WO2001032871A2 - Follistatin-related protein zfsta4 - Google Patents

Abstract

The present invention relates to polynucleotide and polypeptide molecules for zfsta4, a novel member of the follistatin family. The polypeptides, and polynucleotides encoding them are useful for binding to members of the TGF-β family and mediating central nervous system, reproductive, hematopoietic and bone-related activities. The present invention also includes antibodies to the zfsta4 polypeptides.

Description

Description FOLLISTATIN-RELATED PROTEIN ZFSTA4

BACKGROUND OF THE INVENTION

Follistatin is a monomeric, glycosylated protein originally identified in porcine follicular fluid as a potent inhibitor of pituitary follicle-stimulating hormone (FSH) synthesis and secretion, follistatin was later shown to exert some of its biological effects by specifically binding the FSH-inducer activin. Other follistatin family members include follistatin related protein or FRP (Zwijsen et al., Eur. J. Biochem. 225:937-46, 1994), SPARC, also known as osteonectin or BM-40 or the human ortholog of mouse TSC-36 (Lane and Sage, ibid.), agrin (Patthy and Nikolics, TINS 16:76-81, 1993), hevin (Girard and Springer, Immunity 2:113-23, 1995), the Flik protein of chickens (Amthor et al., Dev. Biology 178:343-62, 1996) and the rat brain protein SCI (Mendis et al., Brain Res. 730:95-106, 1996). Follistatins, however, are thought to be more than "activin binders" since follistatin deficient mice prepared by gene targeting have a more complex and different phenotype than activin gene knock- out animals (Mazuk et al., Nature 374:360-3, 1995 and Mazuk et al., Nature 374:356-9. 1995).

Activins and inhibins are potent activators and inhibitors, respectively, of pituitary FSH secretion and are members of the TGF-β family of peptide growth factors (Mather et al., Proc. Soc. Exp. Biol. Med. 215:209-22, 1997). The activin and inhibin family of hormones, while originally described as gonadally produced regulators of pituitary FSH secretion, are now known to have a broad range of effects within and outside of the reproductive system (Mather et al., ibid.). Inhibins consist of a common alpha subunit which is covalently linked to one of two different beta subunits (inhibin A:ot/B_A; inhibin B:α/B_B); activins are covalently linked dimers of the two B-subunits and therefore exist in three different forms (activin A:B_A/B_A; activin

B:B_B/B_B; activin AB:B_A/B_B). Activin and inhibin bind to follistatin with high affinity, and although the structure of the activin binding site has not been completely defined, preliminary data (Inouye et al., Biochem. Biophys. Res. Commun. 179:352-8. 1991) suggest that residues in the first amino terminal cysteine-rich follistatin domain are involved in hormone binding. Activin binding to follistatin is thus thought to limit its biological effects by sequestration of the peptide hormone. Thus, the broad range of biological actions of the activins and inhibins, and possibly other members of the TGF- β family as well, may be regulated by binding to proteins of the follistatin family. Different binding proteins may be involved for each TGF-β family member as follistatin binds activin with high affinity (nM), inhibin with lower affinity, and does not appear to bind TGF-β at all (Mather et al., ibid.). Follistatin family members may regulate the activity of other growth factors as well, for example, SPARC or BM-40 have been shown to bind platelet derived growth factor (PDGF-AB, PDGF-BB) (Lane and Sage, FASEB J. 8:163-73, 1994).

This application provides a new member of the follistatin family, zfsta4, which is likely to play a major role in regulating the biological activities of the TGF-β family of growth factors. Like other members of the follistatin family, zfsta4 may play a broad role in development and differentiation, pathogenesis of atherosclerosis, regulation of the gonadal-pituitary-hypothalamic axis, tooth and bone formation, regulation of gonadal hormone production, spermatogenesis, hypothalmic oxytocin secretion, proliferation and differentiation of erythroid progenitors, hematopoiesis, host defense and neuron survival.

The present invention provides such polypeptides for these and other uses that should be apparent to those skilled in the art from the teachings herein.

SUMMARY OF THE INVENTION Within one aspect the invention provides an isolated polypeptide comprising a follistatin homology domain, wherein said follistatin homology domain comprises amino acid residues 65 to 133 of the amino acid sequence of SEQ ID NO:2. Within one embodiment the polypeptide further comprises an alpha helical linker region that resides in a carboxyl-terminal position relative to said follistatin homology domain, wherein said alpha helical linker region comprises amino acid residues 134 to 173 of the amino acid sequence of SEQ ID NO:2. Within another embodiment the polypeptide further comprises an alpha helical linker region and a calmodulin homology domain that resides in a carboxyl-terminal position relative to said follistatin homology domain, wherein said alpha helical linker region and calmodulin homology domain comprises amino acid residues 134 to 250 of the amino acid sequence of SEQ

ID NO:2. Within yet another embodiment the polypeptide further comprises an alpha helical linker region, a calmodulin homology domain, and two I-set Ig domains that reside in a carboxyl-terminal position relative to said follistatin homology domain, wherein said alpha helical linker region, calmodulin homology domain, and I-set Ig domains comprise amino acid residues 134 to 432 of the amino acid sequence of SEQ ID NO:2. In still another embodiment the polypeptide further comprises an alpha helical linker region, a calmodulin homology domain, two I-set Ig domains, and a carboxy-terminal domain that resides in a carboxyl-terminal position relative to said follistatin homology domain, wherein said alpha helical linker region, calmodulin homology domain, two I-set Ig domains, and carboxy-terminal domain comprises amino acid residues 134 to 842 of said amino acid sequence of SEQ ID NO:2. In a further embodiment the polypeptide further comprises a hydrophilic linker region that resides in an amino-terminal position relative to said follistatin homology domain, wherein said hydrophobic linker region comprises amino acid residues 23 to 64 of the amino acid sequence of SEQ ID NO:2. Within another embodiment the polypeptide further comprises a secretory signal sequence that resides in an amino-terminal position relative to said hydrophobic linker region, wherein said secretory signal sequence comprises amino acid residues 1 to 22 of the amino acid sequence of SEQ ID NO:2.

The invention also provides an isolated polypeptide having an amino acid sequence that is at least 80% identical to the amino acid sequence of SEQ ID NO:2, wherein said isolated polypeptide specifically binds with an antibody to which a polypeptide having the amino acid sequence of SEQ ID NO:2 specifically binds. Within one embodiment isolated polypeptide has an amino acid sequence that is at least 90% identical to the amino acid sequence of SEQ ID NO:2. Within another embodiment any difference between said amino acid sequence and said corresponding amino acid sequence of SEQ ID NO:2 is due to one or more conservative amino acid substitutions. Within a further embodiment the the amino acid percent identity is determined using a FAST A program with ktup=l, gap opening penalty=10, gap extension penalty=l, and substitution matrix=blosum62, with other parameters set as default.

The invention also provides an isolated polypeptide comprising amino acid residues 23-842 of SEQ ID NO:2, and an isolated polypeptide comprising the amino acid sequence of SEQ ID NO: 2.

The invention further provides an isolated polypeptide selected from the group consisting of: a) a polypeptide consisting of the sequence of amino acid residues from residue 23 to residue 64 of SEQ ID NO:2; b) a polypeptide consisting of the sequence of amino acid residues from residue 65 to residue 133 of SEQ ID NO:2; c) a polypeptide consisting of the sequence of amino acid residues from residue 134 to residue 173 of SEQ ID NO:2; d) a polypeptide consisting of the sequence of amino acid residues from residue 65 to residue 173 of SEQ ID NO:2; e) a polypeptide consisting of the sequence of amino acid residues from residue 65 to residue 250 of SEQ ID NO:2; f) a polypeptide consisting of the sequence of amino acid residues from residue 65 to residue 334 of SEQ ID NO:2; g) a polypeptide consisting of the sequence of amino acid residues from residue 65 to residue 432 of SEQ ID NO:2; h) a polypeptide consisting of the sequence of amino acid residues from residue 65 to residue 842 of SEQ ID NO:2; i) a polypeptide consisting of the sequence of amino acid residues from residue 134 to residue 250 of SEQ ID NO:2; j) a polypeptide consisting of the sequence of amino acid residues from residue 134 to residue 334 of SEQ ID NO:2; k) a polypeptide consisting of the sequence of amino acid residues from residue 134 to residue 432 of SEQ ID NO:2; and 1) a polypeptide consisting of the sequence of amino acid residues from residue 134 to residue 842 of SEQ ID NO:2.

Also provided is an isolated polypeptide as described above, further comprising an affinity tag or binding domain. The invention further provides a fusion protein comprising a secretory signal sequence having the amino acid sequence of amino acid residues 1-22 of SEQ ID NO:2, wherein said secretory signal sequence is operably linked to an additional polypeptide. Also provided is a fusion protein consisting essentially of a first portion and a second portion joined by a peptide bond, said first portion comprising a polypeptide as described above; and said second portion comprising another polypeptide.

Within another aspect the invention provides an isolated polynucleotide molecule that encodes a polypeptide as described above. Within one embodiment the polynucleotide molecule encodes a polypeptide further comprising an alpha helical linker region that resides in a carboxyl-terminal position relative to said follistatin homology domain, wherein said alpha helical linker region comprises amino acid residues 134 to 173 of the amino acid sequence of SEQ ID NO:2. Within another embodiment the polynucleotide encodes a polypeptide further comprising an alpha helical linker region and a calmodulin homology domain that resides in a carboxyl- terminal position relative to said follistatin homology domain, wherein said alpha helical linker region and calmodulin homology domain comprise amino acid residues 134 to 250 of the amino acid sequence of SEQ ID NO:2. Within yet another embodiment the polynucleotide encodes a polypeptide further comprising an alpha helical linker region, a calmodulin homology domain, and two I-set Ig domains that reside in a carboxyl-terminal position relative to said follistatin homology domain, wherein said alpha helical linker region, calmodulin homology domain, and I-set Ig domains comprise amino acid residues 134 to 432 of the amino acid sequence of SEQ ID NO:2. Within yet another embodiment the polynucleotide encodes a polypeptide further comprising an alpha helical linker region, a calmodulin homology domain, two I-set Ig domains, and a carboxy-terminal domain that resides in a carboxyl-terminal position relative to said follistatin homology domain, wherein said alpha helical linker region, calmodulin homology domain, two I-set Ig domains, and carboxy-terminal domain comprise amino acid residues 134 to 842 of said amino acid sequence of SEQ ID NO:2. Within a related embodiment the polypeptide further comprises an affinity tag or binding domain.

The invention also provides an isolated polynucleotide molecule, wherein said polynucleotide molecule is a degenerate nucleotide sequence encoding a polypeptide as described above.

The invention further provides an isolated polynucleotide encoding a polypeptide having an amino acid sequence that is at least 80% identical to the amino acid sequence of SEQ ID NO:2, wherein said isolated polypeptide specifically binds with an antibody to which a polypeptide having the amino acid sequence of SEQ ID NO:2 specifically binds. Within another embodiment the isolated polypeptide has an amino acid sequence that is at least 90% identical to the amino acid sequence of SEQ ID NO:2. Within a related embodiment any difference between said amino acid sequence and said corresponding amino acid sequence of SEQ ID NO:2 is due to one or more conservative amino acid substitutions. Within yet another embodiment the amino acid percent identity is determined using a FAST A program with ktup=l, gap opening penal ty= 10, gap extension penalty=l, and substitution matrix=blosum62, with other parameters set as default.

The invention also provides an isolated polynucleotide molecule comprising the nucleotides 183-2632 of SEQ ID NO:l, and an isolated polynucleotide molecule comprising the nucleotide sequence of nucleotides 107 to 2632 of SEQ ID NO: l.

Also provided by the invention is an isolated polynucleotide molecule of SEQ ID NO:l. The invention further provides an isolated polynucleotide selected from the group consisting of: a) a polynucleotide consisting of nucleotides 107-172 of SEQ

ID NO:l; b) a polynucleotide consisting of nucleotides 173-297 of SEQ ID NO:l; c) a polynucleotide consisting of nucleotides 298-505 of SEQ ID NO:l; d) a polynucleotide consisting of nucleotides 506-625 of SEQ ID NO:l; e) a polynucleotide consisting of nucleotides 298-625 of SEQ ID NO:l; f) a polynucleotide consisting of nucleotides 298-856 of SEQ ID NO:l; g) a polynucleotide consisting of nucleotides 298-1108 of SEQ ID NO:l; h) a polynucleotide consisting of nucleotides 298-1402 of SEQ ID NO:l; i) a polynucleotide consisting of nucleotides 298-2632 of SEQ ID NO:l; j) a polynucleotide consisting of nucleotides 506-856 of SEQ ID NO:l; k) a polynucleotide consisting of nucleotides 506-1108 of SEQ ID NO: 1 ; 1) a polynucleotide consisting of nucleotides 506-1402 of SEQ ID NO:l; and m) a polynucleotide consisting of nucleotides 506-2632 of SEQ ID NO:l. The invention also provides a polynucleotide encoding a fusion protein comprising a secretory signal sequence having the amino acid sequence of amino acid residues 1-22 of SEQ ID NO:2, wherein said secretory signal sequence is operably linked to an additional polypeptide. Also provided is a polynucleotide molecule encoding a fusion protein consisting essentially of a first portion and a second portion joined by a peptide bond, said first portion comprising a polypeptide as described above; and said second portion comprising another polypeptide.

Within another aspect the invention provides an expression vector comprising the following operably linked elements: a transcription promoter; a polynucleotide molecule that encodes a polypeptide according to claim 1; and a transcription terminator. Within one embodiment the expression further comprises a secretory signal sequence operably linked to said polypeptide. Within another embodiment the polynucleotide encodes a polypeptide covalently linked amino terminally or carboxy terminally to an affinity tag. Also provided is a cultured cell into which has been introduced an expression vector comprising the following operably linked elements: a transcription promoter; a polynucleotide molecule that encodes a polypeptide as described above; and a transcription terminator, wherein said cultured cell expresses said polypeptide encoded by said polynucleotide segment. The invention further provides a method of producing a polypeptide comprising: culturing a cell into which has been introduced an expression vector comprising the following operably linked elements: a transcription promoter; a polynucleotide molecule that encodes a polypeptide as described above; and a transcription terminator; whereby said cell expresses said polypeptide encoded by said polynucleotide segment; and recovering said expressed polypeptide.

Within another aspect the invention provides an antibody or antibody fragment that specifically binds to a polypeptide as described above. Within one embodiment the antibody is selected from the group consisting of: a) polyclonal antibody; b) murine monoclonal antibody; c) humanized antibody derived from b); and d) human monoclonal antibody. Within a related embodiment the antibody fragment is selected from the group consisting of F(ab'), F(ab), Fab', Fab, Fv, scFv, and minimal recognition unit. Within another embodiment is provided an anti-idiotype antibody that specifically binds to the antibody described above.

The invention also provides a polypeptide as described above, in combination with a pharmaceutically acceptable vehicle.

These and other aspects of the invention will become evident upon reference to the following detailed description of the invention. 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 or detection 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., Meth. Enzymol. 198:3. 1991), glutathione S transferase (Smith and Johnson, Gene 67:31. 1988), Glu-Glu affinity tag (Grussenmeyer et al, Proc. Natl. Acad. Sci. USA 82:7952- 4, 1985), substance P, Flag™ peptide (Hopp et al., Biotechnology 6:1204-10, 1988), streptavidin binding peptide, or other antigenic epitope or binding domain. See, in general, Ford et al., Protein Expression and Purification 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 polymoφhism 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.

The term "complement/anti-complement pair" denotes non-identical moieties that form a non-covalently associated, stable pair under appropriate conditions. For instance, biotin and avidin (or streptavidin) are prototypical members of a complement/anti-complement pair. Other exemplary complement/anti-complement pairs include receptor/ligand pairs, antibody/antigen (or hapten or epitope) pairs, sense/antisense polynucleotide pairs, and the like. Where subsequent dissociation of the complement/anti-complement pair is desirable, the complement/anti-complement pair preferably has a binding affinity of <10^ M^~l.

The term "complements 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 5' ATGCACGGG 3' is complementary to 5' CCCGTGCAT 3'.

The term "contig" denotes a polynucleotide that has a contiguous stretch of identical or complementary sequence to another polynucleotide. Contiguous sequences are said to "overlap" a given stretch of polynucleotide sequence either in their entirety or along a partial stretch of the polynucleotide. For example, representative contigs to the polynucleotide sequence 5'-ATGGCTTAGCTT-3' are 5'- TAGCTTgagtct-3' and 3'-gtcgacTACCGA-5'.

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).

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

The term "operably linked", when referring to DNA segments, indicates that the segments are arranged so that they function in concert for their intended puφoses, e.g., transcription initiates in the promoter and proceeds through the coding segment to the terminator. The term "ortholog" denotes a polypeptide or protein obtained from one species that is the functional counteφart of a polypeptide or protein from a different species. Sequence differences among orthologs are the result of speciation.

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 vitro, or prepared from a combination of natural and synthetic molecules. Sizes of polynucleotides are expressed as base pairs (abbreviated "bp"), nucleotides ("nt"), or kilobases ("kb"). Where the context allows, the latter two terms may describe polynucleotides that are single-stranded or double-stranded. When the term is applied to double-stranded molecules it is used to denote overall length and will be understood to be equivalent to the term "base pairs". It will be recognized by those skilled in the art that the two strands of a double-stranded polynucleotide may differ slightly in length and that the ends thereof may be staggered as a result of enzymatic cleavage; thus all nucleotides within a double-stranded polynucleotide molecule may not be paired. Such unpaired ends will in general not exceed 20 nt in length.

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

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.

The term "receptor" denotes a cell-associated protein that binds to a bioactive molecule (i.e., a ligand) and mediates the effect of the ligand on the cell. Membrane-bound receptors are characterized by a multi-peptide structure comprising an extracellular ligand-binding domain and an intracellular effector domain that is typically involved in signal transduction. Binding of ligand to receptor results in a conformational change in the receptor that causes an interaction between the effector domain and other molecule(s) in the cell. This interaction in turn leads to an alteration in the metabolism of the cell. Metabolic events that are linked to receptor-ligand interactions include gene transcription, phosphorylation, dephosphorylation, increases in cyclic AMP production, mobilization of cellular calcium, mobilization of membrane lipids, cell adhesion, hydrolysis of inositol lipids and hydrolysis of phospholipids. In general, receptors can be membrane bound, cytosolic or nuclear; monomeric (e.g., thyroid stimulating hormone receptor, beta-adrenergic receptor) or multimeric (e.g., PDGF receptor, growth hormone receptor, IL-3 receptor, GM-CSF receptor, G-CSF receptor, erythropoietin receptor and IL-6 receptor).

The term "secretory signal sequence" denotes 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.

The term "splice variant" is used herein to denote alternative forms of RNA transcribed from a gene. Splice variation arises naturally through use of alternative splicing sites within a transcribed RNA molecule, or less commonly between separately transcribed RNA molecules, and may result in several mRNAs transcribed from the same gene. Splice variants may encode polypeptides having altered amino acid sequence. The term splice variant is also used herein to denote a protein encoded by a splice variant of an mRNA transcribed from a gene. 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%.

A new member of the follistatin family of proteins, zfsta4, has been identified from a human urinary tract library. The zfsta4 protein exhibits the characteristic amino terminal cysteine-rich follistatin domain (Hohenester et al.,

EMBO J. 16:3778-86, 1997) found in other follistatin family members such as follistatin related protein or FRP (Zwijsen et al., ibid.), SPARC, also known as osteonectin or BM-40 or the human ortholog of mouse TSC-36 (Lane and Sage, ibid.), agrin (Patthy and Nikolics, ibid.), hevin (Girard and Springer, ibid.), the Flik protein of chickens (Amthor et al., ibid.), follistatin-related protein zfsta2 (Conklin et al., WO 00/22126, April 20, 2000), and the rat brain protein SCI (Mendis et al., ibid.).

The present invention is based in part upon the discovery of a novel DNA sequence that encodes a polypeptide having homology to the family of follistatins. The zfsta4 polynucleotide sequence is disclosed in SEQ ID NO:l and encodes a multi-domain secreted protein of 842 amino acids (SEQ ID NO:2). Sequence analysis of a deduced amino acid sequence of zfsta4, as represented by SEQ ID NO:2, indicates the presence of a 22 amino acid residue signal sequence (amino acid residues 1-22 of SEQ ID NO:2, nucleotides 107-172 of SEQ ID NO:l), followed by a predominantly hydrophilic short linker domain that has no known homology (amino acid residues 23-64 of SEQ ID NO:2, nucleotides 173-297 of SEQ ID NO:l), a follistatin homology domain (amino acid residues 65-133 of SEQ ID NO:2, nucleotides 298-505 of SEQ ID NO:l), an alpha-helical linker region (amino acid residues 134-173 of SEQ ID NO:2, nucleotides 506-625 of SEQ ID NO:l), a calmodulin domain (amino acid residue 177-250 of SEQ ID NO:2, nucleotides 635-856 of SEQ ID NO: l), an I-set IG domain #1 (amino acid residues 251-334 of SEQ ID NO:2, nucleotides 857-1108 of SEQ ID NO: 1), an I-set IG domain #2 (amino acid residues 335-432 of SEQ ID NO:2, nucleotides 1109-1402 of SEQ ID NO:l) and a C-terminal domain with no known homology (amino acid residues 433-842 of SEQ ID NO:2, nucleotides 1403-2632 of SEQ ID NO:l). Those skilled in the art will recognize that predicted domain boundaries are approximations based on primary sequence content, and may vary slightly; however, such estimates are generally accurate to within ±5 amino acid residues.

The follistatin homology domain is predicted to fold into a structure similar to that determined for the follistatin homology domain in SPARC (Swiss-Prot SPRC_HUMAN, PDB 1BMO, also known as BM-40 or osteonectin, Hohenester et al., 1997). This is a beta haiφin structure, followed by a small hydrophobic core of alpha/beta structure. Unlike SPARC, which is glycosylated at Asn99, there is no predicted glycosylation site in zfsta4. Based on the disulfide bonding pattern in SPARC, the disulfide pairings in zfsta4 are inferred as follows: Cys65-Cys76, Cys70- Cys87, Cys89-Cysl l9, Cys93-Cysl l2, and Cysl01-Cysl33, of SEQ ID NO:2. The zfsta4 follistatin homology domain has 49% identity to the follistatin domain in human follistatin related protein (Swiss-Prot FRP_HUMAN, Zwijsen et al., Eur. J. Biochem. 255:937-46, 1994; Tanaka et al., Int. Immunol. 10:1305-14, 1998); the mouse orthologue of this protein is known as TSC-36 (Swiss-Prot FRP_MOUSE, Shibanuma et al„ Eur. J. Biochem. 217:13-19, 1993).

The follistatin homology domain has substantial sequence similarity to the Kazal family (Bode and Huber., Eur. J. Biochem. 204, 433-51, 1992) of serine proteinase inhibitors. Serine proteinase inhibitors regulate the proteolytic activity of target proteinases by occupying the active site and thereby preventing occupation by normal substrates. Although serine proteinase inhibitors fall into several unrelated structural classes, they all possess an exposed loop (variously termed an "inhibitor loop", a "reactive core", a "reactive site", a "binding loop") which is stabilized by intermolecular interactions between residues flanking the binding loop and the protein core (Bode and Huber, ibid.).

Interaction between inhibitor and enzyme produces a stable complex which disassociates very slowly, producing either a virgin or a modified inhibitor which is cleaved at the scissile bond of the binding loop. Based on analogy with the crystal structures for the proteinase inhibitors PEC-60 (PDB 1PCE), and ovomucoid (PDB 1OVO), the putative proteinase binding site in the follistatin homology domain of zfsta4 comprises the amino acid residue 93 (Cys) (P3), residue 94 (Arg) (P2), residue 95 (Pro) (PI), residue 96 (Ser) (PL), and residue 97 (Tyr) (P2') of SEQ ID NO:2. The scissile bond of the binding loop will therefore reside between the PI and PI' at residue 95 (Pro) and 96 (Ser) of SEQ ID NO:2.

The calmodulin homology domain is predicted to fold into a structure similar to that determined for the EC (EF-hand calcium binding; calmodulin-like) domain in SPARC (Hohenester et al., EMBO J. 16:3778-86, 1997). Calmodulin (Swiss-Prot CALM_HUMAN, PDB 1CLI) is an all alpha-helical protein which binds calcium ions through the loops of helix-loop-helix substructures known as EF hands.

Calmodulin has two structurally similar regions, each containing two EF hands, linked by a connecting helical segment. As is used herein "calmodulin homology domain" is meant to describe one of these two regions. The calmodulin homology domain of zfsta4 is predicted to contain two EF hand motifs, and therefore two potential calcium ion binding sites. Based on motif analysis, the loops of these two EF hands are predicted to reside between amino acid residue 187 (Asp) and amino acid residue 199 (Leu) of SEQ ID NO:2, and between amino acid residue 226 (Asp) and amino acid residue 238 (Phe) of SEQ ID NO:2. The last residue of the EF hand loop is always hydrophobic: in zfsta4 these residues are amino acid residue 199 (Leu) and amino acid residue 238 (Phe) of SEQ ID NO:2. In terms of sequence homology, the calmodulin homology domain of zfsta4 has 32% identity at the amino acid level, to the double EF hand segment of human protein phosphatase PPEF-2 Sherman et al., Proc. Natl. Acad. Sci. U.S.A. 94:11639-44, 1997 (GenBank accession AF023456). The zfsta4 calmodulin homology domain has no detectable sequence homology to the calmodulin domain of SPARC.

The second EF hand of the calmodulin domain of SPARC is stabilized by a disulfide bond spanning the EF hand loop. When the two Cys residues in this EF hand were mutated to Leu residues, a 100-fold decrease in calcium ion affinity was noted (Hohenester et al., Nat. Struct. Biol.. 3:67-73, 1996). The present application also provides a mutated form of zfsta4 where the second EF hand is stabilized by replacing amino acid residue 225 (Asp) and amino acid residue 241 (Ala) of SEQ ID NO:2, with cysteine residues. It is thought that this mutated form will have higher calcium binding affinity.

Between the follistatin and calmodulin homology domains is a short segment called the alpha-helical linker which may form a linker between the two segments. This linker is predicted to have an alpha helical structure from amino acid residue 144 (Gly) through amino acid residue 166 (Asp) of SEQ ID NO:2. At the C- terminus of this linker are two basic residues which could be the location of a proteolysis site. Processing at this site of the secreted protein would release domains B and C, containing the follistatin homology domain, from the rest of the protein.

Amino acid residue 140 (Cys) of SEQ ID NO:2 of the alpha-helical linker peptide may form a disulfide bond with amino acid residue 216 (Cys) of SEQ ID

NO:2, which precedes the second EF hand in the calmodulin homology domain of zfsta4.

The I-set IG domains #1 and #2 of zfsta4 are predicted to fold into a structure similar to that determined for the telokin peptide Garcia et al., Am. J. Respir. Cell Mol. Biol. 16:489-94, 1997 and Holden et al., J. Mol. Biol. 227:840-51. 1992 (Swiss-Prot KMLS_HUMAN and PDB 1TLK). The telokin peptide falls into the class of immunoglobulins (Bork et al., J. Mol. Biol. 242:309-20, 1994) which are all beta proteins folding into a beta-sandwich like structure. These have two beta sheets comprising 3+4 beta strands. Furthermore, the telokin peptide has been sub-classified as an "I" set immunoglobulin (IG) domain. Other proteins with I set immunoglobulin domains include titin, vascular and neural cell adhesion molecules, and twitchin. In zfsta4 domains I-set IG #1 and #2 there may be two intra-domain disulfide bonds, one between cysteine residues 270 and 321 of SEQ ID NO:2 in I-set IG domain #1 and cysteine residues 362 and 413 of SEQ ID NO:2 in I-set IG domain #2. The C-terminal domain of zfsta4 shows no recognizable sequence or structural similarity to any known protein. This segment may serve to anchor the protein to the extracellular matrix, or to the cell surface membrane. Northern blot analysis of various human tissues resulted in a transcript of approximately 4 kb seen in predominately in testis with reduced expression in brain, kidney, pancreas, prostate and adrenal gland. RNA Dot Blot analysis indicated that zfsta4 was expressed ubiquitously. The present invention further provides polynucleotide molecules, including DNA and RNA molecules, encoding zfsta4 proteins. The polynucleotides of the present invention include the sense strand; the anti-sense strand; and the DNA as double-stranded, having both the sense and anti-sense strand annealed together by their respective hydrogen bonds. A representative DNA sequence encoding a zfsta4 protein is set forth in SEQ ID NO: 1. DNA sequences encoding other zfsta4 proteins can be readily generated by those of ordinary skill in the art based on the genetic code. Counteφart RNA sequences can be generated by substitution of U for T.

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 zfsta4 polypeptide 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, zfsta4 polypeptide-encoding polynucleotides comprising nucleotide 1 to nucleotide 2526 of SEQ ID NO:4 and their RNA equivalents are contemplated by the present invention. Table 1 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.

Table 1

The degenerate codons used in SEQ ID NO:4, encompassing all possible codons for a given amino acid, are set forth in Table 2.

Table 2

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 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." In general, see, Grantham, et al., Nuc. Acids Res. 8:1893-912, 1980; Haas, et al. Curr. Biol. 6:315-24, 1996; Wain-Hobson, et al., Gene 13:355-64, 1981; Grosjean and Fiers, Gene 18:199-209, 1982; Holm, Nuc. Acids Res. 14:3075-87, 1986; Ikemura, J. Mol. Biol. 158:573-97, 1982. As used herein, the term "preferential codon usage" or "preferential codons" is a term of art referring to protein translation codons that are most frequently used in cells of a certain species, thus favoring one or a few representatives of the possible codons encoding each amino acid (See Table 2). For example, the amino acid threonine (Thr) may be encoded by ACA, ACC, ACG, or ACT, but in mammalian cells ACC is the most commonly used codon; in other species, for example, insect cells, yeast, viruses or bacteria, different Thr codons may be preferential. Preferential codons for a particular species can be introduced into the polynucleotides of the present invention by a variety of methods known in the art. Introduction of preferential 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. Sequences containing preferential codons can be tested and optimized for expression in various species, and tested for functionality as disclosed herein.

Within preferred embodiments of the invention the isolated polynucleotides will hybridize to similar sized regions of SEQ ID NO:l, other polynucleotide probes, primers, fragments and sequences recited herein or sequences complementary thereto. Polynucleotide hybridization is well known in the art and widely used for many applications, see for example, Sambrook et al., Molecular Cloning: A Laboratory Manual. Second Edition, Cold Spring Harbor, NY, 1989; Ausubel et al., eds., Current Protocols in Molecular Biology, John Wiley and Sons, Inc., NY, 1987; Berger and Kimmel, eds., Guide to Molecular Cloning Techniques, Methods in Enzymology, volume 152, 1987 and Wetmur, Crit. Rev. Biochem. Mol. Biol. 26:227-59, 1990. Polynucleotide hybridization exploits the ability of single stranded complementary sequences to form a double helix hybrid. Such hybrids include DNA-DNA, RNA-RNA and DNA-RNA.

Hybridization will occur between sequences which contain some degree of complementarity. Hybrids can tolerate mismatched base pairs in the double helix, but the stability of the hybrid is influenced by the degree of mismatch. The T_m of the mismatched hybrid decreases by 1°C for every 1-1.5% base pair mismatch. Varying the stringency of the hybridization conditions allows control over the degree of mismatch that will be present in the hybrid. The degree of stringency increases as the hybridization temperature increases and the ionic strength of the hybridization buffer decreases. Hybridization buffers generally contain blocking agents such as Denhardt's solution (Sigma Chemical Co., St. Louis, Mo.), denatured salmon sperm DNA, milk powders (BLOTTO), heparin or SDS, and a Na⁺ source, such as SSC (IX SSC: 0.15 M NaCl, 15 mM sodium citrate) or SSPE (IX SSPE: 1.8 M NaCl, 10 mM NaH₂PO₄, 1 mM EDTA, pH 7.7). By decreasing the ionic concentration of the buffer, the stability of the hybrid is increased. Typically, hybridization buffers contain from between 10 mM-1 M Na⁺. Premixed hybridization solutions are also available from commercial sources such as Clontech Laboratories (Palo Alto, CA) and Promega Coφoration (Madison, WI) for use according to manufacturer's instruction. Addition of destabilizing or denaturing agents such as formamide, tetralkylammonium salts, guanidinium cations or thiocyanate cations to the hybridization solution will alter the T_m of a hybrid. Typically, formamide is used at a concentration of up to 50% to allow incubations to be carried out at more convenient and lower temperatures. Formamide also acts to reduce non-specific background when using RNA probes.

Stringent hybridization conditions encompass temperatures of about 5- 25°C below the thermal melting point (T_m) of the hybrid and a hybridization buffer having up to 1 M Na⁺. Higher degrees of stringency at lower temperatures can be achieved with the addition of formamide which reduces the T_m of the hybrid about 1°C for each 1% formamide in the buffer solution. Generally, such stringent conditions include temperatures of 20-70°C and a hybridization buffer containing up to 6X SSC and 0-50% formamide. A higher degree of stringency can be achieved at temperatures of from 40-70°C with a hybridization buffer having up to 4X SSC and from 0-50% formamide. Highly stringent conditions typically encompass temperatures of 42-70°C with a hybridization buffer having up to IX SSC and 0-50% formamide. Different degrees of stringency can be used during hybridization and washing to achieve maximum specific binding to the target sequence. Typically, the washes following hybridization are performed at increasing degrees of stringency to remove non- hybridized polynucleotide probes from hybridized complexes.

The above conditions are meant to serve as a guide and it is well within the abilities of one skilled in the art to adapt these conditions for use with a particular polypeptide hybrid. The T_m for a specific target sequence is the temperature (under defined conditions) at which 50% of the target sequence will hybridize to a perfectly matched probe sequence. Those conditions that influence the T_m include, the size and base pair content of the polynucleotide probe, the ionic strength of the hybridization solution, and the presence of destabilizing agents in the hybridization solution. Numerous equations for calculating T_m are known in the art, see for example (Sambrook et al., ibid.; Ausubel et al., ibid.; Berger and Kimmel, ibid, and Wetmur, ibid.) and are specific for DNA, RNA and DNA-RNA hybrids and polynucleotide probe sequences of varying length. Sequence analysis software such as Oligo 4.0 and Primer Premier, as well as sites on the Internet, are available tools for analyzing a given sequence and calculating T_m based on user defined criteria. Such programs can also analyze a given sequence under defined conditions and suggest suitable probe sequences. Typically, hybridization of longer polynucleotide sequences, >50 bp, is done at temperatures of about 20-25°C below the calculated T_m. For smaller probes, <50 bp, hybridization is typically carried out at the T_m or 5-10°C below. This allows for the maximum rate of hybridization for DNA-DNA and DNA-RNA hybrids.

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 zfsta4 RNA. Such tissues and cells are identified by Northern blotting

(Thomas. Proc. Natl. Acad. Sci. USA 77:5201. 1980), and include brain, testes, kidney, pancreas, prostate and adrenal gland. Total RNA can be prepared using guanidinium isothiocyanate extraction followed by isolation by centrifugation in a CsCl gradient (Chirgwin et al., Biochemistry 18:52-94, 1979). Poly (A)+ RNA is prepared from total RNA using the method of Aviv and Leder .Proc. Natl. Acad. Sci. USA 69:1408-12,

1972). Complementary DNA (cDNA) is prepared from poly(A)⁺ RNA using known methods. In the alternative, genomic DNA can be isolated. Polynucleotides encoding zfsta4 polypeptides are then identified and isolated by, for example, hybridization or PCR. A full-length clone encoding a zfsta4 polypeptide 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 zfsta4, receptor fragments, or other specific binding partners.

The polynucleotides of the present invention can also be synthesized using automated equipment. The current method of choice is the phosphoramidite method. If chemically synthesized double stranded DNA is required for an application such as the synthesis of a gene or a gene fragment, then each complementary strand is made separately. The production of short genes (60 to 80 bp) is technically straightforward and can be accomplished by synthesizing the complementary strands and then annealing them. For the production of longer genes (>300 bp), however, special strategies must be invoked, because the coupling efficiency of each cycle during chemical DNA synthesis is seldom 100%. To overcome this problem, synthetic genes (double-stranded) are assembled in modular form from single-stranded fragments that are from 20 to 100 nucleotides in length. Gene synthesis methods are well known in the art. See, for example, Glick and Pasternak, Molecular Biotechnology. Principles & Applications of Recombinant DNA, ASM Press, Washington, D.C., 1994; Itakura et al., Annu. Rev. Biochem. 53: 323-56, 1984; and Climie et al., Proc. Natl. Acad. Sci. USA 87:633-7, 1990.

The present invention further provides counteφart polypeptides and polynucleotides from other species (orthologs). These species include, but are not limited to mammalian, avian, amphibian, reptile, fish, insect and other vertebrate and invertebrate species. Of particular interest are zfsta4 polypeptides from other mammalian species, including murine, porcine, ovine, bovine, canine, feline, equine, and other primate polypeptides. Orthologs of human zfsta4 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 zfsta4 as disclosed herein. Suitable sources of mRNA can be identified by probing Northern blots with probes designed from the sequences disclosed herein. A library is then prepared from mRNA of a positive tissue or cell line. A zfsta4-encoding cDNA can then be isolated by a variety of methods, such as by probing with a complete or partial human 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 zfsta4 sequence 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 zfsta4 polypeptide. Similar techniques can also be applied to the isolation of genomic clones. Those skilled in the art will recognize that the sequence disclosed in

SEQ ID NO:l represents a single allele of human zfsta4 and that allelic variation and alternative splicing are expected to occur. Allelic variants of this sequence can be cloned by probing cDNA or genomic libraries from different individuals according to standard procedures. Allelic variants of the DNA sequence shown in SEQ ID NO:l, including those containing silent mutations and those in which mutations result in amino acid sequence changes, are within the scope of the present invention, as are proteins which are allelic variants of SEQ ID NO:2. cDNAs generated from alternatively spliced mRNAs, which retain the properties of the zfsta4 polypeptide are included within the scope of the present invention, as are polypeptides encoded by such cDNAs and mRNAs. Splice variants are known in the follistatin family, follistatin exists in at least three forms (32,000, 35,000 and 39,000 Da) due to alternative splicing. Allelic variants and splice variants of these sequences can be cloned by probing cDNA or genomic libraries from different individuals or tissues according to standard procedures known in the art. The present invention also provides isolated zfsta4 polypeptides that are substantially homologous to the polypeptides of SEQ ID NO:2 and their orthologs. The term "substantially homologous" is used herein to denote polypeptides having 50%, preferably 60%, more preferably at least 80%, sequence identity to the sequences shown in SEQ ID NO:2 or their orthologs. Such polypeptides will more preferably be at least 90% identical, and most preferably 95% or more identical to SEQ ID NO:2 or its orthologs. Percent sequence identity is determined by conventional methods. See, for example, Altschul et al., Bull. Math. Bio. 48: 603-16, 1986 and Henikoff and Henikoff, Proc. Natl. Acad. Sci. USA 89:10915-9, 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 "blosum 62" scoring matrix of Henikoff and Henikoff

(ibid.) as shown in Table 3 (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] > < H r- H

S H CN ro H 1

En If) CN CN O

1 rH ro CN CN

1 1 1

Λ [ H ro CN

1 1 1 I

Pm U. CN CN rH ro rH

1 1 1

S m o CN H H H H rH 1 1 1 1

:--; Lf. H ro H O rH ro CN CN

1 1 1 1 1 1

J ^ CN CN O ro CN rH CN rH H

CO ¹ 1 i 1 1 1 u H <Φ CN ro H o ro CN K-i ro rH ro

3 1 1 1 1 1 1

H K CO ro 00 H CN CN rH CN CN CN ro

1 1 1 1 1 1 1 1 1

O SD CN CN ro ro CN o CN CN ro ro

1 1 1 1 1 1 1 1 1 1

W IT) CN O ro ro rH CN ro H o H ro CN CN 1 1 1 1 1 1 1 1 1 1 α -n CN CN O ro CN H o ro rH o H CN H CN 1 1 1 1 1 1 1 1 1 u σ. ro ro ro rH rH ro CN ro rH H CN CN rH

1 1 1 1 1 1 1 1 1 1 1 1 1 1

Q o ro o CM rH ro rH ro ro H O rH >* ro ro

1 1 1 1 1 1 1 1 1 1 1 1 1 a KD rH ro o o O H ro ro o CN ro CN H O <* CN ro

1 1 1 1 1 1 1 1

C-J m O CN rH o CN O ro CN CN rH ro CN H H ro CN ro 1 1 1 1 1 1 1 1 1 1 1 1 1

H CN CM o H H O CN rH rH H H CN H τ- O ro CN O

1 1 1 1 1 1 1 1 1 1 1 1 o. a P U α W o a H .-. W S P- A ω H is H >

Lf) LO O CN Sequence identity of polynucleotide molecules is determined by similar methods using a ratio as disclosed above.

Those skilled in the art appreciate that there are many established algorithms available to align two amino acid sequences. The "FASTA" similarity search algorithm of Pearson and Lipman is a suitable protein alignment method for examining the level of identity shared by an amino acid sequence disclosed herein and the amino acid sequence of a putative variant zfsta4. The FASTA algorithm is described by Pearson and Lipman, Proc. Natl. Acad. Sci. USA 85:2444, 1988, and by Pearson, Meth. Enzvmol. 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 re-scored 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. 48:444, 1970; Sellers, SIAM J. Appl. Math. 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, Meth. Enzvmol. 183:63. 1990.

FASTA can also be used to determine the sequence identity of nucleic acid molecules using a ratio 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 nucleic acid molecules that encode a polypeptide having one or more "conservative amino acid substitutions," compared with the amino acid sequence of SEQ ID NO:2. Conservative amino acid substitutions can be based upon the chemical properties of the amino acids. That is, variants can be obtained that contain one or more amino acid substitutions of SEQ ID NO:2, in which an alkyl amino acid is substituted for an alkyl amino acid in a zfsta4 amino acid sequence, an aromatic amino acid is substituted for an aromatic amino acid in a zfsta4 amino acid sequence, a sulfur-containing amino acid is substituted for a sulfur- containing amino acid in a zfsta4 amino acid sequence, a hydroxy-containing amino acid is substituted for a hydroxy-containing amino acid in a zfsta4 amino acid sequence, an acidic amino acid is substituted for an acidic amino acid in a zfsta4 amino acid sequence, a basic amino acid is substituted for a basic amino acid in a zfsta4 amino acid sequence, or a dibasic monocarboxylic amino acid is substituted for a dibasic monocarboxylic amino acid in a zfsta4 amino acid sequence.

Among the common amino acids, for example, a "conservative amino acid substitution" is illustrated by a substitution among amino acids within each of the following groups: (1) glycine, alanine, valine, leucine, and isoleucine, (2) phenylalanine, tyrosine, and tryptophan, (3) serine and threonine, (4) aspartate and glutamate, (5) glutamine and asparagine, and (6) lysine, arginine and histidine.

The BLOSUM62 table 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, Proc. Natl. Acad. Sci. USA 89:10915, 1992). Accordingly, 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. Although it is possible to design amino acid substitutions based solely upon chemical properties (as discussed above), the language "conservative amino acid substitution" preferably 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. According to this system, preferred conservative amino acid substitutions are characterized by a BLOSUM62 value of at least 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).

Conservative amino acid changes in a zfsta4 gene can be introduced by substituting nucleotides for the nucleotides recited in SEQ ID NO:l. Such "conservative amino acid" variants can be obtained, for example, by oligonucleotide- directed mutagenesis, linker-scanning mutagenesis, mutagenesis using the polymerase chain reaction, and the like (see Ausubel (1995) at pages 8-10 to 8-22; and McPherson (ed.), Directed Mutagenesis: A Practical Approach (IRL Press 1991)). The ability of such variants to modulate cellular interactions or other properties of the wild-type protein as described herein, can be determined using a standard methods, such as the assays described herein. Alternatively, a variant zfsta4 polypeptide can be identified by the ability to specifically bind anti-zfsta4 antibodies. Conservative amino acid changes in a zfsta4 gene can be introduced by substituting nucleotides for the nucleotides recited in SEQ ID NO:l. Such "conservative amino acid" variants can be obtained, for example, by oligonucleotide- directed mutagenesis, linker-scanning mutagenesis, mutagenesis using the polymerase chain reaction, and the like (see Ausubel (1995) at pages 8-10 to 8-22; and McPherson (ed.), Directed Mutagenesis: A Practical Approach (IRL Press 1991)). To select for variants having the properties of the wild-type protein can be done using standard methods, such as the assays described herein. Alternatively, a variant zfsta4 polypeptide can be identified by the ability to specifically bind anti-zfsta4 antibodies. 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 small 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 affinity tag. Polypeptides comprising affinity tags can further comprise a proteolytic cleavage site between the zfsta4 polypeptide and the affinity tag. Preferred such sites include thrombin cleavage sites and factor Xa cleavage sites.

The present invention further provides a variety of other polypeptide fusions. For example, a zfsta4 polypeptide can be prepared as a fusion to a dimerizing protein as disclosed in U.S. Patents Nos. 5,155,027 and 5,567,584. Preferred dimerizing proteins in this regard include immunoglobulin constant region domains.

Immunoglobulin-zfsta4 polypeptide fusions can be expressed in genetically engineered cells to produce a variety of multimeric zfsta4 analogs. Auxiliary domains can be fused to zfsta4 polypeptides to target them to specific cells, tissues, or macromolecules. For example, a zfsta4 polypeptide or protein could be targeted to a predetermined cell type by fusing a zfsta4 polypeptide to a ligand that specifically binds to a receptor on the surface of the target cell. In this way, polypeptides and proteins can be targeted for therapeutic or diagnostic puφoses. A zfsta4 polypeptide can be fused to two or more moieties, such as an affinity tag for purification and a targeting domain. Polypeptide fusions can also comprise one or more cleavage sites, particularly between domains. See, Tuan et al., Conn. Tiss. Res. 34_ιl-9, 1996.

The proteins of the present invention can also comprise non-naturally occurring amino acid residues. Non-naturally occurring amino acids include, without limitation, tr- -.-3-methylproline, 2,4-methanoproline, c.-.-4-hydroxyproline, trαn-.-4- hydroxyproline, N-methyl-glycine, α//o-threonine, methylthreonine, hydroxyethyl- cysteine, hydroxyethylhomocysteine, nitroglutamine, homo-glutamine, pipecolic acid, thiazolidine carboxylic acid, dehydroproline, 3- and 4-methylproline, 3,3-dimethyl- proline, tert-leucine, norvaline, 2-azaphenylalanine, 3-azaphenylalanine, 4- azaphenylalanine, and 4-fluoro-phenylalanine. Several methods are known in the art for incoφorating non-naturally occurring amino acid residues into proteins. For example, an in vitro 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 commercially available enzymes and other reagents. Proteins are purified by chromatography. See, for example, Robertson et al., J. Am. Chem. Soc. 113:2722, 1991; Εllman et al., Methods Enzvmol. 202:301, 1991; Chung et al., Science 259:806-9. 1993; and Chung et al., Proc. Natl. Acad. Sci. USA 90:10145-9, 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: 19991-8, 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 occurring amino acid(s) (e.g., 2-azaphenylalanine, 3-azaphenylalanine, 4-azaphenylalanine, or 4-fluorophenyl- alanine). The non-naturally occurring amino acid is incoφorated into the protein in place of its natural counteφart. See, Koide et al., Biochem. 33:7470-6, 1994. Naturally occurring amino acid residues can be converted to non-naturally occurring species by in vitro 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).

A limited number of non-conservative amino acids, amino acids that are not encoded by the genetic code, non-naturally occurring amino acids, and unnatural amino acids may be substituted for zfsta4 amino acid residues.

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: 1081-5, 1989; Bass et al., Proc. Natl. Acad. Sci. USA 88:4498-502, 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. See also, Hilton et al., J. Biol. Chem. 271:4699-708, 1996. Sites of ligand-receptor interaction can also be determined by physical analysis of structure, as determined by such techniques as nuclear magnetic resonance, crystallography, electron diffraction or photoaffinity labeling, in conjunction with mutation of putative contact site amino acids. See, for example, de Vos et al., Science 255:306-12, 1992; Smith et al., J. Mol. Biol. 224:899-904, 1992; Wlodaver et al., FEBS Lett. 309:59-64, 1992. The identities of essential amino acids can also be inferred from analysis of homologies with related follistatins.

Multiple amino acid substitutions can be made and tested using known methods of mutagenesis and screening, such as those disclosed by Reidhaar-Olson and Sauer (Science 241:53-7, 1988) or Bowie and Sauer (Proc. Natl. Acad. Sci. USA 86:2152-6, 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. 30:10832-7, 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 46:145, 1986; Ner et al., DNA 7:127, 1988). Variants of the disclosed zfsta4 DNA and polypeptide sequences can be generated through DNA shuffling as disclosed by Stemmer, Nature 370:389-91, 1994, Stemmer, Proc. Natl. Acad. Sci. USA 91_.: 10747-51, 1994 and WIPO Publication WO 97/20078. Briefly, variant DNAs are generated by in vitro homologous recombination by random fragmentation of a parent DNA followed by reassembly using PCR, resulting in randomly introduced point mutations. This technique can be modified by using a family of parent DNAs, such as allelic variants or DNAs 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 herein can be combined with high- throughput, automated screening methods to detect activity of cloned, mutagenized polypeptides in host cells. Mutagenized DNA molecules that encode active 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 herein, one of ordinary skill in the art can identify and/or prepare a variety of polypeptide fragments or variants of SEQ ID NO:2 that retain the properties of the wild-type zfsta4 protein. Such polypeptide fragments may include the N-terminal region, the follistatin and/or calmodulin homology domains, I-set IG domains #1 and/or #2, the alpha-helical linker and the C-terminal region. Amino acid truncations or additions can also occur.

Within the polypeptides of the present invention are 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 81:3998-4002, 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 219:660-6. 1983. Antibodies that recognize short, linear epitopes are particularly useful in analytic and diagnostic applications that employ denatured protein, such as Western blotting (Tobin, Proc. Natl. Acad. Sci. USA 76:4350-4356, 1979), or in the analysis of fixed cells or tissue samples. Antibodies to linear epitopes are also useful for detecting fragments of zfsta4, such as might occur in body fluids or cell culture media.

The present invention also provides polypeptide fragments or peptides comprising an epitope-bearing portion of a zfsta4 polypeptide described herein. Such fragments or peptides may comprise an "immunogenic epitope," which is a part of a protein that elicits an antibody response when the entire protein is used as an immunogen. Immunogenic epitope-bearing peptides can be identified using standard methods (see, for example, Geysen et al., Proc. NatT Acad. Sci. USA 81:3998, 1983). Antigenic, epitope-bearing polypeptides of the present invention are useful for raising antibodies, including monoclonal antibodies, that specifically bind to a zfsta4 protein. In contrast, polypeptide fragments or peptides may comprise an

"antigenic epitope," which is a region of a protein molecule to which an antibody can specifically bind. Certain epitopes consist of a linear or contiguous stretch of amino acids, and the antigenicity of such an epitope is not disrupted by denaturing agents. It is known in the art that relatively short synthetic peptides that can mimic epitopes of a protein can be used to stimulate the production of antibodies against the protein (see, for example, Sutcliffe et al., Science 219:660, 1983). Accordingly, antigenic epitope- bearing peptides and polypeptides of the present invention are useful to raise antibodies that bind with the polypeptides described herein.

Antigenic epitope-bearing peptides and polypeptides preferably contain at least 4 to 10 amino acids, at least 10 to 15 amino acids, or about 15 to about 30 amino acids of SEQ ID NO:2. Such epitope-bearing peptides and polypeptides can be produced by fragmenting a zfsta4 polypeptide, or by chemical peptide synthesis, as described herein. Moreover, epitopes can be selected by phage display of random peptide libraries (see, for example, Lane and Stephen, Curr. Opin. Immunol. 5:268 (1993), and Cortese et al., Curr. Opin. Biotechnol. 7:616, 1996). Standard methods for identifying epitopes and producing antibodies from small peptides that comprise an epitope are described, for example, by Mole, "Epitope Mapping," in Methods in Molecular Biology, Vol. 10, Manson (ed.), pages 105-116 (The Humana Press, Inc. 1992), Price, "Production and Characterization of Synthetic Peptide-Derived Antibodies," in Monoclonal Antibodies: Production, Engineering, and Clinical Application, Ritter and Ladyman (eds.), pages 60-84 (Cambridge University Press 1995), and Coligan et al. (eds.), Current Protocols in Immunology, pages 9.3.1 - 9.3.5 and pages 9.4.1 - 9.4.11 (John Wiley & Sons 1997). 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. For any zfsta4 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 1 and 2 above. Moreover, those of skill in the art can use standard software to devise zfsta4 variants based upon the nucleotide and amino acid sequences described herein. Accordingly, the present invention includes a computer-readable medium encoded with a data structure that provides at least one of SEQ ID NO:l, SEQ ID NO:2, and SEQ ID NO:4. 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 zfsta4 polypeptides of the present invention, including full-length polypeptides, biologically active fragments, and fusion polypeptides, 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 zfsta4 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 zfsta4 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 zfsta4, or may be derived from another secreted protein (e.g., t-PA) or synthesized de novo. The secretory signal sequence is operably linked to the zfsta4 DNA sequence, i.e., the two sequences are joined in the correct reading frame and positioned to direct the newly synthesized 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 secretory 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).

Alternatively, the secretory signal sequence contained in the polypeptides of the present invention is used to direct other polypeptides into the secretory pathway. The present invention provides for such fusion polypeptides. A signal fusion polypeptide can be made wherein a secretory signal sequence derived from amino acid residues 1-22 of SEQ ID NO:2 is be operably linked to another polypeptide using methods known in the art and disclosed herein. The secretory signal sequence contained in the fusion polypeptides of the present invention is preferably fused amino-terminally to an additional peptide to direct the additional peptide into the secretory pathway. Such constructs have numerous applications known in the art. For example, these novel secretory signal sequence fusion constructs can direct the secretion of an active component of a normally non-secreted protein. Such fusions may be used in vivo or in vitro to direct peptides through the secretory pathway.

Cultured mammalian cells are suitable hosts 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 7:603, 1981 : Graham and Van der Eb, Virology 52:456. 1973), electroporation (Neumann et al., EMBQ J. L841-5, 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, and viral vectors (Miller and Rosman, BioTechniques 7:980-90, 1989; Wang and Finer, Nature Med. 2:714-6, 1996). 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. 36:59-72, 1977) and Chinese hamster ovary (e.g. CHO-K1; ATCC No. CCL 61) cell lines. Additional suitable cell lines 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. 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. Alternative markers that introduce an altered phenotype, such as green fluorescent protein, or cell surface proteins such as CD4, CD8, Class I MHC, placental alkaline phosphatase may be used to sort transfected cells from untransfected cells by such means as FACS sorting or magnetic bead separation technology. Other higher eukaryotic cells can also be used as hosts, including plant cells, insect cells and avian cells. The use of Agrobacterium rhizogenes as a vector for expressing genes in plant cells has been reviewed by Sinkar et al., J. Biosci. (Bangalore) ϋ:47-58, 1987. Transformation of insect cells and production of foreign polypeptides therein is disclosed by Guarino et al., U.S. Patent No. 5,162,222 and WIPO publication WO 94/06463. Insect cells can be infected with recombinant baculovirus, commonly derived from Autographa californica nuclear polyhedrosis virus (AcNPV). DNA encoding the zfsta4 polypeptide is inserted into the baculoviral genome in place of the AcNPV polyhedrin gene coding sequence by one of two methods. The first is the traditional method of homologous DNA recombination between wild-type AcNPV and a transfer vector containing the zfsta4 flanked by AcNPV sequences. Suitable insect cells, e.g. SF9 cells, are infected with wild-type AcNPV and transfected with a transfer vector comprising a zfsta4 polynucleotide operably linked to an AcNPV polyhedrin gene promoter, terminator, and flanking sequences. See, King and Possee, The Baculovirus Expression System: A Laboratory Guide. London, Chapman & Hall; O'Reilly et al., Baculovirus Expression Vectors: A Laboratory Manual. New York, Oxford University Press., 1994; and, Richardson, Ed., Baculovirus Expression Protocols. Methods in Molecular Biology, Totowa, NJ, Humana Press, 1995. Natural recombination within an insect cell will result in a recombinant baculovirus which contains zfsta4 driven by the polyhedrin promoter. Recombinant viral stocks are made by methods commonly used in the art.

The second method of making recombinant baculovirus utilizes a transposon-based system described by Luckow (Luckow et al., J. Virol. 67:4566-79, 1993). This system is sold in the Bac-to-Bac kit (Life Technologies, Rockville, MD). This system utilizes a transfer vector, pFastBacl™ (Life Technologies) containing a Tn7 transposon to move the DNA encoding the zfsta4 polypeptide into a baculovirus genome maintained in E. coli as a large plasmid called a "bacmid." The pFastBacl™ transfer vector utilizes the AcNPV polyhedrin promoter to drive the expression of the gene of interest, in this case zfsta4. However, pFastBacl™ can be modified to a considerable degree. The polyhedrin promoter can be removed and substituted with the baculovirus basic protein promoter (also known as Peer, p6.9 or MP promoter) which is expressed earlier in the baculovirus infection, and has been shown to be advantageous for expressing secreted proteins. See, Hill-Perkins and Possee, J. Gen. Virol. 71:971-6, 1990; Bonning et al., J. Gen. Virol. 75:1551-6, 1994; and, Chazenbalk, G.D., and Rapoport, J. Biol. Chem. 270:1543-9, 1995. In such transfer vector constructs, a short or long version of the basic protein promoter can be used. Moreover, transfer vectors can be constructed which replace the native zfsta4 secretory signal sequences with secretory signal sequences derived from insect proteins. For example, a secretory signal sequence from Ecdysteroid Glucosyltransferase (EGT), honey bee Melittin (Invitrogen, Carlsbad, CA), or baculovirus gp67 (PharMingen, San Diego, CA) can be used in constructs to replace the native secretory signal sequence. In addition, transfer vectors can include an in-frame fusion with DNA encoding an epitope tag at the C- or N-terminus of the expressed zfsta4 polypeptide, for example, a Glu-Glu epitope tag (Grussenmeyer et al., ibid.). Using a technique known in the art, a transfer vector containing zfsta4 is transformed into E. coli, and 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 jrugiperda cells, e.g. Sf9 cells. Recombinant virus that expresses zfsta4 is subsequently produced. Recombinant viral stocks are made by methods commonly used the art.

The recombinant virus is used to infect host cells, typically a cell line derived from the fall armyworm, Spodoptera frugiperda. See, in general, Glick and Pasternak, Molecular Biotechnology: Principles and Applications of Recombinant DNA, ASM Press, Washington, D.C., 1994. Another suitable cell line is the High FiveO™ cell line (Invitrogen) derived from Trichoplusia ni (U.S. Patent #5,300,435). Commercially available serum-free media are used to grow and maintain the cells. Suitable media are Sf900 H™ (Life Technologies) or ESF 921™ (Expression Systems) for the Sf9 cells; and Ex-cellO405™ (JRH Biosciences, Lenexa, KS) or Express FiveO™ (Life Technologies) for the T. ni cells. 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. The recombinant virus-infected cells typically produce the recombinant zfsta4 polypeptide at 12-72 hours post-infection and secrete it with varying efficiency into the medium. The culture is usually harvested 48 hours post- infection. Centrifugation is used to separate the cells from the medium (supernatant). The supernatant containing the zfsta4 polypeptide is filtered through micropore filters, usually 0.45 μm pore size. Procedures used are generally described in available laboratory manuals (King and Possee, ibid.; O'Reilly et al., ibid.; Richardson, C. D., ibid.). Subsequent purification of the zfsta4 polypeptide from the supernatant can be achieved using methods described herein.

Fungal cells, including yeast cells, can also be used within the present invention. Yeast species of particular interest in this regard include Saccharomyces 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, Schizosaccharomyces pombe, Kluyveromyces lactis, Kluyveromyces fragilis, Ustilago maydis, Pichia pastoris, Pichia methanolica, Pichia guillermondii and Candida maltosa are known in the art. See, for example, Gleeson et al., J. Gen. Microbiol. 132:3459-65, 1986 and Cregg, U.S. Patent No. 4,882,279. 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.

For example, the use of Pichia methanolica as host for the production of recombinant proteins is disclosed by Raymond, U.S. Patent No. 5,716,808, Raymond, U.S. Patent No. 5,736,383, Raymond et al., Yeast 14: 11-23, 1998, and in WIPO Publication Nos. WO 97/17450, WO 97/17451, WO 98/02536, and WO 98/02565. DNA molecules for use in transforming P. methanolica will commonly be prepared as double-stranded, circular plasmids, which are preferably linearized prior to transformation. 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. methanolica alcohol utilization gene {AUGI 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 {AUGI and AUG2) are deleted. For production of secreted proteins, host cells deficient in vacuolar protease genes {PEP4 and PRB1) are preferred. Electroporation is used to facilitate the introduction of a plasmid containing DNA encoding a polypeptide of interest into P. methanolica cells. It is preferred to transform P. methanolica cells by electroporation using an exponentially decaying, pulsed electric field having a field strength of from 2.5 to 4.5 kV/cm, preferably about 3.75 kV/cm, and a time constant (τ) of from 1 to 40 milliseconds, most preferably about 20 milliseconds.

Prokaryotic host cells, including strains of the bacteria Escherichia coli, Bacillus 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 expressing a zfsta4 polypeptide in bacteria such as E. coli, 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.

The adenovirus system can also be used for protein production in vitro. 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. Alternatively, 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., Cvtotechnol. 15:145-55. 1994). With either protocol, an expressed, secreted heterologous protein can be repeatedly isolated from the cell culture supernatant. Within the infected 293 cell production protocol, non-secreted proteins may also be effectively obtained.

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. P. methanolica cells are cultured in a medium comprising adequate sources of carbon, nitrogen and trace nutrients at a temperature of about 25°C to 35°C. Liquid cultures are provided with sufficient aeration by conventional means, such as shaking of small flasks or sparging of fermentors. A preferred culture medium for P. methanolica is YEPD (2% D-glucose, 2% Bacto™ Peptone (Difco Laboratories, Detroit, MI), 1% Bacto™ yeast extract (Difco Laboratories), 0.004% adenine and 0.006%) L-leucine). It is preferred to purify the polypeptides of the present invention to

>80% purity, more 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 is substantially free of other polypeptides, particularly other polypeptides of animal origin.

Expressed recombinant zfsta4 polypeptides (or chimeric zfsta4 polypeptides) can be purified using fractionation and/or conventional purification methods and media. Ammonium sulfate precipitation and acid or chaotrope extraction may be used for fractionation of samples. Exemplary purification steps may include hydroxyapatite, size exclusion, FPLC and reverse-phase high performance liquid chromatography. Suitable chromatographic media include derivatized dextrans, agarose, cellulose, polyacrylamide, specialty silicas, and the like. PEI, DEAE, QAE and Q derivatives are preferred. Exemplary chromatographic media include those media derivatized with phenyl, butyl, or octyl groups, such as Phenyl-Sepharose FF (Pharmacia), Toyopearl butyl 650 (Toso Haas, Montgomeryville, PA), Octyl-Sepharose

(Pharmacia) and the like; or polyacrylic resins, such as Amberchrom CG 71 (Toso Haas) and the like. Suitable solid supports include glass beads, silica-based resins, cellulosic resins, agarose beads, cross-linked agarose beads, polystyrene beads, cross- linked polyacrylamide resins and the like that are insoluble under the conditions in which they are to be used. These supports may be modified with reactive groups that allow attachment of proteins by amino groups, carboxyl groups, sulfhydryl groups, hydroxyl groups and/or carbohydrate moieties. Examples of coupling chemistries include cyanogen bromide activation, N-hydroxysuccinimide activation, epoxide activation, sulfhydryl activation, hydrazide activation, and carboxyl and amino derivatives for carbodiimide coupling chemistries. These and other solid media are well known and widely used in the art, and are available from commercial suppliers. Methods for binding receptor polypeptides to support media are well known in the art. Selection of a particular method is a matter of routine design and is determined in part by the properties of the chosen support. See, for example, Affinity Chromatography: Principles & Methods. Pharmacia LKB Biotechnology, Uppsala, Sweden, 1988.

The polypeptides of the present invention can be isolated by exploitation of physical properties of the zfsta4 sequence or properties of coupled tags or epitopes. For example, immobilized metal ion adsoφtion (IMAC) chromatography can be used to purify histidine-rich proteins, including those comprising polyhistidine tags. Briefly, a gel is first charged with divalent metal ions to form a chelate (Sulkowski, Trends in Biochem. 3:1-7, 1985). Histidine-rich proteins 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. Other methods of purification include purification of glycosylated proteins by lectin affinity chromatography and ion exchange chromatography (Methods in Enzvmol.. Vol. 182, "Guide to Protein Purification", M. Deutscher, (ed.), Acad. Press, San Diego, 1990, pp.529-39). Within additional embodiments of the invention, a fusion of the polypeptide of interest and an affinity tag (e.g., maltose-binding protein, FLAG tag, Glu-Glu tag, an immunoglobulin domain) may be constructed to facilitate purification. An exemplary purification method of protein constructs having an N-terminal or C- terminal affinity tag involves using an antibody to the affinity tag epitope to purify the protein using chromatography methods known in the art. SDS-PAGE, Western analysis, amino acid analysis and N-terminal sequencing can be done to confirm the identity of the purified protein.

Protein refolding (and optionally reoxidation) procedures may be advantageously used. It is preferred to purify the protein to >80% purity, more preferably to >90% purity, even more preferably >95%, 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 protein is substantially free of other proteins, particularly other proteins of animal origin. Proteins/polypeptides which bind zfsta4 (such as a zfsta4-binding receptor) can also be used for purification of zfsta4. The zfsta4-binding protein/polypeptide is immobilized on a solid support, such as beads of agarose, cross- linked agarose, glass, cellulosic resins, silica-based resins, polystyrene, cross-linked polyacrylamide, or like materials that are stable under the conditions of use. Methods for linking polypeptides to solid supports are known in the art, and include amine chemistry, cyanogen bromide activation, N-hydroxysuccinimide activation, epoxide activation, sulfhydryl activation, and hydrazide activation. The resulting medium will generally be configured in the form of a column, and fluids containing zfsta4 polypeptide are passed through the column one or more times to allow zfsta4 polypeptide to bind to the ligand-binding or receptor polypeptide. The bound zfsta4 polypeptide is then eluted using changes in salt concentration, chaotropic agents (guanidine HC1), or pH to disrupt ligand-receptor binding.

Moreover, using methods described in the art, polypeptide fusions, or hybrid zfsta4 proteins, are constructed using regions or domains of the inventive zfsta4 in combination with other polypeptides, in particular, those of other follistatin family proteins (e.g. FRP, SPARC, agrin or hevin), or heterologous proteins (Sambrook et al., ibid., Altschul et al., ibid., Picard, Cur. Opin. Biology, 5:511-5, 1994, and references therein). These methods allow the determination of the biological importance of larger domains or regions in a polypeptide of interest. Such hybrids may alter reaction kinetics, binding, constrict or expand the substrate specificity, or alter tissue and cellular localization of a polypeptide, and can be applied to polypeptides of unknown structure.

Fusion proteins can be prepared by methods known to those skilled in the art by preparing each component of the fusion protein and chemically conjugating them. Alternatively, a polynucleotide encoding both components of the fusion protein in the proper reading frame can be generated using known techniques and expressed by the methods described herein. For example, part or all of a domain(s) conferring a biological function may be swapped between zfsta4 of the present invention with the functionally equivalent domain(s) from another family member, such as FRP. Such domains include, but are not limited to, the secretory signal sequence, follistatin homology domain, calmodulin homology domain, I-set IG domains #1 and #2, the N or C-terminal domains and the alpha helical linker, for example. Such fusion proteins would be expected to have a biological functional profile that is the same or similar to polypeptides of the present invention or other known follistatin family proteins described herein, depending on the fusion constructed. Moreover, such fusion proteins may exhibit other properties as disclosed herein. Zfsta4 polypeptides or fragments thereof may also be prepared through chemical synthesis. zfsta4 polypeptides may be monomers or multimers; glycosylated or non-glycosylated; pegylated or non-pegylated; and may or may not include an initial methionine amino acid residue. Polypeptides, especially polypeptides of the present invention, can also be synthesized as described by Merrifield, J. Am. Chem. Soc. 85:2149, 1963, Stewart et al., "Solid Phase Peptide Synthesis" (2nd Edition), (Pierce Chemical Co., Rockford, IL, 1984) and Bayer & Rapp Chem. Pent. Prot. 3:3, 1986 and Atherton et al., Solid Phase Peptide Synthesis: A Practical Approach, IRL Press, Oxford, 1989, for example.

As described above, the disclosed polypeptides can be used to construct zfsta4 variants and functional fragments of zfsta4. Such variants and extracellular domain fragments are considered to be zfsta4 agonists. Another type of zfsta4 agonist is provided by anti-idiotype antibodies, and fragments thereof, which mimic the extracellular domain of zfsta4. Moreover, recombinant antibodies comprising anti- idiotype variable domains that mimic the zfsta4 extracellular domain can be used as agonists (see, for example, Monfardini et al, Proc. Assoc. Am. Physicians 108:420, 1996). zfsta4 agonists can also be constructed using combinatorial libraries. Methods for constructing and screening phage display and other combinatorial libraries are provided, for example, by Kay et al., Phage Display of Peptides and Proteins (Academic Press 1996), Verdine, U.S. Patent No. 5,783,384, Kay, et. al., U.S. Patent No. 5,747,334, and Kauffman et al., U.S. Patent No. 5,723,323.

The invention also provides antagonists, which either bind to zfsta4 polypeptides or, alternatively, to a receptor to which zfsta4 polypeptides bind, thereby inhibiting or eliminating the function of zfsta4. Such zfsta4 antagonists would include antibodies; oligonucleotides which bind either to the zfsta4 polypeptide or to its receptor; natural or synthetic analogs of zfsta4 polypeptides which retain the ability to bind the receptor but do not result in either ligand or receptor signaling. Such analogs could be peptides or peptide-like compounds. Natural or synthetic small molecules which bind to receptors of zfsta4 polypeptides and prevent signaling are also contemplated as antagonists. As such, zfsta4 antagonists would be useful as therapeutics for treating certain disorders where blocking signal from either a zfsta4 ligand or receptor would be beneficial. Zfsta4 can also be used to identify inhibitors (antagonists) of its activity.

Test compounds are added to the assays disclosed herein to identify compounds that inhibit the activity of zfsta4. In addition to those assays disclosed herein, samples can be tested for inhibition of zfsta4 activity within a variety of assays designed to measure receptor binding or the stimulation/inhibition of zfsta4-dependent cellular responses. For example, zfsta4-responsive cell lines can be transfected with a reporter gene construct that is responsive to a zfsta4-stimulated cellular pathway. Reporter gene constructs of this type are known in the art, and will generally comprise a zfsta4-DNA response element operably linked to a gene encoding an assayable protein, such as luciferase. DNA response elements can include, but are not limited to, cyclic AMP response elements (CRE), hormone response elements (HRE) insulin response element (IRE) (Nasrin et al., Proc. Natl. Acad. Sci. USA 87:5273-7, 1990) and serum response elements (SRE) (Shaw et al. Cell 56: 563-72, 1989). Cyclic AMP response elements are reviewed in Roestler et al., J. Biol. Chem. 263 (19):9063-6; 1988 and Habener, Molec. Endocrinol. 4 (8): 1087-94; 1990. Hormone response elements are reviewed in Beato, Cell 56:335-44; 1989. Candidate compounds, solutions, mixtures or extracts are tested for the ability to inhibit the activity of zfsta4 on the target cells as evidenced by a decrease in zfsta4 stimulation of reporter gene expression. Assays of this type will detect compounds that directly block zfsta4 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 zfsta4 binding to receptor using zfsta4 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 zfsta4 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 also provides isolated and purified zfsta4 polynucleotide probes and or primers. The probes and/or primers can be RNA or DNA. DNA can be either cDNA or genomic DNA. Polynucleotide probes and primers are single or double-stranded DNA or RNA, generally synthetic oligonucleotides, but may be generated from cloned cDNA or genomic sequences or its complements. 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 polynucleotide probes can be used when a small region of the gene is targeted for analysis. For gross analysis of genes, a polynucleotide probe may comprise an entire exon or more. Such probes can also be used in hybridizations to detect the presence or quantify the amount of zfsta4 gene or mRNA transcript in a sample. zfsta4 polynucleotide probes could be used to hybridize to DNA or RNA targets for diagnostic puφoses, using such techniques such as fluorescent in situ hybridization (FISH) or immunohistochemistry. Polynucleotide probes could be used to identify genes encoding zfsta4-like proteins. Such probes can also be used to screen libraries for related zfsta4 sequences. Such screening would be carried out under conditions of lower stringency which would allow identification of sequences which are substantially homologous, but not requiring complete homology to the probe sequence. Such methods and conditions are well known in the art, see, for example, Sambrook et al., Molecular Cloning: A Laboratory Manual. Second Edition, Cold Spring Harbor, NY, 1989. Such stringency conditions are described herein. Libraries may be made of genomic DNA or cDNA. Polynucleotide probes are also useful for Southern, Northern, or slot blots, colony and plaque hybridization and in situ hybridization. Mixtures of different zfsta4 polynucleotide probes can be prepared which would increase sensitivity or the detection of low copy number targets, in screening systems.

Nucleic acid molecules can be used to detect the expression of a zfsta4 gene in a biological sample. In a basic assay, a single-stranded probe molecule is incubated with RNA, isolated from a biological sample, under conditions of temperature and ionic strength that promote base pairing between the probe and target zfsta4 RNA species. After separating unbound probe from hybridized molecules, the amount of hybrids is detected. A method of detecting the presence of zfsta4 RNA in a biological sample is provided, comprising the steps of: a) contacting a zfsta4 nucleic acid probe under stringent hybridizing conditions with either i) test RNA molecules isolated from the biological sample, or ii) nucleic acid molecules synthesized from the isolated RNA molecules, wherein the probe has a nucleotide sequence comprising a portion of the nucleotide sequence of SEQ ID NOs: 1 or 3, or their complements, and b) detecting the formation of hybrids of the nucleic acid probe and either the test RNA molecules or the synthesized nucleic acid molecules, wherein the presence of the hybrids indicates the presence of zfsta4 RNA is the biological sample.

Well-established hybridization methods of RNA detection include northern analysis and dot/slot blot hybridization (see, for example, Ausubel ibid, at pages 4-1 to 4-27, and Wu et al. (eds.), "Analysis of Gene Expression at the RNA

Level," in Methods in Gene Biotechnology, pages 225-39, CRC Press, Inc., 1997). Nucleic acid probes can be detectably labeled with radioisotopes such as ²P or ³⁵S. Alternatively, zfsta4 RNA can be detected with a nonradioactive hybridization method (see, for example, Isaac (ed.), Protocols for Nucleic Acid Analysis by Nonradioactive Probes. Humana Press, Inc., 1993). Typically, nonradioactive detection is achieved by enzymatic conversion of chromogenic or chemiluminescent substrates. Illustrative nonradioactive moieties include biotin, fluorescein, and digoxigenin. Zfsta4 oligonucleotide probes are also useful for in vivo diagnosis. As an illustration, ^l8F-labeled oligonucleotides can be administered to a subject and visualized by positron emission tomography (Tavitian et al., Nat. Med. 4:467, 1998).

Numerous diagnostic procedures take advantage of the polymerase chain reaction (PCR) to increase sensitivity of detection methods. Standard techniques for performing PCR are well-known (see, generally, Mathew (ed.), Protocols in Human

Molecular Genetics, Humana Press, Inc., 1991; White (ed.), PCR Protocols: Current

Methods and Applications, Humana Press, Inc., 1993; Cotter (ed.), Molecular

Diagnosis of Cancer, Humana Press, Inc., 1996; Hanausek and Walaszek (eds.), Tumor Marker Protocols, Humana Press, Inc., 1998; Lo (ed.), Clinical Applications of PCR,

Humana Press, Inc., 1998 and Meltzer (ed.), PCR in Bioanalysis, Humana Press, Inc.,

1998). PCR primers can be designed to amplify a sequence encoding a particular zfsta4 region, such as the follistatin homology domain, encoded by about nucleotide

298 to nucleotide 505 of SEQ ID NO:l, and the calmodulin domain, encoded by about nucleotide 635 to nucleotide 856 of SEQ ID NO: 1.

One variation of PCR for diagnostic assays is reverse transcriptase-PCR (RT-PCR). In the RT-PCR technique, RNA is isolated from a biological sample, reverse transcribed to cDNA, and the cDNA is incubated with zfsta4 primers (see, for example, Wu et al. (eds.), "Rapid Isolation of Specific cDNAs or Genes by PCR," in Methods in Gene Biotechnology, pages 15-28, CRC Press, Inc. 1997). PCR is then performed and the products are analyzed using standard techniques.

As an illustration, RNA is isolated from a biological sample using, for example, the guanidinium-thiocyanate cell lysis procedure described above. Alternatively, a solid-phase technique can be used to isolate mRNA from a cell lysate. A reverse transcription reaction can be primed with the isolated RNA using random oligonucleotides, short homopolymers of dT, or zfsta4 anti-sense oligomers. Oligo-dT primers offer the advantage that various mRNA nucleotide sequences are amplified that can provide control target sequences. zfsta4 sequences are amplified by the polymerase chain reaction using two flanking oligonucleotide primers that are typically 20 bases in length.

PCR amplification products can be detected using a variety of approaches. For example, PCR products can be fractionated by gel electrophoresis, and visualized by ethidium bromide staining. Alternatively, fractionated PCR products can be transferred to a membrane, hybridized with a detectably-labeled zfsta4 probe, and examined by autoradiography. Additional alternative approaches include the use of digoxigenin-labeled deoxyribonucleic acid triphosphates to provide chemiluminescence detection, and the C-TRAK colorimetric assay. Another approach is real time quantitative PCR (Perkin-Elmer Cetus, Norwalk, CT). A fluorogenic probe, consisting of an oligonucleotide with both a reporter and a quencher dye attached, anneals specifically between the forward and reverse primers. Using the 5' endonuclease activity of Taq DNA polymerase, the reporter dye is separated from the quencher dye and a sequence-specific signal is generated and increases as amplification increases. The fluorescence intensity can be continuously monitored and quantified during the PCR reaction.

Another approach for detection of zfsta4 expression is cycling probe technology (CPT), in which a single-stranded DNA target binds with an excess of DNA-RNA-DNA chimeric probe to form a complex, the RNA portion is cleaved with RNase H, and the presence of cleaved chimeric probe is detected (see, for example, Beggs et al., J. Clin. Microbiol. 34:2985, 1996 and Bekkaoui et al., Biotechniques 20:240, 1996). Alternative methods for detection of zfsta4 sequences can utilize approaches such as nucleic acid sequence-based amplification (NASBA), cooperative amplification of templates by cross-hybridization (CATCH), and the ligase chain reaction (LCR) (see, for example, Marshall et al., U.S. Patent No. 5,686,272 (1997), Dyer et al., J. Virol. Methods 60:161, 1996; Ehricht et al., Eur. J. Biochem. 243:358. 1997; and Chadwick et al., J. Virol. Methods 70:59, 1998). Other standard methods are known to those of skill in the art. Zfsta4 probes and primers can also be used to detect and to localize zfsta4 gene expression in tissue samples. Methods for such in situ hybridization are well-known to those of skill in the art (see, for example, Choo (ed.), In Situ Hybridization Protocols, Humana Press, Inc., 1994; Wu et al. (eds.), "Analysis of Cellular DNA or Abundance of mRNA by Radioactive In Situ Hybridization (RISH)," in Methods in Gene Biotechnology, pages 259-278, CRC Press, Inc., 1997; and Wu et al. (eds.), "Localization of DNA or Abundance of mRNA by Fluorescence In Situ Hybridization (RISH)," in Methods in Gene Biotechnology, pages 279-289, CRC Press, Inc., 1997).

Various additional diagnostic approaches are well-known to those of skill in the art (see, for example, Mathew (ed.), Protocols in Human Molecular

Genetics, Humana Press, Inc., 1991; Coleman and Tsongalis, Molecular Diagnostics, Humana Press, Inc., 1996; and Elles, Molecular Diagnosis of Genetic Diseases, Humana Press, Inc., 1996).

An assay system that uses a ligand-binding receptor (or an antibody, one member of a complement/ anti-complement pair) or a binding fragment thereof, and a commercially available biosensor instrument (BIAcore, Pharmacia Biosensor,

Piscataway, NJ) may be advantageously employed. Such receptor, antibody, member of a complement/anti-complement pair or fragment is immobilized onto the surface of a receptor chip. Use of this instrument is disclosed by Karlsson, J. Immunol. Methods 145:229-40, 1991 and Cunningham and Wells, J. Mol. Biol. 234:554-63, 1993. A receptor, antibody, member or fragment is covalently attached, using amine or sulfhydryl chemistry, to dextran fibers that are attached to gold film within the flow cell. A test sample is passed through the cell. If a ligand, epitope, or opposite member of the complement/anti-complement pair is present in the sample, it will bind to the immobilized receptor, antibody or member, respectively, causing a change in the refractive index of the medium, which is detected as a change in surface plasmon resonance of the gold film. This system allows the determination of on- and off-rates, from which binding affinity can be calculated, and assessment of stoichiometry of binding.

Ligand-binding receptor polypeptides can also be used within other assay systems known in the art. Such systems include Scatchard analysis for determination of binding affinity (see Scatchard, Ann. NY Acad. Sci. 51: 660-72, 1949) and calorimetric assays (Cunningham et al., Science 253:545-48, 1991; Cunningham et al„ Science 245:821-25, 1991).

Genomic scanning places zthfr4 on chromosome 5. Probes and primers generated from the sequences disclosed herein can be used to map the zfsta4 gene to human chromosome 5. Radiation hybrid mapping is a somatic cell genetic technique developed for constructing high-resolution, contiguous maps of mammalian chromosomes (Cox et al., Science 250:245-50, 1990). Partial or full knowledge of a gene's sequence allows one to design PCR primers suitable for use with chromosomal radiation hybrid mapping panels. Radiation hybrid mapping panels are commercially available which cover the entire human genome, such as the Stanford G3 RH Panel and the GeneBridge 4 RH Panel (Research Genetics, Inc., Huntsville, AL). These panels enable rapid, PCR-based chromosomal localizations and ordering of genes, sequence- tagged sites (STSs), and other nonpolymoφhic and polymoφhic markers within a region of interest. This includes establishing directly proportional physical distances between newly discovered genes of interest and previously mapped markers. The precise knowledge of a gene's position can be useful for a number of puφoses, including: 1) determining if a sequence is part of an existing contig and obtaining additional surrounding genetic sequences in various forms, such as YACs, BACs or cDNA clones; 2) providing a possible candidate gene for an inheritable disease which shows linkage to the same chromosomal region; and 3) cross-referencing model organisms, such as mouse, which may aid in determining what function a particular gene might have. Sequence tagged sites (STSs) can also be used independently for chromosomal localization. An STS is a DNA sequence that is unique in the human genome and can be used as a reference point for a particular chromosome or region of a chromosome. An STS is defined by a pair of oligonucleotide primers that are used in a polymerase chain reaction to specifically detect this site in the presence of all other genomic sequences. Since STSs are based solely on DNA sequence they can be completely described within an electronic database, for example, Database of Sequence Tagged Sites (dbSTS), GenBank, (National Center for Biological Information, National Institutes of Health, Bethesda, MD http://www.ncbi.nlm. nih.gov), and can be searched with a gene sequence of interest for the mapping data contained within these short genomic landmark STS sequences.

The present invention also contemplates use of such chromosomal localization for diagnostic applications. Briefly, the zfsta4 gene, a probe comprising zfsta4 DNA or RNA or a subsequence thereof, can be used to determine if the zfsta4 gene is present on a particular chromosome or if a mutation has occurred. Detectable chromosomal aberrations at the zfsta4 gene locus include, but are not limited to, aneuploidy, gene copy number changes, insertions, deletions, restriction site changes and rearrangements. Such aberrations can be detected using polynucleotides of the present invention by employing molecular genetic techniques, such as restriction fragment length polymoφhism (RFLP) analysis, short tandem repeat (STR) analysis employing PCR techniques, and other genetic linkage analysis techniques known in the art (Sambrook et al., ibid.; Ausubel et. al., ibid.; Marian, Chest 108:255-65, 1995).

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 (iii) 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 include genomic DNA, cDNA, and RNA. The polynucleotide probe or primer can be RNA or DNA, and will comprise a portion of SEQ ID NO:l, the complement of SEQ ID NO:l, 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 polymoφhism (RFLP) analysis, short tandem repeat (STR) analysis employing PCR techniques, ligation chain reaction (Barany, PCR Methods and Applications L'5-16, 1991), ribonuclease protection assays, and other genetic linkage analysis techniques known in the art (Sambrook et al., ibid. ; Ausubel et. al., ibid. ; 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's 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 polymoφhism (SSCP) analysis (Hayashi, PCR Methods and Applications 1:34-8, 1991). The invention also provides anti-zfsta4 antibodies. Antibodies to zfsta4 can be obtained, for example, using as an antigen the product of a zfsta4 expression vector, or zfsta4 isolated from a natural source. Particularly useful anti-zfsta4 antibodies "bind specifically" with zfsta4. Antibodies are considered to be specifically binding if the antibodies bind to a zfsta4 polypeptide, peptide or epitope with a binding r ι 7 1 R affinity (K_a) of 10 M^" or greater, preferably 10 M^" or greater, more preferably 10

M -1 or greater, and most preferably 109 M -1 or greater. The binding affinity of an antibody can be readily determined by one of ordinary skill in the art, for example, by

Scatchard analysis (Scatchard, Ann. NY Acad. Sci. 51:660, 1949). Suitable antibodies include antibodies that bind with zfsta4 in particular domains, such as the zfsta4 follistatin homology domain (amino acid residues 65 to about 133 of SEQ ID NO:2), the calmodulin homology domain (located at about amino acid residues 175 to 250 of SEQ ID NO:2), or I-set IG domains #1 or #2 (located at about amino acid residues 251 to334 of SEQ ID NO:2 or amino acid residues 335 to 432 of SEQ ID NO:2).

Anti-zfsta4 antibodies can be produced using antigenic zfsta4 epitope- bearing peptides and polypeptides. Antigenic epitope-bearing peptides and polypeptides of the present invention contain a sequence of at least nine, preferably between 15 to about 30 amino acids contained within SEQ ID NO:2. However, peptides or polypeptides comprising a larger portion of an amino acid sequence of the invention, containing from 30 to 50 amino acids, or any length up to and including the entire amino acid sequence of a polypeptide of the invention, also are useful for inducing antibodies that bind with zfsta4. It is desirable that the amino acid sequence of the epitope-bearing peptide is selected to provide substantial solubility in aqueous solvents {i.e., the sequence includes relatively hydrophilic residues, while hydrophobic residues are preferably avoided). Moreover, amino acid sequences containing proline residues may be also be desirable for antibody production.

Polyclonal antibodies to recombinant zfsta4 protein or to zfsta4 isolated from natural sources can be prepared using methods well-known to those of skill in the art. See, for example, Green et al., "Production of Polyclonal Antisera," in Immunochemical Protocols (Manson, ed.), pages 1-5 (Humana Press 1992), and Williams et al., "Expression of foreign proteins in E. coli using plasmid vectors and purification of specific polyclonal antibodies," in DNA Cloning 2: Expression Systems, 2nd Edition, Glover et al. (eds.), page 15 (Oxford University Press 1995). The immunogenicity of a zfsta4 polypeptide can 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 zfsta4 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.

Although polyclonal antibodies are typically raised in animals such as horses, cows, dogs, chicken, rats, mice, rabbits, hamsters, guinea pigs, goats or sheep, an anti-zfsta4 antibody of the present invention may also be derived from a subhuman primate antibody. General techniques for raising diagnostically and therapeutically useful antibodies in baboons may be found, for example, in Goldenberg et al., international patent publication No. WO 91/11465, and in Losman et al., Int. J. Cancer 46:310, 1990. Antibodies can also be raised in transgenic animals such as transgenic sheep, cows, goats or pigs, and may be expressed in yeast and fungi in modified forms as will as in mammalian and insect cells.

Alternatively, monoclonal anti-zfsta4 antibodies can be generated. Rodent monoclonal antibodies to specific antigens may be obtained by methods known to those skilled in the art (see, for example, Kohler et al, Nature 256:495 (1975), Coligan et al. (eds.), Current Protocols in Immunology, Vol. 1, pages 2.5.1-2.6.7 (John Wiley & Sons 1991), Picksley et al., "Production of monoclonal antibodies against proteins expressed in E. coli," in DNA Cloning 2: Expression Systems, 2nd Edition, Glover et al. (eds.), page 93 (Oxford University Press 1995)). Briefly, monoclonal antibodies can be obtained by injecting mice with a composition comprising a zfsta4 gene product, verifying the presence of antibody production by removing a serum sample, removing the spleen to obtain B-lymphocytes, fusing the B-lymphocytes with myeloma cells to produce hybridomas, cloning the hybridomas, selecting positive clones which produce antibodies to the antigen, culturing the clones that produce antibodies to the antigen, and isolating the antibodies from the hybridoma cultures. In addition, an anti-zfsta4 antibody of the present invention may be derived from a human monoclonal antibody. Human monoclonal antibodies are obtained from transgenic mice that have been engineered to produce specific human antibodies in response to antigenic challenge. In this technique, elements of the human heavy and light chain locus are introduced into strains of mice derived from embryonic stem cell lines that contain targeted disruptions of the endogenous heavy chain and light chain loci. The transgenic mice can synthesize human antibodies specific for human antigens, and the mice can be used to produce human antibody-secreting hybridomas. Methods for obtaining human antibodies from transgenic mice are described, for example, by Green et al., Nat. Genet. 7:13, 1994, Lonberg et al., Nature 368:856, 1994, and Taylor et al., Int. Immun. 6:579, 1994.

Monoclonal antibodies can be isolated and purified from hybridoma cultures by a variety of well-established techniques. Such isolation techniques include affinity chromatography with Protein-A Sepharose, size-exclusion chromatography, and ion-exchange chromatography (see, for example, Coligan at pages 2.7.1-2.7.12 and pages 2.9.1-2.9.3; Baines et al., "Purification of Immunoglobulin G (IgG)," in Methods in Molecular Biology, Vol. 10, pages 79-104 (The Humana Press, Inc. 1992)).

For particular uses, it may be desirable to prepare fragments of anti- zfsta4 antibodies. Such antibody fragments can be obtained, for example, by proteolytic hydrolysis of the antibody. Antibody fragments can be obtained by pepsin or papain digestion of whole antibodies by conventional methods. As an illustration, antibody fragments can be produced by enzymatic cleavage of antibodies with pepsin to provide a 5S fragment denoted F(ab')₂. This fragment can be further cleaved using a thiol reducing agent to produce 3.5S Fab' monovalent fragments. Optionally, the cleavage reaction can be performed using a blocking group for the sulfhydryl groups that result from cleavage of disulfide linkages. As an alternative, an enzymatic cleavage using pepsin produces two monovalent Fab fragments and an Fc fragment directly. These methods are described, for example, by Goldenberg, U.S. patent No. 4,331,647, Nisonoff et al., Arch Biochem. Biophys. 89:230, 1960, Porter, Biochem. J. 73:119, 1959, Edelman et al., in Methods in Enzymology Vol. 1, page 422 (Academic

Press 1967), and by Coligan, ibid.

Other methods of cleaving antibodies, such as separation of heavy chains to form monovalent light-heavy chain fragments, further cleavage of fragments, or other enzymatic, chemical or genetic techniques may also be used, so long as the fragments bind to the antigen that is recognized by the intact antibody.

For example, Fv fragments comprise an association of V_H and V_L chains. This association can be noncovalent, as described by Inbar et al., Proc. Nat'l Acad. Sci. USA 69:2659, 1972. Alternatively, the variable chains can be linked by an intermolecular disulfide bond or cross-linked by chemicals such as gluteraldehyde (see, for example, Sandhu, Crit. Rev. Biotech. 12:437, 1992).

The Fv fragments may comprise V_H and V_L chains which are connected by a peptide linker. These single-chain antigen binding proteins (scFv) are prepared by constructing a structural gene comprising DNA sequences encoding the V_H and V_L domains which are connected by an oligonucleotide. The structural gene is inserted into an expression vector which is subsequently introduced into a host cell, such as E. coli. The recombinant host cells synthesize a single polypeptide chain with a linker peptide bridging the two V domains. Methods for producing scFvs are described, for example, by Whitlow et al., Methods: A Companion to Methods in Enzymology 2:97, 1991, also see, Bird et al., Science 242:423, 1988, Ladner et al., U.S. Patent No. 4,946,778, Pack et al., Bio/Technology 11:1271, 1993, and Sandhu, supra.

As an illustration, a scFV can be obtained by exposing lymphocytes to zfsta4 polypeptide in vitro, and selecting antibody display libraries in phage or similar vectors (for instance, through use of immobilized or labeled zfsta4 protein or peptide). Genes encoding polypeptides having potential zfsta4 polypeptide binding domains can be obtained by screening random peptide libraries displayed on phage (phage display) or on bacteria, such as E. coli. Nucleotide sequences encoding the polypeptides can be obtained in a number of ways, such as through random mutagenesis and random polynucleotide synthesis. These random peptide display libraries can be used to screen for peptides which interact with a known target which can be a protein or polypeptide, such as a ligand or receptor, a biological or synthetic macromolecule, or organic or inorganic substances. Techniques for creating and screening such random peptide display libraries are known in the art (Ladner et al., U.S. Patent No. 5,223,409, Ladner et al., U.S. Patent No. 4,946,778, Ladner et al., U.S. Patent No. 5,403,484, Ladner et al., U.S. Patent No. 5,571,698, and Kay et al., Phage Display of Peptides and Proteins (Academic Press, Inc. 1996)) and random peptide display libraries and kits for screening such libraries are available commercially, for instance from Clontech (Palo Alto, CA), Invitrogen Inc. (San Diego, CA), New England Biolabs, Inc. (Beverly, MA), and Pharmacia LKB Biotechnology Inc. (Piscataway, NJ). Random peptide display libraries can be screened using the zfsta4 sequences disclosed herein to identify proteins which bind to zfsta4.

Another form of an antibody fragment is a peptide coding for a single complementarity-determining region (CDR). CDR peptides ("minimal recognition units") can be obtained by constructing genes encoding the CDR of an antibody of interest. Such genes are prepared, for example, by using the polymerase chain reaction to synthesize the variable region from RNA of antibody-producing cells (see, for example, Larrick et al., Methods: A Companion to Methods in Enzymology 2: 106, 1991), Courtenay-Luck, "Genetic Manipulation of Monoclonal Antibodies," in Monoclonal Antibodies: Production, Engineering and Clinical Application, Ritter et al. (eds.), page 166 (Cambridge University Press 1995), and Ward et al., "Genetic Manipulation and Expression of Antibodies," in Monoclonal Antibodies: Principles and Applications. Birch et al., (eds.), page 137 (Wiley-Liss, Inc. 1995)).

Alternatively, an anti-zfsta4 antibody may be derived from a "humanized" monoclonal antibody. Humanized monoclonal antibodies are produced by transferring mouse complementary determining regions from heavy and light variable chains of the mouse immunoglobulin into a human variable domain. Typical residues of human antibodies are then substituted in the framework regions of the murine counteφarts. The use of antibody components derived from humanized monoclonal antibodies obviates potential problems associated with the immunogenicity of murine constant regions. General techniques for cloning murine immunoglobulin variable domains are described, for example, by Orlandi et al., Proc. Nat'l Acad. Sci. USA 86:3833, 1989. Techniques for producing humanized monoclonal antibodies are described, for example, by Jones et al., Nature 321:522. 1986, Carter et al., Proc. Nat'l Acad. Sci. USA 89:4285, 1992, Sandhu, Crit. Rev. Biotech. 12:437, 1992, Singer et al., J. Immun. 150:2844, 1993, Sudhir (ed.), Antibody Engineering Protocols (Humana

Press, Inc. 1995), Kelley, "Engineering Therapeutic Antibodies," in Protein Engineering: Principles and Practice, Cleland et al. (eds.), pages 399-434 (John Wiley & Sons, Inc. 1996), and by Queen et al., U.S. Patent No. 5,693,762 (1997).

Polyclonal anti-idiotype antibodies can be prepared by immunizing animals with anti-zfsta4 antibodies or antibody fragments, using standard techniques. See, for example, Green et al., "Production of Polyclonal Antisera," in Methods In Molecular Biology: Immunochemical Protocols, Manson (ed.), pages 1-12 (Humana Press 1992). Also, see Coligan, ibid, at pages 2.4.1-2.4.7. Alternatively, monoclonal anti-idiotype antibodies can be prepared using anti-zfsta4 antibodies or antibody fragments as immunogens with the techniques, described above. As another alternative, humanized anti-idiotype antibodies or subhuman primate anti-idiotype antibodies can be prepared using the above-described techniques. Methods for producing anti-idiotype antibodies are described, for example, by e, U.S. Patent No. 5,208,146, Greene, et. al., U.S. Patent No. 5,637,677, and Varthakavi and Minocha, J. Gen. Virol. 77:1875, 1996.

Antibodies or polypeptides herein can also be directly or indirectly conjugated to 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 can be used to identify or treat tissues or organs that express a corresponding anti-complementary molecule (receptor or antigen, respectively, for instance). More specifically, zfsta4 polypeptides or anti-zfsta4 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 may 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 may be directly or indirectly attached to the polypeptide or antibody, and include bacterial or plant toxins (for instance, diphtheria toxin, Pseudomonas exotoxin, ricin, abrin and the like), as well as therapeutic radionuclides, such as iodine-131, rhenium- 188 or yttrium-90 (either directly attached to the polypeptide or antibody, or indirectly attached through means of a chelating moiety, for instance). Polypeptides or antibodies may also be conjugated to cytotoxic drugs, such as adriamycin. For indirect attachment of a detectable or cytotoxic molecule, the detectable or cytotoxic molecule can be conjugated with a member of a complementary/ anticomplementary pair, where the other member is bound to the polypeptide or antibody portion. For these puφoses, biotin/streptavidin is an exemplary complementary/ anticomplementary pair.

Zfsta4 polynucleotides can be used as a reporter gene. The zfsta4 polynucleotide is co-transfected with a gene of interest. A detectable molecule as described above is used to detect expression of co-transfected zfsta4 as an indicator of transfection success.

Expression of zfsta4 mRNA is largely confined to brain, kidney, pancreas, adrenal gland, prostate and testis with low level expression seen in a wide variety of other tissues. This is consistent with the reported distribution of follistatin gene transcripts and transcripts of a number of other follistatin family members. This distribution suggests that zfsta4 may play a role in neuron regeneration and repair within the CNS. Injury to the adult mammalian brain or spinal cord generates a cascade of cellular events leading to inflammation, proliferation of astrocytes, angiogenesis, and formation of a glial-mesodermal scar (Logan et al., Brain Res. 587:216-25. 1992; Wang et al., Brain Res. Bull. 36:607-9, 1995 and Lindholm et al., J. Cell. Biol. 117:395-400, 1992). Production of scar tissue within the CNS provides a physical barrier for regeneration of neurons and is thought to limit the ability of the adult CNS to recover after injury. Scar tissue formation in the CNS is thought to be dependent on the localized TGF-β stimulated production of extracellular matrix components, similar to what is seen for scar tissue formation in the periphery. Indeed, TGF-β mRNA and protein have been localized to astrocytes at the site of damage in the CNS (Logan et al., ibid., Wang et al., ibid, and Lindholm et al., ibid.) suggesting that a follistatin family member, such as zfsta4, facilitates neuron regeneration and establishment of new synaptic contacts by sequestering TGF-β. SCI, a member of the follistatin family, is expressed in brain astrocytes following injury (Mendis et al., Brain Res. 730:95-106, 1996) and follistatin related protein (FRP) is secreted by glioma cells in culture (Zwijsen et al., Eur. J. Biochem. 225:937-46, 1994). Proteins that can sequester TGFβ and stimulate neuron regeneration would be useful in treatment of peripheral neuropathies by increasing spinal cord and sensory neurite outgrowth. Such polypeptides, agonists and antagonists can be included in therapeutic treatment to regenerate neurite outgrowths following strokes, brain damage caused by head injuries, and paralysis caused by spinal injuries. Application may also be made in treating neurodegenerative diseases such as Alzheimer's disease, Parkinson's disease and multiple sclerosis by stimulating neuronal outgrowths. Additional applications would include repair of transected axons which are common in lesions of multiple sclerosis.

Zfsta4 polypeptides, agonists or antagonists thereof may be therapeutically useful for treating brain and spinal cord injuries. To verify the presence of this capability in zfsta4 polypeptides, agonists or antagonists of the present invention, such zfsta4 polypeptides, agonists or antagonists are evaluated with respect to their ability to stimulate neuron regeneration and establish new synaptic contacts according to procedures known in the art, see for example Mendis et al., Brain Res. 730.95-106, 1996; Lindholm et al., J. Cell Biol. 117:395-400, 1992 and Logan et al.,

Brain Res. 587:216-25, 1992. If desired, zfsta4 polypeptide performance in this regard can be compared to other follistatins such as SCI and FRP and the like. In addition, zfsta4 polypeptides or agonists or antagonists thereof may be evaluated in combination with one or more follistatins to identify synergistic effects. If desired, zfsta4 performance in this regard can be compared to other anti-inflammatory compounds, such as dexamethasone and hydrocortisone and the like.

In addition to its potential role in treatment of injuries to the CNS, zfsta4 may also have a role in host defense. Human marrow stromal cells have been shown to be reactive with anti-activin A antibodies and the production of the B_A-subunit mRNA is increased in these cells by a number of pro-inflammatory cytokines/regulators such as interleukin lα, lipopolysaccaride, tumor necrosis factor- , or 12-O- tetradecanoylphorbol 13-acetate (Shao et al., Cytokine 10:227-35, 1998). In contrast to the stimulatory effects of these agents, the anti-inflammatory compounds dexamethasone and hydrocortisone inhibited the constitutive and cytokine-stimulated expression of activin B_A-mRNA (Shao et al., ibid.).

Application of the polypeptides of the present invention may be made to inhibit inflammatory response, stimulate a reduction in the number and activity of inflammatory cells, and diminish edema and inflammation. Such anti-inflammatory polypeptides would find application in the treatment of acute inflammation conditions, bursitis, chronic inflammatory demyelinating polyneuropathy, various forms of contact dermatitis, contact vulvovaginitis, myositis, sepsis and ulcerative colitis. Use as therapeutic agents could also be made for treating acute renal failure, pancreatitis and neonatal bronchopulmonary dysplasia. Application can also be made for ocular injuries, such as corneal injury from burns or penetration of a foreign body or ocular inflammatory diseases such as uveitis.

Application may also be made to alleviate chronic itching and inflammation associated with dermatological conditions and skin diseases such as eczema, neurodermatitis, allergy, psoriasis, xerosis, insect bites, and burns, such as thermal, chemical and radiation burns, particularly sunburns.

Symptoms associated with gout, asthma, caφal tunnel syndrome, systemic lupus erythematosus, multiple sclerosis and myasthenia gravis may also be alleviated using the compounds of the present invention. zfsta4 polypeptide, agonist or antagonist-mediated removal of bioactive activin from sites of inflammation would be a useful therapy for treatment of a wide variety of inflammatory disorders. To verify the presence of this capability in zfsta4 polypeptides, or polypeptide fragments thereof, such polypeptides and polypeptide fragments are evaluated with respect to their ability to inhibit acute inflammation. Such methods are known in the art, in particular, zfsta4 polypeptides can be tested for anti- inflammatory activity in the carrageenan-induced rat footpad edema model (Winter et al., J. Pharmac. Exp. Ther. 141:369-76, 1963 and Miele et al., Nature 335:726-30, 1988). Other models include the endotoxin-induced uveitis (EIU) model (Chan et al., Arch. Ophthalmol. 109:278-81, 1991), Oxazolone-induced inflammation model (Lloret and Moreno, Biochem. Pharmocol. 44:1437, 1992), croton oil-induced inflammation model, PMA-induced inflammation model (Miele et al., ibid.), and dextran-induced edema assay for anti-inflammatory agents (Ialenti et al., Agents Actions 29:48-9, 1990 and Rosa and Willoughby, J. Pharm. Pharmac. 23:297-8, 1971). Efficacy for treating diseases such as rheumatoid arthritis can be evaluated using indicators which would include a reduction in inflammation and relief of pain or stiffness, and in animal models indications would be derived from macroscopic inspection of joints and change in swelling of hind paws. If desired, zfsta4 polypeptide performance in this regard can be compared to other anti-inflammatory agents, in particular, dexamethasone and hydrocortisone. In addition, zfsta4 polypeptides may be evaluated in combination with one or more anti-inflammatory agents to identify synergistic effects. The recent conformation of the sequence identity of erythroid differentiation factor (EDF) and B_A subunit of activins and inhibins (Murata et al., Proc. Natl. Acad. Sci. USA 85:2434, 1988) suggests a role for zfsta4 in regulating hematopoiesis and differentiation of erythroid progenitors. EDF exhibits potent differentiation-inducing activity towards cultured erythroleukemia cells and enhances the growth of normal erythroid precursor cells in vitro and in vivo (Yu et al., Nature. 330:765, 1987, Shiozaki, et al., Biochem. Biophys. Res. Commun. 165:1155, 1989) and activin A EDF is expressed in activated macrophages (Eramaa et al., J. Exp. Med. 176:1449-52, 1992). Continuous intraperitoneal administration of follistatin to normal mice resulted in a decrease of erythroid progenitors in bone marrow and spleen (Shiozaki et al, Proc. Natl. Acad. Sci. USA 89:1553-6, 1992) demonstrating that follistatin modulates murine erythropoiesis. In humans, moreover, the follistatin related gene is a target of chromosomal rearrangement in a B-cell chronic lymphocytic leukemia (Hayette et al.. Oncogene 16:2949-54. 1998).

EDF-binding proteins such as zfsta4 polypeptides, agonists or antagonists would provide a useful therapeutic for modulating hematopoiesis and differentiation of erythroid progenitors. To verify the presence of this capability, zfsta4 polypeptides and agonists of the present invention are evaluated with respect to their ability to alter erythropoiesis by decreasing erythroid progenitors in bone marrow and spleen, according to procedures known in the art. zfsta4 antagonists can be evaluated with respect to enhancing hematopoiesis and differentiation of erythroid progenitors by inactivating follistatin and follistatin-like molecules. If desired, zfsta4 performance in this regard can be compared to other follistatins or hematopoietic factors such as erythropoietin or thrombopoietin and the like. In addition, zfsta4 polypeptides or agonists or antagonists thereof may be evaluated in combination with one or more follistatins to identify synergistic effects.

The pleiotropic actions of activins and inhibins on the gonadal/hypothalamic/pituitary axis would indicate that follistatins, such as zfsta4, would be useful in treatment of fertility disorders such characterized by abnormalities in hormone production. Activin A, for example, has been shown to stimulate hypothalamic oxytocin secretion (Sawchenko et al., Nature 334:615-7; 1988). Oxytocin specifically stimulates uterine contraction near term. Proteins which bind activin A would serve as useful therapeutics for delaying birth in pre-term pregnancies. Folliculogenesis is a physiological event characterized by moφhological and functional changes of the follicle. Of these events, antrum formation is considered the milestone of this pathway, a process that is governed by the pituitary hormone FSH. Since FSH is required for normal function of the ovaries, and activin is required for activation of FSH synthesis and secretion, it is not suφrising that follistatin is an important regulator of ovarian function. Follistatin mRNA is present in primordial follicles and its levels are dramatically increased in granulosa cells of the growing secondary or tertiary follicles and then decreases in the pre-ovulatory follicles (Shimasaki et al., Mol. Endocrinol. 3:651-9, 1989). Recent in vitro assay systems have also shown that activin is directly folliculogenic in immature mice but not in adults, the inhibition of folliculogenesis in adults was, furthermore, reversed by follistatin (Yokota et al., Endocrinology 138:4572-6. 1997). The balance between activin and follistatin appears to be critical for normal ovarian function as overexpression of mouse follistatin in female transgenic mice had a number of reproductive defects (Guo et al., Mol. Endocrinol. 12: 96-106, 1998). Follistatins, such as zfsta4, play a role in regulating folliculogenesis by affecting proliferation or differentiation of follicular cells, affecting cell-cell interactions, modulating hormones involved in the process, and the like. The role of sex steroids, such as FSH, on target tissues and organs, e.g., uterus, breast, adipose, bones and liver, has made modulators of their activity desirable for therapeutic applications. Such applications include treatments for precocious puberty, endometriosis, uterine leiomyomata, hirsutism, infertility, pre-menstrual syndrome (PMS), amenorrhea, and as contraceptive agents.

The level and ratio of gonadotropin and steroid hormones in the blood can be used to assess the existence of hormonal imbalances associated with diseases, as well as determine whether normal hormonal balance has been restored after administration of a therapeutic agent. Determination of estradiol, progesterone, LH, and FSH levels, for example, from serum is known by one of skill in the art. Such assays can be used to monitor the effects on hormone levels after administration of zfsta4 in vivo, or in a transgenic mouse model where the zfsta4 gene is expressed or the murine ortholog is deleted.

The zfsta4 polypeptides, agonists and antagonists of the present invention may be used directly or incoφorated into therapies for treating reproductive disorders. As a hormone-modulating molecule, zfsta4 polypeptides, agonists and antagonists can have therapeutic application for treating, for example, breakthrough menopausal bleeding, as part of a therapeutic regime for pregnancy support, or for treating symptoms associated with polycystic ovarian syndrome (PCOS), PMS and menopause. In addition, other in vivo rodent models are known in the art to assay effects of zfsta4 polypeptides, agonists and antagonists on, for example, polycystic ovarian syndrome (PCOS).

Activin, inhibin and follistatin are also found in the testes. mRNA encoding follistatin is located in many germ cells including type B spermatogonia, primary spermatocytes and spermatids at steps 1 to 11 (Meinhardt et al., J. Reprod. Fertil. 112:233-41, 1998). It is also found in Sertoli cells and endothelial cells but not in Leydig cells. Immunohistochemistry with anti-follistatin antibodies showed that the protein was localized to spermatids at all stages and it was also localized to endothelial and Leydig cells. This widespread localization, together with follistatin' s capacity to neutralize the activity of activin, suggests that follistatin modulates spermatogenesis and a range of other testicular functions. The balance between activin and follistatin plays an important role in normal reproduction in males was shown in mouse follistatin transgenic mice: males exhibited variable degrees of Leydig cell hypeφlasia, spermatogenesis was arrested, and seminiferous tubules degenerated which lead to infertility. This suggests that reproductive disorders due to an excess of activin or other TGF-beta family members would be amenable to treatment with members of the follistatin family. Additionally, follistatin antagonists would be useful in treatment regimes to enhance male fertility.

In vivo assays for evaluating the effect of zfsta4 polypeptides, agonists and antagonists on testes are well known in the art. For example, compounds can be injected intraperitoneally for a specific time duration. After the treatment period, animals are sacrificed and testes removed and weighed. Testicles are homogenized and sperm head counts are made (Meistrich et al., Exp. Cell Res. 99:72-78, 1976).

Other activities, for example, chemotaxic activity that may be associated with proteins of the present invention can be analyzed. For example, late stage factors in spermatogenesis may be involved in egg-sperm interactions and sperm motility. Activities, such as enhancing viability of cryopreserved sperm, stimulating the acrosome reaction, enhancing sperm motility and enhancing egg-sperm interactions may be associated with the proteins of the present invention. Assays evaluating such activities are known (Rosenberger, J. Androl. H: 89-96, 1990; Fuchs, Zentralbl Gynakol

11:117-120, 1993; Neurwinger et al., Andrologia 22:335-9, 1990; Harris et al., Human Reprod. 3:856-60. 1988; and Jockenhovel. Andrologia 22:171-178, 1990; Lessing et al., Fertil. Steril. 44:406-9 (1985); Zaneveld, In Male Infertility Chapter 11, Comhaire Ed., Chapman & Hall, London 1996). These activities are expected to result in enhanced fertility and successful reproduction.

Zfsta4 polypeptides, agonists or antagonists would provide a useful therapeutic for modulating reproductive hormones. To verify the presence of this capability, zfsta4 polypeptides, agonists and antagonists of the present invention are evaluated with respect to their ability to regulate hormones associated with reproduction, according to procedures known in the art. For example, Guoqetal, Mol, Endocrinol. 12:96-106, 1998 describes RIA measurement of serum LH, FSH, testosterone, estradiol, activin and follistatin. Zfsta4 polypeptides and agonists would be useful for treating male and female reproductive disorders. Zfsta4 antagonists would also be useful as contraceptives. If desired, zfsta4 performance in this regard can be compared to other follistatins and the like. In addition, zfsta4 polypeptides or agonists or antagonists thereof may be evaluated in combination with one or more follistatins. to identify synergistic effects.

Zfsta4 polypeptides, agonists and antagonists of the present invention may also be used in applications for enhancing fertilization during assisted reproduction in humans and in animals. Such assisted reproduction methods are known in the art and include artificial insemination, in vitro fertilization, embryo transfer, and gamete intrafallopian transfer. Such methods are useful for assisting those who may have physiological or metabolic disorders that prevent or impede natural conception. Such methods are also used in animal breeding programs, e.g., for livestock, racehorses, domestic and wild animals, and could be used as methods for the creation of transgenic animals. Zfsta4 polypeptides, agonists or antagonists could be used in the induction of ovulation, either independently or in conjunction with a regimen of gonadotropins or agents such as clomiphene citrate or bromocriptine (Speroff et al., Induction of ovulation, Clinical Gynecologic Endocrinology and Infertility, 5^th ed., Baltimore, Williams & Wilkins, 1994). Zfsta4 polypeptides, agonists and antagonists can also be used in stimulation of spermatogenesis, independently or in conjunction with other gonadotropins or sex steroids such as testosterone. As such, proteins of the present invention can be administered to the recipient prior to fertilization or combined with the sperm, an egg or an egg-sperm mixture prior to in vitro or in vivo fertilization. Such proteins can also be mixed with oocytes or sperm prior to cryopreservation to enhance viability of the preserved tissues for use in assisted reproduction. The formation of bone and teeth and is a multi-step process that is known to be initiated and promoted by members of the TGF-β superfamily, including TGF-βs and bone moφhogenic proteins (BMPs). Accumulating evidence suggests that activin and follistatin play regulatory roles in both tooth and bone formation. The temporal-spatial expression of activin and follistatin in pre-odontoblasts suggests that activin is required for proliferation of these cells, while odontoblast terminal differentiation is mediated, at least partly, by follistatin inactivation of these proliferative effects (Heikinheimo et al., J. Dent. Res. 76:1625-36; 1997, Heikinheimo et al., Eur. J. Oral Sci. 106:167-73; 1998). Follistatin is also expressed in bone (Inoue et al., Calcif. Tiss. Int. 55:395-7, 1994) and activin and follistatin have been detected by immunohistochemistry in healing fractures in the rat (Nagamine et al., J. Orthopaed. Res. 16:314-21, 1998). Follistatin has also been detected in developing bone and the expression of follistatin and activin A genes during demineralized bone matrix-induced endochondral bone development suggests a cooperative interaction between follistatin and activin during bone formation (Funaba, et al., Endocrinology 137:4250-9, 1996).

Such therapeutic agents may be used for repair of bone defects and deficiencies, such as those occurring in closed, open and non-union fractures; prophylactic use in closed and open fracture reduction; promotion of bone healing in plastic surgery; stimulation of bone ingrowth into non-cemented prosthetic joints and dental implants; elevation of peak bone mass in pre-menopausal women; treatment of growth deficiencies; treatment of periodontal disease and defects, and other tooth repair processes; increase in bone formation during distraction osteogenesis; and treatment of other skeletal disorders, such as age-related osteoporosis, post-menopausal osteoporosis, glucocorticoid-induced osteoporosis, diabetes-associated osteoporosis or disuse osteoporosis and arthritis. The compounds of the present invention can also be useful in repair of congenital, trauma-induced or surgical resection of bone (for instance, for cancer treatment), and in cosmetic surgery. Further uses include limiting or treating cartilage defects or disorders and stimulation of wound healing and tissue repair.

Well established animal models are available to test in vivo efficacy of modulators of bone formation. For example, the hypocalcemic rat or mouse model can be used to determine the effect of test compounds on serum calcium, and the ovariectomized rat or mouse can be used as a model system for osteoporosis. Bone changes seen in these models and in humans during the early stages of estrogen deficiency are qualitatively similar.

Molecules that are capable of modulating the effects of members of the TGF-β family, such as zfsta4 polypeptides, agonists or antagonists, would provide molecules useful for tooth and bone formation. To verify the presence of this capability, zfsta4 polypeptides, agonists and antagonists of the present invention are evaluated with respect to their ability to stimulate tooth or bone formation according to procedures known in the art. If desired, zfsta4 performance in this regard can be compared to other follistatins and the like. In addition, zfsta4 polypeptides or agonists or antagonists thereof may be evaluated in combination with one or more follistatins. to identify synergistic effects. Follistatin and activin also appear likely to play a role in the pathogenesis of atherosclerosis. In vascular wall cells, activin-A has been shown to inhibit endothelial cell growth and promote smooth muscle cell growth (Kojima et al., Exp. Cell Res. 206:152-6; 1993, McCarthy and Bicknell, J. Biol. Chem. 268:23066-71; 1993) and has been shown to produce a modest inhibition of scavenger receptor, SRB1, expression and foam cell formation in THP-1 macrophages (Kozaki et al., Arterioscler. Thromb. Vase. Biol. 17:2389-94; 1997). These effects are antagonized by follistatin. Activin-A, follistatin and bone moφhogenic protein-2, are produced by human atherosclerotic lesions and expression of the first two has been localized to the neointima of the diseased arteries (Inoue et al., Biochem. Biophys. Res. Commun. 205:441-8; 1994). These data suggest that the relative balance between activin, and its binding protein, follistatin, may be important in initiation and progression of atherosclerotic lesions.

Zfsta4 polypeptides, agonists or antagonists would be useful for neutralizing the activities of TGF-β family members. Such molecules would provide a novel therapy for treatment of restenosis after angioplasty. Additionally, TGF-β neutralizers would be useful for the treatment of atherosclerosis. Use of such molecules would also be applicable for treatment of stroke.

Follistatin and activin appear to play important roles in development. Two classes of TGF-β family members are believed to determine the dorsal/ventral pattern of the mesoderm in early development in Xenopus laevi. The first are related to activin and induce the formation of the dorsal mesoderm, which gives rise to muscle and the notocord (Asashima et al., Roux's Arch. Dev. Biol. 198:330-5, 1990) and the second are related to the bone moφhogenic proteins (BMPs) which inhibit dorsal mesoderm formation and induce cells to take on ventral fates, such as blood cells

(Maeno et al., Dev. Biol. 161: 522-9, 1994). Follistatin can block the activities of activin (Fukui et al., Dev. Biol. 159:131-9, 1993) and BMPs (Iemura et al., Proc. Natl. Acad. Sci. USA 95:9337-42, 1998) in these systems. Taken together, these findings suggest that zfsta4, as a member of the follistatin family of TGF-beta binding proteins, may useful as a therapy for a wide range of developmental disorders.

An alternative in vivo approach for assaying proteins of the present invention involves viral delivery systems. Exemplary viruses for this puφose include adenovirus, heφesvirus, 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 acid (for a 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 with a large number of available vectors containing different promoters. Also, 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 incoφorated into the viral DNA by direct ligation or by homologous recombination with a co- transfected plasmid. In an exemplary system, the essential El gene has been deleted from the viral vector, and the virus will not replicate unless the El gene is provided by the host cell (the human 293 cell line is exemplary). 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 secretory 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.

Polynucleotides encoding zfsta4 polypeptides are useful within gene therapy applications where it is desired to increase or inhibit zfsta4 activity. If a mammal has a mutated or absent zfsta4 gene, the zfsta4 gene can be introduced into the cells of the mammal. In one embodiment, a gene encoding a zfst2 polypeptide is introduced in vivo in a viral vector. Such vectors include an attenuated or defective DNA virus, such as, but not limited to, heφes simplex virus (HSV), papillomavirus, Epstein Barr virus (EBV), adenovirus, adeno-associated virus (AAV), and the like. Defective viruses, which entirely or almost entirely lack viral genes, are preferred. A defective virus is not infective after introduction into a cell. Use of defective viral vectors allows for administration to cells in a specific, localized area, without concern that the vector can infect other cells. Examples of particular vectors include, but are not limited to, a defective heφes simplex virus 1 (HSV1) vector (Kaplitt et al., Molec. Cell. Neurosci. 2:320-30, 1991); an attenuated adenovirus vector, such as the vector described by Stratford-Perricaudet et al., J. Clin. Invest. 90:626-30, 1992; and a defective adeno-associated virus vector (Samulski et al., J. Virol. 6 _:3096-101, 1987; Samulski et al., J. Virol. 63:3822-8, 1989).

In another embodiment, a zfsta4 gene can be introduced in a retroviral vector, e.g., as described in Anderson et al., U.S. Patent No. 5,399,346; Mann et al. Cell 33:153, 1983; Temin et al., U.S. Patent No. 4,650,764; Temin et al., U.S. Patent No.

4,980,289; Markowitz et al., J. Virol. 62:1120, 1988; Temin et al., U.S. Patent No.

5,124,263; international Patent Publication No. WO 95/07358, published March 16, 1995 by Dougherty et al.; and Kuo et al., Blood 82:845, 1993. Alternatively, the vector can be introduced by lipofection in vivo using liposomes. Synthetic cationic lipids can be used to prepare liposomes for in vivo transfection of a gene encoding a marker (Feigner et al., Proc. Natl. Acad. Sci. USA 84:7413-7, 1987; Mackey et al., Proc. Natl. Acad. Sci. USA 85:8027-31, 1988). The use of lipofection to introduce exogenous genes into specific organs in vivo has certain practical advantages. Molecular targeting of liposomes to specific cells represents one area of benefit. More particularly, directing transfection to particular cells represents one area of benefit. For instance, directing transfection to particular cell types would be particularly advantageous in a tissue with cellular heterogeneity, such as the pancreas, liver, kidney, and brain. Lipids may be chemically coupled to other molecules for the puφose of targeting. Targeted peptides (e.g., hormones or neurotransmitters), proteins such as antibodies, or non- peptide molecules can be coupled to liposomes chemically.

It is possible to remove the target cells from the body; to introduce the vector as a naked DNA plasmid; and then to re-implant the transformed cells into the body. Naked DNA vectors for gene therapy can be introduced into the desired host cells by methods known in the art, e.g., transfection, electroporation, microinjection, transduction, cell fusion, DEAE dextran, calcium phosphate precipitation, use of a gene gun or use of a DNA vector transporter. See, e.g., Wu et al., J. Biol. Chem. 267:963-7, 1992; Wu et al., J. Biol. Chem. 263: 14621-4, 1988.

Antisense methodology can be used to inhibit zfsta4 gene transcription, such as to inhibit cell proliferation in vivo. Polynucleotides that are complementary to a segment of a zfsta4-encoding polynucleotide (e.g., a polynucleotide as set froth in SEQ ID NO:l) are designed to bind to zfsta4-encoding mRNA and to inhibit translation of such mRNA. Such antisense polynucleotides are used to inhibit expression of zfsta4 polypeptide-encoding genes in cell culture or in a subject.

Transgenic mice, engineered to express the zfsta4 gene, and mice that exhibit a complete absence of zfsta4 gene function, referred to as "knockout mice" (Snouwaert et al., Science 257:1083, 1992), may also be generated (Lowell et al., Nature 366:740-42, 1993). These mice may be employed to study the zfsta4 gene and the protein encoded thereby in an in vivo system.

Transgenic mice can be engineered to over-express the human or murine zfsta4 gene in all tissues or under the control of a tissue-specific or tissue-preferred regulatory element. These over-producers of zfsta4 can be used to characterize the phenotype that results from over-expression, and the transgenic animals can serve as models for human disease caused by excess zfsta4. Transgenic mice that over-express zfsta4 also provide model bioreactors for production of zfsta4 in the milk or blood of larger animals. Methods for producing transgenic mice are well-known to those of skill in the art [see, for example, Jacob, "Expression and Knockout of Interferons in Transgenic Mice," in Overexpression and Knockout of Cytokines in Transgenic Mice, Jacob (ed.), pages 111-124 (Academic Press, Ltd. 1994), Monastersky and Robl (eds.), Strategies in Transgenic Animal Science (ASM Press 1995), and Abbud and Nilson, "Recombinant Protein Expression in Transgenic Mice," in Gene Expression Systems: Using Nature for the Art of Expression, Fernandez and Hoeffler (eds.), pages 367-397 (Academic Press, Inc. 1999)].

For example, a method for producing a transgenic mouse that expresses a zfsta4 gene can begin with adult, fertile males (studs) [B6C3fl, 2-8 months of age (Taconic Farms, Germantown, NY)], vasectomized males (duds) [B6D2fl, 2-8 months, (Taconic Farms)], prepubescent fertile females (donors) [B6C3fl, 4-5 weeks, (Taconic Farms)] and adult fertile females (recipients) [B6D2fl, 2-4 months, (Taconic Farms)]. The donors are acclimated for one week and then injected with approximately 8 IU/mouse of Pregnant Mare's Serum gonadotrophin (Sigma Chemical Company; St. Louis, MO) I.P., and 46-47 hours later, 8 IU/mouse of human Chorionic Gonadotropin (hCG (Sigma)) I.P. to induce superovulation. Donors are mated with studs subsequent to hormone injections. Ovulation generally occurs within 13 hours of hCG injection. Copulation is confirmed by the presence of a vaginal plug the morning following mating.

Fertilized eggs are collected under a surgical scope. The oviducts are collected and eggs are released into urinanalysis slides containing hyaluronidase (Sigma). Eggs are washed once in hyaluronidase, and twice in Whitten's W640 medium [described, for example, by Menino and O'Claray, Biol. Reprod. 77:159 (1986), and Dienhart and Downs, Zygote 4:129 (1996)] that has been incubated with 5% CO2, 5%

02, and 90% N2 at 37°C. The eggs are then stored in a 37°C/5% CO2 incubator until microinjection.

Ten to twenty micrograms of plasmid DNA containing a zfsta4 encoding sequence is linearized, gel-purified, and resuspended in 10 mM Tris-HCl (pH 7.4), 0.25 mM EDTA (pH 8.0), at a final concentration of 5-10 nanograms per microliter for microinjection. Plasmid DNA is microinjected into harvested eggs contained in a drop of W640 medium overlaid by warm, CO2-equilibrated mineral oil. The DNA is drawn into an injection needle (pulled from a 0.75mm ID, 1mm OD borosilicate glass capillary), and injected into individual eggs. Each egg is penetrated with the injection needle, into one or both of the haploid pronuclei. Picoliters of DNA are injected into the pronuclei, and the injection needle withdrawn without coming into contact with the nucleoli. The procedure is repeated until all the eggs are injected. Successfully microinjected eggs are transferred into an organ tissue-culture dish with pre-gassed W640 medium for storage overnight in a 37°C/5% CO2 incubator.

The following day, two-cell embryos are transferred into pseudopregnant recipients. The recipients are identified by the presence of copulation plugs, after copulating with vasectomized duds. Recipients are anesthetized and shaved on the dorsal left side and transferred to a surgical microscope. A small incision is made in the skin and through the muscle wall in the middle of the abdominal area outlined by the ribcage, the saddle, and the hind leg, midway between knee and spleen. The reproductive organs are exteriorized onto a small surgical drape. The fat pad is stretched out over the surgical drape, and a baby serrefine (Roboz, Rockville, MD) is attached to the fat pad and left hanging over the back of the mouse, preventing the organs from sliding back in.

With a fine transfer pipette containing mineral oil followed by alternating W640 and air bubbles, 12-17 healthy two-cell embryos from the previous day's injection are transferred into the recipient. The swollen ampulla is located and holding the oviduct between the ampulla and the bursa, a nick in the oviduct is made with a 28 g needle close to the bursa, making sure not to tear the ampulla or the bursa.

The pipette is transferred into the nick in the oviduct, and the embryos are blown in, allowing the first air bubble to escape the pipette. The fat pad is gently pushed into the peritoneum, and the reproductive organs allowed to slide in. The peritoneal wall is closed with one suture and the skin closed with a wound clip. The mice recuperate on a 37°C slide warmer for a minimum of four hours. The recipients are returned to cages in pairs, and allowed 19-21 days gestation. After birth, 19-21 days postpartum is allowed before weaning. The weanlings are sexed and placed into separate sex cages, and a 0.5 cm biopsy (used for genotyping) is snipped off the tail with clean scissors.

Genomic DNA is prepared from the tail snips using, for example, a Qiagen Dneasy kit following the manufacturer's instructions. Genomic DNA is analyzed by PCR using primers designed to amplify a zfsta4 gene or a selectable marker gene that was introduced in the same plasmid. After animals are confirmed to be transgenic, they are back-crossed into an inbred strain by placing a transgenic female with a wild-type male, or a transgenic male with one or two wild-type female(s). As pups are born and weaned, the sexes are separated, and their tails snipped for genotyping. To check for expression of a transgene in a live animal, a partial hepatectomy is performed. A surgical prep is made of the upper abdomen directly below the zyphoid process. Using sterile technique, a small 1.5-2 cm incision is made below the sternum and the left lateral lobe of the liver exteriorized. Using 4-0 silk, a tie is made around the lower lobe securing it outside the body cavity. An atraumatic clamp is used to hold the tie while a second loop of absorbable Dexon (American Cyanamid; Wayne, N.J.) is placed proximal to the first tie. A distal cut is made from the Dexon tie and approximately 100 mg of the excised liver tissue is placed in a sterile petri dish. The excised liver section is transferred to a 14 ml polypropylene round bottom tube and snap frozen in liquid nitrogen and then stored on dry ice. The surgical site is closed with suture and wound clips, and the animal's cage placed on a 37°C heating pad for 24 hours post operatively. The animal is checked daily post operatively and the wound clips removed 7-10 days after surgery. The expression level of zfsta4 mRNA is examined for each transgenic mouse using an RNA solution hybridization assay or polymerase chain reaction.

In addition to producing transgenic mice that over-express zfsta4, it is useful to engineer transgenic mice with either abnormally low or no expression of the gene. Such transgenic mice provide useful models for diseases associated with a lack of zfsta4. As discussed above, zfsta4 gene expression can be inhibited using anti-sense genes, ribozyme genes, or external guide sequence genes. To produce transgenic mice that under-express the zfsta4 gene, such inhibitory sequences are targeted to murine zfsta4 mRNA. Methods for producing transgenic mice that have abnormally low expression of a particular gene are known to those in the art [see, for example, Wu et al., "Gene Underexpression in Cultured Cells and Animals by Antisense DNA and RNA Strategies," in Methods in Gene Biotechnology, pages 205-224 (CRC Press 1997)].

An alternative approach to producing transgenic mice that have little or no zfsta4 gene expression is to generate mice having at least one normal zfsta4 allele replaced by a nonfunctional zfsta4 gene. One method of designing a nonfunctional zfsta4 gene is to insert another gene, such as a selectable marker gene, within a nucleic acid molecule that encodes murine zfsta4. Standard methods for producing these so- called "knockout mice" are known to those skilled in the art [see, for example, Jacob, "Expression and Knockout of Interferons in Transgenic Mice," in Overexpression and Knockout of Cytokines in Transgenic Mice, Jacob (ed.), pages 111-124 (Academic Press, Ltd. 1994), and Wu et al., "New Strategies for Gene Knockout," in Methods in Gene Biotechnology, pages 339-365 (CRC Press 1997)].

Polynucleotides and polypeptides of the present invention will additionally find use as educational tools as a laboratory practicum kits for courses related to genetics and molecular biology, protein chemistry and antibody production and analysis. Due to its unique polynucleotide and polypeptide sequence molecules of zfsta4 can be used as standards or as "unknowns" for testing puφoses. For example, zfsta4 polynucleotides can be used as an aid, such as, for example, to teach a student how to prepare expression constructs for bacterial, viral, and/or mammalian expression, including fusion constructs, wherein zfsta4 is the gene to be expressed; for determining the restriction endonuclease cleavage sites of the polynucleotides; determining mRNA and DNA localization of zfsta4 polynucleotides in tissues (i.e., by Northern and Southern blotting as well as polymerase chain reaction); and for identifying related polynucleotides and polypeptides by nucleic acid hybridization.

Zfsta4 polypeptides can be used educationally as an aid to teach preparation of antibodies; identifying proteins by Western blotting; protein purification; determining the weight of expressed zfsta4 polypeptides as a ratio to total protein expressed; identifying peptide cleavage sites; coupling amino and carboxyl terminal tags; amino acid sequence analysis, as well as, but not limited to monitoring biological activities of both the native and tagged protein (i.e., receptor binding, signal transduction, proliferation, and differentiation) in vitro and in vivo. Zfsta4 polypeptides can also be used to teach analytical skills such as mass spectrometry, circular dichroism to determine conformation, especially of the four alpha helices, x-ray crystallography to determine the three-dimensional structure in atomic detail, nuclear magnetic resonance spectroscopy to reveal the structure of proteins in solution. For example, a kit containing the zfsta4 can be given to the student to analyze. Since the amino acid sequence would be known by the professor, the protein can be given to the student as a test to determine the skills or develop the skills of the student, the teacher would then know whether or not the student has correctly analyzed the polypeptide. Since every polypeptide is unique, the educational utility of zfsta4 would be unique unto itself.

The antibodies that bind specifically to zfsta4 can be used as a teaching aid to instruct students how to prepare affinity chromatography columns to purify zfsta4, cloning and sequencing the polynucleotide that encodes an antibody and thus as a practicum for teaching a student how to design humanized antibodies. The zfsta4 gene, polypeptide or antibody would then be packaged by reagent companies and sold to universities so that the students gain skill in art of molecular biology. Because each gene and protein is unique, each gene and protein creates unique challenges and learning experiences for students in a lab practicum. Such educational kits, containing the zfsta4 gene, polypeptide or antibody, are considered within the scope of the present invention.

The zfsta4 polypeptides are also contemplated for pharmaceutical use. Pharmaceutically effective amounts of zfsta4 polypeptides, agonists or zfsta4 antagonists of the present invention can be formulated with pharmaceutically acceptable carriers for parenteral, oral, nasal, rectal, topical, transdermal administration or the like, according to conventional methods. Formulations may further include one or more diluents, fillers, emulsifiers, preservatives, buffers, excipients, and the like, and may be provided in such forms as liquids, powders, emulsions, suppositories, liposomes, transdermal patches and tablets, for example. Slow or extended-release delivery systems, including any of a number of biopolymers (biological-based systems), systems employing liposomes, and polymeric delivery systems, can also be utilized with the compositions described herein to provide a continuous or long-term source of the zfsta4 polypeptide or antagonist. Such slow release systems are applicable to formulations, for example, for oral, topical and parenteral use. The term "pharmaceutically acceptable carrier" refers to a carrier medium which does not interfere with the effectiveness of the biological activity of the active ingredients and which is not toxic to the host or patient. One skilled in the art may formulate the compounds of the present invention in an appropriate manner, and in accordance with accepted practices, such as those disclosed in Remington's Pharmaceutical Sciences, Gennaro (ed.), Mack Publishing Co., Easton, PA 1990.

As used herein a "pharmaceutically effective amount" of a zfsta4 polypeptide, agonist or antagonist is an amount sufficient to induce a desired biological result. The result can be alleviation of the signs, symptoms, or causes of a disease, or any other desired alteration of a biological system. For example, an effective amount of a zfsta4 polypeptide is that which provides either subjective relief of symptoms or an objectively identifiable improvement as noted by the clinician or other qualified observer. Effective amounts of the zfsta4 polypeptides can vary widely depending on the disease or symptom to be treated. The amount of the polypeptide to be administered and its concentration in the formulations, depends upon the vehicle selected, route of administration, the potency of the particular polypeptide, the clinical condition of the patient, the side effects and the stability of the compound in the formulation. Thus, the clinician will employ the appropriate preparation containing the appropriate concentration in the formulation, as well as the amount of formulation administered, depending upon clinical experience with the patient in question or with similar patients. Such amounts will depend, in part, on the particular condition to be treated, age, weight, and general health of the patient, and other factors evident to those skilled in the art. Typically a dose will be in the range of 0.1-100 mg/kg of subject. Doses for specific compounds may be determined from in vitro or ex vivo studies in combination with studies on experimental animals. Concentrations of compounds found to be effective in vitro or ex vivo provide guidance for animal studies, wherein doses are calculated to provide similar concentrations at the site of action. The dosages of the present compounds used to practice the invention include dosages effective to result in the desired effects. Estimation of appropriate dosages effective for the individual patient is well within the skill of the ordinary prescribing physician or other appropriate health care practitioner. As a guide, the clinician can use conventionally available advice from a source such as the Physician's Desk Reference, 48^th Edition, Medical Economics Data Production Co., Montvale, New Jersey 07645-1742 (1994).

Preferably the compositions are presented for administration in unit dosage forms. The term "unit dosage form" refers to physically discrete units suitable as unitary dosed for human subjects and animals, each unit containing a predetermined quantity of active material calculated to produce a desired pharmaceutical effect in association with the required pharmaceutical diluent, carrier or vehicle. Examples of unit dosage forms include vials, ampules, tablets, caplets, pills, powders, granules, eyedrops, oral or ocular solutions or suspensions, ocular ointments, and oil-in-water emulsions. Means of preparation, formulation and administration are known to those of skill, see generally Remington's ibid.

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

EXAMPLES

Example 1

Extension of EST Sequence

The novel zfsta4 polypeptide-encoding polynucleotides of the present invention were initially identified by querying an EST database for follistatin homologs.

In addition to EST sequence, PCR was also used to generate a full length nucleotide sequence encoding a zfsta4 polypeptide. A PCR panel containing the cDNA from 18 tissue samples was screened for the zfsta4. PCR reactions were set up using oligos

ZC23578 (SEQ ID NO:8) and ZC23580 (SEQ ID NO:9) as primers. The amplification was carried out as follows: 1 cycle at 94°C for 2 minutes, 35 cycles of 94°C for 30 seconds, 70°C 30 seconds and 72°C for 30 seconds, followed by a 5 minute extension at

72°C. Positive tissue samples included prostate and testis. 5' RACE was performed using testis tissue as template to obtain the full length zfsta4 nucleotide sequence which was confirmed by sequencing on an ABIPRISM ™ model 377 DNA sequencer (Perkin- Elmer Cetus, Norwalk, Ct.) using the ABI PRISM™ Dye Terminator Cycle Sequencing

Ready Reaction Kit (Perkin-Elmer Coφ.) according to manufacturer's instructions.

Sequencing reactions were carried out in a Hybaid OmniGene Temperature Cycling

System (National \Labnet Co., Woodbridge, NY). SEQUENCHER™ 3.0 sequence analysis software (Gene Codes Coφoration, Ann Arbor, MI) was used for data analysis. The resulting 2746 bp nucleotide sequence encoding a zfsta4 polypeptide is provided in

SEQ ID NO: 1.

Example 2 Tissue Distribution

Human Multiple Tissue Northern Blots (MTN I, MTN fl and MTN -H; Clontech) were probed to determine the tissue distribution of human zfsta4 expression. An approximately 115 bp PCR derived probe (SEQ ID NO:6) was amplified using a human testis Marathon™ cDNA library (Clontech) as a template and oligonucleotide ZC23578 (SEQ ID NO:7) and ZC23588 (SEQ ID NO:8) as primers. The amplification was carried out as follows: 1 cycle at 94°C for 1.5 minutes, 35 cycles of 94°C for 15 seconds and 60°C 30 seconds, followed by 1 cycle at 72°C for 10 minutes. The PCR products were visualized by agarose gel electrophoresis and the 115 bp PCR product was purified using a Gel Extraction Kit (Qiagen, Chatsworth, CA) according to manufacturer's instructions. The probe was radioactively labeled using the Red-prime π labeling kit (Amersham, Arlington Heights, IL) according to the manufacturer's instructions. The probe was purified using a NUCTRAP push column (Stratagene). EXPRESSHYB (Clontech) solution was used for prehybridization and as a hybridizing solution for the Northern blots. Hybridization and washes were done under appropriately stringent conditions. A strong transcript of approximately 4 kb was seen predominately in testis with reduced expression in brain, kidney, pancreas, prostate and adrenal gland.

A RNA Master Dot Blot (Clontech) that contained RNAs from various tissues that were normalized to 8 housekeeping genes was also probed and hybridized as described above. Zfsta4 was expressed ubiquitously.

From the foregoing, it will be appreciated that, although specific embodiments of the invention have been described herein for puφoses of illustration, various modifications may be made without deviating from the spirit and scope of the invention. Accordingly, the invention is not limited except as by the appended claims.

Claims

CLAIMSWhat is claimed is:

1. An isolated polypeptide comprising a follistatin homology domain, wherein said follistatin homology domain comprises amino acid residues 65 to 133 of the amino acid sequence of SEQ ID NO:2.

2. An isolated polypeptide of claim 1, wherein said polypeptide further comprises an alpha helical linker region that resides in a carboxyl-terminal position relative to said follistatin homology domain, wherein said alpha helical linker region comprises amino acid residues 134 to 173 of the amino acid sequence of SEQ ID NO:2.

3. An isolated polypeptide of claim 1, wherein said polypeptide further comprises an alpha helical linker region and a calmodulin homology domain that resides in a carboxyl-terminal position relative to said follistatin homology domain, wherein said alpha helical linker region and calmodulin homology domain comprises amino acid residues 134 to 250 of the amino acid sequence of SEQ ID NO:2.

4. An isolated polypeptide of claim 1, wherein said polypeptide further comprises an alpha helical linker region, a calmodulin homology domain, and two I-set Ig domains that reside in a carboxyl-terminal position relative to said follistatin homology domain, wherein said alpha helical linker region, a calmodulin homology domain, and I-set Ig domains comprise amino acid residues 134 to 432 of the amino acid sequence of SEQ ID NO:2.

5. An isolated polypeptide of claim 1, wherein said polypeptide further comprises an alpha helical linker region, a calmodulin homology domain, two I-set Ig domains, and a carboxy-terminal domain that resides in a carboxyl-terminal position relative to said follistatin homology domain, wherein said alpha helical linker region, a calmodulin homology domain, two I-set Ig domains, and carboxy-terminal domain comprises amino acid residues 134 to 842 of said amino acid sequence of SEQ ID NO:2.

6. An isolated polypeptide of claim 1, wherein said polypeptide further comprises a hydrophilic linker region that resides in an amino-terminal position relative to said follistatin homology domain, wherein said hydrophobic linker region comprises amino acid residues 23 to 64 of the amino acid sequence of SEQ ID NO:2.

7. An isolated polypeptide of claim 6, wherein said polypeptide further comprises a secretory signal sequence that resides in an amino-terminal position relative to said hydrophobic linker region, wherein said secretory signal sequence comprises amino acid residues 1 to 22 of the amino acid sequence of SEQ ID NO:2.

8. An isolated polypeptide having an amino acid sequence that is at least 80% identical to the amino acid sequence of SEQ ID NO:2, wherein said isolated polypeptide specifically binds with an antibody to which a polypeptide having the amino acid sequence of SEQ ID NO:2 specifically binds.

9. An isolated polypeptide of claim 8, wherein said isolated polypeptide has an amino acid sequence that is at least 90% identical to the amino acid sequence of SEQ ID NO:2.

10. An isolated polypeptide of claim 8, wherein any difference between said amino acid sequence and said corresponding amino acid sequence of SEQ ID NO:2 is due to one or more conservative amino acid substitutions.

11. An isolated polypeptide of claim 8, wherein the amino acid percent identity is determined using a FASTA program with ktup=l, gap opening penalty=10, gap extension penalty=l, and substitution matrix=blosum62, with other parameters set as default.

12. An isolated polypeptide comprising amino acid residues 23 to 842 of SEQ ID NO:2.

13. An isolated polypeptide comprising the amino acid sequence of SEQ ID NO:2.

14. An isolated polypeptide selected from the group consisting of: a) a polypeptide consisting of the sequence of amino acid residues from residue 23 to residue 64 of SEQ ID NO:2; b) a polypeptide consisting of the sequence of amino acid residues from residue 65 to residue 133 of SEQ ID NO:2; c) a polypeptide consisting of the sequence of amino acid residues from residue 134 to residue 173 of SEQ ID NO:2; d) a polypeptide consisting of the sequence of amino acid residues from residue 65 to residue 173 of SEQ ID NO:2; e) a polypeptide consisting of the sequence of amino acid residues from residue 65 to residue 250 of SEQ ID NO:2; f) a polypeptide consisting of the sequence of amino acid residues from residue 65 to residue 334 of SEQ ID NO:2; g) a polypeptide consisting of the sequence of amino acid residues from residue 65 to residue 432 of SEQ ID NO:2; h) a polypeptide consisting of the sequence of amino acid residues from residue 65 to residue 842 of SEQ ID NO:2; i) a polypeptide consisting of the sequence of amino acid residues from residue 134 to residue 250 of SEQ ID NO:2; j) a polypeptide consisting of the sequence of amino acid residues from residue 134 to residue 334 of SEQ ID NO:2; and k) a polypeptide consisting of the sequence of amino acid residues from residue 134 to residue 432 of SEQ ID NO:2.

15. An isolated polypeptide according to claim 1, further comprising an affinity tag or binding domain.

16. A fusion protein comprising a secretory signal sequence having the amino acid sequence of amino acid residues 1-22 of SEQ ID NO:2, wherein said secretory signal sequence is operably linked to an additional polypeptide.

17. A fusion protein consisting essentially of a first portion and a second portion joined by a peptide bond, said first portion comprising a polypeptide according to claim 1 ; and said second portion comprising another polypeptide.

18. An isolated polynucleotide molecule that encodes a polypeptide according to claim 1.

19. An isolated polynucleotide molecule according to claim 18, encoding a polypeptide further comprising an alpha helical linker region that resides in a carboxyl- terminal position relative to said follistatin homology domain, wherein said alpha helical linker region comprises amino acid residues 134 to 173 of the amino acid sequence of SEQ ID NO:2.

20. An isolated polynucleotide of claim 18, wherein said polynucleotide encodes a polypeptide further comprising an alpha helical linker region and a calmodulin homology domain that resides in a carboxyl-terminal position relative to said follistatin homology domain, wherein said alpha helical linker region and calmodulin homology domain comprises amino acid residues 134 to 250 of the amino acid sequence of SEQ ID NO:2.

21. An isolated polynucleotide of claim 18, wherein said polynucleotide encodes a polypeptide further comprising an alpha helical linker region, a calmodulin homology domain, and two I-set Ig domains that reside in a carboxyl-terminal position relative to said follistatin homology domain, wherein said alpha helical linker region and a calmodulin homology domain, and I-set Ig domains comprise amino acid residues 134 to 432 of the amino acid sequence of SEQ ID NO:2.

22. An isolated polynucleotide of claim 18, wherein said polynucleotide encodes a polypeptide further comprising an alpha helical linker region, a calmodulin homology domain, two I-set Ig domains, and a carboxy-terminal domain that resides in a carboxyl-terminal position relative to said follistatin homology domain, wherein said alpha helical linker region, a calmodulin homology domain, two I-set Ig domains, and carboxy- terminal domain comprises amino acid residues 134 to 842 of said amino acid sequence of SEQ ID NO:2.

23. An isolated polynucleotide according to claim 22, wherein said polypeptide further comprises an affinity tag or binding domain.

24. An isolated polynucleotide molecule, wherein said polynucleotide molecule is a degenerate nucleotide sequence encoding a polypeptide according to claim 1.

25. An isolated polynucleotide encoding a polypeptide having an amino acid sequence that is at least 80% identical to the amino acid sequence of SEQ ID NO: 2, wherein said isolated polypeptide specifically binds with an antibody to which a polypeptide having the amino acid sequence of SEQ ID NO:2 specifically binds.

26. An isolated polynucleotide of claim 25, wherein said isolated polypeptide has an amino acid sequence that is at least 90% identical to the amino acid sequence of SEQ ID NO: 2.

27. An isolated polynucleotide of claim 25, wherein any difference between said amino acid sequence and said corresponding amino acid sequence of SEQ ID NO:2 is due to one or more conservative amino acid substitutions.

28. An isolated polynucleotide of claim 25, wherein the amino acid percent identity is determined using a FASTA program with ktup=l, gap opening penal ty= 10, gap extension penalty=l, and substitution matrix=blosum62, with other parameters set as default.

29. An isolated polynucleotide molecule comprising nucleotides 183-2632 of SEQ ID NO: 1.

30. An isolated polynucleotide molecule comprising the nucleotide sequence of nucleotides 107 to 2632 of SEQ ID NO: 1.

31. An isolated polynucleotide molecule of SEQ ID NO: 1.

32. An isolated polynucleotide selected from the group consisting of: a) a polynucleotide consisting of nucleotides 107-172 of SEQ ID NO:l; b) a polynucleotide consisting of nucleotides 173-297 of SEQ ID NO: 1 ; c) a polynucleotide consisting of nucleotides 298-505 of SEQ ID NO: 1 ; d) a polynucleotide consisting of nucleotides 506-625 of SEQ ID NO: 1 ; e) a polynucleotide consisting of nucleotides 298-625 of SEQ ID NO: 1 ; f) a polynucleotide consisting of nucleotides 298-856 of SEQ ID NO: 1 ; g) a polynucleotide consisting of nucleotides 298- 1108 of SEQ ID NO: 1 ; h) a polynucleotide consisting of nucleotides 298-1402 of SEQ ID NO: 1 ; i) a polynucleotide consisting of nucleotides 298-2632 of SEQ ID NO: 1 ; j) a polynucleotide consisting of nucleotides 506-856 of SEQ ID NO: 1 ; k) a polynucleotide consisting of nucleotides 506- 1108 of SEQ ID NO: 1 ; 1) a polynucleotide consisting of nucleotides 506-1402 of SEQ ID NO: 1 ; and m) a polynucleotide consisting of nucleotides 506-2632 of SEQ ID NO: 1.

33. A polynucleotide encoding a fusion protein comprising a secretory signal sequence having the amino acid sequence of amino acid residues 1-22 of SEQ ID NO:2, wherein said secretory signal sequence is operably linked to an additional polypeptide.

34. A polynucleotide molecule encoding a fusion protein consisting essentially of a first portion and a second portion joined by a peptide bond, said first portion comprising a polypeptide according to Claim 1 ; and said second portion comprising another polypeptide.

35. An expression vector comprising the following operably linked elements: a transcription promoter; a polynucleotide molecule that encodes a polypeptide according to claim 1; and a transcription terminator.

36. An expression vector according to claim 35 further comprising a secretory signal sequence operably linked to said polypeptide.

37. An expression vector according to claim 35, wherein said polynucleotide encodes a polypeptide covalently linked amino terminally or carboxy terminally to an affinity tag.

38. A cultured cell into which has been introduced an expression vector comprising the following operably linked elements: a transcription promoter; a polynucleotide molecule that encodes a polypeptide according to claim 1; and a transcription terminator, wherein said cultured cell expresses said polypeptide encoded by said polynucleotide segment.

39. A method of producing a polypeptide comprising: culturing a cell into which has been introduced an expression vector comprising the following operably linked elements: a transcription promoter; a polynucleotide molecule that encodes a polypeptide according to claim 1; and a transcription terminator; whereby said cell expresses said polypeptide encoded by said polynucleotide segment; and recovering said expressed polypeptide.

40. An antibody or antibody fragment that specifically binds to a polypeptide according to claim 1.

41. An antibody according to claim 40, wherein said antibody is selected from the group consisting of: a) polyclonal antibody; b) murine monoclonal antibody; c) humanized antibody derived from b); and d) human monoclonal antibody.

42. An antibody fragment according to claim 41, wherein said antibody fragment is selected from the group consisting of F(ab'), F(ab), Fab', Fab, Fv, scFv, and minimal recognition unit.

43. An anti-idiotype antibody that specifically binds to said antibody of claim 40

44. A polypeptide according to claim 1, in combination with a pharmaceutically acceptable vehicle.