Vector Containing DNA Encoding Mature Human Protein S

This invention relates to Protein S and the use of recombinant DNA techniques to produce Protein S.

The coagulation cascade is a series of reactions, involving a number of plasma proteins, which occur in response to an injury. One of these plasma proteins, Protein S, helps ensure that blood does not clot in places other than the specific site of the injury. Protein S is a cofactor for the activated Protein C-induced cleavage of activated Factors V and VIII in the blood clotting pathway.

Recently Comp et al., (1984, J. Clin. Invest. 74: 2082) have suggested that Protein S deficiency may result in recurrent thrombotic disease; and Schwarz et al. (1984, Blood 64: 1297) have suggested that recurrent thrombotic disease may be the result of an inherited deficiency of Protein S.

In general, the invention features a vector containing DNA encoding mature human Protein S. As will be apparent from the detailed description below, "mature human Protein S" refers to the biologically active Protein S molecule without the leader sequence; thus the term as used herein is broad enough to include a DNA sequence (genomic or, more preferably, cDNA) which encodes at least the mature human Protein S, and which can also encode human or other (e.g., bovine) leader sequences, or hybrid leader sequences.

The vectors of the invention are used to transfect mammalian cells, e.g., rodent (e.g., mouse) cells to produce biologically active human Protein S, which can be used both therapeutically and diagnostically. Protein S can be used to treat congenital Protein S deficiency, and can also be administered in conjunction with Protein C, of which Protein S is a cofactor, to enhance its antithrombolytic activity. Protein S and Protein C can also be administered together in conjunction with the thrombolytic enzyme tissue plasminogen activator (t-PA) to enhance its activity and advantageously reduce the amount of t-PA required.

Other features and advantages of the invention will be apparent from the following description of the preferred embodiments thereof, and from the claims.

The drawings will first briefly be described.

Drawings

FIG. 1 is a restriction map for the full length bovine protein S clone, pBLS-2400;

FIG. 2-1 to 2-2 are the nucleotide sequence of cDNA coding for bovine protein S, and the amino acid sequence of protein S; Gla stands for .gamma.-carboxyglutamic acid; the Asp in position 93 is hydroxylated to .beta.-hydroxyaspartic acid; the Asp in position 458 is glycosylated (arrowhead); the arrow denotes the start of the mature plasma protein; the polyadenylation signal is underlined;

FIG. 3 is a restriction map for human Protein S cDNA clones; (A) diagram representing the isolated .lambda. gt 11 cDNA clones coding for human Protein S; (B) partial restriction map for Protein S cDNA; the thick line represents those sequences that code for Protein S;

FIGS. 4A-1 to 4B-2 are the nucleotide sequence of cDNA coding for human Protein S, and its predicted amino acid sequence; the arrow denotes the start of the mature plasma protein; the eleven NH.sub.2 -terminal glutamic acid residues are assumed to be carboxylated (Gla); there are three potential glycosylation sites; i.e., Asn residues 458, 468 and 489; two putative polyadenylation signals are underlined;

FIGS. 5, 6, and 7 are diagrammatic representations of the preparation of pieces 1, 2, and 3 respectively, used for construction of pUC9XBHS;

FIG. 8 is a diagrammatic representation of the construction of pUC9XBHS; and

FIG. 9 is a diagrammatic representation of the construction of expression vectors, CSM and CSS, containing the human Protein S encoding DNA.

General Approach

In the example described below, the amino acid sequence of bovine protein S was determined and a cDNA sequence encoding the complete bovine Protein S isolated from a clone which hybridized to oligonucleotides prepared to a portion of mature bovine Protein S. A fragment derived from this clone was then used as a probe to identify a clone containing a cDNA sequence encoding the complete mature human Protein S. In order to facilitate its expression in mammalian cells, the DNA encoding the leader sequence of the bovine Protein S cDNA was spliced onto the human Protein S cDNA. These steps are now described in detail.

Protein S fragments were generated by enzymatic and chemical cleavages and the fragments purified by gel filtration and reverse phase high performance liquid chromatography (HPLC) as described by Dahlbach et al. (1986, J. Biol. Chem., in Press). Amino acid sequence was determined on a Beckman 890 C spinning cup sequencer, or on an Applied Biosystem gas phase sequencer; and the amino acid composition of peptides on a Beckman 6300 amino acid analyzer.

These data allowed identification and placement of 600 amino acids, contained in nine fragments, out of a total of 634. The order of the fragments in the sequence was tentatively determined. Approximately 130 fragments were analyzed either in the spinning cup or the gas phase sequencer. Most of Protein S was sequenced at least twice.

Two mixed oligonucleotide probes were synthesized; probe 1 was a 23-mer coding for amino acid residues 85-93 and probe 2 a 20-mer coding for residues 595-600: ##STR1##

A bovine liver cDNA library (in pBR322) was screened by standard techniques (Grunstein and Hogness 1975, Proc. Nat.'l. Acad. Sci. 72:3961). A total of 10.sup.5 recombinants were screened with the two oligonucleotide probes. One clone, pBLS-200, was identified using probe 1. Sequence analysis showed it to have an insert that coded for amino acid residues 44-104. With probe 2, six clones were identified, but two had inserts less than 150 nucleotides long and were not examined further. The other four clones contained identical inserts coding for the 105 carboxyterminal amino acids and 76 nucleotides of the 3'nontranslated region. One of the four identical clones, pBLS-400, was used together with pBLS-200 to rescreen the library. Fifteen clones hybridized to pBLS-400, one to pBLS-200, and one to both. The latter, clone pBLS-2400 (shown in FIG. 1) carried a cDNA insert of approximately 2.4 kb suggesting that it contains the entire protein S coding sequence. In order to determine the nucleotide sequence, 60 .mu.g of this plasmid DNA was digested with Pst 1. The cDNA insert was isolated, self-ligated, sonicated, and endrepaired. Subfragments of 300-600 base pairs were isolated and ligated into the Smal site of the replicative form of the phage M 13 mp8, which was subsequently used to transfect E. coli K 12-JM 101 cells. Forty-two white plaques, picked at random, were used for preparation of template DNA and sequenced, yielding a continuous sequence of 2379 nucleotides (FIG. 2).

In order to keep an open reading frame, an extra nucleotide had to be introduced around position 595-600. This area was sequenced several times on both strands and unambigously gave the sequence TGTAAAATGGT (position 592-601). Protein sequencing in this region gave the sequence CMCys-Lys-Asn-Gly, which indicates that one nucleotide (A) had been artifactually deleted at position 595-599, presumably during the preparation of the cDNA library.

A human fetal liver cDNA library in lambda gt 11 was prepared by a modification of the procedure of Gubler and Hoffman (1983, Gene 25:263), similar to that described by Lapeyre and Amalric (1985, Gene, 37:215). The library contained over 6.times.10.sup.7 recombinants with inserts averaging 1800 nucleotides in length. Plaques from the amplified library were screened using standard techniques. Nick translated DNA fragments from a bovine Protein S cDNA clone, pBLS-2400 (FIG. 1) and a human Protein C cDNA clone obtained from Robert Wydro of Integrated Genetics, Inc. (Wydro et al., described in detail in (Wydro et al. application Ser. No. 51,366 assigned to the same assignee as this application) were used as probes.

Approximately 1.times.10.sup.6 plaques of the human .lambda.gtll library were screened using the SphI-SacI fragment (coding for amino acids -29 to 620) from pBLS-2400 (FIG. 1) Four positive clones, designated M26, M31, M33 and M48 (FIG. 3) were plaque purified and characterized. These clones were confirmed as coding for human protein S by comparison to the bovine Protein S cDNA sequence, to which they showed significant homology. The EcoRI-Xbal fragment (coding for amino acids 311 to 403) from the the clone M48S was then used to screen 18 million plaques from the .lambda.gtll library. Thirty-nine positives were identified. To distinguish those human clones that contained inserts coding for the 5' region of the mRNA, successive hydridizations were performed with DNA fragments of pBLS-2400 containing 5' end sequences, e.g., the SphI-AvaII fragment. Four clones hydridized with the 5' coding sequences from the bovine cDNA. Of these, three were subsequently shown to be identical. These clones, typified by M117S (FIG. 3) lacked only the coding region for the first 26 amino acids of the leader peptide, as evidenced by a comparison to the bovine sequence.

The total cloned Protein S cDNA contained 3,344 bp and coded for amino acids -15 of the leader sequence to the termination codon after Ser (amino acid 635). The nucleotide sequence of the human Protein S cDNA and the derived amino acid sequence are given in FIG. 4. The cDNA contained a 1,148 bp 3' untranslated region, which was 826 bp longer than its bovine counterpart. The homology between the bovine and the human Protein S cDNA clones extends through all of the 3' untranslated region of the bovine cDNA up to, but not including the poly(A) tail. Northern blot data suggests the existence of only one major molecular weight species of human Protein S mRNA. Although the possibility of extended 5' untranslated sequences cannot be excluded, the 3' untranslated region could account for all of the size difference between the human and bovine Protein S mRNA's.

The entire cloned human Protein S cDNA sequence when aligned with the bovine cDNA sequence contains a 244 bp 5' sequence which differs to a large degree from the bovine sequence. The overall homology between the nucleic acid sequences of the mature human and bovine Protein S is 87.5% and homology between the protein sequences is 81.6%.

In order to insert human Protein S cDNA into expression vectors, unique restriction enzyme sites, as well as a complete leader sequence, were engineered onto the 5' and 3' ends of the human Protein S clone.

The restriction enzyme site Xho I is very rare in nature and is not present in the human Protein S coding sequence. This restriction enzyme had been used previously to insert genes into vectors containing the transforming BPV virus (Wei et al., application Ser. No. 782,686, filed Oct. 1, 1985, assigned to the same assignee as this application, and hereby incorporated by reference). Since the human Protein S cDNA did not encode the full leader peptide, and this peptide may be necessary for post-translational modifications, the leader sequence of the bovine Protein S was spliced onto the human Protein S cDNA, preserving as much of the human cDNA as possible. The bovine leader sequence also may have closer homology than the human sequence to the corresponding rodent sequences, providing potential expression and processing advantages where the mammalian expression system used is of rodent origin.

Due to the nature of distribution of the restriction enzyme sites in the bovine and human Protein S cDNA sequences, a multiple step construction was necessary to engineer the bovine leader sequence onto the human cDNA with the necessary Xho I sites at the 5' and 3' ends of the final construction. The procedure used is illustrated in FIGS. 5 through 8, and was carried out as follows.

Referring to FIG. 5, the bovine Protein S cDNA clone pBLS2400 was restricted at a unique site with the enzyme AvrII. To prevent self-ligation of the linear DNA during the next set of experiments, the terminal phosphates were removed with bacterial alkaline phosphatase (BAP). Two synthetic oligonucleotides, BD382 (5'-CTAGGACCCTCATC-3') and BD 383 (5'-TCGAGATGAGGGTC-3'), containing terminal phosphate groups, were annealed and ligated in the presence of the linear pBLS2400. The desired construction contains a dimer of the annealed oligonucleotides inserted into the AvrII site of pBLS2400, generating an XhoI site. This construction, named pBLS2400X, was digested with XhoI and TaqI, and an 87 bp 5'-end piece isolated from low-melting point (LMP) agarose gel using a modification of the procedure described by Langridge et al. (1980, Anal. Bioch: 103:264-271), as described below. This 87 bp fragment contains the initiator methionine and encodes the leader signal sequence of bovine Protein S. It is designated "Piece 1".

QN.sup.+ -aqueous and QN.sup.+ -butanol were prepared as follows. 2 g of hexadecyltrimethylammonium bromide (QNBr) were dissolved in 100 ml n-butanol, 100 ml H.sub.2 O and 50 .mu.l antifoam A(Sigma), vigorously shaken, and allowed to separate overnight. The upper layer, defined as QN.sup.+ -butanol, and the lower layer, defined as QN.sup.+ -aqueous, were bottled separately.

The procedure includes the following steps.

A band is cut out from a low melting point (LMP) agarose gel; melted at 70.degree. C. for about 3 min; an equal volume of QN.sup.+ -aqueous and QN.sup.+ butanol added, the solution vortexed and centrifuged in a microfuge for 1-2 min; the butanol (upper) phase removed to a fresh tube; the QN.sup.+ -aqueous-agarose phase reheated to 70.degree. C. for 2-3 min, and re-extracted once with QN.sup.+ -butanol; 1/4 volume of 0.4M NaCl added to the combined butanols, the mixture vortexed and spun briefly in a microfuge; the aqueous (lower) layer transfered to a fresh tube; extraction with chloroform; and the DNA precipitated with ethanol.

To join the human Protein S cDNA with the bovine Protein S Leader (Piece 1), a TaqI site was created in the human cDNA at an appropriate site. To accomplish this mutation, the human Protein S cDNA clone 117S-1 (the 2,200 bp EcoRI fragment from M117S subcloned into pUC18) was digested with EcoRI and HindIII. A 580 bp fragment was isolated from a LMP agarose gel and further digested with RsaI. (The 1600 bp fragment was also isolated from an LMP agarose gel and used in the construction of Piece 3, as described below.) The 380 bp fragment generated was isolated from an LMP agarose gel and ligated into the vector mp18 which was previously digested with SmaI and HindIII, to give mp18-2ori. The single-stranded form of mp18-2ori was isolated and the appropriate base pair changed by site directed mutagenesis using a synthetic oligonucleotide BW363, as described below. The mutation was identified by hybridization analysis and a clone (termed mp18-2mut) isolated in bulk quanitities.

In order to construct the additional TaqI site in the human Protein S cDNA, 3.4 .mu.g of mp18-2ori and 1.4 .mu.g of kinased oligonucleotide BW 363 (5'-TGAAGCCTGTTGCTTCGACAAAACTCTTCC-3') were annealed in 30 .mu.l of 0.2M Tris-HCl, pH7.5; 0.1M MgCl.sub.2 ; 0.5M NaCl and 10 mM DTT by heating to 65.degree. C. for 10 minutes and then slowly cooled to room temperature over a two hour period.

The second strand of the phage DNA was synthesized by adding 50 .mu.l of buffer (20 mM Tris,pH7.5; 10 mM MgCl.sub.2 ; 10 mM DTT; 2.5 mM each of dATP, dCTP, dGTP and dTTP), along with 5 .mu.l of 20 mM ATP, 1 .mu.l T4 Ligase (400,000 units), 3.5 .mu.l E. coli DNA polymerase, Klenow fragment (14 units), and 10.5 .mu.l of distilled water. The mixture was incubated at room temperature for 15 minutes, then overnight at 15.degree. C.

In order to reduce the background of vector which has not been made double stranded by the above priming and extension reaction, about 20 .mu.l of the above extension reaction was mixed with 58 .mu.l of water, and 20 .mu.l of buffer containing: 1.5M NaCl; 0.25M Na Acetate, pH 4.6; 15% glycerol, 1 .mu.l of E.coli tRNA(2.mu./.mu.l), and 1 .mu.l of Sl nuclease (2.65 units). After incubation at 37.degree. C. for 30 minutes the reaction was terminated by the addition of 1 .mu.l of 500 mM EDTA and 50 .mu.l of TE (Tris 10 mM, pH 8; EDTA 1 mM). The mixture was phenol-cholorform extracted and the DNA precipitated from the aqueous phase with ethanol. The precipitated DNA was dissolved in 10 .mu.l of TE.

This Sl treated extension was transformed into competent JM101 E. coli cells. Mutants were identified using filter lifts on the plates containing the plaques from the Sl treated extension reaction. The synthetic oligonucleotide BW363 was kinased to a high specific activity with .sup.32 P-ATP and used as a hybridization probe in a solution containing: 0.74M NaCl; 0.05M NaH.sub.2 PO.sub.4, pH 7.4; 5 mM EDTA; 5.times.Denhardt's solution; 0.1% SDS; and 100 .mu.g/ml Torula RNA. Hybridization was at 40.degree. C. overnight. The filters were washed with several changes of 1.times.SSC/0.1% SDS at 57.degree. C. The resulting filters were autoradiographed, the autoradiograph aligned to the culture plate, and a strongly hybridizing plaque picked and grown overnight in 5 ml of medium. A plasmid mini-prep was performed to isolate the double stranded Rf form of the phage and the mutant confirmed by DNA sequence analysis. The phage DNA was digested with EcoRI and HindIII and a 400 bp fragment containing the new mutation isolated from a LMP agarose gel. This DNA fragment was cloned into pUC18 previously digested with EcoRI and HindIII. This new plasmid was termed pUC18mut2S.

DNA consisting of a 430 bp TaoI-PvuII fragment containing an internal unrestricted TaqI site was generated from a 490 bp EcoRI-PvuII DNA fragment of pUC18mut2S as follows: Separate aliquots containing 1 .mu.g of the 490 bp DNA fragment in 10 .mu.l total volume were digested at 55.degree. C. with 0.8 units of TaqI for 15, 20, 25, 30 and 35 minutes respectively; four separate aliquots each containing 4 .mu.g of the 490 bp DNA fragment in 27 .mu.l were restricted at 65.degree. C. for 15 minutes with 0.8, 1.6, 2.4, and 3.2 units respectively of TaqI; four separate aliquots each containing 4 .mu.g of the 490 bp DNA fragment in a total of 27 .mu.l were restricted at 65.degree. C. for 15 minutes with 3.2 units of TaqI; and two separate aliquots each containing 4 .mu.g of the 490 bp DNA fragment in a total of 27 .mu.l were restricted at 65.degree. C. for 15 minutes with 3.9 units of TaqI. After enzymatic treatment, the DNA of each aliquot was electophoresed in a 0.9% LMP agarose gel, bands containing the 430 bp fragment were excised from the gel and pooled. To insure purity of the fragment and facilitate joining to Piece 3 (below), the 430 bp DNA fragment was digested with HindIII and the resulting 330 bp fragment isolated. This fragment was designated "Piece 2".

To complete the contruction, the 3' end of the clone needed to be altered to establish an XhoI cloning site. To accomplish this, two oligonucleotides were made which, when annealed, constitute a linker that can ligate to a 3' EcoRI site, replacing the missing nucleotides and the termination codon, and add an XhoI site. This was performed using the 1600 bP HindIII-EcoRI fragment isolated from 117S-1 as described above, and the synthetic oligonucleotides EC347 (5'-AATTCTTAAGCTCGAGC-3') and EC348 (5'-GCTCGAGCTTAAG-3').

The oligonucleotides were annealed and ligated to the 1600 bp HindIII-EcoRI fragment using techniques analogous to those described above. After ligation, the DNA was digested with HindIII and XhoI and a 1600 bp DNA fragment isolated from a LMP agarose gel. This DNA fragment was cloned into pUC9X (pUC9 digested with SmaI and an XhoI linker added) which had previously been digested with HindIII and XhoI. Chimeric plasmids were identified by restriction analysis, and one clone isolated and grown in bulk. This plasmid was then digested with HindIII and XhoI and the 1600 bp insert isolated from LMP agarose gel. This 1600 bp HindIII-XhoI DNA fragment is designated "Piece 3".

Piece 1 and Piece 2 were ligated together, digested with XhoI and HindIII, electrophoresed on a 1.4% LMP agarose gel, and the 430 bp DNA fragment excised from the gel. This fragment was then ligated into pUC9X previously digested with XhoI and HindIII. The chimeric plasmid was identified by restriction analysis and a clone (termed pUC9X-Hl-2S) isolated.

pUC9X-Hl-2S was restricted with HindIII and XhoI and the 430 bp DNA fragment isolated from a LMP agarose gel. This fragment was ligated to Piece 3 and the resulting DNA digested with XhoI and electrophoresed in a 0.8% LMP agarose gel. The 2,000 bp DNA fragment was excised from the gel, and cloned into pUC9X which had been digested with XhoI and treated with BAP. The desired clone was identified by restriction analysis and one clone (termed pUC9XBHS) isolated in bulk quantities.

The modified cDNA in pUC9XBHS can be inserted into any suitable mammalian expression vector, most preferably rodent epithelioid cells. Preferred expression vectors are the BPV vectors described in Wei et al. supra, and Hsiung et al. 1984, J. Molec. and App. Genet. 2:497. The vectors (FIG. 9) include a mouse metallothionein promoter (MT) from which inserted genes can be transcribed, and bovine papilloma virus DNA (BPV) to effect transfection of mammalian cells. CLH3axBPV also includes late promoter poly-adenylation sequences derived from SV40, which can affect expression from a gene inserted into the vector. The illustrated expression plasmids also include a portion of the E. coli plasmid pML, which permits shuttling between procaryotic and eucaryotic systems. No selection is required for the maintenance of these plasmids in host cells, and they are maintained in high (approximately 100 copies/cell) copy number.

The XhoI-linkered Protein S sequence in pUC9XBHS was isolated by digestion with XhoI and excision of the insert band from an agarose gel. This fragment was then cloned into the XhoI site of each of the two BPV vectors. These vectors were then transformed into E. coli strain MC1061 and grown in bulk culture. The DNA was purified by CsCl banding before transfection into mammalian cells. The 5' to 3' orientation of the Protein S insert within the expression vectors was checked by numerous sets of restriction enzyme digestions. The final vector constructs are shown in FIG. 9. As a matter of convenience, the names of the expression vectors have been shortened to CSM and CSS for CL28XhoBPVProS and CLH3axBPVProS, respectively.

On 5 separate days, two sets of transfections into mouse C127 cells were carried out as follows:

Mouse C127 cells (commercially available) were maintained in Dulbecco's modified Eagle's medium (DME) supplemented with 10% fetal calf serum and 10 mM glutamine as described in Hsiung et al., id. DNA transfections were carried out by the method described in Wilger et al. 1977, Cell 11:233, as modified by Hsiung et al., id. Ten to twenty micrograms of calcium phosphate precipitated DNA was incubated for 6-8 hours at 37.degree. C. with 1.times.10.sup.6 cells in fresh culture medium. The medium was removed and the cells treated with 20% glycerol in 10 mM phosphate-buffered saline (PBS), pH 7.0, for 1-2 minutes at room temperature, washed twice with PBS and fresh DME added. The cells (designated CL28X) were then incubated at 37.degree. C. and the medium replaced after 24 hours and every 3-4 days thereafter.

Ten days after transfection the culture medium was tested in an ELISA assay for human Protein S. In this assay a 1/10,000 dilution of rabbit anti-human Protein S antiserum was used to coat microtiter wells (Immulon), 200 .mu.l per well, overnight at room temperature. After washing the wells extensively with buffer containing PBS/Tween, the wells were "blocked" for 2 hours with 200 .mu.l of 2% bovine serum albumin (BSA) in PBS (phosphate buffered saline). After washing, the wells were challenged with 200 .mu.l of culture medium from the CSM transfected cells. The culture medium was incubated in the wells for four hours at room temperature. After washing, 200 .mu.l of goat anti-human Protein S was added to the wells and incubated overnight at room temperature. The wells were again washed, and 200 .mu.l of rabbit anti-goat IgG antibodies conjugated with horseradish peroxidase (HRP) was added. Incubation was for 2 hours at room temperature before washing and adding 200 .mu.l of the HRP substrate, orthophenylene diamine (OPD). The HRP reaction was terminated by the addition of 50 .mu.l of 3M H.sub.2 SO.sub.4. The microtiter wells were then read with a microtiter well reader set at 492 nm. The results, given in the table below, demonstrate that Protein S is produced by the transfected cells. (The transfected cells are isolates CL28XAlA.sup.(2), CL28XBlA.sup.(3), CL28XBlB.sup.(4), CL28XB2C.sup.(5), and CL28XB2A.sup.(6) in the table.).

Human Protein S produced in transfected cells as described above is purified from the cells and/or the culture medium according to conventional techniques, e.g., as described in Comp et al. (1984, J. Clin. Invest. 74, 2082). The human Protein S so produced, because it is derived from mammalian cells, is free of C4-binding proteins as well as bacterial and viral pathogens.

Human Protein S can be lyophilized and reconstituted in saline prior to use. Protein S in saline can be administered on a regular basis (e.g., weekly) to human patients congenitally deficient in Protein S, to inhibit the recurrent thrombosis seen in these patients. Administration is by the modes of administration used for other anti-thrombolytic agents, e.g., tissue plasminogen activator. Most preferably, administration is by bolus injection or by intravenous infusion. The following examples are illustrative.

For long-term maintainance of a human patient deficient in Protein S, 150 mg of lyophilized Protein S are dissolved in saline and placed in the chamber of a syringe, which is used to inject a bolus of Protein S into the patient intravenously; treatment is carried out once per week.

For infusion treatment for the rapid lysis of coronary thrombi, about 100 mg/hr of lyophilized t-PA, about 1.0 mg/hr of lyophilized Protein C, and about 2.0 mg/hr of lyophilized Protein S are dissolved in saline and infused at the same time, together or via separate infusion lines, intravenously over a period of about 3 hours.

For infusion treatment for the slow lysis of deep vein thrombi, about 10 mg/hr of lyophilized t-PA, about 1.0 mg/hr of lyophilized Protein C, and about 2.0 mg/hr of lyophilized Protein S dissolved in saline are infused, at the same time, together or via separate infusion lines, intravenously over a period of about 12-24 hours.

Antibodies, either polyclonal or monoclonal, can be raised to Protein S and used in conventional immunoassay methods to quantify Protein S in biological fluids, e.g., serum or urine, of patients suspected of having a protein S deficiency.

E. coli cells containing the plasmid CSM have been deposited with the American Type Culture Collection, Rockville, Md., and assigned ATCC Accession Number 67163.

Applicants' assignee, Integrated Genetics, Inc., acknowledges its responsibility to replace this culture should it die before the end of the term of a patent issued hereon, 5 years after the last request for a culture, or 30 years from the date of deposit, whichever is the longer, and its responsibility to notify the depository of the issuance of such a patent, at which time all restrictions on the availability to the public of the material so deposited will be irrevocably removed. Until that time the deposit will be made available to the Commissioner of Patents under the terms of 37CFR Section 1-14 and 35USC Section 112.

Other embodiments are within the following claims.