WO1998022492A1 - Nucleotide and protein sequences of liver activin/inhibin and methods based thereon - Google Patents

Abstract

The present invention relates to the discovery, identification and characterization of a novel subgroup of activins/inhibins, herein referred to as liver activins, that during vertebrate adult life are specifically expressed in the liver. The invention relates to the activin βC and βE subunits and other members of the liver activin subgroup. The invention encompasses liver activin nucleotides; liver activin genomic regulatory elements that regulate the expression of liver activins; host cell expression systems; liver activin proteins, fusion proteins, polypeptides and peptides; antibodies to liver activin protein, transgenic animals that express a liver activin; recombinant knock-out animals that do not express liver activin(s); antagonists and agonists of the liver activins; and other compounds that modulate liver activin gene expression activity to regulate cell growth and/or differentiation and to diagnose and treat abnormalities related thereto. In addition, the present invention encompasses methods and compositions for the regulation of cell growth and/or differentiation including, but not limited to, stimulating liver regeneration, bone growth and hematopoiesis.

Description

NUCLEOTIDE AND PROTEIN SEQUENCES OF LIVER ACTIVIN/INHIBIN AND METHODS BASED THEREON

1. INTRODUCTION The present invention relates to the discovery, identification and characterization of a novel subgroup of activins/inhibins, herein referred to as liver activins, that during vertebrate adult life are specifically expressed in the liver. The invention relates to the activin β_c and β_E subunits and other members of the liver activin subgroup. The invention encompasses liver activin nucleotides; liver activin genomic regulatory elements that regulate the expression of liver activins; host cell expression systems; liver activin proteins, fusion proteins, polypeptides and peptides; antibodies to liver activin protein, transgenic animals that express a liver activin; recombinant knock-out animals that do not express liver activin (s); antagonists and agonists of the liver activins; and other compounds that modulate liver activin gene expression activity to regulate cell growth and/or differentiation and to diagnose and treat abnormalities related thereto. In addition, the present invention encompasses methods and compositions for the regulation of cell growth and/or differentiation including, but not limited to, stimulating liver regeneration, bone growth and hematopoiesis.

2. BACKGROUND OF THE INVENTION

The activins and inhibins are dimeric proteins having structural and functional similarity to TGF-/3 superfamily members. (Vale, . , et al., in Peptide Growth Factors and Their Receptors (Part II) Sporn, M.B. and Roberts, A.B., eds., Springer-Verlag, Heidelberg, pp. 211-248). Like TGF-β superfamily members, activin/inhibin is synthesized as a larger precursor that consists of signal peptide, amino- terminal propeptide and mature growth factor domains. During the synthesis of activin (and other members of the TGF- superfamily) , two precursor chains associate to form a disulfide-bonded dimer with latent activity. The full length dimer is cleaved at an endoproteolytic cleavage motif, which separates the propeptide and mature growth factor domains of the disulfide bonded precursor chains. The propeptide dimer and the mature growth factor dimer typically remain non- covalently associated. It is believed that the propeptide dimer of activins and other TGF-3 superfamily members is necessary and sufficient to achieve latency. The mature growth factor domain of activin contains a cysteine residue motif that is conserved among all known members of the TGF-3 superfamily (Gray, A.M., et al., 1990, Science 247:1328-1330; Huylebroeck, D. , et al., 1990, Mol. Endocrinol. 4:1153-1165; Schwall, R.H., et al., 1988, Mol. Endocrinol. 2:1237-1242). Additionally, the mature growth factor domain of activins characterized thus far share -30% amino acid sequence homology with TGF-31.

Activins/Inhibins have been found to elicit diverse effects in modifying the growth and differentiation of various types of target cells. This complicated regulation of cellular growth is a characteristic of TGF- superfamily members. For a comprehensive review of TGF- effects, see Roberts, A.B., et al., 1990, in Handbook of Experimental Pharmacology, 95 (Part 1) Sporn, M.B. and Roberts, A.B., eds., Springer-Verlag, Heidelberg, pp. 419-472. The autocrine and paracrine activities of TGF- are believed to be regulated by different mechanisms. One regulatory strategy involves the temporal and spatial control of TGF-β gene expression. A second strategy involves the production and storage of TGF-3 as a latent complex that is activated only under certain physiological and pathological conditions, such as, for example, tissue morphogenesis and wound healing. See e.g.. Pircher, R. , et al., 1986, Biochem. Biophys. Res. Commun. 136:30-37; Miyazono, K. , et al., 1988, J. Biol. Chem. 263:6407-6415; Wakefield, L.M. , et al. , 1988, J. Biol. Chem. 263:7646-7654). It is likely that activin/inhibin activities are regulated by similar strategies. There is evidence to suggest that activin/inhibin activity is regulated by association with binding proteins, such as, for example, follistatin, which inhibits activin activity.

Activins and inhibins consist of structurally related polypeptide chains known as and β subunits. Presently, two unique but related β subunits have been identified ( β_A and β_B) , which are shared by both inhibin and activin. Consequently, there are three known activins (A, B, and AB) that consist of different combinations of β subunits (e.g., β_Aβ_A, β_Bβ_B , and β_Aβ_B) , and there are two known inhibins, each consisting of one β subunit and one α subunit ( β_A and β_B) (Mason, A. J. , et al., 1985, Nature 318:659-663; Mason, A.J., et al., 1986, Biochem. Biophys. Res. Commun. 135:957-964; Forage, R.G. , et al., 1986, Proc. Natl. Acad. Sci. U.S.A. 83:3091-3095; Esch, F. , et al., 1987, Mol. Endocrinol. 1:388- 396) .

Recently, a cDNA encoding a third activin β subunit, designated activin β_c has been isolated from a human liver cDNA library. Hotten et al., 1995, Biochem. Biophys. Res. Commun., 206:608-613. The product of this clone has yet to be isolated or recombinantly expressed, and its function is currently unknown. There is ~50% amino acid identity between the conceptualized mature part of human activin β_c and the corresponding regions of human activin β_A and activin β_B ; this is significantly more identity than exists between β_c and other known members of the TGF-β superfamily.

A fourth activin, designated activin β_D was recently isolated from a Xenopus liver cDNA library (Oda, S. , et al., 1995, Biochem. Biophys. Res. Commun., 210:581-588). Northern analysis indicates that β_O is expressed during early stages of Xenopus embryological development. Microinjection of a synthetic mRNA transcribed from the activin β_D cDNA into Xenopus embryo blastomeres led to the formation of a secondary body axis, whereas injection of mRNA into animal pole caps led to mesoderm induction. These activities are consistent with the known activities of activin β_A and β_B during development. It has not been determined whether the human activin β_c and Xenopus activin β_O subunits assemble into homodimers, heterodimers, or both. Their recent isolation suggests, however, that the number of combinatorial possibilities for both activin and inhibin may in fact be relatively large. Inhibin initially was identified as a gonadal peptide that regulated the release of gonadal hormones from the anterior pituitary-e.g. , follicular stimulating hormone or FSH (See e.g., deJong, F.H. , et al, 1985, Mol. Cell. Endocrinol. 42:95-103; Dubas, M. , et al., 1983, Mol. Cell. Endocrinol. 31:187-198; Channing, C.P., et al., 1985, Proc. Soc. Experim. Biol. Medicine 178, 339-361; Robertson, D.M. , Foulds, et al., 1985, Biochem. Biophys. Res. Commun. 126:220- 226; Rivier, J. , et al., 1985, Proc. 7th Intl. Cong. Endocrinol. 655:1141-1144; Miyamoto, K. , et al., 1985,

Biochem. Biophys. Res. Commun. 129:396-403; Rivier, J. , et al., 1985, Biochem. Biophys. Res. Commun. 133:120-127). The major gonadal sites of inhibin synthesis are Sertoli cells (testis) and granulosa cells (ovary) . In the male gonad, inhibin antagonizes spermatogonial DNA synthesis and decreases sperm number. Inhibin synthesis by Sertoli cells fluctuates during spermatogenesis, and inhibin binding to different populations of germ cells has been shown to change during the various stages of spermatogenesis. Consequently, interactions between Sertoli cells and germ cells may be mediated in a paracrine fashion by inhibin. Inhibin (with luteinizing hormone) also stimulates steroid hormone production by Leydig cells in vitro. In the female gonad, inhibin antagonizes oocyte meiosis, increases ovarian follicle number, and stimulates steroidogenesis by thecal cells. In mice of both genders, inhibin appears to be an important negative regulator of gonadal stromal cell proliferation (Matzuk, M.M. , et al., 1992, Nature 360:313- 319) , i.e. , inhibin-deficient mice develop mixed or incompletely differentiated gonadal stromal tumors that consist of granulosa cells (female mice) and Sertoli cells (male mice) . Finally, inhibin regulates the function of other tissues besides the gonad and the pituitary (See e.g. , Sakai, R. , et al., 1992, Biochem. Biophys. Res. Commun. 188:921-926. Inhibin A and B are expressed during development in the placenta and during adult life in the 5 spleen and nervous system. In the placenta inhibin may function as an important regulator of human chorionic gonadotropin.

Activin was also initially identified as a gonadal hypophysiotropic hormone, i.e. , activin A (β_Aβ_A) was

10 originally identified in porcine ovarian fluid as a molecule that stimulates FSH release by pituitary cells (ovarian follicle development and maturation is initiated by FSH secretion from the pituitary) (Vale, W. , et al., 1986, Nature 321:776-779; Ling, N., et al., 1986, Nature 321:779-782;

15 Ling, N. , et al, . 1986, Biochem. Biophys. Res. Commun.

138:1129-1137). Recent evidence indicates that activin A can act on oocytes, granulosa cells, and thecal cells, suggesting that the molecule has an intra-ovarian autocrine paracrine effect. Activin A and B have an extensive anatomical

20 distribution, and they have been implicated to play a role in many biological processes besides ovarian follicle maturation-e.g. , pituitary growth hormone and adrenocortical tropic hormone secretion (Sadatsuki, M. , et al., 1993, Hum. Reprod. 8:1392-1395), hypothalamic oxytocin secretion

25 (Sawchenko, P.E., et al., 1988, Nature 334:615-617), somatostatin induction (Woodruff, T. , et al., 1995, Annu. Rev. Physiol. 57:219-244, lymphocyte proliferation (Hedger, M.P., et al., 1989, Mol. Cell. Endocrinol. 61:133-138), erythropoiesis (EP Publ. No. 210,461, published February 4,

30 1987; Yu, et al., 1987, Nature 330:765-767; Eto, Y. , et al., 1987, Biochem. Biophys. Res. Commun. 142:1095-1103; Brox eyer, H.E., et al., 1988, Proc. Natl. Acad. Sci. U.S.A. 85:9052-9056; Murata et. al., 1988, Proc. Natl. Acad. Sci. USA., 85:2334-2438; Schwall, R.H. , et al., 1991, Meth.

35 Enzymol. 198:340-346; Burton et al., U.S. Patent No.

5,071,834, issued December 10, 1991; Shiozaki, M. , et al. , 1992, Biochem. Biophys. Res. Commun. 183:273-279; Nishimura, M. , et al., 1991, Biochem. Biophys. Res. Commun. 181:1042- 1047), cardiac morphogenesis (Olson, E. et al., 1996, Science 272:671-672), neuronal survival (Schubert, D. , et al., 1990, Nature 344:868-870; Hashimoto, M. , et al., 1990, Biochem. Biophys. Res. Commun. 173:193-200), megakaryocyte differentiation (Fugimoto, K. et al., 1991, Biochem Biophys. Res. Commun. 174:1163-1168), spermatogenesis (Mather, J.P. , et al., 1990, Endocrinology 127:3206-3214), bone formation (Oue, Y., et al., 1994, Bone 15:361-366; Inoue, S., et al., 1994, Calcif. Tissue Int. 55:395-397; Luyten, F.P., et al.,

1994, Exp. Cell Res. 210:224-229; Ogawa, Y. , et al., 1992, J. Biol. Chem. 267:14233-14237; Centrella, M. , et al., 1991, Mol. Cell. Biol. 11:250-258), pancreatic insulin secretion, patterning events in vertebrate embryos (Wakefield, et al., 1988, J. Biol. Chem. 263:7646-7654; Gentry, L.E., et al.,

1987, Molec. Cell Biol. 7:3418-3427; Ogawa, Y. , et al., 1992, J. Biol. Chem. 267:2325-2328), reproductive endocrinology (Pircher, R. , et al., 1984, Cancer Res. 44:5538-5543; Gentry, L.E., et al., 1988, Molec. Cell Biol. 8:4162-4168), somatic growth (Sadatsuki et al., 1993, Hum. Reprod. 8:1392-1395;

Wakefield, L.M. , et al., 1989, Growth Factors 1:203-218) and induction of mesoderm during early embryonic development (Smith, J.C., et al., 1990, Nature 345:729-731; van Den Eijnden-Van Raaij , A. J.M. , 1990, Nature 345:732-734; Green, J.B.A., et al., 1990, Nature 347:391-394; Thomsen, G. , et al., 1990, Cell 63:485-493; Sokol, S., and Melton, D.A. , 1991, Nature 351:409-411; Johansson, B.M. , and Wiles, M.V. ,

1995, Mol. Cell. Biol. 15:141-151). Additionally, previous studies have suggested that activin A may function as an autocrine inhibitor of hepatocyte DNA synthesis in vitro

(Yasuda, H. , et al., 1993, J. Clin. Invest. 92:1491-1496) and an inhibitor of liver regeneration in vivo (Kogure, K. et al., 1995, Gastroenterology 108:1136-1142; and Schwall, R. H., et al., 1993, Hepatology 18:347-356). These studies demonstrate that activin A, synthesis in hepatocytes can be induced by treatment of these cells with either epidermal growth factor or hepatocyte growth factor. From the data compiled thus far it is clear that activins and inhibins do not act alone in effecting cell and tissue physiology, but rather act in concert with other small regulatory molecules. In this regard, a recent study has shown that activin A and BMP-4 work synergistically in embryonic development to induce mammalian mesoderm formation and hematopoiesis. While the gonadal protein follistatin has been reported to be a high affinity activin-binding protein, it does not appear to be involved in activin signal transduction per se.

Erythropoiesis is the production of red cells that occurs continuously throughout the life of vertebrates to offset cell destruction, thereby enabling sufficient numbers of red blood cells to be available in the blood for proper oxygenation. The generation of red blood cells occurs in the bone marrow and is under the control of the hormone erythropoietin. The amount of erythropoietin in circulating plasma increases during states of reduced oxygen transport in blood plasma. This condition, termed hypoxia, may be caused by large losses of blood, resulting from, for example, hemorrhaging, anemia, radiation exposure that destroys red blood cells, reduction in oxygen intake resulting from high altitudes, or prolonged unconsciousness.

Dependent on the cause of anemia, many different types of pharmaceuticals have been employed to relieve anemic conditions, such as, for example, iron preparations (iron- deficiency related anemia) , vitamin B₁₂ and folic acid (malignant anemia) , and adrenocortical steroids such as corticoids (hemolytic anemia) . Steroid hormones are known to have powerful erythropoietic stimulating action and are regarded as effective medicines; however, such hormones exhibit strong side effects and are generally undesirable for administration over long time periods.

Erythropoietin has also recently been proposed as an effective drug for alleviating anemia. U.S. Pat. No.

4,703,008, issued October 27, 1987 describes the recombinant production of commercially viable quantities of erythropoietin.

Activins A and B have been found to have erythropoietic- stimulating activity. See EP Publ. No. 210,461, published February 4, 1987; Eto, et al, 1987, Biochem. et Biophys. Res. Commun. 142:1095-1103; Murata et. al., 1988, Proc. Natl. Acad. Sci. USA., 85:2334-2438, Yu et al., 1987, Nature 330:765-767, and for activin B see Burton et al., U.S. Patent No:5, 071,834, issued December 10, 1991. Activin binding sites have been identified on a number of activin-responsive cells, and chemical crosslinking experiments suggest that multiple binding molecules exist on the cell surface. Additionally, several activin receptors have been isolated and characterized recently. The first of these to be identified (Mathews, L.S., et al., 1991, Cell 65:973-982) was a 494 amino acid residue integral membrane protein with the characteristics of serine/threonine-specific protein kinase. Since then, a number of TGF-β type I and type II signaling receptors have been shown to bind activin (Attisano, L. , et al., Cell 75:671-680; Ebner, R. , et al., 1993, Science 262:900-902; ten Dijke, P., et al., Science, 264:101-104) .

3. SUMMARY OF THE INVENTION The present invention relates to the discovery, identification and characterization of nucleotides that encode a novel subgroup of activins/ inhibins, herein referred to as liver activins, and the discovery that these nucleotides correspond to mRNA that during adult life, is specifically expressed in the liver. More specifically, the invention relates to members of the liver activin subgroup that are defined, herein, as those activins that during adult life are expressed in the liver and demonstrate~65% identity at the amino acid level in substantial stretches of activins β_c and/or β_E .

Liver activins, like other activin subfamily members, occur naturally as homodimers or heterodimers and regulate cell growth and/or differentiation through binding to cell surface receptors. The amino acid sequence organization of liver activins is found in other activin family and TGF-β superfamily members, and while showing significant homology to activins β_A and β_B , liver activins are most closely related to activin β_D .

For clarity of discussion, the invention is described in the subsections below, by way of example, for activin β_c, depicted in FIG. 1 and activin β_E , depicted in FIG. 2 (murine) and FIG. 5 (human) . However, the principles may be analogously applied to other members of the liver activin subgroup.

The activin β_c and activin β_E mRNA transcripts, which are about 2 kb and 3.2 kb long, respectively, are specifically and highly expressed in adult vertebrate liver tissue. The murine activin β_c and β_E cDNAs, described herein, encode proteins of 352 and 350 amino acids, respectively (FIG. 1 and FIG. 2) . The human activin β_E encodes a protein of 350 amino acids. Activin β_c and β_E have a N-terminal signal peptide followed by a propeptide region and a mature growth factor region containing nine cysteine residues, the spacing of which is conserved among TGF-3 superfamily members.

The invention encompasses: (a) nucleotides that encode vertebrate liver activin, including activin β_c and β_E gene products; (b) nucleotides that encode portions of the liver activin that correspond to its functional domains, and the polypeptide products specified by such nucleotide sequences, including, but not limited to, the N-terminal signal peptide, the propeptide domain, and/or the mature growth factor domain of activin β_c and β_E ; (c) nucleotides that encode mutants of the liver activin in which all or a part of one of the domains is deleted or altered, and the polypeptide products specified by such nucleotide sequences, including, but not limited to, nonfunctional proteins in which all or a portion of the N-terminal signal peptide, the propeptide domain, endoproteolytic motif, or mature growth factor domain is deleted; and (d) nucleotides that encode fusion proteins containing the liver activin or one of its domains (e.g. , the mature growth factor region) fused to another polypeptide.

The invention also encompasses genomic fragments containing the murine activin β_c and β_E genes. Several genomic clones have been isolated and a map of the murine genomic region is shown in Figure 4A. Genomic nucleotide sequences that regulate the expression of the liver activin genes, such as those positioned between the murine activin β_c and β_E genes, are also encompassed by the present invention (FIG. 4B) .

The invention also encompasses agonists and antagonists of liver activins, including small molecules, large molecules, mutant liver activins that compete with native liver activin and antibodies; nucleotide sequences that can be used to inhibit the liver activin gene expression (e.g. , antisense and ribozyme molecules, and gene or regulatory sequence replacement constructs) or to enhance the liver activin gene expression (e.g. , expression constructs that place the liver activin gene under the control of a strong promoter system) .

In addition, the invention encompasses transgenic animals that overexpress a liver activin transgene. Such animals may be developed to serve as bioreactors for production of liver activins. The expression of liver activins via transgenic livestock has the advantage over protein production in recombinant bacteria and yeast in the ability to produce proteins which are correctly post- transtationally processed and modified. Transgenic animals may also be developed that do not express liver activin ("knock-out mice") .

In addition, the present invention encompasses methods and compositions for the diagnostic evaluation, typing and prognosis of cell growth and/or differentiation disorders, and for the identification of subjects having a predisposition to such conditions. For example, liver activin nucleic acids can be used as diagnostic hybridization probes or as primers for diagnostic PCR analysis to identify liver activin gene mutations, and polymorphisms, variations as well as regulatory defects in the liver activin gene. The present invention further provides for diagnostic kits that allow the practice of such methods. The liver activin protein, polypeptides, and derivatives (including fragments) disclosed herein may be used for the production of anti-liver activin antibodies which may be used diagnostically in immunoassays for the detection or measurement of liver activin protein in a patient sample. Anti-liver activin antibodies may be used, for example, for the diagnostic detection or measurement of liver activin protein in blood serum or biopsied cells and tissues.

In addition, the present invention encompasses methods and compositions for the regulation of cell growth and/or differentiation, including, but not limited to, stimulating liver regeneration, bone growth or hematopoiesis. Therapeutic compounds of the invention include but are not limited to liver activin proteins, polypeptides, and derivatives (including fragments) thereof; antibodies thereto; nucleic acids encoding the liver activin proteins, polypeptides, and derivatives; and liver activin antisense nucleic acids.

Further, the present invention also relates to methods for the use of liver activin genes and/or liver activin gene products to identify compounds which modulate liver activin gene expression and/or liver activin gene product activity, i.e., act as agonists or antagonists. Such compounds can be used as agents to control cell growth and/or differentiation, such as, for example, therapeutic agents that stimulate cell growth and differentiation, including liver regeneration, hematopoiesis and/or bone growth.

The present invention also relates to methods of production of the liver activin protein, polypeptides, derivatives and antibodies, such as for example, by recombinant means.

This invention is based, in part, on the surprising discovery, of a subgroup of activins that are specifically and highly expressed in adult vertebrate liver, the identification and cloning of liver activin ^_c and β__B cDNAs from libraries prepared from adult vertebrate liver mRNA, characterization of the nucleotide sequence of activin β__c and β__E , expressing activin β_c in a cell-free transcription and translation system and determining that β_c product stimulates liver regeneration, stimulates bone growth in a critical defect model, stimulates hematopoiesis and specifically binds the latent TGF-β binding protein-3 (LTBP-3) in vitro.

3.1 DEFINITIONS As used herein, the following terms, whether used in the singular or plural, will have the meanings indicated: Liver activin gene, nucleotides or coding sequences: means nucleotide sequences encoding liver activin/ inhibin β subunit protein, polypeptide or derivatives (including fragments) of liver activin/ inhibin β subunit protein or liver/activin inhibin β subunit fusion proteins. Liver activin nucleotide sequences encompass liver activin/ inhibin DNA, including genomic DNA (e.g. the β__c or β__E gene) or cDNA, or RNA.

Liver activin genomic fragment: means a genomic DNA fragment comprising nucleotide sequences positioned adjacent to, i.e.. upstream or downstream, of the liver activin coding regions. Such nucleotide sequences can contain regulatory elements, i.e. , promoters, enhances, etc., which regulate the expression of liver activin genes. Liver activin: means monomers or dimers containing liver activin/ inhibin β subunit protein. Polypeptides or derivatives (including fragments) of liver activin/ inhibin β subunit protein are referred to as liver activin polypeptides or liver activin derivatives. Fusions of liver activin/ inhibin β subunit proteins, or liver activin/ inhibin β subunit polypeptides or inhibin β subunit derivatives (including fragments) to an unrelated protein are referred to herein as liver activin fusion proteins. A functional liver activin refers to a protein which forms an active activin/ inhibin monomer, homodimer or heterodimer in vivo or in vitro and which is capable of regulating cellular growth and/or differentiation.

4. DESCRIPTION OF THE FIGURES

FIG. 1. The full length mouse activin β_c cDNA sequence. The nucleotide sequence of mouse activin β_c is shown, with the predicted amino acid sequence below the coding sequence. A typical signal peptide is predicted to include the first 20 amino acid residues. The position of four potential sites of N-linked glycosylation are denoted by a double underline. A cluster of basic residues (KHRVRRR, at amino acid residues 230-236) may serve as an endoproteolytic cleavage site(s) that defines the predicted propeptide and mature growth factor domains.

FIG. 2. The full length mouse activin β_E cDNA sequence. The nucleotide sequence of mouse activin β_E is shown, with the predicted amino acid sequence below the coding sequence. A typical signal peptide is predicted to include the first 21 amino acid residues. The position of a single potential site of N-linked glycosylation is denoted by a double underline. A cluster of basic residues (RARRR, amino acid residues 232- 236) may serve as an endoproteolytic cleavage site (or sites) that defines the predicted propeptide and mature growth factor domains.

FIG. 3. Comparison of the predicted mouse activin β_c and β_E amino acid sequences with other activin group members.

Alignment of the C-terminal acid sequence of mouse activin β_c and β_E (beginning at amino acid 236 in both instances) with corresponding regions of human activin β_A, human activin β_B , and Xenopus activin β_O . The nine invariant cysteines characteristic of the activins are shown in bold. Dashes denote gaps introduced to maximize the alignment. Species are indicated as follows: h, human; m, mouse; and x, Xenopus .

FIG. 4A. A map representing the genomic organization of the mouse activin β_c and mouse activin β_E genes. For both loci, the 5' and 3' untranslated regions are shown as black rectangles, sequences that encode the predicted propeptide domains are shown as open boxes, and sequences that encode the predicted mature growth factor regions are shown as patterned boxes. Partial restriction analysis and partial DNA sequence analysis of overlapping genomic clones indicates that the size of the intron separating activin β_c exon 1 and 2 is ~12-kbp. In contrast, DNA sequencing has shown that the size of the intron separating activin β_E exon 1 and 2 is 234- bp. FIG. 4B and 4C. Nucleotide sequence of the genomic region located between the murine activin β_c and β_E genes.

FIG. 4D, 4E, 4F and 4G. Sequence of the mouse activin genetic loci. The sequence includes nucleotide sequences upstream of the activin β_c gene, activin β_c/exon 1, partial sequence of activin β_c/intron 1, activin _c/exon 2, the activin β_E regulatory region, activin 3_E/exon 1, activin β_E/ intron 1, activin β_E/exon 2, and the sequence located 3' of the activin β_E gene.

FIG. 5. The full length human activin β_E cDNA sequence. The nucleotide sequence of human activin β_E is shown, with the predicted amino acid sequence below the coding sequence. A typical signal peptide is predicted to include the first 17 amino acid residues. The position of a single potential site of N-linked glycosylation is denoted by a double underline. A cluster of basic residues (RARRR, amino acid residues 232- 236 may serve as an endoproteolytic cleavage site (or sites) that defines the predicted propeptide and mature growth factor domains (an arrow separates the two predicted domains) . Conserved cysteines in the predicted mature growth factor region are shown in bold type.

FIG. 6. An adult mouse multiple tissue Northern blot (Clontech) as evaluated using specific activin _c and β_E CDNA probes (top and middle panels, respectively) . RNA samples were loaded as follows: lane 1, heart; lane 2, brain; lane 3, spleen; lane 4, lung; lane 5, liver; lane 6, skeletal muscle; lane 7, kidney; and lane 8, testis. RNA size was determined by the manufacturer using RNA standards (9.5, 7.5, 4.4, 2.4, and 1.35 kb) . Relative RNA loading was evaluated by hybridization with a glucose 6-dehydrogenase cDNA probe (bottom panel) .

FIG. 7. Expression of the full length mouse activin β_c plasmid expression construct in a cell-free transcription and translation system demonstrated by immunoprecipitation of β_c .

FIG. 8. Evidence that mouse activin β_c stimulates bone growth in a critical defect animal model. Microscopic focus of bone (arrows) that formed following direct in vivo gene transfer using a gene activated matrix.

FIG. 9. Evidence that mouse activin β_c promotes hematopoiesis of human bone marrow cultures in vitro.

5. DETAILED DESCRIPTION OF THE INVENTION Liver activins/inhibins, described for the first time herein, comprise an activin/ inhibin subgroup during vertebrate adult life are specifically and highly expressed in liver. Liver activins, like other activin group members, occur naturally as homodimers or heterodimers which regulate cell growth and/or differentiation through binding to cell surface signaling receptors. The amino acid sequence organization of liver activins is similar to that found in other activins and TGF-β superfamily members and consists of an N-terminal signal peptide that is followed by a propeptide region and a mature growth factor region containing nine cysteine residues, the spacing of which is conserved among TGF-β superfamily members. A endoproteolytic cleavage motif separates the propeptide and mature growth factor domains. While showing significant homology to activins β_A and β_B , liver activins are most closely related to activin β_O.

The invention encompasses the use of liver activin genes and gene products, both transcriptional and translational including, but not limited to, DNA, mRNA, antisense RNA, ribozymes and protein. The invention further encompasses the use of liver activin polypeptides, and derivatives; antibodies to liver activins (which can, for example, act as liver activin agonists or antagonists) ; and antagonists that inhibit activin activity or expression, or agonists that activate activin activity or increase its expression in the diagnosis, treatment, and/or regulation of cell growth and/or differentiation, including, but not limited to stimulating liver regeneration and/or bone growth in vertebrates, including humans. In addition, liver activin nucleotides and liver activin proteins are useful for the identification of compounds effective in regulating cell growth and/or differentiation . The invention also encompasses nucleotide sequences that regulate liver activin gene expression such as those found adjacent to liver activin genes in the genome. Such nucleotide sequences can be used, for example, in expression vectors to promote liver specific expression of either activin genes, or any gene of interest for which liver specific expression is desired.

In particular, the invention described in the subsections below encompasses liver activin proteins, polypeptides and derivatives thereof corresponding to functional domains of the liver activin (e.g. , N-terminal signal peptide, propeptide, and/or mature growth factor domains) ; mutated, truncated or deleted liver activins (e.g. a liver activin with one or more functional domains or portions thereof deleted, such as the propeptide domain, endoproteolytic cleavage motif and/or the mature growth factor domain) ; liver activin fusion proteins (e.g. a liver activin or a functional domain of a liver activin, such as the mature growth factor domain, fused to an unrelated protein or peptide such as an immunoglobulin constant region, i.e. , IgFc) ; nucleotide sequences encoding such products; and host cell expression systems that can produce such liver activin products. The invention also encompasses antibodies and anti-idiotypic antibodies (including Fab fragments) , antagonists and agonists of the liver activin, as well as compounds or nucleotide constructs that inhibit expression of the liver activin gene (transcription factor inhibitors, antisense and ribozyme molecules, or gene or regulatory sequence replacement constructs) , or promote expression of liver activin (e.g. , expression constructs in which liver activin coding sequences are operatively associated with expression control elements such as promoters, promoter/enhancers, etc.). The invention also relates to host cells and animals genetically engineered to express the liver activin (or mutants thereof) or to inhibit or "knockout" expression of the animal's endogenous liver activin. Transgenic animals overexpressing a liver activin transgene may be developed to serve as bioreactors for production of activin.

The liver activin proteins, polypeptides or derivatives; liver activin fusion proteins; liver activin nucleotide sequences; antibodies; antagonists; and agonists can be useful when detecting mutant liver activin or inappropriately expressed liver activin for the diagnosis of cell growth and differentiation disorders. The liver activin proteins, polypeptides or derivatives; liver activin fusion proteins; liver activin nucleotide sequences; host cell expression systems; antibodies; antagonists; agonists; and genetically engineered cells and animals can also be used to screen for drugs that regulate cell growth and differentiation and/or treatment of disorders related thereto.

Finally, the liver activin protein products (especially derivatives such as peptides corresponding to the mature growth factor domain) and fusion protein products (especially liver activin fusion proteins, i.e. , fusions of the liver activin or a domain of the liver activin, e.g. , the mature growth factor region to an IgFc) , antibodies and anti- idiotypic antibodies (including Fab fragments) , antagonists or agonists can all be used to regulate cell growth and/or differentiation and/or therapy of disorders related thereto. For example, the administration of an effective amount of an inactive liver activin (e.g. , one in which the endoproteolytic motif has been modified so as to prevent formation of polypeptide corresponding to the mature growth factor) or an anti-idiotypic antibody (or its Fab) that mimics a liver activin would bind to and occupy liver activin receptors, thereby reducing activin receptor availability to the active forms and prevent or reduce the signal cascade which ultimately impacts upon cellular growth and/or differentiation. Nucleotide constructs encoding such liver activin products can be used to genetically engineer host cells to express such liver activin products in vivo; these genetically engineered cells function as "bioreactors" in the body, delivering a continuous supply of liver activin protein, polypeptide or derivative that will bind to and occupy activin receptor sites. Nucleotide constructs encoding functional liver activin protein, polypeptide or derivative; mutant liver activins; and antisense and ribozyme molecules can all be used in "gene therapy" approaches for the modulation of liver activin expression and/or activity in the regulation of cell growth and differentiation and/or treatment of disorders related thereto. Studies described herein demonstrate that β_c product stimulates liver regeneration, bone growth in a critical defect animal model and hematopoiesis. Such studies indicate that the liver activins may be useful both as a cytokine for regulation of cell growth and differentiation and/or as a hormone for regulation of physiological process such as, for example, hematopoiesis and bone growth. Additionally, these studies demonstrate that β_c specifically binds the latent TGF-β binding protein-3 (LTBP-3) in vitro. Thus, the invention also encompasses pharmaceutical formulations and methods for regulating cell growth and differentiation and/or treating disorders related thereto. This invention is based, in part, on the surprising discovery of a subgroup of activins that during vertebrate adult life are specifically and highly expressed in liver. This discovery was made possible using synthesized degenerate oligonucleotide primers based on conserved coding sequences in TGF-β superfamily members, bone morphogenetic protein-2 and -4, to amplify mouse genomic DNA and cDNA. The products of these PCR reactions were cloned, labeled and used as probes to screen a mouse liver cDNA library. One of these probes hybridized to a series of overlapping cDNA clones that, in composite, encode a protein homologous to activin members of the TGF-β superfamily. This gene has been designated activin β_c . Northern blot analysis of multiple adult tissues indicates that activin β_c is highly and specifically expressed in the liver as a transcript of approximately 2.0 kb (See FIG. 6). The gene encoding activin β_c was cloned into an expression vector, and this construct was then expressed in a cell-free transcription and translation system.

While cloning the mouse activin β_c genomic locus, a genomic sequence located approximately 5.0 kb downstream of the activin β_c gene was observed to hybridize under low stringency conditions to the probe corresponding to activin β_c cDNA. This genomic sequence was subcloned, labeled and used as a probe to screen a mouse liver cDNA library under highly stringent conditions. Sequence analysis of a composite of overlapping cDNA clones obtained from this screen indicate that they correspond to a gene encoding a novel activin β subunit that shows strong sequence homology to activin β_c; this gene has been designated activin β_E . Northern blot analysis of multiple adult tissues indicates that activin β_E is highly and specifically expressed in the liver as a transcript of approximately 3.2 kb (See FIG. 6).

The mouse activin β_E cDNA was labeled and used as a probe to screen an adult human liver cDNA library. The probe hybridized to two overlapping cDNA clones. Sequence analysis revealed a predicted open reading frame of 1050 nucleotides coding for a protein with an estimated molecular weight of 39 kD and one potential N-linked glycosylation site. Sequence comparison analysis indicates that the protein is homologous to murine activin β_E . The complete nucleotide sequence and deduced amino acid sequence of the human activin β_E is depicted in FIG. 5.

Various aspects of the invention are described in greater detail in the subsections below.

5.1 THE LIVER ACTIVIN GENE

The liver activin nucleotide sequences of the invention include: (a) the DNA sequences shown in FIG. 1, FIG. 2 and FIG. 5 or contained in the cDNA clones pbC, pbE or pbHE within E_. coli strain DH5α as deposited with the American Type Culture Collection (ATCC) ; (b) nucleotide sequences that encode the amino acid sequences shown in FIG. 1, FIG. 2 and FIG. 5, or the liver activin amino acid sequences encoded by the cDNA clones pbC, pbE or pbHE as deposited with the ATCC; (c) any nucleotide sequence that hybridizes to the complement of the DNA sequence shown in FIG. 1, FIG. 2 and FIG. 5 or contained in the cDNA clones pbC, pbE or pbHE as deposited with the ATCC under highly stringent conditions, e.g. , hybridization to filter-bound DNA in 0.5 M NaHP0₄, 7% sodium dodecyl sulfate (SDS) , 1 mM EDTA at 65°C, and washing in O.lxSSC/0.1% SDS at 68°C (Ausubel F.M. et al., eds., 1989, Current Protocols in Molecular Biology, Vol. I, Green Publishing Associates, Inc., and John Wiley & sons, Inc., New York, at p. 2.10.3) and encodes a functionally equivalent gene product; and (d) any nucleotide sequence that hybridizes to the complement of the DNA sequences that encode the amino acid sequence shown in FIG. 1, FIG. 2 and FIG. 5 or contained in cDNA clones pbC, pbE or pbHE as deposited with the ATCC under less stringent conditions, such as moderately stringent conditions, e.g. , washing in 0.2xSSC/0.1% SDS at 42 °C (Ausubel et al., 1989, supra) , yet which still encodes a functionally equivalent liver activin gene product. Functional equivalents of liver activin include naturally occurring liver activins present in the same or other species, and mutant liver activins, whether naturally occurring or engineered. The invention also includes degenerate variants of sequences (a) through (d) .

The invention also includes nucleic acid molecules, preferably DNA molecules, that hybridize to, and are therefore the complements of, the nucleotide sequences (a) through (d) in the preceding paragraph. Such hybridization conditions may be highly stringent or less highly stringent, as described above. In instances wherein the nucleic acid molecules are deoxyoligonucleotides ("oligos") , highly stringent conditions may refer, e.g. , to washing in 6xSSC/0.05% sodium pyrophosphate at 37 °C (for 14-base oligos) , 48°C (for 17-base oligos) , 55°C (for 20-base oligos) , and 60°C (for 23-base oligos) . These nucleic acid molecules may encode or act as liver activin antisense molecules, useful, for example, in gene regulation (for and/or as antisense primers in amplification reactions of liver activin nucleic acid sequences) . With respect to liver activin regulation, such techniques can be used to regulate, for example, cell growth and/or differentiation, including, but not limited to, stimulation of liver regeneration and/or bone growth. Further, such sequences may be used as part of ribozyme and/or triple helix sequences, also useful for liver activin gene regulation. Still further, such molecules may be used as components of diagnostic methods whereby, for example, the presence of a particular liver activin allele responsible for causing a cell growth and/or differentiation disorder may be detected.

In addition to the liver activin nucleotide sequences described above, full length liver activin cDNA or gene sequences present in the same species, and/or homologs of the liver activin gene present in other species, can be identified and readily isolated, without undue experimentation, by molecular biological techniques well known in the art. The identification of homologs of liver activins in related species can be useful for developing animal model systems more closely related to humans for purposes of drug discovery. cDNA libraries or genomic DNA libraries derived from the organism of interest can be screened by hybridization using the nucleotides described herein as hybridization or amplification probes. For example, the liver activin nucleotide sequences described above may be used as probes to identify and isolate additional human liver activin homologs from genomic DNA isolated from human cells and/or from cDNA generated from human cells or tissue known, or suspected to express the liver activin gene, such as for example, adult liver tissue, or ATCC human liver cell lines 7004, 7120, 7128, 7137, 7154, 7274, 7284, 7290, 7297, 7503, 7705, 7821, 7837, 7859 or 7864. Furthermore, genes at other genetic loci within the genome that encode proteins which have extensive homology to one or more domains of the liver activin gene product can also be identified via similar techniques. In the case of cDNA libraries, such screening techniques can identify clones derived from alternatively spliced transcripts in the same or different species.

Screening can be performed by filter hybridization using duplicate filters. The labeled probe can contain at least 15-30 base pairs of the liver activin nucleotide sequence, as shown in FIG. 1, FIG. 2 and FIG. 5. The hybridization washing conditions used should be of a lower stringency when the cDNA library is derived from an organism different from the type of organism from which the labeled sequence was derived. With respect to the cloning of a human liver activin homolog, using murine liver activin probes, for example, hybridization can, for example, be performed at 65°C overnight in Church's buffer (7% SDS, 250 mM NaHP0₄ , 2μM EDTA, 1% BSA). Washes can be done with 2XSSC, 0.1% SDS at 65°C and then at 0.1XSSC, 0.1% SDS at 65 °C.

Low stringency conditions are well known to those of skill in the art, and will vary predictably depending on the specific organisms from which the library and the labeled sequences are derived. For guidance regarding such conditions see, for example, Sambrook et al., 1989, Molecular Cloning, A Laboratory Manual, Cold Springs Harbor Press, N.Y.; and Ausubel et al., 1989, Current Protocols in Molecular Biology, Green Publishing Associates and Wiley Interscience, N.Y.

Alternatively, the labeled liver activin nucleotide probe may be used to screen a genomic library derived from the organism of interest, again, using appropriately stringent conditions. The identification and characterization of human genomic clones is helpful for designing diagnostic tests and clinical protocols for regulating cell growth and/or differentiation and/or treating cell growth and/or differentiation disorders in human patients. For example, sequences derived from regions adjacent to the intron/exon boundaries of the human gene can be used to design primers for use in amplification assays to detect mutations within the exons, introns, splice sites

(e.g. splice acceptor and/or donor sites), etc., that can be used in diagnostics.

Further, other liver activins and other gene homologs may be isolated from nucleic acids of the organism of interest by performing PCR using two degenerate oligonucleotide primer pools designed on the basis of amino acid sequences within the liver activin gene product disclosed herein. Such primers may be used to amplify sequences of interest from any RNA or DNA source, preferably a cDNA library. PCR can be carried out, e.g. , by use of a Perkin-Elmer Cetus thermal cycler and Taq polymerase (Gene Amp^") . The DNA being amplified can include mRNA or cDNA or genomic DNA prepared from any vertebrate species. One can choose to synthesize several different degenerate primers for use in the PCR reactions. It is also possible to vary the stringency of hybridization conditions used in priming the PCR reactions, to allow for greater or lesser degrees of nucleotide sequence homology between the liver activin gene being cloned and the known liver activin gene. Other means for primer dependent amplification of nucleic acids are known to those of skill in the art and can be used. The PCR product also may be subcloned, sequenced, and used to probe a Northern blot so as to ensure that the amplified sequences represent the sequences of a liver activin gene. The PCR fragment may then be used to isolate a full length cDNA clone by a variety of methods. For example, the amplified fragment may be labeled and used to screen a cDNA library, such as a bacteriophage cDNA library. Alternatively, the labeled fragment may be used to isolate genomic clones via the screening of a genomic library. PCR technology may also be utilized to isolate full length cDNA sequences. For example, RNA may be isolated, following standard procedures, from an appropriate cellular or tissue source (i.e. , one known, or suspected, to express the liver activin gene, such as, for example, embryonic tissue, adult liver tissue, or ATCC human liver cell lines 7004, 7120, 7128, 7137, 7154, 7274, 7284, 7290, 7297, 7503, 7705, 7821, 7837, 7859 or 7864). A reverse transcription reaction may be performed on the RNA using an oligonucleotide primer specific for the most 5 ' end of the amplified fragment for the priming of first strand synthesis. The resulting RNA/DNA hybrid may then be "tailed" with guanine using a standard terminal transferase reaction, the hybrid may be digested with RNAase H, and second strand synthesis may then be primed with a poly-C primer. Thus, cDNA sequences upstream of the amplified fragment may easily be isolated. For a review of cloning strategies which may be used, see e.g. , Sambrook et al., 1989, supra .

Alternatively, the liver activin gene of the present invention may be isolated through an exon trapping system, using genomic DNA (Nehls et al., 1994, Oncogene 9(8):2169- 2175; Verna et al., 1993, Nucleic Acids Res. 21 (22) :5198:5202; Auch et al., 1990, Nucleic Acids Res. 18(22) :6743-6744) .

The liver activin gene sequences may additionally be used to isolate mutant liver activin gene alleles. Such mutant alleles may be isolated, for example, but not by way of limitation, from individuals either known or proposed to have a genotype which contributes to cell growth and/or differentiation disorders. Mutant alleles and mutant allele products may then be utilized in the therapeutic and diagnostic systems described below. Additionally, such liver activin gene sequences can be used to detect liver activin gene regulatory (e.g. , promoter or promoter/enhancer) defects which can affect cell growth and/or differentiation.

A cDNA of a mutant liver activin gene may be isolated, for example, by using PCR, a technique which is well known to those of skill in the art. In this case, the first cDNA strand may be synthesized by hybridizing an oligo-dT oligonucleotide to mRNA isolated from liver tissue in an individual putatively carrying the mutant liver activin allele, and by extending the new strand with reverse transcriptase. The second strand of the cDNA is then synthesized using an oligonucleotide that hybridizes specifically to the 5' end of the normal gene. Using these two primers, the product is then amplified via PCR, cloned into a suitable vector, and subjected to DNA sequence analysis through methods well known to those of skill in the art. By comparing the DNA sequence of the mutant liver activin allele to that of the normal liver activin allele, the mutation (s) responsible for the loss or alteration of function of the mutant liver activin gene product can be ascertained. Alternative hybridization or amplification assays for detecting abnormalities in the amplified liver activin gene fragment, include, but are not limited to, Southern analysis, single stranded conformational polymorphism analysis (SSCP) , denaturing gradient gel analysis, and PCR analysis.

Alternatively, a genomic library can be constructed using DNA obtained from an individual suspected of or known to carry the mutant liver activin allele, or a cDNA library can be constructed using RNA from liver or other tissue known, or suspected, to express the mutant liver activin allele. The normal liver activin gene or any suitable fragment thereof may then be labeled and used as a probe to identify the corresponding mutant liver activin allele in such libraries. Clones containing the mutant liver activin gene sequences may then be purified and subjected to sequence analysis according to methods well known to those of skill in the art .

Additionally, an expression library can be constructed utilizing cDNA synthesized from, for example, RNA isolated from embryonic tissue or adult liver tissue known, or suspected, to express a mutant liver activin allele in an individual suspected of or known to carry such a mutant allele. In this manner, gene products made by the putatively mutant tissue may be expressed and screened using standard antibody screening techniques in conjunction with antibodies raised against the normal liver activin gene product, as described, below, in Section 5.3. (For screening techniques, see, for example, Harlow, E. and Lane, eds., 1988, "Antibodies: A Laboratory Manual", Cold Spring Harbor Press, Cold Spring Harbor.) In cases where a liver activin mutation results in an expressed gene product with altered function (e.g. , as a result of a missense or a frameshift mutation) , a polyclonal set of antibodies to liver activin are likely to cross-react with the mutant liver activin gene product. Library clones detected via their reaction with such labeled antibodies can be purified and subjected to sequence analysis according to methods well known to those of skill in the art. The invention also encompasses nucleotide sequences that encode mutant liver activin proteins, polypeptides, derivatives (including peptide fragments) of the liver activin, truncated liver activins, and liver activin fusion proteins. These include, but are not limited to, truncated nucleotide sequences in which all or part of the nucleotides corresponding to the 3¹ untranslated region are deleted; nucleotide sequences encoding the mutant liver activins described in section 5.2 infra; polypeptides or peptides corresponding to the N-terminal signal peptide domain, propeptide domain, and mature growth factor domain of the liver activin or portions of these domains; and truncated liver activins in which one or two of the domains is deleted, e.g. , a mature growth factor domain lacking the N-terminal signal peptide and/or propeptide domain, or a truncated, nonfunctional liver activin lacking all or a portion of the endoproteolytic cleavage motif and/or N-terminal signal peptide domain, propeptide domain, and/or mature growth factor domain. Nucleotides encoding fusion proteins may include, but are not limited to, full length liver activin, truncated liver activin, polypeptides or derivatives (including fragments) of liver activin fused to an unrelated protein or peptide, such as, for example, a mature growth factor domain sequence; an Ig Fc domain which increases the stability and half life of the resulting fusion protein (e.g. , liver activin-Ig) in the bloodstream; or an enzyme, fluorescent protein, or luminescent protein which can be used as a marker.

The invention also encompasses (a) DNA vectors that contain any of the foregoing liver activin coding sequences and/or their complements (i.e. , antisense) ; (b) DNA expression vectors that contain any of the foregoing liver activin coding sequences operatively associated with a regulatory element that directs the expression of the coding sequences; and (c) genetically engineered host cells that contain any of the foregoing liver activin coding sequences operatively associated with a regulatory element that directs the expression of the coding sequences in the host cell. As used herein, regulatory elements include but are not limited to inducible and non-inducible promoters, enhancers, operators and other elements known to those skilled in the art that drive and regulate expression. Such regulatory elements include, but are not limited to, the cytomegalovirus hCMV immediate early gene, the early or late promoters of SV40 adenovirus, the lac system, the trp system, the TAC system, the TRC system, the major operator and promoter regions of phage A, the control regions of fd coat protein, the promoter for 3-phosphoglycerate kinase, the promoters of acid phosphatase, and the promoters of the yeast α-mating factors.

In addition, the invention encompasses genomic nucleotide sequences found adjacent to the liver activin genes. Such genomic nucleotide sequences include those containing regulatory elements that act to drive and regulate expression of liver activins. The regulatory elements may be identified using routine techniques well known to those skilled in the art. For example, the genomic nucleotide sequences located upstream of the liver activin genes can be cloned adjacent to a reporter gene such as a CAT gene. The transcriptional activity of the nucleotide sequences may be determined by assaying for reporter gene activity. To more precisely define the regulatory elements, deletion mutants can be generated and tested for reporter gene activity.

5.1.1 THE ACTIVIN ft. GENE The cDNA sequence (SEQ. ID. No. 1) and deduced amino acid sequence (SEQ. ID. No. 2) of murine activin β_c are shown in FIG. 1. The murine activin β_c signal sequence extends from amino acid residue 1 to about 20 of FIG. 1. A cluster of basic residues (KHRVRRR, amino acid 230-236) may serve as an endoproteolytic cleavage site that defines the predicted propeptide and mature growth factor domains. The propeptide domain consists of approximately 216 residues and extends from about amino acid residue 21 to about 236 of FIG. 1; and the mature growth factor domain consists of approximately 116 residues and extends from about amino acid residue 237 to about 352 of FIG. 1. Sequences derived from the murine activin β_c cDNA clone were used to screen mouse genomic DNA and liver cDNA libraries under low stringency conditions and to clone the gene corresponding to β_E , as described in the examples, infra. 5.1.2 THE ACTIVIN β_v GENE

The cDNA sequence (SEQ. ID. No. 3) and deduced amino acid sequence (SEQ. ID. No. 4) of murine activin β_E are shown in FIG. 2. The murine activin β_E signal sequence extends from amino acid residue 1 to about 21 of FIG. 2. A cluster of basic residues (RARRR, amino acid residues 232-236) may serve as an endoproteolytic cleavage site (or sites) that defines the predicted propeptide and mature growth factor domains. The propeptide domain consists of approximately 215 residues and extends from about amino acid residue 22 to about 236 of FIG. 2; and the mature growth factor domain of activin β_E subunit consists of approximately 113 residues and extends from about amino acid residue 237 to about 349 of FIG. 2.

5.1.3 THE HUMAN ACTIVIN 6_r GENE

The murine activin β_E cDNA clone was used to screen a human liver cDNA library under low stringency conditions and the cDNA sequence (SEQ. ID. NO. 5) and deduced amino acid sequence (SEQ. ID. NO. 6) of an isolated human activin β_E are shown in FIG. 5. Sequence analysis revealed a predicted open reading frame of 1050 nucleotides coding for a protein with an estimated molecular weight of 39 kD. The putative methionine start codon is located at nucleotide 77. A canonical CXCX sequence, preceding the stop translation signal (TAG) was also identified.

The human activin β_E subunit is organized like other TGF- β superfamily members. The predicted initiation Met residue is followed by a characteristic signal peptide of -20 amino acids. Downstream of the signal peptide, predicted propeptide and mature growth factor domains were identified.

The domains were separated by a cluster of basic amino acids, which create one or more putative endoproteolytic cleavage sites. If these cleavage site(s) are in fact used, then the propeptide and mature growth factor domains of the human activin β_E subunit will consist of -215 and 114 amino acids, respectively. The predicted carboxy-terminal amino acid sequences of mouse and human activin β_E are highly homologous, with >95% amino acid identity over the mature growth factor region. Both coding sequences contain a total of nine cysteine residues. We found that the spacing of the cysteines and the presence of other conserved amino acids is most characteristic of the activins and, to a lesser extent, TGF-β itself. In tabulating percenacid sequence identity within the mature growth factor region, human activin β_E showed 45% identity with human activin β_A, 47% identity with human activin β_B , 64% identity with human activin β_c , and 62% identity with Xenopus activin β_D . In contrast, only 20-40% amino acid sequence identity was found when human activin β_E was compared with other members of the TGF-β superfamily. To our knowledge, this is the first report of the isolation of activin β_E from human.

5.1.4 THE MOUSE ACTIVIN GENETIC LOCUS

A genetic map of the mouse activin genetic locus is shown in FIG. 4A. The nucleotide sequence of the murine activin genetic locus derived from a genomic mouse clone is shown in FIG. 4C. Our studies suggest that in certain respects activin β_c and β_E are more like one another than other known activins: (a) , these genes are physically linked, with only 5-kbp of genomic DNA separating the 3' end of activin β_c from the 5' end of activin β_E ; (b) , the organization of both genetic loci is similar, consisting of two coding exons separated by a single intervening sequence; (c) , the predicted carboxy-terminal sequences of both activins are highly homologous, with 62% amino acid identity over the mature growth factor region; (d) , both genes appear to be specifically and highly expressed in adult mouse liver and (e) the results of several in vitro assays show that homodimers of activin β_c and ^β _E stimulate the proliferation of HepG2 cells. When taken together, our data suggest that the activin β_c and β_E subunit genes may have evolved from a common ancestor that underwent a tandem duplication. These structural and functional data differentiate activin β_c and β_E subunits from other known activins, suggesting that these subunits represent a unique subgroup of TGF-β-like molecules.

5.2 LIVER ACTIVIN PROTEINS AND POLYPEPTIDES

Liver activin protein, polypeptides and derivatives (including fragments) ; mutated, truncated or deleted forms of the liver activin; and/or liver activin fusion proteins can be prepared for a variety of uses. By way of example but not limitation, these include the generation of antibodies, which can be used in diagnostic assays; and as pharmaceutical reagents useful in the regulation of cell growth and/or differentiation and/or treatment of disorders related thereto. The sequence of liver activins β_c and β_E begins with a methionine in a DNA sequence context consistent with a translation initiation site, followed by a typical hydrophobic signal sequence of peptide secretion. The predicted propeptide and mature growth factor domains of mouse activin β_c and β_E are approximately 216 and 115 amino acids and 215 and 113 amino acids, respectively. The predicted propeptide and mature growth factor domains of human activin β_E are approximately 215 and 114 amino acids. Like all known TGF-β superfamily members, except for MIS, activin β_c and β_E have a cluster of basic residues approximately 120 amino acids from the C-terminus, which form an endoproteolytic cleavage site at the junction of the propeptide domain and mature growth factor domain. (Roberts B, Sporn MB 1900, The Transforming Growth Factor-βs in Handbook of Experimental Pharmacology, 95 (Part I) Sporn MB, Roberts AB eds., Springer-Verlag, Heidelberg, pp. 419-472; Vale W, Hsueth A, River J, Yu J 1990 in Peptide Growth Factors and Their Receptors, 95 (Part II) Sporn MB, Roberts AB eds, Springer-Verlag, Heidelberg, pp. 211-248; and Centrella et al., 1994, Endocr Rev 15:27-38). An alignment between the mature growth factor domain of liver activin β_c and _E and other activins and other TGF-β superfamily members is shown in FIG. 3. The liver activin mature growth factor domain contains nine cysteine residues which demonstrate a pattern of spacing that is conserved among TGF-β superfamily members. While showing significant homology to activins β_A and β_B , liver activins β_c and β_E are most closely related to activin β_D .

FIG. 1 shows the amino acid sequence of the murine liver activin β_c subunit. The signal sequence extends from amino acid 1 to 20. A cluster of basic residues (KHRVRRR, amino acid 230-236) may serve as an endoproteolytic cleavage site(s) that defines the predicted propeptide and mature growth factor domains. The propeptide domain consists of approximately 216 residues and extends from about amino acid residue 21 to about 236. The mature growth factor domain consists of approximately 115 residues and extends from about amino acid residue 237 to about 352 of FIG. 1. FIG. 2 shows the amino acid sequence of the murine liver activin β_E protein. The signal sequence extends from amino acid residue 1 to 21. A cluster of basic residues (RARRR, amino acid residues 232-236) may serve as an endoproteolytic cleavage site (or sites) that defines the predicted propeptide and mature growth factor domains. The propeptide domain consists of approximately 215 residues and extends from about amino acid residue 22 to about 236. The mature growth factor domain consists of approximately 113 residues and extends from about amino acid residue 237 to about 349.

FIG. 5 shows the amino acid sequence of human liver activin β_E subunit. The human activin β_E subunit is organized like other TGF- super family members. The predicted initiator MET residue is followed by a characteristic signal peptide of - 20 amino acids. A cluster of basic amino acids were identified which create endoproteolytic cleavage site(s) that define the predicted propeptide and mature growth factor domains. Once cleaved, the propeptide and mature growth factor domains of human activin β_E will consist of approximately 215 and 114 amino acids, respectively. Potential N-linked glycosylation sites (i.e. , amino acid sequence motif N-X-S or N-X-T) are found in the propeptide domains of both murine β_c and β_E liver activin subunits and human β_E liver activin subunits. Four potential N-linked glycosylation sites can be identified in the murine β_c sequence shown in FIG. 1 (see tripeptide motifs starting at amino acid residues 111, 143, 161 and 173), whereas one potential N-linked glycosylation site can be identified in the murine β_E sequence shown in FIG. 2 (see the tripeptide motif starting at amino acid residue 198) .

The liver activin amino acid sequences of the invention include the amino acid sequence shown in FIG. 1 (SEQ. ID. No. 2), FIG. 2 (SEQ. ID. No. 4), and FIG. 5 (SEQ. ID. NO.: 6) and the amino acid sequence encoded by cDNA clone pbC as deposited with the ATCC, cDNA clone pbE as deposited with the ATCC, and cDNA clone pbHE as deposited with the ATCC. Further, other murine and human liver activins and liver activin homologs are encompassed by the invention. In fact, any liver activin protein encoded by the liver activin nucleotide sequences described in Section 5.1, above, is within the scope of the invention. Such animals may be produced for use as "bioreactors" for production of liver activins.

The invention also encompasses proteins that are functionally equivalent to the liver activin encoded by the nucleotide sequences described in Section 5.1, as judged by any of a number of criteria, including but not limited to (a) antigenicity, i.e. , the ability to bind to an anti-liver activin antibody; (b) immunogenicity, i.e. , the ability to generate an antibody which is capable of binding a liver activin protein or polypeptide; (c) the ability to bind (or compete with liver activin for binding) to a substrate for liver activin; (d) ability to ultimerize with liver activin; and (e) the ability to bind a liver activin receptor, the binding affinity for an activin receptor, the resulting biological effect of binding to an activin receptor, e.g. , signal transduction, a change in cellular metabolism (e.g. , ion flux, tyrosine phosphorylation) or change in phenotype when the liver activin equivalent is present in an appropriate cell type (such as the stimulation or inhibition of cellular growth and/or differentiation) . Such functionally equivalent liver activin proteins include but are not limited to additions or substitutions of amino acid residues within the amino acid sequence encoded by the liver activin nucleotide sequences described above, in Section 5.1, but which result in a silent change, thus producing a functionally equivalent gene product. Amino acid substitutions may be made on the basis of similarity in polarity, charge, solubility, hydrophobicity, hydrophilicity, and/or the amphipathic nature of the residues involved. For example, nonpolar (hydrophobic) amino acids include alanine, leucine, isoleucine, valine, proline, phenylalanine, tryptophan, and methionine; polar neutral amino acids include glycine, serine, threonine, cysteine, tyrosine, asparagine, and glutamine; positively charged (basic) amino acids include arginine, lysine, and histidine; and negatively charged (acidic) amino acids include aspartic acid and glutamic acid. Included within the scope of the invention are liver activin proteins, polypeptides, and derivatives (including fragments) which are differentially modified during or after translation, e.g. , by glycosylation, acetylation, phosphorylation, amidation, derivatization by known protecting/blocking groups, proteolytic cleavage, linkage to an antibody molecule or other cellular ligand, etc. Any of numerous chemical modifications may be carried out by known techniques, including, but not limited to, specific chemical cleavage by cyanogen bromide, trypsin, chymotrypsin, papain, V8 protease, NaBH₄ ; acetylation, formylation, oxidation, reduction; metabolic synthesis in the presence of tunicamycin; etc. In a specific embodiment, the compositions of the invention are conjugated to other molecules to increase their water-solubility (e.g. , polyethylene glycol) , half-life, or ability to bind targeted tissue (e.g. , bisphosphonates and fluorochromes to target the proteins to bony sites) .

Furthermore, nonclassical amino acids or chemical amino acid analogs can be introduced as a substitution or addition into the liver activin sequence. Non-classical amino acids include but are not limited to the D-isomers of the common amino acids, α-amino isobutyric acid; 4-aminobutyric acid, Abu; 2-amino butyric acid, γ-Abu; e-Ahx, 6-amino hexanoic acid; Aib, 2-amino isobutyric acid; 3-amino propionic acid; ornithine; norleucine; norvaline; hydroxyproline; sarcosine; citrulline; cysteic acid; t-butylglycine; t-butylalanine; phenylglycine; cyclohexylalanine; β-alanine; fluoro-amino acids; "designer" amino acids such as -methyl amino acids; Cα-methyl amino acids; Nα-methyl amino acids; and amino acid analogs in general. Furthermore, the amino acid can be D (dextrorotary) or L (levorotary) .

While random mutations can be applied to liver activin DNA (using techniques well known to those skilled in the art) and the resulting mutant liver activins tested for activity, site-directed mutations of the liver activin coding sequence can be engineered (again using techniques well known to those skilled in the art) to generate mutant liver activins with higher or lower binding affinity for binding proteins, and/or increased function, e.g. , higher binding affinity for liver activin receptors, and/or greater signalling capacity; or decreased function, e.g. , lower binding affinity for liver activin receptors, and/or decreased signalling capacity.

For example, the alignment of liver activins and other activin group members is shown in FIG. 3. Mutant liver activins can be engineered so that regions of identity are maintained, whereas the variable residues are altered, e.g. , by deletion or insertion of an amino acid residue (s) or by substitution of one or more different amino acid residues. Conservative alterations at the variable positions can also be engineered in order to produce a mutant liver activin that retains function, e.g. , liver activin receptor binding affinities or signalling capability or both. Non- conservative changes can be engineered at these variable positions to alter function, e.g. , liver activin binding affinities or signalling capability, or both. Alternatively, where alteration of function is desired, deletion or non- conservative alterations of the conserved regions can be engineered. For example, deletion or non-conservative alterations (substitutions or insertions) of amino acid residues 21-236 (FIG. 1) of murine liver activin β_c, amino acid residues 22-236 (FIG. 2) of murine liver activin β_E, amino acid residues 17 to about 236 (FIG. 5) of human liver activin β_E or amino acid residues 237-352 (FIG. 1) of murine liver activin β_c, amino acid residues 237-349 (FIG. 2) of murine activin β_E or amino acid residues 237-350 (FIG. 5) of human liver activin β_E can be engineered to produce a mutant liver activin that is unable to form dimers with and/or β subunits, and thus is unable to achieve a functional structural conformation, or that binds liver activin receptor but is signalling-incompetent. Non-conservative alterations to the cysteine residues in the mature growth factor domain shown in FIG. 3 can be engineered to produce mutant liver activins with altered structure and thus altered binding affinity for liver activin receptor.

Other mutations to the liver activin coding sequence can be made to generate liver activins that are better suited for expression, scale up, etc. in the host cells chosen. For example, cysteine residues can be deleted or substituted with another amino acid in order to eliminate disulfide bridges; or N-linked glycosylation sites can be altered or eliminated to achieve expression of a homogeneous product that is more easily recovered and purified from yeast hosts which are known to hyperglycosylate N-linked sites. To this end, a variety of amino acid substitutions at one or both of the first or third amino acid positions of any one or more of the glycosylation recognition sequences which occur in the liver activin propeptide domain (N-X-S or N-X-T) will prevent glycosylation of the propeptide at the modified tripeptide sequence. See, e.g. , Miyajima et al., 1986, EMBO J. 5(6) .1193-1197.

Peptides corresponding to one or more domains of liver activin (e.g. , N-terminal signal peptide, propeptide, or mature growth factor) ; truncated or deleted liver activins (e.g. , molecules in which signal and/or propeptide domain is deleted) ; and fusion proteins in which the full length liver activin protein, polypeptide or derivative (including fragment) , or truncated liver activin is fused to an unrelated protein are also within the scope of the invention. Altered activins can be designed on the basis of the liver activin nucleotide and liver activin amino acid sequences disclosed in this Section and in Section 5.1, above. The fusion protein may also be engineered to contain a cleavage site located between a liver activin sequence and the non- liver activin protein sequence, so that the liver activin protein may be more conveniently cleaved away from the non- liver activin moiety. Such fusion proteins include, but are not limited to, IgFc fusions, which stabilize the liver activin protein or peptide and prolong half-life in vivo; or fusions to an enzyme, fluorescent protein, or luminescent protein which provide a marker function.

While the liver activin proteins, polypeptides and derivatives (including fragments) can be chemically synthesized (e.g. , see Creighton, 1983, Proteins: Structures and Molecular Principles, W.H. Freeman & Co., N.Y.), large polypeptides derived from the liver activin, and the full length liver activin itself, may advantageously be produced by recombinant DNA technology using techniques well known in the art for expressing gene sequences and/or nucleic acid coding sequences. Such methods can be used to construct expression vectors containing the liver activin nucleotide sequences described in Section 5.1 and appropriate transcriptional and translational control signals. These methods include, for example, in vitro recombinant DNA techniques, synthetic techniques, and in vivo genetic recombination. See, for example, the techniques described in Sambrook et al., 1989, supra ; Ausubel et al., 1989, supra ; Caruthers et al., 1980, Nuc. Acids Res. Symp. Ser. 7:215-233; Crea and Horn, 1980, Nuc. Acids Res. 9(10):2331; Matteucci and Caruthers, 1980, Tetrahedron Letters 21:719; and Chow and Kempe, 1981, Nuc. Acids Res. 9 (12) : 2807-2817. Alternatively, RNA capable of encoding liver activin nucleotide sequences may be chemically synthesized using, for example, synthesizers. See, for example, the techniques described in "Oligonucleotide Synthesis", 1984, Gait, M.J. ed. , IRL Press, Oxford, which is incorporated by reference herein in its entirety.

A variety of host-expression vector systems may be utilized to express the liver activin nucleotide sequences of the invention. The liver activin protein, polypeptide or derivative can be recovered from the culture, i.e. , from the host cell in cases where the liver activin peptide or polypeptide is not secreted, and from the culture media in cases where the liver activin protein, polypeptide or derivative is secreted by the cells. However, the expression systems also encompass engineered host cells that express the liver activin or functional equivalents in situ.

The expression systems that may be used for purposes of the invention include, but are not limited to, bacterial cell systems (e.g. , E. coli, B. subtilis) transformed with recombinant bacteriophage DNA, plasmid DNA or cosmid DNA expression vectors containing liver activin nucleotide sequences; yeast cell systems (e.g. , Saccharomyces , Pichia) transformed with recombinant yeast expression vectors containing the liver activin nucleotide sequences; insect cell systems infected with recombinant virus expression vectors (e.g. , baculovirus) containing the liver activin sequences; plant cell systems infected with recombinant virus expression vectors (e.g. , cauliflower mosaic virus, CaMV; tobacco mosaic virus, TMV) or transformed with recombinant plasmid expression vectors (e.g. , Ti plasmid) containing liver activin nucleotide sequences; or mammalian cell systems (e.g.. COS, CHO, BHK, 293, 293T, 3T3) harboring recombinant expression constructs containing promoters derived from the genome of mammalian cells (e.g. , metallothionein promoter) or from mammalian viruses (e.g. , the adenovirus late promoter; the vaccinia virus 7.5K promoter). In bacterial systems, a number of expression vectors may be advantageously chosen depending upon the use intended for the liver activin gene product being expressed. When a large quantity of such a protein is to be produced, as for example during drug manufacturing or to raise antibodies, vectors which direct the expression of high levels of fusion protein products that are readily purified may be desirable. Such vectors include, but are not limited, to the E. coli expression vector pUR278 (Ruther et al., 1983, EMBO J. 2:1791), in which the liver activin coding sequence may be ligated individually into the vector in frame with the lacZ coding region so that a fusion protein is produced; pIN vectors (Inouye & Inouye, 1985, Nucleic Acids Res. 13:3101- 3109; Van Heeke & Schuster, 1989, J. Biol. Chem. 264:5503- 5509) ; the pcDNA (Invitrogen) eukaryotic vectors, and the like. pGEX vectors may also be used to express foreign polypeptides as fusion proteins with glutathione S-transferase (GST) . In general, such fusion proteins are soluble and can easily be purified from lysed cells by adsorption to glutathione-agarose beads followed by elution in the presence of free glutathione. The pGEX vectors are designed to include thro bin or factor Xa protease cleavage sites so that the cloned target gene product can be released from the GST moiety.

In an insect system, Autographa californica nuclear polyhidrosis virus (AcNPV) is used as a vector to express foreign genes. The virus grows in Spodoptera frugiperda cells. The liver activin gene coding sequence may be cloned individually into non-essential regions (for example, the polyhedrin gene) of the virus and placed under control of an AcNPV promoter (for example, the polyhedrin promoter) .

Successful insertion of liver activin gene coding sequence will result in inactivation of the polyhedrin gene and production of non-occluded recombinant virus (i.e., virus lacking the proteinaceous coat coded for by the polyhedrin gene) . These recombinant viruses are then used to infect Spodoptera frugiperda cells in which the inserted gene is expressed. (E.g. , see Smith et al., 1983, J. Virol. 46:584; Smith, U.S. Patent No. 4,215,051).

In mammalian host cells, a number of viral-based expression systems may be utilized. In cases where an adenovirus is used as an expression vector, the liver activin nucleotide sequence of interest may be ligated to an adenovirus transcription/translation control complex, e.g.. the late promoter and tripartite leader sequence. This chimeric gene may then be inserted in the adenovirus genome by in vitro or in vivo recombination. Insertion in a non- essential region of the viral genome (e.g. , region El or E3) will result in a recombinant virus that is viable and capable of expressing the liver activin gene product in infected hosts. E.g. , See Logan & Shenk, 1984, Proc. Natl. Acad. Sci. USA 81:3655-3659. Specific initiation signals may also be required for efficient translation of inserted liver activin nucleotide sequences. These signals include the ATG initiation codon and adjacent sequences. In cases where an entire liver activin gene or cDNA, including its own initiation codon and adjacent sequences, is inserted into the appropriate expression vector, no additional translational control signals may be needed. However, in cases where only a portion of the liver activin coding sequence is inserted, exogenous translational control signals, including, perhaps, the ATG initiation codon, must be provided. Furthermore, the initiation codon must be in phase with the reading frame of the desired coding sequence to ensure translation of the entire insert. These exogenous translational control signals and initiation codons can be of a variety of origins, both natural and synthetic. The efficiency of expression may be enhanced by the inclusion of appropriate transcription enhancer elements, transcription terminators, etc. See Bittner et al., 1987, Methods in Enzymol. 153:516-544. In addition, a host cell strain may be chosen which modulates the expression of inserted sequences, or which modifies and processes the gene product in the specific fashion desired. Such modifications (e.g. , glycosylation) and processing (e.g. , cleavage) of protein products may be important for the function of the protein. Different host cells have characteristic and specific mechanisms for the post-translational processing and modification of proteins and gene products. Appropriate cell lines or host systems can be chosen to ensure the correct modification and processing of the foreign protein expressed. Such mammalian host cells include but are not limited to CHO, VERO, BHK, HeLa, COS, MDCK, 293, 293T, 3T3 , WI38, and in particular, cell lines generated from embryonic tissue or adult liver. For long-term, high-yield production of recombinant proteins, stable expression is preferred. For example, cell lines which stably express the liver activin sequences described above may be engineered. Rather than using expression vectors which contain viral origins of replication, host cells can be transformed with DNA controlled by a selectable marker and appropriate expression control elements (e.g. , promoter, enhancer sequences, transcription terminators, polyadenylation sites, etc. Following the introduction of the foreign DNA, engineered cells may be allowed to grow for 1-2 days in an enriched media, and then are switched to a selective media. The selectable marker in the recombinant plasmid confers resistance to the selection and allows cells to stably integrate the plasmid into their chromosomes and grow to form foci, which, in turn, can be cloned and expanded into cell lines. This method may advantageously be used to engineer cell lines which overexpress the liver activin gene product in vitro.

A number of selection systems may be used, including but not limited to the herpes simplex virus thymidine kinase (Wigler, et al., 1977, Cell 11:223), hypoxanthine-guanine phosphoribosyltransferase (Szybalska & Szybalski, 1962, Proc. Natl. Acad. Sci. USA 48:2026), and adenine phosphoribosyltransferase (Lowy, et al., 1980, Cell 22:817) genes can be employed in tk^~, hgprt^" or aprt^" cells, respectively. Also, antimetabolite resistance can be used as the basis of selection for the following genes: dhfr, which confers resistance to methotrexate (Wigler, et al. , 1980, Natl. Acad. Sci. USA 77:3567; O'Hare, et al., 1981, Proc. Natl. Acad. Sci. USA 78:1527); gpt, which confers resistance to ycophenolic acid (Mulligan & Berg, 1981, Proc. Natl. Acad. Sci. USA 78:2072); neo, which confers resistance to the aminoglycoside G-418 (Colberre-Garapin, et al., 1981, J. Mol. Biol. 150:1); and hygro, which confers resistance to hygromycin (Santerre, et al., 1984, Gene 30:147).

Alternatively, any fusion protein may be readily purified by utilizing an antibody specific for the fusion protein being expressed. For example, a system described by Janknecht et al. allows for the ready purification of non- denatured fusion proteins expressed in human cell lines (Janknecht, et al., 1991, Proc. Natl. Acad. Sci. USA 88:8972- 8976) . In this system, the gene of interest is subcloned into a vaccinia recombination plasmid such that the gene's open reading frame is translationally fused to an amino- terminal tag consisting of six histidine residues. Extracts from cells infected with recombinant vaccinia virus are loaded onto Ni²⁺- nitriloacetic acid-agarose columns, and histidine-tagged proteins are selectively eluted with imidazole-containing buffers.

The liver activin gene products can also be expressed in transgenic animals. Animals of any species, including, but not limited to, mice, rats, rabbits, guinea pigs, pigs, micro-pigs, goats, and non-human primates, e.g. , baboons, monkeys, and chimpanzees may be used to generate liver activin transgenic animals.

Any technique known in the art may be used to introduce the liver activin transgene into animals to produce the founder lines of transgenic animals. Such techniques include, but are not limited to, pronuclear microinjection (Hoppe, P.C. and Wagner, T.E., 1989, U.S. Pat. No. 4,873,191); retrovirus mediated gene transfer into germ lines (Van der Putten et al., 1985, Proc. Natl. Acad. Sci., USA 82:6148-6152); gene targeting in embryonic stem cells (Thompson et al., 1989, Cell 56:313-321); and electroporation of embryos (Lo, 1983, Mol Cell. Biol. 3:1803-1814). For a review of such techniques, see Gordon, 1989, Transgenic Animals, Intl. Rev. Cytol. 115:171-229, which is incorporated by reference herein in its entirety. The present invention provides for transgenic animals that carry the liver activin transgene in all their cells, as well as animals which carry the transgene in some, but not all their cells, i.e. , mosaic animals. The transgene may be integrated as a single transgene or in concatamers, e.g. , head-to-head tandems or head-to-tail tandems. The transgene may also be selectively introduced into and activated in a particular cell type by following, for example, the teaching of Lasko et al. (Lasko, M. et al., 1992, Proc. Natl. Acad. Sci. USA 89:6232-6236). The regulatory sequences required for such a cell-type specific activation will depend upon the particular cell type of interest and will be apparent to those of skill in the art. When it is desired that the liver gene transgene be integrated into the chromosomal site of the endogenous liver activin gene, gene targeting is preferred. Briefly, when such a technique is to be utilized, vectors containing nucleotide sequences homologous to the endogenous liver activin gene are designed for the purpose of integrating, via homologous recombination with chromosomal sequences, and disrupting the function of the endogenous liver activin gene product. The transgene may also be selectively introduced into a particular cell type such as liver, thus inactivating the endogenous liver activin gene in only that cell type, by following, for example, the teaching of Gu et al. (Gu, et al., 1994, Science 265:103-106). The regulatory sequences required for such a cell-type specific inactivation will depend upon the particular cell type of interest, and will be apparent to those of skill in the art. Once transgenic animals have been generated, the expression of the recombinant liver activin gene may be assayed utilizing standard techniques. Initial screening may be accomplished by Southern blot analysis or PCR techniques to analyze animal tissues for integration of the transgene. The tissue level of transgene mRNA expression may also be assessed using techniques which include, but are not limited to, Northern blot analysis of tissue samples obtained from the animal, in situ hybridization analysis, and RT-PCR. Samples of liver activin gene-expressing tissue may also be evaluated immunocytochemically using antibodies specific for the liver activin transgene product.

5.3 ANTIBODIES TO LIVER ACTIVIN PROTEINS Antibodies that specifically recognize one or more epitopes of liver activin, or epitopes of conserved variants of liver activin, or peptide fragments of the liver activin, are also encompassed by the invention. Such antibodies include, but are not limited to, polyclonal antibodies, monoclonal antibodies (mAbs) , humanized or chimeric antibodies, single chain antibodies, Fab fragments, F(ab')₂ fragments, fragments produced by a Fab expression library, anti-idiotypic (anti-Id) antibodies, and epitope-binding fragments of any of the above. The antibodies of the invention may be used, for example, in the detection of the liver activin in a biological sample and, therefore, may represent part of a diagnostic or prognostic technique whereby patients may be tested for abnormal amounts of liver activin. Such antibodies may also be utilized in conjunction with, for example, compound screening schemes, as described below in Section 5.5, for the evaluation of the effect of test compounds on expression and/or activity of the liver activin gene product. Additionally, such antibodies can be used in conjunction with the gene therapy techniques, described below in Section 5.6, to, for example, evaluate the normal and/or engineered liver activin-expressing cells prior to their introduction into the patient. Such antibodies may additionally be used as part of a method to regulate and/or inhibit liver activin activity. Such antibodies, therefore, may be utilized as part of a treatment method to achieve cell 5 growth and/or differentiation.

Polyclonal antibodies are heterogeneous populations of antibody molecules derived from the sera of the immunized animals. For the production of antibodies, various host animals may be immunized by injection with the liver activin

10 protein; a liver activin polypeptide or liver activin derivative (e.g. , one corresponding to a functional domain of the liver activin, such as the mature growth factor domain) ; a truncated liver activin polypeptide (liver activin in which one or more domains, e.g. , the N-terminal signal peptide or

15 propeptide domain has been deleted) ; a functional equivalent of liver activin; or mutants of a liver activin. Such host animals may include, but are not limited to, rabbits, mice, and rats, to name but a few. Various adjuvants may be used to increase the immunological response, depending on the host

20 species, including, but not limited to, Freund's (complete and incomplete) , mineral gels such as aluminum hydroxide, surface active substances such as lysolecithin, pluronic polyols, polyanions, peptides, oil emulsions, keyhole limpet hemocyanin, dinitrophenol, and potentially useful human

25 adjuvants such as BCG (bacille Calmette-Guerin) and Corynebacterium parvum.

Monoclonal antibodies, which are homogeneous populations of antibodies to a particular antigen, may be obtained by any technique which provides for the production of antibody

30 molecules by continuous cell lines in culture. These include, but are not limited to, the hybridoma technique of Kohler and Milstein, (1975, Nature 256:495-497; and U.S. Patent No. 4,376,110), the human B-cell hybridoma technique (Kosbor et al., 1983, Immunology Today 4:72; Cole et al.,

35 1983, Proc. Natl. Acad. Sci. USA 80:2026-2030), and the EBV- hybridoma technique (Cole et al., 1985, Monoclonal Antibodies And Cancer Therapy, Alan R. Liss, Inc., pp. 77-96). Such antibodies may be of any immunoglobulin class including IgG, IgM, IgE, IgA, IgD and any subclass thereof. The hybridoma producing the mAb of this invention may be cultivated in vitro or in vivo. Production of high titers of mAbs in vivo makes this the presently preferred method of production. In addition, techniques developed for the production of "chimeric antibodies" (Morrison et al., 1984, Proc. Natl. Acad. Sci., 81:6851-6855; Neuberger et al., 1984, Nature, 312:604-608; Takeda et al., 1985, Nature, 314:452-454), by splicing the genes from a mouse antibody molecule of appropriate antigen specificity together with genes from a human antibody molecule of appropriate biological activity, can be used. A chimeric antibody is a molecule in which different portions are derived from different animal species, such as those having a variable region derived from a murine mAb and a human immunoglobulin constant region.

Alternatively, techniques described for the production of single chain antibodies (U.S. Patent 4,946,778; Bird, 1988, Science 242:423-426; Huston et al., 1988, Proc. Natl. Acad. Sci. USA 85:5879-5883; and Ward et al., 1989, Nature 334:544-546) can be adapted to produce single chain antibodies against liver activin gene products. Single chain antibodies are formed by linking the heavy and light chain fragments of the Fv region via an amino acid bridge, resulting in a single chain polypeptide.

Antibody fragments which recognize specific epitopes may be generated by known techniques. For example, such fragments include but are not limited to: the F(ab')₂ fragments which can be produced by pepsin digestion of the antibody molecule and the Fab fragments which can be generated by reducing the disulfide bridges of the F(ab')₂ fragments. Alternatively, Fab expression libraries may be constructed (Huse et al., 1989, Science, 246:1275-1281) to allow rapid and easy identification of monoclonal Fab fragments with the desired specificity.

Antibodies to the liver activin can, in turn, be utilized to generate anti-idiotype antibodies that "mimic" the liver activin or to identify the liver activin cell surface receptor, using techniques well known to those skilled in the art. See, e.g. , Greenspan & Bona, 1993, FASEB J 7 (5) :437-444; and Nissinoff, 1991, J. Immunol. 147(8) :2429- 2438. For example, antibodies which bind to the liver activin mature growth factor domain and competitively inhibit the binding of the liver activin receptor to the liver activin can be used to generate anti-idiotypes that "mimic" the mature growth factor region and, therefore, bind and neutralize liver activin receptor. Such neutralizing anti- idiotypes or Fab fragments of such anti-idiotypes can be used in therapeutic regimens that either interfere with, stimulate, or neutralize the effect of liver activin on cell growth and/or differentiation.

5.4 DIAGNOSIS OF CELL GROWTH AND/OR DIFFERENTIATION ABNORMALITIES

A variety of methods can be employed for the diagnostic and prognostic evaluation of cell growth and/or differentiation disorders, and for the identification of subjects having a predisposition to such disorders. Such disorders include, but are not limited to, disorders in hematopoiesis, erythroid differentiation, ovarian follicular maturation, pituitary growth hormone and adrenocortical tropic hormone secretion, hypothala ic oxytocin secretion, lymphocyte proliferation, neuronal survival, spermatogenesis, bone and growth formation, hepatocyte DNA synthesis, pancreatic insulin secretion, somatostatin induction, cardiac morphogenesis, megakaryocytic differentiation and the induction of mesoderm during early embryonic development.

Cell growth and/or differentiation disorders which may be diagnosed and prognosed according to the methods of the present invention include but are not limited to disorders in hematopoiesis, erythropoiesis, ovarian follicular maturation, pituitary growth hormone and adrenocortical tropic hormone secretion, hypothalamic oxytocin secretion, lymphocyte proliferation, neuronal survival, spermatogenesis, bone formation, hepatocyte DNA synthesis, pancreatic insulin secretion, and the induction of mesoderm during early embryonic development.

Such methods may, for example, utilize reagents such as the liver activin nucleotide sequences described in Section 5.1, and liver activin antibodies, as described, in Section 5.3. Specifically, such reagents may be used to, for example: (1) detect the presence of liver activin gene mutations, or detect either over- or under-expression of liver activin mRNA relative to the normal cell growth and/or differentiation state; (2) detect either an over- or an under-abundance of liver activin gene product relative to the non-cell growth and/or non-differentiation disorder state; and (3) detect perturbations or abnormalities in the signal transduction pathway mediated by liver activin ligand. The methods described herein may be performed, for example, with pre-packaged diagnostic kits comprising at least one specific liver activin nucleotide sequence or liver activin antibody reagent described herein. Such kits may be conveniently used, e.g. , in clinical settings, to diagnose patients exhibiting cell growth and/or differentiation disorder abnormalities.

For the detection of liver activin gene mutations, any nucleated cell can be used as a starting source for genomic DNA. To detect liver activin gene expression or liver activin gene products, any cell type or tissue in which the liver activin gene is expressed, such as, for example, adult liver, may be utilized.

Nucleic acid-based detection techniques are described, below, in Section 5.4.1. Peptide detection techniques are described, below, in Section 5.4.2. 5.4.1 DETECTION OF THE LIVER ACTIVIN/ INHIBIN GENE AND TRANSCRIPTS

Mutations within the liver activin gene can be detected by utilizing a number of techniques. DNA from any nucleated cell can be used as the starting point for such assay techniques. DNA may be isolated according to standard nucleic acid preparation procedures, which are well known to those of skill in the art.

DNA may be used in hybridization or amplification assays of biological samples to detect abnormalities involving liver activin gene structure, including point mutations, insertions, deletions, trinucleotide repeat expansions and chromosomal rearrangements. Such assays may include, but are not limited to, Southern analysis, single stranded conformational polymorphism analysis (SSCP) , denaturing gradient gel analysis, and PCR analysis.

Diagnostic methods to detect liver activin gene-specific mutations can involve, for example, incubating nucleic acids (including recombinant DNA molecules, cloned genes or degenerate variants thereof) , obtained from a sample, e.g. , derived from a patient sample or other appropriate cellular source, with one or more labeled nucleic acid probes including recombinant DNA molecules, cloned genes or degenerate variants thereof, as described in Section 5.1, under conditions favorable for the specific annealing of these reagents to their complementary sequences within the liver activin gene. Preferably, the lengths of these nucleic acid reagents are at least 15 to 30 nucleotides. After incubation, all non-annealed nucleic acids are removed from the nucleic acid: liver activin molecule hybrid. The presence of nucleic acids which have hybridized, if any such molecules exist, is then detected. For one such detection scheme, the nucleic acid from the cell type or tissue of interest is immobilized, for example, to a solid support such as a membrane, or a plastic surface such as a microtiter plate or polystyrene beads. In this case, after incubation, non- annealed, labeled nucleic acid reagents of the type described in Section 5.1 are easily removed with simple washing procedures. Detection of the annealed, labeled liver activin nucleic acid reagents is accomplished using standard techniques well-known to those in the art. The structure and/or level of expression of liver activin gene sequences to which the nucleic acid reagents have annealed can be compared to the annealing pattern expected from normal in order to determine whether a liver activin gene mutation is present. Alternative diagnostic methods to detect liver activin gene specific nucleic acid molecules, in patient samples or other appropriate cell sources, may involve their amplification, e.g. , by PCR (the experimental embodiment set forth in Mullis, K.B., 1987, U.S. Patent No. 4,683,202), followed by the evaluation of the amplified molecules using techniques well known to those of skill in the art. The resulting amplified sequences can be compared, for example, to normal in order to determine whether a liver activin gene mutation exists.

Additionally, well-known genotyping techniques can be performed to identify individuals carrying liver activin gene mutations. Such techniques include, for example, the use of restriction fragment length polymorphisms (RFLPs) , which involve sequence variations in one of the recognition sites for the specific restriction enzyme used. Additionally, improved methods for analyzing DNA polymorphisms, which can be utilized for the identification of liver activin gene mutations, have been described which capitalize on the presence of variable numbers of short, tandemly repeated DNA sequences between the restriction enzyme sites. For example, Weber (U.S. Pat. No. 5,075,217, which is incorporated herein by reference in its entirety) describes a DNA marker based on length polymorphisms in blocks of (dC-dA)n- (dG-dT)n short tandem repeats. The average separation of (dC-dA)n- (dG-dT) n blocks is estimated to be 30,000-60,000 bp. Markers which are so closely spaced exhibit a high frequency co-inheritance, and are extremely useful in the identification of genetic mutations, such as, for example, mutations within the liver activin gene, and the diagnosis of diseases and disorders related to liver activin mutations.

Also, Caskey et al. (U.S. Pat. No. 5,364,759, which is incorporated herein by reference in its entirety) describe a DNA profiling assay for detecting short tri and tetra nucleotide repeat sequences. The process includes extracting the DNA of interest, such as the liver activin gene, amplifying the extracted DNA, and labeling the repeat sequences to form a genotypic map of the individual's DNA. The level of liver activin gene expression can also be assayed by detecting and measuring liver activin transcription. For example, RNA from a cell type or tissue known, or suspected, to express the liver activin gene, such as adult liver, may be isolated and tested with hybridization or PCR techniques such as those described above. The isolated cells can be derived from cell culture or directly from a patient. The analysis of cells taken from culture may be a necessary step in the assessment of cells to be used as part of a cell-based gene therapy technique or, alternatively, to test the effect of compounds on the expression of the liver activin gene. Such analyses may reveal both quantitative and qualitative aspects of the expression pattern of the liver activin gene, including activation or inactivation of the liver activin gene expression.

In one embodiment of such a detection scheme, cDNAs are synthesized from the RNAs of interest (e.g. , by reverse transcription of the RNA molecule into cDNA) . A sequence within the cDNA is then used as the template for a nucleic acid amplification reaction, such as a PCR amplification reaction, or the like. Oligonucleotide primers for reverse transcription and nucleic acid amplification are chosen from among the liver activin nucleic acid reagents described in Section 5.1. The preferred lengths of such nucleic acid reagents are at least 9-30 nucleotides. To detect the amplified product, the nucleic acid amplification may be performed using radioactively or non-radioactively labeled nucleotides. Alternatively, Southern analysis may be performed, or enough amplified product may be generated such that the product may be visualized by standard ethidium bromide staining or by utilizing any other suitable nucleic acid staining method.

Additionally, it is possible to perform such liver activin gene expression assays "in situ", i.e.. directly upon histological sections (fixed and/or frozen) of patient tissue obtained from biopsies or resections, such that no nucleic acid purification is necessary. Oligonucleotides such as those described in Section 5.1 may be used as probes and/or primers for such in situ procedures (See, for example, Nuovo, G.J., 1992, "PCR In Situ Hybridization: Protocols And Applications", Raven Press, NY).

Alternatively, if a sufficient quantity of the appropriate cells can be obtained, Northern analysis can be performed to determine the level of mRNA expression of the liver activin gene.

5.4.2 DETECTION OF THE LIVER ACTIVIN GENE PRODUCTS

Antibodies directed against wild type or mutant liver activin gene products or conserved variants or peptide fragments thereof, as discussed above in Section 5.3, may also be used as diagnostics and prognostics for cell growth and/or differentiation disorders as described herein. Such diagnostic methods may be used to detect abnormalities in the level of liver activin gene expression, or abnormalities in the structure and/or temporal, tissue, cellular, or subcellular location of the liver activin; and may be performed in vivo or in vitro, such as, for example, on serum or biopsy tissue.

For example, antibodies directed to epitopes of the liver activin mature growth factor domain can be used in vivo to detect the distribution and level of the liver activin protein in the body. Such antibodies can be labeled, e.g. , with a radio-opaque or other appropriate compound, and then injected into a subject, in order to visualize binding to the liver activin expressed in the body, using methods such as X-rays, CAT-scans, or MRI . Labeled antibody fragments, e.g. , the Fab or single chain antibody comprising the smallest portion of the antigen binding region, are preferred for this purpose to promote crossing the blood-brain barrier and to permit labeling of liver activins in brain tissue.

Alternatively, immunoassays can be utilized on serum and/or autopsy samples in vitro to assess liver activin expression. Such biopsy and autopsy assays can utilize antibodies directed to various liver activin epitopes, e.g. , epitopes within the N-terminal signal peptide, propeptide and/or mature growth factor domains. The use of labeled antibodies will yield useful information regarding translation, intracellular transport, and secretion of the liver activin, and can identify defects in processing.

The tissues or cell types to be analyzed will generally include those which are known, or suspected, to express the liver activin gene, such as, for example, the adult liver. The protein isolation methods employed herein may, for example, be those described in Harlow and Lane (Harlow, E. and Lane, D. , 1988, "Antibodies: A Laboratory Manual", Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York) , which is incorporated herein by reference in its entirety. The isolated cells can be derived from cell culture or from a patient. The analysis of cells taken from culture may be a necessary step toward the development of a cell-based gene therapy protocol or, alternatively, to test the effect of compounds on liver activin gene expression and/or activity.

For example, antibodies, or fragments of antibodies, such as those described above in Section 5.3, may be used to quantitatively or qualitatively detect the presence of liver activin gene products or conserved variants or peptide fragments thereof. This can be accomplished, for example, by immunofluorescence techniques employing a fluorescently labeled antibody (see below, this Section) coupled with light microscopic or fluorimetric detection.

The antibodies (or fragments thereof) useful in the present invention may, additionally, be employed histologically, as in immunofluorescence, immunoelectron microscopy or non-immuno assays, for in situ detection of liver activin gene products or conserved variants or peptide fragments thereof.

In situ detection may be accomplished by screening blood serum or alternatively by removing a histological specimen from a patient, and applying thereto a labeled antibody of the present invention. For histological examination, the antibody (or fragment) is preferably applied by standard immunohistochemical methods. By this procedure, it is possible to determine not only the presence of the liver activin gene product, or conserved variants or peptide fragments, or the binding of liver activin to a receptor, but also its distribution in the examined tissue. Using the present invention, those of ordinary skill will readily perceive that any of a wide variety of histological methods

(such as staining procedures) can be modified in order to achieve such in situ detection.

Immunoassays for liver activin gene products, or conserved variants or peptide fragments thereof, will typically comprise incubating a sample, such as a biological fluid, a tissue extract, freshly harvested cells, or lysates of cells which have been incubated in cell culture, in the presence of a detectably labeled antibody capable of identifying liver activin gene products, or conserved variants or peptide fragments thereof, and detecting the bound antibody by any of a number of techniques well-known in the art.

The biological sample may be brought in contact with and immobilized onto a solid phase support or carrier such as nitrocellulose, or other solid support which is capable of immobilizing cells, cell particles or soluble proteins. The support may then be washed with suitable buffers followed by treatment with the labeled liver activin antibody. The solid phase support may then be washed with the buffer a second time to remove unbound antibody. The amount of bound label on solid support may then be detected by conventional means. 5 By "solid phase support or carrier" is intended any support capable of binding an antigen or an antibody. Well- known supports or carriers include glass, polystyrene, polypropylene, polyethylene, dextran, nylon, amylases, natural and modified celluloses, polyacrylamides, gabbros,

10 and magnetite. The nature of the carrier is that it can be either soluble (to some extent) or insoluble in aqueous environments for the purposes of the present invention. The support material may have virtually any possible structural configuration so long as the coupled molecule is capable of

15 binding to an antigen or antibody. Thus, the support configuration may be spherical, as in a bead, or cylindrical, as in the inside surface of a test tube, or the external surface of a rod. Alternatively, the surface may be flat such as a sheet, test strip, etc. Preferred supports include

20 polystyrene beads. Those skilled in the art will know many other suitable carriers for binding antibody or antigen, or will be able to ascertain the same by use of routine experimentation.

The binding activity of a given lot of liver activin

25 antibody may be determined according to well known methods. Those skilled in the art will be able to determine operative and optimal assay conditions for each determination by employing routine experimentation.

One of the ways in which the liver activin antibody can

30 be detectably labeled is by linking the same to an enzyme for subsequent use in an enzyme immunoassay (EIA) (Voller, A. , "The Enzyme Linked Immunosorbent Assay (ELISA)", 1978, Diagnostic Horizons 2:1-7, Microbiological Associates Quarterly Publication, Walkersville, MD) ; Voller, A. et al.,

35 1978, J. Clin. Pathol. 31:507-520; Butler, J.E., 1981, Meth. Enzymol. 73:482-523; Maggio, E. (ed.), 1980, Enzyme Immunoassay, CRC Press, Boca Raton, FL, ; Ishikawa, E. et al., (eds.), 1981, Enzyme Immunoassay, Kgaku Shoin, Tokyo). The enzyme that is bound to the antibody will react with an appropriate substrate, preferably a chromogenic substrate, in such a manner as to produce a chemical moiety which can be detected, for example, by spectrophotometric, fluorimetric or visual means. Enzymes that can be used to detectably label the antibody include, but are not limited to, malate dehydrogenase, staphylococcal nuclease, delta-5-steroid isomerase, yeast alcohol dehydrogenase, alpha- glycerophosphate, dehydrogenase, triose phosphate isomerase, horseradish peroxidase, alkaline phosphatase, asparaginase, glucose oxidase, beta-galactosidase, ribonuclease, urease, catalase, glucose-6-phosphate dehydrogenase, glucoamylase and acetylcholinesterase. The detection can be accomplished by colorimetric methods which employ a chromogenic substrate for the enzyme. Detection may also be accomplished by visual comparison of the extent of enzymatic reaction of a substrate in comparison with similarly prepared standards.

Detection may also be accomplished using any of a variety of other immunoassays. For example, by radioactively labeling the antibodies or antibody fragments, it is possible to detect liver activin through the use of a radioimmunoassay (RIA) (see, for example, Weintraub, B. , Principles of Radioimmunoassays, Seventh Training Course on Radioligand Assay Techniques, The Endocrine Society, March, 1986, which is incorporated by reference herein) . The radioactive isotope can be detected by such means as the use of a gamma counter, scintillation counter or phosphoimager, or by autoradiography. It is also possible to label the antibody with a fluorescent compound. When the fluorescently labeled antibody is exposed to light of the proper wavelength, its presence can then be detected due to fluorescence. Among the most commonly used fluorescent labeling compounds are fluorescein isothiocyanate, rhodamine, phycoerythrin, phycocyanin, allophycocyanin, o-phthaldehyde and fluorescamine . The antibody can also be detectably labeled using fluorescence emitting metals such as ¹⁵²Eu, or others of the lanthanide series. These metals can be attached to the antibody using such metal chelating groups as diethylenetriaminepentacetic acid (DTPA) or ethylenediaminetetraacetic acid (EDTA) .

The antibody also can be detectably labeled by coupling it to a chemiluminescent compound. The presence of the chemiluminescent-tagged antibody is then determined by detecting the presence of luminescence that arises during the course of a chemical reaction. Examples of particularly useful chemiluminescent labeling compounds are luminol, isoluminol, theromatic acridinium ester, imidazole, acridinium salt and oxalate ester. Likewise, a bioluminescent compound may be used to label the antibody of the present invention. Bioluminescence is a type of chemiluminescence found in biological systems, in which a catalytic protein increases the efficiency of the chemiluminescent reaction. The presence of a bioluminescent protein is determined by detecting the presence of luminescence. Important bioluminescent compounds for purposes of labeling are luciferin, luciferase and aequorin.

5.5 SCREENING ASSAYS FOR COMPOUNDS THAT MODULATE LIVER ACTIVIN EXPRESSION OR ACTIVITY

The following assays are designed to identify compounds that interact with (e.g. , bind to) liver activin (including, but not limited to, the mature growth factor domain of liver activin) , compounds that interfere with or enhance the interaction of liver activin with its cognate signaling receptors, and to compounds which modulate the activity of the liver activin gene (i.e. , modulate the level of liver activin gene expression) or modulate the level of liver activin activity. Assays may also be used that identify compounds which bind to liver activin gene regulatory sequences (e.g. , promoter sequences) and which may modulate liver activin gene expression. See e.g. , Platt, K.A. , 1994, J. Biol. Chem. 269:28558-28562, which is incorporated herein by reference in its entirety.

The compounds that may be screened in accordance with the invention include, but are not limited to, peptides, antibodies and fragments thereof, and other organic compounds (e.g. , peptidomimetics) that bind to the propeptide or mature growth factor domain of liver activin. Peptides, antibodies or fragments thereof, and other organic compounds that mimic the mature growth factor domain of the liver activin (or a portion thereof) , which bind to and "neutralize" natural liver activin receptor and thereby either mimic the activity triggered by the liver activin (i.e. , agonists) or inhibit the activity triggered by the liver activin (i.e. , antagonists) may also be screened in accordance with the invention.

Such compounds may include, but are not limited to, peptides such as, for example, soluble peptides, including but not limited to those found in random peptide libraries; (see, e.g. , Lam, K.S. et al. , 1991, Nature 354:82-84;

Houghten, R. et al., 1991, Nature 354:84-86). Such compounds may also be found in combinatorial chemistry-derived molecular libraries made of D- and/or L- configuration amino acids; phosphopeptides (including, but not limited to members of random or partially degenerate, directed phosphopeptide libraries; see, e.g.. Songyang, Z. et al., 1993, Cell 72:767- 778) ; antibodies (including, but not limited to, polyclonal, monoclonal, humanized, anti-idiotypic, chimeric or single chain antibodies, and FAb, F(ab')₂ and FAb expression library fragments, and epitope-binding fragments thereof) ; and small organic or inorganic molecules.

Numerous experimental methods may be used to select and detect proteins that bind with liver activins and thereby modulate liver activin expression or activity, including, but not limited to, protein affinity chromatography, affinity blotting, immunoprecipitation, cross-linking, and library based methods such as protein probing, phage display and the two-hybrid system. See generally, Phizicky, E. , et al., 1995, Microbiol. Rev. 59:94-123. For example, the two-hybrid system may be used to detect interaction between liver activins and candidate proteins for which genes encoding the candidate proteins are available by constructing the appropriate hybrids and testing for reporter gene activity. If an interaction is detected using the two-hybrid method, deletions can be made in the DNA encoding the candidate interacting protein or the liver activin protein to identify a minimal domain for interaction. Alternatively, the two- hybrid system can be used to screen available organismal and/or mammalian tissue specific libraries of activation domain hybrids to identify proteins that bind to a liver activin. These screens result in the immediate availability of the cloned gene for any new protein identified. In addition, since multiple clones that encode overlapping regions of protein are often identified, the minimal domain for interaction may be readily apparent from the initial screen. Computer modelling and searching technologies permit identification of compounds, or the improvement of already identified compounds, that can modulate liver activin expression or activity. Having identified such a compound or composition, the active sites or regions are identified. Such active sites might typically be ligand or receptor binding sites, such as the interaction domains of liver activin with its signaling receptor. The active site can be identified using methods known in the art including, for example, a comparison of the conceptual amino acid or nucleotide sequences. It can also be identified by study of complexes of the relevant ligand and receptor. In the latter case, chemical or X-ray crystallographic methods can be used to find the active site.

Next, the three dimensional geometric structure of the active site is determined. This can be done by known methods, including X-ray crystallography, which can determine a complete molecular structure at high resolution. On the other hand, solid or liquid phase NMR can be used to determine certain intra-molecular distances. Other experimental methods can be used to obtain partial or complete geometric structures. The geometric structures may also be obtained with a complexed ligand or receptor, natural or artificial, which may increase the accuracy of the active site structure determined.

If an incomplete or insufficiently accurate structure is determined, the methods of computer based numerical modelling can be used to complete the structure or improve its accuracy. Any recognized modelling method may be used, including parameterized models specific to particular biopolymers, such as proteins or nucleic acids. These models include molecular dynamics models based on computing molecular motions, statistical mechanics models based on thermal ensembles, or combined models. For most types of models, standard molecular force fields, representing the forces between constituent atoms and groups, are necessary, and can be selected from force fields known in physical chemistry. The incomplete or less accurate experimental structures can serve as constraints on the complete and more accurate structures computed by these modeling methods.

Finally, having determined the structure of the active site, either by experimentation, by modeling, or by a combination of the two, candidate modulating compounds can be identified by searching databases containing compounds along with information on their molecular structure. Such a search seeks compounds having structures that match the determined active site structure and that interact with the groups defining the active site. Such a search can be manual, but is preferably computer assisted. These compounds found from this search are potential liver activin modulating compounds.

Alternatively, these methods can be used to identify improved modulating compounds from an already known modulating compound or ligand or receptor, such as, for example, follistatin, latent TGF-β binding proteins ("LTBP's") , or activin receptor. The composition of the known compound, ligand, or receptor can be modified and the structural effects of modification can be determined using the experimental and computer modeling methods described above applied to the new composition. The altered structure is then compared to the active site structure of the compound to determine if an improved fit or interaction results. In this manner systematic variations in composition, such as by varying side groups, can be quickly evaluated to obtain modified modulating compounds, ligands, or receptors of improved specificity or activity.

Further experimental and computer modeling methods useful to identify modulating compounds based upon identification of the active sites of liver activins, accessory proteins, liver activin receptors, and related transduction and transcription factors will be apparent to those of skill in the art.

Examples of molecular modeling systems are the CHARMm and QUANTA programs (Polygen Corporation, Waltham, MA) . CHARMm performs the energy minimization and molecular dynamics functions. QUANTA performs the construction, graphic modeling and analysis of molecular structure. QUANTA allows interactive construction, modification, visualization, and analysis of the behavior of molecules with each other. A number of articles review computer modeling of drugs interactive with specific proteins, such as Rotivinen, et al., 1988, Acta Pharmaceutical Fennica 97:159-166; Ripka, New Scientist 54-57 (June 16, 1988); McKinaly and Rossmann, 1989, Annu. Rev. Pharmacol. Toxiciol. 29:111-122; Perry and Davies, OSAR: Quantitative Structure-Activity Relationships in Drug Design pp. 189-193 (Alan R. Liss, Inc. 1989) ; Lewis and

Dean, 1989 Proc. R. Soc. Lond. 236:125-140 and 141-162; and, with respect to a model receptor for nucleic acid components, Askew, et al., 1989, J. Am. Chem. Soc. 111:1082-1090. Other computer programs that screen and graphically depict chemicals are available from companies such as BioDesign, Inc. (Pasadena, CA.), Allelix, Inc. (Mississauga, Ontario, Canada) , and Hypercube, Inc. (Cambridge, Ontario) . Although these are primarily designed for application to drugs specific to particular proteins, they can be adapted to design of drugs specific to regions of DNA or RNA, once that region is identified. Although described above with reference to design and generation of compounds which could alter binding, one could also screen libraries of known compounds, including natural products or synthetic chemicals, and biologically active materials, including proteins, for compounds which are liver activin inhibitors or activators.

Compounds identified via assays such as those described herein may be useful, for example, in elaborating the biological function of the liver activin gene product, and for regulating cell growth and/or differentiation.

5.5.1 IN VITRO SCREENING ASSAYS FOR

COMPOUNDS THAT BIND TO LIVER ACTIVIN

In vitro systems may be designed to identify compounds capable of interacting with (e.g. , binding to) liver activin (including, but not limited to, the mature growth factor domain of liver activin) . Compounds identified may be useful, for example, in modulating or disrupting the activity of wild type and/or mutant liver activin gene products or in elaborating the biological function of the liver activin.

The principle behind assays that identify compounds that bind to liver activin involves preparing a reaction mixture of liver activin and test compound under conditions that allow the two components to interact and bind, thus forming a complex which can be detected in the reaction mixture and/or purified. The liver activin species used can vary depending upon the goal of the screening assay. For example, where agonists of natural liver activin are sought, the full length liver activin, or a truncated liver activin, e.g. , as described above, can be utilized.

The screening assays can be conducted in a variety of ways. For example, one method to conduct such an assay would involve anchoring the liver activin protein, polypeptide, or fusion protein onto a solid phase and detecting liver activin/test compound complexes anchored on the solid phase at the end of the reaction. In one embodiment of such a method, the liver activin reactant may be anchored onto a solid surface, and the test compound, which is not anchored, may be labeled, either directly or indirectly.

In practice, microtiter plates may conveniently be utilized as the solid phase. The anchored component may be immobilized by non-covalent or covalent attachments. Non- covalent attachment may be accomplished by simply coating the solid surface with a solution of the protein and drying. Alternatively, an immobilized antibody, preferably a monoclonal antibody, specific for the protein to be immobilized may be used to anchor the protein to the solid surface. The surfaces may be prepared in advance and stored. In order to conduct the assay, the nonimmobilized component is added to the coated surface containing the anchored component. After the reaction is complete, unreacted components are removed (e.g. , by washing) under conditions such that any complexes formed will remain immobilized on the solid surface. The detection of complexes anchored on the solid surface can be accomplished in a number of ways. Where the previously nonimmobilized component is pre-labeled, the detection of label immobilized on the surface indicates that complexes were formed. Where the previously nonimmobilized component is not pre-labeled, an indirect label can be used to detect complexes anchored on the surface, e.g. , using a labeled antibody specific for the previously nonimmobilized component (the antibody, in turn, may be directly labeled or indirectly labeled with a labeled anti-Ig antibody) .

Alternatively, a reaction can be conducted in a liquid phase, the reaction products separated from unreacted components, and complexes detected, e.g. , using an immobilized' antibody specific for liver activin protein, polypeptide, derivative or fusion protein or the test compound to anchor any complexes formed in solution, and a labeled antibody specific for the other component of the possible complex to detect anchored complexes.

5.6 METHODS FOR TREATMENT AND REGULATION OF CELL GROWTH AND/OR DIFFERENTIATION BY USE OF LIVER ACTIVIN ANTIBODIES, ANTAGONISTS, AGONISTS OR GENE PRODUCTS

The invention encompasses methods and compositions for regulating cell growth and/or differentiation and/or treating cell growth and/or differentiation disorders.

Any method which neutralizes or enhances liver activin activity can be used to regulate cell growth and/or differentiation.

For example, the administration of soluble peptides, proteins, fusion proteins, or antibodies (including anti- idiotypic antibodies) that bind to and "neutralize" circulating liver activin can be used to regulate cell growth and/or differentiation. To this end, peptides corresponding to the extracellular domain of the liver activin receptor, deletion mutants of liver activins, polypeptides or derivatives, or polypeptides containing liver activin mutants or one or more liver activin domains fused to another polypeptide (e.g. , an IgFc polypeptide) can be utilized. Alternatively, anti-idiotypic antibodies or Fab fragments of antiidiotypic antibodies that mimic the liver activin and bind to the activin receptor can be used (see Section 5.3, supra) . Such liver activin proteins, polypeptides, derivatives, fusion proteins, anti-idiotypic antibodies or Fabs are administered to a subject in amounts sufficient to neutralize the interaction between liver activin and liver activin receptor.

Liver activin peptides corresponding to the propeptide domain such as, for example, having the amino acid sequence of β_c shown in FIG. 1 from about amino acid residue number 21 to about 236, having the amino acid sequence of β_E shown in FIG. 2 from about amino acid residue number 22 to about 236 may be used or having the amino acid sequence of β_E shown in FIG. 5 from about amino acid residue number 18 to about 236 may be used. Alternatively, liver activin peptides corresponding to the mature growth factor domain, such as, for example, having the amino acid sequence of β_c shown in FIG. 1 from about amino acid residue number 237 to about 354, or having the amino acid sequence of β_E shown in FIG. 2 from about amino acid residue number 237 to about 352 or having the amino acid sequence of β_E shown in FIG. 5 from about 237- 350 to about may be used. Additionally, liver activin deletion mutants in which all or part of the propeptide domain and/or mature growth factor domain are missing may also be used. The liver activin receptor, and soluble or bound polypeptides containing all or part of the extracellular domain of the liver activin receptor may also be used. The liver activin protein, polypeptide or derivative and the liver activin receptor proteins or polypeptides may also be fused to heterologous proteins to increase stability and/or to increase or decrease the liver activin activity.

Cell growth and/or differentiation processes which may be regulated according to the methods of the present invention include, but are not limited to, hematopoiesis, erythroid differentiation, ovarian follicular maturation, pituitary growth hormone and adrenocortical tropic hormone secretion, hypothalamic oxytocin secretion, lymphocyte proliferation, neuronal survival, spermatogenesis, bone growth and formation, hepatocyte DNA synthesis, pancreatic insulin secretion, somatostatin induction, cardiac morphogenesis, egakaryocytic differentiation and the induction of mesoderm during early embryonic development.

5.6.1 METHOD OF TREATMENT, REGENERATION, AND GROWTH OF LIVER TISSUE BY USE OF LIVER ACTIVIN GENE PRODUCT ANTIBODY OR ANTAGONIST

Any method which neutralizes liver activin or inhibits expression of the liver activin gene can be used to enhance the growth or regeneration of liver tissue. The liver activin antibodies and liver activin gene product antagonists may be used to enhance the growth or regeneration of liver tissue in a variety of situations. In some cases, a patient's liver may be damaged but not beyond repair. For example, and not by way of limitation, excessive consumption of alcohol often leads to cirrhosis of the liver. Hepatocyte destruction can be arrested by discontinuation of alcohol consumption, but recovery may require regeneration of the liver. In such cases, the natural regeneration process may be impaired due to extensive liver damage. In any event, treatment of the patient with pharmaceutical compositions, as described below in Section 5.8, comprising liver activin antibodies or liver activin product antagonists will enhance regeneration and thereby speed recovery.

In some situations, treatment may require transplanting all or a section of the liver of a donor. Regeneration of both a living donor's and a recipient's liver during such transplantation treatments will be aided by administering pharmaceutical compositions, as described below in Section 5.8, comprising a liver activin antibody and/or liver activin gene product antagonist.

In other situations, an artificial liver, produced according to the methods described below, for example, may be implanted into a patient suffering from liver disease. It may be sufficient and desirable to implant such an artificial liver at a stage where it has not yet attained the biological capacity of a normal liver. To increase the capacity of such an implant, the growth rate can be enhanced by administering pharmaceutical compositions, as described below in Section 5.8, comprising liver activin antibody and/or liver activin gene product antagonist.

In cases where a patient's natural liver is damaged or diseased, it may be left intact or only partially removed but still require support from implanted artificial liver tissue or liver tissue transplanted from a donor. Pharmaceutical compositions comprising antibodies or liver activin gene product antagonists can be used in such cases to enhance the growth of the patient's natural liver tissue, as well as the implanted or transplanted liver tissue.

In vitro liver tissue cultures have a variety of uses. In treating patients suffering from liver damage or disease, for example, the liver tissue cultures can be used to support or replace the natural liver, by direct implantation or as part of an extracorporeal liver device. The regenerative capacity of the liver tissue cultures may be tested using the transgenic mouse system developed by Rhim et al. (1995, Proc. Natl. Acad. Sci. USA 92:4942-4946. In such a system, liver cells are transplanted into the diseased livers of transgenic animals and assessed for their replicative potential.

In addition, such liver tissue cultures can serve as models for testing the toxicity of drugs and other compounds. Functional in vitro liver tissue may be generated, for example, and not by way of limitation, using the three- dimensional tissue culture system described in U.S. Pat. No. 5,266,480, which is incorporated herein by reference in its entirety. According to this system, hepatocytes may be isolated by conventional methods (Berry and Friend, 1969, J. Cell Biol. 43:506-520) which can be adapted for human liver biopsy or autopsy material. Briefly, a canula is introduced into the portal vein or a portal branch, and the liver is perfused with calcium-free or magnesium-free buffer until the tissue appears pale. The organ is then perfused with a proteolytic enzyme such a collagenase solution at an adequate flow rate. This should digest the connective tissue framework. The liver is then washed in buffer and the cells are dispersed. The cell suspension may be filtered through a 70 μm nylon mesh to remove debris. Hepatocytes may be selected from the cell suspension by two or three differential centrifugations.

For perfusion of individual lobes of excised human liver, HEPES buffer may be used. Perfusion of collagenase in HEPES buffer may be accomplished at the rate of about 30 ml/minute. A single cell suspension is obtained by further incubation with collagenase for 15-20 minutes at 37 °C (Guguen-Guillouzo and Guillouzo, eds., 1986, "Isolated and Culture Hepatocytes", Paris, INSERM, and London, John Libbey Eurotext, pp. 1-12; 1982, Cell Biol. Int. Rep. 6:625-628). The isolated hepatocytes may then be used to inoculate the three dimensional stroma. The inoculated stroma can be cultured according to the teachings of U.S. Pat. No. 5,266,480, as described for bone marrow and skin, in order to replicate the hepatocytes in vitro, in a system comparable to physiologic conditions. In addition, the growth rate and development of the three-dimensional liver culture, or of liver tissue cultures grown by alternative methods, may be enhanced by adding liver activin antibodies and/or gene product antagonists to the growth medium. This should result in an increased functional expression by the hepatocytes.

5.6.2 STIMULATION OF HEMATOPOIESIS

BY USE OF LIVER ACTIVIN GENE PRODUCT

Any method which elevates liver activin concentration and/or activity can be used to stimulate hematopoiesis. Using these methods, the liver activin gene and nucleotide sequences described in Section 5.1 and proteins, polypeptides, and derivatives described in Section 5.2, above may be used to stimulate hematopoiesis.

According to the invention, the liver activin compositions described herein, therapies directed towards elevating the concentration of blood cells such as hematopoietin therapy. In a specific embodiment, the liver activin compositions described herein are used in erythropoietin therapy, which is directed toward supplementing the oxygen carrying capacity of blood. Liver activin treatment within the scope of the invention includes but is not limited to patients generally requiring blood transfusions, such as, for example, trauma victims, surgical patients, dialysis patients, and patients with a variety of blood composition-affecting disorders, such as hemophilia, cystic fibrosis, pregnancy, menstrual disorders, early anemia of prematurity, spinal cord injury, space flight, aging, various neoplastic disease states, and the like. Examples of patient conditions that require supplementation of the oxygen carrying capacity of blood and which are within the scope of this invention, include but are not limited to: treatment of blood disorders characterized by low or defective red blood cell production, anemia associated with chronic renal failure, stimulation of reticulocyte response, development of ferrokinetic effects (such as plasma iron turnover effects and marrow transit time effects) , erythrocyte mass changes, stimulation of hemoglobin C synthesis, and increasing levels of hematocrit in vertebrates. The invention also provides for treatment to enhance the oxygen-carrying capacity of an individual, such as for example, an individual encountering hypoxic environmental conditions. The invention also encompasses combining the liver activin compositions described in Sections 5.1 and 5.2 with other proposed or conventional hematopoietic therapies. Thus, for example, the liver activin can be combined with compounds that singly exhibit erythropoietic stimulatory effects, such as erythropoietin, testosterone, progenitor cell stimulators, insulin-like growth factor, prostaglandins, serotonin, cyclic AMP, prolactin, and triiodothyzonine. Also encompassed are combinations with compounds generally used to treat aplastic anemia, such as methenolene, stanozolol, and nandrolone; to treat iron-deficiency anemia, such as iron preparations; to treat malignant anemia, such as vitamin B₁₂ and/or folic acid; and to treat hemolytic anemia, such as adrenocortical steroids, e.g. , corticoids. See e.g., Resegotti et al., 1981, Panminerva Medica, 23:243-248; Kurtz, 1982, FEBS Letters, 14a: 105-108; McGonigle et al., 1984, Kidney Int., 25:437-444; and Pavlovic-Kantera, 1980, Expt. Hematol., 8(supp. 8) 283-291.

Compounds that enhance the effects of or synergize with erythropoietin are also useful as adjuvants herein, and include but are not limited to, adrenergic agonists, thyroid hormones, androgens, hepatic erythropoietic factors, erythrotropins , and erythrogenins, See for e.g., Dunn, "Current Concepts in Erythropoiesis" , John Wiley and Sons (Chichester, England, 1983); Weiland et al., 1982, Blut, 44:173-175; Kalmani, 1982, Kidney Int., 22:383-391; Shahidi, 1973, New Eng. J. Med., 289:72-80; Urabe et al., 1979, J. Exp. Med., 149:1314-1325; Billat et al., 1982, Expt.

Hematol., 10:133-140; Naughton et al., 1983, Acta Haemat, 69:171-179; Cognote et al. in abstract 364, Proceedings 7th Intl. Cong, of Endocrinology (Quebec City, Quebec, July 1-7, 1984); and Roth an et al., 1982, J. Surg. Oncol., 20:105-108. Methods for stimulating hematopoiesis comprise administering a hematopoietically effective amount (i.e, an amount which effects the formation of blood cells) of a pharmaceutical composition containing liver activin to a patient. The liver activin is administered to the patient by any suitable technique, including but not limited to, parenteral, sublingual, topical, intrapulmonary and intranasal, and those techniques further discussed in Section 5.8. The pharmaceutical composition optionally contains one or more members of the group consisting of erythropoietin, testosterone, progenitor cell stimulators, insulin-like growth factor, prostaglandins, serotonin, cyclic AMP, prolactin, triiodothyzonine, methenolene, stanozolol, and nandrolone, iron preparations, vitamin B₁₂, folic acid and/or adrenocortical steroids. The liver activin and cotreatment drug(s) are suitably delivered by separate or by the same administration route, and at the same or at different times, depending, e.g. , on dosing, the clinical condition of the patient, etc.

5.6.3 METHOD OF STIMULATING BONE GROWTH

BY USE OF LIVER ACTIVIN GENE PRODUCT

Any method which elevates liver activin concentration and/or activity can be used to effectuate bone growth (i.e,. bone mass) . Such approaches can be used to treat bone fractures, defects, and disorders which result in weakened bones such as osteoporosis, osteomalacia, and age-related loss of bone mass. According to the invention, bone growth is enhanced by local and/or systemic administration of a pharmaceutical composition of liver activin in an osteogenically effective amount (i.e., an amount which effects the formation and development of mature bone) . The compositions of the invention described in Section 5.1 and Section 5.2 above, may optionally contain osteogenically effective amounts of other bone growth promoting compounds, including beta-type transforming growth factors ("TGF- s") (e.g., TGF-^β 1,-2,-3, proteins, polypeptides, derivatives (including fragments) , and homologous proteins thereof) and/or bone morphogenic proteins ("BMPs") (e.g., BMP-, -2 , -3 , -4 , -5, -6, or -7 proteins, polypeptides, and homologous proteins thereof) and/or osteogenic proteins (including polypeptides, and homologous proteins thereof) and/or parathyroid hormone. BMPs and TGF- βs may be prepared by methods known in the art (see e.g., PCT/US87/01537 and 4,774,322 which are incorporated herein by reference in their entirety) . Alternatively, TGF-3s are available from commercial sources (R&D Systems, Minneapolis, Minn. ) .

In an alternative embodiment, the liver activin compositions of the invention include other therapeutics such as for example, in the case of treatment of osteoporosis, fluoride, calcitonin, vitamin D metabolites, estrogen, and parathyroid hormone.

Methods for inducing local bone growth include administering an osteogenically effective amount of liver activin, optionally in combination with TGF-β, BMP, bone marrow, or proteins extracted therefrom, in a pharmaceutically acceptable carrier.

One embodiment of the invention encompasses a method for inducing systemic bone growth, comprising administering an osteogenically effective amount of liver activin optionally including TGF-β, BMP, bone marrow, or proteins extracted therefrom, in a pharmaceutically acceptable carrier.

Another embodiment provides a method for regulating bone growth which comprises initially administering an effective amount of TGF-β and BMPs and sequentially administering an effective amount of activin. Preferably, to induce, endochondral bone growth, combinations of BMP, TGF-β, may be locally administered to induce cartilage modeling. Activin may be subsequently administered subcutaneously and/or systemically to induce mature bone formation and differentiation of the cartilage model induced by TGF-β and BMP. According to this method, the activin would enhance mineralization of the endochondral bone. It is expected that the quantity of mature bone formed by this method would be greater than the quantity of mature bone formed by either TGF-β, BMPs, or liver activin alone, or combinations of TGF-β and BMPs, or combinations of liver activin and BMPs.

Alternative, nonlimiting methods for regulating bone growth and maturation comprises locally administering an osteogenically effective amount of activin with TGF-β, BMP, and/or bone marrow or proteins extracted therefrom to induce bone growth, followed by systemically administration of an osteogenically effective amount of liver activin.

5.7 METHODS FOR REGULATING LIVER ACTIVIN ACTIVITY AND/OR TREATMENT OF LIVER DISEASE BY AFFECTING LIVER ACTIVIN GENE EXPRESSION

Any method which inhibits or activates expression of the liver activin gene (either transcription or translation) can be used to regulate cell growth and/or differentiation activities affected by liver activin genes. Described below are methods whereby cell growth and/or differentiation may be regulated and cell growth and/or differentiation disorders may be treated with the nucleic acid sequences described in Section 5.1, above.

In certain cases, an increase in liver activin activity may be desired to facilitate liver activin regulated mechanisms, such as, for example, to stimulate liver regeneration and/or bone growth. Furthermore, certain cell growth and/or differentiation disorders may be brought about, at least in part, by the absence or reduction of the level of liver activin gene expression. As such, an increase in the level of gene expression would bring about the amelioration of symptoms relating to this disorder. Techniques for increasing liver activin gene expression levels are discussed in Section 5.7.1, below.

In some cases, the reduction in the level and/or activity of liver activin gene products would decrease or prevent the inhibitory effect of liver activin, and it would therefore be desirable to reduce liver activin gene product. Techniques for the reduction of liver activin, gene expression levels are discussed in Section 5.7.2, below.

5.7.1 RESTORATION OR INCREASE IN

LIVER ACTIVIN EXPRESSION OR ACTIVITY

With respect to an increase in the level of normal liver activin gene expression and/or liver activin gene product activity, liver activin nucleic acid sequences can be utilized for the regulation of cell growth and/or differentiation and/or treatment of cell growth and/or differentiation disorders. Treatment can be administered, for example, in the form of gene replacement therapy. Specifically, one or more copies of a normal liver activin gene or a portion of the liver activin gene that directs the production of a liver activin gene product exhibiting normal function, may be inserted into the appropriate cells within a patient or animal subject, using vectors which include, but are not limited to, adenovirus, adeno-associated virus, retrovirus and herpes virus vectors, in addition to other particles that introduce DNA into cells, such as liposomes and gene activated matrices. Methods of transferring nucleic acids into mammalian repair cells using gene activated matrices is further described in U.S. application Serial No. 08/631,334, filed April 12, 1996, the contents of which is hereby incorporated by reference in its entirety.

Because the liver activin gene is expressed in the liver, such gene replacement therapy techniques should be capable of delivering liver activin gene sequences to cells in this tissue within patients, or, alternatively, should involve direct administration of such liver activin gene sequences to the site of the cells in which the liver activin gene sequences are to be expressed. Alternatively, targeted homologous recombination can be utilized to correct the defective endogenous liver activin gene in the appropriate tissue; e.g. , liver.

Additional methods which may be utilized to increase the overall level of liver activin gene expression and/or liver activin activity include the introduction of appropriate liver activin-expressing cells, preferably autologous cells, into a patient at positions and in numbers which are sufficient to ameliorate the symptoms of cell growth and/or differentiation disorders. Such cells may be either recombinant or non-recombinant . Among the cells which can be administered to increase the overall level of liver activin gene expression in a patient are normal cells, which express the liver activin gene. The cells can be administered at the anatomical site in the liver, or as part of a tissue graft located at a different site in the body. Such cell-based gene therapy techniques are well known to those skilled in the art, see, e.g. , Anderson, et al., U.S. Patent No. 5,399,349; Mulligan & Wilson, U.S. Patent No. 5,460,959.

5.7.2 METHODS FOR DECREASING

LIVER ACTIVIN EXPRESSION OR ACTIVITY

A variety of techniques may be utilized to reduce or inhibit the expression of liver activin genes using molecules derived from the sequences described in Section 5.1, above.

The level of endogenous liver activin gene expression can be reduced, e.g. , using antisense or ribozyme approaches to inhibit or prevent translation of the liver activin mRNA transcripts; triple helix approaches to inhibit transcription of the liver activin gene; or targeted homologous recombination to inactivate or "knock out" the liver activin gene or its endogenous promoter. One or more copies of a nucleotide sequence may be inserted into the appropriate cells within a patient or animal subject, using, for example but not by way of limitation, adenovirus, adeno-associated virus, retrovirus and herpes virus vectors, in addition to other particles that introduce nucleic acids into cells, such as liposomes and gene activated matrices.

Because the liver activin gene is expressed in the adult liver, delivery techniques should be preferably designed to target this tissue. Alternatively, the antisense, ribozyme or DNA constructs described herein could be administered directly to the site containing the target cells; e.g. , the adult liver.

Antisense approaches involve the design of oligonucleotides (either DNA or RNA) that are complementary to liver activin mRNA. The antisense oligonucleotides will bind complementary liver activin mRNA transcripts and prevent translation. Absolute complementarily, although preferred, is not required. A sequence "complementary" to a portion of an RNA, as referred to herein, means a sequence having sufficient complementarily to be able to hybridize with the RNA, forming a stable duplex. In the case of double-stranded antisense nucleic acids, a single strand of the duplex DNA may thus be tested, or triplex formation may be assayed. The ability to hybridize will depend on both the degree of complementarily and the length of the antisense nucleic acid. Generally, the longer the hybridizing nucleic acid, the more base mismatches with an RNA it may contain and still form a stable duplex (or triplex, as the case may be) . One skilled in the art can ascertain a tolerable degree of mismatch by use of standard procedures to determine the melting point of the hybridized complex.

Oligonucleotides that are complementary to the 5 ' end of the message, e.g. , the 5' untranslated sequence up to and including the AUG initiation codon, should work most efficiently at inhibiting translation. However, sequences complementary to the 3 ' untranslated sequences of mRNAs have recently shown to be effective at inhibiting translation of mRNAs as well. See, generally, Wagner, R. , 1994, Nature 372:333-335. Thus, oligonucleotides complementary to either the 5'- or 3'- non- translated, non-coding regions of the liver activin genes could be used in an antisense approach to inhibit translation of endogenous liver activin mRNA. Oligonucleotides complementary to the 5' untranslated region of the mRNA should include the complement of the AUG start codon. Antisense oligonucleotides complementary to mRNA coding regions are less efficient inhibitors of translation but could be used in accordance with the invention. Whether designed to hybridize to the 5'-, 3 ' - or coding region of liver activin mRNA, antisense nucleic acids should be at least six nucleotides in length, and are preferably oligonucleotides ranging from 6 to about 50 nucleotides in length. In specific aspects the oligonucleotide is at least 10 nucleotides, at least 17 nucleotides, at least 25 nucleotides or at least 50 nucleotides.

Regardless of the choice of target sequence, it is preferred that in vitro studies are first performed to quantitate the ability of the antisense oligonucleotide to inhibit gene expression. It is preferred that these studies utilize controls that distinguish between antisense gene inhibition and nonspecific biological effects of oligonucleotides. It is also preferred that these studies compare levels of the target RNA or protein with that of an internal control RNA or protein. Additionally, it is envisioned that results obtained using the antisense oligonucleotide are compared with those obtained using a control oligonucleotide. It is preferred that the control oligonucleotide is of approximately the same length as the test oligonucleotide and that the nucleotide sequence of the oligonucleotide differs from the antisense sequence no more than is necessary to prevent specific hybridization to the target sequence .

The oligonucleotides can be DNA or RNA or chimeric mixtures or derivatives or modified versions thereof, single- stranded or double-stranded. The oligonucleotide can be modified at the base moiety, sugar moiety, or phosphate backbone, for example, to improve stability of the molecule, hybridization, etc. The oligonucleotide may also be conjugated with other appended groups such as peptides (e.g. , for targeting host cell receptors in vivo) ; agents facilitating transport across the cell membrane (see, e.g. , Letsinger et al., 1989, Proc. Natl. Acad. Sci. U.S.A. 86:6553-6556; Lemaitre et al., 1987, Proc. Natl. Acad. Sci. 84:648-652; PCT Publication No. WO88/09810, published December 15, 1988) or the blood-brain barrier (see, e.g. , PCT Publication No. WO89/10134, published April 25, 1988); hybridization-triggered cleavage agents (See, e.g. , Krol et al., 1988, BioTechniques 6:958-976); and intercalating agents (see, e.g. , Zon, 1988, Pharm. Res. 5:539-549).

The antisense oligonucleotide may comprise at least one modified base moiety which is selected from the group including, but not limited to, 5-fluorouracil, 5-bromouracil, 5-chlorouracil, 5-iodouracil, hypoxanthine, xantine, 4-acetylcytosine, 5- (carboxyhydroxylmethyl) uracil, 5-carboxymethylaminomethyl-2-thiouridine, 5-carboxymethylaminomethyluracil, dihydrouracil, beta-D- galactosylqueosine, inosine, N6-isopentenyladenine, 1-methylguanine, 1-methylinosine, 2 , 2-dimethylguanine, 2-methyladenine, 2-methylguanine, 3-methylcytosine, 5-methylcytosine, N6-adenine, 7-methylguanine, 5-methylaminomethyluracil , 5-methoxyaminomethyl-2-thiouracil , beta-D-mannosylqueosine, 5 ' -methoxycarboxymethyluracil , 5-methoxyuracil, 2-methylthio-N6-isopentenyladenine, uracil-5-oxyacetic acid (v) , wybutoxosine, pseudouracil, queosine, 2-thiocytosine, 5-methyl-2-thiouracil, 2-thiouracil, 4-thiouracil, 5-methyluracil, uracil-

5-oxyacetic acid methylester, uracil-5-oxyacetic acid (v) , 5-methy1-2-thiouracil , 3- ( 3-amino-3-N-2-carboxypropy1) uracil, (acp3)w, and 2 , 6-diaminopurine.

The antisense oligonucleotide may also comprise at least one modified sugar moiety selected from the group including but not limited to arabinose, 2-fluoroarabinose, xylulose, and hexose. In yet another embodiment, the antisense oligonucleotide comprises at least one modified phosphate backbone selected from the group consisting of a phosphorothioate, a phosphorodithioate, a phosphoramidothioate, a phosphoramidate, a phosphordiamidate, a methyIphosphonate, an alkyl phosphotriester, and a formacetal or analog thereof.

In yet another embodiment, the antisense oligonucleotide is an α-anomeric oligonucleotide. An α-anomeric oligonucleotide forms specific double-stranded hybrids with complementary RNA, in which, contrary to the usual β-units, the strands run parallel to each other (Gautier et al., 1987, Nucl. Acids Res. 15:6625-6641). The oligonucleotide is a 2'- O-methylribonucleotide (Inoue et al., 1987, Nucl. Acids Res. 15:6131-6148), or a chimeric RNA-DNA analogue (Inoue et al., 1987, FEBS Lett. 215:327-330).

Oligonucleotides of the invention may be synthesized by standard methods known in the art, e.g. , by use of an automated DNA synthesizer (such as are commercially available from Biosearch, Applied Biosystems, etc.). As examples, phosphorothioate oligonucleotides may be synthesized by the method of Stein et al. (1988, Nucl. Acids Res. 16:3209), and methyIphosphonate oligonucleotides can be prepared by use of controlled pore glass polymer supports (Sarin et al., 1988, Proc. Natl. Acad. Sci. U.S.A. 85:7448-7451), etc. While antisense nucleotides complementary to the liver activin coding region sequence could be used, those complementary to the transcribed untranslated region are most preferred. For example, antisense oligonucleotides having the following sequences can be utilized in accordance with the invention:

a) 5'- CATTGCTGGAGGATAGCC-3 ' which is complementary to nucleotides 134 to 151 in FIG. 1.

b) 5'- CATTGCTGGAGGATAGCCTCCTT -3' which is complementary to nucleotides 129 to 151 in FIG. 1. c) 5'- CATTGCTGGAGGATAGCCTCCTTAAAGGA -3' which is complementary to nucleotides 123 to 151 in FIG. 1.

d) 5 ' - CATTGCTGGAGGATAGCCTCCTTAAAGGACCCAAAl-3 ' which is complementary to nucleotides 117 to 151 in FIG. 1.

e) 5'- TGCTGGAGGATAGCCTCCTTAAAGGACCC -3' which is complementary to nucleotides 120 to 148 in FIG. 1.

f) 5'- AGGATAGCCTCCTTAAAGGACCC -3' which is complementary to nucleotides 120 to 142 in FIG. 1.

g) 5 ' - GCCTCCTTAAAGGACCC -3 ' which is complementary to nucleotides 120 to 136 in FIG. 1.

h) 5 ' - CATGCTCCAGGTAGATG -3 ' which is complementary to nucleotides 202 to 218 in FIG. 2.

i) 5'- CATGCTCCAGGTAGATGGCTCTG -3' which is complementary to nucleotides 196 to 218 in FIG. 2.

j) 5'- CATGCTCCAGGTAGATGGCTCTGCTTTGA -3' which is complementary to nucleotides 190 to 218 in FIG. 2.

k) 5'- CATGCTCCAGGTAGATGGCTCTGCTTTGAGAACCT -3' which is complementary to nucleotides 184 to 218 in FIG. 2.

1) 5'- CTTCATGCTCCAGGTAGATGGCTCTGCTTTGAGAA -3' which is complementary to nucleotides 187 to 221 in FIG. 2.

m) 5 ' - GCTCCAGGTAGATGGCTCTGCTTTGAGAA -3 ' which is complementary to nucleotides 187 to 215 in FIG. 2.

n) 5 ' - GGTAGATGGCTCTGCTTTGAGAA -3 ' which is complementary to nucleotides 187 to 209 in FIG. 2. o) 5'- TGGCTCTGCTTTGAGAA -3' which is complementary to nucleotides 187 to 203 in FIG. 2.

p) 5'- TGCTTTGAGAA -3' which is complementary to nucleotides 187 to 197 in FIG. 2.

q) 5'- CATGCTCCTGGTGAATGG -3' which is complementary to nucleotides 60 to 78 in FIG. 5.

r) 5'- CATGCTCCTGGTGAATGGTGGTT -3' which is complementary to nucleotides 55 to 78 in FIG. 5.

s) 5'- CATGCTCCTGGTGAATGGTGGTTCAGGAG -3' which is complementary to nucleotides 49 to 78 in FIG. 5.

t) 5'- CATGCTCCTGGTGAATGGTGGTTCAGGAG -3' which is complementary to nucleotides 43 to 78 in FIG. 5.

u) 5 ' - GCTCCTGGTGAATGGTGGTTCAG -3 ' which is complementary to nucleotides 47 to 75 in FIG. 5.

v) 5'- CCTGGTGAATGGTGGTT -3' which is complementary to nucleotides 44 to 71 in FIG. 5.

w) 5'- GGTGAATGGTG -3' which is complementary to nucleotides 41 to 68 in FIG. 5.

The antisense molecules should be delivered to cells which express the liver activin in vivo, e.g. , hepatocytes. A number of methods have been developed for delivering antisense DNA or RNA to cells, e.g. , antisense molecules can be injected directly into the tissue site, or modified antisense molecules, designed to target the desired cells (e.g. , antisense linked to peptides or antibodies that specifically bind receptors or antigens expressed on the target cell surface) can be administered systemically. It is often difficult to achieve intracellular concentrations of the antisense sufficient to suppress translation of endogenous mRNAs. Therefore a preferred approach utilizes a recombinant DNA construct in which the antisense oligonucleotide is placed under the control of a strong pol III or pol II promoter. The use of such a construct to transfect target cells in the patient will result in the transcription of sufficient amounts of single stranded RNAs that form complementary base pairs with the endogenous liver activin transcripts and thereby prevent translation of the liver activin mRNA. For example, a vector can be introduced in vivo such that it is taken up by a cell and directs the transcription of an antisense RNA. Such a vector can remain episomal or become chromosomally integrated, as long as it can be transcribed to produce the desired antisense RNA. Such vectors can be constructed by recombinant DNA technology methods standard in the art. Vectors can be plasmid, viral, or others known in the art, used for replication and expression in mammalian cells. Expression of the sequence encoding the antisense RNA can be by any promoter known in the art to act in mammalian, preferably human, cells. Such promoters can be inducible or constitutive. Such promoters include but are not limited to: the SV40 early promoter region (Bernoist and Chambon, 1981, Nature 290:304-310), the promoter contained in the 3' long terminal repeat of Rous sarcoma virus (Yamamoto et al., 1980, Cell 22:787-797), the herpes thymidine kinase promoter (Wagner et al., 1981, Proc. Natl. Acad. Sci. U.S.A. 78:1441- 1445) , the regulatory sequences of the etallothionein gene (Brinster et al., 1982, Nature 296:39-42), etc. Any type of plasmid, cosmid, YAC or viral vector can be used to prepare the recombinant DNA construct which can be introduced directly into the tissue site; e.g. , the liver. Alternatively, viral vectors can be used which selectively infect the desired tissue; (e.g. , for brain, herpesvirus vectors may be used) , in which case administration may be accomplished by another route (e.g. , systemically) . Ribozyme molecules designed to catalytically cleave liver activin mRNA transcripts can also be used to prevent translation of liver activin mRNA and expression of liver activin. (See, e.g. , PCT International Publication WO90/11364, published October 4, 1990; Sarver et al., 1990, Science 247:1222-1225). While ribozymes that cleave mRNA at site specific recognition sequences can be used to destroy liver activin mRNAs, the use of hammerhead ribozymes is preferred. Hammerhead ribozymes cleave mRNAs at locations dictated by flanking regions that form complementary base pairs with the target mRNA. The sole requirement is that the target mRNA have the following sequence of two bases: 5 ' -UG- 3 ' . The construction and production of hammerhead ribozymes is well known in the art and is described more fully in Haseloff and Gerlach, 1988, Nature, 334:585-591. For example, there are over 110 and over 80 potential hammerhead ribozyme cleavage sites within the nucleotide sequence of murine β_c and β_E , respectively (FIG. 1 and FIG. 2) . Preferably the ribozyme is engineered so that the cleavage recognition site is located near the 5' end of the liver activin mRNA, i.e. , to increase efficiency and minimize the intracellular accumulation of non-functional mRNA transcripts.

For example, hammerhead ribozymes having the following sequences can be utilized in accordance with the invention:

a) 5 ' -UGGAGGAUAGCCUCCUUAAAGGACCCAAAGCAGGNNNNCCUGAGNA GUCAGCCUAUAUUCAGGGUCUCA-3 ' which will cleave between nucleotides 117 and 118 in FIG. 1.

b) 5 ' -GGAGCAAGGAGGAGGCCAAAGCCGGNNNNCCGGAGNA GUCUGCUGGAGGAUAGCCUCCUUA-3 ' which will cleave between nucleotides 148 and 149 in FIG. 1.

c) 5 ' -GUUCACUACUGUGGUUGGAGUCAGGAACAAAGCUGUNNNNA

CAGAGNAGUCAAGAGCCAGGAGCAAGGA-3 ' which will cleave between nucleotides 175 and 176 in FIG. 1. d) 5 ' -ACAUGAAGCGGGUCUGCCGGACCUCCAAAGCUGANNNNUCAGAGNA GUCGCCACUGGCCAUUCUACCAGAG-3 ' which will cleave between nucleotides 526 and 527 in FIG. 1.

e) 5 ' -UUCUGGUACCAAAGCGGGNNNNCCCGAGNA

GUCCUCCAGAGUAAGGUG-3 ' which will cleave between nucleotides 742 and 743 in FIG. 1.

f) 5 ' -AACCUCGAAUGGGUGGCAAAGCCGCNNNNGCGGAGNA GUCUUUUAUGAGUCUGAUUU-3 ' which will cleave between nucleotides 169 and 170 in FIG. 2.

g) 5 ' -UGGAAGCUUCAUGCUCCAAAGCAGANNNNUCUGAGNA GUCGUAGAUGGCUCUG-3 ' which will cleave between nucleotides 208 and 209 in FIG. 2.

h) 5 ' -UGCCCACAGCAGUAUUAGCCAAAGCGGUNNNNACCGAGNA GUCAGCUGGGGUUUUGGAAGCUUCAU-3 ' which will cleave between nucleotides 238 and 239 in FIG. 2.

i) 5 ' -UUGGGCCCCCACAGGACGGGCAAAGCCGANNNNUCGGAGNA GUCGCAGAUCUUCUACUCUGCACCC-3 ' which will cleave between nucleotides 289 and 290 in FIG. 2.

j) 5 ' -CGGCUGCAGGAUCCAAUCCCGCCACCCCAAAGCGGCNNNNGCCG AGNAGUCCUCCUGGAAGUCUAC-3 ' which will cleave between nucleotides 989 and 990 in FIG. 2.

k) 5 ' -CTTGACCACATTGCCAAAGCGGUNNNNACCGAGNA GUCUAUGGUCAAGGUAGAGGAG-3 ' which will cleave between nucleotides 1206 and 1207 in FIG. 2.

1) 5 ' -CCGCAAAGCAGANNNNUCUGAGNAGUCGCUCCU -3' which will cleave between nucleotides 75 and 76 in FIG. 5.

m) 5 ' -GGACAAAGCAGANNNNUCUGAGNAGUCCAGGGA -3 ' which will cleave between nucleotides 88 and 89 in FIG. 5. n) 5 ' -UCGCAAAGCAGANNNNUCUGAGNAGUCCAGUGC -3' which will cleave between nucleotides 123 and 124 in FIG. 5.

o) 5 ' -ACACAAAGCAGANNNNUCUGAGNACUCACCCUG -3' which will cleave between nucleotides 139 and 140 in FIG. 5.

p) 5 ' -CCACAAAGCAGANNNNUCUGAGNAGUCGAGGGA -3' which will cleave between nucleotides 158 and 159 in FIG. 5.

The ribozymes of the present invention also include RNA endoribonucleases (hereinafter "Cech-type ribozymes") , such as the one which occurs naturally in Tetrahymena thermophila (known as the IVS, or L-19 IVS RNA) and which has been extensively described by Thomas Cech and collaborators (Zaug, et al., 1984, Science, 224:574-578; Zaug and Cech, 1986, Science, 231:470-475; Zaug, et al., 1986, Nature, 324:429- 433; published International patent application No. WO 88/04300 by University Patents Inc.; Been and Cech, 1986, Cell, 47:207-216). The Cech-type ribozymes have an eight base pair active site which hybridizes to a target RNA sequence, whereafter cleavage of the target RNA takes place. The invention encompasses those Cech-type ribozymes which target eight base-pair active site sequences that are present in the liver activin RNA.

As in the antisense approach, the ribozymes can be composed of modified oligonucleotides (e.g. for improved stability, targeting, etc.) and should be delivered to cells which express the liver activin in vivo. A preferred method of delivery involves using a DNA construct "encoding" the ribozyme under the control of a strong constitutive pol III or pol II promoter, so that transfected cells will produce sufficient quantities of the ribozyme to destroy endogenous liver activin messages and inhibit translation. Because ribozymes unlike antisense molecules, are catalytic, a lower intracellular concentration is required for efficiency. Endogenous liver activin gene expression can also be reduced by inactivating or "knocking out" the liver activin gene or its promoter using targeted homologous recombination. (E.g. , see Smithies et al., 1985, Nature 317:230-234; Thomas & Capecchi, 1987, Cell 51:503-512; Thompson et al., 1989 Cell 5:313-321; each of which is incorporated by reference herein in its entirety) . For example, a mutant, non-functional liver activin (or a completely unrelated DNA sequence) flanked by DNA homologous to the endogenous liver activin gene (either the coding regions or regulatory regions of the liver activin gene) can be used, with or without a selectable marker and/or a negative selectable marker, to transfect cells that express liver activin in vivo. Insertion of the DNA construct, via targeted homologous recombination, results in inactivation of the liver activin gene. Such approaches are particularly suited in the agricultural field where modifications to ES (embryonic stem) cells can be used to generate animal offspring with an inactive liver activin

(e.g. , see Thomas & Capecchi 1987 and Thompson 1989, supra) . However this approach can be adapted for use in humans provided the recombinant DNA constructs are directly administered or targeted to the required site in vivo using appropriate viral vectors.

Alternatively, endogenous liver activin gene expression can be reduced by targeting deoxyribonucleotide sequences complementary to the regulatory region of the liver activin gene (i.e. , the liver activin promoter and/or enhancers) to form triple helical structures that prevent transcription of the liver activin gene in target cells in the body. (See generally, Helene, C. 1991, Anticancer Drug Des. , 6(6):569- 84; Helene, C. , et al., 1992, Ann, N.Y. Acad. Sci., 660:27- 36; and Maher, L. J. , 1992, Bioassays 14 (12) : 807-15) . In yet another embodiment of the invention, the activity of liver activin can be reduced using a "dominant negative." To this end, constructs which encode defective liver activin, such as, for example, mutants lacking all or a portion of the endoproteolytic motif, and/or mature growth factor domain, can be used in gene therapy approaches to diminish the activity of the liver activin on appropriate target cells. For example, nucleotide sequences that direct host cell expression of liver activin in which all or a portion of the propeptide domain, endoproteolytic motif, and/or mature growth factor domain, is altered or missing can be introduced into cells in the liver (either by in vivo or ex vivo gene therapy methods described above) . Alternatively, targeted homologous recombination can be utilized to introduce such deletions or mutations into the subject's endogenous liver activin gene in the liver. The engineered cells will express non-functional liver activin (i.e. , a ligand that may be capable of binding, but which is incapable of inducing signal transduction) .

5.8 PHARMACEUTICAL PREPARATIONS AND METHODS OF ADMINISTRATION

The liver activin proteins and polypeptides as well as compounds that affect liver activin gene expression or gene product activity can be administered to a patient at therapeutically effective doses to regulate, treat or ameliorate cell growth and/or differentiation disorders. A therapeutically effective dose refers to that amount of the compound sufficient to result in regulation and/or amelioration of symptoms of cell growth and/or differentiation disorders.

5.8.1 EFFECTIVE DOSE

Toxicity and therapeutic efficacy of such compounds can be determined by standard pharmaceutical procedures in cell cultures or experimental animals, e.g. , for determining the LD₅₀ (the dose lethal to 50% of the population) and the ED₅₀ (the dose therapeutically effective in 50% of the population) . The dose ratio between toxic and therapeutic effects is the therapeutic index and it can be expressed as the ratio LD₅₀/ED₅₀. Compounds which exhibit large therapeutic indices are preferred. While compounds that exhibit toxic side effects may be used, care should be taken to design a delivery system that targets such compounds to the site of affected tissue in order to minimize potential damage to uninfected cells and, thereby, reduce side effects.

The data obtained from the cell culture assays and animal studies can be used in formulating dosage range for use in humans. The dosage of such compounds lies preferably within a range of circulating concentrations that include the ED₅₀ with little or no toxicity. The dosage may vary within this range depending upon the dosage form employed and the route of administration utilized. For any compound used in the method of the invention, the therapeutically effective dose can be estimated initially from cell culture assays. A dose may be formulated in animal models to achieve a circulating plasma concentration range that includes the IC₅₀ (i.e. , the concentration of the test compound which achieves a half-maximal inhibition of symptoms) as determined in cell culture. Such information can be used to more accurately determine useful doses in humans. Levels in plasma may be measured, for example, by high performance liquid chromatography.

5.8.2 FORMULATIONS AND USE Pharmaceutical compositions for use in accordance with the present invention may be formulated in conventional manner using one or more physiologically acceptable carriers or excipients.

Thus, the compounds and their physiologically acceptable salts and solvates may be formulated for administration by inhalation or insufflation (either through the mouth or the nose) or oral, buccal, parenteral or rectal administration. Administration can be systemic or local.

For oral administration, the pharmaceutical compositions may take the form of, for example, tablets or capsules prepared by conventional means with pharmaceutically acceptable excipients such as binding agents (e.g. , pregelatinized maize starch, polyvinylpyrrolidone or hydroxypropyl methylcellulose) ; fillers (e.g. , lactose, microcrystalline cellulose or calcium hydrogen phosphate) ; lubricants (e.g. , magnesium stearate, talc or silica) ; disintegrants (e.g. , potato starch or sodium starch glycolate) ; or wetting agents (e.g. , sodium lauryl sulphate) . The tablets may be coated by methods well known in the art. Liquid preparations for oral administration may take the form of, for example, solutions, syrups or suspensions, or they may be presented as a dry product for constitution with water or other suitable vehicle before use. Such liquid preparations may be prepared by conventional means with pharmaceutically acceptable additives such as suspending agents (e.g. , sorbitol syrup, cellulose derivatives or hydrogenated edible fats) ; emulsifying agents (e.g. , lecithin or acacia); non-aqueous vehicles (e.g. , almond oil, oily esters, ethyl alcohol or fractionated vegetable oils) and preservatives (e.g. , methyl or propyl-p-hydroxybenzoates or sorbic acid) . The preparations may also contain buffer salts and flavoring-, coloring- and sweetening- agents as appropriate.

Preparations for oral administration may be suitably formulated to give controlled release of the active compound. For buccal administration the compositions may take the form of tablets or lozenges formulated in conventional manner .

For administration by inhalation, the compounds for use according to the present invention are conveniently delivered in the form of an aerosol spray from pressurized packs or a nebulizer, with the use of a suitable propellant, e.g.. dichlorodifluoromethane, trichlorofluoromethane , dichlorotetrafluoroethane, carbon dioxide or other suitable gas. In the case of a pressurized aerosol, the dosage unit may be determined by providing a valve to deliver a etered amount. Capsules and cartridges of, e.g. , gelatin for use in an inhaler or insufflator, may be formulated containing a powder mix of the compound and a suitable powder base such as lactose or starch.

The compounds may be formulated for parenteral administration by injection, e.g. , by bolus injection or continuous infusion. Formulations for injection may be presented in unit dosage form, e.g. , in ampules or in multi- dose containers, with an added preservative. The compositions may take such forms as suspensions, solutions or emulsions in oily or aqueous vehicles, and may contain formulatory agents such as suspending, stabilizing and/or dispersing agents. Alternatively, the active ingredient may be in powder form for constitution with a suitable vehicle, e.g. , sterile pyrogen-free water, before use. Further discussion of vehicles for parenteral administration may be found in E.W. Martin, "Remington's Pharmaceutical Sciences" (Mack Pub. Co. ) .

The compounds may also be formulated in rectal compositions such as suppositories or retention enemas, e.g. , containing conventional suppository bases such as cocoa butter or other glycerides.

In addition to the formulations described previously, the compounds may also be formulated as a depot preparation. Such long acting formulations may be administered by implantation (for example subcutaneously or intramuscularly) or by intramuscular injection. Thus, for example, the compounds may be formulated with suitable polymeric or hydrophobic materials (for example as an emulsion in an acceptable oil) or ion exchange resins, or as sparingly soluble derivatives, for example, as a sparingly soluble salt.

The compositions may, if desired, be presented in a pack or dispenser device, which may contain one or more unit dosage forms containing the active ingredient. The pack, may for example, comprise metal or plastic foil, such as a blister pack. The pack or dispenser device may be accompanied by instructions for administration.

In one embodiment, it may be desirable to administer the pharmaceutical compositions of the invention locally to the area in need of treatment. This may be achieved, for example, and not by way of limitation, by local infusion during surgery; topical application, e.g. , in conjunction with a wound dressing after surgery; injection, from a catheter, with a suppository, or from an implant, said implant being of a porous, non-porous, or gelatinous material, including membranes, such as sialastic membranes, or fibers. In a specific embodiment, the porous end of dental or orthopedic implants, which is inserted into the implant site, may be coated with osteoinductive proteins such as liver activin, activin A, activin B, BMP, and TGF-β , to enhance attachment of the implant device to the bone. Alternatively, the implants may be coated with DNA encoding osteoinductive proteins such as liver activin, activin A, activin B, BMP and TGF-β.

6. EXAMPLE: IDENTIFICATION AND CHARACTERIZATION

OF THE ACTIVIN/INHIBIN ft AND β_F GENE

The identification and characterization of activin β_c and β_E subunits, two members of a novel subgroup of activins, is described herein. cDNA clones encoding two members of the transforming growth factor (TGF-β) superfamily have been isolated and characterized in the mouse. Both cDNAs appear to code for activin β subunits—i. e. , one gene is likely the mouse homolog of human activin β_c, while the second is novel and has been designated activin β_E . In addition, a human activin β_E cDNA clone has been identified and characterized.

Our studies suggest that in certain respects activin β_c and β_E are more like one another than other known activins: (a) , these genes are physically linked, with only 5-kbp of genomic

DNA separating the 3 ' end of activin β_c from the 5 ' end of activin β_E , (b) , the organization of both genetic loci is similar, consisting of two coding exons separated by a single intervening sequence; (c) , the predicted carboxy-terminal sequences of both activins are highly homologous, with 62% amino acid identity over the mature growth factor region;

(d) , both genes appear to be specifically and highly expressed in adult mouse liver and (e) the results of several in vitro assays show that homodimers of activin β_c and β_E stimulate the proliferation of HepG2 cells. When taken together, our data suggest that the activin β_c and β_E subunit genes may have evolved from a common ancestor that underwent a tandem duplication. These structural and functional data differentiate activin β_c and β_E subunits from other known activins, suggesting that these subunits represent a unique subgroup of TGF-β-like molecules.

6.1 MATERIALS AND METHODS

PCR. Genomic DNA templates were isolated from NIH3T3 cells using standard methods (Sambrook et al., 1989, supra) . RNA and cDNA templates were prepared as described by Chen et al. (1993, J. Biol. Chem., 268:27,381-27,389). Primer sequences were as follows: upstream primer (forward), 5' GT(ACGT) GG(ACGT) TGG AA(CT) GA(CT) TGG AT 3'; and downstream primer (reverse complement), 5' CA (ACGT)CC (AG) CA (ACGT)CC (CT)TC (ACGT)AC (ACGT)AC 3'. Note that the individual codons have been separated by a space and that degeneracy was achieved by adding a cocktail of nucleotides during synthesis, as shown in parentheses, at selected wobble positions. For both genomic DNA and cDNA templates, PCR proceeded through 35 cycles of denaturation (94°C, 1.5 min), annealing (50°C, 1 min, 10 sec) , and elongation (72°C, 1 min, 10 sec) . cDNA Cloning. Aliquots of 50,000 plaque forming units/plate (Mouse Liver 5 ' -Stretch Plus cDNA library in Lambda gtll, from adult BALB/c male mice, Clonetech Laboratories, Inc., Palo Alto, CA) and fresh overnight Y1088 cells (grown in LB medium supplemented with 0.2% maltose in lOmM MgS0₄) were mixed, incubated for 15 min at 37 °C, mixed again with 8 ml of liquid top layer agarose consisting of NZY Broth (Gibco-BRL Life Technologies Inc., Grand Island, NY) plus 0.75% agarose, and then spread evenly onto freshly poured 150 mm NZY-agar plates. Standard methods were used to prepare plaque-lifts and perform filter hybridization (42 °C, in 50% formamide, 5 x SSC buffer, 1 x Denhardt's solution, 10 mM Tris, pH 7.5, 0.1% SDS, and 100 μg/ml salmon sperm DNA). Sambrook et al. , 1989, supra . Filters were washed progressively to high stringency (0.1 x SSC buffer, 0.1% SDS, 65°C) . cDNA probes were radiolabelled by the random priming method using commercially available reagents and protocols (Random Primed DNA Labeling Kit, Amersham Life Sciences, Inc. , Arlington Heights, IL) . Purified phage clones were cultured with Y1088 cells, and phage DNA was extracted using the protocols and reagents with the Wizard™ Lambda Prep kit (Pro ega, Madison, WI) . Phage DNA was digested with EcoRI, the recombinant insert was purified and subcloned into pBluescript plasmid vector (Stratagene, La Jolla, CA) using standard protocols, and the insert was sequenced using Sequenase (version 2.0, United States Biochemical Company, Cleveland, OH) according to the manufacturer's protocols. Sequence alignment and identity were determined using software from Genetics Computer Group, Madison, WI (GrowTree™, version 8.1, UPMGA method of analysis).

Isolation and Characterization of Mouse Genomic Clones. A genomic library in the Lambda Fix II vector (made from mouse strain 129SVJ liver DNA, Stratagene) was plated at 50,000 plaques/plate, and 20 nitrocellulose replicas were screened at high stringency as described above using a labeled β_c cDNA as the probe. Filters were washed progressively to high stringency also as described above and autoradiographed. Positive phage plaques were re-screened until purified, and DNA was prepared as described in the

Wizard™ Lambda Pre kit (Promega) . Five unique clones from a pool of ten isolates, each with more than 10-kbp of genomic DNA, were isolated. Notl digests of all unique clones were subcloned into Notl digested pBluescript plasmid vector (Stratagene) . Genomic inserts were characterized by Southern analysis and by DNA sequencing.

Isolation and Characterization of a Human Activin β_F Clone. The mouse activin β_E plasmid pbE #2 was used as a probe to screen a cDNA library. Aliquots of 50,000 plaque forming units/plate (Human Liver 5 ^λ -Stretch Plus cDNA library in Lambda gtll, from adult human liver, Clonetech Laboratories, Inc., Palo Alto, CA) and fresh overnight Y1088 cells (grown in LB medium supplemented with 0.2% maltose in 10 mM MgS0₄) were mixed, incubated for 15 min at 37°C, mixed again with 8 ml of liquid top layer agarose consisting of NZY Broth (Gibco-BRL Life Technologies Inc., Grand Island, NY) plus 0.75% agarose, and then spread evenly onto freshly poured 150 mm NZY-agar plates. Standard methods were used to prepare plaque-lifts and perform filter hybridization (37°C, in 50% formamide, 5xSSC buffer, 1 x Denhardt's solution, 10 mM Tris, pH 7.5, 0.1% SDS, and 100 μg/ml salmon sperm DNA) . Filters were washed progressively to low stringency (0.5 x SSC buffer, 0.1% SDS, 55°C) . The cDNA probe was radiolabeled by the random priming method using commercially available reagents and protocols (Random Primed DNA Labeling Kit, Amersham Life Sciences, Inc., Arlington Heights, IL) . Purified phage clones were cultured with Y1088 cells, and phage DNA was extracted using the protocols and reagents with the Wizard^S Lambda Prep kit (Promega, Madison, WI) . Phage DNA was digested with Eco RI, the recombinant insert was purified and subcloned into a pBluescript plasmid vector (Stratagene, La Jolla, CA) using standard protocols, and the insert was sequenced using Sequenase (version 2.0, United States Biochemical Company, Cleveland, OH) according to the manufacturer's protocols. Sequence alignment and identity were determined using software from the Genetics Computer Group, Madison, WI (GrowTree^S, version 8.1, UPMGS method of analysis).

Northern analysis. An adult mouse multiple tissue Northern blot was purchased from Clonetech. According to product information, each lane of the blot contains about 2 μg of poly (A) RNA. Specific cDNA probes, consisting solely of untranslated sequence from the 3 ' untranslated region of activin β_c (nucleotides 1230-1555) and β_E (nucleotides 1822- 2130) , were generated by subcloning and then ³²P-labeled by random priming and used for hybridization (42 °C, overnight) in 50% formamide, 5 x SSPE buffer (pH 7.4, BioWhittaker,

Walkersville, MD) , 10 x Denhardt's, 2% SDS, 100 μg/ml salmon sperm DNA. Blots were washed progressively from hybridization buffer to 0.1 x SSC, 0.1% SDS (55°C) , and then placed against x-ray film (X-Omat AR film, Eastman Kodak Co., Rochester, NY) with intensifying screens at -86°C for the indicated times. Expression Plasmids. In order to screen for function, expression plasmids for activin β_c, β_E , and β_A subunits were prepared as described below. A hemagglutinin ("HA") epitope tag (Fang et al., 1996, Proc. Natl. Acad. Sci. U.S.A., 93:5753-5758) was added to the coding sequence of each expression plasmid. To insure the integrity of the final construct, all three expression plasmids were subjected to restriction mapping and DNA sequence analysis.

An expression plasmid coding for the mouse activin β_c subunit was assembled into the pcDNA3 vector (Invitrogen Corp., San Diego, CA) . Prior to subcloning, the HA epitope was added to the 3 ' end of the activin β_c coding sequence by PCR. Primer sequences were as follows: upstream primer (forward), 5' CCG AAT TCC ATG GCC TCC TTG CTC CTG GC 3 ' , and downstream primer (reverse complement), 5' GGG AAT TCA AGC GTA ATC CGG AAC GTA ACT ACA CCC GCA GGC CTC GAC CAC GGG ATC C

3 ' . Note that the downstream primer has been designed so that the PCR product codes for the HA epitope and has an EcoRI site at its 3' end. Plasmid from a cDNA clone containing the entire β_c coding region was used as a source of template DNA. PCR proceeded through 25 cycles of denaturation (94°C, 1.0 min), annealing (65°C, 1 min, 10 sec) , and elongation (72 °C, 1 min, 10 sec) . The PCR product (1.1-kbp) was digested with EcoRI, purified by agarose gel electrophoresis, and subcloned into the pcDNA3 vector. An expression plasmid coding for the mouse activin β_B subunit was also assembled into the pcDNA3 vector. An HA epitope was added to the 3 ' end of the coding sequence essentially as descried above. Primer sequences were as follows: upstream primer (forward), 5' CCC AAT TCC TGG AGC ATG AAG CTT CCA AA 3 ' , and downstream primer (reverse complement) , 5 ' GGG AAT TCA AGC GTA ATC CGG AAC ATC GTA TGG GTA GCT GCA GCC ACA GGC CTC TAC 3 ' . Plasmid from a cDNA clone containing the entire β_E coding region was used as a source of template DNA, and PCR was performed essentially as described above. The PCR product (1.1-kbp) was digested with EcoRI, purified by agarose gel electrophoresis, and subcloned 5 into the pcDNA3 vector.

The activin β_A subunit was obtained by PCR, using an ovary cDNA library (Clonetech) as a source of template DNA. Primer sequences were as follows: upstream primer (forward) , 5 ' GCT GCC AGG ATG CCC TTG CTT 3 ' , and downstream primer

10 (reverse complement), 5' CTA TGA GCA CCC ACA CTC 3'. PCR proceeded through 30 cycles of denaturation (94 °C, 1.5 min) , annealing (60°C, 1 min, 10 sec) , and elongation (72°C, 1 min, 10 sec). The PCR product (1.3-kbp) was purified by agarose gel electrophoresis and then cloned into the TA cloning

15 vector (Invitrogen Corp., San Diego, CA) . An expression plasmid coding for the mouse activin β_A subunit was then assembled into the pcDNA3 vector. An HA epitope was added to the 3 ' end of the coding sequence essentially as described above. The plasmid pbE #2 was used as a source of template

20 DNA, and PCR was performed essentially as described above. The PCR product was digested with EcoRI, purified by agarose gel electrophoresis, and subcloned into the pcDNA3 vector.

Expression Plasmids. To obtain homodimers of recombinant activin β_c, β_E , and β_A subunits, transient

25 transfection was performed using standard protocols.

Sambrook et al., 1989, supra . On the day of transfection, subconfluent 293T cells were fed fresh Dulbecco's Modified Eagles' Medium supplemented with 10% fetal calf serum (reagents from Gibco-BRL Life Technologies) . Ten μg of

30 expression plasmid DNA were added to 0.5 ml of 0.25 M CaCl₂ solution, and the mixture was added drop-wise to 0.5 ml Hepes-buffered saline (270 mM NaCI, 1.5 mM NaP0₄, 50 mM Hepes, pH 7.05). This mixture was kept at room temperature for 10 min and then added drop-wise to the medium of freshly fed

35 293T cells. After an overnight incubation at 37°C, the medium was replaced and cells were cultured for an additional 48 h. Medium samples were collected in the presence of protease inhibitors (leupeptin, 1 mg/ml, aprotinin, 1 mg/ml, soybean trypsin inhibitor, 10 mg/ml, and pepstatin, 1 mg/ml; reagents from Sigma Chemical Co., St. Louis, MO), and stored at 4°C until use.

6.2 RESULTS AND DISCUSSION A pair of degenerate PCR primers was designed based on conserved sequences in human bone morphogenetic protein-2 and -4 (VGWNDWI and WEGCGC) : upstream primer (forward), 5' GT(ACGT) GG(ACGT) TGG AA(CT) GA(CT) TGG AT 3'; and downstream primer (reverse complement), 5' CA (ACGT)CC (AG)CA (ACGT)CC (CT)TC (ACGT)AC (ACGT)AC 3'. Note that individual codons have been separated by a space and that degeneracy was achieved by adding a cocktail of nucleotides, as shown in parentheses, at selected wobble positions. PCR products were expected to be derived from within a single coding exon, and both cDNA and genomic DNA were used as templates. PCR was performed using a series of annealing temperatures until a band of the expected size of ~300-bp was obtained. The resultant PCR products were cloned into the TA cloning vector

(Invitrogen) and then used to transform E. coli. Thousands of transformed E. coli colony forming units were available for screening, and several known TGF-β superfamily members eventually were identified, including bone morphogenetic protein-2 and -4, activin β_A, activin β_B , and growth/differentiation factor-6 and -7 (not shown) .

These efforts also led to the isolation of an apparently unique PCR fragment of 280 nucleotides that we designated DP- 2. The DP-2 fragment was used as a probe to screen 500,000 plaque forming units from a mouse liver cDNA library. Three unique, overlapping cDNA clones that hybridized to DP-2 under high stringency conditions were purified from a pool of more than 20 positive isolates. DNA sequence analysis (all 3 clones) revealed an open reading frame of 1056 nucleotides within a 1837 nucleotide cDNA sequence (FIG. 1) . The predicted sequence consisted of 352 amino acids with an estimated molecular weight (molw) of 39 kD and 4 potential N- linked glycosylation sites. A Met codon (nucleotide 148) in a favorable context for translation initiation was provisionally designated the translational start site. Upstream of this ATG codon, we found four in-frame stop codons but no additional ATG codons. We also found a canonical TGF-/3 CXCX sequence (Roberts, et al. 1990, The transforming growth factor betas. In: Sporn MB, Roberts AB (eds) Handbook of Experimental Pharmacology, 95 (Part I) , Springer-Verlag, Heidelberg, pp. 419-472) preceding the stop translation signal (TAG) . Based on sequence comparisons (described below) , the 1056 nucleotide open reading frame sequence has provisionally been designated mouse activin β_c .

While cloning the mouse activin β_c genomic locus, we found a 300-bp BamHI genomic DNA fragment that was located downstream of the activin β_c gene and yet hybridized on a Southern blot under low stringency conditions with a β_c cDNA probe (data not shown) . This BamHI fragment was subcloned into the plasmid vector pBluescript (Stratagene) and used as a probe to screen an additional 500,000 plaque forming units from the mouse liver cDNA library under high stringency conditions. Unique, overlapping cDNA clones, which represented a sequence of 2130-bp, were purified from a pool of more than 20 positive isolates. Sequence analysis (all 3 clones) revealed an open reading frame of 1050 nucleotides, a conceptual sequence of 350 amino acids, a predicted molecular weight of 39-kD, and 1 potential N-linked glycosylation site (FIG. 2B) . A Met codon (nucleotide 216) in a favorable context for translation initiation was provisionally designated the translational start site. Upstream of this ATG codon, we found a single in-frame stop codon but no additional ATG codons. A canonical CXCX sequence preceding the stop translation signal (TAG) was also identified. As described below, a sequence comparison analysis suggests that the 1050 nucleotide open reading frame sequence codes for a novel activin β subunit, which has been designated mouse activin β_E . The mouse activin β_c and β_E subunits are organized like other TGF-β superfamily members (Roberts, et al., 1990, The transforming growth factor betas. In: Sporn MB, Roberts AB (eds) Handbook of Experimental Pharmacology, 95 (Part I) , Springer-Verlag, Heidelberg, pp. 419-472; Kingsley, D.M. , 1994, Genes Develop 8:133-146). In both instances the initiator Met residue is followed by a characteristic signal peptide of -20 amino acids. Downstream of the signal peptide both sequences appeared to be organized into propeptide and mature growth factor domains, with putative endoproteolytic cleavage sites (i.e., clustered basic amino acids; see FIG. 1 and FIG. 2) , at the junction of these domains. If these cleavage sites are in fact used, then the propeptide and mature growth factor domains of the mouse activin β_c subunit will 'consist of approximately 216 and 116 amino acids, respectively, and the propeptide and mature growth factor domains of the mouse activin β_E subunit will consist of approximately 215 and 114 amino acids, respectively. Except for Mullerian inhibitory substance, all known members of the TGF-β superfamily have a cluster of basic residues approximately 120 amino acids from the C-terminus. At least in the case of the TGF-β's (Roberts et al., 1990, The transforming growth factor betas. In: Sporn MB, Roberts AB (eds) Handbook of Experimental Pharmacology, 95 (Part I) , Springer-Verlag, Heidelberg, pp. 419-472; Kingsley, D.M. , 1994, Genes Develop 8:133-146), inhibins (Vale et al., 1990, The activins and inhibins. In: Sporn MB, Roberts AB (eds) Peptide Growth Factors and Their Receptors, 95 (Part II) . Springer-Verlag, Heidelberg, pp. 211-248; Mason et al., 1985, Nature 318:659-663), and bone morphogenetic protein-2 the mature form of the protein is known to be generated by endoproteolytic cleavage at these sites.

Several genomic clones were isolated under high stringency conditions in a screen of 1,000,000 plaque forming units from a mouse genomic library using a β_c cDNA as the probe. Partial restriction and DNA sequence analyses of the overlapping inserts of these clones indicated that -5-kbp of mouse genomic DNA separates the 3 ' end of activin β_c and the 5' end of activin β_E , with the activin β_E locus downstream of activin β_c (FIG. 4) . The organization of both genetic loci is similar, consisting of two coding exons separated by a single intervening sequence. Partial restriction analysis and partial DNA sequence analysis of overlapping genomic clones indicated that the size of the activin β_c intron is ~12-kbp. In contrast, DNA sequencing has shown that the size of the activin β_E intron is 234-bp. The 234-bp activin β_E intron sequence has been deposited in GenBank (accession number) . The predicted carboxy-terminal amino acid sequences of mouse activin β_c and β_E are highly homologous, with 61% amino acid identity over the mature growth factor domain. Both coding sequences contain a total of nine cysteine residues; spacing of the cysteines and the presence of other conserved amino acids is most characteristic of the activin subfamily and, to a lesser extent, TGF3l-5. In tabulating percent amino acid sequence identity within the mature growth factor domain, mouse activin β_c (the 1056 nucleotide coding sequence) showed 52% identity with human activin β_A, 50% identity with human activin β , 95% identity with human activin β_c, and 63% identity with Xenopus activin β_D . In comparison, mouse activin β_E (the 1050 nucleotide coding sequence) showed 45% identity with human activin β_A, 47% identity with human activin β_B , 64% identity with human activin β_c, and 62% identity with Xenopus activin β_O. In contrast, we found only 20-40% amino acid sequence identity when our mouse coding sequences were compared with other members of the TGF-β superfamily. These data suggest that the 1056 nucleotide coding sequence most likely is the mouse homolog of human activin β_c. In contrast, the 1050 nucleotide coding sequence appears to be a novel activin subunit gene, which we have designated activin β_E . To our knowledge, this is the first report of the isolation of activin β_E from any species. The relationships also suggest that activin β_A and β_B are most related to one another, while activin β_{c ι} β_D, and β_E represent a second subset of related sequences. Alignment of the mature growth factor region of the five activin β subunits is shown in FIG. 3. We found that activin β_A and β_B share 63% amino acid identity, while activin β_c, β_D, and β_E share 62% identity. The five activin β subunits share about 50% identity overall.

The pattern of mouse activin β_c and β_E gene expression was assessed by Northern analysis of several adult mouse tissues. In contrast to activin A and B, which are widely expressed in adult animals and humans (See, e.g., Meunier, H. et al., 1988, Proc. Natl. Acad. Sci. USA, 85:247-251), hybridization of a multiple tissue Northern blot with a specific probe from the 3 ' untranslated region of activin β_c identified a single major transcript of -2.0-kb in the liver (FIG. 6) . This band was identified with as little as 3 h exposure to x-ray film. Expression of the activin β_c gene by other tissues (heart, brain, spleen, lung, kidney, testis, and skeletal muscle) was not detected even if the blot was exposed to x-ray film for as long as 3 days (not shown) . The activin β_E gene was also highly expressed in liver. Re- hybridization of the same multiple tissue Northern blot with as specific probe from the 3 ' untranslated region of activin β_E identified a single liver transcript of -3.2-kb. This band was also identified with as little as 3 h exposure to x-ray film, and expression in other tissues was not detected even after exposure to x-ray film for up to 3 days (not shown) .

The murine activin β_E clone was used as a probe to screen 500,000 plaque forming units from a human liver cDNA library underlow stringency conditions. Two overlapping cDNA clones (representing a sequence of 1,764-bp) were purified from a pool of more than 20 positive isolates. Sequence analysis revealed a predicted open reading frame of 1050 nucleotides coding for a protein with an estimated molecular weight of 39 kD and 1 potential N-linked glycosylation site (Fig. 5) . A MET initiator codon (beginning with nucleotide 77) is believed to be the site for translation initiation. A canonical CXCX sequence, preceding the stop translation signal (TAG) , was also identified. A polyadenylation signal has yet to be identified downstream of the stop codon. Sequence comparison analysis suggests that the 1050-bp open reading frame sequence codes for human activin β_E .

The human activin β_E subunit is organized like other TGF- β superfamily members. The predicted initiator MET residue is followed by a characteristic signal peptide of -20 amino acids. Downstream of the signal peptide, a predicted propeptide and mature growth factor domains have been identified. The domains are separated by a cluster of basic amino acids, creating one or more putative endoproteolytic cleavage sites. Cleavage at these sites will produce a propeptide and mature growth factor of -215 and 114 amino acids, respectively.

The predicted carboxy-terminal amino acid sequences of mouse and human activin β_E are highly homologous, with >95% amino acid identity over the mature growth factor region. Both coding sequences contain a total of nine cysteine residues. The spacing of the cysteines and the presence of other conserved amino acids is most characteristic of the activins and to a lesser extent, TGF-β itself. In tabulating percent amino acid sequence identity within the mature growth factor region, human activin β_E showed 45% identity with human activin β_A, 47% identity with human activin β_B, 64% identity with human activin β_c, and 62% identity with Xenopus activin β_D. In contrast, only 20-40% amino acid sequence identity was found when human activin β_E was compared with other members of the TGF-β superfamily.

This example reports the isolation and characterization of two murine and one human cDNA that code for members of the TGF-β superfamily. The cDNAs, which are closely related, code for β subunits of activin. One of these subunits is likely to be the mouse homolog of human activin β_c, while the second and third are novel and have been designated the mouse and human activin β_E subunit. Taken together, the high degree of sequence identity between activin β_c and β_E , their clustered genomic organization, and their specific and strong expression pattern in adult liver suggest that these genes evolved from a common ancestor that underwent a tandem duplication. These data also suggest that the activin β_c and β_E subunits may represent a unique subgroup of TGF-β-like molecules. To date, a total of five activin/inhibin β subunits and two inhibin subunits have been isolated and characterized. Whereas the activins were originally isolated from gonadal fluids by measuring the regulation of FSH release from primary rat anterior pituitary cells in vitro (Vale et al., 1986, Nature 321:776-779; Ling et al., 1986, Nature 321:779- 782; Ling et al., 1986, Biochem Biophys Res Commun 138:1129- 1137) , activins have now been independently purified based on a number of different activities, including, but not limited to, erythroid differentiation (Yu et al., 1987, Nature 330:765-767; Eto et al., Biochem Biophys Res Commun 142:1095- 1103), nerve cell survival (Schubert, D. , et al., 1990, Nature 344:868-870), mesoderm induction in Xenopus laevis embryos (Slack J.M. , 1994, Curr Biol 4:116-126), megakaryocyte differentiation (Fugimoto, K. , 1991, Biochem. Biophys. Res. Commun. 174:1163-1168), cardiac morphogenesis (Olson, E., et al,. 1996, Science 272:671-676), bone growth (Luyten, F., et al., 1994, Exp. Cell Res. 210:224-229; Ogawa, Y., et al., 1992, J. Biol. Chem. 267:2325-2328), and somatostatin induction (Woodruff, T. et al., 1995, Annu. Rev. Physiol. 57:219-244). Three activins (A, B, and AB) have been well characterized to date, each consisting of different combinations of β subunits (β_Aβ_A, β_Bβ_B , and β_Aβ_B , respectively) that assemble into disulfide linked dimers. It is not yet clear if the activin β_c, β_D, and β_E subunits assemble into homodimers, heterodimers, or both. The recent isolation of these subunits suggests, however, that the number of combinatorial possibilities for activin may in fact be relatively large. On the other hand, inhibin was initially identified as a gonadal peptide that inhibited FSH release from the anterior pituitary (Vale et al., 1990, The activins and inhibins. In: Sporn MB, Roberts AB (eds) Peptide Growth Factors and Their Receptors, 95 (Part II) . Springer-Verlag, Heidelberg, pp. 211-248; Mason et al., 1985, Nature 318:659- 663; Rivier, J. , et al., 1985, Biochem. Biophys. Res. Commun. 133:120-127). The major gonadal sites of inhibin synthesis are Sertoli and granulosa cells. Inhibin A and B are also expressed in the placenta, and in the adult spleen and nervous system. In mice of both genders, inhibin appears to be an important negative regulator of gonadal stromal cell proliferation—i.e. , inhibin-deficient mice develop mixed or incompletely differentiated gonadal stromal tumors that consist of granulosa cells (female mice) and Sertoli cells (male mice) (Matzuk, M. , et al., 1992, Nature 360:313-319).

7. EXAMPLE: DIRECT TRANSFER AND EXPRESSION OF

ACTIVIN ft, EXPRESSION PLASMID DNA CAUSES NEW BONE FORMATION IN VIVO

Direct in vivo gene transfer was used to determine if activin β_c homodimers possess osteoinductive activity. Degradable matrices containing plasmid DNA (gene-activated matrices, or GAMs) were implanted into segmental gaps in the adult rat femur. Implantation of a GAM containing activin β_c expression plasmid DNA resulted in the biological response of new bone filling the gap. This experiment demonstrates that overexpression of activin β_c causes new bone to form in vivo.

7.1 MATERIALS AND METHODS Animal model. To create a 5 mm osteotomy, four 1.2 mm diameter pins were screwed into the femoral diaphysis of normal adult Sprague-Dawley rats under general anesthesia and with constant irrigation. A surgical template guided parallel pin placement, which was confirmed by fluorography (pins were set 3.5 mm from the edge of the fixator plate and 2.5 mm apart). An external fixator plate (30 x 10 x 5 mm) was then secured on the pins. External fixator plates were fabricated with aluminum alloy on a CNC mill to ensure high tolerances. Prefabricated fasteners with associated lockwashers and threaded pins were made of stainless steel. All fixator parts were sterilized with ethylene oxide gas prior to surgery. Five mm segmental defects were created at mid-shaft with a Hall Micro 100 oscillating saw (Zimmer Inc., Warsaw, IN) . Collagen sponges were held in the osteotomy gap until surrounded by clotted blood; preliminary studies showed that this maneuver fixed the sponge within the osteotomy site. The skin incision was closed with staples. The fixator provided the necessary stability so that animal ambulation was unlimited.

Preparation of Plasmid DNA. An established bacterial cell strain that contains an activin β_c expression plasmid was grown in liquid culture according to established protocols. Plasmids were purified in an endotoxin-free manner using reagents and protocols from Qiagen Inc.

Preparation of Gene-Activated Collagen Sponges. To 1.5 ml of a PLGA/chloroform solution (3% w/v) , 50/50 polylactic polyglycolic acid co-polymer, average MW 90,000, inherent viscosity 1.07) was added 0.2 ml of a solution containing expression plasmid DNA encoding activin β_c (1 mg DNA, 10 mM Tris, 0.1 mM EDTA, pH 7.4). The solution was emulsified by vortexing for 2 minutes followed by sonicating for 30 seconds at about 0°C using a microtip probe-type sonicator at 55 Watts output .

Radiography. Weekly plain film radiographs (posterior- anterior view) were obtained while animals were awake using a portable X-ray unit (GE, model 100) .

Histology. Tissues were prepared for light microscopy using standard methods of tissue fixation, demineralization, paraffin embedding, sectioning, and staining.

7.2 RESULTS

Osteotomy Model. Whereas osteotomy repair in the rat is completed by 9 weeks post-surgery, the manner of repair depends on the size of the gap: a 2 mm gap heals by bony union, but a 5 mm gap heals by fibrous nonunion. Control animals maintained for up to 13 weeks post-surgery confirmed the observation that 5 mm gaps typically heal by fibrous nonunion. Weekly plain film radiography and histology demonstrated that bone did not form in animals that received either a 5 mm osteotomy alone (n=3) , a 5 mm osteotomy plus a PLGA matrix (n=10) , or a 5 mm osteotomy plus a PLGA matrix containing plasmid DNA (n=23) . All 36 control gaps healed by deposition of fibrous tissue. Control femurs exhibited focal periosteal new bone formation (a complication of pin placement) . A focal, transient inflammatory response (lymphocytes and macrophages) in gap tissues was also observed post-surgery.

Activin ft-. Gene Transfer. A full length mouse activin β_c cDNA was isolated by library screening and subcloned into the pcDNA3 eukaryotic expression vector (Invitrogen) . To specifically detect recombinant proteins, the 3' end of the activin β_c coding sequence was modified by addition of a hemagglutinin (HA) epitope. Recombinant activin β_c was expressed from this construct (pGAM3) using an in vitro transcription and translation protocol. Immunoprecipitation studies also established the ability of the HA epitope to be recognized by an anti-HA polyclonal antibody.

Biosynthesis of recombinant β_c was evaluated following transient transfection of cultured 293T cells with pGAM3 plasmid DNA: as demonstrated by immunoprecipitation, activin β_c molecules were assembled into homodimers, secreted, and processed as expected. FIG. 6.

PLGA matrices containing pGAM3 DNA were placed in the gap of three adult rats maintained for 4-24 weeks. Microscopic foci of new bone, originating from both surgical margins, were observed as early as 4 weeks after surgery. Consistent with a classic description of bone formation by autoinduction, these foci consisted of bony plates surfaced by large cuboidal osteoblasts and supported by a cellular connective tissue composed of pleomorphic spindled fibroblasts and capillary vessels. The radiographic appearance of the contralateral (unoperated) femur was unchanged in all three cases (not shown) , implying that the effects of gene transfer and activin β_c overexpression were limited to the osteotomy gap. These data demonstrate that activin β_c homodimers possess osteoinductive activity.

8. EXAMPLE: RECOMBINANT ACTIVIN β_c HOMODIMERS

STIMULATE HEMATOPOIESIS IN VIVO

Human bone marrow biopsies were cultured in the presence of media conditioned with recombinant activin β_c to determine the effect, if any, of activin β_c homodimers on hematopoiesis.

The experiment demonstrates that activin β_c homodimers are capable of stimulating hematopoiesis.

8.1 MATERIALS AND METHODS Production of Activin ft, Conditioned Media. Transient transfection was performed as described using pGAM3 plasmid DNA, 293T cells, and standard protocols. On the day of transfection, sub-confluent 293T cells were fed fresh Dulbecco's Modified Eagles 's Medium supplemented with 10% fetal calf serum (reagents from Gibco-BRL Life Technologies, Inc.). Ten μg of Plasmid DNA were added to 0.5 ml of 0.25 M CaCl₂ solution, and the mixture was added drop-wise to 0.5 ml Hepes-buffered saline (270 mM NaCI, 1.5 mM NaP0₄, 50 mM Hepes, pH 7.05). This mixture was kept at room temperature for 10 min and then added drop-wise to the medium of freshly fed 293T cells. After an overnight incubation at 37°C, the medium was replaced and cells were cultured for an additional 48 h. For metabolic labeling studies, transfected cells were re-fed with Dulbecco's Modified Eagle's Medium lacking methionine and cysteine and supplemented with 5% dialyzed fetal calf serum (all reagents from Gibco-BRL Life Technologies, Inc.) and 200 mCi[³⁵S] Met and [³⁵S] Cys (Translabel Cys and Met, ICN) . Cells were incubated for an additional 24 h at 37 °C, medium samples were collected in the presence of protease inhibitors (Leupeptin, I mg/ml, Aprotinin, 1 mg/ml, Soybean trypsin inhibitor, 10 mg/ml, and Pepstatin, 1 mg/ml; Sigma), and stored at 4°C until use.

Human Bone Marrow Cultures. Human bone marrow aspirates were obtained from normal volunteers after informed consent. Long-term bone marrow cultures were established exactly as described by Gartner and Kaplan (1980, Proc. Natl. Acad. Sci. USA, 4756-4759). For immunolocalization, two to three coverslips were placed in each culture and removed weekly for immunocytochemistry . Determination of hematopoietic progenitor cell frequency and the total number of hematopoietic progenitor cells was performed as described previously (Long et al., 1990, J. Clin. Invest. 86:1387- 1395) . Assays were performed in triplicate.

8.2 RESULTS Activin ft Conditioned Media Stimulates Hematopoiesis.

Strong trends or statistically significant stimulation of erythropoiesis, granulopoiesis, and megakaryopoiesis were observed when activin β_c conditioned media was added to human marrow aspirate cultures (FIG. 9) . In addition, the activin β_c conditioned media caused a two-fold increase in the total number of hematopoietic progenitor cells. These data demonstrate that activin β_c homodimers are capable of stimulating hematopoiesis.

9. EXAMPLE: ACTIVIN β_c HOMODIMERS BIND

LATENT TGF-β BINDING PROTEIN-3

The latent TGF-β binding protein (LTBP) is a component of the large latent TGF-β complex produced by platelets and other cells. Genes coding for three LTBPs have been isolated to date; two gene products (LTBP-1 and LTBP-2) have been shown to bind latent TGF-βl via a disulfide bond, whereas

LTBP-3 was isolated only recently and its binding potential has yet to be determined. To address this question, we transiently co-transfected cultured cells with expression plasmids coding for mouse LTBP-3 and human TGF-ftl, and studied complex formation using immunoprecipitation, SDS-PAGE and autoradiography (Yin et al., submitted). The results indicated that LTBP-3 also covalently binds latent TGF-ftl during assembly of the latent complex. The co-transfection strategy was also used to show for the first time that the site of disulfide bond formation is the 8-cysteine TGF-bp structural motif in LTBP-3) . These results were both confirmed and extended by additional co-transfection studies that employed TGF-bp expression plasmids from mouse LTBP-2. Together, these studies identify a new functional role for the TGF-bp motif during assembly of latent TGF-ftl complexes. The results have important implications for TGF-β biosynthesis, storage, and targeting. LTBP-3 has a total of three TGF-bp structural motifs, only two of which bind TGF-ftl with affinity. We therefore sought to determine if the third structural motif was able to bind related molecule—i.e. , other members of the TGF-β superfamily. In particular, we sought to determine if TGF-β- like molecules with propeptides that have a conserved pattern of cysteine residue number and spacing would also bind LTBP- 3. To address this question, a similar experimental design was employed: we transiently co-transfected cultured cells with expression plasmids coding for mouse LTBP-3 and activin /3_C, and studied complex formation using immunoprecipitation, SDS-PAGE and autoradiography. The experiment demonstrates that the third 8-cysteine TGF-bp structural motif of LTBP-3 covalently binds activin β_c. The results demonstrate for the first time that LTBP-3 is a binding protein for activin PC homodimers, and they extend the functional role for the TGF- bp motif during activin β_c homodimer assembly.

9.1 MATERIALS AND METHODS Assembly of pLTBP-3 fl. Yin et al. (1995) assembled and characterized a mammalian expression vector that codes for mouse LTBP-3 (pLTBP-3 fl; see Table 1, #1) . Following transient transfection, this study showed that mouse LTBP-3 is precipitated by polyclonal antiserum #274, which recognizes a 12 amino acid epitope (GESVASKHAIYA) within the upstream, proline- and glycine-rich sequence of mouse LTBP-3.

Assembly of pLTBP-3/pept. Yin et al. assembled and characterized the pLTBP-3/pept expression vector. To this end, pLTBP-3 fl was digested with EcoRI and Hindlll to yield two DNA fragments that were cleanly resolved by agarose gel electrophoresis. One fragment, consisting of an unwanted portion of LTBP-3 coding sequence, was discarded. The second fragment—which included the intact plasmid backbone, CMV promoter, stop translation signal, and polyadenylation element—was designated pLTBP-3/pept (see Table 1, #2). The Hindlll site is located 646 bp downstream of the Ltbp-3 initiator Met codon. Therefore, pLTBP-3/pept codes for an amino-terminal fragment of mouse LTBP-3 that includes the normal Kozak consensus and signal peptide; a short amino- terminal sequence unique to LTBP-3; two EGF-like repeats; and a portion of the upstream proline- and glycine-rich domain. The latter includes the antibody #274 epitope (GESVASKHAIYA) . As described below, the pLTBP-3/pept expression vector functioned in 293T cells in essentially the same way as pLTBP-3 fl—i.e., pLTBP-3/pept was used to generate a series of defined LTBP-3 peptide fragments whose potential to bind TGF-^βl could be directly tested by immunoprecipitation with antibody #274.

Assembly of pLTBP-3/TGF-bpA. pLTBP-3/pept facilitated the assembly of a plasmid encoding a defined fragment of mouse LTBP-3. The following is a detailed description of how this plasmid, pLTBP-3/TGF-bpA, was assembled (see Table 1, #3; also see Fig. 1). PCR was used to generate a cDNA fragment that coded for the TGF-bpA structural motif from LTBP-3. The upstream PCR oligonucleotide primer contained a Nhel site in frame with 6/12 codons of the 274 peptide

(KHAIYA) plus the 5 ' portion of the cDNA sequence that coded for the EGF-like repeats (the sequence was obtained from GenBank™/EMBL Data Bank accession L40459) . The downstream PCR primer contained an Xhol site plus the 3 ' portion of the TGF-bpA cDNA sequence. The PCR template consisted of pLTBP-3 fl plasmid DNA, and PCR was conducted as described (Chen et al., 1993). Following amplification, the PCR fragment was digested with Nhel and Xhol, purified by agarose gel electrophoresis, and ligated with the pLTBP-3/pept expression vector after it had been digested with the same restriction enzymes and purified. Digestion with Xhol preserved the stop translation and polyadenylation elements within the vector. Competent bacteria were transformed by the ligation mixture, and colony forming units were isolated and used to prepare plasmid DNA. To insure its integrity, pLTBP-3/TGF-bpA was characterized by DNA sequencing. Assembly of pGAM3 and transfection were performed as described above.

Immunoprecipita ion. For immunoprecipitation, 10 ml (neat) of antibody was added to 1-5 ml of radiolabelled medium proteins. The mixture was incubated with shaking at room temperature for 2 h. Protein A-Sepharose CL-4B beads were added (200 ml, 10% suspension) , and this mixture was incubated with shaking for 2 h at room temperature. Immunoprecipitated proteins were pelleted by brief centrifugation, and the pellet was washed 4 times with PBS-TD buffer (0.38 mM NaCI, 2.7 mM KCI, 8.1 mM Na₂HP0₄ , 1.5 mM

KH₂P0₄, 1% Triton X-100, 0.5% deoxycholate) and 3 times with PBS. Protein loading dye (2x) was then added, with or without reducing agent (10% 2-mercaptoethanol) , and the samples were boiled for 5 min. Samples were then fractionated using 4-18% gradient SDS-polyacrylamide gel electrophoresis (Yin et al., 1995). Cold molecular weight markers (200-14.3 kDa, Rainbow mix, Amersham Corp.) were used to estimate molecular weight. The gel was dried and exposed to film for the indicated time at room temperature. 9 . 2 RESULTS

293T mammalian cells were programmed by transient co- transfection with pLTBP-3/TGF-bpA and pGAM3 , and then radiolabelled with [³S] Cys and [³⁵S] Met for 16 h. Medium proteins were harvested, and immunoprecipitation studies were carried out.

Neither LTBP-3 nor activin β_c were identified in conditioned medium of untransfected 293T cells that were radiolabelled and immunoprecipitated with the anti-LTBP-3 and anti-HA antibodies. Following immunoprecipitation with the anti-HA antibody, conditioned medium of cells transfected with pGAM3 alone contained the activin propeptide and mature growth factor but not the LTBP-3 TGF-bpA repeat. Under reducing conditions, the propeptide monomer migrated as a single band of -40 kDa, and mature activin β_c migrated as a single band of -14 kDa. Following immunoprecipitation with the anti-HA antibody, the conditioned medium of 293T cells transfected with pLTBP3/TGF-bpA alone contained the bpA motif (a peptide of -30 kDa) but not the latent form of activin β_c . A final set of controls was based on immunoprecipitation of conditioned media with the 'wrong antibody¹: activin β_c could not be identified if conditioned medium was immunoprecipitated with the anti-LTBP-3 antibody, and, conversely, the bpA motif could not be identified if the medium sample from conditioned medium of 293T cells transfected with pLTBP3/TGF-bpA was immunoprecipitated with the anti-HA antiserum. These control data have been obtained independently three times, attesting to the reproducibility of the transient co-transfection system. Co-transfection experiments provided evidence of a specific interaction between the latent form of activin β_c and the TGF bpA structural motif of LTBP-3. For both the anti- LTBP-3 and anti-HA immunoprecipitates, the bpA motif again migrated as a band of -30 kDa, the activin β_c propeptide migrated as a dispersed band of -40 kDa, and mature activin β_c migrated as a single band of -14 kDa.

Ill Therefore, the data establishes that LTBP-3 binds activin β_c homodimers via a covalent reducible crosslink —i.e., a disulfide bond that is stable to boiling in SDS. Moreover, co-immunoprecipitation by both anti-LTBP-3 and anti-HA antibodies provides strong evidence that the binding interaction between LTBP-3 and activin β_c was specific and, therefore, not due to an artifact of the transient co- transfection system.

10. DEPOSIT OF MICROORGANISMS

The following organisms were deposited with the American Type Culture Collection (ATCC) , 12301 Parklawn Drive, Rockville, Maryland 20852.

Strain Designation Containing Accession No. pbHE human activin β_E pbC murine activin β_c pbE murine activin β_E

The present invention is not to be limited in scope by the exemplified embodiments or deposited organisms which are intended as illustrations of single aspects of the invention, and any clones, DNA or amino acid sequences which are functionally equivalent are within the scope of the invention. Indeed, various modifications of the invention in addition to those described herein will become apparent to those skilled in the art from the foregoing description and accompanying drawings. Such modifications are intended to fall within the scope of the appended claims.

It is also to be understood that all base pair sizes given for nucleotides are approximate and are used for purposes of description. MICROORGANISMS

Optional Sheet in connection with the microorganism referred to on page 1 1₂ , lines 10-35 of the description

A. IDENTIFICATION OF DEPOSIT "

Further deposits are identified on an additional sheet

Name of depositary institution ' American Type Culture Collection

Address of depositary institution (including postal code and country)

12301 Parklawn Drive ockville, MD 20852 US

Date of deposit ' November 20, 1996 Accession Number ' 97804

B. ADDITIONAL INDICATIONS ' (leave blank if not applicable) This information is continued on a separate attached sheet

C. DESIGNATED STATES FOR WHICH INDICATIONS ARE MADE " _(if.ι»ω«__J»«.«- *_w««is_U-»)

D. SEPARATE FURNISHING OF INDICATIONS ' (leave blank if not applicable)

The indications listed below will be submitted to the International Bureau later ' (Specify the general nature of the indications e g "Accession Number of Deposit")

E. D This sheet was received with the International application when filed (to be checked by the receiving Office)

(Authorized Officer)

D The date of receipt (from the applicant) by the International Bureau "

(Authorized Officer)

Form PCT/RO/134 (January 1981 )

1 12.1 Form PCT/RO/134 (cont.)

American Type Culture Collection

12301 Parklawn Drive Rockville. MD 20852 US

Accession No. Date of Deposit

97805 November 20, 1996

97806 November 20, 1996

1 12.2

Claims

WHAT IS CLAIMED IS:

1. An isolated nucleic acid molecule containing the nucleotide sequence of liver activin 3_E gene.

2. An isolated nucleic acid molecule which encodes a liver activin β__{E I} or a fragment thereof, having a nucleotide sequence that:

(a) encodes the amino acid sequence shown in

FIG. 2 or the amino acid sequence encoded by the cDNA contained in cDNA clone as deposited with the ATCC having accession No. ; or

(b) hybridizes under stringent conditions to the nucleotide sequence of (a) or to its complement.

3. An isolated nucleotide sequence encoding a liver activin polypeptide corresponding to the N-terminal signal peptide, propeptide, or mature growth factor domain or the mature growth factor domain is deleted.

4. An isolated nucleotide sequence encoding a chimeric protein comprising the polypeptide of Claim 3 fused to a heterologous polypeptide.

5. A nucleotide vector containing the nucleotide sequence of Claim 1, 2, 3, or 4.

6. An expression vector containing the nucleotide sequence of Claim 1, 2, 3, or 4 in operative association with a nucleotide regulatory sequence that controls expression of the nucleotide sequence in a host cell.

7. An expression vector containing the nucleotide sequence shown in FIG. 1, or encoding the amino acid sequence shown in FIG. 1 or the amino acid sequence encoded by the cDNA contained in cDNA clone as deposited with the

ATCC having accession No. ; in operative association

113 with a nucleotide regulatory sequence that controls expression of the nucleotide sequence in a host cell.

8. The expression vector of Claim 6, in which said regulatory sequence is selected from the group consisting of the cytomegalovirus hCMV immediate early gene, the early or late promoters of SV40 adenovirus, the lac system, the trp system, the TAC system, the TRC system, the major operator and promoter regions of phage λ, the control regions of fd coat protein, the promoter for 3-phosphoglycerate kinase, the promoters of acid phosphatase, and the promoters of the yeast α-mating factors.

9. The expression vector of Claim 7, in which said regulatory sequence is selected from the group consisting of the cytomegalovirus hCMV immediate early gene, the early or late promoters of SV40 adenovirus, the lac system, the trp system, the TAC system, the TRC system, the major operator and promoter regions of phage λ, the control regions of fd coat protein, the promoter for 3-phosphoglycerate kinase, the promoters of acid phosphatase, and the promoters of the yeast α-mating factors.

10. A genetically engineered host cell that contains the nucleotide sequence of Claim 1, 2, 3, or 4.

11. A genetically engineered host cell that contains the nucleotide sequence of Claim 1, 2, 3 , or 4 in operative association with a nucleotide regulatory sequence that controls expression of the nucleotide sequence in the host cell.

12. The genetically engineered host cell of Claim 11 in which the host cell is a hepatocyte.

13. A genetically engineered host cell that contains the expression vector of Claim 7.

114

14. The genetically engineered host cell of Claim 13 in which the host cell is a hepatocyte.

15. An isolated liver activin protein.

16. An isolated liver activin protein having the amino acid sequence shown in FIG. 1 or 2 , or the amino acid sequence encoded by the cDNA contained in cDNA clone as deposited with the ATCC having accession No. , or the amino acid sequence encoded by the cDNA contained in cDNA clone as deposited with the ATCC having accession No.

17. A polypeptide having an amino acid sequence corresponding to the N-terminal signal peptide, propeptide, mature growth factor domain of liver activin protein, or a deletion mutant of the liver activin protein in which the N- terminal signal peptide, propeptide, or the mature growth factor domain is deleted.

18. A chimeric protein comprising the polypeptide of Claim 17 fused to a heterologous polypeptide.

19. An antibody that immunospecifically binds the liver activin protein of Claim 16 or 17.

20. An antibody that immunospecifically binds the polypeptide of Claim 17.

21. A method for diagnosing cell growth or differentiation disorders in a vertebrate comprising measuring liver activin gene expression in a patient sample.

22. The method of Claim 21 in which expression is measured by detecting mRNA transcripts of the liver activin gene.

115

23. The method of Claim 21 in which expression is measured by detecting the liver activin gene product.

24. A method for diagnosing cell growth or differentiation disorders in a vertebrate, comprising detecting a liver activin gene mutation contained in the genome of the vertebrate.

25. A method for regulating cell growth or differentiation in a vertebrate, comprising administering liver activin compound to the vertebrate in an amount sufficient to stimulate cell growth or differentiation.

26. The method of Claim 25 in which the compound is a polypeptide corresponding to the mature growth factor domain of the liver activin or a portion of the mature growth factor domain that binds liver activin receptor, a deletion mutant liver activin protein lacking the endoproteolytic motif, propeptide or mature growth factor domain, or a chimeric fusion protein comprising the mature growth factor domain of the liver activin, or a portion of the mature growth factor domain that binds liver activin receptor deletion mutant fused to a heterologous polypeptide.

27. The method of Claim 25 in which the compound is an antagonist that binds to the liver activin receptor and inhibits activation of the receptor.

28. The method of Claim 25 in which the compound binds to endogenous liver activin and neutralizes liver activin activity.

29. The method of Claim 28 in which the compound is delivered to the vertebrate by administering a genetically engineered host cell that expresses and secretes the polypeptide or fusion protein in the vertebrate.

116

30. The method of Claim 28 in which the compound is an anti-idiotypic antibody, or an Fab portion thereof, that mimics the mature growth factor domain of the liver activin and neutralizes endogenous liver activin receptor.

31. A method for regulating cell growth or differentiation in a vertebrate, comprising administering a compound to the mammal in an amount sufficient to inhibit expression of the liver activin in vivo.

32. The method of Claim 31 in which the compound is delivered to the liver.

33. The method of Claim 31 in which the compound is an antisense oligonucleotide that inhibits translation of mRNA transcripts that encode the liver activin.

34. The method of Claim 31 in which the compound is a ribozyme that inhibits translation of mRNA transcripts that encode the liver activin.

35. The method of Claim 31 in which the compound is an oligonucleotide that forms a triple helix with the regulatory region of the liver activin gene and inhibits transcription.

36. The method of Claim 31 in which the compound is a recombinant DNA construct that inactivates the liver activin gene or its regulatory region via targeted homologous recombination.

37. A method for regulating cell growth or differentiation in a vertebrate, comprising administering a compound to the vertebrate in an amount sufficient to inhibit signal transduction induced by binding of liver activin to an activin receptor.

117

38. A method for treating liver disorder, comprising administering an effective amount of liver activin gene product antagonist to a patient in need of such therapy.

39. A method for regulating cell growth or differentiation, comprising administering a compound to a vertebrate in an amount sufficient to up regulate expression of a liver activin in the vertebrate.

40. The method according to Claim 42 in which the compound is delivered to the liver.

41. The method of Claim 42 in which the vertebrate expresses a defective liver activin and the compound comprises a nucleotide construct encoding a functional liver activin controlled by a regulatory region that directs expression of the functional liver activin in target cells in the vertebrate.

42. The method of Claim 42 in which the vertebrate expresses a mutant liver activin and the compound comprises a nucleotide construct encoding a wild-type liver activin that corrects the endogenous mutation via targeted homologous recombination.

43. A method of enhancing the growth or regeneration of liver tissue comprising treating the liver tissue with an effective amount of liver activin gene product agonist.

44. The method of Claim 39 in which the liver tissue is extracorporeal .

45. The method of Claim 39 in which the liver tissue is intracorporeal .

118

46. A method for inducing bone growth in a vertebrate comprising administering an osteogenically effective amount of liver activin compound to the vertebrate.

47. The method of Claim 46 in which the compound comprises a polypeptide corresponding to the mature growth factor domain of the liver activin or a portion of the mature growth factor domain that binds liver activin receptor, a deletion mutant liver activin protein lacking the endoproteolytic motif, propeptide or mature growth factor domain, or a chimeric fusion protein comprising the mature growth factor domain of the liver activin, or a portion of the mature growth factor domain that binds liver activin receptor deletion mutant fused to a heterologous polypeptide.

48. The method of Claim 46 further comprising a compound selected from the group consisting of a TGF-3, a BMP, and bone marrow.

49. The method of Claim 46, wherein the bone growth is systemic.

50. The method of Claim 46, wherein the bone growth is local.

51. A method for stimulating hematopoiesis in a vertebrate comprising administering a hematopoietically effective amount of liver activin compound to the vertebrate.

52. The method of Claim 51 in which the compound comprises a polypeptide corresponding to the mature growth factor domain of the liver activin or a portion of the mature growth factor domain that binds liver activin receptor, a deletion mutant liver activin protein lacking the endoproteolytic motif, propeptide or mature growth factor domain, or a chimeric fusion protein comprising the mature growth factor domain of the liver activin, or a portion of

119 the mature growth factor domain that binds liver activin receptor deletion mutant fused to a heterologous polypeptide.

53. The method of Claim 51 further comprising a compound selected from the group consisting of an iron preparation, vitamin B12 , folic acid, an adrenocortical steroid, an erythropoietin, a testosterone, a progenitor cell stimulator, insulin-like growth factor, a prostaglandin, serotonin, cyclic AMP, prolactin, triidothyzonine, methenolene, stanozolol, nandrolone, and androgen and a erythrogenin.

54. The method of Claim 51 wherein the compound contains erthropoietin.

55. The method of Claim 51 wherein the compound stimulates erythropoiesis.

120