US 2011 009 1543A1 (19) United States (12) Patent Application Publication (10) Pub. No.: US 2011/009 1543 A1 Prior et al. (43) Pub. Date: Apr. 21, 2011

(54) ORAL IDELIVERY OF MODIFIED cation No. 60/334,059, filed on Nov.30, 2001, provi TRANSFERRINFUSION PROTEINS sional application No. 60/406,977, filed on Aug. 30, 2002. (75) Inventors: Christopher P. Prior, King of Publication Classification Prussia, PA (US); Homayoun Sadeghi, King of Prussia, PA (US); (51) Int. Cl. A69/48 (2006.01) Andrew Turner, King of Prussia, A638/40 (2006.01) PA (US) A6IP3L/2 (2006.01) Assignee: BOREXIS A6IP35/00 (2006.01) (73) A6IP3/04 (2006.01) PHARMACEUTICAL A6IP3L/00 (2006.01) CORPORATION, Prussia, PA A6IP 7/06 (2006.01) (US) A6IP 9/00 (2006.01) A6IP37/06 (2006.01) (21) Appl. No.: 10/515,232 A6IP 25/28 (2006.01) A6IP3/10 (2006.01) (22) PCT Fled: Aug. 28, 2003 A638/28 (2006.01) A638/26 (2006.01) (86) PCT NO.: PCT/USO3/26778 A6IR 9/00 (2006.01) A6IR 9/28 (2006.01) S371 (c)(1), (52) U.S. Cl...... 424/463: 514/21.2: 514/3.7: 514/19.3: (2), (4) Date: Feb. 22, 2006 514/4.8: 514/2.3: 514/13.5; 514/16.4: 514/17.9: 514/17.7: 514/6.9; 514/5.9; 514/7.2: 424/400; Related U.S. Application Data 424/474 (63) Continuation-in-part of application No. 10/378,094, (57) ABSTRACT filed on Mar. 4, 2003, now Pat. No. 7,176,278, which is a continuation-in-part of application No. 10/231,494, Pharmaceutical compositions containing modified fusion filed on Aug. 30, 2002, now abandoned. proteins of transferrin and therapeutic proteins or peptides with increased serum half-life or increased serum stability are (60) Provisional application No. 60/406,977, filed on Aug. disclosed. Preferred fusion proteins include those modified so 30, 2002, provisional application No. 60/460,829, that the transferrin moiety exhibits no or reduced glycosyla filed on Apr. 8, 2003, provisional application No. tion, but does exhibit binding to iron and/or the transferrin 60/315,745, filed on Aug. 30, 2001, provisional appli receptor. Such fusion proteins may be administered orally. Patent Application Publication Apr. 21, 2011 Sheet 1 of 6 US 2011/009 1543 A1

F.G. 1 Alignment of N and C domains of Transferrin to show IrOn Binding. residues. N --VED-RVREYSEE RDHMKSVPSDGPSVASIKKASYLCE C PE EEESR ups up a E. E. A E.

f SEGSKEETYSSDSGFQis s s EW L i E.R.E.E.E.RSSSSSSSSSCDLPEPSKPLEKAYANESSADGTDFPQ

- C VTECSGNEva ETEREDTVCLALADENTYEKE EE N ------C LEACTFRRP ( ); similarity 2: identity . . Amino acid residues involved in iron binding (). N domain . C domain Asp 63 . Asp 392 Tyr 95 - Tyr 426 Tyr 188 . o Tyr 514 His 249 His 585 Indirectly involved in iron binding (t) Lys 296 Arg 632 Binding of carbonate ion (t) Thr 120 Thr 452 Arg 124 Arg 456 Ala .126 A a 458 Gly 127 Gly 459 Patent Application Publication Apr. 21, 2011 Sheet 2 of 6 US 2011/009 1543 A1

NZ NZ - SSSŠ N

------

Patent Application Publication Apr. 21, 2011 Sheet 3 of 6 US 2011/009 1543 A1

-WZ'5013

Patent Application Publication Apr. 21, 2011 Sheet 4 of 6 US 2011/009 1543 A1

SS

G07uglumH G07,!qq2\}I07! 98 #017ºsnoW807Ð SJ0H 90;?u!A08 6076?d Patent Application Publication Apr. 21, 2011 Sheet 5 of 6 US 2011/009 1543 A1

Z8G |089 |289 099

22 Z2 22

Patent Application Publication Apr. 21, 2011 Sheet 6 of 6 US 2011/009 1543 A1

FG, 3 PEPTIDE DELIVERY USING RECOMBINANT TRANSFERRIN 279 289

33 US 2011/009 1543 A1 Apr. 21, 2011

ORAL DELVERY OF MODIFIED 0007 Serum transferrin (Tf) is a monomeric glycoprotein TRANSFERRINFUSION PROTEINS with a molecular weight of 80,000 daltons that binds iron in the circulation and transports it to various tissues via the RELATED APPLICATIONS transferrin receptor (TfR) (Aisen et al. (1980) Ann. Rev. Biochem. 49: 357-393; MacGillivray et al. (1981) J. Biol. 0001. This application claims the benefit of U.S. Provi Chem. 258:3543-3553, U.S. Pat. No. 5,026,651). Tf is one of sional Application 60/460,829, filed on Apr. 8, 2003, U.S. the most common serum molecules, comprising up to about application Ser. No. 10/378,094, filed Mar. 4, 2003, and U.S. 5-10% of total serum proteins. Carbohydrate deficient trans Provisional Application 60/406,977, filed Aug. 30, 2002, all ferrin occurs in elevated levels in the blood of alcoholic of which are herein incorporated by reference in their entirety. individuals and exhibits a longer half life (approximately 14-17 days) than that of glycosylated transferrin (approxi FIELD OF THE INVENTION mately 7-10 days). See van Eijket al. (1983) Clin. Chim. Acta 132:167-171, Stibler (1991) Clin. Chem. 37:2029-2037 0002 The present invention relates to orally administer (1991), Arndt (2001) Clin. Chem. 47(1):13-27 and Stibler et able therapeutic proteins or peptides with extended serum al. in “Carbohydrate-deficient consumption'. Advances in stability or serum half-life, particularly to therapeutic pro the Biosciences, (Ed Nordmann et al.), Pergamon, 1988, Vol. 71, pages 353-357). teins or peptides fused to or inserted in a transferrin molecule 0008. The structure of Tf has been well characterized and modified to reduce or inhibit glycosylation. the mechanism of receptor binding, iron binding and release and carbonate ion binding have been elucidated (U.S. Pat. BACKGROUND OF THE INVENTION Nos. 5,026,651, 5,986,067 and MacGillivray et al. (1983).J. 0003. Therapeutic proteins or peptides in their native state Biol. Chem. 258(6):3543-3546). 0009 Transferrin and antibodies that bind the transferrin or when recombinantly produced are typically labile mol receptor have also been used to deliver or carry toxic agents to ecules exhibiting short periods of serum stability or short in tumor cells as cancer therapy (Baselga and Mendelsohn, vivo circulatory half-lives. In addition, these molecules are 1994), and transferrin has been used as a non-viral gene often extremely labile when formulated, particularly when therapy vector to deliver DNA to cells (Frank et al., 1994: formulated in aqueous solutions for diagnostic and therapeu Wagner et al., 1992). The ability to deliver proteins to the tic purposes. central nervous system (CNS) using the transferrin receptor 0004 Few practical solutions exist to extend or promote as the entry point has been demonstrated with several proteins the stability in vivo or in vitro of proteinaceous therapeutic and peptides including CD4 (Walus et al., 1996), brain molecules. Polyethylene glycol (PEG) is a substance that can derived neurotrophic factor (Pardridge et al., 1994), glial be attached to a protein, resulting in longer-acting, Sustained derived neurotrophic factor (Albeck et al.), a vasointestinal activity of the protein. If the activity of a protein is prolonged peptide analogue (Bickel et al., 1993), a beta-amyloid peptide by the attachment to PEG, the frequency that the protein (Saito et al., 1995), and an antisense oligonucleotide (Par needs to be administered may be decreased. PEG attachment, dridge et al., 1995). however, often decreases or destroys the protein's therapeutic 0010 Transferrin fusion proteins have not, however, been activity. While in some instance PEG attachment can reduce modified or engineered to extend the in vivo circulatory half immunogenicity of the protein, in other instances it may life of a therapeutic protein nor peptide or to increase bio increase immunogenicity. availability by reducing or inhibiting glycosylation of the Tf 0005. Therapeutic proteins or peptides have also been sta moiety nor to reduce or prevent iron and/or Tf receptor bind bilized by fusion to a protein capable of extending the in vivo 1ng. circulatory half-life of the therapeutic protein. For instance, SUMMARY OF THE INVENTION therapeutic proteins fused to albumin or to antibody frag 0011. As described in more detail below, the present ments may exhibit extended in vivo circulatory half-life when invention includes orally administerable modified Tffusion compared to the therapeutic protein in the unfused State. See proteins comprising at least one therapeutic protein, polypep U.S. Pat. Nos. 5,876,969 and 5,766,883. tide or peptide entity, wherein the Tfportion is engineered to 0006 Another serum protein, glycosylated human trans extend the serum half-life or bioavailability of the molecule. ferrin (Tf) has also been used to make fusions with therapeu The invention also includes pharmaceutical formulations and tic proteins to target delivery to the interior of cells or to carry compositions formulated for oral administration comprising agents across the blood-brain barrier. These fusion proteins the fusion proteins, methods of extending the serum stability, comprising glycosylated human Tf have been used to target serum half-life and bioavailability of a therapeutic protein by nerve growth factor (NGF) or ciliary neurotrophic factor fusion to modified transferrin, nucleic acid molecules encod (CNTF) across the blood-brain barrier by fusing full-length ing the modified Tffusion proteins, and the like. Moreover, Tf to the agent. See U.S. Pat. Nos. 5,672,683 and 5,977.307. the present invention relates to methods of treating a patient In these fusion proteins, the Tf portion of the molecule is with a modified Tffusion protein by oral administration. glycosylated and binds to two atoms of iron, which is required Further, the present invention relates to methods of treating a for Tfbinding to its receptor on a cell and, according to the patient with a modified Tffusion protein by intranasal admin inventors of these patents, to target delivery of the NGF or istration. Additionally, the present invention relates to meth CNTF moiety across the blood-brain barrier. Transferrin ods of treating a patient with a modified Tffusion protein by fusion proteins have also been produced by inserting an pulmonary administration. HIV-1 protease target sequence into Surface exposed loops of glycosylated transferrin to investigate the ability to produce BRIEF DESCRIPTION OF THE DRAWINGS another form of Tffusion for targeted delivery to the inside of 0012 FIG. 1 shows an alignment of the N and C Domains a cell via the Tf receptor (Ali et al. (1999) J. Biol. Chem. of Human (Hu) transferrin (Tf) (SEQ ID NO:3) with simi 274(34):24.066-24073). larities and identities highlighted. US 2011/009 1543 A1 Apr. 21, 2011

0013 FIG. 2A-2B shows an alignment of transferrin understood by one of ordinary skill in the art to which this sequences from different species (SEQIDNOs: 3 and 48-54). invention belongs. Although any methods and materials simi Light shading: Similarity; Dark shading: Identity. lar or equivalent to those described herein can be used in the 0014 FIG.3 shows the location of a number of Tf surface practice or testing of the present invention, the preferred exposed insertion sites for therapeutic proteins, polypeptides methods and materials are described. or peptides. DEFINITIONS DETAILED DESCRIPTION 0019. As used herein, an "amino acid corresponding to’ or General Description an “equivalent amino acid in a transferrin sequence is iden tified by alignment to maximize the identity or similarity 0015. It has been discovered that a therapeutic protein between a first transferrin sequence and at least a second (e.g., a polypeptide, antibody, or peptide, or fragments and transferrin sequence. The number used to identify an equiva variants thereof) can be stabilized to extend the in vivo circu lent amino acid in a second transferrin sequence is based on latory half-life and/or retain the therapeutic protein's activity the number used to identify the corresponding amino acid in for extended periods of time in vivo by genetically fusing or the first transferrin sequence. In certain cases, these phrases chemically conjugating the therapeutic protein, polypeptide may be used to describe the amino acid residues in human or peptide to all or a portion of modified transferrin sufficient transferrin compared to certain residues in rabbit serum trans to extend its halflife in serum. The modified transferrin fusion ferrin. proteins include a transferrin protein or domain covalently 0020. As used herein, the term “biological activity” refers linked to a therapeutic protein or peptide, wherein the trans to a function or set of activities performed by a therapeutic ferrin portion is modified to contain one or more amino acid molecule, protein or peptide in a biological context (i.e., in an substitutions, insertions or deletions compared to a wild-type organism or an in vitro facsimile thereof). Biological activi transferrin sequence. In one embodiment, Tffusion proteins ties may include but are not limited to the functions of the are engineered to reduce or prevent glycosylation within the therapeutic molecule portion of the claimed fusion proteins, Tfor a Tf domain. In other embodiments, the Tf protein or Tf such as, but not limited to, the induction of extracellular domain(s) is modified to exhibit reduced or no binding to iron matrix secretion from responsive cell lines, the induction of or carbonate ion, or to have a reduced affinity or not bind to a hormone secretion, the induction of chemotaxis, the induc Tf receptor (TfR). tion of mitogenesis, the induction of differentiation, or the 0016. The present invention therefore includes transferrin inhibition of cell division of responsive cells. A fusion protein fusion proteins, therapeutic compositions comprising the or peptide of the invention is considered to be biologically fusion proteins, and methods of treating, preventing, or ame active if it exhibits one or more biological activities of its liorating diseases or disorders by administering the fusion therapeutic protein's native counterpart. proteins. A transferrinfusion protein of the invention includes 0021. As used herein, “binders' are agents used to impart at least a fragment or variant of a therapeutic protein and at cohesive qualities to the powdered material. Binders, or least a fragment or variant of modified transferrin, which are “granulators' as they are sometimes known, impart a cohe associated with one another, preferably by genetic fusion siveness to the tablet formulation, which insures the tablet (i.e., the transferrin fusion protein is generated by translation remaining intact after compression, as well as improving the of a nucleic acid in which a polynucleotide encoding all or a free-flowing qualities by the formulation of granules of portion of a therapeutic protein is joined in-frame with a desired hardness and size. Materials commonly used as bind polynucleotide encoding all or a portion of modified trans ers include starch; gelatin; sugars, such as sucrose, glucose, ferrin) or chemical conjugation to one another. The therapeu dextrose, molasses, and lactose; natural and synthetic gums, tic protein and transferrin protein, once part of the transferrin such as acacia, sodium alginate, extract of Irish moss, panwar fusion protein, may be referred to as a "portion”, “region' or gum, ghatti gum, mucilage of isapol husks, carboxymethyl “moiety” of the transferrin fusion protein (e.g., a “therapeutic cellulose, methylcellulose, polyvinylpyrrolidone, Veegum, protein portion or a “transferrin protein portion'). microcrystalline cellulose, microcrystalline dextrose, amy 0017. In one embodiment, the invention provides a trans lose, and larch arabogalactan, and the like. ferrin fusion protein comprising, or alternatively consisting 0022. As used herein, the term “carrier” refers to a diluent, of a therapeutic protein and a modified serum transferrin adjuvant, excipient, or vehicle with which a composition is protein. In other embodiments, the invention provides a trans administered. Such pharmaceutical carriers can be sterile liq ferrin fusion protein comprising, or alternatively consisting uids, such as water and oils, including those of petroleum, of a biologically active and/or therapeutically active frag animal, vegetable or synthetic origin, such as peanut oil. ment of a therapeutic protein and a modified transferrin pro soybean oil, mineral oil, sesame oil and the like. tein. In other embodiments, the invention provides a transfer 0023. As used herein, "coloring agents' are agents that rinfusion protein comprising, or alternatively consisting of a give tablets a more pleasing appearance, and in addition help biologically active and/or therapeutically active variant of a the manufacturer to control the product during its preparation therapeutic protein and modified transferrin protein. In fur and help the user to identify the product. Any of the approved ther embodiments, the invention provides a transferrin fusion certified water-soluble FD&C dyes, mixtures thereof, or their protein comprising a therapeutic protein, and a biologically corresponding lakes may be used to color tablets. A color lake active and/or therapeutically active fragment of modified is the combination by adsorption of a water-soluble dye to a transferrin. In another embodiment, the therapeutic protein hydrous oxide of a heavy metal, resulting in an insoluble form portion of the transferrin fusion protein is the active form of of the dye. the therapeutic protein. 0024. As used herein, "diluents' are inert substances 0018. Unless defined otherwise, all technical and scien added to increase the bulk of the formulation to make the tific terms used herein have the same meaning as commonly tablet a practical size for compression. Commonly used dilu US 2011/009 1543 A1 Apr. 21, 2011

ents include calcium phosphate, calcium sulfate, lactose, within the host cell nucleic acid in which the element is not kaolin, mannitol, Sodium chloride, dry starch, powdered ordinarily found. As an example, a signal sequence native to Sugar, silica, and the like. a yeast cell but attached to a human Tf sequence is heterolo 0025. As used herein, “disintegrators' or “disintegrants' gOuS. are Substances that facilitate the breakup or disintegration of 0032. As used herein, an "isolated nucleic acid sequence tablets after administration. Materials serving as disinte refers to a nucleic acid sequence which is essentially free of grants have been chemically classified as starches, clays, other nucleic acid sequences, e.g., at least about 20% pure, celluloses, algins, or gums. Other disintegrators include Vee preferably at least about 40% pure, more preferably about gum HV, methylcellulose, agar, bentonite, cellulose and 60% pure, even more preferably about 80% pure, most pref wood products, natural sponge, cation-exchange resins, alg erably about 90% pure, and even most preferably about 95% inic acid, guar gum, citrus pulp, cross-linked polyvinylpyr pure, as determined by agarose gel electrophoresis. For rolidone, carboxymethylcellulose, and the like. example, an isolated nucleic acid sequence can be obtained 0026. As used herein, the term “dispersibility” or "dis by standard cloning procedures used in genetic engineering to persible” means a dry powder having a moisture content of relocate the nucleic acid sequence from its natural location to less than about 10% by weight (% w) water, usually below a different site where it will be reproduced. The cloning about 5% w and preferably less than about 3% w; a particle procedures may involve excision and isolation of a desired size of about 1.0-5.0 um mass median diameter (MMD), nucleic acid fragment comprising the nucleic acid sequence usually 1.0-4.0 um MMD, and preferably 1.0-3.0 um MMD; encoding the polypeptide, insertion of the fragment into a a delivered dose of about >30%, usually >40%, preferably vector molecule, and incorporation of the recombinant vector >50%, and most preferred >60%; and an aerosol particle size into a host cell where multiple copies or clones of the nucleic distribution of 1.0-5.0 um mass median aerodynamic diam acid sequence will be replicated. The nucleic acid sequence eter (MMAD), usually 1.5-4.5 um MMAD, and preferably may be of genomic, cDNA, RNA, semi-synthetic, synthetic 1.5-4.0 um MMAD. origin, or any combinations thereof. 0027. As used herein, the term “dry” means that the com 0033. As used herein, two or more DNA coding sequences position has a moisture content Such that the particles are are said to be joined’ or “fused when, as a result of in-frame readily dispersible in an inhalation device to form an aerosol. fusions between the DNA coding sequences, the DNA coding This moisture content is generally below about 10% by sequences are translated into a fusion polypeptide. The term weight (% w) water, usually below about 5% w and preferably “fusion' in reference to Tffusions includes, but is not limited less than about 3% w. to, attachment of at least one therapeutic protein, polypeptide 0028. As used herein, “effective amount’ means an or peptide to the N-terminal end of Tf, attachment to the amount of a drug or pharmacologically active agent that is C-terminal end of Tf, and/or insertion between any two amino sufficient to provide the desired local or systemic effect and acids within Tf. performance at a reasonable benefit/risk ratio attending any 0034. As used herein, “lubricants’ are materials that per medical treatment. form a number of functions in tablet manufacture, such as As used herein, “flavoring agents' vary considerably in their improving the rate of flow of the tablet granulation, prevent chemical structure, ranging from simple esters, alcohols, and ing adhesion of the tablet material to the surface of the dies aldehydes to carbohydrates and complex volatile oils. Syn and punches, reducing interparticle friction, and facilitating thetic flavors of almost any desired type are now available. the ejection of the tablets from the die cavity. Commonly used 0029. As used herein, the terms “fragment of a Tf protein’ lubricants include talc, magnesium Stearate, calcium Stearate, or “Tf protein, or “portion of a Tf protein’ refer to an amino Stearic acid, and hydrogenated vegetable oils. Typical acid sequence comprising at least about 5%, 10%, 20%, 30%, amounts of lubricants range from about 0.1% by weight to 40%, 50%, 60%, 70%, 80%, 90%, 95%, 96%, 97%, 98%, about 5% by weight. 99% or 100% of a naturally occurring Tf protein or mutant 0035. As used herein, “Modified transferrin' as used thereof. herein refers to a transferrin molecule that exhibits at least one 0030. As used herein, the term “gene' refers to any seg modification of its amino acid sequence, compared to wild ment of DNA associated with a biological function. Thus, type transferrin. genes include, but are not limited to, coding sequences and/or 0036. As used herein, “Modified transferrin fusion pro the regulatory sequences required for their expression. Genes tein’ as used herein refers to a protein formed by the fusion of can also include non-expressed DNA segments that, for at least one molecule of modified transferrin (or a fragment or example, form recognition sequences for other proteins. variant thereof) to at least one molecule of a therapeutic Genes can be obtained from a variety of Sources, including protein (or fragment or variant thereof). cloning from a source of interest or synthesizing from known 0037. As used herein, the terms “nucleic acid' or “poly or predicted sequence information, and may include nucleotide' refer to deoxyribonucleotides or ribonucleotides sequences designed to have desired parameters. and polymers thereof in either single- or double-stranded 0031. As used herein, a "heterologous polynucleotide' or form. Unless specifically limited, the terms encompass a "heterologous nucleic acid' or a "heterologous gene' or a nucleic acids containing analogues of natural nucleotides that "heterologous sequence' or an “exogenous DNA segment' have similar binding properties as the reference nucleic acid refers to a polynucleotide, nucleic acid or DNA segment that and are metabolized in a manner similar to naturally occur originates from a source foreign to the particular host cell, or, ring nucleotides. Unless otherwise indicated, a particular if from the same source, is modified from its original form. A nucleic acid sequence also implicitly encompasses conserva heterologous gene in a host cell includes a gene that is endog tively modified variants thereof (e.g. degenerate codon Sub enous to the particular host cell, but has been modified. Thus, stitutions) and complementary sequences as well as the the terms refer to a DNA segment which is foreign or heter sequence explicitly indicated. Specifically, degenerate codon ologous to the cell, or homologous to the cell but in a position Substitutions may be achieved by generating sequences in US 2011/009 1543 A1 Apr. 21, 2011

which the third position of one or more selected (or all) 0045. As used herein, “tablets' are solid pharmaceutical codons is Substituted with mixed-base and/or deoxyinosine dosage forms containing drug Substances with or without residues (Batzer et al. (1991) Nucleic Acid Res. 19:5081; Suitable diluents and prepared either by compression or mold Ohtsuka et al. (1985) J. Biol. Chem. 260:2605-2608; Cassol ing methods well known in the art. Tablets have been in et al. (1992); Rossolini et al. (1994) Mol. Cell. Probes 8:91 widespread use since the latter part of the 19" century and 98). The term nucleic acid is used interchangeably with gene, their popularity continues. Tablets remain popular as a dosage cDNA, and mRNA encoded by a gene. form because of the advantages afforded both to the manu 0038. As used herein, a DNA segment is referred to as facturer (e.g., simplicity and economy of preparation, stabil “operably linked when it is placed into a functional relation ity; and convenience in packaging, shipping, and dispensing) ship with another DNA segment. For example, DNA for a and the patient (e.g., accuracy of dosage, compactness, port signal sequence is operably linked to DNA encoding a fusion ability, blandness of taste, and ease of administration). protein of the invention if it is expressed as a preprotein that Although tablets are most frequently discoid in shape, they participates in the secretion of the fusion protein; a promoter may also be round, oval, oblong, cylindrical, or triangular. or enhancer is operably linked to a coding sequence if it They may differ greatly in size and weight depending on the stimulates the transcription of the sequence. Generally, DNA amount of drug Substance present and the intended method of sequences that are operably linked are contiguous, and in the administration. They are divided into two general classes, (1) case of a signal sequence or fusion protein both contiguous compressed tablets, and (2) molded tablets or tablet triturates. and in reading phase. However, enhancers need not be con In addition to the active or therapeutic ingredient or ingredi tiguous with the coding sequences whose transcription they ents, tablets contain a number or inert materials or additives. control. Linking, in this context, is accomplished by ligation A first group of Such additives includes those materials that at convenient restriction sites or at adapters or linkers inserted help to impart satisfactory compression characteristics to the in lieu thereof. formulation, including diluents, binders, and lubricants. A 0039. As used herein, “pharmaceutically acceptable' second group of such additives helps to give additional desir refers to materials and compositions that are physiologically able physical characteristics to the finished tablet, such as tolerable and do not typically produce an allergic or similar disintegrators, colors, flavors, and Sweetening agents. untoward reaction, Such as gastric upset, dizziness and the 0046. As used herein, the term “therapeutically effective like, when administered to a human. Typically, as used herein, amount” refers to that amount of the transferrinfusion protein the term “pharmaceutically acceptable” means approved by a comprisingatherapeutic molecule which, when administered regulatory agency of the Federal or a state government or to a subject in need thereof, is sufficient to effect treatment. listed in the U.S. Pharmacopeia or other generally recognized The amount of transferrin fusion protein which constitutes a pharmacopeia for use in animals, and more particularly in “therapeutically effective amount will vary depending on humans. the therapeutic protein used, the severity of the condition or 0040. As used herein, “physiologically effective amount disease, and the age and body weight of the Subject to be is that amount delivered to a Subject to give the desired pal treated, but can be determined routinely by one or ordinary liative or curative effect. This amount is specific for each drug skill in the art having regard to his/her own knowledge and to and its ultimate approved dosage level. this disclosure. 0047. As used herein, “therapeutic protein’ refers to pro 0041 As used herein, the term “powder” means a compo teins, polypeptides, peptides or fragments or variants thereof, sition that consists of finely dispersed solid particles that are having one or more therapeutic and/or biological activities. free flowing and capable of being readily dispersed in an Therapeutic proteins encompassed by the invention include inhalation device and Subsequently inhaled by a Subject so but are not limited to proteins, polypeptides, peptides, anti that the particles reach the lungs to permit penetration into the bodies, and biologics. The terms peptides, proteins, and alveoli. Thus, the powder is said to be “respirable.” Preferably polypeptides are used interchangeably herein. Additionally, the average particle size is less than about 10 microns (um) in the term “therapeutic protein’ may refer to the endogenous or diameter with a relatively uniform spheroidal shape distribu naturally occurring correlate of a therapeutic protein. By a tion. More preferably the diameter is less than about 7.5 um polypeptide displaying a “therapeutic activity” or a protein and most preferably less than about 5.0 um. Usually the that is “therapeutically active' is meant a polypeptide that particle size distribution is between about 0.1 um and about 5 possesses one or more known biological and/or therapeutic um in diameter, particularly about 0.3 um to about 5um. activities associated with a therapeutic protein Such as one or 0042. As used herein, the term “promoter” refers to a more of the therapeutic proteins described herein or other region of DNA involved in binding RNA polymerase to ini wise known in the art. As a non-limiting example, a “thera tiate transcription. peutic protein' is a protein that is useful to treat, prevent or 0043. As used herein, the term “recombinant” refers to a ameliorate a disease, condition or disorder. Such a disease, cell, tissue or organism that has undergone transformation condition or disorder may be in humans or in a non-human with a new combination of genes or DNA. animal, e.g., Veterinary use. As used herein, the term 'subject' can be a human, a mam 0048. As used herein, the term “transformation” refers to mal, or an animal. The Subject being treated is a patient in the transfer of nucleic acid (i.e., a nucleotide polymer) into a need of treatment. cell. As used herein, the term “genetic transformation” refers 0044 AS used herein, a targeting entity, protein, polypep to the transfer and incorporation of DNA, especially recom tide or peptide refers to a molecule that binds specifically to a binant DNA, into a cell. As used herein, the term “transfor particular cell type normal (e.g., lymphocytes) or abnormal mant” refers to a cell, tissue or organism that has undergone e.g., (cancer cell) and therefore may be used to target a Tf transformation. fusion protein or compound (drug, or cytotoxic agent) to that 0049. As used herein, the term “transgene' refers to a cell type specifically. nucleic acid that is inserted into an organism, host cell or US 2011/009 1543 A1 Apr. 21, 2011

vector in a manner that ensures its function. As used herein, receptor), endocytosis then occurs whereby the TfR/Tf com the term “transgenic’ refers to cells, cell cultures, organisms, plex is transported to the endosome, at which point the local bacteria, fungi, animals, plants, and progeny of any of the ized drop in pH results in release of bound iron and the preceding, which have received a foreign or modified gene recycling of the TfR/Tf complex to the cell surface and and in particular a gene encoding a modified Tffusion protein release of Tf (known as apoTfin its un-iron bound form). by one of the various methods of transformation, wherein the Receptor binding is through the C domain of Tf. The two foreign or modified gene is from the same or different species glycosylation sites in the C domain do not appear to be than the species of the organism receiving the foreign or involved in receptor binding as unglycosylated iron bound Tf modified gene. does bind the receptor. 0050. As used herein, the term “Variants or variant refers to a polynucleotide or nucleic acid differing from a reference 0056. Each Tf molecule can carry two iron ions (Fe"). nucleic acid or polypeptide, but retaining essential properties These are complexed in the space between the N1 and N2, C1 thereof. Generally, variants are overall closely similar, and, in and C2 Sub domains resulting in a conformational change in many regions, identical to the reference nucleic acid or the molecule. Tf crosses the blood brain barrier (BBB) via the polypeptide. As used herein, “variant” refers to a therapeutic Tf receptor. protein portion of a transferrinfusion protein of the invention, 0057. In human transferrin, the iron binding sites com differing in sequence from a native therapeutic protein but prise at least amino acids Asp 63 (Asp 82 of SEQID NO: 2 retaining at least one functional and/or therapeutic property which includes the native Tf signal sequence), Asp 392 (Asp thereofas described elsewhere herein or otherwise known in 411 of SEQID NO: 2), Tyr 95 (Tyr 114 of SEQID NO: 2), Tyr the art. 426 (Tyr 445 of SEQID NO: 2), Tyr 188 (Tyr207 of SEQID 0051. As used herein, the term “vector” refers broadly to NO: 2), Tyr 514 or 517 (Tyr 533 or Tyr 536 SEQ ID NO: 2), any plasmid, phagemid or virus encoding an exogenous His 249 (His 268 of SEQID NO: 2), and His 585 (His 604 of nucleic acid. The term is also be construed to include non SEQID NO:2) of SEQID NO:3. The hingeregions comprise plasmid, non-phagemid and non-viral compounds which at least N domain amino acid residues 94-96, 245-247 and/or facilitate the transfer of nucleic acid into virions or cells, such 316-318 as well as C domain amino acid residues 425-427, as, for example, polylysine compounds and the like. The 581-582 and/or 652-658 of SEQ ID NO: 3. The carbonate vector may be a viral vector that is suitable as a delivery binding sites comprise at least amino acids Thr 120 (Thr 139 vehicle for delivery of the nucleic acid, or mutant thereof, to of SEQID NO: 2), Thr 452 (Thr 471 of SEQID NO: 2), Arg a cell, or the vector may be a non-viral vector which is suitable 124 (Arg 143 of SEQID NO: 2), Arg 456 (Arg475 of SEQID for the same purpose. Examples of viral and non-viral vectors NO: 2), Ala 126 (Ala 145 of SEQID NO: 2), Ala 458 (Ala 477 for delivery of DNA to cells and tissues are well known in the of SEQID NO: 2), Gly 127 (Gly 146 of SEQID NO: 2), and art and are described, for example, in Ma et al. (1997, Proc. Gly 459 (Gly 478 of SEQID NO: 2) of SEQ ID NO: 3. Natl. Acad. Sci. U.S.A. 94:12744-12746). Examples of viral 0058. In one embodiment of the invention, the modified vectors include, but are not limited to, a recombinant vaccinia transferrin fusion protein includes a modified human trans virus, a recombinant adenovirus, a recombinant retrovirus, a ferrin, although any animal Tf molecule may be used to recombinant adeno-associated virus, a recombinant avian produce the fusion proteins of the invention, including human pox virus, and the like (Cranage et al., 1986, EMBO 3. Tf variants, cow, pig, sheep, dog, rabbit, rat, mouse, hamster, 5:3057-3063: International Patent Application No. WO94/ echnida, platypus, chicken, frog, hornworm, monkey, as well 17810, published Aug. 18, 1994: International Patent Appli as other bovine, canine and avian species. All of these Tf cation No. WO94/23744, published Oct. 27, 1994). Examples sequences are readily available in GenBank and other public of non-viral vectors include, but are not limited to, liposomes, databases. The human Tf nucleotide sequence is available polyamine derivatives of DNA, and the like. (see SEQ ID NOS 1, 2 and 3 and the accession numbers 0052. As used herein, the term “wild type” refers to a described above and available at www.ncbi.nlm.nih.gov/) polynucleotide or polypeptide sequence that is naturally and can be used to make genetic fusions between Tf or a occurring. domain of Tf and the therapeutic molecule of choice. Fusions 0053 Transferrin and Transferrin Modifications may also be made from related molecules Such as lacto trans 0054 Any transferrin may be used to make modified Tf ferrin (lactoferrin) GenBank Acc: NM 002343) or melan fusion proteins of the invention. As an example, the wild-type otransferrin (GenBank Acc. NM 013900, murine melan human Tf (Tf) is a 679 amino acid protein of approximately otransferrin). 75 kDa (not accounting for glycosylation), with two main 0059 Melanotransferrin is a glycosylated protein found at domains, N (about 330 amino acids) and C (about 340 amino high levels in malignant melanoma cells and was originally acids), which appear to originate from a gene duplication. See named human melanoma antigen p97 (Brown et al., 1982, GenBank accession numbers NM001063, XM002793, Nature, 296: 171-173). It possesses high sequence homology M12530, XM039845, XM 0398.47 and S95936 (www.ncbi. with human serum transferrin, human lactoferrin, and nlm.nih.gov/), all of which are herein incorporated by refer chicken transferrin (Brown et al., 1982, Nature, 296: 171 ence in their entirety, as well as SEQID NOS 1, 2 and 3. The 173; Rose et al., Proc. Natl. Acad. Sci. USA, 1986, 83: 1261 two domains have diverged over time but retain a large degree 1265). However, unlike these receptors, no cellular receptor of identity/similarity (FIG. 1). has been identified for melanotransferrin. Melanotransferrin 0055. Each of the N and C domains is further divided into reversibly binds iron and it exists in two forms, one of which two subdomains, N1 and N2, C1 and C2. The function of Tf is bound to cell membranes by a glycosyl phosphatidylinosi is to transport iron to the cells of the body. This process is tol anchor while the other form is both soluble and actively mediated by the Tf receptor (TfR), which is expressed on all secreted (Baker et al., 1992, FEBS Lett, 298: 215-218: Ale cells, particularly actively growing cells. TfR recognizes the many et al., 1993, J. Cell Sci., 104: 1155-1162: Food et al., ironbound form of Tf(two molecules of which are bound per 1994, J. Biol. Chem. 274: 701 1-7017). US 2011/009 1543 A1 Apr. 21, 2011

0060 Lactoferrin (Lf), a natural defense iron-binding pro glycosylation patterns or iron binding properties Substan tein, has been found to possess antibacterial, antimycotic, tially like that of a native or wild-type N domain or lobe. In a antiviral, antineoplastic and anti-inflammatory activity. The preferred embodiment, the C domain or lobe is modified so protein is present in exocrine secretions that are commonly that it is not glycosylated and does not bind iron by Substitu exposed to normal flora: milk, tears, nasal exudate, Saliva, tion of the relevant C domain regions or amino acids to those bronchial mucus, gastrointestinal fluids, cervico-vaginal present in the corresponding regions or sites of a native or mucus and seminal fluid. Additionally, Lifis a major constitu wild-type N domain. ent of the secondary specific granules of circulating polymor 0066. As used herein, a Tf moiety comprising “two N phonuclear neutrophils (PMNs). The apoprotein is released domains or lobes’ includes a Tf molecule that is modified to on degranulation of the PMNs in septic areas. A principal replace the native C domain or lobe with a native or wild-type function of Lf is that of Scavenging free iron in fluids and N domain or lobe or a modified N domain or lobe or contains inflamed areas so as to suppress free radical-mediated dam a C domain that has been modified to function substantially age and decrease the availability of the metal to invading like a wild-type or modified N domain. Analysis of the two microbial and neoplastic cells. In a study that examined the domains by overlay of the two domains (Swiss PDB Viewer turnover rate of 'I Lf in adults, it was shown that Lf is 3.7b2, Iterative Magic Fit) and by direct amino acid align rapidly taken up by the liver and spleen, and the radioactivity ment (ClustalW multiple alignment) reveals that the two persisted for several weeks in the liver and spleen (Bennett et domains have diverged over time. Amino acid alignment al. (1979), Clin. Sci. (Lond.) 57: 453-460). shows 42% identity and 59% similarity between the two 0061. In one embodiment, the transferrin portion of the domains. However, approximately 80% of the N domain transferrin fusion protein of the invention includes a transfer matches the C domain for structural equivalence. The C rin splice variant. In one example, a transferrin splice variant domain also has several extra disulfide bonds compared to the can be a splice variant of human transferrin. In one specific N domain. embodiment, the human transferrin splice variant can be that 0067 Alignment of molecular models for the N and C of Genbank Accession AAA61140. domain reveals the following structural equivalents:

N domain 4-24 36-72 94-136 138-139 149-164 168-173 178-198 219-2SS 259-260 263-268 271-27S 279-280 283-288 309-327 (1-330) 75-88 2OO-214 29O-304 C domain 340-361 365-415 425-437 470-471 475-490 492-497 SOf-542 SSS-591 S93-594 S97-6O2 605-609 614-615 620-64O 645-663 (340-679) 439-468

0062. In another embodiment, the transferrin portion of The disulfide bonds for the two domains align as follows: the transferrinfusion protein of the invention includes a lacto ferrin splice variant. In one example, a human serum lactof errin splice variant can be a novel splice variant of a neutro phil lactoferrin. In one specific embodiment, the neutrophil lactoferrin splice variant can be that of Genbank Accession AAA59479. In another specific embodiment, the neutrophil lactoferrin splice variant can comprise the following amino acid sequence EDCIALKGEADA (SEQ ID NO: 8), which includes the novel region of splice-variance. 0063. In another embodiment, the transferrin portion of the transferrin fusion protein of the invention includes a mel anotransferrin variant. 0064. Modified Tffusions may be made with any Tf pro tein, fragment, domain, or engineered domain. For instance, fusion proteins may be produced using the full-length Tf sequence, with or without the native Tf signal sequence. Tf Bold aligned disulfide bonds fusion proteins may also be made using a single Tf domain, Italics bridging peptide such as an individual N or C domain or a modified form of Tf comprising 2N or 2C domains (see U.S. Provisional Appli 0068. In one embodiment, the transferrin portion of the cation 60/406,977, filed Aug. 30, 2002, which is herein incor transferrin fusion protein includes at least two N terminal porated by reference in its entirety). In some embodiments, lobes of transferrin. In further embodiments, the transferrin fusions of a therapeutic protein to a single C domain may be portion of the transferrin fusion protein includes at least two produced, wherein the C domain is altered to reduce, inhibit N terminal lobes of transferrin derived from human serum or prevent glycosylation. In other embodiments, the use of a transferrin. single Ndomain is advantageous as the Tf glycosylation sites 0069. In another embodiment, the transferrin portion of reside in the C domain and the N domain, on its own. A the transferrin fusion protein includes, comprises, or consists preferred embodiment is the Tffusion protein having a single of at least two C terminal lobes of transferrin. In further N domain which is expressed at a high level. embodiments, the transferrin portion of the transferrin fusion 0065. As used herein, a C terminal domain or lobe modi protein includes at least two C terminal lobes of transferrin fied to function as an N-like domain is modified to exhibit derived from human serum transferrin. US 2011/009 1543 A1 Apr. 21, 2011

0070 Inafurther embodiment, the C terminal lobe mutant ferrin mutant having a mutation wherein the mutant does not further includes a mutation of at least one of ASn413 and retain the ability to bind metal ions. In an alternate embodi Asnó11 of SEQIDNO:3 which does not allow glycosylation. ment, the transferrin portion of the transferrin fusion protein (0071. In some embodiments, the Tfor Tfportion will be of includes a recombinant transferrin mutant having a mutation sufficient length to increase the in vivo circulatory half-life, wherein the mutant has a weaker binding avidity for metal serum stability, in vitro solution stability or bioavailability of ions than wild-type serum transferrin. In an alternate embodi the therapeutic protein compared to the in vivo circulatory ment, the transferrin portion of the transferrin fusion protein half-life, serum stability, invitro solution stability or bioavail includes a recombinant transferrin mutant having a mutation ability of the therapeutic protein in an unfused state. Such an wherein the mutant has a stronger binding avidity for metal increase in stability, serum half-life or bioavailability may be ions than wild-type serum transferrin. about a 30%, 50%, 70%, 80%, 90% or more increase over the 0076. In another embodiment, the transferrin portion of unfused therapeutic protein. In some cases, the modified the transferrin fusion protein includes a recombinant trans transferrin fusion proteins exhibit a serum half-life of about ferrin mutant having a mutation wherein the mutant does not 10-20 or more days, about 12-18 days or about 14-17 days. retain the ability to bind to the transferrin receptor. In an 0072. When the C domain of Tf is part of the fusion pro alternate embodiment, the transferrin portion of the transfer tein, the two N-linked glycosylation sites, amino acid resi rin fusion protein includes a recombinant transferrin mutant dues corresponding to N413 and N611 of SEQID NO:3 may having a mutation wherein the mutant has a weaker binding be mutated for expression in a yeast system to prevent glyco avidity for the transferrin receptor than wild-type serum Sylation or hypermannosylation and extend the serum half transferrin. In an alternate embodiment, the transferrin por life of the fusion protein and/or therapeutic protein (to pro tion of the transferrin fusion protein includes a recombinant duce asialo-, or in Some instances, monosialo-Tf or disialo transferrin mutant having a mutation wherein the mutant has Tf). In addition to Tfamino acids corresponding to N413 and a stronger binding avidity for the transferrin receptor than N611, mutations may be to the adjacent residues within the wild-type serum transferrin. N-X-S/T glycosylation site to prevent or substantially reduce 0077. In another embodiment, the transferrin portion of glycosylation. See U.S. Pat. No. 5,986,067 of Funk et al. It the transferrin fusion protein includes a recombinant trans has also been reported that the N domain of Tf expressed in ferrin mutant having a mutation wherein the mutant does not Pichia pastoris becomes O-linked glycosylated with a single retain the ability to bind to carbonate ions. In an alternate hexose at S32 which also may be mutated or modified to embodiment, the transferrin portion of the transferrin fusion prevent such glycosylation. protein includes a recombinant transferrin mutant having a 0073. Accordingly, in one embodiment of the invention, mutation wherein the mutant has a weaker binding avidity for the transferrin fusion protein includes a modified transferrin carbonate ions than wild-type serum transferrin. In an alter molecule wherein the transferrin exhibits reduced glycosyla nate embodiment, the transferrin portion of the transferrin tion, including but not limited to asialo-monosialo- and fusion protein includes a recombinant transferrin mutant hav disialo-forms of Tf. In another embodiment, the transferrin ing a mutation wherein the mutant has a stronger binding portion of the transferrin fusion protein includes a recombi avidity for carbonate ions than wild-type serum transferrin. nant transferrin mutant that is mutated to prevent glycosyla 0078. In another embodiment, the transferrin portion of tion. In another embodiment, the transferrin portion of the the transferrin fusion protein includes a recombinant human transferrin fusion protein includes a recombinant transferrin serum transferrin mutant having a mutation in at least one mutant that is fully glycosylated. In a further embodiment, the amino acid residue selected from the group consisting of transferrin portion of the transferrin fusion protein includes a Asp63, Gly65, Tyr95, Tyr188, His249, Asp392, Tyr426, recombinant human serum transferrin mutant that is mutated TyrS14, TyrS17 and His585 of SEQID NO:3, wherein the to prevent glycosylation, wherein at least one of ASn413 and mutant retains the ability to bind metal ions. In an alternate Asnó11 of SEQID NO:3 are mutated to an amino acid which embodiment, a recombinant human serum transferrin mutant does not allow glycosylation. In another embodiment, the having a mutation in at least one amino acid residue selected transferrin portion of the transferrin fusion protein includes a from the group consisting of Asp63, Gly65, Tyr95, Tyr188, recombinant human serum transferrin mutant that is mutated His249, Asp392, Tyra,26, TyrS14, TyrS17 and His585 of SEQ to prevent or substantially reduce glycosylation, wherein ID NO:3, wherein the mutant has a reduced ability to bind mutations may be to the adjacent residues within the N-X-S/T metal ions. In another embodiment, a recombinant human glycosylation site. Moreover, glycosylation may be reduced serum transferrin mutant having a mutation in at least one or prevented by mutating the serine or threonine residue. amino acid residue selected from the group consisting of Further, changing the X to proline is known to inhibit glyco Asp63, Gly65, Tyr95, Tyr188, His249, Asp392, Tyr426, Sylation. TyrS17 and His585 of SEQIDNO:3, wherein the mutant does 0.074. In other embodiments of the invention, the iron not retain the ability to bind metal ions. binding is retained and the iron binding ability of Tf may be 0079. In another embodiment, the transferrin portion of used to deliver a therapeutic protein or peptide(s) to the inside the transferrin fusion protein includes a recombinant human of a cell, across an epithelial or endothelial cell membrane serum transferrin mutant having a mutation at Lys206 or and/or across the BBB. These embodiments that bind iron His207 of SEQID NO:3, wherein the mutant has a stronger and/or the Tf receptor will often be engineered to reduce or binding avidity for metal ions than wild-type human serum prevent glycosylation to extend the serum half-life of the transferrin (see U.S. Pat. No. 5,986,067, which is herein therapeutic protein. The N domain alone will not bind to TfR incorporated by reference in its entirety). In an alternate when loaded with iron, and the ironbound C domain will bind embodiment, the transferrin portion of the transferrin fusion TfR but not with the same affinity as the whole molecule. protein includes a recombinant human serum transferrin 0075. In another embodiment, the transferrin portion of mutant having a mutation at Lys206 or His207 of SEQ ID the transferrin fusion protein includes a recombinant trans NO:3, wherein the mutant has a weaker binding avidity for US 2011/009 1543 A1 Apr. 21, 2011

metal ions than wild-type human serum transferrin. In a fur ther embodiment, the transferrin portion of the transferrin -continued fusion protein includes a recombinant human serum transfer rin mutant having a mutation at Lys206 or His207 of SEQID TABLE OF AMINO ACIDS NO:3, wherein the mutant does not bind metal ions. ONE 0080. Any available technique may be used to make the LETTER THREE-LETTER fusion proteins of the invention, including but not limited to AMINO ACID SYMBOL SYMBOL molecular techniques commonly available, for instance, Glycine G Gly those disclosed in Sambrook et al. Molecular Cloning: A Histidine H His Laboratory Manual, 2nd Ed., Cold Spring Harbor Laboratory Isoleucine I Ile Leucine L Leu Press, 1989. When carrying out nucleotide substitutions Lysine K Lys using techniques for accomplishing site-specific mutagenesis Methionine M Met that are well known in the art, the encoded amino acid Phenylalanine F Phe changes are preferably of a minor nature, that is, conservative Proline P Pro Serine S Ser amino acid Substitutions, although other, non-conservative, Threonine T Thr Substitutions are contemplated as well, particularly when pro Tryptophan W Trp ducing a modified transferrin portion of a Tffusion protein, Tyrosine Y Tyr e.g., a modified Tffusion protein exhibiting reduced glyco Valine V Wall Sylation, reduced iron binding and the like. Specifically con templated are amino acid substitutions, Small deletions or insertions, typically of one to about 30 amino acids; insertions I0084 Iron binding and/or receptor binding may be between transferrin domains; Small amino- or carboxyl-ter reduced or disrupted by mutation, including deletion, Substi minal extensions, such as an amino-terminal methionine resi tution or insertion into, amino acid residues corresponding to due, or small linker peptides of less than 50, 40, 30, 20 or 10 one or more of TfN domain residues Asp63, Tyr95, Tyr188, residues between transferrin domains or linking a transferrin His249 and/or C domain residues Asp 392, Tyr 426, Tyr 514 protein and a therapeutic protein or peptide; or a small exten and/or His 585 of SEQID NO:3. Iron binding may also be sion that facilitates purification, such as a poly-histidine tract, affected by mutation to amino acids Lys206, His207 or an antigenic epitope or a binding domain. Arg632 of SEQID NO:3. Carbonate binding may be reduced 0081 Examples of conservative amino acid substitutions or disrupted by mutation, including deletion, Substitution or are Substitutions made within the same group Such as within insertion into, amino acid residues corresponding to one or the group of basic amino acids (such as arginine, lysine, more of TfN domain residues Thr120, Arg124, Ala 126, Gly histidine), acidic amino acids (such as glutamic acid and 127 and/or C domain residues Thr 452, Arg 456, Ala 458 aspartic acid), polar amino acids (such as glutamine and and/or Gly 459 of SEQID NO:3. A reduction or disruption of asparagine), hydrophobic amino acids (such as leucine, iso carbonate binding may adversely affect iron and/or receptor leucine, Valine), aromatic amino acids (such as phenylala binding. nine, tryptophan, tyrosine) and Small amino acids (such as I0085 Binding to the Tf receptor may be reduced or dis glycine, alanine, serine, threonine, methionine). rupted by mutation, including deletion, Substitution or inser 0082) Non-conservative substitutions encompass substi tion into, amino acid residues corresponding to one or more of tutions of amino acids in one group by amino acids in another group. For example, a non-conservative substitution would Tf N domain residues described above for iron binding. include the Substitution of a polar amino acid for a hydropho I0086. As discussed above, glycosylation may be reduced bic amino acid. For a general description of nucleotide Sub or prevented by mutation, including deletion, Substitution or stitution, see e.g. Ford etal. (1991), Prot. Exp. Pur. 2: 95-107. insertion into, amino acid residues corresponding to one or Non-conservative Substitutions, deletions and insertions are more of Tf C domain residues around the N-X-S/T sites particularly useful to produce TF fusion proteins of the inven corresponding to C domain residues N413 and/or N611 (See tion that exhibit no or reduced binding of iron, no or reduced U.S. Pat. No. 5,986,067). For instance, the N413 and/or N611 binding of the fusion protein to the Tf receptor. may be mutated to Glu residues. 0083. In the polypeptide and proteins of the invention, the I0087. In instances where the Tf fusion proteins of the following system is followed for designating amino acids in invention are not modified to prevent glycosylation, iron accordance with the following conventional list: binding, carbonate binding and/or receptor binding, glycosy lation, iron and/or carbonate ions may be stripped from or cleaved off of the fusion protein. For instance, available deg lycosylases may be used to cleave glycosylation residues TABLE OF AMINO ACIDS from the fusion protein, in particular the Sugar residues ONE attached to the Tf portion, yeast deficient in glycosylation LETTER THREE-LETTER enzymes may be used to prevent glycosylation and/or recom AMINO ACID SYMBOL SYMBOL binant cells may be grown in the presence of an agent that Alanine A. Ala prevents glycosylation, e.g., tunicamycin. Arginine R Arg I0088. The carbohydrates on the fusion protein may also be Asparagine N ASn reduced or completely removed enzymatically by treating the Aspartic Acid D Asp Cysteine C Cys fusion protein with deglycosylases. Deglycosylases are well Glutamine Q Gln known in the art. Examples of deglycosylases include but are Glutamic Acid E Glu not limited to galactosidase, PNGase A, PNGase F, glucosi dase, mannosidase, fucosidase, and Endo H deglycosylase. US 2011/009 1543 A1 Apr. 21, 2011

0089 Nevertheless, in certain circumstances, it may be thereof. In one embodiment, the transferrin fusion protein of preferable for oral delivery that the Tfportion of the fusion the invention contains a therapeutic protein or polypeptide protein be fully glycosylated fused to the N terminus of a transferrin molecule. In an 0090. Additional mutations may be made with Tf to alter alternate embodiment, the transferrin fusion protein of the the three dimensional structure of Tf, such as modifications to invention contains a therapeutic protein fused to the C termi the hinge region to prevent the conformational change needed nus of a transferrin molecule. The present invention also for iron biding and Tf receptor recognition. For instance, provides transferrin fusion protein containing a therapeutic mutations may be made in or around N domain amino acid protein or polypeptide or protion thereof fused to a modified residues 94-96, 245-247 and/or 316-318 as well as C domain transferrin morlecule or portion thererof. amino acid residues 425-427, 581-582 and/or 652-658. In 0098. In other embodiments, the transferrin fusion protein addition, mutations may be made in or around the flanking of the inventions contains a therapeutic protein fused to both regions of these sites to alter Tf structure and function. the N-terminus and the C-terminus of modified transferrin. In 0091. In one aspect of the invention, the transferrin fusion another embodiment, the therapeutic proteins fused at the N protein can function as a carrier protein to extend the halflife and C-termini are the same therapeutic proteins. In an alter or bioavailability of the therapeutic proteinas well as, in some nate embodiment, the therapeutic proteins fused at the N- and instances, delivering the therapeutic protein inside a cell and/ C-termini are different therapeutic proteins. In another alter or across the blood brain barrier. In an alternate embodiment, nate embodiment, the therapeutic proteins fused to the N- and the transferrin fusion protein includes a modified transferrin C-termini are different therapeutic proteins which may be molecule wherein the transferrin does not retain the ability to used to treat or prevent the same disease, disorder, or condi cross the blood brain barrier. tion. In another embodiment, the therapeutic proteins fused at 0092. In another embodiment, the transferrin fusion pro the N- and C-termini are different therapeutic proteins which tein includes a modified transferrin molecule wherein the may be used to treator prevent diseases or disorders which are transferrin molecule retains the ability to bind to the transfer known in the art to commonly occur in patients simulta rin receptor and transport the therapeutic peptide inside cells. neously. In an alternate embodiment, the transferrin fusion protein 0099. In addition to modified transferrin fusion protein of includes a modified transferrin molecule wherein the trans the inventions in which the modified transferrin portion is ferrin molecule does not retain the ability to bind to the fused to the N terminal and/or C-terminal of the therapeutic transferrin receptor and transport the therapeutic peptide protein portion, transferrin fusion protein of the inventions of inside cells. the invention may also be produced by inserting the therapeu 0093. In further embodiments, the transferrin fusion pro tic protein or peptide of interest (e.g., a therapeutic protein or tein includes a modified transferrin molecule wherein the peptide as disclosed herein, or, for instance, a single chain transferrin molecule retains the ability to bind to the transfer antibody that binds a therapeutic protein or a fragment or rin receptor and transport the therapeutic peptide inside cells variant thereof) into an internal region of the modified trans and retains the ability to cross the blood brain barrier. In an ferrin. Internal regions of modified transferrin include, but are alternate embodiment, the transferrin fusion protein includes not limited to, the iron binding sites, the hinge regions, the a modified transferrin molecule wherein the transferrin mol bicarbonate binding sites, or the receptor binding domain. ecule retains the ability to cross the blood brain barrier, but 0100. Within the protein sequence of the modified trans does not retain the ability to bind to the transferrin receptor ferrin molecule a number of loops or turns exist, which are and transport the therapeutic peptide inside cells. stabilized by disulfide bonds. These loops are useful for the 0094) Modified Transferrin Fusion Proteins insertion, or internal fusion, of therapeutically active pep 0095. The fusion proteins of the invention may contain tides, particularly those requiring a secondary structure to be one or more copies of the therapeutic protein or polypeptide functional, or therapeutic proteins to generate a modified attached to the N-terminus and/or the C-terminus of the Tf transferrin molecule with specific biological activity. protein. In some embodiments, the therapeutic protein or 0101. When therapeutic proteins or peptides are inserted polypeptide is attached to both the N- and C-terminus of the into or replace at least one loop of a Tf molecule, insertions Tf protein and the fusion protein may contain one or more may be made within any of the Surface exposed loop regions, equivalents of the therapeutic protein or polypeptide on either in addition to other areas of Tf. For instance, insertions may or both ends of Tf. In other embodiments, the therapeutic be made within the loops comprising Tfamino acids 32-33, protein or polypeptide is inserted into known domains of the 74-75, 256-257, 279-280 and 288-289 (Aliet al., supra) (See Tf protein, for instance, into one or more of the loops of Tf FIG.3). As previously described, insertions may also be made (see Alietal. (1999).J. Biol. Chem. 274(34):24066-24073). In within other regions of Tf such as the sites for iron and other embodiments, the therapeutic protein or therapeutic bicarbonate binding, hinge regions, and the receptor binding peptide is inserted between the N and C domains of Tf. domain as described in more detail below. The loops in the Tf 0096 Generally, the transferrin fusion protein of the protein sequence that are amenable to modification/replace invention may have one modified transferrin-derived region ment for the insertion of proteins or peptides may also be used and one therapeutic protein-derived region. Multiple regions for the development of a screenable library of random peptide of each protein, however, may be used to make a transferrin inserts. Any procedures may be used to produce nucleic acid fusion protein of the invention. Similarly, more than one inserts for the generation of peptide libraries, including avail therapeutic protein may be used to make a transferrin fusion able phage and bacterial display systems, prior to cloning into protein of the invention, thereby producing a multi-functional a Tfdomain and/or fusion to the ends of Tf. In other embodi modified Tffusion protein. ments, the library is made directly in or on the ends of a Tf 0097. The present invention provides transferrin fusion peptide as described below. protein containing a therapeutic protein or polypeptide or 0102 The N-terminus of Tf is free and points away from portion thereof fused to a transferrin molecule or portion the body of the molecule. Fusions of proteins or peptides on US 2011/009 1543 A1 Apr. 21, 2011

the N-terminus may therefore be a preferred embodiment. 0109 As known in the art, “similarity” between two poly Such fusions may include a linker region, such as but not nucleotides or polypeptides is determined by comparing the limited to a poly-glycine stretch, to separate the therapeutic nucleotide or amino acid sequence and its conserved nucle protein or peptide from Tf. Attention to the junction between otide or amino acid Substitutes of one polynucleotide or the leader sequence, the choice of leader sequence, and the polypeptide to the sequence of a second polynucleotide or structure of the mRNA by codon manipulation/optimization polypeptide. Also known in the art is “identity” which means (no major stem loops to inhibit ribosome progress) will the degree of sequence relatedness between two polypeptide increase secretion and can be readily accomplished using or two polynucleotide sequences as determined by the iden tity of the match between two strings of such sequences. Both standard recombinant protein techniques. identity and similarity can be readily calculated (Computa 0103) The C-terminus of Tfappears to be more buried and tional Molecular Biology, Lesk, A. M., ed., Oxford Univer secured by a disulfide bond 6 amino acids from the C-termi sity Press, New York, 1988: Biocomputing: Informatics and nus. In human Tf, the C-terminal amino acid is a proline Genome Projects, Smith, D. W., ed., Academic Press, New which, depending on the way that it is orientated, will either York, 1993: Computer Analysis of Sequence Data, Part I, point a fusion away or into the body of the molecule. A linker Griffin, A. M., and Griffin, H. G., eds., Humana Press, New or spacer moiety at the C-terminus may be used in some Jersey, 1994; Sequence Analysis in Molecular Biology, von embodiments of the invention. There is also a proline near the Heinje, G., Academic Press, 1987; and Sequence Analysis N-terminus. In one aspect of the invention, the proline at the Primer, Gribskov, M. and Devereux, J., eds., M. Stockton N- and/or the C-termini may be changed out. In another Press, New York, 1991). aspect of the invention, the C-terminal disulfide bond may be 0110. While there exist a number of methods to measure eliminated to untether the C-terminus. identity and similarity between two polynucleotide or 0104. In yet other embodiments, small molecule therapeu polypeptide sequences, the terms “identity” and “similarity” tics may be complexed with iron and loaded on a modified Tf are well known to skilled artisans (Sequence Analysis in protein fusion for delivery to the inside of cells and across the Molecular Biology, von Heinje, G., Academic Press, 1987: BBB. The addition of a targeting peptide or, for example, a Sequence Analysis Primer, Gribskov, M. and Devereux, J., single chain antibody (SCA) can be used to target the payload eds., MStockton Press, New York, 1991; and Carillo, H., and to a particular cell type, e.g., a cancer cell. Lipman, D., SIAM.J. Applied Math., 48: 1073 (1988). Meth 0105 Nucleic Acids ods commonly employed to determine identity or similarity 0106 Nucleic acid molecules are also provided by the between two sequences include, but are not limited to those, present invention. These encode a modified Tffusion protein disclosed in Guide to Huge Computers, Martin J. Bishop, ed., comprising a transferrin protein or a portion of a transferrin Academic Press, San Diego, 1994, and Carillo, H., and Lip protein covalently linked or joined to atherapeutic protein. As man, D., SIAM J. Applied Math. 48: 1073 (1988). discussed in more detail below, any therapeutic protein may 0111 Preferred methods to determine identity are be used. The fusion protein may further comprise a linker designed to give the largest match between the two sequences region, for instance a linker less than about 50, 40, 30, 20, or tested. Methods to determine identity and similarity are codi 10 amino acid residues. The linker can be covalently linked to fied in computer programs. Preferred computer program and between the transferrin protein or portion thereof and the methods to determine identity and similarity between two therapeutic protein. Nucleic acid molecules of the invention sequences include, but are not limited to, GCG program pack may be purified or not. age (Devereux, et al., Nucl. Acid Res. 12(1):387 (1984)), 0107 Host cells and vectors for replicating the nucleic BLASTP. BLASTN, FASTA (Atschul, et al., J. Mol. Biol. acid molecules and for expressing the encoded fusion pro 215:403 (1990)). The degree of similarity or identity referred teins are also provided. Any vectors or host cells may be used, to above is determined as the degree of identity between the whether prokaryotic or eukaryotic, but eukaryotic expression two sequences, often indicating a derivation of the first systems, in particular yeast expression systems, may be pre sequence from the second. The degree of identity between ferred. Many vectors and host cells are known in the art for two nucleic acid sequences may be determined by means of such purposes. It is well within the skill of the art to select an computer programs known in the art such as GAP provided in appropriate set for the desired application. the GCG program package (Needleman and Wunsch J. Mol. 0108 DNA sequences encoding transferrin, portions of Biol. 48:443-453 (1970)). For purposes of determining the transferrin and therapeutic proteins of interest may be cloned degree of identity between two nucleic acid sequences for the from a variety of genomic or cDNA libraries known in the art. present invention, GAP is used with the following settings: The techniques for isolating Such DNA sequences using GAP creation penalty of 5.0 and GAP extension penalty of probe-based methods are conventional techniques and are O3. well known to those skilled in the art. Probes for isolating 0112 Codon Optimization such DNA sequences may be based on published DNA or 0113. The degeneracy of the genetic code permits varia protein sequences (see, for example, Baldwin, G. S. (1993) tions of the nucleotide sequence of a transferrin protein and/ Comparison of Transferrin Sequences from Different Spe or therapeutic protein of interest, while still producing a cies. Comp. Biochem. Physiol. 106B/1:203-218 and all ref polypeptide having the identical amino acid sequence as the erences cited therein, which are hereby incorporated by ref polypeptide encoded by the native DNA sequence. The pro erence in their entirety). Alternatively, the polymerase chain cedure, known as “codon optimization' (described in U.S. reaction (PCR) method disclosed by Mullis et al. (U.S. Pat. Pat. No. 5,547.871 which is incorporated herein by reference No. 4,683,195) and Mullis (U.S. Pat. No. 4,683.202), incor in its entirety) provides one with a means of designing Such an porated herein by reference may be used. The choice of altered DNA sequence. The design of codon optimized genes library and selection of probes for the isolation of such DNA should take into account a variety of factors, including the sequences is within the level of ordinary skill in the art. frequency of codon usage in an organism, nearest neighbor US 2011/009 1543 A1 Apr. 21, 2011 frequencies, RNA stability, the potential for secondary struc pPPC0005, pSeCHSA, pScNHSA, pC4 and derivatives ture formation, the route of synthesis and the intended future thereof. Useful yeast plasmid vectors also include pRS403 DNA manipulations of that gene. In particular, available 406, pRS413-416 and the Pichia vectors available from Strat methods may be used to alter the codons encoding a given agene Cloning Systems, La Jolla, Calif. 92037, USA. Plas fusion protein with those most readily recognized by yeast mids pRS403, pRS404, pRS405 and pRS406 are Yeast when yeast expression systems are used. Integrating plasmids (YIps) and incorporate the yeast select 0114. The degeneracy of the genetic code permits the able markers HIS3, TRP1, LEU2 and URA3. Plasmid same amino acid sequence to be encoded and translated in spRS413-41.6 are Yeast Centromere plasmids (YCps). many different ways. For example, leucine, serine and argi 0119 Such vectors will generally include a selectable nine are each encoded by six different codons, while valine, marker, which may be one of any number of genes that exhibit proline, threonine, alanine and glycine are each encoded by a dominant phenotype for which a phenotypic assay exists to four different codons. However, the frequency of use of such enable transformants to be selected. Preferred selectable synonymous codons varies from genome to genome among markers are those that complement host cell auxotrophy, pro eukaryotes and prokaryotes. For example, synonymous vide antibiotic resistance or enable a cell to utilize specific codon-choice patterns among mammals are very similar, carbon sources, and include LEU2 (Broach et al. ibid.), while evolutionarily distant organisms such as yeast (such as URA3 (Botstein et al., Gene 8: 17, 1979), HIS3 (Struhl et al., S. cerevisiae), bacteria (Such as E. coli) and insects (such as D. ibid.) or POT1 (Kawasaki and Bell, EP171,142). Other suit melanogaster) reveal a clearly different pattern of genomic able selectable markers include the CAT gene, which confers codon use frequencies (Grantham, R., et al., Nucl. Acid Res., chloramphenicol resistance on yeast cells. Preferred promot 8, 49-62 (1980); Grantham, R., et al., Nucl. Acid Res., 9, ers for use in yeast include promoters from yeast glycolytic 43-74 (1981); Maroyama, T., et al., Nucl. Acid Res., 14, genes (Hitzeman et al., J Biol. Chem. 225: 12073-12080, 151-197 (1986); Aota, S., et al., Nucl. Acid Res., 16,315-402 1980; Alber and Kawasaki, J. Mol. Appl. Genet. 1: 419-434, (1988); Wada, K., et al., Nucl. Acid Res., 19 Supp., 1981 1982; Kawasaki, U.S. Pat. No. 4,599,311) or alcohol dehy 1985 (1991); Kurland, C. G., FEBS Lett., 285, 165-169 drogenase genes (Young et al., in Genetic Engineering of (1991)). These differences in codon-choice patterns appear to Microorganisms for Chemicals, Hollaender et al., (eds.), p. contribute to the overall expression levels of individual genes 355, Plenum, N.Y., 1982: Ammerer, Meth. Enzymol. 101: by modulating peptide elongation rates. (Kurland, C. G., 192-201, 1983). In this regard, particularly preferred promot FEBS Lett., 285, 165-169 (1991); Pedersen, S., EMBO.J., 3, ers are the TPI1 promoter (Kawasaki, U.S. Pat. No. 4,599, 2895-2898 (1984); Sorensen, M.A., J. Mol. Biol. 207, 365 311) and the ADH2-4 (see U.S. Pat. No. 6.291,212 promoter 377 (1989); Randall, L. L., et al., Eur. J. Biochem., 107, (Russell et al., Nature 304: 652-654, 1983). The expression 375-379 (1980); Curran, J. F., and Yarus, M., J. Mol. Biol. units may also include a transcriptional terminator. A pre 209, 65-77 (1989); Varenne, S., et al., J. Mol. Biol., 180, ferred transcriptional terminator is the TPI1 terminator (Al 549-576 (1984), Varenne, S., et al., J. Mol, Biol., 180,549 ber and Kawasaki, ibid.). Other preferred vectors and pre 576 (1984); Garel, J.-P. J. Theor. Biol., 43,211-225 (1974); ferred components such as promoters and terminators of a Ikemura, T. J. Mol. Biol. 146, 1-21 (1981); Ikemura, T. J. yeast expression system are disclosed in European Patents EP Mol. Biol. 151,389-409 (1981)). 0258067, EP 0286424, EP0317254, EP 0387319, EP 0115 The preferred codon usage frequencies for a syn 0386222, EP 0424117, EP 0431880, and EP 1002095; Euro thetic gene should reflect the codon usages of nuclear genes pean Patent Publications EP 0828759, EP 0764209, EP derived from the exact (or as closely related as possible) 0749478, and EP 0889949; PCT Publication WO 00/44772 genome of the cell/organism that is intended to be used for and WO94/04687; and U.S. Pat. Nos. 5,739,007: 5,637,504; recombinant protein expression, particularly that of yeast 5,302,697; 5,260.202; 5,667,986; 5,728,553; 5,783,423; species. As discussed above, in one preferred embodiment the 5,965,386; 6150,133; 6,379,924; and 5,714,377; which are human Tf sequence is codon optimized, before or after modi herein incorporated by reference in their entirety. fication as herein described for yeast expression as may be the I0120 In addition to yeast, modified fusion proteins of the therapeutic protein nucleotide sequence(s). present invention can be expressed in filamentous fungi, for 0116 Vectors example, Strains of the fungi Aspergillus. Examples of useful 0117 Expression units for use in the present invention will promoters include those derived from Aspergillus nidulans generally comprise the following elements, operably linked glycolytic genes, such as the adh3 promoter (McKnight et al., in a 5' to 3' orientation: a transcriptional promoter, a secretory EMBO J. 4: 2093-2099, 1985) and the tpiA promoter. An signal sequence, a DNA sequence encoding a modified Tf example of a suitable terminator is the adh3 terminator (McK fusion protein comprising transferrin protein or a portion of a night et al., ibid.). The expression units utilizing Such com transferrin protein joined to a DNA sequence encoding a ponents may be cloned into vectors that are capable of inser therapeutic protein or peptide of interest and a transcriptional tion into the chromosomal DNA of Aspergillus, for example. terminator. As discussed above, any arrangement of the thera 0121 Mammalian expression vectors for use in carrying peutic protein or peptide fused to or within the Tfportion may out the present invention will include a promoter capable of be used in the vectors of the invention. The selection of directing the transcription of the modified Tffusion protein. Suitable promoters, signal sequences and terminators will be Preferred promoters include viral promoters and cellular pro determined by the selected host cell and will be evident to one moters. Preferred viral promoters include the major late pro skilled in the art and are discussed more specifically below. moter from adenovirus 2 (Kaufman and Sharp, Mol. Cell. 0118 Suitable yeast vectors for use in the present inven Biol. 2: 1304-13199, 1982) and the SV40 promoter (Subra tion are described in U.S. Pat. No. 6,291,212 and include mani et al., Mol. Cell. Biol. 1: 854-864, 1981). Preferred YRp7 (Struhl et al., Proc. Natl. Acad. Sci. USA 76: 1035 cellular promoters include the mouse metallothionein 1 pro 1039, 1978), YEp13 (Broach et al., Gene 8: 121-133, 1979), moter (Palmiter et al., Science 222: 809–814, 1983) and a plDB249 and plDB219 (Beggs, Nature 275:104-108, 1978), mouse VK (see U.S. Pat. No. 6,291.212) promoter (Grant et US 2011/009 1543 A1 Apr. 21, 2011

al., Nuc. Acids Res. 15:5496, 1987). A particularly preferred fusion protein of the invention. In addition to the transformed promoter is a mouse V. (see U.S. Pat. No. 6,291.212) pro host cells themselves, the present invention also includes a moter (Loh et al., ibid.). Such expression vectors may also culture of those cells, preferably a monoclonal (clonally contain a set of RNA splice sites located downstream from the homogeneous) culture, or a culture derived from a mono promoter and upstream from the DNA sequence encoding the clonal culture, in a nutrient medium. If the polypeptide is transferrin fusion protein. Preferred RNA splice sites may be secreted, the medium will contain the polypeptide, with the obtained from adenovirus and/or immunoglobulin genes. cells, or without the cells if they have been filtered or centri 0122. Also contained in the expression vectors is a poly fuged away. adenylation signal located downstream of the coding Host cells for use in practicing the present invention include sequence of interest. Polyadenylation signals include the eukaryotic cells, and in some cases prokaryotic cells, capable early or late polyadenylation signals from SV40 (Kaufman of being transformed or transfected with exogenous DNA and and Sharp, ibid.), the polyadenylation signal from the aden grown in culture, such as cultured mammalian, insect, fungal, ovirus 5 E1B region and the human growth hormone gene plant and bacterial cells. terminator (DeNoto et al., Nucl. Acid Res. 9: 3719-3730, 0128 Fungal cells, including species of yeast (e.g., Sac 1981). A particularly preferred polyadenylation signal is the charomyces spp., Schizosaccharomyces spp., Pichia spp.) V. (see U.S. Pat. No. 6,291.212) gene terminator (Loh et al., may be used as host cells within the present invention. ibid.). The expression vectors may include a noncoding viral Examples of fungi including yeasts contemplated to be useful leader sequence, such as the adenovirus 2 tripartite leader, in the practice, of the present invention as hosts for expressing located between the promoter and the RNA splice sites. Pre the transferrin fusion protein of the inventions are Pichia ferred vectors may also include enhancer sequences. Such as (some species of which were formerly classified as the SV40 enhancer and the mouse (see U.S. Pat. No. 6,291, Hansenula), Saccharomyces, Kluyveromyces spergillus, 212) enhancer (Gillies, Cell 33: 717-728, 1983). Expression Candida, Torulopsis, Torulaspora, Schizosaccharomyces, vectors may also include sequences encoding the adenovirus Citeromyces, Pachysolen, Zygosaccharomyces, Debaromy VARNAS. ces, Trichoderma, Cephalosporium, Humicola, Mucor, Neu 0123 Transformation rospora, Yarrowia, Metschunikowia, Rhodosporidium, Leu 0.124 Techniques for transforming fungi are well known cosporidiunz, Botryoascus, Sporidiobolus, Endomycopsis, in the literature, and have been described, for instance, by and the like. Examples of Saccharomyces spp. are S. cerevi Beggs (ibid.), Hinnen et al. (Proc. Natl. Acad. Sci. USA 75: siae, S. italicus and S. rouxii. Examples of Kluyveromyces 1929-1933, 1978), Yelton et al., (Proc. Natl. Acad. Sci. USA spp. are K. fragilis, K. lactis and K. marxianus. A suitable 81: 1740-1747, 1984), and Russell (Nature 301: 167-169, Torulaspora species is T. delbrueckii. Examples of Pichia 1983). Other techniques for introducing cloned DNA spp. are P angusta (formerly H. polymorpha), P anomala sequences into fungal cells, such as electroporation (Becker (formerly H. anomala) and P. pastoris. and Guarente, Methods in Enzymol. 194: 182-187, 1991) I0129. Particularly useful host cells to produce the Tf may be used. The genotype of the host cell will generally fusion proteins of the invention are the methylotrophic Pichia contain a genetic defect that is complemented by the select pastoris (Steinlein et al. (1995) Protein Express. Purif. 6:619 able marker present on the expression vector. Choice of a 624). Pichia pastoris has been developed to be an outstanding particular host and selectable marker is well within the level host for the production of foreign proteins since its alcohol of ordinary skill in the art. oxidase promoter was isolated and cloned; its transformation 0.125 Cloned DNA sequences comprising modified Tf was first reported in 1985. P. pastoris can utilize methanol as fusion proteins of the invention may be introduced into cul a carbon Source in the absence of glucose. The P. pastoris tured mammalian cells by, for example, calcium phosphate expression system can use the methanol-induced alcohol oxi mediated transfection (Wigler et al., Cell 14: 725, 1978; Cor dase (AOX1) promoter, which controls the gene that codes for saro and Pearson, Somatic Cell Genetics 7: 603, 1981; the expression of alcohol oxidase, the enzyme which cata Graham and Van der Eb, Virology 52: 456, 1973.) Other lyzes the first step in the metabolism of methanol. This pro techniques for introducing cloned DNA sequences into mam moter has been characterized and incorporated into a series of malian cells, such as electroporation (Neumann et al., EMBO P. pastoris expression vectors. Since the proteins produced in J. 1: 841-845, 1982), or lipofection may also be used. In order P. pastoris are typically folded correctly and secreted into the to identify cells that have integrated the cloned DNA, a select medium, the fermentation of genetically engineered P. pas able marker is generally introduced into the cells along with toris provides an excellent alternative to E. coli expression the gene or cDNA of interest. Preferred selectable markers for systems. A number of proteins have been produced using this use in cultured mammalian cells include genes that confer system, including tetanus toxin fragment, Bordatella pertus resistance to drugs, such as neomycin, hygromycin, and sis pertactin, human serum albumin and lysozyme. methotrexate. The selectable marker may be an amplifiable 0.130 Strains of the yeast Saccharomyces cerevisiae are selectable marker. A preferred amplifiable selectable marker another preferred host. In a preferred embodiment, a yeast is the DHFR gene. A particularly preferred amplifiable cell, or more specifically, a Saccharomyces cerevisiae host marker is the DHFR (see U.S. Pat. No. 6.291,212) cDNA cell that contains a genetic deficiency in a gene required for (Simonsen and Levinson, Proc. Natl. Acad. Sci. USA 80: asparagine-linked glycosylation of glycoproteins is used. S. 2495-2499, 1983). Selectable markers are reviewed by Thilly cerevisiae host cells having Such defects may be prepared (Mammalian Cell Technology, Butterworth Publishers, using standard techniques of mutation and selection, Stoneham, Mass.) and the choice of selectable markers is well although many available yeast strains have been modified to within the level of ordinary skill in the art. prevent or reduce glycosylation or hypermannosylation. Bal 0.126 Host Cells lou et al. (J. Biol. Chem. 255: 5986-5991, 1980) have 0127. The present invention also includes a cell, prefer described the isolation of mannoprotein biosynthesis mutants ably a yeast cell transformed to express a modified transferrin that are defective in genes which affect asparagine-linked US 2011/009 1543 A1 Apr. 21, 2011

glycosylation. Gentzsch and Tanner (Glycobiology 7:481 can then be amplified and used to infect insect cell cultures to 486, 1997) have described a family of at least six genes produce large amounts of the desired protein. (PMT1-6) encoding enzymes responsible for the first step in 0.135 Tffusion proteins of the present invention may also O-glycosylation of proteins in yeast. Mutants defective in one be produced using transgenic plants and animals. For or more of these genes show reduced O-linked glycosylation example, sheep and goats can make the therapeutic protein in and/or altered specificity of O-glycosylation. their milk. Or tobacco plants can include the protein in their 0131 To optimize production of the heterologous pro leaves. Both transgenic plant and animal production of pro teins, it is also preferred that the host strain carries a mutation, teins comprises adding a new gene coding the fusion protein Such as the S. cerevisiae pep4 mutation (Jones, Genetics 85: 23-33, 1977), which results in reduced proteolytic activity. into the genome of the organism. Not only can the transgenic Host strains containing mutations in other protease encoding organism produce a new protein, but it can also pass this regions are particularly useful to produce large quantities of ability onto its offspring. the Tffusion proteins of the invention. 0.136 Secretory Signal Sequences 0132) Host cells containing DNA constructs of the present 0.137 The terms “secretory signal sequence' or “signal invention are grown in an appropriate growth medium. As sequence' or 'secretion leader sequence' are used inter used herein, the term “appropriate growth medium’ means a changeably and are described, for example in U.S. Pat. No. medium containing nutrients required for the growth of cells. 6,291,212 and U.S. Pat. No. 5,547,871, both of which are Nutrients required for cell growth may include a carbon herein incorporated by reference in their entirety. Secretory Source, a nitrogen Source, essential amino acids, vitamins, signal sequences or signal sequences or secretion leader minerals and growth factors. The growth medium will gener sequences encode secretory peptides. A secretory peptide is ally select for cells containing the DNA construct by, for an amino acid sequence that acts to direct the Secretion of a example, drug selection or deficiency in an essential nutrient mature polypeptide or protein from a cell. Secretory peptides which is complemented by the selectable marker on the DNA are generally characterized by a core of hydrophobic amino construct or co-transfected with the DNA construct. Yeast cells, for example, are preferably grown in a chemically acids and are typically (but not exclusively) found at the defined medium, comprising a carbon Source, e.g. Sucrose, a amino termini of newly synthesized proteins. Very often the non-amino acid nitrogen Source, inorganic salts, vitamins and secretory peptide is cleaved from the mature protein during essential amino acid Supplements. The pH of the medium is secretion. Secretory peptides may contain processing sites preferably maintained at a pH greater than 2 and less than 8, that allow cleavage of the signal peptide from the mature preferably at pH 5.5-6.5. Methods for maintaining a stable pH protein as it passes through the secretory pathway. Processing include buffering and constant pH control. Preferred buffer sites may be encoded within the signal peptide or may be ing agents include Succinic acid and Bis-Tris (Sigma Chemi added to the signal peptide by, for example, in vitro mutagen cal Co., St. Louis, Mo.). Yeast cells having a defect in a gene CS1S. required for asparagine-linked glycosylation are preferably 0.138. Secretory peptides may be used to direct the secre grown in a medium containing an osmotic stabilizer. A pre tion of modified Tffusion proteins of the invention. One such ferred osmotic stabilizer is sorbitol supplemented into the secretory peptide that may be used in combination with other medium at a concentration between 0.1 M and 1.5 M., pref secretory peptides is the alpha mating factor leader sequence. erably at 0.5 M or 1.0 M. Secretory signal sequences or signal sequences or secretion 0.133 Cultured mammalian cells are generally grown in leader sequences are required for a complex series of post commercially available serum-containing or serum-free translational processing steps which result in secretion of a media. Selection of a medium appropriate for the particular protein. If an intact signal sequence is present, the protein cell line used is within the level of ordinary skill in the art. being expressed enters the lumen of the rough endoplasmic Transfected mammaliancells are allowed to grow for a period reticulum and is then transported through the Golgi apparatus of time, typically 1-2 days, to begin expressing the DNA to secretory vesicles and is finally transported out of the cell. sequence(s) of interest. Drug selection is then applied to Generally, the signal sequence immediately follows the ini select for growth of cells that are expressing the selectable tiation codon and encodes a signal peptide at the amino marker in a stable fashion. Forcells that have been transfected terminal end of the protein to be secreted. In most cases, the with an amplifiable selectable marker the drug concentration signal sequence is cleaved off by a specific protease, called a may be increased in a stepwise manner to select for increased signal peptidase. Preferred signal sequences improve the pro copy number of the cloned sequences, thereby increasing cessing and export efficiency of recombinant protein expres expression levels. sion using viral, mammalian or yeast expression vectors. In 0134 Baculovirus/insect cell expression systems may Some cases, the native Tf signal sequence may be used to also be used to produce the modified Tffusion proteins of the express and secrete fusion proteins of the invention. invention. The BacPAKTM Baculovirus Expression System 0.139 Linkers (BDBiosciences (Clontech)) expresses recombinant proteins 0140. The Tf moiety and therapeutic protein moiety(s) of at high levels in insect host cells. The target gene is inserted the modified transferrin fusion proteins of the invention can into a transfer vector, which is cotransfected into insect host be fused directly or using a linker peptide of various lengths cells with the linearized BacPAK6 viral DNA. The BacPAK6 to provide greater physical separation and allow more spatial DNA is missing an essential portion of the baculovirus mobility between the fused proteins and thus maximize the genome. When the DNA recombines with the vector, the accessibility of the therapeutic protein portion, for instance, essential element is restored and the target gene is transferred for binding to its cognate receptor. The linker peptide may to the baculovirus genome. Following recombination, a few consist of amino acids that are flexible or more rigid. For viral plaques are picked and purified, and the recombinant example, a linker Such as but not limited to a poly-glycine phenotype is verified. The newly isolated recombinant virus stretch. The linker can be less than about 50, 40, 30, 20, or 10 US 2011/009 1543 A1 Apr. 21, 2011

amino acid residues. The linker can be covalently linked to protein containing that therapeutic protein as the therapeutic and between the transferrin protein or portion thereof and the protein portion of the fusion can be assayed, e.g., by means therapeutic protein. well-known in the art, such as, for example, reducing and 0.141. Detection of Tf Fusion Proteins non-reducing gel chromatography, protein affinity chroma 0142 Assays for detection of biologically active modified tography, and affinity blotting. Other methods will be known transferrin-therapeutic protein fusions may include Western to the skilled artisan and are within the scope of the invention. transfer, protein blot or colony filter as well as activity based 0146 Isolation/Purification of Modified Transferrin assays that detect the fused therapeutic protein. A Western Fusion Proteins transfer filter may be prepared using the method described by 0147 Secreted, biologically active, modified transferrin Towbin et al. (Proc. Natl. Acad. Sci. USA 76: 4350-4354, fusion proteins may be isolated from the medium of host cells 1979). Briefly, samples are electrophoresed in a sodium dode grown under conditions that allow the secretion of the bio cylsulfate polyacrylamide gel. The proteins in the gel are logically active fusion proteins. The cell material is removed electrophoretically transferred to nitrocellulose paper. Pro from the culture medium, and the biologically active fusion tein blot filters may be prepared by filtering supernatant proteins are isolated using isolation techniques known in the samples or concentrates through nitrocellulose filters using, art. Suitable isolation techniques include precipitation and for example, a Minifold (Schleicher & Schuell, Keene, N. fractionation by a variety of chromatographic methods, H.). Colony filters may be prepared by growing colonies on a including gel filtration, ion exchange chromatography and nitrocellulose filter that has been laid across an appropriate affinity chromatography. growth medium. In this method, a Solid medium is preferred. 0.148. A particularly preferred purification method is The cells are allowed to grow on the filters for at least 12 affinity chromatography on an iron binding or metal chelating hours. The cells are removed from the filters by washing with column or an immunoaffinity chromatography using an anti an appropriate buffer that does not remove the proteins bound body directed against the transferrin or therapeutic protein or to the filters. A preferred buffer comprises 25 mM Tris-base, peptide portion of the polypeptide fusion. The antibody is 19 mM glycine, pH 8.3, 20% methanol. preferably immobilized or attached to a solid support or sub 0143 Fusion proteins of the invention may also be strate. A particularly preferred substrate is CNBr-activated detected by assaying for the activity of the therapeutic protein Sepharose (Pharmacia LKB Technologies, Inc., Piscataway, moiety. Such assays are readily available, including but not N.J.). By this method, the medium is combined with the limited to, those assays described in Table 1. Specifically, antibody/substrate under conditions that will allow binding to transferrin fusion proteins of the invention may be assayed for occur. The complex may be washed to remove unbound mate functional activity (e.g., biological activity or therapeutic rial, and the transferrin fusion protein is released or eluted activity) using the assay referenced in the “Exemplary Activ through the use of conditions unfavorable to complex forma ity Assay” column of Table 1. Additionally, one of skill in the tion. Particularly useful methods of elution include changes art may routinely assay fragments of a therapeutic protein in pH, wherein the immobilized antibody has a high affinity corresponding to a therapeutic protein portion of a fusion for the ligand at a first pH and a reduced affinity at a second protein of the invention, for activity using assays referenced (higher or lower) pH; changes in concentration of certain in its corresponding row of Table 1. Further, one of skill in the chaotropic agents; or through the use of detergents. art may routinely assay fragments of a modified transferrin 0149 Labeled Modified Transferrin Fusion Proteins protein for activity using assays known in the art. 0150 Transferrin fusion proteins of the present invention 0144. For example, in one embodiment where one is may also be labeled with a radioisotope or other imaging assaying for the ability of a transferrin fusion protein of the agent and used for in vivo diagnostic purposes. Preferred invention to bind or compete with a therapeutic protein for radioisotope imaging agents include iodine-125 and techne binding to an anti-therapeutic polypeptide antibody and/or tium-99, with technetium-99 being particularly preferred. anti-transferrin antibody, various immunoassays known in Methods for producing protein-isotope conjugates are well the art can be used, including but not limited to, competitive known in the art, and are described by, for example, Eckelman and non-competitive assay systems using techniques such as et al. (U.S. Pat. No. 4,652,440), Parker et al. (WO 87/05030) radioimmunoassays, ELISA (enzyme linked immunosorbent and Wilber et al. (EP203,764). Alternatively, the transferrin assay), Sandwich immunoassays, immunoradiometric fusion proteins may be bound to spin label enhancers and assays, gel diffusion precipitation reactions, immunodiffu used for magnetic resonance (MR) imaging. Suitable spin sion assays, in situ immunoassays (using colloidal gold, label enhancers include stable, sterically hindered, free radi enzyme or radioisotope labels, for example), western blots, cal compounds such as nitroxides. Methods for labeling precipitation reactions, agglutination assays (e.g., gel agglu ligands for MR imaging are disclosed by, for example, Coff tination assays), complement fixation assays, immunofluo man et al. (U.S. Pat. No. 4,656,026). For administration, the rescence assays, protein A assays, and immunoelectrophore labeled transferrin fusion proteins are combined with a phar sis assays, etc. In one embodiment, antibody binding is maceutically acceptable carrier or diluent, Such as sterile detected by detecting a label on the primary antibody. In saline or sterile water. Administration is preferably by bolus another embodiment, the primary antibody is detected by injection, preferably intravenously. detecting binding of a secondary antibody or reagent to the 0151. Production of Fusion Proteins primary antibody. In a further embodiment, the secondary 0152 The present invention further provides methods for antibody is labeled. Many means are known in the art for producing a modified fusion protein of the invention using detecting binding in an immunoassay and are within the nucleic acid molecules herein described. In general terms, the Scope of the present invention. production of a recombinant form of a protein typically 0145. In a further embodiment, where a binding partner involves the following steps. (e.g., a receptor or a ligand) of a therapeutic protein is iden 0153. A nucleic acid molecule is first obtained that tified, binding to that binding partner by a transferrin fusion encodes a transferrin fusion protein of the invention. The US 2011/009 1543 A1 Apr. 21, 2011

nucleic acid molecule is then preferably placed in operable (0160. In further embodiments, a modified transferrin linkage with Suitable control sequences, as described above, fusion protein of the invention may contain at least afragment to form an expression unit containing the protein open read or variant of a therapeutic protein, and/or at least a fragment ing frame. The expression unit is used to transform a suitable or variant of an antibody. In a further embodiment, the trans host and the transformed host is cultured under conditions ferrin fusion proteins can contain peptide fragments or pep that allow the production of the recombinant protein. Option tide variants of proteins or antibodies wherein the variant or ally the recombinant protein is isolated from the medium or fragment retains at least one biological ortherapeutic activity. from the cells; recovery and purification of the protein may The transferrin fusion proteins can contain therapeutic pro not be necessary in Some instances where some impurities teins that can be peptide fragments or peptide variants at least may be tolerated. about 4, at least 5, at least 6, at least 7, at least 8, at least 9, at 0154 Each of the foregoing steps can be accomplished in least 10, at least 11, at least 12, at least 13, at least 14, at least a variety of ways. For example, the construction of expression 15, at least 20, at least 25, at least 30, at least 35, or at least vectors that are operable in a variety of hosts is accomplished about 40, at least about 50, at least about 55, at least about 60 using appropriate replicons and control sequences, as set or at least about 70 or more amino acids in length fused to the forth above. The control sequences, expression vectors, and Nand/or C termini, inserted within, or inserted into a loop of transformation methods are dependent on the type of host cell a modified transferrin. used to express the gene and were discussed in detail earlier 0.161. In another embodiment, the modified transferrin and are otherwise known to persons skilled in the art. Suitable fusion molecules contain a therapeutic protein portion that restriction sites can, if not normally available, be added to the can be fragments of a therapeutic protein that include the full ends of the coding sequence so as to provide an excisable gene length protein as well as polypeptides having one or more to insert into these vectors. A skilled artisan can readily adapt residues deleted from the amino terminus of the amino acid any host/expression system known in the art for use with the Sequence. nucleic acid molecules of the invention to produce a desired (0162. In another embodiment, the modified transferrin recombinant protein. fusion molecules contain a therapeutic protein portion that 0155 As discussed above, any expression system may be can be fragments of a therapeutic protein that include the full used, including yeast, bacterial, animal, plant, eukaryotic and length protein as well as polypeptides having one or more prokaryotic systems. In some embodiments, yeast, mamma residues deleted from the carboxy terminus of the amino acid lian cell culture and transgenic animal or plant production Sequence. systems are preferred. In other embodiments, yeast systems (0163. In another embodiment, the modified transferrin that have been modified to reduce native yeast glycosylation, fusion molecules contain a therapeutic protein portion that hyper-glycosylation or proteolytic activity may be used. can have one or more amino acids deleted from both the 0156 Therapeutic Molecules amino and the carboxy termini. 0157 Any therapeutic molecule may be used as the fusion (0164. In another embodiment, the modified transferrin partner to Tf according to the methods and compositions of fusion molecules contain a therapeutic protein portion that is the present invention. As used herein, a therapeutic molecule at least about 80%, 85%, 90%, 95%, 96%, 97%, 98% or 99% is typically a protein or peptide capable of exerting a benefi identical to a reference therapeutic protein set forth herein, or cial biological effect in vitro or in vivo and includes proteins fragments thereof. In further embodiments, the transferrin or peptides that exert a beneficial effect in relation to normal fusion molecules contain a therapeutic protein portion that is homeostasis, physiology or a disease state. Therapeutic mol at least about 80%, 85%, 90%, 95%, 96%, 97%, 98% or 99% ecules do not include, fusion partners commonly used as identical to reference polypeptides having the amino acid markers or protein purification aids, such as bacterial galac sequence of N- and C-terminal deletions as described above. tosidases (see for example, U.S. Pat. No. 5,986,067 and 0.165. In another embodiment, the modified transferrin Aldred et al. (1984) Biochem. Biophys. Res. Commun. 122: fusion molecules contain the therapeutic protein portion that 960-965). For instance, a beneficial effect as related to a is at least about 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% disease state includes any effect that is advantageous to the or 100%, identical to, for example, the native or wild-type treated Subject, including disease prevention, disease stabili amino acid sequence of a therapeutic protein. Fragments, of Zation, the lessening or alleviation of disease symptoms or a these polypeptides are also provided. modulation, alleviation or cure of the underlying defect to 0166 The therapeutic proteins corresponding to a thera produce an effect beneficial to the treated subject. peutic protein portion of a modified transferrin fusion protein 0158. A modified transferrin fusion protein of the inven of the invention, Such as cell Surface and secretory proteins, tion includes at least a fragment or variant of a therapeutic can be modified by the attachment of one or more oligosac protein and at least a fragment or variant of modified serum charide groups. The modification referred to as glycosylation transferrin, which are associated with one another, preferably can significantly affect the physical properties of proteins and by genetic fusion. can be important in protein stability, secretion, and localiza 0159. In one embodiment, the transferrin fusion protein tion. Glycosylation occurs at specific locations along the includes a modified transferrin molecule linked to a neurop polypeptide backbone. There are usually two major types of harmaceutical agent. In another embodiment, the modified glycosylation: glycosylation characterized by O-linked oli transferrin fusion protein includes transferrin at the carboxyl gosaccharides, which are attached to serine or threonine resi terminus linked to a neuropharmaceutical agent at the amino dues; and glycosylation characterized by N-linked oligosac terminus. In an alternate embodiment, the modified transfer charides, which are attached to asparagine residues in an rin fusion protein includes transferrin at the amino terminus Asn-X-Ser/Thr sequence, where X can be an amino acid linked to a neuropharmaceutical agent at the carboxy termi except proline. Variables Such as protein structure and cell nus. In specific embodiments, the neuropharmaceutical agent type influence the number and nature of the carbohydrate is either nerve growth factor or ciliary neurotrophic factor. units within the chains at different glycosylation sites. Gly US 2011/009 1543 A1 Apr. 21, 2011

cosylation isomers are also common at the same site within a majority of the residues of the complete or mature polypep given cell type. For example, several types of human inter tide are removed from the C-terminus. Whether a particular feron are glycosylated. polypeptide lacking the N-terminal and/or, C-terminal resi 0167. Therapeutic proteins corresponding to a therapeutic dues of a reference polypeptide retains therapeutic activity protein portion of a transferrinfusion protein of the invention, can readily be determined by routine methods described as well as analogs and variants thereof, may be modified so herein and/or otherwise known in the art. that glycosylation at one or more sites is altered as a result of 0172 Peptide fragments of the therapeutic proteins can be manipulation(s) of their nucleic acid sequence by the host cell fragments comprising, or alternatively, consisting of an in which they are expressed, or due to other conditions of their amino acid sequence that displays a therapeutic activity and/ expression. For example, glycosylation isomers may be pro or functional activity (e.g. biological activity) of the polypep duced by abolishing or introducing glycosylation sites, e.g., tide sequence of the therapeutic protein of which the amino by Substitution or deletion of amino acid residues, such as acid sequence is a fragment. Substitution of glutamine for asparagine, or unglycosylated 0173 The peptide fragments of the therapeutic protein recombinant proteins may be produced by expressing the may comprise only the N- and C-termini of the protein, i.e., proteins in host cells that will not glycosylate them, e.g. in the central portion of the therapeutic protein has been deleted. glycosylation-deficient yeast. These approaches are known in Alternatively, the peptide fragments may comprise non-adja the art. cent and/or adjacent portions of the central part of the thera 0168 Therapeutic proteins and their nucleic acid peutic protein. sequences are well known in the art and available in public 0.174. Other polypeptide fragments are biologically active databases such as Chemical Abstracts Services Databases fragments. Biologically active fragments are those exhibiting (e.g., the CAS Registry), GenBank, and GenSeq. The Acces activity similar, but not necessarily identical, to an activity of sion Numbers and sequences referred to below are herein a therapeutic protein used in the present invention. The bio incorporated by reference in their entirety. logical activity of the fragments may include an improved 0169. In other embodiments, the transferrin fusion pro desired activity, or a decreased undesirable activity. teins of the invention are capable of a therapeutic activity 0.175 Generally, variants of proteins are overall very simi and/or biologic activity, corresponding to the therapeutic lar, and, in many regions, identical to the amino acid sequence activity and/or biologic activity of the therapeutic protein of the therapeutic protein corresponding to a therapeutic pro listed in the corresponding row of Table 1 and elsewhere in tein portion of a transferrin fusion protein of the invention. this application. (See, e.g., the “Biological Activity” and Nucleic acids encoding these variants are also encompassed “Therapeutic Protein X columns of Table 1.) In further by the invention. embodiments, the therapeutically active protein portions of 0176 Further therapeutic polypeptides that may be used in the transferrin fusion proteins of the invention are fragments the invention are polypeptides encoded by polynucleotides or variants of the reference sequences cited herein. which hybridize to the complement of a nucleic acid molecule 0170 The present invention is further directed to modified encoding an amino-acid sequence of a therapeutic protein Tffusion proteins comprising fragments of the therapeutic under stringent hybridization conditions which are known to proteins herein described. Even if deletion of one or more those of skill in the art. (see, for example, Ausubel, F. M. et al., amino acids from the N-terminus of a protein results in modi eds., 1989 Current protocol in Molecular Biology, Green fication or loss of one or more biological functions of the Publishing Associates, Inc., and John Wiley & Sons Inc., therapeutic protein portion, other therapeutic activities and/or New. York). Polynucleotides encoding these polypeptides are functional activities (e.g., biological activities, ability to mul also encompassed by the invention. timerize, ability to bind a ligand) may still be retained. For 0177. By a polypeptide-having an amino acid sequence at example, the ability of polypeptides with N-terminal dele least, for example, 95% “identical to a query amino acid tions to induce and/or bind to antibodies which recognize the sequence of the present invention, it is intended that the amino complete or mature forms of the polypeptides generally will acid sequence of the Subject polypeptide is identical to the be retained with less than the majority of the residues of the query sequence except that the Subject polypeptide sequence complete polypeptide removed from the N-terminus. may include up to five amino acid alterations per each 100 Whether aparticular polypeptide lacking N-terminal residues amino acids of the query amino acid sequence. In other of a complete polypeptide retains such immunologic activi words, to obtaina polypeptidehaving anamino acid sequence ties can be assayed by routine methods described herein and at least 95% identical to a query amino acid sequence, up to otherwise known in the art. It is not unlikely that a mutant 5% of the amino acid residues in the Subject sequence may be with a large number of deleted N-terminal amino acid resi inserted, deleted, or substituted with another amino acid. dues may retain some biological or immunogenic activities. These alterations of the reference sequence may occur at the In fact, peptides composed of as few as six amino acid resi amino- or carboxy-terminal positions of the reference amino dues may often evoke an immune response. acid sequence or anywhere between those terminal positions, 0171 Also as mentioned above, even if deletion of one or interspersed either individually among residues in the refer more amino acids from the N-terminus or C-terminus of a ence sequence, or in one or more contiguous groups within therapeutic protein results in modification or loss of one or the reference sequence. more biological functions of the protein, other functional 0.178 As a practical matter, whether any particular activities (e.g., biological activities, ability to multimerize, polypeptide is at least about 80%, 85%, 90%, 95%, 96%, ability to bind a ligand) and/or therapeutic activities may still 97%, 98% or 99% identical to, for instance, the amino acid be retained. For example the ability of polypeptides with sequence of a transferrin fusion protein of the invention or a C-terminal deletions to induce and/or bind to antibodies fragment thereof (such, as the therapeutic protein portion of which recognize the complete or mature forms of the the transferrin fusion protein or the transferrin portion of the polypeptide generally will be retained when less than the transferrin fusion protein), can be determined conventionally US 2011/009 1543 A1 Apr. 21, 2011 using known computer programs. A preferred method for of ubiquitination, and they may be cyclic, with or without determining the best overall match between a query sequence branching. Cyclic, branched, and branched cyclic polypep (a sequence of the present invention) and a Subject sequence, tides may result from posttranslation natural processes or also referred to as a global sequence alignment, can be deter may be made by synthetic methods. Modifications include mined using the FASTDB computer program based on the acetylation, acylation, ADP-ribosylation, amidation, cova algorithm of Brufiaget al. (Comp. App. Biosci 245 (1990)). lent attachment of flavin, covalent attachment of a heme 0179 The polynucleotide variants of the invention may moiety, covalent attachment of a nucleotide or nucleotide contain alterations in the coding regions, non-coding regions, derivative, covalent attachment of a lipid or lipid derivative, or both. Polynucleotide variants containing alterations which covalent attachment of phosphotidylinositol, cross-linking, produce silent Substitutions, additions, or deletions, but do cyclization, disulfide bond formation, demethylation, forma not alter the properties or activities of the encoded polypep tion of covalent cross-links, formation of cysteine, glycosy tide may be used to produce modified Tffusion proteins. lation, GPI anchor formation, hydroxylation, iodination, Nucleotide variants produced by silent substitutions due to methylation, myristylation, oxidation, pegylation, pro the degeneracy of the genetic code can be utilized. Moreover, teolytic processing, phosphorylation, prenylation, racemiza polypeptide variants in which less than about 50, less than 40, tion, sulfation, transfer-RNA mediated addition of amino less than 30, less than 20, less than 10, or 5-50, 5-25, 5-10, acids to proteins such as arginylation, and ubiquitination. 1-5, or 1-2 amino acids are substituted, deleted, or added in (See, for instance, PROTEINS STRUCTURE AND MOLECULAR any combination can also be utilized. Polynucleotide variants PROPERTIES, 2nd Ed., T. E. Creighton, W. H. Freeman and can be produced for a variety of reasons, e.g., to optimize Company, New York (1993); POST TRANSLATIONAL codon expression for a particular host (change codons in the COVALENT MODIFICATION OF PROTEINS, B. C. human mRNA to those preferred by a host, such as, yeast or Johnson, Ed., Academic Press, New York, pgs. 1-12 (1983); E. coli as described above). Seifter et al. (1990) Meth. Enzymol. 182:626-646: Rattan et 0180. In other embodiments, the therapeutic protein moi al., Ann. N.Y. Acad. Sci. 663:48-62. ety has conservative Substitutions compared to the wild-type 0183 Therapeutic molecules that may be fused to or sequence. By "conservative Substitutions' is intended Swaps inserted into Tf include, but are not limited to, hormones, within groups such as replacement of the aliphatic or hydro matrix proteins, immunosuppressants, bronchodilators, car phobic amino acids Ala, Val, Leu and Ile; replacement of the diovascular agents, enzymes, CNS agents, neurotransmitters, hydroxyl residues Ser and Thr; replacement of the acidic receptor proteins or peptides, growth hormones, growth fac residues Asp and Glu; replacement of the amide residues Asn tors, antiviral peptides, fusogenic inhibitor peptides, cytok and Gln, replacement of the basic residues Lys, Arg, and His; ines, lymphokines, monokines, interleukins, colony Stimulat replacement of the aromatic residues Phe, Tyr, and Trp, and ing factors, differentiation factors, angiogenic factors, replacement of the Small-sized amino acids Ala, Ser, Thr, receptor ligands, cancer-associated proteins, antineoplastics, Met, and Gly. Guidance concerning how to make phenotypi viral peptides, antibiotic peptides, blood proteins, antagonist cally silent amino acid substitutions is provided, for example, proteins, transcription factors, anti-angiogenic factors, in Bowie et al., “Deciphering the Message in Protein antagonist proteins or peptides, receptor antagonists, anti Sequences: Tolerance to Amino Acid Substitutions. Science bodies, single chain antibodies and cell adhesion molecules. 247: 1306-1310 (1990). In specific embodiments, the Different therapeutic molecules may be combined into a polypeptides of the invention comprise, or alternatively, con single fusion protein to produce abi or multi-functional thera sist of fragments or variants of the amino acid sequence of a peutic molecule. Different molecules may also be used in therapeutic protein described herein and/or serum transferrin, combination to produce a fusion protein with a therapeutic and/modified transferrin protein of the invention, wherein the entity and a targeting entity. fragments or variants have 1-5, 5-10, 5-25, 5-50, 10-50 or 0.184 Cytokines are soluble proteins released by cells of 50-150 amino acid residue additions, substitutions, and/or the immune system, which act nonenzymatically through deletions when compared to the reference amino acid specific receptors to regulate immune responses. Cytokines sequence. In further embodiments, the amino acid substitu resemble hormones in that they act at low concentrations tions are conservative. Nucleic acids encoding these polypep bound with high affinity to a specific receptor. The term tides are also encompassed by the invention. “cytokine' is used herein to describe naturally occurring or 0181. The modified fusion proteins of the present inven recombinant proteins, analogs thereof, and fragments thereof tion can be composed of amino-acids joined to each other by which elicit a specific biological response in a cell which has peptide bonds or modified peptide bonds and may contain a receptor for that cytokine. Cytokines preferably include amino acids other than the 20 gene-encoded amino acids. The interleukins such as interleukin-2 (IL-2) (GenBank Acc. No. polypeptides may be modified by either natural processes, S77834), IL-3 (GenBank Acc. No. M14743), IL-4 (GenBank Such as post-translational processing, or by chemical modi Acc. No. M23442), IL-5 (GenBank Acc. No. J03478), IL-6 fication techniques which are well known in the art. Such (GenBank Acc. No. M14584), IL-7 (GenBank Acc. No. modifications are well described in basic texts and in more NM 000880), IL-10 (GenBank Acc. No. NM 000572), detailed monographs, as well as in a Voluminous research IL-12 (GenBank Acc. No. AF180562 and GenBank Acc. No. literature. AF180563), IL-13 (GenBankAcc. No. U10307), IL-14 (Gen 0182 Modifications can occur anywhere in a polypeptide, Bank Acc. No. XM 170924), IL-15 (GenBank Acc. No. including the peptide backbone, the amino acid side-chains X91233), IL-16 (GenBank Acc. No. NM 004.513), IL-17 and the amino or carboxy termini. It will be appreciated that (GenBank Acc. No. NM 002190) and IL-18 (GenBankAcc. the same type of modification may be present in the same or No. NM 001562), hematopoietic factors such as granulo varying degrees at several sites in a given polypeptide. Also, cyte-macrophage colony stimulating factor (GM-CSF) (Gen a given polypeptide may contain many types of modifica Bank Acc. No. X03021), granulocyte colony stimulating fac tions. Polypeptides may be branched, for example, as a result tor (G-CSF) (GenBank Acc. No. X03656), platelet activating US 2011/009 1543 A1 Apr. 21, 2011

factor (GenBank Acc. No. NM 000437) and erythropoietin FGF17 (GenBank Acc. No. NM 003867), FGF18 (Gen (GenBank Acc. No. X02158), tumor necrosis factors (TNF) Bank Acc. No. AF075292), FGF19 (GenBank Acc. No. such as TNFC. (GenBank Acc. No. X02910), lymphokines NM 005117), FGF20 (GenBank Acc. No. NM 019851), such as lymphotoxin-C. (GenBank Acc. No. X02911), lym FGF21 (GenBank Acc. No. NM 019113), FGF22 (Gen photoxin-B (GenBank Acc. No. L11016), leukoregulin, mac Bank Acc. No. NM 020637), and FGF23 (GenBank Acc. rophage migration inhibitory factor (GenBank Acc. No. No. NM 020638), angiogenin (GenBank Acc. No. M25639), and neuroleukin (GenBank Acc. No. K03515), M11567), brain-derived neurotrophic factor (GenBank Acc. regulators of metabolic processes such as leptin (GenBank No. M61176), ciliary neurotrophic growth factor (GenBank Acc. No. U43415), interferons such as interferon C. (IFNC) Acc. No. X60542), transforming growth factor-O. (TGF-C.) (GenBank Acc. No. M54886), IFNB (GenBank Acc. No. (GenBank Acc. No. X70340), TGF-B (GenBank Acc. No. V00534), IFNY (GenBank Acc. No. J00219), IFNC. (Gen X02812), nerve growth factor-O. (NGF-C.) (GenBank Acc. Bank Acc. No. NM 002177), thrombospondin 1 (THBS1) No. NM 010915), NGF-?3 (GenBank Acc. No. X52599), (GenBank Acc. No. NM 003246), THBS2 (GenBank Acc. tissue inhibitor of metalloproteinase 1 (TIMP1) (GenBank No. L12350), THBS3 (GenBank Acc. No. L38969), THES4 Acc. No. NM 003254), TIMP2 (GenBank Acc. No. (GenBank Acc. No. NM 003248), and chemokines. Prefer NM 003255), TIMP3 (GenBankAcc. No. U02571), TIMP4 ably, the modified transferrin-cytokine fusion protein of the (GenBank Acc. No. U76456) and macrophage stimulating 1 present invention displays cytokine biological activity. (GenBank Acc. No. L11924). 0185. The term “hormone' is used herein to describe any 0187. The term “matrix protein' is used hereinto describe one of a number of biologically active Substances that are proteins or peptides that are normally found in the extracel produced by certain cells or tissues and that cause specific lular matrix. These proteins may be functionally important biological changes or activities to occur in another cell or for strength, filtration, or adhesion. Matrix proteins prefer tissue located elsewhere in the body. Hormones preferably ably include collagens such as collagen I (GenBank Acc. No. include GLP-1 of glucagon preproprotein (GenBank Acc. Z74615), collagen II (GenBank Acc. No. X16711), collagen No. NM 002045), proinsulin (GenBank Acc. No. V00565), III (GenBank Acc. No. X14420), collagen IV (GenBank Acc. insulin (GenBankAcc. No. NM 000207), growth hormone 1 No. NM 001845), collagen V (GenBank Acc. No. (GenBank Acc. No. V00520), growth hormone 2 (GenBank NM 000393), collagen VI (GenBank Acc. No. Acc. No. F006060), growth hormone release factor (Gen NM 058175), collagen VII (GenBank Acc. No. L02870), Bank Acc. No. NM 021081), insulin-like growth factor I collagen VIII (GenBankAcc. No. NM 001850), collagen IX (GenBank Acc. No. M27544), insulin-like growth factor II (GenBankAcc. No. X54412), collagenX(GenBankAcc. No. (GenBank Acc. No. NM 000612), insulin-like growth factor X60382), collagen XI (GenBank Acc. No. J04177), and col binding protein 1 (IGFBP-1) (GenBank Acc. No. M59316), lagen XII (GenBank Acc. No. U73778), laminin proteins IGFBP-2 (GenBank Acc. No. X16302), IGFBP-3 (GenBank such as LAMA2 (GenBank Acc. No. NM 000426), LAMA3 Acc. No. NM 000598), IGFBP-4 (GenBank Acc. No. (GenBank Acc. No. L34155), LAMA4 (GenBank Acc. No. Y12508), IGFBP-5 (GenBank Acc. No. M65062), IGFBP-6 NM 002290), LAMB1 (GenBank Acc. No. NM 002291), (GenBank Acc. No. NM 002178), IGFBP-7 (GenBank Acc. LAMB3 (GenBank Acc. No. L25541), LAMC1 (GenBank No. NM 001553), chorionic gonadotropin 13 chain (Gen Acc. No. NM 002293), nidogen (GenBank Acc. No. Bank Acc. No. NM 033142), chorionic gonadotropin a NM 002508), C-tectorin (GenBank Acc. No. chain (GenBank Acc. No. NM 000735), luteinizing hor NM 005422), B-tectorin (GenBank Acc. No. mone B (GenBank Acc. No. X00264), follicle-stimulating NM 058222), and fibronectin (GenBank Ace. No. X02761). hormone B (GenBank Acc. No. NM 000510), thyroid 0188 The term “blood proteins” are traditionally defined stimulating hormone B (GenBank Acc. No. NM 000549), as those sourced from plasma, many now commonly pro prolactin (GenBank Acc. No. NM 000948), pro-opiomel duced by recombinant means, and include, but are not limited anocortin (GenBank Acc. No. VO1510), corticotropin to native serum proteins, derivatives, fragments and mutants (ACTH), B-lipotropin, C.-melanocyte stimulating hormone or variants thereof, blood clotting factors, derivatives, (C-MSH), Y-lipotropin, B-MSH, B-endorphin, and corticotro mutants, variants and fragments (including factors VII, VIII, pin-like intermediate lobe peptide (CLIP). IX, X), protease inhibitors (antithrombin 3, alpha-1 antit 0186 The term “growth factor” is used herein to describe rypsin), urokinase-type plasminogen activator, immunoglo any protein or peptide that binds to a receptor to stimulate cell bulins, von Willebrand factor and von Willebrand mutants, proliferation. Growth factors preferably include platelet-de fibronectin, fibrinogen, thrombin and hemoglobin. rived growth factor-C. (PDGF-C) (GenBank Acc. No. 0189 The term “enzyme” is used herein to describe any X03795); PDGF-B (GenBankAcc. No. X02811), steroid hor protein or proteinaceous Substance which catalyzes a specific mones, epidermal growth factor (EGF) (GenBank Acc. No. reaction without itself being permanently altered or NM 001963), fibroblast growth factors such as fibroblast destroyed. Enzymes preferably include coagulation factors growth factor 1 (FGF1) (GenBank Acc. No. NM 000800), such as F2 (GenBank Acc. No. XM 170688), F7 (GenBank FGF2 (GenBank Acc. No. NM 002006), FGF3 (GenBank Acc. No. XM 027508), F8 (GenBank Acc. No. Acc. No. NM 005247), FGF4 (GenBank Acc. No. XM 013124), F9 (GenBank Acc. No. NM 000133), F10 NM 002007), FGF5 (GenBank Acc. No. M37825), FGF6 (GenBank Acc. No. AF503510) and others, matrix metallo (GenBank Acc. No. X57075), FGF7 (GenBank Acc. No. proteinases such as matrix metalloproteinase I (GenBank NM 002009), FGF8 (GenBankAcc. No. AHO06649), FGF9 Acc. No. MMP1) (GenBank Acc. No. NM 002421), MMP2 (GenBank Acc. No. NM 002010), FGF10 (GenBank Acc. (GenBank Acc. No. NM 004530), MMP3 (GenBank Acc. No. AB002097), FGF11 (GenBank Acc. No. NM 004112), No. NM 002422), MMP7 (GenBank Acc. No. FGF12 (GenBank Acc. No. NM 021032), FGF13 (Gen NM 002423), MMP8 (GenBank Acc. No. NM 002424), Bank Acc. No. NM 004114), FGF14 (GenBank Acc. No. MMP9 (GenBank Acc. No. NM 004994), MMP10 (Gen NM 004115), FGF16 (GenBank Acc. No. AB009391), Bank Acc. No. NM 002425), MMP12 (GenBank Acc. No. US 2011/009 1543 A1 Apr. 21, 2011

NM 002426), MMP13 (GenBank Acc. No. X75308), (LDL) receptor, an acetylated LDL receptor, a tumor necrosis MMP20 (GenBank Acc. No. NM 004771), adenosine factor C. receptor, a transforming growth factor B receptor, a deaminase (GenBank Acc. No. NM 000022), mitogen acti cytokine receptor, an immunoglobulin Fc receptor, a hor vated protein kinases such as MAPK3 (GenBank Acc. No. mone receptor, a glucose receptor, a glycolipid receptor, and XM 055766), MAP2K2 (GenBankAcc. No. NM 030662), a glycosaminoglycan receptor. In other embodiments, ligand MAP2K1 (GenBank Acc. No. NM 002755), MAP2K4 binding proteins include CD2 (M14362), CD3G (NM (GenBank Acc. No. NM 003010), MAP2K7 (AF013588), 000073), CD3D (NM 000732), CD3E (NM 000733), and MAPK12 (NM 002969), kinases such as JNKK1 (Gen CD37 (JO4132), CD28 (NM 006139), CD4 (GenBank Acc. Bank Acc. No. U17743), JNKK2 (GenBank Acc. No. No. NM 000616), CD1A (GenBank Acc. No. M28825), AF014.401), JAK1 (M64174), JAK2 (NM 004972), and CD1B (GenBank Acc. No. NM 001764), CD1C (GenBank JAK3 (NM 000215), and phosphatases such as PPM1A Acc. No. NM 001765), CD1D (GenBank Acc. No. (GenBank Acc. No. NM 021003) and PPM1D (GenBank NM 001766), CD80 (GenBank Acc. No. NM 005191), Acc. No. NM 003620). GNB3 (GenBank Acc. No. AF501884), CTLA-4 (GenBank 0190. The term “transcription factors' is used herein to Acc. No. NM 005214), intercellular adhesion molecules describe any protein or peptide involved in the transcription such as ICAM-1 (NM 000201), ICAM-2 (NM 000873), of protein-coding genes. Transcription factors may include and ICAM-3 (NM 002162), tumor necrosis factor receptors Sp1. Sp2 (GenBank Acc. No. NM 003110), Sp3 (GenBank such as TNFRSF1A (GenBank Acc. No. X55313), Acc. No. AY070137), Sp4(GenBankAcc. No. NM 003112) TNFR1SFB (GenBank Acc. No. NM 001066), TNFRSF9 NFYB (GenBank Acc. No. NM 006166), Hap2 (GenBank (GenBank Acc. No. NM 001561), TNFRSF10B (GenBank Acc. No. M59079), GATA-1 (GenBank Acc. No. Acc. No. NM 003842), TNFRSF11B (GenBank Acc. No. NM 002049), GATA-2 (GenBank Acc. No. NM 002050), NM 002546), and TNFRSF13B (GenBank Acc. No. GATA-3 (GenBank Acc. No. X55122), GATA-4 (GenBank NM 006573), and interleukin receptors such as IL2RA Acc. No. L34357), GATA-5, GATA-6 (GenBank Acc. No. (GenBank Acc. No. NM 000417), IL2RG (GenBank Acc. NM 005257), FOG2 (NM 012082), Eryf1 (GenBank Acc. No. NM 000206), IL4R (GenBank Acc. No. AF421855), No. X17254), TRPS1 (GenBank Acc. No. NM 014112), IL7R (GenBank Acc. No. NM 002185), IL9R (GenBank NF-E2 (GenBank Acc. No. NM 006163), NF-E3, NF-E4, Acc. No. XM 015989), and IL13R (GenBank Acc. No. TFCP2 (GenBank Acc. No. NM 005653), Oct-1 (GenBank X95302). Preferably, the Tf-ligand-binding protein fusion of Acc. No. X13403), homeobox proteins such as HOXB2 the present invention displays the biological activity of the (GenBank Acc. No. NM 002145), HOX2H (GenBank Acc. ligand-binding protein. No. X16665), hairless homolog (GenBank Acc. No. 0193 The term “cancer-associated proteins’ is used NM 005144), mothers against decapentaplegic proteins herein to describe proteins or polypeptides whose expression such as MADH1 (GenBank Acc. No. NM 005900), is associated with cancer or the maintenance of controlled cell MADH2 (GenBank Acc. No. NM 005901), MADH3 (Gen growth, Such as proteins encoded by tumor Suppressor genes Bank Acc. No. NM 005902), MADH4 (GenBank Acc. No. or oncogenes. Cancer-associated proteins may include p16 NM 005359), MADH5 (GenBank Acc. No. AF009678), (GenBank Acc. No. AH005371), p53 (GenBank Acc. No. MADH6 (GenBank Acc. No. NM 005585), MADH7 (Gen NM 000546), p63 (GenBank Acc. No. NM 003722), p73 Bank Acc. No. NM 005.904), MADH9 (GenBank Acc. No. (GenBank Acc. No. NM 05427), BRCA1 (GenBank Acc. NM 005905), and signal transducer and activator of tran No. U14680), BRCA2 (GenBank Acc. No. NM 000059), scription proteins such as STAT1 (GenBank Acc. No. CTBP interacting protein (GenBank Acc. No. U72066), XM 010893), STAT2 (GenBank Acc. No. NM 005419), DMBT1 (GenBank Acc. No. NM 004.406), HRAS (Gen STAT3 (GenBank Acc. No. AJ012463), STAT4 (GenBank Bank Acc. No. NM 005343), NCYM (GenBank Acc. No. Acc. No. NM 003151), STAT5 (GenBank Acc. No. NM 006316), FGR (GenBankAcc. No. NM 005248), myb L41142), and STAT6 (GenBank Acc. No. NM 003153). (GenBank Acc. No. AF104863), raf1 (GenBank Acc. No. 0191 In yet another embodiment of the invention, the NM 002880), erbB2 (GenBank Acc. No. NM 004448), therapeutic molecule is a non-human or non-mammalian pro VAV (GenBank Acc. No. X16316), c-fos (V GenBank Acc. tein. For example, HIV gp120, HIV Tat, surface proteins of No. 01512), c-fes (GenBank Acc. No. X52192), c-jun (Gen other viruses such as hepatitis, herpes, influenza, adenovirus Bank Acc. No. NM 002228), MAS1 (GenBank Acc. No. and RSV, other HIV components, parasitic surface proteins M13150), pim-1 (GenBank Acc. No. M16750), TIF1 (Gen Such as malarial antigens, and bacterial Surface proteins are Bank Acc. No. NM 003852), c-fms (GenBank Acc. No. preferred. These non-human proteins may be used, for X03663), EGFR (GenBank Acc. No. NM 005228), erbA example, as antigens, or because they have useful activities. (GenBank Acc. No. X04707), c-Src tyrosine kinase (Gen For example, the therapeutic molecule may be streptokinase, Bank Acc. No. XM 044659), c-abl (GenBank Acc. No. staphylokinase, asparaginase, or other proteins with useful M14752), N-ras (GenBank Acc. No. X02751), K-ras (Gen enzymatic activities. Bank Acc. No. M54968), jun-B (GenBank Ace. No. 0.192 In an alternative embodiment, the therapeutic mol M29039), c-myc (GenBank Acc. No. AH001511), RB1 ecule is a ligand-binding protein with biological activity. (GenBank Acc. No. M28419), DCC (GenBank Acc. No. Such ligand-binding proteins may, for example, (1) block X76132), APC (GenBank Acc. No. NM 000038), NF1 receptor-ligand interactions at the cell Surface; or (2) neutral (GenBank Acc. No. M89914), NF2 (GenBank Acc. No. ize the biological activity of a molecule in the fluid phase of Y18000), and bcl-2 (GenBank Acc. No. M13994). the blood, thereby preventing it from reaching its cellular 0194 “Fusogenic inhibitor peptides’ is used herein to target. In some embodiments, the modified transferrin fusion describe peptides that show antiviral activity, anti-membrane proteins include a modified transferrin molecule fused to a fusion capability, and/or an ability to modulate intracellular ligand-binding domain of a receptor selected from the group processes, for instance, those involving coiled-coil peptide consisting of, but not limited to, a low density lipoprotein structures. Antiviral activity includes, but is not limited to, the US 2011/009 1543 A1 Apr. 21, 2011 20 inhibition of HIV-1, HIV-2, RSV, SIV, EBV, measles virus, box elder, Sycamore, maple, elm, etc., dust mites, bee venom, influenza virus, or CMV transmission to uninfected cells. food allergens, animal dander, and other insect venoms. Additionally, the antifusogenic capability, antiviral activity or 0.198. Other therapeutic molecules include microbial vac intracellular modulatory activity of the peptides merely cines which include viral, bacterial and protozoal vaccines requires the presence of the peptides and specifically does not and their various components such as Surface antigens. These require the stimulation of a host immune response directed include vaccines which contain glycoproteins, proteins or against Such peptides. Antifusogenic refers to a peptide's peptides derived from these proteins. Such vaccines are pre ability to inhibit or reduce the level of membrane fusion pared from Staphylococcus aureus, Streptococcus pyogenes, events between two or more moieties relative to the level of Streptococcus pneumoniae, Neisseria meningitidis, Neis membrane fusion which occurs between said moieties in the seria gonorrhoeae, Salmonella spp., Shigella spp., Escheri chia coli, Klebsiella spp., Proteus spp., Vibrio cholerae, absence of the peptide. The moieties may be, for example, cell Campylobacter pylori, Pseudomonas aeruginosa, Haemo membranes or viral structures, such as viral envelopes or pili. philus influenzae, Bordetella pertussis, Mycobacterium The term “antiviral peptide', as used herein, refers to the tuberculosis, Legionella pneumophila, Treponema pallidum, peptide's ability to inhibit viral infection of cells or some viral chlamydia, tetanus toxoid, diphtheria toxoid, influenza activity required for productive viral infection and/or viral viruses, adenoviruses, paramyxoviruses (mumps, measles), pathogenesis, via, for example, cell-cell fusion or free virus rubella viruses, polio viruses, hepatitis viruses, herpes infection. Such infection may involve membrane fusion, as viruses, rabies virus, HIV-1, HIV-2, RSV and papilloma occurs in the case of enveloped viruses, or some other fusion viruses. event involving a viral structure and a cellular structure. Fuso 0199 Preferred fusion molecules may contain anti-HIV genic inhibitor peptides and antiviral peptides often have viral peptides, anti-RSV peptides, human growth hormone, a amino acid sequences that are derived from greater than one and/or J3 interferons, erythropoietin (EPO), EPO like pep viral protein (e.g., an HIV-1, HIV-2. RSV, and SIV-derived tides, granulocyte-colony stimulating factor (GCSF), granu polypeptide). locyte-macrophage colony-stimulating factor (GMCSF), 0.195 Examples of fusogenic inhibitor peptides and anti insulin, insulin-like growth factor (IGF), thrombopoeitin, viral peptides can be found in WO 94/2820, WO 96/19495, peptides corresponding to the CDR of an antibody, Islet Neo WO 96/40191, WO 01/64013 and U.S. Pat. Nos. 6,333,395, genesis Associated Protein (INGAP), calcitonin, angiostatin, 6,258,782, 6,228,983, 6,133,418, 6,093,794, 6,068,973, endostatin, interleukin-2, growth hormone releasing factor, 6,060,065, 6,054,265, 6,020.459, 6,017,536, 6,013,263, human parathyroid hormone, anti-tumor necrosis factor 5,464,933, 5,346,989, 5,603,933, 5,656,480, 5,759,517, (TNF) peptides, interleukin-1 (IL-1) receptor and/or single 6,245,737; 6,326,004, and 6,348,568; all of which are herein chain antibodies. incorporated by reference. In a preferred embodiment, anti 0200 Fusion proteins of the invention may also be pre fusogenic peptides are selected from the group consisting of pared to include peptides or polypeptides derived from pep HIV T-20 (FWNWLSAWKDLELLEQENKEQQNQSEE tide libraries to screen for molecules with new or novel func ILSHILSTY, SEQ ID NO: 4), HIV T-1249, RSV T786 tions. Such peptide libraries may include those commercially (VYPSDEYDASISQVNEEINQALAYIRKADELLENV, or publicly available, e.g., American Peptide Co. Inc., Cell SEQ ID NO. 5), RSV T1584 (AVSKVLHLEGEVNKIK Sciences Inc., Invitrogen Corporation, Phoenix Pharmaceu SALLSTNKAVVSLSNGVSVLTSKVLDLKNYIDKQL, ticals Inc., United States Biological, as well as those produced SEQ ID NO: 6) and RSV T112 (VFPSDEFDA by available technologies, e.g., bacteriophage and bacterial SISQVNEKINQSLAFIRESDELLHNV, SEQ ID NO: 7). display libraries made using standard procedures. 0196. Examples of other types of peptides, include frag 0201 In yet other embodiments of the invention, Tffusion ments of therapeutic proteins as described herein, in particu proteins may be prepared by using therapeutic protein moi lar, fragments of human proteins that retain at least one activ eties as known in the art and exemplified by the peptides and ity of the parent molecule. Peptides that may be used to proteins currently approved by the Food and Drug Adminis produce modified Tffusion proteins of the invention also tration at (www.fda.gov/cber/efoi/approve.htm) as well as include mimetic peptides and peptides that exhibit a biologi PCT Patent Publication Nos. WO 01/79258, WO 01/77137, cal activity of a therapeutic protein but differ in sequence or WO 01/79442, WO 01/79443, WO 01/794.44 and WO three-dimensional structure from a full-length therapeutic 01/79480, all of which are herein incorporated by reference in protein. As a non-limited example, peptides include erythro their entirety. poietin mimetic peptides disclosed by Johnson et al. (2000) (0202 Table 1 (adapted from PCT International Publica Nephrol. Dial. Transplant 15(9): 1274-7, Kuai et al. (2000).J. tion No. WO 01/79444) provides a non-exhaustive list of Pept. Res. 56(2):59-62, Barbone et al. (1999) Nephrol. Dial. therapeutic proteins that correspond to a therapeutic protein Transplant. 14 Supp 2:80-4, Middleton et al. (1999).J. Biol. portion of a modified transferrin fusion protein of the inven Chem. 274(20): 14163-9, Johnson et al. (1998) Biochemistry tion. The “Therapeutic Protein X column discloses thera 37(10:3699-710, Johnson et al. (1997) Chem. Biol. 12:939 peutic protein molecules followed by parentheses containing 50, Wrighton et al. (1997) Nat. Biotechnol. 15(12): 1261-5, Scientific and brand names that comprise or alternatively con Livnah et al. (1996) Science 273:464-71, and Wrighton et al., sist of that therapeutic protein molecule or a fragment or (1996) Science 273:458-64. variant thereof. “Therapeutic protein X’ as used herein may 0197) Therapeutic molecules also include allergenic pro refer either to an individual therapeutic protein molecule (as teins and digested fragments thereof. These include pollen defined by the amino acid sequence obtainable from the CAS allergens from ragweed, rye, June grass, orchard grass, Sweet and Genbank accession numbers), or to the entire group of Vernal grass, red top grass, timothy grass, yellow dock, wheat, therapeutic proteins associated with a given therapeutic pro corn, Sagebrush, blue grass, California annual grass, pig tein molecule disclosed in this column. The Exemplary Iden weed, Bermuda grass, Russian thistle, mountain cedar, oak, tifier column provides Chemical Abstracts Services (CAS) US 2011/009 1543 A1 Apr. 21, 2011

Registry Numbers (published by the American Chemical in their entirety. The PCT/Patent Reference column provides Society) and/or Genbank Accession Numbers (e.g., Locus U.S. Patent numbers, or PCT International Publication Num ID, NP-XXXXX (Reference Sequence Protein), and bers corresponding to patents and/or published patent-appli cations that describe the therapeutic protein molecule all of XP-XXXXX (Model Protein) identifiers available through which are herein incorporated by reference in their entirety. the National Center for Biotechnology Information (NCBI) The Biological Activity column describes biological activi webpage (www.ncbi.nlm.nih.gov) that correspond to entries ties associated with the therapeutic protein molecule. The in the CAS Registry or Genbank database which contain an Exemplary Activity Assay column provides references that amino acid sequence of the protein molecule or of a fragment describe assays which may be used to test the therapeutic or variant of the therapeutic protein molecule. In addition and/or biological activity of a therapeutic protein or a trans GenSeq Accession numbers and/or journal publication cita ferrinfusion protein of the invention comprisingatherapeutic tions are given to identify the exemplary amino acid sequence protein X portion. These references are also herein incorpo for some polypeptides. rated by reference in their entirety. “The Preferred Indication 0203 The summary pages associated with each of these Y’ column describes disease, disorders, and/or conditions CAS and Genbank and GenSeq Accession Numbers as well that may be treated, prevented, diagnosed, or ameliorated by as the cited journal publications are available (e.g., PubMed therapeutic protein X or a transferrin fusion protein of the ID number (PMID)) and are herein incorporated by reference invention comprising a therapeutic protein X portion. US 2011/009 1543 A1 Apr. 21, 2011 22

IHT8IVIL US 2011/009 1543 A1 Apr. 21, 2011 23

US 2011/009 1543 A1 Apr. 21, 2011 25 US 2011/009 1543 A1 Apr. 21, 2011 26 US 2011/009 1543 A1 Apr. 21, 2011 27 US 2011/009 1543 A1 Apr. 21, 2011 28 US 2011/009 1543 A1 Apr. 21, 2011 29 US 2011/009 1543 A1 Apr. 21, 2011 30 US 2011/009 1543 A1 Apr. 21, 2011 31 US 2011/009 1543 A1 Apr. 21, 2011 32

©Stil US 2011/009 1543 A1 Apr. 21, 2011 33 US 2011/009 1543 A1 Apr. 21, 2011 34 US 2011/009 1543 A1 Apr. 21, 2011 35

- - - US 2011/009 1543 A1 Apr. 21, 2011 36

OXO:CHOXO

US 2011/009 1543 A1 Apr. 21, 2011 38 US 2011/009 1543 A1 Apr. 21, 2011 39 US 2011/009 1543 A1 Apr. 21, 2011 40 US 2011/009 1543 A1 Apr. 21, 2011 41 US 2011/009 1543 A1 Apr. 21, 2011 42 US 2011/009 1543 A1 Apr. 21, 2011 43 US 2011/009 1543 A1 Apr. 21, 2011 44 US 2011/009 1543 A1 Apr. 21, 2011 45 US 2011/009 1543 A1 Apr. 21, 2011 46 US 2011/009 1543 A1 Apr. 21, 2011 47 US 2011/009 1543 A1 Apr. 21, 2011 48 US 2011/009 1543 A1 Apr. 21, 2011 49 US 2011/009 1543 A1 Apr. 21, 2011 50 US 2011/009 1543 A1 Apr. 21, 2011 51 US 2011/009 1543 A1 Apr. 21, 2011 52 US 2011/009 1543 A1 Apr. 21, 2011 53 US 2011/009 1543 A1 Apr. 21, 2011 54 US 2011/009 1543 A1 Apr. 21, 2011 55 US 2011/009 1543 A1 Apr. 21, 2011 56 US 2011/009 1543 A1 Apr. 21, 2011 57 US 2011/009 1543 A1 Apr. 21, 2011 58 US 2011/009 1543 A1 Apr. 21, 2011 59 US 2011/009 1543 A1 Apr. 21, 2011 60 US 2011/009 1543 A1 Apr. 21, 2011 61 US 2011/009 1543 A1 Apr. 21, 2011 62 US 2011/009 1543 A1 Apr. 21, 2011 63 US 2011/009 1543 A1 Apr. 21, 2011 64 US 2011/009 1543 A1 Apr. 21, 2011 65 US 2011/009 1543 A1 Apr. 21, 2011 66 US 2011/009 1543 A1 Apr. 21, 2011 67 US 2011/009 1543 A1 Apr. 21, 2011 68 US 2011/009 1543 A1 Apr. 21, 2011 69 US 2011/009 1543 A1 Apr. 21, 2011 70

US 2011/009 1543 A1 Apr. 21, 2011 72

"ISEJ "ISEJ TÁZOTO US 2011/009 1543 A1 Apr. 21, 2011 73 US 2011/009 1543 A1 Apr. 21, 2011 74 US 2011/009 1543 A1 Apr. 21, 2011 75 US 2011/009 1543 A1 Apr. 21, 2011 76 US 2011/009 1543 A1 Apr. 21, 2011 77 US 2011/009 1543 A1 Apr. 21, 2011 78 US 2011/009 1543 A1 Apr. 21, 2011 79 US 2011/009 1543 A1 Apr. 21, 2011 80 US 2011/009 1543 A1 Apr. 21, 2011 81 US 2011/009 1543 A1 Apr. 21, 2011 82 US 2011/009 1543 A1 Apr. 21, 2011 83 US 2011/009 1543 A1 Apr. 21, 2011 84 US 2011/009 1543 A1 Apr. 21, 2011 85 US 2011/009 1543 A1 Apr. 21, 2011 86 US 2011/009 1543 A1 Apr. 21, 2011 87 US 2011/009 1543 A1 Apr. 21, 2011 88 US 2011/009 1543 A1 Apr. 21, 2011 89 US 2011/009 1543 A1 Apr. 21, 2011 90 US 2011/009 1543 A1 Apr. 21, 2011 91 US 2011/009 1543 A1 Apr. 21, 2011 92

pureºunuILIII US 2011/009 1543 A1 Apr. 21, 2011 93