NZ238583A

NZ238583A - Targetted cytotoxic anticancer conjugates and

Info

Publication number: NZ238583A
Application number: NZ23858391A
Authority: NZ
Inventors: Janet Ahern; David C Heimbrook; Allen I Oliff; Steven M Stirdivant
Original assignee: Merck & Co Inc
Priority date: 1991-06-18
Filing date: 1991-06-18
Publication date: 1994-02-25

Description

<div class="application article clearfix" id="description"> 238583 Priority Druojo): . .I0r.3.'. L Complile Specification Filed: JXk.ff-r-.1 Class: .CsCl [Cr-l Si I .. AkiMsftlm: 25 FEB TO Publication Data: ....7. P.O. Jcurna!, No: ... )3TX? NEW ZEALAND PATENTS ACT, 1953 No.: Date: COMPLETE SPECIFICATION METHOD OF TREATING BLADDER CANCER CELLS We, MERCK & CO., INC., a corporation duly organized and existing under the laws of the State of New Jersey, USA, of 126 East Lincoln Avenue, Rahway, New Jersey, USA hereby declare the invention for which we pray that a patent may be granted to us, and the method by which it is to be performed, to be particularly described in and by the following statement N.z. PATWWT OFPttiE '4.18 JUN1991 *€CEiy» - l - (followed by page la) 9584P/2002a 238583 10 i 18146Y 15 20 BACKGROUND OF THE INVENTION Traditional cancer chemotherapy relies on 25 the ability of drugs to kill tumor cells in cancer patients. Unfortunately, these same drugs frequently kill normal cells as well as the tumor cells. The extent to which a cancer drug kills tumor cells rather than normal cells is an indication of the compound's 30 degree of selectivity for tumor cells. One method of 23 8 5 9584P/2002a - 2 - 18146IA increasing the tumor cell selectivity of cancer drugs is to deliver drugs preferentially to the tumor cells while avoiding normal cell populations. Another term for the selective delivery of chemotherapeutic agents 5 to specific cell populations is "targeting". Drug targeting to tumor cells can be accomplished in several ways. One method relies on the presence of specific receptor molecules found on the surface of tumor cells. Other molecules, referred to as 10 "targeting agents", can recognize and bind to these cell surface receptors. These "targeting agents" include, e.g., antibodies, growth factors, or hormones. "Targeting agents" which recognize and bind to specific cell surface receptors are said 15 to target the cells which possess those receptors. For example, bladder tumor cells possess a protein on their surfaces called the epidermal growth factor receptor. Transforming growth factor-alpha (TGF-alpha) recognizes and binds to the EGF receptor on 20 bladder tumor cells. TGF-alpha is therefore, a "targeting agent" for these tumor cells. "Targeting agents" by themselves do not kill tumor cells. Other molecules including cellular 25 poisons or toxins can be linked to "targeting agents" to create hybrid molecules that possess both tumor cell targeting and cellular toxin domains. These hybrid molecules function as tumor cell selective poisons by virtue of their abilities to target tumor 30 cells and then kill those cells via their toxin component. Some of the most potent cellular poisons used in constructing these hybrid molecules are 238583 9584P/2002a - 3 - 18146IA bacterial toxins that inhibit protein synthesis in mammalian cells. Pseudomonas exotoxin A is one of these bacterial toxins., and has been used to construct hybrid "targeting - toxin" molecules (U.S. Patent 5 4,545,985). Pseudomonas exotoxin A intoxicates mammalian cells by first binding to the cell's surface, then entering the cell cytoplasm and inactivating elonga-10 tion factor 2 which is a cellular protein required for protein synthesis. Pseudomonas exotoxin A has been used to construct anticancer hybrid molecules using monoclonal antibodies and protein hormones. However, one problem with these hybrid molecules 15 is that they exhibit toxicity towards normal cells. At least part of the toxicity associated with hybrid molecules containing pseudomonas exotoxin A is due to the ability of pseudomonas exotoxin A by itself to bind to and enter many types of mammalian cells. 20 Therefore, hybrid molecules formed between pseudomonas exotoxin A and specific "targeting agents" can bind to many normal cells in addition to the cells recognized by the "targeting agent". One method of dealing with this problem is to modify pseudomonas 25 exotoxin A so that it is no longer capable of binding to normal cells. This can be accomplished by removing that portion of the pseudomonas exotoxin A molecule which is responsible for its cellular binding activity. A truncated form of the pseudomonas exotoxin A 30 molecule has been prepared which retains the ability to inactivate elongation factor 2 but no longer is capable of binding to mammalian cells. This modified 233583 9584P/2002a - 4 - 18146IA pseudomonas exotoxin A molecule is called pseudomonas exotoxin - 40 or PE^q (Hwang, fit &1. , Cell 48:129-136 1987). 5 PE40 kas ^een lin^ed to several targeting molecules including TGF-alpha (Chaudhary, fit al. , PNAS USA M:4583-4542 1987). In the case of TGF-alpha, hybrid molecules containing PE^q and TGF-alpha domains are capable of specifically binding to tumor 10 cells that possess EGF receptors and intoxicating these cells via inhibiting protein synthesis. In order for this hybrid molecule to efficiently bind to the EGF receptor it must assume the proper conformation. Efficient receptor binding is also 15 dependent on having the "targeting domain" properly exposed so that it is accessible for binding. When TGF-alpha and PE^q hybrid molecules are produced as fusion proteins in bacteria using recombinant DNA techniques the majority of hybrid molecules exhibit 20 poor EGF receptor binding activity. DISCLOSURE STATEMENT 25 1. U.S. patent 4.545.985 teaches that pseudomonas exotoxin A can be chemically conjugated to an antibody or to epidermal growth factor. While this patent further teaches that these conjugates can be used to kill human tumor cells, these chemically 30 linked toxins have been shown to have undesirable, nonspecific levels of activity. 2385 8 3 9584P/2002a - 5 - 18146IA 2. U.S. patent 4.664.911 teaches that antibodies can be conjugated to the A chain or the B chain of ricin which is a toxin obtained from plants. Patent 4,664,911 further teaches that these conjugates 5 can be used to kill human tumor cells. 3. U.S. patent 4.675.382 teaches that hormones such as melanocyte stimulating hormone (MSH) can be linked to a portion of the diphtheria toxin 10 protein via peptide bonds. Patent 4,675,382 further teaches that the genes which encode these proteins can be joined together to direct the synthesis of a hybrid fusion protein using recombinant DNA techniques. This fusion protein has the ability to bind 15 to cells that possess MSH receptors. 4. Murphv. et al. . PNAS USA £3.: 8258-8262 1986, Genetic construction, expression, and melanoma-selective cytotoxicity of a diphtheria toxin-related 20 alpha-melanocyte-stimulating hormone fusion protein. This article teaches that a hybrid fusion protein produced in bacteria using recombinant DNA technology and consisting of a portion of the diphtheria toxin protein joined to alpha-melanocyte- 25 stimulating hormone will bind to and kill human melanoma cells. 5. Allured, et al.. PNAS USA 83:1320-132* 1986, Structure of exotoxin A of Pseudomonas 30 aeruginosa at 3.0 Angstrom. This article teaches the three dimensional structure of the pseudomonas exotoxin A protein. 9584P/2002a - 6 - 18146IA 6. Hwang. et al.. Cell 48:129-136 1987, Functional Domains of Pseudomonas Exotoxin Identified by Deletion Analysis of the Gene Expressed in E. Coli This article teaches that the pseudomonas exotoxin A protein can be divided into three distinct functional domains responsible for: binding to mammalian cells, translocating the toxin protein across lysosomal membranes, and ADP ribosylating elongation factor 2 inside mammalian cells. This article further teaches that these functional domains correspond to distinct regions of the pseudomonas exotoxin A protein. 7. Chaudharv. et al.. PNAS USA £4:4538-4542 1987, Activity of a recombinant fusion protein between transforming growth factor type alpha and Pseudomonas toxin. This article teaches that hybrid fusion proteins formed between PE-40 and transforming growth factor-alpha and produced in bacteria using recombinant DNA techniques will bind to and kill human tumor cells possessing epidermal growth factor receptors. 8. European patent application 0 261 671. published 30 March 1988, teaches that a portion of the pseudomonas exotoxin A protein can be produced which lacks the cellular binding function of the whole pseudomonas exotoxin A protein but possesses the translocating and ADP ribosylating functions of the whole pseudomonas exotoxin A protein. The portion of the pseudomonas exotoxin A protein that retains the translocating and ADP ribosylating functions of the whole pseudomonas exotoxin A protein is 23 8 5 9584P/2002a - 7 - 18146IA called pseudomonas exotoxin - 40 or PE-40. PE-40 consists of amino acid residues 252-613 of the whole pseudomonas exotoxin A protein as defined in Gray, et al., PNAS USA 81:2645-2649 1984. This patent application further teaches that PE-40 can be linked to transforming growth factor-alpha to form a hybrid fusion protein produced in bacteria using recombinant DNA techniques. 9. Kellev. et al.. PNAS USA 3980-3984 1988, Interleukin 2-diphtheria toxin fusion protein can abolish cell-mediated immunity in vivo. This article teaches that a hybrid fusion protein produced in bacteria using recombinant DNA technology and consisting of a portion of the diphtheria toxin protein joined to interleukin 2 functions in mice to suppress cell mediated immunity. 10. Bailon. Biotechnology, pp. 1326-1329 Nov. 1988. Purification and Partial Characterization of an Interleukin 2-Pseudomonas Exotoxin Fusion Protein. This article teaches that hybrid fusion proteins formed between PE-40 and interleukin 2 and produced in bacteria using recombinant DNA techniques will bind to and kill human cell lines possessing interleukin 2 receptors. 11. Edwards. et al.. Mol. Cell. Biol. 9: 2860-2867 1989 describe the preparation of the modified TGF-alpha - PE^q hybrid molecules that have been found to have utility in treating bladder tumor cells. 8 23 8583 12. Heimbrook. et al.. Proc. Natl. Acad. Sci. USA 87: 4697-4701 1990 describe the in vivo efficacy of modified TGF-alpha - PE^q in significantly prolonging the survival of mice containing human tumor cell xenografts. OBJECTS OF THE INVENTION It is an object of the present invention to provide a method for selectively killing bladder tumor cells. A further object is to provide a hybrid molecule of enhanced potency formed between TGF-alpha and modified PE^q molecules. Another object of the present invention is to provide pharmaceutical compositions containing as active ingredient a hybrid molecule containing a PE4Q domain (or region) wherein the PE^q domain has been modified to improve binding of the hybrid protein to the epidermal growth factor receptor on bladder tumor cells. These and other objects of the present invention will be apparent from the following description. \ "2 AUGI993 ' ' - 9 - 23 85 8 3 SUMMARY OF INVENTION Our New Zealand Patent Specification No. 232564 provides a hybrid molecule comprising a modified PE40 domain bonded to a protein targeting domain. The modified PE40 domain improves the receptor binding activity of this hybrid molecule. Substitution of other neutral amino acids such as, e.g. alanine, for the cysteine residues in PE40, or deletion of cysteine residues, improves binding of the hybrid molecule to the receptors recognized by the targeting domain. The hybrid molecules bind more efficiently to targeted receptors on human tumor cells than hybrid molecules having unmodified PE40- The present invention provides that hybrid molecules comprising a modified PE40 domain bonded to a TGF-alpha targeting domain have utility in killing bladder tumor cells. DETAILED DESCRIPTION OF THE INVENTION Hybrid molecules formed between TGF-alpha and PE4.Q are characterized in three primary assay systems, these assays include: 1 - ADP ribosylation of elongation factor 2 which measures the enzymatic activity of TGF-alpha - PE^q that inhibits mammalian cell protein synthesis, 2 - inhibition of radiolabeled EGF binding to the EGF receptor on membrane vesicles from A431 cells which measures the EGF receptor binding activity of TGF-alpha - PE4f). and 3 - cell proliferation as assessed by conversion of 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetra- zolium bromide (MTT) to formazan which is used to measure the survival' of tumor cells following //^^h A °*\ fv *2. * ri ^ "2 AUGJ993 ~} 238583 9584P/2002a - 10 - 18146IA exposure to TGF-alpha - PE^q. These assays are performed as previously described (Dominic, £t al. . Infection and Immunity lf>: 832-841 1977, Cohen, gi ai-, J- Biol. Chem. 257:1523-1531 1982, Riemen, 5 £l al-, Peptides 8:877-885 1987, Mosmann J. Immunol. Methods £5.:55-63 1983). To create new TGF-alpha - PE^g hybrid molecules with superior receptor binding character-10 istics we first produced a series of recombinant DNA molecules that encoded either TGF-alpha - PE^q or specifically modified versions of TGF-alpha - PE^q. The original or parental TGF-alpha - PE^q gene was molecularly cloned in a bacterial TAC expression 15 plasmid vector (pTAC TGF57-PE40) using distinct segments of cloned DNA as described in Example 1. The pTAC TGF57-PE40 DNA clone was used as the starting reagent for constructing specifically modified versions of TGF-alpha - PE^q DNA. The 20 specific modifications of the pTAC TGF57-PE40 DNA involve site specific mutations in the DNA coding sequence required to replace two or four of the cysteine codons within the PE^q domain of the pTAC TGF57-PE40 DNA with codons for other amino acids. 25 Alternatively, the site specific mutations can be engineered to delete two or four of the cysteine codons within the PE40 domain of pTAC TGF57-PE40. The site specific mutations in the pTAC TGF57-PE*n DNA were constructed using the methods of Winter, 30 gi al., Nature 299:756-758 1982. Specific examples of the mutated pTAC TGF57-PE40 DNAs are presented in Example 3. The amino acid sequence of the hybrid ^ J 0 J 9584P/2002a - 11 - 18146IA protein encoded by the pTAC TFG57-PE40 DNA is presented in Table 3. The four cysteine residues in the PE^q domain of the parental TGF-alpha -PE,Q hybrid protein are designated residues 5 Cys265, Cys287, Cys372, and Cys379 (Table 3). Amino aeid residues in the PE^q domain are numbered as defined in Gray, al, PNAS USA £1: 2645-2649 (1984). The modified TGF-alpha - PE40 hybrid proteins generated from the specifically 10 mutated pTAC TGF57-PE40 DNA contain substitutions or deletions of the two N-terminal PE^q residues [Cys265 and Cys287] or the two C-terminal residues [Cys372 and Cys379], or both [Cys2^, Cys287, Cys372, and Cys379]. To simplify the 15 nomenclature for describing the modified hybrid proteins produced from these mutated pTAC TGF57-PE40 DNAs we have designated the amino acid residues at the N-terminal positions the "A" locus and the residues at the C-terminal positions the "B" locus. When 20 cysteine residues are present at the two N-terminal P£40 positions as in parental TGF-alpha - PE40 hybrid molecule, the locus is capitalized (i.e. "A"). When these cysteines are substituted with other neutral amino acids such as, for example, glycine, 25 alanine, phenylalanine, valine, leucine, isoleucine, tyrosine, histidine, tryptophan, serine, threonine or methionine, or deleted from the N-terminal positions, the locus is represented by a lower case "a". Similarly, if the amino acid res-idues at the two 30 C-terminal positions are cysteines the locus is represented by an upper case "B" while a lower case "bM represents this locus when the amino acid 23 8583 residues at these positions are substituted with other amino acids or deleted. Thus, when only the two cysteine residues at positions 265 and 287 in the PE40 domain of TGF-alpha -PE40 are substituted with other amino acids or deleted, the modified hybrid proteins are designated TGF-alpha -PE40 aB. When only the two cysteine residues at positions 372 and 379 in the PE40 domain are substituted with other amino acids or deleted, the modified hybrid proteins are designated TGF-alpha - PE40 Ab. When all four cysteine residues in the PE40 domain are substituted with other amino acids or deleted, the modified hybrid proteins are designated TGF-alpha - PE40 ab. In a similar fashion the parental TGF-alpha -PE40 hybrid protein with cysteines at amino acid residue positions 265, 287, 372 and 379 can be designated TGF^Vr^v alpha - PE40 AB. .6-' f, i\ Both the TGF-alpha - PE4Q AB hybrid ' protein and the modified TGF-alpha - PE40 hybrid \ <jJ proteins are produced in E. coli using the TAC ex- E pression vector system described by Linemeyer, et- al.. Bio-Technology 960-965 1987. The recombinant hybr id . proteins produced in these bacteria are harvested and purified by lysing the bacteria in guanidine hydrochloride followed by the addition of sodium sulphite and sodium tetrathionate. This reaction mixture is subsequently dialyzed and urea is added to solubilize proteins that have precipitated out of solution. The mixture is next centrifuged to remove insoluble proteins and the recombinant hybrid TGF-alpha - PE^q proteins are separated using ion exchange chromatography followed by size exclusion chromatography, followed once again by ion exchange chromatography. The purified TGF-alpha - PE4Q hybrid proteins are next exposed to reducing agents such as beta-mercaptoethanol in order to permit disulfide bonds to form within the hybrid protein between pairs of cysteine residues. Finally, the refolded hybrid proteins are subjected to size exclusion and ion 2 3 8 5 8 3 9584P/2002a - 13 - 18146IA exchange chromatography to isolate highly pure TGF-alpha - PE^q protein. The precise details of this purification scheme are described in Example 4. Once purified and refolded the biologic activity of 5 these hybrid proteins can be characterized using the ADP ribosylation, EGF receptor binding, and cell proliferation assays described above. Alternatively, and preferably, the hybrid 10 proteins TGF-alpha - PE^q AB, TGF-alpha - PE^q Ab, TGF-alpha - PE^q aB and TGF-alpha - PE^q ab are produced in transformed bacteria. The bacteria are harvested and the cell paste is lysed and treated, preferably by centrifugation, to remove debris and 15 undesired proteins. The desired hybrid protein then is precipitated by addition of a sulfate salt, preferably (NH^^SO^, to the supernatant liquid. The precipitate is sulfitolyzed, refolded by addition of excess 15-mercaptoethanol, concentrated and 20 separated by ion-exchange chromatography and metal-chelating chromatography. Specific details are disclosed in Example 5. An important utility of TGF-alpha modified 25 PE^q lies in its ability to bind to and kill human bladder tumor cells. The anti-cancer proteins described herein have utility in killing bladder cancer cells and are used for this purpose in the form of a solution or suspension in a physiologically 30 acceptable liquid such as, for example, sterile water, water for injection, saline or, preferably, buffered saline or buffered saline containing a 23 85 8 - 14 - carrier protein such as, for example, human serum albumin, e.g., phosphate buffered saline or PBS containing human serum albumin. The solution or suspension contains from substantially 0.1 mg to 5 substantially 10 mg of anti-cancer hybrid protein per 60 ml of physiologically acceptable liquid. More preferably, it contains from substantially 0.5 mg to substantially 5 mg per 60 ml, and most preferably, it contains from substantially 2 mg to substantially 4 mg 10 per 60 ml of physiologically acceptable liquid. The method of treating bladder cancer cells consists in contacting the bladder cancer cells with the solution or suspension containing the anti-cancer 15 proteins described herein for a period of from less than an hour, for example, about 30 minutes, to a period of several hours, for example, up to about four hours, at ambient temperature. In the case of laboratory animals the solution or suspension is 20 administered via a trans-urethral catheter. While the use of TGF-alpha modified PE^q hybrid molecules is described herein and in the following examples, it is to be understood that the 25 scope of the present invention includes as targeting agents TGF-alpha, EGF, other members of the EGF family of peptide hormones that bind to the EGF receptor on bladder tumor cells, Shope fibroma virus growth factor, and vaccinia virus growth factor and that 30 the toxin to which the targeting agent is coupled also includes PE^q, diphtheria toxin, ricin toxin \ 23 8 5 9584P/2002a - 15 - 18146IA or other members of the ADP-ribosylating class of mammalian cell poisons. The following examples illustrate the 5 present invention without, however, limiting the same thereto. All of the enzymatic reactions required for molecular biology manipulations, unless otherwise specified, were carried out as described in Maniatis, et al. (1982) In: Molecular Cloning: A Laboratory 10 Manual, Cold Spring Harbor Press. Example 1 15 Construction of recombinant DNA clones containing TGF-alpha - PE^q DNA The TGF-alpha DNA segment was constructed using three sets of synthetic oligonucleotides as 20 described by Defeo-Jones, fit al-. Molecular and Cellular Biology £:2999-3007 1988. This synthetic TGF-alpha gene was cloned into pUC-19. DNA from the pUC-19 clone containing recombinant human TGF-alpha was digested with Sph I and Eco RI. The digestion 25 generated a 2.8 kb DNA fragment containing all of pUC-19 and the 5' portion of TGF-alpha. The 2.8 kb fragment was purified and isolated by gel electrophoresis. An Eco RI to Sph I oligonucleotide cassette was synthesized. This synthetic cassette 30 had the sequence indicated below: 5'-CGGACCTCCTGGCTGCGCATCTAGG-3« 3•-GTACGCCTGGAGGACCGACGCGTAGATCCTTAA-5' 2 3 8 5 8 3 9584P/2002a - 16 - 18146IA For convenience, this oligonucleotide cassette was named 57. Cassette 57 was annealed and ligated to the TGF-alpha containing 2.8 kb fragment forming a circularized plasmid. Clones which con-5 tained the cassette were identified by hybridization to radiolabeled cassette 57 DNA. The presence of human TGF-alpha was confirmed by DNA sequencing. Sequencing also confirmed the presence of a newly introduced Fsp I site at the 31 end of the TGF-alpha 10 sequence. This plasmid, named TGF-alpha-57/pUC-19, was digested with HinD III and Fsp I which generated a 168 bp fragment containing the TGF-alpha gene (TGF-alpha-57). A separate preparation of pUC-19 was digested with HinD III and Eco RI which generated 15 a 2.68 kb pUC-19 vector DNA. The PE4Q DNA was isolated from plasmid pVC 8 (Chaudhary, &1. , PNAS USA 84:4538-4542 1987). pVC 8 was digested using Nde I. A flush end was then generated on this DNA by using the standard conditions of the Klenow reaction 20 (Maniatis, supra, p.113). The flush-ended DNA was then subjected to a second digestion with Eco RI to generate a 1.3 kb Eco RI to Nde I (flush ended) fragment containing PE^q. The TGF-alpha-57 HinD III to Fsp I fragment (168 bp) was ligated to the 25 2.68 kb pUC-19 vector. Following overnight incuba-bion, the 1.3 kb EcoRI to Nde I (flush ended) PE^q DNA fragment was added to the ligation mixture. This second ligation was allowed to proceed overnight. The ligation reaction product was then used to trans-30 form JM 109 cells. Clones containing TGF-alpha-57 PE^q in pUC-19 were identified by hybridization to radiolabeled TGF-alpha-57 PE^q DNA and the DNA from 2 ^ 8 R R v# U Q nj 9584P/2002a - 17 - 18146IA this clone was isolated. The TGF-alpha-57 PE^g was removed from the pUC-19 vector and transferred to a TAC vector system described by Linemeyer, fii al.. Bio-Technology 5.:960-965 1987). The TGF-alpha-57 5 PE4o *n was digested with HinD III and Eco RI to generate a 1.5 kb fragment containing TGF-alpha-57 A flush end was generated on this DNA fragment using standard Klenow reaction conditions (Maniatis, £t al-. loc. cit.). The TAC vector 10 was digested with HinD III and Eco RI. A flush end was generated on the digested TAC vector DNA using standard Klenow reaction conditions (Maniatis, et al.. loc. cit. The 2.7 kb flush ended vector was isolated using gel electrophoresis. The flush ended 15 TGF-alpha-57 PE^g fragment was then ligated to the flush ended TAC vector. The plasmid generated by this ligation was used to transform JM 109 cells. Candidate clones containing TGF-alpha-57 PE^q were identified by hybridization as indicated above and 20 sequenced. The clone containing the desired construction was named pTAC TGF57-PE40. The plasmid generated by these manipulations is depicted in Table 1. The nucleotide sequence of the amino acid codons of the TGF-alpha - PE^q fusion protein encoded in 25 the pTAC TGF-57-PE40 DNA are depicted in Table 2. The amino acid sequence encoded by the TGF-57-PE40 gene is shown in Table 3. 238 9584P/2002a - 18 - 18146IA Example 2 Construction of modified versions of recombinant TGF-alpha - PE4q containing DNA clones: 5 Substitution of alanines for cysteines. TGF-alpha - PE4Q aB: The clone pTAC TGF57-PE40 was digested with 10 SphI and BamHI and the 748 bp Sphl-BamHI fragment (specifying the C-terminal 4 amino acids of TGF-alpha, 4 linker amino acids and the N-terminal 240 amino acids of PE40) was isolated. M13 mpl9 vector DNA was cut with SphI and BamHI and the vector DNA was 15 isolated. The 748 bp Sphl-BamHI TGF-alpha - PE4Q fragment was ligated into the M13 vector DNA overnight at 15°C. Bacterial host cells were transformed with this ligation mixture, candidate clones were isolated and their plasmid DNA was sequenced to 20 insure that these clones contained the proper recombinant DNAs. Single stranded DNA was prepared for mutagenesis. An oligonucleotide (oligo #132) was synthesized and used in site directed mutagenesis 25 to introduce a Hpal site into the TGF-alpha - PE4q DNA at amino acid position 272 of PE4q: 5' CTGGAGACGTTAACCCGTC 3' (oligo #132) 30 One consequence of this site directed mutagenesis was the conversion of residue number 272 in PE4q from phenylalanine to leucine. The 23 8 5 9584P/2002a - 19 - 18146IA mutagenesis was performed as described by Winter, fit &1., Nature, 111:756-758 1982. A candidate clone containing the newly 5 created Hpal site was isolated and sequenced to validate the presence of the mutated genetic sequence. This clone was then cut with SphI and Sail. A 198 bp fragment specifying the C-terminal 5 amino acids of TGF-alpha and the N-terminal 61 amino acids of PE^q 10 and containing the newly introduced Hpal site was isolated and subcloned back into the parent pTAC TGF57-PE40 plasmid at the Sphl-Sall sites. Bacterial host cells were transformed, a candidate clone was isolated and its plasmid DNA was sequenced to insure 15 that this clone contained the proper recombinant DNA. For convenience this clone was named pTAC TGF57-PE40-132. pTAC TGF57-PE40-132 was digested with SphI and Hpal and a 3.96 Kb DNA fragment was isolated. A synthetic oligonucleotide cassette (oligo #153) 20 spanning the C-terminal 5 amino acids of TGF-alpha and the N-terminal 32 amino acids of PE^q and containing SphI and Hpal compatible ends was synthesized and ligated to the digested pTAC TGF57-PE40-132: 25 5' CGGACCTCCTGGCCATGGCCGAAGAGGGCGGCAGCCTGGCCGCGCTGACCGCGCA 3' GTACGCCTGGAGGACCGGTACCGGCTTCTCCCGCCGTCGGACCGGCGCGACTGGCGCGT CCAGGCTGCACACCTGCCGCTGGAGACGTT 3' 30 GGTCCGACGTGTGGACGGCGACCTCTGCAA 5' {oligo #153) 238583 9584P/2002a - 20 - 18146IA This oligonucleotide cassette incorporated a change in the TGF-alpha - PE^q DNA so that the codon specifying alanine at residue 51 was eliminated and the codon specifying cysteine at residue 264 of PE^q 5 now specified alanine. For convenience this plasmid DNA was called pTAC TGF57-PE40-132.153. Bacterial host cells were transformed with pTAC TGF57-PE40-132, 153 DNA. Candidate clones were identified by hybridization, isolated and their plasmid DNA was sequenced 10 to insure that it contained the proper recombinant DNA. pTAC TGF57-PE40-132,153 DNA was digested with Hpal and Sail and a 3.95 Kb vector DNA was 15 isolated. A synthetic oligonucleotide cassette (oligo #142) spanning amino acid residues 272 to 309 of PE^0 and containing Hpal and Sail compatible ends was synthesized and ligated to the 3.95 Kb pTAC TGF/PE40 132,153 DNA. 20 25 5' AACCCGTCATCGCCAGCCGCGCGGCTGGGAACAACTGGAGCAGGCTGGCTATCCGGTGC 3' TTGGGCAGTAGCGGTCGGCGCGCCGACCCTTGTTGACCTCGTCCGACCGATAGGCCACG AGCGGCTGGTCGCCCTCTACCTGGCGGCGCGGCTGTCGTGGAACCAGG 3• TCGCCGACCAGCGGGAGATGGACCGCCGCGCCGACAGCACCTTGGTCCAGCT 5' (oligo #142) This oligonucleotide cassette changed the codon specifying cysteine at residue 287 so that this codon now specified alanine. For convenience this mutated plasmid DNA was called pTAC TGF57-PE40-30 132,153,142. Bacterial host cells were transformed with this plasmid and candidate clones were identified by hybridization. These clones were isolated and 238583 10 - 21 - their plasmid DNA was sequenced to insure that it contained the proper recombinant DNA. The pTAC TGF57-PE40-132,153,142 plasmid encodes the TGF-alpha - variant with both N-terminal cysteines at locus "A" replaced by alanines. Therefore, following the nomenclature described previously this modified version of TGF-alpha - PE40 is called a TGF-alpha-PE^q aB. The amino acid sequence encoded by the TGF-alpha-PE^Q aB gene is shown in Table 4. TGF-alpha - PE40 Ab: The clone pTAC TGF57-PE40 was digested with SphI and BamHI and the 748 bp Sphl-BamHI fragment 15 (specifying the C-terminal 4 amino acids of TGF-alpha 4 linker amino acids and the N-terminal 240 amino acids of PE^q) was isolated. M13 mpl9 vector DNA was cut with SphI and BamHI and the vector DNA was isolated. The 748 bp Sphl-BamHI TGF-alpha - PE^q 20 fragment was ligated into the M13 vector DNA overnight at 15°C. Bacterial host cells were transformed with this ligation mixture, candidate clones were isolated and their plasmid DNA was sequenced to insure that these clones contained the proper recombinant DNAs. 25 Single stranded DNA was prepared for mutagenesis. An oligonucleotide (oligo #133) was /fy. 4 synthesized and used in site directed mutagenesis typ'V* Ii ^ c4% introduce a Bstell site into the TGF-alpha - FE,n t , 30 DNA at amino acid position 369 of PE 5' GACGTGGTGACCCTGAC 3« (oligo #133) \ \ 2 3 8 5 9584P/2002a - 22 - 18146IA One consequence of this mutagenesis was the conversion of the serine residue at position 369 of PE^q to a threonine. 5 A DNA clone containing the newly created Bstell site was identified, isolated and sequenced to ensure the presence of the proper recombinant DNA, This clone was next digested with Apal and Sail restriction enzymes. A 120 bp insert DNA fragment 10 containing the newly created Bstell site was isolated and ligated into pTAC TGF57-PE40 that had also been digested with Apal and Sail. Bacterial host cells were transformed, and a candidate clone was isolated and sequenced to insure that the proper recombinant 15 DNA was present. This newly created plasmid DNA was called pTAC TGF57-PE40-133. It was digested with Bstell and Apal and 2.65 Kb vector DNA fragment was isolated. 20 a Bstell to Apal oligonucleotide cassette (oligo #155) was synthesized which spanned the region of TGF-alpha - PE^0 deleted from the pTAC TGF57-PE40-133 clone digested with Bstell and Apal restriction enzymes. This cassette also specified 25 the nucleotide sequence for Bstell and Apal compatible ends. 5' GTGACCCTGACCGCGCCGGTCGCCGCCGGTGAAGCTGCGGGCC 3' GGACTGGCGCGGCCAGCGGCGGCCACTTCGACGC 5' lolioo #155) 30 This oligonucleotide cassette changed the codons for cysteines at residues 372 and 379 of - 23 - 2 3 # 5 $ j PE^q to codons specifying alanines. Oligonucleotide cassette #155 was ligated to the 2.65 Kb vector DNA fragment. Bacterial host cells were transformed and candidate clones were isolated and sequenced to insure 5 that the proper recombinant DNA was present. This newly created DNA clone was called pTAC TGF57-PE40-133,155. It encodes the TGF-alpha - PE^q variant with both cysteines at locus "B: replaced by alanines. Therefore, following the nomenclature described previ-10 ously this modified version of TGF-alpha - PE^q is called a TGF-alpha - PE40 Ab. The amino acid sequence encoded by the TGF-alpha-PE^Q Ab gene is shown in Table 5. 15 TGF-alpha - PE4Q ab: The pTAC-TGF57-PE40-132,153,142 plasmid encoding TGF-alpha - PE4Q aB was digested with Sail and Apal and the resultant 3.8 Kb vector DNA fragment 20 was isolated. The pTAC TGF57-PE40-133,155 plasfeid encoding TGF-alpha - PE^q Ab was also digested with Sail and Apal and the resultant 140 bp DNA fragment containing the cysteine to alanine changes at aminoyV "V acid residues 372 and 379 of PE^q was isolated. J? j; 25 These two DNAs were ligated together and used to ^»8 /j- transform bacterial host cells. Candidate clones were identified by hybridization with a radiolabeled ^ 140 bp DNA from pTAC TGF57-PE40-133,155 . Plasmid DNA from the candidate clones was isolated and sequenced*- """" 30 to insure the presence of the proper recombinant DNA. This newly created DNA clone was called pTAC TGF57-PE40-132,153,142,133,155. This plasmid encodes 23858J - 24 - the TGF-alpha - PE^q variant with all four cysteines at loci "A" and "B" replaced by alanines. Therefore, following the nomenclature described previously this modified version of TGF-alpha - PE^q is called a 5 TGF-alpha - PE^0 ab. The amino acid sequence encoded by the TGF-alpha-PE^Q ab gene is shown in Table 6. 10 Example 3 Construction of modified versions of recombinant TGF-alpha-PE^Q containing DNA clones: Deletion of cysteine residues 15 TGF-alpha-PE40 aB, TGF-alpha-PE4Q Ab, and TGF-alpha-PE^g ab can also be constructed by removing the cysteine residues at locus "A" and/or locus "B". Construction of these versions of 20 TGF-alpha-PE^Q are accomplished identically as described in Example 2 except that: for TGF-alpha-PE^0 aB oligonucleotide cassette 153 is changed ^ such that the alanine codon intended for position ^'::v 265 is deleted and oligonucleotide cassette 142 isjpf rwsQV » 25 changed such that the alanine codon intended for ^./ position 287 is deleted. For TGF-alpha-PE^Q Ab K ® oligonucleotide cassette 155 is changed such that the alanine codons intended for residues 372 and 37° are deleted. For TGF-alpha-PE^Q ab the DNA fragments 30 used to construct this recombinant gene are taken from the TGF-alpha-PE40 aB and TGF-alpha-PE^Q Ab gene described in this example. 23 9584P/2002a - 25 - 18146IA EXAMPLE 4 Production and isolation of recombinant TGF-alpha- PE^q fusion proteins: 5 Production of fusion protein Transformed E- coli JM-109 cells were 10 cultured in 1L shake flasks in 500 ml LB-Broth in the presence of 100 mg/ml ampicillin at 37°C. After the A600 spectrophotometric absorbance value reached 0.6, isopropyl B-D-thio-galactopyranoside was added to a final concentration of 1 mM. After 2 hours the cells 15 were harvested by centrifugation. S-Sulphonation of fusion protein 20 The cells were lysed in 8M guanidine hydrochloride, 50 mM Tris pH 8.0, 1 mM EDTA by stirring at room temperature for 2 hours. The lysis mixture was brought to 0.4 M sodium sulphite and 0.1M sodium tetrathionate by adding solid reagents and the 25 pH was adjusted to 9.0 with 1M NaOH. The reaction was allowed to proceed at room temperature for 16 hours. 30 Preparation for chromatography The protein solution was dialysed against a 10,000 fold excess volume of ImM EDTA at 4°C. The 238583 9584P/2002a - 26 - 18146IA mixture was then brought to 611 urea, 50 mM Tris pH 8.0, 50 mM NaCl at room temperature and stirred for 2 hours. Any undissolved material was removed by centrifugation at 32,000 x g for 30 minutes. DEAE F.F. Sepharose Chromatography The cleared supernatant from the previous 10 step was applied to a 26 x 40 cm DEAE Fast Flow column (Pharmacia LKB Biotechnology Inc.) equilibrated with 6M urea, 50 mM Tris pH 8.0, 50 mM NaCl at a flow rate of 1 ml/minute. The column was washed with the equilibration buffer until all unabsorbed 15 materials were removed as evidenced by a UV 280 spectrophotometric absorbance below 0.1 in the equilibration buffer as it exits the column. The adsorbed fusion protein was eluted from the column with a 1000 ml 50-350 mM NaCl gradient and then 20 concentrated in a stirred cell Amicon concentrator fitted with a YM-30 membrane. 25 Sephacrvl S-300 The concentrated fusion protein (8 mis) was applied to a 2.6 x 100 cm Sephacryl S-300 column (Pharmacia LKB Biotechnology Inc.) equilibrated with 6M urea, 50 mM Tris pH 8.0, 50 mM NaCl at a flow rate 30 of 0.25 ml/minute. The column was eluted with additional equilibration buffer and 3 ml fractions collected. Fractions containing TGF-alpha - PE^q activity were pooled. 238583 9584P/2002a - 27 - 18146IA Q-sepharose Chromatography The pooled fractions from the S-300 column were applied to a 1.6 x 40 cm Q-sepharose column 5 (Pharmacia LKB Biotechnology, Inc.) equilibrated with 6M urea, 50 mM Tris pH 8.0, 50 mM NaCl at a flow rate of 0.7 ml/minute. The column was washed with the equilibration buffer and then eluted with a 600 ml 50-450 mM NaCl gradient. The fractions containing 10 the TGF-alpha - PE^q activity were pooled and then dialysed against 50 mM glycine pH 9.0 and stored at -20°C. 15 Refolding A sample of the protein was thawed and diluted to a spectrophotometric absorbance at UV A280 = 0.1 in 50 mM glycine pH 10.5. Beta-mercaptoethanol 20 was added to give a 4:1 molar ratio over the theoretical number of S-sulphonate groups present in the protein sample. The reaction was allowed to proceed for 16 hours at 4°C after which time the solution was dialysed against a 10,000 fold excess of physiologic-25 ally buffered saline and stored at -20°C. EXAMPLE 5 30 Production and Isolation of Recombinant TGF-alpha - PE^q Fusion Proteins E. coli strain JM-109, containing the appropriate TGF-alpha - PE^q plasmid, was cultured 23 8 5 9584P/2002a - 28 - 18146IA at 37°C in complex medium (Bauer, fit al. , Biotechnology and Bioengineering !& 933-41 (1974)) with antibiotic at 100 mg/ml. TGF-PE40 expression was induced upon addition of 1 mM isopropylthiogalactoside 5 after the culture had attained an absorbance at 600 nm of 2.5. The culture was harvested by cross-flow filtration following a nine hour induction period, and frozen at -70°C. 10 The cell paste was thawed on ice in 4 volumes of 50 mM sodium phosphate, pH 7.8, to form a suspension that was passed through 4 layers of cheesecloth and then twice through a Matin-Gaulin press at 9,000 psi. The filtered suspension was 15 centrifuged in a Sorvall GS-3 rotor at 9000 rpm (13,000 x g) for 30 minutes to remove debris. Saturated ammonium sulfate solution was added to the supernatant liquid dropwise with stirring to a 20% saturation (250 ml/1) at room temperature. The 20 suspension was stirred at 4°C for 0.5-1 hour and then centrifuged in the GS-3 rotor at 9000 rpm (13,000 x g) for 20 minutes. Saturated ammonium sulfate was added to the 25 supernatant liquid with stirring to a 357. concentration (230 ml/1 supernatant). The ammonium sulfate containing solution was stirred at 4°C for 0.5 - 1 hour and then centrifuged as above. The pellet was resuspended in 50 mM sodium phosphate, 507c NH^SO^, 30 pH 7.5 at 1/4 of the starting volume, stirred as above and centrifuged in the Sorvall SA-600 at 5,000 rpm (3,600 x g) for 15 minutes in polypropylene tubes. 238583 9584P/2002a - 29 - 18146IA The supernatant liquid was discarded and the pellets resuspended at 10 mg protein/ml in 50 mM Tris, 6M guanidine-HCl, pH 9.0 at room temperature. 5 Na2S0g was added to a concentration of 0.4M and Na2S40^ was added to a concentration of 0.1M. The pH was checked; if not 9.0, an appropriate adjustment is made with HC1 or NaOH. After stirring overnight at room temperature, the sulfito-10 lyzed protein was dialyzed exhaustively against 50 mM Gly-Cl, pH 9.0 at 4°C. The protein was then diluted to 0.1 mg/ml in 50 mM Gly-Cl, pH 10.5 and a 40-fold molar excess 15 of J3-mercaptoethanol (87 mM J3-Me at 0.1 mg/ml) was added. The mixture was stirred at 4°C for about 15 hours, and the refolded protein was dialyzed for about 15 hours at 4°C against 20 mM Tris-Cl, 50 mM NaCl, pH 8.0. The protein was then loaded onto a Q-Sepharose 20 column pre-equilibrated in 20 mM Tris-Cl, 50 mM NaCl, pH 8.0, at 4°C, using about 0.3 ml resin/mg protein, and eluted with a linear salt gradient from 50 mM to 500 mM NaCl in 20 mM Tris-Cl, pH 8.0 (gradient size = 6-10 column volumes). 25 The column fractions were analyzed and pooled by A280 UV absorption, gel electrophoresis and Western blots. A metal-chelating column was; prepared by treating chelating Sepharose 4B with 30 CuSO^ using 0.3 to 1 ml resin/mg protein. The column was equilibrated with 50 mM Tris-acetate, 1M +2 NaCl, pH 7.0. To assure that no Cu was elutmg, 9584P/2002a - 30 - 18146IA a second metal-free column of chelating Sepahrose 4B 2+ was installed downstream of the Cu -charged column. The Q-Sepharose sample pool was diluted 1:2 5 in 50 mM Tris-acetate, 1M NaCl, pH 7.0, and loaded onto the metal-chelating column at room temperature. The column was washed with one column volume of equilibration buffer, and the protein eluted with a linear gradient of 0 to 70 mM imidazole, maintained 10 at pH 7.0, in the equilibration buffer (gradient size 10 to 40 column volumes). The column fractions were analyzed and pooled by A28q UV absorption, gel electrophoresis 15 and Western blots. EXAMPLE 6 20 Eight human bladder carcinoma cell lines were obtained from the American Type Culture Collection (ATCC) as frozen ampoules. They were immediately cultured and passaged as monolayers according to the instructions provided by ATCC. 25 After characterizing the growth rate of each cell line, cells were plated in 96-well plates at the appropriate dilution to form sub-confluent layers in control wells at the end of the assay. The next clay these sub-confluent cultures, maintained either on 30 serum-free MEM-a, RPMI 1640 or McCoy's 5A medium, were utilized in a standard cell kill assay (Mosmann, J. Immunol. Methods 55-63, 1983; Edwards et al., 25 238583 9584P/2002a - 31 - 18146IA Mol. Cell. Biol. 1: 2860-2867, 1989). Each cell line was seeded into 96-well plates at 10,000 viable cells per well. Twenty-four hours later, the cells were washed once and placed in serum-free medium contain-5 ing the test compound under study. Forty-eight hours later the number of surviving cells was quantitated by using an MTT [3-(3,4-dimethylthiazol-2-yl)-2,5-diphenyltetrazalium bromide] assay as described by Mosmann, supra. The activity of the toxin against 10 each cell line was assessed, and the data are summarized in the following table, with activity against A431 (vulva carcinoma) cells presented for comparison. 15 Activity of TGF-alpha-PE40 ab (EX. 5) Against Human Bladder Carcinoma Cell Lines E£50(pM)* 130 Cell Line J-82 RT-4 180 20 5637 180 SCaBER 230 UMUC-3 830 T-24 840 TCCSUP 7,000 HT1197 11,500 A431 79 ^concentration (picomoles/liter) that reduces number of cells surviving after 48 hours to 50% of number of control cells. 30 23 8 5 9584P/2002a - 32 - 18146IA EXAMPLE 7 Comparison of Several Cancer Cell Lines Against TGF-alpha - PE^q AB, TGF-alpha - PE^q ab 5 of EX. 4 and TGF-alpha - PE^ ab of EX. 5 EC50's [pM] AB (EX. 4) fEX. 5) SOUAMOUS CELL A—431 39 378 163 A-431 146 355 161 A-431 94 314 183 A-431 77 297 207 HeLa 8356 310088 3988 SCC-4 227 861 445 SCC-9 443 647 218 SCC-15 106 392 193 SCC-25 39 147 67 GLIOBLASTOMA U138MG 20889 >316nM 216609 U373MG >316nM >316nM 204064 BREAST ADENOCARCINOMA MOA-MB-468 78 527 253 BT-20 58 207 94 MCF-7 >316nM >316nM >316nM COLON ADENOCARCINOMA HT-29 7605 786 669 NORMAL CELL LINES CHO >316nM >316nM >316nM NR-6 >316nM >316nM >316nM 238 9584P/2002a - 33 - 18146IA imll2 atggctgcagcagtggtgtcccattttaatgactgcccagattcccacactcagttctgcttccatggaacatgcagg 5 tttttggtgcaggaggacaagccggcatgtgtctgccattctgggtacgttggtgcgcgctgtgagcatgcggacctc ctggctgctatggccgaagagggcggcagcctggccgcgctgaccgcgcaccaggcttgccacctgccgctggagact ttcacccgtcatcgccagccgcgcggctgggaacaactggagcagtgcggctatccggtgcagcggctggtcgccctc tacctggcggcgcggctgtcgtggaaccaggtcgaccaggtgatccgcaacgccctggccagccccggcagcggcggc gacctgggcgaagcgatccgcgagcagccggagcaggcccgtctggccctgaccctggccgccgccgagagcgagcgc 10 ttcgt ccggcagggcaccggcaacgacgaggccggcgcggccaacgccgacgt ggt gagcctgacctgcccggtcgcc gccggtgaatgcgcgggcccggcggacagcggcgacgccctgctggagcgcaactatcccactggcgcggagttcctc ggcgacggcggcgacgtcagcttcagcacccgcggcacgcagaactggacggtggagcggctgctccaggcgcaccgc caactggaggagcgcggctatgtgttcgtcggctaccacggcaccttcctcgaagcggcgcaaagcatcgtcttcggc ggggtgcgcgcgcgcagccaggacctcgacgcgatctggcgcggtttctatatcgccggcgatccggcgctggcctac 15 ggctacgcccaggaccaggaacccgacgcacgcggccggatccgcaacggtgccctgctgcgggtctatgtgccgcgc tcgagcctgccgggcttctaccgcaccagcctgaccctggcc3cgccggaggcggcgggcgaggtcgaacggctgatc ggccatccgctgccgctgcgcctggacgccatcaccggccccgaggaggaaggcgggcgcctggagaccattctcggc tggccgctggccgagcgcaccgtggtgattccctcggcgatccccaccgacccgcgcaacgtcggcggcgacctcgac ccgtccagcatccccgacaaggaacaggcgatcagcgccctgccggactacgccagccagcccggcaaaccgccgcgc 20 gaggacctgaagtaa 25 238583 9584P/2002a - 34 - 18146IA table 3 TGF-alpha-PE^ AMINO ACID SEQUENCE 5 -4 -3 -2 -I'TGFa1 6 16 Met Ala Ala Ala'Val Val Ser His Phe Asn Asp Cys Pro Asp Ser His Thr Gin Phe Cys 26 36 Phe His Gly Thr Cys Arg Phe Leu Val Gin Glu Asp Lys Pro Ala Cys Val Cys His Ser 46 TGFa501 1PE252 10 Gly Tyr Val Gly Ala Arg Cys Glu His Ala Asp Leu Leu Ala'Ala Met Ala Glu'Glu Gly 263 273 Gly Ser Leu Ala Ala Leu Thr Ala His Gin Ala Cys His Leu Pro Leu Glu Thr Phe Thr 283 293 Arg His Arg Gin Pro Arg Gly Trp Glu Gin Leu Glu Gin Cys Gly Tyr Pro Val Gin Arg 15 303 313 Leu Val Ala Leu Tyr Leu Ala Ala Arg Leu Ser Trp Asn Gin Val Asp Gin Val He Arg 323 333 Asn Ala Leu Ala Ser Pro Gly Ser Gly Gly Asp Leu Gly Glu Ala lie Arg Glu Gin Pro 343 353 20 Glu Gin Ala Arg Leu Ala Leu Thr Leu Ala Ala Ala Glu Ser Glu Arg Phe Val Arg Gin 363 373 Gly Thr Gly Asn Asp Glu Ala Gly Ala Ala Asn Ala Asp Val Val Ser Leu Thr Cys Pro 383 393 Val Ala Ala Gly Glu Cys Ala Gly Pro Ala Asp Ser Gly Asp Ala Leu Leu Glu Arg Asn 25 403 413 Tyr Pro Thr Gly Ala Glu Phe Leu Gly Asp Gly Gly Asp Val Ser Phe Ser Thr Arg Gly 423 433 Thr Gin Asn Trp Thr Val Glu Arg Leu Leu Gin Als Hi? Arg Gin Leu Glu Gl" Aiq f?iv /l*n 45 ^ 30 Tyr Val Phe Val Gly Tyr His Gly Thr Phe Leu Glu Ala Ala Gin Ser He Val Phe Gly 463 473 Gly Val Arg Ala Arg Ser Gin Asp Leu Asp Ala lie Trp Arg Gly Phe Tyr lie Ala Gly 2 3 8'- 9584P/2002a - 35 - 18146IA TABLE 3 CONT'D TGF-alpha-PE^ AMINO ACID SEQUENCE 5 483 493 Asp Pro Ala Leu Ala Tyr Gly Tyr Ala Gin Asp Gin Glu Pro Asp Ala Arg Gly Arg lie 503 513 Arg Asn Gly Ala Leu Leu Arg Val Tyr Val Pro Arg Ser Ser Leu Pro Gly Phe Tyr Arg 523 533 10 Thr Ser Leu Thr Leu Ala Ala Pro Glu Ala Ala Gly Glu Val Glu Arg Leu lie Gly His 543 553 Pro Leu Pro Leu Arg Leu Asp Ala lie Thr Gly Pro Glu Glu Glu Gly Gly Arg Leu Glu 563 573 Thr. lie Leu Gly Trp Pro Leu Ala Glu Arg Thr Val Val He Pro Ser Ala lie Pro Thr 15 583 593 Asp Pro Arg Asn Val Gly Gly Asp Leu Asp Pro Ser Ser He Pro Asp Lys Glu Gin Ala 603 613 lie Ser Ala Leu Pro Asp Tyr Ala Ser Gin Pro Gly Lys Pro Pro Arg Glu Asp Leu Lys 20 25 23 8 5 9584P/2002a - 36 - 18146IA TABLE 4 TGF-alpha-PE^-aB AMINO ACID SEQUENCE 5 -4 -3 -2 -1' TGFa1 6 16 Met Ala Ala Ala'Val Val Ser His Phe Asn Asp Cys Pro Asp Ser His Thr Gin Phe Cys 26 36 Phe His Gly Thr Cys Arg Phe Leu Val Gin Glu Asp Lys Pro Ala Cys Val Cys His Ser 46 TGFa50 • ' PE252 10 Gly Tyr Val Gly Ala Arg Cys Glu His Ala Asp Leu Leu Ala'Met Ala Glu'Glu Gly Gly 264 274 Ser Leu Ala Ala Leu Thr Ala His Gin Ala Ala His Leu Pro Leu Glu Thr Leu Thr Arg 284 294 His Arg Gin Pro Arg Gly Trp Glu Gin Leu Glu Gin Ala Gly Tyr Pro Val Gin Arg Leu 15 304 314 Val Ala Leu Tyr Leu Ala Ala Arg Leu Ser Trp Asn Gin Val Asp Gin Val lie Arg Asn 324 334 Ala Leu Ala Ser Pro Gly Ser Gly Gly Asp Leu Gly Glu Ala lie Arg Glu Gin Pro Glu 344 354 20 Gin Ala Arg Leu Ala Leu Thr Leu Ala Ala Ala Glu Ser Glu Arg Phe Val Arg Gin Gly 364 374 Thr Gly Asn Asp Glu Ala Gly Ala Ala Asn Ala Asp Val Val Ser Leu Thr Cys Pro Val 384 394 Ala Ala Gly Glu Cys Ala Gly Pro Ala Asp Ser Gly Asp Ala Leu Leu Glu Arg Asn Tyr 25 404 414 Pro Thr Glu Ala Glu Phe Leu Gly Asp Gly Gly Asp Val Ser Phe Ser Thr Arg Gly Thr 424 434 Gin Asn Trp Thr Val Glu Arg Leu Leu Gin Ala His Arg Gin Leu Glu Glu Ara Gly Tyr 444 454 30 Val Phe Val Gly Tyr His Gly Thr Phe Leu Glu Ala Ala Gin Ser lie Val Phe Gly Gly 464 474 Val Arg Ala Arg Ser Gin Asp Leu Asp Ala lie Trp Arg Gly Phe Tyr lie Ala Gly Asp 238583 9584P/2002a - 37 - 18146IA TABLE 4 CONT'D TGF-a1phs-PE40 aB AMINO ACID SEQUENCE 5 484 494 Pro Ala Leu Ala Tyr Gly Tyr Ala Gin Asp Gin Glu Pro Asp Ala Arg Gly Arg lie Arg 504 514 Asn Gly Ala Leu Leu Arg Val Tyr Val Pro Arg Ser Ser Leu Pro Gly Phe Tyr Arg Thr 524 534 10 Ser Leu Thr Leu Ala Ala Pro Glu Ala Ala Gly Glu Val Glu Arg Leu lie Gly His Pro 544 554 Leu Pro Leu Arg Leu Asp Ala lie Thr Gly Pro Glu Glu Glu Gly Gly Arg Leu Glu Thr 564 574 lie Leu Gly Trp Pro Leu Ala Glu Arg Thr Val Val lie Pro Ser Ala He Pro Thr Asp 15 584 594 Pro Arg Asn Val Gly Gly Asp Leu Asp Pro Ser Ser lie Pro Asp Lys Glu Gin Ala lie 604 614 Ser Ala Leu Pro Asp Tyr Ala Ser Gin Pro Gly Lys Pro Pro Arg Glu Asp Leu Lys 20 25 30 238583 9584P/2002a - 38 - 18146IA table 5 TGF-a1pha-PE4() Ab AMINO ACIO SEQUENCE 5 -4 -3 -2 -1'TGFa1 6 16 Met Ala Ala Ala'Val Val Ser His Phe Asn Asp Cys Pro Asp Ser His Thr Gin Phe Cys 26 36 Phe His Gly Thr Cys Arg Phe Leu Val Gin Glu Asp Lys Pro Ala Cys Val Cys His Ser 50. . 252 46 TGFa ' 'PE 10 Gly Tyr Val Gly Ala Arg Cys Glu His Ala Asp Leu Leu Ala'Ala Met Ala Glu'Glu Gly 263 273 Gly Ser Leu Ala Ala Leu Thr Ala His Gin Ala Cys His Leu Pro Leu Glu Thr Phe Thr 283 293 Arg His Arg Gin Pro Arg Gly Trp Glu Gin Leu Glu Gin Cys Gly Tyr Pro Val Gin Arg 15 303 313 Leu Val Ala Leu Tyr Leu Ala Ala Arg Leu Ser Trp Asn Gin Val Asp Gin Val lie Arg 323 333 Asn Ala Leu Ala Ser Pro Gly Ser Gly Gly Asp Leu Gly Glu Ala lie Arg Glu Gin Pro 343 353 20 Glu Gin Ala Arg Leu Ala Leu Thr Leu Ala Ala Ala Glu Ser Glu Arg Phe Val Arg Gin 363 373 Gly Thr Gly Asn Asp Glu Ala Gly Ala Ala Asn Ala Asp Val Val Thr Leu Thr Ala Pro 383 393 Val Ala Ala Gly Glu Ala Ala Gly Pro Ala Asp Ser Gly Asp Ala Leu Leu Glu Arg Asn 25 403 413 Tyr Pro Thr Gly Ala Glu Phe Leu Gly Asp Gly Gly Asp Val Ser Phe Ser Thr Arg Gly 423 433 Thr Gin Asn Trp Thr Val Glu Arg Leu Leu Gin Ala His Arg Gl" Leu Glu Glu Arg Gly 443 453 30 Tyr Val Phe Val Gly Tyr His Gly Thr Phe Leu Glu Ala Ala Gin Ser lie Val Phe Gly 463 473 Gly Val Arg Ala Arg Ser Gin Asp Leu Asp Ala lie Trp Arg Gly Phe Tyr lie Ala Gly 23 8 5 8 3 9584P/2002a - 39 - 18146IA TABLE 5 CONT'D TGF-alpha-PE^ Ab AMINO ACID SEQUENCE 5 483 493 Asp Pro Ala Leu Ala Tyr Gly Tyr Ala Gin Asp Gin Glu Pro Asp Ala Arg Gly Arg lie 503 513 Arg Asn Gly Ala Leu Leu Arg Val Tyr Val Pro Arg Ser Ser Leu Pro Gly Phe Tyr Arg 523 533 10 Thr Ser Leu. Thr Leu Ala Ala Pro Glu Ala Ala Gly Glu Val Glu Arg Leu lie Gly His 543 553 Pro Leu Pro Leu Arg Leu Asp Ala lie Thr Gly Pro Glu Glu Glu Gly Gly Arg Leu Glu 563 573 Thr He Leu Gly Trp Pro Leu Ala Glu Arg Thr Val Val He Pro Ser Ala He Pro Thr 15 583 593 Asp Pro Arg Asn Val Gly Gly Asp Leu Asp Pro Ser Ser lie Pro Asp Lys Glu Gin Ala 603 613 lie Ser Ala Leu Pro Asp Tyr Ala Ser Gin Pro Gly Lys Pro Pro Arg Glu Asp Leu Lys 20 25 30 9584P/2002a - 40 - 18146IA table 6 TGF-alpha-PE ab AMINO ACID SEQUENCE 40 -4 -3 -2 -1 ' TGFa1 6 16 Met Ala Ala Ala'Val Val Ser His Phe Asn Asp Cys Pro Asp Ser His Thr Gin Phe Cys 26 36 Phe His Gly Thr Cys Arg Phe Leu Val Gin Glu Asp Lys Pro Ala Cys Val Cys His Ser 46 TGFa501 »PE252 Gly Tyr Val Gly Ala Arg Cys Glu His Ala Asp Leu Leu Ala'Met Ala Glu'Glu Gly Gly 264 274 Ser Leu Ala Ala Leu Thr Ala His Gin Ala Ala His Leu Pro Leu Glu Thr Leu Thr Arg 284 294 His Arg Gin Pro Arg Gly Trp Glu Gin Leu Glu Gin Ala Gly Tyr Pro Val Gin Arg Leu 304 314 Val Ala Leu Tyr Leu Ala Ala Arg Leu Ser Trp Asn Gin Val Asp Gin Val lie Arg Asn 324 334 Ala Leu Ala Ser Pro Gly Ser Gly Gly Asp Leu Gly Glu Ala lie Arg Glu Gin Pro Glu 344 354 Gin Ala Arg Leu Ala Leu Thr Leu Ala Ala Ala Glu Ser Glu Arg Phe Val Arg Gin Gly 364 374 Thr Gly Asn Asp Glu Ala Gly Ala Ala Asn Ala Asp Val Val Thr Leu Thr Ala Pro Val 384 394 Ala Ala Gly Glu Ala Ala Gly Pro Ala Asp Ser Gly Asp Ala Leu Leu Glu Arg Asn Tyr 404 414 Pro Thr Gly Ala Glu Phe Leu Gly Asp Gly Gly Asp Val Ser Phe Ser Thr Arg Gly Thr 424 434 Gin Asn Trp Thr Val Glu Arg Leu Leu Gin Ala His Arg Gin Leu Glu Glu Arci Gly Tyr 444 454 Val Phe Val Gly Tyr His Gly Thr Phe Leu Glu Ala Ala Gin Ser He Val Phe Gly Gly 464 474 Val Arg Ala Arg Ser Gin Asp Leu Asp Ala lie Trp Arg Gly Phe Tyr He Ala Gly Asp 238583 9584P/2002a " 41 " 18146IA TABLE 6 CONT'D TGF-alpha-PE40 ab AMINO ACID SEQUENCE 5 484 494 Pro Ala Leu Ala Tyr Gly Tyr Ala Gin Asp Gin Glu Pro Asp Ala Arg Gly Arg He Arg 504 514 Asn Gly Ala Leu Leu Arg Val Tyr Val Pro Arg Ser Ser Leu Pro Gly Phe Tyr Arg Thr 524 534 10 Ser Leu Thr Leu Ala Ala Pro Glu Ala Ala Gly Glu Val Glu Arg Leu lie Gly His Pro 544 554 Leu Pro Leu Arg Leu Asp Ala lie Thr Gly Pro Glu G1u Glu Gly Gly Arg Leu Glu Thr 564 574 He Leu Gly Trp Pro Leu Ala Glu Arg Thr Val Val lie Pro Ser Ala lie Pro Thr Asp 1 5 584 594 Pro Arg Asn Val Gly Gly Asp Leu Asp Pro Ser Ser He Pro Asp Lys Glu Gin Ala lie 604 614 Ser Ala Leu Pro Asp Tyr Ala Ser Gin Pro Gly Lys Pro Pro Arg Glu Asp Leu Lys 20 25 30 - 42 - 23 85$ j WHAT WE CLAIM IS: 1. A composition comprising a physiologically acceptable liquid containing a concentration of a hybrid protein TGF-alpha - PE40 aB, TGF-alpha - PE40 Ab or TGF-alpha - PE40 ab (all as hereinbefore defined) that is effective to kill bladder cancer cells. 2. A composition according to Claim 1 that contains from substantially 0.1 mg to substantially 10 mg of the hybrid protein per 60 ml. 3. A composition according to Claim 1 that contains from substantially 0.5 mg to substantially 5 mg of the hybrid protein per 60 ml. ss'-i £• ' A 4. A composition according to Claim 1 thafcV ^ ■'* I?: o rr contains from substantially 2 mg to substantially 4 ^ of the hybrid protein per 60 ml. g DEC1993 5. A composition according to any one of % </div>

Claims

<div class="application article clearfix printTableText" id="claims"> Claims 1 to 4 wherein the physiologically acceptable liquid is phosphate buffered saline optionally r-"- containing a carrier protein. 6. A composition according to Claim 5 wherein the optional carrier protein is human serum albumin. - 43 - 238583 7. Hybrid proteins TGF-alpha - pe4q aB, TGF-alpha - pe40 Ab or TGF-alpha - pe4q ab (all as hereinbefore defined) of enhanced potency prepared by a process comprising precipitating with sulfate one of the foregoing hybrid proteins that has been expressed in a transformed cell, and purifying the hybrid protein by affinity chromatography using a metal-chelating column, and not subjecting the hybrid protein to treatment with urea. 8. A hybrid protein according to Claim 7 wherein the precipitating sulfate is (NH4)2S04. DATED THIS ^ A. J. PARK & SON PER MT^ FOD TUC *DOUO/\MTS n.z. patent office -8 JUL 1993 received </div>