AU645077C - Cloning and production of human von willebrand factor GPIb binding domain polypeptides and methods of using same - Google Patents

Description

CLONING AND PRODUCTION OF HUMAN VON WILLEBRAND FACTOR GPIb BINDING DOMAIN POLYPEPTIDES AND METHODS OF USING SAME

Background of the Invention

This application is a continuation-in-part of U.S. Serial No. 487,767, filed March 2, 1990, the contents of which are hereby incorporated by reference into the present disclosure.

Throughout this application various publications are referenced within parentheses. The disclosures of these publications in their entireties are hereby incorporated by reference in this application in order to more fully describe the state of the art to which this invention pertains.

This invention relates to the cloning and production of human von Willebrand Factor analogs and methods of using such analogs.

Structural Features of von Willebrand Factor

Von Willebrand Factor (vWF) is a large plasma protein which is synthesized in the endothelial cells which form the inner surface lining of the blood vessel wall, and by megakarocytes, the precursor of platelets. Large amounts of vWF are found in platelet α-granules, whose contents are released into the blood upon platelet activation. Newly synthesized vWF in endothelial cells may enter the blood via two alternative pathways. Part is secreted constitutively into the blood, mainly as disulfide-linked di ers or small multimers of a 250,000 dalton subunit. Alternatively, part enters secretory organelles called Weibel-Palade bodies. The vWF stored in Weibel-Palade bodies is highly multimeric, ranging in size from that of a dimer to multimers of 50 or ore subunits, and can be released from the cells by treatment with secretatogues, such as thrombin. The highly multimeric vWF is the most effective in promoting platelet adhesion.

The gene encoding vWF has been isolated and shown to be greater than 150 kb in length. It is composed of over 20 exons. The vWF mRNA is approximately 9000 bases in length and encodes a pre-pro-vWF of 2813 amino acids. Residues 1- 22 form a processed leader sequence which presumably is cleaved after entry of the protein into the rough endoplasmic reticulum. The N-terminal portion of the pro- vWF (741 amino acids) is the pro-peptide which is not present in mature vWF. This peptide is present in the blood and has been shown to be identical to a blood protein previously known as von Willebrand Antigen II (vW Agll) . The pro-peptide is essential for the multimerization of vWF. Cells into which a vWF cDNA containing only mature vWF sequences have been introduced produce only di ers. No function is known for the propeptide/vW Agll.

DNA sequence analysis has demonstrated that the pro-vWF precursor is composed of repeated domain subunits. Four different domains have been identified. Mature vWF consists of three A type, three B type, and two C type domains. There are also two complete and one partial D type domain. The pro-peptide consists of two D type domains, leading to the speculation that it may have associated functions.

Mature vWF is a multivalent molecule which has binding sites for several proteins. One of the binding sites recognizes the platelet glycoprotein lb (GPIb) . Using proteolytic digests this site has been localized to the region between amino acid residues 449 and 728 of mature vWF. In addition, vWF has at least two collagen binding sites, at least two heparin binding sites, a Factor VIII binding site, and a RGD site which binds to the platelet GP lib/Ilia receptor.

Involvement Of vWF In Platelet Adhesion To Subendothelium

Evidence that vWF, and specifically, the binding of vWF to the platelet GPIb receptor, is essential for normal platelet adhesion, is based on both clinical observations and in vitro studies. Patients with the bleeding disorder von Willebrand Disease (vWD) have reduced levels of vWF or are completely lacking in vWF. Alternatively, they may have defective vWF. Another disorder, Bernard-Soulier Syndrome (BSS) , is characterized by platelets lacking GPIb receptors.

The in vitro system which most closely approximates the environment of a damaged blood vessel consists of a perfusion chamber in which an everted blood vessel segment (rabbit aorta, human post-mortem renal artery, or the human umbilical artery) is exposed to flowing blood. After stripping off the layer of endothelial cells from the vessel, blood is allowed to flow through the chamber. The extent of platelet adhesion is estimated directly by morpho etry or indirectly using radiolabeled platelets. Blood from patients with VWD or BSS does not support platelet adhesion in this system while normal blood does, indicating the need for vWF and platelet GPIb. Moreover, addition of monoclonal antibodies to GPIb prevents platelet adhesion as well. The vWF-dependence of platelet adhesion is more pronounced under conditions of high shear rates, such as that present in arterial flow. Under conditions of low shear rates, platelet adhesion may be facilitated by other adhesion proteins, such as fibronectin. Possibly, the adhesive forces provided by these other proteins are not adequate to support adhesion at high shear forces, and vWF dependence becomes apparent. Also, the multimeric nature of the vWF may provide for a stronger bond by binding more sites on the platelet.

About 20% of patients from whom clots have been removed by angioplasty or by administration of tissue plasminogen activator (tPA) suffer re-occlusion. This is presumably the result of damage to the endothelium during the treatment which results in the adhesion of platelets to the affected region on the inner surface of the vessel. This is followed by the aggregation of many layers of platelets and fibrin onto the previously adhered platelets, forming a thrombus.

To date none of the anti-platelet aggregation agents described in the literature prevent the initial platelet adhesion to the exposed sub-endothelium thereby preventing subsequent clot formation.

The subject invention provides non-glycosylated, biologically active polypeptides which comprise the vWF (von Willebrand Factor) GPIb binding domain. These polypeptides may be used inter alia to inhibit platelet adhesion and aggregation in the treatment of subjects with conditions such as cerebrovascular disorders and cardiovascular disorders. This invention also provides expression plasmids encoding these polypeptides as well as methods of producing by transforming a bacterial cell and recovering such polypeptides. In addition, the subject invention provides methods of treating and preventing cerebrovascular, cardiovascular and other disorders using these polypeptides to inhibit platelet aggregation. giinnn^-ry of the Invention

This invention provides a non-glycosylated, biologically active polypeptide having the amino acid sequence:

X-A-[Cys ser Arg Leu Leu Asp Leu Val Phe Leu Leu Asp Gly Ser Ser Arg Leu Ser Glu Ala Glu Phe Glu Val Leu Lys Ala Phe Val Val Asp Met Met Glu Arg Leu Arg lie Ser Gin Lys Trp Val Arg Val Ala Val Val Glu Tyr His Asp Gly Ser His Ala Tyr lie Gly Leu Lys Asp Arg Lys Arg Pro Ser Glu Leu Arg Arg lie Ala Ser Gin Val Lys Tyr Ala Gly Ser Gin Val Ala Ser Thr Ser Glu Val Leu Lys Tyr Thr Leu Phe Gin lie Phe Ser Lys lie Asp Arg Pro Glu Ala Ser Arg lie Ala Leu Leu Leu Met Ala Ser Gin Glu Pro Gin Arg Met Ser Arg Asn Phe Val Arg Tyr Val Gin Gly Leu Lys Lys Lys Lys Val lie Val lie Pro Val Gly lie Gly Pro His Ala Asn Leu Lys Gin lie Arg Leu lie Glu Lys Gin Ala Pro Glu Asn Lys Ala Phe Val Leu Ser Ser Val Asp Glu Leu Glu Gin Gin Arg Asp Glu lie Val Ser Tyr Leu Cys]-B-COOH

wherein X is NH₂-methionine- or NH₂-;

A is a sequence of at least 1, but less than 35 amino acids, which sequence is present in naturally occurring vWF, the carboxy terminal amino acid of which is the tyrosine #508 shown in Figure 12;

B is a sequence of at least 1, but less than 211 amino acids, which sequence is present in naturally occurring vWF, the amino terminal amino acid of which is the aspartic acid 696 shown in Figure 12; and

the two cysteines included within the bracketed sequence are joined by a disulfide bond. In addition, the subject invention provides a method of producing any of the above-described polypeptides which comprises transforming a bacterial cell with an expression plasmid encoding the polypeptide, culturing the resulting bacterial cell so that the cell produces the polypeptide encoded by the plasmid, and recovering the polypeptide so produced.

Furthermore, the subject invention provides a pharmaceutical composition comprising an amount of any of the above- described polypeptides effective to inhibit platelet aggregation and a pharmaceutically acceptable carrier. The subject invention also provides a method of inhibiting platelet aggregation which comprises contacting platelets with an amount of any of the above-described polypeptides effective to inhibit platelet aggregation. In addition, the subject invention provides methods of treating, preventing or inhibiting disorders such as cerebrovascular or cardiovascular disorders or thrombosis, comprising administering to the subject an amount of any of the above- described polypeptides effective to treat or prevent such disorders.

The subject invention also provides a method for recovering a purified, biologically active above-described polypeptide which comprises:

(a) producing in a bacterial cell a first polypeptide having the amino acid sequence of the polypeptide but lacking the disulfide bond;

(b) disrupting the bacterial cell so as to produce a lysate containing the first polypeptide; (c) treating the lysate so as to obtain inclusion bodies containing the first polypeptide;

5 (d) contacting the inclusion bodies from step

(c) so as to obtain the first polypeptide in soluble form;

(e) treating the resulting first polypeptide 10 so as to form the biologically active polypeptide;

(f) recovering the biologically active polypeptide so formed; and

15

(g) purifying the biologically active poly¬ peptide so recovered.

Briβf Description of the Figures

Figure 1: Construction of pyWIP

This figure shows the construction of plasmid pvWIP. A series of vWF cDNA clones in λ gtll (isolated from a human endothelial cell cDNA library) were isolated. One cDNA clone covering the entire GPIb binding domain was subcloned into the EcoRI site of pUC19. The resulting plasmid, pvWIP, contains a 2.5 kb cDNA insert.

Figure 2: Construction of pyWF-VAl

This figure shows the construction of plasmid pvWF-VAl. A synthetic oligomer containing an ATG initiation codon located before the amino acid glu-437 (i.e., the 437th amino acid in the vWF protein shown in Figure 12) was ligated to plasmid pvWIP digested with Ndel and Bsu36I. The resulting plasmid was designated pvWF-VAl, and has been deposited in E. coli strain Sφ930 under ATCC Accession No. 68530.

Figure 3: Construction of PVWF-VBI

This figure shows the construction of plasmid pvWF-VBI. A synthetic oligomer containing an ATG initiation codon located before the amino acid phe-443 (see Figure 12) was ligated to plasmid pvWlp digested with Ndel and Bsu36I. The resulting plasmid was designated pvWF-VBI.

Figure 4: Construction of PVWF-VA2

This figure shows the construction of plasmid pvWF-VA2. A synthetic oligomer containing a TAA termination codon located after the amino acid lys-728 (see Figure 12) was ligated to plasmid pvWF-VAl digested with Hindlll and Xmal. The resulting plasmid was designated pvWF-VA2.

Figure 5: Construction of pyWF-VB2

This figure shows the construction of plasmid pvWF-VB2. A synthetic oligomer containing a TAA termination codon was ligated to plasmid pvWF-VBI digested with Hindlll and Xmal. The resulting plasmid was designated pvWF-VB2.

Figure 6: Construction of pyWF-VA3

This figure shows the construction of plasmid pvWF-VA3. An Ndel-EcoRV fragment was isolated from plasmid pvWF-VA2 and ligated to plasmid pMF-945 (constructed as described in Figure 11) digested with Ndel and PvuII. The plasmid obtained was designated pvWF-VA3. The plasmid expresses VA, a vWF GPIb binding domain polypeptide which includes amino acids 437 to 728 (see Figure 12) under the control of the deo P P₂ promoter.

Figure 7: Construction of PVWF-VB3

This figure shows the construction of plasmid pvWF-VB3. An Ndel-EcoRV fragment was isolated from plasmid pvWF-VB2 and ligated to plasmid pMF-945 digested with Ndel and PvuII. The plasmid obtained was designated pvWF-VB3. The plasmid expresses VB, a vWF GPIb binding domain polypeptide which includes amino acids 443 to 728 under the control of the deo P₁P promoter.

Figure 8: Construction of pyWF-VC3

This figure shows the construction of plasmid pvWF-VC3. A synthetic linker was ligated to pvWF-VA3 digested with Ndel and Tthllll. The plasmid obtained was.designated pvWF-VC3, and has been deposited with the ATCC under ATCC Accession No. 68241. The plasmid expresses VC (also referred to as VCL or VC3) , a vWF GPIb binding domain polypeptide which includes amino acids 504 to 728 (see Figure 12) under the control of the deo P_XP₂ promoter.

Figure 9: Construction of PVWF-VD3

This figure shows the construction of plasmid pvWF-VD3. A synthetic linker was ligated to pvWF-VA3 digested with Ndel and Tthllll. The plasmid obtained was designated pvWF-VD3. The plasmid expresses VD, a vWF GPIb binding domain poly¬ peptide which includes amino acids 513 to 728 (see Figure 12) under the control of the deo PχP promoter.

Figure 10: Relative Alignment of Plasmids Expressing vWF- GPIb Binding Domain Polypeptides

This figure shows the relative alignment of the plasmids expressing the vWF-GPIb binding domain polypeptides. Also shown on the top two lines are representations of the vWF cDNA and the location of the GPIb binding domain coding region within the cDNA.

Figure 11: Construction of Plasmid PMF-945

This figure shows the construction of plasmid pMF-945. Plasmid pEFF-920 (in Escherichia coli S 930, ATCC Accession No. 67706) was cleaved with Bglll and Ndel, and the large fragment was isolated. This fragment was ligated to the small 540 bp fragment produced by cleaving plasmid pMF-5534 (ATCC Accession No. 67703) with Bglll and Ndel. This produces plasmid pMF-945 which harbors the PAR sequence and in 5' and 3• order the deo P_χP promoter sequences, the modified deo ribosomal binding site with an enhancer sequence, a pGH analog coding sequence and the T_{2 2} transcription termination sequences. Plasmid pMF-945 is a high level expressor of pGH analog protein.

Figure 12: Translated cDNA Sequence of Mature Human vWF

This figure which consists of Figures 12A, 12B, 12C, 12D, 12E, 12F, 12G and 12H shows the translated cDNA sequence of mature human von Willebrand Factor.

This sequence was compiled using the data disclosed by Verweij, C.L., et al., EMBO Journal 5: 1839-1847 (1986) and Sadler, J.E., et al., Proc. Natl. Acad. Sci. 82.: 6394-6398 (1985) . This nucleotide sequence commences with nucleotide number 2519 (where nucleotide 1 relates to the start of the coding sequence for the signal peptide) and terminates with nucleotide 8668, a total of 6150 nucleotides encoding mature vWF consisting of 2050 amino acids. The translated amino acid sequence commences with amino acid number 1 and terminates with amino acid number 2050. The corresponding nucleotide and amino acid designations are used throughout this application.

Figure 13: Translated Sequence of VC. the vWF GPIb Binding Domain pplypepfrifle Express^ fey

Plasmids PVWF-VC3 and PVWF-VCL

This figure shows the translated sequence of the von Willebrand Factor GPIb binding domain polypeptide expressed by plasmids pvWF-VC3 (ATCC Accession No. 68241) and pvWF-VCL (ATCC Accession No. 68242) .

The first codon ATG encoding the translation initiation codon methionine has been added to the nucleotide sequence corresponding to nucleotides 4028 to 4702 of the sequence of Figure 12. This sequence encodes a polypeptide containing 225 amino acids (plus the initiation methionine) corresponding to amino acid Leu 504 to amino acid Lys 728 of Figure 12, i.e. 226 amino acids in total.

Figure 14: Construction of pyWF-VCL

This figure shows the construction of plasmid pvWF-VCL. Plasmid pvWF-VC3 was digested with Hindlll and Styl and the 860 base pair fragment isolated. This fragment was ligated with the large fragment isolated from the Hindlll-Styl digest of plasmid pMLK-100. The resulting plasmid was designated pvWF-VCL and deposited in E.coli 4300(F") with the ATCC under ATCC Accession No. 68242. This plasmid expresses VCL, the same vWF GPIb binding domain polypeptide as pvWF-VC3 (methionine plus amino acids 504-728) , however under control of the λP_L promoter and the deo ribosomal binding site.

Figure 15: Construction of Plasmid PVWF-VE2

Plasmid pvWF-VA2 was digested with Ndel and PstI and the large fragment isolated. Synthetic oligomers No. 2 and No. 3 (shown in Figure 16) were treated with T4 polynucleotide kinase. The large pvWF-VA2 fragment was then ligated with synthetic oligomers No. 1 and No. 4, (shown in Figure 16) and with kinased oligomers No. 2 and No. 3. The resulting plasmid was designated pvWF-VE2.

Figure 16: Synthetic Oligomers Used in Construction of PVWF-VE2.

This figure shows the four synthetic linkers (Nos. 1-4) used in construction of pvWF-VE2. Ficnirβ 17: Construction of Plasmid PVWF-VE3

Plasmid pvWF-VE2 was digested with Ndel and Hindlll and the small 770 bp fragment isolated and ligated with the large fragment isolated from the Ndel-Hindlll digest of plasmid pMLK-7891. The resulting plasmid was designated pvWF-VE3.

Figure 18: Construction of Plasmid PVWF-VEL

Plasmid pvWF-VE3 was digested with XmnI, treated with bacterial alkaline phosphatase (BAP) , and further digested with Ndel and Hindlll. Plasmid pMLK-100 was digested with Ndel and Hindlll and treated with BAP. The two digests were mixed and ligated, producing plasmid pvWF-VEL which expresses the DNA sequence corresponding to amino acids 469-728 of mature vWF under the control of the λP_L promoter and the ell ribosomal binding site.

I_ _Q_________: The Effect of VCL on BJV-Induced Aggregation in Human Platelet Rich Plasma (PRP.

This figure provides the results of a standardized von Willebrand Factor (vWF)-dependent aggregation assay using human PRP.

Figure 20: The Effegt ς>f V L QT. pJV-Indμpefl Aggregation in R3t ppp This figure provides the results of a standardized von Willebrand Factor (vWF)-dependent aggregation assay using rat PRP. Detailed Description of the Invention

The plasmids pvWF-VC3, pvWF-VCL and pvWF-VAl were deposited in Escherichia coli pursuant to, and in satisfaction of, the requirements of the Budapest Treaty on the International Recognition of the Deposit of Microorganisms for the Purposes of Patent Procedure with the American Type Culture Collection (ATCC) , 12301 Parklawn Drive, Rockville, Maryland 20852 under ATCC Accession Nos. 68241, 68242 and 68530, respectively.

This invention provides a non-glycosylated, biologically active polypeptide having the amino acid sequence:

X-A-[Cys Ser Arg Leu Leu Asp Leu Val Phe Leu Leu Asp Gly Ser Ser Arg Leu Ser Glu Ala Glu Phe Glu Val Leu Lys Ala Phe Val Val Asp Met Met Glu Arg Leu Arg lie Ser Gin Lys Trp Val Arg Val Ala Val Val Glu Tyr His Asp Gly Ser His Ala Tyr lie Gly Leu Lys Asp Arg Lys Arg Pro Ser Glu Leu Arg Arg lie Ala Ser Gin Val Lys Tyr Ala Gly Ser Gin Val Ala Ser Thr Ser Glu Val Leu Lys Tyr Thr Leu Phe Gin lie Phe Ser Lys lie Asp Arg Pro Glu Ala Ser Arg lie Ala Leu Leu Leu Met Ala Ser Gin Glu Pro Gin Arg Met Ser Arg Asn Phe Val Arg Tyr Val Gin Gly Leu Lys Lys Lys Lys Val lie Val lie Pro Val Gly lie Gly Pro His Ala Asn Leu Lys Gin lie Arg Leu lie Glu Lys Gin Ala Pro Glu Asn Lys Ala Phe Val Leu Ser Ser Val Asp Glu Leu Glu Gin Gin Arg Asp Glu lie Val Ser Tyr Leu Cys]-B-COOH

wherein X is NH₂-methionine- or NH -;

A is a sequence of at least 1, but less than 35 amino acids, which sequence is present in naturally occurring human vWF, the carboxy terminal amino acid of which is the tyrosine #508 shown in Figure 12; B is a sequence of at least 1, but less than 211 amino acids, which sequence is present in naturally occurring human vWF, the amino terminal amino acid of which is the aspartic acid #696 shown in Figure 12; and

the two cysteines included within the bracketed sequence are joined by a disulfide bond. The bracketed sequence comprises amino acids #509-#695 of Figure 12.

In one embodiment, this polypeptide has the amino acid sequence:

X-[Leu His Asp Phe Tyr Cys Ser Arg Leu Leu Asp Leu Val Phe Leu Leu Asp Gly Ser Ser Arg Leu Ser Glu Ala Glu Phe Glu Val Leu Lys Ala Phe Val Val Asp Met Met Glu Arg Leu Arg lie Ser Gin Lys Trp Val Arg Val Ala Val Val Glu Tyr His Asp Gly Ser His Ala Tyr lie Gly Leu Lys Asp Arg Lys Arg Pro Ser Glu Leu Arg Arg lie Ala Ser Gin Val Lys Tyr Ala Gly Ser Gin Val Ala Ser Thr Ser Glu Val Leu Lys Tyr Thr Leu Phe Gin lie Phe Ser Lys lie Asp Arg Pro Glu Ala Ser Arg lie Ala Leu Leu Leu Met Ala Ser Gin Glu Pro Gin Arg Met Ser Arg Asn Phe Val Arg Tyr Val Gin Gly Leu Lys Lys Lys Lys Val He Val He Pro Val Gly He Gly Pro His Ala Asn Leu Lys Gin He Arg Leu He Glu Lys Gin Ala Pro Glu Asn Lys Ala Phe Val Leu Ser Ser Val Asp Glu Leu Glu Gin Gin Arg Asp Glu He Val Ser Tyr Leu Cys Asp Leu Ala Pro Glu Ala Pro Pro Pro Thr Leu Pro Pro Asp Met Ala Gin Val Thr Val Gly Pro Gly Leu Leu Gly Val Ser Thr Leu Gly Pro Lys]-COOH

wherein X is NH₂- or NH₂-methionine-, preferably

NH₂-methionine-.

The bracketed sequence comprises amino acids #504-#728 of Figure 12. One skilled in the art to which the subject invention pertains can readily make such polypeptides using recombinant or non-recombinant DNA techniques.

The polypeptides may be constructed using recombinant DNA technology. One means for obtaining the polypeptides is to express nucleic acid encoding the polypeptides in a suitable host, such as a bacterial, yeast, or mammalian cell, using methods well known in the art, and recovering the polypeptide after it has been expressed in such a host.

Examples of vectors that may be used to express the nucleic acid encoding the polypeptides are viruses such as bacteriophages (such as phage lambda) , cosmids, plasmids, and other recombination vectors. Nucleic acid molecules are inserted into vector genomes by methods well known in the art. For example, using conventional restriction enzyme sites, insert and vector DNA can both be exposed to a restriction enzyme to create complementary ends on both molecules which base pair with each other and are then ligated together with a ligase. Alternatively, linkers can be ligated to the insert DNA which correspond to a restriction site in the vector DNA, which is then digested with the restriction enzyme which cuts at that site. Other means are also available.

Vectors comprising nucleic acid encoding the polypeptides may be adapted for expression in a bacterial cell, a yeast cell, or a mammalian cell which additionally comprise the regulatory elements necessary for expression of the nucleic acid in the bacterial, yeast, or mammalian cells so located relative to the nucleic acid encoding the polypeptide as to permit expression thereof. Regulatory elements required for expression include promoter sequences to bind RNA polymerase and transcription initiation sequences for ribosome binding. For example, a bacterial expression vector may include a promoter such as the λ P_L or deo promoters and for transcription initiation the C_{ιτ or} deo ribosomal binding sites. Such vectors may be obtained commercially or assembled from the sequences described by methods well known in the art, for example the methods described above for constructing vectors in general.

In addition, non-recombinant techniques such as chemical synthesis, synthetic DNA or cDNA may be used to obtain the above-described polypeptides. One means of isolating the polypeptide is to probe a human genomic library with a natural or artificially designed DNA probe, using methods well known in the art. DNA and cDNA molecules which encode the polypeptides may be used to obtain complementary genomic DNA, cDNA or RNA from human, mammalian or other animal sources, or to isolate related cDNA or genomic clones by the screening of cDNA or genomic libraries.

The subject invention further provides a pharmaceutical composition comprising an amount of any of the above- described polypeptides effective to inhibit platelet aggregation and a pharmaceutically acceptable carrier.

As used herein, the term "pharmaceutically acceptable carrier" encompasses any of the standard pharmaceutical carriers. Such carriers are well known in the art and may include, but are in no way and are not intended to be limited to, any of the standard pharmaceutical carriers such as phosphate buffered saline solutions, water, emulsions such as oil/water emulsion, and various types of wetting agents. Other carriers may also include sterile solutions, tablets, coated tablets, and capsules.

Typically such carriers contain excipients such as starch. milk, sugar, certain types of clay, gelatin, stearic acid or salts thereof, magnesium or calcium stearate, talc, vegetable fats or oils, gums, glycols, or other known excipients. Such carriers may also include flavor and color additives or other ingredients. Compositions comprising such carriers are formulated by well known conventional methods.

The composition has an amount sufficient to result in a blood concentration of 0.06 to 58 μM, preferably between about 0.06 and 29 μM, for example 0.23 to 23 μM. Expressed in different terms, the amount should be 0.1 to 100 mg/Kg body weight, preferably 0.1 to 50 mg/Kg body weight, for example 0.4 to 40 mg/kG body weight.

The administration of the composition may be effected by any of the well known methods, including but not limited to intravenous, intramuscular, subcutaneous and oral administration.

This invention also provides a method of inhibiting platelet aggregation which comprises contacting platelets with an amount of any of the above-described polypeptides effective to inhibit platelet aggregation so as to inhibit platelet aggregation.

This invention also provides expression plasmids encoding the above-described polypeptides. In one embodiment, the expression plasmid encoding the polypeptide with the brack- eted sequence, i.e. amino acids #504-#728 of Figure 12, is designated pvWF-VC3 and is deposited under ATCC Accession No. 68241. In another embodiment, the expression plasmid encoding a polypeptide with the bracketed sequence, i.e. amino acids #504-#728 of Figure 12, is designated pvWF-VCL and is deposited under ATCC Accession No. 68242. The expression plasmids of this invention further comprise suitable regulatory elements positioned within the plasmid relative to the DNA encoding the polypeptide so as to effect expression of the polypeptide in a suitable host cell, such as promoter and operators, e.g. deo P_j^ and λ P_L0_L, ribosomal binding sites, e.g. deo and C_XI, and repressors. Other suitable regulatory elements include, for example, the lac, trp, tac, and lpp promoters (European Patent Application Publication No. 0303972, published February 22, 1989).

The suitable regulatory elements are positioned within the plasmid relative to the DNA encoding the polypeptide so as to effect expression of the polypeptide in a suitable host cell. In preferred embodiments of the invention, the regulatory elements are positioned close to and upstream of the DNA encoding the polypeptide.

The expression plasmids of this invention may be introduced into suitable host cells, preferably bacterial host cells. Preferred bacterial host cells are Escherichia coli cells. Examples of suitable Escherichia coli cells are strains S 930 or 4300, but other Escherichia coli strains and other bacteria can also be used as host cells for the plasmids. Such bacteria include Pseudomonas aeruσinosa and Bacillus subtilis.

The bacteria used as hosts may be any strain including auxotrophic (such as A1645) , prototrophic (such as A4255) , and lytic strains; F⁺ and F^" strains; strains harboring the cl⁸⁵⁷ represser sequence of the λ prophage (such as A1645 and A4255) ; and strains deleted for the deo repressors and the deo gene (see European Patent Application Publication No. 0303972, published February 22, 1989). Escherichia coli strain A4255 (F⁺) has been deposited under ATCC Accession No. 53468, and Escherichia coli strain A1645 has been deposited under ATCC Accession No. 67829.

The invention provides a bacterial cell which comprises these expression plasmids. In one embodiment, the bacterial cell is an Escherichia coli cell. In preferred embodiments, the invention provides an Escherichia coli cell containing the plasmid designated pvWF-VAl, deposited in E. coli strain Sφ930 with the ATCC under ATCC Accession No. 68530; pvWF- VA3; pvWF-VB3; pvWF-VC3, deposited in E. coli strain Sø930 with the ATCC under ATCC Accession No. 68241; pvWF-VD3; or pvWF-VCL, deposited in E. coli strain 4300(F") with the ATCC under ATCC Accession No. 68242.

All the E. coli host strains described above can be "cured" of the plasmids they harbor by methods well-known in the art, e.g. the ethidium bromide method described by R.P. Novick in Bacteriol. Review 33. 210 (1969) .

In addition, the subject invention provides a method of producing any of the above-described polypeptides which comprises transforming a bacterial cell with an expression plasmid encoding the polypeptide, culturing the resulting bacterial cell so that the cell produces the polypeptide encoded by the plasmid, and recovering the polypeptide so produced.

Furthermore, the invention provides a method of treating a subject with a cerebrovascular disorder which comprises administering to the subject an amount of any of the polypeptides of the invention effective to inhibit platelet aggregation.

Also provided is a method of treating a subject with a cardiovascular disorder which comprises administering to the subject an amount of a polypeptide effective to inhibit platelet aggregation. Examples of cardiovascular disorders susceptible to treatment include acute myocardial infarction or angina.

Further, the subject invention provides method of inhibiting platelet aggregation in a subject prior to, during, or after the subject has undergone angioplasty, thrombolytic treatment, or coronary bypass surgery which comprises administering to the subject an amount of a polypeptide of the invention effective to inhibit platelet aggregation.

The invention also provides a method of maintaining blood vessel patency in a subject prior to, during, or after the subject has undergone coronary bypass surgery, which comprises administering to the subject an amount of a poly¬ peptide of the invention effective to inhibit platelet aggregation.

The invention also provides a method of treating a subject having cancer which comprises administering to the subject an amount of a polypeptide of the invention effective to retard tumor metastasis.

The invention also provides a method of inhibiting thrombosis in a subject which comprises administering to the subject an amount of a polypeptide of the invention effective to inhibit the thrombosis. The thrombosis may be associated with an inflammatory response.

In addition, the subject invention provides a polypeptide of the invention bound to a solid matrix.

The invention also provides a method of treating a subject suffering from platelet adhesion to damaged vascular surfaces which comprises administering to the subject an amount of the polypeptide of the invention effective to inhibit platelet adhesion to damaged vascular surfaces.

The invention also provides a method of preventing platelet adhesion to a prosthetic material or device in a subject which comprises administering to the subject an amount of the polypeptide of the invention effective to prevent platelet adhesion to the material or device.

The invention also provides a method of inhibiting re- occlusion in a subject following angioplasty or thrombolysis which comprises administering to the subject an amount of the polypeptide of the invention effective to inhibit re- occlusion.

The invention also provides a method of preventing vaso- occlusive crises in a subject suffering from sickle cell anemia which comprises administering to the subject an amount of the polypeptide of the invention effective to prevent vaso-occlusive crises.

The invention also provides a method of preventing arteriosclerosis in a subject which comprises administering to the subject an amount of the polypeptide of the invention effective to prevent arteriosclerosis.

The invention also provides a method of thrombolytic treatment of thrombi-containing, platelet-rich aggregates in a subject which comprises administering to the subject an amount of the polypeptide of the invention effective to treat thrombi-containing, platelet-rich aggregates.

The invention also provides a method of preventing platelet activation and thrombus formation due to high shear forces in a subject suffering from stenosed or partially obstructed arteries which comprises administering to the subject an amount of the polypeptide of the invention effective to prevent platelet activation and thrombus formation.

The invention also provides a method of preventing thrombin- induced platelet activation in a subject which comprises administering to the subject an amount of the polypeptide of the invention effective to prevent thrombin-induced platelet activation.

The invention also provides a method of preventing stenosis as a result of smooth muscle proliferation following vascular injury in a subject which comprises administering to the subject an amount of the polypeptide of the invention effective to prevent stenosis.

The invention also provides a method for recovering a purified, biologically active polypeptide of the invention which comprises:

(a) producing in a bacterial cell a first polypeptide having the amino acid sequence of the polypeptide but lacking the disulfide bond;

(b) disrupting the bacterial cell so as to produce a lysate containing the first polypeptide;

(c) treating the lysate so as to obtain inclusion bodies containing the first polypeptide;

(d) contacting the inclusion bodies from step (c) so as to obtain the first polypeptide in soluble form;

(e) treating the resulting first polypeptide so as to form the biologically active polypeptide;

(f) recovering the biologically active polypeptide so formed; and

(g) purifying the biologically active poly¬ peptide so recovered.

Step (e) may comprise contacting the polypeptide with a thiol-containing compound and disulfide so as to refold and reoxidize the polypeptide. Preferably, the thiol-containing compound is glutathione, thioredoxin, β-mercaptoethanol or cysteine.

The contacting of step (d) may be effected in the presence of a denaturant such as guanidine hydrochloride or urea.

The recovery of the polypeptide in step (f) may comprise removing the denaturant by dialysis.

In step (g) , the biologically active polypeptide may be purified by cation exchange chromatography.

The first polypeptide may also be purified by cation exchange chromatography after step (d) . The examples which follow are set forth to aid in understanding the invention but are not intended to, and should not be so construed as to, limit its scope in any way. The examples do not include detailed descriptions for conventional methods employed in the construction of vectors, the insertion of genes encoding polypeptides of interest into such vectors or the introduction of the resulting plasmids into bacterial hosts. Such methods are well-known to those skilled in the art and are described in numerous publications including Sambrook, Fritsch and

Maniatis, Molecular Cloning; A Laboratory Manual. 2nd

Edition, Cold Spring Harbor Laboratory Press, USA, (1989) .

EXAMPLES

All the references to map positions correspond to the identically numbered positions along the translated nucleotide sequence of mature human von Willebrand Factor shown in Figure 12.

Ex mple j: CJQning anfl Expression pf vWf <?Plfr Binding Domain Polypeptides

<?P A Cloning of Human vwy gpifr pjnςtjng ppφajn

A human endothelial cDNA library (obtained from CLONTECH Laboratories, Inc.) in λ gtll was screened for human vWF positive sequences using two synthetic DNA probes. The probes were synthesized according to the published DNA sequence (Sadler et al., Proc. Nat. Acad. Sci. £2.: 6394-8 (1985) and Verweij et al., EMBO J. 3_: 1829-47 (1986)) of human vWF (flanking 5* end and 3¹ end of the vWF domain known to bind the GPIb receptor) (see Figure 12) .

The synthetic probes have the following sequences: Seouence Fugl p __. s

AAATCTGGCAGTGCTCAGGGTCACTGGGATTCAAGGTGAC 3863-3902 CCAGGACGAACGCCACATCCAGAACCATGGAGTTCCTCTT 4700-4739

A series of vWF cDNA clones covering the entire GPIb binding domain were identified and isolated. The cDNA fragments were subcloned into EcoRI site of pUC-19 (New England Biolabs, Inc.). One of the subclones, designated pvWIP (Figure 1), contains a 2.5 Kb insert. This 2.5 Kb insert covers the entire GPIb binding domain extending from 550 bp upstream of the GPIb binding site to 1100 bp downstream of the GPIb binding site. (The subclone pvWIP has also been designated pvWF-lP) .

Manipulation of DNA Coding for the vWF GPIb Binding Domain

In order to obtain expression of the GPIb binding domain in Escherichia coli under the regulation of the deo P_jP₂ promoter, the cDNA fragment of vWF, derived from plasmid pvWIP was used for further manipulations as described below. As indicated previously, the vWF tryptic digest fragment that binds the GPIb receptor is from amino acid Val 449 to amino acid Lys 728.

A. Subcloning of the 5' end of vWF GPIb binding domain and addition of a translation initiation codon ATG.

Plasmid pvWIP has two convenient restriction sites at the 5' end. Bsu36I which cuts at the DNA sequence corresponding to amino acid Ser (445) , and Tthllll which cuts at amino acid Asp (514) . Synthetic fragments of various size were designed that insert an ATG translation initiation codon at the 5* end as well as additional amino acids. This was done first, in order to maximize the chances of obtaining high levels of expression. Second, they are a first step towards reducing the size of the vWF GPIb binding domain peptide down to the minimal size needed, possibly eliminating collagen and heparin binding sites which may ultimately interfere with the function of the product.

Al. Amino acid Glu 437 at 5' end

Synthetic oligomers with the sequences:

5^» - TATGGAGGTGGCTGGCCGGCGTTTTGCC - 3'

3^» - ACCTCCACCGACCGGCCGCAAAACGGAGT - 5'

Ndel Bsu36I

were ligated to plasmid pvWF-lP digested with Ndel and Bsu36I (see Figure 2) . The plasmid obtained was designated pvWF-VAl. Plasmid pvWF-VAl has been maintained in E. coli strain Sφ930 and was deposited under ATCC Accession No. 68530.

A2. Amino acid Phe 443 at 5' end

Synthetic oligomers with the sequences:

5' - TATGTTTGCC - 3'

3' - ACAAACGGAGT - 5^»

were ligated to plasmid pvWIP digested with Ndel and Bsu36I (see Figure 3) . The plasmid obtained was designated pvWF-VBI. B. Subcloning of the 3' end of vWF GPIb binding domain, introduction of translation stop codon.

Bl. Introduction of stop codon in plasmid pvWF-VAl

A synthetic oligomer with the sequence:

5•-CCGGGGCTCTTGGGGGTTTCGACCCTGGGGCCCAAGTAAGATATCA-3• 3'-CCGAGAACCCCCAAAGCTGGGACCCCGGGTTCATTCTATAGTTCGA-5•

was ligated to an X al and Hindlll digested plasmid pvWF-VAl (see Figure 4) . The plasmid obtained was designated pvWF-VA2. This newly constructed plasmid contains a translation termination codon

TAA adjacent to amino acid 728 (Lys) and EcoRV site.

B2. Introduction of translation stop codon in plasmid PVWF-VBI

A synthetic oligomer with the sequence:

5'- CCGGGGC_TCTTGGGGTTTCGACCCTGGGGCCCAAGTAAGATATCA - 3' 3' - CCGAGAACCCCAAAGCTGGGACCCCGGGTTCATTCTATAGTTCGA - 5'

was ligated to plasmid pvWF-VBI digested with Xmal and Hindlll. The plasmid obtained was designated pvWF-VB2 (see Figure 5) .

Expression of the vWF GPIb binding domain in Escherichia coli

In order to obtain expression of the vWF GPIb binding domain various expression plasmids were constructed based on a deo P_jP constitutive promoter system.

1. Expression of a vWF GPIb binding domain polypeptide including amino acid Glu 437 to amino acid Lvs 728

(based on plasmid PVWF-VA2)

An Ndel-EcoRV fragment was isolated from plasmid pvWF-VA2 and ligated into plasmid pMF-945 (see Figure 11) digested with Ndel and PvuII (see Figure 6) . The plasmid obtained was designated as pvWF-VA3 and was maintained in Escherichia coli strain S093O.

2. Expression of a vWF GPIb binding domain polypeptide including amino acid Phe 443 to amino acid Lvs 728 (based on plasmid pvWF-VB2.

An Ndel-EcoRV fragment was isolated from plasmid pvWF-VB2 and ligated into plasmid pMF-945 digested with Ndel and PvuII (see Figure 7) . The plasmid obtained was designated as pvWF-VB3 and was maintained in Escherichia coli strain Sφ930.

3. Expression of a vWF GPIb binding domain polypeptide including amino acid Leu 504 to amino acid Lvs 728 (based on expression plasmid PVWF-VA3)

A synthetic oligomer with the sequence:

5' - TATGTTGCACGATTTCTACTGCAGCAGGCTACTGGACC - 3¹ 3¹ - ACAACGTGCTAAAGATGACGTCGTCCGATGACCTGGA - 5' Ndel Tthllll was ligated to plasmid pvWF-VA3 digested with Ndel and Tthllll. The plasmid obtained was designated as pvWF-VC3 (see Figure 8) . Plasmid pvWF-VC3 is maintained in Esche¬ richia coli strain S 30 and has been deposited with the ATCC under Accession No. 68241 (also see Figure 13) .

4. Expression of a vWF GPIb binding domain polypeptide including amino acid Leu 513 to amino acid Lvs 728 (based on expression plasmid PVWF-VA3)

A synthetic oligomer with the sequence:

5' - TATGCTGGACC - 3' 3'- ACGACCTGGA - 5'

Ndel Tthllll

was ligated to plasmid pvWF-VA3 digested with Ndel and Tthllll. The plasmid obtained was designated pvWF-VD3 (see Figure 9) . Plasmid pvWF-VD3 is maintained in Escherichia coli strain Sφ930.

Expression of vWF-GPIb binding domain polypeptides

The relative alignment of the expression plasmids is shown in Figure 10. Plasmids pvWF-VA3, pvWF-VB3, pvWF-VC3 and pvWF-VD3 in Escherichia coli strain Sφ930 were used in order to analyze the levels of expression of the various vWF-GPIb binding domain peptides. The clones obtained were grown in LB medium containing Amp (100 μg/ml) at 37°C for 48 hours.

After 48 hours growth bacterial cells were harvested and centrifuged for 2 minutes at 10,000 RPM. Pellets were dissolved in 1/10 volume of 50 mM Tris-HCl pH=8.0. Sample buffer (containing SDS and 6-mercaptoethanol) was added. Samples were boiled for 10 minutes and loaded on a 10% SDS polyacrylamide gel. The expression of the vWF GPIb binding domain polypeptides in clones pvWF-VA3, pvWF-VB3 and pvWF- VD3 was low relative to the bacterial total proteins. The vWF polypeptides from these clones were detectable by Western blot analysis using commercially available polyclonal vWF antibody (Dekopatts a/s, Glostrup, Denmark) . However, clones originated from Escherichia coli strain Sφ930 transformed with plasmid pvWF-VC3 expressed the vWF GPIb binding domain polypeptide (amino acid Leu 504 to amino acid Lys 728 plus methionine) at high levels (as a major band) detectable upon Coomassie staining.

Escherichia coli strain S093O harboring plasmid pvWF-VC3 was deposited with the ATCC under Accession No. 68241. Subsequently, an inducible plasmid was constructed which contains the same vWF coding region as pvWF-VC3, expressed under the control of the λ P_L promoter and the deo ribosomal binding site (see Figure 14) . This new plasmid, designated pvWF-VCL, proved to be a high expressor of VCL, the vWF GPIb binding domain polypeptide (methionine plus amino acid Leu 504 to amino acid Lys 728) . This plasmid was deposited in Escherichia coli strain 4300 with the ATCC under Accession No. 68242. Escherichia coli strain 4300, constructed from Escherichia coli strain ATCC Accession No. 12435, is a wild- type, F^", biotin dependent strain, harboring the λ cI857 temperature-sensitive represser. (A third plasmid construct harboring the same vWF coding region under the control of the λ promoter and the ell ribosomal binding site did not express any vWF peptide detectable by Coomassie staining.)

The Ndel-Hindlll insert of pvWF-VCL can be conveniently subcloned into other expression vectors such as commercially available pUC19 for production of a series of polypeptides which include the same amino acid sequence from amino acid 509 (cys) to amino acid 695 (cys) and have the same biological activity.

Examolβ 2: Fermentation of Bacteria Expressing vWF GPIb Binding Domain Polypeptides

During scale-up fermentations of clone pvWF-VC3 it was found that the host tends to lose the plasmid due to instability. The loss of plasmids caused a reduction in vWF GPIb binding domain polypeptide expression. It was found necessary to maintain continuous selective pressure (i.e., continuous addition of A picillin) in order to maintain plasmid copy number and to maintain the expression levels. Large scale fermentation was carried out for 12 hours.

Fermentation was carried out in the following growth medium:

Fructose (50%) was added to the growth medium at final concentration of 150 ml/liter and Ampicillin (100 mg/ml) was pumped continuously into the fermentor (total of 8 ml/liter) . Fermentation was carried out for 12 hours at 37^βC.

Purification of polypeptides

Cells were harvested after 12 hours fermentation and centri- fuged. The bacterial pellet obtained was resuspended in buffer [50 mM Tris pH=8.0, 50 mM NaCl, 1 mM EDTA, ImM DTT (dithiothreitol) , 1 mM PMSF (Phenylmethylsulfonyl fluoride) and 10% Glycerol] . After additional centrifugation and sonication the vWF GPIb binding domain polypeptide was found in the pellet.

The vWF GPIb binding domain polypeptide was further purified by solubilization of the pellet in 8M Urea containing 10 mM DTT, 25 mM Tris pH=8 and 1 mM EDTA at room temperature. The solubilized pellet was fractionated on a DEAE cellulose ion exchange column chromatography. (Elution buffer as above except 0.5 mM DTT).

The vWF GPIb binding domain polypeptide was eluted at 150 mM NaCl. After dilution to 50 mM NaCl (in the above buffer) the partially purified peptide waε loaded on a -Sepharose column. Elution from the Q-Sepharose column was carried out at various NaCl concentrations (step elution) . The vWF GPIb binding domain peptide was pooled in four peaks which eluted at 100 mM, 200 mM, 250 mM and 500 mM NaCl. All four peaks were dialyzed against 150 mM NaCl and 50 mM Tris pH-8 for 36 hours. During the dialysis the Urea concentration of the dialysis solution was reduced in a linear gradient from 6M Urea to no Urea.

Example 3: Biological Activity of vWF GPIb Binding Domain Polypeptides

Platelet Aggregation Assays

vWF preparation:

Human plasma-derived vWF was purified from human outdated blood bank plasma according to J. Loscalzo and R.I. Handin, Biochemistry £2.: 3880-3886 (1984) . The purified plasma- derived vWF was concentrated by Amicon 100,000 cut-off filter membrane, to a final concentration of 0.25 mg/ml.

Asialo-vWF preparation:

The purified plasma-derived human vWF was desialyated according to L. DeMarco and S. Shapiro, J. Clin. Invst. ££.: 321-328 (1981) with the following modifications:

1. The Neuraminidase used was from Vibrio cholera type

II (Sigma) .

2. The reaction mixture contained 0.2 Units enzyme/ mg protein and protease inhibitors according to the following concentrations: Benzamidine (20 mM) ,

Leupeptin (15 μg/ml) and Aprotinin 20 (U/ ml) . The asialo-vWF was used for platelet aggregation with¬ out any further purification.

Platelet aggregation - Induced bv Asialo-vWF

As stated above, soluble vWF does not bind to platelets via the GPIb receptor. Asialo-vWF, obtained by neuraminidase treatment to remove sialic acid residues, readily binds to platelets via GPIb. Presumably, the desialation lowers the net negative charge on the vWF, allowing it to bind to the negatively charged GPIb receptor. Asialo-vWF binding to platelets causes activation, release of ADP, and GP lib/ Ilia mediated aggregation. Platelet aggregation induced by asialo-vWF was carried out with 200 μl of PRP (Platelet-rich plasma) (Fujimura Y. , et al., J. Biol. Che . 261: 381-385 (1986) ) and 39 μg/ml of asialo-vWF (final concentration) in a Lu i aggregometer. The results of inhibition of platelet aggregation with VC, the vWF GPIb binding domain polypeptide, are summarized in Table I.

VC (also referred to as VCL or VC3) is a vWF GPIb binding domain polypeptide which includes methionine plus amino acids 504-728 (see Figure 12) .

vWF-Ristocetin induced platelet aggregation

Ristocetin-induced platelet aggregation in the presence of purified human intact vWF was carried out with washed human platelets according to Fujimura Y. et al., J. Biol. Chem. 261: 381-385 (1986).

The results of inhibition of platelet aggregation induced by ristocetin in the presence of intact vWF are summarized in Table II. Additional results using these assays are described in Example 5.

Tft?K I

Inhibition of Asialo-vWF Induced Platelet Aggregation (In PRP) by VC, a vWF GPIb Binding Domain Polypeptide

TABLB II

Inhibition of Ristocetin Induced Platelet Aggregation by VC, a vWF GPIb Binding Domain Polypeptide

Exa ple 4 :

An Improved Method of Obtaining Pure. Oxidized. Folded and Biologically Active vWF GPIb Binding Domain Polypeptide

In Example 2, fermentation of cells harboring plasmid pvWF- VC3 was described. Subsequently, a preferred plasmid, pvWF- VCL was constructed as described in Example 1 and maintained in E.coli strain A4300. This host/plasmid system was fermented essentially as known in the art for vectors containing a gene expressed under control of the λP_L promoter, see, for example coassigned EPO Patent Publication No. 173,280, published March 5, 1986, Example 5, pages 73-74 (without added biotin, thiamine, trace elements, and ampicillin) . In this improved method of purification of vWF GPIb binding domain polypeptide, a cell pellet of the above fermentation of A4300/pvWF-VCL was used.

In this improved method a purer and more active polypeptide is produced than by the method disclosed in Example 2. The general scheme of the downstream process consists of steps A through H as follows:

A. Cell disruption and suspension of pellet: A pellet containing the vWF GPIb binding domain polypeptide is obtained as described in Example 2, by sonication and centrifugation of a cell suspension in 50mM Tris-HCl pH=8, 50mM NaCl, ImM EDTA, ImM DTT, ImM PMSF, and 10% Glycerol.

The pellet containing the inclusion bodies is dissolved at about 10% w/v in a solution such that the final concentrations after dissolution are 8M urea, 20mM DTT, 20mM HEPES pH 8, and lOOmM NaCl. The resulting solution may be further purified by ion exchange chromatography as described below. Alternatively, the inclusion bodies may be solubilized in a buffer containing 6M guanidine hydrochloride followed by buffer exchange to urea. The inclusion bodies may also be dissolved at different concentrations of urea, guanidine hydrochloride or any other denaturant or in the absence of denaturants, for example, at extremes of pH.

B. Cation exchange chromatographv: This step eliminates most of the contaminants and produces the vWF GPIb binding domain polypeptide at greater than 90% purity. Any cation exchange (e.g. carboxymethyl) method may be used in this step, but CM-Sepharose fast flow (Pharmacia) chromatography is preferred. The functional group may be carboxymethyl, a phospho group or sulphonic groups such as sulphopropyl. The matrix may be based on inorganic compounds, synthetic resins, polysaccharides, or organic polymers; possible matrices are agarose, cellulose, trisacryl, dextran, glass beads, oxirane acrylic beads, acrylamide, agarose/polyacrylamide copolymer (Ultrogel) or hydrophilic vinyl polymer (Fractogel) . In a specific embodiment, the polypeptide is loaded onto a CM-Sepharose FF column equilibrated with 8M urea, ImM DTT, 20mM HEPES pH 8, lOOmM NaCl. Pure polypeptide elutes in 8M urea, ImM DTT, 20mM HEPES pH 8 and 200mM NaCl. Up to about 30 OD₂₈₀ units of solubilized inclusion bodies may be loaded per ml CM-Sepharose FF. At this ratio the eluted polypeptide typically has a concentration of 4-5 OD₂₈₀/ml.

C. Oxidation/Refolding: The polypeptide solution eluted from the cation exchange step above is treated with 6M guanidine hydrochloride (GuCl) to disrupt any aggregates. The polypeptide is then diluted to 0.05 OD₂₈₀/ml in 2M GuCl, pH 5-11, preferably 20mM HEPES pH 8, O.lmM GSSG (glutathione, oxidized form) . This mixture is allowed to stand overnight at room temperature. The products are analyzed by gel filtration on fast protein liquid chromatography (FPLC) such as Superose 12 before proceeding. Analysis shows that this protein concentration reproducibly yields about 30% correctly oxidized monomers, and 70% S-linked dimers and multimers, as well as reduced and incorrectly oxidized monomers. A higher protein concentration gives a higher absolute yield of correctly oxidized monomers but a lower percentage yield due to increased formation of S-linked dimers and multimers. For example, a protein concentration of 0.1 OD₂₈₀/ml yields only 20% correctly oxidized monomers. Reducing the concentration to 0.025 OD₂₈₀/ml yields 35-40% correctly oxidized monomers but a lower absolute yield per liter oxidation. Oxidations may also be performed in urea instead of in GuCl, or in any other denaturant or in the absence of denaturants under appropriate buffer conditions in which, for example, pH, ionic strength, and hydrophobicity are varied. The preferred concentration of urea is in the range 0.5M to 10M, preferably 4M, and the preferred oxidant is GSSG in the range 0.0ImM to 5mM preferably O.lmM. Other oxidants such as CuCl₂ may be used or alternatively no oxidant may be added, thereby utilizing air oxidation only. For scale-up, 4M urea is the presently preferred solution for the oxidation step.

Concentration: The oxidation products are concentrated, preferably to about OD₂₈₀=1 by a tangential flow ultra- filtration system with a 30K cutoff membrane, such as a "MINITAN" or "PELLICON" system of Millipore. The filtrate is quite clear as the material is relatively clean and most of the contaminants are large enough not to pass through the 30K membrane. It is thus possible to reuse the filtrate for performing oxidations. This results in considerable savings since large volumes of 2M GuCl are quite expensive. No difference in the oxidation products of oxidations performed in reused versus freshly prepared 2M GuCl was detectable by FPLC analysis.

Dialysis: It is necessary to reduce the GuCl or urea concentration to less than lOmM. This is achieved by dialysis against 20mM HEPES pH8, lOOmM NaCl. The dialysis was performed in dialysis tubing with 2-3 changes of buffer, but may be alternatively performed by diafiltration against the same buffer in a tangential flow ultrafiltration system with a 10K MW cutoff membrane.

During dialysis, as the concentration of GuCl (or urea or other denaturant) decreases, a white precipitate forms. This precipitate contains about 80% of the protein yielded by step D comprising S-S linked dimers, reduced and incorrectly oxidized monomer and some contaminants which coeluted from the cation exchange step. The supernatant is nearly 100% correctly oxidized and refolded monomer at a concentration of 0.2 OD₂₈₀/ml, which is about 20% of the protein yield of step D. This selective precipitation of contaminants and undesirable forms of the protein as a result of dialysis was surprising and not predictable. The yield of correctly oxidized monomer can be greatly increased by recovery from the precipitate. This is done as follows: the solution is clarified by centrifugation. The supernatant is saved, and the pellet is treated with DTT to reduce S-S bonds and reoxidized as described above. The pellet is dissolved in a minimal volume of 6M GuCl, 20ml HEPES pH 8, 150mM NaCl, 20mM DTT. The solution was passed through Sephadex G25 in a buffer similar to the dissolution buffer but containing only ImM DTT (instead of 20mM) . The eluate is then diluted to OD₂₈₀=0.05 and treated as in steps C, D and E above. This procedure may be repeated more than once as long as additional purified monomer is obtained. All of the supernatants are then combined.

F. Cation exchange: The combined supernatant of the dialysate of step E is concentrated by binding to CM Sepharose in 20mM HEPES pH8, lOOmM NaCl. Elution is with 20mM HEPES pH8, 400mM NaCl. The eluate is exclusively monomeric despite the high salt concentration. Concentrations of up to 3 mg/ml have been achieved by this method and that is not the upper limit. This step can alternatively be performed with Heparin-Sepharose which also binds the purified monomer in lOmM Tris pH 7.4, 150mM NaCl. Elution from Heparin-Sepharose is performed using lOmM Tris-HCl pH 7.4, 500mM NaCl.

G. Dialysis: The product of the previous step is dialyzed against 20mM HEPES pH8, 150mM NaCl.

H. Storage: At this stage the purified vWF GPIb binding domain polypeptide may be lyophilized. Upon reconstitution in a volume of water equal to the volume before lyophilization, the resultant solution contains exclusively monomeric protein showing no traces of dimers or other multimers on FPLC.

In a specific embodiment of this method the following procedure was performed: a) 10 gm inclusion bodies (comprising 0.43 g net dry weight) were dissolved in a final volume of 100ml 8M urea, 20mM DTT, 20mM HEPES pH 8, 100 mM NaCl.

b) The protein was loaded onto a CM Sepharose column equilibrated with 8M urea, ImM DTT, 20mM HEPES pH 8, lOOmM NaCl. The protein eluted at 200mM NaCL in 8M urea, 20mM HEPES pH 8, ImM DTT, and was saved.

c) The saved eluate of the previous step was treated with 6M GuCl to eliminate any aggregates, and was then diluted to 0.05 OD₂₈₀/ml in 2M GuCl, 20mM HEPES pH 8, O.lmM GSSG. Oxidation was performed overnight at room temperature. (Note that the oxidation step can be performed in the presence of urea instead of GuCl.)

d) The oxidation products were concentrated to OD₂₈₀=1 by ultrafiltration on a "MINITAN" unit containing a 3OK membrane.

e) The concentrate of the previous step was dialyzed with three buffer changes against 20mM HEPES pH 8, lOOmM NaCl. During dialysis, as the GuCl concentration decreased, a white precipitate formed which was removed by centrifugation and reprocessed once as described above. The supernatants were combined.

f) The combined supernatants were concentrated by binding to CM Sepharose in 20mM HEPES pH 8, lOOmM NaCl. The polypeptide was eluted in 20mM HEPES, pH 8, 400mM NaCl and stored at 4^βC.

g) The saved eluate from the previous step was dialyzed against 20mM HEPES pH 8, 150mM NaCl at 4°C.

h) After dialysis, the purified vWF GPIb binding domain polypeptide, designated VCL, was lyophilized.

Analysis of VCL

1. Amino acid sequence analysis of VCL purified as described above revealed that the N-terminal sequence is Met-Leu-His-Asp-Phe which is the expected sequence according to Figure 12 with the addition of an N-terminal methionine residue.

2. Examination of VCL on polyacrylamide gels revealed that VCL electrophoreses at lower apparent molecular weight under non-reducing conditions than under reducing conditions (beta-mercaptoethanol) . This shift from compact to less compact configuration is consistent with the reduction of a disulfide bond. Such an intramolecular bond is formed between the cysteines at positions 509 and 695. (The shift in molecular weight is not large enough to be consistent with the reduction of an intermolecular bond.)

Example 5:

Biological Activity of VCL. a vWF GPIb binding domain polypeptide

The vwF GPIb binding domain polypeptide produced as described in Example 4 was designated VCL and was assayed for biological activity as described below.

1. Ristocetin induced platelet aggregation (RIPA)

RIPA assay was performed as described in Example 3 in a reaction mix containing 2xlO^θ platelets/ml, lμg/ml plasma vWF, and lmg/ml ristocetin. A series of concentrations of VCL was tested and the IC₅₀ of VCL in 3 assays was determined to be 0.2-0.3μM. 100% inhibition was achieved with about lμM VCL.

2. Asialo vWF induced platelet aggregation

Asialo vWF induced platelet aggregation assay was performed as described in Example 3 with 200μl platelet-rich plasma (PRP) and lOμg/ l asialo-vWF in a Lu i aggregometer. A series of concentrations of VCL was tested and the IC₅₀ of VCL in this assay was determined to be 0.15μM, and complete inhibition by 0.5 μM.

3. Effect of VCL on preformed aggregates

The effect of VCL on preformed aggregates made by RIPA was tested. Aggregates were formed as in paragraph (1) above in the absence of VCL. Addition of VCL to a concentration of 0.5μM disrupted the aggregates instantaneously. 4. Inhibition of thrombin induced Platelet aggregation

Thrombin induced platelet aggregation assay was performed using 0.025 unit/ml thrombin and stractan prepared platelets. A series of concentrations of VCL was tested and the IC₅₀ of VCL in this assay was determined to be 0.3μM. This is a surprising effect, since in a parallel experiment, VCL was not effective in inhibiting direct binding of [¹²⁵I] labelled thrombin to platelets.

5. Effect on platelet deposition under conditions of flow

In a model system consisting of an everted denuded human umbilical artery in a flow cell, platelet deposition may be determined. Whole human blood flows over the artery fragment. After 10-15 minutes, the flow is stopped and platelet deposition is determined microscopically. The IC₅₀ of VCL in this system was determined to be about lμM.

All the above results are summarized in Table III.

The inhibitory activity of VCL on ristocetin-induced or asialo vWF-induced platelet aggregation, ristocetin-induced vWF binding, and platelet adhesion was lost upon reduction of the disulfide bond between the cysteines at positions 509 and 695. In some experiments, the reduced VCL precipitated out of solution.

TABLE τττ

Biological Activity of VCL, a vWF GPIb Binding Domain Polypeptide.

Example 5 Assay μM VCL (Paragraph No.)

Ristocetin induced IC₅₀=0.2-0.3 platelet aggregation

Asialo vWF induced IC₅₀=0.15 platelet aggregation

Thrombin induced IC₅₀=0.3 platelet aggregation

Dissolution of 0.5 preformed aggregates

Platelet deposition ICso-1 under conditions of flow

Example 6:

Construction of Plasmid PVWF-VEL It was decided to construct a plasmid which expresses a slightly longer portion of the vWF GPIb binding domain than pvWF-VCL. The construction is shown in Figures 15-18 and described in the brief descriptions of the figures.

A. Construction of PVWF-VE2

Plasmid pvWF-VA2 (constructed as shown in Figure 4) was digested with Ndel and PstI and the large fragment isolated. Four synthetic oligomers shown in Figure 16 were prepared. Nos. 2 and 3 were treated with T4 polynucleotide kinase to add 5' phosphate. The above mentioned large fragment of pvWF-VA2 was ligated as shown in Figure 15 with the four oligomers (two kinased, and two non-kinased) . The resulting plasmid shown in Figure 15 was designated pvWF-VE2.

B. Construction of plasmid pyWF-VE3

Plasmid pvWF-VE2 was digested with Ndel and Hindlll and the 770 bp fragment containing the vWF GPIb binding domain was isolated. Plasmid pMLK-7891 was also digested with Ndel and Hindlll and the large fragment was isolated. The resulting plasmid, shown in Figure 17, was designated pvWF-VE3.

C. Construction of plasmid PVWF-VEL

Plasmid pvWF-VE3 was digested with XmnI, dephosphorylated with bacterial alkaline phosphatase (BAP) and then digested with Ndel and Hindlll. Plasmid pMLK-100 was digested with Ndel and Hindlll and dephosphorylated with BAP. The two digests were then ligated to yield plasmid pvWF-VEL as shown in Figure 18. This plasmid expresses the DNA sequence corresponding to amino acids 469-728 of mature vWF under the control of the λP_L promoter and the cll ribosomal binding site. The protein probably also includes an additional N- terminal methionine residue. A conservative base change was introduced into ala-473 changing GCC to GCA which also encodes alanine. This introduced an SphI site into the gene by changing GCCTGC to GCATGC.

Expression of pvWF-VEL in E. coli 4300(F^") yields a 29 kD protein which reacts strongly with a monoclonal anti-vWF antibody and will be referred to herein as VEL.

Example 7 :

Pharmaceutical Uses of vWF GPIb Binding Domain Polypeptide

Examples 1 and 4 describe the production and purification of a novel vWF GPIb binding domain polypeptide designated VCL. Some of the uses envisaged for VCL or for other vWF GPIb binding domain polypeptides are described below. Pharmaceutical compositions containing VCL or such other polypeptides may be formulated with a suitable pharmaceutically acceptable carrier using methods and carriers well known in the art.

1. The VCL composition described above may be used for prevention of platelet adhesion to damaged vascular surfaces

(see Example 5, sub-section 5).

2. The VCL composition described above may be used for disruption of platelet-rich aggregates (see Example 5, subsection 3) .

3. The VCL composition described above may be used for prevention of re-occlusion following angioplasty or thrombolysis (see Bellinger et al., PNAS, USA, £_£: 8100-8104 (1987) , Prevention of occlusive coronary artery thrombosis by a murine monoclonal antibody to porcine von Willebrand Factor) .

4. The VCL composition described above may be used for prevention of platelet activation and thrombus formation due to high shear forces such as in stenosed or partially obstructed arteries or at arterial bifurcations (see Peterson et al.. Blood £: 625-628 (1987), Shear-induced platelet aggregation requires von Willebrand Factor and platelet membrane glycoproteins lb and Hb-IIIa) . 5. The VCL composition described above may be used for prevention of thrombosis and re-occlusion after angioplasty or thrombolysis due to thrombin activation of platelets (see Fuster et al., J. Am. Coll. Cardiol. 1£: 78A-84A (1988), Antithrombotic therapy after myocardial reperfusion in acute myocardial infarction) .

6. The VCL composition described above may be used for prevention of platelet adhesion to and aggregation on prosthetic materials (see Badimon et al., J. of Biomaterials Applications, 5.: 27-48 (1990), Platelet interaction to prosthetic materials - role of von Willebrand Factor in Platelet Interaction to PTFE) .

7. The VCL composition described above may be used for prevention of intramyocardial platelet aggregation in patients with unstable angina (see Davies et al.. Circulation 73.: 418-427 (1986), Intramyocardial platelet aggregation in patients with unstable angina suffering sudden ischemic cardiac death) .

8. The VCL composition described above may be used for prevention of vasospasm and vasoconstriction following arterial injury caused by angioplasty, thrombolysis or other causes (see Lam et al.. Circulation 7£: 243-248 (1987), Is vasospasm related to platelet deposition?) .

9. The VCL composition described above may be used for prevention of restenosis following angioplasty or thrombolysis (see McBride et al., N. Eng. J. of Med. 318: 1734-1737 (1988), Restenosis after successful coronary angioplasty) .

10. The VCL composition described above may be used for prevention of vaso-occlusive crises in sickle-cell anemia (see Wick et al., J. Clin. Invest. £_: 905-910 (1987), Unusually large von Willebrand multimers increase adhesion of sickle erythrocytes to human endothelial cells under controlled flow) .

11. The VCL composition described above may be used for prevention of thrombosis associated with inflammatory response (see Esmon, Science 235: 1348-1352 (1987) , The regulation of natural anticoagulant pathways) .

12. The VCL composition described above may be used for prevention of arteriosclerosis (see Fuster et al. , Circulation Res. SI: 587-593 (1982), Arteriosclerosis in normal and von Willebrand pigs) .

13. The VCL composition described above may be used as an antimetastatic agent (see Kitagawa et al.. Cancer Res. 49: 537-541 (1989) , Involvement of platelet membrane glycoprotein lb and lib/Ilia complex in thrombin-dependent and -independent platelet aggregations induced by tumor cells) .

1. In Vitro gfrudiey

For these studies VCL, or vehicle control, was made up fresh in sterile water (2.2 mg/ml stock).

A. Platelet Aggregation (PRP)

This is a standardized von Willebrand Factor (vWF)-dependent aggregation assay using human or rat platelet rich plasma (PRP) . The addition of various concentrations of unfractionated Bothrops jararaca venom (BJV) , which includes botrocetin and an additional thrombin-like component, or ristocetin results in an aggregatory response in the absence of any additional agent. Using ristocetin (1.5 mg/ml) as the agonist 43 μg/ml VCL abolished the aggregation of human PRP. Ristocetin up to 5.0 mg/ml did not cause measurable aggregation of rat PRP. Using this assay system with BJV as the agonist VCL at 83 μg/ml slightly inhibited the response of human PRP (Figure 19) but not rat PRP (Figure 20) .

It is concluded that ristocetin is not a suitable agonist for inducing vWF dependent aggregation in the rat. Further, it is not possible to monitor the effects of VCL ex vivo using BJV-induced aggregation of rat PRP. However, VCL does inhibit vWF-dependent aggregation in human PRP in vitro.

B. Platelet Thrombin Receptor Assay

This assay measures the inhibition of thrombin-induced platelet pro-coagulant expression and is briefly described below. Human washed platelets are incubated in a buffer which contains CaCl₂, factor Xa, prothrombin, and human alpha-thrombin for 60 min at 28°C. At the end of this period an aliquot is transferred into a buffer containing S2238 and EDTA (to prevent any further thrombin generation) . The S2238 reaction is terminated after 15 minutes at room temperature with acetic acid and the absorbance at 405 nm read. The amount of S2238 cleavage directly due to the added human alpha-thrombin is. estimated by including a control which contains no prothrombin and this value is subtracted from all results. VCL was tested in this assay at a final concentration of 0.1 mg/ml.

This assay is sensitive to both thrombin inhibitors and thrombin receptor antagonists. In the presence of VCL the thrombin generation was 114% of control (n-2) .

We therefore conclude that VCL is not a thrombin receptor antagonist in this system.

2. In Vivo Studies

Arterial T-__pπfl.psjs Mod l (Rat)

This method is essentially a modification of the model of Shand et al., Thromb. Res. 4J> 505-515 (1987). The method we use routinely is outlined below.

Rats are labelled with ^13,1In platelets and ¹²⁵I fibrinogen. The dorsal aorta is clamped, using modified Spencer-Wells forceps, for 1 minute. After a 45 minute reperfusion period the damaged vessel is removed, washed in citrate and counted. Results are expressed as mg blood equivalents. Differences in radiolabel accumulation between placebo and drug-treated animals are calculated and expressed as a percentage inhibition.

For the purpose of the evaluation of VCL the route of administration was by bolus intravenous injection. VCL was used at doses of 2mg/kg (n=5) and 4mg/kg (n=3). It was administered 1 minute prior to clamping. The vessel was then reperfused for twenty minutes. The antithrombotic effect was assessed at the 20 minute end point of the reperfusion. The shortening of the reperfusion time (as compared to routine) was designed to save compound. Appropriate vehicle controls (n-5 for both doses) were assessed.

It can be seen that under these conditions VCL inhibits thrombus formation in this model (Table IV) . The inhibition is seen for the platelet (^IIΣIn) components of the thrombus at the 4mg/kg dose. The other changes do not reach statistical significance, thus VCL shows antithrombotic efficacy in this rat arterial model.

In conclusion, VCL exhibits an antithrombotic effect in thiε rat model of arterial thrombosis which may be dose dependent.

3. Discussion

From the present data it appears that the VCL interacts with the human platelet vWF receptor and hence inhibits platelet aggregation in human PRP. There is however a marked difference between species (rat vs. human) when comparing inhibition of platelet aggregation. The species specificity of this effect and the causal mechanism were not investigated further. At a practical level this meant we were unable to analyze ex vivo samples in order to correlate the effects of VCL on aggregation and arterial thrombosis.

Hence the analysis and interpretation of the in vivo efficacy of VCL as an antithrombotic in the rat arterial thrombosis model is complicated by this factor. Despite GPIb possessing binding sites for both vWF and thrombin it would appear that any effects of VCL on thrombin binding to GPIb do not translate into antagonism of thrombin-induced pro-coagulant expression.

Overall VCL shows an antithrombotic effect in the rat arterial thrombosis model. This inhibition may be due to its interference with the binding of vWF to its receptor.

TABLE IV

The Effect of VCL on Arterial Thrombus Formation in the Rat Dorsal Aorta

The results are expressed as mean percentage inhibition ± standard error. The number of experiments in the treated groups are denoted in the table and in all cases were compared to a group of 5 control animals. Statistical analysis was performed on the raw data prior to transformation to percentage inhibition. NS = not statistically significant.

Claims (1)

1. What is claimed is:

   1. A non-glycosylated, biologically active polypeptide having the amino acid sequence: 5

   X-A-[Cys Ser Arg Leu Leu Asp Leu Val Phe Leu Leu Asp Gly Ser Ser Arg Leu Ser Glu Ala Glu Phe Glu Val Leu Lys Ala Phe Val Val Asp Met Met Glu Arg Leu Arg He Ser Gin Lys Trp Val Arg Val Ala Val Val Glu Tyr His Asp Gly Ser His

   10 Ala Tyr He Gly Leu Lys Asp Arg Lys Arg Pro Ser Glu Leu Arg Arg He Ala Ser Gin Val Lys Tyr Ala Gly Ser Gin Val Ala Ser Thr Ser Glu Val Leu Lys Tyr Thr Leu Phe Gin He Phe Ser Lys He Asp Arg Pro Glu Ala Ser Arg He Ala Leu Leu Leu Met Ala Ser Gin Glu Pro Gin Arg Met Ser Arg Asn

   15 Phe Val Arg Tyr Val Gin Gly Leu Lys Lys Lys Lys Val He Val He Pro Val Gly He Gly Pro His Ala Asn Leu Lys Gin He Arg Leu He Glu Lys Gin Ala Pro Glu Asn Lys Ala Phe Val Leu Ser Ser Val Asp Glu Leu Glu Gin Gin Arg Asp Glu He Val Ser Tyr Leu Cys]-B-COOH

   20 wherein X is NH₂-methionine- or NH₂-;

   A is a sequence of at least 1, but less than 35 amino acids, which sequence is present in naturally

   25 occurring human vWF, the carboxy terminal amino acid of which is the tyrosine #508 shown in Figure 12;

   B is a sequence of at least 1, but less than 211 30 amino acids, which sequence is present in naturally occurring human vWF, the amino terminal amino acid * of which is the aspartic acid #696 shown in Figure

   12; and

   35 the two cysteines included within the bracketed sequence are joined by a disulfide bond.

   2. A polypeptide of claim 1 having the amino acid sequence:

   X-[Leu His Asp Phe Tyr Cys Ser Arg Leu Leu Asp Leu Val Phe Leu Leu Asp Gly Ser Ser Arg Leu Ser Glu Ala Glu Phe Glu Val Leu Lys Ala Phe Val Val Asp Met Met Glu Arg Leu Arg He Ser Gin Lys Trp Val Arg Val Ala Val Val Glu Tyr His Asp Gly Ser His Ala Tyr He Gly Leu Lys Asp Arg Lys Arg Pro Ser Glu Leu Arg Arg He Ala Ser Gin Val Lys Tyr Ala Gly Ser Gin Val Ala Ser Thr Ser Glu Val Leu Lys Tyr Thr Leu Phe Gin He Phe Ser Lys He Asp Arg Pro Glu Ala Ser Arg He Ala Leu Leu Leu Met Ala Ser Gin Glu Pro Gin Arg Met Ser Arg Asn Phe Val Arg Tyr Val Gin Gly Leu Lys Lys Lys Lys Val He Val He Pro Val Gly He Gly Pro His Ala Asn Leu Lys Gin He Arg Leu He Glu Lys Gin Ala Pro Glu Asn Lys Ala Phe Val Leu Ser Ser Val Asp Glu Leu Glu Gin Gin Arg Asp Glu He Val Ser Tyr Leu Cys Asp Leu Ala Pro Glu Ala Pro Pro Pro Thr Leu Pro Pro Asp Met Ala Gin Val Thr Val Gly Pro Gly Leu Leu Gly Val Ser Thr Leu Gly Pro Lys]-COOH

   wherein X is NH₂-methionine- or NH -.

   A pharmaceutical composition comprising an amount of a polypeptide of claim 1 or 2 effective to in¬ hibit platelet aggregation and a pharmaceutically acceptable carrier.

   4. A method of inhibiting platelet aggregation which comprises contacting platelets with an amount of a polypeptide of claim 1 or 2 effective to inhibit platelet aggregation so aε to inhibit platelet aggregation. 5. An expression plasmid encoding a polypeptide of claim 1.

   »

   6. An expression plasmid encoding a polypeptide of

   * 5 claim 2 designated pvWF-VC3 and deposited under

   ATCC Accession No. 68241.

   7. An expression plasmid encoding a polypeptide of claim 2 designated pvWF-VCL deposited under ATCC

   10 Accesεion No. 68242.

   8. A bacterial cell which comprises the expression plasmid of claim 5, 6, or 7.

   15 9. An Escherichia coli cell of claim 8.

   10. A method of producing a polypeptide of claim 1 which comprises transforming a bacterial cell with an expression plasmid encoding the polypeptide,

   20 culturing the resulting bacterial cell so that the cell produces the polypeptide encoded by the plas¬ mid, and recovering the polypeptide so produced.

   11. A method of producing a polypeptide of claim 2 25 which comprises transforming a bacterial cell with an expression plasmid encoding the polypeptide, culturing the resulting bacterial cell so that the cell produces the polypeptide encoded by the plas¬ mid, and recovering the polypeptide so produced. 30

   12. A method of treating a subject with a cerebrovascu-

   * lar disorder which comprises administering to the subject an amount of a polypeptide of claim 1 or 2 effective to inhibit platelet aggregation.

   35 13. A method of treating a subject with a cardiovascu¬ lar disorder which comprises administering to the subject an amount of a polypeptide of claim 1 or 2 effective to inhibit platelet aggregation.

   14. A method of treating a subject in accordance with claim 13, wherein the cardiovascular disorder comprises acute myocardial infarction.

   15. A method of treating a subject in accordance with claim 13, wherein the cardiovascular disorder comprises angina.

   16. A method of inhibiting platelet aggregation in a subject prior to, during, or after the subject has undergone angioplasty, thrombolytic treatment, or coronary bypass surgery which comprises administer¬ ing to the subject an amount of a polypeptide of claim 1 or 2 effective to inhibit platelet aggregation.

   17. A method of maintaining blood vessel patency in a subject prior to, during, or after the subject has undergone coronary bypass surgery, which compriseε administering to the subject an amount of a poly¬ peptide of claim 1 or 2 effective to inhibit plate¬ let aggregation.

   18. A method of treating a subject having cancer which compriseε administering to the subject an amount of a polypeptide of claim 1 or 2 effective to retard tumor metastasis.

   19. A method of inhibiting thrombosis in a subject which compriseε administering to the subject an amount of a polypeptide of claim 1 or 2 effective to inhibit the thrombosis.

   20. A polypeptide in accordance with claim 1 or 2 bound to a solid matrix.

   21. A method of treating a subject suffering from platelet adhesion to damaged vascular surfaceε which comprises administering to the subject an amount of the polypeptide of claim 1 or 2 effective to inhibit platelet adhesion to damaged vaεcular surfaces.

   22. A method of preventing platelet adhesion to a prosthetic material or device in a subject which comprises administering to the subject an amount of the polypeptide of claim 1 or 2 effective to prevent platelet adhesion to the material or device.

   23. A method of inhibiting re-occlusion in a subject following angioplasty or thrombolysis which com¬ prises administering to the subject an amount of the polypeptide of claim 1 or 2 effective to inhib- it re-occlusion.

   24. A method of preventing vaso-occlusive crises in a subject suffering from sickle cell anemia which compriseε administering to the εubject an amount of the polypeptide of claim 1 or 2 effective to pre¬ vent vaεo-occluεive crises.

   25. A method of preventing arteriosclerosis in a εub¬ ject which compriseε administering to the subject an amount of the polypeptide of claim 1 or 2 effec- tive to prevent arteriosclerosis.

   26. A method of thrombolytic treatment of thrombi- containing, platelet-rich aggregates in a εubject which comprises administering to the subject an amount of the polypeptide of claim 1 or 2 effective to treat thrombi-containing, platelet-rich aggre¬ gates.

   27. A method of preventing platelet activation and thrombus formation due to high shear forces in a subject suffering from stenosed or partially ob¬ structed arteries which comprises administering to the subject an amount of the polypeptide of claim 1 or 2 effective to prevent platelet activation and thrombus formation.

   28. A method of preventing thrombin-induced platelet activation in a subject which comprises ad inister- ing to the εubject an amount of the polypeptide of claim 1 or 2 effective to prevent thrombin-induced platelet activation.

   29. A method of preventing stenosis as a result of smooth muscle proliferation following vascular injury in a subject which comprises administering to the subject an amount of the polypeptide of claim 1 or 2 effective to prevent stenosis.

   30. A method of claim 19, wherein the thrombosis is asεociated with an inflammatory response.

   31. A method for recovering a purified, biologically active polypeptide of claim 1 which comprises: (a) producing in a bacterial cell a firεt polypeptide having the amino acid sequence of the polypeptide but lacking the disulfide bond;

   (b) disrupting the bacterial cell so as to produce a lysate containing the first polypeptide;

   (c) treating the lysate so as to obtain inclusion bodieε containing the firεt polypeptide;

   (d) contacting the inclusion bodies from step (c) so as to obtain the first polypeptide in soluble form;

   (e) treating the resulting first polypeptide so as to form the biologically active polypeptide;

   (f) recovering the biologically active polypeptide so formed; and

   (g) purifying the biologically active polypeptide so recovered.

   32. A method of claim 31, wherein the treatment step (e) compriseε contacting the polypeptide with a thiol-containing compound and disulfide.

   33. A method of claim 32, wherein the thiol-containing compound is glutathione, thioredoxin, B- mercaptoethanol or cyεteine. 34. A method of claim 31, wherein the contacting of εtep (d) iε effected in the presence of a denaturant.

   35. A method of claim 34, wherein the denaturant is guanidine hydrochloride or urea.

   36. A method of claim 34, wherein the recovery of the polypeptide of step (f) compriseε removing the denaturant by dialysis.

   37. A method of claim 31, wherein in step (g) the polypeptide is purified by cation exchange chroma- tography.

   38. A method of claim 37, wherein the first polypeptide is purified by cation exchange chromatography after step (d) .