KR930005917B1 - Method for preparing erythropoietin - Google Patents

Abstract

No content.

Description

[Name of invention]

Method for preparing erythropoietin

[Brief Description of Drawings]

Table 1 shows the nucleotide sequences of 87 base pair exons of the human EPO gene;

1 shows electrophoresis indicating detection of EPO mRNA in liver mRNA of a human fetus;

Table 2 shows the amino acid sequences of EPO proteins deduced from the nucleotide sequence of lambda-HEPOFL 13, which is one of the clones of human EPO cDNA;

Table 3 shows the nucleotide sequence of EPO cDNA in lambda-HEPOFL 13 (shown schematically in FIG. 2) and the amino acid sequence deduced therefrom;

Figure 3 shows the relative positions of the insert DNA of four independent human EPO genomic clones;

4 shows a map of the apparent intron and exon structure of the human EPO gene;

Table 4 shows the DNA sequences of the EPO gene shown in Figure 4;

5a, b and c show the formation of the vector 91023 (B);

Figure 6 shows the SDS polyacrylamide gel analysis of EPO prepared in COS-1 cells compared to native EPO;

Table 5 shows the nucleotides of the EPO cDNA clone in human EPO cDNA clone, lambda-HEPOFL 6 and the amino acid sequences deduced therefrom;

Table 6 shows the nucleotides of EPO cDBA in human EPO cDNA clone, lambda-HEPOFL 8 and amino acid sequences deduced therefrom;

Table 7 is the same as Table 3, showing the nucleotides of the human EPO cDNA clone, EPO cDNA in lambda-HEPOFL 13 and the amino acid sequences deduced therefrom;

7 shows a schematic illustration of the plasmid pRK 1-4 for expression of EPO:

8 shows a schematic illustration of plasmid pdBPV-MMTneo (342-12).

Detailed description of the invention

[Technical Field]

The present invention relates to a cloning gene for human erythropoietin that provides very high expression expression titers, the expression of this gene and the in vitro production of active human erythropoietin.

Background Technology

Erythropoietin (hereinafter abbreviated as EPO) is a circulating glycoprotein that stimulates red blood cell formation in higher organisms (see Garnet et al, Compt. Rend. 143: 384 (1906)). Therefore, EPO is sometimes referred to as erythropoiesis-stimulating factor.

Human erythrocytes live about 120 days. Therefore, about 1/120 of the total red blood cells are destroyed in the reticuloendothelial system every day. At the same time, a relatively constant number of erythrocytes are produced daily so that the concentration of erythrocytes remains constant (Guyton, Textbook of Medical Physiology, PP 56-60, W.B.Saunders Co., Philadel Pha (1976)).

Erythrocytes are produced by maturation and differentiation of erythrocytes in the bone marrow, and EPO is a factor that induces differentiation into erythrocytes by acting on less differentiated cells (Guyton, same as above).

EPO is a promising therapeutic agent for the clinical treatment of anemia, especially renal anemia. Unfortunately, the use of EPO is not yet used in actual treatment due to its low availability.

In order to use EPO as a therapeutic agent, the problem of antigenicity should be taken into consideration. Therefore, it is preferable to manufacture EPO from human-derived raw materials. For example, blood or urine obtained from patients suffering from diseases such as aplastic anemia that secrete large amounts of EPO can be used. However, the supply of these raw materials is limited. See, eg, White et al., Rec. Progr. Horm. Res., 16: 219 (1960); Espada et al., Biochem, Med., 3: 475 (1970): Fisher, Pharmacol, Rev., 24: 459 (1972) and Gordon, Vitam. Horm. (N. Y.) 31: 105 (1973). These documents are for reference.

The manufacture of EPO products is generally carried out through the enrichment and purification of urine from patients with high EPO concentrations, such as patients with aplastic anemia and the like. See, for example, US Pat. Nos. 4,397,840: 4,303,650 and 3,865,801. These documents are for reference. This limited supply of urine is a barrier to the practical use of EPO, and therefore, there is a strong demand for the manufacture of EPO products from urine in healthy people. The problem with using urine obtained from healthy people is that the content of EPO is lower than that of anemia patients. In addition, the urine of healthy people contains some inhibitory factors that inhibit the formation of erythrocytes at high enough concentrations, and only after a significant purification can a satisfactory therapeutic effect be obtained from EPO.

In addition, EPO can be recovered from sheep plasma, and EPO isolated from sheep plasma provides a satisfactory role and stable water-soluble formulation. Goldwasser. Control CeIIular Dif. Develop., Part A; pp 487-494, Alan R. See Liss, Inc., N. Y. (1981). These documents are for reference. However, positive EPO can act as an antigen in the human body.

Therefore, EPO is a preferred therapeutic agent, but known separation and purification techniques using natural sources are inadequate for mass production of these compounds.

Sugimoto et al., US Pat. No. 4,377,513, discloses the in vivo proliferation of human lymphocytes, including Namalwa, BAII-1, NAII-1, TAII-1 and JBL. A method for mass production of EPO is described.

Reports of others producing EPO by genetic engineering techniques appear in the literature. See, for example, WO 085/02610 and WO 0/03079 published after the priority date of the present application. However, the details or chemical properties of this product have not yet been published. On the other hand, the present invention provides details regarding the mass production of proteins that exhibit the biological properties of human EPO. In addition, this technique allows the production of proteins that may be chemically different from genuine human EPO but with apparently similar properties (in some cases improved). For convenience, proteins that exhibit the biological properties of human EPO are subsequently abbreviated as EPO, regardless of their chemical identity.

Overview of the Invention

The present invention relates to the cloning of genes expressing surprisingly high titers of human EPO, their expression and the in vitro mass production of active human EPO therefrom. Also described are suitable expression vectors, expression cells, purified schemes and associated treatments for EPO production.

As explained in more detail below, EPO is obtained in partially purified form, which is further purified uniformly. Digestion with trypsin yields specific fragments. These fragments are purified and sequenced. Based on these sequences, EPO oligonucleotides are designed and synthesized. This oligonucleotide is used to screen the human genome library from which the EPO gene is isolated.

The EPO gene is also identified based on its DNA sequence corresponding to many trypsin digested protein fragments sequenced. Fragments of genomic clones are used to demonstrate that hybridization can detect EPO mRNA in human fetal (20 week) mRNAs. Prepare a liver cDNA library of human fetus. Screen. Three EPO cDNA clones are obtained (after screening more than 750,000 preparations). Two of these clones represent the full length as judged by the complete coding sequence and the actual 5'- and 3'-untranslated sequences. These cDNAs were transformed by the SV-40 virus into monkey cells (COS-1 cell line; Gluzman, CeII 23: 175-182 (1981)) and Chinese hamster ovary cells (CHO cell line; URIaub, G and Chasin). , LA Proc. Natl. Acad. Sci. USA 77: 4216-4280 (1980)). EPO prepared from COS cells is EPO which is biologically active in vitro and in vivo. In addition, EPO prepared from CHO cells is also biologically active in vitro and in vivo.

EPO cDNA clones have an interesting open reading frame of 14-15 amino acids (aa) with initiators and terminators 20-20 nucleotides (nt) upstream of the coding portion. Representative samples of this collie infected with cloned EPO cDNA have been deposited with the accession number ATCC 40153 in the American Type Culture Collection (ATCC) in Rockville, Maryland, United States, and are available from ATCC. This sample was deposited on September 12, 1986, with the accession number of KFCC-10265 to the Korean spawn association.

The present invention relates to the production of EPO by cloning of EPO genes and by in vitro strict expression of these genes.

Methods useful for the preparation of premixed products have been fully reported in the patent light scientific literature. Generally, such techniques include the isolation or synthesis of the desired gene sequences and the expression of these sequences in prokaryotic or eukaryotic cells using techniques commonly available to those skilled in the art. Once the designated gene is isolated and purified and inserted into the transfer vector (ie, cloned), it is clear that sufficient amounts can be obtained. Vectors having cloned genes thereof can be used in suitable microorganisms or cell lines, eg bacteria. Bacterials can be cloned by propagating microorganisms or cell lines by transferring to yeast, mammalian cells such as COS-1 (monkey kidney) and CHO (Chinese hamster ovary), insect cell lines, etc., and separating the vectors from these microorganisms or cell lines by known methods. can do.

Therefore, a continuously available gene source is provided for manipulation, modification, and transfer to other vectors or to other locations within the same vector.

Translational translation of prokaryotic or eukaryotic genes allows cloning with suitable directional and translational frames to synthesize protein precursors having amino acid sequences encoded by cloned genes linked from the prokaryotic or eukaryotic genes to Met or amino-terminal sequences Transduced genes can be expressed by transferring them to appropriate sites of the transfection vector. In other cases, transcriptional or translational initiation signals can be supplied by appropriate genomic fragments of the cloned genes. If a protein precursor is produced, a variety of specific protein cleavage techniques can be used to cleave the protein precursor at the required points and to purify it by known methods to obtain the desired amino acid sequence. In some cases, proteins containing the desired amino acid sequence may be prepared without the need for specific cleavage techniques and released from the cells into extracellular growth medium.

[Isolation of Genomic Clones of Human EPO]

As described below, human EPO is homogeneously purified from the urine of patients suffering from aplastic anemia. The purified EPO was completely digested with protease trypsin to obtain fragments, which were separated by reverse phase high performance liquid chromatography, recovered from gradient fractions and subjected to microsequence analysis. The sequence of trypsin digestion fragments corresponds to the underlined portions in Tables 2 and 3, for example. These are selected from suitable fragments, preferably Val-Asn-Phe-Tyr-Ala-Trp-Lys and Val-Tyr-Ser-Asn-Phe-Leu-Arg, to synthesize oligonucleotide probes against their sequences. . Oligonucleotide pools of 17nt length and 32-degenerate from the former trypsin digest fragment and 14nt long and 48-deflated from the latter trypsin digest fragment from 18nt length and 128-fold condensation Each pool is obtained. The human genomic DNA library in Ch4A vector 22 is screened by modifying the in situ amplification method 47 of the right and o 'molly to prepare filters for screens and using 17mer pools in 32-fold.

As used herein, the Arabic numerals (1) to (61) in parentheses are the references listed in order at the end of this specification.

17mer, phage hybridized phages were taken out and collected in small groups, parcels probed with 14mer and 18mer pools-plaque purified phages hybridized to 17mer, IBmer and 14mer pools, and fragments were removed from the dideoxy chains of Sanger and Kulson (23) (1977) Subclones into the sequencing M13beta according to the termination method. The sequence of the portion of one of the clones that hybridizes to 17mer in 32-implantation is shown in Table 1. This DNA sequence contains nucleotides that can accurately code a trypsin digestion fragment used to infer oligonucleotides of a 17mer pool of oligonucleotides within an open translation frame. Moreover, DNA sequencing shows that the 17mer hybridization moiety is contained within a 87bp exon surrounded by a latent splice receptor and donor sites (represented by a and d in Table 1, respectively).

The fact that these two clones (here lambda-HEP01 and lambda-HEP02) are EPO genomic clones is confirmed by sequencing another exon containing information encoding other trypsin digest fragments.

TABLE 1

TABLE 2

TABLE 3

Isolation of EPO cDNA Clone

Next, the EPO cDNA is isolated from a suitable human cDNA library using the genomic DNA of EPO obtained above as a probe. According to our study, human cDNA libraries made from fetal liver mRNA were most suitable for this purpose. That is, the present inventors prepared a 95nt single strand prepared from the M13 clone containing a portion of the 87bp exon described in Table 1, and used it as a probe for Northern analysis of mRNA between human fetuses (20 weeks) ( 56). As shown in FIG. 1, strong signals can be detected in the liver myNA of the sun. This band, EPO mRNA, is accurately identified by using the same probe to screen the bacteriophage lambda cDNA library of fetal liver mRNA (25). Several hybridized clones are obtained with a positive frequency of about 1 for the screened 250,000 recombinants. The complete nucleotide and inferred amino acid sequences of these clones (lambda-HEPOFL 13 and lambda-HEPOFL 8) are shown in Tables 3 and 7. Information encoding EPO is contained within 594 nt of the 5 'half of cDNA, including a highly hydrophobic 27 amino acid leader and a mature protein consisting of 166 amino acids.

Identification of the N-terminus of the mature protein is shown in Table 1) and aplastic anemia, as reported by Gold and Western (26), Water and Sitkovsky (27), and Yanagawa (21). It is based on the N-terminal sequence of the protein excreted in the urine of the patient. It is not currently known whether this N-terminus (Ala-Pro-Pro-Arg---) represents the actual N-terminus found in the EPO in circulation or which cleavage occurs in the kidneys or urine.

The underlined amino acid sequences in Tables 2 and 3 show the N-terminal part obtained from trypsin digested fragment or protein sequence information. The deduced amino acid sequence exactly matches the sequenced trypsin digested fragments, confirming that the isolated cDNA encodes human EPO.

[Structure and Sequence of Human EPO Gene]

The relative positions of the insert DNAs of the four independent human EPO genomic clones are shown in FIG. These cloned DNAs are hybridized with oligonucleotide probes and various probes prepared from two kinds of EPO cDNA clones in which the EPO gene is located within about 3.3 kb. Complete sequencing of this portion (see Example 4) and comparison with cDNA clones yield maps of the introns and exon structures of the EPO gene shown in FIG. The EPO gene is divided into five exons. Part of exon I, all of exon II, III and IV and part of exon V contain information encoding the protein. The rest of exons I and V encode untranslated sequences on the 5 'and 3' sides, respectively.

[Transient Expression of EPO in COS Cells]

COS cell expression experiments were performed to demonstrate that the cDNA obtained can express biologically active EPO in an in vitro cell culture system (58). Vector p91023 (B) used for this experimental expression is described in Example 5. This vector contains the alate virus late promoter, the SV4O polyadenylation sequence, the replication origin of SV4O, the SV4O enhancer and the adenovirus VA genes. Insertion of lambda-HEPOFL 13 (see Table 7) cDNA is inserted downstream of the adenovirus main rate promoter of the p91023 (B) vector. This new vector is named pPTFL 13.

Approximately 24 hours after infection with COS-1 cells of M6 strain (Horowitz et al, J. Mol. Appl. Genet. 2: 147-149 (1983)), cells were washed and washed in serum-free medium. Correct, cells are harvested after about 48 hours.

Shewood, JB et al. For EPO quantitative radioimmunoassay (55) is used to examine the titer of EPO released into the culture supernatant. According to this method, as shown in Table 8 (Example 5), deductively active EPO is expressed. In addition, the biological activity of EPO prepared from COS-1 cells was also examined. In another experiment, a vector containing EPO cDNA from lambda-HEPOFL 13 is infected with COS-1 cells and the medium is harvested as described above-EPO in medium is harvested with ³ H-thymidine and CPU-E (12 And quantitative analysis by two in vitro biological assays of (29) and two in vivo assays using hypoxic mice and starved mice (30,31) (see Table 9, Example 7). These results demonstrate that biologically active EPO is produced in COS-1 cells. Western blotting using polyclonal anti-EPO antibodies showed that EPO produced by COS cells had the same mobility as native EPO prepared from human urine on SDS-polyacrylamide gels (Example 8). This is because the fact indicates that the degree of glycosylation of EPO produced by COS-1 is similar to that of natural EPO.

In addition, the present invention can use different vectors having different promoters, which are particularly preferred for the expression of human EPO cDNA, in COS cells or in other mammalian or eukaryotic cells. Examples of other promoters useful in practicing the present invention include SV4O early and late promoters, mouse metallothionein gene promoters, promoters found in long terminal repeat units of avian or mammalian retroviruses, baclovirus polyhedron gene promoters, and the like. There is this. Examples of other cells useful for practicing the present invention include E. coli. Mammalian cells such as collie, yeast, CHO (Chinese hamster ovary), Cl27 (monkey epithelium), 3T3 (mouse fibroblast) and CV-1 (African green monkey kidney). Such other promoters and / or cell types allow for control of the time and extent of EPO expression, EPO production of cell specificity, or mass growth of EPO producing cells under inexpensive and easily controlled conditions.

Examples of EPO expression in CHO, Cl27 and 3T3 are described in Examples 10 and 11 (CHO) and 13 (Cl27 and 3T3).

Recombinant EPO prepared in CH0 cells as in Example 11 is purified by known column chromatography methods. Relative amounts of sugars present in glycoproteins were determined by two independent methods [(i) Reinhold, Methods in Enzymol. 50: 244-249 (methanol decomposition) and (ii) Takemoto, H. et al, Anal. Biochem. 145: 245 (1985) (with pyridyl amination, with independent sialic acid crystals). The results obtained in each of these methods are very consistent. Several measured values thus obtained yield the following average values. The value of N = acetylglucosamine is set to 1 for comparison.

* N-acetinneuraminic acid is unstable and its value varies in the range of about 0.4 to about 1. It should be noted that significant amounts of fucose and N-acetylgalactosamine are repeatedly observed in two independent sugar assay methods. The presence of N-acetyl galactosamine indicates the presence of zero-linked glycosylation in the protein. The presence of zero-linked glycosylation is also confirmed by SDS-PAGE analysis of glycoproteins and degradation of glycoproteins with combinations of various glycosidase degrading enzymes. In particular, enzymatic removal of all N-linked carbohydrates of the glycoprotein using the enzyme peptide endo-FN-glycosidase, as determined by SDS-PAGE analysis upon subsequent degradation by neuraminidase The molecular weight is further reduced.

Rat in vitro biological activity of purified recombinant EPO. Crystal, Exp. Hematol. 11: 649 (1983) (Spherical Cell Proliferation Bioassay) to analyze and determine protein based on amino acid composition data. By determination of aberrations, the in vitro inactivation of purified recombinant EPO is calculated to be at least 200,000 columns / mg protein. The average value is about 275,000 to 300,000 units / mg-protein. Furthermore, values of more than 300,000 are also observed. The in vivo activity ratio observed in recombinants (polycyticemia assay, Kazal and Erslev, Am. Clinical Lab. Sci., Vol, B, p. 91 (1975)) / in vitro activity ratio is 0.7-1.3.

It is interesting to compare the above glycoprotein characteristics with those of recombinant CHO-producing EPO materials already reported in International Publication No. WO85 / 02610 (June 20, 1985) after the priority date of the present application. A corresponding comparative sugar assay is described on page 65 of the application, wherein the fucose and N-acetyl galactosamine were reported as 0, and the hexose: N-acetyl galactosamine ratio was reported as I5.09: 1. The absence of N-acetyl galactosamine in glycoproteins reported in this international publication indicates the absence of 0-linked glycosylation. In contrast, the recombinant CHO-producing EPO of the present invention contains a significant amount of fucose and N-acetyl galactosamine which can be observed repeatedly, and contains less than one tenth the relative amount of hexose. And, the presence of 0-linked glycosylation is characteristic. Moreover, the high specific activity of the CHO-derived recombinant EPO of the present invention described above may be directly related to its characteristic glycosylation pattern.

Biologically active EPO prepared by prokaryotic or eukaryotic cell expression of the cloned EPO gene of the present invention can be used by physicians and / or veterinarians for in vivo treatment of mammals. The amount of active ingredient is, of course, determined by the severity of the disease to be treated, the route of administration chosen and the inactivation of the active EPO, and finally by the doctor or veterinarian's prescription. The amount of active EPO determined by the doctor's prescription is named "EPO therapeutically effective amount." For example, the daily effective amount of EPO for the treatment of hypoproliferative anemia caused by positive chronic kidney disease is 10 units / kg for 15-40 days. Eschbach et al., J. Clin. Invest., 74: 434 (1984).

Active EPO can be administered by any route suitable for the condition to be treated. Preferably the EPO is injected into the bloodstream of the mammal to be treated. Preferred routes may be determined by those skilled in the art depending on the condition to be treated.

Active EPO can be administered as a pure or nearly pure compound, but is preferably administered as a pharmaceutical preparation or formulation.

The formulations of the invention for livestock and for human use contain the above-mentioned active EPO protein together with one or more pharmaceutically acceptable carriers and any other therapeutic ingredients. The carrier must be "acceptable" and compatible with the other ingredients of the formulation and must be harmless to the recipient. The formulation should not contain oxidants and other substances known to be incompatible with the peptide. The formulation is convenient in unit dosage form. It may be prepared by any method known in the pharmacy art. All methods include the step of combining the active ingredient with a carrier consisting of one or more accessory ingredients. In general, formulations are prepared by homogeneously and precisely combining the active ingredient with a liquid carrier or a finely ground solid carrier or both, and if necessary shaping the product into the desired formulation. Formulations suitable for parenteral administration conveniently contain a sterile aqueous solution of the active ingredient, preferably with the hull of the recipient and an isotonic solution. Such formulations may be conveniently prepared by dissolving the solid active ingredient in water to form an aqueous solution and sterilizing it, and are placed in a unit or virtue container, such as a sealed ampoule or vial.

EPO / cDNA as used herein includes EPO / cDNA encoding alleles of the mature EPO / cDNA gene and the EPO protein, preceded by ATG codes. One allele is described in Tables 3-6. EPO proteins include 1-methionine derivatives of the EPO protein (Met-EPO) and alleles of the EPO protein. The mature EPO protein described by the sequence in Table 2 is sequence Ala. Pro. Pro. It begins with Arg .... The starting point is indicated by the number "1" in Table 2. Met-EPO has the sequence Met. Ala. Pro. Pro. Start with Arg.

Glycoproteins produced as a result of the expression of EPO / cDNA are sufficiently contemplated to be subjected to the action of proteolytic enzymes present in the transformant or outside the expression system and include all of amino acid sequences 1-166 which are read from the DNA sequence. There may be things you are not doing. For example, desorption of C-terminal Arg, for example, has been found for CHO expressing EPO. However, this type of glycoprotein also has in vivo activity and is included in the object of the present invention. Desorption of C-terminal Arg has also been found in human urine EPO, and it has been confirmed that in vivo activity is maintained.

The following examples are provided to aid the understanding of the present invention, the actual scope of the present invention is as shown in the following claims. The following examples may be modified without departing from the scope of the present invention. All temperatures are expressed in degrees Celsius and cannot be modified. The micron toe is micro, for example the symbol of microliter, micromole, etc. is "μ", for example μl, μm or the like.

EXAMPLE

Example I

[Isolation of Genomic DNA Clones of EPO]

Miyake et al. (Miyake, et al., J. Biol. Chem., 252: 5558 (1977), except that inactivation of neuraminidase by incubation at 80 ° C. for 5 minutes instead of phenol treatment). Purify EPO from the urine of patients with aplastic anemia. The final step of purification is fractionation on a C-4 Vydac HPLC column (The Separations Group) for at least 100 minutes using a 0-95% acetonitrile concentration gradient containing 0.1% trifluoroacetic acid (TFA). The location of the EPO in the concentration gradient is determined by gel electrophoresis and N-terminal sequencing of the main perk (21, 26, 27). EPO elutes in about 53% acetonitrile and represents about 40% of the protein subjected to reverse phase-HPLC. Fractions containing EPO were evaporated to 100 μl, adjusted to pH 7.0 with ammonium bicarbonate, and then digested with 2% tosyl phenylalanylchloromethylketone (TPCK) -treated tripsil (Washton) at 37 ° C. for 18 hours. . Trypsin digest is reversed phase HPLC as described above. Observe the optical density at 280 and 214 nm. Well separated peaks are grown to dryness and direct N-terminal amino acid sequencing (58) using an Applied Biosystems Model 480A gas phase sequencer. The sequences obtained are underlined in Tables 2 and 3. As described above, two of these trypsin digest fragments are selected for the synthesis of oligonucleotide probes. 17mer in 32-fold conjugation from VaI-Asn-Phe-Tyr-Ala-Trp-Lys (amino acids 46-52 of Tables 2 and 3) sequence;

And 18 mer in 128-fold axis;

Is manufactured. 14mer2 pool in each 48-fold axis differing in first position of leucine codon from the Val-Tyr-Ser-Asn-Phe-Leu-Arg (amino acids 144-150 in Tables 2 and 3) sequence

Is manufactured. Polynucleotide kinases (New England Biolabs) and gamma ³² P-ATP (New England Nuclear) are used to label the 5'-end of the oligonucleotide with ³² P. The specific activity of oligonucleotides is varied within the range of 100-3000 Ci / mmolol oligonucleotides. A modification of the in situ proliferation method originally described by Ho et al (47) (1978) screens the human genomic DNA library in bacteriophage lambda (Lawn et al., 22). About 3.5 * 10 ⁵ phages are plated at a density of 6000 phages per 150 mm Petri dish (NZCYM medium) and incubated at 37 ° C. until small (about 0.5 mm) plaques are visible to the naked eye. After cooling for 1 hour at 4 ° C., duplicate replicas of plaque patterns are transferred to nylon membrane (New England Nuclear) and incubated overnight at 37 ° C. in new NZCYM plates. The filter is denatured and neutralized by immersing in a thin film of 0.5 N NaOH-1 M NaCl and 0.5 M Tris pH 8-1 M NaCl for 10 minutes, respectively. After vacuum baking at 80 ° C. for 2 hours, the filter is washed with 5 * SSC, 0.5% SDS for 1 hour and removed by gently wiping the cell pieces on the filter surface with a wet tissue.

This wiping reduces the probe's nonspecific binding to the filter. Rinse the filter with H ₂ O and convert for 4-8 hours at 48 ° C. in 3M tetramethyl ammonium chloride, 10 mM NaPO ₄ (PH 6.8), 5 * Denhart, 0.5% SDS and 10 mM EDTA. ³² P-labeled 17mer was added at 0.1 picomolar / ml concentration and hybridization was performed at 48 ° C. for 72 hours. After hybridization, the filter was removed at room temperature for 2 * SSC (0.3M NaCl-0.03N Na citrate, pH7) seconds. Wash strongly and wash with 3M TMACI-10 mM NaPO ₄ (pH 6.8) for 1 hour at room temperature and 5-15 minutes at hybridization temperature.

Automatic radiographing for two days with an enhanced screen detects about 120 strong overlapping signals. Positive samples are collected and pooled in 8 pools, replated and rescreened with 3 specimens using 127mer for the third filter, one half of 14mer pools for each of the two filters. Plating and hybridization of 17mer is as described above except that hybridization of 14mer is performed at 37 ° C. After autoradiography is taken, the 17mer filter is operated for 20 minutes at room temperature in 50% formamide to remove the probe and the filter is rehybridized with an 18mer probe at 52 ° C. Both two independent phages hybridize to three probes. DNA from one of these phages (here lambda HEPO 1) is fully digested with Sau3A and subcloned into M13 for DNA sequencing using Sanger and Coulson's dideoxy chain termination method (23) (1977). . The nucleotide sequence and the deduced amino acid sequence of the open translation frame encoding the EPO trypsin digestion fragment (underlined portion) are described herein. Intron sequences are shown in lowercase. Exon sequence 87nt is capitalized. Sequences matching the standard Splice receptor (a) and donor (b) sites are underlined (see Table 4).

TABLE 4

Example 2

[Northern Analysis of Liver mRNA in Human Fetuses]

5 μg of human fetal liver mRNA (obtained from the liver of a 20-week-old fetus) and adult liver mRNA were subjected to residue residues on a 0.8% agarose formaldehyde gel, followed by the method of Derman et al. (CeII, 23: 731 (1981). Transfer to nitrocellulose. Single-stranded probes are prepared from M13 templates containing the inserts described in Table 1. The primer is 20mer derived from trypsin digestion fragment, like the original 17mer probe. The probe was digested with Sma I (prepared a 95nt long desired probe with a 74nt coding sequence) and purified from the M13 template by chromatography of small fragments with 0.1N NaOH-0.2M NaCl on a Sepharose C14B column. Except as described in Anderson et al., PNSS, (50) (1984). Filters are hybridized to about 5 * 10 ⁶ cpm of the probe for 12 hours at 68 ° C, washed in 2 * SSC at 68 "C, and then exposed to the enrichment screen for 6 days. 1200nt single marker mRNA (with arrows) Mark) appears in the adjacent lane (FIG. 1).

As a result, one band is observed which hybridizes with the probe to human fetal liver mRNA.

Example 3

[Isolation of Liver cDNA in Fetuses]

As a result of the Northern analysis, it is predicted that human EPO cDNA can be separated by constructing a cDNA library from fetal liver mRNA and performing a double screen. Thus, the same probes as described in Example 2 were prepared and beta lambda-Ch2lA (Toole et al., Nature, (25) () using the standard plaque screen (Benton Davis, Science, (54) (1978)) method. 1984) is used to screen the fetal liver cDNA library. Screening 1 × 10 6 plaques separates three independent positive cDNA clones (here lambda-HEPOFL 6 (1350 bp), lambda-HEPOFL 8 (700 bp) and lambda-HEPOFL 13 (1400 bp)). Subcloning into M13 sequenced the entire insertion portion of lambda-HEPOFL 13 and lambda-HEPOFL 6 (Tables 7 and 5, respectively). Only a portion of lambda-HEPOFL 8 is sequenced and the rest are assumed to be identical to the other 2 clones (Table 6). Untranslated sequences on the 5 'and 3' pages are shown in lowercase letters. The coding part is capitalized.

Referring to Table 7 (and Table 3), the inferred amino acid sequence described above of the nucleotide sequence is numbered with 1 as the first amino acid of the mature protein. Presumed leader peptides are written in all capital letters. Cysteine residues in mature proteins are additionally labeled SH and strong N-linked glycosylation sites are marked with asterisks. Underlined amino acids represent residues identified by N-terminal protein sequencing or sequencing of trypsin digested fragments of EPO such as Example 1. The amino acids underlined by the dashed line represent residues in the amino acid sequence of the trypsin digest fragment which cannot be clearly determined. cDNA clones Lambda HEPOFL 6, Lambda-HEPOFL 8 and Lambda-HEPOFL 13 are deposited in the American Type Culture Collection (ATCC) of Rockville, Maryland, USA under the accession numbers ATCC 40156, ATCC 40152 and ATCC 40153, respectively, and are available from ATCC. have. In addition, on September 12, 1986, they were deposited with the Korean spawn association with accession numbers of KFCC-10268, KFCC-10264 and KFCC-10265.

TABLE 5

TABLE 6

TABLE 7

Example 4

[Genome Structure of EPO Gene]

The relative size and position of four independent genomic clones (lambda-HEPO 1,2,3 and 6) from the Hae III / Alu I library are indicated by overlapping lines in FIG. The thick line shows the location of the EPO gene. The location and scale (Kb) of known restriction endonuclease cleavage sites is described. The EPO gene containing portion is fully sequenced from two strands through a series of deletions resulting from the action of aromatic exonuclease III on the portion containing the EPO gene. 4 shows the 5 exons encoding EPO mRNA. The exact 5'-bound of exon I is currently unknown. The protein coding portion of exon is shown in black. The complete nucleotide sequence of this portion is shown in Table 4. Known limits of each exon are indicated by right angle straight lines. Genomic clone lambda. HEPO 1, lambda-HEPO 2, lambda-HEPO 3 and lambda-HEPO 6 are assigned to the American Type Culture Collection (ATCC) of Rockville, Maryland, USA, under the accession numbers ATCC 40154, ATCC 40155, ATCC 40150 and ATCC 40151, respectively. It has been deposited and is available from ATCC. In addition, as of September 12, 1986, they were deposited with the accession numbers of KFCC-10266, KFCC-10267, KFCC-10262 and KFCC-10263, respectively.

Example 5

[Formation of Vector p91023 (B)]

The transformation vector is pAdD26SVpA (3) reported by Kaufman et al. (Mol. CeII Biol. 2: 1304 (1982)). The structure of this vector is shown in FIG. 5A. Briefly, this plasmid contains the mouse dihydrofolate reductase (DFHR) cDNA gene under the transcriptional control of the adenovirus 2 (Ad2) main rate promoter. The 5'-conjugation site is in the adenovirus DNA and the 3'-conjugation site derived from the immunoglobulin gene is between the Ad2 week rate promoter and the DFHR coding sequence. The SV4O early poly adenylation site is downstream of the DFHR coding sequence. The prokaryotic-derived portion of pAdD26SVpA (3) is from pSVOD (Melton et al., CeII, 27. 279 (1981)) and does not contain the pBR 322 sequence known to inhibit replication in mammalian cells (Lusky et. al., Nature, 293: 79 (1981).

pAdD26SVpA (3) is converted to plasmid pCVDVL 2 as shown in FIG. 5A. pAdD26SVpA (3) is converted to plasmid pAdD26SVpA (3) (d) by deleting one of the two Pst I sites in pAdD26SVpA (3). SV4O poly-adenylation sequence after partial digestion with Pst I using an insufficient amount of enzyme to yield only one Pst I site cleaved, treated with Klenow and ligated for re-ization Screen for deletion of the Pst I site located at 3'- of the.

Subsequently, adenovirus triad leader and virus-associated gene (VA gene) are inserted into pAdD26SVpA (3) (d) as follows. First, pAdD26SVpA (3) (d) is cleaved with Pvu III to make open linear molecules in the 3'-part of the trielement containing the three-minute leader. Then, pJAW43 (Zain et al., CeII, 16: 851 (1979)) was digested with Xho I, treated with Klenow, then with PVU II, and acrylamide gel (6% in tris borate buffer: Maniatis et al., Homolog) to isolate the 140 bp fragment containing the second part of the third leader. The 140 bp fragment is ligated to pAdD26SVpA (3) (d) digested with Pvu II. Ligation products are used to hedge this coli into tetracycline resistant strains and to screen colonies by the Grunstein-Hognes method using 32P-labeled probes that hybridize to 140 bp fragments- DNA from positively hybridized colonies. Prepare and test whether the re-formed Pvu II site is the 5 'or 3' side of the inserted 140bp DNA having specificity for Example 2 and 3 adenovirus rate readers. The exact orientation of the Pvu II site is on the 5'-side of the 140 bp insert. This plasmid is named tTPL in Figure 5a.

The SV4O DNA was digested with Ava II, and the base pairs at both ends were completely formed by the Cleno fragment of Pol I. The Xho I linker was ligated to this fragment, and the Xho I site was opened by Xho I digestion, followed by gel electrophoresis. The Ava IID fragment of SV4O containing the SV4O enhancer sequence was obtained by separating the fourth large (D) fragment. This fragment is ligated into Xho I-cleaved pTPL to give plasmid pCVSVL2-TPL. The direction of the SV4O D fragment in pCVSVL2-TPL is such that the SV4O rate promoter is the same as the adenovirus main rate promoter.

To introduce the adenovirus-associated (VA) gene into pCVSVL2-TPL, first, plasmid pBR 322 containing adenovirus type 2 Hind III B fragment is formed. Adenovirus type 2 DNA is digested with Hind III, and fragment B is separated by gel electrophoresis. This fragment is inserted into pBR 322 previously digested with Hind III. After transforming this coli to ampicillin resistance, the transformants are screened for ampicillin resistance as an indicator for insertion of the Hind III B fragment and the inserted direction is determined by restriction enzyme digestion. pBR 322-Ad Hind III B contains the adenovirus 2 hung Hind III B fragment in the orientation described in Figure 5b.

As shown in FIG. 5B, the VA gene is conveniently obtained by digesting plasmid pBR 322-Ad Hind III B with HPa I, adding EcoRI linker, digesting with EcoRI, and recovering the 1.4 kb fragment. Thereafter, the fragment having the EcoRI-adhesive end is ligated to the EcoRI site of pCVSVL2-PTL previously digested with EcoRI. The coli HB 101 was transformed, the tetracycline resistant strain was selected, and the DNA having the specificity for the VA gene was added. Screen colonies by filter hybridization. DNA is prepared from positively hybridized colonies and characterized by digestion with restriction endonucleases. The resulting plasmid is named p91023.

As shown in FIG. 5C, p91023 was removed from the EcoRI site by complete cleavage of p91023 with EcoRI, one at about 7 kb. The other gets two DNA fragments of about 1,3 kb. The latter fragment contains the VA gene. Both ends of the two fragments are filled with Klenow fragments of Pol I and the two fragments ligated with each other. The plasmid p91023 (A), which contains the VA gene and is similar to p91023 but deleted two EcoRI sites, is identified by the Grunstein-Hognes method and known restriction site analysis, screening with VA gene fragments.

Remove one Pst I site of p91023 (A) and replace it with EcoRI site. p91023 (A) is completely cleaved with Pst I and treated with the Klenow fragment of Pol I to obtain base paired ends. The EcoRI linker is ligated to the base paired Pst I site of p91023 (A). The linear p91023 (A) with EcoRI linker attached to the Pst I site of the complete base pair is isolated from the non-ligated linker, completely digested with EcoRI and then religated. The plasmid p91023 (B) as shown in FIG. 5C was recovered, and the recovered plasmid p91023 (B) was recovered, and the recovered plasmid p91023 (B) had a structure similar to that of p91023 (A), but with the former Pst. Make sure you have the EcoRI site instead of the I site. Plasmid p91023 (B) has been deposited under the accession number ATCC 39754 in the American Type Culture Collection (ATCC) of Rockville, Maryland, and is available from ATCC.

Example 6

[Example 1 Expression of EPO cDNA in COS-1 Cells]

cDNA clones (lambda-HEPOFL 6 and lambda-HEPOFL 13: Example 3) were inserted into plasmid p91023 (B), respectively, to form pPTFL 6 and pPTFL 13. Each 8 μg of purified DNA is used to infect 5 * 10 ⁶ COS cells by the DEAE-dextran method (described later). After 12 hours, cells are washed, treated with chlorokin (0.1 mM) for 2 hours, then washed again and exposed to 10 ml medium containing 10% fetal calf serum for 24 hours. Change medium to 4 ml serum free medium and harvest after 48 hours.

Production of immunologically active EPO is quantitated by radioimmunoassay of Shewood and Goldworther (55). Antibodies are provided by Dr. Judith Shewood—Iodide tracers are prepared from the homologous EPO described in Example 1. The sensitivity of the assay is about 1 ng / ml. The results are shown in Table 8 below.

TABLE 8

PTFL 13 has been deposited with the accession number ATCC39990 at American Type Culture Collection, Rockville, Maryland, and is available from ATCC. PTFL 13 was deposited on September 12, 1986 with the accession number of KFCC-10260 to the Korean spawn association.

Example 7

[Example 2 Expression of EPO cDNA in COS-1 Cells]

EPO cDNA (lambda-HEPOFL 13) was inserted into the p91023 (B) vector. After infection with COS-1 cells, harvest was performed as described above (Example 6) except that chlorokin treatment was omitted.

In vitro biologically active EPO is measured by colony formation assay using hepatocytes of mouse embryos as a source of CFU-E or by 3H-thymidine uptake analysis using splenocytes from mice injected with phenylhydrazine. The sensitivity of these assays is about 25 mU / ml. In vivo biologically active EPO is measured using hypoxic mice or starved rat methods. The sensitivity of this assay is about 10 mU / ml. No activity is detected when digested with mock condition medium. The results of EPO expressed by clone EPOPL 13 are shown in Table 9 below. Reported activity uses commercially available quantitative EPO (Toyobo, Inc.) as standard and is expressed in units / ml.

TABLE 9

EPO secreted from COS-1 cells infected with type I EPO cDNA

Example 8

[SDS Polyacrylamide Gel Analysis of EPO from COS Cells]

180 ng of EPO generated in the medium of COS cells infected with EPO (lambda-HEPOFL 13) cDNA of Vectar 91023 (B) (described above) was electrophoresed on a 10% SDS Laemry polyacrylamide gel and electrophoresis on nitrocellulose paper. (Towbin et al., Proc. Natl. Acad. Sci- USA 76: 4350 (1979)). The filter is probed with the anti-EPO antibody described in Example 6, washed and then reprobeed with ¹² 5l-staph A protein. Automatic radiographing of the filter for 2 days. Natural homogeneous EPO is described in Example 1 and electrophoresed before (iod B) or after (lane C) iodide (see FIG. 6). Markers used are ³⁵ S labeled methionine, serum albumin (68,000d) and ovalbumin (45,000d).

Example 9

[Formation of RK1-4]

Polasmid PSVEDHFR (Subramani et al., Mol.) Containing SV 40 early partial promoter, SV 40 enhancer, small t antigen intron and SV 40 polyadenylation sequence adjacent to the dihydrofolate reductase (DHFR) gene in mice. Bam Hl-PvuII fragment is isolated from CeII Biol. 1: 854-864 (1981).

The remaining fragments are obtained from Vector P91023 (A) (described above) as follows. P91023 (A) was digested with PStl at one PStl site near the adenovirus promoter to linearize the plasmid, ligated to synthetic PStl and converted into EcoR I converters for re-establishment (PstI-EcoRI − at the original PstI site). Formation of Pst I moiety, 91023 (B ′)), treatment with large fragments of DNA polymerase I to disrupt the PstI site, ligating into a synthetic EcoR I linker and re-establishing (EcoRI site formation at the original Pst I site: 91023 (B)). Each of the two resulting plasmids 91023 (B) and 91023 (B ') is digested with Xba and EcoRI to obtain two fragments (F and G). By combining the fragment “F from p91023 (B) and fragment 6 from p91023 (B ′), or fragment G from p91023 (B) and fragment“ F from p91023 (B ′), EcoRI is attached to the original PstI site. Two new plasmids with the -PstI site or PstI -EcoR I site are obtained. The rlasmid with the Pstl-EcoRI site whose PstI site is closest to the adenovirus main rate promoter is named p91023 (C).

Vector p91023 (C) is completely digested with Xhol and base paired by filling the linear DNA with the resulting sticky ends with large fragments of DNA polymerase I of Iroli. The DNA is ligated with a 340 bp HindIII-EcoRI fragment containing the SV 40 enhancer as follows: the replication origin of SV 40 and the HindIII-PvuII fragment from SV40 containing the enhancer were plasmid c lac (Little et at., Mol). Biol, Med. 1: 473-488 (1983). A c lac vector is prepared by digesting c lac DNA with BamH I, filling its sticky end with a large fragment of DNA polymerase I, and then digesting the DNA with HindIII.

The resulting BamH I site is obtained by ligation of the resulting plasmid (c SVHPlaI) to the PvuII base pair ends. EcoRI-HindIII fragments are prepared from cSVHPlac and ligated to EcoRI-HindIII fragments of pSVOD (Mellonet at, described above) containing the replication origin of the plasmid, and the resulting plasmid pSVHPOd is selected. A 340 bP EcoRI-dinindrII fragment of pSVHPOd containing an SV 40 origin / enhancer is prepared and base-paired at both ends with large fragments of DNA polymerase I followed by base pairing of the XhoI digestion described above.

Example 10

[Expression of EPO in CHO Cells-Method I]

DNA (20 μg) from plasmid pPTFL 13 described above (Example 6) was digested with restriction endonuclease Clal to linearize the plasmid and intact dihydrorofolate reduct induced by the adenovirus main rate promoter. Plasmid pAdD26SVp (A) 1 (2 μg) containing the kinase (DHFR) gene (Kaufman and Sharp, Mol. And CeII Biol. 2: 1304-1319 (1982)). Let's do it. This ligation ONA was used to infect DHFR-responsive CN0 cells (DUKX-BII, Chasin LA and URIaub G. (1980) PNAS 77.4216-422D) and grow for 2 days, resulting in nucleotide deficient and 10% dialysis. Cells in which at least one DHFR gene is combined are selected in alpha medium supplemented with fetal calf serum. After growing for two weeks in a selective medium, colonies are removed from the plate, collected into groups of 10 to 100 colonies per pool, replated in eggs and medium lacking nucleotides, and grown to saturation. EPO analyzes were performed on the contaminant hole RIA from the pool grown prior to methotrexate selection. Pools showing positive EPO production are grown in the presence of methotrexate (D-02uM), subcloned and reanalyzed.

One EPD Cla 4α 4.02-7 subcloned from the EPO Cla 4α 4.02 pool yields 460 ng / ml EPO medium containing 0.02 uM MTX (Table 10). EPO Cla 4α 4.02-7 is a cell line of choice for EPO production and has been deposited with the Accession Number of ATCC CRI 8695 in the American Type Culture Collection (ATCC). Generally. This clone is being selected in stages by gradually increasing the concentration of MTX, and provides taxation to produce high concentrations of EPO.

About Pulume which is voice in RIA. Methotrexate resistant colonies obtained from Han Chinese basin grown in the presence of methotrexate (0.02 uM) are re-analyzed by RIA in the pool for EPO. Non-positive cultures are subcloned and grown in increasing concentrations of methotrexate.

Stepwise methotrexate (MTX) selection is performed by an iterative cycle of selecting cells to survive by culturing cells in the presence of increasing concentrations of betotrexate. EPO is measured by RIA and in vitro biological activity in each cycle of culture supernatant. The methotrexate concentrations used for each step increase were 0.02 uM, 0.1 uM and 0.5 uM. As shown in Table 10, after one round of selection in 0.02 uM MTX, a significant amount of EPO is produced in the culture medium.

TABLE 10

Concentration of EPO produced in the medium

Example 11

[Expression of EPO in CHO Cells-Method II]

DNA from clone lambda HEPOFL 13 is digested with Eco RI, and a small RI fragment containing the EPO gene is subcloned into the Eco RI site of plasmid RKI-4 (Example 9). This DNA (RKFL 13) is used to directly (without digestion) DHFR-dependent CHO cells, and to select and propagate as described in Example 10 above.

RKFL 13 DNA is inserted into CHO cells by protoplast fusion and microinjection. Plasmid RKFL 13 has been deposited with the accession number ATCC 39989 in the American Type Culture Collection (ATCC) of Rockville, Maryland, and is available from ATCC. RKFL 13 was deposited on September 12, 1986, with the accession number of KFCC-10259 to the Korean spawn association.

TABLE 11

Concentration of EPO produced in the medium

Preferred single colony clones have been deposited with ATCC CRI 8695 in the American Type Culture Collection (Aroll CC) of Rockville, Maryland, and are available from ATCC.

Example 12

[Expression of EPO Genomic DNA in COS-1 Cells]

The vector used for the expression of EPO genomic clones is pSVOd (MeIIon et al., Described above). DNA from pSVOd is completely digested with HindIII and base paired with large fragments of DNA polymerase I. EPO genomic clone lambda-HEPO3 is completely digested with EcoR I and HindIII, and the 4.0 kb fragment containing the EPO gene is isolated and base paired as above. Nucleotide sequences from the HindIII site of this fragment to the portion just beyond the polyadenylation signal are shown in FIGS. 4 and 4. The EPO gene fragment is inserted into the pSVOd plasmid fragment and the recombinants with the correct orientation of both orientations are isolated and identified. Plasmid CZ2-1 has the EPO gene in the "a" direction (ie, where the 5 'end of the EPO is close to the SV 40 origin) and plasmid CZ1-3 is in the opposite direction (direction "b").

Plasmids CZ1-3 and CZ2-1 are infected with COS-1 cells as described in Example 7, media is harvested and analyzed for immunologically reactive EPO. About 31 ng / ml of EPO is detected in the culture supernatant of CZ2-1, and 16-31 ng / ml is detected from CZ1-3.

Genomic clones HEPO1, HEPO2 and HEPO6 can be inserted into COS cells and transduced in a similar manner.

Example 13

[Formation of Heng-Shin-Dong pBPVEPO in Cl27 and 373 Cells]

A plasmid containing the EPO cDNA sequence and linked to complete bovine papillomavirus DNA under the transcriptional control of the mouse metallothionein promoter was prepared as follows:

[pEP049f]

Plasmid SP6 / 5 purchased from promega Biotec. This plasmid is completely digested with EcoRI and 1340 bp EcoRI fragment from lambda-HEPOFL 13 is inserted by DNA ligase. The plasmid whose 5 'end of the EPO gene is closest to the SP6 promoter (as determined by Bgl I and HindIII digestion) is named pEPO 49F. In this direction, the BamH I site of the pSP6 / 5 polylinker is directly adjacent to the 5'-end of the EPO gene.

[pMMTneo △ BPV]

Plasmid pdBPV-MMTneo (342-12) (Law et al., Mol. And CeII Biol. 3: 2110-2115 (1983)) described in FIG. 8 was completely digested with BamH I to contain the two-segment-BPV genome. Large fragments of about Bkb in length and small fragments of about 6.5 kb containing the replication origin and ampicillin resistance gene, metallothionein promoter, neomycin resistance gene and SV40 polyadenylation signal of pML2 are prepared. The digested DNA was re-reclassified by DNA ligase and plasmids containing only 6.Bkb fragments were identified by EcoRI and BamH I restriction endonuclease digestion. One such plasmid is named pMMTneoΔBPV.

[pEPOl5a]

pMMTneoΔBPV is completely digested with Bg1II. pEP049f is completely digested with BamHI and B9g1II, and about 700 bp fragment containing the complete EPO encoding portion is prepared by gel separation. Bg1II digested pMMTneoΔBPV and 700bp BamH I / Bg1II EPO fragments are ligated and the resulting plasmids containing the EPO cDNA are identified by colonizing hybridization with oligonucleotide d (GGTCATCTGTCCCCTGTCC) probes specific for the EPO gene. Among the plasmids positive by hybridization analysis, the plasmid with EPO cDNA (pEP015a) in the direction in which the 5 'end of the EPO cDNA is closest to the metallothionein promoter is identified by EcoRI and KPnI.

[pBPV-EPO]

Plasmid pEP015a is completely digested with BamH I to linearize the plasmid. Blasmid pdBPV-MMTneo (342-12) was completely digested with BamH I to produce two fragments of 6.5 and Bkb. Bkb fragments containing the complete bovine papilloma virus genome were gel isolated. pEP015a / BamH I and Bkb BamH I fragments are 'ligated together' and plasmids containing BPV fragments (PBPV-EPO) are identified by colony hybridization using oligonucleotide probe α (P-CCACACCCGGTACACA-OH) specific for the BPV genome. . pBPV-EPO DNA

Digestion with HindIII shows that the transcriptional direction of the BPV genome is the same as that of the metalcothionein promoter (same as pdBPV-MMTneo (342-12) in FIG. 8) -plasmid pdBPV-MMTneo (342-12). Is deposited with the accession number ATCC 37224 in the American Type Culture Collection (ATCC) of Rockville, Maryland, and is available therefrom.

[Expression of quality]

The following method is used to express EPO.

[Method 1]

DNA pBPV-EPO was prepared and subjected to standard calcium phosphate precipitation technique (Grahm et at., Virology, 52: 456-67 (1973)) using about 25 μg about I × 10 ⁶ c127 (Lowy et al., J. of Virol., 26: 291-98 (1978)). Five hours after infection, the infection medium is removed, the cells are shocked and washed with glycerol, and then fresh α-medium containing 10% fetal calf serum is added. After 48 hours, cells were antitrypsinic and divided in a 1: 10 ratio in DME medium containing 500 μg / ml G418 (Southern et al., Mol. Appl. Genet: 327-41 (1952)) and cells Incubate for 2-3 weeks. G418 resistant colonies are each isolated in microtiter wells and grown to subsaturation in the presence of G418. The cells are washed and fresh medium containing 10% fetal calf serum is added and the medium is harvested after 24 hours. When the control medium is tested, the EPO is shown to be positive by radioimmunoassay or in vitro bioassay.

[Method 2]

C127 or 3T3 cells are simultaneously infected with 25 μg pBPV-EPO and 2 μg pSV2neo (Southern et al., Described above) as described in Method 1. This is more than about 10-fold molarity of pBPV-EPO. After infection is the same as in the method I.

[Method 3]

Cl27 cells are infected with 30 μg pBPV-EPO as described in Method 2. After infection and dispersion (1:10), fresh medium is changed every 3 days. After about two weeks, spots of BPV transformed cells are evident. Each spot is placed in a 1 cm well microtiter plate and grown into a monosaturated monolayer and analyzed for EPO activity or antigenicity in a control medium.

Example 14

[EPO Stagnation]

COS-cell control media (12 L) with an EPO concentration of 200 μg / L or less are concentrated to 6 (10 mL) using a 10,000-molecular-blocking ultrafiltration membrane such as Millipore pelican immobilized with 5 square feet of membrane. Analyze by RIA as described Residues from the extrafiltration are diafiltered against 4 mM 10 mM sodium phosphate buffered at pH 7.0 Concentrated and diafiltered control medium is 2.5 mg EPO in 380 mg total protein. It contains.

The EOP solution is further concentrated to 186 ml and the precipitated protein is removed by centrifugation at 110,000 * g for 30 minutes.

The supernatant containing EPO (2.Omg) was adjusted to pH 5.5 with 50% acetic acid. After 30 min stirring at 4 ° C., the precipitate is removed by centrifugation at 13,000 * g for 30 min.

Carbonylmethyl Sepharose Chromatography

Centrifuge supernatant (20 ml) containing 200 μg EPO (24 mg total protein) is equilibrated in 10 mM sodium acetate (pH 5.5) and placed in a column packed with CM-Sepharose (20 ml) washed with 40 ml of the same buffer. . EPO bound to CM-Sepharose is eluted with 100 mL of a concentration gradient NaU (0-1) dissolved in 10 mM sodium phosphate (pH 5.5). Fractions containing EPO (50 μg total to 2 mg total protein) are pooled and concentrated to 2 ml using a Amicon YMIO filter membrane.

[Phase-HPLC]

The concentrated fraction from CM-Sephazooz containing EPO is further purified by reverse phase-HPLC using Vydac C-4 column. 1 ml / min in a column paralleled with EPO to 10% dissolved B (Solvent A is 0.1% CF ₃ CO ₂ H dissolved in water; Reagent B is 0.1% CF ₃ CO ₂ H dissolved in CF ₃ CN). Add at flow rate—Column is washed with 10% B for 10 minutes and EPO eluted with a linear concentration gradient B (10-70% at 60 minutes). Aliquots containing EPO are pooled (-40 μg of EPO to 120 μg total protein) and lyophilized. The dried EPO was redissolved in 0.1 M Tris-HCl at pH 7.5 containing 0.15 M Nacl and rechromatated in reverse phase HPLC. Fractions containing EPO are pooled and analyzed by SDS-polyacrylamide (10%) gel electrophoresis (Lameli, UKNature). The collected EPO fraction contains 15.5 μg EPO in 25 μg total protein.

The present invention has been described in more detail by way of preferred specific examples thereof. However, one of ordinary skill in the art may further modify and improve with reference to the specification and drawings without departing from the scope of the invention as set forth in the claims.

[references]

Claims (9)

Production of erythropoietin characterized by isolating glycoproteins having erythropoietin activity by culturing mammalian cells transformed with a recombinant DNA vector containing genomic DNA encoding human erythropoietin in a suitable medium. Way. The method of claim 1 wherein the genomic DNA sequence encoding human erythropoietin is all or part of the following DNA sequence.

The method of claim 1, wherein the genomic DNA sequence encoding human erythropoietin is all or part of the following DNA sequence.

The method of claim 1, wherein the mammalian cell is a COS, 3T3 or Cl27 cell. The method of claim 1, wherein the mammalian cell is a Chinese hamster ovary (CHO) cell. The method of claim 1, wherein the DNA sequence is contained in a vector containing bovine papilloma virus DNA. The method of claim 1, wherein the culture medium contains fetal serum. The method of claim 1, wherein the host cell is a mammalian cell. The method of claim 8, wherein the mammalian host cell is a COS, CHO, Cl27 or 3T3 cell.

Images