WO2007014167A2 - Compositions for and methods of treating epithelial diseases with growth factors - Google Patents

Abstract

The invention relates to compositions and methods for treating epithelial diseases such as mucositis and inflammatory bowel disease in a subject. The invention also relates to long-acting therapeutic agents capable of promoting survival of an oral keratinocyte or an intestinal epithelial cell.

Description

COMPOSITIONS FOR AND METHODS OF TREATING EPITHELIAL DISEASES WITH GROWTH FACTORS

CROSS-REFERENCE TO RELATED APPLICATIONS

[001] This application claims the benefit of U.S. Provisional Application No. 60/702,065, filed July 22, 2005 in the United States Patent and Trademark Office, the disclosure of which is herein incorporated by reference in its entirety.

FIELD OF THE INVENTION

[002] The invention relates to compositions for and methods of treating epithelial diseases, such as mucositis and inflammatory bowel disease, using growth factors.

BACKGROUND OF THE INVENTION

[003] The mucosa is a mucus-secreting membrane lining all body cavities or passages that communicate with the exterior, and includes oral mucosa, gastric mucosa and intestinal mucosa. The gastric and intestinal mucosa are present in the gastrointestinal tract and comprise cells at the surface that serve various functions, including secretion and absorption. The mucosa is a very sensitive tissue, the ulceration of which characterizes several diseases and disorders.

[004] One such ulcerative disease is mucositis. During and following chemotherapy and radiation therapy for treatment of cancer, patients often develop mucositis, an acute reaction causing ulcerations in the mucosa of the mouth and the gastrointestinal tract. Because of these lesions, mucositis typically causes decreased food intake and diarrhea. These adverse effects on the quality of life often result in dose limitation or disruption of radiation treatment or chemotherapy, which leads to suboptimal therapy for the patient.

[005] Currently, the best known biologic for treatment of mucositis is keratinocyte growth factor (KGF, also known as Kepivance®, or palifermin, marketed by Amgen), which must be administered intravenously. KGF reduces oral mucositis, but its efficacy in mucositis of the small and large intestine is not yet clear.

[006] Inflammatory bowel disease (IBD) is another category of diseases characterized by ulcerations of the mucosa. IBD refers to disorders in which one or more portions of the gastrointestinal tract are inflamed. The two most common types of IBD are ulcerative colitis and Crohn's disease. Ulcerative colitis involves ulceration of the lining of the large intestine. Crohn's disease, on the other hand, may cause inflammation anywhere throughout the gastrointestinal tract and usually involves the entire intestinal wall.

[007] Current treatment regimens for both ulcerative colitis and Crohn's disease include anti-inflammatory drugs such as topical and systemic corticosteroids, and immunosuppressive agents. These immunosuppressive agents produce undesirable side effects. The KGF palifermin must be given three days prior to and then three days after the start of radiation or chemotherapy. When this drug is given only at the time of radiation or chemotherapy and thereafter, the patient fares worse. Thus, it has significant limitations.

SUMMARY OF THE INVENTION

[008] The present invention provides compositions and methods that can be used in treatment of epithelial disorders. Subjects that would benefit from the stimulation or support of epithelial cell survival, cell proliferation, or both include those suffering from mucositis, inflammatory bowel disease, wound healing disorders, and other illnesses that involve mucosal epithelium.

[009] The present invention is directed to compositions comprising an effective amount of a first therapeutic agent for treatment of a mucosal disorder in a subject and a pharmaceutically acceptable carrier, wherein the therapeutic agent comprises at least a first growth factor or a variant or an active fragment thereof, wherein the growth factor is other than keratinocyte growth factor (KGF) alone or hepatocyte growth factor alone.

[010] In some embodiments, the growth factor comprises insulin-like growth factor-I (IGF-I). Non-limiting examples of other growth factors include (i) members of the ErbB ligand family such as betacellulin (BTC), heregulinl-beta (HRGl -beta), TGF- alpha, HB-EGF, epiregulin, EGF, amphiregulin (AR), heregulinl -alpha (HRGl-alpha); (ii) hepatocyte growth factor (HGF); (iii) other insulin-like growth factors (IGFs); and keratinocyte growth factors (e.g., KGFl, KGF2). In addition, the invention also provides for variants and fragments of these growth factors.

[011] In some embodiments, the invention also provides for long-acting therapeutic agents that incorporate one or more growth factors such as, for example, one or more ErbB ligands (including variants and fragments thereof), or one or more of IGF-I, HGF, and KGFs, including variants and fragments or any of these. The long-acting therapeutic agent comprising a growth factor or a variant or an active fragment thereof and a fusion partner, is such that (i) the growth factor or variant or fragment promotes survival and/or proliferation of a mucosal cell and (ii) the second molecule (i.e., the fusion partner) confers an extended half life to the growth factor, when the long-acting agent is administered to a subject.

[012] In some embodiments, and in addition to the growth factor (which may or may not be a long-acting growth factor), the compositions can further comprise a second therapeutic agent. In one embodiment, the second therapeutic agent is a second growth factor. In other embodiments, the second therapeutic agent is another biologic or small molecule.

[013] In some embodiments, the growth factor, or the long-acting therapeutic agent comprising a growth factor or a variant or an active fragment thereof and a fusion partner, is such that (i) the growth factor or variant or fragment promotes survival and/or proliferation of a mucosal cell and (ii) the second molecule (i.e., the fusion partner) confers an extended half life to the growth factor in a subject.

[014] The invention provides that the long-acting therapeutic agent is such that the second molecule (i.e., the fusion partner) can comprise a polymer, a polypeptide, a succinyl group, a lipid group, serum albumin, or any combination of these.

[015] In some embodiments, the long-acting therapeutic agent is such that the polymer comprises either a polyethylene glycol moiety (PEG), or a polypeptide that comprises at least a portion of an Fc molecule (or one of its variants), or a combination of both.

[016] In addition, the invention also provides methods of treating an epithelial disease in a subject by (i) providing one of the compositions of the invention, comprising a therapeutic agent such as, for example, a growth factor or a long-acting agent selected from an ErbB ligand, IGF-I, KGFs (e.g., KGFl, KGF2), HGF, and (ii) administering an effective amount of the therapeutic agent to the subject.

[017] In some embodiments, the method of treatment further comprises administering chemotherapy, radiation therapy or another biologic or small molecule to the subject before, concurrently or after the administration of the growth factor composition. [018] The method of treatment can be used to treat a subj ect suffering from epithelial disease comprising mucositis, inflammatory bowel disease, or any combination of epithelial diseases. In some embodiments, the method of treatment is used to treat oral mucositis, hi other embodiments, the method of treatment is used to treat intestinal mucositis. Non-limiting examples of the inflammatory bowel diseases that can be treated with this method include Crohn's disease, ulcerative colitis, and the like.

[019] The invention provides embodiments wherein the compositions further comprise a biologic or small molecule such as an TNF-alpha inhibitor, or any other combination of anti-inflammatory agents and/or immunomodulatory agents.

[020] In some embodiments, the method of treatment uses a composition comprising a growth factor and/or long-acting therapeutic agent that can be administered by one or more routes. Non-limiting examples of routes of administration for the compositions of the invention, which can be used for growth factors (and variants and/or fragments thereof), long-acting therapeutic agents and other biologic or small molecule, are intravenous, oral, subcutaneous, and intramuscular administration, as well as transdermal, buccal, and intranasal administration, by inhalation, by suppository or by implantation. In some embodiments, any combination of one or more of these routes for administration of one or more of the compositions of the invention can be used, hi some embodiments, the compositions of the invention can be administered via oral adminstration as a mouth wash or as an oral gel, and rectal administration as an enema.

[021] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.

BRIEF DESCRIPTION OF THE FIGURES

[022] FIG. 1 shows the amino acid alignment of betacellulin 22218788_33871113, betacellulin NP_001720_NM_001729, and betacellulin 15079597_l 5079596. The alignment was performed by CLUSTAL FORMAT for T- COFFEE Version_1.37, CPU=0.00 sec, SCORE=76, Nseq=3, Len=178.

[023] FIG. 2 shows the amino acid alignment of betacellulin

CLN00902377_expressed_Met (mature human betacellulin, corresponding to residues 32- 111, preceded by a Met residue); betacellulin NP_001720_NM_001729; SEQ. ID NOS. 3, 14, 17, and 18 from US Patent No. 5,886,141; and SEQ ID NOS. 1 and 2 from US Patent No. 6,232,288. The alignment was performed by the freeware CLUSTAL FORMAT for T-COFFEE Version_1.37, CPU=0.00 sec, SCORE=75, Nseq=8, Len=178.

[024] FIG. 3 shows the results of a high-throughput phospho-AKT assay on HT- 29 cells in which a splice variant of human neuregulin-1 (FPT NRGl -beta 3 SV, herein referred to as NRGl-β3sv87) was identified, as further described in Examples 1 and 2.

[025] FIG. 4 shows a profile of the effect of human betacellulin clone 736345 in assays for various biological activities, as further described in Example 3. The assays shown include two that measure the phosphorylation of protein kinase AKT in HT-29 cells (HT-29pAKTl) and in RIE (RIEpAKTl) cells. Other assays tested using betacellulin include assays that measure non-activated B cell proliferation (BPro4), ability to stimulate glucose uptake by adipocytes (Gu2Gy3T3), unactivated monocyte proliferation (MonPro4), NK cell proliferation and/or survival (NKGIo), T-cell proliferation (TPro4), activated primary B-cell proliferation (aBPro4), activated primary monocyte proliferation (aMonPro3), and activated primary T-cell proliferation (aTPro4).

[026] FIG. 5 shows a profile of the effect of human NRGl-β3sv87 (CLN00541754) in assays for various biological activities, as further described in Example 3.

[027] FIG. 6 shows the results of a betacellulin activation assay in which the levels of phospho-AKT in human adenocarcinoma (HT-29) cells after treatment with tagged and untagged betacellulin were measured by an ELISA as a function of absorbance. The betacellulin supernatant, or conditioned medium harvested from cells transfected with a betacellulin cDNA of our library, was used either neat (dilution factor 1), or diluted to 0.2, 0.02, 0.004 of the initial concentration. Control supernant was harvested from cells that were not transfected with any betacellulin cDNA.

[028] FIG. 7 shows a comparison of AKT phosphorylation inducing-activities of various growth factors, namely: (i) conditioned media harvested from 293 cells expressing clonal NRGl-β3sv87 (clone 00891196), (ii) HRG-βl, (iii) HGF, and (iv) IGF-I. The left panel shows results of pAkt assays done in HT-29 cells. The right panel shows results of pAlct assays done in RIE cells. Assays were done in duplicate with clonal supernatants (clone 00891196) harvested from 293 cells. Background is indicated by the horizontal solid line. [029] FIG. 8 shows the results of a test of the activation of phospho-AKT in rat intestinal epithelium (RIE) cells by various concentrations of recombinant hepatocyte growth factor alone (HGF) and heregulin B alone, as further described in Example 5.

[030] FIG. 9 illustrates the results of a test of the activation of phospho-AKT in HT-29 cells by recombinant HGF and recombinant IGF-I, purchased from R&D Systems (Minneapolis, MN), as further described in Example 6. IGF-I and HGF are capable of promoting the phosphorylation of AKT in a dose-dependent manner.

[031] FIG. 10 illustrates the effect of various recombinant factors on the phosphorylation of HT-29 cells, as further described in Example 7. The factors tested were: heparin-binding EGF (HB-EGF), betacellulin (BTC), epiregulin (EPR), transforming growth factor-alpha (TGF-alpha), epidermal growth factor (EGF), hepatocyte growth factor (HGF), amphiregulin (AR), heregulin A, heregulin B, and insulin-like growth factor- 1 (IGF-I).

[032] FIG. 1 IA and FIG. 1 IB illustrate the effect of 1 nM and 100 nM, respectively, of various recombinant factors on the phosphorylation of HT-29 cells, as further described in Example 7. The factors tested were: heparin-binding EGF (HB-EGF), betacellulin (BTC), epiregulin (EPR), transforming growth factor-alpha (TGF-alpha), epidermal growth factor (EGF), hepatocyte growth factor (HGF), amphiregulin (AR), heregulin A, heregulin B, and insulin-like growth factor-1 (IGF-I), as further described in Example 7.

[033] FIG. 12A and FIG. 12B show the results of two phospho-AKT cell assays (Experiment 1 and Experiment 2, respectively), done in rat intestinal epithelial cells (RIE) as further described in Example 8. The assays tested the same factors as in Example 7 for their effect on the phosphorylation of AKT in RIE cells at various different doses.

[034] FIG.13 A and FIG.13B illustrates the effect of 1 nM and 100 nM, respectively, of various recombinant factors on the phosphorylation of RIE cells, as further described in Example 8. The factors tested were: heparin-binding EGF (HB-EGF), betacellulin (BTC), epiregulin (EPR), transforming growth factor-alpha (TGF-alpha), epidermal growth factor (EGF), hepatocyte growth factor (HGF), amphiregulin (AR), heregulin A, heregulin B, and insulin-like growth factor-1 (IGF-I), as further described in Example 7. [035] FIG. 14 illustrates the results of a test of the activation of phospho-AKT in human oral keratinocytes in response to recombinant keratinocyte growth factor (KGF), HGF, BTC, HRG-b, and IGF-I, as further described in Example 9.

[036] FIG. 15 shows the results of a test of the change in cell proliferation of human oral keratinocytes, measured in relative light units (RLU), in the presence of recombinant KGF, HGF, BTC, HRG-b, and IGF-I, as further described in Example 10.

[037] FIG. 16 shows the results of a Western blot-based analysis of betacellulin in the plasma at 2 min, 30 min, 2 hr, and 18 hr after injection of betacellulin-Fc fusion protein (BTC-Fc), PEGylated betacellulin (PEG-BTC), and unmodified betacellulin (BTC), as further described in Examples 12 and 13. PEG-BTC and BTC-Fc were cleared from mouse plasma significantly slower than unmodified betacellulin.

DETAILED DESCRIPTION OF THE INVENTION

[038] Definitions

[039] Unless defined herein, terms used herein have their ordinary meanings, and can be further understood in the context of the specification.

[040] The terms "polypeptide" and "protein" are used interchangeably to refer to a polymer of amino acid residues, and are not limited to a minimum length. Thus, peptides, oligopeptides, dimers, multimers, and the like, whether produced biologically, recombinantly, or synthetically and whether composed of naturally occurring or non- naturally occurring amino acids, are included within the definition. Both full-length proteins and fragments thereof are encompassed by the definition. The terms also include co-translational (e.g., signal peptide cleavage) and post-translational modifications of the polypeptide, such as, for example, dissulfide-bond formation, glycosylation, acetylation, phosphorylation, proteolytic cleavage (e.g., cleavage by furins or metalloproteases), and the like. Furthermore, for purposes of the present invention, a "polypeptide" refers to a protein that includes modifications, such as deletions, additions, and substitutions (generally conservative in nature, as would be known to a person in the art), to the native sequence, as long as the protein maintains the desired activity. These modifications may be deliberate, as through site-directed mutagenesis, or may be accidental, such as through mutations of hosts that produce the proteins, or errors due to PCR amplification or other recombinant DNA methods. Recombinant, as used herein to describe a nucleic acid molecule, means a polynucleotide of genomic, cDNA, viral, semisynthetic, and/or synthetic origin, and which, by virtue of its origin or manipulation, is not associated with all or a portion of the polynucleotide with which it is associated in nature. The term recombinant as used with respect to a protein or polypeptide, means a polypeptide produced by expression of a recombinant polynucleotide. The term recombinant, when used with respect to a host cell, means a host cell into which a recombinant polynucleotide has been introduced.

[041] The terms "nucleic acid molecule," "nucleotide," "polynucleotide," and "nucleic acid" are used interchangeably herein to refer to polymeric forms of nucleotides of any length. They can include both double- and single-stranded sequences and include, but are not limited to, cDNA from viral, prokaryotic, and eukaryotic sources; mRNA; genomic DNA sequences from viral (e.g. DNA viruses and retroviruses) or prokaryotic sources; RNAi; cRNA; antisense molecules; ribozymes; and synthetic DNA sequences. The term also captures sequences that include any of the known base analogs of DNA and RNA.

[042] A "growth factor" is a protein that binds receptors located, for example, on the surface of a cell or in intracellular vesicles, and subsequently activates cellular proliferation and/or differentiation. Many growth factors are quite versatile and can act to stimulate cellular division in a wide variety of cell types, while others are specific to a particular cell-type. The term "growth factors," as used herein, includes modified derivatives and peptide fragments thereof, and include, for example, fibroblast growth factors (FGFs), interleukins (IL) 1 through 12, keratinocyte growth factors (KGFl, KGF2), colony stimulating factors (CSFs), epidermal growth factor (EGF) and the ErbB ligand protein family, and insulin-like growth factors (IGFs).

[043] As used herein, an "ErbB ligand" refers to a molecule in which at least a portion of the molecule comprises an ErbB ligand (i.e., a member of the EGF-like family of proteins which bind one or more ErbB receptors) or a fragment thereof. Non-limiting examples of ErbB ligands are betacellulin (BTC), epidermal growth factor (EGF), Epigen, amphiregulin (AR), transforming growth factor alpha (TGF-α), heparin-binding EGF (HB- EGF), epiregulin (EPR), and any of the multiple neuregulin isoforms and splice variants (e.g., NRG-I, NRG-2, NRG-3, or NRG-4). A receptor is defined by the International Union of Pharmacology Committee on Receptor Nomenclature and Drug Classification (NC-IUPHAR) as a protein, or a complex of proteins, which recognizes physiologically relevant ligands that can regulate the protein to mediate cellular events. [044] A "ligand" is any molecule that binds to a specific site on another molecule, including but not limited to receptors. For example, a ligand may be an extracellular molecule that, upon binding to another molecule, usually initiates a cellular response, such as activation of a signal transduction pathway.

[045] A "fragment" is any portion or subset of the corresponding polypeptide or polynucleotide molecule. Thus, for example, a "fragment of albumin" refers to a polypeptide subset of albumin and a "fragment of Fc" refers to a polypeptide subset of an Fc molecule. The term "fragment" is not intended to limit the portion or subset to any minimum or maximum length.

[046] A "variant" of an ErbB ligand is meant to refer to a ligand substantially similar in structure and biological activity to either the native ErbB ligand or to a fragment thereof, but not identical to such molecule or fragment thereof. A variant is not necessarily derived from the native molecule and may be obtained from any of a variety of similar or different cell lines. The term "variant" is also intended to include genetic alleles, aptamers, and glycosylation variants. Thus, provided that two growth factors (e.g. ErbB ligands, IGF-I, HGF, KGFs) possess a similar structure and biological activity, they are considered variants as that term is used herein even if the composition or the secondary, tertiary, or quaternary structure of one of the ligands is not identical to that found in the other.

[047] "Long-acting" in relation to the growth factors of the invention (e.g. ErbB ligands, IGF-I, HGF, KGFs) refers to a growth factor with a pharmacokinetic half-life that is longer than the half-life of the corresponding growth factor alone. Similarly, the term "extended half-life" as used herein is a relative term that refers to a longer pharmacokinetic half-life in one form of a molecule relative to another form. The term "pharmacokinetic half-life" refers to the extent of time that it takes, after administration of the growth factor of interest, for the concentration of the growth factor to decrease to one half of its initial concentration (i.e., that reached upon administration) in the blood, plasma or other specified tissue. The term "first molecule" is used herein to refer to the existence of a plurality of molecules (i.e., a "second molecule" or a "third molecule") and is not intended to indicate, for example, the location or order of the molecule in relation to other molecules.

[048] A "fusion polypeptide" is one comprising amino acid sequences derived from two or more different polypeptides. For example, a "long-acting betacellulin fusion protein" is a fusion polypeptide comprising a betacellulin polypeptide, or an active variant or fragment thereof, and a fusion partner, or an active variant or fragment thereof. The fusion polypeptide hence comprises the protein of interest linked (e.g., recombinant^ or by synthetic methods) to a second polypeptide, termed a "fusion partner." Examples of commonly used fusion partners include, among others, albumin, Fc molecules, polypeptides comprising oligomerization domains, and various domains of the constant regions of the heavy or light chains of a mammalian immunoglobulin.

[049] The terms "albumin" and "albumin molecule" refer to any one of a group of proteins that are soluble in water and moderately concentrated salt solution, and that are coagulable on heating. Suitable albumins will be familiar to those skilled in the relevant art. In addition, these proteins may be modified by proteolysis, sequence modification using molecular biological methods, and by binding to lipids or carbohydrates.

[050] The term "Fc molecule" as used herein includes native and mutein forms of polypeptides derived from the Fc region of an antibody comprising any or all of the constant heavy (CH) domains of the Fc region. An antibody or an immunoglobulin is a protein that is capable of recognizing and binding to a specific antigen. Antibodies can generated by the immune system, synthetically, or recombinantly, and include polyclonal and monoclonal antibody preparations, as well as preparations including hybrid antibodies, altered antibodies, chimeric antibodies, hybrid antibody molecules, F(ab')2 and F(ab) fragments; Fv molecules (for example, noncovalent heterodimers), dimeric and trimeric antibody fragment constructs; minibodies, human antibodies, humanized antibody molecules, and any functional fragments obtained from such molecules, wherein such fragments retain specific binding. Antibodies are commonly known in the art. Antibodies may recognize, for example, polypeptide or polynucleotide antigens. The term includes active fragments, including for example, an antigen-binding fragment of an immunoglobulin, a variable and/or constant region of a heavy chain, a variable and/or constant region of a light chain, a complementarity-determining region (cdr), and a framework region. An antibody CH3 domain refers to the CH3 portion of an Fc molecule. Truncated forms of such polypeptides containing the hinge region that promotes dimerization are also included. As defined herein, an Fc molecule that is defective in effector function is one that does not induce antibody-dependent cell-mediated cytoxicity (ADCC). [051] The term "polymer" means any compound that is made up of two or more monomeric units covalently bonded to each other, where the monomeric units may be the same or different, such that the polymer may be a homopolymer or a heteropolymer. Representative polymers include peptides, polysaccharides, nucleic acids, and the like, where the polymers can be naturally occurring or synthetic.

[052] The term "succinyl group" as used herein refers to the acyl residue derived from succinic acid or (l,4-dioxobutyl)-l-carboxylic acid.

[053] The term "oligomerization domain" refers to a portion of a fusion partner at which the formation of an oligomer may occur; i.e., there is sufficient structure to allow oligomerization. The oligomers can be of any subunit stoichiometry, including, for example dimerization and tetramerization domains. The oligomerization domain may comprise a coiled-coil domain (such as a tetranectin coiled-coil domain, a coiled-coil domain in a cartilage oligomeric matrix protein, an angiopoietin coiled-coil domain, or a leucine zipper domain), a collagen or a collagen-like domain (such as collagen, mannose- binding lectin, lung surfactant protein A, lung surfactant protein D, adiponectin, ficolin, conglutinin, macrophage scavenger receptor, or emilin), or a dimeric immunoglobulin domain (such as an antibody CH3 domain).

[054] The term "small molecule" includes any chemical or other moiety, other than polypeptides and nucleic acids, that can act to affect biological processes. Small molecules can include any number of therapeutic agents presently known and used, or can be small molecules synthesized in a library of such molecules for the purpose of screening for biological function(s). hi certain embodiments, the biologic or small molecule may comprise tumor necrosis factor-alpha inhibitor, an anti-inflammatory agent, or an immunomodulatory agent. The term "anti-inflammatory agent" as used herein refers to compounds that counteract or suppress the inflammatory process. Examples of antiinflammatory agents include, without limitation, glucocorticoids, anti-rejection drugs such as cyclosporin, cytotoxic agents such as cyclophosphamide, anti-metabolites such as methotrexate and azathioprine, and TNF-alpha receptor and IL-I receptor antagonists. The term "immunomodulatory agent" refers to substances that act to modulate the immune system of the subject being treated herein. \

[055] A "biologic" is a product which is naturally produced in some form by living organisms, whether modified or unmodified, whether in whole or a fragment thereof. A biologic may be prepared from a living source, such as animal tissue. The term includes, but is not limited to, a polynucleotide, polypeptide, antibody, cell, virus, toxin, vaccine, blood component or derivative, and fusion protein. A "biologic" may be used to treat an animal, including a human.

[056] A "composition" or "pharmaceutical composition" herein refers to a composition that usually contains an excipient, such as a pharmaceutically acceptable carrier that is conventional in the art and that is suitable for administration into a subject for therapeutic, diagnostic, or prophylactic purposes. It can include a cell culture, in which the polypeptide or polynucleotide is present in the cells and/or in the culture medium. In addition, compositions for topical (e.g., oral mucosa, respiratory mucosa) and/or oral administration can form solutions, suspensions, tablets, pills, capsules, sustained-release formulations, oral rinses, or powders, as known in the art and described herein. The compositions also can include stabilizers and preservatives. For examples of carriers, stabilizers and adjuvants, University of the Sciences in Philadelphia (2005) Remington: The Science and Practice of Pharmacy with Facts and Comparisons, 21st ed.

[057] As used herein, the term "pharmaceutically acceptable carrier" encompasses any of the standard pharmaceutical carriers, such as a phosphate buffered saline solution, water, and emulsions, such as an oil/water or water/oil emulsion, and various types of wetting agents.

[058] A "disease" is a pathological condition, for example, one that can be identified by symptoms or other identifying factors as diverging from a healthy or a normal state. The term "disease" includes disorders, syndromes, conditions, and injuries. Diseases include, but are not limited to, proliferative, inflammatory, immune, metabolic, infectious, and ischemic diseases.

[059] Cellular "proliferation" is an increase in cell number via the growth and reproduction of similar cells.

[060] The term "promoting" as used herein refers to increasing or improving cell proliferation and/or survival. The term "survival" as used herein refers to the viability of the cells, such as intestinal epithelial cells and "oral keratinocytes," which are epidermal cells that secrete the protein keratin and are found in the mouth skin of a subject. An "intestinal epithelial cell" is an epithelial cell located in or obtained from the intestine of a subject, such as, for example, a human or a rat.

[061] The term "epithelial cell," as used herein, refers to a cell located in a cellular layer covering the free surface (cutaneous, mucous or serous) of an organ or lining a tube or cavity in an animal body, and is consistent with the art-recognized definition of epithelial cells in epithelium. See, for example, the definition in Taber's Encyclopedic Medical Dictionary, Edition 12, (1973) F. A. Davis Company, publisher.

[062] The term "epithelial disease" refers to both epidermal and mucosal diseases, and encompasses, but is not limited to, wounds, mucosal disorders, asthma, dermatitis, Barrett's esophagus, psoriasis, and the like.

[063] The term "mucosal disorders," as used herein, applies to disorders of the gastrointestinal tract mucosa, and includes, but is not limited to, mucositis, inflammatory bowel disease, ulcers, aphthous ulcers, aphthous stomatitis, Behcet's disease, celiac disease, Menetrier's disease, microvillous inclusion disease, sore mouth, blistering (vesiculobullous) disorders, red and white patches, burning mouth syndrome, lichen planus, pemphigoid, hand-foot syndrome, tufting enteropathy and chronic peptic ulcer disease, and the like.

[064] The term "inflammatory bowel disease" (IBD) refers to a general classification for a chronic and relapsing inflammation of the gastrointestinal tract. IBD is associated with substantial morbidity and mortality, and currently has no cure. The two most prominent forms of IBD are ulcerative colitis and Crohn's disease.

[065] The term "mucositis" refers to a gastrointestinal (GI) ulcerative disease that can benefit from treatments that promote epithelial wound healing.

[066] Ulcerative colitis (UC) typically involves only the mucosa of the colon. The mucosa, comprising the coat which lines the gastrointestinal (GI) tract and faces its lumen, actually is a multilayer structure comprising an epithelial cell layer, a lamina propria, and a muscularis mucosa. The intestinal epithelial cells are the cells at the innermost surface of the gut that serve various functions, including secretion and absorption. The lamina propria is a connective tissue layer housing the blood vessels, lymphatics, and nerves. And the muscularis mucosa is a thin smooth muscle layer below the lamina propria. The mucosa is a very sensitive tissue, the ulceration of which characterizes several disorders. UC includes colitis that begins in the rectum and involves the bowel contiguously, with ulcers and hemorrhaging predominating in the active phase.

[067] "Crohn's disease" or regional enteritis may involve both the small and large intestines. Furthermore, Crohn's disease involves all four coats of the gut namely the mucosa, submucosa, tunica muscularis and tunica serosa; thus the term "transmural colitis" has been used to describe this disease when present in the colon. [068] The terms "subject," "individual," "host," and "patient" are used interchangeably herein to refer to a living animal, including a human and a non-human animal. The subject may, for example, be an organism possessing immune cells capable of responding to antigenic stimulation, or possessing cells responding to stimulatory and inhibitory signal transduction through cell surface receptor binding. The subject can be a mammal, such as a human or a non-human mammal, for example, non-human primates, dogs, cats, pigs, cows, sheep, goats, horses, rats, and mice. The term "subject" does not preclude individuals that are entirely normal with respect to a disease, or normal in all respects.

[069] "Treatment" or "treating" as used herein, covers any administration or application of remedies for disease in a mammal, including a human, and includes inhibiting the disease. It also includes arresting disease development and relieving the disease, such as by causing regression or restoring or repairing a lost, missing, or defective function, or by stimulating an inefficient or absent process. A therapeutic agent is any agent used for treatment of a condition.

[070] The term "anti-inflammatory agent," as used herein, refers to compounds that counteract or suppress the inflammatory process. Examples of anti-inflammatory agents include, without limitation, glucocorticoids, anti-rejection drugs such as cyclosporin, cytotoxic agents such as cyclophosphamide, anti-metabolites such as methotrexate and azathioprine, and TNF-alpha receptor and interleukin-1 (IL-I) receptor antagonists.

[071] The term "immunomodulatory agent" refers to substances that act to modulate the immune system of the subject being treated herein. To modulate refers to the production, either directly or indirectly, of an increase or a decrease, a stimulation, inhibition, interference, or blockage in a measured activity when compared to a suitable control. A modulator of a polypeptide or polynucleotide, or an agent are terms used interchangeably herein to refer to a substance that affects, for example, increases, decreases, stimulates, inhibits, interferes with, or blocks, a measured activity of the polypeptide or polynucleotide, when compared to a suitable control.

[072] One considers an amount of agent to be a therapeutically "effective amount" if that amount of agent will produce a desirable result upon administration to a subject; this amount will vary depending on various factors, such as the dosage to be administered and the route of administration. [073] The term "chemotherapy" refers to the administration of chemical compounds or drugs that are used in the treatment of cancer, for example, to kill cancer cells and/or lessen the spread of the disease (i.e., metastasis), and it does not exclude photodynamic therapy. "Radiation therapy" is a term commonly used in the art to refer to multiple types of radiation therapy, including internal and external radiation therapy, radioimmunotherapy, and the use of various types of radiation including X-rays, gamma rays, alpha particles, beta particles, photons, electrons, neutrons, radioisotopes, and other forms of ionizing radiation. As used herein, the term "radiation therapy" is inclusive of all of these types of radiation therapy, unless otherwise specified.

[074] The description herein is put forth to provide those of ordinary skill in the art with a detailed description of how to make and how to use the present invention, and is not intended to limit the scope of what the inventors regard as their invention, nor is it intended to represent that the experiments set forth are all or the only experiments performed.

[075] While the present invention is described with reference to the specific embodiments thereof, it should be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the true spirit and scope of the invention. In addition, many modifications can be made to adapt to a particular situation, material, composition of matter, process, process step or steps, to the objective, spirit, and scope of the present invention. AU such modifications are intended to be within the scope of the claims appended hereto.

[076] Unless defined otherwise, the meanings of all technical and scientific terms used herein are those commonly understood by one of ordinary skill in the art to which this invention belongs.

[077] With respect to ranges of values, the invention encompasses each intervening value between the upper and lower limits of the range to at least a tenth of the lower limit's unit, unless the context clearly indicates otherwise. Further, the invention encompasses any other stated intervening values. Moreover, the invention also encompasses ranges including either or both of the upper and lower limits of the range, unless specifically excluded from the stated range.

[078] It must be noted that, as used herein and in the appended claims, the singular forms "a," "or," and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to "a subject polypeptide" includes a plurality of such polypeptides and reference to "the agent" includes reference to one or more agents and equivalents thereof known to those skilled in the art, and so forth.

[079] Further, all numbers expressing quantities of ingredients, reaction conditions, % purity, polypeptide and polynucleotide lengths, and so forth, used in the specification, are modified by the term "about," unless otherwise indicated. Accordingly, the numerical parameters set forth in the specification and claims are approximations that may vary depending upon the desired properties of the present invention. At the very least, and not as an attempt to limit the application of the doctrine of equivalents, each numerical parameter should at least be construed in light of the number of reported significant digits, applying ordinary rounding techniques. Nonetheless, the numerical values set forth in the specific examples are reported as precisely as possible. Any numerical value, however, inherently contains certain errors from the standard deviation of its experimental measurement.

[080] The specification is most thoroughly understood in light of the cited references, all of which are hereby incorporated by reference in their entireties.

High-Throughput Screening for Epithelial Cell Protecting Agents

[081] Methods are provided for screening for secreted factors that may be able to confer epithelial cell protection through either accelerated cell proliferation or by protecting intestinal crypt cells and oral keratinocytes from death. The method may screen for secreted factors that increase the amounts of phosphorylated serine-threonine protein kinases AKT (also known as the protein kinase B family) or extracellular-signal-regulated kinases (also known as the ERK family). Both families of proteins have been implicated in a number of cellular functions, including cell survival and cell proliferation. The method may use any epithelial cell culture such as, for example, rat intestine epithelium (RIE) cells, human adenocarcinoma (HT-29) cells, and human oral keratinocytes (HOK).

[082] An embodiment of the provided screening method can use ELISA to detect the phosphorylation of AKTl (a measure of AKT activation) in 96-well format, such as the PathScan® Phospho-AKTl (Ser473) Sandwich ELISA Kit (Cell Signaling Technology®, MA). After incubating the epithelial cells with secreted protein, cell lysing buffer can be added, the lysed mixture can be transferred to a Phospho-AKT antibody- coated plate, and the phosphorylated AKT protein can be captured by the coating antibody. Following extensive washing, an anti-AKTl monoclonal antibody can be added to detect the captured phospho-AKTl protein. An horseradish peroxidase (HRP)-linked anti-mouse antibody can then be used to recognize the bound detection antibody. An HRP substrate, for example, 3,3',5,5' tetramethylbenzidine (TMB) can be added to develop color. The magnitude of optical density for the developed color is proportional to the quantity of phosphorylated protein. By varying the antibody used to capture the phosphorylated protein, similar assays are done to assess the phosphorylation of additional AKTs and ERK family. For example, the phosphorylation of ERK in response to the growth factors of the invention was also measured using a Phospho-ERK ELISA kit from Cell Signaling Technologies, Catalog #7315). Thus, methods are provided for screening secreted factors for those that increase the levels of phosphorylated survival and/or proliferation supporting mediators of those factors in epithelial cells .

[083] Using these methods, the inventors determined that some compounds found to increase the levels of phosphorylated AKT or ERK under these conditions are several ErbB ligand proteins, such as betacellulin (Table 5), a neuregulin-1 splice variant (herein named NRGl-β3sv87), heregulin B, as well as hepatocyte growth factor (Table 6). As such, the present invention further relates to methods for treating epithelial cell damage of the mucosa using growth factors that promote cellular survival and/or mucosal cellular proliferation, such as betacellulin, heregulin B, a neuregulin-1 splice variant (e.g. NRGl- β3sv87), insulin-like growth factor, and hepatocyte growth factor.

The ErbB Ligand Family of Proteins

[084] The family of ligands for the ErbB receptors (herein referred to as the "ErbB ligand family," and its members as ErbB ligands) is named after the cellular homologue of the viral erb gene, which in turn is one of three first RNAs of seven replication-defective leukaemia virus (DLV) strains originally identified as having the capacity to transform erythroblasts (hence the name erb) (Roussel, M. et al., Nature, 281 : 452-5 (1979)).

[085] Epidermal growth factor (EGF) is the prototype member of the ErbB ligand family. EGF binds the human EGF-Receptor 1 (HERl /ErbB 1 /EGFR) tyrosine kinase. Three other mammalian genes encoding receptors structurally similar to HERl (ErbBl) have been identified and named HER2 (or ErbB2), HER3 (or ErbB3), and HER4 (or ErbB4). EGF is the prototype member of a growing family of protein ligands, each of which binds a different set of receptors of the HER family; what these ligands have in common is the EGF domain, a consensus sequence consisting of six spatially conserved cysteine (C) residues (CX7 CX4-5 CXlO-13 CXCX8 C) that form three intramolecular dissulfide bonds (Cl to C3, C2 to C4, and C5 to C6). EGF contains six copies of the EGF domain. The other EGF family members contain only one, and one EGF domain is both necessary and sufficient for binding to and activation of a HER. In addition to their ability to promote wound-healing, human genetic studies and targeted mutations in animal models indicate that EGF/HER family members are key players in multiple other biological processes. For example, EGFs dictate both neuronal and epithelial lineage differentiation during embryogenesis and some variants reportedly associate with schizophrenia, whereas sustained and inappropriate self-activation of HERs reportedly mediates signaling pathways that promote both epithelial cell survival and growth as well as angiogenesis in a significant proportion of lung and breast tumors.

[086] Currently, the mammalian EGF family of ligands includes three groups of proteins. Group 1 members are capable of activating cells singly expressing HERl/ErbBl. These are: EGF (Savage et al, J. Biol. Chem. 247 : 7612-7621 (1972)), transforming growth factor-a (TGF-alpha) (Marquardt et al., Science, 223 : 1079-1082 (1984)), epigen (Strachan L. et al. J Biol Chem. 276:18265-18271 (2001)), and amphiregulin (Shoyab et al., Science, 243 : 1074-1076 (1989)). Group 2 members can activate cells singly expressing either HERl/ErbBl or HER4/ErbB4, and includes heparin- binding EGF-like growth factor (HB-EGF) (Higashiyama et al., Science, 251 : 936-939 (1991)), epiregulin (Toyoda et al., J. Biol. Chem., 270 : 7495-7500 (1995)), and betacellulin (BTC) (Shing et al., Science, 259 : 1604-1607 (1993)). Group 3 is the largest, and its members are capable of activating cells singly expressing either the HER3/ErbB3 or the HER4/ErbB4 receptor; this group includes the neuregulin (NRG) subfamily, which in humans is the product of four genes: NRGl (Marchionni et al., Nature, 362 : 312-318 (1993), NRG2 (Higashiyama et al., J Biochem 122(3):675-80 (1997); Chang et al., Nature, 387: 509-512 (1997); Carraway et al., Nature, 387 : 512-515 (1997)), NRG3 (Zhang et al., Proc. Natl. Acad. Sci. (USA), 94 :9562-9567(1997)), and NRG4 (Harari et al., Oncogene, 18 : 2681-2689 (1999)). No direct HER2/ErbB2 ligand has been identified to date, although HER2/ErbB2 reportedly is indirectly activated by NRGs and BTC (Harris, CR. et al., Exp. Cell Res. 284:2-13 (2003)) upon heterodimerization with other HER family members. [087] The remarkable growth of the mammalian EGF family has been primarily attributed to the continuing discovery of novel exons, as well as alternative promoters and transcription initiation sites, as part of the gene encoding NRGl. At least 16 NRGl splice variants have been identified to date. The human NRGl gene is also unusually long. It is in the order of 1.4 megabases long, comprising approximately l/2000th of the genome, but less than 0.3% of this span encodes a protein.

[088] ErbB ligand family proteins regulate the proliferation and differentiation of many tissue types. Overexpression of ErbB receptors has been identified in a wide variety of tumors including breast, colorectal, ovarian, and non-small cell lung cancers. Most studies on the ErbB ligand family are focused on proliferation and differentiation.

[089] Betacellulin

[090] As noted above, betacellulin increased the level of AKT phosphorylation in the screening assay described above. Betacellulin — a Group 2 ErbB ligand family protein — is a type I membrane protein that is translated as a transmembrane precursor molecule and proteolytically cleaved to a mature extracellular soluble form. The protease ADAM 10 can effect betacellulin shedding to the soluble form (Sanderson MP et al, J. Biol. Chem. 280:1826(2005)). Betacellulin exists primarily as a monomer. The molecule folds into a configuration comprising an A loop, a B loop, and a C loop; the latter is involved in receptor binding. Soluble betacellulin comprises 80 amino acids. The human betacellulin gene is located on chromosome 4 at band 4ql3-q21.

[091] Betacellulin contains one EGF-like domain, and its carboxyl terminal region has approximately 50% homology with transforming growth factor-alpha. Betacellulin has been reported to be a mitogen for retinal pigment epithelial cells and vascular smooth muscle cells (Shing et al., Science 259:1604 (1993)). Betacellulin acts on ErbB/HER receptors, though the exact receptors it may be working on in intestinal epithelial cells are unclear — perhaps ErbBl or ErbB4 (Slikowski MX et al., FEBS letters 447:227(1999)). A similar role has been noted for neuregulin-1, also called heregulin betal (Suarez et al., J. Biol. Chem. 18257-18264 (2001)). Betacellulin Expression and Purification

[092] The invention provides compositions comprising betacellulin, and methods of use for betacellulin. In one embodiment, the betacellulin is isolated human betacellulin, optionally an active fragment of human betacellulin, either modified or unmodified. The modification can include addition of an N-terminal methionine residue for facilitation of expression in a prokaryotic expression system such as in E. coli. One skilled in the art would be familiar with several methods for producing betacellulin, some of which are described in detail in, for example, the U.S. Patent Application No. 11/442,244, filed May 30, 2006; and in the PCT Application entitled "Methods of and Compositions for Stimulation of Glucose Uptake and Treatment of Diseases" filed May 30, 2006; both of which are incorporated herein by reference in their entirities. m one embodiment, recombinant betacellulin can be purified as described for rat betacellulin by Dunbar et al. at the Cooperative Research Centre for Tissue Growth and Repair, CSIRO Health Sciences and Nutrition, Adelaide, Australia (Dunbar, AJ. et al., J. MoI. Endo. 27:239-247 (2001)); and by Folkman and Shing in U. S. Patent No. 5,328,986 for example. Betacellulin can also be expressed in, and purified from, E. coli using a cleavable fusion protein strategy. Insoluble fusion protein can be collected as inclusion bodies and dissolved in urea under reducing conditions, re- folded, and purified by gel filtration chromatography and C4 RP-HPLC. Both full-length and a truncated fragment of betacellulin can be obtained by proteolytically cleaving the fusion protein with Factor Xa; the biologically active fragment can be separated from full-length betacellulin by heparin- affinity chromatography.

[093] In one embodiment, betacellulin can also be expressed in mammalian cells (e.g. CHO cells, 293 cells, PerC6® cells (Crucell, Netherlands)), hi another embodiment, betacellulin can be isolated from mammalian tissues. It has been reported that betacellulin is synthesized by several tissue types, including pancreas, small intestine, kidney, and liver tissue, and tumor cell types, including a mouse beta tumor and the MCF-7 cell line (Sasada, R. et al., Biochem. Biophys. Res. Comm. 190:1173-1179 (1993)). High levels of expression have been observed in the pancreas and small intestine. Heregulin-Beta and the Neuregulin Family: Genes, Structure and Nomenclature

[094] The inventors have further discovered that another human polynucleotide, and the neuregulin polypeptide it encodes, both of which are described in the Tables and Sequence Listing herein, are also useful for production of therapeutic agents for treatment of epithelial diseases in mammals, such as in humans. Thus, the invention provides a protein, herein named NRGl-β3sv87 (clone 00891196 and clone 00541754), identified as a naturally occurring splice variant of human neuregulin-1, for treating epithelial diseases. In certain embodiments, the invention provides for the use of this protein (FIG. 3 and Tables 1 through 3) and nucleotide sequence from the encoding gene in compositions for and methods of treating epithelial diseases, such as inflammatory bowel disease and mucositis. The NRGl-β3sv87 of the invention has one EGF-domain and is predicted to have altered biological properties relative to known members of the neuregulinl family. For example, it has a shorter primary structure and is predicted to lack both N- and O- linked glycosylation, all three properties being advantageous at least for the purpose of making and delivering the protein therapeutically.

[095] The neuregulins are the largest subclass of ligands for the HER/ErbB receptors. The exact receptors they may be working through, or binding to, in the intestinal epithelial cells and human oral keratinocytes are unclear. The receptors may include HER3/ErbB3, HER4/ErbB4, ErbB2/ErbB3 heterodimers, and/or ErbB2/ErbB4 heterodimers (Sliwkowski et al., FEBS letters, 447:227 (1999)). Neuregulins are a family of structurally related glycoproteins that comprise products from four distinct, but related genes: NRGl, NRG2, NRG3, and NRG4. Through alternative splicing and/or the use of alternative promoters, the NRGl gene has been reported to encode more than 16 soluble or transmembrane proteins. The extracellular domain of the transmembrane NRGl isoforms can be proteolytically cleaved to release soluble growth factors. All NRGl isoforms contain an EGF domain that is required for their direct binding to the HER3/ErbB3 or HER4/ErbB4 receptor tyrosine kinases. The HER3/ErbB3 or HER4/ErbB4 subsequently recruits and heterodimerizes with HER2/ErbB2, resulting in tyrosine phosphorylation and NRGl signaling. Much less is known about the products of the NRG2, 3 and 4 genes than about the biological functions of the NRGl isoforms.

[096] The nomenclature of the NRGl family reflects the diversity of isoforms encoded by the human NRGl gene, which has been mapped to the short arm of chromosome 8. Furthermore, the fact that the same proteins were simultaneously identified by different groups on the basis of different bioassays also contributed to the multitude of names reportedly used to identify the same NRGl isoforms. Recently, however, NRGl isoforms have been classified on the basis of their domain structure into three major subtypes, and a consensus name for each protein has been reported.

[097] It was proposed by Meyer et al. (1997) that alternative splicing of the gene encoding murine NRGl results in transcripts encoding three major types of isoforms with distinct domain structures (types I through III). Although all contain an EGF domain, different isoforms vary in the particular combination of exons or "building blocks" encoding N-terminal domains, Ig-like regions, spacer segments for glycosylation, stalks placed between the EGF domain and the transmembrane domain(s), and cytoplasmic tails; each of these is encoded by different exons. Among these domains, three structural features are reported to be particularly important in differentiating NRGl isoforms in terms of their function and biological activities. These are: the type of EGF domain (α or β); the N-terminal sequence (type I, II or III); and whether the protein is initially synthesized as a transmembrane or non-transmembrane protein (Falls, D.L. et al., Exp. Cell. Res. 284:14-30 (2003)).

[098] Type I NRGl isoforms comprise all heregulin (HRG) isoforms; a rat equivalent is the neu differentiation factor (NDF), and a chicken equivalent is the acetylcholine receptor inducing activity (ARIA) polypeptide. All heregulins contain an N- terminal immunoglobulin (Ig)-like domain prior to the EGF domain, which is followed by a specific hydrophobic stretch and a unique C-terminal domain. The EGF domain in type I NRGl can be either the α or β variant, depending on whether the transcript includes "exon 7 or 8", respectively (Meyer et al. 1997). Multiple heregulin isoforms were cloned during the quest for the activator of the HER2/ErbB2 receptor and initially named HRG-D, HRG- D l, HRG-D2, HRG-D2-like and HRG-D3 (Holmes et al. 1992). However, it was later discovered that HER2 activation is indirect, requiring ligand-induced heterodimerization with another HER for full heregulin responsiveness. Type II NRGl isoforms comprise glial growth factor-2 (GGF2) and its variants; Type II NRGIs contain a Kringle-like domain, an N-terminal signal peptide, and an Ig-like domain prior to the EGF domain (D variant). Type III NRGl isoforms, also known as cysteine-rich domain neuregulins (CRD-NRG), comprise sensory and motor neuron-derived factor (SMDF) and its variants. NRGIs of this type lack Ig-like domains and glycosylation spacer regions, and contain a unique N terminus with a Cys-rich hydrophobic stretch prior to the EGF domain (D variant). It has been reported that some type III NRGl exhibit a topology unique among all known ligands for receptor tyrosine kinases, which reportedly is also different from what had been predicted by protein sequence analysis. Indeed, when expressed in fibroblasts, the type III NRGl isoforms that have a transmembrane domain C-terminal to the EGF domain, showed two transmembrane domains, and both C- and N-terminal domains were reported to be cytoplasmic. Even upon proteolytic cleavage, much of the protein (with exception of the EGF domain) remains intramembrane and/or intracellular. As a result, type III NRGIs reportedly function in juxtacrine signaling pathways, in contrast to the majority of the other neuregulins and HER/ErbB ligands which typically are shed and specialized for paracrine signaling.

[099] Other types of NRGl isoforms have also been described by RT-PCR and 5- prime RACE of adult and fetal human brain cDNA libraries. Steinthorsdottir et al. (Gene 343:97-105 (2004)) identified 10 novel alternatively spliced NRGl transcripts. They identified an additional variant by database analysis. The transcripts encode proteins with six different N-terminal domains and variability in the spacer region downstream of the Ig- like domain.

[0100] NRGl isoforms (such as NRGl -beta, also known as heregulin-beta) show distinct spatial and temporal expression patterns. These proteins play important roles during development of both the nervous system and the heart. They have been shown to regulate the selective expression of neurotransmitter receptors in neurons and at the neuromuscular junction, and promote the differentiation and development of Schwann cells from neural crest stem cells. NRGIs have also been shown to be involved in the establishment of the oligodendroglial lineage (Buonanno et al., Curr. Opin. Neurobiol. 11 :287 (2001); Adlkofer, K. and Lai, C, GHa 29:104 (2000); Garratt et al., BioEssays 22:987 (2000)). Hepatocyte Growth Factor

[0101] Another factor which was found capable of increasing the phosphorylation of protein kinase AKT using the disclosed assay is hepatocyte growth factor (HGF, also known as hepatopoietin A). HGF was reportedly identified as a potent mitogen for hepatocytes in primary culture. Subsequently, HGF was also shown to be mitogenic for a variety of cell types, including endothelial and epithelial cells, melanocytes, and keratinocytes. Recently, HGF was found to be identical to scatter factor, a fibroblast- derived soluble factor that promotes the dissociation of epithelial and vascular endothelial cell colonies in monolayer cultures by stimulating cell migration, hi addition to its mitogenic and motogenic activities, HGF is also a paracrine mediator of epithelial morphogenesis and can induce formation of well-organized tubules by MDCK epithelial cells grown in collagen gels.

[0102] HGF has now been identified as the ligand for the tyrosine kinase receptor encoded by the met proto-oncogene. Native human HGF is secreted as a biologically inactive single-chain pro-peptide that is cleaved by an extracellular serum serine protease. Treatment for Epithelial Disorders: IBD, Mucositis and Other Ailments of the GI Tract

[0103] The invention provides a composition comprising a therapeutic agent for treatment of epithelial diseases, such as mucositis, in a subject, wherein the therapeutic agent comprises either one growth factor or a combination of two or more growth factors that promote proliferation and/or survival of epithelial cells (e.g. oral keratinocytes), provided that the growth factor is not a keratinocyte growth factor (KGF) alone. The expression "provided that" indicates a required condition or exception. In certain embodiments, the composition further comprises a pharmaceutically acceptable carrier.

[0104] Moreover, in certain embodiments, the growth factor comprises a keratinocyte growth factor (KGF), hepatocyte growth factor (HGF), a member of the ErbB ligand family (such as betacellulin, heregulin-β, or NRGl-β3sv87), insulin-like growth factor-I (IGF-I), or any combination of two or more of them.

[0105] In one embodiment, the growth factor promotes proliferation of oral keratinocytes. This growth factor may comprise a member of the ErbB ligand family (such as betacellulin, heregulin-β, or NRGl-β3sv87), IGF-I, or a combination of two or more of them. In certain embodiments, the combination further comprises a KGF (e.g., KGFl, KGF2).

[0106] In another embodiment, the growth factor promotes survival of oral keratinocytes. This growth factor may comprise HGF, BTC, HRGb, NRGl-β3sv87, IGF- I, a KGF or a combination of two or more of them. In other embodiments, the growth factor promotes both proliferation and survival of oral keratinocytes. The growth factor may comprise HGF, BTC, HRGb, NRGl-β3sv87, IGF-I, a KGF, or a combination of two or more of them.

[0107] The invention further provides a composition comprising a therapeutic agent for treatment of mucositis in a subject, wherein the therapeutic agent comprises one growth factor or a combination of two or more growth factors that promote proliferation and/or survival of intestinal epithelial cells, provided that the growth factor is not KGF alone. In certain embodiments of the composition, the growth factor comprises HGF, BTC, HRGb, NRGl-β3sv87, IGF-I, or a combination of two or more of them. The combination may comprise KGF. Moreover, in certain embodiments, the composition further comprises a pharmaceutically acceptable excipient. [0108] In one embodiment, the invention also provides a long-acting therapeutic agent comprising (i) a first molecule that comprises a growth factor that promotes survival and/or proliferation of oral keratinocytes and/or of intestinal epithelial cells; and (ii) a second molecule that confers an extended half-life to the first molecule in a subject. In certain embodiments, the extended half-life exhibited by the long-acting therapeutic agent may be at least 0.5 hr, or 1 hr, or 2 hr, or 3 hr, or 4 hr, or 5 hr longer than the half-life of the first molecule in the subject. In certain embodiments, the growth factor comprises KGF, HGF, BTC, HRGb, NRGl-β3sv87, IGF-I, or a combination of two or more of them.

[0109] The invention also provides a method of treating an epithelial disease in a subject comprising providing any of the disclosed compositions (e.g. growth factors, long- acting therapeutic agents), and administering the agent(s) to the subject. In certain embodiments, the method of treatment further comprises administering chemotherapy, radiation therapy, or another biologic or small molecule to the subject. In certain embodiments, the biologic or small molecule may comprise tumor necrosis factor-alpha inhibitor, an anti-inflammatory agent, or an immunomodulatory agent. In some embodiments, anti-inflammatory agents, which are compounds that counteract or suppress the inflammatory process, can be glucocorticoids, anti-rejection drugs such as cyclosporin, cytotoxic agents such as cyclophosphamide, anti-metabolites such as methotrexate and azathioprine, and/or TNF-alpha receptor and IL-I receptor antagonists.

[0110] The epithelial disease that the method treats can be mucositis, which may include oral and/or intestinal mucositis. The epithelial disease can also be an inflammatory bowel disease, such as Crohn's disease or ulcerative colitis. In certain embodiments the composition is delivered as a mouth wash or an oral gel. The composition may also be delivered via an enema. Excipients and Formulations

[0111] In some embodiments, the compositions are provided in formulation with pharmaceutically acceptable carriers, a wide variety of which are known in the art. See, e.g., Gennaro, A.R. (2003) Remington: The Science and Practice of Pharmacy with Facts and Comparisons: DrugfactsPlus. 20th ed. Lippincott Williams & Williams; Ansel, H.C., et al., eds. (2004) Pharmaceutical Dosage Forms and Drug Delivery Systems 8th ed. Lippincott Williams & Wilkins; Kibbe, A.H., ed. (2000) Handbook of Pharmaceutical Excipients, 3rd ed. Pharmaceutical Press; and University of the Sciences in Philadelphia (2005) Remington: The Science and Practice of Pharmacy with Facts and Comparisons, 21st ed.). In one embodiment, the pharmaceutically acceptable carrier(s), includes vehicles, adjuvants, excipients, encapsulating material, auxiliary substances, or diluents, which are readily available to the public. Also readily available to the public are pharmaceutically acceptable auxiliary substances, such as pH adjusting and buffering . agents, tonicity adjusting agents, stabilizers, wetting agents and the like; one or more of which can be present in a composition of the invention. In some embodiments, suitable vehicles are selected from water, saline, dextrose, glycerol, ethanol, or the like, and combinations thereof. In addition, if desired, the vehicle can contain minor amounts of auxiliary substances such as wetting or emulsifying agents or pH buffering agents.

[0112] The U.S. Department of Health and Human Services of the Food and Drug Administration provides guidelines for estimating starting doses that are applicable for initial clinical trials on the basis of results obtained with animal tests. Thus, the publication 'Guidance for Industry and Reviewers: Estimating the Safe Starting Does in Clinical Trials for Therapeutics in Adult Healthy Volunteers' (published in December 2002) can be used, along with other guidelines available to those of skill in the art, in order to properly design the concentration and dosages of the compositions provided in the invention. The following methods and excipients are merely exemplary and are in no way limiting.

[0113] The invention also provides embodiments for pharmaceutical dosage forms in which the compositions of the invention can be administered in the form of their pharmaceutically acceptable salts; alternatively, they can also be used alone or in appropriate association, as well as in combination, with other pharmaceutically active compounds. The subject compositions are formulated in accordance to the mode of potential administration. Administration of the agents can be achieved in various ways, including oral, buccal, intranasal, rectal, enteral, parenteral, topical (e.g. gastrointestinal mucosa, oral mucosa, ocular mucosa, respiratory mucosa, genital mucosa, bladder mucosa), intraperitoneal, intradermal, transdermal, intramuscular, subcutaneous, intravenous, intra-arterial, intracardiac, intraventricular, intracranial, intratracheal, intrathecal administration, and the like; or otherwise by implanted catheter or pump, or provided via inhalation.

[0114] In one embodiment, agents that can be administered by injection refer to a formulation of the agent that will render it appropriate for parenteral administration, for example, intravenous, intraperitoneal, subcutaneous, intramuscular, intrathecal, intraorbital, intracapsular, intraspinal, intrasternal injection, or for local injection to a site of injury, damage or disorder. The injectable agent may comprise additionally to an effective amount of agent any pharmaceutically and/or physiologically acceptable solution, such as phosphate buffered saline that may be chosen by the physician handling the case according to standards known in the art. Thus, the subject compositions can be formulated into preparations in solid, semi-solid, liquid or gaseous forms, such as tablets, capsules, powders, granules, ointments, solutions, suppositories, injections, inhalants, and aerosols.

[0115] In one embodiment, agents for oral administration (i.e., an oral agent) can form solutions, suspensions, tablets, pills, granules, capsules, sustained release formulations, oral rinses, or powders. In some embodiments of oral preparations, the agents, polynucleotides, and polypeptides can be used alone or in combination with appropriate additives, for example, with conventional additives, such as lactose, mannitol, corn starch, or potato starch; with binders, such as crystalline cellulose, cellulose derivatives, acacia, corn starch, or gelatins; with disintegrators, such as corn starch, potato starch, or sodium carboxymethylcellulose; with lubricants, such as talc or magnesium stearate; and if desired, with diluents, buffering agents, moistening agents, preservatives, and flavoring agents. In addition, in one embodiment the composition may be administered intranasally using an inhalant. This composition will be formulated to allow for administration of pharmaceutically effective amounts to the lungs while minimizing damage to pulmonary tissue.

[0116] In one embodiment, the growth factors (e.g. ErbB ligands, IGF-I, HGF, KGF), including all their variants and modifications described above, can also be delivered in time-release formulations (e.g. lipid and amino acid-based microspheres and microparticles) or delivery devices.

[0117] In one embodiment, the compositions include compositions which comprise a gel matrix, such as, for example, one of the hydrogel matrices known to those of skill in the art. Non-limiting examples of gel matrices include a collagen matrix which can comprise a poloxamer or an alginate.

[0118] In one embodiment, the growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF; short or long-acting) is formulated for oral delivery. Non-limiting examples of formulations that can be used for delivery of betacellulin and/or other growth factors (e.g. ErbB ligands, IGF-I, HGF, KGF), include those formulations prepared for delivery of drugs via inhaler pumps, or via any other device for delivery of powders or aerosols which are known to those skilled in the art, such as those prepared by methods similar to those described in U.S. Patent Nos. 5740794, 5997848, 6051256, 6737045, RE37872, and RE38385; or those described in U.S. Patent Nos. 5352461, 5503852, 6071497, and 6331318; and in U.S. Published Applications 20040096403, 20060040953, each of which is incorporated herein by reference in its entirety for all that it teaches regarding diketopiperazines and microparticle-mediated drug delivery in general. In one embodiment, the growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) is delivered intranasally via an inhaler. In one embodiment, the growth factor (e.g. ErbB ligands, IGF- I, HGF, KGF) is delivered via an inhaler in a powder formulation. In another embodiment, the composition may be administered via a mouth wash, an oral gel, or an enema.

[0119] In one embodiment, the growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) is formulated for oral delivery as a pill, capsule, or an equivalent thereof, which is absorbed through a gastrointestinal membrane. For example, the growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) is formulated for oral delivery using one of the methods described in U. S. Patents 7,005,141; 6,906,030; or 6,663,898.

[0120] The composition or formulation to be administered will, in any event, contain a quantity of the agent adequate to achieve the desired state in the subject being treated.

[0121] In one embodiment, the invention provides growth factors (e.g. ErbB ligands, IGF-I, HGF, KGF) that are formulated for the purposes of being provided (e.g., sold, stored, manufactured, prescribed, and the like) as parts of a kit. A kit refers to components packaged or marked for use together. In one embodiment, the invention provides a kit containing one or more growth factors (e.g. ErbB ligands, IGF-I, HGF, KGF), optionally another biologic or small molecule, and a carrier, and these two or three components may be provided in two or three separate containers. In another example, a kit can contain any two or more components in one container, and a third component and any additional components in one or more separate containers. Optionally, a kit further contains instructions for combining and/or administering the components so as to formulate a composition suitable for administration to a subject.

[0122] Actual methods of preparing dosage forms are known, or will be apparent, to those skilled in the art (see Gennaro, A.R. (2003) Remington: The Science and Practice of Pharmacy with Facts and Comparisons: DrugfactsPlus. 20th ed.; and University of the Sciences in Philadelphia (2005) Remington: The Science and Practice of Pharmacy with Facts and Comparisons, 21st ed.). The composition or formulation to be administered will, in any event, contain a quantity of the agent adequate to achieve the desired state in the subject being treated.

[0123] In one embodiment, therapeutic formulations that comprise betacellulin and/or another of the growth factors (e.g. ErbB ligands, IGF-I, HGF, KGF; short or long- acting) of the invention can be prepared for storage by mixing these proteins, having the desired degree of purity, with optional physiologically acceptable carriers, excipients, or stabilizers (Remington's Pharmaceutical Sciences, supra), in the form of lyophilized cake, dry powder, suspensions, aqueous solutions, and the like. In one embodiment, acceptable carriers, excipients or stabilizers are nontoxic to recipient subjects at the dosages and concentrations employed, and include buffers such as phosphate, citrate, and other organic acids; antioxidants including ascorbic acid; low molecular weight (less than about 10 residues) polypeptides; proteins, such as serum albumin, gelatin, or immunoglobulins; hydrophilic polymers such as polyvinylpyrrolidone; amino acids such as glycine, glutamine, asparagine, arginine or lysine; monosaccharides, disaccharides, and other carbohydrates including glucose, lactose, mannose, or dextrins; chelating agents such as EDTA; sugar alcohols such as mannitol or sorbitol; salt-forming counterions such as sodium; and/or nonionic surfactants such as Tween, Pluronics or polyethylene glycol.

[0124] In one embodiment, one or more of the growth factors (e.g. ErbB ligands, IGF-I, HGF, KGF; short or long-acting) described herein can be complexed or bound to a polymer to increase its/their circulatory half-life for therapeutic administration. Non- limiting examples of polyethylene polyols and polyoxyethylene polyols useful for this purpose include polyoxyethylene glycerol, polyethylene glycol, polyoxyethylene sorbitol, polyoxyethylene glucose, or the like. In one embodiment, the glycerol backbone of polyoxyethylene glycerol is the same backbone occurring in, for example, animals and humans in mono-, di-, and triglycerides. Polymeric molecules are described in greater detail below.

[0125] According to one embodiment, the growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) is a long-acting growth factor (e.g. a long-acting ErbB ligand, a long-acting IGF-I, a long-acting HGF, a long-acting KGF) comprising (i) a first molecule that comprises an activity of the growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) and a (ii) second molecule that confers an extended half-life to the first molecule in a subject.

[0126] According to one embodiment, the first molecule of this long-acting growth factor is an ErbB ligand that interacts with an ErbB receptor, such as ErbBl or ErbB4 receptor. The interaction means that the two molecules form a complex that is relatively stable under physiologic conditions. Moreover, an ErbB receptor, such as an ErbBl receptor and an ErbB4 receptor, is a receptor that specifically interacts with one or more ErbB ligands and/or fragments thereof. In some embodiments, the ErbB ligand is betacellulin, a betacellulin variant, and/or a fragment thereof.

[0127] hi one embodiment, the long-acting growth factor (e.g. ErbB ligands, IGF- I, HGF, KGF) has an extended half-life in the subject that is at least 0.5 hours, or 1 hour, or 2 hours, or 3 hours, or 4 hours, or 5 hours longer than the half-life of the first molecule.

[0128] hi one embodiment, the second molecule of the long-acting growth factor (e.g. long-acting ErbB ligands, IGF-I, HGF, or KGF) comprises a polypeptide, an albumin molecule, a succinyl group, and/or a polymer.

[0129] In one embodiment, the polypeptide comprises a portion of an Fc molecule.

[0130] In one embodiment, the albumin molecule comprises an albumin, one or more fragments of albumin, a peptide that binds albumin, a molecule that conjugates with a lipid, or another molecule that binds albumin, hi one embodiment, to bind means that two or more molecules form a complex that is relatively stable under physiologic conditions, hi other words, a molecule forms a complex with albumin that is relatively stable under physiologic conditions. Conjugate is defined to encompass a molecule that is bound, either covalently or noncovalently, to another molecule. In one embodiment, for example, the albumin molecule is bound to a lipid molecule. The expression 'another molecule that binds albumin' as used in this context refers to any molecule other than a peptide that binds albumin.

[0131] In one embodiment, the polymer comprises a polyethylene glycol moiety (PEG). Optionally, the polyethylene glycol moiety is either a branched or linear chain polymer. Furthermore, in embodiments in which the polymer (e.g., PEG) is, directly or indirectly, covalently bound to the polypeptide, such covalent bond may be either permanent or transient/reversible.

[0132] hi one embodiment, upon administration of the long-acting growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) to a subject, the polymer is released from the polypeptide (i.e., the drug); the kinetics and the conditions of such release may vary with physiological and pathological paramenters such as plasma, cellular and tissue pH, redox potential, and the like. Non-limiting examples of methods for transiently, or reversibly, pegylating drugs, including polypeptide-based drugs, are provided in U.S. Patents numbers 4,935,465 (issued in June 19, 1990) and 6,342,244 (issued January 29, 2002); and in U.S. published applications number US2006/0074024. One skilled in the art would typically find more details about PEG-based reagents in, for example, published applications WO2005047366, US2005171328, or those listed on the NEKTAR PEG Reagent Catalog® 2005-2006 (Nektar Therapeutics, San Carlos, CA).

[0133] In one embodiment, the second molecule of the long-acting growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) comprises an oligomerization domain, m one embodiment, the second molecule of the long-acting growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) comprises a molecule with improved receptor binding in a lysosome. Improved receptor binding refers to increased binding (i.e., increased affinity or avidity) to the receptor relative to the growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) alone.

[0134] DNA Mutations and Amino Acid Sequence Variants

[0135] The present invention further relates to variants of the nucleic acid molecules of the present invention, which encode portions, analogs, or derivatives of the growth factors of the invention.

[0136] Thus, non-limiting examples of a fragment, derivative, or analog of the growth factors of the invention can be (i) one in which one or more of the amino acid residues are substituted with one or more conserved or non-conserved amino acid residue(s); such a substituted amino acid residue may or may not be one encoded by the genetic code; (ii) one in which one or more of the amino acid residues includes a substituent group; (iii) one in which the mature polypeptide is fused with another compound, such as a compound to increase the half-life of the polypeptide (for example, polyethylene glycol); or (iv) one in which the additional amino acids are fused to the above form of the polypeptide, such as an IgG Fc fusion region peptide, a leader or secretory sequence, a sequence employed to express or purify the above form of the polypeptide, or a proprotein sequence. Such fragments, derivatives, and analogs are deemed to be within the scope of those skilled in the art from the teachings herein.

[0137] Li one embodiment, growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) variants can occur naturally, which encompasses splice variants (see, for example, Ogata, T. et al. Endocrinology 146: 4673-81. (2005); Dunbar AJ and Goddard C, Growth Factors 18:169-75 (2000)); as well as natural allelic variants. Allelic variants include one of several alternate forms of a gene occupying a given locus on a chromosome of an organism, as described in, for example, Genes II, Lewin, B., ed., John Wiley & Sons, New York (1985), and the products of recombination, hi one embodiment, non-naturally occurring variants can also be produced using mutagenesis techniques known in the art.

[0138] Accordingly, in one embodiment, allelic variants include those produced by nucleotide substitutions, deletions, or additions. The substitutions, deletions, or additions can involve one or more nucleotides. The variants can be altered in coding regions, non- coding regions, or both. Alterations in the coding regions can produce conservative or non-conservative amino acid substitutions (discussed in more detailed below), deletions or additions. These can take the form of silent substitutions, additions, or deletions which do not alter the properties or activities of the described growth factor (e.g. ErbB ligands, IGF- I, HGF, KGF), or portions thereof.

[0139] In an embodiment, the invention provides nucleic acid molecules encoding mature growth factors (e.g. ErbB ligands, IGF-I, HGF, KGF), including those with cleaved signal peptide or leader sequences. One embodiment includes an isolated nucleic acid molecule comprising a polynucleotide having a nucleotide sequence at least 70% identical, at least 80% identical, at least 90% identical, or at least 95% identical to one or more of the growth factors of the invention (e.g., betacellulin), or a biologically active fragment of one or more of such ligands.

[0140] hi one embodiment, a biologically active fragment of an growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) is one having structural, regulatory, or biochemical functions of a naturally occurring molecule or any function related to or associated with a cellular, metabolic or physiological process. Biologically active polynucleotide fragments are those exhibiting activity similar, but not necessarily identical to, an activity of a polynucleotide of the present invention.

[0141] Jn one embodiment, a biologically active polypeptide or fragment thereof includes one that can participate in a biological reaction, including, but not limited to, activation of one or more ErbB receptors, promoting epithelial cell survival, inhibiting epithelial cell apoptosis and/or necrosis, promoting wound healing, or a combination of any of the above. In another embodiment, a biologically active polypeptide is one that can serve as an epitope or immunogen to stimulate an immune response, such as production of antibodies; or that can participate in modulating the immune response. In one embodiment, the biological activity can include an improved desired activity, or a decreased undesirable activity.

[0142] In addition, in another embodiment, an entity demonstrates biological activity when it participates in a molecular interaction with another molecule, such as hybridization, when it has therapeutic value in alleviating a disease condition, when it has prophylactic value in inducing an immune response, when it has diagnostic and/or prognostic value in determining the presence of a molecule, such as a biologically active fragment of a polynucleotide that can, for example, be detected as unique for the polynucleotide molecule, or that can be used as a primer in a polymerase chain reaction.

[0143] A polynucleotide having a nucleotide sequence at least, for example, 95% identical to a reference nucleotide sequence encoding a growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) is one in which the nucleotide sequence is identical to the reference sequence except that it may include up to five point mutations per each 100 nucleotides of the reference nucleotide sequence. In other words, to obtain a polynucleotide having a nucleotide sequence at least 95% identical to a reference nucleotide sequence, up to 5% of the nucleotides in the reference sequence may be deleted or substituted with another nucleotide, or a number of nucleotides up to 5% of the total nucleotides in the reference sequence may be inserted into the reference sequence. These mutations of the reference sequence may occur at the 5' or 3' terminal positions of the reference nucleotide sequence or anywhere between those terminal positions, interspersed either individually among nucleotides in the reference sequence or in one or more contiguous groups within the reference sequence.

[0144] In one embodiment, whether any particular nucleic acid molecule is at least 70%, 80%, 90%, or 95% identical to the growth factors of the invention including betacellulin can be determined conventionally using known computer programs such as the Bestfit program (Wisconsin Sequence Analysis Package, Version 8 for Unix, Genetics Computer Group, Madison, WI). Bestfit uses the local homology algorithm of Smith and Waterman, Advances in Applied Mathematics 2: 482-489 (1981), to find the best segment of homology between two sequences. When using Bestfit or any other sequence alignment program to determine whether a particular sequence is, for instance, 95% identical to a reference sequence according to the present invention, the parameters are set such that the percentage of identity is calculated over the full length of the reference nucleotide sequence and that gaps in homology of up to 5% of the total number of nucleotides in the reference sequence are allowed.

[0145] In one embodiment, one or more of the nucleic acid molecules are at least 70%, 80%, 90%, or 95% identical to the growth factors of the invention, including betacellulin, irrespective of whether they encode a polypeptide having an growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) activity as described herein. Even where a particular nucleic acid molecule does not encode a polypeptide having activity, one of skill in the art would know how to use the nucleic acid molecule, for instance, as a hybridization probe or a polymerase chain reaction (PCR) primer. Uses of the nucleic acid molecules of the present invention that do not encode a polypeptide having activity include, inter alia, isolating the gene or allelic variants thereof in a cDNA library; and in situ hybridization (for example, fluorescent in situ hybridization (FISH)) to metaphase chromosomal spreads to provide the precise chromosomal location of the growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) genes, as described in Verna et al, Human Chromosomes: A Manual of Basic Techniques, Pergamon Press, New York (1988); and Northern blot analysis for detecting their betacellulin mRNA expression in specific tissues.

[0146] In another embodiment, one or more nucleic acid molecules have sequences at least 70%, 80%, 90%, or 95% identical to a nucleic acid sequence of a growth factor of the invention (e.g. ErbB ligands, IGF-I, HGF, KGF) and encode a polypeptide having polypeptide activity, that is, a polypeptide exhibiting activity similar but not necessarily identical, to an activity of the growth factors of the invention, as defined above. In one embodiment, for example, the growth factors (e.g. ErbB ligands, IGF-I, HGF, KGF) of the present invention can stimulate epithelial cell survival, proliferation or both.

[0147] In another embodiment, and due to the degeneracy of the genetic code, one of ordinary skill in the art will immediately recognize that a large number of the nucleic acid molecules having a sequence at least 70%, 80%, 90%, or 95% identical to the nucleic acid sequence of one or more of the growth factors of the invention will encode a polypeptide having activity. In fact, since multiple degenerate variants of these nucleotide sequences encode the same polypeptide, this will be clear to the skilled artisan even without performing the above described comparison assay. It will be further recognized in the art that a reasonable number of nucleic acid molecules that are not degenerate variants will also encode a polypeptide having activity. Thus, the skilled artisan is fully aware of amino acid substitutions that are either less likely or not likely to significantly affect protein function (for example, replacing one aliphatic amino acid with a second aliphatic amino acid), as further described below.

[0148] In one embodiment, protein engineering can be employed to improve or alter the characteristics of the growth factors of the invention. Recombinant DNA technology known to those skilled in the art can be used to create novel mutant proteins or "muteins" including single or multiple amino acid substitutions, deletions, additions, or fusion molecules, m one embodiment, such modified polypeptides can show desirable properties, such as enhanced activity or increased stability. In one embodiment, such modified polypeptides can be purified in higher yields and show better solubility than the corresponding natural polypeptide, at least under certain purification and storage conditions. In one embodiment, non-limiting examples of betacellulin muteins are given in US Patent No. 6,825,165 (for example, SEQ ID NO. 1, 2, and 38 referred to therein).

[0149] In one embodiment the invention provides that, for many proteins, including the extracellular domain of a membrane associated growth factor of the invention or the mature form(s) of a secreted growth factor of the invention, such as an ErbB ligand, one or more amino acids can be deleted from the N-terminus or C-terminus without substantial loss of biological function. One skilled in the art knows that, for instance, Ron et al., J. Biol. Chem., 268:2984-2988 (1993), reported modified KGF proteins that had heparin binding activity even if 3, 8, or 27 amino-terminal amino acid residues were missing. Similarly, many examples of biologically functional C-terminal deletion muteins are known. For instance, interferon gamma increases in activity as much as ten fold when 8-10 amino acid residues are deleted from the carboxy terminus of the protein, see, for example, Dobeli et al., J. Biotechnology, 7:199-216 (1988).

[0150] In one embodiment, even if deletion of one or more amino acids from the N-terminus or C-terminus of a protein results in modification or loss of one or more biological functions of the protein, other biological activities may still be retained. Thus, the ability of the shortened protein to induce and/or bind to antibodies which recognize the complete or mature from of the protein generally will be retained when less than the majority of the residues of the complete or mature protein are removed from the N- or C- terminus. Whether a particular polypeptide lacking N- or C-terminal residues of a complete protein retains such immunologic activities can be determined by routine methods described herein and otherwise known in the art. Accordingly, in one embodiment, the present invention further provides polypeptides having one or more residues deleted from the amino terminus of the amino acid sequences of the growth factors of the invention.

[0151] In one embodiment, it also will be recognized by one of ordinary skill in the art that some amino acid sequences of the growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) polypeptides of the invention can be varied without significant effect on the structure or function of the protein. If such differences in sequence are contemplated, it should be remembered that there will be critical areas on the protein which determine activity.

[0152] Li one embodiment, the invention includes variations of the growth factors (e.g. ErbB ligands, IGF-I, HGF, KGF) which show substantial growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) activity as described herein or which include regions of the growth factors (e.g. ErbB ligands, IGF-I, HGF, KGF) such as the protein portions discussed below. Such mutants include deletions, insertions, inversions, repeats, and type substitutions, selected according to general rules known in the art, so as have little effect on activity. For example, guidance concerning how to make phenotypically silent amino acid substitutions is provided in Bowie, J.U. et al., Science, 247:1306-1310 (1990), wherein the authors indicate that there are two main approaches for studying the tolerance of an amino acid sequence to change. The first method relies on the process of evolution, in which mutations are either accepted or rejected by natural selection. The second approach uses genetic engineering to introduce amino acid changes at specific positions of a cloned gene and selections, or screens, to identify sequences that maintain functionality.

[0153] These studies report that proteins are surprisingly tolerant of amino acid substitutions. The authors further indicate which amino acid changes are likely to be permissive at a certain position of the protein. For example, most buried amino acid residues require nonpolar side chains, whereas few features of surface side chains are generally conserved. Other such phenotypically silent substitutions are described in Bowie, et al., supra, and the references cited therein. Typically seen as conservative substitutions are the replacements, one for another, among the aliphatic amino acids Ala, VaI, Leu, and He; hydrophobic substitutions Leu, Iso, and VaI, interchange of the hydroxyl residues Ser and Thr, exchange of the acidic residues Asp and GIu, substitution between the amide residues Asn and GIn, exchange of the basic residues Lys, His, and Arg, replacements between the aromatic residues Phe, Tip, and Tyr, and between small amino acid substitutions Ala, Ser, Thr, Met, and GIy.

[0154] In one embodiment, amino acids involved in growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) functions can be identified by methods known in the art, such as site-directed mutagenesis or alanine-scanning mutagenesis, see, for example, Cunningham, B.C. and Wells, J.A., Science, 244:1081-1085 (1989). The latter procedure introduces single alanine mutations, hi one embodiment, the resulting mutant molecules are then tested for biological activity including, but not limited to, receptor binding, or in vitro or in vivo promotion of epithelial cell survival and/or proliferation, wound healing, or any combination of these activities.

[0155] In one embodiment, substitutions of charged amino acids with other charged or neutral amino acids can produce proteins with highly desirable improved characteristics, such as less aggregation. Aggregation may not only reduce activity but also be problematic when preparing pharmaceutical formulations, because, for example, aggregates can be immunogenic, Pinckard, R.N. et al., Clin. Exp. Immunol., 2:331-340 (1967); Robbins, D.C. et al., Diabetes, 36:838-845 (1987); Cleland, J.L. et al., Crit. Rev. Therapeutic Drug Carrier Systems, 10:307-377 (1993).

[0156] In one embodiment, replacing amino acids can also change the selectivity of the binding of a ligand to cell surface receptors. For example, Van Ostade, X. et al., Nature, 361:266-268 (1993) describes mutations resulting in selective binding of TNF-α to only one of the two known types of TNF receptors. In one embodiment, sites that are important for ligand-receptor binding can also be determined by structural analysis such as crystallization, nuclear magnetic resonance, or photoaffϊnity labeling, for example, Smith, LJ. et al., J. MoI. Biol, 224:899-904 (1992) and de Vos, A.M. et al., Science, 255:306- 312 (1992).

[0157] In one embodiment, applying some of these common principles to the ErbB ligand betacellulin, we note that the sequence includes eight cysteine residues, located at amino acid positions number 7, number 28, number 69, number 77, number 82, number 93, number 95, and number 104. In one embodiment, the invention provides mutant betacellulin molecules with one or more cysteine residues mutated to, for example, serine residues, hi one embodiment, these constructs can be cloned into any expression suitable vector, as known in the art, for example, the pTT5-G vector. [0158] In another embodiment, analyzing these muteins provides an understanding of the disulfide bond pattern of betacellulin and may identify a protein with improved properties, for example, improved expression and secretion from mammalian cells, decreased aggregation of the purified protein, and the potential to produce active recombinant betacellulin, when expressed in E. coli.

Fusion Polypeptides

[0159] As discussed above, the inventors have found that certain growth factors (e.g., KGF, HGF, BTC, HRGb, NRGl-β3sv87, IGF-I) can promote the survival and/or proliferation of epithelial cells. It can therefore be desirable to increase the half-life of these growth factors in vivo to produce a more sustained in vivo activity. Gene manipulation techniques have enabled the development and use of recombinant therapeutic proteins with fusion partners that impart desirable pharmacokinetic properties. Several different fusion partners have been used to produce fusion molecules. For example, recombinant human serum albumin fused with synthetic heme protein has been reported to reversibly carry oxygen (Chuang, V.T. et al., Pharm Res., 19:569-577 (2002)). The long half-life and stability of human serum albumin (HSA) makes it an attractive candidate for fusion to short-lived therapeutic proteins (U.S. Patent No. 6,686,179). Thus, in one embodiment, the fusion partner comprises albumin. The albumin can include human serum albumin or a peptide that binds to or conjugates with a lipid or other molecule that binds albumin. These fusion partners can include any variant of or any fragment of such.

[0160] Oligomerization offers functional advantages to a fusion protein, including multivalency, increased binding strenght, and the combined function of different domains. These features are seen in natural proteins and may also be introduced by protein engineering. Accordingly, the invention provides a growth factor fusion molecule, wherein the fusion partner comprises an oligomerization domain, for example, a dimerization domain. Suitable oligomerization domains include coiled-coil domains, including alpha-helical coiled-coil domains; collagen domains; collagen-like domains, and dimeric immunoglobulin domains. Suitable coiled-coil polypeptide fusion partners of the invention include tetranectin coiled-coil domain, the coiled-coil domain of cartilage oligomeric matrix protein; angiopoietin coiled-coil domains; and leucine zipper domains. Growth factors fusion molecules with collagen or collagen-like oligomerization domains as fusion partner may comprise, for example, those found in collagens, mannose binding lectin, lung surfactant proteins A and D, adiponectin, ficolin, conglutinin, macrophage scavenger receptor, and emilin.

[0161] The Fc receptor of human immunoglobulin G subclass 1 (IgGl) has also been used as a fusion partner for a therapeutic molecule. It has been recombinantly linked to two soluble p75 tumor necrosis factor (TNF) receptor molecules. This fusion protein has been reported to have a longer circulating half-life than monomeric soluble receptors, and to inhibit TNF-alpha-induced proinflammatory activity in the joints of patients with rheumatoid arthritis (Goldenberg, M.M. Clin Ther., 21:75-87 (1999)). This fusion protein has been used clinically to treat rheumatoid arthritis, juvenile rheumatoid arthritis, psoriatic arthritis, and ankylosing spondylitis (Nanda, S. and Bathon, J.M., Expert Opin. Pharmacother., 5: 1175-1186 (2004)). Thus, in one embodiment, the fusion partner can comprise an Fc fragment.

[0162] Fusion partners have also been produced comprising the first two domains of the human CD4-polypeptide and various domains of the constant regions of the heavy or light chains of mammalian immunoglobulins. See, for example, EP A 394,827; Traunecker, A. et al., Nature, 331:84-86 (1988). Fusion molecules that have a disulfide- linked dimeric structure due to the IgG part can also be more efficient in binding and neutralizing other molecules than, for example, a monomeric growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) polypeptide or polypeptide fragment alone. See, for example, Fountoulakis, M. et al., J. Biochem., 270:3958-3964 (1995).

[0163] Thus, the invention provides polypeptide fusion partners for the growth factors of the invention. In one embodiment, the fusion partners may be part of a fusion molecule, for example, a polynucleotide or polypeptide, which represents the joining of all or portions of more than one gene. As such, the invention can provide a nucleic acid molecule with a second nucleotide sequence that encodes a fusion partner. This second nucleotide sequence can be operably linked to the first nucleotide sequence. For example, a fusion protein can be the product obtained by splicing strands of recombinant DNA and expressing the hybrid gene.

[0164] In one embodiment, a fusion molecule can be made by genetic engineering, for example, by removing the stop codon from the DNA sequence of a first protein, then appending the DNA sequence of a second protein in frame. The DNA sequence will then be expressed by a cell as a single protein. In one embodiment, this is accomplished by cloning a cDNA into an expression vector in frame with an existing gene. The invention also provides fusion molecules with heterologous and homologous leader sequences, fusion molecules with a heterologous amino acid sequence, and fusion molecules with or without N-terminal methionine residues. The fusion partners of the invention can be either N-terminal fusion partners or C-terminal fusion partners.

[0165] Li one embodiment, fusion polypeptides can be secreted from the cell by the incorporation of leader sequences that direct the protein to the membrane for secretion. These leader sequences can be specific to the host cell, and are known to skilled artisans; they are also cited in the references. Thus, the invention includes appropriate restriction enzyme sites for cloning the various fusion polypeptides into the appropriate vectors. In addition to facilitating the secretion of these fusion molecules, the invention provides for facilitating their production. This can be accomplished in a number of ways, including producing multiple copies, employing strong promoters, and increasing their intracellular stability, for example, by fusion with beta-galactosidase.

[0166] In one embodiment, the fusion partners can include linkers, i.e., fragments of synthetic DNA containing a restriction endonuclease recognition site that can be used for splicing genes. These can include polylinkers, which contain several restriction enzyme recognition sites. A linker may be part of a cloning vector. It can be located either upstream or downstream of the therapeutic protein, and it can be located either upstream or downstream of the fusion partner.

[0167] In one embodiment, the first molecule can comprise any growth factor of the invention (e.g. ErbB ligands, IGF-I, HGF, KGF) , or one or more of its fragments, that can be purchased from suppliers, such as R&D System (Minneapolis, MN). In one embodiment, the first molecule can, for example, be an ErbB ligaήd, or a fragment thereof, for example one chosen from the molecules listed in Tables 1 through 6 of Example 13 .

[0168] In one embodiment, the second molecule can facilitate production, secretion, and/or purification of the fusion molecule. In one embodiment, second molecules suitable for use in the invention include, for example, a polymer, a polypeptide, a succinyl group, or an albumin molecule. In one embodiment, the second molecule can comprise an oligomerization domain or a molecule with improved receptor binding in a lysosome.

[0169] In one embodiment, a long-acting growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) polypeptide of the invention can be prepared by attaching polypeptides or branch point amino acids to the growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) polypeptide. For example, the polypeptide may be a carrier protein that serves to increase the circulation half-life of the growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) polypeptide (i.e., in addition to the advantages achieved via a growth factor fusion molecule). In one embodiment, such polypeptides do not create neutralizing antigenic response, or other adverse responses. Such polypeptides can be selected from serum album (such as human serum albumin), an additional antibody or portion thereof, for example the Fc region, or other polypeptides, for example poly-lysine residues. As described herein, the location of attachment of the polypeptide may be at the N-terminus, or C-terminus, or other places in between, and also may be connected by a chemical linker moiety to the selected growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF).

[0170] Such modified polypeptides can show, for example, enhanced activity or increased stability. In addition, they may be purified in higher yields and show better solubility than the corresponding natural polypeptide, at least under certain purification and storage conditions. In one embodiment, a human serum albumin-ErbB ligand fusion molecule may be prepared as described herein and as further described in U.S. Patent No. 6,686,179.

[0171] In one embodiment, the invention also provides for facilitating the purification of these fusion proteins. Fusion with a selectable marker can, for example, facilitate purification by affinity chromatography. For example, fusion with the selectable marker glutathione S-transferase (GST) produces polypeptides that can be detected with antibodies directed against GST, and isolated by affinity chromatography on glutathione- sepharose; the GST marker can then be removed by thrombin cleavage. Polypeptides that provide for binding to metal ions are also suitable for affinity purification. For example, a fusion protein that incorporates Hisn, where n is between three and ten, inclusive, for example, a 6xHis-tag can be used to isolate a protein by affinity chromatography using a nickel ligand.

Other Polymer-based Modifications, Derivatizations, Pegylations

[0172] According to one embodiment, conjugates of the growth factors (e.g. ErbB ligands, IGF-I, HGF, KGF) can be prepared using glycosylated, non-glycosylated or de- glycosylated growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) and fragments or variants thereof. Suitable chemical moieties for derivatization of growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) and variants of growth factor include, for example, polymers, such as water soluble polymers described herein. [0173] In one embodiment, polymers, including water soluble polymers, are useful in the present invention as the polypeptide to which each polymer is attached will not precipitate in an aqueous environment, such as a physiological environment, hi one embodiment, polymers employed in the invention will be pharmaceutically acceptable for the preparation of a therapeutic product or composition. One skilled in the art will be able to select the desired polymer based on such considerations as whether the polymer/protein conjugate will be used therapeutically and, if so, the desired dosage, circulation time and resistance to proteolysis.

[0174] In one embodiment, polymers (e.g., water soluble polymers) can be of any molecular weight. In one embodiment, polymers can be branched or unbranched. In one embodiment, the polymers each can have an average molecular weight of between about 2 kDa to about 100 kDa. In another embodiment, the average molecular weight of each polymer is between about 5 kDa and about 50 kDa. In another embodiment, the average molecular weight of each polymer is between about 12 kDa and about 25 kDa. Generally, the higher the molecular weight or the more branches, the higher the porymeπprotein ratio. In an embodiment, other sizes may be used, depending on the desired therapeutic profile, for example the duration of sustained release; the effects, if any, on biological activity; the ease in handling; the degree or lack of antigenicity and other known effects of a polymer on a modified growth factor of the invention.

[0175] In one embodiment, suitable, clinically acceptable, water soluble polymers include, but are not limited to, polyethylene glycol (PEG), polyethylene glycol propionaldehyde, copolymers of ethylene glycol/propylene glycol, monomethoxy- polyethylene glycol, carboxymethylcellulose, dextran, polyvinyl alcohol (PVA), polyvinyl pyrrolidone, poly-l,3-dioxolane, poly-l,3,6-trioxane, ethylene/maleic anhydride copolymer, poly (β-amino acids) (either homopolymers or random copolymers), poly(n- vinyl pyrrolidone) polyethylene glycol, polypropylene glycol homopolymers (PPG) and other polyakylene oxides, polypropylene oxide/ethylene oxide copolymers, polyoxyethylated polyols (POG) (for example, glycerol) and other polyoxyethylated polyols, polyoxyethylated sorbitol, or polyoxyethylated glucose, colonic acids or other carbohydrate polymers, Ficoll or dextran and mixtures thereof.

[0176] In one embodiment, polyethylene glycol encompasses any of the forms that have been used to derivatize other proteins, such as mono-(Cl-ClO) alkoxy- or aryloxy- polyethylene glycol. In one embodiment, polyethylene glycol propionaldehyde may have advantages in manufacturing due to its stability in water.

[0177] hi one embodiment, polymers employed in the present invention are attached to a growth factor of the invention with consideration of effects on functional or antigenic domains of the polypeptide, hi one embodiment, chemical derivatization can be performed under any suitable condition used to react a protein with an activated polymer molecule, hi one embodiment, activating groups that can be used to link the polymer to the active moieties include the following: sulfone, maleimide, sulfhydryl, thiol, triflate, tresylate, azidirine, oxirane and 5-pyridyl.

[0178] In one embodiment, one (or more) polymers is attached to a growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) polypeptide of the invention at the alpha (α) or epsilon (ε) amino groups of amino acids. In one embodiment, the polymer(s) is(are) attached to a reactive thiol group. In one embodiment, the polymer(s) is(are) attached to any reactive group of the protein that is sufficiently reactive to become attached to a polymer group under suitable reaction conditions. Thus, in one embodiment, a polymer can be covalently bound to a growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) polypeptide of the invention via a reactive group, such as a free amino or carboxyl group. In one embodiment, the amino acid residues having a free amino group may include lysine residues and the N-terminal amino acid residue. In one embodiment, amino acids having a free carboxyl group may include aspartic acid residues, glutamic acid residues and the C- terminal amino acid residue, hi one embodiment, amino acids having a reactive thiol group include cysteine residues.

[0179] hi one embodiment, the invention provides methods of preparing growth factors (e.g. ErbB ligands, IGF-I, HGF, KGF) conjugated with polymers, including growth factor fusion molecules conjugated with polymers, such as water soluble polymers, including: (a) reacting a protein with a polymer under conditions whereby the protein becomes attached to one or more polymers and (b) obtaining the reaction product.

[0180] Reaction conditions for each conjugation are well known by those skilled in the art, and may be selected from any of those known in the art or those subsequently developed, but should be selected to avoid or limit exposure to reaction conditions such as temperatures, solvents, and pH levels that would inactivate the protein to be modified, hi general, the optimal reaction conditions for the reactions will be determined case-by-case based on known parameters and the desired result. For example, the larger the ratio of polymeπprotein conjugate, the greater the percentage of conjugated product. The optimum ratio (in terms of efficiency of reaction in that there is no excess unreacted protein or polymer) can be determined by factors such as the desired degree of derivatization (for example, mono-, di-, tri- etc.), the molecular weight of the polymer selected, whether the polymer is branched or unbranched and the reaction conditions used. In one embodiment, the ratio of polymer (for example, PEG) to growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) polypeptide will range from 1:1 to 100:1. Molar ratios of activated polymer to protein of 2000:1 can also be used, depending on the concentration of the protein.

[0181] hi one embodiment, one or more purified polymer conjugates can be prepared from each mixture by standard purification techniques, including among others, dialysis, salting-out, ultrafiltration, ion-exchange chromatography, gel filtration chromatography and electrophoresis.

[0182] In one embodiment, one may specifically prepare an N-terminal chemically modified protein. One may select a polymer by, for example, its molecular weight and/or its branching, the proportion of polymers to protein (or peptide) molecules in the reaction mix, the type of reaction to be performed, and the method of obtaining the selected N- terminal chemically modified protein. The method of obtaining the N-terminal chemically modified protein preparation (i.e., separating this moiety from other mono-derivatized moieties if necessary) may be by purification of the N-terminal chemically modified protein material from a population of chemically modified protein molecules.

[0183] hi one embodiment, selective N-terminal chemical modification can be accomplished by reductive alkylation that exploits differential reactivity of different types of primary amino groups (lysine versus the N-terminal) available for derivatization in a particular protein. In one embodiment, the present invention contemplates the chemically derivatized growth factor (e.g. ErbB ligands, IGF-I, HGF, KGF) polypeptide to include mono- or poly- (for example, 2-4) PEG moieties. "Pegylation" may be carried out by any of the pegylation reactions known in the art. There are a number of PEG attachment methods available to those skilled in the art. See, for example, U.S. Patents numbers 4,935,465 (issued in June 19, 1990) and 6,342,244 (issued January 29, 2002); U.S. published applications number US2006/0074024 EP 0401 384; Malik, F. et al., Exp. Hematol., 20:1028-1035 (1992); Francis, Focus on Growth Factors, 3(2):4-10 (1992); EP 0 154 316; EP 0 401 384; WO 92/16221; WO 95/34326; and the other publications cited herein that relate to pegylation.

[0184] Pegylation by acylation generally involves reacting an active ester derivative of polyethylene glycol with an growth factor (e.g. ErbB ligands, IGF-I, HGF₅ KGF) polypeptide of the invention. In one embodiment, the activated PEG ester is PEG esterified to N-hydroxysuccinimide (NHS). In one embodiment, the linkage between the therapeutic protein and a polymer such as PEG is an amide, carbamate, urethane, and the like. See, for example, Chamow, S.M. Bioconjugate Chem., 5 (2):133-140 (1994). Pegylation by acylation will generally result in a poly-pegylated protein. In one embodiment, the resulting product is substantially only (for example, >95%) mono, di- or tri-pegylated. hi another embodiment, some species with higher degrees of pegylation can be formed in amounts depending on the specific reaction conditions used.

[0185] Pegylation by alkylation generally involves reacting a terminal aldehyde derivative of PEG with the protein in the presence of a reducing agent. For the reductive alkylation reaction, the polymer(s) selected should have a single reactive aldehyde group. An exemplary reactive PEG aldehyde is polyethylene glycol propionaldehyde, which is water stable, or mono Cl-ClO alkoxy or aryloxy derivatives thereof (see for example, U.S. Pat. No. 5,252,714).

[0186] Reference will now be made in detail to some exemplary embodiments consistent with the invention, examples of which are illustrated in the accompanying figures.

Examples

[0187] The examples, which are intended to be purely exemplary of the invention and should therefore not be considered to limit the invention in any way, also describe and detail aspects and embodiments of the invention discussed above. The examples are not intended to represent that the experiments below are all or the only experiments performed. Efforts have been made to ensure accuracy with respect to numbers used (for example, amounts, temperature, etc.) but some experimental errors and deviations should be accounted for. Unless indicated otherwise, parts are parts by weight, molecular weight is weight average molecular weight, temperature is in degrees Centigrade, and pressure is at or near atmospheric. Example 1: High-Throughput Protein Production System

[0188] A human cDNA library was developed using RIKEN technology. The full- length clones of the library were obtained from over 70 human tissue sources. From this library, an automated process was developed to obtain the secreted proteins that were to be screened with various bioassays. The automated process, having the capacity to prepare in the order of 2000 samples per week, was used in several high-throughput in vitro bioassays, including one for AKT activation. Betacellulin (clone 736345), and the neuregulin 1-beta 3 splice variant NRGl-β3sv87 cDNA (clone 541754), were among the proteins identified as giving a positive signal in the latter assay. Preparation of cDNA Library

[0189] The cDNA clones of the invention were derived from a library of total RNA isolated from normal and diseased tissues, as well as from normal and treated cells, for example, stimulated peripheral blood mononuclear cells (PBMC). These RNA samples were transcribed into cDNA using technology described by RIKEN and others, including methods of capturing the 5' ends of DNA ("CAP trapping") and methods to eliminate secondary structure in the mRNA using trehalose so that the entire molecule can be reverse transcribed (WO 02/28876; WO 02/070720; U.S. Patent No. 6,627,399; U.S. Patent No. 6,458,756; U.S. Patent No. 6,372,437; U.S. Patent No. 6,365,350; U.S. Patent No. 3,344,345; U.S. Patent No. 6,342,387, U.S. Patent No. 6,333,156; U.S. Patent No. 6,294,337; U.S. Patent No. 6,265,569; U.S. Patent No. 6,221,599; U.S. Patent No. 6,174,669; U.S. Patent No. 6,143,528; U.S. Patent No. 6,074,824; and U.S. Patent No. 6,013,488).

[0190] Libraries of the transcribed cDNA were compiled, and samples of approximately three 384-well plates from each library were sequenced at their 5' end. Using the diversity of the library as represented by the sample as the criteria, the 5¹ ends of as many as 10,000 clones from each library were sequenced. This 5¹ end sequence information was the basis of an analysis that provided a clustered organization of the clones. The clusters were based on a map of the human genome including all known human genes and all known human expressed sequence tags. Multiple sequences mapping to the same locus were identified as belonging to one cluster. A cluster may include splice variants. Further, samples of some of the members of the transcribed cDNA libraries were compiled, and sequenced at their 3' end, as well as their 5' end. A subset of these possessed contiguous 5' end sequence and 3' end sequence. These were assembled into full length sequences, and served to identify the betacellulin clones and the NRGl-β3sv87 cDNA clone 541754, described herein. High-Throughput Cell Transfection and Collection of Conditioned Media

[0191] Human cells (human kidney epithelial 293 cells obtained from ATCC; ATCC® Number: CRL- 1573 ™)) were seeded in 96-well plates and grown until reaching 70-80% confluency. A plurality of purified cDNAs obtained from the library generated as described above were transiently transfected into the cells by the Lipofectamine (Invitrogen Corporation, CA) method following the instructions provided by the manufacturer. Subsequently, cells were washed with PBS and grown in serum free medium for up to 7 days. Conditioned media were then collected from each well on multiple days, filtered and concentrated. On average, 1.6 ml of conditioned media were obtained per cDNA per week for the AKT bioassay. These conditioned media were used, for example, in assays for AKT and/or ERK phosphorylation, cell survival and cell proliferation tests.

Example 2: High-Throughput Assay of AKT Phosphorylation

[0192] High-throughput conditioned media, prepared as described in Example 1, were used for the AKT bioassay as follows. Preparation of the AKT-Containing Cells of the Assay

[0193] To prepare the cells whose AKT phosphorylation levels were measured (i.e. HT-29 and RIE cells), the following methods were used.

[0194] HT-29 cells were obtained from ATCC (ATCC® Number: HTB-38 ™) and grown in McCoy's media (CellGro Cat #10-050-CV) with 10% serum (CellGro Cat #35-010-CV), 1% penicillin/streptomycin (P/S CellGro Cat #30-002-CI). Upon receipt, the cells were rapidly thawed and plated in one T- 150 flask with 30 ml of growth media following the instructions from ATCC. When they reached confluence, the cells were split. For splitting the cells, cells were washed once with D-PBS (without Ca2+ and Mg2+, CellGro Cat #25-053-CI). 2.5 ml of trypsin (0.25% Trypsin-EDTA CellGro Cat #25-053-CL) were added and the cells were incubated at room temperature for approximately five to ten minutes, until the majority of the cells were coming off the plate after light tapping when examined under the microscope. 10 ml of growth media were added to stop the trypsinization. The cells were counted with the hemocytometer, and plated at a density of 1 x 106 cells/25 cm2 area of tissue culture flask (i.e., 6 x 106 cells/T- 150 flask). For the initial splits, as many cells as possible were frozen (following standard procedures at a concentration of 20 x 106 cells/vial). Cells were then used for experiments.

[0195] To prepare the RIE cells, these were first obtained from ATCC (ATCC® Number: CRL- 1592™). As per ATCCs instructions, the cells were expanded in growth media consisting of DMEM (high glucose - CellGro Cat #10-013-CV) with 10% serum (CellGro Cat #35-010-CV), 1% penicillin/streptomycin (P/S CellGro Cat #30-002-CI) and 0.1 unit/ml of human insulin (Humulin Lilly Cat #002-8215-01). Upon receiving, the cells were rapidly thawed and plated in a T- 150 flask with 30 ml of the growth media following the instructions from ATCC. When they reached confluence, the cell cultures were split. For splitting, the cells were washed one time with D-PBS (without calcium and magnesium, CellGro Cat #25-053-CI). 2.5 ml of trypsin (0.25% Trypsin-EDTA CellGro Cat #25-053-CL) were then added and the cells were incubated at room temperature for approximately 3-4 minutes until the majority of the cells were coming off after light tapping when examined under the microscope. 10 ml of growth media were added to stop the trypsinization. The cells were counted with the hemocytometer, and plated at a density of 1 x 106 cells/25 cm2 area of tissue culture flask (i.e., 6 x 106 cells/T-150 flask). For the initial splits, as many cells as possible were frozen for stock maintenance (following standard procedures at a concentration of 10x106 cells/vial). The cells were then used for experiments. Quantification of AKT Phosphorylation in HT-29 cells

[0196] HT-29 cells, prepared as described above, were plated at a density of 75,000 cells/well on 96-well plates, transferred to a cell culture incubator at 37oC and 5% CO2, where they were kept overnight. The next day, the cells were washed with PBS once. The cells were then starved in 100 μl of DMEM containing 1OmM HEPES and 0.1% BSA at 37oC for eight hours with no insulin. The plates were then washed one time with 100 μl PBS (with Ca2+ and Mg2+) and 50 μl fresh DMEM containing 10 niM HEPES and 0.1% bovine serum albumin (BSA) without insulin were added to negative control wells. At the same time, 50 μl of IGF-I controls were added from the control plate to column 12 of the assay plate (at varying concentrations of IGF-I ranging from 0.02 ug/ml to 100ng/ml). 50 μl of media conditioned by cells expressing one of the various cDNA clones from the library described above (e.g., betacellulin, or NRGl-β3sv87) (in DMEM + 5% serum) were added to the rest of the assay plate (col. 1-11) and incubated at 37⁰C for 30 minutes. The plate was washed once with 390 μl PBS (with Ca²⁺ and Mg²⁺). The fluid was aspirated out and, after 80 μl of lysis buffer were added, the plate was shaken for 2 minutes. Quantification of AKT Phosphorylation in RIE cells

[0197] RDE cells, prepared as described above, were plated at a density of 60,000 cells/well on a 96-well plate in the growth media. The cells were incubated at 37oC in 5% CO2 overnight. The next day, the cells were washed with PBS once. The cells were then starved in 100 μl of DMEM containing 10 mM HEPES and 0.1% BSA at 37 oC for eight hours with no insulin. The plates were then washed one time with 100 μl PBS (with Ca2+ and Mg2+). Then, 50 μl of fresh DMEM containing 10 mM HEPES and 0.1% BSA without insulin were added to negative control wells. Simultaneously, 50 μl of test conditioned media and a recombinant protein control (rIGF-1), all in DMEM containing 5% serum, were added to different wells as follows: empty- vector control, Hrg-D 1, NRGl-β3sv87, HGF, IGF-I, and rIGF-1 at 100ng/ml (R&D Systems). The RIE cells were incubated with these test substances at 37 oC for 30 minutes. The plate was washed once with 390 μl PBS (with Ca2+ and Mg2+). The fluid was aspirated out and, after 80 μl of lysis buffer were added, the plate was shaken for 2 minutes. Subsequently, the cells were processed and analyzed using a phospho-AKTl ELISA plate as described next.

[0198] The phosphorylation of AKT was measured using a Phospho-AKTl ELISA plate (PathScan® Phospho-AKTl (Ser473) Sandwich ELISA Kit (Cell Signaling Technology®, MA)). 50 μl of each supernatant were transferred to a Phospho-AKTl ELISA plate, which already contained 50 μl of sample diluent. The plate was incubated at 37 oC overnight. The following day, the plate was washed with the wash buffer four times. Then, 100 μl AKTl monoclonal detection antibody were added and the plate was incubated at 37 oC for one hour. The plate was then washed with buffer four times. 100 μl of anti-mouse IgG HRP-linked antibody were added and the plate was incubated at 37oC for 30 minutes. The plate was then washed with buffer four times. Then 100 μl of 3,3',5,5' tetramethylbenzidine (TMB), the HRP-substrate, were added and the plate was incubated at 37 oC for 10 minutes. 100 μl of stop solution were then added and the plate was read on a Fusion Multilabel plate reader (Perkin Elmer), using a software program (ELISA Assay), PMT Voltage of 1000, Read time 0.2 seconds, Gain 1, Excitation 450 nm. [0199] The results of the high-throughput screenings, which included the identification of human betacellulin (clone 736345), human hepatocyte growth factor (clone 657926), and the human neuregulin-1 splice variant NRGl-β3sv87 (clone 00541754) clones, are summarized in Tables 5 and 6, and FIG. 3. The results of the high- throughput screening using betacellulin clone 736745 to treat either HT-29 or RIE cells, and whose clone profile is illustrated in FIG. 4, are displayed in Table 5. The activity percentages in HT-29 cells were 41.6% and 29.5%, while the Z-values for the assays were

1

0.52 and 0.32. Using RIE cells, the activity percentage as a result of betacellulin treatment was 10.7% and 13.1%. Furthermore, the Z-values for the assay plate were 0.7 and 0.6, indicating these were reliable assays (Z-values greater than 0.5 are considered reliable). As seen in Table 5, the standard deviation from the median using the HT-29 assay was 4.6 and 3.2. Thus, the results are significant that betacellulin increases the phosphorylation of AKT in HT-29 cells. Furthermore, as shown for example in FIG. 6, this activity was confirmed in follow-up assays, using both tagged and untagged betacellulin prepared from clonally isolated DNA in HT-29 cells as described in Example 4.

[0200] As the betacellulin concentration increased, the absorbance increased, indicating increasing phosphorylation of AKT. Moreover, recombinant betacellulin has the same effect and is quite potent with an EC50 of 0.4 nM. For NRGl-β3sv87 mediated AKT activation in HT-29 cells, the activity percentages (as compared to rIGF-1) were 31.7% with a standard deviation of 4.01. The z values for the assays were 0.71, 0.81, 0.56, and 0.77.

Example 3: Activity Profile for Betacellulin and NRGl-β3sv87 Clones Identified by High-throughput Assays for AKT Activation

[0201] FIG. 4 and FIG. 5 illustrate the results of various additional assays aimed at further characterizing the activity profile of the betacellulin and NRGl-β3sv87 clones identified in the high-throughput assay; the profile assays also used culture supernatants, i.e., conditioned media collected from 293 cells expressing either betacellulin (clone 736345) or NRGl-β3sv87 (clone 00891196). The assays, which are described below in more detail, measure the effect of these proteins on (i) tumor cell proliferation, (ii) immune cell proliferation, and (iii) phosphoenolpyruvate carboxykinase 1 (PCKl) expression. The high-throughput phosphoAKT assay on HT-29 cells (HT-29pAKTl) was included to provide a relative control. [0202] Detailed descriptions of each assay are provided below. Results were considered significant if they were at least two standard deviations from the media. As seen in FIG. 4, with standard deviations from the median of 3.0 and 2.7, the results for the betacellulin RTEpAKT 1 assay, indicating that betacellulin induces an increase in RTE cell phosphorylation, were significant and confirmed the results of the high-throuput screening assays. On the other hand, as seen in FIG. 5, the NRGl-β3sv87 CM did not demonstrate significant activity (potency) in any of the "CellType"Glo and "CellType"Pro assays. However, the standard deviation from the median using the HT-29 assay was 10.71 and 1.78. Thus, the results are significant that NRGl-β3sv87 also increases the phosphorylation of AKT in HT-29 cells.

[0203] As shown below (see, for example, in FIG. 7), this activity was confirmed in other assays in HT-29 and RTE cells, using untagged NRGl-β3sv87 prepared in 293 cells from clonally isolated DNA.

Effect of NRGl-β3sv87 on Tumor Cell Proliferation: Preparation of the Tumor Cell Lines

[0204] The objective of the tumor cell proliferation assay is to assess the effect of NRGl-β3sv87 on the proliferation of tumor cells. The assay is herein referred to as "CellType"Glo assay, in which "CellType" is replaced by the name of the particular cell type used in each particular assay. The assay is based on the principle that ATP levels increase with increased cell number upon cell proliferation. Cell proliferation can be measured, for example, by measuring ATP bioluminescence (Crouch et al. (1993), further described in Promega's CellTitreGlo Technical Bulletin No. 288). For example, ATP levels can be measured by measuring the intensity of luminescence produced in the presence of luciferase and luciferin. This assay may, for example, be performed in a 96- well plate format.

[0205] The following human tumor cell lines were used in the assay: A549 (lung carcinoma; ATCC® Number: CCL-185™), Colo205 (colorectal adenocarcinoma; ATCC® Number: CCL-222™), MDA-MB231 (estrogen receptor negative breast cancer, ATCC® Number: HTB-26™), PC3 (prostate adenocarcinoma; ATCC® Number: CRL-

1435™), U251, ,U87114 and WTl 1 (brain glioblastoma; ATCC® Number: CRL- 11544TM)₁

[0206] All cell lines were obtained from either NCI or the American Type Culture Collection, ATCC (Rockville, MD) and were cultured in RPMI- 1640 medium supplemented with 25 mM HEPES, 0.25% sodium bicarbonate, 10% fetal bovine serum, and 100 ug/ml kanamycin, unless otherwise specifically described below.

[0207] The human lung carcinoma cell line A549 was obtained from the American Type Culture Collection (ATCC) (Manassas, VA). A549 cells were routinely cultured in DMEM basal media (Invitrogen Corporation, Carlsbad, CA) supplemented with 10% fetal calf serum (Invitrogen Corporation, Carlsbad, CA), penicillin 100 units per ml, and streptomycin 100 micrograms per ml (Invitrogen Corporation, Carlsbad, CA). Cells were routinely passaged by trypsinization and dilution when they reached 90% confluence. Effect Of NRGl-β3sv87 on Immune Cell Proliferation

[0208] The objective of the NKGIo and the aMonPro4 assays is to assess the effect of NRGl-β3sv87 on the proliferation and/or survival of peripheral blood cells.

Preparation of Natural Killer Cells

[0209] Mouse peripheral blood natural killer (NK) cells were purified from the spleens of C57BL6 10 week old female mice using the NK cell isolation kit according to the manufacturer's instructions (Miltenyi Biotechnology Inc., Auburn CA). Approximately 30,000 purified NK cells were incubated with purified MGD-CSF at concentrations from 0.01 to 10 ug/ml. NK cell numbers were determined using the CellTiter-Glo Luminescent Cell Viability Assay Kit (Promega # G7571).

[0210] Human NK cells were isolated and purified from blood enriched in buffy coat cells obtained from the Stanford Blood Center (Palo Alto, CA). The blood was diluted approximately 1:5 with PBS and Ficoll (Ficoll-Paque Plus, Amersham Biosciences; Piscataway, NJ) added (12.5 ml/tube) to multiple 50 ml conical tubes, each with 25 ml of diluted blood. The Ficoll/blood mixture was centrifuged at 450 x g for 30 minutes. The peripheral blood mononuclear cell (PBMC) layers were removed, washed with DPBS IX without calcium and magnesium (Mediatech, Inc., Prince William Co. VA) and pelleted at 1000 RPM for 10 minutes. The PBMCs were washed three times in PBS by centrifugation at 1350 RPM for 10 minutes and resuspended in 40 ml PBS with 0.5% fetal calf serum (Gibco (Invitrogen), Carlsbad CA) and 2 mM EDTA (Sigma Aldrich, St. Louis MO) (PBSFE).

[0211] NK cells were enriched from the PBMCs with a human NK Cell Isolation Kit π (Miltenyi Biotechnology Inc., Auburn CA), as recommended by the manufacturer. This enrichment step utilized the "deplete" program of an autoMACS™ Separator (Miltenyi Biotechnology Inc., Auburn CA); the negative fraction, representing enriched NK cells, was collected from outlet port "negl." These cells were centrifuged at 1350 RPM for 10 minutes, the cell pellets resuspended in DMEM with 10% fetal calf serum, and diluted to a concentration of 1 x 10⁶ cells/ml. The cells were incubated with the control and test agents described below for four days at 37⁰C in an atmosphere of 7% CO2 in 96 well round bottom plates at a cell concentration of 5 x 10⁴ cells in 50 μl DMEM with 10% fetal calf serum per well. Effect of NRGl-β3sv87 on NK Cell Proliferation and/or Survival

[0212] The effect of control and test agents on the proliferation and/or survival of NK cells prepared in this manner was determined in the NKGIo screening assay by assessing the number of viable cells in the culture based on quantitation of ATP by measuring luciferase activity as described in Promega CellTitreGlo Technical Bulletin No. 288 (Promega, Madison WT). Quantitative results were read on an Lmax plate reader (Molecular Devices, Sunnyvale CA) at room temperature for 0.6 second/well. The ATP content of the wells was measured four days after plating. The results are shown in FIG. 5.

Preparation of Monocytes

[0213] Human primary monocytes were purified from PBMC using a protocol modified from a previously-described method (de Almeida, et al., Mem. Inst Oswaldo Cruz 95:221-223, 2000). To isolate human PBMC from blood, the buffy coat was diluted in a six-fold volume of PBS, then overlain onto 20-ml Ficoll in a 50 ml tube. The tubes were centrifuged at 2,000 rpm at 22°C for 20 minutes without the use of the centrifuge brake. The PMBC cells were collected from the interface, washed with PBS twice then resuspended in RPMI 5% FBS and filtered through a BD Falcon cell strainer. To purify the primary untouched monocytes from PBMC, six ml of the PBMC suspension (containing 70-120 x 106 cells) were carefully and slowly overlain onto 10 ml hyperosmotic Percoll. The cells were centrifuged at a speed of 580xg for 15 minutes without the use of the centrifuge brake. Cells at the interface were collected and washed with 50 ml of RPMI 5% FBS. This purified monocyte cell pellet was resuspended in 50 ml RPMI 5% FBS. Effect Of NRGl-β3sv87 on Monocyte Proliferation

[0214] The objective of the aMonPro4 assay was to assess the effect of NRGl- β3sv87 on the proliferation of human activated primary monocytes. Mouse IgG2a can be used to activate monocytes. [0215] The aMonPro4 assay was performed by incubating approximately 30,000 purified monocytes with conditioned medium from NRGl-β3sv87 transfected 293 cells (NRGl-β3sv87 CM) obtained as described above . After four days of incubation in RPMI with 5% FBS, monocyte proliferation was determined using the CellTiter-Glo Luminescent Cell Viability Assay Kit (Promega # G7571).

[0216] According to Cell Titer GIo assay instructions, the Cell Titer GIo buffer was thawed and equilibrated to room temperature. 10 ml of Cell Titer GIo buffer were added into an amber bottle containing the substrate. The bottle was mixed and then 100 μl of Cell Titer GIo were added to each well of cells. The plate was covered with foil and mixed for 15 minutes. 100 μl of supernatant were transferred to a new white plate and proliferation was measured with a luminometer using a 1 second integration time. The results are shown in FIG. 5.

Effect of NRGl-β3sv87 on PCKl Gene Expression on Rat H4IIE Hepatoma Cells (PCKlbDNAh4IIE Assay)

[0217] EGF reportedly affects glucose metabolism, including glycolysis and gluconeogenesis (Nowak G, Schnellmann R.G. Am J Physiol. 269:C1317-25 (1995)). Phosphoenolpyruvate carboxykinase 1 (PCKl) is a gluconeogenic enzyme, subject to regulation by multiple factors. The induction of PCKl mRNA by glucagon and its second messenger cAMP, synergized by glucocorticoids such as dexamethasone, as well as the de-induction by insulin, reportedly are part of the multi-hormonal regulation of the metabolic adaptations that occur during feeding, fasting or diabetes (Iynedjian P. B., Biochem J. 386:113-8 (2005)). .

[0218] The objective of the PCKlbDNAH4IIE assay is to determine if the novel EGF-family member of the invention, i.e. the NRGl-β3sv87, regulates dexamethasone (dex)-induced PCKl mRNA expression. It can be performed, for example, in rat H4IIE hepatoma cell lines (ATCC^® Number: CRL-1548 ™) using the Genospectra branched DNA (bDNA) detection method according to the manufacturer's instructions (Wu et al., 2003). When this assay was performed with NRGl-β3sv87 conditioned media no significant effect on dex-induced PCKl mRNA expression was observed.

[0219] The invention also provides other methods for recombinant expression of NRGl-β3sv87 in both mammalian and non-mammalian cells. Non-limiting examples are provided below. Example 4: Subcloning and Transient Expression of Betacellulin and NRGl-β3sv87 in Mammalian Cells

[0220] Based on the initial results from the high-throughput screening, the biological activity of the proteins encoded by the various clones identified as hits in the high-throughput AKT assays were subcloned for protein expression and purification into other mammalian expression vectors. To this end, complementary DNA encoding the NRGl-β3sv87 polypeptide was subcloned into the expression vectors pTT5 and pCDNA- pDEST40, transfected and expressed as both a tagged and untagged protein in human kidney epithelial 293 cells obtained from ATCC (ATCC^® Number: CRL-1573 ™). Protein levels were quantified by measuring the levels of the tag, for example, a V5His tag, by quantitative Western blot analysis . The expression vectors were transfected into adherent 293 cells using the transfection agent Fugene 6 (Roche, Nutley NJ) in complete DMEM [DMEM with 10% fetal bovine serum (FBS) and penicillin/streptomycin (100 μg/ml, 100 U/ml)], and incubated at 37⁰C in 5% CO₂ for 40 hours, after which the cells were washed with PBS and incubated for an additional 48 hours in complete DMEM. Cell supernatants were harvested, cleared of cell debris by centrifugation, and tested for biological activity (untagged cDNA) and protein expression (V5 tagged cDNA) by Western blot assay using an anti-V5 antibody. The new un-tagged NRGl-β3sv87 expression vector is herein named clone 00891196. The same method was followed for subcloning the human betacellulin cDNA of clone 736345. The resulting cDNAs were used to transfect different cell lines for recombinant protein expression (e.g. 293 cells), and the media conditioned by these cells was used for subsequent experiments to further characterize the activity of these growth factors exerted when they were used to treat epithelial cells (e.g., RIE, HT-29, HOK).

Example 5. Hepatocyte Growth Factor Increases Phosphorylation of AKT in Rat Intestine Epithelium Cells

[0221] As stated earlier, hepatocyte growth factor was also found to increase the phosphorylation of protein kinase AKT in rat intestinal epithelial (REE) cells. Cells were treated and plated according to the protocol in Example 2. HGF-containing conditioned media (50 ul; in DMEM + 5% serum) was added to the plate. [0222] The results of the high-throughput screening using HGF clone 657926, are displayed in Table 6. The activity percentages were 12.4% and 11.8%, while the standard deviations from the mean were 2.9 and 2.5, which are both significant. Also, the assay plate Z-values were 0.7 and 0.6, indicating the assay results were reliable.

[0223] In agreement with the results obtained with clonal supernatants, and as seen in FIG. 8, in RIE cells, as the concentration of recombinant HGF increases (R&D Systems), the absorbance increases, indicating the phosphorylation of AKT is increasing as well. Thus, HGF was shown by two different assays to have a dose-dependent effect on the phosphorylation of AKT in rat epithelial cells.

Example 6. Hepatocyte Growth Factor and Insulin-like Growth Factor-1 Increase Phosphorylation of AKT in Human Adenocarcinoma (HT-29) cells

[0224] To determine if HGF was also capable of increasing the phosphorylation of AKT in human epithelial cells, HT-29 cells were prepared and analyzed following the protocol of Example 2. The results of the assay using HGF clone 657926 are shown in Table 6 (upper rows). The activity percentages were 18.2% and 23.1%. The standard deviations from the median were 7.9 and 21.2, both of which are significant. Finally, the assay plate Z-values were 0.3 and 0.8, which indicate that the latter assay was very reliable. These results therefore indicate that HGF increases the phosphorylation of AKT in human epithelial HT-29 cells.

[0225] The activity of HGF was confirmed in follow-up phospho-AKT assays using commercially avaiable recombinant HGF (cat # 294-HGN, R&D Systems, Minneapolis, MN). As shown in FIG. 9, recombinant HGF and recombinant IGF-I (also a hit in the high-throughput AKT phosphorylation assay) both demonstrated increased phosphorylation of AKT in HT-29 cells. The absorbance — and therefore the phosphorylation of AKT — increases as the concentration of both rHGF and rIGF-1 increases. Thus, HGF and IGF-I are positive factor in the high-throughput assay and therefore are possible agents for treating mucosal epithelial cell damage. Example 7. AKT Phosphorylating Activity of Recombinant Growth Factors in HT- 29 cells

[0226] To compare the activities of different ErbB ligand family members and other growth factors, various commercially available recombinant factors were screened using the HT-29 assay protocol. Cells were prepared according to Example 2. Each of the recombinant test proteins was purchased from R&D Systems (Minneapolis, MN): heparin-binding EGF growth factor (HB-EGF) (Cat. # 259-HE), betacellulin (BTC) (Cat. #261-CE), epiregulin (EPR) (Cat. #1195-EP), transforming growth factor-alpha (TGF- alpha) (Cat. # 239- A), epidermal growth factor (EGF) (Cat. # 236-EG), hepatcyte growth factor (HGF) (Cat. # 294-HGN), amphiregulin (AR) (Cat. # 262- AR), heregulin A (Cat. # 296-HR), heregulin B (Cat. #396-HB), and insulin-like growth factor-1 (Cat. # 291-G1) .

[0227] As shown in FIG. 10, heregulin B showed the greatest increase in phosphorylation of AKT in HT-29 cells with increasing concentration. IGF-I and HGF also showed an increase in phosphorylation. The remaining factors were all about equal at increasing concentrations. The results were further confirmed using only two different concentrations of these growth factors (1 nM and 100 nM), as shown in FIG. 11. Example 8. AKT Phosphorylating Activity of Recombinant Growth Factors in RIE Cells

[0228] The same recombinant growth factors were also tested for their relative activities using the RIE phospho-AKT assay. Cells were prepared according to Example 2 and were tested with the same factors as in Example 6. The results are shown in FIGS. 12A and 12B, which illustrate two separate experiments using this protocol. In both experiments, IGF-I had the greatest increase in absorbance as concentration increased, followed by heregulin B. The next highest values were obtained with HGF, followed by TGF-alpha, and EGF. The remaining factors— AR, HB-EGF, BTC, epiregulin, and heregulin A — showed approximately equal activities. The results were further confirmed using only two different concentrations of these growth factors (1 nM and 100 nM), as shown in FIG. 13. Example 9. AKT Phosphorylation in Human Oral Keratinocytes

[0229] We also tested the ability of several growth factors to stimulate AKT phosphorylation in epithelial cells derived not from the intestine (as the RIE and HT-29 cells are), but from the oral surface. To this end, we used human oral keratinocytes (HOK). Cells were plated at a cell density of 15,000 cells/well in a 96-well plate. The cells were incubated at 37⁰C overnight and then washed with PBS once. The cells were then starved with 100 uL Keratinocyte Basal Medium containing 102 mM calcium for eight hours. The media was aspirated and then 50 uL fresh Keratinocyte Basal Medium was added. 50 uL test proteins were added to the rest of the assay plate (columns 1-11) and the samples incubated at 37⁰C for 30 minutes. Keratinocyte growth factor (KGF), hepatocyte growth factor (HGF), betacellulin (BTC)₃ heregulin B (HRGb), and insulin-like growth factor-1 (IGF-I) were tested. The plate was then washed once with 390 uL PBC (with Ca²⁺ and Mg²⁺). The fluid was aspirated out and 80 uL lysis buffer was added. The plate was then shaken for 2 minutes.

[0230] The phosphorylation of AKT was then measured using a phospho-AKTl ELISA plate as in Example 2. As seen in FIG. 14, HGF and IGF-I were approximately equal and had the highest absorbance value of all of the factors. The next highest value was with HRG-b, followed by BTC and finally KGF. Thus, these results indicate that HGF increases phosphorylation in HOK cells and thus can be effective on promoting the survival and/or proliferation of epithelial cells from the oral mucosa. Example 10. Effect of ErbB Ligands and Other Growth Factors on Cell Proliferation of Human Oral Keratinocytes

[0231] Our experiments further showed that, KGF, BTC, HRGb, and IGF-I are all capable of supporting HOK proliferation, whereas HGF is not. Primary oral keratinocytes (p5) were plated at 15,000 cells/well of a 96-well plate and grown in KBM (Cambrex, Cat. # CC-3104) and BEGM Singlequots (Cambrex, Cat. # CC-4175 without retinoic acid and GA (gentamycin/amphoterecin)) with calcium at the final concentration of 102 mM. The cells were starved for eight hours in KBM alone, the media was aspirated, and fresh media containing the appropriate concentration of growth factors were added. The cells were grown for 48 hours and then cellular proliferation (as measured by ATP incorporation) was determined using the Cell Titer GIo assay (Promega, Madison WI).

[0232] According to Cell Titer GIo instructions, the Cell Titer GIo buffer was thawed and equilibrated to room temperature. 10 ml of Cell Titer GIo buffer was added into an amber bottle containing the substrate. The bottle was mixed and then 100 ul of Cell Titer GIo was added to each well of cells. The plate was covered with foil and mixed for 15 minutes. 100 ul of supernatant was transferred to a new white plate and proliferation was measured with a luminometer using a 1 second integration time.

[0233] FIG. 15 shows the results of a proliferation assay using HOK cells and the various factors. As seen in FIG. 15, KGF, BTC₅ HRGb, and IGF-I all indicate high levels of proliferation. HGF, on the other hand, had little proliferative effect on HOK cells. These results together with those of Example 7 would suggest that HGF may be involved in cellular survival rather than proliferation. Example 11 : Production of Recombinant Human Betacellulin

[0234] Recombinant human betacellulin cDNA may be expressed in a number of different conventional expression systems, whether in eukaryotic cells or prokaryotic cells, to produce the recombinant protein, using methods such as those described in U.S. 5,886,141.

[0235] In order to obtain larger amounts of betacellulin for in vivo testing, we produced recombinant human betacellulin by conventional techniques by expression of a pET24/BTC expression vector in E. coli (hereafter referred to as "BTC made internally from E. coli expression"). First, we created a BTC construct in the vector pET24(+) (Novagen, EMD Biosciences Inc, San Diego, CA) without the His-Tag (which was removed during subcloning), which encoded an active recombinant human betacellulin fragment corresponding to amino acid residues Asp³²-Tyr^{π i} preceded by an initial methionine (Met) residue. The vector was transformed into E. coli Rosetta™(DE3) cells (Novagen) according to conventional methods. Individual transformants were isolated and grown according to the pET24 vector manufacturer's instructions (see pET System Manual, 10^th and 11^th Editions, Novagen). The BTC was then purified from inclusion bodies in bacterial lysates by affinity chromatography on ToyoPearl AF-Blue resin, followed by hydrophobic interaction chromatography on Phenyl-Sepharose 6 Fast Flow (high sub). Details of the process are provided below. AU standard chemicals were obtained from Sigma-Aldrich Chemical Co. (St. Louis, MO).

[0236] In the initial fermentation step, Rosetta™ (DE3) cells were grown in Luria Bertani (LB) broth (supplemented with 50 μg/ml of kanamycin and 34 μg/ml of chloramphenicol) at 37°C in standard bacterial fermentation vessels, with agitation, to an optical density of about 5 at the wavelength of about 600 nm. This was followed by four hours of induction of expression of rhBTC protein in the presence of 1 mM isopropyl β-D- thiogalactopyranoside (Sigma Chemical Co., St. Louis, MO).

[0237] The process of harvesting and solubilization of inclusion bodies to obtain the BTC protein was done as follows. BTC, produced as insoluble inclusion bodies in the bacteria, was purified as follows. Cells were harvested by centrifugation and the cell pellets resuspended in 20 mM Tris-HCl at pH 8.0 containing 10 mM EDTA and 1% Triton X-100 in a volume of that was equal to 0.1 volume of the initial culture medium. Thereafter, cells were lysed by pressure homogenization (with a Microfluidizer), and the inclusion bodies (IB) recovered by centrifugation at 20,000 x g for 15 min at 4⁰C. The IB pellets were washed twice with the same volume of 20 mM Tris-HCl at pH 8.0 containing 10 mM EDTA and 1% Triton X-100 and resuspended to 3 mg of pellet per ml of solubilization buffer (100 mM Tris-HCl at pH8.0 containing 7 M guanidine hydrochloride and 5 mM dithiothreitol). The BTC protein was extracted from the IB by incubation at 4⁰C for an average of one hour without agitation.

[0238] The next step entailed re-folding of the recombinant BTC, which proceeded as follows. After extraction, the solubilized protein concentration was adjusted to 2.5 mg/ml and diluted 25-fold further with refolding buffer (50 mM Tris-HCl at pH 8.0 containing 2 M urea, 0.5 mM oxidized glutathione, 1 mM reduced glutathione, and 0.1 M arginine) and incubated for approximately 20 hr at 4⁰C, during which period the BTC was renatured or refolded. Refolding was terminated by adjusting the pH to 5.0 with concentrated 3 M sodium acetate (pH 4.75). The refolded BTC protein was dialyzed against phosphate buffered saline (PBS) (without calcium and magnesium) diluted 1 :3 in purified water. The dialyzate containing the refolded BTC was clarified by centrifugation at 5,000 x g.

[0239] Next, BTC was purified by chromatography. Refolded BTC was applied to a Toyopearl AF-Blue HC-650 column (1.6 cm x 20 cm) (Tosoh Bioscience LLC, Montgomeryville, PA ) equilibrated with 10 mM potassium phosphate buffer pH 7.0 buffer containing 50 mM NaCl (Buffer A). Proteins were eluted at 3 ml/min with a continuous gradient of Buffer A to Buffer B (10 mM potassium phosphate buffer at pH 7.0 containing 1.5 M NaCl ) established over 20 column volumes {i.e., a linear gradient of 0 to 1.5 M NaCl). The desired BTC-containing fractions were collected and pooled. Ammonium Sulfate was added to a final concentration of 1.3M for further purification by hydrophobic interaction chromatography over a Phenyl Sepharose™ 6 FF/high sub (1.6 cm x 20 cm) (GE Healthcare, Piscataway, NJ) equilibrated with 10 mM potassium phosphate buffer at pH 7.0 containing 1.5 M NH₄SO₄ (Buffer C). The BTC protein was eluted with a continuous gradient of Buffer C to Buffer D (10 mM potassium phosphate buffer pH 7.0 containing 50 mM NaCl) established over 25 column volumes at the flow rate of 3 ml/min. The fraction containing the purified BTC protein (as determined by conventional SDS-PAGE and Coomassie blue/Silver Stain protein visualization techniques) was concentrated by tangential flow filtration and the concentrate was dialyzed against PBS (without Ca^2-1VMg²⁺). [0240] Removal of endotoxin was accomplished by further purification by Cellufme™ ET clean (Chisso Corporation, Tokyo, Japan) chromatography (Sakata, M. et al. American Biotechnol. Lab. 20:36 (2002)) following the manufacturer's instructions. Briefly, the dialyzed BTC was applied to a Cellufme™ ET clean column (10 x 0.9 cm (I.D.); 9.6 ml) equilibrated with PBS, and collected in the flow through at the flow rate of 0.5 ml/min. The final BTC solution (in PBS without Ca²⁺ZMg²⁺) typically contained less than 2 E.U./mg of protein, as assessed by the Limulus amoebocyte lysate (LAL) assay (Cambrex, Walkersville, MD). Example 12: Betacellulin Fusion Proteins Have Extended Half-Lives

[0241] In this study, we demonstrated that pharmacokinetic properties of betacellulin could be improved by conjugating betacellulin with polyethylene glycol (PEG) or by fusing betacellulin to the Fc region of an immunoglobulin. Part A: PEGylation of Betacellulin

[0242] Human betacellulin expressed in E. coli and purified as previously described (see Example 11) was pegylated as follows. A number of test reaction conditions were tested for two PEG reagents namely, mPEG-SMB-20K and mPEG- ButyrALD-20K (Nektar Therapeutics, Huntsville, AL) in order to identify conditions that provide the highest yield of active, mono-PEGylated betacellulin. For mPEG-SMB-20K, 18 reactions were performed as shown in the table below, varying betacellulin concentration (1 or 2.5 mg/mL), molar ratio of betacellulin: PEG (1:1, 1:2 or 1:5), buffer (potassium phosphate pH 7.0, potassium phosphate pH 7.5, or borate pH 9.0). Aliquots were taken at 30 min, 1 hr, 4 hr, and 24 hr to monitor reaction progress.

Reaction Conditions for PEGylation of BTC with mPEG-SMB-20K

[0243] The progress of the PEGylation reaction was monitored by separating aliquots of each reaction at different time points by SDS-PAGE (4-12% Bis-Tris), and staining the proteins with Coomassie blue, following methods standard in the art. PEG addition was observed as decreased migration of the protein in the gel; unreacted BTC migrated to just above the dye front, monoPEGylated BTC migrated to near the about 51 kDa molecular weight marker, and multiply PEGylated species ran between the about 64 kDa and the about 191 kDa markers. The reactions proceeded quickly, with significant product observed even at 30 min. A variety of multiply PEGylated species were observed at 24 hr.

[0244] For PEGylation of betacellulin with the reagent mPEG-ButyrALD-20K, 18 reactions were performed, varying betacellulin concentration (1 or 2.5 mg/mL), molar ratio of betacellulin: PEG (1:1, 1:2, or 1:5), and buffer (potassium phosphate pH 7.0, potassium phosphate pH 6.0, or acetate pH 5.0). In all cases, a five-fold molar excess (versus betacellulin) of sodium cyanoborohydride was used. Aliquots were taken at 30 min, 1 hr, 4 hr, and 24 hr to monitor reaction progress.

[0245] Reaction progress was also monitored by Coomassie blue stained SDS- PAGE (4-12% Bis-Tris). PEG addition to betacellulin was observed as decreased migration in the gel. These reactions proceeded more slowly and approached completion at about 24 hr. As expected, this reagent produced mostly mono-PEGylated betacellulin, which migrated near the 51 kDa molecular weight marker. At 24 hr, all the PEGylation the reactions were quenched by addition of excess glycine. The mPEG-SMB-20K and mPEG-ButyrALD-20K reactions were pooled and fractionated by size exclusion chromatography using S75 and S200 columns (Amersham Pharmacia Biotech, GE Healthcare Bio-Sciences Corp., Piscataway, NJ). Peaks corresponding to PEGylated betacellulin were pooled, diluted to 40 microM (based on absorbance at 280 nm), and tested for activity.

[0246] Betacellulin activity, in terms of ErbB receptor activation, was determined using an in vitro HeLa 229 (ATCC number CCL2.1) cell based binding assays and a phospho-EGFR pY1068 ELISA based assays according to the manufacturer's instructions (Cat. Number: KHR9081, BioSource International, Inc. Camarillo, California). Under these reaction and assay conditions, the activity of the PEGylated betacellulin produced using the mPEG-SMB-20K reagent was approximately 3-fold lower than the activity of unreacted betacellulin, while the activity of the PEGylated betacellulin produced using the mPEG-ButyrALD-20K reagent was reduced by less than 50%. Activit of PEG lated Betacellulin

Part B: Betacellulin-Fc Fusion Protein

[0247] Murine betacellulin (containing amino acid residues 1 - 111 of the full- length protein) was fused to the Fc portion of the human immunoglobin IgGl. The fusion construct was subcloned into pERESpuro3 expression vector (Cat# 6986-1, Clontech Laboratories, Inc., Mountain View, CA). The vector was stably transfected into CHO-S cells using standard transfection methods, and the protein was produced using a 10 L Wave fermenter (Cat# BASE2050EH, Biotech, LLC; Somerset, New Jersey) and CD- CHO medium (Cat# 10743-029, Invitrogen hie, Carlsbad, California). After eight days of culturing under these conditions, the cell supernatants were harvested. The fusion protein mouse BTC-human Fc was purified by affinity chromatography using Protein A Sepharose 4 Fast Flow resin (Cat# 17-5280-02, GE Healthcare, Piscataway, NJ) following the manufacturer's recommendations and dialyzed in PBS. The activity of the purified mouse BTC-human Fc fusion protein (betacellulin-Fc fusion) was also tested by the phospho-ErbB receptor assay described above. Part C: Pharmacokinetic Assay of PEGylated and Fc-Fusion Betacellulin

[0248] To determine whether PEGylation or Fc fusion affects the pharmacokinetic properties of betacellulin, unreacted betacellulin, betacellulin-Fc fusion, and PEGylated betacellulin were prepared, administered to mice, and monitored for disappearance from the bloodstream.

[0249] The PEGylation reaction conditions for the betacellulin protein used in this test were as follows: 2.5 mg/mL betacellulin, 5-fold molar excess of mPEG-ButyrALD- 2OK and sodium cyanoborohydride, potassium phosphate pH 7.0 buffer, and 24 hr reaction time followed by quenching with excess glycine pH 7.0. The reaction products were prepared for injection by overnight dialysis against 2x PBS. The success of the reaction was confirmed by Coomassie-stained SDS-PAGE gels, as described in Parts B and C above. The concentration of the PEG-BTC, the BTC-Fc (prepared as described in Part C), and the BTC (prepared as described in Example 11) protein solutions used for this test was determined by Bradford assay. Samples were prepared for injection by diluting each of the betacellulin protein solutions to 0.125 mg/mL in PBS supplemented with 0.1% BSA (Sigma #A3059, St. Louis MO).

[0250] Eight-week old C57B1/6 mice were injected intravenously with 200 microliter of BTC, PEG-BTC, or BTC-Fc at a dose of 1 mg/kg BTC, and blood samples were collected at 2, 30, 120, and 1440 min. For each betacellulin type tested, six mice were injected with the test material. Then, three of the six mice were bled from the retro- orbital sinus at 2 min and then again by cardiac puncture at the 120 min time point. For the other three mice, blood was collected from the retro-orbital sinus at 30 min and then by cardiac puncture at 1440 min. AU blood samples were collected into plasma collection "Microtainer" tubes with EDTA from Becton Dickinson (Cat# 365973, Franklin Lakes, New Jersey) and then spun immediately to obtain plasma.

[0251] Human betacellulin concentrations in the BTC and PEG-BTC plasma samples and murine betacellulin concentrations in the BTC-Fc plasma samples were determined using ELISA assays. Standard curves were generated using 0.34 - 250 pM of murine and human betacellulin. The plasma samples were diluted 10, 100, and 5000- fold in 10% FCS/ PBS solution to ensure that the signal was in the linear region of the standard curve. ELISA concentrations, determined for each plasma sample at 2 min, 30 min, 120 min and 1080 min post-injection, were calculated to be as follows:

Pharmacokinetics of Betacellulin

[0252] To prepare samples for the Western blot, 3.25 microliter plasma from each mouse from the same group at each timepoint was pooled. Plasma aliquots were separated in nonreducing Tris-Tricine gels (10-20%), and the proteins visualized by standard Western blot analysis. The results are shown in FIG. 16. Human betacellulin was detected using R&D Systems (Minneapolis, MN) antibody #261, and BTC-Fc was detected using an HRP-labeled anti-human Fc antibody ( Cat#209-035-088; Jackson LnmunoResearch, West Grove, PA) combined with an ECL detection system GE Healthcare, Piscataway, NJ). PEG-BTC migrated at approximately 45 kDa, unreacted BTC migrated at approximately 10 kDa, and the location of BTC-Fc is as shown on the left in FIG. 16.

[0253] From the results of both the ELISA and Western blot analyses, we determined that both PEG-BTC and BTC-Fc were cleared from mouse plasma significantly more slowly than unmodified betacellulin and thus have an extended pharmacokinetic half-life.

Example 13: Summary of Structural Properties of Betacellulin and other Growth Factors of the Invention: Protein Sequence, Nucleotide Sequence, and Protein Domains

[0254] In Table 1, "Protein and Nucleotide Sequence Identification," we provide some additional characteristics of a subset of the betacellulin polypeptides and other ErbB ligands of the invention. Each polypeptide is identified by the internal reference designation (FP ID), as shown in the first column. The nucleotide sequence identification number for the open reading frame of the nucleic acid sequence (Nl) is shown in the second column. The amino acid sequence identification number for the polypeptide sequence (Pl) is shown in the third column. The nucleotide sequence identification number for the entire nucleic acid sequence that contains UTR (NO) is shown in the fourth column. The fifth column shows an internal clone reference designation (Clone ID). The sixth column list annotations for some of the proteins.

Table 1: Protein and Nucleotide Sequence Identification

[0255] The Pfam system is an organization of protein sequence classification and analysis, based on conserved protein domains. We performed a Pfam analysis of betacellulin and other ErbB ligands to gather more information about their structure and possible activity. The Pfam system can be publicly accessed in a number of ways (for review and links to publicly available websites see Finn, R.D. et al. Nucleic Acids Res. 34:D247-D251, (2006)). Protein domains are portions of proteins that have a tertiary structure and sometimes have enzymatic or binding activities; multiple domains can be connected by flexible polypeptide regions within a protein. Pfam domains can comprise the N-terminus or the C-terminus of a protein, or can be situated at any point in between. The Pfam system identifies protein families based on these domains and provides an annotated, searchable database that classifies proteins into families.

[0256] hi Table 2, "Pfam Coordinates and Annotations of Betacellulin and other ErbB Ligand Sequences," we provide the FP IDs of the proteins (FP ID) in the first column. The second column lists the Source ID. The third column lists the Pfam domains of each polypeptide (Pfam). The fourth column lists the coordinates of each Pfam domain, in terms of amino acid residues, beginning with "1" at the N-terminus of the full-length polypeptide. The fifth column lists an annotation from a public database.

Table 2: Pfam Protein Coordinates and Annotations of Betacellulin and other ErbB Ligand Sequences

[0257] In Table 3, "Transmembrane Domain Coordinates for Betacellulin and other ErbB Ligands," we provide some physical properties of a subset of proteins described throughout the specification. The first column lists the FP ID. The second column shows the cluster ID. The third column classifies betacellulin as a type 1 single transmembrane domain (STM) membrane protein. The fourth column shows the predicted length of each polypeptide, expressed as the number of amino acid residues. The fifth column specifies the result of an internally developed algorithm that predicts whether a sequence is secreted (Tree Vote), with "1" being a high probability that the polypeptide is secreted and "0" being a low probability that the polypeptide is secreted. The sixth column lists the number of transmembrane regions (TM). The seventh column list the amino acid coordinates of the transmembrane domains.

Table 3: Transmembrane Domain Coordinates for Betacellulin and other ErbB Li ands

[0258] In Table 4, "Signal-Peptide and Non-Transmembrane Domain Coordinates for Betacellulin and other ErbB ligands," we provide some additional physical characteristics for these proteins. The first column lists the FP ID. The second column lists the coordinates of the non-transmembrane regions (Non-TM Coordinates.). The third column lists a signal peptide (or secretory leader) position of each polypeptide (Signal Peptide coordinates) based on positions of the starting and end amino acid residues. The fourth column lists the corresponding mature protein coordinates, which are the amino acid residues of the mature polypeptide after cleavage of the signal peptide (or secretory leader) sequence of each polypeptide (Mature Protein coordinates). The fifth and six columns list possible alternative signal peptide and mature protein coordinates, respectively.

Table 4: Signal-Peptide and Non-Transmembrane Domain Coordinates for Betacellulin and other ErbB li ands

[0259] In Table 5 we summarize some of the results of the high-throughput screening assay for phosphorylation of AKT in response to betacellulin, as set forth in Examples 1 and 2. The results show data from assays using both rat intestinal epithelial (RIE) cells and human adenocarcinoma (HT-29) cells.

Table 5: Results of High-Throughput Screening Assay for Phosphorylation of AKT in Response to Betacellulin

[0260] In Table 6, we summarize the results of the high-throughput screening assay for phosphorylation of AKT in response to hepatocyte growth factor (HGF), as set forth in Examples 1, 2 and 5. The results show data from assays using both RIE cells and HT-29 cells. Table 6. Results of High-Throughput Screening Assay for Phosphorylation of AKT in Response to He atoc te Growth Factor

INDUSTRIAL APPLICABILITY

[0261] The compositions and methods of the invention are useful in the treatment and prevention of epithelial diseases. They can be used, for example, to treat or prevent mucosal diseases, including mucositis and inflammatory bowel disease.

Claims

1. A composition comprising an effective amount of a first therapeutic agent for treatment of a mucosal disorder in a subject and a pharmaceutically acceptable carrier, wherein the therapeutic agent comprises at least a first growth factor or a variant or an active fragment thereof, wherein the growth factor is other than keratinocyte growth factor (KGF) alone or hepatocyte growth factor alone.

2. The composition of claim 1, wherein the growth factor comprises insulin-like growth factor-I (IGF-I).

3. The composition of claim 1 , wherein the growth factor comprises a member of the ErbB ligand family.

4. The composition of claim 3, wherein the member of the ErbB ligand family comprises betacellulin (BTC).

5. The composition of claim 3, wherein the member of the ErbB ligand family comprises heregulinl-beta (HRGl -beta).

6. The composition of claim 3, wherein the member of the ErbB ligand family comprises transforming growth factor-alpha (TGF-alpha).

7. The composition of claim 3, wherein the member of the ErbB ligand family comprises heparin binding-epidermal growth factor (HB-EGF).

8. The composition of claim 3, wherein the member of the ErbB ligand family comprises epiregulin.

9. The composition of claim 3, wherein the member of the ErbB ligand family comprises EGF.

10. The composition of claim 3, wherein the member of the ErbB ligand family comprises amphiregulin (AR).

11. The composition of claim 3 , wherein the member of the ErbB ligand family comprises heregulinl -alpha (HRGl -alpha).

12. The composition of claim 1, further comprising a second therapeutic agent.

13. The composition of claim 12, wherein the second therapeutic agent is a second growth factor.

14. The composition of claim 13, wherein the second therapeutic agent comprises keratinocyte growth factor (KGF).

15. The composition of claim 13, wherein the second therapeutic agent comeseses hepatocyte growth factor (HGF).

16. The composition of claim 1, wherein the subject suffers from inflammatory bowel disease.

17. The composition of claim 1 , wherein the subject suffers from oral mucositis.

18. Use of the composition of any of claims 1 - 15 for treatment of a mucosal disorder in. a subject.

19. A long-acting therapeutic agent comprising a growth factor or a variant or an active fragment thereof and a fusion partner, wherein the growth factor or variant or fragment promotes survival and/or proliferation of a mucosal cell and a second molecule that confers an extended half life to the growth factor in a subject.

20. The long-acting therapeutic agent of claim 19, wherein the growth factor comprises keratinocyte growth factor (KGF).

21. The long-acting therapeutic agent of claim 19. wherein the growth factor comprises hepatocyte growth factor (HGF).

22. The long-acting therapeutic agent of claim 19, wherein the growth factor comprises insulin-like growth factor-I (IGF-I).

23. The long-acting therapeutic agent of claim 19, wherein the growth factor comprises a member of the ErbB ligand family.

24. The long-acting therapeutic agent of claim 23, wherein the member of the ErbB ligand family comprises betacellulin (BTC).

25. The long-acting therapeutic agent of claim 23, wherein the member of the ErbB ligand family comprises heregulinl-β (HRGl -beta).

26. The long-acting therapeutic agent of claim 23, wherein the member of the ErbB ligand family comprises transforming growth factor-alpha (TGF-alpha).

27. The long-acting therapeutic agent of claim 23, wherein the member of the ErbB ligand family comprises epiregulin,

28. The long-acting therapeutic agent of claim 23, wherein the member of the ErbB ligand family comprises heparin brading-epidermal growth factor (HB-EGF).

29. The long-acting therapeutic agent of claim 23, wherein the member of the ErbB ligand family comprises EGF.

30. The long-acting therapeutic agent of claim 23, wherein the member of the ErbB ligand family comprises amphiregulin (AR).

31. The long-acting therapeutic agent of claim 19, wherein the second molecule comprises a polymer, a polypeptide, a succinyl group, a lipid group, or serum albumin.

32. The long-acting therapeutic agent of claim 31 , wherein the polymer comprises a polyethylene glycol moiety (PEG).

33. The long-acting therapeutic agent of claim 31, wherein the polypeptide comprises at least a portion of an Fc molecule.

34. The long-acting therapeutic agent of claim 33, wherein the Fc molecule is a variant Fc molecule,

35. A method for treating an epithelial disease in a subject comprising:

(a) providing the composition of any of claims 1 - 15 or the long-acting therapeutic agent of any of claims 19 — 34; and

(b) administering an effective amount of the therapeutic agent to the subject.

36. The method of claim 35, further comprising administering chemotherapy, radiation therapy or another biologic or small molecule to the subject

37. The method of claim 35, wherein the epithelial disease comprises mucositis,

38. The method of claim 37, wherein the mucositis comprises oral mucositis or intestinal mucositis,

39. The method of claim 35, wherein the epithelial disease comprises an inflammatory bowel disease.

40. The method of claim 39, wherein the inflammatory bowel disease comprises Crohn's disease.

41. The method of claim 39, wherein the inflammatory bowel disease comprises ulcerative colitis,

42. The method of claim 35, wherein the composition is administered as a mouth wash, as an oral, gel, or as an enema.

43_. The method of claim 36, wherein the biologic or small molecule comprises a tumor necrosis factor-alpha (TNF-alpha) inhibitor.

44. The method of claim 36, wherein, the biologic or small molecule comprises an. anti- inflammatory agent or an immunomodulatory agent. 45, The method of claim 35, wherein the composition or long-acting therapeutic agent is administered intravenously, orally, subcutaneously, intramuscularly, transdermalIy, buccally, intranasally, by inhalation, by suppository or implantation.