DD283839A5 - METHOD FOR THE PRODUCTION OF A HEFESTAMM WITHOUT CARBOXYTEPTIDASE-YSC ALPHA ACTIVITY - Google Patents

Abstract

The invention relates to a method for producing a yeast strain without carboxypeptidase ysca activity, which is used for the production of heterologous proteins. The yeast strain of the present invention is transformed with a hybrid vector consisting of a yeast promoter operatively linked to a DNA sequence encoding a heterologous protein {yeast strain production; without carboxypeptidase ysca activity; Hybrid vector transformed; yeast promoter; DNA sequence; Polypeptide}

Description

-2 ~ 283 839 Field of application of the invention

Application area is the DNA Recombinant Technologio. There are genetically manipulated yeast cells .lorgostoPt, which are used to produce heterologous protein in with pharmacologically valuable properties

Characteristic of the known state of the art

Recently, a whole number of heterologous proteins have been expressed in yeast after the yeast cells have been transformed with suitable expression vectors, these vectors being DNA sequences encoding these proteins, e.g. B. α-interferon (IFNa, Hitzemann et al., (1981) Nature 294, 717-722), lysozyme (Oberto et al., [1985] Gene 40, 57-65) a-amylase (Sato et al., (1986) Gene SO , 247-257), tissue plasminogen activator (t-PA, European Patent Application No. 143,081), or desulphatohirudin (European Patent Application No. 225,633). However, in many cases, the heterologous proteins are not synthesized in pure form but as mixtures that are partially synthesized For example, the expression of the human atrial natriuretic peptide (hANP) in yeast results in the deposition of two forms of mature hANPs that are expressed in their C-terminus The main form did not have the last two amino acids of the protein (Arg 160 and Tyr 151), while the minor component formed the full-length protein, similar results were after expression of epidermal growth factor (BGF) in yeast e obtained (George Nascimento et al. (1988) Biochemistry 27, 797-802) where the precipitated expropriation products were heterogeneous in the sense that either the last (Arg 53) or the last two amino acids (Leu 52 and Arg53) were missing and no full length EGF was produced. A polypeptide that has recently received considerable attention from molecular biologists is hirudin, an anticoagulant found naturally in leeches (Hirudo medicinalis). Hirudin is not a single polypeptide but a class of the same effect-inducing polypeptides consisting of at least four members known as hirudin variant 1 (HV 1), hirudin variant 2 (HV2) (see European Patent Application No. 158,564), Hirodin Variant PA (see PCT Application No.86 / 03.493) and "des- (Val) _r Hirudin" (see European Patent Application No. 158986) These variants differ from each other by a number of amino acids (in particular the Sequence at the N-terminus of the HV1 Val-Val-Tyr, that of the HV2 and PA Ile-Thr-Tyr and that of the "des- (Val) ₂ -hirudin" Thr-Tyr), however, collectively have an accumulation of hydrophobic Amino acids at the N-terminus and polar amino acids at the C-terminus, a tyrosine residue (Tyr * ³ ) in the form of a sulfate monoester, three disulfide bridges and the anticoagulant effect.

More recently, c-DNA and synthetic genes encoding hirudin varinous have been cloned and expressed in microwavewtons. Although the expression products on Tyr ⁶³ do not have a sulfate monoester group and were therefore termed desulphatohirudins, they were found to have approximately the same biological activity as the natural sulfated hirudins.

The desuliato hirudin variant HV1 was expressed in Escherichia coli (European Patent Applications Nos. 158,654 and 168,342) and in Saccharomyces cerevisiae (European Patent Applications Nos. 168,342,200,655,225,633 and 252,854). Similarly, desulfato-hirudin HV2 has been expressed in Escherichia coli (European Patent Application No. 158,564) and in Saccharomyces cerevisiae (European Patent Application No. 200,655; PCT Application No. 86 / 01,224), and des- (Va | i ₂ -desulfato-hirudin has been described in U.S. Pat Escherichia coli (European Patent Application No. 158996).

In general, the expression efficiency and the yields of hirudin derivatives are higher when S cerevisiae air host microorganism is chosen. However, independent of the particular yeast host strain used, the expression product proved to be a heterogeneous mixture of desulfato-hirudin derivatives that differed from each other in sequence at the C-terminus. So z. For example, it was found that the culture broths obtained from cultured yeast strains containing the hirudin variant HV1 strain contained desulfato-hirudin HV1 contaminated with substantial quantities of analogous compounds in which the amino acid GIn * ^{5 was expressed} at C Terminus or the amino acids Leu ⁶⁴ and GIn ⁶⁵ at the C-terminus were absent.

The separation of mixtures containing the proteins of full length of the amino acid sequence and desulfato-hirudin, hANP or EGF and derivatives thereof with a truncated sequence at the C-terminus, into the individual components and the purification of these mixtures and components to homogeneity difficult and time consuming. In view of the costs to be incurred, there is therefore a need for improved processes which make it possible to economically produce homogeneous proteins such as desulfato-hirudin in yeast.

Object of the invention

The invention provides a method for the detection of a yeast strain without carboxypeptidase ysca activity, which enables an economic production of homogeneous heterologous proteins.

Op. Ment of the essence of the invention

It is an object of the present invention to provide a method for producing genetically manipulated yeast cells which is suitable for producing homogeneous heterologous proteins.

While it is apparent that the isolation of DP derivatives of heterologous proteins with truncated sequences at the C-terminus from yeast strains transformed from culture broths containing the corresponding DNA sequence encoding these proteins, due to the action of endogenous yeast proteases on the primary Expression product, e.g. Integral desulfato hirudin, the particular protease (s) responsible for degradation at the C-terminus has not been identified. The most important yeast proteases that are generally involved in protein degradation are the endopeptidases yscA and yscB and the

Corboxypeptidases yscY and yscS The use of yeast strains lacking the protease A, B, Y, and / or S activity may in part prevent the statistical proteolysis of foreign gene products, e.g. As the desulfato-hirudins reduce. However, substantial amounts of proteins lacking one or two amino acids at the C-terminus are still observed. Surprisingly, it has now been found that yeast mutants deficient in the carboxypoptidase ysca activity are unable to cleave amino acids from the C-terminus of heterologous proteins and therefore result in integral (authentic) proteins. Carboxypeptidase ysca is a membrane bound exopeptidase and, as is well known, plays an important role in the maturation of killer factor and mating factor α (see J.C. Wagner and D.Wolf (1987) FEBS Letters 221, 423). It is the expression product of the KEX1 gene. According to the published data (see P.S.Rendueles and D.H. Wolf [1988], FEMS Microbiol. Rev. 54, 17), the effect of ysca strong on the basic amino acid residues at the C-terminus is limited (Arg, Lys). Taking this data into account and the fact that the amino acids of desulfato-hirudin are not basic at the C-terminus (eg GIn and Leu in the desulfato-hirudin variant HV1), it is highly surprising and an unexpected result that By using strains of yeast mutants without carboxypeptidase ysca activity, it is possible to produce homogeneous desulfato-hirudin without a lengthy separation of the C-terminal truncated desulfato-hirudin analogs that would be necessary. The same applies to other heterologous proteins z. Hap, EGF, or the connective tissue activating peptide III (CTAP-III, Mullenbach et al., (1986) J. Biol. Chem. 261, 719-722), which can be produced to their full length when a strain of a yeast mutant , which has no carboxypeptidase ysca activity, is used for transformation with a suitable vector.

Accordingly, this invention relates to an improved process for producing a yeast strain lacking carboxypeptidase ysca activity for producing heterologous yeast proteins in homogeneous torr which consists of a yeast strain having no carboxypeptidase ysca activity and a hybrid vector derived from a yeast promoter which is operably linked to a DNA sequence coding for said heterologous protein, has been transformed, and to isolate said heterologous protein.

More particularly, the invention relates to an improved method for producing a yeast strain lacking carboxypeptidase ysca activity for producing a protein heterologous to yeast in a homogeneous form consisting of cultivating said yeast stem with a hybrid vector derived from a yeast promoter operably linked to a first DNA sequence encoding a signal paptide which is linked in a suitable reading frame to a second DNA sequence encoding said heterologous protein, and from a DNA sequence encoding the termination signals for yeast transcription, has been transformed and to isolate said heterologous proteins. Heterologous proteins which can be prepared by the improved method of the invention are those proteins which are susceptible to degradation at the C-terminus after translation by carboxypeptidase ysca following their expression in yeast. Such heterologous proteins are characterized by two terminal amino acids at the C-terminus which can be selected from the group: Lys, Arg, Tyr, Ala, Leu, Gin, Glu, Asp, As η and Ser. The preferred heterologous proteins are those proteins mentioned above, especially desulfatochirudin. In the most preferred embodiment, this invention relates to an improved method of producing a yeast strain lacking carboxypeptidase ysca activity for the production of desulphatohirudin, which consists in culturing a yeast strain which has no carboxpeptidase ysca activity and has been transformed with a hybrid vector which consists of a desulphatohirudin expression cassette of a yeast promoter operable on a first DNA sequence; ¹ which encodes a signal peptide and is bound in a suitable reading frame to a second DNA sequence coding for desulphatohirudin and from a DNA sequence containing yeast transformation termination signals and to isolate desulphatohirudin. The term "desulphatirudin" is used to include the desulphatohirudin derivatives described in the literature or obtainable from transformed microorganism strains containing a DNA encoding desulphatohirudin. Variants HV 1, HV 1 (modified a, b), HV 2, HV 2 (modified a, b and c), PA, variants of PA and of (Val ₂ ) -desulphatohirudin Preferred are such desulphatohirudins which have the following formula:

(I) X ₁ TyrThr Asp Cys Thr Glu Ser Gly GIn Asn Leu Cys Leu Cys Glu Gly Ser Asn VaI Cys Gly Gin Gly Asn X ₂ Cys Le Le Gly Ser Asp Gly GIu X ₃ Asn Gin Cys VaI Thr GIy Glu GIy Thr Pro X ₄ Pro GIn Ser X ₆ Asn Asp Gly Asp Phe Glu Glu Pro Glu X ₆

in the

a) X _{1 represents} the dipeptide residue VaI-VaI and X ₂ , X ₃ and X _{4 are} each Lys, X _{5 is} His and X _{6 is} the peptide residue Glu-Tyr-Leu-Gln (HV-I), or

b) X ₂ lie odor GIu and X) and X ₃ -Xe are as defined under a) (HV 1 modified a), or

c) X _{3 is} He or GIu and X ₁ , X ₂ and X ₄ -X _{6 are} as defined under a) (HV 1 modified a), or

d) X _{4 is} He or GIu and X ₁ -X ₃ and Xs-X, as defined under a) (HV 1 modified a), or

e) X _{5 is} Leu or Asp and X ₁ -X ₄ and X _{6 are} as defined under a) (HV 1 modifies a), or

f) X ₆ is selected from the group consisting of Glu-Tyr, Glu-Tyr-Leu, Glu-Asp-Leu-Gln, Glu-Glu-Leu-Gln, Glu-Tyr-Lys-Arg, Glu-Asp-Lys -Arg, Glu-Lys-Leu-Gln, Ser-Phe-Arg-Tyr, Trp-Glu-Leu-Aig, Glu-Tyr-Leu-Gln-Pro and Glu-Tyr-Leu-Gln-Arg and X ₁ - X _{5 are} as defined under a) (HV 1 modifies D), or

g) wherein X _{1 represents} Thr and X ₂ -X _{6 are} as defined under a) (des- (Val) _2- desulfatohirudin), or the formula

(II) Y ₁ Tyr Thr Asp Cys Thr Glu Ser Gly GIn Asn Leu Cys Leu Cys GIu Gly Ser Asn VaI Cys Gly Lys Gly Asn Lys Cys He Leu Gly Ser AsnGlyLysGlyAsriGlnCysValThrGlyGluGlyThrProYjProGluSerHisAsnAsnGlyAsp Phe GIu GluL Pro GIu GIu Y ₃ Leu GIn

in which

a) Y _{1 represents} the N-terminal dipeptide radical Ile-Thr, Y _{2 is} Asn and Y _{3 is} Tyr (HV 2), or

b) Y ₂ Lys, Arg or His are listed and Y ₁ and Y _{2 are} as defined under a) (HV2 modified a), or

c) Y _{3 is} GIu or Asp and Y ₁ and Y _{2 are} as defined under a) (HV 2 modified b), or

d) Yi is the N-terminalon dipeptide moiety VaI-VaI and Y ₂ and Y _{3 are} as defined under a) (HV2 modified c), or the formula

(III) LieThrTyrThrAspCysThrGluSerGlyGlnAsnLeuCysLouCysGluGlySerAsnValCysGlyLysGlyAsnLyBCyslleLouGly SerGine Gly LysApAsn GIn CysValThrGy GIu GIy for Pro Lys ProGIn Ser HisAsn Gin GIyAsp Phe GIu Pro He Pro GIu Asp Ala Tyr Asp GIu

and (PA) and variants of this (PA) which are characterized in that the primary structure is shortened by 1 to 2 amino acids at the N-terminus or by 18, 10, 9, 6, 4 or 2 amino acids at the C-terminus. The most bovorzugtu desulphatohirudin compound is that of formula (I) ₁ in which Xi to _Xn as described under a) are defined. The yeast host strains and the components of the hybrid vectors are those specified below. The transformed yeast strains are cultured using methods known in the art. Thus, the transformed yeast strains of this invention are cultured in a liquid medium containing assimilable sources of carbon, nitrogen and inorganic salts.

Many carbon sources are usable. Examples of preferred carbon sources are assimilable carbohydrates, e.g. Glucose, maltose, mannitol, fructose or lactose, or acetates, e.g. As sodium acetate, which can be used either alone or in geeigi.dten mixtures. Suitable nitrogen sources are for. B. amino acids, eg. ß. Casamino acids, peptides and proteins and their degradation products such as tryptone, peptone, meat extracts, yeast extracts, malt extract, corn starch and ammonium salts such as ammonium chloride, sulfate or nitrate, which can be used either alone or in suitable mixtures. Inorganic salts, ^d: e can be used are, among others, sulfates, chlorides, phosphates and carbonates of sodium, potassium, magnesium and calcium, in addition to the nutrient medium may also contain growth promoting substances. Substances that promote growth include trace elements such as Eison, zinc, manganese and similar or specific amino acids. Because of the incompatibility between the endogenous two-micron DNA and hybrid vectors carrying their replica, the Hof cells transformed with such hybrid vectors tend to slough off the latter. Such yeast cells must be grown under selective conditions, ie, conditions under which expression of a gene contained in a plasmid is required for growth. The most selective markers (markers) currently used and present in the vectors of the invention (infra) are genes encoding enzymes for amino acid or purine biosynthesis. This makes it necessary to use synthetic minimal drugs lacking the corresponding amino acid or purine base. However, genes conferring resistance to a particular biocide may as well be used (eg, a gene conferring resistance to the aminoglucoside G418). Yeast cells transformed with -. Sectoron containing antibiotic resistance genes are grown in complex nutrient media containing the appropriate antibiotic, thereby achieving higher growth rates and higher cell densities.

Hybrid vectors containing the complete two micron DNA (including a functional replication site) are maintained stable in strains of Saccharomyces cerevisiao lacking the endogenous two micron plasmid (so-called cir ° strains), so that the cultures do not -selective growth conditions can be carried out, for. B. in a complex nutrient medium. Yeast cells containing hybrid plasmids with a constitutive promoter (e.g., ADH I, GAPDH) express the DNA encoding a heterologous protein driven by this promoter without induction. However, if this DNA is controlled by a regulatory promoter (eg, PGK or PHO5), the composition of the growth medium must be adjusted to obtain maximum yields of mRNA transcripts, i. H. When the PHO5 promoter is used, the medium used for growth must contain a low concentration of inorganic phosphates to derepress this promoter.

The cultivation is carried out using known techniques. The conditions for growth such as temperature, pH of the medium and fermentation time are selected in such a way that maximum yields of heterologous proteins are achieved. A selected yeast strain is preferably grown under aerobic conditions in suspension cultures with shaking or stirring at a temperature between about 25 ° C and 35 ° C, preferably at about 28 ° C, at a pH between 4 and 7, e.g. At about pH 5, and for at least 1 to 3 days, preferably until satisfactory protease yields are obtained.

The heterologous proteins expressed in yeast can be accumulated within the cells or delivered into the culture medium. In the case of desulphatohirudin, regardless of the yeast strain, promoter and signal peptide used, most of the protein produced is delivered into the culture medium while only a small portion remains associated with the cells. The exact ratio (delivered compounds / cell-associated compounds) will depend on the fermentation conditions and the method of recovery used. Generally it is about 8: 1 or above. Accordingly, the discharged desulphatohirudin is always strongly dominant.

The heterologous protein can be recovered from the culture medium by conventional ancestors. Thus, the first step is usually the separation of the cells from the culture fluid by centrifugation. The resulting supernatant fluid can be enriched in heterologous proteins by adding polyethyleneimine to thereby remove most of the non-protein mixture and precipitate the proteins by saturating the solution with ammonium sulfate. Host proteins, if present, may also be precipitated by acidification with acetic acid (eg, 0.1%, pH = 4-5). Further enrichment of the hiterologic protein can be achieved by extracting the acetic acid supernatant with n-butanol. Other purification steps include, for. Demineralization, chromatography methods such as ion exchange chromatography, gel filtration chromatography, partition chromatography, HPLC, reversed-phase HPLC, and the like. The Abtreni.ung the components of the protein mixture is also effected by dialysis, due to the charges by gel electrophoresis or carrier-free electrophoresis, on the basis of molecular size by means of suitable Sephadex columns, by affinity chromatography, for. With antibodies, particularly with anthonoclonal antibodies, or with thrombin coupled to a suitable affinity chromatography carrier, or by other methods, especially those known from the literature.

If the hematopoietic protein is not excreted or if it is desired to isoprene an additional heterologous protein isolated from cells, i. which has been accumulated intracellularly or is associated with cells or in the poriplabatic space, some additional purification steps become necessary. Thus, if the hematopoietic protein has accumulated within the cells, then the first step to such recovery is in its release from the cell interior. In most procedures, the cell wall is first removed by enzymatic degradation with glucosidases (infra). Thereafter, the spheroplasts thus prepared are washed with detergents, e.g. with Triton, treated. On the other hand cind mechanical forces, z. Shear forces (e.g., X-Press or French Prosse) or shaking with glass beads to disrupt the cells. In cases where the heterologous protein is secreted by the host cells into the periplasmic space, a simplified protocol can be used: The heterologous protein is obtained without lyso of the cells by enzymatic removal of the cell wall or by treatment with chemical reagents, e.g. As thiol compounds or EDTA, which lead to damage to the cell wall, thereby allowing to set the product in freedom won. In the case of desulphatohirudin, the test may be carried out with anti-hirudin or anti-desulphatohirudin antibodies (eg monoclonal antibodies obtainable from hybridoma cells), the thrombin assay (MU Bergmeyer, ed., Methods in Enzymatic Analysis , Dand 2, p.314-316, Verlag Chemie, Weinheim IBRD), 1983) or the blood coagulation test (F.Markwardt et al., 11982], Thromb. Haemost., 47,226) are used to detect hirudin activity in fractions obtained during the purification procedures. Analogous test methods known from the literature may be used to detect other hoterologic proteins.

The transformed yeast cells according to the invention can be produced by recombinant DNA technology with the following steps:

Preparing a hybrid vector consisting of a yeast promoter operably linked to a DNA sequence encoding a heterologous protein, in particular a hybrid vector consisting of a Hofe promoter operable on a first DNA sequence bound for a signal peptide linked in a suitable reading frame to a second DNA sequence encoding a heterologous protein and a DNA sequence containing yeast transcription termination signals;

if necessary, providing a mutant strain of yeast which has no carboxypeptidase ysca activity,

- Transformation of the resulting mutant yeast strain with the designated hybrid vector and

- Selection of the transformed yeast cells from the untransformed yeast cells.

The yeast hybrid vectors produced according to the invention consist of a yeast promoter operably linked to a DNA sequence encoding a heterologous protein. Preferred hybrid vectors consist of a yeast promoter linked to a first DNA sequence encoding a signal peptide and bound in a suitable reading frame to a second DNA sequence encoding a heterologous protein such as desulphatohirudin, and from a DNA sequence containing yeast transcription termination signals.

The yeast promoter is anyway a regulatory promoter such as PHO5, MF1 or GAL1 promoter or a constitutive promoter. In the case of desulphatohirudin expression, a constitutive promoter is preferred. The constitutive yeast promoter is preferably derived from a highly expressive, e.g. A gene encoding a glycolytic enzyme such as the promoter of enolase, glyceraldehyde-S-phosphate dehydrogeriase-IGAPDH), 3-phosphoglycerate kinase (PGK), hexokinase, pyruvate decarboxylase, phosphofructokinase , The glucose-6-phosphate isomerase, 3-phosphoglycerate mutase, pyruvate kinase, triosephosphate isomerase, phosphoglucose isomerase and glucokinase gene, furthermore the ADHI or TRPI promoter and a truncated promoter of the acid phosphatase PH05, of which the upstream activation region was separated. Particularly preferred are the GAPDH promoter and functionally fragments thereof starting at a nucleofectide between -550 and -180, in particular at nucleotide -540, -263 or -198, and terminating at the nucleoide-D of the GAPDH gene, and truncated constitutive PH05 promoters starting at the nucleotide between -200 and -160, especially at -173, and at the nucleotide -9 of the PHO5 gene.

The DNA sequence encoding a signal peptide ("signal sequence") is preferably derived from a yeast gene encoding a polypeptide which has been deposited in a conventional manner DNA of the leech may be obtained, or other signal sequences from heterologous proteins that are commonly deposited, may also be selected. "" Yeast signal sequons are, for example, the yeast invertase signal and prepro sequences, the α-factor , Pheromone peptidase (KEX 1), the "killer toxin" and suppressible acid phosphatase (PHO 5) genes, and the glucoamylase signal sequence from Aspergillus awamori. Alternatively, fused signal sequences may be constructed by expressing a portion of the signal sequence (if any) of the gene naturally linked to the promoter used (eg, to ΡΗΩ5) with a portion of the hirudin signal sequence or a ligated to other heterologous protein. There are those combinations preferred that an exact cleavage between the signal sequence and z. B. allow the amino acid sequence of Desulsulfatohirudins. Additional sequences, e.g. Pro or spacer sequences, which may or may not have specific processing signals, may also be included in the constructs to facilitate accurate processing of the precursor molecules. On the other hand, fused proteins can be generated which contain internal processing labels that allow good maturation in vivo and in vitro. The processing signals contain z. A Lys-Arg residue recognized by a yeast endopepidase that sits in the Golgi membranes. Preferred signal sequences of the invention are those of the yeast PHO5 gene encoding a signal peptide of the formula

Met Phe Lys Ser VaI VaI Tyr Ser Me Leu Ala AIa Ser Leu Ala Asn AIa as well as those of the yeast invertase gene encoding a signal peptide of the formula Met Leu Lei Gin AIa Phe Leu Phe Leu Leu Alp GIy Phe Ala Ala Lys Let Ser AIa encode.

Genomic DNA sequences encoding heterologous procelins can be obtained from naturally occurring sources or copy DNA (cDNA) can be isolated from the corresponding complementary mRNA or by chemical or enzymatic methods in a manner known per se.

Dio DNA-3equeiu, which codes for desulphatohirudin, is ζ. B. bokai.iit odor k; inn be isolated from genomic DNA of the leech or a complementary, double-stranded DNA of Desulfatohirudins (Dosulfatohirudin ds cDNA) is produced by the Dosulfatohlrudin rviRNA, or a gene coding for the Aminosäureii-sequence dos desulphatohirudin , is produced by chemical and enzymatic processes.

A DNA sequence containing yeast transcriptional termination signal is preferably the flanking probe sequence at the 3 'end of a yeast gene containing appropriate translation terminator and polyadenylation primers. Suitable flanking sequence on the 3'-end are z. Those of the yeast gene that are naturally linked to the promoter used. The preferred flanking sequence is that of the yeast PHO5 gene. The yeast promoter, the target DNA sequence coding for the signal peptide, the DNA sequence encoding a heterologous protein, and the DNA sequence containing Hofe transcription termination signals are operably linked together that is, they adjoin one another to such a Woiso that their normal functions are maintained. The arrangement is such that the promoter will effect appropriate expression of the heterologous gene (optionally with a preceding signal sequence), transcription termination signals, appropriate termination of transcription and polysdenylation, and the optionally present signal sequence in an appropriate reading frame in such a manner and manner Linked to the heterologous gene is that the last kudon of the signal sequence is bound directly to the first codon of the gene coding for the heterologous protein and enters a deposition of the protein. When the promoter and signal sequence have been derived from different genes, the P'omotor is preferably linked to the signal sequence between the major mRNA start and the ATG of the gene naturally linked to the promoter. The signal sequence should have its own ATG for initiation of translation. The connection of these sequences can be effected by synthetic oligodeoxynucleotide linkers carrying the recognition sequence of an endonuclease. In addition to the expression cassette, the hybrid vectors of the invention have a yeast replication site. Accordingly, the hybrid vectors will consist of a DNA segment derived from the two-micron DNA containing the start for the replication or, if yeast strains free of the two-micron DNA are used, from the entire two-micron DNA. micron DNA. The last described type of vectors is preferred. The preferred hybrid vectors according to the invention ent ent the entire two-micron DNA in uninterrupted form, ie the two-micron DNA is cut once with a restriction He donuclease, and the linearized DNA is before recirculation with other components dos vectors connected. The 11 site of restriction is chosen so as to ensure the normal function of the REP1, REP2 and FLP genes and the OHI, IR1 and IR2 * ) of the two micron DNA. Optionally, the restriction site is chosen such that the D gene of the '.' Two micron DNA is also kept intact. Preferred restriction sites are the unique Pst I site located within the D gene and the unique Hpal and SnaBI sites located outside of all of the named genes and sites. Preferably, the hybrid vectors prepared according to the invention contain one or more, in particular one or two, selective yeast genetic markers and markers and a replication source for a bacterial host, in particular Escherichia coli.

As far as the selective yeast genetic markers are concerned, any marker gene can be used which facilitates the selection of transformants due to the phenotypic expression of the marker gene. Suitable markers for yeast are e.g. those that express resistance to antibiotics or, as in the case of auxotrophic yeast mutants, genes that complement host damage. Corresponding particles exceed e.g. Resistance to the antibiotics G418, hygromycin or bleomycin, or produce prototrophy in an auxo'.phen yeast mutant, e.g. As the URA3, LEU 2-, LYS 2- or TRP1 genes. Since the amplification of the hybrid vectors is best performed in E. coli, a genetic marker of E. coli and a replication site for E. coli are advantageously included. These can be derived from E. coli plasmids, e.g. PBR322 or a PUC plasmid, e.g. PUC 18 or pUC 19, containing both a replication site for E. coli and a genetic marker of E. coli that has resistance to antibiotics, e.g. B. ampicillin transfers. The hybrid vectors prepared according to the invention are prepared by methods known in the art, e.g. By linking the expression cassette containing a yeast promoter operably linked to a DNA sequence encoding a hetrologic protein and DNA fragments containing yeast selective genetic markers and a bacterial host, and Sources of replication for yeast and for a bacterial host in the predetermined order.

The yeast strains prepared according to the invention have no carboxypeptidase ysca activity. Preferably, the yeast strains are double, triple and quadruple mutants, ie they lack a further peptidase activity. A wide variety of proteinases, such as those mentioned above, have been characterized in the yeast Saccharomyces cerevisiae (See T. Achstetter and DH Wolf [1985] Yeast 1,139-157). Mutants lacking most of these proteases were isolated and biochemically assayed. The consequences of the absence of certain proteases have been clarified and some properties have been found to be useful in the isolation of novel mutants lacking in proteases. Since spontaneous mutation frequencies are low, yeast is usually with mutage .-. S, such as X-rays or ultraviolet light or chemical mutagens treated, are remarkably effective and mutations at a rate of 1 x 10 ^"4 to 1 χ 10" ³ per gene without delivering a large percentage of killed cells. The proteases lacking in the yeast strains according to the invention do not necessarily exert necessary functions in the cell metabolism; therefore, mutations that completely destroy the activity of these proteins are not fatal. Any type of mutant for the proteases mentioned (ysca, yscB, yscA, yscY and yscS) can be isolated by itself after mutagenesis. The isolation and selection are carried out in accordance with colony screening assays which are well known in the art.

A second and more efficient method of introducing deficiency for one or more proteases into the yeast genome is site-directed mutagenesis or gene cleavage or gene replacement (see H. Rudolph et al., Gene [1985] 36, 87-05 ). If the genetic sequence is known, what, for. For example, in the case of the protease yscB, the carboxypeptidase yscY and the carboxypeptidase ysca, the genomic protease gene may be defective by insertion, substitution or deletion using the well-known method of site-directed mutagenesis (see, for example, BMJ Zoller and M. Schmidt [1983] Methods Enzymol. 100, 488), which include the preparation of a suitably engineered mutagenic oligodeoxyribonucleotide initiator iPrimer). On the other hand, the genomic protease gene can be replaced with foreign DNA, or this foreign DNA can be introduced at a suitable restriction site of the protease gene. To z. B.

To generate an I (efo-mutant that does not have peptidase ysca-Aklivity (kex ~ -mutanto) is introduced fromclo DNA at a suitable restriction site found in dom genomic KEX1 gene. In case of the yeast used, a defect in a chromosomal gene encoding an enzyme that encodes amino acid or purine biosynthesis, a corresponding intact gene can be introduced into the chromosomal KEXI gene, thereby producing prototrophy in the auxotrophic yeast strain and, at the same time, the genotype of KEX1 The gene replacement or directed mutagenesis methods have been widely used in the art and are perfectly reproducible.

The presently used method of producing strains deficient in several proteases, e.g. Strains lacking ysca and yscB activity have a meiotic crossing and subsequent tetrad analysis. The tetrads derived from the diploid cells are dissected by standard genetic engineering techniques. A statistical selection of the four spores of a tetrado allows the construction of double or multiple mutants in subsequent intersections. As an alternative system, statistical spore analysis may also be used: Since mutants lacking single proteases, and even double mutants are available from yeast genetic cell bankon, triple and quadruple mutants can be reproducibly combined by known, sequential meiotic crossover seductions.

As a starting point suitable strains of Saccharomyces cerevisiae are, inter alia, for. The kex 1 strain 96, de; from Yeast Genetic Stock Center, Berkeley, to yeast peptidase A (yscA) negative strains AB 103 (ATCC 20673) and AB 110 (ATCC20796) to yeast peptidase B (yscB) negative strains HT246, H426 and H449 (the latter is additionally cir °) deposited with the German Collection of Microorganisms, Braunschweig, FRG under accession numbers 4084, 4231 and 4413 respectively, yeast peptidase B.Y and S (yscB, yscY and yscSl-nogative strains BYS 232-31-42 and yeast peptidases A, B, Y and S (yscA, yscB, yscY and yscS) negative strains ABYS.

As stated above, the endogenous two-micron DNA is absent in the preferred yeast strains according to the invention.

The two micron plasmid is a high copy number self-replicating extrachromosomal DNA element present in most strains of Saccharomyces cerevisiae. The most prominent structural features of the two-micron plasmid are two 559 bp inverted repeat units (IR1 and IR2), which divide the plasmid into two DNA regions of different lengths. Homologous recombination between these two identical IR sequences results in the formation of two molecule isomers (Form A and Form B). The stability of the two micron plasmid is given by three t-unions encoded in the plasmid. The products of the REI 'I and REP2 genes are trans-acting proteins required for the stable distribution of the two-micron plasmid. Of these two, REP1 may be the more important because the efficiency of the distribution is dependent on the gene dosage of the product of the REP 1 gene (A.Cashmore et al., 1986, Mal. Gen. Gen., 201, 154). These two proteins act through and at the STB (REP3) site, an important cis-acting element of the plasmid (M.Jayaram et al., (1985) Mol. Cell. Biol. 5, 2466-2475; B. Viet et al. [1985] Mol. Cell. Bi.5,2190-2196).

Such cir ° strains of Saccharomyces cerevisiae are known, or may be prepared by methods known in the art (see, e.g., BCP Hollenberg! W-), Curr.Top. Microbiol. Immun.96,119). The following alternative methods for producing cir ° strains are based on the assumption that healing of the two micron plasmid by a second plasmid involves increasing the dosage of the STB site to titrate out the REP1 and RFP2 proteins. This relative reduction in REP1 and REP2 proteins would lead to instability of the endogenous two micron plasmid. Preferably, the second plasmid used has a defect in the REP-1 gene or this gene is completely absent. An example of such a plasmid is pDP38, which lacks an inverted repeating unit (IR 2) other than the REP1 gene. As a result, its high copy number expression becomes dependent on the complementation of the REP1 protein by the endogenous two micron plasmid. It contains two selective yeast tags: URA3, which is used in both high and low copy number situations, and dl.EU 2, which is only applicable in high copy number cases (E. Et hart et al. [1968], J.Backteriol.625). A yeast strain, which is Ura "and Leu", is transformed with the plasmid pDP38 and selected for Ura ⁺ colonies. The selection on uracil-free plates' Ura selection) leads to a much better transformation frequency than the selection on leucine-free plates (Leu selection), since the URA3 gene is expressed much better than the defective dLEU 2 gene , A single colony is selected and streaked on a leu selection plate, yielding colonies of different shape and size. Some of the smallest colonies are then in turn streaked on Ura selection plates and transferred in the Abklatschverfahren the plates on Leu selection plates. Those colonies are selected which can grow under the Ura selection conditions, but only grow very slowly under leu selection conditions. The growth on the urinary selection plates indicates that the plasmid pDP38 is still present and that the rather slow growth under the conditions of the leu selection is not due to the loss of this plasmid and the lack of growth under the conditions of the liver. Selection implies that the plasmid pDP38 is unable to complete this marker. This last observed fact can be explained in two ways: A. The LEU 2 gene in the plasmid pDP38 has been mutated or B. the plasmid can not complete leu2, as it can not increase its copy number, suggesting that the two- Micron plasmid is not available (ie was lost) to complete the REP1 gene product.

These two possibilities can be easily distinguished from each other. First, the minimal growth observed in these colonies (compared to zero absolute growth of cells without pDP38) indicates that some level of LEU 2 expression is present. The second point can be examined directly because in the absence of the two micron plasmid, pDP38 will only act as a plasmid of the ARS type, i. H. it is then very unstable so that most of the colonies will lose this plasmid after several generations. Accordingly, when a single colony is streaked onto an YDP plate and individual colonies are picked and transferred in the washdown to uracil-free plates, only some of these colonies will grow under the conditions of Ura selection. The non-growing colonies are examined by hybridization experiments for the presence of pUC and two micron sequences. Colonies that do not give a hybridization label are free of plasmid pDP38 and endogenous two micron plasmids (cir ° strains). The cir ° strains thus obtained may be further treated as described above to give yeast mutant strains which have no peptidase activity, in particular no yscα activity, and which additionally lack the two-micron DNA.

In addition to the lack of ysca activity in the preferred yeast strains of the invention, peptidases further selected from the group include: yscA, yscB, yscY, and yecS, and still lack the two '' micron DNA. The most preferred yeast strains have no ysca and yscY activity, and you may also lack yscB and ysoS activity or yscA and yscB activity, and they do not have two-megron DNA.

The yeast strains prepared according to the invention can advantageously be used for the production of heterologous proteins. Surprisingly, it has been found that kox 1 strains according to this invention carrying a hybrid vector containing a gene coding for a protein heterologous to yeast produce this protein in homogeneous form, i. H. that any by-products truncated at the C-terminus are missing.

More particularly, this invention relates to a method of producing a yeast strain which has no carboxyneptidase ysca activity and which has been transformed with a hybrid vector consisting of a yeast promoter operably linked to a DNA sequence encoding a hortoroprotein protoin.

Suitable yeast host strains include other strains of Saccharomyces cerevisiae lacking carboxypeptidase ysca activity and optionally the activity of other peptidases, from which endogenous two micron plasmids have been detected (Curing, supra).

The method of producing the genomic transformed yeast strains is to transform a yeast strain without carboxypeptidase-ysc α activity with said hybrid vector.

The transformation of yeasts with the hybrid vectors according to the invention can be carried out by the method described by Hinnen ot al. Natl., Acad., USA. [1978] 75, 129, 29). This method can be divided into three steps:

(1) Removal of the yeast cell walls or parts thereof using different preparations of glucosidaseri, e.g. From digestive juices of snails (eg Glusulase® or Helicase *) or enzyme mixtures obtained from microorganisms (eg Zymolyase®) in osmotically stabilized solutions (eg 1M sorbitol).

(2) Treatment of the "naked" yeast cells (spheroplasts) with the DNA vector in the presence of PEG (polyethylene glycol) and Ca ** ions.

(3) Regeneration of the cell walls and selection of the transformed cells in a solid agar layer. This regeneration is best done by embedding the spheroplasts in agar. So z. B. melted agar (about 5O ⁰ C) mixed with the spheroplasts. Upon cooling the solution to the temperature for growth of the yeast (about 3O ⁰ C) is obtained a solid layer. This agar layer prevents rapid diffusion and the loss of essential macromolecules from the spheroplasts, thus facilitating the regeneration of the cell walls. Regeneration of the cell walls, however, can also be accomplished by spreading the spheroplasts onto the surface of preformed agar layers (albeit at a lower efficiency).

Preferably, the agar is prepared for regeneration in such a way that the regeneration and selection of the transformed cells can be performed at the same time. Since yeast genes encoding the enzymes of the metabolic rate of amino acid or nucleotide bosynthesis are generally used as selective labels (markers) (supra), production is preferably carried out in minimal media for agar-based yeast. When very high regeneration efficiencies are required, the following two-step protocol is advantageous: (1) regenerating the cell walls in a rich complex medium; and (2) selecting the transformed cells by replicating the cell layer onto selective agar plates

According to this invention, desulphatohirudin of the formula (IV) Xi Tyr Thr Asp Cys Thr Glu Ser Gly GIn Asn Leu Cys Leu Cys GIu Gly Ser Asn VaI Cys Gly Gin Gly Asn X ₃ Cys He Leu Gly Ser Asp Gly GIu X ₃ Asn GIn Cys VaI Thr GIy GIu GIy Thr Pro X ₄ Pro GIn Ser X ₆ Asn Asp GIy Asp Phe GIu Glu le Pro GIu X ₆ .

where Xi is the dipeptide moiety VaI-VaI, X ₂ , X ₃ and X _{4 are} Lys, X _{5 is} His and X _{8 is} Glu-Tyr-Lys-Arg, Ser-Phe-Arg-Tyr or Trp-Glu-Leu -Arg can be.

The known heterologous proteins which can be prepared by the process according to the invention can be used in a manner known per se, e.g. As in the therapy and prophylaxis of diseases in humans and animals. So z. For example, human ANP has natriuretic, diuretic and vasorelaxant effects and can be used in the regulation of cardiovascular homeostasis.

The desulfato-hirudin derivatives, analogous to the naturally occurring hirudin, can be used for thrombosis therapy and prophylaxis, acute shock therapy, coagulopathy therapy, and the like, as described in European Patent Application No. 168,342 ,

In particular, this invention relates to the preparation of the transformed yeast strains for the production of heterologous proteins, as described in the Examples.

embodiment

In the following experimental part, various embodiments of this invention will be described with reference to the accompanying drawings, in which:

Fig. 1: shows chromatographic curves of desuifatohirudins derived from transformed KEX1 and kex 1 strains of

S. cerevisiae were harvested

 Fig. 2: is a schematic diagram showing the in vitro synthesis of the hirudin HV 1 gene including the PHO5 signal sequence with the preferred stapled codons. The 21 oligonucleotides used were indicated by numbered lines and dashed lines.

Fig. 3: Schematic representation of the construction of the plasmid pDP33

 Fig.4: illustrates in the scheme the construction of the plasmids pDP34 and pDP38. Fig. 5: Schematic representation of the construction of the expression plasmid pDP34 / GAPEL-YHIR.

Fig.6: Schematic representation of the construction of the expression platform pPP34 / PHO5 (-173) -YHIR. Fig.7: Schematically illustrates the construction of the plos. Nld pDP92.

Fig. 8: shows chromatographic curves of hirudin from Wll strain / i and don hirudin mutants HV1 -KR, HV1-WQLR and HV1-SFRY from cultures of S. cerevisiae BYSKEX 1 and BYSkex 1.

Example V. Crossing of S. cerevlslae kexi mutants 96 with the S.cerevlslae strain BYS and analysis of the spores for the secretory capacity of the α-factor and the carboxypeptylase ysca-Aktlvltat (KEX 1-Qen) The strain 96 of S. cerevisiae mutant kex1 (a, kex1, ade2, thr1), obtained from Genetic Stock Center, Berkeley, USA, for yeast is inserted into the yeast strain S. cerevisiae BYS232-31 -42 (α, prb 1 -1, pre 1 -1, eps 1 -3, Iys2, leu2, his7) (T.Achstetter and DHWolf, EMBO J. [1985] 4,173-177; DHWolf and C.Ehmann, J. Bacteriol. [1981] 147,418 -426), which carries the AIIeIe for the wild type KEX1. Diploid heterozygous cells of genotype kex1 / KEX1 are isolated from this cross. The tetrads derived from the diploid cells are read by standard genetic engineering techniques (DCHawthorne and RKMortimer, Genetics (1960) 45, 1085-1110; Methods in Yeast Genetics, F. Sherman et al., Ed., Cold Spring Harbor Laboratory, NY, 1986) The four spores of each nibble are tested for their ability to segregate the α-factor, to distinguish between KEX1 wild-type and kex1 mutant colonies, the pheromone-supersensitive test strain S is used. cerevisiae RC629 (a, sst-2, ade2-1, ura1, his6, met1, can 1, cyh2, rme) (RKChan, CKotte, Mol. Cell. Biol. [1982] 2,11-20 RK Chan, CK Otto, Mol. Cell Biol. [1982] 2: 21-29) As expected from the behavior of features encoded by individual nuclear genes, two spores of each tetrad are distinguished from all the analyzed tetrads the a-factor decreases while the other two spores release the α-factor, colonies of the wild-type KEX1 of the α-crossing type inhibit this s growth of the test strain to a large extent and create a large courtyard around them, as they are able to fully process the precursor of the α-factor and so four active a-factor molecules from a. Generate precursor molecule. In contrast, the colonies of the kex1 mutant inhibit the growth of the test strain to a lesser extent, thus creating a small courtyard around them, as they are only able to generate a crossing a-factor molecule from a precursor molecule.

The spores of several complete tetrads identified as defective for the kex 1 gene using the pheromone-tostane set described above are finally assayed for their specific activity of carboxypeptidase ysca. The cells are grown, membranes are prepared and for their carboxypeptidase ysca activity using Cbz-Tyr-Lys-Arg as the substrate as described (JCWagner and DHWolf, FEBS Lett. 221 [1987], 2,423-426). The fact that kex1 mutant cells lack carboxypeptidase ysca activity indicates that KEX1 is the structural gene of this enzyme. This suggests that carboxypeptidase ysca is indeed involved in the processing of the α-factor at the carboxyl terminus.

Example 2: Classification of Confirmed kex 1 Mutants for Additional Lack of Proteases yscB, yscY and yscS Kex 1 mutants of S. cerevisiae are deficient in other proteases (proteinase yscB, carboxypaptidase yscY and carboxypeptidase yscY) and additionally require growth factors classified.

Cellular material from kex 1 mutants prepared from the stationary phase in YPD (Difco) medium is suspended in 200 μM of 2OmM Tris-HCl buffer, pH = 7.2, in an Eppendorf tube and up to two-thirds of the volume Glasperlon with a diameter of 0.4 mm added. This suspension is thoroughly mixed three times a minute on a vortex mixer and cooled occasionally with ice.

By centrifugation (5 minutes), the crude extract is obtained as the supernatant. These extracts are dialyzed against 0.1 M imidazole HCL buffer, pH 5.2 with 0.1 mM ZnCl ₂ to activate proteases and to remove free amino acids from the extracts.

The activities of the proteinase yscB are measured according to the Azocoll test (REUlane et al., J. Biol. Chem. [1976J 251, 367], E. Cabib et al., Biochem., Biophys., Res., Commun., [1973] 50,186 T. Saheki et al., Eur. J. Biochem. [1974] 42, 621). Following protein concentration measurements, an aliquot of each sample is made up to 100 μΙ with 0.1 M sodium phosphate buffer (NaPi), pH = 7.0, to adjust for the same amount of protein needed. To the protein solution is added a suspension of 500μΙ Azocoll (240mg in 10ml 0.1M NaPi buffer, pH = 7.0). These mixtures are incubated with stirring for one hour at 30 ° C. After the addition of 500 μl of 10% trichloroacetic acid, whereby the reaction is stopped, the mixtures are centrifuged twice and the absorption spectra of the supernatants are measured at 520 nm.

The activities of the carboxypeptidases yscY and yscS are measured using the chromogenic substrate Cbz-Gln-Leu (compare DHWolf et al., FEBS Lett. [1978] 91, 59; DHWolf et al., Eur. J. Bioc '. iem [1977] 73,553). The dialyzed extracts are divided into three portions and to either of them is added either phenylmethylsulfonyl fluoride (PMSF) at a final concentration of 1 mM or EDTA at a final concentration of 5 mM to selectively block the two protease activities. Namely, PMSF inhibits the activity of carboxypeptidase yscY and EDTA inhibits that of carboxypeptidase yscS. The mixtures with the inhibitors are each incubated for one hour at 25 ° C to complete the inhibition. After determining the protein concentration, two aliquots of inhibitor and an aliquot without inhibitor as a control from each sample are made up to 50 μΙ with 0.1 M NaPi buffer, pH 7.4, to adjust for equal protein levels. To these protein solutions, the following assay solutions are added:

500μΙ examination solution I:

L-amino acid oxidase 0.24 mg / ml

Horseradish peroxidase 0.40mg / ml

0.0ImMMnCl ₂

in 0.1 M NaPi buffer, pH 7.4

50μΙ examination solution Il

o-dianisidine 2mg / ml

in water

500μΙ Examination Solution III: 20mMCbz-Gly-Leu

in 0.2 M potassium phosphate buffer, pH = 7.4

-ίo- 2 ^ 3

The gemisols are incubated for one hour at 28 ⁰ C and after the addition of 100μΙ 20% Triton X-100- to stop the reaction - the absorbents are milled at 405nm.

For purposes of subsequent transformation, an auxotrophic amino acid leucine strainer is streaked by Roplica technique on minimal plates supplied with adenine, threonine, lysine and histidine and with or without leucine. The assay described above isolates mutants termed S. cerevisiae BYSkex 1, which have a quadruple proto-insal deficiency (α, prb-1, prc-1, cps-3, kex 1) and additionally require leucine.

Example 3: Transformation of the Saccharomyces cerevlslae mutant with fourfold protease deficiency

The plasmid pJDB207 / PH05-HIR (European Patent Application No. 2225633) is in the mutant with the four-fold protease deficiency BYSkex 1 (α, prb-1, prc-1, cps-3, kex-1, leu2) and in the Wild strain KEX1 BYS232-31 -42 (α, prb-1, prc-1, cps-3, lys2, leu2, his7) was introduced as a control using the transf nmation protocol described by Hinnen et al. has been described. (Proc Nat Nat. Acad., USA, 1978, 75, 192S).

The two different yeast strains are harvested in the early logarithmic stage in 100 ml of YPD medium (OD ₆₀₀ = 0.2), washed with 25 ml of 0.8 M sorbitol and re-tended in 5 ml of the same sorbitol solution. To these 5 ml of cell suspensions are added 30 μM zymolyase (ZYMÖLYASE-100 from Arthrobacter luteus, Seikagaku Kogyo Co., Ltd., 10 mg / ml in 0.8M sorbitol). These mixtures are each gently shaken in 100 ml shake flasks on the shaker (110 revolutions / min) at 30 ° C for about 30 minutes. An aliquot of 5 μl is withdrawn at 5 minute intervals, diluted with 10 ml of distilled water and the absorbances of the dilute solutions measured at 600 nm to control the progressive course of spheroplast formation. In order to obtain good spheroplast formation, the difference in absorbances before and after the addition of zymolyase must be greater than a factor of 10. The cells transformed into spheroplasts are washed twice with 0.8 M sorbitol, resuspended in 25 ml of HE 30 medium (2 M sorbitol in YPD medium) and shaken in a 100 ml shake flask for one hour at 30 ° C. on a shaker with gentle shaking ( 110 revolutions / minute). The cells are centrifuged (3000 rpm, 5 minutes) and carefully resuspended in 1 ml HE-31 solution (10 mM Tris-HCl, pH 7.5, 0.1 mM CaCl ₂ , 0.9 M sorbitol). 4 μg of plasmid DNA are added to this 100 μl cell suspension. This mixture is incubated for 15 minutes at room temperature. To each tube, 1 ml of 20% polyethylene glycol (PEG) 4000 is added and incubated for a further 30 minutes at room temperature. It is then centrifuged (3000 rpm, 3 minutes) and resuspended in 500 μ10.8 M sorbitol. The spheroplasts with DNA are supplemented with 10 ml regeneration agar (1M mannitol, 6.8 g / l yeast nitrogen base w / o AA [Difco], 10 g / l L-apparagine, 1.0 g / l L-histiidine, 1 , 0g / l adenine, i, 0g / l threonine, 1.0g / l lysine and 3% agar) and poured as a topsheet onto plates having a base layer of the same composition. The plates are 96 hours be> 3O ⁰ C incubated until the transformed colonies appear.

A single colony of transformed yeast is now taken out and grown in synthetic minimal medium for yeast (8.4 g / L yeast nitrogen base w / o AA, 10 g / L L-asparagine and 1.0 g / L L-histidine), which also contains adenine, threonine, lysine and histidine but no leucine. The colony of the mutant BYSkex 1 with fourfold protease deficiency and that of the control strain BYSKEX 1 with triple protease deficiency are designated as Saccharomyces cerevisiae BYSkex 1 / PJDB207 / PH05-HIR and Saccharomyces cerevisia * BYSKEX1 / pJDB207 / PH05-HIR.

Example 4: Cultivation of Saccharomyces cerevlslae transformants with pJDB207 / PH05-HIR cells of the Saccharomyces cerevisiae transformants BYSkex 1 / pJOB207 / PHO-HIR and BYSKEX 1 / pJDB 207 / PH 05-HIR are prepared as precultures in 10 ml complete yeast medium HE41! 48 ⁰ C. (4.5 g / l casamino acids, 4 g / l yeast extract, 20g / l sucrose, 20 g / l glucose, 3.6 g / l [NH ₄ I ₂ SO _4, 0.2 g / l MgSO ₄ · 7H ₂ 0,0,013g / l CaCl ₂ · H ₂ O and 1 ml / l Spurer.elementgemisch:

10g / l FeSO ₄ · 7H ₂ 0.50 g / l ZnSO ₄ · 7H ₂ 0,3,3g / l CuSO ₄ · 5H ₂ 0.3g / l MnSO ₄ · H ₂ 0.2 g / l CoCI ₂ 6H ₂ O and 1 g / l (NH ₄ I ₆ Mo ₇ O ₂₄ .4H ₂ O) and cultured for approximately 48 hours until the stationary phase is reached. The harvested cells are washed in 0.9% NaCl. 50 ml of the above-described synthetic yeast medium are inoculated with 5% of an inoculum.

The cultures are seeded to a cell density of OD ₆ Oo = 0.3 and agitated for up to 72 hours at 1 "C and 250 rpm.

The complete yeast medium HE41 is used to measure the background absorbance in the HPLC. to suppress .lalyse.

Example 5: Analysis of hirudin-65 and its degradation product at the carboxyl termini hirudin-64 and hirudln-63 from fermentation cultures of Saccharomyces cerevisiae transformants BYSkex1 / pJDB207 / PH05-HIR and BYSKEX1 / pJDB207 / PH05-HIR using reverse phase HPLC

Samples of liquid yeast cultures are prepared by centrifugation to produce a clear solution, which is then diluted 1:10 (v / v) with 1 M acetic acid and subjected to HPLC analysis under the following conditions. A HIBAR (MERCK) column (4mm χ 125mm) is filled with the reverse phase pore-filled silica gel (Type 71252, 300-5-C18, MACHEREY-NAGEL), a spherical stationary phase having a particle diameter of 5 μm and a porosity from 300A. The ends of the column are equipped with stainless steel frits. Mobile phase A is prepared from water (Nanopure ™, BARNSTEAD) with 0.1% (v / v) trifluoroacetic acid Mobile phase B is prepared from 20% mobile phase A and 80% (v / v) acetonitrile (HPLC-pure , FLUKA) with 0.08% (v / v) trifluoroacetic acid The chromatographic separations are performed at a flow rate of 1.5 ml / min using the following gradient and the eluted fractions characterized by their absorbance at 216 nm.

90 10 79 21 79 21 67 33 0 100 0 100 90 10 90 10

9 17 20 22 24 32/0

A standard solution for calibrating the system is prepared by gcitrating 1 mg of pure desulphatohirudin in 1 ml of water. 50 μL of this standard solution is injected into the column and chromatographed as described to calibrate the system.

Figure 1 shows the curves of analysis with reverse-phase liquid chromatography of hirudin harvested from cultures of S. cerevisiae BYSkex1 / pJDB207 / PH05-HIR and S.cerevisiae BYSKEX1 / pJDB207 /. ~> H05-HIR. The conditions for dioxygen chromatography were the same as described above. It is clearly evident that in contrast to the strain BYSKEX1 / pJDB207 / PH05-HIR of 3r. Peptidase ysca negative strain BYSkexi / pJDB207 / PH05-HIR produce a dosulfatohirudin product (HIR-65), the substantially free C-terminus truncated byproducts desulphatohirudin HIR-64 lacking the C-terminus amino acid GIn, and HIR-63, which lacks the amino acids Leu and GIn at the C-terminus, is.

Example 6: In Vitro Synthesis of the Hirudln HVI Gene with Preferred Yeast Codons The hirudin expression cassette coding sequence is constructed with preferred yeast codons (B. Hall, J. Biol. Chem.

(1982) 257, 3026) that optimal translation of the hirudin mRNA is guaranteed. The coding sequence contains the PH05 signal sequence fused in-frame to the coding sequence of desulphatohirudin HV1. The 5 'end of the synthetically produced DNA contains the cohesive ends of the EcoRI restriction site.

At the 3 'end, the stop codon TAG immediately follows the cohesive end of the BamHI site. The sequence of the 257 bp EcoRI-BamHI DNA fragment is shown in FIG.

Figure 2 also shows the strategy for the synthesis of the double-stranded DNA fragment in vitro. It

For example, 21 oligonucleotides are synthesized using the phosphoramidite method (MH'Caruthers in: Chemical and Enzymatic Synthesis of Gene Fragments, publisher HG Gassen and A. Lang, Verlag Chemie, Weinheim / FRG, 1982) using an automated synthesizer from Applied Bicsystems, Model 380 B, is used. The sequence of the individual oligodeoxynucleotides is shown in FIG. The overlaps are unique. The lyophilisiortan oligonucleotides are redissolved in 50 mM Tris-HCl buffer, pH = 8.0, at a concentration of 10ΜΜΜ / μΙ. The 21 oligonucleotides are arranged in two groups: (A) Nos. 1 to 11, which represent the 5 'portion of the DNA fragment, and (B) Nos. 12 to 21, for the 3' half. The two groups are implemented separately. 1OpMoI of each of the oligonucleotides of one group are mixed. The oligonucleotides are phosphorylated in 25 mM Tris-20μΙ hlCI buffer, pH = 8,0,1OmM MgCl _2, 1OmM NaCl, 3 mM DTT, 0.4 mM ATP and 8 units Polynucluotid kinase (Boehringer) for 1 hour at 37 ⁰ C. After 30 minutes. At room temperature, the two mixtures (A and B) are each heated to 95 ^° C. on a water bath for 5 minutes. The samples are then allowed to cool slowly to room temperature in this water bath overnight. The molten oligonucleotide mixtures A and B are then stored on ice.

The plasmid pBR 322 is completely cut with EcoRI and BamHI. The large 4 kb fragment is isolated using a preparative 0.6% agarose gel. The DNA is recovered by electroelution, purified by ion exchange chromatography on DEb2 and ethanol precipitation as described in Example 7. Dia DNA is remelted in water at a concentration of 0,4ρΜοΙ / μΙ.

10μΙ of the molten oligonucleotide mixture A (each 5μmol of each of oligonucleotides 1 to 11), 9.5μΙ of mixture B (each 5μmol of each of oligonucleotides 12 to 21), 0.4pMoi of the 4kb EcoRI-BamHI fragment of p3R322 and 400 units of T4 DNA ligase (Biolabs) are incubated bsi 15 ⁰ C for 16 hours.

ΙΟμΙ aliquots are used to transform competent E. coli HB101 Ca "" cells. 12 transformed, ampicillin resistant colonies are individually grown in LB medium containing 100 μg / ml ampicillin. The plasmid DNA is prepared by the method of Holmes et al. (Anal., Biochem 114 [1981], 193) and analyzed by EcoRI and BamHI restriction digestion. Plasmid DNAs carrying the 257 bp insert of E. coli BamHI are further analyzed by DNA sequencing of both strands. Oligonucleotides 3, 11, 12, 14 and 20 (see Figure 2) are used as primers for sequencing. A clone with the correct sequence of both DNA strands is selected and designated as pBR 322 / YHIR.

Example 7: Construction of the piasmide pJDB207 / GAPFL-YHIR

pJDB207 / GAPFL-YHIR is a yeast plasmid for expressing the desulfatorhirudin variant HV1 under the control of a short, constitutive promoter of the yeast glyceraldehyde S-phosphate dehydrogenase (GAPDH) gene. The coding sequence of desulphatohirudin consists of preferred yeast codrns.

Vi \ ig of Piasmids pBR322 / YHIR are degraded with restriction Endonucloasen uamHlund EcoRI. The 257bp EcoRI-BamHi fragment is separated from other DNA fragments using a 1.2% agarose gel. The DNA bands were stained with ethidium bromide and visualized under UV light at 360 nm. The band of 25V bp DNA was cut out of the gel and placed in 0.2 χ TBE buffer (TBE: 9 mM Tris base, 9 mM boric acid, 2.5 mM EDTA, pH = 8 3) within 45 minutes at 100 mA removed by electroelution. After switching the polarity for 45 seconds, the ONS solution is separated and adjusted to a concentration of 0.15 M with NaC. The DNA is absorbed on 100μΙ DE 52-ion exchanger (Whatman) and Γηΐΐ400μΙ buffer with high salt concentration OOmMTris-HCl, pH = 8.0 1mM EDTA, 1.51V NaCl) eluted. The DNA is precipitated by ethanol and resuspended in water at a concentration of 0.1 pmol / μΙ.

Plasmid pJDB2O7 / GAI FL-HIR (European Patent Application No. 225,333) contains the synthetic gene for desulphatohirudin (based on the use of E. coli condon) which is in frame with the yeast acid phosphatase signaling sequence (PH05). is merged. The gene is expressed on the binary vector pJDB207 under the control of a short, constitutive yeast glycerol-3-phosphate dehydrogenase (GAPFL) promoter. 10 μg of plasmid pJDB207 / GAPFL-HIR are digested with Sail and EcoRI, and the 478 bp SalI-EcoRI fragment contains the Sall-BamHI-pBR322-Toil and don GAPFL promoter. The DNA fragment is isolated using a 0.8% preparative agaroso gel, eluted by electroelution and purified by chromatography on DE 52 and precipitation with ethanol. The DNA is resuspended in water at a concentration of 0.1 pmol / μΙ. 5pg of pJDB207 / GAPFL-HIR are degraded with Sail and BamHI. The large 6.7kb vector fragment is isolated as described above.

0.2 pmoles of the 478 bp SalI EcoRi promoter Fragmants, 0.2 pmol of the 257 bp EcoRI-BamHI fragment containing the PH 05 signal sequence and the synthetic hirudin gene (yeast codons), and 0.1 pmol of the 6.7 kb vector fragments are digested in ΙΟμm of 6 mM Tris-HCl, pH = 7.5, 10 mM MgCl ₂ , 5 mM OTT, 1 mM ATP and 200 units of T4 DNA ligase, (Biolabs) 6 Hours at 15 ⁰ C ligated. An aliquot of 1μΙ of the ligation mixture is used to transform competent E. coli HB 101 cells. 12 transformed ampicillin-resistant colonies are independently grown on LB medium containing 100 μg ampicillin per ml. The Pia jtnid DNA is prepared by the method of Holmes et al. (see above) and analyzed by Sall / HindIII double degradation. A single clone with the expected restriction pattern is referred to as pJDB207 / GAPFL-YHIR. In an analogous manner, construction may be performed on a 543 bp SalI-EcoRI promoter fragment of plasmid pJDB207 / GAPEL-HIR (European Patent Application No. 225,633). The new plasmid so prepared is designated PÜDB207 / GAPEL-YHIR.

Example 8: Construction of plasmid pJDB207 / PH05 (-173) -HIR

pJDB207 / PH05 (-173) -HIR is a yeast plasmid for the exudation of the desulphatohirudin variant HV1 under the control of a short PHO6 promoter. The P! L05 (-173) promoter element consists of the nucleotide sequence of the yeast PH05 promoter from position -9 to -173 (BstEII restriction site), but has no regulatory sequences ( UAS). The PH05 (-173) promoter therefore behaves like a constitutive promoter.

Plasmid pJDB207 / PH05 (Eco) -HIR (EP 225,633) contains the full-length regulated PH05 promoter with an EcoRI site introduced at position -8 for the ATG of the PH05 signal sequence, and the coding sequence desulphatohirudin followed by the PH05 transcription termination fragment. In this example, the replacement of the regulated PH05 promoter by the short PH05 (-173) promoter element is described.

20 μg of the plasmid pJDB207 / P H05 (Eco) -HIR are digested with BstEII. The cohesive ends of the restriction fragments are cleaved by reaction with KlonowONS polymerase (1 unit / Mg DNA) in 200 μM 60 mM Tris-HCl, pH = 7.5, 10 mM MgCl ₂ , 0.1 mM of each dATP, dCTP. dGTP, TTP filled at room temperature within 30 minutes. After dur extraction with phenol, the DNA is precipitated with ethanol.

4.16 μg of the BamHI linker (5'-CGGATCCG-S ', Biolabs) are dissolved in 100 μl 6OmM Tris-HCl, pH = 7.5, 10 mM MgCl ₂ , 5 mM DTT, 0.5 mM ATP and 18 units of the T4 polynucleotide. Kiraise (Boehringer) was phosphorylated at 37 ° C for 45 minutes. After heating to 75 ^° C. for 10 minutes, the reaction mixture is slowly cooled to room temperature. The fused oligonucleotide linkers are stored at -2O ⁰ C.

4 pmol of the fragment? [BstEII] / blunt end of plasmid pJDB207 / PH05 (Eco) -HIR are incubated for 16 hours at 15 ⁰ C with a 100-fold excess of the phosphorylated and fused BamHI linker in 20ϋμ! 6 mM Tris-HCl, pH = 7.5, 10 mM MgCl ₂ , 5 mM DTT, 3.5 mM ATP and 800 units of T4 DNA ligase (Biolabs). Subsequently, the ligase is inactivated within 10 minutes at 85 ⁰ C and the excess linker removed by precipitation of the DNA in the presence of 1OmM EDTA, 30OmM sodium acetate, pH = 6.0 and 0.5 * »volumes of isopropanol, the DNA with BamHI and EcoRI degraded. The DNA fragments are separated in a preparative 0.8% agarose gel. The 172 bp BamHI-EcoRI promoter fragment is recovered from the gel by electroelution and precipitation with ethanol. The DNA is resuspended at a concentration of 0.1 ρΜοΙ / μΙ.

The plasmid pJDB207 / PH05 (Eco) -HIR is digested with EcoRI and HindIII. The 643 bp EcoRI-HindIII fragment is isolated as described above. The DNA fragment contains the PH05 signal sequence fused in-frame to the desulphatohirudin coding sequence and the PHOδ transcription termination fragment. The plasmid is also cut with HindIII and BamHI. The 6.6kb vector fragment is isolated.

0.2 pmol of the 172 bp BamHI-EcoRI fragment and the 643 bp EcoRI-HindIII fragment and 0.1 pMoldes of the 6.6 kb vector fragment are dissolved in 10 μl of 60 mM Tris-HCl, pH-7.5, 0.1 mM MgCl _2, 5 mM DTT, 1 mM ATP and 400 units of T4 DNA ligase (Biolabs) for 6 hours at 15 ⁰ C. An aliquot of 1 μM of the ligation mixture is added to ΊΟΟμΙ of calcium-treated, transformation-competent E. coli HB101 cells.

12 transformed, ampicillin resistant colonies are grown in LB medium containing 100pg / ml ampicillin. The plasmid DNA is prepared by BamHI and ScII / HindIII digestion and analyzed. A clone with the expected restriction fragments is selected and designated pJDB207 / PH05 (-173) -HIR.

Example 9: Construction of plasmid pDP34

Yeast-derived 2-micron DNA covalently closed to the circle is isolated from Saccharomyces cerevisiae strain S288C. The cells v / t with the 5 .mu.g / mlZymoylase (100000 units ^ g) within 20 minutes at 37 ⁰ C incubation to degrade the cell walls. C \ z Sphäropiasten are lysed with 2% SDS. Thereafter EDTA is added to a final concentration of 25 mM, ethidium bromide up to 1 mg / ml and cesium chloride to a final density of 1.55 g / ml. Plasmid DNA is separated from the chromosomal DNA by ultracentrifugation 42stündiges bei42000U / min and 15 ⁰ C. The 2-micron plasmid DNA is excised from the graduate by syringe. The ethidium bromide is removed by extraction with NaCl-saturated isopropanol and the Pksmid DNA is finally precipitated by ethanol. The purified two micron pldsmid DNA is then linearized with PdtI and cloned into the PstI site of pUC19 (J. Norrander et al., Gene [1983] 26, 101) to produce the plasmin pDP31.

The plasmid pJDB2O7 is digested with the restriction enzyme Kpnl and Hpal. The O, 55kb Hpal-Kpn I fragment thus produced contains the intersection between the two micron Sequen / and the defective promoter of the dLEU 2 gene. The plasmid pUC7 / LEU 2 contains the genomic yeast fragment of 2.2 kb length Xho I-Sal I of the LEU 2 gene (A. Amdreadis et al., Cell (1982) 31, 319), which clones in the Sall site of the plasmid pUC7 Vieira et al., Gene (1982, 119, 259) The plasmid pUC7 / LEU 2 is cleaved with Kpn I and Hpa I. The 4.25 kb lango Kpn I-Hpa I fragment is added to the 0.55 kb This results in the plasmid pDP30, wherein the fusion product of the original two-micron / dLEU 2 is located in front of the LEU 2 gene with its full terminator as in plasmid pJDB207 Hpa I and Sal I are digested and the 1.85 kb fragment containing the complete LEU 2 gene is purified and cloned into the 8.7 kb Sal I-Hpa I fragment of plasmid pDP31 The plasmid prepared thereby pDP33 (see Figure 3) is degraded by partial digestion with Hind III in the presence of 50 μg / ml ethidium bromide (M. Esterlund et al., Gene (1982) 20, 121) and with 1.17 k b-Hind HI fragment containing the URA3 gene ligated (M. Rose et al. Gene (1984) 29,113). The insertion of the URA3 gene is selected by transformation into the E. coli strain pyrF (Rose et al., Supra). A positive clone is called plasmid pDP34 (see Figure 4).

pDP34 is a binary yeast E. coli vector with the ampicillin resistance tag for E. coli and the selective tags URA3 and dLEU 2 for yeast. It contains the complete zwi-micron sequence in the Α-form and is proficient for REP1, REP2 and FLP.

Example 10: Cloning of hirudin expresson cassettes in pDP34

The plasmid pDP34 is digested with BamHI. The cohesive ends of the restriction site are filled in by a reaction with Klenow DNA polymerase (Maniatis, T., et al., Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory, 1982.) DNA is further cut with Sail and The 11.8kb vector fragment is isolated via a 0.6% preparative agarose gel The DNA is recovered by electroelution and precipitation with ethanol Between the sites of the blunt ends of Sail and [BamHI] various expression cassettes are cloned into the pDP34 vector fragment The plasmid pJDB207 / GAFPL-YHIR is digested with Hind III, and the cohesive ends are blunt-ended by Klcnow DNA polymerase, which is precipitated by ethanol and further digested with Sal I. The 1.1 kb fragment of the blunt end of Sal I- [Hind III] contains the complete expression cassette with pBR322 sequences, the GAPFL promoter, the PHO5 signal sequence, which is aligned in frame to the coding sequence (preferred yeast codons) of the D esulphate hirudin and the PHO5 transcription termination fragment. The 1.1 kb fragment is isolated on a 0.8% preparative agarose gel, recovered from the gel by electroelution and purified by ion exchange chromatography on DE52 and precipitation with ethanol. 0.2 pmol of the 1.1 kb fragment and 0.1 pmol of the 11.8 kb vector fragment are dissolved in 10 μl of 60 mM Tris-HCl, pH = 7.5, 10 mM MgCl ₂ , 5 mM DTT, 3.5 mM ATP and 400 units T4 DNA ligase (Biolabs) within 16 hours at 15 ° C ligated together. An aliquot of 1 μΙ is used to transform E. coli HB101 Ca ²⁺ cells. 5 transformed, ampicillin resistant colonies are analyzed. The plasmid DNA is digested with BamH I and SalI / BamHI. One clone with the correct restriction fragments is selected and designated pDP34 / GAPFL-YHIR (see Figure 5). '

In an analogous manner, the blunt-ended 1.2 kb fragment of SalI (Hind III) of pJDB 207 / GAPFL-YHIR (see Example 7) is cloned into the pDP34 vector, whereby the plasmid pDP34 / GAPEL-YHIR is obtained.

The plasmid pJDB207 / PHO5 (-173) -HIR is digested with Sail and EcoRI. The 448bp fragment of Sall-EcoRI is isolated as described. The DNA fragment contains the Sal I-Bam H I portion of pBR322 and the short, constitutive promoter PHO5 (-173) (see Example 8). The plasmid pJDB 207 / GAPFL-YHIR is degraded with HindIII. The cohesive ends are converted to blunt ends by Klenow DNA polymerase. The DNA is further degraded with EcoRI. The truncated end fragment of EcoRI- [Hir, d III] mi *. 642 bp is isolated. It contains the PHO5 signal sequence, the coding sequence of desulphatohirudin (with the preferred yeast codons) and the PHO6 transcription termination fragment. 0.2 pmol of each of the 448 bp Sall-EcoRI. fragments and the 642bp fragments of the blunt. At the end of EcoRI and 0.1 pmol, the 11.8 kb vector fragments of the blunt end of Sal I- [BamH I] are ligated mininim. Aliquots of this ligation mixture were used to transform E. coli HB101 Ca * ^T cells. The plasmid DNA of 12 transformants is analyzed by digestion with BamH I and Sal I / BamH I. A clone with the correct plasmid is selected and designated pDP34 / PHO3 (-173) -YHIR (see Figure 6).

Example 11: Further expression plasmids for the hirudin gene with E. coli codons Expression plasmids containing the synthetic gene for the desulphatohi.-udin variant HV1 based on the use of E. coli codons are applied to analogous Way, as described in Example 10, constructed.

The 1.1 kb fragment Sall- (HindIII) at the blunt end of pJDB207 / GAPFL-HIR (European Patent Application 225,633) is isolated and cloned into the vector pDP34 The expression plasmid produced thereby is pDP34 / GAPFL-HIR, which consists of the synthetic Gene for desulfetohirudin based on preferred E. coli codons expressed under the control of the constitutive GAPFL promoter.

For a similar construction, the 1.1 kb SaII (HindIII) radicle at the blunt end of the pJDB207 / PHO5 (-173) -HIR (see Example 8) is cloned into pDP34 The plasmid pDP34 / PHO5 ( -1731-HIR contains the synthetic gene for desulphatohirudin under the control of the short, constitutive PH05 (-173) promoter.

Example 12: The Hirudin Expression Cassette Cloned into the Plasmid ρΓ) Ρ92 Vectors having the full two micron sequence do not necessarily express all functions of the original two micron loop of the yeast. Open reading frames can be destroyed by cloning. So far no function was known for the gene product of the "D" reading frame, the only PstI site within this gene was to clone the dLEU 2 gene (JDBeggs, Nature [1978] 275, 104-109) or to insert the pUC19 Vector portion as used in plasmid pDP31 (see Example 9) Recently, it has been suggested that the D-gene product regulates the expression of the FLP gene product (JA Murray et al., EMBO-J (1987), 6 , 4205) In order to fully exploit the advantage of the two-micron circle, a vector was constructed which is proficient for all two-micron functions, including the D-gene product a) Construction of the plasmid pDP92 (see Figure 7).

The plasmid pDP31 (Example 9) is abyebaut with Pst I and Hpal to give three fragments. The plasmid pK19 (conferring kanamycin resistance; RDPridmore, Gene | 1987) 56, 309-312) is linearized with SmaI. The DNA Fragmonte from both degradation steps are extracted with phenol and precipitated with ethanol. The DNA fragments are mixed together and ligated together. The egg mixture is transformed into competent E. coli JM109 cells (DJ Hanahan, Mol. Biol. 119831166, 557-580). (C.Yanisch-Perron et al., Gene [1985] 33.103-119), 2 hours at 37 ⁰ C in LB medium expressed and Dana. , on LB agar plates supplemented with 50 μg / ml kanamycin, 30 μg / ml XGal and 7 μg / ml IPTG. Twelve white, kanamycin resistant colonies were grown. The plasmid DNA is analyzed by degradation with XbaI and BamHI / KpnI. A single clone that has lost the pUC 19 portion of the pDP31, restored the two micron D-type lector by relegating the Pst I site, and inserted the blunt end of the pK 19 plasmid into the Hpa I-Ot, is referred to as pDP91. The plasmid contains the small Hpal-PstI and small PstI-HpaI fragments of the two-micron plasmid, which are cloned into the SmaI site of pK19. Relocating the Pstl sites restores the D reading frame. The URA3 gene is isolated on a 1.17 kb Hind III fragment from plasmid pDP34 (see Example 9) and cloned into the single Hi.idIII site of plasmid pUC12. A clone with the URA3 gene inserted with the same orientation as the ampicillin resistance gene is called pUC12 / URA3. The plasmid pDP91 and the plasmid pUC12 / URA3 are digested with SalI and BamHI, from which each two fragments are obtained. The DNA fragments are ligated in the mixture and used to transform competent E. coli JM109 cells. The cells were plated on LB agar plates spiked with 100 μg / ml ampicillin, 30 μg / ml XGal and 7 μg / ml IPTG.

12 white, ampicillin resistant colonies are grown. The plasmid DNA is analyzed by digestion with Hind III and Pvu II. A single clone is referred to as pDP92, which contains the entire two micron sequence, which is proficient for all its functions, and also contains the URA3 gene cloned into the pUC vector. b) Cloning of the hirudin expression cassettes into the pDP92

Analogously to Example 10, pDP92 is degraded with BamHI. The cohesive ends are filled in by a reaction with Klenow DNA polymerase. The DNA is further cut with Sal I. The 10.2kb vector fragment is isolated. The 1.1 kb Sall- [Hindllll blunt end fragment of plasmid pJDB207 / GAPFL-YHIR is isolated and ligated to the vector fragment.

6 transformed anti-ampicillin resistant colonies are analyzed. The plasmid DNA is digested with BamHI, PstI and Sall / BamHI. A clone with the expected restriction fragments is selected and designated pDP92 / GAPFL-YHIR. In a similar manner, the plasmid pDP92 / GAPEL-YHIR is prepared using the 1.2 kb SaI l- [HindIII] long blunt end fragment from pJr ': 207 / GAPEL-YHIR (see Example 7).

Plasmid pDP92 / PHO0 (-173) is constructed as described in Example 10 using the 10.2kb SaI I- [BamHI] vector fragment from the blunt end of pDO92 (see above).

Example 13: Construction of a Saccharomyces cerevisiae host strain without the two micron DNA In order to remove the endogenous two micron plasmid, in a first step the URA3 gene of strain HT246 (OSM4084; a, leu 2-3, leu 2-112, prb) introduced a deletion to make the strain uracil auxotrophic. HT246 is supplemented with 1 μg of the plasmid YeP 13 (JR Broach, JNStrathern, JB Hicks, Gene [1979], 8, 211-123) using the method described by Hinnen et al. (Proc Natl Acad., USA 75 [1978], 1929). lOpg desP ^ asmids pUC l2ura3A, which has a deletion in the URA3 gene (Ch.Sengstag, A. Hinnen, Nucl. Acids Res. 15 [1987], 233-246) are added together with the plasmid YEP13. Approximately 3,000 phototrophic leucine transformants are resuspended in 5 ml of minimal medium (Difco Yeast Nitrogen Base without amino acids to which ~% glucose, 0.1% leucine, 0.1% uracil, and 0.25% fluorouritic acid have been added) in a small shake flask and incubated for 60 hours at 3O ⁰ C and 180 rev / min. The growing transformants are resistant to the toxic analog fluorouritic acid and therefore have a substitution for the chromasomal URA3 gene by the Ura 3Δ. The grown cells are plated on complete medium ausgesv cozy consisting of (g / l): Peptone 20, yeast extract 10, glucose 20. After the culturing for 48 hours at 3O ⁰ C on Minimainledium (Difco) (yeast nitrogen base without Amino acids) abgeklatscht. 2% glucose and 1% leucine were added to the minimal medium to detect uracil auxotrophic cells. Several auxotrophs were read and examined for the loss of plasmid YEP13, which carries the leucine auxotrophy. A single colony (designated Tr889) which requires leucine and uracil is read and used for further experiments.

Tr899 is transformed with the plasmid pDP38 (obtained from plasmid pDP34 by digestion with Sphl and re-alloying of the thus generated 8.4kb fragment, see Figure 4), which has both marker genes LEU 2 and URA3 (transformation protocol see above ). The transformed yeast cells are first selected on yeast medium plates without uracil, to which leucine has been added, and then transferred to Mirimalmedium plates without leucine but with uracil by grafting. 10 low growing colonies are picked and grown individually in liquid complete medium (see above) for approximately one hundred generations. By doing so, the cells lose the pDP38 plasmid and to a certain percentage also the endogenous two micron plasmid.

10 colonies needing uracil and leucine are picked, the DNA prepared and the DNA completely digested by Pst I and probed on Southern blots with yeast ³² P-labeled two micron DNA. An isolated colony without any hybridization signal is designated H449 (a, leu-2-3, leu 2-112, ura3A, prb, cir °), an isogenic, two micron DNA free (cir °) form of the yeast strain HT246 ,

Example 14: Preparation of a kex1 variant of S. can vlslaeH449 by cleavage of the genomic KEX1-Qens The carboxypeptidase yscu activity is derived from the S. cerevisiae strain H449 (prb, Ieu2, ura3, cir °) by cleavage of the genomic KEX1 -Genseleminiert. For this purpose, the KEX 1 gene is identified in flini genomic library for yeast and cloned into a suitable vector. The URA3 gene, which serves as a selective marker, is inserted into the structural gene of KEX1 to cleave its reading frame. The hybrid plasmid DNA consisting of the URA3 gene flanked on both sides by the KEX I sequences is introduced into the Ura yeast strain H449. The sequence homology of the KEX1 gene on the plasmid and on the chromosome allows for in vivo recombination, thereby transforming yeast cells from Ura "to Ura * and simultaneously from KEX1 to kex 1. Synthesizing strains with a cleaved kex 1 gene no functioning yscci protein.

The gene encoding KEX1 is isolated from a yeast genomic library (in the Contromer dan Pendulum Vector pCS 19, C, Songstag et al., Nucl. Acids Res. 15 11987], 233) by colony hybridization with a KEX1 specific oligonucleotide Probe cloned. The sequence of the following oligonucleotide

5'-GTCGAATCCGGCCCTTTTAGGGTGAATTCa-S '

is derived from the published KEX1 sequence (see A. Dmochowska et al., Cell (1987) 50, 573) and selected from the overall sequence for its particularly low homology to the sequence yscY. It hybridizes with the KEX 1-DN ¹ S sti. upstream from the EcoR V restriction site used to insert the URA3 gene. The same oligonucleotide can be used as a primer for sequencing to confirm the insertion of the URA3 fragment. This synthetic oligonucleotide is radiolabeled and used to screen the library.

Approximately 10000 clones (5 x 2000 clones / plate, (0 = 140mm) are screened by colony hybridization (see, DEWoods et al., Proc Natl Acad., U.S.A., 79, 1982, 565 and ASWhitehead et al Ac., USA, 80 (1983), 5387) Two repetitive screening runs isolate five independent, positive clones, one of which is designated pKEX 1. In order to obtain a KEX1-specific Hind III BamHI fragment (1380bp) is digested, the pKEX1 plasmid is degraded with the appropriate two endonucleases, the appropriate fragment is transferred to the Bluescript vector M13 + with the SK polylinkar, and this clone is cloned using a universal primer for the M13 Vector sequenced to confirm the KEX1 fragment and designated pKEX 1M13.

The plasmid pUC 12 / URA3 (see Example 12a) containing the URA3 gene cloned as a Hind III fragment is digested with HindIII to isolate the 1179 bp Hind HI fragment (see M. Rose et al., Gene 29 [1984], 11G). The cohesive ends of the fragment are filled in with Klenow polymerase to create blunt ends. On the other hand, the plasmid pKEX 1 M13 is linearized by digestion with the endonuclease EcoR V, whereby the KEX1 HindIII BamH I fragment is cut and dephosphorylated with alkaline phosphatase to prevent self-ligation. Subsequently, these two DNA fragments are miteinanciur mixed, ligated and transferred to the E. coli strain JM103 by transfection. The transformants are analyzed by double digestion with Hind III and BamH I, whereby the size of the considered Hind HI-BamHI fragments increases from 1380 bp to 2 550. The DNA of a correct clone is sequenced using the oligonucleotide primer described above as a sequencing initiator to confirm the insertion of the URA3 fragment into the KEX 1 gene at the position of the Eco RV cleavage site. The plasmid is referred to as pKEX1 M13-URA3. The DNA of pKEX 1 M13-URA3 is digested with Hind III and BamHI to excise the KEX 1-URA3 hybrid fragment and then incorporated into the S. cerevisiae strain H449 without separation from the vector by the method described above (see Example 3). , After isolation of the membrane fraction from the Ura3 * -transformanton, the activity of ysca is measured using a chromogenic substrate (see Example 1). A single transformant having no protease activity of ysca is termed S. cerevisiae H449kex1.

Example 15: Transformation of S. cerevisiae strain H449kex 1

The Saccharomyces cerevisiae strain H449kex 1 is mixed with the plasmids

pDP34 / PHO5 (-173) -HIR

PDP34 / GAPFL-HIR

pDP34 / GAPEL-HIR

pDP34 / PHO5 (-173) -YHIR

pDP34 / GAPFL YHIR

pDP34 / GAPEL-YHIR

pDP92 / PHO5 (-173) -YHIR

PÜP92 / GAPFL YHIR

PDP92 / GAPELYHIR

using the transformation protocol of Hinnen et al. (see above). The transformed yeast cells are selected on plates with minimal yeast medium supplemented with leucine but containing no uracil.

Single transformed yeast cells have been isolated and designated as:

Saccharomyces cerevisiae H449kex1, pDP34 / PHO5 (-173) -HIR

Saccharomyces cerevisiae H449kex1 / pDP34 / GAPFL-HIR

Saccharomyces cerevisiae H449kex1 / pDP34 / GAPEL-HIR

Saccharomyces cerevisiae H449kex1 / pDP34 / PHO5 (-173) -YHIR

daccharomyces cerevisiae H449kex1 / pDP34 / GAPFL-YHIR

Saccharomyces cerevisiae H449kex1 / pDP34 / GAPEL-YHIR

Saccharomyces cerevisiae H449kex1 / pDP92.PHO5 (-173) -YHIR

Saccharomyces cerevisiae H449kex1 / pDP92 / GAPFL-YHIR

Saccharomyces cerevisiae H449kex1 / pDP92 / GAPEL-YHIR

Example 16: Fermentation of transformed yeast strains on a laboratory scale

Cells of Sacchoromyces cerevlslae H449kex1 / pDP34 / PH05 (-173) -YHIR and of Saccharomyces cerevislao H449kex1 / pDP34 / GAPFL-YHIR are grown sequentially in two pre-cultures of 10 ml minimal mode, composed as follows:

(G / l): Difco yeast nitrogen base 6.7 asparagine 10 leucine 1 glucose 20

The first pre-culture is incubated at 28 ⁰ C and 180U / min 60 hours. The second preculture is inoculated with 2% of the first preculture and incubated for 24 hours at 28 ° C and 180 rpm.

The medium of the main culture is composed as follows:

(G / l) yeast extract 49 glucose 5 fructose 57 NH ₄ NO ₃ 0.5 MgSO ₄ χ 7H ₂ O 1.0 CaCO ₃ 5.0 Ca ₃ (PO ₄ I ₂ 2.0

The main culture is inoculated with about 2 x 10 ⁶ Zelien / ml and incubated up to 72 hours at 28 ⁰ C and 180 rev / min. About 1 × 10 ⁹ cells / ml are obtained at the end of the fermentation. At several times during the fermentation, aliquots are removed from the culture, the cells are removed by centrifugation and the supernatant of the cultures is analyzed for desulphatohirudin by HPLC (see above).

Example 17: Preparation of desuliatohirudin variant HV1 on a 50-l scale

A working cell bank on the two micron DNA free Saccharomyces cerevisiae strain H449kex 1 / pDP34 / GAPFL-YHIR was used as a source of inoculum for the production of desulphatohirudin on a 50-l scale. Ampoules with the working cell bank are stored in the vapor phase of a liquid nitrogen container. The contents of an ampule are used to inoculate a shake flask with a culture with a selective medium composed as follows:

(G / l): Yeast Nitrogen Base 8.4 L-asparagine monohydrate 11.4 L-histidine 1.0 L-leucine 0.1 D-glucose monohydrate 20.0

The 500 ml flask contains 100 ml of medium and 48 hours at 28 ⁰ C incubated on a rotary shaker with a shaking speed of 180U / min.

The second preculture in a shake flask contains the same medium, of which 600 ml are contained in a 2 liter flask with four baffles. The inoculum level from the first pre-culture is 5% (30 ml) and the flasks are incubated on an orbital shaker at 120U / min hours at 28 ⁰ C.

A third preculture is fermented in a 50 L stainless steel bioreactor equipped with 4 baffles and a single 115 mm diameter turbine disc stirrer. The medium described above is also used for this culture, the starting volume is 30I. A single 2 L flask with 600 ml of culture is used to inoculate the culture in the 50 L reactor (2%). The fermentation lasts for about 42 hours at a temperature of 28 ° C. The stirring speed is 600 rpm, the aeration amount 1 vvm, and the reactor is operated at a pressure of 0.3 bar.

A similar 50 l bioreactor, which was additionally equipped for continuous addition, is used for the production of desulphatohirudin on a production scale. For this step (30I) a medium of the following composition is used:

(G / l):

Meat peptone (Merck) 5.0

Yeast Extract 30.0

Ammonium sulfate 6.0

Magnesium sulfate heptahydrate 1.0

Sodium chloride 0.1

Potassium dihydrogen phosphate 1.0

DGIucose, monohydrate 10.0

Optionally, the inoculum concentration from the third preculture stage is 2%. The fermentation lasts 48 hours at a temperature of 28 ⁰ C, the Ruhr speed is set to 750U / min. The overpressure is initially set at 0.3 bar, but may be increased to 1.0 bar during fermentation to keep the Mongu dissolved oxygen above 20% of saturation. The initial air flow used is 0.25vvm, but this is increased to 1 wm after nine hours to ensure adequate oxygenation.

The pH falls during the early stage of the fermentation to a value of 5.0, which is maintained by an automatic supply of ammonium hydroxide.

In simple batch culture, the concentration of biomass that is obtained, and consequently also the dsulfate hirudin titre, will depend on the amount of available carbon source that was initially used in the fermenter. This in turn is determined by the oxygen transfer capacity of the bioreactor and the need to prevent the production of ethanol by the growing yeast. These limitations can be overcome by using feed-in technology. Thus, quite little glucose is added to the initial medium, but glucose is added to achieve the formation of a considerably higher concentration and a desulphatohirudin titer of about three times at the end, which would be achieved by batch culture. In practice, the feed is gradually increased at intervals and finally a feed rate of 175 g / l glucose monohydrate is set. Small amounts of silicone-based antifoam are added, if necessary, to control foaming. A part of the effluent gas from the fermenter is analyzed to obtain information on oxygen uptake and carbon dioxide evolution. The dissolved oxygen concentration is measured on-line using a sterilizable electrode.

Throughout the process, samples are taken at 6-hour intervals to monitor the evolution of glucose and ethanol levels, the desulphatohirudin titer is further monitored by bioassay and HPLC, and sterility is monitored.

As indicated by HPLC, the produced desulphatohirudin is essentially free of analogs with truncated C-terminus. At the end of the fermentation process desulphatohirudin can be recovered from the supernatant of the culture.

Example 18: Recovery of desulphatohirudin from S.cerevlslae from a culture in the 50 l scale The culture broth (see Example 17) is mixed with Amberlite XAD-7 and is allowed to absorb for 4 hours at 25 ⁰ C are. The cells are separated from the resin in a column. After washing with 1M NaCl, the resin is eluted with Tris buffer (50 mM, pH = 7.0 to 8.5). The major fraction (30I) is adjusted to a pH of 2.9 and applied to an S-Sepharose column (Amicon PA, equilibrated with ammonium formate buffer 25nM, pH = 2.9) with a reaction volume of 21. After washing with ammonium formiate buffer (4OmM, pH = 3.6), elution with ammonium formate buffer (5OmM, pH - 3.8) is performed. The main fraction of the eluate (101) is concentrated using a Filtron Minisette Ullrafiltration System equipped with an il3k membrane. A 0.51 portion of the clear protein solution thus obtained is applied to a Bio-Gel P-δ (fine) column (Amicon GF, equilibrated with 0.5% acetic acid) having a column volume of 1.51. Elution onolgt with 0.5% acetic acid. The main fraction of the eluate (11) is concentrated by means of ultrafiltration and then added to a high-speed Q-Sepharose column (Amicon PA, equilibrated with ammonium formate buffer, 25 mM, pH = 2.9) with a bed volume of 21 , Elution is carried out with ammonium formate buffer (50 mM, pH = 4.2). The main fraction of El jates is concentrated by ultrafiltration and then dialyzed against water. The clear aqueous solution thus obtained is lyophilized. The solid is pure desulphatohirudin which is free from detectable levels of truncated C-terminus analogues.

Example 19: Cleavage of the PRA 1 gene in S. cprevisiae H449

By cleavage of the S. cerevisiae strain H 449 (DSM4413, prb, leu 2, ura 3, cir °) is equipped with a multiple protease deficiency. Gene cleavage is considered advantageous as compared to meiotic crossing, since a mutation is stably introduced while the genetic background of a given strain remains identical.

In a first step, the proteinase yscA activity is eliminated by cleavage of the PRA1 gene (G. Ammerer et al., Mol.

Cell. Biol. [1986, 16, 2490]. PRA1 is isolated from all yeast genomic DNA and digested with Sac I and Pst I. From a preparative 0.6% agarose gel, 2 kb fragments are isolated, recovered by electroelution and ligated into the Pst I-Sac I sites of the polylinker region pUC 19. De ^r vector is transformed into E.coli JM109 and 280 and read out individual colonies.

The colonies are individually grown in wells of microtiter plates with LB + amp medium. Colony hybridization is performed essentially as described (DEWoods et al., (1982) Proc Natl Acad., USA 79, 5651) using the following, ³² P-labeled sondo (oligonucleotide). is used:

5 '-AAGCCTAGTGACCTAGT- 3'

which was derived from a published PRA1 sequence (Ammerer et al. supra), 3 positive clones are read, the DNA of one of them pUC19 / PRA1 is cleaved with Sac I and Xhol and the 1.9 kb fragment, containing the complete PRA1 gene, subcloned into the KS polylinker region of the transcript vector (Oluescript vector) M13 + (Stratagene Cloning Systems, San Diego, CA, USA). A 1.2 kb Hind III fragment containing the complete HPA3 gene (Rose et al., Gene [1984, 129, 1113] becomes the unique Hind III region within the coding region of the PRA1 insert inserted. The plasmid thus prepared is designated M13 + / pra 1 :: URA3. M13 + / pra 1 :: URA3 is digested with SacI / Xhol and the 3.1 kb fragment used without separation from the vector to transform S. cerevisiae H449 as described (see Example 3). Uracil independent transformants are read out, the DNA prepared and degraded by Sac I / Xhol and examined by Southern transfer for the correct cleavage of the PRA1 gene out. A transformant with the correct shift of the Sacl / Xho fragment which hybridizes with PRA1 from i, 9kb to 3.1kb is referred to as Tr 1186. In the next step, Tr 1196-under cleavage of its "PRA 1 gene by pra1 :: URA3-is made dependent on the incorporation of a deletion into the pra 1 :: URA3 gene insert in turn uracil. Tr 1186 is treated with 1 μg of the plasmid YEp 13 (J.R. Broach et al., Gene

[1979] 8,121) together with 10 μg of the plasmid pUC12ura3delta, which has a 200 bp deletion in the URA3 gene (Sungstag, C, et al., Nucl. Acids Res. 15 (1978), 233). 3000 leucine prototrophic yeast transformants are resuspended in 5 ml minimal medium (Difco yeast nitrogen base without amino acids, to which 2% glucoso, 0.1% leucine, 0.1% uracil and 0.25% fluoro-amino acid are added) in a small shake flask and incubated for 60 hours at 3O ⁰ C and 180U / min. Transformants growing under these conditions are resistant to the toxic analog of fluoro-amino acid and therefore have a replacement abundance in the pra 1 :: URA3 region by ura3delta. The grown cells are streaked on complete medium consisting of (g / l) peptone 20, yeast extract 10, glucose 20, and after culturing at 30 ° C. within 48 hours in the form of replica streaked to minimal mode (Difco yeasts). Nitrogen base without amino acids, 2% glucose and 0.1% leucine added) to detect the uracil auxotrophs. A number of auxotrophs are read and examined for the loss of plasmid YEp13, which carries leucine auxotrophy. A single colony designated Tr 1195-which requires leucine and uracil is read and used for further experiments.

Example 20: Cleavage of the PRC1 Gone in Tr 1195

Next, the activity of carboxypeptidase yscY in Tr 1195 is eliminated. PRC1, which codes for yscY (JH Rothman et al., 1986, Proc. Natl. Acad., U.S.A., 83,3248), is obtained from the genomic library for yeast in the centromeric vector pCS 19 (C. Sengstag et al ], Nucleic Acids Research 15,233) by colony hybridization with two synthetic oligonucleotides

Ö'-GAAAGCATTCACCAGTTTACTATGTGG -3 'and

5'-CGAAl GGATCCCAACGGGTTTCTCC -3 '

which correspond to the 5 'and 3' ends of the coding sequence of the PRC1 isolated. A positive clone is referred to as pCS 19 / cpy 8. The DNA of pCS 19 / cpy8 is digested with CIa I / Pvu II, applied to a 0.6% preparative agarose gel, and a 2.6 kb fragment is isolated and recovered by electroelution. This CIa I / Pvu II fragment containing the entire PRC1 gene is further cloned into the Narl / SmaI site of pUC19. The resulting plasmid pUC19 / PRC1 is cleaved at the unique site to which the 1.2 kb URA3 fragment (see Example 19) is ligated. To this end, the cohesive ends of Hind III of the URA3-containing fragment are filled in ointr reaction with Klenow DNA polymerase to fit into the Stu I site at the blunt end of pUC / PRC1.

The resulting plasmid pUC 19 / prc :: URA3 is degraded with AatII and the Aat II fragment without separation from the vector for the transformation of S. cerevisiae Tr 1195 used, as has been described. A uracil-independent transformant Tr 1206 is examined by Southern transfer for correct cleavage of the PRC1 gene and then re-uracil-dependent as described (see Example 19). The strain of S. cerevisiae thus obtained is designated Tr 1210 (pra 1, prb1, prd, ura3, leu2, cir °).

Example 21: Generation of a kex1 variant of S.corevisiaeTr 1210

The activity of carboxypeptidase ysc is eliminated from the S. cerevisiae Tr 1210 strain by cleaving the genomic KEX1-Csn as described (see Example 14). The strain of S. cerevisiae thus obtained, which has no protease activity of ysca, is referred to as Tr 1302 (pra1, prb1, prc1, ura3, leu2, cir0, kex1).

Example 22: Construction of hirudin mutants HV1-KR, HVI-SFRY and HV1-WELR In order to further investigate the degradation at the C-terminus of heterologous proteins with different amino acids at the C-terminus by ysca, hirudin variants were created which are in their amino acid composition at the C-terminus. As a general method, in vitro site-directed mutagenesis is used, in principle as described (Bio-Rad-Muta-Gen M13 Kit, Bio-Rad, Richmond, Ca. USA).

The following mutants were constructed:

1. HV1-KR, corresponds to [Lys ^ -Arg ^ l-Hirudin variant 1

2. I-IV1-SFRY, corresponds to [Ser ^ -Phe ^ -Arg ^ -Tyr ^ l-HVI

3. HV1 -WELR, corresponds to ITrp ^ -Glu ^ -Leu ^ -Arg ^ l HV 1.

The amino acid sequences of all these hirudin Mui nten correspond to the hirudin variant 1 his to the amino acid 61 and 63, respectively.

In the HV1-SEPY variant, the C-tarmin is identical to the natriuretic atrium peptide (Vlasuk et al., J. Biol. Chem. [1986] 261, 4789-4796 / and in HV1-WELR the C-term corresponds to The term epidermal growth factor (C. George-Nascimento et al., 1988, Biochemistry 27, 797-802), both of which are known to be degraded by protease-containing wild yeast strains at the C-terminus.

In a first step, plasmid pJDB207 / GAPEL-HIR (as described in European Patent Application No. 225,633) is digested with Sal I-Hind III to give a 1.2 kb fragment containing the hirudin expression Cassette in full length contains. The 1.2 kb SaI I-Hind III fragment is subcloned into Sal I-Hind III cut with the Bluescript M13 + with the polylinker SK. The resulting plasmid iVi13 + / HV1 is transfected into E. coli cells CJ236 to incorporate uracil as described (Bio-Rad Muta-Gene M13 kit, supra). From transfection E. coli CJ 236 cells are isolated from single-stranded DNA using M13 helper phage (Stratagene, supra).

To construct the variant HV1-KR, the mutagenic primer is used

5'-CCGGA ^ GAATACAAGAG nTAGGATCCT-3 '. For variant HV1-SFRY, the mutagenic primer is used

5'-GAAGAAATCCCGGAATCnTCAGATACTAGGATCCTGGTACG-S '. For HV1-WELR, the mutagenic primer becomes

5'-GAAGAAATCCCGGAATGGGAACTGAGATAGGATCCTGGTACg-S '

used. First, 20OpMoI of each primer in a total volume of 30μΙ containing 3pl 1 MTris HCl, pH = 8.0; 0.3μΙ 1 mgCI ₂ , 0.75μΙ0.2Μ DTT ₁ O ₁ SpI 2OmM ATP contains, phosphorylated.4.5 units of T4 polynucleotide kinase are added and this mixture is maintained at 37 ° C for 45 minutes and at 65 ° for 10 minutes C incubated.

Subsequently, the phosphorylated oligonucleotides L.iter are attached to the DNA of the template under the following conditions: DNA containing 0.1 pmol of uracil, which was obtained from M13 + / HV1, is incubated with 2 pmoles of phosphorylated primer in a total volume of the addition buffer of 10 μM (20 mM Tris-HCl, pH = 7.4; 2 mM MgCl ₂ , 50 mM NaCl). These mixtures are heated in a water bath to 8O ⁰ C and then allowed to cool slowly until room temperature is sufficient.

Under the following conditions dsn complementary strands are then generated: 10μΙ of each of the mixtures werdenmit for attachment 'Ml2mMdNTP's 0,75Ml20mMATP, 0,75Ml0,5MTris-HCl, pH = 7.4; 0,75Ml0,1MMgCI _2, 2,15Ml0! , 2MDTT, 1 unit of T4 DNA polymerase and 2 units of T4 DNA ligase. These reaction mixtures are first incubated 5 minutes on ice, then incubated for 5 minutes at 25 ⁰ C and finally 30 minutes at 37 ⁰ C irkubiert. The resulting duplex DNAs were transformed into E. coli JM101 cells, a strain that effectively removes the uracil-containing template, allowing the mutant complementary strand to replicate (Bio-Rad, supra). Plasmids are prepared and assayed for the absence of the Pst I site, which should only be present in the last two codons of the unmutated wild-type HV1 DNA.

The correct mutagenesis is further accomplished by sequencing the new hirudin mutants using the following primer:

5'-GAAGGTACCCCGAAACCGCa-S '

wherein the primer of the hirudin coding sequence is approximately 20 bp upstream of the mutagenic primers.

Finally, the three mutant SaI I-Hind HI fragments are excised from the Bluescript M13 + and ligated back into the SalI-HindHI sites of pJDB207. The three new plasmids are called

1. pJDB207 / GAPEL-HV1-KR

2. pJDB2O7 / GAPEL-HV1-Sf-RY and

3. pJDB207 / GAP £ LHV1-WEIR.

On the other hand, and in order to be able to transform cir ° strains such as Tr1302 (see above, Example 21), the hirudin variants were subcloned into the complete 2 micron vector pDP34 (see Example 9, supra). pDP34 is cut at the single BamH 1 site and the cohesive ends are padded with Klenow DNA polymerase to create blunt ends. The SaI I-Hind HI fragments from Bluescript M13 +, which encode the mutant hirudin sequences (see above), are also blunt-ended and ligated into the blunt-ended (BamH 1) sites of pDP34. The resulting plasmids are designated as

1. pDP34 / GAPEL-HV1-KR

2. pDP34 / GAPEL-HV1-SERY and

3. PDP34 / GAPEL-HV1-WELR.

Example 23. Transformation of S. cerevisiae strains BYSKEX1 and BYSkexi with pJDB207 / GAPEL-HIR, pJDB207 / GAPEL-HV1-KR, pJDB207 / GAPEL-HV1-SFRY, and pJDB207 / GAPEL-HV1-WELR and S. cerevisiae strains Tr1210 and TM302 with pDP34 / GAPEL-HV1 -KR, pDP34 / GAPEL-HV1-SERY and pDP34 / GAPEL-HV1-WELR

pJDB207 / GAPEL-HV1-KR, pJDB207 / GAPEL-HV1-SFRY, pJDB207 / GAPEL-HV1-WELR and pJDB207 / GAPEL-HIR are transformed into S. cerevisiae BYSKEX1 (-DSM 4583) and BYSkexi using the standard protocol and I .eucine selection transformed (see Example 3). In the same manner, pDP34 / GAPF.L-HV 1 -KR, pDP34 / GAPEL-HV 1 -SERY and pDP34 / GAPEL-HV 1-WELR are transformed into S. cerevisiae strains Tr 1210 and Tr 1302. The strains thus obtained were designated as follows:

S.cerevisiae BYSKEX 1 / pJDB207 / GAPEL-HIR

S.cerevisiae BYSKEX 1 / pJDB207 / GAPEL-HV 1-KR

S. cerevisiae BYSKEX1 / pJDB 207 / GAPEL-HV 1 -SFRY

S.cerevisiae BYSKEX 1 / pJDB207 / GAPEL-HV 1-WELR

S. cerevisiae BYSkex1 / pJD8207 / GAPEL-HIR

S. cerevisiae BYSkex 1 / pJDB207 / GAPEL-HV 1 -KR

S. cerevisiae BYSkex1 / pJDB207 / GAPEL-HV 1-SFRY

S. cerevisiae BYSkex1 / pJDB207 / GAPEL-HV 1-WELR

S. cerevisiae Tr1210 / pDP34 / GAPEL-HV1-KR

S. cerevisiae Tr 1210 / pDP34 / GAPEL-HVi-SFRY

S.cerevisiaeTr1210 / pDP34 / GAPEL HV1 WELR

S. cerevisiae Tr 1302 / pDP34 / GAPEL-HV1-KR

S.cerevisiaeTM302 / pDP34 / GAPEL HV1 SFRY

S.cerevisiaeTM302 / pDP34 / GAPEL HV1 WELR

The fermentation of the S. cerevisiae strains is carried out in minimal medium (pre- and main culture) as has been described (Beise Beispiol 17).

Example 24 Analysis of the Hirudin Varlante HV1 and its Mutants from the Fermentatlonskulturen of Saccharomycos

cerevisiae transformants BYSkexi and BYSKEX I using reverse-phase HPLC. After 72 hours of fermentation, the samples are prepared from the liquid yeast cultures by centrifugation to produce a clear solution containing 1:10 (v / v) with 1 M acetic acid and analyzed by HPLC under the following conditions.

A HIBAR (Merck) column (4mm χ 125mm) is filled with the Umkek'phase, a spherical silica gel from Macherey-Nagel, a spherical stationary phase with a particle diameter of 5μηι and a porosity of 100A. The ends of the columns are equipped with stainless steel frits. Mobile phase A is prepared from water (Nanopure®, BARNSTEAD) with 0.1% (v / v) trifluoroacetic acid. Mobile phase B is prepared from 20% mobile phase A and 80% (v / v) acetonitrile (HPLC grade, FLUKA) with 0.075 vol% trifluoroacetic acid. The chromatographic separations are carried out at a flow rate of 1.5 ml / min with the following gradient. The eluates are evaluated by absorption at 216nm.

t (min)% A% B

0 00 10 I 79 21 G 79 21 17 64 36 20 0 100 22 0 100 24 90 10 32/0 90 10

To calibrate the system, prepare a standard solution by dissolving 1 mg of pure desulphatohirudin in 1 ml of water. 50μl of this standard solution is injected into the column and chromatographed as described to calibrate the system.

Figure 8 presents the results showing that S. cerevisiae produced full-length BYSkex 1 -hirudins (either wild-type HV1-hirudin or HV1-KR, HV1-SFRY, HV1-WELR) mutants having a have retention time other than the degradation products of hirudin produced by S. cerevisiae BYSKEX 1. The wild-type hirudin HV1 shows the typical mixture of the full-length HV1 (retention time 17.05 min) and the two degradation products "HIR-64" (retention time 17.8 min) and "HIR-63" (retention time 15.33 min) in the BYSKEX 1. BYSkex 1 generates only "HIR-65" (retention time 17.05min) In the case of mutation HV1-KR, BYSKEX1 produces exclusively the degradation product, which lacks two terminal amino acids (= "HIR-63", retention time 15.9 min) while BYSkex 1 produces the full-length HV1-KR (retention time 14.4min).

In the case of the mutant HV1 -WELR BYSKEX 1 shows the degradation product with a retention time of 18.6 min, BYSkex 1 preferably produces full-length HV1-WELR (retention time 17.3 min) and only traces of the degradation product. HV1-SFRY is found only in traces with full length (retention time 17.1 min), in BYSkex 1 only intact HV1-SRY is found (retention time 17.1 min). The large peak (retention time 15.7 min) is not related to HV1-SFRY.

Corresponding results are obtained when the KEX1-S. cerevisiae strains Tr 1210 / pDP34 / GAPEL-HV 1 -KR, Tr 1210 / pDP34 / GAPEL-HV1-SERY and Tr1210 / pDP34 / GAPEL-HV1-WELR with the corresponding kex1 Strains Tr1302 / pDP34 / GAPEL-HV1-KR, Tr 1302 / pDP34 / GAPEL-HV1-SERY and Ti 1302 / pDP34 / GAPEL-HV1-WELR.

Deposit of microorganisms

The following microorganism strains have been deposited with the German Collection of Microorganisms (DSM), Mascher or Weg 1 b, D-3300 Braunschweig (deposit date and accession numbers are given):

Saccharomyces cerevisiae K449: February 18, 1988, DSM4413

Escherichia coli JMiO9 / pDP38: February 19, 1988, DSM4414

Escherichia coli JM109 / pDP34: March 14, 1988, DSM4473

Saccharomyces cerevisiae BYS 232-31-42: May 6, 1988, DSM4583

Claims (17)

A process for producing a yeast strain lacking carboxypeptidase ysca activity which has been transformed with a hybrid vector consisting of a yeast promoter operably linked to a DNA sequence encoding a heterologous protein, characterized in that Yeast strain is transformed with such a hybrid vector. 2. A method for producing a yeast strain according to claim 1, characterized in that the hybrid vector consists of a yeast promoter operably linked to a first DNA sequence which codes for a signal peptide and bound in a suitable reading frame to a second DNA sequence which encodes a heterologous protein and a DNA sequence containing signals for the termination of transcription in yeast. 3. A method for producing a yeast strain according to claim 1, characterized in that the heterologous protoin desulphatohirudin. 4. A method for producing a yeast strain according to claim 1, characterized in that it also has no peptidase activity selected from the following: yscA, aysB, yscY and yscS. 5. A method for producing a yeast strain according to claim 1, characterized in that it has no yscY activity. 6. A method for producing a yeast strain according to claim 1, characterized in that it has no yscY, yscB and yscS activity. 7. A method for producing a yeast strain according to claim 1, characterized in that it has no yscY, yscA and yscB activity. 8. A method for producing a yeast strain according to claim 1, characterized in that it does not contain endogenous two-micron DNA and was transformed with a hybrid vector, that of the complete two-micron DNA, including the intact REP1, REP2 and FLP Gone and the intact ORI, STB, IR1 and IR2 locations. 9. A method for producing a yeast strain according to claim 8, characterized in that it has no yscY activity. 10. A method for producing a yeast strain according to claim 8, characterized in that it has no yscY, yscB and yscS activity. 11. A method for producing a yeast strain according to claim 8, characterized in that it has no yscY, yscA and yscB activity. 1.?. Process for the preparation of a yeast strain according to claim 1, characterized in that the hybrid vector consists of a yeast promoter selected from the group consisting of: MF1 promoter, GAL1 promoter, a promoter of a gene coding for a glycolytic enzyme, ADHI promoter Promoter, TRPI promoter and the PH05 promoter, from which, if necessary, the upstream activation sites were cleaved. 13. A method for producing a yeast strain according to claim 2, characterized in that the hybrid vector consists of a first DNA sequence which was selected from the following group: Hirudin signal sequence, the signal and prepro sequences of the yeast invertase, the a factor, pheromone, pheromone peptidase (KEX 1), "killertoxin" and repressive acid phosphatase (PHO 5) genes, and the glucoamylase signal sequence of Aspergillus awamori. 14. A method for producing a yeast strain according to claim 3, characterized in that the hybrid vector consists of a second DNA sequence coding for a Desulfatohirudin derivative selected from the group consisting of: Desulfatohirud: n variant HV1, HV'l (modified a, b), HV2, HV2 (modified a, b and c), PA, variants of PA and (Val ₂ ) -desulphatohirudin. 15. A method for producing a yeast strain according to claim 1, characterized in that the hybrid vector comprises one or more selective genetic marker for a yeast. 16. A method for producing a yeast strain according to claim 1, characterized in that the hybrid vector comprises a selective genetic marker and a site for the replication of a bacterial host. 17. A method for producing a yeast strain according to claim 1, characterized in that the yeast strain is Saccharomyces cerevisiae. For this 11 pages drawings