Fusion Polypeptides Comprising Human Serum Albumin, Nucleic Acids Encoding Same, and Recombinant Expression Thereof

The present invention relates to new biologically active polypeptides, their preparation and pharmaceutical compositions containing them.

More particularly, the present invention relates to essentially recombinant polypeptides composed of an active part derived from a natural or artificial polypeptide having a therapeutic activity and coupled to an albumin or to a variant of albumin. It is understood that the therapeutic activity of the polypeptides of the invention can be either direct (treatment of diseases), or indirect (and for example capable of being used in the prevention of diseases, in the design of vaccines, in medical imaging techniques and the like).

It is understood in the following text that the albumin variants designate any protein with a high plasma half-life which is obtained by modification (mutation, deletion and/or addition), by genetic engineering techniques, of a gene encoding a given isomorph of human serum albumin, as well as any macromolecule with a high plasma half-life obtained by in vitro modification of the protein encoded by such genes. Albumin being highly polymorphic, numerous natural variants have been identified and classified ›Weitkamp L. R. et al., Ann. Hum. Genet. 37 (1973) 219!.

The aim of the present invention is to prepare artificial proteins which are biologically active and can be used pharmaceutically. Indeed, numerous polypeptides possessing one or more potential therapeutic activities cannot be exploited pharmaceutically. This may have various reasons, such as especially their low stability in vivo, their complex or fragile structure, the difficulty of producing them on an industrially acceptable scale and the like. Likewise, some polypeptides do not give the expected results in vivo because of problems of administration, of packaging, of pharmacokinetics and the like.

The present invention makes it possible to overcome these disadvantages. The present invention indeed provides new molecules which permit an optimal therapeutic exploitation of the biological properties of these polypeptides. The present invention results especially from the demonstration that it is possible to couple genetically any active structure derived from a biologically active polypeptide to another protein structure consisting of albumin, without impairing the said biological properties thereof. It also results from the demonstration by the Applicant that human serum albumin makes it possible efficiently to present the active structure to its sites for interaction, and that it provides a high plasma stability for the polypeptide of the invention. The polypeptides of the invention thus make it possible to maintain, in the body, a given biological activity for a prolonged period. They thus make it possible to reduce the administered doses and, in some cases, to potentiate the therapeutic effect, for example by reducing the side effects following a higher administration. The polypeptides of the invention make it possible, in addition, to generate and to use structures derived from biologically active polypeptides which are very small and therefore very specific for a desired effect. It is understood that the peptides having a biological activity, which are of therapeutic interest, may also correspond to non-natural peptide sequences isolated for example from random peptide libraries. The polypeptides of the invention possess, moreover, a particularly advantageous distribution in the body, which modifies their pharmacokinetic properties and favours the development of their biological activity and their use. In addition, they also have the advantage of being weakly or non-immunogenic for the organism in which they are used. Finally, the polypeptides of the invention can be expressed (and preferentially secreted) by recombinant organisms, at levels permitting their industrial exploitation.

One subject of the present invention therefore relates to polypeptides containing an active part derived from a polypeptide having a therapeutic activity, coupled to an albumin or a variant of albumin.

In a specific embodiment, the peptides possessing a therapeutic activity are not of human origin. For example, there may be mentioned peptides, or their derivatives, possessing properties which are potentially useful in the pathologies of the blood and interstitial compartments, such as hirudin, trigramine, antistatine, tick anticoagulant peptides (TAP), arietin, applagin and the like.

More particularly, in the molecules of the invention, the polypeptide having a therapeutic activity is a polypeptide of human origin or a molecular variant. For example, this may be all or part of an enzyme, an enzyme inhibitor, an antigen, an antibody, a hormone, a factor involved in the control of coagulation, an interferon, a cytokine ›the interleukins, but also their variants which are natural antagonists of their binding to the receptor(s), the SIS (small induced secreted) type cytokines and for example the macrophage inflammatory proteins (MIPs), and the like!, of a growth factor and/or of differentiation ›and for example the transformant growth factors (TGFs), the blood cell differentiation factors (erythropoietin, M-CSF, G-CSF, GM-CSF and the like), insulin and the growth factors resembling it (IGFs), or alternatively cell permeability factors (VPF/VEGF), and the like!, of a factor involved in the genesis/resorption of bone tissues (OIF and osteospontin for example), of a factor involved in cellular motility or migration ›and for example autocrine motility factor (AMF), migration stimulating factor (MSF), or alternatively the scatter factor (scatter factor/hepatocyte growth factor)!, of a bactericidal or antifungal factor, of a chemotactic factor ›and for example platelet factor 4 (PF4), or alternatively the monocyte chemoattracting peptides (MCP/MCAF) or neutrophil chemoattracting peptides (NCAF), and the like!, of a cytostatic factor (and for example the proteins which bind to galactosides), of a plasma (and for example von Willebrand factor, fibrinogen and the like) or interstitial (laminin, tenascin, vitronectin and the like) adhesive molecule or extracellular matrices, or alternatively any peptide sequence which is an antagonist or agonist of molecular and/or intercellular interactions involved in the pathologies of the circulatory and interstitial compartments and for example the formation of arterial and venous thrombi, cancerous metastases, tumour angiogenesis, inflammatory shock, autoimmune diseases, bone and osteoarticular pathologies and the like.

The active part of the polypeptides of the invention may consist for example of the polypeptide having a whole therapeutic activity, or of a structure derived therefrom, or alternatively of a non-natural polypeptide isolated from a peptide library. For the purposes of the present invention, a derived structure is understood to mean any polypeptide obtained by modification and preserving a therapeutic activity. Modification should be understood to mean any mutation, substitution, deletion, addition or modification of genetic and/or chemical nature. Such derivatives may be generated for various reasons, such as especially that of increasing the affinity of the molecule for its binding sites, that of improving its levels of production, that of increasing its resistance to proteases, that of increasing its therapeutic efficacy or alternatively of reducing its side effects, or that of conferring on it new biological properties. As an example, the chimeric polypeptides of the invention possess pharmacokinetic properties and a biological activity which can be used for the prevention or treatment of diseases.

Particularly advantageous polypeptides of the invention are those in which the active part has:

(a) the whole peptide structure or,

(b) a structure derived from (a) by structural modification (mutation, substitution addition and/or deletion of one or more residues) and possessing a therapeutic activity.

Among the structures of the (b) type, there may be mentioned more particularly the molecules in which certain N- or O-glycosylation sites have been modified or suppressed, the molecules in which one or more residues have been substituted, or the molecules in which all the cystein residues have been substituted. There may also be mentioned molecules obtained from (a) by deletion of regions not involved or not highly involved in the interaction with the binding sites considered, or expressing an undesirable activity, and molecules containing, compared to (a), additional residues such as for example an N-terminal methionine and/or a signal for secretion and/or a joining peptide.

The active part of the molecules of the invention can be coupled either directly or via an artificial peptide to albumin. Furthermore, it may constitute the N-terminal end as well as the C-terminal end of the molecule. Preferably, in the molecules of the invention, the active part constitutes the C-terminal part of the chimera. It is also understood that the biologically active part may be repetitive within the chimera. A schematic representation of the molecules of the invention is given in FIG. 1.

Another subject of the invention relates to a process for preparing the chimeric molecules described above. More specifically, this process consists in causing a eukaryotic or prokaryotic cellular host to express a nucleotide sequence encoding the desired polypeptide, and then in harvesting the polypeptide produced.

Among the eukaryotic hosts which can be used within the framework of the present invention, there may be mentioned animal cells, yeasts or fungi. In particular, as regards yeasts, there may be mentioned yeasts of the genus Saccharomyces, Kluyveromyces, Pichia, Schwanniomyces, or Hansenula. As regards animal cells, there may be mentioned COS, CHO and C127 cells and the like. Among the fungi capable of being used in the present invention, there may be mentioned more particularly Aspergillus ssp, or Trichoderma ssp. As prokaryotic hosts, the use of bacteria such as Escherichia coli, or belonging to the genera Corynebacterium, Bacillus, or Streptomyces is preferred.

The nucleotide sequences which can be used within the framework of the present invention can be prepared in various ways. Generally, they are obtained by assembling, in reading phase, the sequences encoding each of the functional parts of the polypeptide. The latter may be isolated by the techniques of persons skilled in the art, and for example directly from cellular messenger RNAs (mRNAs), or by recloning from a complementary DNA (cDNA) library, or alternatively they may be completely synthetic nucleotide sequences. It is understood, furthermore, that the nucleotide sequences may also be subsequently modified, for example by the techniques of genetic engineering, in order to obtain derivatives or variants of the said sequences.

More preferably, in the process of the invention, the nucleotide sequence is part of an expression cassette comprising a region for initiation of transcription (promoter region) permitting, in the host cells, the expression of the nucleotide sequence placed under its control and encoding the polypeptides of the invention. This region may come from promoter regions of genes which are highly expressed in the host cell used, the expression being constitutive or regulatable. As regards yeasts, it may be the promoter of the gene for phosphoglycerate kinase (PGK), glyceraldehyde-3-phosphate dehydrogenase (GPD), lactase (LAC4), enolases (ENO), alcohol dehydrogenases (ADH), and the like. As regards bacteria, it may be the promoter of the right-hand or left-hand genes from the lambda bacteriophage (P.sub.L, P.sub.R), or alternatively the promoters of the genes for the tryptophan (P.sub.trp) or lactose (P.sub.lac) operons. In addition, this control region can be modified, for example by in vitro mutagenesis, by the introduction of additional control elements or of synthetic sequences, or by deletions or substitutions of the original control elements. The expression cassette may also comprise a region for termination of transcription which is functional in the host envisaged, positioned immediately downstream of the nucleotide sequence encoding a polypeptide of the invention.

In a preferred mode, the polypeptides of the invention result from the expression, in a eukaryotic or prokaryotic host, of a nucleotide sequence and from the secretion of the product of expression of the said sequence into the culture medium. It is indeed particularly advantageous to be able to obtain, by the recombinant route, molecules directly in the culture medium. In this case, the nucleotide sequence encoding a polypeptide of the invention is preceded by a "leader" sequence (or signal sequence) directing the nascent polypeptide in the secretory pathways of the host used. This "leader" sequence may be the natural signal sequence of the biologically active polypeptide in the case where the latter is a naturally secreted protein, or that of the stabilizing structure, but it may also be any other functional "leader" sequence, or an artificial "leader" sequence. The choice of one or the other of these sequences is especially guided by the host used. Examples of functional signal sequences include those of the genes for the sexual pheromones or the "killer" toxins of yeasts.

In addition to the expression cassette, one or several markers which make it possible to select the recombinant host may be added, such as for example the URA3 gene from the yeast S. cerevisiae, or genes conferring the resistance to antibiotics such as geneticin (G418) or to any other toxic compound such as certain metal ions.

The unit formed by the expression cassette and by the selectable marker can be introduced directly into the considered host cells, or previously inserted in a functional self-replicating vector. In the first case, sequences homologous to regions present in the genome of the host cells are preferably added to this unit; the said sequences then being positioned on each side of the expression cassette and of the selectable gene so as to increase the frequency of integration of the unit into the genome of the host by targetting the integration of the sequences by homologous recombination. In the case where the expression cassette is inserted in a replicative system, a preferred replication system for yeasts of the genus Kluyveromyces is derived from the plasmid pKD1 originally isolated from K. drosophilarum; a preferred replication system for yeasts of the genus Saccharomyces is derived from the 2.mu. plasmid from S. cerevisiae. Furthermore, this expression plasmid may contain all or part of the said replication systems, or may combine elements derived both from the plasmid pKD1 and the 2.mu. plasmid.

In addition, the expression plasmids may be shuttle vectors between a bacterial host such as Escherichia coli and the chosen host cell. In this case, a replication origin and a selectable marker functioning in the bacterial host are required. It is also possible to position restriction sites surrounding the bacterial and unique sequences on the expression vector: this makes it possible to suppress these sequences by cutting and religation in vitro of the truncated vector before transformation of the host cells, which may result in an increase in the number of copies and in an increased stability of the expression plasmids in the said hosts. For example, such restriction sites may correspond to sequences such as 5'-GGCCNNNNNGGCC-3' SEQ ID NO:19 (SfiI) or 5'-GCGGCCGC-3' (NotI) in so far as these sites are extremely rare and generally absent from an expression vector.

After construction of such vectors or expression cassette, the latter are introduced into the host cells selected according to the conventional techniques described in the literature. In this respect, any method permitting the introduction of a foreign DNA into a cell can be used. This may be especially transformation, electroporation, conjugation, or any other technique known to persons skilled in the art. As an example of yeast-type hosts, the various strains of Kluyveromyces used were transformed by treating the whole cells in the presence of lithium acetate and polyethylene glycol, according to the technique described by Ito et al. ›J. Bacteriol. 153 (1983) 163!. The transformation technique described by Durrens et al. ›Curr. Genet. 18 (1990) 7! using ethylene glycol and dimethyl sulphoxide was also used. It is also possible to transform the yeasts by electroporation, according to the method described by Karube et al. ›FEBS Letters 182 (1985) 90!. An alternative procedure is also described in detail in the examples below.

After selection of the transformed cells, the cells expressing the said polypeptides are inoculated and the recovery of the said polypeptides can be carried out, either during the cell growth for the "continuous" processes, or at the end of growth for the "batch" cultures. The polypeptides which are the subject of the present invention are then purified from the culture supernatant for their molecular, pharmacokinetic and biological characterization.

A preferred expression system for the polypeptides of the invention consists in using yeasts of the genus Kluyveromyces as host cell, transformed by certain vectors derived from the extrachromosomal replicon pKD1 originally isolated from K. marxianus var. drosophilarum. These yeasts, and in particular K. lactis and K. fragilis are generally capable of stably replicating the said vectors and possess, in addition, the advantage of being included in the list of G.R.A.S. ("Generally Recognized As Safe") organisms. Favoured yeasts are preferably industrial yeasts of the genus Kluyveromyces which are capable of stably replicating the said plasmids derived from the plasmid pKD1 and in which has been inserted a selectable marker as well as an expression cassette permitting the secretion, at high levels, of the polypeptides of the invention.

The present invention also relates to the nucleotide sequences encoding the chimeric polypeptides described above, as well as the eukaryotic or prokaryotic recombinant cells comprising such sequences.

The present invention also relates to the application, as medicinal products, of the polypeptides according to the present invention. More particularly, the subject of the invention is any pharmaceutical composition comprising one or more polypeptides or nucleotide sequences as described above. The nucleotide sequences can indeed be used in gene therapy.

The present invention will be more fully described with the aid of the following examples, which should be considered as illustrative and non-limiting.

The representations of the plasmids indicated in the following figures are not plotted to scale and only the restriction sites important for the understanding of the clonings carried out have been indicated.

FIG. 1A is a schematic representation of the chimera of the HSA-PEPTIDE type; FIG. 1B is a schematic representation of a chimera of the PEPTIDE-HSA type; and FIG. 1C is a schematic representation of a chimera of the PEPTIDE-HSA-PEPTIDE type. Abbreviations used: M/LP, translational initiator methionine residue, optionally followed by a signal sequence for secretion; HSA, mature albumin or one of its molecular variants; PEP, peptide of natural or artificial origin possessing a given therapeutic property. The PEP sequence may be present several times in the FIG. 1A, B or C molecules. The black arrow indicates the N-terminal end of the mature protein.

FIGS. 2(a) and 2(b), together, comprise an example of a nucleotide sequence (SEQ ID NO:1) and an amino acid sequence (SEQ ID NO:2) of a HindIII restriction fragment encoding a chimeric protein of the prepro-HSA-PEPTIDE type. The black arrows indicate the end of the "pre" and "pro" regions of HSA. The MstII restriction site is underligned and the codon specifying the termination of translation is in bold characters.

FIG. 3: Restriction map for the plasmid pYG105 and generic strategy for construction of the plasmids for expression of the chimeric proteins of the present invention. Abbreviations used: P, transcriptional promoter; T, transcriptional terminator; IR, inverted repeat sequences of the plasmid pKD1; LP, signal sequence for secretion; Ap.sup.r and Km.sup.r designate the genes for resistance to ampicillin (E. coli) and to G418 (yeasts), respectively.

FIGS. 4A, 4B, 4C, 4D, and 4E collectively show examples of nucleotide sequences of MstII-HindIII restriction fragments derived from the von Willebrand factor. FIG. 4A is a representation of the structure of the MstII-HindIII fragment of the plasmid pYG1248 (SEQ ID NOS:3 and 4). FIG. 4B is a representation of the structure of the MstII-HindIII fragment of the plasmid pYG1214 (SEQ ID NOS:5 and 6). FIG. 4C is a representation of the MstII-HindIII fragment of the plasmid pYG1206; in this particular chimera, the Leu694 residue of the vWF is also the last residue (Leu585) of the HSA. FIG. 4D is a representation of the MstII-HindIII fragment of the plasmid pYG1223 (SEQ ID NOS:9 and 10). The numbering of the amino acids corresponds to the numbering of the mature vWF according to Titani et al. ›Biochemistry 25 (1986) 3171-3184!. The MstII and HindIII restriction sites are underlined and the translation termination codon is in bold characters. FIG. 4E is a nucleotide sequence (SEQ ID NO:3) of the MstII-HindIII restriction fragment of the plasmid pYG1248. The numbering of the amino acids (right-hand column) corresponds to the mature chimeric protein HSA-vWF470.fwdarw.713 (829 residues). The Thr470, Leu494, Asp498, Pro502, Tyr508, Leu694, Pro704 and Pro708 residues of the mature vWF are underlined.

FIGS. 5A, 5B, and 5C collectively show the characterization of the material secreted after 4 days of culture (erlenmeyers) of the strain CBS 293.91 transformed with the plasmids pYG1248 (plasmid for expression of a chimera of the HSA-vWF Thr470.fwdarw.Val713) and pKan707 (control plasmid). In this experiment, the polypeptides for FIGS. 5A, 5B, and 5C were run on the same gel (8.5% SDS-PAGE) and then treated separately.

FIG. 5A shows the results of coomassie blue staining of a molecular weight standard (lane 2); of a supernatant equivalent to 50 .mu.l of the culture transformed with the plasmid pKan707 in YPL medium (lane 1); the plasmid pYG1248 in YPD medium (lane 3) and the plasmid pYG1248 in YPL medium (lane 4).

FIG. 5B shows the results of immunological characterization of the secreted material after using mouse antibodies directed against human vWF. The lanes are the same as described for FIG. 5A except that biotinilated molecular weight standards were used (lane 2).

FIG. 5C shows the results of immunological characterization of the secreted material after using rabbit antibodies directed against human albumin: supernatant equivalent to 50 .mu.l of the culture transformed with the plasmid pKan707 in YPL medium (lane 1), the plasmid pYG1248 in YPD medium (lane 2) the plasmid pYG1248 in YPL medium (lane 3).

FIGS. 6A and 6B show the kinetic analysis of secretion of a chimera of the invention by the strain CBS 293.91 transformed with the plasmid pYG1206 (HSA-vWF Leu694-Pro708).

In FIG. 6A, coomassie blue staining was employed. Lane 1 is the molecular weight standard, lane 2 is the supernatant equivalent to 2.5 .mu.l of a "Fed Batch" culture in YPD medium after 24 hours of growth; lane 3 is the supernatant of the same culture after 40 hours; and lane 4 is the supernatant of the same culture after 46 hours of growth.

FIG. 6B shows the results of immunological characterization of the secreted material after using mouse antibodies directed against the human vWF. The lanes are the same as in FIG. 6A except that biotinilated molecular weight standards were used.

FIG. 7: Characterization of the material secreted by K. lactis transformed with the plasmids pKan707 (control plasmid, lane 2), pYG1206 (lane 3), pYG1214 (lane 4) and pYG1223 (lane 5); molecular weight standard (lane 1). The deposits correspond to 50 .mu.l of supernatant from a stationary culture after growing in YPD medium, running on an 8.5% acrylamide gel and staining with coomassie blue.

FIG. 8: Nucleotide sequence (SEQ ID NO:11) and amino acid sequence (SEQ ID NO:12) of the MstII-HindIII restriction fragment of the plasmid pYG1341 (HSA-UK1.fwdarw.135). The limit of the EGF-like domain (UK1.fwdarw.46) present in the MstII-HindIII restriction fragment of the plasmid pYG1340 is indicated. The numbering of the amino acids corresponds to the mature chimeric protein SAU-UK1.fwdarw.135 (720 residues).

FIG. 9: Secretion of the HSA-UK1-46 and HSA-UK1-135 chimeras by the strain CBS 293.91 respectively transformed with the plasmids pYG1343 (HSA-UK1-46) and pYG1345 (HSA-UK1-135), after 4 days of growth (YPL+G418 medium). The deposits (equivalent to 50 .mu.l of culture) are run on an 8.5% PAGE-SDS gel and stained with coomassie blue: supernatant from a clone transformed with the plasmids pKan707 (lane 1), pYG1343 (lane 3) or pYG1345 (lane 4); molecular weight standard (lane 2).

FIG. 10: Nucleotide sequence (SEQ ID NO:13) and amino acid sequence (SEQ ID NO:14) of the MstII-HindIII restriction fragment of the plasmid pYG1259 (HSA-G.CSF). The limit of the G-CSF part (174 residues) is indicated. The ApaI and SstI (SstI) restriction sites are underlined. The numbering of the amino acids corresponds to the mature chimeric protein HSA-G.CSF (759 residues).

FIG. 11(a) and 11(b) together comprise the nucleotide sequence (SEQ ID NO:15) and amino acid sequence (SEQ ID NO:16) of the HindIII restriction fragment of the plasmid pYG1301 (chimera G.CSF-Gly.sub.4 -HSA). The black arrows indicate the end of the "pre" and "pro" regions of HSA. The ApaI, SstI (SacI) and MstII restriction sites are underlined. The G.CSF (174 residues) and HSA (585 residues) domains are separated by the synthetic linker GGGG. The numbering of the amino acids corresponds to the mature chimeric protein G.CSF-Gly4-SAH (763 residues). The nucleotide sequence between the translation termination codon and the HindIII site comes from the HSA complementary DNA (cDNA) as described in Patent Application EP 361 991.

FIGS. 12A, 12B, and 12C collectively show the characterization of the material secreted after 4 days of culture (erlenmeyers) of the strain CBS 293.91 transformed with the plasmids pYG1266 (plasmid for expression of a chimera of the HSA-G.CSF type) and pKan707 (control plasmid). In this experiment, the polypeptides for FIGS. 12A, 12B, 12C were run on the same gel (8.5% SDS-PAGE) and then treated separately.

FIG. 12A shows the results of coomassie blue staining of a molecular weight standard (lane 2); supernatant equivalent to 100 .mu.l of culture transformed with the plasmid pKan707 in YPL medium (lane 1); the plasmid pYG1266 in YPD medium (lane 3) and the plasmid pYG1266 in YPL medium (lane 4).

FIG. 12B shows the results of immunological characterization of the material secreted after using primary antibodies directed against human G-CSF. The lanes are as described above for FIG. 12A.

FIG. 12C shows the results of immunological characterization of the material secreted after using primary antibodies directed against human albumin. The lanes are as described above for FIG. 12A.

FIGS. 13A and B collectively show the characterization of the material secreted after 4 days of culture (erlenmeyers in YPD medium) of the strain CBS 293.91 transformed with the plasmids pYG1267 (chimera HSA-G.CSF), pYG1303 (chimera G.CSF-Gly.sub.4 -HSA) and pYG1352 (chimera HSA-Gly.sub.4 -G.CSF) after running on an 8.5% SDS-PAGE gel.

FIG. 13A shows the results of coomassie blue staining of a supernatant equivalent to 100 .mu.l of the culture transformed with the plasmid pYG1303 (lane 1), the plasmid pYG1267 (lane 2), and the plasmid pYG1352 (lane 3). Lane 4 is the molecular weight standard.

B, immunological characterization of the material secreted after using primary antibodies directed against the human G-CSF: same legend as in A.

FIG. 14: Nucleotide sequence (SEQ ID NO:17) and amino acid sequence (SEQ ID NO:18) of the MstII-HindIII restriction fragment of the plasmid pYG1382 (HSA-Fv'). The VH (124 residues) and VL (107 residues) domains of the Fv' fragment are separated by the synthetic linker (GGGGS).sub.x3. The numbering of the amino acids corresponds to the mature chimeric protein HSA-Fv' (831 residues).

FIGS. 15A and 15B collectively show the characterization of the secretions of the chimera HSA-Fv' by the strain CBS 293.91 transformed with the plasmid pYG1383 (LAC4) after 4 days of growth in erlenmeyers at 28.degree. C. in YPD medium (lane 2), and in YPL medium (lane 3). Lane 1 shows the molecular weight standard. The deposits, equivalent to 200 .mu.l of culture (precipitation with ethanol), are run on a PAGE-SDS gel (8.5%).

FIG. 15A shows the results of coomassie blue staining of the gel.

FIG. 15B shows the results of immunological characterization of the material secreted after using primary antibodies directed against HSA.

FIG. 16: Assay of the in vitro antagonistic activity of the agglutination of human platelets fixed with formaldehyde: IC50 of the hybrids HSA-vWF694-708, ›HSA-vWF470-713 C471G, C474G! and ›HSA-vWF470-704 C471G, C474G! compared with the standard RG12986. The determination of the dose-dependent inhibition of the platelet agglutination is carried out according to the method described by C. Prior et al. ›Bio/Technology (1992) 10 66! using an aggregameter recording the variations in optical transmission, with stirring, at 37.degree. C. in the presence of human vWF, botrocetin (8.2 mg/ml) of the test product at various dilutions. The concentration of the product which makes it possible to inhibit the control agglutination (in the absence of product) by half is then determined (IC50).

FIG. 17: Activity on the in vitro cellular proliferation of the murine line NFS60. The radioactivity (.sup.3 H-thymidine) incorporated into the cellular nuclei after 6 hours of incubation is represented on the y-axis (cpm); the quantity of product indicated on the x-axis is expressed in molarity (arbitrary units).

FIG. 18: Activity on granulopoiesis in vivo in rats. The number of neutrophils (average for 7 animals) is indicated on the y-axis as a function of time. The products tested are the chimera HSA-G.CSF (pYG1266), 4 or 40 mg/rat/day), the reference G-CSF (10 mg/rat/day), the recombinant HSA purified from Kluyveromyces lactis supernatant (HSA, 30 mg/rat/day, cf. EP 361 991), or physiological saline.

The methods conventionally used in molecular biology, such as the preparative extractions of plasmid DNA, the centrifugation of plasmid DNA in caesium chloride gradient, electrophoresis on agarose or acrylamide gels, purification of DNA fragments by electroelution, extractions of proteins with phenol or phenol-chloroform, DNA precipitation in saline medium with ethanol or isopropanol, transformation in Escherichia coli, and the like are well known to persons skilled in the art and are widely described in the literature ›Maniatis T. et al., "Molecular Cloning, a Laboratory Manual", Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., 1982; Ausubel F. M. et al. (eds), "Current Protocols in Molecular Biology", John Wiley & Sons, New York, 1987!.

The restriction enzymes were provided by New England Biolabs (Biolabs), Bethesda Research Laboratories (BRL) or Amersham and are used according to the recommendations of the suppliers.

The pBR322 and pUC type plasmids and the phages of the M13 series are of commercial origin (Bethesda Research Laboratories).

For the ligations, the DNA fragments are separated according to their size by electrophoresis on agarose or acrylamide gels, extracted with phenol or with a phenol/chloroform mixture, precipitated with ethanol and then incubated in the presence of phage T4 DNA ligase (Biolabs) according to the recommendations of the manufacturer.

The filling of the protruding 5' ends is carried out by the Klenow fragment of DNA polymerase I of E. coli (Biolabs) according to the specifications of the supplier. The destruction of the protruding 3' ends is carried out in the presence of phage T4 DNA polymerase (Biolabs) used according to the recommendations of the manufacturer. The destruction of the protruding 5' ends is carried out by a controlled treatment with S1 nuclease.

Site-directed mutagenesis in vitro with synthetic oligodeoxynucleotides is carried out according to the method developed by Taylor et al. ›Nucleic Acids Res. 13 (1985) 8749-8764! using the kit distributed by Amersham.

The enzymatic amplification of DNA fragments by the so-called PCR technique ›Polymerase-catalyzed Chain Reaction, Saiki R. K. et al., Science 230 (1985) 1350-1354; Mullis K. B. and Faloona F. A., Meth. Enzym. 155 (1987) 335-350! is carried out using a "DNA thermal cycler" (Perkin Elmer Cetus) according to the specifications of the manufacturer.

The verification of the nucleotide sequences is carried out by the method developed by Sanger et al. ›Proc. Natl. Acad. Sci. U.S.A., 74 (1977) 5463-5467! using the kit distributed by Amersham.

The transformations of K. lactis with DNA from the plasmids for expression of the proteins of the present invention are carried out by any technique known to persons skilled in the art, and of which an example is given in the text.

Except where otherwise stated, the bacterial strains used are E. coli MC1060 (lacIPOZYA, X74, galU, galK, strA.sup.r), or E. coli TG1 (lac, proA,B, supE, thi, hsdD5/FtraD36, proA.sup.+ B.sup.+, lacI.sup.q, lacZ, M15).

The yeast strains used belong to the budding yeasts and more particularly to yeasts of the genus Kluyveromyces. The K. lactis MW98-8C (a, uraA, arg, lys, K.sup.+, pKD1.sup.o) and K. lactis CBS 293.91 strain were particularly used; a sample of the MW98-8C strain was deposited on 16 Sep. 1988 at Centraalbureau voor Schimmelkulturen (CBS) at Baarn (the Netherlands) where it was registered under the number CBS 579.88.

A bacterial strain (E. coli) transformed with the plasmid pET-8c52K was deposited on 17 Apr. 1990 with the American Type Culture Collection under the number ATCC 68306.

The yeast strains transformed with the expression plasmids encoding the proteins of the present invention are cultured in erlenmeyers or in 21 pilot fermenters (SETRIC, France) at 28.degree. C. in rich medium (YPD: 1% yeast extract, 2% Bactopeptone, 2% glucose; or YPL: 1% yeast extract, 2% Bactopeptone, 2% lactose) with constant stirring.

The plasmid pYG404 is described in Patent Application EP 361 991. This plasmid contains a HindIII restriction fragment encoding the prepro-HSA gene preceded by the 21 nucleotides naturally present immediately upstream of the initiator ATG for translation of the PGK gene of S. cerevisiae. The nucleotide sequence of this restriction fragment is included in that of FIG. 2. The MstII site localized in the coding sequence, three residues from the codon specifying the end of translation is particularly useful as site for cloning a biologically active peptide which it is desired to couple in translational phase at the C-terminus of HSA. In a specific embodiment, it is useful to use peptides whose sequence is encoded by an MstII-HindIII restriction fragment of the type: 5'-CCTTAGGCTTA ›3xN! .sub.p TAAGCTT-3' (SEQ ID NO:20), the sequence encoding the biologically active peptide (p residues) is ›3xN!.sub.p). The ligation of this fragment to the HindIII-MstII restriction fragment corresponding to the entire gene encoding HSA, with the exception of the three C-terminalmost amino acids (leucin-glycine-leucin residues) generates a HindIII restriction fragment containing a hybrid gene encoding a chimeric protein of the HSA-PEPTIDE type (FIG. 1, panel A), immediately preceded by the "prepro" export region of HSA. In another embodiment, the biologically active peptide may be present more than once in the chimera.

In a specific embodiment, the combined techniques of site-directed mutagenesis and PCR amplification make it possible to construct hybrid genes encoding a chimeric protein resulting from the translational coupling between a signal peptide (and for example the prepro region of HSA), a sequence including the biologically active peptide and the mature form of HSA or one of its molecular variants. These hybrid genes are preferably bordered in 5' of the translational initiator ATG and in 3' of the translational stop codon by HindIII restriction sites and encode chimeric proteins of the PEPTIDE-HSA type (FIG. 1, panel B). In a still more specific embodiment, the biologically active peptide may be present more than once in the chimera.

The combined techniques of site-directed mutagenesis and PCR amplification described in Examples 1 and 2 make it possible to construct hybrid genes encoding a chimeric protein resulting from the translational coupling between the mature form of HSA, or one of its molecular variants, and a biologically active peptide coupled to the N- and C-terminal ends of HSA. These hybrid genes are preferably bordered in 5' of the translational initiator ATG and in 3' of the translational stop codon by HindIII restriction sites and encode chimeric proteins of the PEPTIDE-HSA-PEPTIDE type (FIG. 1, panel C), immediately preceded by the "prepro" export region of HSA. In a still more specific embodiment, the biologically active peptide may be present more than once in the chimera.

The chimeric proteins of the preceding examples can be expressed in yeasts using functional, regulatable or constitutive promoters such as, for example, those present in the plasmids pYG105 (LAC4 promoter of Kluyveromyces lactis), pYG106 (PGK promoter of Saccharomyces cerevisiae), pYG536 (PHO5 promoter of S. cerevisiae), or hybrid promoters such as those described in Patent Application EP 361 991. The plasmids pYG105 and pYG106 are particularly useful here because they permit the expression of the genes encoded by the HindIII restriction fragments as described in the preceding examples and cloned into the HindIII site and in the productive orientation (defined as the orientation which places the "prepro" region of albumin proximally relative to the promoter for transcription), using promoters which are functional in K. lactis, regulatable (pYG105) or constitutive (pYG106). The plasmid pYG105 corresponds to the plasmid pKan707 described in Patent Application EP 361 991 in which the HindIII restriction site which is unique and localized in the gene for resistance to geneticin (G418) has been destroyed by site-directed mutagenesis while preserving an unchanged protein (oligodeoxynucleotide 5'-GAAATGCATAAGCTCTTGCCATTCTCACCG-3')(SEQ ID NO:21). The SalI-SacI fragment encoding the URA3 gene of the mutated plasmid was then replaced with a SalI-SacI restriction fragment containing an expression cassette consisting of the LAC4 promoter of K. lactis (in the form of a SalI-HindIII fragment) and the terminator of the PGK gene of S. cerevisiae (in the form of a HindIII-SacI fragment). The plasmid pYG105 is mitotically very stable in the Kluyveromyces yeasts and a restriction map thereof is given in FIG. 3. The plasmids pYG105 and pYG106 differ from each other only in the nature of the promoter for transcription encoded by the SalI-HindIII fragment.

The transformation of the yeasts belonging to the genus Kluyveromyces, and in particular the strains MW98-8C and CBS 293.91 of K. lactis is carried out for example by the technique for treating whole cells with lithium acetate ›Ito H. et al., J. Bacteriol. 153 (1983) 163-168!, adapted as follows. The growth of the cells is carried out at 28.degree. C. in 50 ml of YPD medium, with stirring and up to an optical density of 600 nm (OD.sub.600) of between 0.6 and 0.8; the cells are harvested by centrifugation at low speed, washed in a sterile solution of TE (10 mM Tris HCl pH 7.4; 1 mM EDTA), resuspended in 3-4 ml of lithium acetate (0.1M in TE) in order to obtain a cellular density of about 2.times.10.sup.8 cells/ml, and then incubated at 30.degree. C. for 1 hour with moderate stirring. Aliquots of 0.1 ml of the resulting suspension of competent cells are incubated at 30.degree. C. for 1 hour in the presence of DNA and at a final concentration of 35% polyethylene glycol (PEG.sub.4000, Sigma). After a heat shock of 5 minutes at 42.degree. C., the cells are washed twice, resuspended in 0.2 ml of sterile water and incubated for 16 hours at 28.degree. C. in 2 ml of YPD medium in order to permit the phenotypic expression of the gene for resistance to G418 expressed under the control of the P.sub.kl promoter (cf. EP 361 991); 200 .mu.l of the cellular suspension are then plated on selective YPD dishes (G418, 200 .mu.g/ml). The dishes are incubated at 28.degree. C. and the transformants appear after 2 to 3 days of cell growth.

After selection on rich medium supplemented with G418, the recombinant clones are tested for their capacity to secrete the mature form of the chimeric proteins. Few clones, corresponding to the strain CBS 293.91 or MW98-8C transformed by the plasmids for expression of the chimeras between HSA and the biologically active part, are incubated in YPD or YPL medium at 28.degree. C. The cellular supernatants are recovered by centrifugation when the cells reach the stationary growth phase, optionally concentrated 10 times by precipitation for 30 minutes at -20.degree. C. in a final concentration of 60% ethanol, and then tested after electrophoresis on an 8.5% SDS-PAGE gel, either directly by staining the gel with coomassie blue, or after immunoblotting using primary antibodies directed against the biologically active part or a rabbit polyclonal serum directed against HSA. During the experiments for immunological detection, the nitrocellulose filter is first incubated in the presence of specific primary antibodies, washed several times, incubated in the presence of goat antibodies directed against the primary antibodies, and then incubated in the presence of an avidin-peroxidase complex using the "ABC kit" distributed by Vectastain (Biosys S. A., Compiegne, France). The immunological reaction is then revealed by the addition of 3,3'-diamino benzidine tetrahydrochloride (Prolabo) in the presence of hydrogen peroxide, according to the recommendations of the manufacturer.

The plasmid pET-8c52K contains a fragment of the vWF cDNA encoding residues 445 to 733 of human vWF and therefore includes several crucial determinants of the interaction between vWF and the platelets on the one hand, and certain elements of the basal membrane and the sub-endothelial tissue on the other, and especially the peptides G10 and D5 which antagonize the interaction between vWF and GP1b ›Mori H. et al., J. Biol. Chem. 263 (1988) 17901-17904!. This peptide sequence is identical to the corresponding sequence described by Titani et al. ›Biochemistry 25, (1986) 3171-3184!. The amplification of these genetic determinants can be carried out using the plasmid pET-8c52K, for example by the PCR amplification technique, using as primer oligodeoxynucleotides encoding contiguous residues localized on either side of the sequence to be amplified. The amplified fragments are then cloned into vectors of the M13 type for their verification by sequencing using either the universal primers situated on either side of the multiple cloning site, or oligodeoxynucleotides specific for the amplified region of the vWF gene of which the sequence of several isomorphs is known ›Sadler J. E. et al., Proc. Natl. Acad. Sci. 82 (1985) 6394-6398; Verweij C. L. et al., EMBO J. 5 (1986) 1839-1847; Shelton-Inloes B. B. et al., Biochemistry 25 (1986) 3164-3171; Bonthron D. et al., Nucleic Acids Res. 17 (1986) 7125-7127!. Thus, the PCR amplification of the plasmid pET-8c52K with the oligodeoxynucleotides 5'-CCCGGGATCCCTTAGGCTTAACCTGTGAAGCCTGC-3' (SEQ ID NO:22) (Sq1969, the MstII site is underlined) and 5'-CCCGGGATCCAAGCTTAGACTTGTGCCATGTCG-3' (SEQ ID NO:23) (Sq2029, the HindIII site is underlined) generates an MstII-HindIII restriction fragment including the Thr470 to Val713 residues of vWF (FIG. 4, panel E). The ligation of this fragment to the HindIII-MstII restriction fragment corresponding to the entire gene encoding HSA, with the exception of the three C-terminalmost amino acids (cf. FIG. 2) generates a HindIII restriction fragment containing a hybrid gene encoding a chimeric protein of the HSA-PEPTIDE type (FIG. 1, panel A), immediately preceded by the "prepro" export region of HSA. This restriction fragment is cloned in the productive orientation and into the HindIII site of the plasmid pYG105, which generates the expression plasmid pYG1248 (HSA-vWF470-713).

In another embodiment, the binding site of vWF is a peptide including the Thr470 to Asp498 residues of the mature vWF. This sequence including the peptide G10 (Cys474-Pro488) described by Mori et al. ›J. Biol. Chem. 263 (1988) 17901-17904! and capable of antagonizing the interaction of human vWF with the GP1b of the human platelets. The sequence corresponding to the peptide G10 is first included in an MstII-HindIII restriction fragment (FIG. 4, panel B), for example by PCR amplification of the plasmid pET-8c52K with the oligodeoxynucleotides Sq1969 and 5'-CCCGGGATCCAAGCTTAGTCCTCCACATACAG-3' (SEQ ID NO:24) (Sq1970, the HindIII site is underlined), which generates an MstII-HindIII restriction fragment including the peptide G10, and whose sequence is: 5'-CCTTAGGCTTAACCTGTGAAGCCTG-CCAGGAGCCGGGAGGCCTGGTGGTGCCTCCCACAGATGCCCCGGT GAGCCCC-ACCACTCTGTATGTGGAGGACTAAGCTT-3' (SEQ ID NO:25) (the sequence encoding the peptide G10 is in bold characters). The ligation of this fragment to the HindIII-MstII restriction fragment corresponding to the entire gene encoding HSA, with the exception of the three C-terminalmost amino acids (cf. FIG. 2) generates a HindIII restriction fragment containing a hybrid gene encoding a chimeric protein of the HSA-PEPTIDE type (FIG. 1, panel A), immediately preceded by the "prepro" export region of HSA. This restriction fragment is cloned in the productive orientation into the HindIII site of the plasmid pYG105, which generates the expression plasmid pYG1214.

In another embodiment, the site for binding of vWF to GP1b is directly designed with the aid of synthetic oligodeoxynucleotides, and for example the oligodeoxynucleotides 5'-TTAGGCCTCTGTGACCTTGCCCCTGA-AGCCCCTCCTCCTACTCTGCCCCCCTAAGCTTA-3' (SEQ ID NO:26) and 5'-GATCTAAG-CTTAGGGGGGCAGAGTAGGAGGAGGGGCTTCAGGGGCAAGGTCACAGAGGCC-3' (SEQ ID NO:27). These oligodeoxynucleotides form, by pairing, a MstII-BglII restriction fragment including the MstII-HindIII fragment (FIG. 4, panel C) corresponding to the peptide D5 defined by the Leu694 to Pro708 residues of vWF. The ligation of the MstII-HindIII fragment to the HindIII-MstII restriction fragment corresponding to the entire gene encoding HSA with the exception of the three C-terminalmost amino acids (cf. FIG. 2) generates a HindIII restriction fragment containing a hybrid gene encoding a chimeric protein of the HSA-PEPTIDE type (FIG. 1, panel A), immediately preceded by the "prepro" export region of HSA. This restriction fragment is cloned in the productive orientation into the HindIII site of the plasmid pYG105, which generates the expression plasmid pYG1206.

Useful variants of the plasmid pET-8c52K are deleted by site-directed mutagenesis between the peptides G10 and G5, for example sites for binding to collagen, and/or to heparin, and/or to botrocetin, and/or to sulphatides and/or to ristocetin. One example is the plasmid pMMB9 deleted by site-directed mutagenesis between the residues Cys509 and Ile662. The PCR amplification of this plasmid with the oligodeoxynucleotides Sq1969 and Sq2029 generates an MstII-HindIII restriction fragment (FIG. 4, panel D) including the Thr470 to Tyr508 and Arg663 to Val713 residues and in particular the peptides G10 and D5 of vWF and deleted in particular of its site for binding to collagen localized between the residues Glu542 and Met622 ›Roth G. J. et al., Biochemistry 25 (1986) 8357-8361!. The ligation of this fragment to the HindIII-MstII restriction fragment corresponding to the entire gene encoding HSA, with the exception of the three C-terminalmost amino acids (cf. FIG. 2) generates a HindIII restriction fragment containing a hybrid gene encoding a chimeric protein of the HSA-PEPTIDE type (FIG. 1, panel A), immediately preceded by the "prepro" export region of HSA. This restriction fragment is cloned in the productive orientation into the HindIII site of the plasmid pYG105, which generates the expression plasmid pYG1223.

In other embodiments, the use of combined techniques of site-directed mutagenesis and PCR amplification makes it possible to generate at will variants of the MstII-HindIII restriction fragment of panel A of FIG. 4 but deleted of one or more sites for binding to sulphatides and/or to botrocetin and/or to heparin and/or to collagen, and/or substituted by any residue involved in the vWF-associated emergence of IIB type pathologies.

In other useful variants of the plasmid pET-8c52K, mutations are introduced, for example by site-directed mutagenesis, in order to replace or suppress all or part of the set of cysteines present at positions 471, 474, 509 and 695 of the human vWF. Specific examples are the plasmids p5E and p7E in which the cysteins present at positions 471 and 474, on the one hand, and at positions 471, 474, 509 and 695, on the other hand, have been respectively replaced by glycine residues. The PCR amplification of these plasmids with the oligodeoxynucleotides Sq2149 (5'-CCCGGGATCCCTTAGGCTTAACCGGTGAAGCCGGC-3' (SEQ ID NO:28), the MstII site is underlined) and Sq2029 makes it possible to generate MstII-HindIII restriction fragments including the Thr470 to Val713 residues of the natural vWF with the exception that at least the cystein residues at positions 471 and 474 were mutated to glycine residues. The ligation of these fragments to the HindIII-MstII restriction fragment corresponding to the entire gene encoding HSA with the exception of the three C-terminalmost amino acids (cf. FIG. 2) generates a HindIII restriction fragment containing a hybrid gene encoding a chimeric protein of the HSA-PEPTIDE type (FIG. 1, panel A), immediately preceded by the "prepro" export region of HSA. These restriction fragments are cloned in the productive orientation into the HindIII site of the plasmid pYG105, which generates the expression plasmids pYG1283 (chimera HSA-vWF470-713, C471G, C474G) and pYG1279 (chimera HSA-vWF470-713, C471G, C474G, C509G, C695G).

Other particularly useful mutations affect at least one residue involved in vWF-associated type IIB pathologies (increase in the intrinsic affinity of vWF for GP1b), such as the residues Arg543, Arg545, Trp550, Val551, Val553, Pro574 or Arg578 for example. The genetic recombination techniques in vitro also make it possible to introduce at will one or more additional residues into the sequence of vWF and for example a supernumerary methionine between positions Asp539 and Glu542.

In a specific embodiment, the sites for binding of vWF to the components of the sub-endothelial tissue, and for example collagen, are generated by PCR amplification of the plasmid pET-8c52K, for example with the oligodeoxynucleotides Sq2258 (5'-GGATCCTTAGGGCTGTGCAGCAGGCTACTGGACCTGGTC-3' (SEQ ID NO:29), the MstII site is underlined) and Sq2259 (5'-GAATTCAAGCTTAACAGAGGTAGCTAA-CGATCTCGTCCC-3' (SEQ ID NO:30), the HindIII site is underlined), which generates an MstII-HindIII restriction fragment encoding the Cys509 to Cys695 residues of the natural vWF. Deletion molecular variants or modified variants are also generated which contain any desired combination between the sites for binding of vWF to the sulphatides and/or to botrocetin and/or to heparin and/or to collagen and/or any residue responsible for a modification of the affinity of vWF for GP1b (vWF-associated type II pathologies). In another embodiment, the domain capable of binding to collagen may also come from the vWF fragment which is between the residues 911 and 1114 and described by Pareti et al. ›J. Biol. Chem. (1987) 262: 13835-13841!. The ligation of these fragments to the HindIII-MstII restriction fragment corresponding to the entire gene encoding HSA with the exception of the three C-terminalmost amino acids (cf. FIG. 2) generates HindIII restriction fragments containing a hybrid gene encoding a chimeric protein of the HSA-PEPTIDE type (FIG. 1, panel A), immediately preceded by the "prepro" export region of HSA. These restriction fragments are cloned in the productive orientation into the HindIII site of the plasmid pYG105, which generates the corresponding expression plasmids, and for example the plasmid pYG1277 (HSA-vWF509-695).

The chimeras present in the culture supernatants corresponding to the CBS 293.91 strain transformed, for example with the expression plasmids according to Examples E.7.1. and E.7.2., are characterized in a first instance by means of antibodies specific for the HSA part and for the vWF part. The results of FIGS. 5 to 7 demonstrate that the yeast K. lactis is capable of secreting chimeric proteins between HSA and a fragment of vWF, and that these chimeras are immunologically reactive. It may also be desirable to purify some of these chimeras. The culture is then centrifuged (10,000 g, 30 min), the supernatant is passed through a 0.22 mm filter (Millipore) and then concentrated by ultrafiltration (Amicon) using a membrane whose discrimination threshold is situated at 30 kDa. The concentrate obtained is then dialysed against a Tris-HCl solution (50 mM pH 8) and then purified on a column. For example, the concentrate corresponding to the culture supernatant of the CBS 293.91 strain transformed with the plasmid pYG1206 is purified by affinity chromatography on Blue-Trisacryl (IBF). A purification by ion-exchange chromatography can also be used. For example, in the case of the chimera HSA-vWF470-713, the concentrate obtained after ultrafiltration is dialysed against a Tris-HCl solution (50 mM pH 8), and then loaded in 20 ml fractions onto a cation-exchange column (5 ml) (S Fast Flow, Pharmacia) equilibrated in the same buffer. The column is then washed several times with the Tris-HCl solution (50 mM pH 8) and the chimeric protein is then eluted from the column by an NaCl gradient (0 to 1M). The fractions containing the chimeric protein are then pooled and dialysed against a 50 mM Tris-HCl solution (pH 8) and then reloaded onto the S Fast Flow column. After elution of the column, the fractions containing the protein are pooled, dialysed against water and freeze-dried before characterization: for example, sequencing (Applied Biosystem) of the protein ›HSA-vWF470-704 C471G, C474G! secreted by the yeast CBS 293.91 gives the N-terminal sequence expected for HSA (Asp-Ala-His . . . ), demonstrating a correct maturation of the chimera immediately at the C-terminus of the doublet of residues Arg-Arg of the "pro" region of HSA (FIG. 2). The essentially monomeric character of the chimeric proteins between HSA and vWF is also confirmed by their elution profile on a TSK 3000 column ›Toyo Soda Company, equilibrated with a cacodylate solution (pH 7) containing 0.2M Na.sub.2 SO.sub.4 !: for example the chimera ›HSA-vWF 470-704 C471G, C474G! behaves under the conditions like a protein with an apparent molecular weight of 95 kDa, demonstrating its monomeric character.

A fragment corresponding to the amino-terminal fragment of urokinase (ATF: EGF-like domain+ringle domain) can be obtained from the corresponding messenger RNA of cells of certain human carcinoma, for example using the RT-PCR kit distributed by Pharmacia. An MstII-HindIII restriction fragment including the ATF of human urokinase is given in FIG. 8. The ligation of the HindIII-MstII fragment of the plasmid pYG404 to this MstII-HindIII fragment makes it possible to generate the HindIII fragment of the plasmid pYG1341 which encodes a chimeric protein in which the HSA molecule is genetically coupled to the ATF (HSA-UK1.fwdarw.135). Likewise, the plasmid pYG1340 contains a HindIII fragment encoding a chimera composed of HSA immediately followed by the first 46 residues of human urokinase (HSA-UK1.fwdarw.46, cf. FIG. 8). The cloning in the productive orientation, of the HindIII restriction fragment of the plasmid pYG1340 (HSA-UK1.fwdarw.46) into the HindIII site of the plasmids pYG105 (LAC4) and pYG106 (PGK) generates the expression plasmids pYG1343 and pYG1342 respectively. Likewise, the cloning, in the productive orientation, of the HindIII restriction fragment of the plasmid pYG1341 (HSA-UK1.fwdarw.135) into the HindIII site of the plasmids pYG105 (LAC4) and pYG106 (PGK) generates the expression plasmids pYG1345 and pYG1344 respectively.

After selection on rich medium supplemented with G418, the recombinant clones are tested for their capacity to secrete the mature form of the chimeric proteins HSA-UK. A few clones corresponding to the strain K. lactis CBS 293.91, which is transformed with the expression plasmids according to Example E.9.1., are incubated in selective complete liquid medium at 28.degree. C. The cellular supernatants are then tested after electrophoresis on an 8.5% acrylamide gel, either directly by staining of the gel with coomassie blue, or after immunoblotting using as primary antibodies a rabbit polyclonal serum directed against human albumin or against human urokinase. The results of FIG. 9 demonstrate that the hybrid proteins HSA-UK1.fwdarw.46 and HSA-UK1.fwdarw.135 are particularly well secreted by the yeast Kluyveromyces.

After centrifugation of a culture of the CBS 293.91 strain transformed with the expression plasmids according to Example E.8.1., the culture supernatant is passed through a 0.22 mm filter (Millipore) and then concentrated by ultrafiltration (Amicon) using a membrane whose discrimination threshold is situated at 30 kDa. The concentrate obtained is then adjusted to 50 mM Tris-HCl starting with a stock solution of 1M Tris-HCl (pH 7), and then loaded in 20 ml fractions onto an anion-exchange column (3 ml) (D-Zephyr, Sepracor) equilibrated in the same buffer. The chimeric protein (HSA-UK1.fwdarw.46 or HSA-UK1.fwdarw.135) is then eluted from the column by a gradient (0 to 1M) of NaCl. The fractions containing the chimeric protein are then pooled and dialysed against a 50 mM Tris-HCl solution (pH 6) and reloaded onto a D-Zephyr column equilibrated in the same buffer. After elution of the column, the fractions containing the protein are pooled, dialysed against water and freeze-dried before characterization of their biological activity and especially with respect to their ability to displace urokinase from its cellular receptor.

An MstII-HindIII restriction fragment including the mature form of human G-CSF is generated, for example according to the following strategy: a KpnI-HindIII restriction fragment is first obtained by the enzymatic PCR amplification technique using the oligodeoxynucleotides Sq2291 (5'-CAAGGATCC-AAGCTTCAGGGCTGCGCAAGGTGGCGTAG-3' (SEQ ID NO:31), the HindIII site is underlined) and Sq2292 (5'-CGGGGTACCTTAGGCTTAACCCCCCTG-GGCCCTGCCAGC-3' (SEQ ID NO:32), the KpnI site is underlined) as primer on the plasmid BBG13 serving as template. The plasmid BBG13 contains the gene encoding the B form (174 amino acids) of mature human G-CSF, which is obtained from British Bio-technology Limited, Oxford, England. The enzymatic amplification product of about 550 nucleotides is then digested with the restriction enzymes KpnI and HindIII and cloned into the vector pUC19 cut with the same enzymes, which generates the recombinant plasmid pYG1255. This plasmid is the source of an MstII-HindIII restriction fragment which makes it possible to fuse G-CSF immediately downstream of HSA (chimera HSA-G.CSF) and whose nucleotide sequence is given in FIG. 10.

It may also be desirable to insert a peptide linker between the HSA part and G-CSF, for example in order to permit a better functional presentation of the transducing part. An MstII-HindIII restriction fragment is for example generated by substitution of the MstII-ApaI fragment of the plasmid pYG1255 by the oligodeoxynucleotides Sq2742 (5'-TTAGGCTTAGGTGGTGGCGGT-ACCCCCCTGGGCC-3' (SEQ ID NO:33), the codons encoding the glycine residues of this particular linker are underlined) and Sq2741 (5'-CAGGGGGGTACCGCCACCACCTAAGCC-3') (SEQ ID NO:34) which form, by pairing, an MstII-ApaI fragment. The plasmid thus generated therefore contains an MstII-HindIII restriction fragment whose sequence is identical to that of FIG. 10 with the exception of the MstII-ApaI fragment.

The ligation of the HindIII-MstII fragment of the plasmid pYG404 to the MstII-HindIII fragment of the plasmid pYG1255 makes it possible to generate the HindIII fragment of the plasmid pYG1259 which encodes a chimeric protein in which the B form of the mature G-CSF is positioned by genetic coupling in translational phase at the C-terminus of the HSA molecule (HSA-G.CSF).

An identical HindIII restriction fragment, with the exception of the MstII-ApaI fragment, may also be easily generated and which encodes a chimeric protein in which the B form of the mature G-CSF is positioned by genetic coupling in translational phase at the C-terminus of the HSA molecule and a specific peptide linker. For example, this linker consists of 4 glycine residues in the HindIII fragment of the plasmid pYG1336 (chimera HSA-Gly.sub.4 -G.CSF).

The HindIII restriction fragment of the plasmid pYG1259 is cloned in the productive orientation and into the HindIII restriction site of the expression plasmid pYG105, which generates the expression plasmid pYG1266 (HSA-G.CSF). In another exemplification, the cloning of the HindIII restriction fragment of the plasmid pYG1259 in the productive orientation and into the HindIII site of the plasmid pYG106 generates the plasmid pYG1267. The plasmids pYG1266 and pYG1267 are mutually isogenic with the exception of the SalI-HindIII restriction fragment encoding the LAC4 promoter of K. lactis (plasmid pYG1266) or the PGK promoter of S. cerevisiae (plasmid pYG1267).

In another exemplification, the cloning in the productive orientation of the HindIII restriction fragment of the plasmid pYG1336 (chimera HSA-Gly.sub.4 -G.CSF) into the HindIII site of the plasmids pYG105 (LAC4) and pYG106 (PGK) generates the expression plasmids pYG1351 and pYG1352 respectively.

In a specific embodiment, the combined techniques of site-directed mutagenesis and PCR amplification make it possible to construct hybrid genes encoding a chimeric protein resulting from the translational coupling between a signal peptide (and for example the prepro region of HSA), a sequence including a gene having a G-CSF activity, and the mature form of HSA or one of its molecular variants (cf. chimera of panel B, FIG. 1). These hybrid genes are preferably bordered in 5' of the translational initiator ATG and in 3' of the translational stop codon by HindIII restriction sites. For example the oligodeoxynucleotide Sq2369 (5'-GTTCTACGCCACCTTGCGCAG-CCCGGTGGAGGCGGTGATGCACACAAGAGTGAGGTTGCTCATCGG-3' (SEQ ID NO:35) the residues underlined (optional) correspond in this particular chimera to a peptide linker composed of 4 glycine residues) makes it possible, by site-directed mutagenesis, to put in translational phase the mature form of the human G-CSF of the plasmid BBG13 immediately upstream of the mature form of HSA, which generates the intermediate plasmid A. Likewise, the use of the oligodeoxynucleotide Sq2338 ›5'-CAGGGAGCTGGCAGGGCCCAGGGGGGTTCGACGAAACACACCCCTGGAATAAGCCGAGCT-3' (SEQ ID NO:36) (non-coding strand), the nucleotides complementary to the nucleotides encoding the first N-terminal residues of the mature form of the human G-CSF are underlined! makes it possible, by site-directed mutagenesis, to couple in translational reading phase the prepro region of HSA immediately upstream of the mature form of the human G-CSF, which generates the intermediate plasmid B. A HindIII fragment encoding a chimeric protein of the PEPTIDE-HSA type (cf. FIG. 1, panel B) is then generated by combining the HindIII-SstI fragment of the plasmid B (joining prepro region of HSA+N-terminal fragment of the mature G-CSF) with the SstI-HindIII fragment of the plasmid A ›joining mature G-CSF-(glycine).sub.x4 - mature HSA!. The plasmid pYG1301 contains this specific HindIII restriction fragment encoding the chimera G.CSF-Gly.sub.4 -HSA fused immediately downstream of the prepro region of HSA (FIG. 11). The cloning of this HindIII restriction fragment in the productive orientation and into the HindIII site of the plasmids pYG105 (LAC4) and pYG106 (PGK) generates the expression plasmids pYG1302 and pYG1303 respectively.

After selection on rich medium supplemented with G418, the recombinant clones are tested for their capacity to secrete the mature form of the chimeric proteins between HSA and G-CSF. A few clones corresponding to the strain K. lactis CBS 293.91 transformed with the plasmids pYG1266 or pYG1267 (HSA-G.CSF), pYG1302 or pYG1303 (G.CSF-Gly.sub.4 -HSA) or alternatively pYG1351 or pYG1352 (HSA-Gly.sub.4 -G.CSF) are incubated in selective complete liquid medium at 28.degree. C. The cellular supernatants are then tested after electrophoresis on an 8.5% acrylamide gel, either directly by staining the gel with coomassie blue, or after immunoblotting using as primary antibodies rabbit polyclonal antibodies directed against the human G-CSF or a rabbit polyclonal serum directed against human albumin. The results of FIG. 12 demonstrate that the hybrid protein HSA-G.CSF is recognized both by antibodies directed against human albumin (panel C) and human G-CSF (panel B). The results of FIG. 13 indicate that the chimera HSA-Gly.sub.4 -G.CSF (lane 3) is particularly well secreted by the yeast Kluyveromyces, possibly because of the fact that the presence of the peptide linker between the HSA part and the G-CSF part is more favourable to an independent folding of these 2 parts during the transit of the chimera in the secretory pathway. Furthermore, the N-terminal fusion (G.CSF-Gly.sub.4 -HSA) is also secreted by the yeast Kluyveromyces (FIG. 13, lane 1).

After centrifugation of a culture of the CBS 293.91 strain transformed with the expression plasmids according to Example E.9.1., the culture supernatant is passed through a 0.22 mm filter (Millipore) and then concentrated by ultrafiltration (Amicon) using a membrane whose discrimination threshold is situated at 30 kDa. The concentrate obtained is then adjusted to 50 mM Tris-HCl from a 1M stock solution of Tris-HCl (pH 6), and then loaded in 20 ml fractions onto an ion-exchange column (5 ml) (Q Fast Flow, Pharmacia) equilibrated in the same buffer. The chimeric protein is then eluted from the column by a gradient (0 to 1M) of NaCl. The fractions containing the chimeric protein are then pooled and dialysed against a 50 mM Tris-HCl solution (pH 6) and reloaded onto a Q Fast Flow column (1 ml) equilibrated in the same buffer. After elution of the column, the fractions containing the protein are pooled, dialysed against water and freeze-dried before characterization: for example, the sequencing (Applied Biosystem) of the protein HSA-G.CSF secreted by the yeast CBS 293.91 gives the N-terminal sequence expected for HSA (Asp-Ala-His . . . ), demonstrating a correct maturation of the chimera immediately at the C-terminus of the doublet of residues Arg-Arg of the "pro" region of HSA (FIG. 2).

An Fv' fragment can be constructed by genetic engineering techniques, and which encodes the variable fragments of the heavy and light chains of an immunoglobulin (Ig), linked to each other by a linker peptide ›Bird et al., Science (1988) 242: 423; Huston et al., (1988) Proc. Natl. Acad. Sci. 85: 5879!. Schematically, the variable regions (about 120 residues) of the heavy and light chains of a given Ig are cloned from the messenger RNA of the corresponding hybridoma, for example using the RT-PCR kit distributed by Pharmacia (Mouse ScFv module). In a second stage, the variable regions are genetically coupled by genetic engineering via a synthetic linkage peptide and for example the linker (GGGGS).sub.x3. An MstII-HindIII restriction fragment including the Fv' fragment of an immunoglobulin secreted by a murine hybridoma is given in FIG. 14. The ligation of the HindIII-MstII fragment of the plasmid pYG404 to this MstII-HindIII fragment makes it possible to generate the HindIII fragment of the plasmid pYG1382 which encodes a chimeric protein in which the HSA molecule is genetically coupled to the Fv' fragment of FIG. 14 (chimera HSA-Fv'). The cloning in the productive orientation of the HindIII restriction fragment of the plasmid pYG1382 into the HindIII site of the plasmids pYG105 (LAC4) and pYG106 (PGK) generates the expression plasmids pYG1383 and pYG1384 respectively.

After selection on rich medium supplemented with G418, the recombinant clones are tested for their capacity to secrete the mature form of the chimeric protein HSA-Fv'. A few clones corresponding to the strain K. lactis CBS 293.91 transformed with the plasmids pYG1383 or pYG1384 (HSA-Fv') are incubated in selective complete liquid medium at 28.degree. C. The cellular supernatants are then tested after electrophoresis on an 8.5% acrylamide gel, either directly by staining of the gel with coomassie blue, or after immunoblotting using as primary antibodies a rabbit polyclonal serum directed against human albumin, or directly incubated with biotinylated antibodies directed against the immunoglobulins of murine origin. The results of FIG. 15 demonstrate that the hybrid protein HSA-Fv' is recognized both by antibodies directed against human albumin (panel C) and reacts with biotinylated goat antibodies which are immunologically reactive towards mouse immunoglobulins (panel B).

The antagonistic activity of the products is determined by measuring the dose-dependent inhibition of the agglutination of human platelets fixed with paraformaldehyde according to the method described by Prior et al. ›Bio/Technology (1992) 10: 66!. The measurements are carried out in an aggregameter (PAP-4, Bio Data, Horsham, Pa., U.S.A.) which records the variations over time of the optical transmission, with stirring, at 37.degree. C. in the presence of vWF, of botrocetin (8.2 mg/ml) and of the test product at various dilutions (concentrations). For each measurement, 400 ml (8.times.10.sup.7 platelets) of a suspension of human platelets stabilized with paraformaldehyde (0.5%, and then resuspended in ›NaCl (137 mM); MgCl.sub.2 (1 mM); NaH.sub.2 PO.sub.4 (0.36 mM); NaHCO.sub.3 (10 mM); KCl (2.7 mM); glucose (5.6 mM); HSA (3.5 mg/ml); HEPES buffer (10 mM, pH 7.35)! are preincubated at 37.degree. C. in the cylindrical tank (8.75.times.50 mm, Wellcome Distriwell, 159 rue Nationale, Paris) of the aggregameter for 4 min and are then supplemented with 30 ml of the solution of the test product at various dilutions in apyrogenic formulation vehicle ›mannitol (50 g/l); citric acid (192 mg/l); L-lysine monohydrochloride (182.6 mg/l); NaCl (88 mg/l); pH adjusted to 3.5 by addition of NaOH (1M)!, or formulation vehicle alone (control assay). The resulting suspension is then incubated for 1 min at 37.degree. C. and 12.5 ml of human vWF ›American Bioproducts, Parsippany, N.J., U.S.A.; 11% von Willebrand activity measured according to the recommendations for the use of PAP-4 (Platelet Aggregation Profiler.RTM.) with the aid of platelets fixed with formaldehyde (2.times.10.sup.5 platelets/ml), human plasma containing 0 to 100% vWF and ristocetin (10 mg/ml, cf. p. 36-45: vW Program.TM.! are added and incubated at 37.degree. C. for 1 min before adding 12.5 ml of botrocetin solution ›purified from freeze-dried venom of Bothrops jararaca (Sigma) according to the procedure described by Sugimoto et al., Biochemistry (1991) 266: 18172!. The recording of the reading of the transmission as a function of time is then carried out for 2 min with stirring by means of a magnetic bar (Wellcome Distriwell) placed in the tank and with a magnetic stirring of 1,100 rpm provided by the aggregameter. The mean variation of the optical transmission (n.sup.3 5 for each dilution) over time is therefore a measurement of the platelet agglutination due to the presence of vWF and botrocetin, in the absence or in the presence of variable concentrations of the test product. From such recordings, the % inhibition of the platelet agglutination due to each concentration of product is then determined and the straight line giving the % inhibition as a function of the reciprocal of the product dilution in log-log scale is plotted. The IC50 (or concentration of product causing 50% inhibition of the agglutination) is then determined on this straight line. The table of FIG. 6 compares the IC50 values of some of the HSA-vWF chimeras of the present invention and demonstrates that some of them are better antagonists of platelet agglutination than the product RG12986 described by Prior et al. ›Bio/Technology (1992) 10: 66! and included in the assays as standard value. Identical tests for the inhibition of the agglutination of human platelets in the presence of vWF of pig plasma (Sigma) makes it possible, furthermore, to demonstrate that some of the hybrids of the present invention, and especially some type IIB variants, are very good antagonists of platelet agglutination in the absence of botrocetin-type cofactors. The botrocetin-independent antagonism of these specific chimeras can also be demonstrated according to the procedure initially described by Ware et al. ›Proc. Natl. Acad. Sci. (1991) 88: 2946! by displacing the monoclonal antibody .sup.125 I-LJ-IB1 (10 mg/ml), a competitive inhibitor of the binding of vWF to the platelet GPIb ›Handa M. et al., (1986) J. Biol. Chem. 261: 12579! after 30 min of incubation at 22.degree. C. in the presence of fresh platelets (10.sup.8 platelets/ml).

The purified chimeras are tested for their capacity to permit the in vitro proliferation of the IL3-dependant murine line NFS60, by measuring the incorporation of tritiated thymidine essentially according to the procedure described by Tsuchiya et al. ›Proc. Natl. Acad. Sci. (1986) 83 7633!. For each chimera, the measurements are carried out between 3 and 6 times in a three-point test (three dilutions of the product) in a zone or the relation between the quantity of active product and incorporation of labelled thymidine (Amersham) is linear. In each microtitre plate, the activity of a reference product consisting of recombinant human G-CSF expressed in mammalian cells is also systematically incorporated. The results of FIG. 17 demonstrate that the chimera HSA-G.CSF (pYG1266) secreted by the yeast Kluyveromyces and purified according to Example E.9.3. is capable in vitro of transducing a signal for cellular proliferation for the line NFS60. In this particular case, the specific activity (cpm/molarity) of the chimera is about 7 times lower than that of the reference G-CSF (non-coupled).

The activity of stimulation of the HSA-G-CSF chimeras on granulopoiesis in vivo is tested after subcutaneous injection in rats (Sprague-Dawley/CD, 250-300 g, 8-9 weeks) and compared to that of the reference G-CSF expressed using mammalian cells. Each product, tested at the rate of 7 animals, is injected subcutaneously into the dorso-scapular region at the rate of 100 ml for 7 consecutive days, (D1-D7). 500 ml of blood are collected on days D-6, D2 (before the 2nd injection). D5 (before the 5th injection) and D8, and a blood count is performed. In this test, the specific activity (neutropoiesis units/mole injected) of the chimera HSA-G.CSF (pYG1266) is identical to that of the reference G-CSF (FIG. 18). Since this specific chimera has in vitro a specific activity 7 times lower than that of the reference G-CSF (FIG. 17), it is therefore demonstrated that the genetic coupling of G-CSF onto HSA favourably modifies the pharmacokinetic properties thereof.