AU1180692A - Human protein c expression vector - Google Patents

Description

D E S C R I P T I O N

Human Protein C Expression Vector

Technical Field

The present invention relates to DNA fragments encoding human protein and to expression vectors of human protein C including them. The human protein C which is produced using the expression vector according to the present invention or the activated human protein C which is prepared by activa¬ tion thereof can be used as an anticoagulant or fibrinolytic accelerator. In the present invention, the sequence of DNAs is abbreviated with bases constituting individual deoxyribonu- cleotides and, for example, the bases are abbreviated as follows:

Aadenine (representing deoxyadenylic acid) Ccytosine (representing deoxycytidylic acid)

Gguanine (representing deoxyguanylic acid)

Tthy ine (representing deoxythymidylic acid)

Background Art Protein C is one of zymogens of plasma serine protease and activated through limited proteolysis with a complex of thrombin and thrombomodulin on the surface of blood plate¬ lets or endothelial cells of blood vessels to convert into serine protease, namely activated protein C, (abbreviated to APC hereinafter).

APC exhibits its anticoagulative action by selective hydrolysis of the activated factor V and activated factor VIII in the blood coagulation system. This activity is known to be enhanced by protein S. Further, APC has been thought to get its fibrinolysis-accelerating activity by cleaving PAI-1, an inhibitor of tissue plasminogen activators.

The nucleotide sequence of the gene for human protein C was already described by Foster DC (Proc. Natl. Acad. Sci. USA, 82, 4673-4677 (1985) and the nucleotide sequence of human protein C cDNA was shown in Nucleic Acids Res., 13, 5233-5247, (1985).

The amino acid sequence of human protein C is composed of, as shown evidently in Proc. Natl. Acad. Sci. USA, 82, 4673-4677 (1985), the light chain (molecular weight: about 21,000) having Gla domain and epidermal growth factor (EGF)- like domain on the amino terminus and the heavy chain (molecular weight: about 41,000) comprising the activation peptide and catalytic domain where both of the chains are disulfide-bonded (two chain type).

The light chain of human protein C comprises 153 amino acids from Ala as the amino terminal to Leu as the carboxyl terminal, while the heavy chain includes 262 amino .acids from Asp as the amino terminal to Pro as the carboxyl termi- nal, and they are biosynthesized in the form of (H2N-)-light chain-Lys-Arg-heavy chain-(-C00H) in the cells and Lys-Arg is cleaved into two strands while it is secreted from the cells.

The light chain and the heavy chain are linked through a disulfide bond at 141 Cys from the amino terminal of the light chain and 120 Cys from the amino terminal of the heavy chain.

Disclosure of Invention An object of the present invention is to provide DNA fragments which are useful for expression of human protein C, the expression vectors using said fragments and the mammalian cells transformed with said expression vectors. Another object of the present invention is to provide human protein C-encoding DNA fragments of high expression efficiency.

These objects can be achieved by DNA fragments which are prepared by removing n introns (where n = 1 to 6) from the seven introns between eight exons in the protein C gene and ligating the remaining before and behind sequences.

The present invention is DNA fragments encoding human protein C which are prepared by removing n (n =_ 1 to 6) introns from 7 introns distributing between 8 exons in human protein C gene and ligating the remaining before and behind sequences.

The above-stated DNA fragment which is prepared by removing 2 introns between exons 6 and 7 and between exons 7 and 8 in human protein C gene is preferred. Further, the above-stated DNA fragment which is prepared by removing 5 introns between exons 3 and 4, exons 4 and 5, exons 5 and 6, exons 6 and 7, and exons 7 and 8 is preferred. The present invention also includes a DNA fragment encoding human protein C that is constructed by incorporat¬ ing at least one of the following DNA fragments into any one intron, a plurality of introns or partially deleted introns among n (n = 1 to 7) introns included in human protein C gene or the DNA fragments mentioned above: a. a DNA fragment which accelerates transcription activity in mammalian cells. b. a DNA fragment which can be a replication origin of DNA in mammalian cells and c. a DNA fragment of adnovirus VA gene.

Moreover, the present invention is an expression vector which can be transduced into mammalian cells and contains a promoter, the above-described DNA fragment and, when needed, the polyadenylation signal wehre the transcription of the above-stated DNA fragment is commanded by the promoter.

The present invention further includes human protein C- producing mammalian cells which has been transformed by the expression vectors. The cells are suitably selected from the group consisting of BHK cells, CHO cells, HeLa cells, C127 cells and 293 cells.

The cells to be transformed may undergo other transfor¬ mations with other expression vectors, before or after transformation with the expression vectors according to the present invention. For example, after transduction of human protein C expression vector, pKEX2-gpt or pVA-gpt to be stated later can be transduced.

The human protein C gene according to the present invention has almost the same base sequence as the sequence of 8975 bases from 1st to 8975th described in Fig. 4 of Japanese Patent Specification Laid-open No. Tokkaisho 62- 111690 (Proc. Natl. Acad. Sci. USA., 82, 4674 (1985)), but partially differs. The gene comprises eight exons (called exon 1, 2, ...8, respectively) and seven introns (called intron 1, 2, ... 7, respectively) .

The exon 1 is the sequence of seventy bases from 1 A to 70 G in the base sequence shown in Fig. 4. The intron 1 has the same base sequence as the sequence of 1263 bases from 71 G to 1333 G given in Fig. 4 of the above-stated patent specification except that

(1) there is G between 127 G and 128 C;

(2) there is A between 532 G and 533 C; (3) there is C between 733 C and 734 A;

(4) 761 is not C but G;

(5)^' there are G and A in this order between 981 G and

982 C; (6) there is C between 1300 T and 1301 C; (7) there is C between 1318 C and 1319 T;

The exon 2 is the sequence of 167 bases from 1334 A to 1500 A in the base sequence shown in Fig. 4.

The intron 2 has the same base sequence as the sequence of 1462 bases from 1501 G to 2962 G given in Fig. 4 of the above-stated patent specification except that

(1) there is C between 1739 C and 1740 A;

(2) there is G between 2141 G and 2142 C;

(3) there is G between 2398 G and 2399 C; (4) there is C between 2672 G and 2673 C;

(5) there is no 2927 G.

The exon 3 is the sequence of 25 bases from 2963 C to 2987 G in the base sequence shown in Fig. 4.

The intron 3 is the sequence of 92 bases from 2988 G to 3079 G.

The exon 4 is the sequence of 138 bases from 3080 A to 3217 G in the base sequence shown in Fig. 4.

The intron 4 is the sequence of 102 bases from 3218 G to 3319 G. The exon 5 is the sequence of 135 bases from 3320 A to 3454 G except that 3342 T is replaced with G in the base se¬ quence shown in Fig. 4.

The intron 5 is the sequence of 2668 bases from 3455 G to 6122 G. The exon 6 is the sequence of 143 bases from 6123 T to 6265 G in the base sequence shown in Fig. 4.

The intron 6 is the sequence of 873 bases from 6266 G to 7138 G.

The exon 7 is the sequence of 118 bases from 7139 G to 7256 G in the base sequence shown in Fig. 4.

The intron 7 is the sequence of 1129 bases from 7257 G to 8385 G.

The exon 8 is the sequence of 590 bases from 8386 G to 8975 G in the base sequence shown in Fig. 4. The DNA sequence according to the present Invention can be built up from the gene for human protein C.

The removal of an arbitrary intron can be attained by effecting deletion mutagenesis in the intron region employ- ing the site-directed mutagenesis or the so-called cassette mutagenesis techniques.

Further, it can be attained, as will be shown in the examples, by substituting a part of the human protei C genomic DNA with the corresponding region of its cDNA uti- lizing restriction sites in common.

As an example of the DNA sequence according to the present invention, can be cited, as will be given in exam¬ ples, the sequence in which two introns between exons 6 and 7 , and between exons 7 and 8 are deleted, respectively or the sequence in which five introns are deleted between exons 3 and 4, 4 and 5, 5 and 6, 6 and 7, and 7 and 8, respective¬ ly.

In general, the partial mutation of intron base se¬ quence causes no mutation at the protein level and the possibility is low that the bodies having such mutation are excluded in comparison with the mutation in the protein- encoding regions. Therefore, it is thought that there are a few differences in the intron sequences between human races, individual bodies and so on. The DNA fragments according to the present invention are permissible in such partial dif¬ ferences of intron sequences and must not be limited to the above-described.intron sequences.

The DNA fragments encoding human protein C according to the present invention includes at least one intron. This intron can include DNA fragments which exert some actions or effects in mammalian cell nuclei. The present invention covers such recombinant DNA fragments, too.

The DNA fragments having some actions and effects in mammalian cell nuclei mean, for example, enhancer: the DXA fragments accelerating transcription activity, replication origin from an animal virus or adenovirus VA gene.

These incorporations, however, must not exert any influence on the DNA sequence acting as the splicing signal in the intron. Accordingly, such incorporation must be done in the other regions than the splicing signal sequence. Additionally, the DNA fragments to be incorporated must not have the sequence functioning as a splice acceptor.

In the meantime, the partial removal of the sequence can be permitted for convenient recombination before the incorporation, unless the sequence is for splicing signal in the intron. The simultaneous incorporation of DNA fragments which have no special funtion is also permitted.

This DNA fragment incorporation is done by utilizing appropriate restriction sites in the intron and further, by creating suitable restriction sites through the site- directed mutagenesis, when needed. Further, the incorpora¬ tion can be achieved by general recombination technology, for example, utilizing a snthetic DNA as an adaptor. As a DNA fragment accelerating transcription activity in mammalian cells, is cited, for example, enhancers in animal viruses such as SV40, human cytomegalovirus, BK virus, EB virus, herpes simplex virus, or enhancers included in animal genes such as antibody genes. The incorporation is not always limited only to the terminology "enhancer" but any DNA fragments having such activities can be utilized in the present invention. For example, the DNA fragment in HTLV-1 LTR responding to P40x of HTLV-1 to accelerate the transcription activity, the DNA fragment in the promoter region of metallothionein-1 gene, responding to heavy metal ions or glucocorticoid hormones to promote the transcription activity or the like are cited.

As the DNA fragments which can be a DNA replication origin in mammalian cells, are cited, for example, the replication origins in the DNAs of animal viruses such as SV40 or BK virus.

There are two genes, VAI and VAII in the adenovirus VA gene and both of them or either of them can be used. It has been thought that the VA gene is transcribed into VA RNA, which is presumed to increase the expression efficiency.

In order to obtain human protein C using the DNA frag¬ ments according to the present invention, the DNA fragments are incorporated into an expression vector which is suitably selected depending on the host vector system. Such expres¬ sion vectors are, for example, vectors which are prepared by incorporating suitable controlling regions into plas ids originating from bacteria, DNA originating from bacterio- phage, or DNAs of animal virus. Such controlling regions are selected from, for example, adenovirus major late promoter, SV40 early promoter, SV40 late promoter, mouse metallothio- nein-I promoter, MMTV (mouse mammary tumor virus) promoter, RSV (Rous sarcoma virus) promoter, human β-actin promoter, chicken β-actin promoter. The vectors originating from viruses such as bovine papilloma virus are also preferable. As a polyadenylation signal according to the present invention, may be used any signal as long as it works in mammalian cells. Accordingly it may originate from animal viruses or a polyadenylation signal from animal genes (class 2). For example, the polyadenylation signal existing in the early gene region of SV40 can be cited. Or the signal in human protein C gene itself may be used.

The integration of the above-stated DNA fragments into such vectors can be carried out by a known method per se, for example, the method described by Miura Q. et al., J. Clin. Invest., 83, 1958-1604 (1989) and others.

The mammalian cells for the host may be either human cells or cells of any animals other than human and, for example, CHO, C127, BHK, Cosl, Cos7, LM, NIH3T3, 293, HeLa or the like may be cited. Particularly, CHO, BHK and 293 are preferred.

The transfection of the above-stated expression vector into these cells also can be effected by the well-known methods per se in the prior art, for example, described in the following literature: Spandidos D. A. and Wilkie N. M. , Expression of Exogenous DNA in Mammalian Cells, Ed. Hames B. D. and Higgins S. J., Transcription and Translation, IRL Press Oxford pp 1 - 48 and others. The transformed cells are cultured by a usual method under conditions suitable for the cells so that the target protein can be recovered from the culture supernatant.

The concentration of human protein C obtained according to the present invention can be determined by the method disclosed in Japanese Patent Specification Laid-open No. Tokkaisho 61-283868. This method can also determine only human protein C having Gla.

The barium adsorption method and the ion-exchange chromatography can be utilized for purification of the protein C produced according to the present invention, but the affinity column chromatography using an anti-human protein C monoclonal antibody which recognizes the conforma¬ tion change in the Gla domain caused by calcium ion is particularly preferably used (see Japanese Patent Specifica- tion Laid-open No. Tokkaisho 64-85091).

The method is an excellent process because it can pu- rifiy only human protein C having Gla and can use EDTA, a mild eluent.

Human protein C is converted into APC by removal of the activation peptide ranging from the 12th amino acid to the amino terminus (abbreviated to N-terminus hereinafter) from the heavy chain by the protein C activation enzyme.

The protein C activation enzyme is, for example, throm- bin, thrombin-thrombomodulin complex, snake venom or the like.

The use of the expression vector of human protein C solves the problems on contamination with viruses which may cause troubles. Moreover, stabilized supply can be achieved 5 because it becomes unnecessary to demand human blood as a starting substance.

In general, the expression of cDNA in mammalian cells is effected by artificial addition of the splice donor and the acceptor to the noncodlng region on the 5' or 3'-side, 10 but it becomes unnecessary, when the DNA fragments according to the present invention are used.

So far, a variety of modified human protein C have been prepared by partial modification of the sequence of cDNA in the system for expression of human protein C cDNA in mamma¬ ls lian cells.

For example, a cDNA modified at the connecting peptide [Lys (156) Arg (157)] region (see Japanese Patent Laid-open No. Tokkaisho 64-85084) is cited for the purpose of in¬ creased two chain processing efficiency, another cDNA where 20 the activation peptide region is deleted for direct expres¬ sion of activated protein C, and one where the neighborhood is further modified (see Japanese Patent Specification Laid- open No. Tokkaihei 2-2338).

As these modified cDNAs correspond to partial modifica- 25 tion of exon 6 sequence, the part of exon 6 in the DNA fragment according to the present invention is replaced with a DNA fragment modified so as to correspond to the above- stated modification instead of the naturally occurring DNA fragment so that the same modified human protein C as these 30 proteins can be obtained. In other words, direct expression of human protein C of increased two chain processing effi ciency and of activated human protein C can be realized.

Additionally, modified human protein C which are pro¬ duced by other modifications in human protein C cDNA than above-stated examples also can be prepared by using the correspondingly modified DNA fragments in the DNA fragments according to the present invention.

Each gene in mammalian cells is not always expressed constantly at a certain level. Some genes are expressed only in the cells differentiated in a specific tissue, while some other genes are expressed only at a specific differentiation stage or only before the start of differentiation.

Further, some genes change the expression depending on the outside environments. For example, the expression is increased or decreased depending on nutrient conditions such as glucose concentration, temperature, metal ion concentra¬ tion or signal transduction substances in the living body. A living body copes with the surrounding environments by con- trolling the gene expression to maintain its life activity. Multistage fine mechanisms participates in such control of gene expression In mammalian cells. One of them is the control at the transcription stage. The transcription is controlled through mutual action between the promoter/ enhancer region of the gene and the DNA-binding proteins in the nuclei.

In this case, the kinds and amounts of DNA-binding proteins vary reflecting the outside environments and the direction and degree of differentiation. Further, the ex- pression is also controlled by varying the half-life of mRNA. For example, it has been known that the presence of estrogen largely prolongs the half-life of some kind of mRNA.

In addition, the expression is presumably controlled a" the stages of mRNA transport from the nucleus to the cyto¬ plasm, translation and intracellular transport of the formed protein, too.

By the way, almost all genes of mammalian cells have introns. The sequences are spliced out in the nuclei after transcription, and this stage is also thought to give mamma¬ lian cells the chance to control the gene expression.

Thus, it is more advantageous than the expression with cDNA to use genomic DNA in the production of a substance in mammalian cells because the expression can be controlled through the modification of the splicing efficiency or the transporting efficiency from the nuclei into the cytoplasm of the participating mRNA.

The control of expression efficiency can be utilized so that the protein production is inhibited and the protein- synthesizing mechanism is directed to other functions such as cell division at the stage in no need of the protein synthesis, for example, a cell number-increasing step, while the mechanism is directed to the protein synthesis, when becomes necessary. In general, however, the expression efficiency of genomic DNA is lower than that of cDNA in many cases. The DNA fragments according to the present invention has enabled extracellular control of the expression effi¬ ciency to secure a practical level of the efficiency. The DNA fragments according to the present invention is also to provide "a place where the sequence relating to transcription control is to be put". The DNA sequence con¬ trolling the transcription is placed near the promoter to develop its effect and it has been known that the site is on the upper stream of the promoter, but in some cases, it is effective even on the down stream of the transcription initiation point. An antibody enhancer is cited as an exam¬ ple.

In the case of the expression with cDNA, the primary structure of protein is not changed and such DNA sequence needs to be placed between the transcription initiation site and the translation initiation point. In this case, however, there is the possibility of decreased translation efficiency because the distance between the transcription initiation site and the translation initiation site is long.

On the contrary, the DNA fragments according to the present invention do not change the primary structure of protein even when modified by partial substitution or incop- oration as long as modification is done in introns and there is no problem even when a transcription-controlling sequence such as an enhancer is incorporated.

In this case, the sequence from the transcription starting point to the translation-starting point can be made optimal length and sequence order for translation efficien¬ cy.

Meanwhile, similar things are also possible in the case of genomic DNA, but low expression efficiency is an inherent problem, as stated above. When the DNA fragments according to the present invention is used, the sequence of an enhanc¬ er or the like can be integrated, as a practical level of expression efficiency is secured.

The similar conception of a place where the DNA se¬ quence is to be placed also can be applied to incorporation of the replication origin of polyoma viruses such as SV40, or VA gene of adenovirus. The VA gene is a sequence which can increase the protein production, when the major late promoter of adenovirus is used as a promoter.

An expression vector which is prepared by incorporating the sequence into an intron of the DNA fragments according to the present invention is transfected into mammalian cells and screened using the expression of the target protein as an indication. At this time, the cells expressing the target protein can be regarded almost always as to include the VA gene in the chromosome DNA .

On the contrary, in the case of an expression vector where the VA gene is incorporated into the outside of the protein transcription unit, the cells producing the target protein do not always include the VA gene, even when the gene is on the same vector. It is why a complicated recombi¬ nation may happen during the integration of the expression vector into the chromosome DNA in some cases.

As a matter of course, it cannot be accurately concluded that VA gene is incorporated into the chromosome of the protin-expressing cells on the reason that the target pro¬ tein has been expressed, in the case of cotransfection of an expression vector having the transcription unit of the target protein and a plasmid having VA gene, too. By the way, the judgment on whether VA gene is incorpo¬ rated into a cell or not can be made by the Southern hybri¬ dization method or the like. However, in the expression vectors prepared by incorporation of the VA gene into the introns of the DNA fragments according to the present inven- tion, the judgment can be made by examining the expression of the target protein as an indication, and so a method which can simply investigate a plurality of specimens simul¬ taneously such as ELISA can be advantageously used.

Particularly, when the target protein is secretory one, the Southern hybridization method always requires the de¬ struction of the cells for extraction of the DNA, while only collection of the supernatant of cell culture mixture is required enough for the method.

It is a matter of course that VA RNA is formed by tran- scription even in the case where the VA gene integrated in intron is incorporated into the chromosome DNA of the cells, and it is independent from splicing of the mRNA encoding the target protein.

It can be also applied to the case of incorporation of replication origin into the introns of the DNA fragments according to the present invention. For example, when an ex¬ pression vector including a DNA fragment as a replication origin of SV40 is integrated into the chromosome DNA of mammalian cells and T-antigen of SV40 is expressed, the replication of DNA can be started from the replication origin, thus the copies of the transcription units of the target protein nearby are increased, resulting in increase of the expression. In the case, when an expression vector prepared by incorporation of such a DNA fragment as a replication origin into the introns of the DNA fragments according to the present invention is used, it can be concluded that the replaction origin has been also incoporated into the chro- mosome DNA of mammalian cells on the reason that the target protein has been expressed, as in the case of VA gene.

VA gene can work so long as it is incorporated into an arbitrary place in the chromosome of mammalian cells, but the replication origin requires the location near the tran- scription unit (a set of DNA sequence series of promoter region, protein-encoding region, polyA addition signal and so on) .

But, the Southern hybridization method or the like gives only information that the replication origin is some- where in the chromosome of mammalian cells. On the contrary, the use of the DNA fragments according to the present inven¬ tion substantially always ensures the location of the repli¬ cation origin in the transcription unit of the target cells, every when the target protein has been expressed. In addition, use of the DNA fragments according to the present invention enables the replication origin to locate near the center of the transcription unit of the target protein and more efficient gene amplification can be expect¬ ed. By the way, even when a replication origin in introns is incorporated into the chromosome DNA of mammalian cells, it functions as the replication origin in the chromosome, as a matter of course, and it is indifferent from splicing of the mRNA encoding the target protein.

The above-described effects can be also expected in the case where VA gene of adenovirus or the replication origin is incorporated into the introns of genomic DNA, but there is a problem of low expression efficiency of the target protein. But, the application of DNA fragments according to present invention makes the above-stated things possible as ensuring a practical level of expression.

Example

The present invention willed illustrated in more detail with the following examples, but the present invention is not limited to them at all.

Following methods are used in the examples according to present invention.

Cleavage of DNA

The cleavage of 1 it g of plasmid DNA or M13 phage repli¬ cative form (RF) DNA or DNA fragments was carried out in 10 1 of buffer solution using 4 to 10 units of a restriction enzyme and keeping them at the temperature indicated by the maker for 2 hours. The buffer solution was accompanied by the restriction enzyme.

Separation and Collection of DNA fragments

The DNA fragments cleaved with the restriction enzyme were separated by agarose gel electrophoresis with a subma¬ rine type electrophoresis system. The agarose gel containing the target DNA fragments is cut out and the fragments were recovered with GENECLEAN II (Registered Trade Mark) (Bio 101 Co.). The procedure followed the manual appended to the reagent. Unless otherwise noted, 0.8 % agarose gel was used.

Ligation of DNA fragments

It was carried out with the DNA ligation kit (TAKARA SHUZ0 CO. LTD.). The procedure followed the manuals appended to the kit .

Blunting of DNA fragments

It was effected using a blunting kit (TAKARA SHUZO CO. LTD.). The procedure followed the manual appended.

Transformation of Escherichea coli

Less than 20 μ 1 of DNA solution was added to the compe¬ tent cells of E. coli HB101 strain (TAKARA SHUZO Co. Ltd.), placed on ice for 1 hour, and dipped in the water bath at 42' C for 1 minute, and placed on ice, again for 5 minutes. The mixture was added to 1 ml of L-broth and subjected to shake culture for 1 hour, then a part (50 to 300 μ 1 ) of the culture mixture was spread on an ampicillin plate (L-broth, agar 15 g/1, ampicillin 50 μ g/ml) and cultured at 37' C one overnight to form colonies.

Small-scale preparation of plasmid DNA

The alkaline lysis method was employed for preparation (see [Molecular Cloning] T. Manuals, Cold Spring Harbor

Laboratory (1982), P 368). When needed, further purification was effected with GENECLEAN II (Registered Trade Mark) .

Large-scale preparation of plasmid DNA The alkaline lysis method (see [Transcription and Translation] B. D. Hames, IRL press, (1984) P 8) and the CsCl equilibrium density-gradient centrifugation were used. The rotor for ultracentrifugation was Hitachi's RP-67VF vertical rotor. CsCl was removed not by dialysis but 4-tlme dilution with TE (10 mM Tris-HCl pH 8.0 1 mM EDTA) followed by twice precipitation with ethanol.

Determination of nucleotide sequence

A single-stranded DNA was obtained by incorporating the DNA fragment to be determined into bacteriophage M13.

The procedure is described in the manual appended to "M13 cloning kit" made by Amasham Co. Ltd. A large amount of single-stranded DNA was prepared by the method described in the manual appended to "oligonucleotide-directed in vitro mutagenesis system" (Amasham Co.). The resultant single- stranded DNA was used as a temperate and a synthetic 20mer DNA near the region to be examined whose synthetic method will be explained later was used as a primer to carry out the sequencing reaction using Dye Deoxy™ Terminator Taq Se¬ quencing kit (Applied Biosystems Co. Ltd.).

The sample obtained by the reaction was analyzed with Applied Biosystem's 373A DNA sequencer to determine the nucleotide sequence.

Chemical synthesis and purification of DNA fragments

The synthesis was effected using the Applied Biosys¬ tem's 380A DNA synthesizer under the "Tr ON, Auto condi¬ tions. The products were purified using "oligonucleotide purification cartridge" of Applied Biosystem in accordance with the appended manual.

On the preparation of the DNA fragments and the expres¬ sion vectors according to the present invention, following plasmids and DNA fragments were used.

pDX/PC

This is a human protein C expression vector having a full-length protein C cDNA under the control of the adenovi- rus type 2 major late promoter, which is described by Fos¬ ter, D.C. et al., Biochemistry 2_6, 7003-7011 (1987) or in Japanese Patent Specification Laid-open No. Tokkaisho 62- 11690.

The functional part constituting the expression vector is described by Busby S. et al., Nature, 316, 271-273 (1985) and Berkner, K.L. et al., Nuc. Acids Res., 13, 841-857 (1985) .

Human protein C genomic DNA

Human protein C genomic DNA was disclosed in Japanese Patent Specification Laid-open No. Tokkaisho 62-111690.

In the following examples, the DNA fragment spanning from the EcoRI site in the intron between exon 2 and exon 3 to the EcoRI site which is located at 4.4 kbp upstream of it (called R4.4-1 hereinafter), and the 6.2 kbp DNA fragment spanning from the EcoRI site in the intron between exon 2 and exon 3 to the EcoRI site in the intron between exon 7 and exon 8 (called R2-14 hereinafter) were used after sub- cloning them into the EcoRI site of PUC9.

Zem228 and Zem229

These expression vectors are disclosed in Japanese Patent Specification Laid-open No. Tokkaihei 2-2338. In these vectors, the promoter of mouse metallothionein I gene is used and neomycin resistant gene or dihydrofolate reduc- tase gene is used as a selection marker.

pSV2-dhfr It is registered in American Type Culture Collections (ATCC) as No. 37146.

228/PC594

It is a human protein C expression vector which is prepared by incorporating human protein C cDNA into Bamlll site of the above-stated Zem228. The human protein C cDNA is identical with that in the above-stated PDX/PC.

229/PC962 It is a modified human protein C expression vector which is prepared by incorporating a modified human protein C cDNA PC962 into the BamHI site of the above-stated Zem229, The PC962 is disclosed in Japanese Patent Specification Laid-open No. Tokkaisho 64-85084.

Rc/CMV

It is commercially available from Invitrogen Co..

KEX2

The KEX2 gene is a region shown as KEX2 in a plasmid which is disclosed in Japanese Patent Specification Laid- open No. Tokkaihei 2-2338.

pSV2-gpt

It is registered in ATCC as No. 37145.

Example 1

Preparation of human protein C expression vector

(1) Preparation of 228/AC-PC9001

In order to destroy the Ball site in the neomycin resistant gene in 228/PC594, 228/PC594 was digested with AatI and Clal and the resultant DNA fragments were blunted at their ends.

The larger fragment was separated, subjected to intra¬ molecular ligation, then transduced into E. coli HB 101 to increase the amount of the plasmid. Then, the plasmid was digested with Ball and SacII (both of them have the cleavage sites only in protein C cDNA) and the larger Ball-SacII fragment was recovered.

Two plasmids which were prepared by subcloning the above-stated R4.4-1 and R2-14 in PUC9 were digested with Ball and PstI and EcoRI, and SacII and EcoRI, respectively to collect the Ball-EcoRI fragment of about 1.65 kbp and the EcoRI-SacII fragment of about 4.5 kbp. These two DNA fragments and the above-stated Ball-SacII fragment were ligated termolecularly to form 228/AC-PC9001. Thus, the expression vector has human protein C gene from which two introns between exons 6 and 7 and exons 7 and 8 were removed. The expression vector lose most of neomycin resistant gene.

(2) Preparation of 229-PC9002

The above-stated plasmid resulting from subcloning R4.4-1 in PUC9 was digested with Ball, EcoRI and PstI to recover a Ball-EcoRI DNA fragment of about 1.65 kbp (A fragment)

Meanwhile, the above-stated plasmid resulting from subcloning R2-14 in PUC9 was digested with Xcyl, dephospho- rylated with alkaline phosphatase, further digested with EcoRI to recover an EcoRI-Xcyl DNA fragment of about 1.22 kbp (B fragment) .

Subsequently, the A fragment and the B fragment were ligated and digested with Ball to recover a Ball-Xcyl DNA fragment of about 2.87 kbp, then the fragment was phosphory- lated with T4 polynucleotide kinase (TAKARA SHUZO Co. Ltd. ) (C ragment) .

The DNA fragment from Xcyl (No.2902 residue) to BstEII (No.3082 residue) in the human protein C gene where the intron between exon 3 and exon 4 had been removed was chemi- cally synthesized. The sequence was disclosed in the above- stated speci ication of Japanese Patent Laid-open No. Tok¬ kaisho 62-111690 (or Proc. Natl. Acad. Sci. USA, 82, 4674 (1985)).

After only the antisense strand was phosphorylated with T4 polynucleotide kinase, both strands were annealed (D fragment) . 228/PC594 was digested with Sac II and BstEII and a BstEII-SacII DNA fragment of about 0.36 kbp was separated by means of 2% agarose-gel electrophoresis (E fragment) . The ligation between D fragment and E fragment was followed by digestion with SacII to recover a Xcyl-SacII DNA fragment of about 0.45 kbp through the 2 % agarose gel electrophoresis. Then, the fragment was phosphorylated with T4 polynucleotide kinase (F ragment) . Further, 229/PC962 was digested with Ball and SacII and the larger fragment was collected (G fragment) . Finally, the C fragment, the F fragment and the G fragment were termolec- ularly ligated to form 229-PC9002. This vector contains a human protein C gene where other introns than two introns between exon 1 and exon 2, and between exon 2 and exon 3 are removed under the control of the mouse metallothionein 1 promoter and has the transcription unit of dhfr gene on the same plasmid.

(3) Preparation of TZml-PC9002

Zem228 has two EcoRI sites and the partial digestion with EcoRI and agarose gel electrophoresis gave the only one site-digested product.

The DNA fragment was blunted and subjected to self-liga- tion. The product was transduced into E. coli, proliferated, recovered, then completely digested with EcoRI, blunted again, further self-ligated, and transduced into E. coli to proliferate the plasmid. Thus, two EcoRI sites were deleted. Then, a short double-stranded synthetic DNA having BamHI cleavage terminals on both ends and EcoRI site inside was incorporated into unique BamHI site in the plasmid to prepare an expression vector Zem228R having the EcoRI cleav¬ age site immediately downstream of the mouse metallothionein 1 promoter. The Zem228R was digested with EcoRI and Hindlll and subjected to agarose-gel electrophoresis to collect the large fragment. Further, the Hindlll-EcoRI fragment of 1.03 kbp was excised from the above-stated pDX/PC and both of fragments were ligated to give ZmB3. This expression vector has the transcription unit of neomycin resistant gene as well as adenovirus major late promoter.

The ZmB3 was digested with EcoRI and Clal to separate the large fragment having neomycin resistant gene (H frag- ment). Further, 229-PC9002 was digested with Ball and Clal to recover the larger fragment (I fragment).

229/PC962 was digested with EcoRI and Ball and subject¬ ed to agarose-gel electrophoresis to recover the smaller fragment (J fragment). The H fragment, the I fragment and the J fragment were termolecularly ligated to give TZml- PC9002.

This expression vector contains a human protein C gene where other than two introns between exons 1 and 2, and exons 2 and 3, respectively, are removed under the control of the adenovirus major late promoter and has the transcrip¬ tion unit of neomycin resistant gene on the same plasmid.

(4) Preparatio of 228-PC9001

228-PC9001 was obtained by termolecular ligation be- tween the smaller fragment resulting from digestion of

228/AC-PC9001 with Kpnl and SacII, the smaller fragment by digestion of 229-PC9002 with Sfil and Kpnl and the larger fragment from digestion of 228/PC594 with Sfil and SacII.

The expression vector expresses a human protein C gene where two introns between exons 6 and 7, and exons 7 and 8 were removed by the mouse metallothionein I promoter and has the transcription unit of neomycin resistant gene on the same plasmid. (5) Preparation of 228-PC9002

228-PC9002 was formed by ligation between the smaller fragment from digestion of 229-PC9002 with Sfil and SacII and the larger fragment from digestion of 228/PC594 with Sfil and Sac II.

This expression vector expresses a human protein C gene where other than two introns between exons 1 and 2, and exons 2 and 3 were removed by the mouse metallothionein 1 promoter, and has the transcription unit of neomycin resist- ant gene on the same plasmid.

(6) Preparation of TZm4-PC9002

TZm4-PC9002 corresponds to the sequence of the TZml- PC9002 from which the sequence from the top of the second branch of the tripartite leader sequence of adnovlrus to the 37th residue upstream of the translation initiation site of human protein C is deleted. The upstream sequence of human protein C initiation codon is shown by Beckmann et al., Nuc. Acids Res., 13, 5233 (1985). In other words, TZm4-PC9002 has only the first branch of the tripartite leader sequence and lacks the sequences of the second and third branches, the sequence containing the splice donor and the splice acceptor on the downstream of them, and a part of 5'-noncoding region of human protein C cDNA.

This deletion mutagenesis aims at increased translation efficiency by shortening the distance from the transcription start site to the translation start site and increased expression efficiency of human protein .C by removal of introns for reducing the splicing load.

Since the DNA sequence encoding human protein C accord¬ ing to the present invention includes introns, it should be noted that there is in no need of new intron incorporation into the untranslated regions. The deletion mutagenesis was effected by the so-called cassette mutagenesis which will be stated in the following:

TZml-PC9002 was digested with SacII, then PmaCI to separate the larger DNA fragment (K fragment). The double- stranded DNA fragment where the part to be deleted was removed from the sequence from the PmaCI site in the ade¬ novirus major late promoter to the Ball site in the DNA sequence encoding human protein C in TZml-PC9002 was chemi¬ cally synthesized. The synthesis was carried out in three blocks. These fragments were phosphorylated at the 5' terminals, annealed and ligated to give the desired DNA fragment.

The upstream terminal of this fragment had PmaCI cleaved end, and the downstream terminal had 6 base Ball recognition site followed by two base spacer (G, C) and SacII cleaved end.

The synthetic PmaCI-SacII fragment and the K fragment were ligated and the product was transduced into E. coli HB101. The plasmid obtained was digested with Ball and the produced smaller fragment was recovered (L fragment).

The two Ball sites on this plasmid are in the neomycin resistant gene and in the synthetic DNA. Then, TZml-PC9002 was digested with Ball, dephosphorylated with alkaline phosphatase and the larger DNA fragment was recovered and ligated with L-fragment to give TZm4-PC9002.

(7) Preparation of TZm5-PC9002

TZml-PC9002 was digested with Ball, Xbal and Clal to prepare a DNA fragment of 4.66 kbp (M fragment). A DNA fragment ranging from the Ball site at about 30 bp down¬ stream of the initiation codon to the 103 bp upstream of Ball site in human protein C was synthesized in two blocks with Hindlll cleaved end on the upstream terminus. Four strands of synthetic DNA were phosphorylated on the 5' terminals, annealed and ligated. The product was digested with Ball and Hindlll and subjected to 2% agarose-gel elec¬ trophoresis to collect a DNA fragment of 0.11 kbp (N frag¬ ment) . The M fragment and the N fragment were termolecularly ligated with Hindlll-Xbal large fragment of Rc/CMV to give TZm5-PC9002. This is an expression vector whose transcrip¬ tion is driven by cytomegalovirus IE enhancer/promoter.

(8) Preparation of TZm9-PC9002 TZm5-PC9002 was digested with Hindlll and Mlul, de- phosphorylated, and blunted (0 fragment). TZml-PC9002 was digested with Kpnl and Xhol and subjected to 2% agarose-gel electrophoresis to collect a DNA fragment of 0.42 kbp, which was blunted. The fragment was ligated with the 0 fragment and the product was analyzed by restriction enzyme cleavage to select the plasmid into which the adenovirus major late promoter was incorporated in such a direction that it can command the expression of human protein C as TZm9-PC9002.

(9) Preparation of TZml6-PC9002

TZm5-PC9002 was digested with Asp700 and Hindlll to collect the larger fragment (P fragment). Further, TZm9- PC9002 was digested with Nrul and Hindlll to collect the smaller fragment (Q fragment) . Rc/CMV was digested with BanI to collect a DNA fragment of 1.69 kbp, which was digested with Asp700 to collect the larger fragment. This DNA frag¬ ment, P fragment and fragment were termolecularly ligated to prepare TZml6-PC9002, an expression vector having human cytomegalovirus IE enhancer and Adenovirus major late pro- moter.

(10) Preparation of TZm20-PC9002

TZm5-PC9002 was digested with Sful, blunted, addition¬ ally digested with Sfil to recover the larger fragment. This DNA fragment was ligated with the smaller fragment which is obtained by digesting pSV2-gpt with BamHI, blunting the digestion product followed by additional digestion with Sfil to give TZm20-PC9002. This has the same transcription unit as in TZm5-PC9002, but has Eco-gpt gene as a selection marker.

(11) Preparation of TZm31-PC9002

A DNA fragment ranging from adenovirus tripartite leader leader sequence to the Ball cleavage site at about 30 bp downstream of the initiation codon of human protein C gene in TZml-PC9002 was chemically synthesized in four blocks with Hindlll cleaved end on the upstream terminus. Six syn¬ thetic DNAs other than two synthetic DNAs whose 5' terminal was to constitute both terminals of the DNA fragment were phosphorylated on their 5' terminals. After annealing of the complementary chain, ligation was effected and a DNA frag¬ ment of 0.53 kbp was recovered by 2 % agarose-gel electro¬ phoresis. The Hindlll-Ball fragment was to be called R frag- ment. TZm5-PC9002 was digested with Ball and Sfil to collect the smaller fragment (S fragment). Further, TZm5-PC9002 was digested with Hindlll and Sfil to collect a smaller frag¬ ment. The DNA fragment, the R fragment and the S fragment were termolecularly ligated to give TZm31-PC9002. This has the same basic structure as that of TZm5- PC9002, but includes the adenovirus tripartite leader se¬ quence in the 5' untranslated region.

(12) Preparation of TZm5-PC9004 Human adenovirus II DNA (about 36 kbp) was digested with Hindlll to recover a DNA fragment of 5.32 kbp, which was digested with Hpal to collect a DNA fragment of 1.33 kbp. The DNA fragment was subcloned into the Hindlll-HincII site in PUC8 to give PUC-VA. The PUC-VA was digested with EcoRI and Nrul to collect the smaller fragment (T-fra^'gment) . TZm5- PC9002 was digested with Nrul and Sfil to collect the small¬ er fragment (U fragment). Further, TZm5-PC9002 was digested with Sfil and EcoRI to collect the larger fragment. This DNA fragment, the T fragment and the U fragment were subjected to termolecular ligation to give TZm5-PC9004.

This has the same basic structure as that of TZm5- PC9002, but includes adenovirus VAI and VAII genes in the second intron of human protein C.

(13) Preparation of TZm31-PC9004

The smaller fragment which was prepared by digestion of TZm5-PC9002 with Hindlll and Sfil, the larger fragment from digestion of TZm5-PC9004 with Sfil and Bgl II, and the medium fragment from digestion of TZm31-PC9002 with Hindlll and Bglll were termolecularly ligated to give TZm31-PC9004.

This has the same basic structure as that of TZm31- PC9002, but Includes adenovirus VAI and VAII genes in the second intron of human protein C.

(14) Preparation of TZm5-PC9005

TZm5-PC9002 was digested with BstXI to recover the larger fragment, which was ligated with a single stranded synthetic adaptor DNA which converts the upstream terminal of the uppermost BstXI site in the first intron of human protein C gene into the pairing terminal of Mlul, then the terminals were phosphorylated using T4 polynucleotide ki¬ nase. The product was digested with Sfil to recover the larger fragment (V fragment) . TZm5-PC9002 was digested with Sfil and Seal to recover the larger fragment (W fragment). Rc/CMV was digested with BanI to collect a DNA fragment of 1.69 kbp, further digested with Mlul to give a DNA fragment of 0.49 kbp.

This DNA fragment, the V fragment and the W fragment were subjected to termolecular ligation to give TZm5-PC9005, an expression vector including IE enhancer of human cytome¬ galovirus in the first intron of human protein C gene.

(15) Preparation of TZm5-PC9006

Rc/CMV is digested with BanI to recover a DNA fragment of 1.69 kbp followed by digestion with Nrul to obtain a DNA fragment of 0.51 kbp (X fragment). Further, TZm5-PC9002 was digested with BstXI to recover the larger fragment, which was ligated with a single stranded synthetic adaptor DNA which converts the upstream terminal of the uppermost BstXI site in the first intron of human protein C gene into the pairing terminal with the BanI terminal of the X fragment, then the terminals were phosphorylated using T4 polynucleo- tide kinase. The product was digested with Sfil to recover the larger fragment (Y fragment) . TZm5-PC9002 was digested with Seal and Sfil to recover the larger fragment. This DNA fragment, the X fragment and the Y fragment were termolecu¬ larly ligated to give TZm5-PC9006, an expression vector including human cytomegalovirus IE enhancer in the first intron of human protein C gene. It is different from TZm5- PC9005 in the direction of the human cytomegalovirus IE enhancer.

(16) Preparation of TZml-PC9005 and TZml-PC9006

The larger fragment from digestion of TZm5-PC9005 with Tthllll was digested with Apal to recover the smaller frag¬ ment. This DNA fragment, the smaller fragment from digestion of TZml-PC9002 with Clal and Apal and the larger fragment from digestion of TZml-PC9002 with Tthllll and Clal were subjected to termolecular ligation to give TZml-PC9005. This has the same basic structure as that of TZml-PC9002 but includes human cytomegalovirus IE enhancer in the first intron of human protein C gene. TZm5-PC9005 was replaced with TZm5-PC9006 to give TZm5-PC9006.

(17) Preparation of TZm9-PC9005 and TZm9-PC9006

The larger fragments from digestion of TZm5-PC9005 and TZm5-PC9006 with Hindlll and Sfil were ligated with the smaller fragment from digestion of TZm9-PC9002 with Hindlll and Sfil, respectively, to give TZm9-PC9005 and TZm9-PC9006. They have the same basic structure as that of TZm9-PC9002, but include human cytomegalovirus IE enhancer in the first Intron of the human protein C gene.

(18) Preparation of 228-PC9005 and 228-PC9006

The medium fragment from digestion of TZm5-PC9005 or TZm5-PC9006 with Ball and SacII, the larger fragment from digestion of 228-PC9002 with SacII and Sfil and the smaller fragment from digestion of 228-PC9002 with Ball and Sfil were termolecularly ligated to give 228-PC9005 and 228- PC9006, respectively. They are equal to 228-PC9002 in their basic structure, but have human cytomegalovirus IE enhancer in the first intron of the human protein C gene.

(19) Preparation of pkEX2-gpt

KEX2/Zem228 was digested with BamHI to cut out the KEX2 gene and cloned into the BamHI site of the ZmB3. The frag- ment into which the gene is incorporated in the direction that the transcription of KEX2 gene is commanded by the adenovirus promoter was selected as ZmB3-KEX2. The ZmB3-KEX2 was digested with Kpnl and Aatll and the resultant smaller fragment was blunted (Z fragment). pSV2-gpt was digested with EcoRI, dephosphorylated and blunted. The DNA fragment was ligated with the Z fragment to give pKEX2-gpt.

(20) Preparation of pVA-gpt

The PUC-VA was digested with Smal and Nrul and the resultant smaller fragment was ligated with the fragment which was prepared by digestion of pSV2-gpt with EcoRI, followed by dephosphorylation and blunting to give pVA-gpt

Example 2

Expression of human protein C in mammalian cells

BHK-21 cells (ATCC CCL-10) or 293 cells (ATCC CRL1573) were cultured in a Falcon's 3003 petri dish containing 10 ml of the culture medium which is prepared by adding streptomy¬ cin and penicillin G to inactivated 10 % FCS-eRDF (Kyokuto Pharmaceuticals Co. Ltd.)-5 μ g/ml vitamin K-^ so that they reach 100 μ g/ml and 100 units/ml, respectively. The expression vector was used 10 μg each petri dish. Only in the case of 228/AC-PC9001, a mixture of 8 μ g of the expression vector and 2 μg of pSV2-dhfr was employed. The expression vector was mixed with 10 μg of salmon sperm DNA, 25 μ l of 2M CaCl₂, and was adjusted to 200 μ l with TE (ImM Tris • HC1 0.05 mM EDTA, pH 7.5).

To the solution, was added dropwise under stirring 200 μ l of 2xHBS (280 mM NaCl, 50 mM Hepes, 1.5 mM NaH₂P0₄, pH 7.12) and the mixture was allowed to stand at room tempera¬ ture for 30 minutes. The medium was removed from the petri dish in which the cell confluency reached 60 to 80 % and 3 ml of the culture medium containing 100 μ M of chloroquine was added.

The DNA-containing mixture was added dropwise to the petri dish and it was kept at 37' C in 5 % C0₂ incubator for 4 hours. Then, the culture mediur was removed, 1 ml of glycerol solution (prepared by adding 15 % of glycerol to eRDF medium) was added to the dish, it was allowed to stand at room temperature for 1 minute, then the glycerol solution was aspirated off, the dish was rinsed with 3 ml of PBS (-) (Nissui Pharmaceuticals Co. Ltd.) twice, added to 10 ml of the medium and the culture was continued in the 5 % C0₂ incubator at 37' C.

The method stated here is known as the calcium phos- phate coprecipitation method and its basic technology is shown by Wigler, et. al., Cell., 14, 725 (1978) and Van der Eb, et.al., Virology 52, 456 (1973). The next day, the cells were trypsinized, diluted 30 to 90 times and the culture was continued in the above-stated medium in Falcon 3025 dishes. At that time, 1 mg/ml of G418 (Gibco) was added to the medium when neomycin resistant gene was used as a marker; 1 μg/ml of methotrexate (Sigma), to dhfr gene; 6 μg/ml of mycophenolic acid, 15 μg/ml of hypoxanthine and 10 μg/ml of thymidine, to Eco-gpt gene. Thereafter the selective medium was used for the culture.

Colonies were formed in 12 days (BHK cells) or 18 days (293 cells) and they were transferred to Coaster 3424 dishes using a cloning cylinder. They were transferred to Falcon 3003 dishes and the culture solution was exchanged, when the cells became confluent, the cultivation was continued for additional 24 hours to collect the culture supernatant.

When needed, the cloning of the cell strain was carried out by the limiting dilution mehtod. In other words, a coaster 3599 dish (96 wells) was used, 200 μ1 of the selec- tive medium was added to each well, one cell was added to each well, and cultivation was continued as the medium solution was exchanged once every 4 days. The cells which formed one colony in one well were transferred to a Coaster 3424 dish to continue cultivation, further transferred to Falcon 3003 dish, and the medium solution was exchanged, when the cells reached confluence, cultured for additionally 24 hours to recover the culture supernatant.

Further, pVA-gpt was transfected into the cloned cell strain. The procedure was the same as in the transfection of strain. The procedure was the same as in the transfection of human protein C expression vector except that eRDF medium containing 500 μg/ml of G418, 6 μg/ml of mycophenolic acid, 15 μ g/ml of hypoxanthine, 10 μ g/ml of thymidine, 250 μ g/ml of xanthine, 5 ^g/ml of vitamin K-_ , 100 μ g/ml of streptomy¬ cin, 100 units/ml of penicillin G and 10 % FCS. The trans¬ fection of pKEX2-gpt was also carried out by the same proce¬ dure.

Then, the total concentration of human protein C and the concentration of the Gla-containing human protein C were determined by the ELISA. In this ELISA, the heavy chain- recognizing anti-human protein C monoclonal antibody, JTC-4 was used on the plate side, while JTC-5 (for the activated peptide recognition and determination of the total human protein C) or JTC-1 (for recognition of Gla domain depending on Ca²⁺ and determination of the human protein C bearing Gla normally) were used as a horse radish peroxidase (HRPO)- labeled antibody.

These monoclonal antibodies are described by Waka- bayashi, et. al., J. Biol. Chem., 261, 11097 (1987). Among the results, the examples which gave the highest values will be given in Table 1.

Table 1

HPC means human protein C.

Further, another series of expression experiments gave the results shown in Table 2. Table 2

HPC means human protein C,

Example 3

Increase in two chain processing efficiency in human protein C

The strain which was obtained by transfection of TZm5- PC9002 into 293 cells was cloned by the above-stated proce¬ dure and pKEX2-gpt was transfected by the above-stated meth¬ od. The cells before and after transfection of pKEX2-gpt were cultured in Falcon 3003 dishes until the cells became confluent, rinsed with PBS(-) twice, then 10 ml of a serum- free medium (5 μ g/ml vitamin Kl-eRDF) was added. Two days later, the culture supernatant was recovered, passed through a membrane filter (Millipore, pore size: 0.22 μ m) and concentrated with Centricon 10 (Λmecon, Registered Trade Mark) into 1/20 volume. The sample was subjected to SDS polyacrylamide-gel electrophoresis (10 - 20 % ) under reduc¬ tive conditions to examine the two chain processing effi¬ ciency by the immunoblotting method using anti-human protein C antiseru (American Diagnostic Co.). The efficiency was about 80 % before transfection of pKEX2-gpt and found to be almost 100 % in almost all clones after transfection. No reduction in expression efficiency of human protein C was was caused by transfection of pKEX2-gpt. Example 4

Purification of human protein C and measurement of the activity A human protein C-producing strain prepared from TZml- PC9001 was cultured in the selective medium stated in Exam¬ ple 2 in Falcon 3025 dishes to collect about 600 ml of the culture supernatant.

The supernatant was passed through a filter of 0.45 μ m pore size, then CaCl was added so that the final concentra¬ tion became 5 mM, and an anti-human protein C monoclonal antibody-immobilized column was used to purify human protein C.

The antibody used here is described as 6H2 in Japanese Patent Specification Laid-open No. Tokkaisho 61-134399. Further, the purification is disclosed in Japanese Patent Specification Laid-open No. Tokkaihei 2-163085. Human pro¬ tein C was eluted as a single peak and the amount of the protein was found to be about 300 μ g calculated from the absorbance.

The purified human protein C showed a single band on the non-reduced SDS polyacrylamide gel electrophoresis. Further, It showed a single chain band in addition to the heavy chain band and the light chain band on the reduced SDS polyacrylamide gel electrophoresis. The densitometry showed that the fraction of single chain human protein C was about 20%.

Three micrograms of the purified human protein C were activated with 0.3 μ g of bovine thrombin (Mochida Pharma- ceuticals Co.) at 37" C, then 9-time volume of antithrombin III (150 μ g/ml)-heparin (2 units/ml) was added 5, 15, 30 and 60 minutes later to terminate the reaction.

Fifty microliters of the reaction mixture and 50 μ 1 of 2mM S-2366 (Cabi) were added to each well of a 96 well micro- plate and the absorbance change at 405 nm wavelength was measured with ETY-96 analyzer (TOYO SOKKI Co.).

As shown in Table 3, time dependent increase in hydrol¬ ysis of synthetic substrate S-2366 was observed.

Table 3

Further, the activated human protein C mentioned above was diluted to 10 - 200 ng sample/50 μ 1 0.1% BSA-TBS (pll 7.4). The sample and 50 μ 1 of Sysmex APTT reagent were added to 100 μ 1 of Sysmex control serum I which was kept at 37' C for 2 minutes, stirred, maintained at 37" C for 2 minutes, then stirred together with 100 μ 1 of 25 mM CaCl₂ and APTT was determined by means of Sysmex CA-100 type blood coagula¬ tion analyzer. The activity was compared with that of the activated human protein C which was obtained by purification with the above-stated affinity column from human plasma and activation through a similar method and the product was found to have equal or higher specific activity.

Claims (8)

1. A DNA fragment encoding human protein C wherein the fragment is constructed by removing n (n = 1 to 6) introns from 7 introns distributing between 8 exons in human protein C gene and ligating the remaining before and behind sequences.

2. The DNA fragment according to claim 1 wherein two introns are removed from between exons 6 and 7 and between exons 7 and 8 in human protein C gene.

3. The DNA fragment according to claim 1 wherein five introns are removed from between exons 3 and 4, exons 4 and 5, exons 5 and 6, exons 6 and 7, and exons 7 and 8 in human protein C gene.

4. A DNA fragment encoding human protein C wherein the frag¬ ment is constructed by incorporating at least one of the following DNA fragments Into any one intron, a plurality of introns or partially deleted introns among n (n = 1 to 7) introns included in human protein C gene or the DNA frag- g ent according to claim 1, 2 or 3: a. a DNA fragment which accelerates transcription activity in mammalian cells, b. a DNA fragment which can be a replication origin of DNA in mammalian cells and c. a DNA fragment of adenovirus VA gene.

5. A method for producing human protein C by expression of the DNA fragment according to claim 1, 2, 3 or 4 in mamma¬ lian cells.

6. The human protein C expression vector wherein the vector can be transfected into mammalian cells, includes a promoter and the DNA fragment according to claim 1, 2, 3 or 4, and a polyadenj'lation signal, when necessary, and is commanded the transcription of the DNA sequence of said DNA fragment by said promoter.

7. A mammalian cell producing human protein C wherein the cell is transformed by the expression vector according to claim 6.

8. The mammalian cell according to claim 7 wherein the cell is selected from the group consisting of BHK cell, CHO cell, HeLa cell, C127 cell and 293 cell.