HK1021643A - High level expression of human erythropoiet in genein stably transfected mammalian cells - Google Patents

Description

High level expression of human erythropoietin gene in stably transfected mammalian cells

no marking

The present application is a divisional application of the invention patent application having the title of "method for preparing nucleotide sequence corresponding to human erythropoietin gene fragment" with the application number of 87104424.2 and the application date of 26.6.87.

The present invention relates generally to the field of genetic engineering, particularly to expression of glycoprotein products of recombinant genes, and more particularly to high levels of expression of biologically viable human erythropoietin from stably transfected cells.

Erythropoietin, a hormone that plays an important role in regulating erythropoiesis, the formation of red blood cells, and erythropoietin deficiency resulting from anemia. Due to the lack of pure materials, it has been difficult to study this hormone in detail and to try to use it for replacement therapy.

The production of human red blood cells requires the kidneys to secrete erythropoietin in the form of the mature glycoprotein. At steady state, this hormone is present in circulating blood at a concentration of 10 to 18 milliunits per milliliter (128-230 picograms), and its levels can increase by a thousand fold under severe tissue hypoxic (anoxic) stimulation. The hormone level is increased to trigger the proliferation and differentiation of the stem cell group in the bone marrow, activate the synthesis of hemoglobin in mature red blood cells, accelerate the release of red blood cells from the bone marrow to enter the circulation, thereby increasing the amount of red blood cells and improving the situation of insufficient oxygen supply. Patients lacking erythropoietin, such as chronic renal patients, often suffer from severe anemia.

Erythropoietin is a 34-38 kilodalton glycoprotein, with about 40% of its molecular weight being supplied by carbohydrates. At least one disulfide bridge is required for viability, and little is known about the structure of this hormone, nor is the details of its synthesis fully understood. Recently isolated CDNA and genomic clones have provided the opportunity to analyze regulatory control of erythropoietin production, but have not expressed sufficient quantities of biologically viable human erythropoietin to be used in replacement therapy.

In accordance with the present disclosure, biologically viable human erythropoietin can be expressed at high levels (in excess of two million nominal titer units per liter of supernatant) from stably transfected mammalian cell lines, providing a vast source of pure human erythropoietin for clinical use. This surprisingly high level of erythropoietin expression is achieved by transfecting host cell lines with the ApaI restriction fragment of the human erythropoietin gene. The sense strand of the ApaI restriction fragment has the nucleotide sequence shown in FIG. 1.

FIG. 1 shows an ApaI restriction fragment of 2426 base pairs containing the sequence of the human erythropoietin gene;

FIG. 2 depicts a representative plasmid expression vector (pD11-Ep) containing a 2426 base pair ApaI restriction fragment;

FIG. 3 depicts another expression vector (pBD-Ep) with an ApaI restriction fragment.

In a preferred embodiment, the genetically engineered ApaI restriction fragment shown in FIG. 1 is inserted into a mammalian expression vector as shown in FIGS. 2 and 3, and the expression vector is then introduced into a mammalian cell line to develop stably transfected cells that produce large quantities of biologically viable human erythropoietin. The ApaI restriction fragment of the human erythropoietin gene is selected to provide the highest efficiency in transcription of erythropoietin messenger RNA, efficient translation of the RNA and post-translational modification to produce a biologically viable mature erythropoietin glycoprotein. Specifically, it is important to remove the 5' interfering sequence from the erythropoietin gene, while retaining the enhancer sequence. To preserve potentially valuable enhancer sequences, introns within the apai restriction fragment are also preserved. Some 3 'untranslated sequences on the 3' end of the gene are also retained, optimizing the putative regulatory sequences. The enhanced expression provided by the apai fragment is demonstrated in the following examples, where the fragment in combination with two promoters and two cell lines resulted in stable high level expression of erythropoietin.

The following examples are put forth so as to illustrate the advantages of the invention and to assist one of ordinary skill in making and using the invention. These examples are not intended to limit the scope of the disclosure, or the protection afforded by the patent certificate, in any way. EXAMPLE 1 extraction of Gene clones

The human genomic library constructed in lambda phage (Cell,15:1157-1174,1978) was screened using low stringency hybridization conditions and the oligonucleotide probe mixture described in Cell,38:287-297,1984, hereby incorporated by reference.

Oligonucleotide mixtures were prepared using an Applied Biosystems synthesizer, end-labeled with 32P-ATP and T4 polynucleotide kinase. The synthetic oligonucleotide was designed to correspond to the amino acid sequence of the amino terminal portion:

H2N-Ala-Pro-?-Arg-Leu-Ile-Leu-Asp-Ser-Arg-Val-Leu-Glu-Arg-

Tyr-Leu-Leu-Glu-Ala-Lys-Glu-Ala-Glu-?-Ile-Thr-Asp-Gly-Gly-Ala

24 this sequence was obtained from the human protein purified from the urine of aplastic anemia persons by Yanagawa et al (J.biol.chem.259:2707-2710, 1984). In order to reduce the degree of codon degeneracy in this amino acid sequence, the codon usage rules of Grantham et al (Nucleic Acids Research,8:43-59,1981) and Jaye et al (Nucleic Acids Research,11:2325-2335,1983) were used. These rules take into account the relative rarity of dinucleotide CPG in vertebrate DNA, and the appropriate avoidance of potential A: G mismatches. The amino acid at position 24 is most likely asparagine (J.biol.chem.259:2707-2710, 1984). For the amino acid sequence Val-Pro-Lys-Val-Pro-Val-asparagine, 2 sets of 72 each corresponding to the expected codon sequence were synthesized, the first set being 20 nucleotide probes for TT (C/T) TC (a/g/T) GC (C/T) TC (C/T) TT (a/g/T) -GCTTC, and the second set replacing T with C at position 18. For the amino acid sequence valley-asparagine-isoleucyl-threo-aspartyl-glycine, a set of 23 nucleotide probes with the sequence AGC TCC TCC ATC AGT ATT ATT T (c/t) was constructed.

Plaques hybridized with the oligonucleotide probe were rescreened at low density until purified. After the initial positive phage clones were purified by plaque, Jaco bs et al published further sequence data for the erythropoietin gene (Nature,313:806-810, 1985). Oligonucleotides were constructed from this data and used to verify positive clones, which were digested with EcoRI restriction enzyme, their insert DNA purified by gel electrophoresis and ligated into the previously digested EcoRI restriction enzyme by standard techniquesIn the digested PUC13 plasmid. DNA sequence determination by the dideoxynucleotide chain termination method (Proc. Natl. Acad. Sci. USA,74:5463-5467,1977) using dATP (. alpha. -³⁵S) and a universal 17-nucleotide primer, or using a special oligonucleotide primer in a selected region.³²P-ATP was obtained from ICN and the enzyme was obtained from New England Biolabs or Bethesda Research Laboratories.

About 4.8X 10⁵Individual phages were screened by duplicate nitrocellulose filter hybridization. Three different clones remained positive during plaque purification, with the DNA insert being restriction mapped and partially dideoxy sequenced. Two of the three clones apparently contain the full information of the erythropoietin gene. The restriction map of these clones and the sequence of their 2426 base pair ApaI fragment (see FIG. 1) are essentially identical to the human erythropoietin gene sequence recently published by Jacobs et al (Nature,313:806-810, 1985). The frequency of phages containing the erythropoietin gene proposed in this expanded library was low, about 2X 10⁶1 of each phage-consistent with the statement that the erythropoietin gene is present in single copy in the human chromosomal genome. Southern blot hybridization of the ApaI fragment or other restriction fragment of the erythropoietin gene to total human chromosomal DNA revealed only a single hybridization zone and no other highly homologous DNA regions. Example 2 selection of ApaI restriction fragments

For the construction of the expression vector, the ApaI restriction fragment of the erythropoietin gene was prepared according to the technique described in Proc.Natl.Acad.Sci.USA 74:5463-6467,1977, which is hereby incorporated by reference. Briefly, the extracted phage DNA was digested with the restriction enzyme Apa I (Bethesda Research Laboratories), then separated on a 1% agarose gel, electrophoretically eluted, phenol extracted and ethanol precipitated. This fragment was verified by partial sequencing as in example 1.

Referring to FIG. 1, the inserted ApaI restriction fragment contains a 58 base pair 5 'untranslated sequence (nucleotide 0001-0058) followed by a putative 27 amino acid signal peptide coding sequence, a mature protein, four putative inserts, and a 222 base pair 3' noncoding DNA sequence. At the 5' end of the gene, the ApaI site is 58 base pairs upstream of the start codon ATG (0058) of the protein sequence. This 5' end site (0001) was chosen to avoid a false start immediately upstream of the ApaI restriction site, while preserving regulatory sequences believed to be important for efficient ribosome binding and processing. The 3 'apai site (2426) was chosen to retain the putative processing signal in most of the 3' untranslated region and to remove further downstream sequences. The use of the complete erythropoietin gene, including its intron sequences, is intended to preserve potential regulatory and enhancer sequences within the intron that may play a role in the expression of the erythropoietin gene or in the modification and secretion of proteins. EXAMPLE 3 construction of expression plasmid with ApaI restriction fragment

The apai restriction fragment of 2426 base pairs containing the entire human erythropoietin gene was inserted into two expression vectors, pD11 and pBD, which were based on different mammalian promoters, respectively.

Plasmid expression vector pD11 was derived from the aforementioned plasmid (Nucleic acids research,13:841-857,1985, cited herein) and contained the enhancer sequence and origin of replication of simian kidney virus 40(SV40) and the major late promoter and triple leader sequence of adenovirus 2. The apai fragment of the human erythropoietin genomic sequence (fig. 1) was gel purified, single-stranded ends filled with T4 DNA polymerase, two blunt-ended ends linked with a bamhi linker, and the construct inserted into a single bamhi restriction site of pD11 to transcribe the erythropoietin gene directly from a strong promoter.

The structure of the constructed expression plasmid pD11-Ep carrying the ApaI fragment (Ep) is depicted in FIG. 2. Plasmid pD11 was constructed by inserting the EcoRI (RI) sites of pML with the following elements: the left terminal adenovirus (0-1) containing 350 bp, the SV-40 origin of replication and enhancer sequences (E), the major late promoter of adenovirus (MLP), the triple leader sequence of adenovirus 2 (L1-3), the 5 'splice site of the third leader sequence (5' SS), the 3 'splice site of immunoglobulin (3' SS), and the SV-40 late polyadenylation signal (pA). Cloning the recombinant plasmid in E.Coli HB101, carrying out cesium chloride isopycnic centrifugation purification, wherein the expression plasmid pD11-Ep is about 6500 base pairs, and the constructed product is verified by a restriction map and a partial dideoxy sequencing method.

pBD contains the metallothionein I promoter sequence (MT-1, Glanvi11e, Durham and Palmiter, Nature,292:267-269,1981 hereby incorporated by reference) and the pUC plasmid with the dihydrofolate reductase (DHFR) selectable marker sequence. Referring to FIG. 3, the ApaI restriction fragment (Ep) of 2426 base pairs was inserted into the unique SmaI restriction site of pBD to form expression plasmid pBD-Ep. The DHFR sequence of pBD-Ep is linked to the replication initiation and enhancer sequences of SV-40 and to the hepatitis B virus surface antigen poly (A) sequence. pBD-Ep was constructed substantially as described above for pD 11-Ep.

Although plasmid vectors are used in these confirmatory experiments, it is also contemplated that the Apa i erythropoietin gene fragment may be introduced into the host cell line in a similar manner by using viral or retroviral expression vectors instead. DNA viruses suitable for this purpose include adenovirus or BPV (bovine papilloma virus), while suitable retroviruses are also known and readily available. It is clear that such Retroviral (RNA) vectors will carry the antisense strand of the ApaI restriction fragment, i.e., the RNA transcript sequence corresponding to the sense strand of FIG. 1, or an equivalent thereof. Retroviral vectors will also include sequences for the transchromosomal acting elements required for reverse transcription of the viral genome and integration of the viral DNA into the host genome. EXAMPLE 4 transfection of mammalian cells

Two mammalian cell lines, each transfected with pD11-Ep and pBD-Ep. Mammalian cell lines COS-7 (monkey kidney) and BHK (baby rat kidney) were maintained in modified Dul becco minimal medium containing 10% fetal bovine serum. Cells were transfected by the calcium phosphate method when they were passaged to 50-70% confluence (Virology,52:456-467, 1973). BHK is derived from kidney epithelial cells, which are believed to be most likely to produce erythropoietin in its native state in mammals when anemic or hypoxic, and thus, it is possible for these cells to recognize critical regulatory sequences on the erythropoietin Apa I fragment, efficiently process and produce erythropoietin glycoprotein.

For the temporal expression of cells, 20. mu.g of total DNA, 10. mu.g of plasmid pD11-Ep carrying the erythropoietin gene and 10. mu.g of vector salmon sperm DNA were added to a 100 mm dish, and after 48 hours, the supernatant was collected, centrifuged at 400g for 10 minutes to remove cells and particles, frozen at-20 ℃ and cells were collected separately, and the temporal expression results were transfected with pD11-Ep alone as shown in Table 1, and the data were obtained in three experiments for each cell type.

TABLE 1

Erythropoietin mammalian cell protein (microgram) units per milliliter culture (in vitro bioassay)

BHK 3.4±0.2 270±16

COS-7 3.2±0.4 255±32

Secretion of erythropoietin into the supernatant from mammalian cell lines COS-7 or BHK was observed to be about 80-fold higher than previously reported for erythropoietin-encoding cDNA or for transient expression using the entire erythropoietin gene.

To establish a stable cell line producing high levels of erythropoietin, COS-7 or BHK cells were transfected simultaneously with pD11-Ep and pDHFR-1a plasmids (the latter being a similar mammalian expression vector containing dihydrofolate reductase cDNA). Instead, the transfection procedure was performed using 5. mu.g of pD11-Ep plasmid, 5. mu.g of pDFFR-1 a plasmid and 10. mu.g of vector DNA for co-transfection, followed by incubation for 18-24 hours, and then various concentrations of methotrexate (10 nmol/L to 1 mmol/L) were added to the medium. Cells incorporating the DHFR gene survive on this selection medium. After several days of incubation, colonies resistant to methotrexate were isolated, passaged and the supernatants screened for erythropoietin viability, and approximately half of the detected colonies resistant to methotrexate secreted measurable erythropoietin viability.

To establish a stable cell line with the expression vector pBD-Ep, BHK cells were transfected with the calcium phosphate method, after 18-24 hours, 1. mu.M to 1 mM methotrexate was added to the medium, and after several days of culture, colonies that survived higher methotrexate were isolated, passaged and screened, and in this series all colonies tested against methotrexate secreted a measurable erythropoietin activity.

In order to optimize the expression of transcription units containing the erythropoietin gene and DHFR gene, each BHK cell line secreting high levels of erythropoietin, containing pBD-Ep or pD11-Ep, was passaged several times in increasing concentrations of methotrexate (Nature,316:271-273, 1985). However, rather than applying selective pressure to cell lines by small, stepwise increases in the concentration of methotrexate, these cells are immediately challenged by very high concentrations of methotrexate (i.e., 1 mmol/l). Only a few cells survive, but in these cells the plasmid constructs are incorporated into the DNA at a site that is particularly advantageous for expression (the so-called hot spot) and/or possess copies of many constitutive transcription units. Thus, the highest producing cell lines could be selected in one step, and if high erythropoietin production was maintained for more than 15 passages in the absence of methotrexate selection pressure, these cell lines (including F7.2 and S5.2 listed in Table 2) were considered stable.

The activity of erythropoietin cannot be quantified in cell pellets because there are many factors in the cell extract that inhibit the assay. The results listed in Table 2 are therefore only the erythropoietin protein produced and secreted into the supernatant by the cell lines and the intracellular erythropoietin levels were not analyzed.

TABLE 2

From stably transfected BHK cell lines

Expressed recombinant erythropoietin

Erythropoietin cell line protein (microgram) units per milliliter of supernatant (in vitro bioassay) pBD-EpF1.112.4970F3.432.02500F6.179.66210F7.284.16728

Table 2 (continuation)

The observed erythropoietin secretion in S1.26.4500S2.464.25000S5.282.16400 units (in vitro bioassay) of the erythropoietin cell line pD11-Ep protein (microgram) per ml supernatant corresponds to a nominal yield of up to seven million units per liter. This nominal yield is multiplied by one thousand times the observed yield per ml to give the expected yield per liter for large scale production.

The erythropoietin production observed using the ApaI fragment is more than three hundred times greater than that reported earlier for stable transformed CHO cell lines using different human erythropoietin gene fragments (Lin et al, Proc. Nat1.Acad. Sci. USA,82: 7580-.

Control experiments in these transfectant assays, including supernatants from untransfected cells, and supernatants from parallel cultures of cells transfected with DNA encoding other proteins, including bacterial chloramphenicol transacetylase and human coagulation factor IX-, did not result in measurable erythropoietin activity in either control cultures, mock transfected or transfected with other genes. It should also be noted that in each of the above experiments with the erythropoietin gene, the expression level obtained from the selected cell line is irrelevant whether the selection marker associated with the erythropoietin gene is co-infected therewith or a plasmid containing the ApaI fragment is inserted prior to infection.

Other representative infection methods suitable for use in the present invention include DEAE-dextran mediated transfection techniques, lysozyme fusion or erythrofusion, scrape, direct uptake, osmotic shock or glucoshock, direct microinjection or indirect microinjection, such as erythrocyte mediated techniques and/or methods of placing host cells in an electric current, among others. Transfection is the transfer of genetic information, specifically the coding information of the Apa I restriction fragment of the human erythropoietin gene, into cells using extracted DNA, RNA or synthetic nucleotide polymers, and the transfection techniques listed above are not necessarily perfect, as other means for introducing genetic information into cells will undoubtedly be developed. Prior to transfection, the apai restriction fragments typically may be ligated (ligated) to other nucleic acid sequences, such as promoters, enhancers, and polyadenylation sequences. Although various host cell lines of mammalian origin are described, and particular reference is made to renal epithelial cells, it is contemplated that other eukaryotic and prokaryotic (bacterial or yeast) host cell lines may be used with the present invention. Recent success in introducing mammalian genes into plant cells has provided potential for the use of plant and algal cells. EXAMPLE 5 expression of erythropoietin from transfected cell lines

Erythropoietin secreted into the supernatant of transfected cell lines is biologically viable, secreting large amounts of hormone up to seven thousand units per ml.

The in vitro assay for erythropoietin viability was based on the formation of red colonies (CFU-E, red colony forming cells) on plasma clot medium from mouse bone marrow cells (Bloodcells,4:89-103,1978) and the sensitivity of this assay was approximately 5 milliunits/ml. Erythropoietin, as a standard for the assay, is a preparation obtained by partial purification of plasma from anemic sheep (Connaught, EP, batch No. 3026) and is used in the assay of supernatants from continuous cell lines grown for 24 hours on fresh medium without methotrexate at 2X 10 cells/ml⁹Bone marrow cells, 10% citrate-treated bovine serum, 20% fetal bovine serum, 1% bovine serum albumin and 1.6% fetal bovine extract (Gibco) were assayed in a medium containing 1 to 10. mu.l of supernatant diluted 1: 200 with the medium, and after culturing for 36 to 48 hours, plasma clots were fixed on a glass slide and stained with benzidine for hemoglobin to count the number of erythrocyte colonies. Colonies produced by CFU-E were not detected when erythropoietin was not added. In general, the most suitable colony growth can be observed with 50 milliunits of erythropoietin (0.64 nanogram) per milliliter of culture (per 2X 10)⁴100-150 CFU-E cells were detected for each bone marrow cell). As shown in tables 1 and 2, a large amount of erythropoietin hormone was secreted into the transfected cell line supernatants up to 7000 units per ml, which, assuming that the specific activity of recombinant erythropoietin is equivalent to native erythropoietin (78000 units per mg protein), corresponds to approximately 80 μ g erythropoietin protein per ml.

In addition, multivalent rabbit antisera against human erythropoietin (J.cell. physiol.118:87-96,1984) were used in a competitive radioimmunoassay to detect immunoreactive erythropoietin in supernatants from selected cell lines. The amount of protein detected by radioimmunoassay was equivalent to the protein level estimated by the bioassay. These data indicate that more than about 98% of the erythropoietin protein expressed and secreted by transfected cell lines is viable.

The transfected cells were further examined for recombinant erythropoietin production. To confirm that pillboxes secreted by these cells are natural hormones, selected cell lines were used for in vivo detection of erythropoietin in mice with severe hypoxia-induced erythropoiesis (Nature,191: 1069-. The supernatant liquid secreted by the cell line is detected by a mouse which causes erythrocyte proliferation due to severe hypoxia, and has strong in-vivo biological activity. In experiments with partially purified native erythropoietin, it has long been noted that sialidase treatment completely abolished erythropoietin viability when tested in whole animals (J.biol.chem.247:5159-5160, 1958). This loss of viability is presumably due to enhanced clearance of asialohormone by the liver, since sialidase-treated erythropoietin retains its full viability when tested in vitro. The strong biological viability observed in vivo, indicates that the transfected mammalian cell line is in the post-translational modification phase. Appropriate sugars and terminal sialic acid were added to the erythropoietin protein.

In a separate experiment, the activity of erythropoietin detected in vitro was also neutralized by anti-human erythropoietin antibody added to the culture medium.

Representative transfected cell lines secreted erythropoietin into the supernatant and their proliferative effects on other myeloid progenitor cells were also examined. Testing the effects of recombinant erythropoietin on various progenitor cells derived from human and mouse bone marrow, including: red colony forming cells (CFU-E), red burst-forming cells (BFU-E), granulocyte-macrophage precursors (CFU-GM), and mixed cell colony forming cells (CFU-Mix) (J.cell.physiol.Suppl.1:79-85,1982; J.cell.Phy-sio.118: 87-96,1984). The proliferative effects exhibited by red stem cells on recombinant erythropoietin paralleled the dose relationship exhibited on native erythropoietin. The increase in the concentration of recombinant erythropoietin from that for CFU-GM and CFU-Mix to 10 units per ml of test cell culture did not show any proliferative effect.

The purified recombinant erythropoietin electrophoresed the same as purified erythropoietin in the urine of aplastic anemia persons when analyzed by SDS-PAGE under reducing or non-reducing conditions. These proteins exhibit the same microscopic heterogeneity, with the major components all at a molecular weight of 34 kilodaltons.

While the invention has been described with reference to the embodiments disclosed, it will be understood by those skilled in the art that various changes may be made, equivalents may be substituted, and other changes may be made in the components and methods described above upon reading the above examples. Accordingly, it is intended that the scope of protection of this patent shall be limited only by the terms of the appended claims and equivalents thereof.

Claims (8)

1.A method for expressing biologically active recombinant human erythropoietin which comprises transfecting host cells with DNA, RNA or nucleotide sequence comprising essentially a 2.4kb Apa I restriction fragment of the human erythropoietin gene, contacting the transfected cells with a culture medium, allowing the cells to express erythropoietin, and recovering the expressed erythropoietin.

2. The method of claim 1, wherein the apai restriction fragment is carried on a plasmid or virus.

3. The method of claim 1, wherein the host cell line is a mammalian cell.

4. In a method of expressing biologically viable recombinant human erythropoietin in a cell line contacted with an incubated medium, the improvement which comprises incorporating into said method a cell line capable of producing erythropoietin in the incubated medium, said cell line produced by transfecting a host cell line with DNA, RNA or nucleotide sequences comprising predominantly a 2.4kb Apa I restriction fragment of the human erythropoietin gene.

5. A method for expressing biologically viable recombinant human erythropoietin in a cell line contacted with an incubation medium according to claim 4, wherein said cell line is capable of providing a nominal production of at least two million units of erythropoietin per liter of incubation medium.

6. The method of claim 4, wherein the ApaI restriction fragment is carried on a plasmid.

7. The method of claim 4, wherein the ApaI restriction fragment is carried by the virus.

8. The method of claim 4, wherein the host cell line is a mammalian cell.