IE61775B1 - Process for the production of proteins - Google Patents

Description

Process for the production of proteins The Invention concerns a novel process for the production of proteins, snore especially serine proteases of the urokinase type. Said process includes the use of genetically engineered yeast strains. The invention concerns furthermore novel urokinase-type proteins, DNAs encoding such proteins, hybrid vectors containing such DNAs, said genetically engineered yeast strains and processes for the production of said DNAs, hybrid vectors and yeast strains.

Urokinase or urokinase-type plasminogen activator (hereinafter referred to as ,5u-PA”) is a serine protease which activates plasminogen to plasmin by proteolytic cleavage. Plasmin is a potent protease which is able to degrade the fibrin network of blood clots to form soluble degradation products. u-PA was first isolated from human urine and is also known to be secreted by cultured kidney cells and some tumour cell lines. It is initially produced as a single chain molecule (hereinafter referred to as 5’scu-PA5’) and can be proteolytically converted by the action of plasmin to a two chain form (hereinafter referred to as tcu-PA) in which the two chains remain attached to each other via a disulfide bridge. u~PA has a unique glycosylation site at Asn302.

Since scu-PA has only inferior amidolytic activity against low molecular weight synthetic substrates it was regarded until recently as a true proteolytically inactive precursor of the active enzyme tcu-PA. Most recent results, however, prove that scu-PA efficiently activates plasminogen to plasmin and has a considerably higher selectivity for fibrin than tcu-PA [cf. H.R. Lijnen et al..

J.Biol.Chem. 261, 1253 (1986)]. The mechanisms of the surprising fibrinolytic activity and the clot specificity of scu-PA have been studied [Lijnen et al. supra; D. Collen et al. J.Biol.Chem. 261, 1259 (1986)]. It was demonstrated that scu-PA is not an inactive zymogen but activates plasminogen without being previously transformed into tcu-PA. Unlike tcu-PA, scu-PA is competitively inhibited by a yet unknown plasma component which inhibition is reversed by fibrin or fibrin fragments, i.e. at the clot. While circulating in blood under in vivo conditions scu-PA does therefore not activate plasminogen. Its fibrinolytic activity remains restricted to its proper target. On the contrary, tcu-PA is capable of activating plasminogen at any point within the circulatory system, thereby leading to undesirable side effects such as hemorrhage. These properties render scu-PA the preferred urokinase-type plasminogen activator.

With the advent of recombinant DNA technology it is now possible to produce proteins such as scu-PA on industrial scale. Based on the known structure of genomic u-PA DNA [A. Riccio et al. Nucleic Acids Research 13, 2759 (1985)] and u-PA cDNA [W.E. Holmes et al. Biotechnology 3, 923 (1985)] processes for the production of scu-PA which make use of recombinant DNA technology have been described in the literature. Thus, expression in E. coli has been achieved by W.E. Holmes et al. (supra), P. Jacobs et al. [DNA 4, 139 (1985)], M.E. Winkler et al. [Biochemistry 25, 4041 (1986)], M. Nagai et al. [Gene 36, 183 (1985)]; see also Belgian Patent No. 900 826, Japanese Patent No. 61 181 377, and European Patent Application No. 92 182. The expression of scu-PA in animal cells is disclosed in European Patent Applications No. 92 182 and 154 272. However, all of these known processes suffer from weighty disadvantages: It has been proven difficult to grow animal cells on a large scale which Is a prerequisite for the cheap and profitable manufacture of proteins produced by these cells. The generation time of animals cells is considerably higher than that of microorganisms, thus requiring a prolonged fermentation period in order to obtain a sufficiently high cell density. The cell density, in turn, obtainable in the cultiva3 tion of animal cells is considerably lower than the cell density generally reached in large scale cultivation of microorganisms. Moreover, strain improvements are difficult to achieve as compared to microorganisms. On the other hand, contaminating endotoxins are often found in protein preparations from E. coli. These have to be eliminated by expensive and time-consuming purification steps. Proteins produced by E. coli are necessarily unglycosylated because E. coli is devoid of the enzymatic system which is responsible for the attachment of carbohydrate chains to appropriate sites in the protein molecule. Recombinant scu-PA is therefore, unlike natural scu-PA, unglycosylated when produced by E. coli. It was reported that scu-PA produced by E. coli exists as an amorphous insoluble polymer, due to an imperfect alignment of the disulfide bridges and to an incorrect folding of the protein. At least one additional refolding step involving large quantities of solvent is thus required in order to obtain a biologically active protein (M.E. Winkler et al., supra).

Considering the drawbacks of the known processes there is a continued need for improved methods which render possible the production of biologically active and scu-PA on a large scale. It is an object of the present invention to provide such methods.

It has surprisingly been found that yeast cells transformed with a hybrid vector carrying the human u-PA coding sequence attached to the signal sequence of a yeast gene produce scu-PA which has a biological activity equivalent to that of natural scu-PA and is yeast specifically glycosylated. It is noteworthy that yeast scu-PA is completely active without any in vitro refolding procedures being required.

Accordingly, the invention concerns a method for the production of human single chain urokinase-type plasminogen activator comprising 1 culturing under appropriate nutrient conditions a yeast strain transformed with a hybrid vector comprising a yeast expression control sequence, a DNA segment consisting of a first DNA sequence encoding a signal peptide upstream of and in reading frame with a second DNA sequence coding for mature urokinase-type plasminogen activator, which DNA segment is under transcriptional control of said expression control sequence, and a DNA sequence comprising transcription termination signals of a yeast gene, and isolating said urokinase-type plasminogen activator.

The term DNA sequence coding for mature urokinase-type plasminogen activator is intended to embrace all allelic forms of u-PA which are known to exist in or can be isolated from the human genome. These DNA sequences are devoid of any pre- and/or pro-sequences.

The invention relates especially to a method for the production of scu-PA proteins having the formula Ϊ Ser Asn Glu Leu His Gin Val Pro Ser Asn Cys Asp Cys Leu Asn Gly Gly Thr Cys Val Ser Asn Lys Tyr Phe Ser Asn He His Trp Cys Asn Cys Pro Lys Lys Phe Gly Gly Gin His Cys Glu He Asp Lys Ser Lys Thr Cys Tyr Glu Gly Asn Gly His Phe Tyr Arg Gly Lys Ala Ser Thr Asp Thr Het Gly Arg Pro Cys Leu Pro Trp Asn Ser Ala Thr Val Leu Gin Gin Thr Tyr His Ala His Arg Ser Asp Ala Leu Gin Leu Gly Leu Gly Lys His Asn Tyr Cys Arg Asn Pro Asp Asn Arg Arg Arg Pro Trp Cys Tyr Val Gin Val Gly Leu Lys Pro Leu Val Gin Glu Cys Het Val His Asp Cys Ala Asp Gly Xj x2 Pro Ser Ser Pro Pro Glu Glu Leu Lys Phe Gin Cys Gly Gin Lys Tk w Alia. Leu Arg Pro ¥1 ¥2 ¥3 Tie lie Gly Gly Glu Phe Thr Thr T le Glu Asn Gin Pro Trp Phe Ala Ala I le Tyr Arg Arg His Arg Gly Gly Ser Val Thr Tyr Val Cys Gly Gly Ser Leu He Ser Pro Cys Trp Val He Ser Ala Thr His Cys Phe lie Asp Tyr Pro Lys Lys Glu Asp Tyr He Val Tyr Leu Gly Arg Ser Arg Leu Asn Ser Asn Thr Gin Gly Glu Met Lys Phe Glu Val Glu Asn Leu He Leu His Lys Asp Tyr Ser Ala Asp Thr Leu Ala His His Asn Asp He Ala Leu Leu Lys He Arg Ser Lys Glu Gly Arg Cys Ala Gin Pro Ser Arg Thr He Gin Thr Tie Cys Leu Pro Ser Met Tvr Asn Asp Pro Gin Phe Gly Thr Ser Cys Glu lie Thr Gly Phe Gly Lys Glu Zi Ser z2 Asp Tyr Leu Tyr Pro Glu Gin Leu Lys Met Thr Val Val Lys Leu I le Ser His Arg Glu Cys Gin Gin Pro His Tyr Tyr Gly Ser Glu Val Thr Thr Lys Met Leu Cys Ala Ala Asp Pro Gin 1 Fp Lys Thr Asp Ser Cys Gin Gly Asp Ser Gly Gly Pro Leu Val Cys Ser Leu Gin Gly Arg Met Thr Leu Thr Gly He Val Ser Trp Gly Arg Gly Cys Ala Leu in Lys Asp Lys Pro Gly Val Tyr Thr Arg Val Ser His Phe Leu Pro Trp lie Arg Ser His Thr Lys Glu Glu Asn Gly Leu Ala Leu which X] and Xz independently from each other represent Lys, '1 is Arg, Y2 is Phe, Y3 is Lys, Z, is Asn which is yeast-specifically glycosylated, and Z2 is Thr.

The transformed yeast strains according to the invention are cultured in a liquid medium containing assimilable sources of carbon, nitrogen and inorganic salts.

Various carbon sources are usable. Examples of preferred carbon sources are assimilable carbohydrates, such as glucose, maltose, mannitol or lactose, or an acetate such as sodium acetate, which can be used either alone or in suitable mixtures. Suitable nitrogen sources include, for example, amino acids, such as casamino acids, peptides and proteins and their degradation products, such as tryptone, peptone or meat extracts, furthermore yeast extract, malt extract, corn steep liquor, as well as ammonium salts, such as ammonium chloride, sulphate or nitrate, which can be used either alone or in suitable mixtures. Inorganic salts which may be used include, for example, sulphates, chlorides, phosphates and carbonates of sodium, potassium, magnesium and calcium. Additionally, the nutrient medium may also contain growth promoting substances. Substances which promote growth include, for example, trace elements, such as iron, zinc, manganese and the like, or individual amino acids.

Yeast cells containing hybrid plasmids with a constitutive promoter (e.g. ADHI, GAPDH) express the u-PA gene attached to said promoter without induction. However, if the u-PA gene is under the control of a regulated promoter (e.g. PGK or PHO5) the composition of the growth medium has to be adapted in order to obtain maximum levels of » mRNA transcripts, i.e. when using the PHO5 promoter the growth medium must contain a low concentration of inorganic phosphate for derepression of this promoter.

The cultivation is carried out by employing conventional techniques.

The culturing conditions, such as temperature, pH of the medium and fermentation time are selected in such a way that maximal levels of scu-PA proteins are produced. A chosen yeast strain is preferably grown under aerobic conditions in submerged culture with shaking or stirring at a temperature of about 25° to 35°C, preferably at about 28°C, at a pH value of from 4 to 7, for example at approximately pH 5, and for about 20 to 50 hours, preferably until maximum yields of scu-PA proteins are reached.

The produced scu-PA protein can accumulate within the yeast cells or can be secreted into the periplasmic space. In case the scu-PA protein has accumulated within the cells, the first step for the recovery of the scu-PA protein consists in liberating the protein from the cell interior. In most procedures the cell wall is first removed by enzymatic digestion with glucosidases (infra). Subsequently, the resulting spheroplasts are treated with detergents, such as Triton® X-100. Alternatively, mechanical forces, such as shearing forces (for example X-press, French-press) or shaking with glass beads, are suitable for breaking cells. The resulting mixture is enriched for scu-PA protein by conventional means, such as removal of most of the non-proteinaceous material by treatment with polyethvleneimine, precipitation of the proteins using ammonium r sulphate, gel electrophoresis, dialysis, chromatography, for example, ion exchange chromatography, size-exclusion chromatography, .

HPLC or reverse phase HPLC, molecular sizing on a suitable Sephadex® column, or the like. The final purification of the pre-purified product is achieved, for example, by means of affinity chromatography, for example antibody affinity chromatography, especially monoclonal antibody affinity chromatography using monoclonal anti-u-PA antibodies fixed on an insoluble matrix by methods known in the art, and the like. Advantageously, a detergent, especially a nonionic detergent, such as Triton X-100® or Tween 80®, is added to all buffer solutions used in the purification steps, in order to prevent the adsorption of the scu-PA protein to the vessel surfaces and to improve stability. The detergent may be added to a final concentration of 0.01-1 %.

In the case where the scu-PA protein is secreted by the yeast cell into the periplasmic space, a simplified protocol can be used: The protein is recovered without cell lysis by enzymatic removal of the cell wall or by treatment with chemical agents, e.g. thiol reagents or EDTA, which gives rise to cell wall damages permitting the produced scu-PA protein to be released. In the case where the scu-PA protein is secreted into the culture broth, it can be recovered directly therefrom.

Dependent on the host strain used and the purification methods applied the scu-PA proteins according to the present invention may be contaminated with small amounts of the corresponding two chain forms caused by proteolytic activity released by the host cells. Separation of the two chain form from the desired one chain form (scu-PA) is accomplished by methods known in the art such as by chromatography on benzamidine-Sepharose® [cf. M.E. Winkler et al. Biochemistry 25, 4041 (1986)].

Surprisingly, it was found that the scu-PA proteins according to the present invention differ from scu-PA obtained from E. coli in that they exhibit the biological activity of natural human scu-PA without any refolding procedure being necessary and in that they are yeast specifically glycosylated at Asn302. Owing to the yeast specific glycosylation the scu-PA proteins according to the invention are also distinct from scu-PA isolated from cultured or transformed animal cells and are thus novel.

Accordingly, the invention relates also to scu-PA and mutants 5 thereof having yeast specif?ic glycosylation, in particular having a glycosylation specific for Saccharoaiyces cerevisiae.

The transformed yeast strains according to the invention can be prepared by recombinant DNA techniques comprising the steps of - preparing a structural gene coding for scu-PA or a mutant thereof, - incorporating the obtained structural gene into an appropriate vector, - transforming a suitable host organism with the produced hybrid vector - and selecting transformed hosts from untransformed hosts.

The nucleotide sequences of u-PA cDNA and of genomic u-PA DNA are known [W.E. Holmes et al., Biotechnology 3, 923 (1985); A. Riccio et al. Nucleic Acids Research 13, 2759 (1985)]. Knowing the cDNA and genomic DNA sequences of u-PA the structural gene coding for u-PA or a mutant thereof can be made by methods known in the art. The methods for making these DNAs include isolating mRNA from human cells which produce scu-PA such as human carcinoma cells or human embryo kidney cells, selecting the desired mRNA, e.g. by hybridization with a suitable DNA probe, preparing single stranded DNA complementary to that mRNA, then double stranded complementary DNA (ds cDNA) therefrom, or isolating genomic DNA from human cells and selecting the desired DNA using a suitable DNA probe, and, if required, mutating the cDNA or genomic DNA obtained, or preparing the structural gene by chemical synthesis. Preferably, the structural genes according to the invention are prepared via the mRNA route and, if mutants are desired, by mutagenesis of the primarily obtained u-PA cDNA. Accordingly, the structural gene coding for s. mutant of u-PA can be prepared by excising a portion of the DNA comprising the codon(s) for the undesired amino acid residue(s) from the parental mature u-PA gene and replacing it with a DNA segment wherein said codon(s) has (have) been substituted with codon(s) coding for the desired amino acid residue(s), or accomplishing the deoxyribonucleotide substitution by means of site directed mutagenesis [cf. M.J. Zoller et al. Methods Enzymol. 100, 468 (1983), D. Botstein et al. Science 229, 1193 (1985)].

The hybrid vectors according to the present invention comprise a yeast expression control sequence, a DNA segment consisting of a first DNA sequence encoding a signal peptide upstream of and in reading frame with a second DNA sequence coding for mature urokinase-type plasminogen activator or a mutant thereof, which ΐ! Ο DNA segment is under transcriptional control of said expression control sequence, and a DNA sequence comprising transcription termination signals of a yeast gene.

Yeast expression control sequences are derived from the genomic DNA of yeast, especially of Saccharomyces cerevisiae. Preferably, the expression control sequence of a highly expressed yeast gene is used for the expression of scu-PA. Thus, the promoter of the TRP1 gene, the ADHI or ADHII gene, acid phosphatase (PH05) gene, a promoter of the enolase, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), 3-phosphoglycerate kinase (PGK), hexokinase, pyruvate decarboxylase, phosphofructokinase, glucose-6-phosphate isomerase, 3-phospho= glycerate mutase, pyruvate kinase, triosephosphate isomerase, phosphoglucose isomerase and glucokinase genes, or a promoter of the yeast mating pheromone genes coding for the a~ or α-factor, can be used. It is also possible to use hybrid promoters comprising upstream activation sequences (UAS) of one yeast gene and downstream promoter elements including a functional TATA box of another yeast gene, for example a hybrid promoter including the UAS(s) of the yeast PHO5 gene and downstream promoter elements including a functional TATA box of the yeast GAPDH gene. Preferred vectors of the present invention contain promoters with transcriptional control. Promoters of this type, e.g. the promoter of the PH05 gene and PH05-GAPDH hybrid promoters, can be turned on or off by variation of the growth conditions. For example, the PHO5 promoter can be repressed at will, solely by increasing the concentration of inorganic phosphate in the medium. A further preferred promoter according to the invention is the promoter of the GAPDH gene, especially functional fragments thereof starting at nucleotides between -550 and -180, in particular at nucleotide -540, -263 or -198, and ending at nucleotide -5 of the GAPDH gene.

The DNA sequence encoding a signal peptide (“'signal sequence) is preferably derived from eukaryotic, for example human or yeast, genes coding for polypeptides which are ordinarily secreted.

Suitable signal sequences are, for example, the u-PA signal sequence obtainable from genomic human DNA, yeast signal sequences, such as the signal and prepro sequences of the yeast invertase, a-factor, pheromone peptidase (KEX1), killer toxin and repressible acid phosphatase (PH05) genes and the glucoamvlase signal sequence from Aspergillus awamori. Alternatively, fused signal sequences may be constructed by ligating part of the signal sequence (if present) of the gene naturally linked to the promoter used, with part of the u-PA signal sequence. Those combinations are favoured which allow a precise cleavage between the signal sequence and the mature scu-PA amino acid sequence. Additional sequences, such as pro- or spacersequences which may or may not carry specific processing signals can also be included in the constructions to facilitate accurate processing of precursor molecules. Alternatively fused proteins can be generated containing internal processing signals which allow proper maturation in vivo or in vitro. For example, the processing signals contain a Lys-Arg residue, which is recognized by a yeast endopeptidase located in the Golgi membranes. The preferred signal sequences according to the present invention are those of the yeast PHO5 gene coding for a signal peptide having the formula 2q Met Phe Lys Ser Val Val Tyr Ser lie Leu Ala Ala Ser Leu Ala Asn Ala, of the yeast invertase gene coding for a signal peptide having the formula Met Leu Leu Gin Ala Phe Leu Phe Leu Leu Ala Gly Phe Ala Ala Lys lie Ser Ala. and of the human u-PA gene coding for a signal peptide having the formula Met Arg Ala Leu Leu Ala Arg Leu Leu Leu Cys Val Leu Val Val Ser Asp Ser Lys Gly.

A DNA sequence comprising yeast transcription termination signals is preferably the 3' flanking sequence of a yeast gene which contains proper signals for transcription termination and polyadenylation. Suitable 3' flanking sequences are for example those of the yeast gene naturally linked to the expression control sequence used. The preferred flanking sequences are those of the yeast PHO5 gene.

The hybrid plasmids according to the invention contain apart from the promoter, the signal sequence, the DNA sequence coding for u-PA or a mutant thereof and 3' flanking sequences additional DNA sequence(s) which perform important functions, for example, in the propagation of the cells transformed with said hybrid plasmids. The additional DNA sequence(s) may be derived from prokaryotic andZor eukaryotic cells and may include chromosomal andZor extra-chromosomal DNA sequences. For example, the additional DNA sequences may stem from (or consist of) plasmid DNA, such as bacterial or eukaryotic plasmid DNA, viral DNA andZor chromosomal DNA, such as bacterial, yeast or higher eukaryotic chromosomal DNA. Preferred hybrid plasmids contain additional DNA sequences derived from bacterial plasmids, especially Escherichia coli plasmid pBR322 or pUCl9 related plasmids, bacteriophage λ, yeast 2μ plasmid, andZor yeast chromosomal DNA.

In the preferred hybrid plasmids according to the invention, the additional DNA sequences carry a yeast replication origin and a selective genetic marker for yeast. Hybrid plasmids containing a yeast replication origin, e.g. an autonomously replicating segment(ars), are extrachromosomally maintained within the yeast cell after transformation and are autonomously replicated upon mitosis. Hybrid plasmids containing sequences homologous to yeast 2μ plasmid DNA can be used as well. These hybrid plasmids recombine with the yeast 2μ plasmids already present within the cell or will replicate autonomously on their own if the origin of replication is present. 2μ sequences are especially suitable for high-frequency transformation plasmids and give rise to high copy numbers.

As to the selective gene marker for yeast, any marker gene can be used which facilitates the selection for transformants due to the phenotypic expression of the marker. Suitable dominant markers for yeast are particularly those expressing antibiotic resistance or, in the case of auxotrophic yeast mutants, genes which complement host lesions. Corresponding genes confer, for example, resistance to the antibiotic G418 or hygromycin or provide for prototrophy in an auxotrophic yeast mutant, for example the URA3, LEU2, HIS3 or TRP1 gene.

Advantageously, the additional DNA sequences which are present in the hybrid plasmids according to the invention also include a replication origin and a selective genetic marker for a bacterial host, especially Escherichia coli. There are useful features which are associated with the presence of an E. coli replication origin and an E. coli marker in a yeast hybrid plasmid. Firstly, large amounts of hybrid plasmid DNA can be obtained by growth and amplification in E. coli and, secondly, the construction of hybrid plasmids is conveniently done in E. coli making use of the whole repertoire of cloning technology based on E. coli. E. coli plasmids, such as pBR322 and the like, contain both E. coli replication origin and E. coli genetic markers conferring resistance to antibiotics, for example tetracycline and ampicillin, and are advantagously employed as part of the yeast hybrid vectors.

The additional DNA sequences which contain, for example, replication origin and genetic markers for yeast and a bacterial host (see above) are hereinafter referred to as ‘'vector DNA which together with the yeast promoter and the scu-PA protein coding region is forming a hybrid plasmid according to the invention.

In a preferred embodiment, the present invention relates to hybrid plasmids capable of autonomous replication in a yeast host strain and having selectable markers comprising a yeast expression control sequence, a DNA segment consisting of a first DNA sequence encoding a signal peptide upstream of and in reading frame with a second DNA sequence coding for mature urokinase-type plasminogen activator or a mutant thereof, and a DNA sequence containing transcription termination signals of a yeast gene, said DNA segment being positioned together with transcription start and termination signals as well as encoded translation start and stop signals in said hybrid vector under control of said expression control sequence such that in a transformed yeast strain it is expressed to produce said urokinase-type plasminogen activator or mutants thereof.

The hybrid vectors of the present invention are prepared by methods known in the art, for example by linking a yeast expression control sequence, a DNA segment consisting of the signal peptide coding region and a DNA sequence coding for u-PA or a mutant thereof, the 3' flanking sequence of a yeast gene and vector DNA.

For the preparation of hybrid plasmids, conveniently mapped circular vector DNA, for example bacterial plasmid DNA or the like (see above), having at least one restriction site, preferably two or more restriction sites, can be employed. Advantageously, the vector DNA already contains replication origins and gene markers for yeast and/or a bacterial host. The vector DNA is cleaved using an appropriate restriction endonuclease. The restricted DNA is ligated to the DNA fragment(s) containing the yeast expression control sequence, said DNA segment and said DMA sequence containing transcription termination signals. Prior to or after linking of said DNA fragment(s) (or simultaneously as well), it is also possible to introduce replication origins and/or markers for yeast or a bacterial host. At all events, the restriction and ligation conditions are to be chosen in such a manner that there is no interference with the essential functions of the vector DNA and of the expression control sequence. The hybrid vector may be built up sequentially or by ligating two DNA segments comprising all sequences of interest.

Various techniques may be used to join DNA segments in vitro. Blunt ends (fully base-paired DNA duplexes) produced by certain restriction endonucleases may be directly ligated with T4 DNA ligase. More usually, DNA segments are linked through their single-stranded cohesive ends and covalently closed by a DNA ligase, e.g. T4 DNA ligase. Such single-stranded cohesive termini may be formed by cleaving DNA with another class of endonucleases which produce staggered ends (the two strands of the DNA duplex are cleaved at different points at a distance of a few nucleotides). Single strands can also be formed by the addition of nucleotides to blunt ends or staggered ends using terminal transferase (homopolymeric tailing) or by simply chewing back one strand of a blunt-ended DNA segment with a suitable exonuclease, such as λ exonuclease. A further approach to the production of staggered ends consists in ligating to the blunt-ended DNA segment a chemically synthesized linker DNA which contains a recognition site for a staggered-end forming endonuclease and digesting the resulting DNA with the respective endonuclease.

The components of the hybrid vector according to the invention, such as the yeast promoter, structural gene for u-PA or a mutant thereof including a signal sequence, transcription terminator, the replication system etc., are linked together in a predetermined order to assure proper function. The components are linked through common restriction sites or by means of synthetic linker molecules to assure proper orientation and order of the components.

The transformed yeast strains according to the invention are made in a manner known per se, viz. transforming a yeast strain with a hybrid vector comprising a yeast expression control sequence, a DNA segment consisting of a first DNA sequence encoding a signal peptide upstream of and in reading frame with a second DNA sequence coding for mature urokinase-type plasminogen activator or a mutant thereof, which DNA segment is under transcriptional control of said expression control sequence, and a DNA sequence containing transcription termination signals of a yeast gene. ϊ! 6 Suitable yeast host organisms include species of the genera Kluyveromyces, Candida, Pichia, Saccharomyces, Yarrowia, Torulopsis and related genera (cf. J- Lodder, The Yeasts, Amsterdam 1971), especially strains of Saccharomyces cerevisiae.

The transformation of the yeast host cells is accomplished by methods known in the art. For example, the transformation may be accomplished according to the method described by Hinnen et al [Proc . Natl. Acad. Sci. USA 75, 1929(1978)]. This method can be divided into three steps: (1) Removal of the yeast cell wall or parts thereof. (2) Treatment of the naked’' yeast cells (spheroplasts) with the transforming DNA in the presence of PEG (polyethyleneglycol) and „ 2+ .

Ca ions. (3) Regeneration of the cell wall and selection of the transformed cells in a solid layer of agar.

Preferred methods: ad (1): The yeast cell wall is removed enzymatically using various preparations of glucosidases, such as snail gut juices (e.g. Glusulase® or Helicase®) or enzyme mixtures obtained from microorganisms (e.g. Zymolyase®) in osmotically stabilized solutions (e.g. 1 M sorbitol). ad (2): The yeast spheroplasts aggregate in the presence of PEG and local fusions of the cytoplasmic membranes are induced. The generation of fusion-like conditions is crucial and many transformed yeast cells become diploid or even triploid during the process of transformation. Procedures which allow selection of fused spheroplasts can be used to enrich for transformants, i.e. transformed cells can easily be screened for among preselected fusion products. ad (3): Since yeast cells without cell wall do not divide the cell wall has to be regenerated. This regeneration is conveniently done by embedding the spheroplasts into agar. For example, molten agar (about 50°C) is mixed with the spheroplasts. Upon cooling the solution to yeast growth temperatures (about 30°C), a solid layer is obtained. This agar layer is to prevent rapid diffusion and loss of essential macromolecules from the spheroplasts and thereby facili5 tates regeneration of the cell wall. However, cell wall regeneration may also be obtained (although at lower efficiency) by plating the spheroplasts onto the surface of preformed agar layers.

Preferably, the regeneration agar is prepared in a way to allow regeneration and selection of transformed cells at the same time. 0 Since yeast genes coding for enzymes of amino acid biosynthetic pathways are generally used as selective markers (supra), the regeneration is preferably performed in yeast minima] medium agar.

If very high efficiencies of regeneration are required the following two step procedure is advantageous: (1) regeneration of the cell wall in a rich complex medium, and (2) selection of the transformed cells by replica plating the cell layer onto selective agar plates.

If the hybrid vector does not contain any marker gene the transformed cells can also be identified by means of alternative methods. Such methods include, for example, in situ hybridization with a 2q labeled DNA fragment homologous to sequences of the hybrid vector [e.g. according to Hinnen et al., supra], in situ immunoassays provided that an antibody for the product of the introduced gene is available, or other screening methods which measure gene products encoded by the transforming plasmid(s).

Alternatively, the yeast can be co-transformed with a hybrid vector according to the invention and a second vector containing a genetic marker for yeast. If the two different vectors have DNA sequences in common (these can be bacterial sequences), recombination takes place leading to a fused selectable hybrid molecule.

The transformed yeast strains containing the hybrid plasmids according to the invention can be improved in production of scu-PA or mutants thereof by mutation and selection using methods known in Ε 8 the art. The mutation can be effected, for example, by U.V. irradiation or suitable chemical reagents. Especially preferred is the production of protease deficient mutants, especially yeast mutants, so as to avoid proteolytic degradation of the produced scu-PA or mutants thereof within the cells. Suitable mutants can be selected and isolated by conventional means.

The scu-PA proteins, obtainable according to the present invention, especially the novel scu-PA proteins, exhibit valuable pharmacological properties. They can be used in analogy to known plasminogen activators in humans for the prevention or treatment of thrombosis or other conditions where it is desired to produce local fibrinolytic or proteolytic activity via the mechanism of plasminogen activation, such as arteriosclerosis, myocardial and cerebral infarction, venous thrombosis, thromboembolism, post-surgical -,5 thrombosis, thrombophlebitis and diabetic vasculopathies.

The invention relates also to pharmaceutical compositions that contain a therapeutically effective amount of the active ingredient (scu-PA or a mutant thereof) together with organic or inorganic, solid or liquid pharmaceutically acceptable carriers that are 2q suitable for parenteral, such as intravenous, administration and that do not deleteriously interact with the active ingredients.

There are suitable especially infusion solutions, preferably aqueous solutions or suspensions, it being possible to prepare these before use, for example from lyophilised preparations that contain the active ingredient alone or together with a carrier, such as mannitol, lactose, glucose, albumin and the like. The pharmaceutical composition may be sterilized and, if desired, mixed with adjuncts, for example preservatives, stabilisers, emulsifiers, solubilisers, buffers and/or salts, such as sodium chloride, for regulating the 30 osmotic pressure. Sterilization can be achieved by sterile filtration through filters of small pore size (0.45 pm diameter or smaller) after which the preparation can be lyophilised, if desired. Antibiotics may also be added in order to assist in preserving i a sterility. For example, the scu-PA protein according to the invention is formulated into a sterilized aqeuous solution containing 5 % glucose and optionally stabilisers and salts.

The pharmaceutical compositions according to the present invention are dispensed in unit dosage forms, for example ampoules, comprising to 2000 mg of a pharmaceutically acceptable carrier per unit dosage and about 1 to 20 mg, preferably about 3 to 15 mg, of the active ingredient (scu-PA or a mutant thereof) per unit dosage.

Depending upon the type of the disease and the age and the condition of the patient, the daily dose to be administered for the treatment of a patient weighing approximately 70 kg is in the range from 20 to 150 mg, preferably from 45 to 100 mg, per 24 hours.

The invention also concerns a method for producing a pharmaceutical composition characterised in that a biologically active protein of the present invention is admixed with a pharmaceutically acceptable carrier.

The use of the new proteins for the prophylactic and therapeutic treatment of the human body is also an object of the present invention.

The invention concerns also the novel hybrid plasmids and the yeast strains transformed with said hybrid plasmids and the processes for the production thereof.

The invention relates especially to the scu-PA proteins, hybrid vectors, transformed yeast hosts and to the processes for the production thereof as described in the Examples.

Brief description of the drawings In the following experimental part various embodiments of the present invention are described with reference to the accompanying drawings in which: Fig. 1 is a restriction endonuclease map of human u-PA cDNA.

Fig. 2 illustrates the nucleotide sequence and deduced amino acid sequence of human u-PA cDNA. The first amino acid of the mature protein is underlined.

Fig. 3 schematically illustrates the construction of plasmid pCS16.

Fig. 4 schematically illustrates the construction of plasmid pCSl6/UPA comprising the u-PA cDNA.

Fig. 5 is a schematic diagram showing the construction of plasmid pJDB2O7/PH05-I-UPA (abbreviations used: SS signal sequence; TO t transcription terminator; p promoter; L linker).

Fig. 6 provides the DNA sequence of the promoter region of GAPDH.

Fig. 7 depicts schematically the construction of plasmid p31GAPDL-IT.

Fig. 8 depicts promoter elements of the GAPDH gene used in the 15 present invention.

Fig. 9 shows schematically the construction of plasmid pJDB2O7/GAPDL-I-UPA.

Fig. 10 is a schematic diagram depicting the construction of plasmid pJDB2O7/GAPDL-UPA. 2Q Fig. 11 and Fig. 12 show schematically the construction of plasmid pDP38 via plasmid pDP33.

Fig. 13 provides the DNA sequences of the mutated u-PA genes and the corresponding amino acid sequences of the mutant scu-PA proteins according to the invention. 1 The following Examples serve to illustrate the present invention but should not be construed as a limitation thereof.

Experimental Part Example 1: Construction of plasmid pCS16/UPA comprising the u-PA coding region A) Construction of plasmid pCS16 (see Fig. 3) A 1.5 kb Pstl-BamHI fragment of plasmid pUN121 [B= Nilsson et al. Nucl. Acids Res. 11, 8019-8030 (1983)] comprising the cl gene of phage lambda and part of the tetracyclin resistance gene is cloned 1q into pUC18 [J. Norrander et al., Gene 26, 101-106 (1983)], cut with Pstl and BamHI. The resulting clone is digested with Pstl. The 3' overhanging ends are removed in a reaction with Ti, DNA polymerase [T. Maniatis et al., Molecular Cloning (1982), p. 395] and Xhol linkers (51-CCTGGAGG-3’, Biolabs) are ligated to the blunt ends.

After digestion with Xhol the molecule is recircularised by ligation. An aliquot of the ligation mixture is used to transform Ca'' treated E.colj HB101 cells. The DNA of individual ampicillin resistant, transformed colonies is analysed. One of several correct clones is chosen and referred to as pCSlo.

Construction of plasmid pCS16/UPA (see Fig. 4) Urokinase cDNA is prepared from mRNA obtained from human Hep3 cells [cf. T. Maniatis et al., p. 188-246, supra]. A 1.3 kb Smal-BamHI fragment and a 1 kb BamHI-EcoRI fragment of the u-PA cDNA is cloned into the Sinai, EcoRI sites of pUN121 [B. Nilsson et al., Nucl. Acids Res. 11, 8019-8030 (1983)] to yield plasmid pcUK176. The restriction z □ — endonuclease map of the human u-PA cDNA insert is shown in Fig. 1. The nucleotide sequence and deduced amino acid sequence of the u-PA insert is given in Fig. 2. The u-PA cDNA insert is subcloned in plasmid pCS16. The subcloned cDNA extends from the Smal site in the 5' nontranslated region (Fig. 1) to the PvuII site at nucleotide positions 1439-1444 (Fig· 2) in the 3’ nontranslated region. pg of plasmid pcUK176 are digested with PvuII. The 379 bp PvuII fragment is separated from other fragments on a 1.5 % agarose gel in Tris-borate-EDTA buffer pH 8.3. The DNA is electroeluted, purified by DE52 (Whatman) ion exchange chromatography and precipitated by ethanol. 1.2 pg of single stranded Xhol linkers (5’-CCTCGAGG-3') are phosphorylated at their 5' ends, heated for 10 min at 75°C, self annealed during cooling to room temperature and stored at -20°C. 0.9 pg of the kinased, double stranded Xhol linkers are ligated at an 80-fold molar excess to the blunt ends of the 379 bp PvuII fragment of pcUKl76 (see above) in 20 pi of 60 mM Tris-HCl pH 7.5, mM MgCl2, 5 mM DTT, 3.5 mM ATP and 400 units of Ti, DNA ligase (Biolabs) at 15°C for 16 hours. The mixture is heated for 10 min at 85°C. Excess linker molecules are removed by precipitation with 0.54 volumes of isopropanol in the presence of 10 mM EDTA and 300 mM sodium acetate pH 6.0 for 30 min at room temperature. The DNA is digested with Xhol and BamHI. A 121 bp BamHI-XhoI fragment is isolated on a 1.5 X agarose gel in Tris-borate-EDTA buffer pH 8.3. pg of plasmid pcUK176 are digested with Smal and BamHI. A 1.3 kb Smal-BamHI fragment comprising most of the u-PA coding sequence is isolated. 6 pg of plasmid pCS16 are digested with Smal and Xhol. The 2.7 kb vector fragment is isolated. The DNA fragments are electroeluted from the gel and ethanol precipitated. 0.2 pmoles of the 1.3 kb Smal-BamHI fragment, 0.2 pmoles of the 121 bp BamHI-XhoI fragment (both fragments together comprise the full u-PA coding sequence) and 0.1 pmoles of the 2.7 kb vector fragment are ligated in 10 pi of 60 mM Tris.HCl pH 7.5, 10 mM MgCl2, 5 mM DTT, 3.5 mM ATP and 400 units of Tu DNA ligase at 15°C. One and 3 pi aliquots of the ligation mixture are added to 100 pi of Ca'' treated E.coli HB101 cells. Transformation is carried out as described [A. Hinnen et. al., Proc.Natl. Acad. Sci. USA 75, 1929 (1978)]. ampicillin resistant colonies are grown in LB medium containing 100 mg/1 ampicillin. DNA is isolated according to Holmes et al. 3 [Anal. Biochem. 114, 193 (1981)] and analysed by EcoRI, PvuII and Xhol restriction digests. Plasmid DNA of a single clone with the expected restriction fragments is referred to as pCSl6/UPA.

In an analogous manner the urokinase cDNA from the Smal site (nucleotide positions l-ό, Fig. 2) to the Pstl site at nucleotide positions 1637-1642 (see Fig. 2) is subcloned in plasmid pCS16. Plasmid pcUKl76 is digested with Pstl. The sticky ends are converted to blunt ends in a reaction with T4 DNA polymerase as described by Maniatis et al. (supra, p. 395). The 1.2 kb DNA fragment is isolated. Xhol linkers are added as described above. The DNA is digested with BamHI and Xhol and a 315 bp BamHT-XhoI fragment is isolated. 0.2 pmoles of this fragment are ligated to the 1.3 kb Smal-BamHI fragment and the 2.7 kb vector fragment (see above). The ligation mixture is used to transform E. coli HB101 Ca‘‘ cells. Plasmid DNA of a single clone with the expected restriction fragments is referred to as pCSl6/UPA-l3. This plasmid contains the u-PA cDNA insert from the Smal site in the 5s nontranslated region (Fig. 1) to the Pstl site at nucleotide position 1641 (Fig. 2) in the 3? nontranslated region. The only difference to plasmid pCS16/UPA is the extended 3s nontranslated region.

Example 2‘ Construction of plasmid p31RIT-12 containing the PHO5 promoter and the invertase signal sequence A) Synthesis of oligodeoxyribonucleotides for invertase signal sequence; Four oligodeoxyribonuclotides: I-l, Ϊ-2, Ϊ-3, 1-4 are synthesized by DNA synthesizer (model 38OB Applied Biosystems). After deblocking the synthetic fragments are purified on a 12 % polyacrylamide gel containing 8 M urea. Salt-free pure oligodeoxyribonucleotides are obtained using Sep. Pak (Waters Associates). These fragments constitute a duplex which encodes the invertase signal sequence with the frequently used yeast codons. 4 Hindlll EcoRI MetLeuLeuGlnAlaPheLeuPheLeuLeu 1-1 5'AATTCATGCTTTTGCAAGCTTTCCTTTTCCTTTT 3' 1-2 3'GTACGAAAACGTTCGAAAGGAAAAGGAAAACCGAC 5' AlaGlyPheAlaAlaLysIleSerAla 1-3 5'GGCTGGTTTTGCAGCCAAAATATCTGCATCTTAGCGTC 3' 1-4 3'CAAAACGTCGGTTTTATAGACGTAGAATCGCAGAGCT 5’ Hgal_ Xhol B) Subcloning of the invertase signal sequence in plasmid p31 a) Preparation of vector: 1.5 pg of p31R/SS-7PAA2 (European Patent Application No. 143,081) is digested with 10 U of EcoRI (Boehringer) in 50 pi of 10 mM Tris.HCl pH 7.5, 6 mM MgCli, 100 mM NaCl, 6 mM mercaptoethanol for one hour at 37°C. After adding 1 pi of 2.5 M NaCl, 10 U of Xhol (Boehringer) are added and incubated at 37°C for one hour. The 4.2 kb vector is isolated on a 0.8 % preparative agarose gel. The gel slice is transferred to a Micro Colloidor tube (Sartorius GmbH), covered with 200 pi of TE and electroeluted (electrophoresed at 90 mA for 50 min). The TE solution is collected and precipitated in 2.5 volumes of absolute ethanol after the addition of 0,1 volume 10 x TNE. The DNA pellet is washed with cold 80 % ethanol and dried in vacuum. The DNA is resuspended in 6 pi TE (40 pmoles/pl). b) Annealing oligodeoxyribonucleotides (I-1, 1-2, 1-3, 1-4), kination and ligation with vector A solution containing 10 pmoles of each of the four deoxyribonucleotides in 10 pi of 0.5 M Tris.HCl pH 8 is incubated at 95°C for 5 minutes on a water bath. The water bath is slowly cooled to 30°C over a period of 5 hours. To this annealed mixture is added 2 pi each of 0.1 M MgClz, 0.1 M NaCl, 30 mM DTT, 4 mM ATP and 8 U (1 pi) of polynucleotide kinase (Boehringer)» Kination is carried out at 37°C for one hour. The annealed, kinased oligodeoxyribonucleotides and 60 pmoles of p3lR/SS-TPAA2 EcoRI-Xhol cut vector (1.5 pi) are ligated with 400 U (1 μΐ) of T4 DNA ligase (Biolabs) at 14°C for 17 hours. The reaction is stopped by incubation at 65°C for 10 min. 10 μΐ of this ligation mixture is used for transformation of E.coli HB101 Ca-' cells [M. Dagert and S.D. Ehrlich, Gene 56, 23-28 (1979)]. 20 amp colonies are picked. DNA is prepared by the quick isolation procedure [D.S. Holmes and M. Quigley, Anal. Biochem. 114, 193-197 (1981)]. DNA is digested with EcoRI and Xhol, radiolabelled at the EcoRI end and analysed on a 6 % polyacryalmide gel containing 8 M urea using radiolabelled pBR322 Haelll cut DNA as marker. Correct size bands are observed for DNA obtained from all the 20 clones. One clone is grown in 100 ml LB medium containing 100 pg/rol of ampicillin. Plasmid DNA is isolated and is referred to as p31RIT-12.

Example 3: Construction of plasmid pJDB207/PHO5-I-UPA (Fig. 5) pJDB207/PH05-I-UPA contains the PHO5 promoter, the invertase signal sequence, the coding sequence of mature urokinase and the PHO5 transcription terminator in a tandem array cloned into the pJDB207 yeast expression vector. pg of plasmid pCS16/UPA are digested to completion with 40 units of EcoRI. After phenol extraction and ethanol precipitation the EcoRI digested DNA is further cut by TaqI at 65°C. The resulting fragments are separated on a preparative 1.2 % agarose gel. The 462 bp TaqI-EcoRI fragment is isolated by electroelution from the gel and ethanol precipitation.

An oligodesoxyribonucleotide linker of the formula (I) 5’-CTGCAAGCAATGAACTTCATCAAGTTCCAT-3’ (II) 3’- TCGTTACTTGAAGTAGTTCAAGGTAGC-5’ is ligated to the TaqI site of the DNA fragment. The linker restores the 5’ terminus of the coding sequence of mature u-PA (nucleotides 130-154, Fig. 2) and establishes the in frame fusion to the 6 invertase signal sequence. The 5'-CTGCA sequence of the linker fills the corresponding 3' recessed end of the invertase signal sequence created by Hgal cleavage. 300 pmoles each of the oligodesoxynucleotides I and IT are phosphorylated and annealed. 5.25 pg (600 pinoles) of phosphorylated, double-stranded linker DNA are ligated to 1.7 pg (5.6 pinoles) of the 462 bp TaqI-EcoRI fragment (see above) in 175 pi of 60 mM Tris-HCl pH 7.5, 10 mM MgCl2, 1 mM ATP, 5 mM DTT and 800 units of TA DNA ligase at 15°C for 16 hours. T4 DNA ligase is inactivated for 10 min at 85°C. The excess of linkers is removed by precipitation in the presence of 10 mM EDTA, 300 mM sodium acetate pH 6.0 and 0.54 volumes of isopropanol. The DNA is digested with Pstl. An unique 312 bp fragment is isolated containing the linker attached to DNA sequences coding for u-PA up to nucleotide 436 (Pstl site, see Fig. 2). The DNA fragment is purified by electroelution and precipitation with ethanol.

Plasmid pCS16/UPA is digested with Xhol and Pstl. A 1007 bp Pstl-Xhol fragment is isolated and purified. This fragment contains most of the coding sequence for urokinase.

Plasmid p31RIT-12 (see Example 2) is digested with Sail and Xhol. An 882 bp Sall-Xhol fragment is isolated from the gel by electroelution and ethanol precipitation. The fragment is further digested with BamHI and Hgal. A 591 bp BamHI-Hgal fragment is isolated which contains the PHO5 promoter region and the invertase signal sequence.

Plasmid pJDB207/PH05-TPA 18 (see European Patent Application No. 143,081) is digested with BamHI and Xhol. The 6.8 kb vector fragment is isolated on a preparative 0.6 % agarose gel in Trisacetate buffer pH 8.2, The DNA is electroeluted and precipitated with ethanol. 7 All DNA fragments are resuspended in H2O at a concentration of 0.1 pmoles/μΐ. 0.2 pmoles of the 591 bp BamHI-Hgal fragment. 0.2 pmoles of the 312 bp Hgal-Pstl fragment. 0.2 pmoles of the 1007 bp Pstl-Xhol fragment and 0.1 pmoles of the 6.8 kb BamHI-XhoI vector fragment are ligated for 15 h at 15°C in 10 μΐ of 50 mM Tris-HCl pH 7.5, 10 mM MgCl2, 5 mH DTT, 1 mM ATP and 400 units of T4 DNA ligase. One μΐ of the ligation mixture is used to transform E. coli HB101 Ca' cells. 12 amp colonies are picked and grown in LB medium containing 100 rag/I of ampicillin. DNA is prepared by the quick isolation procedure [D.S. Holmes et al., Anal. Biochem. 114, 193 (1981)]. On restriction digests of the plasmid DNA with Hindlll and EcoRI the expected restriction fragments are observed. Plasmid DNA of a single clone is selected and referred to as pJDB207/PH05-I~UPA, Example 4: Construction of plasmid pJD3207/PH05-UPA This construction comprises the PHO5 promoter, the PH05 signal sequence joined in frame to the coding sequence of mature urokinase and the PHO5 transcription terminator cloned Into the pJDB2O7 yeast vector.

An oligodesoxyribonucleotide linker of the formula (I) 5’-CCAATGCAAGCAATGAACTTCATCAAGTTCCAT-3' (II) 3'-GGTTACGTTCGTTACTTGAAGTAGTTCAAGGTAGC-5' provides 8 nucleotides (5’-CCAATGCA) of the PHO5 signal sequence (from an internal Ball site to its processing site) and establishes an in frame fusion to the coding sequence of mature u-PA (nucleotide positions 130-154. rig. 2). The linker is ligated to the TaqI site of the 462 bp TaqI-EcoRI fragment of pCS16/UPA (see Example 3). Phosphorylation, annealing and ligation is done according to Example 3, The DNA is digested with Ball and Pstl. A 315 bp fragment is isolated. The DNA fragment contains the DNA sequence of the u-PA gene up to the Pstl site at position 436 (Fig. 2).

Plasmid p31 (see European Patent Application No. 100,561) is digested with Ball and BamHI. A 584 bp BamHI-Ball fragment is isolated containing the PHO5 promoter and most of the PHO5 signal sequence. The DNA fragments are purified by electroelution and DE52 chromatography, precipitated with ethanol and resuspended in H2O at a concentration of 0.1 pmoles/pl. 0.2 pmoles of the 584 bp BamHI-Ball fragment, 0.2 pmoles of the 315 bp Ball-Pstl fragment, 0.2 pmoles of the 1007 bp Pstl-Xhol fragment (see Example 3) and 0.1 pmoles of the 6.8 kb BamHI-XhoI vector fragment (see Example 3) are ligated and used to transform R E. coll HB101. 6 amp colonies are picked and grown in LB medium containing 100 mg/1 of ampicillin. Plasmid DNA is prepared by the quick DNA procedure and analysed by BamHI/Pstl double digestion. A single done with the expected restriction fragments is selected and the plasmid DNA is referred to as pJDB207/PH05-UPA.

Example 5: Construction of plasmid pJDB207/PH05~SSUPA This plasmid contains the urokinase gene with its own signal sequence under the control of the PH05 promoter in the yeast expression vector pJDB2O7. pg of plasmid pCS16/UPA-13 (see Example 1) are digested to completion with 100 units of Bgll (nucleotide positions 76-86, Fig. 2). The resulting 3 fragments are separated on a preparative 1 % agarose gel in Tris-borate-EDTA buffer pH 8.3. The 1.7 kb Bgll fragment is electroeluted and precipitated with ethanol.

An oligodesoxyrihonucleotide linker of the formula (X) 5'-AATTCGATTACCAATGAGAGCCCTGC-3s (II) 3'- GCTAATGGTTACTCTCGGG -5’ fi has an EcoRI site linked to 8 nucleotides of the PHO5 5' noncoding region in front of its ATG and nucleotides 70-82 (Fig. 2) of the coding sequence of u-PA including the ATG. The linker is ligated to the Bgll sticky ends of DNA as described in Example 3. The DNA is digested with EcoRI and Pstl.

A 380 bp fragment is isolated comprising the u-PA signal sequence and part of the coding sequence for mature u-PA up to nucleotide 436 (Pstl site, Fig. 2).

Plasmid p31R (see European Patent Application No. 100,561) is digested with BamHI and EcoRI and the 534 bp BamHI-EcoRI fragment is isolated comprising the PHO5 promoter. DNA fragments are electroeluted from the agarose gel, purified by DE52 chromatography, precipitated with ethanol and resuspended in H2O at a concentration of 0.1 pmoles/μΐ. 0.2 pmoles of the 534 bp BamHI-EcoRI fragment, 0.2 pmoles of the 380 bp EcoRI-Pstl fragment, 0.2 pmoles of the 1007 bp Pstl-Xhol fragment (see Example 3) and 0.1 pmoles of the 6.8 kb BamHI-XhoI vector fragment (see Example 3) are ligated and used to transform ++ R E. coli HB101 Ca cells. 12 amp colonies are grown separately in LB medium containing 100 mg/1 of ampicillin. DNA is prepared by the quick DNA procedure and analysed by BamHI and EcoRI. Plasmid DMA from a single clone is selected and referred to as PJD3207/PH05-SSUPA.

Example 6; Construction of plasmid pJDB2O7R/PHO5-UPA The coding sequence of mature urokinase is linked to the PHO5 promoter. No signal sequence is included in this construct which has been made for comparative purposes.

An oligodesoxyribonucleotide linker of the formula (I) 5'-AATTCATGAGCAATGAACTTCATCAAGTTCCAT -3' (II) 3'- GTACTCGTTACTTGAAGTAGTTCAAGGTAGC-5’ contains an ATG and nucleotides 130-154 of u-PA (see Fig. 1) coding for amino acids Seri to Ser9 of mature urokinase. The oligonucleotides are phosphorylated, annealed and ligated to the 462 bp TaqI-EcoRI fragment of pCS16/UPA (see Example 3). The DNA is digested with EcoRI and Pstl. A 315 bp EcoRI-Pstl fragment is isolated comprising the ATG and the coding sequence for mature u-PA up to the Pstl site at position 436 (Fig. 2).

Plasmid p3lR (see European Patent Application No. 100,561) is digested with BamHI and EcoRI and the 534 bp BamHI-EcoRI fragment is isolated on a preparative 1.5 % agarose gel. This fragment contains the PHO5 promoter.

All DNA fragments are electroeluted, purified by DE52 chromatography, precipitated with ethanol and resuspended in H2O at a concentration of 0.1 pmoles/μΐ. 0.2 pmoles each of the 534 bp BamHI-EcoRI fragment, the 315 bp EcoRI-Pstl fragment and the 1007 bp Pstl-Xhol fragment (see Example 3) and 0.1 pmoles of the 6.8 kb BamHI-XhoI vector fragment (see Example 3) are ligated and trans4,4» formed into E. coli HB101 Ca as usual. 8 amp colonies are picked and grown in LB medium containing 100 mg/1 of ampicillin.

Plasmid DNA is prepared and analysed by BamHI and EcoRI restriction digests. Plasmid DNA of a single clone with the expected restriction pattern is referred to as pj'DB2O7R/PHO5-UPA.

Example 7: Cloning of the yeast GAPDH gene with its constitutive promoter a) Construction of a yeast gene library Thirty pg of total high molecular weight yeast DNA [M.V.Olsen et al. J. Mol.Biol. 132, 387 (1979)] from wild type Saccharomyces cerevisiae strain S288C is incubated for 30 min at 37°C with 2 units of EcoRI methylase (New England Biolabs) in 250 μΐ of EcoRI methylation buffer as recommended by the supplier. DNA is precipitated by ethanol, resuspended in 500 μΐ of 25 mM Tris-HCl pH 8.5, mM MgClz (EcoRI* buffer) [H.Meyer, FEBS Lett. 90, 341 (1979)] and digested with EcoRI (Boehringer) until the size distribution of the DNA fragments has a maximum in the 30-50 kb range (a Xhol digest of XDNA provides appropriate 33 kb and 17 kb markers). The yeast DNA digested under EcoRI* conditions is size-fractionated on a sucrose gradient (5-20 % sucrose in 10 mM Tris-HCl pH 7.5, 1 mM JEDTA) for 6 hrs at 38'000 rpm in a SW 40 rotor. Thirty fractions of 0.4 ml each are collected from the top of the gradient. Fraction 16 contains DNA fragments of 30-40 kb in size. The DNA of this fraction (3 pg) is precipitated with ethanol and ligated for 16 hours at 15°C in a total volume of 15 μΐ to 1 pg of cosmid vector pYcl [B.Hohn et al. in Genetic Engineering, Vol. 2, p. 169, New York 1980] linearized by EcoRI. Ligation is carried out with 300 U T4 DNA ligase (New England Biolabs) using the buffer system described by the supplier. The DNA is packaged in vitro into bacteriophage X (B.Hohn in Methods in Enzymology, Vol. 68, p. 299, New York 1979] and the assembled phages are used to transduce E.coli strain HB101 Θ Θ Θ Θ (r^, —lc* ^eu s pro , recA). The efficiency of transduction is about 5000 ampicillin-resistant colonies per pg of pYcl vector.

R 3000 amp colonies are picked and grown individually in the wells of microtiter dishes in LB medium [10 g Bacto-Tryptone (Difco), 5 g Bacto Yeast Extract (Difco), 10 g NaCl] containing 100 pg/ml ampicillin. b) Isolation of the yeast GAPDH gene The gene library described above is screened with a synthetic oligonucleotide [prepared using the phosphotriester method: K. Itakura et. al., J. Am. Chem. Soc. 97 , 7327 (1975); J.F.M. de Rooij et al. , Reel. Trav. Chim. Pays-Bas 98, 537 (1979)] of the following structure: 5'-GCTCCATCTTCCACCGCCCC-31. 10 pg of the oligonuclotide are kinased using 10 pi of y-32P-ATP (3000 Ci/mmol, pCi/pl Amersham) with T4 polynucleotide kinase (Boehringer) in a total volume of 50 pi as described by Maniatis et al. [Molecular Cloning, Cold Spring Harbor Lab., 1982, page 125]. Colony hybridization is performed as described by the same authors (page 312). Positive clones are detected by autoradiography using Kodak X-5 X-ray film. Plasmid DNA isolation (see European Patent Application Nr. 100,561) produces a hybrid clone which contains a 2100 bp Hindlll fragment coding for GAPDH [J.P. Holland et al., J. Biol. Chem. 254, 9839 (1979)]. Final proof for the authenticity of the cloned DNA comes from DNA sequencing experiment using the above mentioned oligonucleotide in combination with the dideoxy sequencing protocol as described by G.F. Hong [Bioscience Reports _1, 243 (1981)] for double stranded DNA. The cloned GAPDH gene has the same sequence as pgap491 published by Holland et al. [J. Biol. Chem. 255, 2596 (1980)]. c) Preparation of the GAPDH promoter The 649 bp TaqI fragment which includes position -27 to -675 from the ATG of the GAPDH gene (see Figure 6) is isolated by digesting the above mentioned hybrid plasmid with TaqI (New England Biolabs), separating the DNA fragments on a 1.2 % soft agarose gel and extracting the DNA by hot phenol. Cloning of the TaqI fragment is done into the Clal site of pBR322: 1 pg of pBR322 is cleaved with three units of Clal (New England Biolabs) as described by the supplier. 300 ng of the phenolized and cut vector is ligated to about 300 ng of insert DNA (649 bp Taq fragment) using 200 U of T4 DNA ligase in a total volume of 20 pi. Transformation is done into E.coli HB101 for ampicillin resistance, plasmid DNA is prepared and analyzed by restriction analysis [TaqI, Dral]. The orientation of 0l>) J the TaqI fragment is established using restriction endonuclease Dral in combination with the BamHI site of the plasmids and a plasmid is selected which has the TaqI site of position -675 close to the Hindlll site of pBR322. This plasmid designated pBR322/GAPDH is linearized using BamHI (New England Biolabs). The DNA is resuspended in 10 mM Tris pH 8.0 at a concentration of 0.5 pg/ml. 16 pg of Sail cleaved DNA are digested with 2 U of exonuclease Bal31 (BRL) in 100 pi of 20 mM Tris pH 8.0, 199 mM NaCl, 12 mM MgCl2s 12 mM CaCl2 and 1 mM EDTA. Aliquots of 2 pg DNA each are withdrawn after 1, 2, 3, 4, 5 and 6 min. of incubation at 30°C and are immediately mixed with 50 pi phenol and 60 pi TNE. After extraction with phenol/ chloroform and ethanol precipitation, the DNA is resuspended in 10 mM Tris pH 8.0 at a concentration of 100 pg/ml. To analyse the extent of exonucleolytic cleavage by Bal31 0.5 pg of DNA from each time point are digested with endonuclease BamHI and analysed on a 1.5 % agarose gel in Tris-borate buffer pH 8.3 (90 saM Tris-HCl pH 8.3, 90 mM boric acid, 2.5 mM EDTA). On the average 100 bp are removed from each end of the fragment per 1 min. of Bal31 digestion.

Bglll linkers (5'-GGAGATCTCC-3‘) are phosphorylated, annealed and ligated to the Bal31 treated DNA fragments. Excess linkers are removed by precipitation with isopropanol. The DNA is digested with Bglll and directly circularized at a concentration of 5 pg/ml in a total volume of 20 pi. The size of the Bal31 shortened TaqI fragment is determined by restriction analysis (using Bglll and Hindlll). Three clones are selected. They contain DNA fragments that extend about 200 bp, 265 bp and 540 bp, respectively, from the ATG upstream into the GAPDH promoter. All three fragments contain the presumptive TATA box at about -140 bp. These clones still contain the origin of replication in the pBR322 derived part of the DNA and are named pGAPDH-F, pGAPDH-E and pGAPDH-D, respectively.

In order to extend the GAPDH promoter elements from the TaqI site at position -27 to a position immediately adjacent to the ATG of the GAPDH gene two synthetic complementary oligonucleotides of the following structure are synthesized: ' CGAATAAACACACATAAATAAAG 3' * TTATTTGTGTGTATTTATTTCTTAA 5' These oligonucleotides provide the genuine GAPDH promoter sequence from position -26 to position -5 with the generation of a terminal EcoRI site. Two pg of plasmids pGAPDH-F, pGAPDH-E and pGAPDH-D are each digested with 6 units of TaqI in 50 pi and the resulting mixture is phenolized, ethanol precipitated and resuspended in 10 pi of water. The synthetic oligonucleotides are annealed by mixing 2 pi of each single strand in 100 pi of a solution containing lOmM Tris-HCl pH 7.5, 10 mM MgClj, 50 mM NaCI, heating for 3 min. to 90°C and slowly cooling the solution to room temperature (within about 3 hours). One pg of the TaqI digested plasmid is mixed with about a twenty fold molar excess of the annealed oligonucleotides in a volume of 20 pi and ligated for about 18 hours using 800 U of T4 DNA ligase. After inactivation of the ligase, the whole mixture is digested with 3 units of Bglll (New England Biolabs). The Bglll-EcoRI fragments of about 200 bp, 265 bp and 540 bp, respectively, are separated on a 1.5 % soft agarose gel, extracted from the gel and ethanol precipitated.

Plasmid p31RIT-12 (see Example 2) is digested with BamHI and EcoRI. The 3.6 kb large vector fragment is isolated. This fragment is used to clone the 200 bp, 265 bp and 540 bp Bglll-EcoRI GAPDH-promoter fragments. Ligation, transformation and plasmid isolation conditions are as described above. Plasmids with the correct insert are referred to as p3lGAPFL-IT, p31GAPEI.-IT and p31GAPDL-IT, respectively (Fig. 7).

The DNA sequences of the Bglll-EcoRI fragments of plasmids p31GAPFL-IT, p31GAPEL-IT and p31GAPDL-IT are shown in Fig. 8. The exact size of these fragments is 202 bp, 267 bp and 544 bp, respectively.

Example 8: Construction of plasmid pJDB207/GAPPL-I-UPA (Fig. 9) This plasmid contains the GAPDH-D promoter, the invertase signal sequence, the coding sequence of mature urokinase and the PHO5 transcription terminator in a tandem array in shuttle vector pJDB2O7.

Plasmid DNA of p31GAPDL-IT is digested with Sail and Hindlll. A 0.8 kb Sall-Hindlll fragment is separated on a preparative 1 % agarose gel. This fragment contains the GAPDH-D promoter and part of the invertase signal sequence. 1q Plasmid pJDB2O7/PH05-I-UPA is digested with Hindlll and BamHI. The 1239 bp Hindlll-BamHI fragment comprises the remaining part of the invertase signal sequence and most of the u-PA coding sequence. pJDB2O7/PH05-I-UPA is also digested with Sail and BamHI. The large, 6.6 kb vector fragment is isolated on a preparative 0.6 % agarose -,5 gel in tris-acetate buffer pH 8.2.

The DNA fragments are isolated by electroelution from the gel, purified by DE52 chromatography and precipitated with ethanol. The 0.8 kb Sall-HindllT fragment, the 1239 bp Hindlll-BamHI fragment and the 6.6 kb Sall-BamHI vector fragment are ligated and used to 4.4. ft 2q transform E. coli HB101 Ca-' cells. Plasmid DNA from 6 amp transformants is analysed by Hindlll and Sail restriction digests.

Plasmid DNA from a single clone with the expected restriction fragments is referred to as pJDB2O7/GAPDL-I-UPA.

In an analogous construction a 493 bp Sall-Hindlll fragment of plasmid p3lGAPFL-IT (see Example 7) is isolated and used for the ligation. The resulting plasmid DNA is referred to as PJDB2O7/GAPFL-I-UPA.

In an analogous manner pJD3207/GA?EI,~I-UPA is constructed. β Example 9: Construction of plasmid pJDB2O7/GAPDL-UPA (Fig. 10) A tandem array comprising the GAPDH-D promoter, the PH05 signal sequence, the urokinase coding sequence and the PH05 transcription terminator is cloned into yeast shuttle vector pJDB2O7. pg of plasmid pJDB207/PH05~UPA are digested with BgllT and Sail. The resulting two fragments are separated on a preparative 0.8 % agarose gel in tris-acetate buffer pH 8.2. The 7.6 kb and 1.1 kb Bglll-Sall fragments are isolated. The 1.1 kb fragment is further digested with Dral. After phenol/ehloroform extraction and ethanol precipitation 3.5 pmoles of the DNA are ligated with a 100-fold excess of a kinased and annealed oligodesoxyribonucleotide linker of the formula: '-AATTCGATTACCAATGTTT-3' 3' - GCTAATGGTTACAAA_5’ with an EcoRI site and 8 nucleotides of the PHO5 5’ non-coding region in front of the ATG and part of the Dral recognition sequence.

After ligation for 16 h at 15°C the ligase is inactivated, excess linkers are removed by isopropanol precipitation (0.54 volumes) in the presence of 300 mM sodium acetate pH 6.0 and 10 mM EDTA, The DNA is digested with Bglll and EcoRI. A 326 bp EcoRI-Bglll fragment is isolated.

Plasmid p31GAPDL-IT is digested with Sail, and EcoRI. A 0.75 kb Sall-EcoRI fragment is isolated.

DNA fragments are isolated by electorelution from the agarose gel, purified by DE52 chromatography and precipitated with ethanol. 0.2 pmoles of the 0.75 kb Sall-EcoRI fragment, 0.4 pmoles of the 326 bp EcoRI-Bglll fragment and 0.1 pmoles of the 7.6 kb Sall-Bglll vector fragment ar ligated for 5.5 h at 15°C. 1 pi of the ligation mixture is used to transform E. coli HB101 Ca'' cells. amp transformants are grown and plasmid DNA is prepared. The Plasmid DNA is analysed by EcoRI restriction digests. Plasmid DNA from a single clone with the expected restriction pattern is referred to as pJDB2O7/GAPDL-UPA.

In an analogous manner, plasmids pJDB207/GAPFL=UPA and pJDB207/GAPEJ.-UPA are constructed.

Example 10: Construction of plasmid pDP38 (Fig. 11 and Fig. 12) micron covalently closed circle DNA is isolated from Saccharomyces cerevisiae strain S288C by digesting the cell wall with 5 pg/ml Zymolyase 100,000 units/pg for 20 min at 37°C, then lysing the cells with 2 % SDS. EDTA is then added to 25 mM, caesium chloride to a final density of 1.55 g/ml, ethidium bromide to 1 mg/ml, and then the whole transferred to an ultracentrifuge tube. Plasmid DNA is separated from the chromosomal DNA by ultra-centrifugation for 42 hours at 42,000 rpm at 15°C. The 2 micron plasmid DNA is cut from the gradient with a syringe. The ethidium bromide is removed with NaCl saturated isopropanol and the plasmid DNA is finally ethanol precipitated. The purified plasmid DNA is then linearised with Pstl and cloned into the Pstl site of pUCl9 [j. Norrander et al., Gene 26, 101 (1983)] to give plasmid pDP31. Plasmid pJDB2O7 is digested with the enzymes Kpnl and Hpal. The resulting 0.55 kb fragment is purified and cloned into the 4.25 kb KpnI-Hpai fragment of plasmid pUC7/LF.U2 [a plasmid containing the yeast genomic 2.2 kb Xhol-Sal! J.EU2 gene [A. Andreadis et al. Cell 31, 319 (1982)] cloned into the Sail site of the plasmid pUC7 [J. Vieira et al. Gene 19, 259 (1982)]]. This results in plasmid pDP30 where the original 2 micron/LEU2 fusion as in plasmid pJDB2O7 is placed in front of the LEU2 gene plus its complete terminator. pDP30 is digested with Hpal and Sail and the 1.85 kb fragment containing the complete I.EU2 gene is purified and cloned into the 8.7 kb Hpal-Sall fragment of plasmid pDP31. The resulting plasmid, pDP33, is linearised with Hindi!! in the presence of 50 pg/ml ethidium bromide [M. Oesterlund et al. Gene 20, 121 (1982)] and ligated with the 1.17 kb Hindlll fragment containing the URA3 gene fM. Rose et al. Gene 29, 113 (1984)]. Positive insertion of the URA3 gene is selected for by transformation into the E. coli strain pyrf [M. Rose et al., supra]. This gives plasmid pDP34. pDP34 is digested with the enzyme Sphl. The resulting 8.4 kb fragment is purified and self ligated to give plasmid pDP38.

Example 11: Construction of plasmids pDP38/GAPDL-I-UPA and pDP38/GAPDL-UPA Vector pDP38 (see Example 10) contains the LEU2 and URA3 genes, pBR322 sequences and part of the yeast 2 micron DMA. Plasmid pDP38 is linearized with BamHI. The resulting sticky ends are filled in a reaction with Klenow DMA polymerase (Maniatis et al., p. 113, supra). The DNA is further digested to completion with Sail and the large 8.4 kb fragment is isolated. pg of plasmid pJDB2O7/GAPDL-I-UPA are partially digested with Hindlll (1 unit/pg DNA) in the presence of 10 pg/ml of ethidium bromide for 28 min at 37°C. The reaction is stopped by the addition of EDTA to a final concentration of 10 mM. The sticky ends of the DNA are filled in a reaction with Klenow DNA polymerase. The DNA is further digested with Sail. A 2.2 kb SalI-[HindIII]/blunt end fragment is isolated.

The purified DNA fragments are ligated and transformed into E. coli HB101 Ca cells. 24 amp colonies are grown up. Plasmid DNA is prepared and analysed by EcoRI and Sail/Hindlll restriction digests. Plasmid DNA of a single clone with the expected restriction fragments is referred to as pDP38/GAPDL-I-UPA.

In an analogous construction a 2.2 kb SalI-[HindIIJ]/blunt end fragment is isolated from pJDB2O7/GAPDL-UPA after complete Hindlll digestion, Klenow DNA polymerase reaction and Sail digestion. This fragment is ligated into pDP38 as described above. A single clone is referred to as pDP38/GAPDL-UPA.

In an analogous way the 2.2 kb SalI-[HindjII]/blunt end fragments are isolated from pJDB2O7/GAPEL-I-UPA, pJD3207/GAPFL-I-UPA, pJDB2O7/GAPEI.-UPA and pJDB2O7/GAPFL-UPA. The purified fragments are ligated into pDP38. The ligation mixture is used to transform E. coli HB101.

Plasmid DNA of a single clone each is referred to as pDP38/GAPEL-I-UPA, pD?38/GA?FL-I-UPAs pDP38/GAPEL~UPA and pDP38/GAPFL-UPA.

Example 12: Transformation of S. cerevisiae strains HT246 and GRF18 Saccharomyces cerevisiae strain HT246 (a, leu2-3, leu2-112, prb) is transformed with plasmids pJDB2O7/PHO5-UPA pJD3207/PH05-I~UPA pJDB2O7IPH05-SSUPA PJD3207R/PH0S-UPA pJDB207/GAPDL-UPA pJDB207/GAPFL-UPA pJDB2O7/GAPDL-I-UPA using the transformation protocol described by Hinnen et al. [Proc. Natl. Acad. Sci. USA 75, 1929 (1978)]. Transformed yeast cells are selected on yeast minimal media plates deficient in leucine. Single transformed yeast colonies are isolated and referred to as Saccharomyces cerevisiae HT2 46/pjDB207/PH05-UPA Saccharomyces cerevisiae HT246/pJDB2O7/PH05-I-UPA Saccharomyces cerevisiae HT246/pJDB2O7/PHO5-SSUPA Saccharomyces cerevisiae hT246/pJDB207R/PH05-UPA Saccharomyces cerevisiae HT246/pJDB2O7/GAPDL-UPA Saccharomyces cerevisiae HT246/pJD3207/GAPFL-UPA Saccharomyces cerevisiae HT2 4 6/pJD32071GAPDL-1-UFA In an analogous manner, Saccharomyces cerevisiae strain. GRF18 (DSM 3665) is transformed with the above mentioned plasmids. The resulting transformed yeast strains are designated Saccharomyces cerevisiae GRF18/pJDB207/PH05-UPA Saccharomyces cerevisiae GRF18/pJDB2O7/PH05-I-UPA Saccharomyces cerevisiae GRF18/pJDB2O7/PH05-SSUPA Saccharomyces cerevisiae GRF181pJDB207 R /PH05-UPA Saccharomyces cerevisiae GRF 18/pJ.DB207/GAPDL-UPA Saccharomyces cerevisiae GRF18/pJDB2O7/GAPFL-UPA Saccharomyces cerevisiae GRF18/pJDB2O7/GAPDL-I-UPA.

Example 13: Transformation of Saccharomyces cerevisiae strain HT350 S. cerevisiae strain HT350 is obtained by crossing strain S. cerevisiae HT246 with a ura-3 deficient strain such as S. cerevisiae HT285 (a, his3~ll, his3-15, leu2-3, leu2-l12, ura3, pep4-3) which results in the following genotype for strain HT350 (a, his3-l1, his3-15, leu2-3. leu2°I12, ura38 prb, pep4-3). Strain HT350 is transformed with the plasmids pDP38/GAPDL-I-UPA, pDP38/GAPFL-I-UPA, pDP38/GAPEL-I-UPA, pDP38/GAPDI.-UPA, pDP38/GAPFL-UPA, and pDP38/GAPEL-UPA, respectively, using the transformation protocol by Hinnen et al. (supra). Transformed yeast cells are selected on yeast minimal media plates deficient in uracil and supplemented with leucine. Single transformed yeast colonies are isolated and referred to as Saccharomyces cerevisiae Saccharomyces cerevisiae Saccharomyces cerevisiae Saccharomyces cerevisiae Saccharomyces cerevisiae Saccharomyces cerevisiae Example 14; Fermentation Saccharomyces cerevisiae Saccharomyces cerevisiae Saccharomyces cerevisiae HT350/pDP38/GAPDL-I-UPA HT 3 50/pD?38/GAPFL-1-UPA HT350/pDP38/GAPEL-I-UPA HT 3 50/pDP38/GAPDL-UPA HT 350/pDP38/GAPFL-UPA HT350/pDP38/GAPF.L-UPA of transformed yeast strains HT246/pJDB2O7/PH05-UPA, HT246/pJDB2O7/PHO5-I~U?A and HT246/pJDB2O7R/PHO5-I-UPA contain plasmids with the full length PHO5 promoter and require derepression of the promoter for expression of scu-PA. Cells of the S. cerevisiae ΗΤ246 transformants are each grown in 10 ml of yeast minimal medium (Difco Yeast Nitrogen Base without amino acids to which 2 % glucose and 20 mg/1 L-histidine are added) in a 50 ml Erlenmeyer flask with shaking at 30°C for 24 h until a density of 5-7 x 107 cells/ ml is reached. The cells of the preculture are then washed in 0.9 % NaCl and 20 % of the preculture cells are used to inoculate 50 ml of a low P^ minimal medium prepared according to the -, q recipe of the Difco Yeast Nitrogen Base medium (without amino acids), but containing 0.03 g/1 ΚΗςΡΟι,, 10 g/1 L-aspargine instead of (NHOzSOt,, 20 g/1 glucose and 1 g/1 L-histidine. The cultures are agitated at 30°C for up to 48 h at 180 revs/min. Final densities of 1 x 108 cells/ml (= Οϋθοο = 4-5) are obtained.

Yeast transformants comprising plasmids with GAPDH promoter of different length express scu-PA constitutively.

Corresponding transformants containing pJDB2O7 derived plasmids are grown in the yeast minimal medium preculture (supra). 20 % of the washed preculture is inoculated into 50 ml of the complex main culture medium consisting of (g/1): peptone 5, yeast extract 10, glucose 20, sucrose 40, (NHh)2SOi» 3, KH2PO1, 2, MgSOi, 0.5, NaCl 0.1, CaCl2 0.1, biotin 10 pg/l. Approximately 1 x 10’ cells/ml (=OD&oo 40-45) are obtained after 48 h of incubation at 30°C and 200 revs/min.

Corresponding transformants comprising pDP38 derived plasmids are cultivated under uracil selection. The cells are grown for 24 h at 30°C and 180 revs/min in a preculture medium consisting of (g/1); casamino acids 4.5, yeast extract 6.5, sucrose 20, glucose 20, (NHii^SOi, 3.6, ΚΗ2ΡΟ4 ls MgSOi, 0.2, CaCl2 0.013, trace elements. % of the washed preculture cells are used to inoculate 50 ml selective main culture medium consisting of (g/1): Yeast Nitrogen base (Difco, without amino acids) 5, L-asparagine 7.5, casamino acids 8.5, methyl-ethylsulfonat 10, adenine 0.05, L-histidine 0.04, L-leucine 0.1, L-trypfcophan 1, Ca-pantothenate 0.03, glucose 30.

Approximately 8 x 10s cells/ml ( = ODeoo ca. 35) are obtained after 48 h of incubation at 30°C and 200 revs/min.

Cells from 2 ml are collected after 24 h and 48 h, respectively, by centrifugation at 3000 rpm for 10 min in Falcon 2070 tubes. The cells are washed once with 0.9 % NaCl and centrifuged. The cell pellet is suspended in lysis buffer (66 mM potassium phosphate pH 7.4, 4 mM Zwittergent (Calbiochem)]. To the cell suspension are added 8 g of glass beads (0.5-0.75 mm in diameter) and the suspension is shaken on a Vortex Mixer [Scientific Instruments Inc. USA) at full speed for 4-5 x 2 min. More than 90 % of the cells are broken by this procedure. Cell debris and glass beads are sedimented by centrifugation for 5 min at 3000 rpm at 4°C. Immediately before testing the biological activity the supernatants are diluted up to 500 fold in 0.1 M Tris-HCl pH 7.4, 0.05 % Tween 80, 0.1 % bovine serum albumin.

Example 15: Determination of biological activity Amidolytic activity of scu-PA may be measured directly using the Kabi (KabiVitrum, Stockholm, Sweden) synthetic tripeptide chromogenie substrate S-2444 (pyro-Glu-Gly-Arg-pNA). For the determination of the content of tcu-PA cleaved at Lysl58 and/or Lysl36 the assay is carried out according to the manufacture's recommendations (without plasmin activation). Scu-PA requires plasmin activation prior to the determination of its amidolytic activity leading to the following modification of the direct assay: 100 pi scu-PA containing samples (see Example 14) are preincubated with 0.01 U human plasmin (Boehringer Mannheim, Germany) for 60 min at 37°C. 5 I.U. aprotinin (Boehringer Mannheim, Germany) are added and the mixture incubated for a further 10 min at 100 m temperature before the chromogenic substrate S-2444 (supra) is added. 3 Quantitation of the activity is done by comparison with the WHO urokinase standard (lot 66/46) and expressed in International Units (I.U.). According to the commercial preparation Ukidan® (Serono, Freiburg, Germany), highly purified urokinase isolated from human urine has a specific activity of 70,000-100,000 I.U./mg protein.

Scu-PA content may also be measured via its plasminogen activation using the synthetic tripeptide substrate S-2251 (KabiVitrum, Stockholm, Sweden). The assay is done according to the manufacturer's recommendation (KabiVitrum, supra) with scu-PA containing samples as plasminogen activator instead of streptokinase.

The amount of antigen present in yeast fermentation broths is estimated using the dot blotting procedure (Bio-Rad, Richmond, USA). A polyclonal rabbit anti-human urine urokinase antibody is used for detection.

Samples are taken at 24 h of fermentation and are pre-treated as described in Example 14. A summary of the plasminogen activating (S-2251 as substrate) activities with different plasmids in 2 different strains is given in Table 1.

Table 1 plasm!d host selection S-2251 activity I.U./2 x 107 cells pJDB207R/PH05-UPA HT246 leu 0 pJDB207/PH05-SSU?A HT246 leu 2.2 pJDB2O7/PH05-I-UPA HT246 leu 10 PJDB2O7/PHO5-UPA HT246 leu 1.4 pJDB2O7/GAPDL-UPA HT246 leu 9 pDP38/GAPDL-X-UPA HT350 leu 3 pDP38/GAPDL-T-UPA HT350 ura 5.7 pDP38/GAPFL-x-UPA HT350 leu 0.4 pDP 38/GAPFL-1-UFA HT350 ura 3.4 4 A comparison of the activities determined in the 3 assays used is given in Table 2.

Indicated are total volumetric titers (per ml culture broth) obtained after 48 h fermentation of strain HT246 using the plasmid pJDB2O7/GAPDL-I-UPA.

Table 2 I.U./ml culture broth Activity on Activity on S-2251 Dot-blotting (estimation) S-2444 without plasmin S-2444 with plasmin 25.8 215 1670 ca. 1000 The results indicate that best scu-PA production in yeast is obtained when the DNA construction includes a yeast signal sequence such as the PHO5 or invertase signal sequence. The results further indicate that scu-PA is expressed in different yeast host-backgrounds, whereby fermentation under selective conditions (either low Pi minimal medium or uracil selection) leads to relatively higher specific productivity per OD.

About 90 % of the expression product is scu-PA as indicated in the activity measurements on S-2444 with or without plasmin activation. Dot-blotting indicates that the amount of antigen present is not exceeding significantly the amount of biological activity measured. Therefore yeast recomh.inant scu-PA is correctly folded as in genuine scu-PA, Example 16: Recovery of scu-PA from yeast cells (30 1 culture broth) S. cerevisiae strain HT246/pJDB207/GAPDL-l-UPA is grown in the same way as S. cerevisiae strain GRF18/pJDB207/GAPFI.~HIR in the production of hirudin (cf. European Patent Application No. 225633). The content of scu-PA Inside the cells (after disintegration, see Example 14) is determined by the indirect ami do]ytic assay using the substrate S-2251 (see Example 15). After 48 h of incubation the cells are harvested by centrifugation in a Sharpies centrifuge (Appareils Centrifuges, Rueil, France) and suspended in an equal volume of lysis buffer [200 mM K2HPOi,5 0.2 % Tween 80]. The cells are then broken in a glass bead mill (Dyno-Mill, KDL, Bachofen AG, Basel; 4500 rpm, 18 1/h). The suspension is diluted 5 times with 150 mM NaCl, 0.05 % Tween 80, and the cell debris are sedimented in the presence of 0.6 °/oo polyethyleneimine at 4°C by centrifugation in a Westfalia separator SB7-47. The slightly turbid supernatant is filtered through a Zetapor Cartridge (0.22 pm pore width) and adjusted to pH 6.05 with IN HC1. 1 1 of cation exchanger S-Sepharose fast flow (Pharmacia) per kg of sedimented yeast cells is added and the suspension is stirred for 1 h at 4°C. The beads are then transferred to a column of diameter 9 cm and washed with 50 mM sodium phosphate pH 7.0, 150 mM NaCl, 0=05 % Synperonic. Scu-PA is eluted at a flow rate of 30 ml/min with the same buffer containing 250 mM NaCl. To the fractions containing the scu-PA [as measured in the direct araidolvtlc assays; cf. Example 15] concanavalin A Sepharose (Pharmacia) is added (1 ml gel per 0,2 mg scu-PA) and the suspension stirred for 45 min at 4°C- After washing with 1M NaCl 10 mM sodium phosphate pH 7.0, 0.05 % Synperonic, the beads are transferred to a column of diameter 4.4 cm and scu-PA is eluted with 0.8 M methyl-Ci-D-mannopyranoside, 150 mM NaCl, 20 mM sodium acetate pH 4.0, 0.05 % Synperonic at a flow rate of 1 1/h. To the fractions containing scu-PA, antiurokinase IgG-Sepharose is added [purified polyclonal rabbit antibody (IgG-fraction) raised against human urinary urokinase] and the suspension is stirred for 45 min at 4°C. The beads are then transferred to a column of diameter 2.2 cm and washed with 1M NaCl, 10 mM sodium phosphate pH 7.0, 0.05 % Synperonic. Scu-PA Is eluted with 0.1M glvcine-HCl pH 2,4S 0.05 % Synperonic at a flow rate of 1 ml/min. The fractions containing scu-PA are adjusted to pH 6 with IN NaOH.

At this stage about 25-30 % of the total activity consists of tcu-PA as identified by the direct amidolytic assay (cf. -Example 15). 6 The effluent from the antibody column is applied to a Mono-S column (1 ml bed volume; Pharmacia) equilibrated with 50 mM sodium phosphate ph 6.0, 0.05 % Synperonic. After washing with the equilibration buffer, scu-PA is eluted with a step-like gradient of the equilibration buffer and a buffer B composed of 500 mM NaCl, 50 mM sodium phosphate pH 7.0, 0.05 % Synperonic at a flow rate of 1 ml/min. By this elution, two activity peaks are observed, one peak eluted at the first step at 30 % of buffer B and the other at the second step at 55 % of buffer B. The direct amidolytic assay revealed that the fraction eluted at 30 % B exhibits still a high amount of tcu-PA, whereas the fraction eluted at 55 % B exhibits as little as 1-3 % tcu-PA.

Yeast scu-PA produced in this manner migrates as a single band of about 51 KD molecular weight in SDS polyacrylamide gel electrophoresis under non-reducing conditions. The scu-PA obtained has a purity of about or more than 95 % as judged by the direct amidolytic assay (cf. Example 15) and by SDS-polyacrylamide gel electrophoresis under reducing conditions.

Example 17: Glycosylation state of yeast scu-PA Glycosylation is determined with a 125I-concanavalin A overlay assay after gel electrophoresis. Concanavalin A is a plant lectin which specifically recognizes mannose residues. Purified scu-PA from yeast as well as u-PA standard Ukidan® (Serono, Freiburg, Germany) are subjected to gel electrophoresis on 10 % gels. Proteins are then fixed to the gel in 7 % acetic acid for 30 min. The gel is washed in lectin buffer having the following composition: 0.15 M NaCl mM Tris-HCl pH 7.4 0.5 mM CaCl2 0.5 mM MnCl2 Washing is continued until the pH reaches about 7,3.

? The gel is then carefully overlayered with 2 ml lectin buffer containing 3 mg hemoglobin, 100 pg concanavalin A and 2 pCi 125Iconcanavalin A.

Another otherwise identical gel is overlayered with the mix mentioned above supplemented with 0.2 M α-methyl-mannoside.

After 3 h incubation in a humidified chamber the gels are extensively washed in lectin buffer, dried and exposed to X-ray films. Without os-methyl-mannoside both Ukidan® and yeast scu-PA are stained with 125I-concanavalin. Their molecular weights are essentially identical. In the presence of o-methyl-mannoside, a competitive inhibitor of lectin-binding, the human urinary urokinase disappears, whereas yeast scu-PA is still visible. This indicates that yeast scu-PA is glycosylated and that its glycosylation is of a different type compared to human urinary urokinase.

Example 18: Further purification of scu-PA A solution of scu-PA (cf. Example 16) dialyzed against 0.05 M Tris-HCl pH 8.0, 0.05 % Tween 80, 0.05 M NaCl is loaded onto 3 ml of a benzamidine-Sepharose column (Pharmacia, Uppsala, Sweden) and washed with the Tris-HCl pH 8.0 buffer containing 1 M NaCl. Scu-PA 2q is fully recovered in the flow-through and in the wash-solution whereas the byproduct tcu-PA can be eluted from the column with 0.05 M Tris-HCl pH 8.0 buffer, containing 1 M L-arginine. The scu-PA obtained has a purity of about or more than 98 %. 8 Example 19: First pharmaceutical composition for parenteral administration A solution for parenteral administration is prepared by dissolving 3 mg of purified scu-PA, 25 mg mannitol and 45 mg sodium chloride in 5 ml sterilized water and admixing the resulting solution with a suitable volume of 5 % glucose solution. The solution is sterilized by filtration through a 0.22 pm membrane filter.

Example 20: Second pharmaceutical composition for parenteral administration (dispersion for injection) 169.3 mg soybean lecithin (soybean phosphatide NC 95, manufacturer: Nattermann, Cologne, Germany; purity 90-96 composition of fatty acids: linoleic acid 61-71 %, linolenic acid 4-7 oleic acid 6-13 %, palmitic acid 10=15 %, stearic acid 1.5-3.5 %) and 92.7 mg pure sodium glycocholafe are dissolved in 752.5 ml of sterilized water. The solution is adjusted to pH 7.4 with 1 N NaOH. 10 mg of lyophilized scu-PA is added. The mixture is stirred until a clear solution has been obtained. The solution is sterilized by filtration through a 0.22 pm membrane filter and filled into ampoules.

Pharmaceutical compositions containing scu-PA mutants as active ingredient are prepared in an analogous manner as described in Examples 19 and 20» 0 Deposition of microorganisms The following microorganism strains were deposited at the Deutsche Sammlung von Mikroorganissnen (DSM), & Grisebachstrasse 8, D-3400 Gottingen, Mascheroder Weg lb, D-3300 Braunschweig (deposition dates and accession numbers given): * Saccharomyces cerevisiae HT246: April 15, 1987; DSM 4084; ** Escherichia coli HBlOl/pCSlo: October 23, 1987; DSM 4294; ** Escherichia coli HB101/p3lR/SS-TPAA2: October 23, 1987; DSM 4295; * ** Escherichia coli HB101/pcUK176: October 23, 1987; DSM 4290; ** Escherichia coli JM109/pDP38: February 19, 1988; DSM 4414.

Claims (14)

Claims

1. A method for the production of human single chain urokinase-type plasminogen activator comprising culturing under appropriate nutrient conditions a yeast strain transformed with a hybrid vector comprising a yeast expression control sequence, a DNA segment consisting of a first DNA sequence encoding a signal peptide upstream of and in reading frame with a second DNA sequence coding for mature urokinase-type plasminogen activator, which DNA segment is under transcriptional control of said expression control sequence, and a DNA sequence comprising transcription termination signals of a yeast gene, and isolating said urokinase-type plasminogen activator.

2. A method for the production of a plasminogen activator according to claim 1 having the amino acid sequence of natural human single chain urokinase-type plasminogen activator in which Asn3O2 is yeast specifically glycosylated.

3. A method according to claim 1 in which the yeast strain is a strain of Saccharomyces cerevisiae.

4. A plasminogen activator which is human single urokinase-type plasminogen activator which is yeast specifically glycoslyated.

5. A human single chain urokinase-type plasminogen activator which is yeast specifically glycosylated whenever prepared by the method according to claim 1.

6. . a yeast hybrid vector comprising a yeast expression control sequence, a DNA segment consisting of a first DNA sequence encoding a signal peptide upstream of and in reading frame with a second DNA sequence coding for mature urokinase-type plasminogen activator , which DNA segment is under transcriptional control of said expression control sequence, and a DNA sequence comprising transcription termination signals of a yeast gene.

7. , A transformed yeast host containing a hybrid vector comprising a yeast expression control sequence, a DNA segment consisting of a first DNA sequence encoding a signal peptide upstream of and in reading frame with ® second DNA sequence coding for mature urokinase-type plasminogen activator, which DNA segment is under transcriptional control of said expression control sequence, and a DNA sequence comprising transcription termination signals of a yeast gene.

8. A human single chain urokinase-type plasminogen activator according to claim 4 for use in a method for the treatment of the human body by prophylaxis or therapy.

9. . A pharmaceutical composition comprising a therapeutically effective amount of a plasminogen activator according to claim 4 together with a pharmaceutically acceptable carrier.

10. Use of a plasminogen activator according to claim 4 for the production of a pharmaceutical composition.

11. · A human single chain urokinase-type plasminogen activator or a mutant thereof substantially as herein described.

12. A method for the production of a human single chain urokinasetype plasminogen activator or s mutant thereof substantially as herein described.

13. A transformed yeast host substantially as herein described.

14. A yeast hybrid vector substantially as herein described.