Binding agent comprising at least one peptide capable of binding myostatin, a polynucleotide sequence encoding the binding agent, an expression vector comprising the polynucleotide, a host cell comprising the expression vector, a pharmaceutical composition comprising the binding agent and uses thereof, a method of detecting or measuring myostatin in a sample and a method of diagnosing a myostatin related disorder in a subject

205265/2

A BINDING AGENT COMPRISING AT LEAST ONE PEPTIDE CAPABLE OF BINDING MYOSTATIN, A POLYNUCLEOTIDE SEQUENCE ENCODING THE

BINDING AGENT, AN EXPRESSION VECTOR COMPRISING THE POLYNUCLEOTIDE, A HOST CELL COMPRISING THE EXPRESSION VECTOR, A

5 PHARMACEUTICAL COMPOSITION COMPRISING THE BINDING AGENT AND USES THEREOF, A METHOD OF DETECTING OR MEASURING MYOSTATIN IN A

SAMPLE, AND A METHOD OF DIAGNOSING A MYOSTATIN RELATED DISORDER IN A SUBJECT

10 FIELD OF THE INVENTION

The present application is a divisional application of Israel Patent No. 169168 having the filing date of December 19, 2003. IL 169168 relates to a binding agent which binds myostatin and inhibits its activity. The present invention relates to further binding agents which bind myostatin and inhibit its activity. 15

BACKGROUND

Myostatin, also known as growth/differentiation factor 8 (GDF-8), is a transforming growth factor-β (TGF-β) family member known to be involved in regulation of skeletal muscle mass. Most members of the TGF^-GDF family are expressed non-specifically in many tissue 20 types and exert a variety of pleitrophic actions. However, myostatin is largely expressed in the cells of developing and adult skeletal muscle tissue and plays an essential role in negatively controlling skeletal muscle growth (McPherron et al. Nature (London) 387, 83-90 (1997)). Recent studies, however, indicate that low levels of myostatin expression can be measured in cardiac, adipose and pre-adipose tissues.

25 The myostatin protein has been highly conserved evolutionarily (McPherron et al. PNAS USA 94:12457-12461 (1997)). The biologically active C- terminal region of myostatin has 100 percent sequence identity between human, mouse, rat, cow, chicken, and turkey sequences. The function of myostatin also appears to be conserved across species as well. This is evident from the phenotypes of animals having a mutation in the myostatin gene. Two breeds of cattle, the 30 Belgian Blue (Hansel R., Muscle Hypertrophy of Genetic Origin and its Use to Improve Beef Production, eds, King, J.W.G. & Menissier, F. (Nijhoff, The Hague, The Netherlands) pp. 437-449) and the Piedmontese (Masoero, G. & Ponjardieu, B. Muscle Hypertrophy of Genetic Origin and its Use to Improve Beef Production, eds, King, J.W.G. & Menissier, F. (Nijhoff, The Hague, The Netherlands) pp. 450-459) are characterized by a "double muscling" phenotype and 35 increase in muscle mass. These breeds were shown to contain mutations in the coding region of the myostatin gene (McPherron et al. (1997) supra). In addition, mice containing a targeted deletion of the gene encoding myostatin (Mstn) demonstrate a dramatic increase in muscle mass without a corresponding increase in fat. Individual muscles of Mstn7" mice weigh approximately 100 to 200 percent more than those of control animals as a result of muscle fiber hypertrophy 40 and hyperplasia (Zimmers et al. Science 296,1486 (2002)). 1 205265/2

Administration of myostatin to certain strains of mice has been shown to create a condition similar to muscle wasting disorders found associated with cancer, AIDS, and muscular dystrophy, for example. Myostatin administered as myostatin-producing CHO cells to athymic nude mice resulted in a wasting effect with a high degree of weight loss, a decrease of as much 5 as 50% of skeletal muscle mass in addition to fat wasting, and severe hypoglycemia ( Zimmers et al. supra).

Loss of myostatin appears to result in the retention of muscle mass and reduction in fat accumulation with aging. It has been shown that age-related increases in adipose tissue mass and decrease in muscle mass were proportional to myostatin levels, as determined by a 10 comparison of fat and muscle mass in Mstn +/+ when compared with Mstn''' adult knockout mice (McFerron et al. J. Clin. Invest 109, 595 (2002)). Mstn''' mice showed decreased fat accumulation with age compared with Mstn +/+ mice.

In addition myostatin may play a role in maintaining blood glucose levels and may influence the development of diabetes in certain cases. It is known that, for example, skeletal 15 muscle resistance to insulin-stimulated glucose uptake is the earliest known manifestation of non-insulin-dependent (type 2) diabetes mellitus (Corregan et al. Endocrinology 128:1682 (1991)). It has now been shown that the lack of myostatin partially attenuates the obese and diabetes phenotypes of two mouse models, the agouti lethal yellow (Ay) (Yen et al. FASEB J. 8:479 (1994)), and obese (Lepob/ob). Fat accumulation and total body weight of the Ay/a, Mstn 20 double mutant mouse was dramatically reduced compared with the Ay/a Mstn +/+ mouse (McFerron et al., (2002) supra). In addition, blood glucose levels in the Ay/a, Mstn mice was dramatically lower than in Ay/aMstn +/+ mice following exogenous glucose load, indicating that the lack of myostatin improved glucose metabolism. Similarly Lepob/ob Mstnmice showed decreased fat accumulation when compared with the Lepob/ob Mstn +/+phenotype. 25 Therefore, there is considerable evidence from the phenotypes of over-expressing and knockout animals that myostatin may play a role in contributing to a number of metabolic disorders including disorders resulting in muscle wasting, diabetes, obesity and hyperglycemia.

Hill et al (2002) JBC 277:40735 "The myostatin propeptide and the follistatin-related gene are inhibitory binding proteins of myostatin in normal serum" describes in vivo binding of 30 myostatin to myostatin propeptide and FLRG.

Hill et al (2003) Mol Endocrin 17(4):1144 "Regulation of Myostatin in Vivo by Growth and Differentiation Factor-Associated Serum Protein-1: A Novel Protein with Protease Inhibitor and Follistatin Domains" describes identification of an additional protein associated with endogenous myostatin in normal mouse and human serum, named growth and differentiation 35 factor-associated serum protein-1 (GASP-1). WO 00/43781 "Growth differentiation factor inhibitors and uses thereof' describes inhibitors of GDF-8 or GDF-11. 2 205265/2

SUMMARY OF THE INVENTION

The present invention is directed to binding agents which bind myostatin and inhibit its activity. The binding agents comprise at least one peptide capable of binding myostatin. The myostatin-binding peptides are preferably between about 5 and about 50 amino acids in length, 5 more preferably between about 10 and 30 amino acids in length, and most preferably between about 10 and 25 amino acids in length. In one embodiment the myostatin-binding peptide comprises the amino acid sequence WMCPP (SEQ ID NO: 633). In another embodiment the 2a WO 2004/058988 PCT/US2003/040781 myostatin binding peptides comprise the amino acid sequence Cam^Wa^WMCPP (SEP ID NO: 352), wherein ab a2 and a3 are selected from a neutral hydrophobic, neutral polar, or basic aminn acid. In another embodiment the myostatin binding peptide comprises the sequence Cb^WbgWMCPP (SEQ ID NO: 353), wherein bi is selected from any one of the amino acids T, 5 I, or R; b2 is selected from any one of R, S, Q; b3 is selected from any one of P, R and Q, and wherein the peptide is beween 10 and 50 amino acids in length, and physiologically acceptable salts thereof. In another embodiment, the myostatin binding peptide comprises the formula: ciCiCsCAC^CfiCciCBWcqWMCPPcmCi!cnCn (SEQ ID NO: 354), wherein: c, is absent or any amino acid; 10 c2 is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; c3 is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; c4 is absent or any amino acid; C5 is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; ce is absent or a neutral hydrophobic, neutral polar, or basic amino acid; 15 c7 is a neutral hydrophobic, neutral polar, or basic amino acid; eg is a neutral hydrophobic, neutral polar, or basic amino acid;

Cg is a neutral hydrophobic, neutral polar or basic amino acid; and cio to Cj3 is any aminn acid; and wherein the peptide is between 20 and 50 amino acids in length, and physiologically acceptable salts thereof. 20 A related embodiment the myostatin binding peptide comprises the formula: didid^d^drisCd-idgWdoWMCPP d10dnd12di3 (SEQ ID NO: 355), wherein dj is absent or any amino acid; d2 is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; 25 dj is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; di is absent or any amino acid; ds is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; d6 is absent or a neutral hydrophobic, neutral polar, or basic amino acid; d7 is selected from any one of the amino acids T, I; or R; 30 d8 is selected from any one of R, S, Q; d9 is selected from any one of P, R and Q, and d]0 to dJ3 is selected from any amino acid, and wherein the peptide is between 20 and 50 amino acids in length, and physiologically acceptable salts thereof. 35 Additional embodiments of binding agents comprise at least one of the following peptides: (1) a peptide capable of binding myostatin, wherein the peptide comprises the sequence WYe,e,Ye,G, (SEQ ID NO: 356) wherein e! is P, S or Y, 40 e2 is C or Q, and 3 WO 2004/058988 PCT/US2003/040781 e3 is G or H, wherein the peptide is between 7 and 50 amino acids in length, and physiologically acceptable salts thereof; (2) a peptide capable of binding myostatin, wherein the peptide comprises the sequence fiEMLfeSLfJdLL, (SEQ ID NO: 455), 5 wherein fr is M or I, f2 is any amino acid, f3 is L or F, f4 is E, Q or D; and wherein the peptide is between 7 and 50 amino acids in length, and physiologically 10 acceptable salts thereof; (3) a peptide capable of binding myostatin wherein the peptide comprises the sequence

Lgig7LLg3gtL· (SEQ ID NO: 456), wherein gjisQ, DorE, g2 is S, Q, D or E, 15 g3 is any amino acid, g4 is L, W, F, or Y, and wherein the peptide is between 8 and 50 amino acids in length, and physiologically acceptable salts thereof; (4) a peptide capable of binding myostatin, wherein the peptide comprises the sequence hih2h3h4h5h6h7hBh9 (SEQ ID NO: 457), wherein 20 hi is R or D, h2 is any amino acid, h3 is A, T S or Q, hi is L or M, hs is L or S, 25 hg is any amino acid, · lb is F or E, hB is W, F or C, h9 is L, F, M or K, and wherein the peptide is between 9 and 50 amino acids in length, and physiologically acceptable salts thereof. 30 In one embodiment, the binding agents of the present invention further comprise at least one vehicle such as a polymer or an Fc domain, and may further comprise at least one linker sequence. In this embodiment, the binding agents of the present invention are constructed so that at least one myostatin-binding peptide is attached to at least one vehicle. The peptide or peptides are attached directly or indirectly through a linker sequence, to the vehicle at the N-terminal, C- 35 terminal or an amino acid sidechain of the peptide. In this embodiment, the binding agents of the present invention have the following generalized structure: (X')a-F*-(X2)b, or multimers thereof; wherein F1 is a vehicle; and X1 and X2 are each independently selected from -(L'VP1; 40 -(L’VP’-dAi-P2; 4 WO 2004/058988 PCT/US2003/040781 -(LVp’-CL^a-P^CLVP3; and -(LVP’-CLVP2-^ -P3-(L4)rP4; wherein P1, P2, P3, and P4 axe peptides capable of binding myostatin; and L1, L2, L3, and L4 are each linkers; and a, b, c, d, e, and f are each independently 0 or 1, 5 provided that at least one of a and b is 1, and physiologically acceptable salts thereof.

In various embodiments of binding agents having this generalized structure, the peptides P1, P2, P3, and P4 can be independently selected from one or more of any of the peptides comprising the sequences provided above. P1, P2, P3, and P4 axe independently selected from one or more peptides comprising any of the following sequences: SEQ ID NO: 633, SEQ ID NO: 10 352, SEQ ID NO: 353, SEQ ID NO: 354, SEQ ID NO: 355, SEQ ID NO: 356, SEQ ID NO: 455, SEQ ID NO: 456, or SEQ ID NO: 457.

In a further embodiment, the binding agents comprise peptides fused to an Fc domain, either directly or indirectly, thereby providing peptibodies. The peptibodies of the present 15 invention display a high binding affinity for myostatin and can inhibit the activity of myostatin as demonstrated both in vitro using cell based assays and in animals.

The present invention also provides nucleic acid molecules comprising polynucleotides encoding the peptides, peptibodies, and peptide and peptibody variants and derivatives of the present invention. 20 The present invention provides pharmaceutically acceptable compositions comprising one or more binding agents of the present invention.

The binding agents of the present invention inhibit myostatin activity in vitro and in vivo. The binding agents of the present invention increase lean muscle mass in a treated animal and decreases fat mass as a percentage of body weight of the animal. The myostatin binding agents of 25 the present invention increase muscular strength in treated animal models.

The present invention provides methods of inhibiting myostatin activity in animals including humans by administering an effective dosage of one or more binding agents to the subject. The present invention provides methods of increasing lean muscle mass in animals including humans by administering an effective dosage of one or more binding agents. The 30 present invention further provides methods of treating myostatin- related disorders by administering an therapeutically effective dosage of one or more myostatin binding agents in a pharmaceutically acceptable composition to a subject. The present invention provides methods of treating muscle wasting disorders including muscular dystrophy, muscle wasting due to cancer, AIDS, rheumatoid arthritis, renal failure, uremia, chronic heart failure, age-related sarcopenia, 35 prolonged bed-rest, spinal chord injury, stroke, bone fracture. The present invention also provides 5 WO 2004/058988 PCT/US2003/040781 methods of treating metabolic disorders including obesity, diabetes, hyperglycemia, and bone loss.

The present invention also provides a method of increasing muscle mass in food animals by administering an effective dosage of one or more myostatin binding agents to the animal. 5 The present invention provides assays utilizing one or more myostatin binding agents to identify and quantitate myostatin in a sample. The assays may be diagnotic assays for measuring or monitoring myostatin levels in individuals with a myostatin related disorder or disease.

BRIEF DESCRIPTION OF THE FIGURES 10 Figure 1 shows myostatin activity as measured by expressed luciferase activity (y-axis) vrs. concentration (x-axis) for the TN8-19 peptide QGHCTRWPWMCPPY' (Seq ID No: 32) and the TN8-19 peptibody (pb) to determine the IC3B for each using the C2C12 pMARE luciferase assay described in the Examples below. The peptibody has a lower IC50 value compared with the peptide. 15 Figure 2 is a graph showing the increase in total body weight for CD1 nu/nu mice treated with increasing dosages of the lx mTN8-19-21 peptibody over a fourteen day period compared with mice treated with a huFc control, as described in Example 8.

Figure 3A shows the increase in the mass of the gastrocnemius muscle mass at necropsy of the mice treated in Figure 2 (Example 8). Figure 3B shows the increase in lean mass as 20 determined by NMR on day 0 compared with day 13 of the experiment described in Example 8. Figure 4 shows the increase in lean body mass as for CD1 nu/nu mice treated with biweekly injections of increasing dosages of lx mTN8-19-32 peptibody as determined by NMR on day 0 and day 13 of the experiment described in Example 8.

Figure 5A shows the increase in body weight for CD1 nu/nu mice treated with biweekly 25 injections of lx mTN8-19-7 compared with 2x mTN8-19-7 and the control animal for 35 days as described in Example 8. Figure 5B shows the increase in lean carcass weight at necropsy for the lx and 2x versions at 1 mg/kg and 3 mg/kg compared with the animals receiving the vehicle (huFc) (controls).

Figure 6A shows the increase in lean muscle mass vrs. body weight for aged mdx mice 30 treated with either affinity matured lx mTN8-19-33 peptibody or huFc vehicle at 10 mg/kg subcutaneously every other day for three months. Figure 6B shows the change in fat mass compared to body weight as determined by NMR for the same mice after 3 months of treatment. 35 6 WO 2004/058988 PCT/US2003/040781 DETAILED DESCRIPTION OF THE INVENTION The present invention provides binding agents capable of binding myostatin and inhibiting its activity. The myostatin binding agents can be used in assays, to identify, quantitate, or monitor the level of myostatin in an animal. The myostatin binding agents of the present 5 invention reduce myostatin activity. The myostatin binding agents of the present invention increase lean muscle mass in animals, decrease fat mass as a percentage of body weight, and increase muscle strength. The myostatin binding agents of the present invention can be used to treat a variety of metabolic disorders in which myostatin plays a role, including muscle wasting disorders such as muscular dystrophies, muscle wasting due to cancer, AIDS, rheumatoid arthritis, 10 renal failure, uremia, chronic heart failure, prolonged-bed-rest, spinal chord injury, stroke, and age-related sarcopenia as well as other metabolic disorders including diabetes, obesity, hyperglycemia, and bone loss, by administering a therapeutic dosage of one or more binding agents in a pharmaceutically acceptable composition to a subject. 15 Myostatin

Myostatin, a growth factor also known as GDF-8, is known to be a negative regulator of skeletal muscle tissue. Myostatin is synthesized as an inactive preproprotein which is activated by proteolyic cleavage (Zimmers et al., supra. (2002)). The precurser protein is cleaved to produce an NH2-terminal inactive prodomain and an approximately 109 amino acid COOH- 20 terminal protein in the form of a homodimer of about 25 kDa, which is the mature, active form (Zimmers et al, supra (2002)). It is now believed that the mature dimer circulates in the blood as an inactive latent complex bound to the propeptide (Zimmers et al, supra (2002).

As used herein the term “full-length myostatin” refers to the full-length human preproprotein sequence described in McPherron et al. supra (1997), as well as related full-length 25 polypeptides including allelic variants and interspecies homologs which are also described in McPherron et al. (1997). As used herein, the term “prodomain” or “propeptide” refers to the inactive NH2-terminal protein which is cleaved off to release the active COOH-terminal protein. As used herein the term “myostatin” or “mature myostatin” refers to the mature, biologically active COOH-terminal polypeptide, in monomer, dimer, multimeric form or other form. 30 “Myostatin” or “mature myostatin” also refers to fragments of the biologically active mature myostatin, as well as related polypeptides including allelic variants, splice variants, and fusion peptides and polypeptides. The mature myostatin COOH-terminal protein has been reported to have 100% sequence identity among many species including human, mouse, chicken, porcine, turkey, and rat (Lee et al., PNAS 98, 9306 (2001)). Myostatin may or may not include additional 7 2Q5265/2 terminal residues such as targeting sequences, or methionine and lysine residues and /or tag or fusion protein sequences, depending on how it is prepared.

As used herein the term “capable of binding to myostatin” or “having a binding affinity for myostatin” refers to a binding agent or peptide which binds to myostatin as demonstrated by as 5 the phage ELISA assay, the BIAcore® or KinExA™ assays described in the Examples below.

As used herein, the term “capable of modifying myostatin activity” refers to the action of an agent as either an agonist or an antagonist with respect to at least one biological activity of myostatin. As used herein, “agonist” or “mimetic”activity refers an agent having biological activity comparable to a protein that interacts with the protein of interest, as described, for 10 example, in International application WO 01/83525, filed May 2,2001.

As used herein, the term “inhibiting myostatin activity” or “having antagonist activity”refers to the ability of a peptide or binding agent to reduce or block myostatin activity or signaling as demonstrated or in vitro assays such as, for example, the pMARE C2C12 cell-based myostatin activity assay or by in vivo animal testing as described below. 15

Structure of Myostatin Binding Agents

In one embodiment, the binding agents of the present invention comprise at least one myostatin binding peptide covalently attached to at least one vehicle such as a polymer or an Fc domain. The attachment of the myostatin-binding peptides to at least one vehicle is intended to 20 increase the effectiveness of the binding agent as a therapeutic by increasing the biological activity of the agent and/or decreasing degradation in vivo, increasing half-life in vivo, reducing toxicity or immunogenicity in vivo. The binding agents of the present invention may further comprise a linker sequence connecting the peptide and the vehicle. The peptide or peptides are attached directly or indirectly through a linker sequence to the vehicle at the N-terminal, C- 25 terminal or an amino acid sidechain of the peptide. In this embodiment, the binding agents of the present invention have the following structure: (X'X-F’-CX2)!), θΓ multimers thereof; wherein F1 is a vehicle; and X1 and X2 are each independently selected from -(lVp1; 30 -(LVP'-CLVP2; -(L1)c-P1-(L2)d-P2-(L3)e-P3; and -(L1)c-P1-(L2)d-P2-(L3)e -P3-(L4)rP4; wherein P1, P2, P3, and P4 are peptides capable of binding myostatin; and 35 8 205265/2 L1, L2, L3, and L4 are each linkers; and a, b, c, d, e, and f are each independently 0 or 1, provided that at least one of a and b is 1.

Any peptide containing a cysteinyl residue may be cross-linked with another Cys-containing peptide, either or both of which may be linked to a vehicle. Any peptide having more than one Cys residue may form an intrapeptide disulfide bond, as well.

In one embodiment, the vehicle is an Fc domain, defined below. This embodiment is referred to as a “peptibody”. As used herein, the term “peptibody” refers to a molecule comprising an antibody Fc domain attached to at least one peptide. The production of peptibodies is generally described in PCT publication WO 00/24782, published May 4, 2000. Exemplary peptibodies are provided as lx and 2x configurations with one copy and two copies of the peptide (attached in tandem) respectively, as described in the Examples below.

Peptides

As used herein the term “peptide” refers to molecules of about 5 to about 90 amino acids linked by peptide bonds. The peptides of the present invention are preferably between about 5 to about 50 amino acids in length, more preferably between about 10 and 30 amino acids in length, and most preferably between about 10 and 25 amino acids in length, and are capable of binding to the myostatin protein.

The peptides of the present invention may comprise part of a sequence of naturally occuring proteins, may be randomized sequences derived from naturally occuring proteins, or may be entirely randomized sequences. The peptides of the present invention may be generated by any methods known in the art including chemical synthesis, digestion of proteins, or recombinant technology. Phage display and RNA-peptide screening, and other affinity screening techniques are particularly useful for generating peptides capable of binding myostatin.

Phage display technology is described, for example, in Scott et al. Science 249: 386 (1990); Devlin et al., Science 249: 404 (1990); U.S. Patent No. 5,223,409, issued June 29, 1993; U.S. Patent No. 5,733,731, issued March 31, 1998; U.S. Patent No. 5,498,530, issued March 12, 1996; U.S. Patent No. 5,432,018, issued July 11, 1995; U.S. Patent No. 5,338,665, issued August 16, 1994; U.S. Patent No. 5,922,545, issued July 13,1999; WO 96/40987, published December 19, 1996; and WO 98/15833, published April 16,1998,. Using phage libraries, random peptide sequences are displayed by fusion with coat proteins of filamentous phage. Typically, the displayed peptides are affinity-eluted either specifically or non-specifically against the target molecule. The retained phages may be enriched 9 WO 2004/058988 PCT/US2003/040781 by successive rounds of affinity purification and repropagation. The best binding peptides are selected for further analysis, for example, by using phage ELISA, described below, and then sequenced. Optionally, mutagenesis libraries may be created and screened to further optimize the sequence of the best binders (Lowman, Ann Rev Biophys Biomol Struct 26:401-24 (1997)). 5 Other methods of generating the myostatin binding peptides include additional affinity selection techniques known in the art. A peptide library can be fused in the carboxyl terminus of the lac repressor and expressed in E.coli. Another E. coZi-based method allows display on the cell’s outer membrane by fusion with a peptidoglycan-associated lipoprotein (PAL). Hereinafter, these and related methods are collectively referred to as “E. coli display.” In another method, 10 translation of random RNA is halted prior to ribosome release, resulting in a library of polypeptides with their associated RNA still attached. Hereinafter, this and related methods are collectively referred to as “ribosome display.” Other methods employ chemical linkage of peptides to RNA. See, for example, Roberts and Szostak, Proc Natl Acad Sci USA, 94: 12297-303 (1997). Hereinafter, this and related methods are collectively referred to as “RNA-peptide 15 screening.” Yeast two-hybrid screening methods also may be used to identify peptides of the invention that bind to myostatin. In addition, chemically derived peptide libraries have been developed in which peptides are immobilized on stable, non-biological materials, such as polyethylene rods or solvent-permeable resins. Another chemically derived peptide library uses photolithography to scan peptides immobilized on glass slides. Hereinafter, these and related 20 methods are collectively referred to as “chemical-peptide screening.” Chemical-peptide screening may be advantageous in that it allows use of D-amino acids and other analogues, as well as nonpeptide elements. Both biological and chemical methods are reviewed in Wells and Lowman, Curr Opin Biotechnol 3: 355-62 (1992).

Additionally, selected peptides capable of binding myostatin can be further improved 25 through the use of “rational design”. In this approach, stepwise changes are made to a peptide sequence and the effect of the substitution on the binding affinity or specificity of the peptide or some other property of the peptide is observed in an appropriate assay. One example of this technique is substituting a single residue at a time with alanine, referred to as an “alanine walk” or an “alanine scan”. When two residues are replaced, it is referred to as a “double alanine walk”. 30 The resultant peptide containing amino acid substitutions are tested for enhanced activity or some additional advantageous property.

In addition, analysis of the structure of a protein-protein interaction may also be used to suggest peptides that mimic the interaction of a larger protein. In such an analysis, the crystal structure of a protein may suggest the identity and relative orientation of critical residues of the 35 protein, from which a peptide may be designed. See, for example, Takasaki et al., Nature Biotech 10 WO 2004/058988 PCT/US2003/040781 15:1266 (1977). These methods may also he used to investigate the interaction between a targeted protein and peptides selected by phage display or other affinity selection processes, thereby suggesting further modifications of peptides to increase binding affinity and the ability of the peptide to inhibit the activity of the protein. 5 In one embodiment, the peptides of the present invention are generated as families of related peptides. Exemplary petides are represented by SEQ ID NO: 1 through 132. These exemplary peptides were derived through an selection process in which the best binders generated by phage display technology were further analyzed by phage ELISA to obtain candidate peptides by an affinity selection technique such as phage display technology as described herein. 10 However, the peptides of the present invention may be produced by any number of known methods including chemical synthesis as described below.

The peptides of the present invention can be further improved by the process of “affinity maturation”. This procedure is directed to increasing the affinity or the activity of the peptides and peptibodies of the present invention using phage display or other selection technologies. 15 Based on a consensus sequence, directed secondary phage display libraries, for example, can be generated in which the “core” amino acids (determined from the consensus sequence) are held constant or are biased in frequency of occurrence. Alternatively, an individual peptide sequence can be used to generate a biased, directed phage display library. Panning of such libraries under more stringent conditions can yield peptides with enhanced binding to myostatin, selective 20 binding to myostatin, or with some additional desired property. However, peptides having the affinity matured sequences may then be produced by any number of known methods including chemical synthesis or recombinantly. These peptides are used to generate binding agents such as peptibodies of various configurations which exhibit greater inhibitory activity in cell-based assays and in vivo assays. 25 Example 6 below describes affinity maturation of the “first round” peptides described above to produce affinity matured peptides. Exemplary affinity matured peptibodies are presented in Tables IV and V. The resultant lx and 2x peptibodies made from these peptides were then further characterized for binding affinity, ability to neutralize myostatin activity, specificity to myostatin as opposed to other ΤΝΕβ family members, and for additional in vitro and in vivo 30 activity, as described below. Affinity-matured peptides and peptibodies are referred to by the prefix “m” before their family name to distinguish them from first round peptides of the same family.

Exemplary first round peptides chosen for further affinity maturation according to the

present invention included the following peptides: TN8-19 QGHCTRWPWMCPPY (SEQ ID 35 NO: 33), and the linear peptides Linear-2 MEMLDSLFELLKDMVPISKA (SEQ ID NO: 104), 11 WO 2004/058988 PCT/TJS2003/040781

Linear-15 HHGWNYLRXGSAPQWFEAWV (SEQ ID NO: 117), Linear-17, RATLLKDFWQLVEGYGDN (SEQ ID NO: 119), Linear-20 YREMSMLEGLLDVLERLQHY (SEQ ID NO: 122), Linear-21 HNSSQMLLSELIMLVGSMMQ (SEQ ID NO: 123), Linear-24 EFFHWLHNI1RSEVNHWLDMN (SEQ ID NO: 126). The affinity matured families of each of 5 these is presented below in Tables IV and V.

The peptides of the present invention also encompass variants and derivatives of the selected peptides which are capable of binding myostatin. As used herein the term “variant” refers to peptides having one or more amino acids inserted, deleted, or substituted into the original amino acid sequence, and which are still capable of binding to myostatin. Insertional and 10 substitutional variants may contain natural amino acids as well as non-naturally occuring amino acids. As used herein the term “variant” includes fragments of the peptides which still retain the ability to bind to myostatin. As used herein, the term “derivative” refers to peptides which have been modified chemically in some manner distinct from insertion, deletion, and substitution variants. Variants and derivatives of the peptides and peptibodies of the present invention are 15 described more fully below.

Vehicles

As used herein the term “vehicle” refers to a molecule that may be attached to one or more peptides of the present invention. Preferably, vehicles confer at least one desired property 20 on the binding agents of the present invention. Peptides alone are likely to be removed in vivo either by renal filtration, by cellular clearance mechanisms in the reticuloendothelial system, or by proteolytic degradation. Attachment to a vehicle improves the therapeutic value of a binding agent by reducing degradation of the binding agent and/or increasing half-life, reducing toxicity, reducing immunogenicity, and/or increasing the biological activity of the binding agent 25 Exemplary vehicles include Fc domains; linear polymers such as polyethylene glycol (PEG), polylysine, dextran; a branched chain polymer (see for example U.S. Patent No. 4,289,872 to Derikenwalter et al., issued September 15,1981; U. S. Patent No. 5,229,490 to Tam, issued July 20, 1993; WO 93/21259 by Frechet et al., published 28 October 1993); a lipid; a cholesterol group (such as a steroid); a carbohydrate or oligosaccharide; or any natural or synthetic protein, 30 polypeptide or peptide that binds to a salvage receptor.

In one embodiment, the myostatin binding agents of the present invention have at least one peptide attached to at least one vehicle (F1, F2) through the N-terminus, C-terminus or a side chain of one of the amino acid residues of the peptide(s). Multiple vehicles may also be used; such as an Fc domain at each terminus or an Fc domain at a terminus and a PEG group at the 35 other terminus or a side chain. 12 205265/2

An Fc domain is one preferred vehicle. As used herein, the term “Fc domain” encompasses native Fc and Fc variant molecules and sequences as defined below. As used herein the term “native Fc” refers to a non-antigen binding fragment of an antibody or the amino acid sequence of that fragment which is produced by recombinant DNA techniques or by enzymatic or 5 chemical cleavage of intact antibodies. A preferred Fc is a fully human Fc and may originate from any of the immunoglobulins, such as IgGl and IgG2. However, Fc molecules that are partially human, or originate from non-human species are also included herein. Native Fc molecules are made up of monomeric polypeptides that may be linked into dimeric or multimeric forms by covalent (z'.e., disulfide bonds) and non-covalent association. The number of 10 intermolecular disulfide bonds between monomeric subunits of native Fc molecules ranges from 1 to 4 depending on class (e.g., IgG, IgA, IgE) or subclass (e.g., IgGl, IgG2, IgG3, IgAl, IgGA2). One example of a native Fc is a disulfide-bonded dimer resulting from papain digestion of an IgG (see Ellison et al. (1982), Nucl Acids Res 10: 4071-9). The term “native Fc” as used herein is used to refer to the monomeric, dimeric, and multimeric forms. 15 As used herein, the term “Fc variant” refers to a modified form of a native Fc sequence provided that binding to the salvage receptor is maintained, as described, for example, in WO 97/34631 and WO 96/32478. Fc variants may be constructed for example, by substituting or deleting residues, inserting residues or truncating portions containing the site. The inserted or substituted residues may also be altered amino acids, such as peptidomimetics or D-amino acids. 20 Fc variants may be desirable for a number of reasons, several of which are described below. Exemplary Fc variants include molecules and sequences in which: 1. Sites involved in disulfide bond formation are removed. Such removal may avoid reaction with other cysteine-containing proteins present in the host cell used to produce the molecules of the invention. For this purpose, the cysteine-containing segment at the N-terminus 25 may be truncated or cysteine residues may be deleted or substituted with other amino acids (e.g., alanyl, seryl). Even when cysteine residues are removed, the single chain Fc domains can still form a dimeric Fc domain that is held together non-covalently. 2. A native Fc is modified to make it more compatible with a selected host cell. For example, one may remove the PA sequence near the N-terminus of a typical native Fc, which may 30 be recognized by a digestive enzyme in E. coli such as proline iminopeptidase. One may also add an N-terminal methionyl residue, especially when the molecule is expressed recombinantly in a bacterial cell such as E. coli. 13 WO 2004/058988 PCT/US2003/040781 3. A portion of the N-tenninus of a native Fc is removed to prevent N-terminal heterogeneity when expressed in a selected host cell. For this purpose, one may delete any of the first 20 amino acid residues at the N-terminus, particularly those at positions 1, 2,3,4 and 5. 4. One or more glycosylation sites are removed. Residues that are typically glycosylated 5 (e.g., asparagine) may confer cytolytic response. Such residues may be deleted or substituted with unglycosylated residues (e.g., alanine). 5. Sites involved in interaction with complement, such as the Clq binding site, are removed. For example, one may delete or substitute the EKK sequence of human IgGl. Complement recruitment may not be advantageous for the molecules of this invention and so may 10 be avoided with such an Fc variant. 6. Sites are removed that affect binding to Fc receptors other than a salvage receptor. A native Fc may have sites for interaction with certain white blood cells that are not required for the fusion molecules of the present invention and so may be removed. 7. The ADCC site is removed. ADCC sites are known in the art. See, for example, 15 Molec Immunol 29 (5):633-9 (1992) with regard to ADCC sites in IgGl. These sites, as well, are not required for the fusion molecules of the present invention and so may be removed. 8. When the native Fc is derived from a non-human antibody, the native Fc may be humanized. Typically, to humanize a native Fc, one will substitute selected residues in the nonhuman native Fc with residues that are normally found in human native Fc. Techniques for 20 antibody humanization are well known in the art.

The term “Fc domain” includes molecules in monomeric or multimeric form, whether digested from whole antibody or produced by other means. As used herein the term “multimer” as applied to Fc domains or molecules comprising Fc domains refers to molecules having two or more polypeptide chains associated covalently, noncovalently, or by both covalent and non- 25 covalent interactions. IgG molecules typically form dimers; IgM, pentamers; IgD, dimers; and IgA, monomers, dimers, trimers, or tetramers. Multimers may be formed by exploiting the sequence and resulting activity of the native Ig source of the Fc or by derivatizing such a native Fc. The term “dimer” as applied to Fc domains or molecules comprising Fc domains refers to molecules having two polypeptide chains .associated covalently or non-covalently. 30 Additionally, an alternative vehicle according to the present invention is a non-Fc domain protein, polypeptide, peptide, antibody, antibody fragment, or small molecule (e.g., a peptidomimetic compound) capable of binding to a salvage receptor. For example, one could use as a vehicle a polypeptide as described in U.S. Patent No. 5,739,277, issued April 14,1998 to

Presta et al. Peptides could also be selected by phage display for binding to the FcRn salvage 35 receptor. Such salvage receptor-binding compounds are also included within the meaning of 14 205265/2 “vehicle”and are within the scope of this invention. Such vehicles should be selected for increased half-life (e.g., by avoiding sequences recognized by proteases) and decreased immunogenicity (e.g., by favoring non-immunogenic sequences, as discovered in antibody humanization). 5 In addition, polymer vehicles may also be used to construct the binding agents of the present invention. Various means for attaching chemical moieties useful as vehicles are currently available, see, e.g., Patent Cooperation Treaty (“PCT”) International Publication No. WO 96/11953, entitled “N-Terminally Chemically Modified Protein Compositions and Methods”. This PCT publication discloses, among other things, the selective attachment of water soluble polymers 10 to the N-terminus of proteins. A preferred polymer vehicle is polyethylene glycol (PEG). The PEG group may be of any convenient molecular weight and may be linear or branched. The average molecular weight of the PEG will preferably range from about 2 kDa to about 100 kDa, more preferably from about 5 kDa to about 50 kDa, most preferably from about 5 kDa to about 10 kDa. The PEG groups will 15 generally be attached to the compounds of the invention via acylation or reductive alkylation through a reactive group on the PEG moiety (e.g., an aldehyde, amino, thiol, or ester group) to a reactive group on the inventive compound (e.g., an aldehyde, amino, or ester group). A useful strategy for the PEGylation of synthetic peptides consists of combining, through forming a conjugate linkage in solution, a peptide and a PEG moiety, each bearing a special functionality 20 that is mutually reactive toward the other. The peptides can be easily prepared with conventional solid phase synthesis as known in the art. The peptides are “preactivated” with an appropriate functional group at a specific site. The precursors are purified and fully characterized prior to reacting with the PEG moiety. Ligation of the peptide with PEG usually takes place in aqueous phase and can be easily monitored by reverse phase analytical HPLC. The PEGylated peptides 25 can be easily purified by preparative HPLC and characterized by analytical HPLC, amino acid analysis and laser desorption mass spectrometry.

Polysaccharide polymers are another type of water soluble polymer which may be used for protein modification. Dextrans are polysaccharide polymers comprised of individual subunits of glucose predominantly linked by al-6 linkages. The dextran itself is available in many 30 molecular weight ranges, and is readily available in molecular weights from about 1 kDa to about 70 kDa. Dextran is a suitable water-soluble polymer for use in the present invention as a vehicle by itself or in combination with another vehicle (e.g., Fc). See, for example, WO 96/11953 and WO 96/05309. The use of dextran conjugated to therapeutic or diagnostic immunoglobulins has been reported; see, for example, European Patent Publication No. 0 315 456. 35 15 205265/2

Dextran of about 1 kDa to about 20 kDa is preferred when dextran is used as a vehicle in accordance with the present invention.

Linkers 5 The binding agents of the present invention may optionally further comprise a “linker” group. Linkers serve primarily as a spacer between a peptide and a vehicles or between two peptides of the binding agents of the present invention. In one embodiment, the linker is made up of amino acids linked together by peptide bonds, preferably from 1 to 20 amino acids linked by peptide bonds, wherein the amino acids are selected from the 20 naturally occurring amino acids. 10 One or more of these amino acids may be glycosylated, as is understood by those in the art. In one embodiment, the 1 to 20 amino acids are selected from glycine, alanine, proline, asparagine, glutamine, and lysine. Preferably, a linker is made up of a majority of amino acids that are sterically unhindered, such as glycine and alanine. Thus, exemplary linkers are polyglycines (particularly (Gly)5, (Gly)s), poly(Gly-Ala), and polyalanines. As used herein, the designation 15 “g” refers to a glycine homopeptide linkers. As shown in Table II, “gn” refers to a 5x gly linker at the N terminus, while “gc” refers to 5x gly linker at the C terminus. Combinations of Gly and Ala are also preferred. One exemplary linker sequence useful for constructing the binding agents of the present invention is the following: gsgsatggsgstassgsgsatg (Seq ID No: 305). This linker sequence is referred to as the “k” or lk sequence. The designations “kc”, as found in Table II, 20 refers to the k linker at the C-terminus, while the designation “kn”, refers to the k linker at the N-terminus.

The linkers of the present invention may also be non-peptide linkers. For example, alkyl linkers such as -NH-(CH2)s-C(O)-, wherein s = 2-20 can be used. These alkyl linkers may further be substituted by any non-sterically hindering group such as lower alkyl (e.g., CrC6) lower acyl, 25 halogen (e.g., Cl, Br), CN, NH2, phenyl, etc. An exemplary non-peptide linker is a PEG linker, and has a molecular weight of 100 to 5000 kDa, preferably 100 to 500 kDa. The peptide linkers may be altered to form derivatives in the same manner as above.

Exemplary Binding Agents 30 The binding agents of the present invention comprise at least one peptide capable of binding myostatin. In one embodiment, the myostatin binding peptide is between about 5 and about 50 amino acids in length, in another, between about 10 and 30 amino acids in length, and in another, between about 10 and 25 amino acids in length. In one embodiment the myostatin binding peptide comprises the amino acid sequence WMCPP (SEQ ID NO: 633). In other 35 embodiment, the myostatin binding peptide comprises the amino acid sequence 16 WO 2004/058988 PCT/US2003/040781

Cai a9~WaiWMCPP (SEQ ED NO: 352), wherein ab a2 and a3 are selected from a neutral hydrophobic, neutral polar, or basic amino acid. In another embodiment the myostatin binding peptide comprises the amino acid sequence Cbib^WbgWMCPP (SEQ ID NO: 353), wherein bj is selected from any one of the amino acids T, I, or R; b2 is selected from any one of R, S, Q; b3 is 5 selected from any one of P, R and Q, and wherein the peptide is beween 10 and 50 amino acids in length, and physiologically acceptable salts thereof.

In another embodiment, the myostatin binding peptide comprises the formula:' cJc2c3C4CSc6Cc7cBWcgWMCPPc10cnc12c13 (SEQ ID NO: 354), wherein: c3 is absent or any amino acid; c2 is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; c3 is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; C4 is absent or any amino acid;

Cs is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; cs is absent or a neutral hydrophobic, neutral polar, or basic amino acid; c7 is a neutral hydrophobic, neutral polar, or basic amino acid;

Cs is a neutral hydrophobic, neutral polar, or basic amino acid; c9 is a neutral hydrophobic, neutral polar or basic amino acid; and cio to c13 is any amino acid; and wherein the peptide is between 20 and 50 amino acids in length, and physiologically acceptable salts thereof. 20 A related embodiment the myostatin binding peptide comprises the formula: did2d3d4d5dsCd7d8Wd9WMCPP d!0dnd12di3 (SEQ ID NO: 355), wherein di is absent or any amino acid; d2 is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; 25 d3 is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; d4 is absent or any amino acid; d5 is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; d6 is absent or a neutral hydrophobic, neutral polar, or basic amino acid; d7 is selected from any one of the amino acids T, I, or R; 30 ds is selected from any one of R, S, Q; . d9 is selected from any one of P, R and Q,. and djo to dj3 is selected from any amino acid, and wherein the peptide is between 20 and 50 amino acids in length, and physiologically acceptable salts thereof. 35 Additional embodiments of binding agents comprise at least one of the following peptides: (1) a peptide capable of binding myostatin, wherein the peptide comprises the sequence WYeie2Ye3G, (SEQ ID NO: 356) wherein et is P, S or Y, 40 e2 is C or Q, and 17 WO 2004/058988 PCT/US2003/040781 e3 is G or H, wherein the peptide is between 7 and 50 amino acids in length, and physiologically acceptable salts thereof. (2) a peptide capable of binding myostatin, wherein the peptide comprises the sequence fiEMLfjSLfifaLL, (SEQ ID NO: 455), 5 wherein fr is M or I, f2 is any amino acid, f3 is L or F, fi is E, Q or D; and wherein the peptide is between 7 and 50 amino acids in length, and physiologically 10 acceptable salts thereof. (3) a peptide capable of binding myostatin wherein the peptide comprises the sequence

LgignLLgigaL. (SEQ ID NO: 456), wherein gi isQ, DorE, g2 is S, Q, D or E, 15 g3 is any amino acid, gi is L, W, F, or Y, and wherein the peptide is between 8 and 50 amino acids in length, and physiologically acceptable salts thereof. (4) a peptide capable of binding myostatin, wherein the peptide comprises the sequence hih^hihshehvhghp (SEQ ID NO: 457), wherein 20 hjisRorD, h2 is any amino acid, h3 is A, T S or Q, hi is L or M, ' h5 is L or S, 25 h6 is any amino acid, h7 is F or E, h8 isW, F or-C, h9 is L, F, M or K, and wherein the peptide is between 9 and 50 amino acids in length, and physiologically acceptable salts thereof. 30 In one embodiment, the binding agents of the present invention further comprise at least one vehicle such as a polymer or an Fc domain, and may further comprise at least one linker sequence. In this embodiment, the binding agents of the present invention are constructed so that at least one myostatin-binding peptide is covalently attached to at least one vehicle. The peptide or peptides are attached directly or indirectly through a linker sequence, to the vehicle at the N- 35 terminal, C-terminal or an amino acid sidechain of the peptide. In this embodiment, the binding agents of the present invention have the following generalized structure: (X^)a-F*-(X2)b, or multimers thereof; wherein F1 is a vehicle; and X1 and X2 are each independently selected from -(L^c-P1; 40 -clVp’-clVp2; 18 WO 2004/058988 PCT/TJS2003/040781 -(L'VP’-CLVp^CLVp3; and -(L^e-P’-CLVP2-^ -P3-(L4)rP4; wherein P}, P2, P3, and P* are peptides capable of binding myostatin; and L1, L2, L3, and L4 are each linkers; and a, b; c, d, e, and f are each independently 0 or 1, 5 provided that at least one of a and b is 1. hi one embodiment of the binding agents having this generalized structure, the peptides P1, P2, P3, and P4 can be selected from one or more of any of the peptides comprising the sequences provided above. Peptides P1, P2, P3, and P4 can be selected from one or more peptides comprising any of the following sequences: SEQ ID NO: 633, SEQ ID NO: 352, SEQ ID NO: 10 353, SEQ ID NO: 354, SEQ ID NO: 355, SEQ ID NO: 356, SEQ ID NO: 455, SEQ ID NO: 456, or SEQ ID NO: 457.

In a further embodiment, the vehicles of binding agents having the general formula above are Fc domains. The peptides are therefore fused to an Fc domain, either directly or indirectly, thereby providing peptibodies. The peptibodies of the present invention display a high binding 15 affinity for myostatin and can inhibit the activity of myostatin as demonstrated by in vitro assays and in vivo testing in animals provided herein.

The present invention also provides nucleic acid molecules comprising polynucleotides encoding the peptides, peptibodies, and peptide and peptibody variants and derivatives of the present invention. Exemplary nucleotides sequences are given below. 20

Variants and Derivatives of Peptides and Peptibodies

The binding agents of the present invention also encompass variants and derivatives of the peptides and peptibodies described herein. Since both the peptides and peptibodies of the present invention can be described in terms of their amino acid sequence, the terms “variants” and 25 “derivatives” can be said to apply to a peptide alone, or a peptide as a component of a peptibody. As used herein, the term “peptide variants” refers to peptides or peptibodies having one or more amino acid residues inserted, deleted or substituted into the original amino acid sequence and which retain the ability to bind to myostatin and modify its activity. As used herein, fragments of the peptides or peptibodies are included within the definition of “variants”. 30 It is understood that any given peptide or peptibody may contain one or two or all three types of variants. Insertional and substitutional variants may contain natural amino acids, as well as non-naturally occuring amino acids or both.

Peptide and peptibody variants also include mature peptides and peptibodies wherein leader or signal sequences are removed, and the resulting proteins having additional amino 35 terminal residues, which amino acids may be natural or non-natural. Peptibodies with an 19 WO 2004/058988 PCT/US2003/040781 additional methionyl residue at amino acid position -1 (Met" ^peptibody) are contemplated, as are peptibodies with additional methionine and lysine residues at positions -2 and -1 (Met^-Lys"1-). Variants having additional Met, Met-Lys, Lys residues (or one or more basic residues, in general) are particularly useful for enhanced recombinant protein production in bacterial host cells. 5 ' Peptide or peptibody variants of the present invention also includes peptides having additional amino acid residues that arise from use of specific expression systems. For example, use of commercially available vectors that express a desired polypeptide as part of glutathione-S-transferase (GST) fusion product provides the desired polypeptide having an additional glycine residue at amino acid position-1 after cleavage of the GST component from the desired 10 polypeptide. Variants which result from expression in other vector systems are also contemplated, including those wherein histidine tags are incorporated into the amino acid sequence, generally at the carboxy and/or amino terminus of the sequence.

In one example, insertional variants are provided wherein one or more amino acid residues, either naturally occurring or non-naturally occuring amino acids, are added to a peptide 15 amino acid sequence. Insertions may be located at either or both termini of the protein, or may be positioned within internal regions of the peptibody amino acid sequence. Insertional variants with additional residues at either or both termini can include, for example, fusion proteins and proteins including amino acid tags or labels. Insertional variants include peptides in which one or more amino acid residues are added to the peptide amino acid sequence or fragment thereof. 20 Insertional variants also include fusion proteins wherein the amino and/or carboxy termini of the peptide or peptibody is fused to another polypeptide, a fragment thereof or amino acids which are not generally recognized to be part of any specific protein sequence. Examples of such fusion proteins are immunogenic polypeptides, proteins with long circulating half lives, such as immunoglobulin constant regions, marker proteins, proteins or polypeptides that facilitate 25 purification of the desired peptide or peptibody, and polypeptide sequences that promote formation of multimeric proteins (such as leucine zipper motifs that are useful in dimer forma tion/stability).

This type of insertional variant generally has all or a substantial portion of the native molecule, linked at the N- or C-terminus, to all or a portion of a second polypeptide. For 30 example, fusion proteins typically employ leader sequences from other species to permit the recombinant expression of a protein in a heterologous host. Another useful fusion protein includes the addition of an immunologically active domain, such as an antibody epitope, to facilitate purification of the fusion protein. Inclusion of a cleavage site at or near the fusion junction will facilitate removal of the extraneous polypeptide after purification. Other useful 20 WO 2004/058988 PCT/US2003/040781 fusions include linking of functional domains, such as active sites from enzymes, glycosylation domains, cellular targeting signals or transmembrane regions.

There are various commercially available fusion protein expression systems that may be used in tire present invention. Particularly useful systems include but are not limited to the 5 glutathione-S-transferase (GST) system (Pharmacia), the maltose binding protein system (NEB, Beverley, MA), the FLAG system (IBI, New Haven, CT), and the 6xHis system (Qiagen, Chatsworth, CA). These systems are capable of producing recombinant peptides and/or peptibodies bearing only a small number of additional amino acids, which are unlikely to significantly affect the activity of the peptide or peptibody. For example, both the FLAG system 10 and the 6xHis system add only short sequences, both of which are known to be poorly antigenic and which do not adversely affect folding of a polypeptide to its native conformation. Another N-tenninal fusion that is contemplated to be useful is .the fusion of a Met-Lys dipeptide at the N-terminal region of the protein or peptides. Such a fusion may produce beneficial increases in protein expression or activity. 15 Other fusion systems produce polypeptide hybrids· where it is desirable to excise the fusion partner from the desired peptide or peptibody. In one embodiment, the fusion partner is linked to the recombinant peptibody by a peptide sequence containing a specific recognition sequence for a protease. Examples of suitable sequences are those recognized by the Tobacco Etch Virus protease (Life Technologies, Gaithersburg, MD) or Factor Xa (New England Biolabs, 20 Beverley, MA).

The invention also provides fusion polypeptides which comprise all or part of a peptide or peptibody of the present invention, in combination with truncated tissue factor (tTF). tTF is a vascular targeting agent consisting of a truncated form of a human coagulation-inducing protein that acts as a tumor blood vessel clotting agent, as described U.S. Patent Nos.: 5,877,289; 25 6,004,555; 6,132,729; 6,132,730; 6,156,321; and European Patent No. EP 0988056. The fusion of tTF to the anti-myostatin peptibody or peptide, or fragments thereof facilitates the delivery of anti-myostatin antagonists to target cells, for example, skeletal muscle cells, cardiac muscle cells, fibroblasts, pre-adipocytes, and possibly adipocytes.

In another aspect, the invention provides deletion variants wherein one or more amino 30 acid residues in a peptide or peptibody are removed. Deletions can be effected at one or both termini of the peptibody, or from removal of one or more residues within the peptibody amino acid sequence. Deletion variants necessarily include all fragments of a peptide or peptibody.

In still another aspect, the invention provides substitution variants of peptides and peptibodies of the invention. Substitution variants include those peptides and peptibodies wherein 35 one or more amino acid residues are removed and replaced with one or more alternative amino 21 WO 2004/058988 PCT/US2003/040781 acids, which amino acids may be naturally occurring or non-naturally occurring. Substitutional variants generate peptides or peptibodies that are “similar” to the original peptide or peptibody, in that the two molecules have a certain percentage of amino acids that are identical. Substitution variants include substitutions of 1, 2, 3,4,5, 6,7, 8, 9,10,15, and 20 amino acids within a 5 peptide or peptibody, wherein the number of substitutions may be up to ten percent of the amino acids of the peptide or peptibody. In one aspect, the substitutions are conservative in nature, however, the invention embraces substitutions that are also non-conservative and also includes unconventional amino acids.

Identity and similarity of related peptides and peptibodies can be readily calculated by 10 known methods. Such methods include, but are not limited to, those described in Computational Molecular Biology, Lesk, AM., ed., Oxford University Press, New York (1988); Biocomputing: Informatics and Genome Projects, Smith, D.W., ed., Academic Press, New York (1993);

Computer Analysis of Sequence Data, Part 1, Griffin, A.M., and Griffin, H.G., eds., Humana Press, New Jersey (1994); Sequence Analysis in Molecular Biology, von Heinje, G., Academic 15 Press (1987); Sequence Analysis Primer, Gribskov, M. and Devereux, J., eds., M. Stockton Press, New York (1991); and Carillo et al., SIAM J. Applied Math., 48:1073 (1988).

Preferred methods to determine the relatedness or percent identity of two peptides or polypeptides, or a polypeptide and a peptide, are designed to give the largest match between the sequences tested. Methods to determine identity are described in publicly available computer 20 programs. Preferred computer program methods to determine identity between two sequences include, but are not limited to, the GCG program package, including GAP (Devereux et al., Nucl. Acid. Res., 12:387 (1984); Genetics Computer Group, University of Wisconsin, Madison, WI, BLASTP, BLASTN, and FASTA (Altschul et al., J. Mol. Biol., 215:403-410 (1990)). The BLASTX program is publicly available from the National Center for Biotechnology Information 25 (NCBI) and’other sources {BLAST Manual, Altschul et al. NCB/NLM/NIH Bethesda, MD 20894; Altschul et al., supra (1990)). The well-known Smith Waterman algorithmmay also be used to determine identity.

Certain alignment schemes for aligning two amino acid sequences may result in the matching of only a short region of the two sequences, and this small aligned region may have very 30 high sequence identity even though there is no significant relationship between the two full-length sequences. Accordingly, in certain embodiments, the selected alignment method will result in an alignment that spans at least ten percent of the full length of the target polypeptide being compared, i.e., at least 40 contiguous amino acids where sequences of at least 400 amino acids are being compared, 30 contiguous amino acids where sequences of at least 300 to about 400 amino 22 WO 2004/058988 PCT/US2003/040781 acids are being compared, at least 20 contiguous amino acids where sequences of 200 to about 300 amino acids are being compared, and at least 10 contiguous amino acids where sequences of about 100 to 200 amino acids are being compared. For example, using the computer algorithm GAP (Genetics Computer Group, University of Wisconsin, Madison, WI), two polypeptides for 5 which the percent sequence identity is to be determined are aligned for optimal matching of their respective amino acids (the “matched span”, as determined by the algorithm). In certain embodiments, a gap opening penalty (which is typically calculated as 3X the average diagonal; the “average diagonal” is the average of the diagonal of the comparison matrix being used; the “diagonal” is the score or number assigned to each perfect amino acid match by the particular 10 comparison matrix) and a gap extension penalty (which is usually 1/10 times the gap opening penalty), as well as a comparison matrix such as PAM 250 or BLOSUM 62 are used in conjunction with the algorithm. In certain embodiments, a standard comparison matrix (see Dayhoff et al., Atlas of Protein Sequence and Structure, 5(3)(1978) for the PAM 250 comparison matrix; Henikoff et al., Proc. Natl. Acad. Sci USA, 89:10915-10919 (1992) for the BLOSUM 62 15 comparison matrix) is also used by the algorithm.

In certain embodiments, for example, the parameters for a polypeptide sequence comparison can be made with the following: Algorithm: Needleman et al., J. Mol. Biol., 48:443-453(1970); Comparison matrix: BLOSUM 62 from Henikoff etal., supra (1992); Gap Penalty: 12; Gap Length Penalty: 4; Threshold of Similarity: 0, along with no penalty for end gaps. 20 In certain embodiments, the parameters for polynucleotide molecule sequence (as opposed to an amino acid sequence) comparisons can be made with the following: Algorithm: Needleman et al., supra (1970); Comparison matrix: matches = +10, mismatch = 0; Gap Penalty: 50: Gap Length Penalty: 3

Other exemplary algorithms, gap opening penalties, gap extension penalties, comparison 25 matrices, thresholds of similarity, etc. may be used, including those set forth in the Program

Manual, Wisconsin Package, Version 9, September, 1997. The particular choices to be made will be apparent to those of skill in the art and will depend on the specific comparison to be made, such as DNA-to-DNA, protein-to-protein, protein-to-DNA; and additionally, whether the comparison is between given pairs of sequences (in which case GAP or BesfFit are generally 30 preferred) or between one sequence and a large database of sequences (in which case FASTA or BLASTA are preferred).

Stereoisomers (e.g., D-amino acids) of the twenty conventional (naturally occuring) amino acids, non-naturally occuring amino acids such as α-, α-disubstituted amino acids, N-alkyl amino acids, lactic acid, and other unconventional amino acids may also be suitable components 23 WO 2004/058988 PCT/US2003/040781 for peptides of the present invention. Examples of non-naturally occuring amino acids include, for example: aminoadipic acid, beta-alanine, beta-aminopropionic acid, aminobutyric acid, piperidinic acid, aminocaprioic acid, amirioheptanoic acid, amimoisobutyric acid, aminopimelic acid, diaminobutyric acid, desmosine, diaminopimelic acid, diaminopropionic acid, N- 5 ethylglycine, N-etbylaspargine, hyroxylysine, allO-hydroxylysine, hydroxyproline, isodesmosine, allo-isoleucine, N-methylglycine, sarcosine, N-methylisoleucine, N-methylvaline, norvaline, norleucine, orithine, 4-hydroxyproline, γ-carboxyglutamate, ε-Ν,Ν,Ν-trimethyllysine, ε-Ν-acetyllysine, O-phosphoserine, N-acetylserine, N-formylmethionine, 3-methylhistidine, 5-hydroxylysine, σ-Ν-methylarginine, and other similar amino acids and amino acids (e.g., 4- 10 hydroxy proline).

Naturally occurring residues may be divided into (overlapping) classes based on common side chain properties: 15 1) neutral hydrophobic: Met, Ala, Val, Leu, He, Pro, Trp, Met, Phe; 2) neutral polar: Cys, Ser, Thr, Asn, Gin, Tyr, Gly; 3) acidic: Asp, Glu; 4) basic: His, Lys, Arg; 5) residues that influence chain orientation: Gly, Pro; and 6) aromatic: Tip, Tyr, Phe.

Substitutions of amino acids may be conservative, which produces peptides having 20 functional and chemical characteristics similar to those of the original peptide. Conservative amino acid substitutions involve exchanging a member of one of the above classes for another member of the same class. Conservative changes may encompass unconventional amino acid residues, which are typically incorporated by chemical peptide synthesis rather than by synthesis in biological systems. These include peptidomimetics and other reversed or inverted forms of 25 amino acid moieties.

Non-conservative substitutions may involve the exchange of a member of one of these classes for a member from another class. These changes can result in substantial modification in the functional and/or chemical characteristics of the peptides. In making such changes, according to certain embodiments, the hydropathic index of amino acids may be considered. Each amino 30 acid has been assigned a hydropathic index on the basis of its hydrophobicity and charge characteristics. They are: isoleucine (+4.5); valine (+4.2); leucine (+3.8); phenylalanine (+2.S); cysteine/cystine (+2.5); methionine (+1.9); alanine (+1.8); glycine (-0.4); threonine (-0.7); serine 24 WO 2004/058988 PCT/US2003/040781 (-0.8); tryptophan (-0.9); tyrosine (-1.3); proline (-1.6); histidine (-3.2); glutamate (-3.5); glutamine (-3.5); aspartate (-3,5); asparagine (-3.5); lysine (-3.9); and arginine (-4,5).

The importance of the hydropathic amino acid index in conferring interactive biological function on a protein is understood in the art. Kyte et al., J. Mol. Biol.,· 157:105-131 (1982). It is 5 known that certain ami-no acids may be substituted for other amino acids having a similar hydropathic index or score and still retain a similar biological activity. In making changes based upon the hydropathic index, in certain embodiments, the substitution of amino acids whose hydropathic indices are within ±2 is included, hr certain embodiments, those which are within ±1 are included, and in certain embodiments, those within ±0.5 are included. 10 It is also understood in the art that the substitution of like amino acids can be made effectively on the basis of hydrophilicity, particularly where the biologically functional peptibody or peptide thereby created is intended for use in immunological embodiments, as in the present case. In certain embodiments, the greatest local average hydrophilicity of a protein, as governed by the hydrophilicity of its adjacent amino acids, correlates with its immunogenicity and 15 antigenicity, i.e., with a biological property of the protein.

The following hydrophilicity values have been assigned to these amino acid residues: arginine (+3.0); lysine (+3.0); aspartate (+3.0 + 1); glutamate (+3.0 ± 1); serine (+0.3); asparagine (+0.2); glutamine (+0.2); glycine (0); threonine (-0.4); proline (-0.5 ± 1); alanine (-0.5); histidine (-0.5); cysteine (-1.0); methionine (-1.3); valine (-1.5); leucine (-1,8); isoleucine (-1.8); tyrosine (- 20 2.3); phenylalanine (-2.5) and tryptophan (-3.4). In making changes based upon similar hydrophilicity values, in certain embodiments, the substitution of ammo acids whose hydrophilicity values are within ±2 is included, in certain embodiments, those which are within ±1 are included, and in certain embodiments, those within ±0.5 are included. One may also identify epitopes from primary amino acid sequences on the basis of hydrophilicity. These 25 regions are also referred to as “epitopic core regions.”

Exemplary amino acid substitutions are set forth in Table 1 below.

Amino Acid Substitutions

Original Residues Exemplary Substitutions Preferred Substitutions Ala Val, Leu, He Val Arg Lys, Gin, Asn Lys Asn Gin, Glu, Asp Gin Asp Glu, Gin, Asp Glu Cys Ser, Ala Ser 25 WO 2004/058988 PCT/US2003/040781

Gin Asn, Glu, Asp Asn Glu Asp, Gin, Asn Asp Gly Pro, Ala Ala His Asn, Gin, Lys, Arg Arg He Leu, Val, Met, Ala, Phe, Norleucine Leu Leu Norleucine, He, Val, Met, Ala, Phe He Lys Arg, 1,4 Diamino-butyric Acid, Gin, Asn Arg Met Leu, Phe, He Leu Phe Leu, Val, He, Ala, Tyr Leu Pro Ala Gly Ser Thr, Ala, Cys Thr Thr Ser Ser Tip Tyr, Phe Tyr Tyr Trp, Phe, Thr, Ser Phe Val He, Met, Leu, Phe, Ala, Norleucine Leu

One skilled in the art will be able to produce variants of the peptides and peptibodies of the present invention by random substitution, for example, and testing the resulting peptide or peptibody for binding activity using the assays described herein. 5 Additionally, one skilled in the art can review structure-function studies or three- dimensional structural analysis in order to identify residues in similar polypeptides that are important for activity or structure. In view of such a comparison, one can predict the importance of amino acid residues in a protein that correspond to amino acid residues which are important for activity or structure in similar proteins. One skilled in the art may opt for chemically similar 10 amino acid substitutions for such predicted important amino acid residues. The variants can then be screened using activity assays as described herein. A number of scientific publications have been devoted to the prediction of secondary structure. See Moult J., Curr. Op. in Biotech., 7(4):422-427 (1996), Chou et al., Biochemistry, 13(2):222-245 (1974); Chou et al., Biochemistry, 113(2):211-222 (1974); Chou et al.,Adv. 15 Enzymol. Relat. Areas Mol. Biol., 47:45-148 (1978); Chou et αΙ.,Αηη. Rev. Biochem., 47:251-276 and Chou et al.,Biophys. J., 26:367-384 (1979). Moreover, computer programs are currently available to assist with predicting secondary structure. One method of predicting secondary structure is based upon homology modeling. For example, two polypeptides or proteins which 26 WO 2004/058988 PCT/US2003/040781 have a sequence identity of greater than 30%, or similarity greater than 40% often have similar structural topologies. The recent growth of the protein structural database (PDB) has provided enhanced predictability of secondary structure, including the potential number of folds within a protein’s structure. See Holm e/·al., Nucl.· Acid. Res., 27(1):244-247 (1999). It has been 5 suggested (Brenner et al., Curr. Op. Struct. Biol., 7(3):369-376 (1997)) that there are a limited number of folds in a given protein and that once a critical number of structures have been resolved, structural prediction will become dramatically more accurate.

Additional methods of predicting secondary structure include “threading” (Jones, D., Curr. Opin. Struct. Biol., 7(3):377-87 (1997); Sippl et al., Structure, 4(1):15-19 (1996)), “profile 10 analysis” (Bowie et al., Science, 253:164-170 (1991); Gribskov et al., Meth. Enzym., 183:146-159 (1990); Gribskov et al., Proc. Nat. Acad. Sci., 84(13):4355-4358 (1987)), and “evolutionary linkage” (See Holm, supra (1999), and Brenner, supra (1997)).

In certain embodiments, peptide or peptibody variants include glycosylation variants wherein one or more glycosylation sites such as a N-linked glycosylation site, has been added to 15 the peptibody. An N-linked glycosylation site is characterized by the sequence: Asn-X-Ser or Asn-X-Thr, wherein the amino acid residue designated as X may be any amino acid residue except proline. The substitution or addition of amino acid residues to create this sequence provides a potential new site for the addition of an N-linked carbohydrate chain. Alternatively, substitutions which eliminate this sequence will remove an existing N-linked carbohydrate chain. 20 Also provided is a rearrangement of N-linked carbohydrate chains wherein one or more N-linked glycosylation sites (typically those that are naturally occurring) are eliminated and one or more new N-linked sites are created.

The invention also provides “derivatives” of the peptides or peptibodies of the present invention. As used herein the term “derivative” refers to modifications other than, or in addition 25 to, insertions, deletions, or substitutions of amino acid residues which retain the ability to bind to myostatin.

Preferably, the modifications made to the peptides of the present invention to produce derivatives are covalent in nature, and include for example, chemical bonding with polymers, lipids, other organic, and inorganic moieties. Derivatives of the invention may be prepared to 30 increase circulating half-life of a peptibody, or may be designed to improve targeting capacity for the peptibody to desired cells, tissues, or organs.

The invention further embraces derivative binding agents covalently modified to Include one or more water soluble polymer attachments, such as polyethylene glycol, polyoxyethylene 27 WO 2004/058988 PCT/TJS2003/040781 glycol, or polypropylene glycol, as described U.S. Patent Nos.: 4,640,835; 4,496,689; 4,301,144; 4,670,417; 4,791,192; and 4,179,337. Still other useful polymers known in the art include monomethoxy-polyethylene glycol, dextran, cellulose, or other carbohydrate based polymers, poly-(N-vinyl pyrrolidone)-polyethylene glycol, propylene glycol homopolymers, a 5 polypropylene oxide/ethylene oxide co-polymer, polyoxyethylated polyols (e.g., glycerol) and polyvinyl alcohol, as well as mixtures of these polymers. Particularly preferred are peptibodies covalently modified with polyethylene glycol (PEG) subunits. Water-soluble polymers may be bonded at specific positions, for example at the amino terminus of the peptibodies, or randomly attached to one or more side chains of the polypeptide. The use of PEG for improving the 10 therapeutic capacity for binding agents, e.g. peptibodies, and for humanized antibodies in particular, is described in US Patent No. 6,133,426 to Gonzales et al., issued October 17,2000.

The invention also contemplates derivatizing the peptide and/or vehicle portion of the myostatin binding agents. Such derivatives may improve the solubility, absorption, biological half-life, and the like of the compounds. The moieties may alternatively eliminate or attenuate 15 any undesirable side-effect of the compounds and the like. Exemplary derivatives include compounds in which: 1. The derivative or some portion thereof is cyclic. For example, the peptide portion may be modified to contain two or more Cys residues (e.g., in the linker), which could cyclize by disulfide bond formation. 20 2. The derivative is cross-linked or is rendered capable of cross-linking between molecules. For example, the peptide portion may be modified to contain one Cys residue and thereby be able to form an intermolecular disulfide bond with a like molecule. The derivative may also be cross-linked through its C-terminus. 3. One or more peptidyl [-C(O)NR-] linkages (bonds) is replaced by a non-peptidyl 25 linkage. Exemplary non-peptidyl linkages are -CH2-carbamate [-CH2-OC(O)NR-], phosphonate, -CH2-sulfonamide [-CH2-S(O)2NR-], urea [-NHC(O)NH-], -CH2-secondary amine, and alkylated peptide [-C(O)NR6- wherein R6 is lower alkyl], 4. The N-terminus is derivatized. Typically, the N-terminus may be acylated or modified to a substituted amine. Exemplary N-terminal derivative groups include -NRRj (other than -NH2), 30 -NRC(O)Rb -NRC(O)ORb -NRS(O)2Rb -NHC(0)NHRb succinimide, or benzyloxycarbonyl-NH- (CBZ-NH-), wherein R and R1 are each independently hydrogen or lower alkyl and wherein the phenyl ring may be substituted with 1 to 3 substituents selected from the group consisting of Ci-C4 alkyl, CrC4 alkoxy, chloro, and bromo. 5. The free C-terminus is derivatized. Typically, the C-terminus is esterified or amidated. 35 For example, one may use methods described in the art to add (NH-CH2-CH2-NH2)2 to 28 WO 2004/058988 PCT/US2003/040781 compounds of this invention at the C-terminus. Likewise, one may use methods described in the art to add -NH2, (or “capping” with an -NHZ group) to compounds of this invention at the C-terminus. Exemplary C-teiminal derivative groups include, for example, -C(O)R2 wherein R2 is lower alkoxy or -NR3R4 wherein R3 and R4 are independently hydrogen or Ci-C8 alkyl (preferably 5 CrC4 alkyl). 6. A disulfide bond is replaced with another, preferably more stable, cross-linking moiety (e.g., an alkylene). See, e.g., Bhatnagar et al., J Med Chem 39: 3814-9 (1996), Alberts et al., Thirteenth Am Pep Symp, 357-9 (1993). 1 7. One or more individual amino acid residues is modified. Various derivatizing agents 10 are known to react specifically with selected side chains or terminal residues, as described in detail below.

Lysinyl residues and amino terminal residues may be reacted with succinic or other carboxylic acid anhydrides, which reverse the charge of the lysinyl residues. Other suitable reagents for derivatizing alpha-amino-containing residues include imidoesters such as methyl 15 picolinimidate; pyridoxal phosphate; pyridoxal; chloroborohydride; trinitrobenzenesulfonic acid; O-methylisourea; 2,4 pentanedione; and transaminase-catalyzed reaction with glyoxylate.

Arginyl residues may be modified by reaction with any one or combination of several conventional reagents, including phenylglyoxal, 2,3-butanedione, 1,2-cyclohexanedione, and ninhydrin. Derivatization of arginyl residues requires that the reaction be performed in alkaline 20 conditions because of the high pKa of the guanidine functional group. Furthermore, these reagents may react with the groups of lysine as well as the arginine epsilon-amino group.

Specific modification of tyrosyl residues has been studied extensively, with particular interest in introducing spectral labels into tyrosyl residues by reaction with aromatic diazonium compounds or tetranitromethane. Most commonly, N-acetylimidizole and tetranitromethane are 25 used to form O-acetyl tyrosyl species and 3-nitro derivatives, respectively.

Carboxyl side chain groups (aspartyl or glutamyl) may be selectively modified by reaction with carbodiimides (R’-N=C=N-R’) such as l-cyclohexyl-3-(2-rnorpholinyl-(4-ethyl) carbodiimide or l-ethyl-3-(4-azonia-4,4-dimethylpentyl) carbodiimide. Furthermore, aspartyl and glutamyl residues may be converted to asparaginyl and glutaminyl residues by reaction with 3 0 ammonium ion s.

Glutaminyl and asparaginyl residues may be deamidated to the corresponding glutamyl and aspartyl residues. Alternatively, these residues are deamidated under mildly acidic conditions. Either form of these residues falls within the scope of this invention. 29 WO 2004/058988 PCT/US2003/040781

Cysteinyl residues can be replaced by amino acid residues or other moieties either to eliminate disulfide bonding or, conversely, to stabilize cross-linking. See, e.g., Bhatnagar et al., (supra).

Derivatization with bifunctional agents is useful for cross-linking the peptides or their 5 functional derivatives to a water-insoluble support matrix or to other macromolecular vehicles:

Commonly used cross-linking agents include, e.g., l,l-bis(diazoacetyl)-2-phenylethane, glutaraldehyde, N-hydroxysuccinimide esters, for example, esters with 4-azidosalicylic acid, homobifunctional imidoesters, including disuccinimidyl esters such as 3,3’- dithiobis(succinimidylpropionate), and bifunctional maleimides such as bis-N-maleimido-1,8-10 octane. Derivatizing agents such as methyl-3-[(p-azidophenyl)dithio]propioimidate yield photoactivatable intermediates that are capable of forming crosslinks in the presence of light. Alternatively, reactive water-insoluble matrices such as cyanogen bromide-activated carbohydrates and the reactive substrates described in U.S. Patent Nos. 3,969,287; 3,691,016; 4,195,128; 4,247,642; 4,229,537; and 4,330,440 are employed for protein immobilization. 15 Carbohydrate (oligosaccharide) groups may conveniently be attached to sites that are known to be glycosylation sites in proteins. Generally, O-linked oligosaccharides are attached to serine (Ser) or threonine (Thr) residues while N-linked oligosaccharides are attached to asparagine (Asn) residues when they are part of the sequence Asn-X-Ser/Thr, where X can be any amino acid except proline. X is preferably one of the 19 naturally occurring amino acids other 20 than proline. The structures of N-linked and O-linked oligosaccharides and die sugar residues found in each type are different. One type of sugar that is commonly found on both is N-acetylneuraminic acid (referred to as sialic acid). Sialic acid is usually the terminal residue of both N-linked and O-linked oligosaccharides and, by virtue of its negative charge, may confer acidic properties to the glycosylated compound. Such site(s) may be incorporated in the linker of 25 the compounds of this invention and are preferably glycosylated by a cell during recombinant production of the polypeptide compounds (e.g., in mammalian cells such as CHO, BHK, COS). However, such sites may further be glycosylated by synthetic or semi-synthetic procedures known in the art.

Other possible modifications include hydroxylation of proline and lysine, phosphorylation 30 of hydroxyl groups of seryl or threonyl residues, oxidation of the sulfur atom in Cys, methylation of the alpha-amino groups of lysine, arginine, and histidine side chains [see, for example, Creighton, Proteins: Structure and Molecule Properties (W. H. Freeman & Co., San Francisco), pp. 79-86 (1983)].

Compounds of the present invention may be changed at the DNA level, as well. The DNA 35 sequence of any portion of the compound may be changed to codons more compatible with the 30 WO 2004/058988 PCT/US2003/040781 chosen host cell. For E. coli, which is the preferred host cell, optimized codons are known in the art. Codons may be substituted to eliminate restriction sites or to include silent restriction sites, which may aid in processing of the DNA in the selected host cell. The vehicle, linker and peptide DNA sequences may be modified to include any of the foregoing sequence changes. 5 Additional derivatives include non-peptide analogs that provide a stabilized structure or lessened biodegradation, are also contemplated. Peptide mimetic analogs can be prepared based on a selected inhibitory peptide by replacement of one or more residues by nonpeptide moieties. Preferably, the nonpeptide moieties permit the peptide to retain its natural confirmation, or stabilize a preferred, e.g., bioactive, confirmation which retains the ability to recognize and bind 10 myostatin. hi one aspect, the resulting analog/mimetic exhibits increased binding affinity for myostatin. One example of methods for preparation of nonpeptide mimetic analogs from peptides is described in Nachman et al., Regul Pept 57:359-370 (1995). If desired, the peptides of the invention can be modified, for instance, by glycosylation, amidation, carboxylation, or phosphorylation, or by the creation of acid addition salts, amides, esters, in particular C-tenninal 15 esters, and N-acyl derivatives of the peptides of the invention. The peptibodies also can be modified to create peptide derivatives by forming covalent or noncovalent complexes with other moieties. Covalently-bound complexes can be prepared by linking the chemical moieties to functional groups on the side chains of amino acids comprising the peptibodies, or at the N- or C-terminus. 20 In particular, it is anticipated that the peptides can be conjugated to a reporter group, including, but not limited to a radiolabel, a fluorescent label, an enzyme (<?.g., that catalyzes a colorimetric or fluorometric reaction), a substrate, a solid matrix, or a carrier (e.g., biotin or avidin). The invention accordingly provides a molecule comprising a peptibody molecule, wherein the molecule preferably further comprises a reporter group selected from the group 25 consisting of a radiolabel, a fluorescent label, an enzyme, a substrate, a solid matrix, and a carrier. Such labels are well known to those of skill in the art, e.g., biotin labels are particularly contemplated. The use of such labels is well known to those of skill in the art and is described in, e.g., U.S. Patent Nos.3,817,837; 3,850,752; 3,996,345; and 4,277,437. Other labels that will be useful include but are not limited to radioactive labels, fluorescent labels and chemiluminescent 30 labels. U.S. Patents concerning use of such labels include, for example, U.S. Patent Nos. 3,817,837; 3,850,752; 3,939,350; and 3,996,345. Any of the peptibodies of the present invention may comprise one, two, or more of any of these labels.

Methods of Making Peptides and Peptibodies 31 205265/2

The peptides of the present invention can be generated using a wide variety of techniques known in the art. For example, such peptides can be synthesized in solution or on a solid support in accordance with conventional techniques. Various automatic synthesizers are commercially available and can be used in accordance with known protocols. See, for example, Stewart and 5 Young {supra)', Tam et al., J Am Chem Soc, 105:6442, (1983); Merrifield, Science 232:341-347 (1986); Barany and Merrifield, The Peptides. Gross and Meienhofer, eds, Academic Press, New York, 1-284; Barany et al., Int J Pep Protein Res, 30:705-739 (1987); andU.S. Patent No. 5,424,398.

Solid phase peptide synthesis methods use a copoly(styrene-divinylbenzene) containing 10 0.1-1.0 mM amines/g polymer. These methods for peptide synthesis use butyloxycarbonyl (t- BOC) or 9-fluorenylmethyloxy-carbonyl(FMOC) protection of alpha-amino groups. Both methods involve stepwise syntheses whereby a single amino acid is added at each step starting from the C-terminus of the peptide (See, Coligan et al., Curr Prot Immunol, Wiley Interscience, 1991, Unit 9). On completion of chemical synthesis, the synthetic peptide can be deprotected to 15 remove the t-BOC or FMOC amino acid blocking groups and cleaved from the polymer by treatment with acid at reduced temperature (e.g., liquid HF-10% anisole for about 0.25 to about 1 hours at 0°C). After evaporation of the reagents, the peptides are extracted from the polymer with 1% acetic acid solution that is then lyophilized to yield the crude material. This can normally be purified by such techniques as gel filtration on Sephadex G-15 using 5% acetic acid as a solvent. 20 Lyophilization of appropriate fractions of the column will yield the homogeneous peptides or peptide derivatives, which can then be characterized by such standard techniques as amino acid analysis, thin layer chromatography, high performance liquid chromatography, ultraviolet absorption spectroscopy, molar rotation, solubility, and quantitated by the solid phase Edman degradation. 25 Phage display techniques can be particularly effective in identifying the peptides of the present invention as described above. Briefly, a phage library is prepared (using e.g. ml 13, fd, or lambda phage), displaying inserts from 4 to about 80 amino acid residues. The inserts may represent, for example, a completely degenerate or biased array. Phage-bearing inserts that bind to the desired antigen are selected and this process repeated through several cycles of reselection 30 of phage that bind to the desired antigen. DNA sequencing is conducted to identify the sequences of the expressed peptides. The minimal linear portion of the sequence that binds to the desired antigen can be determined in this way. The procedure can be repeated using a biased library containing inserts containing part or all of the minimal linear portion plus one or more additional degenerate residues upstream or downstream thereof. These techniques may identify peptides of 35 the invention with still greater binding affinity for myostatin than agents already identified herein. 32 WO 2004/058988 PCT/US2003/040781

Regardless of the maimer in which the peptides are prepared, a nucleic acid molecule encoding each such peptide can be generated using standard recombinant DNA procedures. The nucleotide sequence of such molecules can be manipulated as appropriate without changing the amino acid sequence they encode to account for the degeneracy of the nucleic acid code as well as 5 to account for codon preference in particular host cells.

The present invention also provides nucleic acid molecules comprising polynucleotide sequences encoding the peptides and peptibodies of the present invention. These nucleic acid molecules include vectors and constructs containing polynucleotides encoding the peptides and peptibodies of the present invention, as well as peptide and peptibody variants and derivatives. 10 Exemplary nucleic acid molecules are provided in the Examples below.

Recombinant DNA techniques also provide a convenient method for preparing full length peptibodies and other large polypeptide binding agents of the present invention, or fragments thereof.. A polynucleotide encoding the peptibody or fragment may be inserted into an expression vector, which can in turn be inserted into a host cell for production of the binding 15 agents of the present invention. Preparation of exemplary peptibodies of the present invention are described in Example 2 below. A variety of expression vector/host systems may be utilized to express the peptides and peptibodies of the invention. These systems include but are not limited to microorganisms such as bacteria transformed with recombinant bacteriophage, plasmid or cosmid DNA expression 20 vectors; yeast transformed with yeast expression vectors; insect cell systems infected with virus expression vectors (e.g., baculovirus); plant cell systems transfected with vims expression vectors (e.g., cauliflower mosaic virus, CaMV; tobacco mosaic virus, TMV) or transformed with bacterial expression vectors (e.g., Ti or pBR322 plasmid); or animal cell systems. One preferred host cell line is E.coli strain 2596 (ATCC # 202174), used for expression of peptibodies as described 25 below in Example 2. Mammalian cells that are useful in recombinant protein productions include but are not limited to VERO cells, HeLa cells, Chinese hamster ovary (CHO) cell lines, COS cells (such as COS-7), W138, BHK, HepG2,3T3, RIN, MDCK, A549, PC12, K562 and 293 cells.

The term “expression vector” refers to a plasmid, phage, virus or vector, for expressing a polypeptide from a polynucleotide sequence. An expression vector can comprise a transcriptional 30 unit comprising an assembly of (1) a genetic element or elements having a regulatory role in gene expression, for example, promoters or enhancers, (2) a structural or sequence that encodes the binding agent which is transcribed into mRNA and translated into protein, and (3) appropriate transcription initiation and termination sequences. Structural units intended for use in yeast or eukaryotic expression systems preferably include a leader sequence enabling extracellular 33 WO 2004/058988 PCT/US2003/040781 secretion of translated protein by a host cell. Alternatively, where recombinant protein is expressed without a leader or transport sequence, it may include an amino terminal methionyl residue. This residue may or may not be subsequently cleaved from the expressed recombinant protein to provide a final peptide product. 5 For example, the peptides and peptibodies may be recombinantly expressed in yeast using a commercially available expression system, e.g., the Pichia Expression System (Invitrogen, San Diego, CA), following the manufacturer’s instructions. This system also relies on the pre-proalpha sequence to direct secretion, but transcription of the insert is driven by the alcohol oxidase (AOX1) promoter upon induction by methanol. The secreted peptide is purified from the yeast 10 growth medium using the methods used to purify the peptide from bacterial and mammalian cell supernatants.

Alternatively, the cDNA encoding the peptide and peptibodies may be cloned into the baculovirus expression vector pVL1393 (PharMingen, San Diego, CA). This vector can be used according to the manufacturer’s directions (PharMingen) to infect Spodoptera frugiperda cells in 15 sF9 protein-free media and to produce recombinant protein. The recombinant protein can be purified and concentrated from the media using a heparin-Sepharose column (Pharmacia).

Alternatively, the peptide or peptibody may be expressed in an insect system. Insect systems for protein expression are well known to those of skill in the art. In one such system, Autographa califomica nuclear polyhedrosis virus (AcNPV) can be used as a vector to express 20 foreign genes in Spodoptera frugiperda cells or in Trichoplusia larvae. The peptide coding sequence can be cloned into a nonessential region of the virus, such as the polyhedrin gene, and placed under control of the polyhedrin promoter. Successful insertion of the peptide will render the polyhedrin gene inactive and produce recombinant virus lacking coat protein coat. The recombinant viruses can be used to infect S. frugiperda cells or Trichoplusia larvae in which the 25 peptide is expressed (Smith et al., J Virol 46: 584 (1983); Engelhard et al., Proc Nat Acad Sci (USA) 91: 3224-7 (1994)).

In another example, the DNA sequence encoding the peptide can be amplified by PCR and cloned into an appropriate vector for example, pGEX-3X (Pharmacia). The pGEX vector is designed to produce a fusion protein comprising glutathione-S-transferase (GST), encoded by the 30 vector, and a protein encoded by a DNA fragment inserted into the vector’s cloning site. The primers for PCR can be generated to include for example, an appropriate cleavage site. Where the fusion moiety is used solely to facilitate expression or is otherwise not desirable as an attachment to the peptide of interest, the recombinant fusion protein may then be cleaved from the GST portion of the fusion protein. The pGEX-3X/specific binding agent peptide construct is 35 transformed into E. coli XL-1 Blue cells (Stratagene, La Jolla CA), and individual transformants 34 WO 2004/058988 PCT/LS2003/040781 isolated and grown. Plasmid DNA from individual transformants can be purified and partially sequenced using an automated sequencer to confirm the presence of the desired specific binding agent encoding nucleic acid insert in the proper orientation.

The fusion protein, which may be produced as an insoluble inclusion body in the bacteria, 5 can be purified as follows. Host cells are collected by centrifugation; washed in 0.15 M NaCl, 10 mM Tris, pH 8, 1 mM EDTA; and treated with 0.1 mg/ml lysozyme (Sigma, St. Louis, MO) for 15 minutes at room temperature. The lysate can be cleared by sonication, and cell debris can be pelleted by centrifugation for 10 minutes at 12,000 X g. The fusion protein-containing pellet can be resuspended in 50 mM Tris, pH 8, and 10 mM EDTA, layered over 50% glycerol, and 10 centrifuged for 30 min. at 6000 X g. The pellet can be resuspended in standard phosphate buffered saline solution (PBS) free of Mg++ and Ca++. The fusion protein can be further purified by fractionating the resuspended pellet in a denaturing SDS-PAGE (Sambrook et al., supra). The gel can be soaked in 0.4 M KC1 to visualize the protein, which can be excised and electroeluted in gel-running buffer lacking SDS. If the GST/fusion protein is produced in bacteria as a soluble 15 protein, it can be purified using the GST Purification Module (Pharmacia).

The fusion protein may be subjected to digestion to cleave the GST from the peptide of

the invention. The digestion reaction (20-40 mg fusion protein, 20-30 units human thrombin (4000 U/mg, Sigma) in 0.5 ml PBS can be incubated 16-48 hrs at room temperature and loaded on a denaturing SDS-PAGE gel to fractionate the' reaction products. The gel can be soaked in 0.4 M 20 KC1 to visualize the protein bands. The identity of the protein band corresponding to the expected molecular weight of the peptide can be confirmed by amino acid sequence analysis using an automated sequencer (Applied Biosystems Model 473A, Foster City, CA). Alternatively, the identity can be confirmed by performing HPLC and/or mass spectometry of the peptides.

Alternatively, a DNA sequence encoding the peptide can be cloned into a plasmid 25 containing a desired promoter and, optionally, a leader sequence (Better et al., Science 240:1041-43 (1988)). The sequence of this construct can be confirmed by automated sequencing. The plasmid can then be transformed into E. cali strain MC1061 using standard procedures employing CaC12 incubation and heat shock treatment of the bacteria (Sambrook et al., supra). The transformed bacteria can be grown in LB medium supplemented with carbenicillin, and 30 production of the expressed protein can be induced by growth in a suitable medium. If present, the leader sequence can effect secretion of the peptide and be cleaved during secretion.

Mammalian host systems for the expression of recombinant peptides and peptibodxes are well known to those of skill in the art. Host cell strains can be chosen for a particular ability to process the expressed protein or produce certain post-translation modifications that will be useful 35 in providing protein activity. Such modifications of the protein include, but are not limited to, 35 WO 2004/058988 PCT/US2003/040781 acetylation, carboxylation, glycosylation, phosphorylation, lipidation and acylation. Different host cells such as CHO, HeLa, MDCK, 293, WI38, and the like have specific cellular machinery and characteristic mechanisms for such post-translational activities and can be chosen to ensure the correct modification and processing of the introduced, foreign protein. 5 It is preferable that transformed cells be used for long-term, high-yield protein production. Once such cells are transformed with vectors that contain selectable markers as well as the desired expression cassette, the cells can be allowed to grow for 1-2 days in an enriched media before they are switched to selective media. The selectable marker is designed to allow growth and recovery of cells that successfully express the introduced sequences. Resistant 10 clumps of stably transformed cells can be proliferated using tissue culture techniques appropriate to the cell line employed. A number of selection systems can be used to recover the cells that have been transformed for recombinant protein production. Such selection systems include, but are not limited to, HSV thymidine kinase, hypoxanthine-guanine phosphoribosyltransferase and adenine 15 phosphoribosyltransferase genes, in tk-, hgprt- or aprt- cells, respectively. Also, anti-metabolite resistance can be used as the basis of selection for dhfr which confers resistance to methotrexate; gpt which confers resistance to mycophenolic acid; neo which confers resistance to the aminoglycoside G418 and confers resistance to chlorsulfuron; and hygro which confers resistance to hygromycin. Additional selectable genes that may be useful include trpB, which allows cells to 20 utilize indole in place of tryptophan, or hisD, which allows cells to utilize histinol in place of histidine. Markers that give a visual indication for identification of transformants include anthocyanins, β-glucuronidase and its substrate, GUS, and luciferase and its substrate, luciferin.

Purification and Refolding of Binding Agents 25 In some cases, the binding agents such as the peptides and/or peptibodies of this invention may need to be “refolded” and oxidized into a proper tertiary structure and disulfide linkages generated in order to be biologically active. Refolding can be accomplished using a number of procedures well known in the art. Such methods include, for example, exposing the solubilized polypeptide agent to a pH usually above 7 in the presence of a chaotropic agent. The selection of 30 chaotrope is similar to the choices used for inclusion body solubilization, however a chaotrope is typically used at a lower concentration. Exemplary chaotropic agents are guanidine and urea. In most cases, the refolding/oxidation solution will also contain a reducing agent plus its oxidized form in a specific ratio to generate a particular redox potential which allows for disulfide shuffling to occur for the formation of cysteine bridges. Some commonly used redox couples include 35 cysteine/cystamine, glutathione/dithiobisGSH, cupric chloride, dithiothreitol DTT/dithiane DTT, 36 WO 2004/058988 PCT/US2003/040781 and 2-mercaptoethanol (bME)/dithio-bME. In many instances, a co-solvent may be used to increase the efficiency of the refolding. Commonly used cosolvents include glycerol, polyethylene glycol of various molecular weights, and arginine.

It may be desirable to purify the peptides and peptibodies of the present invention. 5 Protein purification techniques are well known to those of skill in the art. These techniques involve, at one level, the crude fractionation of the proteinaceous and non-proteinaceous fractions. Having separated the peptide and/or peptibody from other proteins, the peptide or polypeptide of interest can be further purified using chromatographic and electrophoretic techniques to achieve partial or complete purification (or purification to homogeneity). Analytical methods particularly 10 suited to the preparation of peptibodies and peptides or the present invention are ion-exchange chromatography, exclusion chromatography; polyacrylamide gel electrophoresis; isoelectric focusing. A particularly efficient method of purifying peptides is fast protein liquid chromatography or even HPLC.

Certain aspects of the present invention concern the purification, and in particular 15 embodiments, the substantial purification, of a peptibody or peptide of the present invention. The term “purified peptibody or peptide” as used herein, is intended to refer to a composition, isolatable from other components, wherein the peptibody or peptide is purified to any degree relative to its naturally-obtainable state. A purified peptide or peptibody therefore also refers to a peptibody or peptide that is free from the environment in which it may naturally occur. 20 Generally, “purified” will refer to a peptide or peptibody composition that has been subjected to fractionation to remove various other components, and which composition substantially retains its expressed biological activity. Where the term “substantially purified” is used, this designation will refer to a peptide or peptibody composition in which the peptibody or peptide forms the major component of the composition, such as constituting about 50%, about - 25 60%, about 70%, about 80%, about· 90%, about 95% or more of the proteins in the composition.

Various methods for quantifying the degree of purification of the peptide or peptibody will be known to those of skill in the art in light of the present disclosure. These include, for example, determining the specific binding activity of an active fraction, or assessing the amount of peptide or peptibody within a fraction by SDS/PAGE analysis. A preferred method for 30 assessing the purity of a peptide or peptibody fraction is to calculate the binding activity of the fraction, to compare it to the binding activity of the initial extract, and to thus calculate the degree of purification, herein assessed by a “-fold purification number.” The actual units used to represent the amount of binding activity will, of course, be dependent upon the particular assay technique chosen to follow the purification and whether or not the peptibody or peptide exhibits a 35 detectable binding activity. 37' WO 2004/058988 PCT/US2003/040781

Various techniques suitable for use in purification will be well known to those of skill in the art. These include, for example, precipitation with ammonium sulphate, PEG, antibodies (immunoprecipitation) and the like or by heat denaturation, followed by centrifugation; chromatography steps such as affinity chromatography.(e.g., Protein-A-Sepharose), ion exchange, 5 gel filtration, reverse phase, hydroxylapatite and affinity chromatography; isoelectric focusing; gel electrophoresis; and combinations of such and other techniques. As is generally known in the art, it is believed that the order of conducting the various purification steps may be changed, or that certain steps may be omitted, and still result in a suitable method for the preparation of a substantially purified binding agent. 10 There is no general requirement that the binding agents of the present invention always be provided in their most purified state. Indeed, it is contemplated that less substantially purified binding agent products will have utility in certain embodiments. Partial purification may be accomplished by using fewer purification steps in combination, or by utilizing different forms of the same general purification scheme. For example, it is appreciated that a cation-exchange 15 column chromatography performed utilizing an HPLC apparatus will generally result in a greater “-fold” purification than the same technique utilizing a low-pressure chromatography system. Methods exhibiting a lower degree of relative purification may have advantages in total recovery of the peptide or peptibody, or in maintaining binding activity of the peptide or peptibody.

It is known that the migration of a peptide or polypeptide can vary, sometimes 20 significantly, with different conditions of SDS/PAGE (Capaldi et al., Biochem Biophys Res Comm, 76: 425 (1977)). It will therefore be appreciated that under differing electrophoresis conditions, the apparent molecular weights of purified or partially purified binding agent expression products may vary. 25 Activity of Myostatin Binding Agents

After the construction of the binding agents of the present invention, they are tested for their ability to bind myostatin and inhibit or block myostatin activity. Any number of assays or animal tests may be used to determine the ability of the agent to inhibit or block myostatin activity. Several assays used for characterizing the peptides and peptibodies of the present 30 invention are described in the Examples below. One assay is the C2C12 pMAJRE-luc assay which makes use of a myostatin-responsive cell line (C2C12 myoblasts) transfected with a luciferase reporter vector containing myostatin/activin response elements (MAKE). Exemplary peptibodies are assayed by pre-incubating a series of peptibody dilutions with myostatin, and then exposing the cells to the incubation mixture. The resulting luciferase activity is determined, and a titration 35 curve is generated from the series of peptibody dilutions. The IC50 (the concentration of 38 WO 2004/058988 PCT/TJS2003/040781 peptibody to achieve 50% inhibition of myostatin activity as measured by luciferase activity) was then determined. A second assay described below is aBIAcore® assay to determine the kinetic parameters ka (association rate constant), kd (dissociation rate constant), and KD (dissociation equilibrium constant) for the myostatin binding agents. Lower dissociation equilibrium constants 5 (Kd, expressed in nM) indicated a greater affinity of the peptibody for myostatin. Additional assays include blocking assays, to determine whether a binding agent such as a peptibody is neutralizing (prevents binding of myostatin to its receptor), or non-neutralizing (does not prevent binding of myostatin to its receptor); selectivity assays, which determine if the binding agents of the present invention bind selectively to myostatin and not to other TGFB family members; and 10 KinEx A™ assays or solution-based equilibrium assays, which also determine Kd and are considered to be more sensitive in some circumstances. These assays are described in Example 3.

Figure 1 shows the IC50 of a peptide compared with the IC50 of the peptibody form of the peptide. This demonstrates that the peptibody is significantly more effective at inhibiting myostatin activity than the peptide alone. In addition, affinity-matured peptibodies generally 15 exhibit improved IC50 and KD values compared with the-parent peptides and peptibodies. The IC50 values for a number of exemplary affinity matured peptibodies are shown in Table VII, Example 7 below. Additionally, in some instances, making a 2x version of a peptibody, where two peptides are attached in tandem, increase the activity of the peptibody both in vitro and in vivo. 20 In vivo activities are demonstrated in the Examples below. The activities of the binding agents include anabolic acitivity increasing lean muscle mass in animal models, as well as decreasing the fat mass with respect to total body weight in treated animal models, and increasing muscular strength in animal models. 25 Uses of the Myostatin Binding Agents

The myostatin binding agents of the present invention bind to myostatin and block or inhibit myostatin signaling within targeted cells. The present invention provides methods and reagents for reducing the amount or activity of myostatin in an animal by administering an effective dosage of one or more myostatin binding agents to the animal. In one aspect, the present 30 invention provides methods and reagents for treating myostatin-related disorders in an animal comprising administering an effective dosage of one or more binding agents to the animal. These myostatin-related disorders include but are not limited to various forms of muscle wasting, as well as metabolic disorders such as diabetes and related disorders, and bone degenerative diseases such as osteoporosis. 39 WO 2004/058988 PCT/US2003/040781

As shown in the Example 8 below, exemplary peptibodies of the present invention dramatically increases lean muscle mass in the CD1 nu/nu mouse model. This in vivo activity correlates to the in vitro binding and inhibitory activity described below for the same peptibodies.

Muscle wasting disorders include dystrophies such as Duchenne’s muscular dystrophy, 5 progressive muscular dystrophy, Becker’s type muscular dystrophy, Dejerine-Landouzy muscular dystrophy, Erb’s muscular dystrophy, and infantile neuroaxonal muscular dystrophy. For example, blocking myostatin through use of antibodies in vivo improved the dystrophic phenotype of the mdx mouse model of Duchenne muscular dystrophy (Bogdanovich et al, Nature 420, 28 (2002)). The peptibodies of the present invention increase lean muscle mass as a percentage of 10 body weight and decreases fat mass as percentage of body weight when administered to an aged mdx mouse model.

Additional muscle wasting disorders arise from chronic disease such as amyotrophic lateral sclerosis, congestive obstructive pulmonary disease, cancer, AIDS, renal failure, and rheumatoid arthritis. For example, cachexia or muscle wasting and loss of body weight was 15 induced in athymic nude mice by a systemically administered myostatin (Zimmers et al., supra).

In another example, serum and intramuscular concentrations of myostatin-immunoreactive protein was found to be increased in men exhibiting AIDS-related muscle wasting and was inversely •related to fat-free mass (Gonzalez-Cadavid et al., PNAS USA 95: 14938-14943 (1998)).

Additional conditions resulting in muscle wasting may arise from inactivity due to disability such 20 as confinement in a wheelchair, prolonged bedrest due to stroke, illness, spinal chord injury, bone fracture or trauma, and muscular atrophy in a microgravity environment (space flight). For example, plasma myostatin immunoreactive protein was found to increase after prolonged bedrest (Zachwieja et al. J Gravit Physiol. 6(2): 11(1999). It was also found that the muscles of rats exposed to a microgravity environment during a space shuttle flight expressed an increased 25 amount of myostatin compared with the muscles of rats which were not exposed (Lalani et al., J.Endocrin 167 (3):417-28 (2000)).

In addition, age-related increases in fat to muscle ratios, and age-related muscular atrophy appear to be related to myostatin. For example, the average serum myostatin-immunoreactive protein increased with age in groups of young (19-35 yr old), middle-aged (36-75 yr old), and 30 elderly (76-92 yr old) men and women, while the average muscle mass and fat-free mass declined with age in these groups (Yarasheski et al. J Nutr Aging 6(5):343-8 (2002)). It has also been shown that myostatin gene knockout in mice increased myogenesis and decreased adipogenesis (Lin et al., Biochem Biophys Res Commun 291(3):701-6 (2002), resulting in adults with increased muscle mass and decreased fat accumulation and leptin secretion. Exemplary peptibodies 35 improve the lean muscle mass to fat ratio in aged mdx mice as shown below. 40 WO 2004/058988 PCT/US2003/040781

In addition, myostatin has now been found to be expressed at low levels in heart muscle and expression is upregulated after cardiomyocytes after infarct (Shanna et al., J Cell Physiol. 180 (1):1-9 (1999)). Therefore, reducing myostatin levels in the heart muscle may improve recovery of heart muscle after infarct. 5 Myostatin also appears to influence metabolic disorders including type 2 diabetes, noninsulin-dependent diabetes mellitus, hyperglycemia, and obesity. For example, lack of myostatin has been shown to improve the obese and diabetic phenotypes of two mouse models (Yen et al. supra). It has been demonstrated in the Examples below that decreasing myostatin activity by administering the inhibitors of the present invention will decreases the fat to muscle 10 ratio in an animal, including aged animal models. Therefore, decreasing fat composition by-administering the inhibitors of the present invention will improve diabetes, obesity, and hyperglycemic conditions in animals.

In addition, increasing muscle mass by reducing myostatin levels may improve bone strength and reduce osteoporosis and other degenerative bone diseases. It has been found, for 15 example, that myostatin-deficient mice showed increased mineral content and density of the mouse humerus and increased mineral content of both trabecular and cortical bone at the regions where the muscles attach, as well as increased muscle mass (Hamrick et al. Calcif Tissue Int 71(1):63-8 (2002)). .

The present invention also provides methods and reagents for increasing muscle mass in 20 food animals by administering an effective dosage of the myostatin binding agent to the animal. Since the mature C-terminal myostatin polypeptide is identical in all species tested, myostatin binding agents would be expected to be effective for increasing muscle mass and reducing fat in any agriculturally important species including cattle, chicken, turkeys, and pigs.

The binding agents of the present invention may be used alone or in combination with 25 other therapeutic agents to enhance their therapeutic effects or decrease potential side effects. The binding agents of the present invention possess one or more desirable but unexpected combination of properties to improve the therapeutic value of the agents. These properties include increased activity, increased solubility, reduced degradation, increased half-life, reduced toxicity, and reduced immunogenicity. Thus the binding agents of the present invention are useful for 30 extended treatment regimes. In addition, the properties of hydrophilicity and hydrophobicity of the compounds of the invention are well balanced, thereby enhancing their utility for both in vitro and especially in vivo uses. Specifically, compounds of the invention have an appropriate degree of solubility in aqueous media that permits absorption and bioavailability in the body, while also having a degree of solubility in lipids that permits the compounds to traverse the cell membrane to 35 a putative site of action, such as a particular muscle mass. 41 205265/2

The binding agents of the present invention are useful for treating a “subject” or any animal, including humans, when administered in an effective dosages in a suitable composition.

In addition, the mystatin binding agents of the present invention are useful for detecting and quantitating myostatin in a number of assays. These assays are described in more detail 5 below.

In general, the binding agents of the present invention are useful as capture agents to bind and immobilize myostatin in a variety of assays, similar to those described, for example, in Asai, ed., Methods in Cell Biology, 37, Antibodies in Cell Biology, Academic Press, Inc., New York (1993). The binding agent may be labeled in some manner or may react with a third molecule 10 such as an anti-binding agent antibody which is labeled to enable myostatin to be detected and quantitated. For example, a binding agent or a third molecule can be modified with a detectable moiety, such as biotin, which can then be bound by a fourth molecule, such as enzyme-labeled streptavidin, or other proteins. (Akerstrom, J Immunol 135:2589 (1985); Chaubert, Mod Pathol 10:585 (1997)). 15 Throughout any particular assay, incubation and/or washing steps may be required after each combination of reagents. Incubation steps can vary from about 5 seconds to several hours, preferably from about 5 minutes to about 24 hours. However, the incubation time will depend upon the assay format, volume of solution, concentrations, and the like. Usually, the assays will be carried out at ambient temperature, although they can be conducted over a range of 20 temperatures.

Non-competitive binding assays:

Binding assays can be of the non-competitive type in which the amount of captured myostatin is directly measured. For example, in one preferred “sandwich” assay, the binding agent can be bound directly to a solid substrate where it is immobilized. These immobilized 25 agents then bind to myostatin present in the test sample. The immobilized myostatin is then bound with a labeling agent, such as a labeled antibody against myostatin, which can be detected. In another preferred “sandwich” assay, a second agent specific for the binding agent can be added which contains a detectable moiety, such as biotin, to which a third labeled molecule can specifically bind, such as streptavidin. (See, Harlow and Lane, Antibodies, A Laboratory Manual, 30 Ch 14. Cold Spring Harbor Laboratory, NY (Ί988)).

Competitive Binding Assays:

Binding assays can be of the competitive type. The amount of myostatin present in the sample is measured indirectly by measuring the amount of myostastin displaced, or competed away, from a binding agent by the myostatin present in the sample. In one preferred competitive 42 WO 2004/058988 PCT/US2003/040781 binding assay, a known amount of myostastin, usually labeled, is added to the sample and the sample is then contacted with the binding agent. The amount of labeled myostastin bound to the binding agent is inversely proportional to he concentration of myostastin present in the sample, (following the protocols found in, for example Harlow and Lane, Antibodies, A Laboratory 5 Manual, Ch 14, pp. 579-583, supra}.

In another preferred competitive binding assay, the binding agent is immobilized on a solid substrate. The amount of myostastin bound to the binding agent may be determined either by measuring the amount of myostatin present in a myostatin/binding agent complex, or alternatively by measuring the amount of remaining uncomplexed myostatin. 10 Other Binding Assays’

The present invention also provides Western blot methods to detect or quantify the ·.. presence of myostatin in a sample. The technique generally comprises separating sample proteins by gel electrophoresis on the basis of molecular weight and transferring the proteins to a suitable solid support, such as nitrocellulose filter, a nylon filter, or derivatized nylon filter. The sample is 15 incubated with the binding agents or fragments thereof that bind myostatin and the resulting complex is detected. These binding agents may be directly labeled or alternatively may be subsequently detected using labeled antibodies that specifically bind to the binding agent.

Diagnostic Assays 20 The binding agents or fragments thereof of the present invention may be useful for the diagnosis of conditions or diseases characterized by increased amounts of myostatin. Diagnostic assays for high levels of myostatin include methods utilizing a binding agent and a label to detect myostatin in human body fluids, extracts of cells or specific tissue extracts. For example, serum levels of myostatin may be measured in an individual over tune to determine the onset of muscle 25 wasting associated with aging or inactivity, as described, for example, in Yarasheski et al., supra. Increased myostatin levels were shown to correlate with average decreased muscle mass and fat-free mass in groups of men and women of increasing ages (Yarasheski et al., supra}. The binding agents of the present invention may be useful for monitoring increases or decreases in the levels of myostatin with a given individual over time, for example. The binding agents can be used in 30 such assays with or without modification. In a preferred diagnostic assay, the binding, agents will be labeled by attaching, e.g., a label or a reporter molecule. A wide variety of labels and reporter molecules are known, some of which have been already described herein. In particular, the present invention is useful for diagnosis of human disease. 43 WO 2004/058988 PCT/US2003/040781 A variety of protocols for measuring myostatin proteins using binding agents of myostatin are known in the art. Examples include enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA) and fluorescence activated cell sorting (FACS).

For diagnostic applications, in certain embodiments the binding agents of the present 5 invention typically will be labeled with a detectable moiety. The detectable moiety can be any one that is capable of producing, either directly or indirectly, a detectable signal. For example, the detectable moiety may be a radioisotope, such as 3H, 14C, 32P, 35S, or 125I, a fluorescent or chemiluminescent compound, such as fluorescein isothiocyanate, rhodamine, or luciferin; or an . z enzyme, such as alkaline phosphatase, Pgalactosidase, or horseradish peroxidase (Bayer et al., 10 Meth Enz, 184: 138 (1990)).

Pharmaceutical Compositions

Pharmaceutical compositions of myostatin binding agents such as peptibodies described herein are within the scope of the present invention. Such compositions comprise a 15 therapeutically or prophylactically effective amount of a myostatin binding agent, fragment, variant, or derivative thereof as described herein, in admixture with a pharmaceutically acceptable agent. In a preferred embodiment, pharmaceutical compositions comprise antagonist binding agents that inhibit myostatin partially or completely in admixture with a pharmaceutically acceptable agent. Typically, the myostastin binding agents will be sufficiently purified for 20 administration to an animal.

The pharmaceutical composition may contain formulation materials for modifying, maintaining or preserving, for example, the pH, osmolarity, viscosity, clarity, color, isotonicity, odor, sterility, stability, rate of dissolution or release, adsorption or penetration of the composition. Suitable formulation materials include, but are not limited to, amino acids (such as 25 glycine, glutamine, asparagine, arginine or lysine); antimicrobials; antioxidants (such as ascorbic acid, sodium sulfite or sodium hydrogen-sulfite); buffers (such as borate, bicarbonate, Tris-HCl, citrates, phosphates, other organic acids); bulking agents (such as mannitol or glycine), chelating agents (such as ethylenediamine tetraacetic acid (EDTA)); complexing agents (such as caffeine, polyvinylpyrrolidone, beta-cyclodextrin or'hydroxypropyl-beta-cyclodextrin); fillers; 30 monosaccharides; disaccharides and other carbohydrates (such as glucose, mannose, or dextrins); proteins (such as serum albumin, gelatin or immunoglobulins); coloring; flavoring and diluting agents; emulsifying agents; hydrophilic polymers (such as polyvinylpyrrolidone); low molecular weight polypeptides; salt-forming counterions (such as sodium); preservatives (such as benzalkonium chloride, benzoic acid, salicylic acid, thimerosal, phenethyl alcohol, 35 methylparaben, propylparaben, chlorhexidine, sorbic acid or hydrogen peroxide); solvents (such 44 WO 2004/058988 PCT/US2003/040781 as glycerin, propylene glycol or polyethylene glycol); sugar alcohols (such as mannitol or sorbitol); suspending agents; surfactants or wetting agents (such as pluronics, PEG, sorbitan esters, polysorbates such as polysorbate 20, polysorbate SO, triton, tromethamine, lecithin, cholesterol, tyloxapal); stability enhancing agents (sucrose or sorbitol); tonicity enhancing agents 5 (such as alkali metal halides (preferably sodium or potassium chloride, mannitol sorbitol); delivery vehicles; diluents; excipients and/or pharmaceutical adjuvants. (Remington’s · Pharmaceutical Sciences, 18th Edition, A.R. Gennaro, ed., Mack Publishing Company, 1990).

The optimal pharmaceutical composition will be determined by one skilled in the art depending upon, for example, the intended route of administration, delivery format, and desired 10 dosage. See for example, Remington’s Pharmaceutical Sciences, supra. Such compositions may influence the physical state, stability, rate of in vivo release, and rate of in vivo clearance of the binding agent.

The primary vehicle or carrier in a pharmaceutical composition may be either aqueous or non-aqueous in nature. For example, a suitable vehicle or carrier may be water for injection, 15 physiological saline solution or artificial cerebrospinal fluid, possibly supplemented with other materials common in compositions for parenteral administration. Neutral buffered saline or saline mixed with serum albumin are further exemplary vehicles. Other exemplary pharmaceutical compositions comprise Tris buffer of about pH 7.0-8.5, or acetate buffer of about pH 4.0-5.5, which may further include sorbitol or a suitable substitute therefore. In one embodiment of the 20 present invention, binding agent compositions may be prepared for storage by mixing the selected composition having the desired degree of purity with optional formulation agents (Remington’s Pharmaceutical Sciences, supra) in the form of a lyophilized cake or an aqueous solution.

Further, the binding agent product may be formulated as a lyophilizate using appropriate excipients such as sucrose. 25 The pharmaceutical compositions can be selected for parenteral delivery. Alternatively, the compositions may be selected for inhalation or for enteral delivery such as orally, aurally, opthalmically, rectally, or vaginally. The preparation of such pharmaceutically acceptable compositions is within the skill of the art.

The formulation components are present in concentrations that are acceptable to the site 30 of administration. For example, buffers are used to maintain the composition at physiological pH or at slightly lower pH, typically within a pH range of from about 5 to about 8.

When parenteral administration is contemplated, the therapeutic compositions for use in this invention may be in the form of a pyrogen-free, parenterally acceptable aqueous solution comprising the desired binding agent in a pharmaceutically acceptable vehicle. A particularly 35 suitable vehicle for parenteral injection is sterile distilled water in which a binding agent is 45 205265/2 formulated as a sterile, isotonic solution, properly preserved. Yet another preparation can involve the formulation of the desired molecule with an agent, such as injectable microspheres, bio-erodible particles, polymeric compounds (polylactic acid, polyglycolic acid), beads, or liposomes, that provides for the controlled or sustained release of the product which may then be delivered 5 via a depot injection. Hyaluronic acid may also be used, and this may have the effect of promoting sustained duration in the circulation. Other suitable means for the introduction of the desired molecule include implantable drug delivery devices.

In another aspect, pharmaceutical formulations suitable for parenteral administration may be formulated in aqueous solutions, preferably in physiologically compatible buffers such as 10 Hanks’ solution, ringer’s solution, or physiologically buffered saline. Aqueous injection suspensions may contain substances that increase the viscosity of the suspension, such as sodium carboxymethyl cellulose, sorbitol, or dextran. Additionally, suspensions of the active compounds may be prepared as appropriate oily injection suspensions. Suitable lipophilic solvents or vehicles include fatty oils, such as sesame oil, or synthetic fatty acid esters, such as ethyl oleate, 15 triglycerides, or liposomes. Non-lipid polycationic amino polymers may also be used for delivery. Optionally, the suspension may also contain suitable stabilizers or agents to increase the solubility of the ompounds and allow for the preparation of highly concentrated solutions.In another embodiment, a pharmaceutical composition may be formulated for inhalation. For example, a binding agent may be formulated as a dry powder for inhalation. Polypeptide or 20 nucleic acid molecule inhalation solutions may also be formulated with a propellant for aerosol delivery. In yet another embodiment, solutions may be nebulized. Pulmonary administration is further described in PCT Application No. PCT/US94/001875 (WO 1994/020069), which describes pulmonary delivery of chemically modified proteins.

It is also contemplated that certain formulations may be administered orally. In one 25 embodiment of the present invention, binding agent molecules that are administered in this fashion can be formulated with or without those carriers customarily used in the compounding of solid dosage forms such as tablets and capsules. For example, a capsule may be designed to release the active portion of the formulation at the point in the gastrointestinal tract when bioavailability is maximized and pre-systemic degradation is minimized. Additional agents can 30 be included to facilitate absorption of the binding agent molecule. Diluents, flavorings, low melting point waxes, vegetable oils, lubricants, suspending agents, tablet disintegrating agents, and binders may also be employed.

Pharmaceutical compositions for oral administration can also be formulated using pharmaceutically acceptable carriers well known in the art in dosages suitable for oral 46 WO 2004/058988 PCT/US2003/040781 administration. Such carriers enable the pharmaceutical compositions to be formulated as tablets, pills, dragees, capsules, liquids, gels, syrups, slurries, suspensions, and the like, for ingestion by the patient.

Pharmaceutical preparations for oral use can be obtained through combining active 5 compounds with solid excipient and processing the resultant mixture of granules (optionally, after grinding) to obtain tablets or dragee cores. Suitable auxiliaries can be added, if desired. Suitable excipients include carbohydrate or protein fillers, such as sugars, including lactose, sucrose, mannitol, and sorbitol; starch from com, wheat, rice, potato, or other plants; cellulose, such as methyl cellulose, hydroxypropylmethyl-cellulose, or sodium carboxymethylcellulose; gums, 10 including arabic· and tragacanth; and proteins, such as gelatin and collagen. If desired, disintegrating or solubilizing agents may be added, such as the cross-linked polyvinyl pyrrolidone, agar, and alginic acid or a salt thereof, such as sodium alginate.

Dragee cores may be used in conjunction with suitable coatings, such as concentrated sugar solutions, which may also contain gum arabic, talc, polyvinylpyrrolidone, carbopol gel, 15 polyethylene glycol, and/or titanium dioxide, lacquer solutions, and suitable organic solvents or solvent mixtures. Dyestuffs or pigments may be added to the tablets or dragee coatings for product identification or to characterize the quantity of active compound, i.e., dosage.

Pharmaceutical preparations that can be used orally also include push-fit capsules made of gelatin, as well as soft, sealed capsules made of gelatin and a coating, such as glycerol or 20 sorbitol. Push-fit capsules can contain active ingredients mixed with fillers or binders, such as lactose or starches, lubricants, such as talc or magnesium stearate, and, optionally, stabilizers. In soft capsules, the active compounds may be dissolved or suspended in suitable liquids, such as fatty oils, liquid, or liquid polyethylene glycol with or without stabilizers.

Another pharmaceutical composition may involve an effective quantity of binding agent 25 in a mixture with non-toxic excipients that are suitable for the manufacture of tablets. By dissolving the tablets in sterile water, or other appropriate vehicle, solutions can be prepared in unit dose form. Suitable excipients include, but are not limited to, inert diluents, such as calcium carbonate, sodium carbonate or bicarbonate, lactose, or calcium phosphate; or binding agents, such as starch, gelatin, or acacia; or lubricating agents such as magnesium stearate, stearic acid, or 30 talc.

Additional pharmaceutical compositions will be evident to those skilled in the art, including formulations involving binding agent molecules in sustained- or controlled-delivery formulations. Techniques for formulating a variety of other sustained- or controlled-delivery means, such as liposome carriers, bio-erodible microparticles or porous beads and depot 35 injections, are also known to those skilled in the art. See for example, PCT/US93/00829 that 47 WO 2004/058988 PCT/US2003/040781 describes controlled release of porous polymeric microparticles for the delivery of pharmaceutical compositions. Additional examples of sustained-release preparations include semipermeable polymer matrices in the form of shaped articles, e.g. films, or microcapsules. Sustained release matrices may include polyesters, hydrogels, polylactides (U.S. 3,773,919, EP 58,481), 5 copolymers of L-glutamic acid and gamma ethyl-L-glutamate (Sidman et al., Biopolymers, 22:547-556 (1983), poly (2-hydroxyethyl-methacrylate) (Langer et al., J. Biomed. Mater. Res., 15:167-277, (1981); Langer et al., Chem. Tech., 12:98-105(1982)), ethylene vinyl acetate (Langer et al., supra) or poly-D(-)-3-hydroxybutyric acid (EP 133,988). Sustained-release compositions also include liposomes, which can be prepared by any of several methods known in the art. See 10 e.g., Eppstein et ai.,PNAS (USA), 82:3688 (1985); EP 36,676; EP 88,046; EP 143,949.

The pharmaceutical composition to be used for in vivo administration typically must be sterile. This may be accomplished by filtration through sterile filtration membranes. Where the composition is lyophilized, sterilization using this method may be conducted either prior to or following lyophilization and reconstitution. The composition for parenteral administration may 15 be stored in lyophilized form or in solution. In addition, parenteral compositions generally are placed into a container having a sterile access port, for example, an intravenous solution bag or vial having a stopper pierceable by a hypodermic injection needle.

Once the pharmaceutical composition has been formulated, it may be stored in sterile vials as a solution, suspension, gel, emulsion, solid, or a dehydrated or lyophilized powder. Such 20 formulations may be stored either in a ready-to-use form or in a form (e.g., lyophilized) requiring reconstitution prior to administration.

In a specific embodiment, the present invention is directed to kits for producing a singledose administration unit. The kits may each contain both a first container having a dried protein and a second container having an aqueous formulation. Also included within the scope of this 25 · invention are kits containing single and multi-chambered pre-filled syringes (e.g., liquid syringes and lyosyringes).

An effective amount of a pharmaceutical composition to be employed therapeutically will depend, for example, upon the therapeutic context and objectives. One skilled in the art will appreciate that the appropriate dosage levels for treatment will thus vary depending, in part, upon 30' the molecule delivered, the indication for which the binding agent molecule is being used, the route of administration, and the size (body weight, body surface or organ size) and condition (the age and general health) of the patient. Accordingly, the clinician may titer the dosage and modify the route of administration to obtain the optimal therapeutic effect. A typical dosage may range from about O.lmg/kg to up to about 100 mg/kg or more, depending on the factors mentioned 48 WO 2004/058988 PCT/US2003/040781 above. In other embodiments, the dosage may range from 0.1 mg/kg up to about 100 mg/kg; or 1 mg/kg up to about 100 mg/kg; or 5 mg/kg up to about 100 mg/kg.

For any compound, the therapeutically effective dose can be estimated initially either in cell culture assays or in animal models such as mice, rats, rabbits, dogs, pigs, or monkeys. An 5 animal model may also be used to determine the appropriate concentration range and route of administration. Such information can then be used to determine useful doses and routes for administration in humans.

The exact dosage will be determined in light of factors related to the subject requiring treatment. Dosage and administration are adjusted to provide sufficient levels of the active 10 compound or to maintain the desired effect. Factors that may be taken into account include the. severity of the disease state, the general health of the subject, the age, weight, and gender of the subject, time and frequency of administration, drug combination(s), reaction sensitivities, and response to therapy. Long-acting pharmaceutical compositions may be administered every 3 to 4 days, every week, or biweekly depending on the half-life and clearance rate of the particular 15 formulation.

The frequency of dosing will depend upon the pharmacokinetic parameters of the binding agent molecule in the formulation used. Typically, a composition is administered until a dosage is reached that achieves the desired effect. The composition may therefore be administered as a single dose, or as multiple doses (at the same or different concentrations/dosages) over time, or as 20 a continuous infusion. Further refinement of the appropriate dosage is routinely made. Appropriate dosages may be ascertained through use of appropriate dose-response data.

The route of administration of the pharmaceutical composition is in accord with known methods, e.g. orally, through injection by intravenous, intraperitoneal, intracerebral (intra-parenchymal), intracerebroventricular, intramuscular, intra-ocular, intraarterial, intraportal, 25 intralesional routes, intramedullary, intrathecal, intraventricular, transdermal, subcutaneous, intraperitoneal, intranasal, enteral, topical, sublingual, urethral, vaginal, or rectal means, by sustained release systems or by implantation devices. Where desired, the compositions may be administered by bolus injection or continuously by infusion, or by implantation device.

Alternatively or additionally, the composition may be administered locally via 30 implantation of a membrane, sponge, or another appropriate material on to which tire desired molecule has been absorbed or encapsulated. Where an implantation device is used, the device may be implanted into any suitable tissue or organ, and delivery of the desired molecule may be via diffusion, timed-release bolus, or continuous administration. 49 WO 2004/058988 PCT/US2003/040781

In some cases, it may be desirable to use pharmaceutical compositions in an ex vivo manner. In such instances, cells, tissues, or organs that have been removed from the patient are exposed to the pharmaceutical compositions after which the cells, tissues and/or organs are subsequently implanted back into the patient. 5 In other cases, a binding agent of the present invention such as a peptibody can be delivered by implanting certain cells that have been genetically engineered, using methods such as those described herein, to express and secrete the polypeptide. Such cells may be animal or human cells, and may be autologous, heterologous, or xenogeneic. Optionally, the cells may be immortalized. In order to decrease the chance of an immunological response, the cells may be 10 encapsulated to avoid infiltration of surrounding tissues. The encapsulation materials are typically biocompatible, semi-permeable polymeric enclosures or membranes that allow the release of the protein produces) but prevent the destruction of the cells by the patient’s immune system or by other detrimental factors from the surrounding tissues.

Pharmaceutical compositions containing the binding agents of the present invention, are 15 administered to a subject to treat any myostatin-related disorders. These include muscle-wasting disorders including but not limited to muscular dystrophy, muscle wasting in cancer, AIDS, muscle atrophy, rheumatoid arthritis, renal failure/uremia, chronic heart failure, prolonged bedrest, spinal chord injury, stroke, and aging related sarcopenia. In addition these compositions are administed to treat obesity, diabetes, hyperglycemia, and increase bone density, 20

The invention having been described, the following examples are offered by way of illustration, and not limitation.

Example 1 25 Identification of myostatin -binding peptides

Three filamentous phage libraries, TN8-IX (5X109 independent transformants), TN12-I (1.4X109 independent transformants), and linear (2.3X109 independent transformants) (Dyax Corp.) were used to select for myostatin binding phage. Each library was incubated on myostatin-coated surfaces and subjected to different panning conditions: non-specific elution, and specific 30 elution using recombinant human activin receptor HB/Fc chimera (R&D Systems, Inc.,

Minneapolis, Minnesota), or myostastin propeptide elution as described below. For all three libraries, the phages were eluted in a non-specific manner for the first round of selection, while the receptor and promyostatin was used in the second and third rounds of selection. The selection procedures were carried out as described below. 50 WO 2004/058988 PCT/DS2003/040781

Preparation of mvostatin

Myostatin protein was produced recombinantly in the E.coli K-12 strain 2596 (ATCC # 202174) as follows. Polynucleotides encoding the human promyostatin molecule were cloned

5 into the pAMG21 expression vector (ATCC No. 98113), which was derived from expression vector pCFM1656 (ATCC No. 69576) and the expression vector system described in United States Patent No. 4,710,473, by following the procedure described in published International Patent Application WO 00/24782. The polynucleotides encoding promyostatin were obtained from a mammalian expression vector. The coding region was amplified using a standard PCR 10 method and the following PCR primers to introduce the restriction site for Ndel and BamHL 5’ primer- 5’-GAGAGAGAGCATATGAATGAGAACAGTGAGCAAAAAG-3’ (SeqlDNo: 292) 3’primer: 5’-AGAGAGGGATCCATTATGAGCACCCACAGCGGTC-3’ (SeqlDNo: 293) ' The PCR product and vector were digested with both enzymes, mixed and ligated. The 15 product of the ligation was transformed into E. coli strain #2596. Single colonies were checked microscopically for recombinant protein expression in the form of inclusion bodies. The plasmid was isolated and sequenced through the coding region of the recombinant gene to verify genetic fidelity.

Bacterial paste was generated from a 10L fermentation using a batch method at 37°C. 20 The culture was induced with HSL at a cell density of 9.6 OD6oo and harvested six hours later at a density of 104 OD600. The paste was stored at -80°C. E.coli paste expressing promyostatin was lysed in a microfluidizer at 16,000psi, centrifuged to isolate the insoluble inclusion body fraction. Inclusion bodies were resuspended in guanidine hydrochloride containing dithiothreitol and solubilized at room temperature. This was then diluted 30 fold in an aqueous buffer. The 25 refolded promyostatin was then concentrated and buffer exchanged into 20mM Tris pH 8.0, and applied to an anion exchange column. The anion exchange column was eluted with an increasing sodium chloride gradient. The fractions containing promyostatin were pooled. The promyostatin produced in E.coli is missing the first 23 amino acids and begins with a methionine before the residue 24 asparagine. To produce mature myostatin, the pooled promyostatin was enzymatically 30 cleaved between the propeptide and mature myostatin C terminal. The resulting mixture was then applied to a C4-ipHPLC column using a increasing gradient of acetonitrile containing 0.1% trifluoroacetic acid. Fractions containing mature myostatin were pooled and dried in a speed-vac.

The recombinant mature myostatin produced from E. coli was tested in the myoblast C2C12 based assay described below and found to be fully active when compared with 51 WO 2004/058988 PCT/US2003/040781 recombinant murine myostatin commercially produced in a mammalian cell system (R&D Systems, Inc., Minneapolis, Minnesota). The X.coZz'-produced mature myostatin was used in the phage-display and screening assays described below.

Preparation of Myostatin-Coated Tubes 5 Myostatin was immobilized on 5 ml Immuno ™ Tubes (NUNC) at a concentration of S ug of myostatin protein in 1 ml of 0.1M sodium carbonate buffer (pH 9.6). The myostatin-coated Immuno™ Tube was incubated with orbital shaking for 1 hour at room temperature. Myostatin-coated Immuno™ Tube was then blocked by adding 5 ml of 2% milk-PBS and incubating at room temperature for 1 hour with rotation. The resulting myostatin-coated Immuno ™ Tube was 10 then washed three times with PBS before being subjected to the selection procedures. Additional Immuno ™ Tubes were also prepared for negative selections (no myostatin). For each panning condition, five to ten Immuno™ Tubes were subjected to the above procedure except that the Immuno™ Tubes were coated with 1ml of 2% BSA-PBS instead of myostatin protein. _ Negative Selection 15 For each panning condition, about 100 random library equivalents for TN8-IX and TN12- I libraries (5X1011 pfu for TN8-IX, and 1.4X1011 pfu for TN12-I) and about 10 random library equivalents for the linear library (2.3X1O10 pfu) were aliquoted from the library stock and diluted to 1 ml with PBST (PBS with 0.05% Tween-20). The 1 ml of diluted library stock was added to an Immuno™ Tube prepared for the negative selection, and incubated for 10 minutes at room 20 temperature with orbital shaking. The phage supernatant was drawn out and addeii to the second Immuno™ Tube for another negative selection step. In this way, five to ten negative selection steps were performed.

Selection for Myostatin Binding

After the last negative selection step above, the phage supernatant was added to the 25 prepared myostatin coated Immuno™ Tubes. The Immuno™ Tube was incubated with orbital shaking for one hour at room temperature, allowing specific phage to bind to myostatin. After the supernatant was discarded, the Immuno™ Tube was washed about 15 times with 2% milk-PBS, 10 times with PBST and twice with PBS for the three rounds of selection with all three libraries (TN8-IX, TN12-I, and Linear libraries) except that for the second round of selections with TN8- 30 IX and TN12-I libraries, the Immuno™ Tube was washed about 14 times with 2% milk-PBS, twice with 2% BSA-PBS, 10 times with PBST and once with PBS.

Non-specific elution

After the last washing step, the bound phages were eluted from the Immuno ™ Tube by adding 1 ml of 100 mM triethylamine solution (Sigma, St. Louis, Missouri) with 10-minute 52 WO 2004/058988 PCT/TJS2003/040781 incubation with orbital shaking. The pH of the phage containing solution was then neutralized with 0.5 ml of 1 M Tris-HCl (pH 7.5).

Receptor (Human Activin Receptor) elution of bound phage

For round 2 and 3, after the last washing step, the bound phages were eluted from the 5 frnmuno ™ Tube by adding 1 ml of 1 μΜ of receptor protein (recombinant human activin receptor HB/Fc chimera, R&D Systems, Inc., Minneapolis, Minnesota) with a 1-hour incubation for each condition.

Propeptide elution of bound phage

For round 2 and 3, after the last washing step, the bound phages were eluted from the 10 Immuno ™ Tube by adding 1 ml of 1 μΜ propeptide protein (made as described above) with a 1-hour incubation for each condition.

Phage Amplification

Fresh E.coli. (XL-1 Blue MRF’) culture was grown to ODsoo = 0-5 in LB media containing 12.5 ug/ml tetracycline. For each panning condition, 20 ml of this culture was chilled 15 on ice and centrifuged. The bacteria pellet was resuspended in 1 ml of the min A salts solution.

Each mixture from different elution methods was added to a concentrated bacteria sample and incubated at 37°C for 15 minutes. 2 ml of NZCYM media (2x NZCYM, 50 ug/ml Ampicillin) was added to each mixture and incubated at 37°C for 15 minutes. The resulting 4 ml solution was plated on a large NZCYM agar plate containing 50 ug/ml ampicillin and incubated 20 overnight at 37°C.

Each of the bacteria/phage mixture that was grown overnight on a large NZCYM agar plate was scraped off in 35 ml of LB media, and the agar plate was further rinsed with additional 35 ml of LB media. The resulting bacteria/phage mixture in LB media was centrifuged to pellet the bacteria away. 50 ul of the phage supernatant was transferred to a fresh tube, and 12.5 ml of 25 PEG solution (20% PEG8000, 3.5M ammonium acetate) was added and incubated on ice for 2 hours to precipitate phages. The precipitated phages were centrifuged down and resuspended in 6 ml of the phage resuspension buffer (250 mM NaCl, 100 mM Tris pH8, 1 mM EDTA). This phage solution was further purified by centrifuging away the remaining bacteria and precipitating the phage for the second time by adding 1.5 ml of the PEG solution. After a centrifugation step, 30 the phage pellet was resuspended in 400 ul of PBS. This solution was subjected to a final centrifugation to rid of remaining bacteria debris. The resulting phage preparation was titered by a standard plaque formation assay (Molecular Cloning, Maniatis et al., 3rd Edition).

Additional rounds of selection and amplification 53 WO 2004/058988 PCT/US2003/040781

In the second round, the amplified phage (1011 pfu) from the first round was used as the input phage to perform the selection and amplification steps. The amplified phage (1011 pfu) from the second round in turn was used as the input phage to perform third round of selection and amplification. After the elution steps of the third round, a small fraction of the eluted phage was 5 plated out as in the plaque formation assay above. Individual plaques were picked and placed into 96 well microtiter plates containing 100 ul of TE buffer in each well. These master plates were incubated at 4°C overnight to allow phages to elute into the TE buffer.

Clonal Analysis

Phage ELISA 10 The phage clones were subjected to phage ELISA and then sequenced. The sequences were ranked as discussed below.

Phage ELISA was performed as follows. An E. Coli XL-1 Blue MRF’culture was grown until OP600 reached 0.5. 30 ul of this culture was aliquoted into each well of a 96 well microtiter plate. 10 ul of eluted phage was added to each well and allowed to infect bacteria for 15 min at 15 room temperature. About 120 ul of LB media containing 12.5 ug/ml of tetracycline and 50 ug/ml of ampicillin were added to each well. The microtiter plate was then incubated with shaking overnight at 37 °C. Myostatin protein (2 ug/ml in 0.1M sodium carbonate buffer, pH 9.6) was allowed to coat onto a 96 well Maxisorp™ plates (NUNC) overnight at 4°C. As a control, a separate Maxisorp™ plate was coated with 2% BSA prepared in PBS. 20 On the following day, liquid in the protein coated Maxisorp™ plates was discarded, washed three times with PBS and each well was blocked with 300 ul of 2% milk solution at room temperature for 1 hour. The milk solution was discarded, and the wells were washed three times with the PBS solution. After the last washing step, about 50 ul of PBST-4% milk was added to each well' of the protein-coated Maxisorp™ plates. About 50 ul of overnight cultures from each 25 well in the 96 well xnicrotiter plate was transferred to the corresponding wells of the myostatin coated plates as well as the control 2% BSA coated plates. The 100 ul mixture in the two kinds of plates were incubated for 1 hour at room temperature. The liquid was discarded from the Maxisorp™ plates, and the wells were washed about three times with PBST followed by two times with PBS. The HRP-conjugated anti-M13 antibody (AmershamPharmacia Biotech) was 30 diluted to about 1:7,500, and 100 ul of the diluted solution was added to each well of the

Maxisorp™ plates for 1 hour incubation at room temperature. The liquid was again discarded and the wells were washed about three times with PBST followed by two time with PBS. 100 ul of LumiGlo™ Chemiluminescent substrate (KPL) was added to each well of the Maxisorp™ plates and incubated for about 5 minutes for reaction to occur. The chemiluminescent unit of the 35 Maxisorp ™ plates was read on a plate reader (Lab System). 54 WO 2004/058988 PCT/US2003/040781

Sequencing of the phage clones

For each phage clone, the sequencing template was prepared by a PCR method. The following oligonucleotide pair was used to amplify a 500 nucleotide fragment: primer #1: 5’- 5 . CGGCGCAACTATCGGTATCAAGCTG-3’ (Seq ID No: 294) and primer #2: 5’- CATGTACCGTAACACTGAGTTTCGTC-3’ (Seq ID No: 295)'. The following mixture was prepared for each clone.

Reagents Volume (pL) / tube distilled H2O 26.25 50% glycerol 10 10X PCR Buffer (w/o MgCl2) 5 25mMMgCl2 4 10 mM dNTP mix 1 100 μΜ primer 1 0.25 100 μΜ primer 2 0.25 Taq polymerase 0.25 Phage in TE (section 4) 3 Final reaction volume 50 . A thennocycler (GeneAmp PCR System 9700, Applied Biosystem) was used to run the following program: [94°C for 5min; 94°C for 30 sec, 55°C for 30 sec, 72°C for 45 sec.] x30 10 cycles; 72°C for 7 min; cool to 4DC. The PCR product from each reaction was cleaned up using the QIAquick Multiwell PCR Purification kit (Qiagen), following the manufacturer’s protocol. The PCR cleaned up product was checked by running 10 ul of each PCR reaction mixed with 1 ul of dye (10X BBXS agarose gel loading dye) on a 1% agarose gel. The remaining product was then sequenced using the ABI377 Sequencer (Perkin Elmer) following the manufacturer 15 recommended protocol.

Sequence Ranking and Analysis

The peptide sequences that were translated from the nucleotide sequences were correlated to ELISA data. The clones that showed high chemiluminescent units in the myostatin-coated wells and low chemiluminescent units in the 2% BSA-coated wells were identified. The 20 sequences that occurred multiple times were identified. Candidate sequences chosen based on these criteria were subjected to further analysis as peptibodies. Approximately 1200 individual clones were analyzed. Of these approximately 132 peptides were chosen for generating the peptibodies of the present invention. These are shown in Table I below. The peptides having 55 WO 2004/058988 PCT/US2003/040781 SEQ ID NO: 1 to 129 were used to generate peptibodies of the same name. The peptides having SEQ ID NO: 130 to 141 shown in Table 1 comprise two or more peptides from SEQ ID NO: 1 to 132 attached by a linker sequence. SEQ ID NO: 130 to 141 were also used to generate peptibodies of the same name. 5 Consensus sequences were determined for the TN-8 derived group of peptides. These are as follows: KDXCXXWHWMCKPX (SeqlDNo: 142) WXXCXXXGFWCXNX (SeqlDNo: 143) IXGCXWWDXXCYXX (Seq ID No: 144) 10 XXWCVSPXWFCXXX (SeqlDNo: 145) XXXCPWFAXXCVDW (Seq ID No: 146)

For all of the above consensus sequences, the underlined “core sequences” from each consensus sequence are the amino acid which always occur at that position. “X” refers to any naturally occurring or modified amino acid. The two cysteines contained with the core sequences were 15 fixed amino acids in the TN8-IX library.

TABLET

PEPTIBODYNAME SEQ-IDNo PEPTIDE SEQUENCE

Myostatin-TN8-Conl 1 KDKCKMWHWMCKPP Myostatin-TN8-Con2 2 KDLCAMWHWMCKPP Myostatin-TN8-Con3 3 KDLCKMWKWMCKPP Myostatin-TN8-Con4 4 KDLCKMWHWMCKPK Myostatin-TN8-Con5 5 WYPCYEFHFWCYDL Myostatin-TN8-Con6 6 WYPCYEGHFWCYDL Myostatin-TN8-Con7 7 IFGCKWWDVQCYQF Myostatin-TN8-Con8 8 IFGCKWWDVDCYQF Myostatin-TN8-Con9 9 ADWCVSPNWFCMVM Myostatin-TN8-Con 10 10 HKFCPWWALFCWDF Myostatin-TN8-1 11 KDLCKMWHWMCKPP Myostatin-TN8-2 12 IDKCAIWGWMCPPL Myostatin-TN8-3 13 WYPCGEFGMWCLNV Myostatin-TN8-4 14 WFTCLWNCDNE Myostatin-TN8-5 15 HTPCPWFAPLCVEW Myostatin-TN8-6 16 KEWCWRWKWMCKPE Myostatin-TN8-7 17 FETCPSWAYFCLDI Myostatin-TN8-8 18 AYKCEANDWGCWWL Myostatin-TN8-9 19 NSWCEDQWHRCWWL Myostatin-TN8-10 20 WSACYAGHFWCYDL Myostatin-TN8-U 21 ANWCVSPNWFCMVM Myostatin-TN8-12 22 WTECYQQEFWCWNL Myostatin-TN8-13 23 ENTCERWKWMCPPK Myostatin-TN8-14 24 WLPCHQEGFWCMNF Myostatin-TN8-15 25 STMCSQWHWMCNPF 56 WO 2004/058988 PCT/US2003/040781

My ostatin-TN8-16 26 IFGCHWWDVDCYQF Myostatin-TN8-17 27 IYGCKWWDIQCYDI Myostatin-TN8-18 28 PDWCEDPDWWCKFW Myostatin-TN8-19 29 QGHCTRWPWMCPPY Myostatin-TN8-20 30 WQECYREGFWCLQT Myostatin-TN8-21 31 WFDCYGPGFKCWSP Myostatin-TN8-22 32 gvrcpkghlwclyp Myostatin-TN8-23 33 HWACGYWPWSCKWV Myostatin-TN8-24 34 GPACHSPWWWCVFG Myostatin-TN8-25 35 TTWCISPMWFCSQQ Myostatin-TN8-26 36 HKFCPPWAIFCWDF Myostatin-TN8-27 37 PDWCVSPRWYCNMW Myostatin-TN8-28 38 VWKCHWFGMDCEPT Myostatin-TN8-29 39 KKHCQIWTWMCAPK Myostatin-TN8-30 40 WFQCGSTLFWCYNL Myostatin-TN8-31 41 WSPCYDHYFYCYTI Myostatixi-TN8-32 42 SWMCGFFKEVCMWV Myostatin-TN8 -33 43 EMLCM1HPVFCNPH Myostatin-TN8-34 44 LKTCNLWPWMCPPL Myostatin-TN8-35 45 WGCKWYEAWCYNK Myostatin-TN8-36 46 PIHCTQWAWMCPPT Myostatin-TN8-37 47 DSNCPWYFLSCVIF Myostatin-TN8-38 48 H1WCNLAMMKCVEM Myostatin-TN8-39 49 NLQCIYELGKCIYF Myostatin-TN8-40 50 AWRCMWFSDVCTPG Myostatin-TN8-41 51 WFRCFLDADWCTSV Myostatin-TN8-42 52 EKICQMWSWMCAPP Myostatin-TN8-43 53 WFYCHLNKSECTEP Myostatin-TN8^44 54 FWRCAIGIDKCKRV Myostatin-TN8-45 55 NLGCKWYEVWCFTY Myostatin-TN8-46 56 IDLCNMWDGMCYPP Myostatin-TN8-47 57 EMPCNIWGWMCPPV Myostatin-TNl 2-1 58 WFRCVLTGIVDWSECFGL Myostatin-TNl 2-2 59 GFSCTFGLDEFYVDCSPF Myostatin-TN12-3 60 LPWCHDQVNADWGFCMLW My ostatin-TNl 2-4 61 YPTCSEKFWIYGQTCVLW Myostatin-TN12-5 62 LGPCPIHHGPWPQYCVYW Myostatin-TN12-6 63 PFPCETHQISWLGHCLSF Myostatin-TN12-7 64 HWGCEDLMWSWHPLCRRP Myostatin-TN12-8 65 LPLCDADMMPTIGFCVAY Myostatin-TN12-9 66 SHWCETTFWMNYAKCVHA Myostatin-TN12-10 67 LPKCTHVPFDQGGFCLWY Myostatin-TN12-11 68 FSSCWSPVSRQDMFCVFY My ostatin-TN 12-13 69 SHKCEYSGWLQPLCYRP Myostatin-TN12-14 70 PWWCQDNYVQHMLHCDSP My ostatin-TN 12-15 71 WFRCMLMNSFDAFQCVSY Myostatin-TN 12-16 72 PDACRDQPWYMFMGCMLG My ostatin-TN 12-17 73 FLACFVEFELCFDS Myostatin-TN12-18 74 SAYCIHESDPYVLCVPL 57 WO 2004/058988 PCT/US2003/040781

Myostatin-TN12-19 75 PSICESYSTMWLPMCQHN Myostatin-TN12-20 76 WLDCHDDSWAWTKMCRSH Myostatin-TN12-21 77 YLNCVMMNTSPFVECVFN Myostatin-TN12-22 78 YPWCDGEMIQQGrrCMFY Myostatin-TN12-23 79 FDYCTWLNGFKDWKCWSR Myostatin-TN12-24 80 LPLCNLKEISHVQACVLF Myostatin-TN12-25 81 SPECAFARWLG1EQCQRD Myostatin-TN12-26 82 . YPQCFNLHLLEWTECDWF Myostatin-TN12-27 83 RWRCEIYDSEFLPKCWFF Myostatin-TN12-28 84 LVGCDNVWHRCKLF Myostatin-TN12-29 85 AGWCHVWGEMFGMGCSAL Myostatin-TN12-30 86 HHECEWMARWMSLDCVGL Myostatin-TN12-31 87 FPMCGIAGMKDFDFCVWY Myostatin-TN12-32 88 RDDCTFWPEWLWKLCERP Myostatin-TN12-33 89 YNFCSYLFGVSKEACQLP Myostatin-TN12-34 90 AHWCEQGPWRYGNICMAY My ostatin-TN 12-35 91 NLVCGKISAWGDEACARA Myostatin-TN12-36 92 HNVCTIMGPSMKWFCWND Myostatin-TN 12-37 93 NDLCAMWGWRNTIWCQNS Myostatin-TN12-38 94 PPFCQNDNDMLQSLCKLL Myostatin-TN12-39 95 WYDCNVPNELLSGLCRLF Myostatin-TN12-40 96 YGDCDQNHWMWPFTCLSL Myostatin-TN12-41 97 GWMCHFDLHDWGATCQPD Myostatin-TN12A2 98 YFHCMFGGHEFEVHCESF Myostatin-TN12-43 99 AYWCWHGQCVRF Myostatin-Linear-1 100 SEHWTFTDWDGNEWWVRPF - Myostatin-Linear-2 101 MEMLDSLFELLKDMVPISKA Myostatin-Linear-3 102 SPPEEALMEWLGWQYGKFT Myostatin-Linear-4 103 SPENLLNDLYILMTKQEWYG Myostatin-Linear-5 104 FHWEEGIPFHVVTPYSYDRM Myostatin-Linear-6 105 KRLLEQEMNDLAELVSGHS Myostatin-Linear-7 106 DTRDALFQEFYEFVRSRLVI Myostatin-Linear-8 107 RMSAAPRPLTYRDIMDQYWH Myostatin-Linear-9 108 NDKAHFFEMFMFDVHNFVES Myostatin-Linear-10 109 QTQAQKIDGLWELLQSIRNQ Myostatin-Linear-11 110 MLSEFEEFLGNLVHRQEA Myostatin-Liiiear-12 111 YTPKMGSEWTSFWHNRJHYL My ostatin-Linear-13 112 LNDTLLRELKMVLNSLSDMK Myostatin-Linear-14 113 FDVERDLMRWLEGFMQSAAT Myostatin-Linear-15 114 HHGWNYLRKGSAPQWFEAWV Myostatin-Linear-16 115 VESLHQLQMWLDQKLASGPH Myostatin-Linear-17 116 RATLLKDFWQLVEGYGDN Myostatin-Linear-18 117 EELLREFYRFVSAFDY My ostatin-Linear-19 118 GLLDEFSHFIAEQFYQMPGG Myostatin-Linear-20 119 YREMSMLEGLLDVLERLQHY Myostatin-Linear-21 120 HNSSQMLLSELIMLVGSMMQ Myostatin-Linear-22 121 WREHFLNSDYIRDKLIAIDG Myostatin-Linear-23 122 QFPFYVFDDLPAQLEYWIA 58 WO 2004/058988 PCT/US2003/040781

Myostatin-Linear-24 123 EFFHWLHNHRSEVNHWLDMN Myostatin-Linear-25 124 EALFQNFFRDVLTLSEREY Myostatin-Linear-26 125 QYWEQQWMTYFRENGLHVQY Myostatin-Linear-27 126 . NQKMMLEDLWRJMTPMFGRS Myostatm-Linear-29 127 FLDELKAELSRHYALDDLDE Myostatin-Linear-30 128 GKLIEGLLNELMQLETFMPD Myostatin-Linear-31 129 ILLLDEYKKDWKSWF Myostatin-2xTN8-19 kc 130 QGHCTRWPWMCPPYGSGSATGGS GSTASSGSGSATGQGHCTRWPWM CPPY Myostatin-2xTN8-con6 131 WYPCYEGHFWCYDLGSGSTASSG SGSATGWYPCYEGHFWCYDL Myostatin-2xTN8-5 kc 132 HTPCPWFAPLCVEWGSGSATGGSG STASSGSGSATGHTPCPWFAPLCV EW Myostatin-2xTN8-18 kc 133 PDWCIDPDWWCKFWGSGSATGGS GSTASSGSGSATGPDWCIDPDWW CKFW Myostatin-2xTN8-ll kc 134 ANWCVSPNWFCMVMGSGSATGG SGSTASSGSGSATGANWCVSPNWF CMVM Myostatin-2xTN8-25 kc 135 PDWCIDPDWWCKFWGSGSATGGS GSTASSGSGSATGPDWCIDPDWW CKFW Myostatin-2xTN8-23 kc 136 HWACGYWPWSCKWVGSGSATGG SGSTASSGSGSATGHWACGYWPW SCKWV Myostatin-TN8-29-19 kc 137 KKHCQIWTWMCAPKGSGSATGGS GSTASSGSGSATGQGHCTRWPWM CPPY Myostatin-TN8-19-29 kc 138 QGHCTRWPWMCPPYGSGSATGGS GSTASSGSGSATGKKHCQIWTWM CAPK Myostatin-TN8-29-19 kn 139 KKHCQIWTWMCAPKGSGSATGGS GSTASSGSGSATGQGHCTRWPWM CPPY Myostatin-TN8-29-19-8g 140 KKHCQIWTWMCAPKGGGGGGGG QGHCTRWPWMCPPY Myostatin-TN8-19-29-6gc 141 QGHCTRWPWMCPPYGGGGGGKK HCQIWTWMCAPK

Example 2

Generating peptibodies 5 Construction of DNA encoding peptide-Fc fusion proteins

Peptides capable of binding myostatin were used alone or in combination with each other to construct fusion proteins in which a peptide was fused to the Fc domain of human IgGl. The 59 205265/2 amino acid sequence of the Fc portion of each peptibody is as follows (from amino terminus to carboxyl terminus):

DKTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDV

SHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVL

5 HQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREPQVYTLPPS RDELTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVL DSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLS LSPGK (Seq ID No: 296)

The peptide was fused in the N configuration (peptide was attached to the N-terminus of 10 the Fc region), the C configuration (peptide was attached to the C-terminus of the Fc region), or the N,C configuration (peptide attached both at the N and C terminus of the Fc region). Separate vectors were used to express N-terminal fusions and C-terminal fusions. Each peptibody was constructed by annealing pairs of oligonucleotides (“oligos”) to the selected phage nucleic acid to generate a double stranded nucleotide sequence encoding the peptide. These polynucleotide 15 molecules were constructed as ApaL to Xhol fragments. The fragments were ligated into either the pAMG21-Fc N-terminal vector for the N-terminal orientation, or the pAMG21-Fc-C-terminal vector for the C-terminal orientation which had been previously digested withZ/?aLI andATioI. The resulting ligation mixtures were transformed by electroporation into E. coli strain 2596 or 4167 cells (a hsdR- variant of strain 2596 cells) using standard procedures. Clones were screened 20 for the ability to produce the recombinant protein product and to possess the gene fusion having a correct nucleotide sequence. A single such clone was selected for each of the modified peptides.

Many of constructs were created using an alternative vector designated pAMG21-2xBs-N(ZeoR) Fc. This vector is simlar to the above-described vector except that the vector digestion was performed with fozwBI. Some constructs fused peptide sequences at both ends of the Fc. In 25 those cases the vector was a composite of pAMG21-2xBs-N(ZeoR) Fc and pAMG21-2xBs-C-Fc. Construction of pAMG21

Expression plasmid pAMG21 (ATCC No. 98113) is derived from expression vector pCFM1656 (ATCC No. 69576) and the expression vector system described in United States Patent No. 4,710,473, by following the procedure described in published International Patent 30 Application WO 00/24782.

Fc N-terminal Vector

The Fc N-terminal vector was constructed using the pAMG21 Fc_Gly5_ Tpo vector as a template. A 5’ PCR primer (below) was designed to remove the Tpo peptide sequence in pAMG 35 Tpo Gly5 and replace it with a polylinker containing ZpnLI and Xhol sites. Using this vector as a template, PCR was performed with Expand Long Polymerase, using the following 5’ primer and a universal 3’ primer: 60 WO 2004/058988 PCT/US2003/040781 5’primer: 5’-ACAAACAAACATATGGGTGCACAGAAAGCGGCCGCAAAAAAA CTCGAGGGTGGAGGCGGTGGGGACA-3’ (Seq ID No: 297) 3’primer: 5’-GGTCATTACTGGACCGGATC-3’ (Seq ID No: 298)

The resulting PCR product was gel purified and digested with restriction enzymes Ndel 5 and BsrGl. Both the plasmid and the polynucleotide encoding the peptide of interest together with its linker were gel purified using Qiagen (Chatsworth, CA) gel purification spin columns.

The plasmid and insert were then ligated using standard ligation procedures, and the resulting ligation mixture was transformed into E. coli cells (strain 2596). Single clones were selected and DNA sequencing was performed. A correct clone was identified and this was used as a vector 10 source for the modified peptides described herein.

Construction of Fc C-terminal Vector

The Fc C-terminal vector was constructed using pAMG21 Fc_Gly5_ Tpo vector as a template. A 3’ PCR primer was designed to remove the Tpo peptide sequence and to replace it with a polylinker containing ApoLI and Xhol sites. PCR was performed with Expand Long 15 Polymerase using a universal 5’ primer and the 3’ primer. 5’ Primer: 5’-CGTACAGGTTTACGCAAGAAAATGG-3’ (Seq ID No: 299) TTCTGTGCACCACCACCTCCACCTTTAC-3’ (Seq ID No: 300)

The resulting PCR product was gel purified and digested with restriction enzymes BsrGl 20 and PamHI. Both the plasmid and the polynucleotide encoding each peptides of interest with its linker were gel purified via Qiagen gel purification spin columns. The plasmid and insert were then ligated using standard ligation procedures, and the resulting ligation mixture was transformed into E. coli (strain 2596) cells. Strain 2596 (ATCC # 202174) is a strain of E. coli K.-12 modified to contain the lux promoter and two lambda temperature sensitive repressors, the 25 cI857s7 and the lac IQ repressor. Single clones were selected and DNA sequencing was performed. A correct clone was identified and used as a source of each peptibody described herein.

Expression in E. coli.

Cultures of each of the pAMG21-Fc fusion constructs in E. coli strain 2596 were grown 30 at 37°C in Terrific Broth medium (See Tartof and Hobbs, “Improved media for growing plasmid and cosmid clones”, Bethesda Research Labs Focus, Volume 9, page 12,1987, cited in aforementioned Sambrook et al. reference). Induction of gene product expression from the luxPR promoter was achieved following the addition of the synthetic autoinducer, N-(3-oxohexanoyl)-DL-homoserine lactone, to the culture medium to a final concentration of 20 nanograms per 35 milliliter (ng/ml). Cultures were incubated at 37°C for an additional six hours. The bacterial cultures were then examined by microscopy for the presence of inclusion bodies and collected by 61 WO 2004/058988 PCT/US2003/040781 centrifugation. Refractile inclusion bodies were observed in induced cultures, indicating that the Fc-fusions were most likely produced in the insoluble fraction in E. coli. Cell pellets were lysed directly by resuspension inLaemmli sample buffer containing 10% β-mercaptoethanol and then analyzed by SDS-PAGE. In most cases, an intense coomassie-stained band of the appropriate 5 molecular weight was observed on an SDS-PAGE gel.

Folding and purifying peptibodies

Cells were broken in water (1/10 volume per volume) by high pressure homogenization (3 passes at 15,000 PSI) and inclusion bodies were harvested by centrifugation (4000 RPM in J-6B for 30 minutes). Inclusion bodies were solubilized in 6 M guanidine, 50 mM Tris, 8 mM DTT, 10 pH 8.0 for 1 hour at a 1/10 ratio at ambient temperature. The solubilized mixture was diluted 25 times into 4 M urea, 20% glycerol, 50 mM Tris, 160 mM arginine, 3 mM cysteine, 1 mM cystamine, pH 8.5. The mixture was incubated overnight in the cold. The mixture was then dialyzed against 10 mM Tris pH 8.5, 50 mM NaCl, 1.5 M urea. After an overnight dialysis the pH of the dialysate was adjusted to pH 5 with acetic acid. The precipitate was removed by 15 centrifugation and the supernatant was loaded onto a SP-Sepharose Fast Flow column equilibrated in 10 mM NaAc, 50 mM NaCl, pH 5,4°C). After loading the column was washed to baseline with 10 mM NaAc, 50 mM NaCl, pH 5.2. The column was developed with a 20 column volume gradient from 50mM -500 mM NaCl in the acetate buffer. Alternatively, after the wash to baseline, the column was washed with 5 column volumes of 10 mM sodium phosphate pH 7.0 and 20 the column developed with a 15 column volume gradient from 0-400 mM NaCl in phosphate buffer. Column fractions were analyzed by SDS-PAGE. Fractions containing dimeric peptibody were pooled. Fractions were also analyzed by gel filtration to determine if any aggregate was present A number of peptibodies were prepared from the peptides of Table I. The peptides were 25 attached to the human IgGl Fc molecule to form the peptibodies in Table Π. Regarding the peptibodies in Table Π, the C configuration indicates that the peptide named was attached at the C-termini of the Fc. The N configuration indicates that the peptide named was attached at the N-termini of the Fc. The N,C configuration indicates that one peptide was attached at the N-tennini and one at the C-termini of each Fc molecule. The 2x designation indicates that the two peptides 30 named were attached in tandem to each other and also attached at the N or the C termini, or both the N,C of the Fc, separated by the linker indicated. Two peptides attached in tandem separated by a linker, are indicated, for example, as Myostatin-TN8-29-19-8g, which indicates that TN8-29 peptide is attached via a (gly)8 linker to TN8-19 peptide. The peptide(s) were attached to the Fc via a (gly)s linker sequence unless otherwise specified. In some instances the peptide(s) were

35 attached via a k linker. The linker designated k or lk refers to the gsgsatggagstassgsgsatg (Seq ID 62 WO 2004/058988 PCT/US2003/040781

No: 301) linker sequence, with kc referring to the linker attached, to the C-teiminus of the Fc, and kn referring to the linker attached to the N-terminus of the Fc. In Table Π below, column 4 refers to the linker sequence connecting the Fc to the first peptide and the fifth column refers to the configuration N or C or both. 5 Since the Fc molecule dimerizes in solution, a peptibody constructed so as to have one peptide will actually be a dimer with two copies of the peptide and two Fc molecules, and the 2X version having two peptides in tandem will actually be a dimer with four copies of the peptide and two Fc molecules.

Since the peptibodies given in Table Π are expressed in E. coli, the first amino acid 10 residue is Met (M). Therefore, the peptibodies in the N configuration are Met-peptide-linker-Fc, or Met-peptide-liriker-peptide-liriker-Fc, for example. Peptibodies in the C configuration are arranged as Met-Fc-liriker-peptide or Met-Fc-linker-peptide-liriker-peptide, for example. Peptibodies in the C,N configuration are a combination of both, for example, Met-peptide-liriker-Fc-liriker-peptide. 15 Nucleotide sequences encoding exemplary peptibodies are provided below in Table H.

The polynucleotide sequences encoding an exemplary peptibody of the present invention includes a nucleotide sequence encoding the Fc polypeptide sequence such as the following:

5’-GACAAAACTCACACATGTCCACCTTGCCCAGCACCTGAACTC

CTGGGGGGACCGTCAGTTTTCCTCTTCCCCCCAAAACCCAAGGACACCCTCA

20 TGATCTCCCGGACCCCTGAGGTCACATGCGTGGTGGTGGACGTGAGCCACG

AAGACCCTGAGGTCAAGTTCAACTGGTACGTGGACGGCGTGGAGGTGCATA ATGCCAAGACAAAGCCGCGGGAGGAGCAGTACAACAGCACGTACCGTGTG GTCAGCGTCCTCACCGTCCTGCACCAGGACTGGCTGAATGGCAAGGAGTAC AAGTGCAAGGTCTCCAACAAAGCCCTCCCAGCCCCCATCGAGAAAAC’CATC

25 TCCAAAGCCAAAGGGCAGCCCCGAGAACCACAGGTGTACACCCTGCCCCCA

TCCCGGGATGAGCTGACCAAGAACCAGGTCAGCCTGACCTGCCTGGTCAAA GGCTTCTATCCCAGCGACATCGCCGTGGAGTGGGAGAGCAATGGGCAGCCG GAGAACAACTACAAGACCACGCCTCCCGTGCTGGACTCCGACGGCTCCTTCT TCCTCTACAGCAAGCTCACCGTGGACAAGAGCAGGTGGCAGCAGGGGAACG

30 TCTTCTCATGCTCCGTGATGCATGAGGCTCTGCACAACCACTACACGCAGAA GAGCCTCTCCCTGTCTCCGGGTAAA-3’ (Seq ID No: 301)

In addition, the polynucleotides encoding the ggggg linker such as the following are included: 35 5’-GGTGGAGGTGGTGGT-3’ (Seq ID No: 302)

The polynucleotide encoding the peptibody also includes the codon encoding the methionine ATG and a stop codon such as TAA. 63 WO 2004/058988 PCT/US2003/040781

Therefore, the structure of the first peptibody in Table Π is TN8-Conl with a C configuration and a (gly)5 linker is as follows: M-Fc-GGGGG-EDKCKMWHWMCKPP (Seq H) No: 303).

Exemplary polynucleotides encoding this peptibody would be:

5’- ATGGACAAAACTCACACATGTCCACCTTGCCCAGCACCTGAA 5 CTCCTGGGGGGACCGTCAGTTTTCCTCTTCCCCCCAAAACCCAAGGACACCC

TCATGATCTCCCGGACCCCTGAGGTCACATGCGTGGTGGTGGACGTGAGCC acgaagaccctgaggtcaagttcaactggtacgtggacggcgtggaggtgc

ATAATGCCAAGACAAAGCCGCGGGAGGAGCAGTACAACAGCACGTACCGT

GTGGTCAGCGTCCTCACCGTCCTGCACCAGGACTGGCTGAATGGCAAGGAG

10 TACAAGTGCAAGGTCTCCAACAAAGCCCTCCCAGCCCCCATCGAGAAAACC

ATCTCCAAAGCCAAAGGGCAGCCCCGAGAACCACAGGTGTACACCCTGCCC

CCATCCCGGGATGAGCTGACCAAGAACCAGGTCAGCCTGACCTGCCTGGTC

AAAGGCTTCTATCCCAGCGACATCGCCGTGGAGTGGGAGAGCAATGGGCAG

CCGGAGAACAACTACAAGACCACGCCTCCCGTGCTGGACTCCGACGGCTCC 15 TTCTTCCTCTACAGCAAGCTCACCGTGGACAAGAGCAGGTGGCAGCAGGGG

AACGTCTTCTCATGCTCCGTGATGCATGAGGCTCTGCACAACCACTACACGC

AGAAGAGCCTCTCCCTGTCTCCGGGTAAAGGTGGAGGTGGTGGTAAGACAA ATGCAAAATGTGGCACTGGATGTGCAAACCGCCG-3’ (Seq ID No: 304) 20 TABLE Π

Peptibody Name Peptide Nucleotide Sequence (Seq ID No) Myostatin-TN8- conl KDKCKMWHWMCKPP AAAGACAAATGCAAAATGTGGCACTG GATGTGCAAACCGCCG (Seq. ID No: 147) 5 giy c Myostatin-TN8- con2 KDLCAMWHWMCKPP AAAGACCTGTGCGCTATGTGGCACTG GATGTGCAAACCGCCG (Seq. ID No: 148) 5 giy c Myostatin-TN8- con3 KDLCKMWKWMCKPP AAAGACCTGTGCAAAATGTGGAAATG GATGTGCAAACCGCCG (Seq ID No: 149) 5 giy c Myostatin-TN8- con4 KDLCKMWHWMCKPK AAAGACCTGTGCAAAATGTGGCACTG GATGTGCAAACCGAAA (Seq ID No: 150) 5 giy c Myostatin-TN8- con5 WYPCYEFHFWCYDL' TGGTACCCGTGCTACGAATTCCACTTC TGGTGCTACGACCTG (Seq ID No: 151) 5 giy c Myostatin-TN8- con5 WYPCYEFHFWCYDL TGGTACCCGTGCTACGAATTCCACTTC TGGTGCTACGACCTG (Seq ID No: 152) 5 giy N Myostatin-TN8- con6 WYPCYEGHFWCYDL TGGTACCCGTGCTACGAAGGTCACTT CTGGTGCTACGACCTG (Seq ID No: 153) 5 giy C Myostatin-TN8- con6 WYPCYEGHFWCYDL TGGTACCCGTGCTACGAAGGTCACTT CTGGTGCTACGACCTG (Seq ID No: 154) 5 giy N Myostatin-TN8- con7 IFGCKWWDVQCYQF ATCTTCGGTTGCAAATGGTGGGACGT TCAGTGCTACCAGTTC (Seq ID No: 155) 5 giy C hftyostatin-TNS- con8 IFGCKWWDVDCYQF ATCTTCGGTTGCAAATGGTGGGACGT TGACTGCTACCAGTTC (Seq ID No: 156) 5 giy C Myostatin-TN8- con8 IFGCKWWDVDCYQF ATCTTCGGTTGCAAATGGTGGGACGT TGACTGCTACCAGTTC (Seq ID No: 157) 5 giy N Myostatin-TN8- ADWCVSPNWFCMVM GCTGACTGGTGCGTTTCCCCGAACTG 5 giy C 64 WO 2004/058988 PCT/US2003/040781 con9 GTTCTGCATGGTTATG (Seq ID No: 158) Myostatin-TNS- conlO HKFCPWWALFCWDF CACAAATTCTGCCCGTGGTGGGCTCT GTTCTGCTGGGACTTC (Seq ID No: 159) 5 giy c Myostatm-TN8-1 KDLCKMWHWMCKPP AAAGACCTGTGCAAAATGTGGCACTG GATGTGCAAACCGCCG (Seq ID No: 160 5 giy c Myostatin-TN8-2 IDKCAIWGWMCPPL ATCGACAAATGCGCTATCTGGGGTTG GATGTGCCCGCCGCTG (Seq ID No: 161) 5giy c Myostatin-TN8-3 WYPCGEFGMWCLNV TGGTACCCGTGCGGTGAATTCGGTAT GTGGTGCCTGAACGTT (Seq ID No: 162) 5 giy c Myostatin-TN8-4 WFTCLWNCDNE TGGTTCACCTGCCTGTGGAACTGCGA CAACGAA (Seq ID No: 163) 5 giy c Myostatin-TN8-5 HTPCPWFAPLCVEW CACACCCCGTGCCCGTGGTTCGCTCC GCTGTGCGTTGAATGG (Seq ID No: 164) 5 giy c Myostatin-TN8-6 KEWCWRWKWMCKPE AAAGAATGGTGCTGGCGTTGGAAATG GATGTGCAAACCGGAA (Seq ID No: 165) s giy c Myostatin-TN8-7 FETCPSWAYFCLDI TTCGAAACCTGCCCGTCCTGGGCTTA CTTCTGCCTGGACATC (Seq ID No: 166) 5 giy c Myostatin-TN8-7 FETCPSWAYFCLDI TTCGAAACCTGCCCGTCCTGGGCTTA CTTCTGCCTGGACATC (Seq ID No: 167) 5 giy N Myostatin-TN8-8 AYKCEANDWGCWWL GCTTACAAATGCGAAGCTAACGACTG GGGTTGCTGGTGGCTG (Seq ID No: 168) 5 giy C Myostatin-TN8-9 NSWCEDQWHRCWWL AACTCCTGGTGCGAAGACCAGTGGCA CCGTTGCTGGTGGCTG (Seq ID No: 169) 5 giy c Myostatin-TN8-10 WSACYAGHFWCYDL TOGTCCGCTTGCTACGCTGGTCACTTC TGGTGCTACGACCTG (Seq ID No: 170) 5 giy c Myostatin-TN8-11 ANWCVSPNWFCMVM GCTAACTGGTGCGTTTCCCCGAACTG GTTCTGCATGGTTATG (Seq ID No: 171) 5 giy c Myostatin-TN8-12 WTECYQQEFWCWNL TGGACCGAATGCTACCAGCAGGAATT CTGGTGCTGGAACCTG (Seq ID No: 172) 5 giy c Myostatin-TN8-13 ENTCERWKWMCPPK GAAAACACCTGCGAACGTTGGAAATG GATGTGCCCGCCGAAA (Seq ID No: 173) 5 giy c Myostatin-TN8-14 WLPCHQEGFWCMNF TGGCTGCCGTGCCACCAGGAAGGTTT CTGGTGCATGAACTTC (Seq ID No: 174) 5 giy c Myostatin-TN8-15 STMCSQWHWMCNPF TCCACCATGTGCTCCCAGTGGCACTG GATGTGCAACCCGTTC (Seq ID No: 175) 5 giy c Myostatin-TN8-16 EFGCHWWDVDCYQF ATCTTCGGTTGCCACTGGTGGGACGT TGACTGCTACCAGTTC (Seq ID No: 176) . 5 giy c Myostatin-TN8-17 IYGCKV7WDIQCYDI ATCTACGGTTGCAAATGGTGGGACAT CCAGTGCTACGACATC (Seq ID No: 177) 5 giy c Myostatin-TN8-18 PD'WCIDPDWWCKFW CCGGACTGGTGCATCGATCCGGACTG GTGGTGCAAATTCTGG (Seq ID No: 178) s giy c Myostatin-TN8-19 QGHCTRWPWMCPPY CAGGGTCACTGCACCCGTTGGCCGTG GATGTGCCCGCCGTAC (Seq ID No: 179) 5 giy c Myostatin-TN8-20 WQECYREGFWCLQT tggcaggaatgctaccgtgaaggttt CTGGTGCCTGCAGACC (Seq ID No: 180) 5 giy c 65 WO 2004/058988 PCT/US2003/040781

Myostatin-TN8-21 WEDCYGPGFKCWSP TGGTTCGACTGCTACGGTCCGGGTTTC AAATGCTGGTCCCCG (Seq ED No: 181) 5 giy c Myostatin-TN8-22 GVRCPKGHLWCLYP GGTGTTCGTTGCCCGAAAGGTCACCT GTGGTGCCTGTACCCG (Seq ID No: 182) 5 giy c Myostatin-TN8-23 HWACGYWPWSCKWV CACTGGGCTTGCGGTTACTGGCCGTG GTCCTGCAAATGGGTT (Seq ID No: 183) 5 giy c Myostatin-TN8-24 GPACHSPWWWCVFG GGTCCGGCTTGCCACTCCCCGTGGTG GTGGTGCGTTTTCGGT (Seq ID No: 184) 5 giy c Myostatin-TN8-25 TIWCISPMWFCSQQ ACCACCTGGTGCATCTCCCCGATGTG GTTCTGCTCCCAGCAG (Seq ED No: 185) 5 giy c Myostatin-TN8-26 HKFCPPWAIFCWDF CACAAATTCTGCCCGCCGTGGGCTAT CTTCTGCTGGGACTTC (Seq ID No: 186) 5 giy N Myostatin-TN8-27 PDWCVSPRWYCNMW CCGGACTGGTGCGTTTCCCCGCGTTG GTACTGCAACATGTGG (Seq ID No: 187) 5 giy N Myostatin-TN8-28 VWKCHWFGMDCEPT GTTTGGAAATGCCACTGGTTCGGTAT GGACTGCGAACCGACC (Seq ID No: 188) 5 giy N Myostatin-TN8-29 KKHCQIWTWMCAPK AAAAAACACTGCCAGATCTGGACCTG GATGTGCGCTCCGAAA (Seq ID No: 189) 5 giy N Myostatin-TN8-30 WFQCGSTLFWCYNL TGGTTCCAGTGCGGTTCCACCCTGTTC TGGTGCTACAACCTG (Seq ED No: 190) 5 giy N My ostatin-TN8-31 WSPCYDHYFYCYTI ’ TGGTCCCCGTGCTACGACCACTACTTC TACTGCTACACCATC (Seq ID No: 191) 5 giy N Myostatin-TN8-32 SWMCGFFKEVCMWV TCCTGGATGTGCGGTTTCTTCAAAGA AGTTTGCATGTGGGTT (Seq ED No: 192) 5 giy N Myostatin-TN8-33 EMLCMIHPVFCNPH gaaatgctgtgcatgatccacccggt TTTCTGCAACCCGCAC (Seq ID No: 193) 5 giy N Myostatin-TN8-34 LKTCNLWPWMCPPL CTGAAAACCTGCAACCTGTGGCCGTG GATGTGCCCGCCGCTG (Seq ID No: 194) 5 giy N Myostatin-TN8-35 WGCKWYEAWCYNK gttgttggttgcaaatggtacgaagc TTGGTGCTACAACAAA (Seq ID No: 195) 5 giy N Myostatin-TN8-36 PIHCTQWAWMCPPT CCGATCCACTGCACCCAGTGGGCTTG GATGTGCCCGCCGACC (Seq ID No: 196) 5 giy N Myostatin-TN8-37 DSNCPWYFLSCVIF GACTCCAACTGCCCGTGGTACTTCCT GTCCTGCGTTATCTTC (Seq ED No: 197) 5 giy N My o statin-TNS-3 8 HIWCNLAMMKCVEM CACATCTGGTGCAACCTGGCTATGAT' GAAATGCGTTGAAATG (Seq ID No: 198) 5 giy N Myostatin-TN8-39 NLQCIYFLGKCIYF AACCTGCAGTGCATCTACTTCCTGGG TAAATGCATCTACTTC (Seq ID No: 199) 5 giy N Myostatin-TN8-40 AWRCMWFSDVCTPG GCTTGGCGTTGCATGTGGTTCTCCGAC GTTTGCACCCCGGGT (Seq ID No: 200) 5 giy N Myostatin-TN8-41 WFRCFLDADWCTSV TGGTTTCGTTGTTTTCTTGATGCTGAT TGGTGTACTTCTGTT (Seq ID No: 201) 5 giy N Myostatin-TN8-42 EKICQMWSWMCAPP GAAAAAATTTGTCAAATGTGGTCTTG GATGTGTGCTCCACCA (Seq ID No: 202) 5 giy N Myostatin-TN8-43 WFYCHLNKSECTEP TGGTTTTATTGTCATCTTAATAAATCT 5 giy N 66 WO 2004/058988 PCT/US2003/040781

* GAATGTACTGAACCA (Seq ED No: 2031 Myostatin-TN8~44 FWCAIGIDKCKRV TTTTGGCGTTGTGCTATTGGTATTGAT AAATGTAAACGTGTT (Seq ID No: 2041 5 giy N Myostatin-TN8-45 NLGCKWYEVWCFTY AATCTTGGTTGTAAATGGTATGAAGT TTGGTGTTTTACTTAT (Seq ID No: 2051 5 giy N My ostatin-TN8-46 IDLCNMWDGMCYPP ATTGATCTTTGTAATATGTGGGATGGT ATOTGTTATCCACCA (Seq ED No: 2061 5 giy N Myostatin-TN8-47 EMPCNIWGWMCPPV GAAATGCCATGTAATATTTGGGGTTG GATGTGTCCACCAGTT (Seq ID No: 207) 5 giy N Myostatin-TN12-1 WFRCVLTGIVDWSECF GL TGGTTCCGTTGCGTTCTGACCGGTATC GTTGACTGGTCCGAATGCTTCGGTCT G (Seq ID No: 208) 5 giy N Myostatin-TN12-2 GFSCTPGLDEFYVDCSP F GGTTTCTCCTGCACCTTCGGTCTGGAC GAATTCTACGTTGACTGCTCCCCGTTC (Seq ID No: 209) 5 giy N Myostatin-TN12-3 LPWCHDQVNADWGFC MLW CTGCCGT.GGTGCCACGACCAGGTTAA CGCTGACTGGGGTTTCTGCATGCTGT GG (Seq ED No: 210) 5 giy N Myostatin-TN12-4 YPTCSEKFWIYGQTCV LW TACCCGACCTGCTCCGAAAAATTCTG GATCTACGGTCAGACGTGCGTTCTGT GG (Seq ID No: 211) 5 giy N Myostatin-TN12-5 LGPCPEHHGPWPQYCV YW CTGGGTCCGTGCCCGATCCACCACGG TCCGTGGCCGCAGTACTGCGTTTACT GG (Seq ID No: 212) 5 giy N Myostatin-TN12-6 PFPCETHQISWLGHCLS F CCGTTCCCGTGCGAAACCCACCAGAT CTCCTGGCTGGGTCACTGCCTGTCCTT C (Seq ED No: 213) 5 giy N Myostatin-TN12-7 HWGCEDLMWSWHPLC KRP CACTGGGGTTGCGAAGACCTGATGTG GTCCTGGCACCCGCTGTGCCGTCGTC CG (Seq ED No: 214) 5 giy N Myostatin-TN12-8 LPLCDADMMPTIGFCV AY CTGCCGCTGTGCGACGCTGACATGAT GCCGACCATCGGTTTCTGCGTTGCTTA C (Seq ED No: 215) 5 giy N Myostatin-TN12-9 SHWCETTFWMNYAKC VHA TCCCACTGGTGCGAAACCACCTTCTG GATGAACTACGCTAAATGCGTTCACG CT (Seq ID No: 216) 5 giy N Myostatin-TN12- 10 LPKCTHVPFDQGGFCL WY CTGCCGAAATGCACCCACGTTCCGTT CGACCAGGGTGGTTTCTGCCTGTGGT AC (Seq ED No: 217) 5 giy N Myostatin-TN12- 11 FSSCWSPVSRQDMFCV FY TTCTCCTCCTGCTGGTCCCCGGTTTCC CGTCAGGACATGTTCTGCGTTTTCTAC (Seq Π) No: 218) 5 giy N Myostatin-TN12- 13 SHKCEYSGWLQPLCYR P TCCCACAAATGCGAATACTCCGGTTG GCTGCAGCCGCTGTGCTACCGTCCG (Seq ID No: 219) 5 giy N Myostatin-TN12- 14 PWWCQDNYVQHMLH CDSP CCGTGGTGGTGCCAGGACAACTACGT TCAGCACATGCTGCACTGCGACTCCC CG (Seq ID No: 220) 5 giy N Myostatin-TN12- 15 WERCMLMNSFDAFQC VSY TGGTTCCGTTGCATGCTGATGAACTCC TTCGACGCTTTCCAGTGCGTTTCCTAC (Seq ID No: 221) 5 giy N Myostatin-TN12- 16 PDACRDQPWYMFMGC MLG CCGGACGCTTGCCGTGACCAGCCGTG GTACATGTTCATGGGTTGCATGCTGG GT (Seq ID No: 222) 5 giy N Myostatin-TN12- FLACFVEFELCFDS TTCCTGGCTTGCTTCGTTGAATTCGAA 5 giy N 67 WO 2004/058988 PCT/US2003/040781

17 CTGTGCTTCGACTCC (Seq ID No: 223) l My ostatin-TN 12-18 SAYCIITESDPYVLCVP L TCCGCTTACTGCATCATCACCGAATCC GACCCGTACGTTCTGTGCGTTCCGCTG (Seq ID No: 224) 5 gly N Myostatin-TN12- 19 PSICESYSTMWLPMCQ HN CCGTCCATCTGCGAATCCTACTCCACC ATGTGGCTGCCGATGTGCCAGCACAA C (Seq ID No: 225) 5 gly N Myostatin-TN12- 20 WLDCHDDSWAWTKM CRSH TGGCTGGACTGCCACGACGACTCCTG GGCTTGGACCAAAATGTGCCGTTCCC AC (Seq ID No: 226) 5 gly N Myostatin-TN12- 21 YLNCVMMNTSPFVEC VPN TACCTGAACTGCGTTATGATGAACAC CTCCCCGTTCGTTGAATGCGTTTTCAA C (Seq ID No: 227) 5 gly N Myostatin-TN12- 22 YPWCDGPMIQQGITCM FY TACCCGTGGTGCGACGGTTTCATGAT CCAGCAGGGTATCACCTGCATGTTCT AC (Seq ID No: 228) 5 gly N Myostatin-TN 12-23 EDYCTWLNGFKDWKC WSR TTCGACTACTGCACCTGGCTGAACGG TTTCAAAGACTGGAAATGCTGGTCCC GT (Seq ID No: 229) 5 gly N Myostatin-TN12- 24 LPLCNLKEISHVQACVL F CTGCCGCTGTGCAACCTGAAAGAAAT CTCCCACGTTCAGGCTTGCGTTCTGTT C (Seq ID No: 230) 5 gly N My ostatin-TN 12-25 SPECAFARWLG3EQCQ RD TCCCCGGAATGCGCTTTCGCTCGTTGG CTGGGTATCGAACAGTGCCAGCGTGA C (Seq ID No: 231) 5 gly N Myostatin-TN12- 26 YPQCFNLHLLEWTECD WF TACCCGCAGTGCTTCAACCTGCACCT GCTGGAATGGACCGAATGCGACTGGT TC (Seq ID No: 232) 5 gly N Myostatin-TN12- 27 RWRCEIYDSEFLPKCW FF CGTTGGCGTTGCGAAATCTACGACTC CGAATTCCTGCCGAAATGCTGGTTCTT C (Seq ID No: 233) 5 gly N Myostatin-TN12-28 . LVGCDNVWHRCKLF CTGGTTGGTTGCGACAACGTTTGGCA CCGTTGCAAACTGTTC (Seq ID No: 234) 5 gly N Myostatin-TN12- 29 AGWCHVWGEMFGMG CSAL GCTGGTTGGTGCCACGTTTGGGGTGA AATGTTCGGTATGGGTTGCTCCGCTCT G (Seq ID No; 235) 5 gly N My ostatin-TN12-30 HHECEWMARWMSLD CVGL CACCACGAATGCGAATGGATGGCTCG TTGGATGTCCCTGGACTGCGTTGGTCT G (Seq ID No: 236) 5 gly N Myostatin-TN12- 31 FPMCGIAGMKDFDFCV WY TTCCCGATGTGCGGTATCGCTGGTAT GAAAGACTTCGACTTCTGCGTTTGGT AC (Seq ID No: 237) 5 gly N Myostatin-TN12- 32 RDDCTFWPEWLWKLC ERP CGTGATGATTGTACTTTTTGGCCAGAA TGGCTTTGGAAACTTTGTGAACGTCC A (Seq ID No: 238) 5 gly N Myostatin-TN 12-33 YNFCSYLFGVSKEACQ LP TATAATTTTTGTTCTTATCTTTTTGGTG TTTCTAAAGAAGCTTGTCAACTTCCA (Seq ID No: 239) 5 gly N Myostatin-TN 12-34 AHWCEQGPWRYGNIC MAY GCTCATTGGTGTGAACAAGGTCCATG GCGTTATGGTAATATTTGTATGGCTTA T (Seq ID No: 240) 5 gly N C Myostatin-TN12- 35 NLVCGKISAWGDEACA RA AATCTTGTTTGTGGTAAAATTTCTGCT TGGGGTGATGAAGCTTGTGCTCGTGC T (Seq ID No: 241) 5 gly N Myostatin-TN12- HNVCTIMGPSMKWFC CATAATGTTTGTACTATTATGGGTCCA 5 gly N 68 WO 2004/038988 PCT/US2003/040781

36 WND TCTATGAAATGGTTTTGTTGGAATGAT (Seq ED No: 242) C Myostatin-TN12- 37 NDLGAMWGWRNTIWC QNS AATGATCTTTGTGCTATGTGGGGTTGG CGTAATACTATTTGGTGTCAAAATTCT (Seq ID No: 243) 5 giy N C Myostatin-TN12- 38 PPFCQNDNDMLQSLCK LL CCACCATTTTGTCAAAATGATAATGA TATGCTTCAATCTCTTTGTAAACTTCT T (Seq ID No: 244) 5 giy N My ostatin-TN 12-39 WYDCNVPNELLSGLCR LF TGGTATGATTGTAATGTTCCAAATGA ACTTCTTTCTGGTCTTTGTCGTCTTTTT (Seq ID No: 245) 5 giy N Myostatin-TN12- 40 YGDCDQNHWMWPFTC LSL TATGGTGATTGTGATCAAAATCATTG GATGTGGCCATTTACTTGTCTTTCTCT T (Seq ID No: 246) 5 giy N C Myostatin-TN12- 41 GWMCHFDLHDWGAT- CQPD GGTTGGATGTGTCATTTTGATCTTCAT GATTGGGGTGCTACTTGTCAACCAGA T (Seq ED No: 247.) 5 giy N Myostatin-TN12- 42 YFHCMFGGHEFEVHCE SF TATTTTCATTGTATGTTTGGTGGTCAT GAATTTGAAGTTCATTGTGAATCTTTT (Seq ID No: 248) 5 giy N C Myostatin-TN12- 43 AYWCWHGQCVRF GCTTATTGGTGTTGGCATGGTCAATGT GTTCGTTTT (Seq ED No: 249) 5 giy N Myostatin-Linear- 1 SEHWTFTDWDGNEW WVRPF TCCGAACACTGGACCTTCACCGACTG GGACGGTAACGAATGGTGGGTTCGTC CGTTC (Seq ID No: 250) s giy N My ostatin-Linear-2 MEMLDSLFELLKDMVP ISKA ATGGAAATGCTGGACTCCCTGTTCGA ACTGCTGAAAGACATGGTTCCGATCT CCAAAGCT (Seq ID No: 251) 5 giy N Myostatin-Linear- 3 SPPEEALMEWLGWQY GKFT TCCCCGCCGGAAGAAGCTCTGATGGA ATGGCTGGGTTGGCAGTACGGTAAAT TCACC (Seq ED No: 252) 5 giy N Myostatin-Linear- 4 SPENLLNDLYILMTKQ EWYG TCCCCGGAAAACCTGCTGAACGACCT OTACATCCTGATGACCAAACAGGAAT GGTACGGT (Seq ED No: 253) 5 giy N Myostatin-Linear- 5 FHWEEGIPFHVVTPYS YDRM TTCCACTGGGAAGAAGGTATCCCGTT CCACGTTGTTACCCCGTACTCCTACGA CCGTATG (Seq ID No: 254) 5 giy N Myostatin-Linear- 6 KRLLEQFMNDLAELVS GHS AAACGTCTGCTGGAACAGTTCATGAA CGACCTGGCTGAACTGGTTTCCGGTC ACTCC (Seq ED No: 255) 5 giy N Myostatin-Linear- 7 DTRDALFQEFYEFVRS RLVI GACACCCGTGACGCTCTGTTCCAGGA ATTCTACGAATTCGTTCGTTCCCGTCT GGTTATC (Seq ED No: 256) 5 giy N Myostatin-Linear- 8 RMSAAPRPLTYRDIMD QYWH CGTATGTCCGCTGCTCCGCGTCCGCTG ACCTACCGTGACATCATGGACCAGTA CTGGCAC (Seq ED No: 257) 5 giy N Myostatin-Linear- 9 NDKAHFFEMFMFDVH NFVES AACGACAAAGCTCACTTCTTCGAAAT GTTCATGTTCGACGTTCACAACTTCGT TGAATCC (Seq Id No: 258) 5 giy N My ostatin-Linear-10 QTQAQKEDGLWELLQS 3RNQ CAGACCCAGGCTCAGAAAATCGACGG TCTGTGGGAACTGCTGCAGTCCATCC GTAACCAG (Seq ED No: 259) 5 giy N Myostatin-Linear- 11 MLSEFEEFLGNLVHRQ EA ATGCTGTCCGAATTCGAAGAATTCCT GGGTAACCTGGTTCACCGTCAGGAAG CT (Seq ID No: 260) 5 giy N Myostatin-Linear- YTPKMGSEWTSFWHN TACACCCCGAAAATGGGTTCCGAATG 5 giy N 69 WO 2004/058988 PCT/US2003/040781

12 RIHYL GACCTCCTTCTGGCACAACCGTATCC ACTACCTG (Seq ID No: 261) Myostatin-Linear- 13 LNDTLLRELKMVLNSL SDMK CTGAACGACACCCTGCTGCGTGAACT GAAAATGGTTCTGAACTCCCTGTCCG ACATGAAA (Seq ID No: 262) 5 giy N Myostatm-Linear- 14 FDVERDLMRWLEGFM QSAAT TTCGACGTTGAACGTGACCTGATGCG TTGGCTGGAAGGTTTCATGCAGTCCG CTGCTACC (Seq ID No: 263) 5 giy N Myostatin-Linear- 15 HHGWNYLRKGSAPQW FEAWV CACCACGGTTGGAACTACCTGCGTAA AGGTTCCGCTCCGCAGTGGTTCGAAG CTTGGGTT (Seq ID No: 264) 5 giy N Myostatin-Linear- 16 VESLHQLQMWLDQKL ASGPH GTTGAATCCCTGCACCAGCTGCAGAT GTGGCTGGACCAGAAACTGGCTTCCG GTCCGCAC (Seq ID No: 265) 5 giy N Myostatin-Linear- 17 RATLLKDFWQLVEGY GDN CGTGCTACCCTGCTGAAAGACTTCTG GCAGCTGGTTGAAGGTTACGGTGACA AC (Seq ID No: 266) s giy N Myostatin-Linear- 18 EELLREFYRFVSAFDY GAAGAACTGCTGCGTGAATTCTACCG TTTCGTTTCCGCTTTCGACTAC (Seq ID No: 267) 5 giy- N Myostatin-Linear- 19 GLLDEFSHFIAEQFYQ MPGG GGTCTGCTGGACGAATTCTCCCACTTC ATCGCTGAACAGTTCTACCAGATGCC GGGTGGT (Seq ID No: 268) 5 giy N Myostatin-Linear- 20 YREMSMLEGLLDVLER LQHY TACCGTGAAATGTCCATGCTGGAAGG TCTGCTGGACGTTCTGGAACGTCTGC AGCACTAC (Seq ID No: 269) 5 giy N Myostatin-Linear- 21 HNSSQMLLSELIMLVG SMMQ CACAACTCCTCCCAGATGCTGCTGTC CGAACTGATCATGCTGGTTGGTTCCA TGATGCAG (Seq ID No: 270) 5 giy N Myostatin-Linear- 22 WREHFLNSDYIRDKLI AIDG TGGCGTGAACACTTCCTGAACTCCGA CTACATCCGTGACAAACTGATCGCTA TCGACGGT (Seq ID No: 271) 5 giy N My ostatin-Linear-23 QFPFYVEDDLPAQLEY WIA CAGTTCCCGTTCTACGTTTTCGACGAC CTGCCGGCTCAGCTGGAATACTGGAT CGCT (Seq ID No: 272) 5 giy· N Myostatin-Linear- 24 EFEHWLHNHRSEVNH WLDMN GAATTCTTCCACTGGCTGCACAACCA CCGTTCCGAAGTTAACCACTGGCTGG ACATGAAC (Seq ID No: 273) 5 giy N Myostatin-Linear- 25 EALFQNFFRDVLTLSER EY GAAGCTCTTTTTCAAAATTTTTTTCGT GATGTTCTTACTCTTTCTGAACGTGAA TAT (Seq ID No: 274) 5 giy N C Myostatin-Linear -26 QYWEQQWMTYFRENG LHVQY CAATATTGGGAACAACAATGGATGAC TTATTTTCGTGAAAATGGTCTTCATGT TCAATAT (Seq ID No: 275) 5 giy N Myostatin-Linear- 27 NQRMMLEDLWRIMTP MFGRS AATCAACGTATGATGCTTGAAGATCT TTGGCGTATTATGACTCCAATGTTTGG TCGTTCT (Seq ID No: 276) 5 giy N C Myostatin-Linear- 29 ELDELKAELSRHYALD DLDE TTTCTTGATGAACTTAAAGCTGAACTT TCTCGTCATTATGCTCTTGATGATCTT GATGAA (Seq ID No: 277) s giy N Myostatin-Linear- 30 GKLIEGLLNELMQLETF MPD GGTAAACTTATTGAAGGTCTTCTTAAT GAACTTATGCAACTTGAAACTTTTATG CCAGAT (Seq ID No: 278) 5 giy N C Myostatin-Linear- 31 ILLLDEYKKDWKSWF. ATTCTTCTTCTTGATGAATATAAAAAA GATTGGAAATCTTGGTTT (Seq ID No: 279) 5 giy N 70 WO 2004/058988 PCT/US2003/040781

Myostatin-2XTN8-19 kc QGHCTRWPWMCPPYG SGSATGGSGSTASSGSG SATGQGHCTRWPWMC PPY CAGGGCCACTGTACTCGCTGGCCGTG. GATGTGCCCGCCGTACGGTTCTGGTT CCGCTACCGGTGGTTCTGGTTCCACTG CTTCTTCTGGTTCCGGTTCTGCTACTG GTCAGGGTCACTGCACTCGTTGGCCA TGGATGTGTCCACCGTAT (Seg ID No: 280) lk N Myostatin- 2XTN8-CON6 WYPCYEGHFWCYDLG SGSTASSGSGSATGWY PCYEGHFWCYDL TGGTATCCGTGTTATGAGGGTCACTTC TGGTGCTACGATCTGGGTTCTGGTTCC ACTGCTTCTTCTGGTTCCGGTTCCGCT ACTGGTTGGTACCCGTGCTACGAAGG TCACTTTTGGTGTTATGATCTG (Seq ID No: 281) 5giy C Myostatin-2XTN8-5 kc HTPCPWFAPLCVEWGS GSATGGSGSTASSGSGS ATGHTPCPWFAPLCVE W CACACTCCGTGTCCGTGGTTTGCTCCG CTGTGCGTTGAATGGGGTTCTGGTTCC GCTACTGGTGGTTCCGGTTCCACTGCT TCTTCTGGTTCCGGTTCTGCAACTGGT CACACCCCGTGCCCGTGGTTTGCACC GCTGTGTGTAGAGTGG (SeqlDNo: 282) lk C Myostatin-2XTN8-18 kc PDWCIDPDWWCKFWG SGSATGGSGSTASSGSG SATGPDWCIDPDWWC KFW CCGGATTGGTGTATCGACCCGGACTG GTGGTGCAAATTCTGGGGTTCTGGTTC CGCTACCGGTGGTTCCGGTTCCACTG CTTCTTCTGGTTCCGGTTCTGCAACTG GTCCGGACTGGTGCATCGACCCGGAT TGGTGGTGTAAATTTTGG (SeqlDNo: 283) lk C Myostatin-2XTN8-11 kc ANWCVSPNWFCMVM GSGSATGGSGSTASSGS GSATGANWCVSPNWF CMVM CCGGATTGGTGTATCGACCCGGACTG GTGGTGCAAATTCTGGGGTTCTGGTTC CGCTACCGGTGGTTCCGGTTCCACTG CTTCTTCTGGTTCCGGTTCTGCAACTG GTCCGGACTGGTGCATCGACCCGGAT TGGTGGTGTAAATTTTGG (SeqlDNo; 284) lk C Myostatin-2XTN8-25 kc PDWCIDPDWWCKFWG SGSATGGSGSTASSGSG SATGPDWCIDFDWWC KFW ACCACTTGGTGCATCTCTCCGATGTG GTTCTGCTCTCAGCAGGGTTCTGGTTC CACTGCTTCTTCTGGTTCCGGTTCTGC AACTGGTACTACTTGGTGTATCTCTCC AATGTGGTTTTGTTCTCAGCAA (Seq ID No: 285) lk C Myostatin-2XTN8-23 kc HWACGYWPWSCKWV GSGSATGGSGSTASSGS GSATGHWACGYWPWS CKWV CACTGGGCATGTGGCTATTGGCCGTG GTCCTGCAAATGGGTTGGTTCTGGTTC CGCTACCGGTGGTTCCGGTTCCACTG CTTCTTCTGGTTCCGGTTCTGCAACTG GTCACTGGGCTTGCGGTTACTGGCCG TGGTCTTGTAAATGGGTT (Seq ID No: 286) lk C Myostatin-TN8-29-19 kc KKHCQIWTWMCAPKG SGSATGGSGSTASSGSG SATGQGHCTRWPWMC PPY AAAAAACACTGTCAGATCTGGACTTG GATGTGCGCTCCGAAAGGTTCTGGTT CCGCTACCGGTGGTTCTGGTTCCACTG CTTCTTCTGGTTCCGGTTCCGCTACTG GTCAGGGTCACTGCACTCGTTGGCCA TGGATGTGTCCGCCGTAT (Seq ED No: 287) lk c Myostatin-TN8-19-29 kc QGHCTRWPWMCPPYG SGSATGGSGSTASSGSG SATGKKHCQIWTWMC CAGGGTCACTGCACCCGTTGGCCGTG GATGTGCCCGCCGTACGGTTCTGGTT CCGCTACCGGTGGTTCTGGTTCCACTG lk c 71 WO 2004/058988 PCT/US2003/040781

APK CTTCTTCTGGTTCCGGTTCTGCTACTG GTAAAAAACACTGCCAGATCTGGACT TGGATGTGCGCTCCGAAA (Seq ID No: 288) Myostatin-TN8-29-19 kn KKHCQIWTWMCAPKG SGSATGGSGSTASSGSG SATGQGHCTRWPWMC PPY AAAAAACACTGTCAGATCTGGACTTG GATGTGCGCTCCGAAAGGTTCTGGTT CCGCTACCGGTGGTTCTGGTTCCACTG CTTCTTCTGGTTCCGGTTCCGCTACTG GTCAGGGTCACTGCACTCGTTGGCCA TGGATGTGTCCGCCGTAT (Seq ID No: 289) lk N Myostatin-TNS- 29-19-8g KKHCQIWTWMCAPKG GGGGGGGQGHCTRWP WMCPPY AAAAAACACTGCCAGATCTGGACTTG GATGTGCGCTCCGAAAGGTGGTGGTG GTGGTGGCGGTGGCCAGGGTCACTGC ACCCGTTGGCCGTGGATGTGTCCGCC GTAT (Seq ID No: 290) 8 gly C Myostatin-TN8- 19-29-6gc QGHCTRWPWMCPPYG GGGGGKKHCQIWTWM CAPK CAGGGTCACTGCACCCGTTGGCCGTG GATGTGCCCGCCGTACGGTGGTGGTG GTGGTGGCAAAAAACACTGCCAGATC TGGACTTGGATGTGCGCTCCGAAA (Seq ID No: 291) 6gly C

Example 3 In vitro Assays 5 C2C12 Cell Based Mvostatin Activity Assay

This assay demonstrates the myostatin neutralizing capability of the inhibitor being tested by measuring the extent that binding of myostatin to its receptor is inhibited. A myostatin-responsive reporter cell line was generated by transfection of C2C12 myoblast cells (ATCC No: CRL-1772) with a pMARE-luc construct. The pMARE-luc construct 10 was made by cloning twelve repeats of the CAGA sequence, representing the myostatin/activin response elements (Dennler et al. EMBO 17: 3091-3100 (1998)) into a pLuc-MCS reporter vector (Stratagene cat # 219087) upstream of the TATA box. The myoblast C2C12 cells naturally express myostatin/activin receptors on its cell surface. When myostatin binds the cell receptors, the Smad pathway is activated, and phosphorylated Smad binds to the response element (Macias- 15 Silva et al. Cell 87:1215 (1996)), resulting in the expression of the lucerase gene. Luciferase activity is then measured using a commercial luciferase reporter assay kit (cat # E4550, Promega, Madison, WI) according to manufacturer’s protocol. A stable line of C2C12 cells that had been transfected with pMARE-luc (C2C12/pMARE clone #44) was used to measure myostatin activity according to the following procedure. 20 Equal numbers of the reporter cells (C2C12/pMARE clone #44) were plated into 96 well cultures. A first round screening using two dilutions of peptibodies was performed with the myostatin concentration fixed at 4 nM. Recombinant mature myostatin was pre-incubated for 2 72 WO 2004/058988 PCT/US2003/040781 hours at room temperature with peptibodies at 40 nM and 400 nM respectively. The reporter cell culture was treated with the myostatin with or without peptibodies for six hours. Myostatin activity was measured by determining the luciferase activity in the treated cultures. This assay was used to initially identify peptibody hits that inhibited the myostatin signaling activity in the 5 reporter assay. Subsequently, a nine point titration curve was generated with the myostatin concentration fixed at 4 nM. The myostatin was preincubated with each of the following nine concentrations of peptibodies: 0.04 mM, 0.4 nM, 4 nM, 20 nM, 40 nM, 200 nM, 400 nM, 2 uM and 4 uM for two hours before adding the mixture to the reporter cell culture. The IC3o values were for a number of examplary peptibodies are provided in Tables HI and for affinity matured 10 peptibodies, in Table VHI. BIAcore® assay

An affinity analysis of each candidate myostatin peptibody was performed on a BIAcore®3000 (Biacore, Inc., Piscataway, NJ), apparatus using sensor chip CM5, and 0.005 15 percent P20 surfactant (Biacore, Inc.) as running buffer. Recombinant mature myostatin protein was immobilized to a research grade CM5 sensor chip (Biacore, Inc.) via primary amine groups using the Amine Coupling Kit (Biacore, Inc.) according to the manufacturer’s suggested protocol.

Direct binding assays were used to screen and rank the peptibodies in order of their ability to bind to immobilized myostatin. Binding assays were carried by injection of two 20 concentrations (40 and 400 nM) of each candidate myostatin-binding peptibody to the immobilized myostatin surface at a flow rate of 50 μΐ/min for 3 minutes. After a dissociation time of 3 minutes, the surface was regenerated. Binding curves were compared qualitatively for binding signal intensity, as well as for dissociation rates. Peptibody binding kinetic parameters including k„ (association rate constant), kd (dissociation rate constant) and KD (dissociation 25 equilibrium constant) were determined using the BIA evaluation 3.1 computer program (Biacore, Inc.). The lower the dissociation equilibrium constants (expressed in nM), the greater the affinity of the peptibody for myostatin. Examples of peptibody KD values are shown in Table HI and Table VI for affinity-matured peptibodies below. 30 Blocking assay on ActRHB/Fc surface

Blocking assays were carried out using immobilized ActRHB/Fc (R&D Systems,

Minneapolis, MN) and myostatin in the presence and absence of peptibodies with the BIAcore® assay system. Assays were used to classify peptibodies as non-neutralizing (those which did not prevent myostatin binding to ActRHB/Fc) or neutralizing (those that prevented myostatin binding 73 WO 2004/058988 PCT/US2003/040781 to ActRBB/Fc). Baseline myostatin-ActRUB/Fc binding was first determined in the absence of any peptibody.

For early screening studies, peptibodies were diluted to 4 nM, 40 nM, and 400 nM in sample buffer and incubated with 4 nM myostatin (also diluted in sample buffer). The 5 peptibody: ligand mixtures were allowed to reach equilibrium at room temperature (at least 5 hours) and then were injected over the immobilized ActRUB/Fc surface for 20 to 30 minutes at a flow rate of 10 uL/min. An increased binding response (over control binding with no peptibody) indicated that peptibody binding to myostatin was non-neutralizing. A decreased binding response (compared to the control) indicated that peptibody binding to myostatin blocked the 10 binding of myostatin to ActRUB/Fc. Selected peptibodies were further characterized using the blocking assay of a full concentration series in order to derive IC50 values (for neutralizing peptibodies) or EC50 (for non-neutralizing peptibodies). The peptibody samples were serially diluted from 200 nM to 0.05 mM in sample buffer and incubated with 4 mM myostatin at room temperature to reach equilibrium (minimum of five hours) before injected over the immobilized 15 ActRUB/Fc surface for 20 to 30 minutes at a flow rate of 10 uL/min. Following the sample injection, bound ligand was allowed to dissociate from the receptor for 3 minutes. Plotting the binding signal vrs. peptibody concentration, the IC50 values for each peptibody in the presence of 4 nM myostatin were calculated. It was found, for example, that the peptibodies TN8-19, L2 and L17 inhibit myostatin activity in cell-based assay, but binding of TN-8-19 does not block 20 myostatin/ActRUB/Fc interactions, indicating that TN-8-19 binds to a different epitope than that observed for the other two peptibodies.

Epitope binning for peptibodies A purified peptibody was immobilized on a BIAcore chip to capture myostatin before 25 injection of a second peptibody, and the amount of secondary peptibody bound to the captured myostatin was determined. Only peptibodies with distinct epitopes will bind to the captured myostatin, thus enabling the binning of peptibodies with similar or distinct epitope binding properties. For example, it was shown that peptibodies TN8-19 and L23 bind to different epitopes on myostatin. 30

Selectivity Assays

These assays were performed using BIAcore® technology, to determine the selectivity of binding of the peptibodies to other TGFB family members. ActRUB/Fc, TGFBRU/Fc and BMPR-1 A/Fc (all obtained from R & D Systems, Minneapolis, MN) were covalently coupled to research 35 grade sensor chips according to manufacturer’s suggested protocol. Because BIAcore assays 74 WO 2004/058988 PCT/US2003/040781 detects changes in the refractive index, the difference between the response detected with injection over the immobilized receptor surfaces compared with the response detected with injection over the control surface in the absence of any peptibody represents the actual binding of Activin A, TGFpl, TGFp3, and BMP4 to the receptors, respectively. With pre-incubation of 5 peptibodies and TGFP molecules, a change (increase or decrease) in binding response indicates peptibody binding to the TGFP family of molecules. The peptibodies of the present invention all bind to myostatin but not to Activin A, TGFpi, TGFP3, or BMP4.

KiriEx A™ Equilibrium Assays 10 Solution-based equilibrium-binding assays using KinExA™ technology (Sapidyne

Instruments, Inc.) were used to determine the dissociation equilibrium (KD) of myostatin binding to peptibody molecules. This solution-based assay is considered to be more sensitive than the BIAcore assay in some instances. Reacti-Gel™ 6X was pre-coated with about 50 ug/ml myostatin for over-night, and then blocked with BSA. 30pM and lOOpM of peptibody samples 15 were incubated with various concentrations (0.5 pM to 5 nM) of myostatin in sample buffer at room temperature for 8 hours before being run through the myostatin-coated beads. The amount of the bead-bound peptibody was quantified by fluorescent (Cy5) labeled goat anti-human-Fc antibody at 1 mg/ml in superblock. The binding signal is proportional to the concentration of free peptibody at equilibrium with a given myostatin concentration. Kd was obtained from the 20 nonlinear regression of the competition curves using a dual-curve one-site homogeneous binding model provided in the KinEx A™ software (Sapidyne Instruments, Inc.). 25

Example 4

Comparison of Myostatin Inhibitors

The ability of three exemplary first-round peptibodies to bind to (KD) and inhibit (IC50) were compared with the KD and ICS0 values obtained for the soluble receptor fusion protein actRUB/Fc (R &D Systems, Inc., Minneapolis, Minn.). The IC50 values were determined using the pMARE luc cell-based assay described in Example 3 and the KD values were determined using the Biacore® assay described in Example 3. TABLE ΙΠ

Inhibitor IC50 (nM) Kd (nM) ActRUB/Fc -83 . -7 2xTN8-19-kc -9 -2 TN8-19 -23 -2 30 75 WO 2004/058988 PCT/US2003/040781 ’ TN8-29 ~26 ~60 TN12-34 -30 Linear-20 -11

The peptibodies have an IC5o that is improved over the receptor/Fc inhibitor and binding affinities which are comparable in two peptibodies with the receptor/Fc. 5 Example 5

Comparison of Ability of Peptide and Peptibody to Inhibit Myostatin

The following peptide sequence: QGHCTRWPWMCPPY (TN8-19) (SEQ ID NO: 33) was used to construct the corresponding peptibody TN8-19(pb) according to the procedure described in Example 2 above. Both the peptide alone and the peptibody were screened for 10 myostatin inhibiting activity using the C2C12 based assay described in Example 3 above. It can be seen from Figure 1 the IC50 (effective concentration for fifty percent inhibition of myostatin) for the peptibody is significantly less than that of the peptide, and thus the ability of the peptide to inhibit myostatin activity has been substantially improved by placing it in the peptibody configuration. 15 Example 6

Generation of Affinity-Matured Peptides and Peptibodies

Several of the first round peptides used for peptibody generation were chosen for affinity maturation. The selected peptides included the following: the cysteine constrained TN8-19, QGHCTRWPWMCPPY (SEQ ID NO: 33), and the linear peptides Linear-2 20 MEMLDSLFELLKDMVPISKA (SEQ ID NO: 104); Linear-15 HHGWNYLRKGSAPQWEEAWV (SEQ. ID NO: 117); Linear-17 RATLLKDFWQLVEGYGDN (SEQID NO: 119); Linear-20 YREMSMLEGLLDVLERLQHY (SEQ ID NO: 122), Linear-21 HNSSQMLLSELIMLVGSMMQ (SEQ ID NO: 123), Linear-24 EFFHWLHNHRSEVNHWLDMN (SEQ ID NO: 126). Based on a consensus sequence, directed 25 secondary phage display libraries were generated in which the “core”amino acids (determined from the consensus sequence) were either held constant or biased in frequency of occurrence. Alternatively, an individual peptide sequence could be used to generate a biased, directed phage display library. Panning of such libraries under more stringent conditions can yield peptides with enhanced binding to myostatin, selective binding to myostatin, or with some additional desired 30 property.

Production of doped oligos for libraries 76 WO 2004/058988 PCT/US2003/040781

Oligonucleotides were synthesized in a DNA synthesizer which were 91% “doped” at the core sequences, that is, each solution was 91% of the represented base (A, G, C, or T), and 3% of each of the other 3 nucleotides. For the TN8-19 family, for example, a 91% doped oligo used for the construction of a secondary phage library was the following:

5 5’-CAC AGT GCA CAG GGT NNK NNK NNK caK ggK caK tgK acK cgK tgK ccK tgK atK tgK ccK ccK taK NNK NNK NNK CAT TCT CTC GAG ATC A-3’ (SEQ ID NO: 634)

I 10 wherein “N” indicates that each of the four nucleotides A, T, C, and G were equally represented, K indicates that G and T were equally represented, and the lower case letter represents a mixture of 91% of the indicated base and 3% of each of the other bases. The family of oligonucleotides prepared in this manner were PCR amplified as described above, ligated into a phagemid vectors, for example, a modified pCESl plasmid (Dyax), or any available phagemid vector according to 15 the protocol described above. The secondary phage libraries generated were all 91% doped and had between 1 and 6.5x 109 independent transformants. The libraries were panned as described above, but with the following conditions:

Round 1 Panning:

Input phage number: 1012 -1013 cfu of phagemid 20 Selection method: Nunc Immuno Tube selection

Negative selection: 2 X with Nunc Immuno Tubes coated with 2% BSA at 10 min. each Panning coating: Coat with 1 μg of Myostatin protein in 1 ml of 0.1M Sodium carbonate buffer (pH 9.6)

Binding time: 3 hours 25 Washing conditions: 6 X 2%-Milk-PBST; 6XPBST; 2XPBS Elution condition: 100 mM TEA elution

Round 2 Panning:'

Input phage number: 1011 cfu of phagemid 30 Selection method: Nunc Immuno Tube selection

Negative selection: 2 X with Nunc Immuno Tubes coated with 2% BSA at 30 min. each Panning coating: Coat with 1 μg of Myostatin protein in 1 ml of 0.1M Sodium carbonate buffer (pH 9.6)

Binding time: 1 hour 35 Washing conditions: 15 X 2%-Milk-PBST, 1 X 2%-Milk-PBST for 1 hr., 10 X 2%-BSA-PBST,

1 X 2%-BSA-PBST for 1 hr., 10 X PBST and 3 X PBS

Elution condition: 100 mM TEA elution

Round 3 Panning: 40 Input phage number: 1010 cfu of phagemid

Selection method: Nunc Immuno Tube selection

Negative selection: 6 X with Nunc Immuno Tubes coated with 2% BSA at 10 min. each Panning coating: Coat with 0.1 μg of Myostatin protein in 1 ml of 0.1M Sodium carbonate buffer (pH 9.6) 45 Binding time: 1 hour 77 WO 2004/058988 PCT/US2003/040781

Washing conditions: 15 X 2%-Milk-PBST, 1 X 2%-Milk-PBST for 1 hr., 10 X 2%-BSA-PBST,

1 X 2%-BSA-PBST for 1 hr., 10 X PBST and 3 X PBS

Elution condition: 100 mM TEA elution 5 Panning of the secondary libraries yielded peptides with enhanced binding to myostatin.

Individual selected clones were subjected phage ELISA, as described above, and sequenced.

The following affinity matured TN8-19 family of peptides are shown in Table IV below. 78 WO 2004/058988 PCT/US2003/040781

Affinity- matured TABLEIV SEQED ?eptide sequence peptibody NO: ΠΪΓΝ8-19-1 305 7ALHGQCTRWPWMCPPQREG mTN8-19-2 306 YPEQGLCIRWPWMCPPQTLA mTN8-19-3 307 GLNQGHCTRWPWMCPPQDSN mTN8-19-4 308 MITQGQCTRWPWMCPPQPSG ΠΪΓΝ8-19-5 309 A.GAQEHCTRWPWMCAPNDWI mTN8-19-6 310 GVNQGQCTRWRWMCPPNGWE mTN8-19-7 311 LADHGQCIRWPWMCPPEGWE mTN8-19-8 312 tLEQAQCTRWPWMCPPQRGG mTN8-19-9 313 TQTHAQCTRWPWMCPPQWEG mTN8-19-10 314 WTQGHCTLWPWMCPPQRWR mTN8-19-ll 315 rYPHDQCIRWPWMCPPQPYP mTN8-19-12 316 SYWQGQCTRWPWMCPPQWRG ΠΪΓΝ8-19-13 317 MWQQGHCTRWPWMCPPQGWG mTN8-19-14 318 EFTQWHCTRWPWMCPPQRSQ ΠΪΓΝ8-19-15 319 LDDQWQCTRWPWMCPPQGPS mTN8-19-16 320 YQTQGLCTRWPWMCPPQSQR mTN8-19-17 321 ESNQGQCIRWPWMCPPQGGW mTN8-19-18 322 WTDRGPCTRWPWMCPPQANG mTN8-19-19 323 VGTQGQCTRWPWMCPPYETG mTN8-19-20 324 PYEQGKCTRWPWMCPPYEVE mTN8-19-21 325 SEYQGLCTRWPWMCPPQGWK mTN8-19-22 326 TFSQGHCTRWPWMCPPQGWG mTN8-19-23 327 PGAHDHCTRWPWMCPPQSRY mTN8-19-24 328 VAEEWHCRRWPWMCPPQDWR ΠΪΓΝ8-19-25 329 VGTQGHCIRWPWMCPPQPAG mTN8-19-26 330 EEDQAHCRSWPWMCPPQGWV mTN8-19-27 331 ADTQGHCTRWPWMCPPQHWF mTN8-19-28 332 SGPQGHCIRWPWMCAPQGWF mTN8-19-29 333 TLVQGHCTRWPWMCPPQRWV ΠΪΓΝ8-19-30 334 GMAHGKCTRWAWMCPPQSWK mTN8-19-31 335 ELYHGQCTRWPWMCPPQSWA mTN8-19-32 336 VADHGHCTRWPWMCPPQGWG mTN8-19-33 337 PESQGHCTRWPWMCPPQGWG mTN8-19-34 338 IPAHGHCTRWPWMCPPQRWR mTN8-19-35 339 FTVHGHCTRWPWMCPPYGWV mTN8-19-36 340 PDFPGHCTRWRWMCPPQGWE mTN8-19-37 341 QLWQGPCTQWPWMCPPKGRY mTN8-19-38 342 HANDGHCTRWQWMCPPQWGG mTN8-19-39 343 ETDHGLCTRWPWMCPPYGAR mTN8-19-40 344 GTWQGLCTRWPWMCPPQGWQ mTN8-19 coni 345 VATQGQCTRWPWMCPPQGWG mTN8-19 con2 346 VATQGQCTRWPWMCPPQRWG 79 WO 2004/058988 PCT/US2003/040781

mTN8 con6-l 347 QREWYPCYGGHLWCYDLHKA mTN8 con6-2 348 ISAWYSCYAGHFWCWDLKQK mTN8 con6-3 349 WTGWYQCYGGHLWCYDLRRK mTN8 con6^4 350 KTFWYPCYDGHFWCYNLKSS mTN8 con6-5 351 ESRWYPCYEGHLWCFDLTET

The consensus sequence derived from the affinity- matured TN-8-19- 1 through Con2 (excluding the mTN8 con6 sequences) shown above is: Caia-^Wa^WMCPP (SEP ID NO: 352). All of these peptide comprise the sequence WMCPP (SEQ ID NO: 633). The underlined amino 10 acids represent the core amino acids present in all embodiments, and ai, a2 and a3 are selected from a neutral hydrophobic, neutral polar, or basic amino acid. In one embodiment of this consensus sequence, Cbib2Wb3WMCPP (SEQ ID NO: 353), bi is selected horn any one of the amino acids T, I, or R; b2 is selected from any one of R, S, Q; and b3 is selected from any one of P, R and Q. All of the peptides comprise the sequence WMCPP (SEQ ID NO: 633). A more 15 detailed analysis of the affinity matured TN8 sequences comprising SEQ ID NO: 352 provides the following formula: c i CQmCdCsCfiCciCflWcqWMCPPci nci i Ci -.Ci 3 (SEQ ID NO: 354), wherein:

Cj is absent or any amino acid; c2 is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; 20 c3 is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; c4 is absent or any amino acid; c5 is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; c6 is absent or a neutral hydrophobic, neutral polar, or basic amino acid; c7 is a neutral hydrophobic, neutral polar, or basic amino acid; 25 c8 is a neutral hydrophobic, neutral polar, or basic amino acid; c9 is a neutral hydrophobic, neutral polar or basic amino acid; and wherein cio to C]3 is any amino acid.

In one embodiment of the above formulation, b7 is selected from any one of the amino acids T, I, or R; b8 is selected from any one of R, S, Q; and b9 is selected from any one of P, R 30 and Q. This provides the following sequence: did2d3d4d5d6Cd7d8Wd9WMCPP d1Dd„d,2d13 (SEQ ID NO: 355). dj is absent or any amino acid; d2 is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; d3 is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; 25 d4 is absent or any amino acid; d5 is absent or a neutral hydrophobic, neutral polar, or acidic amino acid; d6 is absent or a neutral hydrophobic, neutral polar, or basic amino acid; d7 is selected from any one of the amino acids T, I, or R; d8 is selected from any one of R, S, Q;

40 d9 is selected from any one of P, R and Q 80 WO 2004/058988 PCT/US2003/040781 and d10 through dJ3 are selected from any aminn acid.

The consensus sequence of the mTN8 con6 series is WYeie^Ye^G. (SEQ ID NO: 356) wherein ej is P, S or Y; e2 is C or Q, and e3 is G or H. 5

In addition to the TN-19 affinity matured family, additional affinity matured peptides were produced from the linear L-2, L-15, L-17, L-20, L-21, and L-24 first round peptides. These families are presented in Table V below.

Affinity matured peptihody TABLE V Peptide Sequence SEQ ID NO: L2 104 MEMLDSLEELLKDMVPISKA mL2-Conl 357 RMEMLESLLELLKEIVPMSKAG mL2-Con2 358 RMEMLESLLELLKEIVPMSKAR mL2-l 359 RMEMLESLLELLKDIVPMSKPS iriL2-2 360 GMEMLESLHELLQEIVPMSKAP mL2-3 361 RMEMLESLLELLKDIVPISNPP mL2-4 362 RIEMLESLLELLQEIVPISKAE mL2-5 363 RMEMLQSLLELLKDIVPMSNAR mL2-6 364 RMEMLESLLELLKEIVPTSNGT mL2-7 365 RMEMLESLEELLKEIVPMSKAG mL2-8 366 RMEMLGSLLELLKEIVPMSKAR mL2-9 367 QMELLDSLEELLKEIVPKSQPA mL2-10 368 RMEMLDSLLELLKEIVPMSNAR mL2-ll 369 RMEMLESLLELLHEIVPMSQAG rhL2-12 370 QMEMLESLLQLLKEIVPMSKAS mL2-13 371 RMEMLDSLLELLKDMVPMTTGA mL2-14 372 RLEMLESLLELLKDMVPMANAS mL2-15 373 RMEMLESLLQLLNEIVPMSRAR mL2-16 374 RMEMLESLEDLLKELVPMSKGV mL2-17 375 RJEMLESLLELLKDIVPIQKAR mL2-18 376 RMELLESLFELLKDMVPMSDSS mL2-19 377 RMEMLESLLEVLQEIVPRAKGA mL2-20 378 RMEMLDSLLQLLNEIVPMSHAR mL2-21 379 RMEMLESLLELLKDIVPMSNAG mL2-22 380 RMEMLQSLFELLKGMVPISKAG mL2-23 381 RMEMLESLLELLKEIVPNSTAA mL2-24 382 RMEMLQSLLELLKEIVPISKAG mL2-25 383 RIEMLDSLLELLNELVPMSKAR L-15 117 HHGWNYLRKGSAPQWEEAWV mL15-conl 384 QVESLQQLLMWLDQKLASGPQG mL15-l 385 RMELLESLEELLKEMVPRSKAV mL15-2 386 QAVSLQHLLMWLDQKLASGPQH 81 WO 2Q04/058988 PCTZUS2003/040781

mL15-3 387 DEDSLQQLLMWLDQKLASGPQL mL15-4 388 PVASLQQLLriVLDQKLAQGPHA mL15-5 389 EVDELQQLLNWLDHKLASGPLQ mL15-6 390 DVESLEQLLMWLDHQLASGPHG mL15-7 391 QVDSLQQVLLWLEHKLALGPQV mL15-8 392 GDESLQHLLMWLEQKLALGPHG mL15-9 393 QIEMLESLLDLLRDMVPMSNAF mL15-10 394 EVDSLQQLLMWLDQKLASGPQA mL15-ll 395 EDESLQQLLIYLDKMLSSGPQV mL15-12 396 AMDQLHQLLIWLDHKLASGPQA mL15-13 397 RIEMLESLLELLDEIALIPKAW inL15-14 398 EWSLQHLLMWLEHKLASGPDG mL15-15 399 GGESLQQLLMWLDQQLASGPQR mL15-16 400 GVESLQQLLIFLDHMLVSGPHD mL15-17 401 NVESLEHLMMWLERLLASGPYA mL15-18 402 QVDSLQQLLIWLDHQLASGPKR mL15-19 403 EVESLQQLLMWLEHKLAQGPQG mL15-20 404 EVDSLQQLLMWLDQKLASGPHA mL15-21 405 EVDSLQQLLMWLDQQLASGPQK inL15-22 406 GVEQLPQLLMWLEQKLASGPQR mL15-23 407 GEDSLQQLLMWLDQQLAAGPQV mL15-24 408 ADDSLQQLLMWLDRKLASGPHV mL15-25 409 PVDSLQQLLIWLDQKLASGPQG L-17 119 RATLLKDFWQLVEGYGDN mL17-conl 410 DWRATLLKEFWQLVEGLGDNLV mL17-con2 411 QSRATLLKEFWQLVEGLGDKQA mL17-l 412 DGRATLLTEFWQLVQGLGQKEA mL17-2 413 LARATLLKEFWQLVEGLGEKVV mL17-3 414 GSRDTLLKEFWQLWGLGDMQT mL17-4 415 DARATLLKEFWQLVDAYGDRMV mL17-5 416 NDRAQLLRDFWQLVDGLGVKSW mL17-6 417 GVRETLLYELWYLLKGLGANQG mL17-7 418 QARATLLKEFCQLVGCQGDKLS mL17-8 419 QERATLLKEFWQLVAGLGQNMR mL17-9 420 SGRATLLKEFWQLVQGLGEYRW mL17-10 421 TMRATLLKEFWLFVDGQREMQW mL17-ll 422 GERATELNDFWQLVDGQGDNTG mL17-12 423 DERETLLKEFWQLVHGWGDNVA mL17-13 424 GGRATLLKELWQLLEGQGANLV mL17-14 425 TARATLLNELVQLVKGYGDKLV mL17-15 426 GMRATLLQEFWQLVGGQGDNWM xnL17-16 427 STRATLLNDLWQLMKGWAEDRG mL17-17 428 SERATLLKELWQLVGGWGDNFG mL17-18 429 VGRATLLKEFWQLVEGLVGQSR mL17-19 430 EIRATLLKEFWQLVDEWREQPN 82 WO 2004/058988 PCT/US2003/040781

mL17-20 - 431 QLRATLLKEFLQLVHGLGETDS mL.17-21 432 TQRATLLKEFWQLEEGLGGKHV mL17-22 433 HYRATLLKEFWQLVDGLREQGV mL17-23 434 QSRVTLLREFWQLVESYRPIVN mL17-24 435 LSRATLLNEFWQFVDGQRDKRM mL17-25 436 WDRATLLNDFWHLMEELSQKPG mL17-26 437 QERATLLKEFWRMVEGLGKNRG mL17-27 438 NERATLLREFWQLVGGYGVNQR L-20 122 YREMSMLEGLLDVLERLQHY mL20-l 439 tiQRDMSMLWELLDVLDGLRQYS mL20-2 440 TQRDMSMLDGLLEVLDQLRQQR mL20-3 441 TSRDMSLLWELLEELDRLGHQR mL20-4 442 MQHDMSMLYGLVELLESLGHQI mL20-5 443 WNRDMRMLESLFEVLDGLRQQV mL20-6 444 GYRDMSMLEGLLAVLDRLGPQL mL20 coni 445 TQRDMSMLEGLLEVLDRLGQQR mL20 con2 446 WYRDMSMLEGLLEVLDRLGQQR L-21 123 HNSSQMLLSELIMLVGSMMQ mL21-l 447 TQNSRQMLLSDFMMLVGSMIQG mL21-2 448 MQTSRHELLSEFMMLVGSIMHG mL21-3 449 HDNSRQMLLSDLLHLVGTMIQG mL21-4 450 MENSRQNLLRELEMLVGNMSHQ mL21-5 451 QDTSRHMLLREFMMLVGEMIQG mL21 coni 452 dqnsrqmllsdlmilvgsmiqg L-24 126 EFFHWLHNHRSEVNHWLDMN mL24-l 453 · NVFFQWVQKHGRWYQWLDINV mL24-2 454 FDFLQWLQNHRSEVEHWLVMDV

The affinity matured peptides provided in Tables IV and V are then assembed into peptibodies as described above and assayed using the in vivo assays. 5 The affinity matured L2 peptides comprise a consensus sequence of fiEMLfcSLfJjLL, (SEQ ID NO: 455), wherein fj is M or I; f2 is any amino acid; f3 is L or F; and f4isE, QorD.

The affinity matured L15 peptide family comprise the sequence LgiR>LLg3gdL, (SEQ ID NO: 456), wherein gj is Q, D or E, g2 is S, Q, D or E, g3 is any amino acid, and g4 is L, W, F, or 10 Y. The affinity matured L17 family comprises the sequence: hih2h3h4h5h6h7h8h9 (SEQ ID NO: 457) wherein hi is R or D; h2 is any amino acid; h3 is A, T.S or Q; h4 is L or M; h5 is L or S; h6 is any amino acid; h7 is F or E; h8 is W, F or C; and h9 is L, F, M or K. Consensus sequences may also be determined for the mL20, mL21 and mL24 families of peptides shown above.

Peptibodies were constructed from these affinity matured peptides as described above, 15 using a linker attached to the Fc domain of human IgGl, having SEQ ID NO: 296, at the N- 83 WO 2004/058988 PCT/US2003/040781 terminus (N configuration), at the C terminus (C configuration) of the Fc, or at both the N and C terminals (N,C configurations), as described in Example 2 above. The peptides named were attached to the C or N terminals via a 5 glycine (5G), 8 glycine or k linker sequence. In the 2X peptibody version the peptides were linked with linkers such as 5 gly, 8 gly or k. Affinity 5 matured peptides and peptibodies are designated with a small “m” such as mTN8-19-22 for example. Peptibodies of the present invention further contain two splice sites where the peptides were spliced into the phagemid vectors. The position of these splice sites are AQ—peptide—LE. The peptibodies generally include these additional amino acids (although they are not included in the peptide sequences listed in the tables). In some peptibodies the LE amino acids were removed 10 from the peptides sequences. These peptibodies are designated -LE.

Exemplary peptibodies, and exemplary polynucleotide sequences encoding them, are provided in Table VI below. This table includes examples of peptibody sequences (as opposed to peptide only), such as the 2x mTN8-19-7 (SEQ ID NO: 615) and the peptibody with the LE sequences deleted (SEQ ID NO: 617). By way of explanation, the linker sequences in the 2x 15 versions refers to the linker between the tandem peptides. These peptibody sequences contain the Fc, linkers, AQ and LE sequences. The accompanying nucleotide sequence encodes the peptide sequence in addition to the AQ/LE linker sequences, if present, but does not encode the designated linker.

20 TABLE VI

Peptibody Name Peptide Nucleotide Sequence (SEQ ID No) Linker Terra -inus mL2-Conl RMEMLESLLELL KEIVPMSKAG CGTATGGAAATGCTTGAATCTCTTC TTGAACTTCTTAAAGAAATTGTTCC AATGTCTAAAGCTGGT (SEQ ID NO: 458) 5 gly N mL2-Con2 RMEMLESLLELL KEIVPMSKAR CGTATGGAAATGCTTGAATCTCTTC TTGAACTTCTTAAAGAAATTGTTCC AATGTCTAAAGCTCGT (SEQ ID NO: 459)· 5 gly N mL2-l RMEMLESLLELL KDIVPMSKPS CGTATGGAAATGCTTGAATCTCTTC TTGAACTTCTTAAAGATATTGTTCC AATGTCTAAACCATCT (SEQ ID NO: 460) 5 gly N mL2-2 GMEMLESLFELL QEIVPMSKAP GGTATGGAAATGCTTGAATCTCTTT TTGAACTTCTTCAAGAAATTGTTCC AATGTCTAAAGCTCCA' (SEQ ID NO: 461) 5 giy N mL2—3 RMEMLESLLELL KDIVPISNPP CGTATGGAAATGCTTGAATCTCTTC TTGAACTTCTTAAAGATATTGTTCC AATTTCTAATCCACCA (SEQ ID NO: 462) s giy N 84 WO 2004/058988 PCT/US2003/040781

mL2—4 RIEMLESLLELLQ EIVPISKAE CGTATTGAAATGCTTGAATCTCTTC TTGAACTTCTTCAAGAAATTGTTCC AATTTCTAAAGCTGAA (SEQIDNO·. 463) 5 giy N mL2-5 RMEMLQSLLELL KDIVPMSNAR CGTATGGAAATGCTTCAATCTCTTC TTGAACTTCTTAAAGATATTGTTCC AATGTCTAATGCTCGT (SEQIDNO: 464) 5 giy N mL2-6 RMEMLESLLELL KEIVPTSNGT CGTATGGAAATGCTTGAATCTCTTC TTGAACTTCTTAAAGAAATTGTTCC AACTTCTAATGGTACT (SEQ ID NO: 465) 5 giy N mL2—7 RMEMLESLEELL KEIVPMSKAG CGTATGGAAATGCTTGAATCTCTTT TTGAACTTCTTAAAGAAATTGTTCC AATGTCTAAAGCTGGT (SEQ ID NO: 466) 5 giy N mL2-8 RMEMLGSLLELL KEIVPMSKAR CGTATGGAAATGCTTGGTTCTCTTC TTGAACTTCTTAAAGAAATTGTTCC AATGTCTAAAGCTCGT(SEQ ID NO: 467) 5 giy N mL2-9 QMELLDSLEELL KEIVPKSQPA CAAATGGAACTTCTTGATTCTCTTT TTGAACTTCTTAAAGAAATTGTTCC AAAATCTCAACCAGCT (SEQ ID NO: 468) 5 giy N mL2—10 RMEMLDSLLELL KEIVPMSNAR CGTATGGAAATGCTTGATTCTCTTC TTGAACTTCTTAAAGAAATTGTTCC AATGTCTAATGCTCGT (SEQ ID NO: 469) s giy N mL2-ll RMEMLESLLELL HEIVPMSQAG CGTATGGAAATGCTTGAATCTCTTC TTGAACTTCTTCATGAAATTGTTCC AATGTCTCAAGCTGGT (SEQ ID NO: 470) 5 giy N mL2-12 QMEMLESLLQLL KEIVPMSKAS CAAATGGAAATGCTTGAATCTCTTC TTCAACTTCTTAAAGAAATTGTTCC AATGTCTAAAGCTTCT (SEQ ID NO: 471) 5 giy N mL2-13 RMEMLDSLLELL KDMVPMTTGA CGTATGGAAATGCTTGATTCTCTTC TTGAACTTCTTAAAGATATGGTTCC AATGACTACTGGTGCT (SEQ ID NO: 472) 5 giy N mL2-14 RIEMLESLLELLK DMVPMANAS CGTATTGAAATGCTTGAATCTCTTC TTGAACTTCTTAAAGATATGGTTCC AATGGCTAATGCTTCT (SEQIDNO: 473) 5 giy N mL2-15 RMEMLESLLQLL NEIVPMSRAR CGTATGGAAATGCTTGAATCTCTTC TTCAACTTCTTAATGAAATTGTTCC AATGTCTCGTGCTCGT (SEQ ID NO: 474) s giy N mL2—16 RMEMLESLFDLL KELVPMSKGV CGTATGGAAATGCTTGAATCTCTTT TTGATCTTCTTAAAGAACTTGTTCC AATGTCTAAAGGTGTT (SEQ ID NO: 475) 5 giy N mL2—17 RIEMLESLLELLK DIVPIQKAR CGTATTGAAATGCTTGAATCTCTTC TTGAACTTCTTAAAGATATTGTTCC AATTCAAAAAGCTCGT (SEQ ED NO: 476) 5 giy N 85 WO 2004/058988 PCT/US2003/040781

mL2-18 RMELLESLEELLK DMVPMSDSS CGTATGGAACTTCTTGAATCTCTTT TTGAACTTCTTAAAGATATGGTTCC AATGTCTGATTCTTCT (SEQIDNO: 477) 5 giy N mL2-19 RMEMLESLLEVL QEIVPRAKGA CGTATGGAAATGCTTGAATCTCTTC TTGAAGTTCTTCAAGAAATTGTTCC ACGTGCTAAAGGTGCT (SEQID NO: 478) 5 gly N mL2—20 RMEMLDSLLQLL NEEVPMSHAR CGTATGGAAATGCTTGATTCTCTTC TTCAACTTCTTAATGAAATTGTTCC AATGTCTCATGCTCGT (SEQ ID NO: 479) 5 giy N mL2—21 RMEMLESLLELL KDIVPMSNAG CGTATGGAAATGCTTGAATCTCTTC TTGAACTTCTTAAAGATATTGTTCC AATGTCTAATGCTGGT (SEQ ED NO: 480) . 5 gly N mL2—22 RMEMLQSLFELL KGMVPISKAG CGTATGGAAATGCTTCAATCTCTTT TTGAACTTCTTAAAGGTATGGTTCC AATTTCTAAAGCTGGT (SEQID NO: 481) 5 giy N mL2-23 RMEMLESLLELL KEIVPNSTAA CGTATGGAAATGCTTGAATCTCTTC TTGAACTTCTTAAAGAAATTGTTCC AAATTCTACTGCTGCT (SEQ ED NO: 482) 5 gly N mL2-24 RMEMLQSLLELL KEIVPISKAG CGTATGGAAATGCTTGAATCTCTTC TTGAACTTCTTAAAGAAATTGTTCC AATTTCTAAAGCTGGT (SEQIDNO: 483) 5 gly N mL2-25 RIEMLDSLLELLN ELVPMSKAR CGTATTGAAATGCTTGATTCTCTTC TTGAACTTCTTAATGAACTTGTTCC AATGTCTAAAGCTCGT (SEQIDNO: 484) 5 giy N mL17-Conl DWRATLLKEFW QLVEGLGDNLV GATTGGCGTGCTACTCTTCTTAAAG AATTTTGGCAACTTGTTGAAGGTCT TGGTGATAATCTTGTT (SEQ ED NO: 485) 5 giy N mL17-l DGRATLLTEFWQ LVQGLGQKEA GATGGTCGTGCTACTCTTCTTACTG AATTTTGGCAACTTGTTCAAGGTCT TGGTCAAAAAGAAGCT (SEQID NO: 486) 5 gly N inL17-2 LARATLLKEFWQ LVEGLGEKVV CTTGCTCGTGCTACTCTTCTTAAAG AATTTTGGCAACTTGTTGAAGGTCT TGGTGAAAAAGTTGTT (SEQIDNO: 487) 5 gly N mL17-3 GSRDTLLKEFWQ LVVGLGDMQT GGTTCTCGTGATACTCTTCTTAAAG AATTTTGGCAACTTGTTGTTGGTCT TGGTGATATGCAAACT (SEQIDNO: 488) 5 gly N mL17-4 DARATLLKEFWQ LVDAYGDRMV GATGCTCGTGCTACTCTTCTTAAAG AATTTTGGCAACTTGTTGATGCTTA TGGTGATCGTATGGTT (SEQ ID NO: 489) s giy N mL17-5 NDRAQLLRDFWQ LVDGLGVKSW AATGATCGTGCTCAACTTCTTCGTG ATTTTTGGCAACTTGTTGATGGTCT TGGTGTTAAATCTTGG (SEQ ID NO: 490) 5 giy N 86 WO 2004/058988 PCT/US2003/040781

mL17-6 GVKETLLYELWY LLKGLGANQG GGTGTTCGTGAAACTCTTCTTTATG AACTTTGGTATCTTCTTAAAGGTCT TGGTGCTAATCAAGGT (SEQ ID NO: 491) 5 giy N mL17-7 QARATLLKEFCQ LVGCQGDKLS CAAGCTCGTGCTACTCTTCTTAAAG AATTTTGTCAACTTGTTGGTTGTCA AGGTGATAAACTTTCT (SEQ ID NO: 492) 5 giy N mL17-8 QERATLLKEFWQ LVAGLGQNMR CAAGAACGTGCTACTCTTCTTAAA GAATTTTGGCAACTTGTTGCTGGTC TTGGTCAAAATATGCGT (SEQID NO: 493) 5 giy N mL17-9 SGRATLLKEFWQ LVQGLGEYRW TCTGGTCGTGCTACTCTTCTTAAAG aattttggcaacttgttcaaggtct TGGTGAATATCGTTGG (SEQ ID NO: 494) 5 giy N mL17-10 TMRATLLKEFWL FVDGQREMQW ACTATGCGTGCTACTCTTCTTAAAG AATTTTGGCTTTTTGTTGATGGTCA ACGTGAAATGCAATGG (SEQ ID NO: 495) 5 giy N mL17-ll GERATLLNDFWQ LVDGQGDNTG GGTGAACGTGCTACTCTTCTTAATG ATTTTTGGCAACTTGTTGATGGTCA AGGTGATAATACTGGT (SEQID NO: 496) 5 giy N mL17-12 DERETLLKEFWQ LVHGWGDNVA GATGAACGTGAAACTCTTCTTAAA GAATTTTGGCAACTTGTTCATGGTT GGGGTGATAATGTTGCT (SEQID NO: 497) 5 giy N mL17-13 GGRATLLKELWQ LLEGQGANLV GGTGGTCGTGCTACTCTTCTTAAAG AACTTTGGCAACTTCTTGAAGGTCA AGGTGCTAATCTTGTT (SEQ ID NO: 498) 5 giy N mLl'7-14 TARATLLNELVQ LVKGYGDKLV ACTGCTCGTGCTACTCTTCTTAATG AACTTGTTCAACTTGTTAAAGGTTA TGGTGATAAACTTGTT (SEQ ID NO: 499) 5 giy N mL17-15 GMRATLLQEFWQ LVGGQGDNWM GGTATGCGTGCTACTCTTCTTCAAG AATTTTGGCAACTTGTTGGTGGTCA AGGTGATAATTGGATG (SEQID NO: 500) 5 giy N mL17-16 STRATLLNDLWQ' LMKGWAEDRG TCTACTCGTGCTACTCTTCTTAATG atctttggcaacttatgaaaggttg GGCTGAAGATCGTGGT (SEQID NO: 501) 5 giy N mL17-17 SERATLLKELWQ LVGGWGDNFG TCTGAACGTGCTACTCTTCTTAAAG AACTTTGGCAACTTGTTGGTGGTTG GGGTGATAATTTTGGT (SEQ ID NO: 502) 5 giy N mL17—18 VGRATLLKEFWQ LVEGLVGQSR GTTGGTCGTGCTACTCTTCTTAAAG AATTTTGGCAACTTGTTGAAGGTCT TGTTGGTCAATCTCGT (SEQ ID NO: 503) 5 giy N 87 WO 2004/058988 PCT/US2003/040781 2x mTN8-Con6-(N)-1K M-GAQ- WYPCYEGHFWC YDL- GSGSATGGSGST ASSGSGSATG-WYPCYEGHFWC YDL-LE-5G-FC (SEQ ID NO: 504) TGGTATCCGTGTTATGAGGGTCACT TCTGGTGCTACGATCTGGGTTCTGG TTCCACTGCTTCTTCTGGTTCCGGT TCCGCTACTGGTTGGTACCCGTGCT ACGAAGGTCACTTTTGGTGTTATGA TCTG (SEQ ID NO: 505) IK N 2x mTN8-Con6-(C)-1K FC-5G-AQ- WYPCYEGHFWC YDL- GSGSATGGSGST ASSGSGSATG-WYPCYEGHFWC YDL-LE (SEQ ID NO: 506) TGGTATCCGTGTTATGAGGGTCACT TCTGGTGCTACGATCTGGGTTCTGG TTCCACTGCTTCTTCTGGTTCCGGT TCCGCTACTGGTTGGTACCCGTGCT ACGAAGGTCACTTTTGGTGTTATGA TCTG (SEQ ID NO: 507) IK c 2x mTNS—Con7- (N)-1K M-GAQ- IFGCKWWDVQC YQF- GSGSATGGSGST ASSGSGSATG-EFGCKWWDVQC YQF-LE-5G-FC (SEQ ID NO: 508) ATCTTTGGCTGTAAATGGTGGGAC GTTCAGTGCTACCAGTTCGGTTCTG GTTCCACTGCTTCTTCTGGTTCCGG TTCCGCTACTGGTATCTTCGGTTGC AAGTGGTGGGATGTACAGTGTTAT CAGTTT (SEQ ID NO: 509) IK N 2x mTN8-Con7-(C)-1K FC-5G-AQ- IFGCKWWDVQC YQF- GSGSATGGSGST ASSGSGSATG-IFGCKWWDVQC YQF-LE (SEQ ID NO: 510) ATCTTTGGCTGTAAATGGTGGGAC GTTCAGTGCTACCAGTTCGGTTCTG GTTCCACTGCTTCTTCTGGTTCCGG TTCCGCTACTGGTATCTTCGGTTGC AAGTGGTGGGATGTACAGTGTTAT CAGTTT (SEQ ID NO: 511) IK · c 2x mTN8-Con8- (N)-1K M-GAQ- IFGCKWWDVDC YQF- GSGSATGGSGST ASSGSGSATG-EFGCKWWDVDC YQF-LE-5G-FC (SEQ ID NO: 512) ATCTTTGGCTGTAAGTGGTGGGAC GTTGACTGCTACCAGTTCGGTTCTG GTTCCACTGCTTCTTCTGGTTCCGG TTCCGCTACTGGTATCTTCGGTTGC AAATGGTGGGACGTTGATTGTTAT CAGTTT (SEQ ID NO: 513) IK N 2x mTN8-Con8-(C)-1K FC-5G-AQ- IFGCKWWDVDC YQF- GSGSATGGSGST ASSGSGSATG-EFGCKWWDVDC YQF-LE (SEQ ID NO: 514) ATCTTTGGCTGTAAGTGGTGGGAC GTTGACTGCTACCAGTTCGGTTCTG GTTCCACTGCTTCTTCTGGTTCCGG TTCCGCTACTGGTATCTTCGGTTGC AAATGGTGGGACGTTGATTGTTAT CAGTTT (SEQ ID NO: 515) IK c ML15-Conl QVESLQQLLMWL DQKLASGPQG CAGGTTGAATCCCTGCAGCAGCTG CTGATGTGGCTGGACCAGAAACTG GCTTCCGGTCCGCAGGGT (SEQ ID NO: 516) 5 giy c ML15-1 RMELLESLFELLK EMVPRSKAV CGTATGGAACTGCTGGAATCCCTG TTCGAACTGCTGAAAGAAATGGTT CCGCGTTCCAAAGCTGTT (SEQ ID NO: 517) 5 giy c 88 WO 2004/058988 PCT/TJS2003/040781 mL15-2 QAVSLQHLLMW LDQKLASGPQH CAGGCTGTTTCCCTGCAGCACCTGC TGATGTGGCTGGACCAGAAACTGG CTTCCGGTCCGCAGCAC (SEQ ID NO: 518) 5 giy c mL15-3 DEDSLQQLLMWL DQKLASGPQL GACGAAGACTCCCTOCAGCAGCTG CTGATGTGGCTGGACCAGAAACTG GCTTCCGGTCCGCAGCTG (SEQ ID NO: 519) 5 giy c mL15-4 PVASLQQLLIWL DQKLAQGPHA CCGGTTGCTTCCCTGCAGCAGCTGC TGATCTGGCTGGACCAGAAACTGG CTCAGGGTCCGCACGCT (SEQ ID NO: 520) 5 giy c mL15-5 EVDELQQLLNWL DHKLASGPLQ GAAGTTGACGAACTGCAGCAGCTG GTGAACTGGCTGGACCAGAAACTG GCTTCCGGTCCGCTGCAG (SEQ ID NO: 521) 5 giy c mL15-6 DVESLEQLLMWL DHQLASGPHG GACGTTGAATCCCTGGAACAGCTG CTGATGTGGCTGGACCACCAGCTG GCTTCCGGTCCGCACGGT (SEQ ID NO: 522) 5 giy c mL15-7 QVDSLQQVLLWL EHKLALGPQV CAGGTTGACTCCCTGCAGCAGGTT CTGCTGTGGCTGGAACACAAACTG GCTCTGGGTCCGCAGGTT (SEQ ED NO: 523) 5 giy c mL15-8 GDESLQHLLMWL EQKLALGPHG GGTGACGAATCCCTGCAGCACCTG CTGATGTGGCTGGAACAGAAACTG GCTCTGGGTCCGCACGGT (SEQ ED NO: 524) 5 giy c mL15-9 QIEMLESLLDLLR DMVPMSNAF CAGATCGAAATGCTGGAATCCCTG CTGGACCTGCTGCGTGACATGGTTC CGATGTCCAACGCTTTC (SEQ ID NO: 525) 5 giy c mL15-10 EVDSLQQLLMWL DQKLASGPQA GAAGTTGACTCCCTGCAGCAGCTG CTGATGTGGCTGGACCAGAAACTG GCTTCCGGTCCGCAGGCT (SEQ ID NO: 526) 5 giy c mL15-ll EDESLQQLLIYLD KMLSSGPQV GAAGACGAATCCCTGCAGCAGCTG CTGATCTACCTGGACAAAATGCTG TCCTCCGGTCCGCAGGTT (SEQ ED NO: 527)’ 5 giy c mL15-12 AMDQLHQLLIWL DHKLASGPQA GCTATGGACCAGCTGCACCAGCTG CTGATCTGGCTGGACCACAAACTG GCTTCCGGTCCGCAGGCT (SEQ ID NO: 528) 5 giy c mL15-13 RIEMLESLLELLD EIALIPKAW CGTATCGAAATGCTGGAATCCCTG CTGGAACTGCTGGACGAAATCGCT CTGATCCCGAAAGCTTGG (SEQ ED NO: 529) 5 giy c mL15-14 EVVSLQHLLMWL EHKLASGPDG GAAGTTGTTTCCCTGCAGCACCTGC TGATGTGGCTGGAACACAAACTGG CTTCCGGTCCGGACGGT (SEQ ID NO: 530) 5 giy c mL15-15 GGESLQQLLMWL DQQLASGPQR GGTGGTGAATCCCTGCAGCAGCTG CTGATGTGGCTGGACCAGCAGCTG GCTTCCGGTCCGCAGCGT (SEQ ID NO: 531) 5 giy c 89 WO 2004/058988 PCT/US2003/040781 mL15-16 GVESLQQLLIFLD HMLVSGPHD GGTGTTGAATCCCTGCAGCAGCTG CTGATCTTCCTGGACCACATGCTGG TTTCCGGTCCGCACGAC (SEQID NO: 532) Sgiy c mL15-17 NVESLEHLMMW LERLLASGPYA AACGTTGAATCCCTGGAACACCTG ATGATGTGGCTGGAACGTCTGCTG GCTTCCGGTCCGTACGCT (SEQ ID NO: 533) Sgiy c mL 15-18 QVDSLQQLLIWL DHQLASGPKR CAGGTTGACTCCCTGCAGCAGCTG CTGATCTGGCTGGACCACCAGCTG GCTTCCGGTCCGAAACGT (SEQID NO: 534) Sgiy c mL 15-19 EVESLQQLLMWL EHKLAQGPQG . GAAGTTGAATCCCTGCAGCAGCTG CTGATGTGGCTGGAACACAAACTG GCTCAGGGTCCGCAGGGT (SEQID NO: 535) 5 giy c mLl5-20 EVDSLQQLLMWL DQKLASGPHA GAAGTTGACTCCCTGCAGCAGCTG CTGATGTGGCTGGACCAGAAACTG GCTTCCGGTCCGCACGCT (SEQID NO: 536) 5giy c mL15-21 EVDSLQQLLMWL DQQLASGPQK GAAGTTGACTCCCTGCAGCAGCTG CTGATGTGGCTGGACCAGCAGCTG GCTTCCGGTCCGCAGAAA (SEQID NO: 537) Sgiy c mL15—22 GVEQLPQLLMWL EQKLASGPQR GGTGTTGAACAGCTGCCGCAGCTG CTGATGTGGCTGGAACAGAAACTG GCTTCCGGTCCGCAGCGT (SEQID NO: 538) 5giy c mL15-23 GEDSLQQLLMWL DQQLAAGPQV GGTGAAGACTCCCTGCAGCAGCTG CTGATGTGGCTGGACCAGCAGCTG GCTGCTGGTCCGCAGGTT (SEQID NO: 539) Sgiy c mL15—24 ADDSLQQLLMW LDRKLASGPHV GCTGACGACTCCCTGCAGCAGCTG CTGATGTGGCTGGACCGTAAACTG GCTTCCGGTCCGCACGTT (SEQID NO: 540) Sgiy c mL15—25 PVDSLQQLLIWL DQKLASGPQG CCGGTTGACTCCCTGCAGCAGCTG CTGATCTGGCTGGACCAGAAACTG GCTTCCGGTCCGCAGGGT (SEQID NO: 541) Sgiy c mL17-Con2 QSRATLLKEFWQ LVEGLGDKQA CAGTCCCGTGCTACCCTGCTGAAA GAATTCTGGCAGCTGGTTGAAGGT CTGGGTGACAAACAGGCT (SEQID NO: 542) Sgiy c mL17-19 EIRAILLKEFWQL VDEWREQPN GAAATCCGTGCTACCCTGCTGAAA GAATTCTGGCAGCTGGTTGACGAA TGGCGTGAACAGCCGAAC (SEQ ID NO: 543) 5giy c mL17-20 QLRATLLKEFLQL VHGLGETDS CAGCTGCGTGCTACCCTGCTGAAA GAATTCCTGCAGCTGGTTCACGGTC TGGGTGAAACCGACTCC (SEQID NO: 544) Sgiy c mL17-21 TQRATLLKEFWQ LIEGLGGKHV ACCCAGCGTGCTACCCTGCTGAAA GAATTCTGGCAGCTGATCGAAGGT CTGGGTGGTAAACACGTT (SEQID NO: 545) s giy c 90 WO 2004/058988 PCT/US2003/040781 mL17-22 HYRATLLKEFWQ LVDGLREQGV CACTACCGTGCTACCCTGCTGAAA GAATTCTGGCAGCTGGTTGACGGT CTGCGTGAACAGGGTGTT (SEQID NO: 546) 5 giy C mL17-23 QSRVTLLREFWQ LVESYRPIVN CAGTCCCGTGTTACCCTGCTGCGTG AATTCTGGCAGCTGGTTGAATCCTA CCGTCCGATCGTTAAC (SEQ ID NO: 547) 5 giy c mL17-24 LSRATLLNEFWQ FVDGQRDKRM CTGTCCCGTGCTACCCTGCTGAACG AATTCTGGCAGTTCGTTGACGGTCA GCGTGACAAACGTATG (SEQID NO: 548) 5 giy c mL17-25 WDRATLLNDFW HLMEELSQKPG TGGGACCGTGCTACCCTGCTGAAC GACTTCTGGCACCTGATGGAAGAA CTGTCCCAGAAACCGGGT (SEQID NO: 549) 5 giy c mL17-26 QERATLLKEFWR MVEGLGKNRG CAGGAACGTGCTACCCTGCTGAAA GAATTCTGGCGTATGGTTGAAGGT CTGGGTAAAAACCGTGGT (SEQID NO: 550) 5giy c mL17—27 NERATLLREFWQ LVGGYGVNQR ' AACGAACGTGCTACCCTGCTGCGT GAATTCTGGCAGCTGGTTGGTGGTT ACGGTGTTAACCAGCGT (SEQID NO: 551) 5giy c mTN8Con6-l QREWYPCYGGHL WCYDLHKA CAGCGTGAATGGTACCCGTGCTAC GGTGGTCACCTGTGGTGCTACGAC CTGCACAAAGCT (SEQ ID NO: 552) 5 gly c mTN8Con6—2 ISAWYSCYAGHF WCWDLKQK ATCTCCGCTTGGTACTCCTGCTACG CTGGTCACTTCTGGTGCTGGGACCT GAAACAGAAA (SEQ ED NO: 553) 5 gly c mTN8Con6-3' WTGWYQCYGGH LWCYDLRRK · TGGACCGGTTGGTACCAGTGCTAC GGTGGTCACCTGTGGTGCTACGAC CTGCGTCGTAAA (SEQ ED NO: 554) 5 giy c mTN8Con6-4 KTFWYPCYDGHF WCYNLKSS AAAACCTTCTGGTACCCGTGCTAC GACGGTCACTTCTGGTGCTACAAC CTGAAATCCTCC (SEQ ID NO: 545) 5 giy c mTN8Con6—5 ESRWYPCYEGHL WCFDLTET GAATCCCGTTGGTACCCGTGCTAC GAAGGTCACCTGTGGTGCTTCGAC CTGACCGAAACC (SEQ ED NO: 546) s gly c mL24—1 NVEFQWVQKHG RVVYQWLDINV AATGTTTTTTTTCAATGGGTTCAAA AACATGGTCGTGTTGTTTATCAATG GCTTGATATTAATGTT (SEQ ED NO: 557) 5 giy c mL24—2 FDFLQWLQNHRS EVEHWLVMDV TTTGATTTTCTTCAATGGCTTCAAA ATCATCGTTCTGAAGTTGAACATTG GCTTGTTATGGATGTT (SEQ ID NO: 558) 5 giy c mL20-l HQRDMSMLWEL LDVLDGLRQYS CATCAACGTGATATGTCTATGCTTT GGGAACTTCTTGATGTTCTTGATGG TCTTCGTCAATATTCT (SEQ ID NO: 559) 5 gly c mL20-2 TQRDMSMLDGLL EVLDQLRQQR ACTCAACGTGATATGTCTATGCTTG ATGGTCTTCTTGAAGTTCTTGATCA ACTTCGTCAAGAACGT (SEQID 5 gly c 91' WO 2004/058988 PCT/US2003/040781 NO: 560) mL20-3 TSRDMSLLWELL EELDRLGHQR ACCTCCCGTGACATGTCCCTGCTGT GGGAACTGCTGGAAGAACTGGACC GTCTGGGTCACCAGCGT (SEQED NO: 561) 5 giy C mL20-4 MQHDMSMLYGL VELLESLGHQI ATGCAACATGATATGTCTATGCTTT ATGGTCTTGTTGAACTTCTTGAATC TCTTGGTCATCAAATT (SEQ ED NO: 562) 5 gly C mL20-5 WNRDMRMLESL EEYLDGLRQQV TGGAATCGTGATATGCGTATGCTTG AATCTCTTTTTGAAGTTCTTGATGG TCTTCGTCAACAAGTT (SEQ ED NO: 563) 5 giy c mT ,2.0—6 GYRDMSMLEGLL AVLDRLGPQL GGTTATCGTGATATGTCTATGCTTG AAGGTCTTCTTGCTGTTCTTGATCG TCTTGGTCCACAACTT (SEQ ED NO: 564) s giy c mL20 Coni TQRDMSMLEGLL EVLDRLGQQR ACTCAACGTGATATGTCTATGCTTG AAGGTCTTCTTGAAGTTCTTGATCG TCTTGGTCAACAACGT (SEQ ED NO: 565) 5 giy c mL20 Con2 WYRDMSMLEGL LEVLDRLGQQR TGGTACCGTGACATGTCCATGCTG GAAGGTCTGCTGGAAGTTCTGGAC CGTCTGGGTCAGCAGCGT (SEQ ID NO: 566) 5 gly c niL21-l TQNSRQMLLSDF MMLVGSMIQG ACTCAAAATTCTCGTCAAATGCTTC TTTCTGATTTTATGATGCTTGTTGG TTCTATGATTCAAGGT (SEQ ED NO: 567) 5 giy c mL21-2 MQTSREULLSEFM MLVGSIMHG ATGCAAACTTCTCGTCATATTCTTC TTTCTGAATTTATGATGCTTGTTGG TTCTATTATGCATGGT (SEQ ID NO: 568) 5 giy c mL21-3 HDNSRQMLLSDL LHLVGTMIQG CACGACAACTCCCGTCAGATGCTG CTGTCCGACCTGCTGCACCTGGTTG GTACCATGATCCAGGGT (SEQ ID NO: 569) s giy c mL2W MENSRQNLLRELI MLVGNMSHQ ATGGAAAACTCCCGTCAGAACCTG CTGCGTGAACTGATCATGCTGGTTG GTAACATGTCCCACCAG (SEQ ID NO: 570) 5 giy c mL21-5 QDTSRHMLLREF MMLVGEMIQG CAGGACACCTCCCGTCACATGCTG CTGCGTGAATTCATGATGCTGGTTG GTGAAATGATCCAGGGT (SEQ ID NO: 571) 5 giy c mL21 Coni DQNSRQMLLSDL MILVGSMIQG GACCAGAACTCCCGTCAGATGCTG CTGTCCGACCTGATGATCCTGGTTG GTTCCATGATCCAGGGT (SEQ ID NO: 572) 5 giy c ΠΪΓΝ8-19-1 VALHGQCTRWP WMCPPQREG GTTGCTCTTCATGGTCAATGTACTC GTTGGCCATGGATGTGTCCACCAC AACGTGAAGGT (SEQ ID NO: 573) 5 gly c 92 WO 2004/058988 PCT/US2003/040781 mTN8-19-2 ' YPEQGLCTRWPW MCPPQTLA l TATCCAGAACAAGGTCTTTGTACTC GTTGGCCATGGATGTGTCCACCAC AAACTCTTGCT (SEQ ED N: 574) 5 gty C mTN8-19-3 GLNQGHCTRWP WMCPPQDSN GGTCTGAACCAGGGTCACTGCACC CGTTGGCCGTGGATGTGCCCGCCG CAGGACTCCAAC (SEQ ED NO: 575) 5 gly C mTN8-19-4 MITQGQCTRWPW MCPPQPSG ATGATTACTCAAGGTCAATGTACTC GTTGGCCATGGATGTGTCCACCAC AACCATCTGGT (SEQ ID NO: 576) 5 gly C mTN8-19-5 AGAQEHCTRWP WMCAPNDWI GCTGGTGCTCAGGAACACTGCACC CGTTGGCCGTGGATGTGCGCTCCG AACGACTGGATC (SEQ ID NO: 577) 5 gly c mTN8-19-6 GVNQGQCTRWR WMCPPNGWE GGTGTTAACCAGGGTCAGTGCACC CGTTGGCGTTGGATGTGCCCGCCG AACGGTTGGGAA (SEQ ID NO: 578) 5 gly c mTN8-19-7 LADHGQCIRWPW MCPPEGWE CTGGCTGACCACGGTCAGTGCATC CGTTGGCCGTGGATGTGCCCGCCG GAAGGTTGGGAA (SEQ ID NO: 579) 5 gly c mTN8-19-8 ILEQAQCTRWPW MCPPQRGG ATCCTGGAACAGGCTCAGTGCACC CGTTGGCCGTGGATGTGCCCGCCG CAGCGTGGTGGT (SEQ ID NO: 580) 5 gly c mTN8-19-9 TQTHAQCTRWP WMCPPQWEG ACTCAAACTCATGCTCAATGTACTC GTTGGCCATGGATGTGTCCACCAC AATGGGAAGGT (SEQ ED NO: 581) 5 gly c mTKf8-19-10 VVTQGHCTLWP WMCPPQRWR GTTGTTACTCAAGGTCATTGTACTC TTTGGCCATGGATGTGTCCACCACA ACGTTGGCGT (SEQ ED NO: 582) 5 gly c mTN8-19-ll IYPHDQCTRWPW MCPPQPYP ATTTATCCACATGATCAATGTACTC GTTGGCCATGGATGTGTCCACCAC AACCATATCCA (SEQ ID NO: 583) 5 gly c ΠΪΤΝ8-19-12 SYWQGQCTRWP WMCPPQWRG TCTTATTGGCAAGGTCAATGTACTC GTTGGCCATGGATGTGTCCACCAC AATGGCGTGGT (SEQ ID NO: 584) 5 gly c mTN8-19-13 MWQQGHCTRWP WMCPPQGWG ATGTGGCAACAAGGTCATTGTACT CGTTGGCCATGGATGTGTCCACCA CAAGGTTGGGGT (SEQ ID NO: 585) 5 gly c mTN8-19-14 EFTQWHCTRWP WMCPPQRSQ GAATTCACCCAGTGGCACTGCACC CGTTGGCCGTGGATGTGCCCGCCG CAGCGTTCCCAG (SEQ ID NO: 586) 5 gly c mTN8-I9-15 LDDQWQCTRWP WMCPPQGFS CTGGACGACCAGTGGCAGTGCACC CGTTGGCCGTGGATGTGCCCGCCG CAGGGTTTCTCC (SEQ ID NO: 587) 5 gly c mTN8-19-16 YQTQGLCTRWP WMCPPQSQR TATCAAACTCAAGGTCTTTGTACTC GTTGGCCATGGATGTGTCCACCAC AATCTCAACGT (SEQ ID NO: 588) 5 gly c mTNS-19-17 ESNQGQCTRWP WMCPPQGGW GAATCTAATCAAGGTCAATGTACT CGTTGGCCATGGATGTGTCCACCA CAAGGTGGTTGG (SEQ ID NO: 589) 5 gly c 93 WO 2004/058988 PCT/US2003/040781 mTN8-19-18 WTDRGPCTRWP WMCPPQANG TGGACCGACCGTGGTCCGTGCACC CGTTGGCCGTGGATGTGCCCGCCG CAGGCTAACGGT (SEQ ID NO: 590) 5 giy c mTN8-19-19 VGTQGQCTRWP WMCPPYETG GTTGGTACCCAGGGTCAGTGCACC CGTTGGCCGTGGATGTGCCCGCCG TACGAAACCGGT (SEQ ID NO: 591) 5 giy c mTN8-19-20 PYEQGKCTRWP WMCPPYEVE CCGTACGAACAGGGTAAATGCACC CGTTGGCCGTGGATGTGCCCGCCG TACGAAGTTGAA (SEQ ID NO: 592) s giy c mTN8-19-21 SEYQGLCTRWPW MCPPQGWK TCCGAATACCAGGGTCTGTGCACC CGTTGGCCGTGGATGTGCCCGCCG CAGGGTTGGAAA (SEQ ID NO: 593) 5 giy c mTN8—19-22 TFSQGHCIRWPW MCPPQGWG ACCTTCTCCCAGGGTCACTGCACCC GTTGGCCGTGGATGTGCCCGCCGC AGGGTTGGGGT (SEQ ID NO: 594) 5 giy c mTN8-19-23 PGAHDHCTRWP WMCPPQSRY CCGGGTGCTCACGACCACTGCACC CGTTGGCCGTGGATGTGCCCGCCG CAGTCCCGTTAC (SEQ ID NO: 595) 5 giy c mTN8-19-24 VAEEWHCRRWP WMCPPQDWR GTTGCTGAAGAATGGCACTGCCGT CGTTGGCCGTGGATGTGCCCGCCG CAGGACTGGCGT (SEQ ID NO: 596) 5 giy c mTN8-19-25 VGTQGHCTRWP WMCPPQPAG GTTGGTACCCAGGGTCACTGCACC CGTTGGCCGTGGATGTGCCCGCCG CAGCCGGCTGGT (SEQ ID NO: 597) 5 giy c mTN8-19-26 EEDQAHCRSWP WMCPPQGWV GAAGAAGACCAGGCTCACTGCCGT TCCTGGCCGTGGATGTGCCCGCCG CAGGGTTGGGTT (SEQ ID NO: 598) 5 giy c mTN 8-19-27 ADTQGHCTRWP WMCPPQHWF GCTGACACCCAGGGTCACTGCACC CGTTGGCCGTGGATGTGCCCGCCG CAGCACTGGTTC (SEQ ID NO: 599) 5 giy c mTN8-19-28 SGPQGHCTRWPW MCAPQGWF TCCGGTCCGCAGGGTCACTGCACC CGTTGGCCGTGGATGTGCGCTCCG CAGQGTTGGTTC (SEQ ID NO: 600) 5 giy c mTNS-19-29 TLVQGHCTRWP WMCPPQRWV ACCCTGGTTCAGGGTCACTGCACC CGTTGGCCGTGGATGTGCCCGCCG CAGCGTTGGGTT (SEQ ID NO: 601) s giy c mTN8—19-30 GMAHGKCIRWA WMCPPQSWK GGTATGGCTCACGGTAAATGCACC CGTTGGGCTTGGATGTGCCCGCCG CAGTCCTGGAAA (SEQ ID NO: 602) 5 giy c mTN8-19-31 ELYHGQCTRWP WMCPPQSWA GAACTGTACCACGGTCAGTGCACC CGTTGGCCGTGGATGTGCCCGCCG CAGTCCTGGGCT (SEQ ID NO: 603) 5 giy c mTN8-19—32 VADHGHCTRWP WMCPPQGWG GTTGCTGACCACGGTCACTGCACC CGTTGGCCGTGGATGTGCCCGCCG CAGGGTTGGGGT (SEQ ID NO:'604 5 giy c mTN8-19-33 PESQGHCTRWPW MCPPQGWG CCGGAATCCCAGGGTCACTGCACC CGTTGGCCGTGGATGTGCCCGCCG CAGGGTTGGGGT (SEQ ID NO: 605) 5 giy c 94 WO 2004/058988 PCT/US2003/040781 mTN8-19-34- IPAHGHCTRWPW MCPPQRWR ATCCCGGCTCACGGTCACTGCACC CGTTGGCCGTGGATGTGCCCGCCG CAGCGTTGGCGT (SEQ ED NO: 606) 5gty c mTN8-19-35 FTVHGHCTRWP WMCPPYGWV TTCACCGTTCACGGTCACTGCACCC GTTGGCCGTGGATGTGCCCGCCGT ACGGTTGGGTT (SEQ ID NO: 607) 5 giy c mTN8-19-36 PDFPGHCTRWRW MCPPQGWE CCAGATTTTCCAGGTCATTGTACTC GTTGGCGTTGGATGTGTCCACCAC AAGGTTGGGAA (SEQ ID NO: 608) 5 giy c mTN8-19-37 QLWQGPCTQWP WMCPPKGRY CAGCTGTGGCAGGGTCCGTGCACC CAGTGGCCGTGGATGTGCCCGCCG AAAGGTCGTTAC (SEQ ID NO: 609) 5 giy c mTN8-19-38 HANDGHCTRWQ WMCPPQWGG CACGCTAACGACGGTCACTGCACC CGTTGGCAGTGGATGTGCCCGCCG CAGTGGGGTGGT (SEQ ID NO: 610) 5 giy c mTN8-19-39 ETDHGLCTRWPW MCPPYGAR GAAACCGACCACGGTCTGTGCACC CGTTGGCCGTGGATGTGCCCGCCG TACGGTGCTCGT (SEQ ID NO: 611) 5 giy c mTN8-19-40 GTWQGLCTRWP WMCPPQGWQ GGTACCTGGCAGGGTCTGTGCACC CGTTGGCCGTGGATGTGCCCGCCG CAGGGTTGGCAG (SEQ ID NO: 612) 5 giy c mTN8-19 Coni VATQGQCTRWP WMCPPQGWG GTTGCTACCCAGGGTCAGTGCACC CGTTGGCCGTGGATGTGCCCGCCG CAGGGTTGGGGT (SEQ ID NO: 613) 5 giy c mTN8-19 Con2 VATQGQCTRWP WMCPPQRWG GTTGCTACCCAGGGTCAGTGCACC CGTTGGCCGTGGATGTGCCCGCCG CAGCGTTGGGGT (SEQ ID NO: 614) 5 giy c 2X mTN8-19-7 FC-5G-AQ-LADHGQCERWPW MCPPEGWELEGS GSATGGSGSTASS GSGSATGLADHG QCIRWPWMCPPE GWE-LE (SEQID NO: 615) CTTGCTGATCATGGTCAATGTATTC GTTGGCCATGGATGTGTCCACCAG AAGGTTGGGAACTCGAGGGTTCCG GTTCCGCTACCGGCGGCTCTGGCTC CACTGCTTCTTCCGGTTCCGGTTCT GCTACTGGTCTGGCTGACCACGGT CAGTGCATCCGTTGGCCGTGGATG TGCCCGCCGGAAGGTTGGGAACTG GAA (SEQ ID NO: 616) 1K c 2X mTN8-19-7 ST—GG de!2x LE FC-5G-AQ- LADHGQCIRWPW MCPPEGWEGSGS ATGGSGGGASSG SGSATGLADHGQ CIRWPWMCPPEG WE (SEQID NO: 617) CTTGCTGATCATGGTCAATGTATTC GTTGGCCATGGATGTGTCCACCAG AAGGTTGGGAAGGTTCCGGTTCCG CTACCGGCGGCTCTGGCGGTGGCG CTTCTTCCGGTTCCGGTTCTGCTAC TGGTCTGGCTGACCACGGTCAGTG CATCCGTTGGCCGTGGATGTGTCCA CCAGAAGGTTGGGAA (SEQ ID NO: 618) 1K c 95 WO 2004/058988 PCT/US2003/040781

2XmTN8-19-21 FC-5G-AQ-SEYQGLCTRWPW MCPPQGWXLEGS GSATGGSGSTASS GSGSATGSEYQG LCTRWPWMCPPQ GWK-LE (SEQ ID NO: 619) TCTGAATATCAAGGTCTTTGTACTC GTTGGCCATGGATGTGTCCACCAC AAGGTTGGAAACTCGAGGGTTCCG GTTCCGCTACCGGCGGCTCTGGCTC CACTGCTTCTTCCGGTTCCGGTTCT GCTACTGGTTCTGAGTATCAAGGC CTCTGTACTCGCTGGCCATGGATGT GTCCACCACAAGGCTGGAAGCTGG AA (SEQ ID NO: 620) IK C 2XmTN8-19-21 ST—GGdel2x LE FC-5G-AQ- SEYQGLCTRWPW MCPPQGWKGSGS ATGGSGGGASSG SGSATGSEYQGL CTRWPWMCPPQ GWK (SEQ ID NO: 621) TCTGAATATCAAGGTCTTTGTACTC GTTGGCCATGGATGTGTCCACCAC AAGGTTGGAAAGGTTCCGGTTCCG CTACCGGCGGCTCTGGCGGTGGCG CTTCTTCCGGTTCCGOTTCTGCTAC TGGTTCTGAGTATCAAGGCCTCTGT actcgctggccatggatgtgtcca CCACAAGGTTGGAAA (SEQ ID NO: 622) 1K C 2X mTN8-19-22 FC-5G-AQ-TFSQGHCTRWPW MCPPQGWGLEGS GSATGGSGSTASS GSGSATGTFSQG HCTRWPWMCPP QGWG-LE (SEQ ID NO: 623) ACTTTTTCTCAAGGTCATTGTACTC GTTGGCCATGGATGTGTCCACCAC AAGGTTGGGGTCTCGAGGGTTCCG GTTCCGCTACCGGCGGCTCTGGCTC CACTGCTTCTTCCGGTTCCGGTTCT GCTACTGGTACTTTTTCTCAAGGCC ATTGTACTCGCTGGCCATGGATGTG TCCACCACAAGGCTGGGGCCTGGA A (SEQ ID NO: 624) 1K C 2X ΠΪΓΝ8-19-32 l FC-5G-AQ- VADHGHCTRWP WMCPPQGWGLE GSGSATGGSGST ASSGSGSATGVA DHGHCTRWPWM CPPQGWG-LE (SEQ ID NO: 625) GTTGCTGATCATGGTCATTGTACTC GTTGGCCATGGATGTGTCCACCAC AAGGTTGGGGTCTCGAGGGTTCCG GTTCCGCAACCGGCGGCTCTGGCT CCACTGCTTCTTCCGGTTCCGGTTC TGCTACTGGTGTTGCTGACCACGGT CACTGCACCCGTTGGCCGTGGATG TGCCCGCCGCAGGGTTGGGGTCTG GAA (SEQ ED NO: 626) 1K C 2X mTN8-19-32 ST—GG deI2x LE FC-5G-AQ-VADHGHCTRWP WMCPPQGWGGS GSATGGSGGGAS SGSGSATGVADH GHCTRWPWVCPP QGWG (SEQ ID NO: 627) GTTGCTGATCATGGTCATTGTACTC GTTGGCCATGGATGTGTCCACCAC AAGGTTGGGGTGGTTCCGGTTCCG CTACCGGCGGCTCTGGCGGTGGTG CTTCTTCCGGTTCCGGTTCTGCTAC TGGTGTTGCTGACCACGGTCACTGC ACCCGTTGGCCGTGGGTGTGTCCA CCACAAGGTTGGGGT (SEQ ED NO: 628) 1K C 96 WO 2004/058988 PCT/US2003/040781

2X mTN8-19-33 FC-5G-AQ- PESQGHCTRWPW MCPPQGWGLEGS GSATGGSGSTASS GSGSATGPESQG HCTRWPWMCPP QGWGLE (SEQ ED NO: 629) CCAGAATCTCAAGGTCATTGTACTC GTTGGCCATGGATGTGTCCACCAC AAGGTTGGGGTCTCGAGGGTTCCG GTTCCGCTACCGGCGGCTCTGGCTC CACTGCTTCTTCCGGTTCCGGTTCT GCTACTGGTCCGGAATCCCAGGGT CACTGCACCCGTTGGCCGTGGATG TGCCCGCCGCAGGGTTGGGGTCTG GAA (SEQ ID NO: 630) 1K C 2XmTN8-19-33 ST—GGdel2x LE FC-5G-AQ-PESQGHCTRWPW MCPPQGWGGSGS ATGGSGGGASSG SGSATGPESQGH CTRWPWMCP PQGWG (SEQ ED NO: 631) CCAGAATCTCAAGGTCATTGTACTC GTTGGCCATGGATGTGTCCACCAC AAGGTTGGGGTGGTTCCGGTTCCG CTACCGGCGGCTCTGGCGGTGGTG CTTCTTCCGGTTCCGGTTCTGCTAC TGGTCCGGAATCCCAGGGTCACTG CACCCGTTGGCCGTGGATGTGTCC ACCACAAGGTTGGGGT (SEQ ID NO: 632) 1K C

Example 7

In vitro screening of affinity matured peptibodies 5 The following exemplary peptibodies were screened according to the protocols set forth above to obtain the following KD and IC50 values. Table VH shows the range of KD values for selected affinity matured peptibodies compared with 'the parent peptibodies, as determined by KinExA™ solution based assays or BIAcore® assays. These values demonstrate increased binding affinity of the affinity matured peptibodies for myostatin compared with the parent 10 peptibodies. Table VIH shows IC50 values for a number of affinity matured peptibodies. A range of values is given in this table.

TABLE VII

peptibodies Kn TN8-19 (parent) > 1 nM 2xmTN8-19 (parent) > 1 nM lx mTN8-19-7 10 pM 2x mTN8-19-7 12 pM lxmTN8-19-21 6pM 2xmTN8-19-21 6pM lxmTN8-19-32 9 pM lxmTN8-19-33 21 pM 2x mTNS-1.9-33 3 pM lx mTN8-19-22 4pM lx mTN8-19-conl 20 pM 15 97 WO 2004/058988 PCT/US2003/040781

TABLE VIII

Affinity Matured Peptibody IC50(nM) mTN8-19 Coni 1.0-4.4 mTN8-19-2 7.508-34.39 mTN8-19-4 16.74 mTN8-19-5 7.743-3.495 mTN8-19-6 17.26 mTN8-19-7 1.778 mTN8-19-9 22.96-18.77 mTN8-19-10 5.252 - 7.4 mTN8-19-ll 28.66 mTN8-19-12 980.4 mTN8-19-13 20.04 mTN8-19-14 4.065-6.556 mTN8-19-16 4.654 mTN8-19-21 2.767-3.602 mTN8-19-22 1.927-3.258 mTN8-19-23 6.584 mTN8-19-24 1.673-2.927 mTN8-19-27 4.837-4.925 mTNS-19-28 4.387 mTN8-19-29 6.358 mTN8-19-32 1.842-3.348 mTN8-19-33 2.146-2.745 mTN8-19-34 5.028-5.069 mTN8Con6-3 86.81 mTN8Con6-5 2385 mTN8-19-7(-LE) 1.75-2.677 mTN8-19-21(-LE) 2.49 mTN8-19-33(-LE) 1.808 2xmTNS-19-7 0.8572-2.649 2xmTN8-19-9 1.316-1.228 2xmTN8-19-14 1.18-1.322 2xmTN8-19-16 0.9903 -1.451 2xmTN8-19-21 0.828-1.434 2xmTN8-19-22 0.9937-1.22 2xmTN8-19-27 1.601-3.931 2xmTN8-19-7(-LE) 1.077-1.219 2xmTN8-19-21(-LE) 0.8827-1.254 2xmTN8-19-33(-LE) 1.12-1.033 mL2-7 90.24 mL2-9 105.5 raL15-7 32.75 mL15-9 354.2 mL20-2 122.6 mL20-3 157.9 mL20-4 160 98 WO 2004/058988 PCT/US2003/040781

Example 8

In vivo Anabolic Activity of Exemplary Peptibodies

The CD1 nu/nu mouse model (Charles River Laboratories, Massachusettes) was used to 5 determine the in vivo efficacy of the peptibodies of the present invention which included the human Fc region (huFc). This model responded to the inhibitors of the present invention with a rapid anabolic response which was associated with increased dry muscle mass and an increase in myofibrillar proteins but was not associated with accumulation in body water content.

In one example, the efficacy of lx peptibody mTN8-19-21 in vivo was demonstrated by 10 the following experiment. A group of 10 8 week old CD1 nu/nu mice were treated twice weekly or once weekly with dosages of lmg/kg, 3 mg/kg and 10 mg/kg (subcutaneous injection). The control group of 10 8 week old CD1 nu/nu mice received a twice weekly (subcutaneous) injection of huFc (vehicle) at 10 mg/kg. The animals were weighed every other day and lean body mass determined by NMR on day 0 and day 13. The animals are then sacrified at day 14 15 and the size of the gastrocnemius muscle determined. The results are shown in Figures 2 and 3. Figure 2 shows the increase in total body weight of the mice over 14 days for the various dosages of peptibody compared with the control. As can be seen from Figure 2 all of the dosages have show an increase in body weight compared with the control, with all of the dosages showing statistically significant increases over the control by day 14. Figure 3 shows the change in lean 20 body mass on day 0 and day 13 as determined by NMR, as well as the change in weight of the gastrocnemius muscle dissected from the animals at day 14.

In another example, the lx mTN8-19-32 peptibody was administered to CD1 nu/nu mice in a biweekly injection of 1 mg/kg, 3 mg/kg, 10 mg/kg, and 30 mg/kg compared with the huFc control (vehicle). The peptibody- treated animals show an increase in total body weight (not 25 shown) as well as lean body mass on day 13 compared with day 0 as determined by NMR measurement. The increase in lean body mass is shown in Figure 4.

In another example, a lx affinity-matured peptibody was compared with a 2x affinity-matured peptibody for in vivo anabolic efficacy. CD1 nu/nu male mice (10 animals per group) were treated with twice weekly injections of 1 mg/kg and 3 mg/kg of lx mTN8-19-7 and 2x 30 mTN8-19-7 for 35 days, while the control group (10 animals) received twice weekly injections of huFc (3 mg/kg). As shown in Figure 5, treatment with the 2x peptibody resulted in a greater body weight gain and leans carcass weight at necropsy compared with the lx peptibody or control. 35 99 WO 2004/058988 PCT/US2003/040781

Example 9

Increase in muscular strength

Normal age-matched male 4 month old male C57B/6 mice were treated for 30 days with 2 injections per week subcutaneous injections 5 mg/kg per week of 2x mTN8-19-33,2x mTN8-19- 5 7, and huFc vehicle control group (10 animals/group). The animals were allowed to recover without any further injections. Gripping strength was measured on day 18 of the recovery period. Griping strength was measured using a Columbia Instruments meter, model 1027 dsm (Columbus, Ohio). Peptibody treatment resulted in significant increase in gripping strength, with 2x mTN8-19-33 pretreated animals showing a 14 % increase in gripping strength compared with the control- 10 treated mice, while 2x mTN8-19-7 showed a 15% increase in gripping strength compared with the control treated mice.

Example 10 Pharmacokinetics 15 In vivo phamacokinetics experiments were performed using representative peptibodies without the LE sequences. 10 mg/kg and 5mg/kg dosages were administered to CD1 nu/nu mice and the following parameters determined: Cmax (ug/mL), area under the curve (AUC) (ug-hr/mL), and half-life (hr). It was found that the 2x versions of the affinity matured peptibodies have a significantly longer half-life than the lx versions. For example lx affinity matured mTN8- 20 19-22 has a half-life in the animals of about 50.2 hours, whereas 2x mTN8-19-22 has a half- life of about 85.2 hours. Affinity matured lx mTN8-7 has a half-life of about 65 hours, whereas 2x mTN8-19-7 has a half-life of about'106 hours.

Example 11 25 Treatment of mdx Mice

The peptibodies of the present invention have been shown to increase lean muscle mass in an animal and are useful for the treatment of a variety of disorders which involve muscle wasting. Muscular dystrophy is one of those disorders. The mouse model for Duchenne’s muscular dystrophy is the Duchenne mdx mouse (Jackson Laboratories, Bar Harbor, Maine). Aged (10 30 month old) mdx mice were injected with either the peptibody lx mTN8-19-33 (n=8/group) or with the vehicle huFc protein (N=6/group) for a three month period of time. The dosing schedule was every other day, 10 mg/kg, by subcutaneous injection. The peptibody treatment had a positive effect on increasing and maintaining body mass for the aged mdx mice. Significant increases in body weight were observed in the peptibody-treated group compared to the hu-Fc-treated control 35 group, as shown in Figure 6A. In addition, NMR analysis revealed that the lean body mass to fat 100 WO 2004/058988 PCT/US2003/040781 mass ratio was also significantly increased in the aged mdx mice as a result of the peptibody treatment compared with the control group, and that the fat percentage of body weight decreased in the peptibody treated mice compared with the control group, as shown in Figure 6B. 5

The present invention is not to be limited in scope by the specific embodiments described herein, which are intended as single illustrations of individual aspects of the invention, and functionally equivalent methods and components are invention. Indeed, various modifications of the invention, in addition to those shown and described herein will become apparent to those 10 skilled in the art from the foregoing description and accompanying drawings. Such modifications are intended to fall within the scope of the appended claims. 101