system has an El gene deletion, the virus cannot replicate in the host cells. However, the host's tissue (e.g., liver) will express and process (and, if a signal sequence is present, secrete) the heterologous protein. Secreted proteins will enter the circulation in the highly vascularized liver, and effects on the infected animal can be determined.

An alternative method of gene delivery comprises removing cells from the body and introducing a vector into the cells as a naked DNA plasmid. The transformed cells are then re- implanted in the body. Naked DNA vectors are introduced into host cells by methods known in the art, including transfection, electroporation, microinjection, transduction, cell fusion, DEAE dextran, calcium phosphate precipitation, use of a gene gun, or use of a DNA vector transporter. See, Wu et al . , J. Biol . Chem . 263 : 14621-14624 , 1988; Wu et al . , J^". Biol . Chem . 267 : 963-967, 1992; and Johnston and Tang, Meth. Cell Biol . 4_3: 353-365, 1994.

Transgenic mice, engineered to express a zalpha5 gene, and mice that exhibit a complete absence of zalphaδ gene function, referred to as "knockout mice" 53

(Snouwaert et al . , Science 257 : 1083 , 1992), can also be generated (Lowell et al . , Na ture 3_6_6: 740-742 , 1993).

These mice can be employed to study the zalpha5 gene and the protein encoded thereby in an in vivo system. Transgenic mice are particularly useful for investigating the role of zalpha5 proteins in early development in that they allow the identification of developmental abnormalities or blocks resulting from the over- or underexpression of a specific factor. See also, Maisonpierre et al . , Science 277 :55-60, 1997 and Hanahan,

Science 277 :48-50 , 1997. Preferred promoters for transgenic expression include promoters from metallothionein and albumin genes.

Antisense methodology can be used to inhibit zalpha5 gene transcription to examine the effects of such inhibition in vivo . Polynucleotides that are complementary to a segment of a zalpha5 -encoding polynucleotide (e.g., a polynucleotide as set forth in SEQ ID N0:1) are designed to bind to zalpha5 -encoding mRNA and to inhibit translation of such mRNA. Such antisense oligonucleotides can also be used to inhibit expression of zalpha5 polypeptide-encoding genes in cell culture .

Zalpha5 polypeptides may be used therapeutically to stimulate tissue development or repair, or cellular differentiation or proliferation. Specific applications include, without limitation: the treatment of full -thickness skin wounds, including venous stasis ulcers and other chronic, non-healing wounds, particularly in cases of compromised wound healing due to diabetes mellitus, connective tissue disease, smoking, burns, and other exacerbating conditions; fracture repair; skin grafting; within reconstructive surgery to promote neovascularization and increase skin flap survival; to establish vascular networks in transplanted 54

cells and tissues, such as transplanted islets of Langerhans; to treat female reproductive tract disorders, including acute or chronic placental insufficiency (an important factor causing perinatal morbidity and mortality) and prolonged bleeeding; to promote the growth of tissue damaged by periodontal disease; to promote endothelialization of vascular grafts and stents; in the treatment of acute and chronic lesions of the gastrointestinal tract, including duodenal ulcers, which are characterized by a deficiency of microvessels ; to promote angiogenesis and prevent neuronal degeneration due to chronic cerebral ischemia; to accelerate the formation of collateral blood vessels in ischemic limbs; to promote vessel re-endothelialization and to reduce intimal hyperplasia following invasive procedures such as balloon angioplasty and stent placement; to promote vessel repair and development of collateral circulation following myocardial infarction so as to limit ischemic injury; and to stimulate hematopoiesis . The polypeptides are also useful additives in tissue adhesives for promoting revascularization of the healing tissue. Zalpha5 polypeptides can be administered alone or in combination with other vasculogenic or angiogenic agents, including VEGF. For example, basic and acidic FGFs and VEGF have been found to play a role in the development of collateral circulation, and the combined use of zalpha5 with one or more of these factors may be advantageous . VEGF has also been implicated in the survival of transplanted islet cells (Gorden et al . Transplantation 61:436-443, 1997; Pepper, Arterioscl erosis , Throm . and

Vascular Biol . 17:605-619, 1997). Basic FGF has been shown to induce angiogenesis and accelerate healing of ulcers in experimental animals (reviewed by Folkman, Nature Medi cine 1:27-31, 1995) . When using zalpha5 in combination with an additional agent, the two compounds 55

can be administered simultaneously or sequentially as appropriate for the specific condition being treated.

For pharmaceutical use, zalpha5 proteins are formulated for topical or parenteral, particularly intravenous or subcutaneous, delivery according to conventional methods. In general, pharmaceutical formulations will include a zalpha5 polypeptide in combination with a pharmaceutically acceptable vehicle, such as saline, buffered saline, 5% dextrose in water, or the like. Formulations may further include one or more excipients, preservatives, solubilizers , buffering agents, albumin to prevent protein loss on vial surfaces, etc. Methods of formulation are well known in the art and are disclosed, for example, in Remington : The Science and Practice of Pharmacy, Gennaro, ed. , Mack Publishing Co., Easton, PA, 19th ed., 1995. Zalpha5 will preferably be used in a concentration of about 10 to 100 μg/ml of total volume, although concentrations in the range of 1 ng/ml to 1000 μg/ml may be used. For topical application, such as for the promotion of wound healing, the protein will be applied in the range of 0.1-10 μg/cm² of wound area, with the exact dose determined by the clinician according to accepted standards, taking into account the nature and severity of the condition to be treated, patient traits, etc. Determination of dose is within the level of ordinary skill in the art. The therapeutic formulations will generally be administered over the period required for neovascularization, typically from one to several months and, in treatment of chronic conditions, for a year or more. Dosing is daily or intermittently over the period of treatment . Intravenous administration will be by bolus injection or infusion over a typical period of one to several hours. Sustained release formulations can also be employed. In general, a therapeutically effective amount of zalpha5 is an amount sufficient to produce a clinically significant 56

change in the treated condition, such as a clinically significant reduction in time required for wound closure, a significant reduction in wound area, a significant improvement in vascularization, a significant reduction in morbidity, or a significantly increased histological score .

Proteins of the present invention are useful for modulating the proliferation, differentiation, migration, or metabolism of responsive cell types, which include both primary cells and cultured cell lines. Of particular interest in this regard are hematopoietic cells (including stem cells and mature myeloid and lymphoid cells), endothelial cells, smooth muscle cells, fibroblasts, and hepatocytes . Zalpha5 polypeptides are added to tissue culture media for these cell types at a concentration of about 10 pg/ml to about 100 ng/ml. Those skilled in the art will recognize that zalpha5 proteins can be advantageously combined with other growth factors in culture media. Within the laboratory research field, zalpha5 proteins can also be used as molecular weight standards or as reagents in assays for determining circulating levels of the protein, such as in the diagnosis of disorders characterized by over- or under-production of zalpha5 protein or in the analysis of cell phenotype .

Zalphaδ proteins can also be used to identify inhibitors of their activity. Test compounds are added to the assays disclosed above to identify compounds that inhibit the activity of zalpha5 protein. In addition to those assays disclosed above, samples can be tested for inhibition of zalphaδ activity within a variety of assays designed to measure receptor binding or the stimulation/inhibition of zalpha5-dependent cellular responses. For example, zalphaδ-responsive cell lines can be transfected with a reporter gene construct that is responsive to a zalpha5-stimulated cellular pathway. Reporter gene constructs of this type are known in the 57

art, and will generally comprise a zalpha5-activated serum response element (SRE) operably linked to a gene encoding an assayable protein, such as luciferase. Candidate compounds, solutions, mixtures or extracts are tested for the ability to inhibit the activity of zalpha5 on the target cells as evidenced by a decrease in zalpha5 stimulation of reporter gene expression. Assays of this type will detect compounds that directly block zalpha5 binding to cell-surface receptors, as well as compounds that block processes in the cellular pathway subsequent to receptor-ligand binding. In the alternative, compounds or other samples can be tested for direct blocking of zalpha5 binding to receptor using zalpha5 tagged with a detectable label (e.g., ¹²⁵I, biotin, horseradish peroxidase, FITC, and the like) . Within assays of this type, the ability of a test sample to inhibit the binding of labeled zalphaδ to the receptor is indicative of inhibitory activity, which can be confirmed through secondary assays . Receptors used within binding assays may be cellular receptors or isolated, immobilized receptors .

The invention further provides polypeptides that comprise an epitope-bearing portion of a protein as shown in SEQ ID NO: 2. An "epitope" is a region of a protein to which an antibody can bind. See, for example, Geysen et al . , Proc . Natl . Acad . Sci . USA 11:3998-4002,

1984. Epitopes can be linear or conformational , the latter being composed of discontinuous regions of the protein that form an epitope upon folding of the protein. Linear epitopes are generally at least 6 amino acid residues in length. Relatively short synthetic peptides that mimic part of a protein sequence are routinely capable of eliciting an antiserum that reacts with the partially mimicked protein. See, Sutcliffe et al . , Science 219 : 660-666 , 1983. Antibodies that recognize short, linear epitopes are particularly useful in 58

analytic and diagnostic applications that employ denatured protein, such as Western blotting (Tobin, Proc .

Natl . Acad. Sci . USA 21:4350-4356, 1979).

Antigenic, epitope-bearing polypeptides of the present invention are useful for raising antibodies, including monoclonal antibodies, that specifically bind to a zalpha5 protein. Antigenic, epitope-bearing polypeptides contain a sequence of at least six, preferably at least nine, more preferably from 15 to about 30 contiguous amino acid residues of a zalpha5 protein (e.g., SEQ ID NO:2) . Polypeptides comprising a larger portion of a zalphaδ protein, i.e. from 30 to 50 residues up to the entire sequence are included. It is preferred that the amino acid sequence of the epitope- bearing polypeptide is selected to provide substantial solubility in aqueous solvents, that is the sequence includes relatively hydrophilic residues, and hydrophobic residues are substantially avoided. Preferred such regions include residues 19-42, 88-108, 88-116, and 404- 430 of SEQ ID NO : 2.

As used herein, the term "antibodies" includes polyclonal antibodies, monoclonal antibodies, antigen- binding fragments thereof such as F(ab')2 ^an( Fab fragments, single chain antibodies, and the like, including genetically engineered antibodies. Non-human antibodies may be humanized by grafting non-human CDRs onto human framework and constant regions, or by incorporating the entire non-human variable domains

(optionally "cloaking" them with a human-like surface by replacement of exposed residues, wherein the result is a "veneered" antibody) . In some instances, humanized antibodies may retain non-human residues within the human variable region framework domains to enhance proper binding characteristics. Through humanizing antibodies, biological half-life may be increased, and the potential for adverse immune reactions upon administration to 59

humans is reduced. One skilled in the art can generate humanized antibodies with specific and different constant domains (i.e., different Ig subclasses) to facilitate or inhibit various immune functions associated with particular antibody constant domains. Antibodies are defined to be specifically binding if they bind to a zalphaδ polypeptide or protein with an affinity at least

10-fold greater than the binding affinity to control

(non-zalpha5) polypeptide or protein. The affinity of a monoclonal antibody can be readily determined by one of ordinary skill in the art (see, for example, Scatchard,

Ann. NY Acad . Sci . 5L: 660-672, 1949).

Methods for preparing polyclonal and monoclonal antibodies are well known in the art (see for example, Hurrell, J. G. R. , Ed., Monoclonal Hybridoma Antibodies :

Techniques and Appli cations, CRC Press, Inc., Boca Raton,

FL, 1982, which is incorporated herein by reference) . As would be evident to one of ordinary skill in the art, polyclonal antibodies can be generated from a variety of warm-blooded animals such as horses, cows, goats, sheep, dogs, chickens, rabbits, mice, and rats. The immunogenicity of a zalphaδ polypeptide may be increased through the use of an adjuvant such as alum (aluminum hydroxide) or Freund's complete or incomplete adjuvant. Polypeptides useful for immunization also include fusion polypeptides, such as fusions of a zalpha5 polypeptide or a portion thereof with an immunoglobulin polypeptide or with maltose binding protein. The polypeptide immunogen may be a full-length molecule or a portion thereof. If the polypeptide portion is "hapten-like" , such portion may be advantageously joined or linked to a macromolecular carrier (such as keyhole limpet hemocyanin

(KLH) , bovine serum albumin (BSA) or tetanus toxoid) for immunization . Alternative techniques for generating or selecting antibodies include in vi tro exposure of 60

lymphocytes to zalpha5 polypeptides, and selection of antibody display libraries in phage or similar vectors

(e.g., through the use of immobilized or labeled zalphaδ polypeptide) . Human antibodies can be produced in transgenic, non-human animals that have been engineered to contain human immunoglobulin genes as disclosed in

WIPO Publication WO 98/24893. It is preferred that the endogenous immunoglobulin genes in these animals be inactivated or eliminated, such as by homologous recombination.

A variety of assays known to those skilled in the art can be utilized to detect antibodies which specifically bind to zalphaδ polypeptides. Exemplary assays are described in detail in Antibodies : A Labora tory Manual , Harlow and Lane (Eds.), Cold Spring

Harbor Laboratory Press, 1988. Representative examples of such assays include: concurrent immunoelectrophoresis, radio- immunoassays, radio- immunoprecipitations , enzyme- linked immunosorbent assays (ELISA) , dot blot assays, Western blot assays, inhibition or competition assays, and sandwich assays.

Antibodies to zalpha5 may be used for affinity purification of the protein, within diagnostic assays for determining circulating levels of the protein; for detecting or quantitating soluble zalpha5 polypeptide as a marker of underlying pathology or disease; for immunolocalization within whole animals or tissue sections, including immunodiagnostic applications; for immunohistochemistry; and as antagonists to block protein activity in vi tro and in vivo . Antibodies to zalpha5 may also be used for tagging cells that express zalpha5 ,- for affinity purification of zalpha5 polypeptides and proteins; in analytical methods employing FACS; for screening expression libraries; and for generating anti- idiotypic antibodies. Antibodies can be linked to other compounds, including therapeutic and diagnostic agents, 61

using known methods to provide for targetting of those compounds to cells expressing receptors for zalpha5. For certain applications, including in vi tro and in vivo diagnostic uses, it is advantageous to employ labeled antibodies . Suitable direct tags or labels include radionuclides , enzymes, substrates, cofactors, inhibitors, fluorescent markers, chemiluminescent markers, magnetic particles and the like; indirect tags or labels may feature use of biotin-avidin or other complement/anti-complement pairs as intermediates. Antibodies of the present invention may also be directly or indirectly conjugated to drugs, toxins, radionuclides and the like, and these conjugates used for in vivo diagnostic or therapeutic applications (e . g . , inhibition of cell proliferation) . See, in general, Ramakrishnan et al., Cancer Res . 51:1324-1330, 1996.

Polypeptides and proteins of the present invention can be used to identify and isolate receptors. Zalpha5 receptors may be involved in growth regulation in the liver, blood vessel formation, and other developmental processes. For example, zalpha5 proteins and polypeptides can be immobilized on a column, and membrane preparations run over the column (as generally disclosed in Immobilized Affinity Ligand Techniques, Hermanson et al . , eds., Academic Press, San Diego, CA, 1992, pp.195-202). Proteins and polypeptides can also be radiolabeled {Methods Enzymol . , vol. 182, "Guide to

Protein Purification", M. Deutscher, ed. , Academic Press, San Diego, 1990, 721-737) or photoaffinity labeled (Brunner et al . , Ann . Rev. Biochem . 62_.:483-514, 1993 and

Fedan et al . , Biochem. Pharmacol . 11:1167-1180, 1984) and used to tag specific cell-surface proteins. In a similar manner, radiolabeled zalpha5 proteins and polypeptides can be used to clone the cognate receptor in binding 62

assays using cells transfected with an expression cDNA library.

The present invention also provides reagents for use in diagnostic applications. For example, the zalpha5 gene, a probe comprising zalphaδ DNA or RNA, or a subsequence thereof can be used to determine if the zalpha5 gene is present on chromosome 1 or if a mutation has occurred. Detectable chromosomal aberrations at the zalpha5 gene locus include, but are not limited to, aneuploidy, gene copy number changes, insertions, deletions, restriction site changes, and rearrangements. These aberrations can occur within the coding sequence, within introns, or within flanking sequences, including upstream promoter and regulatory regions, and may be manifested as physical alterations within a coding sequence or changes in gene expression level . Analytical probes will generally be at least 20 nucleotides in length, although somewhat shorter probes (14-17 nucleotides) can be used. PCR primers are at least 5 nucleotides in length, preferably 15 or more nt , more preferably 20-30 nt . Short polynucleotides can be used when a small region of the gene is targetted for analysis. For gross analysis of genes, a polynucleotide probe may comprise an entire exon or more. Probes will generally comprise a polynucleotide linked to a signal- generating moiety such as a radionucleotide . In general, these diagnostic methods comprise the steps of (a) obtaining a genetic sample from a patient; (b) incubating the genetic sample with a polynucleotide probe or primer as disclosed above, under conditions wherein the polynucleotide will hybridize to complementary polynucleotide sequence, to produce a first reaction product; and (c) comparing the first reaction product to a control reaction product. A difference between the first reaction product and the control reaction product is indicative of a genetic abnormality in the patient. Genetic samples for use within the present invention 63

include genomic DNA, cDNA, and RNA. The polynucleotide probe or primer can be RNA or DNA, and will comprise a portion of SEQ ID N0:1, the complement of SEQ ID N0:1, or an RNA equivalent thereof. Suitable assay methods in this regard include molecular genetic techniques known to those in the art, such as restriction fragment length polymorphism (RFLP) analysis, short tandem repeat (STR) analysis employing PCR techniques, ligation chain reaction (Barany, PCR Methods and Applica tions 1:5-16, 1991), ribonuclease protection assays, and other genetic linkage analysis techniques known in the art (Sambrook et al . , ibid . ; Ausubel et . al . , ibid . ; A.J. Marian, Chest

108:255-65, 1995) . Ribonuclease protection assays (see, e.g., Ausubel et al . , ibid. , ch. 4) comprise the hybridization of an RNA probe to a patient RNA sample, after which the reaction product (RNA-RNA hybrid) is exposed to RNase. Hybridized regions of the RNA are protected from digestion. Within PCR assays, a patient genetic sample is incubated with a pair of polynucleotide primers, and the region between the primers is amplified and recovered. Changes in size or amount of recovered product are indicative of mutations in the patient. Another PCR-based technique that can be employed is single strand conformational polymorphism (SSCP) analysis (Hayashi, PCR Methods and Applica tions 1:34-38, 1991) .

The polypeptides, nucleic acid and/or antibodies of the present invention may be used in diagnosis or treatment of disorders associated with cell loss or abnormal cell proliferation (including cancer), including impaired or excessive vasculogenesis or angiogenesis and liver disease. In particular, cancers of the liver may be amenable to such diagnosis, treatment or prevention. Elevated levels of zalpha5 polypeptides may be indicative of liver damage. Labeled zalphaδ polypeptides may be used for imaging tumors or other sites of abnormal cell proliferation. 64

Because angiogenesis in adult animals is limited to wound healing and the female reproductive cycle, it is a very specific indicator of pathological processes. Angiogenesis is indicative of, for example, developing solid tumors, retinopathies, and arthritis. In general, the probe or antibody is labeled with a moiety that produces a detectable signal, such as a radionuclide, enzyme, contrast agent, or fluorophore, although labeled second antibodies or other labeled secondary agents can be employed. The probe or antibody can be administered to the patient and detected in vivo by conventional scanning methodologies, or can be used to screen biopsy or other tissue samples in vi tro .

Inhibitors of zalpha5 activity (zalpha5 antagonists) include anti-zalpha5 antibodies and soluble zalpha5 receptors, as well as other peptidic and non- peptidic agents (including ribozymes) . Such antagonists can be used to block the effects of zalpha5 on cells or tissues. Of particular interest is the use of antagonists of zalpha5 activity in cancer therapy. As early detection methods improve it becomes possible to intervene at earlier times in tumor development, making it feasible to use inhibitors of growth factors to block cell proliferation, angiogenesis, and other events that lead to tumor development and metastasis. Inhibitors are also expected to be useful in adjunct therapy after surgery to prevent the growth of residual cancer cells. Inhibitors can also be used in combination with other cancer therapeutic agents. Inhibitors of zalphaδ may also prove useful in the treatment of ocular neovascularization, including diabetic retinopathy and age-related macular degeneration. Experimental evidence suggests that these conditions result from the expression of angiogenic factors induced by hypoxia in the retina. 65

Zalpha5 antagonists are also of interest in the treatment of inflammatory disorders, such as rheumatoid arthritis and psoriasis. In rheumatoid arthritis, studies suggest that VEGF plays an important role in the formation of pannus, an extensively vascularized tissue that invades and destroys cartilage. Psoriatic lesions are hypervascular and overexpress the angiogenic polypeptide IL-8.

Zalpha5 antagonists may also prove useful in the treatment of infantile hemangiomas, which exhibit overexpression of VEGF and bFGF during the proliferative phase .

In addition to antibodies, zalphaδ inhibitors include small molecule inhibitors and angiogenically or mitogenically inactive receptor-binding fragments of zalpha5 polypeptides. Inhibitors are formulated for pharmaceutical use as generally disclosed above, taking into account the precise chemical and physical nature of the inhibitor and the condition to be treated. The relevant determinations are within the level of ordinary skill in the formulation art. Other angiogenic and vasculogenic factors, including VEGF and bFGF, have been implicated in pathological neovascularization. In such instances it may be advantageous to combine a zalphaδ inhibitor with one or more inhibitors of these other factors .

Polynucleotides encoding zalpha5 polypeptides are useful within gene therapy applications where it is desired to increase or inhibit zalpha5 activity. For example, Isner et al . , The Lancet (ibid.) reported that

VEGF gene therapy promoted blood vessel growth in an ischemic limb. Additional applications of zalphaδ gene therapy include stimulation of wound healing and repopulation of vascular grafts. Zalphaδ polypeptides and anti-zalpha5 antibodies can be directly or indirectly conjugated to 66

drugs, toxins, radionuclides and the like, and these conjugates used for in vivo diagnostic or therapeutic applications. For instance, polypeptides or antibodies of the present invention may used to identify or treat tissues or organs that express a corresponding anti- complementary molecule (receptor or antigen, respectively, for instance) . More specifically, zalpha5 polypeptides or anti-zalpha5 antibodies, or bioactive fragments or portions thereof, can be coupled to detectable or cytotoxic molecules and delivered to a mammal having cells, tissues, or organs that express the anti-complementary molecule.

Suitable detectable molecules can be directly or indirectly attached to the polypeptide or antibody, and include radionuclides, enzymes, substrates, cofactors, inhibitors, fluorescent markers, chemiluminescent markers, magnetic particles, and the like. Suitable cytotoxic molecules can be directly or indirectly attached to the polypeptide or antibody, and include bacterial or plant toxins (for instance, diphtheria toxin, Pseudomonas exotoxin, ricin, abrin, saporin, and the like) , as well as therapeutic radionuclides, such as iodine-131, rhenium-188 or yttrium-90. These can be either directly attached to the polypeptide or antibody, or indirectly attached according to known methods, such as through a chelating moiety. Polypeptides or antibodies can also be conjugated to cytotoxic drugs, such as adriamycin. For indirect attachment of a detectable or cytotoxic molecule, the detectable or cytotoxic molecule may be conjugated with a member of a complementary/anticomplementary pair, where the other member is bound to the polypeptide or antibody portion. For these purposes, biotin/streptavidin is an exemplary complementary/anticomplementary pair. Polypeptide-toxin fusion proteins or antibody/fragment-toxin fusion proteins may be used for 67

targeted cell or tissue inhibition or ablation, such as in cancer therapy. Of particular interest in this regard are conjugates of a zalphaδ polypeptide and a cytotoxin, which can be used to target the cytotoxin to a tumor or other tissue that is undergoing undesired angiogenesis or neovascularization. Target cells (i.e., those displaying the zsig51 receptor) bind the zsig51-toxin conjugate, which is then internalized, killing the cell. The effects of receptor-specific cell killing (target ablation) are revealed by changes in whole animal physiology or through histological examination. Thus, ligand-dependent , receptor-directed cyotoxicity can be used to enhance understanding of the physiological significance of a protein ligand. A preferred such toxin is saporin. Mammalian cells have no receptor for saporin, which is non-toxic when it remains extracellular .

In another embodiment, zalpha5-cytokine fusion proteins or antibody/fragment-cytokine fusion proteins may be used for enhancing in vi tro cytotoxicity (for instance, that mediated by monoclonal antibodies against tumor targets) and for enhancing in vivo killing of target tissues (for example, blood and bone marrow cancers) . See, generally, Hornick et al . , Blood 19:4437- 4447, 1997) . In general, cytokines are toxic if administered systemically . The described fusion proteins enable targeting of a cytokine to a desired site of action, such as a cell having binding sites for zalpha5, thereby providing an elevated local concentration of cytokine. Suitable cytokines for this purpose include, for example, interleukin-2 and granulocyte-macrophage colony-stimulating factor (GM-CSF) . Such fusion proteins may be used to cause cytokine- induced killing of tumors and other tissues undergoing angiogenesis or neovascularization. 68

In a further embodiment, a zalpha5 polypeptide or anti-zalpha5 antibody can be conjugated with a radionuclide, particularly with a beta-emitting or gamma- emitting radionuclide, and used to reduce restenosis. For instance, iridium-192 impregnated ribbons placed into stented vessels of patients until the required radiation dose was delivered resulted in decreased tissue growth in the vessel and greater luminal diameter than the control group, which received placebo ribbons. Further, revascularisation and stent thrombosis were significantly lower in the treatment group. Similar results are predicted with targeting of a bioactive conjugate containing a radionuclide, as described herein.

The bioactive polypeptide or antibody conjugates described herein can be delivered intravenously, intra-arterially or intraductally, or may be introduced locally at the intended site of action.

The invention is further illustrated by the following non-limiting examples.

69

Examples Example 1

A full length cDNA for zalphaδ was obtained from a plasmid containing an insert corresponding to an expressed sequence tag (EST) from a fetal liver/spleen library (Hillier et al . , The WashU-Merck EST Project, 1995; EST319951; GenBank Accession No. R86161; obtained from Genome Systems, Inc., St. Louis, MO) . The insert was sequenced and found to consist of 1534 bp with an open reading frame of 1383 bp (including a termination codon) . Analysis of the encoded polypeptide showed significant sequence homology to angiopoietin-1 and angiopoietin-2. A second EST (EST135867; GenBank Accession No. T73510; Hillier et al . , The WashU-Merck EST Project, 1985) was also analyzed and was found to be truncated at the 3' end.

Analysis of tissue distribution was performed by the northern blotting technique using Human Multiple Tissue and Master Dot Blots from Clontech Laboratories, Inc. (Palo Alto, CA) . A probe was obtained by restriction digest of the EST135867 clone with EcoRI and Xhol to remove the insert from the vector. The reaction mixture was electrophoresed on a 2% agarose gel, and the 656 bp band corresponding to the original EST was excised and purified using commercially available gel purification reagents and protocol (QIAEX^® II Gel Extraction kit; Qiagen, Inc., Valencia, CA) . The purified DNA was radioactively labeled with ³²P using a commercially available kit (Rediprime™ DNA labeling system; Amersham Corp., Arlington Heights, IL) . The probe was purified using a commercially available push column (NucTrap® column; Stratagene, La Jolla, CA; see U.S. Patent No. 5,336,412) . A commercially available hybridization solution (ExpressHyb™ Hybridization Solution; Clontech Laboratories, Inc., Palo Alto, CA) was used for prehybridization and hybridization. The final 70

hybridization solution contained 8 ml ExpressHyb™ Hybridization Solution, 80 μl sheared salmon sperm DNA (10 mg/ml; obtained from 5 Prime-3 Prime, Boulder, CO), 48 μl human Cot-1 DNA (1 mg/ml, GibcoBRL, Gaithersburg, MD) and 45 μl labeled probe (^"50 ng @ 305,000 CPM/ng) . Hybridization took place overnight at 55°C. After hybridization the blots were washed in 2X SSC, 0.1% SDS at room temperature; then in 2X SSC, 0.1% SDS at 65°C; followed by a 0. IX SSC, 0.1% SDS wash at 65°C. The blots were exposed to film overnight. Three transcript sizes were observed in liver at approximately 5.8 kb, 2.8 kb, and 1.7 kb, and a faint signal was detected in kidney at approximately 2 kb . The dot blot showed signals in liver and fetal liver and a faint signal in kidney. While not wishing to be bound by theory, this expression pattern suggests a possible role for zalphaδ in liver regeneration .

Example 2 Northern analysis was performed on Northern blots of human tumor RNA (Human Tumor Panel Blot I, Human Tumor Panel Blot II, Human Tumor Panel Blot IVb, and Human Tumor Panel Blot V from Invitrogen Corporation, Carlsbad, CA) . A probe was obtained by PCR using primers ZC16,985 (SEQ ID NO: 5) and ZC17,250 (SEQ ID NO: 6) and EST319951 as template, yielding a DNA fragment corresponding to the coding sequence of zalpha5 with flanking restriction sites. The PCR product was analyzed on a 1% agarose gel, and the band of interest was purified using a commercially available kit (QIAEX^® II gel purification reagents and protocol; Qiagen, Inc.) . The purified DNA was labeled with ³²P using a commercially available kit (Rediprime™ DNA labeling system; Amersham Corp.) . The probe was purified using a push column. Hybridization was carried out in 8 mis of hybridization 71

solution (ExpressHyb™; Clontech Laboratories) containing 80μl sheared salmon sperm DNA (lOmg/ml, 5 Prime-3 Prime; Boulder, CO), 48 μl human Cot-1 DNA (lmg/ml, GibcoBRL) and 23μl labeled probe (700,000 CPM/μl). Hybridization took place overnight at 55°C. The blots were then washed in 2X SSC, 0.1%SDS at room temperature; then in 2X SSC, 0.1% SDS at 60°C; followed by a final wash in 0. IX SSC, 0.1% SDS at 60°C. The blots were exposed to film overnight overnight. Two transcript sizes were observed in liver and liver tumor, corresponding to the smaller transcript sizes in the multiple tissue Northern blot for liver. No other signals were observed.

Example 3 Zalpha5 was mapped to human chromosome 1 using the commercially available GeneBridge 4 Radiation Hybrid Panel (Research Genetics, Inc., Huntsville, AL) . This panel contains PCRable DNAs from each of 93 radiation hybrid clones, plus two control DNAs (the HFL donor and the A23 recipient) . A publicly available world-wide web server (http : //www-genome . wi .mit . edu/cgi- bin/contig/rhmapper .pi) allows mapping relative to the Whitehead Institute/MIT Center for Genome Research's radiation hybrid map of the human genome (the "WICGR" radiation hybrid map) , which was constructed with the GeneBridge 4 Radiation Hybrid Panel.

For the mapping of zalpha5, 20-μl reaction mixtures were set up in a PCRable 96 -well microtiter plate (Stratagene, La Jolla, CA) and used in a thermal cycler (RoboCycler^® Gradient 96; Stratagene). Each of the 95 PCR reactions contained 2 μl buffer (10X KlenTaq PCR reaction buffer; Clontech Laboratories, Inc., Palo Alto, CA) , 1.6 μl dNTPs mix (2.5 mM each, Perkin-Elmer, Foster City, CA) , 1 μl sense primer ZC 17,332 (SEQ ID NO:7), 1 μl antisense primer ZC 17,336 (SEQ ID NO:8), 2 μl of a density increasing agent and tracking dye 72

(RediLoad, Research Genetics, Inc., Huntsville, AL) , 0.4 μl of a commercially available DNA polymerase/antibody mix (5OX Advantage™ KlenTaq Polymerase Mix; Clontech Laboratories, Inc.), 25 ng of DNA from an individual hybrid clone or control and x μl ddH₂0 for a total volume of 20 μl . The mixtures were overlaid with an equal amount of mineral oil and sealed. The PCR cycler conditions were as follows : an initial 5 minute denaturation at 95°C; 35 cycles of a 1 minute denaturation at 95°C, 1 minute annealing at 58°C and 1.5 minute extension at 72 °C; followed by a final extension of 7 minutes at 72 °C. The reaction products were separated by electrophoresis on a 2% agarose gel (Life Technologies, Gaithersburg, MD) . The results showed that zalpha5 maps 2.74 cR_3000 from the framework marker D1S230 on the chromosome 1 WICGR radiation hybrid map. Proximal and distal framework markers were D1S230 and WI-9515 (D1S2423), respectively. The use of surrounding markers positions zalpha5 in the Ip31.1-p22.3 region on the integrated LDB chromosome 1 map (The Genetic Location Database, University of Southhampton, WWW server: http : //cedar . genetics . soton . ac . uk/public_html/) .

Example 4

A full-length mouse zalphaδ DNA sequence was constructed by splicing together two ESTs. ESTs 917237 and 744974 were obtained from the IMAGE Consortium. EST917237 contained the 3' end of the mouse zalpha5 sequence. EST744974 contained both 5' and 3' ends but contained an internal alternative splice. Splicing of the two clones was done by PCR. A first reaction was carried out with primers ZC18,979 (SEQ ID NO: 10) and ZC18,967 (SEQ ID NO: 11) with EST744974 as a template. The second reaction was carried out with primers ZC18,969 (SEQ ID NO:12) and ZC18,968 (SEQ ID NO:13) with EST917237 73

as a template. The reactions were run in a thermocycler with incubation at 94°C for 1% minute; thirty- five cycles at 94°C for 15 seconds, 62°C for 30 seconds, 68°C for 30 seconds; incubation at 68°C for 10 minutes; followed by a 4°C hold. A dilution from both reactions was used as a template in a third reaction with primers ZC18,969 (SEQ ID NO:12) and ZC18,979 (SEQ ID NO:10). Thermocycler conditions were as in the previous reactions. The products of the third reaction were separated by gel electrophoreses, and the 1.36 kb product was purified by adsorption to silica gel particles (QIAEX® II gel extraction kit; Qiagen, Valencia, CA) . The fragment was subcloned into the vector pCR®2.1-T0P0 using a commercially available cloning kit (Invitrogen) . Positive clones were sequenced for PCR errors. The full- length mouse zalpha5 sequence is shown in SEQ ID NO: 14.

Example 5

Expression constructs were made for eukaryotic cell production of N- and C-terminally glu-glu tagged proteins. A mammalian expression vector was constructed with the dihyrofolate reductase gene under control of the SV40 early promoter and SV40 polyadenylation site, and a cloning site to insert the gene of interest under control of the mouse MT-1 promoter and the hGH polyadenylation site. The expression vector was designated pZP-9 and was deposited at the American Type Culture Collection, 12301 Parklawn Drive, Rockville, MD on February 20, 1998 under Accession Number 98668. To facilitate purification of the protein of interest, the pZP-9 vector was modified by addition of the tPA leader sequence (U.S. Patent 5,641,655, incorporated herein by reference) and a GluGlu tag (SEQ ID NO: 9) between the MT-1 promoter and hGH terminator. Expression results in an N-terminally tagged fusion protein comprising the tPA leader. The N- 74

terminally tagged vector was designated pZP9NEE. Another vector was similarly constructed with a C-terminal GluGlu tag (SEQ ID NO: 9) inserted just 5' to the hGH terminator and utilizes the native (or other fused) secretory signal sequence for secretion of the encoded polypeptide of interest; expression resulted in a C-terminally tagged protein. The C-terminal GluGlu tagged vector was designated pZP9CEE.

The expression construct zalpha5CEEpZP9 was created by ligation of a PCR-generated zalpha5 sequence into the vector pZP9CEE. EST 319951 (Example 1) was used as a template in a PCR reaction using oligonucleotide primers ZC16,985 (SEQ ID N0:5) and ZC17,250 (SEQ ID NO:6). PCR conditions were: one cycle at 94° for 1'30"; five cycles at 94° for 10", 32° for 20", 72° for 1'30"; 20 cycles at 94° for 10", 64° for 20", 72° for 1'30"; one cycle at 72° for 7'; followed by a 4° hold. The reaction mixture was gel electrophoresed, and the resulting band was purified using silica gel particles (QIAEX® II gel extraction kit; Qiagen, Valencia, CA) . The DNA was digested with EcoRI and BamHI following standard protocols. The reaction mixture was gel electrophoresed, and the band was re-purified. A portion of the purified DNA was ligated into pZP9CEE that had been previously digested with EcoRI and BamHI following standard protocols. The ligated DNA was electroporated into E. coli host cells (Electromax DH10B™ cells; obtained from

Life Technologies, Inc., Gaithersburg, MD) . The resulting colonies were sequenced for the correct insert sequence. Plasmid from a positive colony was purified using an anion-exchange kit (QIAGEN® Plasmid Maxi Kit, Qiagen Inc., Valencia, CA) according to the manufacturer's directions.

A second construct, zalpha5NEEpZP9 , was created by ligation of a PCR-generated zalphaδ sequence into the vector pZP9NEE. EST 319951 was used as a template in a 75

PCR reaction using primers zcl6,987 and zcl7,548. PCR conditions were as disclosed above for zalpha5CEEpZP9. The reaction mixture was gel electrophoresed, and the resulting band purified using silica gel particles. The DNA was digested with BamHI and Xbal following standard protocols. The reaction mixture was gel electrophoresed, and the band was re-purified. A portion of the purified DNA was ligated into pZP9NEE that had been digested with BamHI and Xbal following standard protocols. The ligated DNA was electroporated into E. coli host cells

(Electromax DH10B™) , sequenced, and purified as above.

Example 6

Three expression vectors were prepared to express zalpha5 polypeptides in insect cells: palpha5 , designed to express a zalpha5 polypeptide with its native leader and no tag; zalpha5CEE pzbv4L, designed to express a zalpha5 polypeptide with a C-terminal Glu-Glu tag; and zalpha5NEE pzbv4L, designed to express a zalpha5 polypeptide with an N-terminal Glu-Glu tag.

To construct palphaδ, a commercially available baculovirus expression vector (pFastBac; Life Technologies, Gaithersburg, MD) was modified to replace the polyhedron promoter with the late activating basic protein promoter (Wilson et al . , Virology 61 : 661-666 ,

1987). The resulting vector was designated pZBV3L. 81.8 ng of a 1394-bp EcoRI/Xba I zalpha5 fragment and 95.6 ng of the EcoRI/Xba I-digested pZBV3L were ligated overnight. The ligation mixture was diluted 3 -fold in TE (10 mM Tris-HCI, pH 7.5 and 1 mM EDTA), and 4 fmol of the diluted ligation mixture was transformed into competent E. coli cells (Library Efficiency DH5α™ competent cells;

Life Technologies) according to the supplier's directions by heat shock for 45 seconds in a 42 °C waterbath. The ligated DNA was diluted in 450 μl of SOC media (2% Bacto™ 76

Tryptone (Difco Laboratories, Detroit, MI), 0.5% Bacto™ Yeast Extract (Difco Laboratories), 10 ml IM NaCl, 1.5 mM KCI, 10 mM MgCl₂, 10 mM MgS0₄ and 20 mM glucose) and plated onto LB plates containing 100 μg/ml ampicillin. Clones were analyzed by restriction digestion, and 1 μl of a positive clone was transformed into 20 μl of competent E. coli host cells (Max Efficiency DHlOBac™ competent cells; Life Technologies) according to the supplier's instructions by heat shock for 45 seconds in a 42°C waterbath. The ligated DNA was diluted in 980 μl SOC media and plated onto Luria Agar plates containing 50 μg/ml kanamycin, 7 μg/ml gentamicin, 10 μg/ml tetracycline, IPTG and halogenated indolyl-β-D- galactoside (bluo-gal; Life Technologies) . The cells were incubated for 48 hours at 37°C. A color selection was used to identify those cells having virus that had incorporated into the plasmid (referred to as a "bacmid") . Those colonies that were white in color were picked for analysis. Bacmid DNA was isolated from positive colonies using a commercially available kit

(QiaVac Miniprepδ system; Qiagen Inc., Valencia, CA) according the manufacturer's directions. Clones were screened for the correct insert by amplifying DNA using primers to the Basic Protein Promoter and to the SV40 terminator via PCR. Those having the correct insert were used to transfect Spodoptera frugiperda (Sf9) cells.

A second vector, designated zalpha5CEEpZBV4L, was then constructed. Vector zalpha5CEEpZP9 (Example 5) was digested with EcoRI and Xbal . The reaction mixture was gel electrophoresed, and the band corresponding to zalpha5CEE was gel-purified as described in Example 5. The purified DNA was ligated into pZBV4L (a modified pFastBac™ expression vector (Life Technologies) containing the late activating Basic Protein promoter) that had been digested with with EcoRI and Xbal . A 77

portion of the ligated DNA was electroporated into E. coli host cells (Electromax DH10B™) . Positive ligations were identified by colony PCR using primers ZC7350 (SEQ ID N0:16) and ZC16,084 (SEQ ID N0:17). PCR conditions were: one cycle at 94° for 1'30"; thirty cycles at 94° for 10", 52° for 20", 72° for 1'30"; one cycle at 72° for 1'30"; followed by a 4 ° hold. Plasmid from a positive colony was purified using an anion-exchange kit as disclosed above. One microliter of zalpha5CEEpZBV4L was used to independently transform 20 μl E. coli Max

Efficiency DHlOBac™ competent cells according to the supplier's instructions by heat shock at 42 °C for 45 seconds. The transformants were then diluted in 980 μl SOC media and plated onto Luria Agar plates as described above. Bacmid DNA was isolated from positive colonies and screened for the correct insert using the PCR method as described above. Those having the correct insert were used to transfect Spodoptera frugiperda (Sf9) cells.

An expression construct designated zalpha5NEEpZBV4L was constructed from zalpha5NEEpZP9 and pZBV4L essentially as disclosed above for zalpha5CEEpZBV4L. One microliter of zalpha5NEEpZBV4L was used to independently transform 20 μl of E. coli Max

Efficiency DHlOBac™ competent cells according to supplier's instructions by heat shock at 42 °C for 45 seconds. The transformants were then diluted in 980 μl SOC media and plated on to Luria Agar plates as described above. Bacmid DNA was isolated from positive colonies and screened for the correct insert using the PCR method as described above. Those having the correct insert were used to transfect Spodoptera frugiperda (Sf9) cells.

Sf9 cells were seeded at 5 x 10⁶ cells per 35- mm plate and allowed to attach for 1 hour at 27°C. Five microliters of bacmid DNA was diluted with 100 μl serum- 78

free media (Sf-900 II SFM; Life Technologies) . Six μl of a 1:1.5 (M/M) formulation of N, N¹, N¹¹, N^ι:cι-tetramethyl-N, N¹, N¹¹, N^Iι:ι:-tetrapalmitylspermine and dioleoyl phosphatidylethanolamine in membrane-filtered water (CellFECTIN™ reagent; Life Technologies) was diluted with 100 μl Sf-900 II SFM. The bacmid DNA and lipid solutions were gently mixed and incubated for 30-45 minutes at room temperature. The media from one plate of cells was aspirated, and the cells were washed once with 2 ml fresh media. Eight hundred microliters of Sf-900 II SFM was added to the lipid-DNA mixture. The wash media was aspirated, and the DNA-lipid mixture was added to the cells. The cells were incubated at 27°C for 4-5 hours. The DNA-lipid mixture was aspirated, and 2 ml of Sf-900 II media was added to each plate. The plates were incubated at 27°C, 90% humidity for 96 hours, after which the virus was harvested.

Sf9 cells were grown in 50 ml Sf-900 II SFM in a 125-ml shake flask to an approximate density of 0.41- 0.52 x 10⁵ cells/ml. They were then infected with 100 μl of virus stock (disclosed above) and incubated at 27°C for 2-3 days, after which the virus was harvested. The titer for palphaδ was 2.9xl0⁷ pfu/ml, for Zalpha5CEEpZBV4L it was 4xl0⁶ pfu/ml, and for Zalpha5NEEpZBV4L it was 1.85xl0⁷ pfu/ml.

Example 7

To express zalpha5 in transgenic animals, zalpha5 cDNA is inserted into the expression vector pHB12-8 (see Fig. 2) . Vector pHB12-8 was derived from P2999B4 (Palmiter et al . , Mol . Cell Biol . 13 :5266-5275,

1993) by insertion of a rat insulin II intron (ca. 200 bp) and polylinker (Fse I/Pme I/Asc I) into the Nru I site. The vector comprises a mouse metallothionein (MT- 1) promoter (ca. 750 bp) and human growth hormone (hGH) untranslated region and polyadenylation signal (ca. 650 79

bp) flanked by 10 kb of MT-1 5' flanking sequence and 7 kb of MT-1 3' flanking sequence. The cDNA is inserted between the insulin II and hGH sequences.

Example 8

Zalpha5 protein produced in a baculovirus or BHK cell system is purified by a combination of cation and anion exchange chromatography followed by size exclusion chromatography. Conditioned cell culture medium is diluted to ~5mS with 20 mM MES, phosphate, or acetate/citrate buffer, pH 5.5-6.5. The diluted medium is loaded onto a cation exchange column (Poros^® 20HS or 50HS (PerSeptive Biosystems, Framingham, MA) or SP Sepharose^® Fast Flow (Pharmacia Biotech, Piscataway, NJ) ) at a ratio of 100-200 volumes medium to resin. The column is washed with 40-50 column volumes of the same buffer used to dilute the medium. Protein is eluted with a 0-1M NaCl gradient over 40-50 column volumes. The resulting eluate is diluted to ~5 mS with 20 mM Tris-HCI or HEPES buffer, pH 7.4-8.5. The diluted eluate is loaded onta a Poros^® 20HQ (PerSeptive Biosystems) column at a ratio of ~50 volumes sample to resin. The column is washed with 40-50 column volumes of the same buffer used to dilute the sample. Protein is eluted with a 0-1M NaCl gradient over 40-50 column volumes. The eluate from the HQ column is concentrated using a centrifugal filter device (Ultrafree™-15 ; Millipore, Bedford, MA) and loaded onto a Sephacryl^® S-200 or S-300 (Pharmacia Biotech) size exclusion column run in PBS.

Example 9

N-terminal glu-glu tagged zalphaδ is produced in BHK cells according to conventional procedures. The tagged protein is purified by affinity chromatography using an antibody to the glu-glu tag. Unless otherwise noted, all operations are carried out at 4°C. 80

BHK cell-conditioned medium is sequentially sterile filtered through a 4 inch, 0.2 mM capsule filter (Opticap™; Millipore, Bedford, MA) and a 0.2 mM capsule filter (Supercap™ 50; Gelman, Ann Arbor, MI). The material is then concentrated about 20x using a tangential flow concentrator (ProFlux® A30; Millipore) fitted with a 3000 kDa cutoff ultrafiltration membrane (S10Y3 membrane; Amicon, Bedford, MA) . The concentrated material is again sterile-filtered with the Gelman filter as described above. A mixture of protease inhibitors is added to the concentrated conditioned media to final concentrations of 2.5 mM ethylenediaminetetraacetic acid (EDTA, Sigma Chemical Co. St. Louis, MO), 0.001 mM leupeptin (Boehringer-Mannheim, Indianapolis, IN), 0.001 mM pepstatin (Boehringer-Mannheim) and 0.4 mM 4-(2- Aminoethyl) -benzenesulfonyl fluoride hydrochloride (Pefabloc®; Boehringer-Mannheim) .

A 100 ml bed volume of immobilized protein G (protein G-Sepharose®; Pharmacia Biotech) is washed 3 times with 100 ml of PBS containing 0.02% sodium azide using a 500 ml Nalgene 0.45 micron filter unit. The gel is washed with 6.0 volumes of 200 mM triethanolamine, pH 8.2 (TEA; Sigma Chemical Co., St. Louis, MO), and an equal volume of anti-glu-glu antibody solution containing 900 mg of antibody is added. After an overnight incubation at 4°C, unbound antibody is removed by washing the resin with 5 volumes of 200 mM TEA as described above. The resin is resuspended in 2 volumes of TEA, transferred to a suitable container, and dimethylpimilimidate-2HCl (Pierce, Rockford, IL) , dissolved in TEA, is added to a final concentration of 36 mg/ml of gel . The gel is rocked at room temperature for 45 minutes, and the liquid is removed using the filter unit as described above. Nonspecific sites on the gel are then blocked by incubating for 10 minutes at room temperature with 5 volumes of 20 mM ethanolamine in 200 mM TEA. The gel is washed with 5 volumes of PBS containing 0.02% sodium azide and stored in this solution at 4°C.

A 25.0 ml sample of anti-GluGlu Sepharose is added to the concentrated culture medium for batch adsorption, and the mixture is gently agitated on a roller culture apparatus (Wheaton Science Products, Millville, NJ) for 18.0 hours at 4°C. The mixture is poured into a 5.0 x 20.0 cm column (Econo-Column® ; Bio- Rad Laboratories, Hercules, CA) , and the gel is washed with 10 column volumes (cv) of phosphate buffered saline

(PBS) at 10-20 cv/hr. The unretained flow-through fraction is discarded. Once the absorbance of the effluent at 280 nM is less than 0.05, flow through the column is reduced to zero, and the anti-GluGlu Sepharose gel is washed batchwise with 2.0 cv of PBS containing 0.4 mg/ml of GluGlu peptide having the sequence GluTyrMetProValAsp (SEQ ID NO: 18) (obtained from AnaSpec, San Jose, CA) . After 1.0 hour at 4°C, flow is resumed, and the eluted protein is collected (peptide elution fraction) . The anti-GluGlu Sepharose gel is then washed with 2.0 column volumes of 0. IM glycine, pH 2.5, and the glycine wash is collected separately. The pH of the glycine-eluted fraction is adjusted to 7.0 by the addition of a small volume of 10X PBS and stored at 4°C for future analysis if needed.

The peptide elution fraction is concentrated to 5.0 ml using a 15,000 molecular weight cutoff membrane concentrator (Millipore, Bedford, MA) according to the manufacturer's instructions. The concentrated peptide elution is separated from free peptide by chromatography on a 1.5 x 50 cm Sephadex^® G-50 (Pharmacia Biotech) column equilibrated in PBS at a flow rate of 1.0 ml/min using a commercially available HPLC system (BioCad™/Sprint™ HPLC system; PerSeptive BioSystems, Framingham, MA) . Two-ml fractions are collected, and the 82

absorbance at 280 nM is monitored. The first peak of material absorbing at 280 nM and eluting near the void volume of the column, which contains the purified zalphaδ protein, is collected. The pure material is concentrated as described above, analyzed by SDS-PAGE and Western blotting with anti-GluGlu antibodies, and samples are taken for amino acid analysis and N-terminal sequencing. The remainder of the sample is aliquoted and stored at - 80°C.

Example 10

The molecular weight predicted for human zalphaδ was 52 kDa. The observed molecular weight of recombinant zalpha5 with an N-terminal Glu-Glu tag made in BHK cells was « 68kd.

Zalpha5 produced by a baculovirus expression system was found to have an amino terminus at residue 17

(Ser) of SEQ ID NO:2. The baculovirus-produced protein included cleavage products of 36.5 kDa and 21.5 kDa, as well as full-length protein of approximately 60 kDa.

A second preparation of zalphaδ protein from baculovirus was electrophoresed on a reducing 4-20% tris- glycine-SDS polyacrylamide gel, transferred to a PVDF membrane, and stained with coomassie blue. The protein ran as a doublet of M_r∞36 kDa. Sequencing indicated the presence of multiple species, the predominant species having an amino terminus at Thr225 and lesser species with amino termini at Leu231, Asn232, and Glu233.

Zalpha5 produced in BHK cells and by baculovirus expression ran as a multimer on size exclusion chromatography.

BHK-produced, N-terminal glu-glu tagged zalpha5 was analyzed by sequencing, peptide mapping, carbohydrate analysis, and disulfide bond determination. For peptide mapping, the protein was digested with several different enzymes to generate a peptide map. 83

100 μg of protein was dissolved in reduction buffer (6 M guanidinium chloride, 130 mM Tris-Cl, 0.25 mM EDTA, pH 7.6) to a concentration of about 1 mg/ml. Dithiothreitol (DTT) was added in a molar ration of 40:1 over cysteines. The mixture was incubated at 37°C for one hour. A 5-fold molar excess of 4 -vinylpyridine over DTT was added, and the mixture was incubated at room temperature for 2 hours, then desalted on a C-18 column (obtained from Vydac/The Separations Group, Inc., Hesperia, CA) and divided into two tubes. One sample was digested with PNGase F (Oxford GlycoSystems, Rosedale, NY) to remove N- linked glycosylation by dissolving the protein in 100 mM sodium phosphate pH 7.5 and incubating overnight at 37°C in PNGase F at lmU/5 μg protein. Both the digested and undigested samples were then divided in two. One set of samples was digested with trypsin, the other with Endoproteinase Glu-C. The peptide mixtures were then analyzed by liquid chromatography/mass spectrometry using a C-18 column, a gradient of 1-81% B (90% acetonitrile, 0.1% trifluoroacetic acid), and an ion trap mass spectrometer. Peptides corresponding to most of the predicted amino acid sequence were identified. The C- terminal peptide was easily identified and showed no heterogeneity. The first N-linked glycosylation site (Asn-Ser-Ser, residues 23-25 of SEQ ID NO: 2) appeared to be unoccupied since the unmodified peptide was clearly present. Two sites of probable O-linked glycosylation were identified in peptides 190-205 of SEQ ID NO: 2

(oligosaccharide attached at either Thrl90 or Serl91) and peptide 161-169 of SEQ ID NO : 2.

Carbohydrate analysis was performed essentially as disclosed by Hardy ("High-pH Anion-Exchange Chromatography of Glycoprotein-Derived Carbohydrates", in Lennarz and Hart , eds . , Guide to Techniques in Glycobiology, Academic Press, 1994) . Three lots of BHK- produced protein (A167F, A215F, and A221F) and one lot of 84

baculovirus-produced protein (A241F) were analyzed. Results are shown below in Table 5.

Table 5 mols/mcιl protein

Monosaccharide A167F A215F A221F A241F N-acetylneuraminic acid 4.7 12.5 8.5 -- Fucose 0 1.5 1.6 1.9

N-acetylgalactosamine 2.6 4.7 5.9 3.6 N-acetylglucosamine 9.6 17.4 17.4 7.2 Galactose 3.6 6.5 4.9 1.9 Mannose 5.2 8.7 6.5 14.1

These results indicate that , in lot A167F, there are approximatel .

yy two complex- -type N- -linked oligosaccharides (biantennary) per molecule, and that, since N-acetylgalacosamie is present, there are probably from one to three O-linked oligosaccharides per molecule. Lots A215F and A221F have more glycosylation than A167F, with about one more N-linked site occupied and 3-5 O- linked sites occupied. Lot A241F probably has a high- mannose N-linked oligosaccharide . For disulfide bond determination, a trypsin digest was done on non-reduced zalpha5. 1 M guanidinium- Cl and 5% acetonitrile were added to the digestion buffer. Peptides from the diest were separated on a C-4 column (Vydac/The Separations Group, Inc.) using a gradient of 5-70% B (ACN, 0.1% TFA) over 120 minutes. Fractions were collected by hand. Each fraction was spotted on a target with alpha-cyano-4 -hyroxycinnamic acid as matrix and analyzed by matrix assisted laser desorption ionization-time-of-flight mass spectrometry . Data were examined for peptides with molecular weights suggesting a predicted peptide with a modification, or peptides with molecular weights that could be readily assigned to the sequence of zalphaδ . N-terminal sequencing (Edman degradation) was done on those 85

fractions . Combining molecular weight data with the sequencing analysis indicated that one fraction contained a single tryptic peptide with two cysteines (residues 390-421 of SEQ ID NO:2) and that a second fraction contained two tryptic peptides each containing one cysteine (residues 239-252 and 253-298 of SEQ ID NO: 2) .

The first peptide was digested with V-8 protease in 50 mM sodium phosphate, pH 7.8. This second digest mixture was spotted on MALDI targets (both non- reduced and reduced) using both alpha-cyano-4- hydroxycinnamic acid and sinapinic acid as matrices. The data indicated the presence of a disulfide bond between the two cysteines (residues 394 and 408 of SEQ ID NO : 2 ) . Material from the second fraction was spotted on MALDI targets (both non-reduced and reduced) using both alpha-cyano-4-hydroxycinnamic acid and sinapinic acid as matrices . The data clearly showed a high molecular weight species (5667 D, corresponding to the two tryptic peptides linked by a disulfide bond between residues 246 and 274 of SEQ ID NO: 2) in the non-reduced fraction, which disappeared upon reduction. Ions at 1589.4D and 4074.4D (corresponding to the individual peptides) appeared when the material was reduced.

86

Example 11

Anti-zalpha5 polyclonal antibodies were raised using the peptide immunogens shown below in Table 6. Peptides were prepared by conventional synthetic techniques. Sequence numbers in Table 6 refer to SEQ ID NO : 2. "Cys" indicates and additional C- or N-terminal cysteine residue.

Table 6 Name Sequence huzalpha5-l 19-42-Cys huzalpha5-2 88-108-Cys huzalpha5-3 Cys-404-430

Peptides huzalpha5-l and huzalpha5-2 were conjugated to maleimide-activated keyhole limpet hemocyanin through the terminal cysteine.

Antibodies were prepared in New Zealand white rabbits. Animals were given initial immunizations of 200 μg/animal with Freund ' s complete adjuvant, and three boosts, at 3 -week intervals, of 100 μg/animal with Freund ' s incomplete adjuvant. Animals were first bled after the second boost, then at 3 -week intervals thereafter. Antisera were affinity-purified on a CNBr- Sepharose-peptide column prepared with 10 mg peptide per gram of Sepharose. Purified antibodies were dialyzed overnight against PBS.

Example 12

Immunocytochemical screening analyses were performed on normal and diseased human liver, monkey liver, normal human kidney, and normal human ileum using a polyclonal rabbit antibody to full-length zalpha5 protein as the primary antibody at dilutions of 1:5,000 and 1:10,000. The detection system employed a commercially available biotinylated second antibody and staining kit (VECTASTAIN® ABC-AP kit; Vector 87

Laboratories, Inc., Burlingame, CA) with an alkaline phosphatase substrate (VECTOR® Red substrate kit; Vector Laboratories) which was used to produce a fuchsia-colored deposit. Negative controls performed on each tissue sample included staining with diluent in the absence of primary antibody. Positive control tissues consisted of cell pellets and monkey liver tissue. Both the positive control cell pellet and the monkey liver stained positive with antibody, and the signal was competed out with addition of zalpha5 peptide.

In normal tissues, zalpha5 was strongly expressed in the space of Disse and within Kupffer cells of the liver, the convoluted tubules and collecting ducts of the kidney (staining concentrated along the brush border of the proximal tubule and along the luminal border or the distal tubules and collecting ducts) , and within nerves of the ileum and liver. Vascular smooth muscle was negative. The location of the signal in the liver was along the interface between the hepatocyte, sinusoidal lining cell, and Kupffer cell. Light microscopy did not reveal whether the sinusoidal lining cell was producing the protein independently of the signal seen in the Kupffer cell and along the hepatocyte cell membrane. Occasional staining was observed along the bile canaliculi between hepatocytes, suggesting that the protein is either membrane-bound by hepatocytes or produced by normal hepatocytes .

In carcinomas, strong expression was seen in both hepatomas and adenocarcinomas . The level of positive staining was variable within the carcinomas and was seen along both the sinusoidal and biliary borders of the cell and within acinar spaces.

In diseases of the liver such as cirrhosis, the positive signal was seen along the opposite border of the hepatocyte as compared to normal liver and was concentrated toward the bile canaliculi rather than the 88

sinusoidal spaces. In tissue in which necrosis was prominent, signals were increased in the cetrilobular region and in areas of necrosis.

Example 13

To identify cells that bound zalpha5, a zalpha5- saporin conjugate was administered to cells, tissue slices, and animals. Saporin was conjugated to zalpha5 using conventional methods. The prostate epithelial cell line pz-HPV7 specifically bound ¹²⁵I-zalpha5 and was killed by saporin- zalpha5 with an LD50 of 10 nM, while saporin alone was at least 100-fold less toxic. Cell killing was abrogated by the addition of unconjugated zalphaδ . Similar results were obtained with prostate tissue slices.

In contrast to results using cells shown to bind zalpha5, saporin-zalphaδ showed no toxicity toward DU-145, a prostate cell line that does not bind 125j_ zalpha5. Saporin and saporin-zalphaδ were injected at 4,

40, and 400 μg/kg into the tail vein of mice at day 1, day 3, and day 8 for saporin and saporin-zAlpha5. On day 9 all animals were sacrificed, and samples were prepared for physioscreen.

Example 14

Chemoattractant activity of zalpha5 was assayed using a matrigel assay. The matrigel invasion assay assesses the ability of a cell to move through a basement membrane matrix coated on a PET membrane containing uniform 8 -micron pores. The layer of matrigel, derived from the Engelbreth-Holm-Swarm tumor as a soluble membrane extract, serves as a basement membrane for in vi tro studies. It occludes the pores of the membrane, blocking non-invasive cells from migrating through the membrane. A cell must move through the matrigel, through 89

the pores, and then re-attach to the uncoated bottom portion of the membrane to be scored as positive for migration. Cells may move spontaneously or to an attractant placed in the bottom well below the insert. The matrigel invasion assay is obtained from

Becton Dickinson, Sparks, MD . Primary cell lines, Human Umbilical Vein Endothelial Cells (Huvec) and Human Uterine Myometrial Microvascular Endothelial Cells (Mvec) (both obtained from Clonetics, San Diego, CA) are maintained and passaged according to instructions provided by the supplier. VEGF was purchased from RD Systems, Minneapolis, MN. The invasion chambers are brought to room temperature in a laminar air- flow hood. Sterile technique is used throughout the assay. The Matrigel is rehydrated with 250 μl of warm media for 2 hours. Assay media for Huvec cells is EBM containing 1.0% fetal bovine serum and GA1000; Mvec media is EGM-2 containing 0.5 % fetal bovine serum, GA1000, and ascorbic acid (obtained from Clonetics) . Cells are removed from tissue culture flasks using trypsinization reagents

(ReagentPack™ reagents; Clonetics) and resuspended in the appropriate media to a concentration of 100,000 cells per ml . Positive control protein and zalpha5 are diluted in assay media such that when lOμl is added to the lower chamber of the kit, the VEGF concentration is 100 ng/ml and the zalphaδ concentration varies. Media (750 μl) is added to the lower chamber of the assay kit followed by control or test protein. The rehydration media placed into the top of the insert is removed, and 0.5 ml of cell suspension is added. The insert is carefully placed into the well such that no air bubbles are trapped beneath the insert. The plate is covered and placed into an incubator at 37° C, 5% C0₂ for 48 hours.

Prior to staining the migrating cells, the media in the top of the insert is removed, and the top surface of the membrane is swabbed twice using a cotton 90

applicator moistened with warm media to remove cells that did not migrate. A commercially available staining kit (Diff-Quick, obtained from VWR Scientific Products, Salt Lake City, UT) is used to fix and stain the cells. The inserts are placed in fixative for 2 minutes followed by stain 1 and then stain 2, 2 minutes each, followed by thorough rinsing in deionized water. The inserts are numbered and allowed to air dry. The membranes are removed using a #15 scalpel blade. The membrane is cut next to the plastic edge and the insert is rotated around the blade leaving only a small piece of the membrane attached and it is removed using forceps. A drop of immersion oil is placed on a glass slide followed by the membrane placed bottom side down. A drop of oil is placed on top followed by a glass coverslip. The number of migrating cells is quantified using an image analysis program (Media Cybernetics, Optimus version 6.2).

Two primary cell lines, Huvec and Mvec, were found to migrate in response to zalpha5. In a representative, blinded experiment, the results shown in Table 7 were obtained using Huvec cells:

Table 7 Condition (n=3) Number of migrating cells Control 363

VEGF (lOOng/ml) 1265

Zalpha5 (lOOng/ml) 644 Zalpha5 (lug/ml) 859

Some of the cells that migrated in response to zalph5 were observed to have a more spread and stellate morphology as compared to the controls.

From the foregoing, it will be appreciated that, although specific embodiments of the invention have been described herein for purposes of illustration, various modifications may be made without deviating from 91

the spirit and scope of the invention. Accordingly, the invention is not limited except as by the appended claims .

Claims

92CLAIMSWhat is claimed is:

1. An isolated polypeptide of at least 15 amino acid residues comprising an epitope-bearing portion of a polypeptide of SEQ ID NO: 2 or SEQ ID NO: 15.

2. The isolated polypeptide of claim 1 comprising at least 15 contiguous residues of SEQ ID NO : 2.

3. The isolated polypeptide of claim 2 wherein said residues are selected from residues 19-42, 88-116, and 404-430.

4. An isolated polypeptide comprising a sequence of amino acids of the formula B_x-L_y-C_z, wherein:

B is at least 70% identical to residues m to n of SEQ ID NO: 2, wherein m is from 15 to 23 and n is from 199-207;

L is at least 70% identical to residues o to p of SEQ ID NO: 2, wherein o is n+1 and p is from 234 to 242;

C is at least 70% identical to residues q to r of SEQ ID NO: 2, wherein q is from p+1 to 246 and r is from 408 to 460; and each of x, y, and z is individually 0 or 1, subject to the limitations that at least one of x and z is 1 and if both x and z are 1, y is 1.

5. The isolated polypeptide of claim 4, wherein B is at least 80% identical to residues m to n of SEQ ID NO: 2, L is at least 80% identical to residues o to p of SEQ ID NO : 2 , and C is at least 80% identical to residues q to r of SEQ ID NO: 2.

6. The isolated polypeptide of claim 4, wherein x is 1 and B comprises residues 23-199 of SEQ ID NO : 2. 93

7. The isolated polypeptide of claim 4, wherein z is 1 and C comprises residues 246-408 of SEQ ID NO : 2.

8. The isolated polypeptide of claim 4, wherein said polypeptide comprises residues X1-X2 of SEQ ID NO : 2 , wherein

XI is selected from the group consisting of 15, 16, 17, 18, 19, 20, 21, 22, 23, 200, 201, 202, 203, 204, 205, 206, 207, 208, 225, 231, 232, 233, 235, 236, 237, 238, 239, 240, 241, 242, 243, and 246, and X2 is from 408 to 460; or

XI is selected from the group consisting of 15, 16, 17, 18, 19, 20, 21, 22, and 23, and X2 is selected from the group consisting of 199, 200, 201, 202, 203, 204, 205, 206, 207, 224, 230, 231, 232, 234, 235, 236, 237, 238, 239, 240, 241, 242, and 408 through 460.

9. The isolated polypeptide of claim 4, wherein said polypeptide comprises a sequence of amino acid residues selected from the group consisting of: residues 17-460 of SEQ ID NO : 2 _; residues 17-238 of SEQ ID NO:2 residues 17-203 of SEQ ID NO : 2 , residues 225-408 of SEQ ID NO : 2 residues 225-460 of SEQ ID NO : 2 , residues 231-408 of SEQ ID NO: 2; and residues 231-460 of SEQ ID NO : 2

10. The isolated polypeptide of claim 4, wherein said polypeptide is glycosylated.

11. The isolated polypeptide of claim 4, wherein said polypeptide comprises cysteine residues at positions corresponding to residues 246, 274, 394, and 408 of SEQ ID N0:2. 94

12. The isolated polypeptide of claim 4, covalently linked to a moiety selected from the group consisting of affinity tags, toxins, radionuclides, enzymes, and fluorophores .

13. The isolated polypeptide of claim 12 wherein said moiety is an affinity tag selected from the group consisting of polyhistidine, protein A, glutathione S transferase, substance P, and an immunoglobulin heavy chain constant region.

14. The isolated polypeptide of claim 13 further comprising a proteolytic cleavage site between said sequence of amino acid residues and said affinity tag.

15. An isolated multimeric protein comprising a first polypeptide according to claim 4 non-covalently associated with a second polypeptide, wherein said protein modulates cell proliferation, differentiation, migration, adhesion, or metabolism.

16. The multimeric protein of claim 15 wherein said first polypeptide comprises a sequence of amino acid residues selected from the group consisting of: residues 17-460 of SEQ ID NO : 2 residues 17-238 of SEQ ID NO:2, residues 17-203 of SEQ ID NO : 2 residues 225-408 of SEQ ID NO : 2 ; residues 225-460 of SEQ ID NO : 2 ; residues 231-408 of SEQ ID NO: 2; and residues 231-460 of SEQ ID NO : 2.

17. The multimeric protein of claim 16 wherein said second polypeptide comprises a sequence of amino acid residues selected from the group consisting of: residues 17-460 of SEQ ID NO : 2 ; 95

residues 17-238 of SEQ ID NO: 2; residues 17-203 of SEQ ID NO : 2 ; residues 225-408 of SEQ ID NO : 2 ; residues 225-460 of SEQ ID NO : 2 ; residues 231-408 of SEQ ID NO : 2 ; and residues 231-460 of SEQ ID NO : 2.

18. The multimeric protein of claim 15, wherein said first and second polypeptides are the same.

19. A protein produced by a method comprising: culturing a cell containing an expression vector comprising the following operably linked elements: a transcription promoter; a DNA segment encoding the polypeptide of claim 4 ; and a transcription terminator; and isolating the protein encoded by said DNA segment and produced by said cell.

20. The protein of claim 19, wherein said DNA construct further comprises a secretory signal sequence operably linked to said DNA segment.

21. The protein of claim 19, wherein said polypeptide comprises residues X1-X2 of SEQ ID NO: 2, wherein

XI is selected from the group consisting of 15, 16, 17, 18, 19, 20, 21, 22, 23, 200, 201, 202, 203, 204, 205, 206, 207, 208, 225, 231, 232, 233, 235, 236, 237, 238, 239, 240, 241, 242, 243, and 246, and X2 is from 408 to 460; or

XI is selected from the group consisting of 15, 16, 17, 18, 19, 20, 21, 22, and 23, and X2 is selected from the group consisting of 199, 200, 201, 202, 203, 204, 205, 206, 207, 224, 230, 231, 232, 234, 235, 236, 237, 238, 239, 240, 241, 242, and 408 through 460. 96

22. The protein of claim 21, wherein said polypeptide comprises a sequence of amino acid residues selected from the group consisting of: residues 23-238 of SEQ ID NO : 2 ; and residues 17-238 of SEQ ID NO : 2.

23. The protein of claim 21, wherein said polypeptide comprises a sequence of amino acid residues selected from the group consisting of: residues 23-460 of SEQ ID NO : 2 ; and residues 17-460 of SEQ ID NO : 2.

24. An expression vector comprising the following operably linked elements: a transcription promoter; a DNA segment encoding the polypeptide of claim 4; and a transcription terminator.

25. The expression vector of claim 24 further comprising a secretory signal sequence operably linked to said DNA segment .

26. The expression vector of claim 24, wherein said polypeptide comprises residues X1-X2 of SEQ ID NO : 2 , wherein

XI is selected from the group consisting of 15, 16, 17, 18, 19, 20, 21, 22, 23, 200, 201, 202, 203, 204, 205, 206, 207, 208, 225, 231, 232, 233, 235, 236, 237, 238, 239, 240, 241, 242, 243, and 246, and X2 is from 408 to 460; or

XI is selected from the group consisting of 15, 16, 17, 18, 19, 20, 21, 22, and 23, and X2 is selected from the group consisting of 199, 200, 201, 202, 203, 204, 205, 206, 207, 224, 230, 231, 232, 234, 235, 236, 237, 238, 239, 240, 241, 242, and 408 through 460. 97

27. The espression vector of claim 26, wherein said polypeptide comprises a sequence of amino acid residues selected from the group consisting of: residues 23-238 of SEQ ID NO : 2 ; and residues 17-238 of SEQ ID NO : 2.

28. The expression vector of claim 26, wherein said polypeptide comprises a sequence of amino acid residues selected from the group consisting of: residues 23-460 of SEQ ID NO : 2 ; and residues 17-460 of SEQ ID NO : 2.

29. A cultured cell into which has been introduced the expression vector of claim 24, wherein said cell expresses the DNA segment and produces a polypeptide encoded by the DNA segment .

30. A method of making a protein comprising: culturing a cell containing a DNA construct comprising the following operably linked elements: a transcription promoter; a DNA segment encoding the polypeptide of claim 4 ; and a transcription terminator; and isolating the protein encoded by said DNA segment and produced by said cell.

31. An antibody that specifically binds to a polypeptide as shown in SEQ ID NO : 2 from residue 18 through residue 460 or to a fragment of said polypeptide.

32. An antibody that specifically binds to the multimeric protein of claim 15.