JPH022338A - Simultaneous development in eucaryotic cell - Google Patents

Abstract

PURPOSE: To produce the objective active type protein by transferring a DNA sequence capable of coding the objective protein together with a DNA sequence capable of coding a protein processing or stabilizing the objective protein into a host cell.

CONSTITUTION: A eucaryotic host cell is transfected or transformed with the first DNA sequence capable of coding the objective protein (preferably a DNA sequence capable of coding a blood plasma serine protease such as t-PA, a factor VII, a factor IX, a factor X, a plasminogen, etc.) and at least one additional DNA sequence capable of coding a protein processing or stabilizing the objective protein (preferably a DNA sequence capable of coding a protease, a protease inhibitor, a γ-carboxylase and a protein binding to the objective protein).

COPYRIGHT: (C)1990,JPO

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention This invention relates generally to the production of proteins, and more particularly to the production of biologically active forms of proteins in commercially viable quantities.

BACKGROUND OF THE INVENTION Recombinant DNA techniques are used for the production of a variety of proteins of therapeutic or economic value, such as enzymes, growth factors and peptide hormones.

These proteins are found in bacterial or fungal cells and-
Layers have recently been produced in cultured mammalian cells.

Because unicellular organisms cannot properly process many human proteins, it is often necessary to use cultured mammalian cells to produce these proteins.

Mammalian cells, for expression of cloned DNA,
Transfection (
Ltransfect). However, all types of mammalian cells are transfected with DNA.
The cells that have been shown to efficiently express the transfected sequences may not otherwise express the different gene products at low levels. Probably. Low expression levels of transfected genes may be due to intracellular and/or extracellular degradation of the protein product, production of inactive forms of the protein, or production of toxic forms of the protein. Low levels of protein or activity may also be due to unstable mRNA sequences, proteolytic activation, or improper, insufficient or incorrect processing by the host cell. Processing steps that may be necessary for the activity or secretion of newly synthesized proteins include specific proteolytic cleavage, polymerization of subunit I, formation of disulfide bonds, post-translational or Includes modifications during translation, and glycosylation.

These problems in protein production reflect the specialized nature of cells derived from higher organisms. Mammalian cells derived from a particular tissue will not adequately produce proteins not normally produced by that tissue. Furthermore, mammalian cell lines adapted to grow in culture will often be derived from tumors or other abnormal M1 tissues, leading to unpredictable protein processing. For example, many research groups have produced human coagulation factors in cultured mammalian cells (Aufman et al.
J, Biol, Chem, 261:9622-
9628.1986; 8n5on et al., Natur
e315: 683-685. 1985
; llagen et al., EP 200.421; Bu
sby et al., Nature 316: 271273
, 1985). Despite efforts to maximize production of biologically active protein through the use of strong promoters, enhancers, and increased gene copy numbers, and despite relatively high levels of factor IX mRNA being harbored. , the level of active factor II produced by these transfected cell lines does not exceed about 5 μg/ml per cell culture medium. In some cases, a precursor form of Factor II is produced, but the mature protein is not effectively secreted from the host cell.

Problems with protein production have traditionally been solved by testing a large number of different cell types and by selecting and screening large numbers of isolates of a particular transformed or transfected strain or cell line. been treated. Such methods are very labor intensive and are not always successful. Therefore, there is a need in the art for a method for systematically and predictably producing recombinant cells capable of expressing the active form of the protein of interest in economically viable amounts. The present invention provides such a system and further provides other related advantages.

(Summary of the Invention) In summary, the present invention provides a method for producing a protein of interest, which method comprises: (a) a first D protein encoding the protein of interest;
introducing the NA sequence and an additional DNA sequence encoding a protein that processes or stabilizes the protein of interest into a mammalian Cypress host cell; (b) allowing expression of the first DNA sequence and the additional DNA sequence; and (c) isolating the protein of interest from the host cell. Introducing the DNA sequence into a host cell may include (a) co-transfection or co-transformation with multiple vectors, each containing a separate expression unit; or transfection with a single vector containing multiple expression units; Or it can be through transformation. If the host cell is a mammalian cell, the introduction step involves a polycistronic message.
transfection with a single vector containing a single expression unit that is transcribed into ssage). If yeast host cells are used, the DNA
A preferred method for introducing sequences includes (a) additional D
transforming a yeast host cell with a single expression unit containing the NA sequence; (b) isolating a host cell that stably provides a processing or stabilizing activity; and (c) transforming the isolated host cell. , transforming with a first DNA sequence containing the protein of interest. Preferably, the initial transformation step results in the integration of a single expression unit into the yeast host cell genome.

Preferred first DNA sequences include DNA sequences encoding plasma serine proteases such as L-PA, Factor ■, Factor ■, Factor X, Protein C1, and plasminogen. Preferred additional DNA sequences include DNA sequences encoding proteases, protease inhibitors, T-carboxylase, and proteins that bind to the protein of interest.

In another aspect of the invention, a DNA sequence encoding said protein of interest and one or more additional DNA sequences encoding one or more proteins that process (or stabilize) said protein of interest are introduced. Nuclear host cells are disclosed. Preferred eukaryotic host cells include mammalian host cells and yeast host cells.

DETAILED DESCRIPTION Before describing the present invention, it may be helpful for understanding the invention to explain some terms that will be used later.

The term "stabilizing" is used to describe the protection of proteins from denaturation. Stabilization is achieved by a number of mechanisms, including inhibition of proteolytic enzymes that might degrade the protein of interest, binding to the protein of interest to protect it from proteolytic enzymes, and due to the activity of proteases. Inhibition of the action of or binding to necessary cofactors or other molecules is included.

Processing As used herein, "processing" means modifying the structure of a protein. Processing of proteins includes the generation of multi-domain proteins or removal of peptides from protein precursors by specific proteolytic cleavage; modification of amino acids, including carboxylation and hydroxylation; and addition of carbohydrates at specific sites. included.

Processing may be necessary for full biological activity of the protein or may be necessary to enable secretion of the protein from the cell.

These are the processes by which cloned DNA is introduced into host cells. Transfection refers to the insertion of DNA into a mammalian cell line. The process of inserting DNA into fungal or bacterial cells is known as transformation. Numerous transfection and transformation methods are known in the art.

As mentioned above, cells containing a cloned DNA sequence consistently produce the protein encoded by the cloned sequence at economically viable levels or in biologically active form. Not exclusively. This is often due to the origin of the cells or their unusual properties. In many cases, it is not possible to obtain cultured cell lines or host cell lines that have the necessary characteristics to enable production of a particular protein at the desired level. In any case, it is not possible, and not always successful, to screen a large number of potential host cell lines.

The present invention relates to a gene or cDNA encoding a protein of interest.
The shortcomings of available cells are overcome by providing a method for introducing a protein into a eukaryotic host cell along with one or more additional DNA sequences encoding one or more proteins that process or stabilize the protein of interest. Overcome. Processing proteins include proteases that cleave a precursor protein at a specific site, resulting in a mature and/or active form of the protein, such as removing a peptide or propeptide, or cleaving a single chain polypeptide to produce multiple products. Contains peptidases that form Other examples of processing proteins are enzymes that modify amino acids, such as T-carboxylase, an enzyme that modifies specific glutamate residues of certain clotting factors and other calcium-containing proteins.

Other processing proteins include enzymes responsible for converting aspartate to β-hydroxyaspartate, a modification necessary for protein C activity; enzymes responsible for adding carbohydrate chains to glycoproteins; and myristoyl Includes proteins responsible for oxidation, hydroxylation of C-terminal amino acids, hydroxylation of proline residues, sulfation, and C-terminal amidation. Stabilizing proteins include protease inhibitors that block proteolytic degradation of the protein of interest; proteins that bind to the protein of interest and make it unavailable as a substrate; protein cofactors, ions, or other molecules required by the protease; proteins that bind to cofactors; as well as proteins that inactivate cofactors. Certain eukaryotic host cells can be transfected or transformed to produce several processing proteins, several stabilizing proteins, or a combination of stabilizing and processing proteins.

The invention is based in part on the unexpected discovery that various processing proteins function outside of their native environment. For example, Factor■, which is an enzyme that normally functions in the blood.
It is disclosed herein that the factor II is capable of activating factor II in cells. Alternatively, yeast KrEX
The products of the two genes were found to function normally in mammalian cells.

The observed processing unexpectedly shows that both the target protein and the processing protein are directed to the same cellular components. The observed functions of these processed proteins are also surprising in that the proteins of interest are not produced in large amounts or in intact or active form in recombinant cells.

By characterizing the natural fluorescent protein of interest and, if necessary, the gene or cDNA encoding it, the nature of the processing steps involved in its biosynthesis can be determined. By characterizing the recombinant protein, It will become clear whether this is being processed correctly and, if not, which processing steps have been omitted.According to the invention, by replacing lost activity or enhancing limited activity, it will become clear. Appropriate processing is provided. This activity is supplied either as a protein that normally processes the protein of interest, or as a related protein that normally performs a similar function in other contexts, such as proteins from different cell types. The latter An example of this case is the use of an enzyme II) gene which provides a processing protein that cleaves Protein C or an activated Protein C precursor into a double-stranded form. Blockage of secretion of the protein of interest can be determined by characterizing the intracellular and secreted forms of the recombinant protein. This method determines missing or restrictive processing steps.

In some cases, the identity of the protein providing the missing activity will be known. In this case, the desired gene or cDNA
is cloned and introduced into a selected host cell.

If the protein responsible for the missing activity is unknown, the protein of interest is analyzed, the nature of the missing processing step is determined, and a protein known to perform that function is selected. Appropriate proteins can be selected from known and available NA sequences encoding proteins with similar activities.

For example, Kettner and Shaw (MeLh, i
n Enz mol.

Ho: 826-842.1981) characterized the specificity of a number of proteases. Alternatively, DNA sequences that coat the required processing proteins can be identified by transfecting cells with a mammalian expression library and selecting cells that do not exhibit the required activity.

If protein stabilization is desired, transfect or! to produce the protein of interest. - Transformed cells are grown in the presence of various stabilizing proteins and assayed for production of intact target protein. Generally, cells are labeled with radioisotopes and proteins are analyzed by radioimmunoprecipitation and gel electrophoresis, or by other conventional methods. The presence of intact target cells in the culture medium
This shows that the stabilizing protein protects the target protein from degradation. Stabilization may also lead to phenotypic changes in the host cell. For example, protease activity can cause cellular encroachment, which can be corrected by inhibiting this determination. This correction is indicated by the presence of a normal (adhesive) phenotype.

Once the desired stabilizing or processing protein has been identified, the DNA sequence encoding it and the DNA sequence encoding the protein of interest are introduced into selected host cells as described below. Expressing cells are selected and screened for production of the desired form of the protein of interest at the desired level. Cells meeting these criteria are then cloned and scaled up for production.

Target proteins that can be produced by the method of the present invention include various plasma serine proteases, such as coagulation factors Factors ■, ■, and X, activated Factor ■ (referred to as Factor ■a), and activated Factor ■. X(
Factor Xa), protein C1 activated protein C, proteins, Mi tissue plasminogen activator (tPA), plasminogen, and analogs and derivatives thereof. However, virtually any desired protein can be produced. These methods are particularly suitable for the production of these serine proteases. This is due to post-translational processing required for these activities and/or secretion by the host cell.

According to the present invention, coagulation factors that require T-carboxylation of specific glutamic acid residues for biological activity are expressed at high levels in cells into which a DNA sequence encoding T-carboxylase has been introduced. be able to. T
- It has been found by the inventors that the carboxylation step is limiting in mammalian cells commonly used for the production of recombinant coagulation factors. Additionally, the present invention allows for the production of biologically active T-carboxylated proteins in non-mammalian cells such as yeast cells.

DNA sequence encoding factor ■ and factor ■
The activated form of coagulation factor 1 can be produced by transfecting cells with a DNA sequence encoding . The precursor form of Factor ■ is activated by Factor ■, and Factor ■a is secreted by the cell. Alternatively, a derivative of factor ■ or Th'Q (
By co-expressing the DNA sequence encoding the body and the DNA sequence encoding Factor 2, a protein having the biological activity of Factor 2a can be produced. DNA sequences encoding analogs and derivatives of Factor II are described in US Patent Application No. 721,311 and No. 810,002, as well as in published European Patent Application EP 200.421. Direct production of the activated form of Factor II has been shown to be of therapeutic value, and therefore it is desirable to produce the activated form directly. Direct production increases yields, reduces the number of manufacturing steps, and eliminates problems associated with activating blood products.

In other embodiments, a DNA sequence encoding a Protein C1-activated Protein C1-modified Protein C or an activated Protein C precursor is inserted into a cell line transfected to express a yeast dish gene. be done. This gene encodes an endopeptidase that cleaves after a dibasic amino acid sequence (Fu
ller et al., in Leiveli, Microbio.
lo, 1987. 273 278゜1986)
. The yeast ■■ gene (D+noch) encodes an enzyme that removes basic amino acids from the C-terminus of protein C light chain.
Owska et al., Juxtaposition+' 50: 573-584.
Processing may be further enhanced by further transfecting the cells with 1987). The modified protein C precursor is described in US patent application Nα924.4.
62 and Nα 130,370, and published European patent application 191EP 266.190. Preferred protein C and activated protein C precursors encompass the amino acid sequence R, -Rt-R3-114-XR5-R, -Rw-R8 at the cleavage site between the light and heavy chains, where R, -R, are Lys or Arg and X is a peptide bond or a spacer arm of 1 to 12, preferably 2 to 8 amino acids.

Activated protein C can also be produced in cell lines transfected with DNA sequences encoding activated protein C precursors, thrombin and thrombomodulin.

Thronchin cleaves the activated protein C precursor (the double-stranded form of protein C) in the presence of thrombomodulin, and the activated protein C is secreted by the cell. Activated protein C precursors are described in U.S. Patent Application No. 924,462 and No. 130,37.
0, as well as published European patent application UP 26
6.190.

Factor Ia and Factor IXa can be produced in cell lines transfected to co-express Factor X. Factor X cleaves Factor ■ or Factor ■, resulting in the secretion of activated coagulation factors.

In other embodiments, transfecting a host cell line to produce both t-PA or t, a PAi analog or derivative and a protease inhibitor.
- Production of PA or t-PA analog or derivative is enhanced. Suitable protease inhibitors in this regard include TIMP (tissue inhibitor of metalloproteases), trypsin inhibitor and aprotinin, with aprotinin being particularly preferred. Analogs or derivatives of j-PA are disclosed in U.S. Patent Application No. 822.005 (corresponding published Japanese Patent Application No. 63-133.988), NQO
O4,995, No.053,412, No.053,
411, No, 058,061, No, 058.2
1? , and N.

125.629, as well as European Patent Application Publication N.

196.920, No. 201,153, and No.
, 240,334. Expression of L-PA at commercially viable levels has hitherto been difficult.

This is because the serine protease activity of LPA results in the detachment of cells from supporting surfaces. This necessitates the use of drugs such as aprotinin in the growth medium. The use of aprotinin in the culture medium also allows the production of a single chain form of t-PA, which is a preferred therapeutic substance. However, aprotinin is expensive and has limited availability.

Cells transfected to produce both plasminogen and protease inhibitors can be used to produce plasminogen in its intact form. Variants of plasminogen (US Patent No. 1tJI No. 05)
3,412) can also be produced. A particularly preferred protease inhibitor in this regard is alpha-1-antitrypsin. However, α-1-
Variants of antitrypsin (U.S. Patent No. 4,711)
, 848) and other protease inhibitors can also be used. Intracellular activation and subsequent degradation of plasminogen has limited the possibility of producing recombinant plasminogen at reasonable levels. Inhibition of plasminogen activity by AAT may ameliorate this problem.

Another example of a combination of protein stabilizing substances is Protein S
Stabilization of protein C by co-expression of von Willebrand factor, stabilization of factor (2) by co-expression of von Willebrand factor, and stabilization of factor (2) by co-expression of tissue factor. Bip
(Baas and -abl, Nature 306
: 387-389, 1983) can also be used to stabilize various proteins. c to code Bip
DNA is Munr.

and Pelham (Cell 46: 291-3
00.1986).

NA sequences useful in carrying out the invention can be cloned by standard methods known in the art. Genomic or cDNA sequences can be used. Many such clones, the DNA encoding t-PA (Pennica et al., Nature, 3
(Narrow: 214-221.1983), DNA encoding factor ■ (llagen et al., supra; lla
Gen et al., Proc.

DNA encoding actor-■ (Kurachi and Davie).

Proc, Natl, 8cad, sci, UsA,
79: 6461-6464. 1982), DNA encoding plasminogen (Malinows
Ki et al., Biochemistr 23: 424
3 4250.1984; Forsgren et al., F.
EBS Lett, 213: 254.1987)
, DNA that coats α-1-antitrypsin (
Long et al., Biochemi Dan■, 23: 48
28-4837.1984; Kawasaki, U.S. Patent No. 4,559,311), DNA encoding Protia C (Foster and Davie,
Proc, Natl, 8cad, sci, LIsA81
: 4766-4770.1984 ; Foste
r etc., Proc, Na order.

^cad, sci, LIsA, 82,: 4673-
4677, 1985), DNA encoding prothrombin (Friezner-Degan et al., Bioch.
emistr, 22: 2087-2097.1
983), DNA encoding factor ■ (To
Ole et al., Nature, 312:342-347.
1984), von Willbrand (1984), vonWill
ebrand) DNA encoding the factor (Lynch
etc., the DNA encoding the child (Spicer et al., Proc
, Natl, ^cad.

Sci, USA, 84: 5148 5152.1
987), and DNA encoding factor-X (L
ey tus et al., Biochemistr25: 5
098-5102.1986) is described in the literature.

By screening cosmid, genomic or cDNA libraries using oligonucleotide probes or cloned DNA fragments designed based on amino acid sequence data, or by using antibodies against the protein of interest (Young and Davis, Proc.
c, NaLl, Acad, Sci, LISA, 80
; 11941198, 1983) by ligand fronting (Sikela and 1lah
n, Proc, Natl, 8cad, Sci.

tlsA, 84: 303B-3042, 1987
) or from expression libraries that are screened for activity.

The cloned DNA sequence is inserted into a suitable expression vector and used to transfect or transform a suitable eukaryotic host.

The expression vector used to practice the invention in mammalian cells comprises a promoter capable of directing transcription of the cDNA or cloned gene introduced into the mammalian cell. Viral promoters are preferred because of their efficiency in directing transcription. A particularly preferred promoter is the major late promoter from adenovirus 2.
1ate) is a promoter. Other suitable promoters include the SV40 promoter (Subramani et al., Mo1. Ce1l Biol, Sat: 854864, 1
981) and MT-1 (metallothionein gene) promoter (Palmiter et al., 5science+ 2
22:809-814.1984). Such expression vectors may further contain one cent RNA splice site, either downstream from the promoter and upstream from the insertion site for the cloned DNA sequence, or can be located within the array itself. Preferred RNA splice sites can be obtained from adenovirus and/or immunoglobulin genes. Furthermore, the expression vector contains a polyadenylation signal located downstream of the insertion site. Viral polyadenylation signals, e.g. SV40
Preferably, early or late polyadenylation signals from the adenovirus 5E+tlJi region are preferred. Expression vectors useful in the practice of the invention also contain non-coding viral leader sequences, such as adenovirus 2 tripartite located between the promoter and the RNA splice site.
) can include a leader. Preferred vectors also include transcriptional enhancer sequences, such as the SV40 enhancer,
and translation enhancer sequences, such as adenovirus UA.
It can contain sequences encoding RN^.

The vector containing the cloned DNA sequence is then subjected to, for example, calcium phosphate-mediated transfection (
Somatic Ce1l Genetics, 7
: 603+1981; Graham and Van
der Eb, u animal ■, , 52: 456+
1973) or electroporation (Neuman
n.

EMBOJ, Sat.: 841-845.1982) into cultured mammalian cells. A small fraction of the cell population integrates the DNA into the genome or maintains the DNA in a non-chromosomal structure. To identify these integrants, a gene conferring a selectable phenotype (selectable marker) is introduced into the cells together with the gene of interest. Preferred selectable markers include genes that confer resistance to drugs such as neomycin, hygromycin, and methotrexate. If calcium phosphate-mediated transfection is used, the selectable marker is introduced into the cell at the same time as the gene of interest on a separate plasmid, or they are introduced on the same plasmid.

When on the same plasmid, the selectable marker and the gene of interest can be under the control of different promoters or the same promoter. The latter arrangement results in what is known as a nidistronic or polycistronic message. This type of construct is known in the art (
For example, European Patent Application Publication No. 117.058
; U.S. Patent No. 4.713.339).

After the cells have taken up the DNA, they are grown for a certain period of time, typically 1 to 2 days, to initiate expression of the target gene. Drug selection is then applied to select for proliferation of cells expressing the selectable marker. When using methotrexate selection, stepwise increases in drug concentration allow selection of increased copy numbers of cloned sequences, resulting in increased expression levels. Clones of these cells can be screened for production of the protein of interest. Common screening methods include immunoassays and activity measurements.

Preferred mammalian cervical cell lines for use in the present invention include CO5 (8TCCCRL 1650), [1I
IK (8TCCCCL 10) and 293 (ATCC1
573) Includes cell lines and derivatives and isolates of these cell lines. However, other cell lines are also preferred for the production of specific proteins. A preferred cell line is tk-ts 1
3 BIIK cell line (Waech Ler and Base
rga, Proc, Natl, Acad, Sci, I
JSA and: 1106-1110.1982),
These are hereinafter referred to as Lk-BIIK cells. Other useful adherent cell lines include Ratllep [(8TCCCRL
1600), Rat l1epH (ATCC
CRL 1548) TC gate K (ATCCCCCL 13
9), human lung cell line (ATCCCCCL75.1), human liver cancer (8TCCHT8-52), l1ep G2
(ATCCIIB 8065), mouse hepatocytes (ATC
CCC29,1), and DUKX cells (Urlaub and Chasin, Proc. Natl.

8cad, sci, UsA, 77: 4266 4
220.1980). Useful expression cell lines include 8tp-20 (ATCCCCCL 89).

MOLT-, 1 (ATCCCI?L 1582), [
3W5147. G, 1.4.0IJAR, 1 (ATCC
CRL 1588), 519415. XXO,
IIU, 1 (ATCCTlB20), EL4. B.U.
1.0UAr1.1 (ATCCTIB 41), Sp
2108g14 (ATCC CRL 1581),
J558L (Oi et al., Proc, Natl.

^cad, sci, UsA, 80: 825 82
9.1983), and Raji (8TCCCCL 86
) is included.

Generally, a host cell system will be selected based on its ability to produce high levels of the protein of interest and its ability to perform at least some of the processing steps necessary for the biological activity of the protein. . In this way, the number of cloned DNA sequences that must be transfected into host cells will be minimized and the overall yield of biologically active protein will be maximized. However, the present invention allows virtually any protein to be produced in virtually any cell line that can be cultured in vitro.

A DNA sequence encoding a protein of interest and a processing and/or stabilizing protein can be introduced into cells on the same vector or on different vectors. It is preferred to use a single vector with one selectable marker to minimize problems that can arise from selectable marker instability. Multiple genes or cDNAs on one vector can be controlled by separate promoters or a single promoter.

In the latter configuration, giving rise to a polycistronic message, multiple genes or cDNAs will be separated by translation termination and initiation signals. many D
In the case of transfection using NA sequences, practical limitations on vector size may require the use of multiple vectors, each with its own selection marker.

Of course, multiple vectors can be used whenever co-expressing a protein of interest and a stabilizing or processing protein.

Other eukaryotic cells such as yeast and filamentous fungi can also be used in the present invention. These lower eukaryotic hosts offer several advantages over mammalian cell systems, including ease of culturing and economy of using existing industrial fermentation capacity. Eukaryotic cells or other eukaryotic cells capable of expressing cloned DNA sequences may be used to produce virtually any protein by manipulating the cells as described herein. I can do it.

For example, baker's yeast [Saccharomyces cerevisiae (S
accharom ces cerevisiae) can be transformed with the cloned foreign DNA sequence and cultured to high cell density;
It will then express the cloned DNA and secrete the foreign protein. However, in some cases the foreign protein is not secreted or is inactive due to lack of proper processing or proteolysis. For example, yeast cells cannot T-carboxylate proteins or add complex mammalian-type carbohydrate chains to glycoproteins. According to the invention, said processing deficiency can be overcome by transforming a yeast host cell with a DNA sequence encoding the deleted protein (eg, T-carboxylase or glycosylation enzyme). Degradation of foreign proteins can be overcome by transforming cells to produce protease inhibitors or proteins that bind to the protein of interest. An example of a suitable binding protein is Bi
Examples include p. Binding proteins can also enhance secretion of foreign proteins by altering their conformation. Furthermore, by transforming yeast cells to produce foreign proteases, we can generate "active forms" of foreign proteins (e.g., mammalian serine proteases) that require specific cleavage for secretion and/or biological activity. The yeast cells may be capable of producing and secreting. By manipulating the junction between the secreted peptide and the mature protein, the mature W1 protein can be released by a coexpressed protease (eg, thrombin).

Eukaryotic microorganisms, such as the yeast Satucharomyces cerevisiae-1 or Aspergillus
Filamentous fungi such as species of can be transformed by known methods. Aspergillus species, for example, Y
Elton et al.'s method (Proc, Natl, Acad,
Sci.

USA, 81: 1740-1747.1984)
can be transformed according to Particularly preferred species of Aspergillus include A. niger (A-Heng Ju);
, A, n1dulans (A, n1dulans), A
, Orysee (8.

■u animals), and A, terreu (A, terreu)
s) is included. Methods for transforming yeast are described, for example, by Begg (Nature, 275:
104-108.1978). Expression vectors for use in yeast include YRp7
(Struh I et al., Proc.

Natl, 8cad, Sci, USA, 76:
1035 1039. 1979).

YEp13 (Broach et al., Gene, 8:
121 133.1979).

Included are pJDB248 and pJDB219 (Beggs, supra), and derivatives thereof. Such vectors generally carry a selectable marker, for example a nutritional marker that allows selection in host strains with starvation mutations, or a PO that allows selection in rich medium-grown tpi-strains.
TI selection markers (Kawasaki and Be11,
EP 171, 142). Preferred promoters for use in yeast expression vectors include the promoter from the yeast glycolytic gene (II i tzem
ay et al., J. Biol. Chem., 255: 1
2073-12080.1980; Alber and K
ainasaki, J., Mo1. A1. Gene
t,, 1.419-434゜19B2 ;
ki, U.S. Pat. No. 4,599,311), or the alcohol dehydrogenase gene (Young et al., Gen.
etic En 1neerin of Mi
croo anisms for chemistry
edited by als11o11aender et al., 335 pages, Pl.
enum, New York, 1982; and 8mmer
er+ Meth in Enz mol.

101:192-201.1983).

To promote secretion and proper processing of target proteins produced in yeast transformants,
Generally a signal sequence can be provided. Preferably, the signal sequence will be obtained from a yeast gene that coats a secreted protein. A particularly preferred signal sequence is the plapro region (Kurjan and H
erskowi tz, Cedan, 30:933943
, 1982: Kurjan et al., U.S. Patent No. 4.
546.082; and Singh, EP 123,
544).

Co-expression of multiple DNA sequences in yeast can be achieved in several ways. Preferably, the expression unit for the processing or stabilizing protein is integrated into the host cell genome and an isolate is selected that stably produces the processing or stabilizing activity. Next, an expression vector containing a DNA sequence encoding the protein of interest is transformed into a host strain. Alternatively, the two expression units can be placed on different autonomously replicating expression vectors with different selection markers, or they can be placed on a single expression vector.

In a preferred embodiment, a DNA sequence encoding a foreign processing protein is inserted into a DNA, A sequence encoding a yeast protein with a similar function. This insertion is
It is designed to replace yeast sequences encoding the processing functions of yeast proteins with foreign sequences. A particularly preferred such yeast protein is the KEX2 gene product. This protein has been analyzed and its catalytic domain and other domains have been characterized (Fuller et al., Yeas LCell Bioio, Col.
Spring 1larbor Laborator
y Cold Spring Barber, New York.

1987, p. 181). Next, foreign DNA sequences and transport domains and cellular localization are introduced.
The resulting hybrid sequence, comprising the yeast sequence encoding the (orization) domain, is ligated to a yeast promoter and terminator. Preferably, the promoter is a strong promoter, such as the TPII promoter. The construct can further contain flanking sequences to direct the expression unit to a specific integration site in the host cell genome.

Proteins produced according to the present invention can be
l-conditioned) culture medium by various methods known in the art,
This method can be selected according to the physicochemical properties of the particular protein. Suitable methods include affinity chromatography, ion exchange chromatography, gel filtration, high performance liquid chromatography, and combinations of these methods.

Proteins produced according to the invention can be used as compositions for pharmaceutical industry, research and other purposes.

For pharmaceutical use, the protein is generally mixed with a physiologically acceptable carrier or diluent, such as sterile water or sterile salt solution, and packaged in individual doses in sterile vials. Alternatively, the protein can be lyophilized in lyophilized form and remelted prior to administration. Administration will generally be by injection or infusion. The pharmaceutical composition may further contain additional proteins or other compounds of medical value, auxiliary agents, local anesthetics, and the like.

Next, the present invention will be explained in more detail with reference to examples, but the scope of the present invention is not limited thereby.

■ Unless otherwise noted, routine molecular biology techniques were used. Restriction endonucleases and other DNA modifying enzymes (e.g., T4 polynucleotide kinase, bovine alkaline phosphatase, DNA polymerase l Kleno-fragment, T4 polynucleotide ligase) were obtained from Bethesda Research Laboratories New England Biolabs or Boehringer Mannheim. , and according to the supplier's instructions, or according to the manufacturer's instructions.
”J'fl, Mo1ecular C1onin
:8LaboratorManual, ColdSpr
ingllarborLab.

Ratory+ Cold Spring Harbor-3 New York, 1982. Oligonucleotides were synthesized on an Applied Hiosystems Model 380 DNA Synthesizer 1 and purified by polyacrylamide gel electrophoresis on denaturing gels.

The composition of plasminogen activator (t-PA) has been described in the literature. For example, Penn1ca et al. (ibid.), Kaufma et al.
n et al. (Mo1. Ce1l Biol, 5:17
50-1759.1985), and Verheijen
etc. (EMBOJ,.

5.3525-3530.1986). A mature cDNA clone comprising the coding sequence for PA was isolated from the Boives melanoma cell line (Rijken and Colen J, Biol, Chem, 256:
7035 7041.1981) as a starting material in the inventor's laboratory. Next, this cDNA was used to create plasmid pDR129.
6 was produced. The E. coli JM83 strain transformed by pDR1296 has been deposited with the American Type Culture Collection as ATCC 53347.

By ligating the cDNA in pDR1296 to the fragment made from the synthesized oligonucleotide,
The cDNA was extended to include the sequence encoding the PA brepro peptide. The synthesized prepro fragment is B
Inserted into pUc8 digested with am111.

Plasmid pi (J9R (Marsh et al., Gene,
32: 481 4861984) was digested with Sma I and 1Iind[l. next,? 7 button position 2
70 (Pvu II) to 5171st (Ilind
The ori region of SV40 up to p
Plasmid Zem67 was created by ligating to lc19R. This plasmid was then cleaved with Bgl II and the human growth hormone gene (De Noto et al., Nuc, Ac1dsRes., 9:3719
3730.1981)
gl[1-BamHI fragment was inserted to create plasmid Zem86 (Fig. 1). Synthesized t-P
The Aprepro sequence was removed from pUc8 by digestion with Bamtl I and Xho II. This fragment was inserted into Zem86 digested with Bgl II to create plasmid Zem88. Plasmid pDR1296 was digested with Bglll and BamHI and t-PA
Ze cDNA was cut into eight fragments and cut with BglTI.
Inserted into m88. The resulting plasmid was named Zem94.

Vector Zem99, comprising the MT-1 promoter, the complete t-PA coding sequence and the human growth hormone (hGH) terminator, was then assembled as shown in FIG. Kpn comprising the MT-1 promoter
I BamHI fragment was converted into TThGHll (Palm
Iter et al., 5science, 222: 809-
814.1983) and inserted into pUclB to generate Zem93.

pBR322 derivative pBX322 (PalmiLer
Plasmid EV142 comprising the MT-1 and hGH sequences in MT-1 and hGH sequences was digested with EcoRI and the MT-1
A fragment comprising the -1 promoter and hGH terminator sequences was isolated. This fragment was cloned into pUc13 digested with EcaRI to create plasmid Ze.
m4 was created. Next, Zem93 was linearized by digestion with Bam1 (I and Sal I).
1 II and 5all and purified the hGH terminator. t, -PA playpro sequence 5
It was extracted from the pUc9 hector as an au3A fragment.

Next, the three DNA fragments were ligated and Zem97
A plasmid having the structure shown in FIG. 2 was selected. Ze
m97 was cut with BglII and pDI? 12
Bgl II-Bamll I t-P from 96
A % piece was inserted. The obtained vector was named Zem99.

Plasmid pSV2-DIIFR (Subramani
et al., supra) with Cfo I and DIIFRc
The fragment comprising the DNA and the 3' appended SV40 sequence was isolated, repaired, and ligated with Bamll I linkers. After digestion with BamHI, total cDNA and SV40
The approximately 800 bp fragment containing the terminator region was purified and ligated into BamHI digested pUc8.

As shown in FIG. 2, Zem67 was digested with BglI and BamHI DIIFR-sv401! Fr.
and ligated to generate plasmid Zem176.

Plasmid Zem93 was digested with 5stl and MT
Plasmid Zem106 was generated in which approximately 600 bp of sequence 5' to I was removed. Plasmid Zem10
6 was digested with EcoRI and the plasmid Zem17
1EcoR containing the D It PR gene from 6
! The fragments were concatenated. The obtained plasmid was named Zts13. Plasmid Zts13 was digested with Ram)lf and ligated to the Bamtl I fragment from plasmid Zem99 containing the entire t,-PA coding region and hGH terminator sequence.

The resulting plasmid was named Zts15.

Zts15 was partially digested with BamtI, repaired, and religated to generate plasmid Zem219 in which the 3' Bam1I site was destroyed (Figure 2).

Next, Zem219 was partially digested with Xho, repaired, and religated to generate plasmid Zem219a in which the 3' Xho 1 site was destroyed (Figure 5).

Next, the DNA sequence encoding aprotinin was added to the
Zem219a instead of the sequence encoding j -p A
inserted into. Zem219a as Bgl I [and Xba
I and the vector fragment was recovered. Plasmid pKFN-414, a pUC-based plasmid containing the synthetic aprotinin gene, was transfected with ^vaI [and
Digested with T and recovered an approximately 150 bp aprotinin fragment. (The sequence of the synthetic aprotinin gene is shown in Figure 15.) This fragment lacks the 5' end of the aprotinin coding sequence. Oligonucleotide ZC1908 (sense strand: 5'-CAT CTA GGCCTG
ATTTCT GTT TGG AACCTCC
AT ACA CTG-3”) and ZC1909 (
Anti-sense strand: 5'-GACCAG TGT
ATG GAG GTT CCA AACAG
BIl! to anneal the 5' end of the aprotinin sequence and ligate this aprotinin sequence in frame with the t-PA prepro sequence. The l■ site was obtained. next,
The aprotinin fragment of approximately tsobp, the annealed oligonucleotide, and the Zem219a fragment were ligated to create Zem.
252 was produced.

Plasmids Zem219a and Zem252 were co-transfected into tk-13+IK cells by the calcium phosphate method. Transfected cells were selected using methotrexate and screened for L-PA production. Positive clones were radiolabeled and the media was screened for aprotinin production using a rabbit anti-aprotinin antibody. Aprotinin producing clones were selected.

A cDNA encoding the major form of knob human alpha-1-antitrypsin (AA'F) was extracted from a human liver cDNA library using a balloon sequence (Kurachi et al., Proc, Nat.
l, Δcad.

Sci, flsA, 78: 6826 6830.
1980; and Chandra et al., Biochem,
Biohs, Res, comm,, 103.75
1 7581981) was isolated by a conventional method using it as a DNA hybridization probe. This library combines human liver cDNA library with plasmid pB
R322 (Boliver et al., Gene, 2:
95 113.197? ) was created by inserting it into the Pst1 site of .

HeheT cDN8 was isolated from this library as a 6Pstl fragment. This fragment was inserted into the Pst1 site of pUc13 to generate plasmid pUcα1. pUc
In αl, the AAT sequence has an Xho I and EcoR1 site in a polylinker at its 3' end. As shown in Figure 4, this cDNA sequence was used to create plasmid pFA.
TPOT decreased by 1. Plasmid pFATPOT has been deposited with the ATCC as S. cerevisiae-218 strain transformant ATCC20699.

Next, AAT cDNA was ligated to the TPII (triose bosphate isomerase gene) terminator in plasmid pMVI11. This plasmid further comprised the TPII promoter and was assembled as follows. Plasmid plc7 (Marsh
et al., Gene, 32: 481 486.1984
) was digested with EcoRI, the ends of the fragment were blunted with DNA polymerase I (Klenow fragment), and the linear DNA was recircularized using T4 DNA ligase. The resulting plasmid was used to transform E. coli RRI. Plasmid DNA was prepared from transformants and screened for loss of the EcoR1 site. The plasmid with the correct restriction pattern was named plc7RI''. The TPII promoter fragment was transformed into plasmid pTPIc as shown in Figure 4.
10 (81ber and Kawasaki, J. Mo.
1ec, 工＾”)]λ”U Genet, , Sat: 41
9-434.1982). This plasmid
Cut at the unique pn1 site in the dwarf gene, B
The TP11 coding region was removed by a131 exonuclease and IEcoR [linker (sequence: GG
AATTCC) was added to the 3' end of the promoter.

Bgl by digestion with Bgl II and EcoR1
A dwarf promoter fragment with II and EcoRI cohesive ends was obtained. Next, this fragment was extracted from plasmid YRp7' (Stin-c
hcomb et al., Nature, 282: 3943
．． 1979).

The resulting plasmid TE32 [! coR1 and Bam
The tetracycline resistance gene portion was removed by cleavage with 1 l l. The linearized plasmid was then recircularized by addition of the EcoR[-Bamll I linker- described above to generate plasmid TEA32.

Plasmid TE 32 was digested with BglI and EcoRI and the approximately 900 bp partial tool promoter fragment was gel purified. Plasmid plc1'1ll (
Arsh et al., supra) was cut with Bgl[I and EcoRI, and the vector fragment was gel purified. Next, T.I.
'11 promoter fragment linearized plc1911
and this mixture was used to transform E. coli RRI. Plasmid DNA was prepared and screened for the presence of a Bgl II-4coRI fragment of approximately qoobp. The correct plasmid was selected and named plcTP[P. Plasmid ρIC71? I'' was digested with 11indllll and Narr and the 2500 bp fragment was gel purified.

A partial dwarf promoter fragment (approximately 900 bp) was removed from plcTPIr' using Nar I and Sph I and gel purified. pFATPOT sph
1 and 1lindll and gel purified a 1750 bp fragment comprising the TPII promoter, 8AT cDNA and TPTI terminator. Next, 1? Ic7RI” fragment, milk promoter fragment, and sin promoter AAT7TP1 from PFATPOT
1 terminator fragments are linked in a ternary ligation to form pMν
R1 was produced (Fig. 4).

Plasmid pMνR containing the entire coding region of AAT from amino acid number 2 for the expression and secretion of α1-antitrypsin by transfected mammalian cells.
The [lamHI-Xba I fragment from 1 was isolated and the 26m
219a (Example 1).

To generate an AAT expression vector, 26m219a was digested with BglII and XbaI and the t-PA coding region was removed. Oligonucleotide ZC1173
(5'-GAT CTT CA-3') and ZC1
174 (5'GAT CTG AA-3'.) was annealed to form the Bgl II-Xho II adapter. Next, the AAT fragment, adapter and floccated vector were ligated by three-way ligation to create Zem235
was produced (Fig. 5).

The adapter aligned the reading frame of the L-PA prepro sequence and the AAT sequence and restored the missing glu codon from the 5' end of the AAT sequence.

Plasmid Zem235 was transfected into tk-BIIK cells by electroporation. Methotrexate selection was applied after 48 hours, and after 2 weeks, multiple clones were picked, expanded, and characterized for levels of α-1-antitrypsin secreted into the medium. AAT at a rate of approximately 20 μg/ml/mouth
Select one clone secreting 235-6 and
It was named.

The cDNA encoding plasminogen was obtained by Mark Markso of the University of Washington, Seattle, Washington.
Obtained from Dr. Malinowski et al. (cited above)
Clones were isolated from a lambda phage library by screening using partial cDNA clones described by . The cDNA sequence is shown in FIG.

λ phage DNA was extracted from positive clones according to standard methods.
11 was made. Lambda phage was subjected to EcoR [partial digestion;
and approximately 2800 bp of Eco containing the entire coding sequence.
The RI fragment was recovered and cloned into the EcoRI site of pUc, 19 to create plasmid pUc19-P1g.

183 bp Ba1l containing the 5' end of the coding sequence
The EcoRI fragment was converted to pHc19-P! pUc18 isolated from g and digested with Sma I and [1con I
A Bam1l site was added to the 5' end. The resulting plasmid was incubated with BamHI and EcoRr.
and the 190 bpljIr piece was isolated.

Plasmid pUc19-P1g was incubated with EcoRI and Eco
Rvte was digested and the fragment containing the central region of the cDNA was isolated.

To obtain the 3' portion of the plasminogen cDNA, a fragment starting from the codon encoding amino acid 541 was inserted into plJ
It was subcloned into c118. The resulting plasmid was digested with EcoRV and X'bal and approximately 660b
We isolated the 3' plasminogen fragment of p.

Three plasminogen cDNA fragments (Bamll I
-EcoRI, EcoR [-EcoRV, and IE
coRV-Xbal) [lamll 1 and X
It was ligated with Zem219b digested with baI to generate plasmid 219b-Plg.

(Zem219b was derived by quenching 26m219a with Bamll I and Xba I, removing the 1-PAcDN8 sequence, and ligating the vector fragments with the [lamll IXba I adapter.)
Next, the Bamtl 1 plasminogen fragment was transferred to Zem2
19b - removed from Pig and Bam1l I
was inserted into Zem228 digested with To create Zem228, Zem67 was combined with l1indlII and Bg
Digested with lTI, pSV2neo (South
ern and Berg+J.

Mo1. A1. Genet, 1:327
-341.1982) H4nd III Ba
The mHI neo+5V40 terminator fragment was inserted. The resulting vector was named 20m220.

Plasmid Zem93 was digested with 5stl and the upstream MT-
1 sequence and recircularize the vector to create Zem1
06 was produced. Zem220 was digested with EcoR[,
The fragment containing the expression unit of the SV40 promoter neo-SV40 terminator was then recovered and ligated to EcoRl-digested Zem106. Z
The resulting hector, designated em223, contained the SV40 and MT-1 promoters in opposite orientations. Zem223 was digested with BamHI and 2
37bp Bcl I-Ram) I I SV40
A terminator fragment was inserted. Z the resulting plasmid
It was named em228 (Figure 3). Plasminogen expression hector derived from Zem228 was transferred to Zem228P.
It was named lg.

Zem228-Pig, essentially Neumann
AAT expressing cell line 235-6 was transfected by electroporation as described by (supra). Colonies were selected and measured for plasminogen production by enzyme-linked immunosorbent assay (EL[SA).

In this measurement, a goat polyclonal antiserum (American) against human plasminogen was used as the capture antibody.
Diagnostic force) was used. Immunoreactive substances were detected with a biotinylated rabbit polyclonal antibody against human plasminogen and avidin-conjugated horseradish peroxidase.

Ten positive clones were placed into 6-well Mi tissue culture dishes. When cells were 80-90% confluent, they were placed in cysteine-deficient medium for 1 hour. Each well (
3 of 50 microcuries in 1 ml culture volume/well)
53-cysteine was added, cells were incubated overnight, and the medium was harvested for measurements.

Cells were washed with cold PBS and extracts were prepared for measuring cytoplasmic plasminogen.

Cells were soaked in 1 μl of RIP A buffer (10 mM T
ris (pH 7,4) 1% deoxycholate, 1% Triton O. 1% SDS, 5mM EDTA, 0.7M
(NaCf)). Cell lysates were freeze-thawed twice on dry ice and at 10.00 rpm.
The mixture was centrifuged for 15 minutes while cooling. Next, radioimmunoprecipitation was performed using the supernatant.

Media and cell extract samples were measured for plasminogen by radioimmunoprecipitation. Sample (0.1 ~
1. (in a volume of 0 ml) and 5-10 μl of the immune system.
Mix with supernatant and incubate for 1 hour on ice, and then incubate 50 to 50 minutes of Staphylococcus aureus (
Stah lococcus aureus) (Pansorbin, Sigma Chemical Co., St. Louis, Mo.
) and incubated on ice for 1 hour.

After centrifugation, the supernatant was mixed with rabbit polyclonal antiserum against human plasminogen (Heringer Mannheim) 5I and incubated on ice for 1 hour. 50μ! of Staphylococcus aureus was added and the mixture was incubated on ice for 1 hour. Pellet the cells and divide the pellet into 0.5
%NP-40 and 0. 1m containing 1% SDS
Washed with 1 liter of PBS. Pellet the washed cells and add 50 μl PBS + 50 μl 2X loading buffer (0.1M Tris (p! 16.8)
, 16% glycerol, 3.2% SOS, 8% β-
Mercaptoethanol, 0.002% bromophenol blue), boiled for 5 minutes, and electrophoresed on a 10% polyacrylamide gel. Proteins were visualized by silver staining and subjected to autoradiography.

Measurements showed that AAT-producing cells secreted full-length plasminogen into the medium. In contrast, control BHK cells (i.e. cells transfected with 219bP1g but not Zem235) did not secrete detectable amounts of plasminogen and was degraded into the cytoplasm. It contained plasminogen (Figure 7).

cONA encoding a portion of human protein C was
Prepared as described by Sler and Davie (supra). In summary, λgtll cDNA
A library was prepared from human liver mRNA according to a conventional method.

Clones were screened using an affinity-purified, 12S1-labeled antibody to human protein C and plate lysate assay (M
aniatis et al., supra) followed by hand generation on a cesium chloride gradient. Phage were prepared from positive clones. Several cDNA DNA inserts were generated using EcoRI and plasmid pUc9 (Vieira
and Messing, Gene, 19.259
268°1982). Restriction fragments were added to phages Hector M13+y+plO and M13m.
pH (Messing.

Meth, in [nz molo, 101
:20-77, 1983) and the dideoxy method (Sanger et al., Proc. Natl.
, ^cad, sci, UsA, 74: 5463
54671977). Known partial sequence of human protein C (Kisiel, J, Cl1n,
InvesL, G4: 761769, 1979
) and contains the DNA corresponding to amino acid 64 of the light chain.
A clone was selected that encodes protein C starting from and extending through the heavy chain to the 3'-uncoated region.

This clone was named λ1lc1375. A second cDNA clone encoding protein C starting at amino acid 24 was identified.

The insert from this clone was subcloned into pUc9 and the plasmid was named pHcλ6L. This clone coats the majority of protein C,
Contains heavy chain coding region, termination codon and 3'-non-coding region.

cDNA from λlIC1375 was converted into α-32P dNT
nick translation using P and this was used to perform nick translation using Phage Sharon 4A (Maniatis
Benton and Davi et al.
s (Science 196: 181-182.1
Woo of the plaque hybridization method of 977)
CMe±L 川”dan dan yi+rule: 381-395.1
The probe was performed using a modified method of 979). Positive clones were isolated and plaque prepared (Foster et al., Proc. Natl., 8 cad.).

Sci, USA, 82: 4673 4677, 1
985). Phage DNA extracted from positive clones
(Silt + avy, etc., upper ≧ y Lρ) 1: j-me
nts with Gene Fusion, Cold Spring Barber Laboratory-11984
) was digested with 1EcoRI or BgIn and the genomic insert was purified and subcloned into pUc9. subcloning the inserted restriction fragments into the M13 vector and sequencing to confirm their origin;
They then established the complete DNA sequence of the gene.

Amino acids 42-- of protein C pre-propeptide
A genomic fragment containing exons corresponding to 19 was isolated and nick translated and 1lep
Gubler and H. using mRNA8 from G2 cells.
offman (Gene, 25: 263 26
A cDNA library prepared by the method of 9.1983) was used as a probe for screening. This cell line is derived from human hepatocytes and has previously been shown to synthesize protein C (
Fair and Bahnak, Blood, 64:
194-204.1984). E of fan λgtll
Ten positive clones containing cDNA inserted into the coR1 site were isolated and screened with an oligonucleotide probe corresponding to the 5'-noncoding region of the protein C gene. One clone was also positive using this probe and its entire nucleotide sequence was determined. This cDNA consists of 70 bp of 5
'-untranslated sequence, the entire coding sequence of human prepro-protein C, and all three corresponding to the second polyadenylation site.
'-contained non-coding regions.

B, Vector D5-11 As shown in Figure 8, from pDHFRIII to vector p
D5 was induced. pDHFRm (Berkner and 5harp+Nuc, Ac1ds Res, 13:
841 857.1985), the Pst1 site immediately upstream from the DIIPR sequence was converted to Bam111 as follows.
It was converted into a part. 10 μg of plasmid was incubated with 5 units of Pstl for 10 min at 37°C in 100 μl of buffer A (10 mM Tris (pH 8), 10 mM
Mgcg,.

Digested in 6mM NaCl, 7mM β-MSI+). DNA was phenol extracted, ethanol precipitated,
and 10mM dCTP and 16U'-7T of T4.
40 pl of buffer B containing DNA (50mM
Tris (pH 8), 7mM MgCfz.

resuspended in 7mM β-MSI+) and incubated at 12°C.
The ink was heated for 60 minutes. After ELOII precipitation, the DNA was dissolved in 14u of Buffer C (10mM Tri
s (pH 8) 10mM MgCl! ,z, 1mM
DTT, 1.4. 2.5 μg of kinased Bamtl [linker] was ligated for 12 hours at 12°C in mM ATP). After phenol extraction and Et011 precipitation, the DNA was diluted with 120I solution D [75tnM
KCI! , 6mM Tris (pH 7,5),
4,9 containing p8R322 and Hector sequences.
kb DNA fragments (10 bags) were isolated from the gel and
Ligate for 2 hours at 12°C in rOttl's Buffer C containing 0 units of T4 polynucleotide ligase,
This was then used to transform E. coli H[1101. Positive colonies were identified by rapid DNA preparation analysis and plasmid DNA (designated pDIIPR') was generated in triplicate.

Next, plasmid ``P D 1'' was prepared as follows. First, 25tttr psV40 CpML
1 (Lusky and BoLchan, Nature
, 293: 79-81.1981) Bam11
SV40 DN digested with Bam1 (I) at one site.
A] was cleaved with 25 units of Boil in 100 µm buffer at 50°C for 60 minutes, then 50 units of BamHl was added and incubated for an additional 60 minutes at 37°C. did. Plasmid pDIIFR' was linearized with BamHI and treated with calf intestinal phosphatase. DNA fragments were separated by agarose gel electrophoresis and 4,9
kb(7) pDl (FR' fragment and 0.2kb SV
40 fragments were isolated. These fragments (200 ng pD
HFR' DNA and 1100n SV40 DNA)
was incubated for 4 hours at 12°C in 104 buffer C containing 100 units of T4 polynucleotide ligase, and the resulting construct (pDl) was used to transform E. coli RRI.

As shown in FIG. 8, in the pBR322iJt region
“Poison” sequences (Lusky and Bo
plasmid pD by removing the
1 was modified.

Plasmids pDl (6.6 μg) and pML-1 (L
usky and Botchan, supra) (4 μg) at 5
0μ! The cells were incubated with 10 units each of EcoRI and Nru I in buffer A for 2 hours at 37°C and then subjected to agarose gel electrophoresis. 1.7k
pD11 of b! Fr piece and 1.8 kb pl'lL −
1 fragment was isolated and incubated for 12 hours in 20 barrels of buffer C containing 100 units of T4 polynucleotide ligase.
Ligate for 2 hours at °C and then transform into E. coli IIIBIOI. Colonies containing the desired construct (referred to as ppDl) were identified by rapid preparation analysis. Next, 10 bags of ppDl were added to 20 units of EcoR.
I and Bgllll at 37° in 50 μm of buffer A.
The mixture was digested at C for 2 hours. The DNA was electrophoresed through agarose and the desired 2.8 kb fragment (fragment C
) was isolated.

To generate the remaining fragment used to create pH5, pDHFRIII was transformed into Sac II (Sst
II) site was converted to either the l1indI [[ or the Kpn 1 site.

10 bags of pDllFRI[I and 20 units of 5st
II for 2 hours at 37°C, followed by phenol extraction and ethanol precipitation. The resuspended DNA was incubated for 12 min in 100p7 buffer B containing 10mM dCTP and 16 units of T4 DN8 helimerase.
Incubated for 60 minutes at °C, phenol extracted, dialyzed, and ethanol precipitated. DNA (5 cods) was mixed with 50 ng of kinase-treated ll1nd[I linker or Kpn I linker and 400 units of T4.
Ligations were performed for 10 hours at 12°C in 20I11 buffer C containing DNA ligase, phenol extracted, and ethanol precipitated. 50μ! After resuspension in Buffer A, the resulting plasmids were digested with 50 units of old nd■ or Kpnl (as appropriate) and electrophoresed through agarose. D made small moon 1 by gel
N A (250 ng) to 400 units of T4'DN
A 30 μl buffer containing ligase? l, inside 12
Ligated for 4 hours at °C and used to transform E. coli RRI. The resulting plasmid is p
DHFRI[I (Ilindl[) and pDllFR
It was named UI (pn I). Next, pDIII'
Rm (Kpn I) as IEcoRI and Kpn I
The 0.4 kb EcoRIKpn I fragment (fragment A) was purified by digestion and agarose gel electrophoresis.

The SV40 enhancer sequence was transformed into pDI['R
[Inserted into (llind m). SV of 50 bags
40 DNA was incubated with 50 units of old ndI[I in 120 d of buffer A for 2 hours at 37°C and the l1indIII CSV40 fragment (517
1-1046bp) was gel purified.

Plasmid pDIIFII III (Hind n
I) (10 pg) was treated with 250 ng of calf intestinal phosphatase for 1 hour at 37°C, phenol extracted and ethanol precipitated. Linearized plasmid (50 ng) was added to 250 ng of old ndlll CSV.
40 fragments were ligated using 200 units of T4 polynucleotide ligase in L6pl's buffer C for 3 hours at 12°C and transformed into E. co. January 18101. 0.9kb KpnI-BglI
I fragment (Fragment B) was isolated from this plasmid.

For the final construction of pD5, fragments A and B (50 ng each)
and long fragment C were ligated using 200 units of T4 polynucleotide ligase for 4 hours at 12°C, and the mixture was 1-transfected into E. coli RRt. Detect positive colonies by rapid preparation analysis;
Then, large-scale preparation of pD5 (Fig. 8) was performed.

C. Expression of Ebec-594 Protiin C cDNA was achieved using vector pDX. This vector was derived from pDll and pD5'. Plasmid pD5' contains the SV40 polyazonylesion signal (i.e. 5VtOBamHI (
2533 bp)-Bcl4 (2770 bp) fragment) is in the late orienta Lion.

That is, pD5' has Bam1 as the gene insertion site.
l Contains an I site. The plasmid pill contains the former nd III (5171 bp in the SV40 genome) Kpn [(294 bp in the SV40 genome) containing enhancer sequences.
p) It differs from pD5 in that the fragment is in the opposite direction.

To generate pDX, EcoRI cleavage, incubation with S1 nuclease and l1cl I linker
By ligating the IEcoR1 site in pDll with B
Converted to cl 1 site. DNA was prepared from positively identified colonies and a 1,9 kb Xho I-PsL I fragment containing the altered restriction site was prepared by agarose electrophoresis. In the second variant,
Ec kinase-treated pD5′ cleaved with Rcll
oRI-Bcl I adapter (oligonucleotide ZC525: 5′-GGAATTCT-3′
, and ZC526: 5'GATCAGAATTC
C-3') to create an ε (oR1 site) as a site for inserting the gene into the expression vector.
The oI-PstI fragment was isolated. The above two DNs
The A fragment was incubated with T4 DNA ligase and transformed into E. coli HBOI, and positive colonies were identified by restriction analysis. Then p
A DNA called DX (Figure 9) was prepared. This plasmid contains a unique EcoR1 site for insertion of foreign genes.

Next, protein C cDNA was inserted into pDX as an EcoRI fragment. Recombinant plasmids were screened by restriction analysis to identify a plasmid with the protein C insert in the correct orientation relative to the promoter element, and this plasmid (designated pDX/PC) was prepared from the correct clone. Because the cDN^ insert in pDX/pc contains an ATG codon in the 5'-noncoding region, this cDNA
Deletion mutation was performed on NA. Three base pair deletions were performed according to standard methods of oligonucleotide-directed mutagenesis. The pDX-based vector containing the modified cDNA was named p594.

The S. cerevisiae-XEX2 gene was isolated from a yeast genomic library by screening transformed mutated cells for the production of α-factor halos on appropriate test cell rounds. One clone was obtained that complemented all the reported defects of the 1ex2 mutation (mating, α-factor production, killer toxin maturation, and sporulation in kex2 homozygous diploid strains). pU this gene
The vector was subcloned into C hector under the control of the yeast GALL promoter. The resulting plasmid, designated p1515, has been deposited with the American Type Culture Collection under ATCC 67569. As shown in FIG. 10, p1515 was digested with ll1ndlII,
A 2.1 kb fragment was then recovered. This fragment is the old ndl
pUc18 cut with l and ligated together to create plasmid pUc
1B/KEX2 was produced. Next, ■ Cheng fragment (2,1k
b) from εx2 to ptlc18/, and convert this to the old ndlI
[Isolated by partial digestion with [] and complete digestion with BamHI. Next, the rest of the ■ order array is p
It was isolated as a 0.43 kb fragment from a Bam1+11 indllll digest of 1515.

The two KIEX2 fragments were then ligated into the BamHI site of vectors Zem228 and Zem229 (Figure 10). (Zem299 is similar to Zem228, but contains the DIIPR gene instead of the neomycin resistance gene. The MT-1 promoter and Sv40 terminator are present on either side of the cloning site.) The resulting plasmids were transformed into KEX2/Zem228 and K, respectively.
It was named EX2/Zem229.

The BHK cell line tk-ts13 was transfected with plasmids p594 and pSV2-DIIPR'/7 by the calcium phosphate method. Transfected cells were selected with 250 nM methotrexate (MTX) and clonal cell lines were plated in single dishes. A clonal cell line secreting protein C at 1.5 pg/cell/day into the medium was selected and PC594-204/BI
It was named IK.

10 pg of KIEX2/Zem22B was transfected into PC594-204/IIHK cell line by calcium phosphate method. Cells were constantly exposed to 250 nM MTX
cultured in the presence of 500 g/rrdl of clones
G41)3 and 12 clonal cell lines were selected.

'MtR, clones were incubated at 353° C. in sostein-free MEM (Gibco) containing 1% fetal bovine serum.
- 24 hours of pulse healing with cysteine. Media was collected and tested for the presence of single-chain Protein C and cleaved double-chain Protein C by immunoprecipitation using a monoclonal antibody against Protein C. 25
0 μl of medium was mixed with 108 g of antibody and the mixture was incubated for 1 hour at 37°C. I00
#5 taph A cell suspension (Pharmacia, Bispower Way, NJ) was added and incubation was continued for 1 hour at 37°C. Cells were pelleted by centrifugation and the pellet was resuspended in TBS.

Pellet the cells again and transfer the pellet to 1% β
-Resuspended in 60 μl of gel buffer containing mercaptoethanol. Heat the suspension for 3 minutes,
Next, electrophoresis was performed on a 5DS-polyacrylamide gel. Proteins were visualized by autoradiography.

The parental cell line pc594-204/BHK exhibited approximately 70% protein C in single chain form, with the remaining 30% in double chain form. PC, one of the cell lines selected in G418
594-204/KEX2-1 produced 95% double-stranded protein C, with the remaining 5% in single-stranded form.

E. Broin C Processing - Main Product To enhance the processing of protiin C into the double-stranded form, two additional arginine residues were introduced into the protein to create a cleavage site consisting of four basic amino acids. Obtained. The resulting mutant precursor of protein C was named PC962. This has the sequence Ser-His-LeuArg-Arg-Lys-Δ at the cleavage site between the light and heavy chains.
Contains in rg-8 sp. Processing at the rg-Asp bond results in a double-stranded protein C molecule.

The cloned cDNA was digested with mutagenic oligonucleotide ZC962 (5'-AGT CACCTG AG
Site-directed mutagenesis (Zoller and Smlth,
DNA 3:479 4B8.1984). 5stl of plasmid p594
The approximately 87 bp fragment was cloned into M13mpH and single-stranded template DNA was isolated. After mutagenesis, correct clones were identified by sequencing. Replicated DNA was isolated and digested with 5stl to isolate the mutated fragment, which was ligated with Sst-cut p594 by a three-part ligation. Clones with the 5stl fragment inserted in the desired orientation were identified by restriction mapping. The resulting expression vector was transformed into pDX/PC.
It was named 962.

Plasmid pDX/PC962 to pSV2-DIIF
II (Subraman i et al., Mo1. Ce1l
Biol., supra: 854-864.1981) and tk-ts13 BIIK cells using the calcium phosphate method (Graham and van der Eb, supra.
(as essentially described by). The transfected cells were treated with 10% fetal bovine serum, IXPsN antibiotic mixture (Gibco
6005640), 2. OmM was enhanced in Dulbecco's Modified Eagle's Medium (MEM) containing L-glutamine and vitamin K (5 μg/ml).

Cells were incubated in 250 nM methotrexate (MTX) for 14 min.
Colonies obtained were selected for days and the resulting colonies were subjected to immunofilter assay (McCracken and Brown,
BioTechniques, 82-87.1984
Screening was conducted in March 2016/Re-Mon). Plates were rinsed with PBS or serum-free medium (Dulbeco + Penicillin-Streptomycin, 5 μg/mfl Vitamin K). Next, use Teflon mesh (Spectrum).
Medical Industries, Los Angeles, CA)
was placed on the cells. After 4 hours incubation at 37°C, filters were removed and 50mM
Tris (pH 7,4).

5mM EDTA, 0.05%NP-40,150
1.mM NaCff in 0.25% gelatin for 30 minutes at room temperature. Filters were incubated in a biotin-labeled sheep polyclonal antibody against protein C (1 μg/rnl in the same buffer) for 1 h at room temperature with shaking. Filters were then washed in this buffer and incubated in avidin-conjugated horseradish peroxidase (Heringer Mannheim) (1:1000 in the same buffer) for 1 hour at room temperature with shaking.

Filters were diluted with 50mM Tris-11Cj2 (pH
7,4), 5mM EDTA1mM NaCl,0
, 25% gelatin, 0.4% Sarkosyl (Sarkos)
yl), washed in 0.05% NP-40 and then in water, and color reagent (60 mg!IRP color former [
Bio-Rad], 20ml methanol, 1ooy hydrogen peroxide/100mj! 50mM Tris(p)17
．． 4), incubated in 180 mM NaCf). The reaction was stopped by transferring the filter to water. The six most strongly reacting colonies were picked by cylinder cloning and grown separately in 10 cm plates.

Protein C production levels from confluent cultures were measured by enzyme-linked immunosorbent assay (ELISA). Affinity-purified polyclonal antibody or heavy chain-specific monoclonal antibody against human protein C (0.1M N8
1 pz/ml in 2CO,J (pH 9,6)? 8
Liquid 1OOj! ! ] was added to each well of a 96-well microtiter plate and the plate was incubated overnight at 4'C. Next, the wells were filled with 0.05%
PBS containing Tween-20 (5n+M phosphate buffer M (pH 7,5), 0.15M NaCf)
and then a solution of 1% bovine serum albumin, 0.05% Tween-20 in PBS 10
Incubated with 0 m. Next, put this plate on P
Rinse several times with BS, air dry, and store at 4°C. 100μ to measure the sample! Each sample was incubated in the coated wells for 1 hour at 37°C, and the wells were incubated in PBS at 0.05 °C.
Rinse with %Tween-20. The plates were then treated with 1% bovine serum albumin and 0.05% Tween.
-20 in PBS containing biotin-conjugated and affinity purified sheep polyclonal antibody to protein C for 1 hour at 37°C. Rinse the wells with PBS;
and 1% bovine serum albumin and 0.05% Twee.
Reincubate with avidin-conjugated alkaline phosphatase in PBS containing n-20 for 1 hour at 37°C. The wells were then rinsed with PBS and alkaline phosphatase activity was determined by the addition of 100 m of phosphatase substrate (Sigma 104; 600 μg/mf) in 10% jetanolamine (pH 9,8) containing 0.3 mM Mg (1!2). was measured at 405 nm.
The absorbance at was read on a microtiter plate reader. The results are shown in Table 1.

"edge" - friend 962 = 1 1.1 2500
2.202 0.8 1250
1.563 1.2 1
350 1.124 1.2
550 0.46-5 1.
2 1550 1.30-6
The 1.2 950 0.80 clone BIIK/962-1 was grown in large scale culture and the sheep polyclonal antibody 71T1 against human protein C was added to 2 mg of CNBr-activated Sepharose 4B (Pharmacia, Biochemistry, Biochem. Hundreds of milligrams of protein C were purified by affinity chromatography on a column prepared by coupling to NJ).

Apply cell culture medium to the column and coat the column with roomR's TBS (50mM Tris (pH 7,5), 15
1fJi? in TBS or pHLL containing 3M KSCN, 5) 1fJi? & (
25mM calcium phosphate (pH 11,5), 0.
2M NaCl, 2% Tween 80.0.5%N
Protein C by [aNff]? I rolled an 8. Western flow analysis shows that approximately 95% of the mutant protein C is in double-stranded form, as opposed to protein C obtained from BHK cells transfected with the original sequence. Approximately 20% were double stranded.

A stable HK cell clone expressing the PC962 mutant protein or a stable HK cell clone expressing wild-type protein C.
Milligram quantities of protein C were purified from three cell clones (p59.4-transfected cells) using a monoclonal antibody column specific for the calcium-3A J' conformation of protein C.

Cell culture medium was added to 5mMCa (, e.

was applied to the column in the presence of Protein C was removed from the column with TBS containing 10mM EDTA. I rolled an 8. Using this purification method, it was possible to purify fully active protein C without subjecting it to denaturing conditions. Purified protein C was purified by SDS/PAG.
E and subsequent silver staining and showed greater than 95% purity.

The PC962 protein produced by BHK was determined for its ability to be activated into a form exhibiting amytolytic and anticoagulant activity. Affinity-purified protein samples were extensively dialyzed against TBS and 0.05%.
Activation was performed for 1 hour at 37°C with 1 volume of 1 unit/d Protac C (Amersham Diagnostics). 1 of 100 Ill in a microtiter well
mM Protein C Substrate (Spectrozyme PC
Amitrysis activity was determined by adding an aliquot of the activation mixture to a Amersham Diagnosis test (A) and measuring the temporal change in ① using a microtiter reader. The anticoagulant activity of the activated protein was determined as described by Sugo et al. (supra). Affinity purified PC9
The 62 protein was shown to be fully active in both amitrysis and anticoagulation measurements. pH11.5
Elution from the antibody column with buffer was shown to yield proteins with higher activity than that obtained using 3M KSCN elution.

Clonal cell lines obtained from transfection of pDX/PC962 into B I (K cells) were isolated by limiting dilution. Plates of MTX-selected colonies (approximately 300 colonies) were trypsinized and counted. , and replated into microtiter wells at an average of 0.5 cells/well. These were grown in selection medium containing 250 nM MTX.

Approximately 50% of the wells contained colonies. Wells containing recognizable colonies (1-2 mm diameter) were measured for protein C levels in the medium by ELISA. For this measurement, fresh medium was added to all wells, the plate was incubated for 75 minutes, then the medium was removed and measured. 75 of 50 ng/m or more
- Minute accumulation (equivalent to more than 1000 ng/ml/day)
Large-scale culture was performed by dividing 5 colonies giving 100 μl into 10 cm plates.

Protein C production levels in these colonies ranged from 1,1 to
It was 2.8 pg/cell/day.

A second plasmid, designated PC229/962, was created by inserting PC962 cDNA into plasmid Zem229. PC962 from pDX/PC962
The EcoRI fragment containing the cDNA was
It was ligated with Bam11 I synthetic oligonucleotide adapter to Zem229 which had been cut with BamHI and treated with phosphatase. The resulting vector is PC229/962.

Plasmid PC229/962 was transfected into BHK cells by the calcium phosphate method. This cell,
Cultured in Dulbecco M'EM containing 10% calf serum and 5 fish/ml vitamin K. The 48 hour transient expression level from this transfection is approximately 25 ng
/ml. After 2 days, the transfected cells were split into selective medium containing 250 nM MTX and
Then, the cells were further cultured for 14 days. Three plates from this transfection, each containing approximately 200 colonies, were screened by immunofilter assay and the 24 most strongly reacting colonies were picked by cylinder cloning. These separately 1
were grown in 0 cm plates and their protein C production levels were measured. Colonies producing 1.1-2.3 pg/cell 7 days were used for the preparation of stable protein C producing cell lines.

0 in expression vector pDX/PC962 and plasmid p
neo was co-transfected into 29311I cells by the calcium phosphate method. Transfected cells were split after 48 h into medium containing 500 pg/ml G418. A 10-day munofilter assay was performed on selective media, and two clones were picked up by cylinder cloning. Protein C production was found to be 1-2 pg/cell/day. Cultures were scaled up and protein C was purified by immunoaffinity chromatography. 95
% of protein C was found to be in double-stranded form.

The structure of the 962 mutant protein prepared from BHK and 293 cells was compared to that of wild type protein C from 293 cells and plasma. Analysis by SDS/PAGE and subsequent silver staining revealed that all recombinant proteins contained heavy and light chains, which co-migrated with those of plasma proteins. The mutant protein produced in either cell type contained significant amounts (approximately 20%) of unprocessed -terminus protein (molecular weight 66,000) of wild-type protein C synthesized in 293 cells. was essentially completely processed into double strands. N-terminal sequence analysis showed that both the light and heavy chains of the recombinant wild-type protein and the BIIK/PC962 mutant protein were properly processed. The degree of T-carboxylation of the recombinant protein was
Measured by two different ELISA systems. The first system recognizes both the gamma-carboxylated and non-carboxylated forms of the protein, and the first system recognizes gamma-carboxylated and non-carboxylated forms of the protein, and the first system
A specific antibody that recognizes only protein C that has undergone a la-induced conformational change is used. Analysis shows that 60% of the recombinant protein C produced in BHK cells and 9% of that produced in 293 cells
0-95% was T-carboxylated to a degree sufficient to be recognized by specific antibodies.

3M! Cervical recombinant proteins were also analyzed for amidolysyl activity and anticoagulant activity, and the results were compared to plasma protein C activity. PC9G2 and 29 from BHK cells
Both wild type protein C from 3 cells showed sufficient amidolytic activity. Is protein C from BHK cells in blood in anti-susceptibility assay? It had virtually the same specific activity as protein C, while both the wild type protein and the PC962 mutant protein from the 2931.111 cells showed approximately 15% higher specific activity. One unit of protein C activity is 1
Defined as the amount in normal human blood f,1,p containing 4H protein C per ml (Gard
iner and Griffin
o1. 13: 265 278.1983) (specific activity=250 units/II1g). Wild-type protein C produced in 293 cells was consistently 300 units).
/mg exceeded.

F, activation peptide of protein C
The protein C cDNA sequence was changed by site-directed mutagenesis to remove the portion encoding e). The modified sequences were then transfected into BHK and 293 cells and stably transfected cells were selected. Active protein C was detected in culture fluid samples from both cell lines.

To remove the activating peptide coding sequence, plasmid p594 was digested with SsL and the approximately 880 bp fragment was purified and purified with M13mplO (Messing,
Methods Oka Mutoshi 101: 20-77, 19
It was inserted into the SsL1 site of 83). The 12 activation peptide codons were mutagenic oligonucleotide ZC829 (
5'-CTG AAA CGACTCATT GAT
-3') using oligonucleotide-directed deletion mutagenesis (Zoller and Smith, DNA 3
:479-"488.1984). Double acid form DNA was prepared from the mutant phage clones and digested with 5stl. Protein C fragment (approximately 840
p594 was isolated (bp) and digested with 5stl.
inserted into. The resulting plasmids were screened for the correct orientation of the 5stl fragment by restriction mapping using Bgl II. The correct plasmid was selected and named pPc829. Plasmid ρPC829 was sequenced to confirm the presence of the desired coding sequence.

Plasmid pPC829, plasmid psVDIIF
RT (Lee et al., Nature, 294:2
28 232.1982) to tk-B11 cells, and pKo-neo (Southern and B
erg, J., Mol.

"Illusion" ↓Tokusako μMiyu, Vol. 327-341.1982)
Calcium phosphate coprecipitation method (Gra
ham and vander Eb, supra).

After 48 hours, medium was harvested and measured for protein C by ELISA. The results are shown in Table 2. At the same time, the culture was mixed at 1:5 with 500 μg/ml of 641
8 (293 cells) or 250 nM methotrexate (Lk-B11 cells). After 10 days in the presence of selective medium, stably transfected colonies were screened for protein C production by immunofilter assay.

Pick positive colonies and use selective medium (500x/m!
G418 or 250 nM methotrexate) for 10 days. The culture solution was colored using a coloring method.
Measured for PC activity. 50mM Tris (p
H7,5), 0.2mM in 150111M NaCQ
Spectrozyme P
Ca (Amersham Diagnostica #336)
Media samples were added to microtiter wells containing looms. Plates were incubated at 37°C and A4° at various time points.

was measured. Representative results from one transfected 293 cell line (referred to as 829-20) are shown in the first
Shown in Figure 1. Media from positive colonies of cell line 829-20 consistently showed higher activity with chromogenic substrates for APC compared to control media incubated with non-transfected 293 cells over the same period (10 days). showed that.

Table 1 Transient expression of activated protein C (ELISA)
BHK 2.7 Processing site sequence Arg-^rg-Lys-^rg
DNA encoding activated protein C having
The sequence was generated by mutagenesis of the wild type protein C sequence. The resulting sequence (designated 1058) was similar to that encoding PC962, but lacked the region encoding the activation peptide.

The protein C sequence present in Plasmid 594 was altered by a single mutagenesis to remove the activation peptide codon and insert an Arg-Arg codon at the processing site. Oligonucleotide ZC1058 (5'
-CGCAGT CACCTG AGA AG8 8＾Δ CGA CTCATTGAT GGG-3
), mutagenesis was performed on the 870 bp Sst E fragment from p594 essentially as described above.

Expression vector pDX/P using mutagenized sequences
C1058 (similar to pDX/I'C962) and transformed this vector into BHK as described above.
cells were co-transfected. The protein was purified on a polyclonal antibody column by elution with pH 11, 511 buffer.

The activity of PC105B protein was compared to that of activated plasma protein C and activated PC962.

Plasma protein C and PC962 (5n/In1) at 1
The cells were activated by treatment with 1/10 volume Protac C (Amersham Diagnostics) for 2 hours. 504
Anticoagulant activity was measured by mixing 50 p1 of human plasma with 50 p1 of activated protein C and incubating the mixture for 150 seconds at 37°C. To this mixture was added 50 μl of activated cephaloplastin (Amersham Scientific Products, McGarrah Park, IL) and the mixture was
Incubate for 300 seconds at 7°C. 100J
l! of 20mM CaCQ2 was added and the clotting time was recorded. The data are shown in Figure 12.

G, BHK Rio-7 (pDX/PC1058) transfected with pDX/PC1058, a highly productive protein C and KEX2
3//BIIK) and transfected with KEX2/Zem229 by the calcium phosphate method.

Transfected cells were selected with 500 μg/ml G418 and 250 nM methotrexate 1.

KEX2-1058//BIIK) J?
The 353-cysteine was pulse labeled with 353-cysteine in cysteine-poor DMEM containing 1% fetal bovine serum for 24 hours. Media was collected and tested for the presence of single chain and cleaved double chain activated protein C by immunoprecipitation using a monoclonal antibody against protein C. 250 I! 1 medium was combined with 10 μg of antibody and the mixture was incubated for 1 hour at 37°C. 100μ! 5taph A cell F='i? ffl (Pharmacia, Bisworth Way, NJ) was added and the mixture was incubated at 37°C for 1 hour. Cells were pelleted by centrifugation and pelleted with 1% β-
60 μl of gel buffer containing mercaptoethanol
(re-suspended). The suspension was heated to 100°C for 3 minutes and electrophoresed on a 5DS-polyacrylamide gel. Proteins were visualized using autoradiography. The KEX2-1058//BIIK clone showed approximately 100% cleavage of the protein into double strands.

Dulbecco MEM supplemented with 10% fetal bovine serum, 250 nM methotrexate and 500 i/ml G418
KEX2-1058/ that grew to confluency in
Activated protein C was produced from the /BHK clone. Confluent cells were incubated with 1 μg/ml febronectin, 2 μg/ml insulin, 5 n
/ mQ transferrin, 5 μg / ml vitamins, IXPSN antibiotic mixture (Kipco 600-564
0), switched to Dulbecco's MEM supplemented with 2.0 mM L-glutamine, 250 nM methotrexate and 500 μg/mQ G418. Media was collected every 1-2 days over 7 days and frozen to -20"C. Frozen media samples were thawed and 0.4
All cell debris was removed through a 5/ITrl filter. Solid calcium chloride was added to a final concentration of 5mM and solid sodium azide 0.02% (w/v)
was added to the final concentration. Activated protein C was removed from the culture medium using a monoclonal antibody column specific for the calcium-induced form of protein C. Apply the treated medium to the column and add 10mM EDT
Protein C was activated by TBS containing A) Iwata. Protein C concentration was measured by absorbance at 280 nm and ELISA.

Protein C activity was measured by coagulation assay.

Affinity-purified plasma protein C at 50m
M Tris+ 500:1 (APC:8C) in 100mM NaCl and 0.1% bovine serum albumin
University of New Mexico, Albuquerque, N, M, W, K15i diluted 1 in the ratio C-C)
Agekistroton contortrix contortrix (k.
contortrix) 37° with protease
The cells were incubated at C for 2 hours. KEX2-1058
//Noaffinity purified activated protein C from BIIK cells was incubated for 2 hours at 37°C. MLA Electra 800Coagul
ation Timer (Medical Laboratory)
Automation, Inc., Pleasantville, NY) clot formation in 4! II was determined. 100111 activated plasma protein C or KEX2-1058 activated protein C was added to the MLA cuvette and warmed for 50 seconds to raise the temperature to 37°C. 100μl Dad
e8ct to n FS (Amersham Scientific.

product) and incubate the test solution for 100 seconds. The time required for clot formation was measured. The results of coagulation measurement are as follows: KEX2-1058//B11
Activated protein C produced by cells in
It was shown to be approximately 100% more active than activated plasma protein C.

The carboxy-terminal sequence of the light 1'+ of EX2-1058 protein C and the carboxy-terminal sequence of the light chain of commercially available protein C (Amersham Tiagosti) were cleaved with CNBr at the unique methionine residue of the light chain. -Comparisons were made by releasing peptides that could be completely sequenced by terminal sequence analysis. 1
% fetal bovine serum, 250 nM methotrexate and 50
Darhe-yME supplemented with 0 μg/ml 6418
Protein C affinity-purified from KHX2-1058//BIIK cells grown in M.
Dithiothreitol (DTT) at a 10-fold molar excess (relative to CysrA groups) in M Tris-11Cj2 (pH 18,3) and guanidine+1 at a final concentration of 6.0 M
Reduction was achieved by addition of Cf. This mixture was heated at 65°C for 4 hours.
Incubated for ~6 hours. Iodoacetic acid (pH 17,
0) or iodoacetamide was added to the reduced protein in a 4-fold molar excess over the molar concentration of DTT, and the mixture was incubated at 37°C for 30 minutes. Q, IM N114) ICOz (pH 8,
5) was dialyzed at 22°C for 24 hours.

The dialyzed solution was applied to an IIPLC Po1y-F column (DuPont) to isolate the light chain. A 500-fold molar excess of CNBr per methionine residue was added to the purified light chain in 70% formic acid for 30 minutes at room temperature in the dark under nitrogen.

The CNBr digest was processed using an American Biosystems Model 470A sequencer (Marins-on-St.).

Croix, MN). The sequences obtained showed that the C-terminal sequences of both commercially available purified Protein C and KEX2-1058 proteins ended with Glu, and the light chains of both proteins were shown to end at amino acid 149.

Protein sequences of H, 1645Zem229R and plasmid p962 (Table 3; amino acids added to the wild-type protein C-encoding sequence are underlined, and spaces between amino acids indicate the light chain sequence and the heavy chain sequence. amino acids Ar instead of the DNA sequence encoding amino acids 9-11 of the activation peptide encoded by
A DNA sequence encoding g-Arg was provided. p164
Cells transfected with 5 secreted activated protein C into the medium. Plasmid p962 was digested with 5all and 5stl and 730bp
The fragment was purified and inserted into M13mplO, which had been digested and linearized with 5ail and 5stl. Synthetic oligonucleotide ffl'Zc1645 (5'-GAA
GACCAA ACA ACA AAA
CGGCTCATT GAT-3'), and ZC550 (5'-TCCCAGTCA CGA
CGT-3') by site-directed in vitro mutagenesis (Zoller and Smlth, supra).
The single-stranded template DNA prepared from the phage obtained above is mutated. Mutant phages were subjected to dideoxy sequencing to confirm mutagenesis. preparing replicative form (rf) DNA from a confirmed mutant clone designated 1645;
A 411 bp fragment is then isolated by digestion with 5stl and Pstl. Plasmid p229/962 (Example 3.E,
) was digested with EcoRI and Pst[ to isolate a 592 bp protein C fragment. Plasmid p229/962
was digested with EcoRI and Ss to isolate a 700 bp protein C fragment. 411bp from 1645rf
Protein C fragment, 411 bp from p229/962
The protein C fragment and the 700 bp protein C fragment were ligated by a four-part ligation to pZem229R, which had been linearized with EcoRI and treated with calf intestinal phosphatase to prevent self-ligation. (Plasmid pZem229R is similar to Zem229, except for partial digestion, blunt-ending with Klenow fill-in, religation, followed by (Bam1l I digestion and BamH
Religation using the I-EcoRl adapter results in E
The coR1 site is destroyed. The correct plasmid was selected and named pPc1645/229R.

2ε [Table WT (594) ~ S-Toshi-K-R-D-7829-
5-11-L-K-11 962-5-11-L-1? - Toni - R-D-T105
8 -1 to -3-H-L-R-+1-? -1645-
s-+TLR-11-1)-T-IE18
80 -3-11-1. -1? -1? - to -1? -D
-T1953 -5-11-L-1? -R-ni-11
1954-5-11-L-R-11-I? -D～E
-D-Q-E-[1-Q-V-D-P-1? -L-1-
D-1-D IE-D-Low E-[]-]Q-V-0-P-1? -L
-1-D---1-D- Low old E-D- (1~R-R- to -1?-L-1-DO
-1 to -G-124-? -L--D R4-ni-R-L-1-D Q-dan-i-ni-11-L-=I]--Plasmid pP
c1645/229R by calcium phosphate method (Graf
Lk- by +am and van der Eb, supra)
Transfected by BIIK cells. Transfected cells were subjected to selection with 111M methotrexate and the medium was IEL'd for protein C.
[SA determined by (Example 3.E,). Positive clones were grown in Dulbecco MEM supplemented with 10% fetal bovine serum and 1 ml of methotrexe-1 until cells reached confluency. 1 confluent cell
Switched to Dulbecco MEM supplemented with % fetal calf serum and 1 omethotrexate.

The medium was collected every 1 or 2 days for 7 days, and -
Stored at 20°C. Frozen media samples were thawed and filtered through a Q, 4.5 gn filter to remove all cell debris. Solid calcium chloride was added to a final concentration of 5mM and solid sodium chloride was added to a final concentration of 0.02% (w/v). Protein C was recovered from the culture medium using a monoclonal antibody column specific for the calcium-induced conformation of protein C.

Apply the treated medium to the column and 10mM 1:
Protein C was eluted with TBS containing DTA. Protein c? The a degree was measured by absorbance at 280 nm and by ELrSA.

activated protein C produced from cells transfected with ppct6s5/229 and PC229/962 produced from transfected cells.
The same amount of protein C and colorimetric measurement (Chrom.

Genetic assay). 40μl
1N of affinity purified protein C diluted in TBS + EDTA was added to each well of a 96-well plate. 40ttl of 2mM Spectr
ozyme PCa (Amersham Diagnostics, New York, NY) was added to each well and incubated at 37°C until full color development. 40
Activity was measured as increase in absorbance at 5 nm. This result indicates that activated protein C produced by cells transfected with pPc1645/229 was higher than activated protein C produced by pC229/962.
It was shown that the activity was 5 to 10% higher.

The DNA sequence encoding Protein C in PCl380 229R and plasmid 1645 was further modified to remove the first, second, seventh and eighth amino acids (Table 3) of the activation peptide. Prepare single-stranded 1645 template DNA and synthesize synthetic oligonucleotide ZC
1880(5'-AAA CGA GACACA G
ACC<GAAG8-3'), and was subjected to site-directed in vitro mutagenesis (Zoller and Smlth, supra) using ZC550. Positive clones were subjected to dideoxy sequencing to confirm mutagenesis. A positive clone was identified and named 1880.

The replicative DNA prepared from clone 1880 was
A 562 bp protein C fragment was isolated by digestion with PsLI and PsLI. EcoR plasmid PC229/962
The 'yoobp protein C fragment was isolated by digestion with I and Pstl. 411bp protein C fragment from 1880rf and 562b from PC229/962
The p fragment and the 700 bp fragment were ligated into pZem229R digested with EcoRI by a four-part ligation. Select the correct plasmid and ρPCl380/22
It was named 9R.

tk-BIIK plasmid pPc1880/229R
Cells were transfected and measured as described above.

J, PCl954 229R and plasmid 1
The coding sequence of the activation peptide present in 645 is modified to remove the second to seventh amino acid codons of the activation peptide; the first and eighth amino acid codons of the activation peptide in 1645 are fused. single chain 1645
Prepare template DNA and synthesize synthetic oligonucleotide Z
C1954(5'-GAG^AG AAA AC
G AGA CCA AAG AAG A1
AAC-3'), and subjected to site-directed in vitro mutagenesis using ZC550. A positive clone is selected and named 1954. The amino acid sequences of the light chain and heavy chain junctions of the encoded proteins are shown in Table 3.

Replicated DNA was prepared from clone 1954 and 5
A mutated protein C fragment of approximately 400 bp is isolated by digestion with stl and Pstl. Plasmid pPC229
/962 with EcoRI and PsLI, or 5stI and Ec
Digested with oRI to obtain each 562 bp Eco
The Rl-PstJ fragment and the 700 bp protein C fragment are single-lit. Approximately 400 bp protein C fragment from 1954rf and 5 from pc229/962
The 62 bp and 700 bp fragments are ligated into pZem229R digested with EcoRI by a four-way ligation. Select the correct plasmid and pPc1954
/ Named 229R.

Plasmid pPC1954/229R is transfected with tk-tuu[i] by calcium phosphate coprecipitation method.

Cells are selected and measured as described above.

K, the coding sequence of the activation peptide present in the control plasmid 1645 of PC1953229R was modified to remove the first to eighth amino acid codons of the activation peptide, and the first and second sets of Arg present in 1645 were modified. - Perform a fusion between Arg-Lys-Arg amino acid codons. Single-stranded 1645 template DNA was prepared and synthetic oligonucleotide ZC1953 (5'ACC
TCA G88 GAA AACGAA GAA
GAA AACGGCTC883'), and ZC
550 for site-directed in vitro mutagenesis. Sequence positive clones to confirm mutagenesis. A positive clone is selected and named 1953. The amino acid sequence at the light chain-heavy chain junction of the encoded protein is shown in Table 3.

Replicated DNA was prepared from clone 1953 and S
Digest with stI and PstI to isolate a mutated protein C fragment of approximately 4oobp. Plasmid PC2
29/962 as EcoRT k Pst I, or 5s
Digested with tI and EcoRI, each containing 562 bp of Ec.
The oRI-Pst 1 fragment and the 700 bp protein C fragment are isolated.

Approximately 400 bp protein C fragment from 1953rf,
and the 562 bp fragment from PC229/962 and 7
The 00bp fragment was added to pZem digested with EcoRr.
It is connected to 229R by fractional connection. The correct sequence is selected and named pPc1953/229R.

Plasmid pPc1953/229R is transfected into Lk-BllK cells by calcium phosphate coprecipitation method. Cells are selected and measured as described above.

Example: Plasmid FIX/pD2 (Busby, etc., Natu
Bam re, 316+271-273.1985)
Digested with tlI and recovered a 1.4 kb factor 1 fragment. This fragment was ligated with pD5' that had been digested with Bamll and treated with calf intestinal phosphatase. The resulting plasmid was transformed into l'lX/po! 5'
It was named.

In order to simultaneously express factor ■ and factor ■,
10■ FIX/pD5' 10μg F■ (56
52463) / pDX (llagen et al., EP
200.421; ATCC 40205) and one DllFRres-pD5'(pD5'-derived plasmid containing the methotrexate resistance DI! FR gene (Levinson et al., EP 177.0
Muni-BHK cells were transfected using Muni-BHK cells. Transfected cells were selected in the presence of methotrexate and measured for production of Factor ■ and Factor ■ by immunofilter assay using antibodies directed against both white matter. Colonies positive for both Factor ■ and Factor ■ were selected, expanded, and pulsed with 353-cysteine. Media proteins and intracellular proteins are immunoprecipitated using Factor 1 and Factor 2 antibodies and analyzed by polyacrylamide gel electrophoresis. Cells co-expressing Factor ■ and Factor ■ produced double-stranded Factor ■ (ie, Factor ■a). On the other hand, cells producing only factor ■ exhibited only the single-chain form of this protein.

S, cerevisiae (S, cerevisiae)
The BARI gene encodes a secreted protein known as Barrier L. The secretory peptide portion of the BARI gene product, or the secretory peptide plus third (C-terminal) domain of the barrier, can be used to promote secretion of foreign proteins produced in S. cerevisiae (MacKay, Wo 87102670,
and MacKay et al., European Patent Application No. 881.
16335.6).

B(1-29)Ala-Ala- as described in European Patent Application No. 88116335.6
1ys-8(1-21) insulin precursor (Mark
Ussen et al., EP 163,529; hereafter referred to as Ml-3
An expression hector was created containing sequences encoding the signal peptide and third domain of the barrier fused to the coding sequence of (referred to as ). psW195 (Figure 14)
This hector, called ., further contains the yeast TPII promoter and terminator. The barrier and insulin sequences are linked at the amino acid sequence +ys-arg. To enable processing of the fusion protein by thrombin, the lys-arg linkage sequence was
(mutated to H). Cleavage at this portion by thrombin will result in secretion of the unfused insulin precursor.

As shown in Figure 14, plasmid psW195 was transformed into sp
1, digested with hI and 5all, comprising a BAIll-insulin fusion and a TPII terminator;
A 7kb fragment is isolated. Add this fragment to sph in advance.
M13mp18 completely digested by 1 and 5all
Connect to. The resulting phage clone was called mpL8-ZV
It is called 172. Oligonucleotide (5'-TCC
TTG GATCCA AGA TTCGTT-3'
) using the uracil method (Kunkel, Proc, N
at, Acad, Sci, USA, 82: 488
492°1985) to mutate mp18 ZV172. The resulting mutants were sequenced to confirm mutagenesis and the positive clone was named ZV172/1083. For convenience, the insert present in ZV172/1083 was subcloned into pUc18. ZV17
1.7kb Sph I-Sa from 2/1083
l I insert was isolated and sph I and Sat
It was ligated into pUc18 which had been completely digested with I. The resulting plasmid pZV180 was completely digested with Sal I. The cohesive ends of linearized pZV180 were blunted with DNA polymerase I (Mlenow fragment) and ligated to kinased Bgl II linkers. Excess linker was removed by digestion with Bgl II. Next, the DN with the linker added
A was completely excised with Sphr and the 1.7 kb insert was isolated. Partial ■Dan Promoter, BARI -Old-
1.7 kb containing 3 fusions and the 1 terminator
The insert of plasmid S 207 sph I-B
pZV187 was created by ligating the amll1 partial promoter to the hector fragment. (pzy207 is p
cPOT (ATCC39685) was derived as follows. i.e. 2 micron sequence and pBR32
The 750 bp Sph I of pcpoT containing 2 sequences
- Bamll I fragment of 186 bp from pBR322 tetracycline resistance gene
Replaced with 1lI fragment; destroyed Sph [site and inserted NotI site into the tetracycline resistance gene;
The raw plasmid was digested with Not I and Bam1l I; and Not I Bam11 pB
Not I-Bamll in place of the R322-derived fragment
Insert the promoter fragment. ] Isolate the thrombin cDNA from the prothrombin cDNA cloned into the Pst1 site into pBR322 (Friezner-Degan et al., supra). p151
The dish gene from 5 (Figure 3) is digested and manipulated to remove the catalytic region and the remaining sequences are ligated to the thrombin cDNA. An expression unit is prepared by linking the hybrid gene to the TPII promoter and terminator. This expression unit is then placed in place of the coding region of the yeast BARI gene.

The resulting construct comprising an expression unit flanked by BARI gene non-coding sequences is used to transform S. cerevisiae. The transformed cells are cultured and screened for thrombin production by enzyme activity measurements. Colonies positive for thrombin production are then transformed with pZV187. The cells are cultured and insulin precursors are isolated from the culture medium.

Although the present invention has been specifically described above, various changes may be made without departing from the scope of the present invention.

[Brief explanation of the drawing]

FIG. 1 shows the production process of plasmid Zem99. FIG. 2 shows the production process of t-PA-expressing Hector Zem219. FIG. 3 shows the production process of plasmid Zem228. FIG. 4 shows the process of subcloning α-1-antitrypsin cDNA. FIG. 5 shows the production process of expression vector Zem235. FIG. 6 shows the nucleotide sequence of plasminogen cDNA along with the encoded amino acid sequence. FIG. 7 shows transfected BHK cells (a),
Figure 3 shows the production of plasminogen in transfected BHK cells (b) co-expressing α-1-antitrypsin and α-1-antitrypsin. Lane 1 is the result of the media sample and lane 2 is the result of the cytoplasmic extract. The arrow indicates the position of native plasminogen (91'd)
shows. FIG. 8 shows the process of constructing vector pD5. The symbols used in the figure are: O-1 is adenovirus 50-1 mapmilunit-E is SV40 enhancer; MLP is adenovirus 2 major late promoter; Ll-3 is adenovirus 2 tripartite l-(tr
ipartite) leader; 5' indicates the 5' splice site; 3' indicates the 3' splice site; p(A) indicates the polyazonylesion signal; and DIlr'R indicates the dihydrofolate reductase gene. FIG. 9 shows the process of constructing vector pDX. The same symbols as in Figure 8 are used. Figure 1O shows S, cerevis.
iae) The process of constructing an expression vector containing the KEX2 gene is shown. FIG. 11 shows the results of measurements of activated protein C in media samples from transfected 293 cells. Figure 12 shows the anticoagulant activity of activated protein C produced in transfected mammalian cells. Figure 13 shows that (
The results of radioimmunoprecipitation of Factor ■ produced by Lane 1) and Factor ■-transfected control cells (Lane 2) are shown. FIG. 14 shows the process of constructing a yeast expression vector containing a DNA sequence encoding a thrombin-cleavable fusion protein. °°old-3°” is insulin precursor D N
A sequence is shown. Figure 15 shows the nucleotide sequence of the synthetic aprotinin gene.

Claims (1)

[Scope of Claims] 1. A true DNA sequence that has been transfected or transformed with a first DNA sequence encoding a protein of interest and at least one additional DNA sequence encoding a protein that processes or stabilizes the protein of interest. Nuclear host cell. 2. The host cell according to claim 1, wherein the eukaryotic host cell is a mammalian host cell or a yeast host cell. 3. The first DNA sequence is t-PA, Factor VII
, Factor IX, Factor The host cell according to claim 1. 4. The host cell of claim 1, wherein the first DNA sequence encodes a serine protease. 5. The host cell according to claim 1, wherein the additional amino acid sequence is selected from the group consisting of a protease, a protease inhibitor, and a respective sequence encoding a protein that binds to the protein of interest. 6. The host cell of claim 1, wherein the first DNA sequence encodes Factor VII and the additional DNA sequence encodes Factor IX. 7. said first DNA sequence encodes protein C or activated protein C, and said additional D
The host cell according to claim 1, wherein the NA sequence is the yeast\KEX2\ gene. 8. the first DNA sequence encodes t-PA;
and the additional DNA sequence encodes a protease inhibitor selected from the group consisting of TIMP, trypsin inhibitor, and aprotinin. 9. Claim 1, wherein the first DNA sequence encodes plasminogen and the additional DNA sequence encodes α-1-antitrypsin or a variant thereof.
Host cells as described in. 10. The host cell of claim 1, wherein the first DNA sequence encodes protein C and the additional DNA sequence encodes protein S. 11. The host cell of claim 1, wherein the first DNA sequence encodes Factor VII and the additional DNA sequence encodes tissue factor. 12. A method for producing a target protein, claims 1 to 11
culturing the host cell according to any one of the above under conditions that allow expression of the first DNA sequence and the additional DNA sequence, and isolating the protein of interest from the host. How to become. 13. The method of claim 12, comprising transfecting or transforming the host cell with a plurality of vectors, each containing a separate expression unit, prior to said culturing step. 14. The method of claim 12, comprising transfecting or transforming the host with a single vector containing multiple expression units prior to said culturing step. 15. The method of claim 12, comprising transfecting the mammalian cell prior to said culturing step with a single vector containing a single expression unit transcribed into a polycistronic message. 16. A DNA sequence encoding a protein having substantially the same biological activity as human activated protein C, the sequence further comprising the amino acid sequence L-R_1-R
_2-R_3-R_4-X-R_5-R_6-R_7-
encodes R_8-H, where L is essentially the light chain of activated protein C, and R_1 to R_8
is Lys or Arg, X is a peptide bond or a spacer peptide of 1 to 12 amino acids, and H is essentially the heavy chain of activated protein C. 17. A mammalian cell transfected with an expression vector capable of integrating into the mammalian host cell DNA and containing a promoter operably linked to the DNA of claim 16. 18. The cells are Saccharomyces cerevisiae (￥S
accharomyces￥￥cerevisiae￥
18. The mammalian cell according to claim 17, further transfected with the KEX1 or KEX2 gene of ).