FI106563B - Process for Preparation of Human Erythropoietin Analogs and DNA Sequences Encoding These - Google Patents

Description

x 106563

METHOD FOR THE PREPARATION OF HUMAN ERYTHROPOEETIN ANGLES AND THE DNA sequences encoding them This application is separated from FI patent application No. 912829.

This invention relates to a process for the preparation of human erythropoietin analogs, to DNA sequences encoding them, and to eukaryotic host cells transfected with them.

Background of the Invention

Erythropoietin is a glycoprotein hormone involved in the maturation of erythrocyte stem cells into erythrocytes. It is essential for the regulation of circulating red-15 cells. Naturally occurring erythro poietin is produced in the liver during the fetal phase and in the kidneys in adults and circulates in the blood and stimulates the formation of red blood cells in the bone marrow. Anemia is almost invariably due to renal dysfunction due to decreased renal production of erythropoietin. Recombinant erythropoietin produced by genetic engineering, comprising expression of a protein product in a host cell transformed with an erythropoietin encoding gene, has been found to be effective in the treatment of anemia due to chronic renal failure.

Until recently, the availability of erythropoietin was very limited. Although the protein is present in human urine, secreted levels are too low to make it a useful source of erythropoietin for therapeutic use. Patients with aplastic anemia have higher levels of urinary erythropoietin compared to healthy subjects, but the limited amount of urine makes this source impractical. Purification of human urinary erythropoietin 2,106563, described by Miyake et al., J. Biol. Chem., 252, 5558 (1977), was used as a starting material for the urine of individuals with aplastic anemia.

Identification, cloning and expression of genes encoding erythropoietin are described in U.S. Patent No. 4,703,008 to Lin et al. Description of purification of recombinant erythropoietin from a nutrient medium that maintains recombinant erythropoietin plasmid containing mammalian hosts. 4,667,016, 10 Lai et al. Expression and recovery of biologically active recombinant erythropoietin from mammalian host cells containing recombinant erythropoietin gene has for the first time made available therapeutically effective amounts of erythropoietin. In addition, knowledge of the gene sequence and availability of higher amounts of purified protein has led to a better understanding of the mode of action of this protein.

The biological activity of a protein depends on its structure. In particular, the primary structure of the protein (i.e., its amino acid sequence) provides information that allows the formation of secondary (i.e., α-helix or β-sheet) and tertiary (three-dimensional folds) structures during and after synthesis of the polypeptide. Degradation of the proper secondary and tertiary structure in the production of mutations or chemical or enzymatic treatments can lead to a decrease in biological activity.

In prokaryotic organisms, the biological activity of proteins is largely regulated by the structures described above. Unlike prokaryotic cell proteins, many surface and secretory proteins of eukaryotic cells are modified by one or more oligosaccharide groups. This modification, termed 35 glycosylation, can dramatically affect the physical properties of proteins 3 106563, and may also be important for protein stability, secretion, and intracellular localization. Proper glycosylation can be essential for biological activity. In fact, some of the gee-5 nit eukaryotic cells expressed in bacteria (e.g., E. coli) lacking cellular functions for glycosylation of proteins, produce proteins that are obtained with little or no activity due to lack of glycosylation.

Glycosylation occurs at specific sites along the polypeptide and is generally of two types: O-linked oligosaccharides attach to serine or threonine residues, while N-linked oligosaccharides attach to asparagine residues as part of the Th , where X can be any amino acid except proline. The structures of the N-linked and O-linked oligosaccharides and the sugar residues that occur in the different types are different. One type of sugar commonly found in both is N-acetylneu-20 ramic acid (hereinafter referred to as sialic acid). Sialic acid is generally a terminal residue in both bound and O-linked oligosaccharides and, due to its negative charge, can give an acidic property to a glycoprotein.

Both human urine-derived erythropoietin and recombinant erythropoietin (expressed in mammalian hand cells) having human erythropoietin amino acid sequence 1-165 contain three N-linked and one 0-linked oligosaccharide molecules comprising 30% to about 40% glycoprotein.

N-linked glycosylation occurs at asparagine residues at positions 24, 38, and 83, while O-linked glycosylation occurs at serine residues at position 126 (Lai et al., J. Biol. Chem. 261, 353116 (1986); Broudy et al., Arch. Biochem. Biophys. 265, 4 106563 329 (1988). The oligosaccharide chains have been shown to be modified by terminal sialic acid residues. Enzymatic treatment of glycosylated erythropoietin to remove all sialic acid residues results in decreased activity in vivo but does not affect activity in vitro (Lowy et al., Nature 185, 102 (1960);

Goldwasser et al., J. Biol. Chem. 249, 4202 (1974)). This behavior has been explained by the rapid elimination of asialoerythropoietin from the circulation by interaction with hepatic 10 asialoglycoprotein binding protein (Morrell et al., J. Biol. Chem. 243, 155 (1968); Briggs et al.,

Am. J. Physiol. 227, 1385 (1974); Ashwell et al., Methods Enzymol. 50, 287 (1978)). Thus, erythropoietin has activity in vivo only when it is sialylated to avoid its binding to hepatic binding protein.

The role of other constituents in erythropoietin oligosaccharide chains is not well defined. It has been shown that non-glycosylated erythropoietin has a greatly reduced activity in vivo compared to the glycosylated form but retains activity in vitro (Dordal et al., Endocrinology 116, 2293 (1985); Lin, supra). However, in other studies, deletion of N-linked or O-linked oligosaccharide chains alone or in combination by mutagenesis of asparagine or serine residues acting as glycosylation sites significantly reduces the activity of modified erythropoietin produced in mammalian cells in vitro (J. Dube et al., Chem. 17516 (1988).

Glycoproteins such as erythropoietin can be resolved into various charged forms using techniques such as isoelectric focusing (IEF). IEF studies of unpurified and partially purified erythropoietin formulations have been reported by several parties (Lu-kowsky et al., J. Biochem. 50, 909 (1972); Shelton et al., 35 Biochem. Med. 12, 45 (1975); Fuhr et al. et al., Biochem. Bio-10 106563 Phys. Res. Comm. 98, 930 (1981)). No more than three or four fractions with erythropoietin activity were detected by IEF in these studies, and none were characterized by carbohydrate content. Also, no correlation between the isoelectric points of the fractions and the biological activity was made.

During urine erythropoietin purification from human urine as described by Miyake et al., Direct. two erythropole-10 ethene fractions of hydroxylapatite chromatography, named II and IIIA, were reported to have the same specific activity. The following carbohydrate analysis of Fractions II and IIIA showed that Fraction II has a higher mean sialic acid content than Fraction IIIA (Dordal et al. Supra).

It is an object of the present invention to provide human erythropoietin as an analog which has multiple glycosylation sites relative to human erythropoietin. The isoforms of these analogs have a certain sialic acid content and biological activity. Pharmaceutical compositions containing such molecules would have therapeutic utility. They can be used to raise hematocrit levels in mammals to increase the production of reticulocytes and red blood cells.

SUMMARY OF THE INVENTION The present invention relates to a process for the preparation of a human erythropoietin analogue having an amino acid sequence different from the amino acid sequence 1-165 or 1-166 of human erythropoietin shown in Figure 1, which increases glycosylation of at least . The method of the invention is characterized in that a) one or more alterations are made to the human erythropoietin amino acid sequence to form an analog having at least one new glycosylation site, and b) the analog is expressed in a host cell where the carbohydrate chain is attached to a new glycosylation site.

The human erythropoietin analogs prepared according to the invention thus have a greater number of sites for 5 carbohydrate attachment than human erythropoietin such as [Asn69] EPO, [Asn125, Ser127] EPO, [Thr125-] EPO and [Pro124, Thr125] EPO.

The invention also relates to DNA sequences encoding the analogs described above.

The invention further relates to eukaryotic host cells transfected with such DNA sequences, wherein the host cells express analogs of human erythropoietin.

Brief Description of the Contents of the Figures Figure 1 shows the amino acid sequence of human erythropoietin. The squares indicate asparagine residues to which the carbohydrate chains are attached and the stars indicate carbohydrate-modified threonine and serine residues. The 20 additional glycosylation sites provided by the analogs of the example have been shown to be mutated to asparagine, serine and threonine.

Figures 2A, 2B and 2C show a series of cloning steps used to generate plasmids for the construction and analysis of human erythropoietin analogs. These analogs have amino acid changes, as shown in Figure 1, which provide additional Gly cosylation sites.

Figure 3 shows a Wes-30 tern blot analysis of erythropoietin having human sequence and the shown erythropoietin analogs in COS cell supernatant. The analogs [Asn9, Ser11] EPO, [Asn69] EPO, [Asn125, Ser127] EPO, and [Pro124, Thr125] EPO are constructed as described in the example. The analogs [Pro125, Thr127] EPO, [Asn126, Ser128] EPO and [Thr125, Ser127] EPO, which do not contain additional carbohydrate-35 chains, are shown for comparison purposes.

7 106563

Figure 4 shows a Western blot analysis of human sequence-derived erythropoietin and shown erythropoietin analogs after COS cell supernatant treatment with N-glycanase. The analogs [Thr125] EPO and [Pro124,5 Thr125] EPO are constructed as described in the example.

The analogs [Val126] EPO, [Pro124] EPO, [Pro125] EPO, [Thr127] EPO, [Pro125, Ser127] EPO and [Thr125, Ser127] EPO are shown for comparison.

Figure 5 shows the isoelectric focusing gel from poles 2, 3, and 4 obtained by Q-Sepharose and C4 reverse phase chromatography on medium maintaining CHO cells transfected with [Thr125 mutation-containing erythropoietin cDNA. Purified recombinant erythropoietin containing a mixture of isoforms obtained using Lain et al. supra, except that DEAE agarose chromatography has been replaced by Q-Sepharose chromatography, it is also shown in the left and right lanes of the gel.

Figure 6 shows the relationship between the amount of sialic acid per 20 erythropoietin isoforms and the specific activity of each isoform in vivo, expressed as mg per erythropoietin polypeptide. In Figure 6A, the concentration of each erythropoietin isoform was determined by the Bradford protein assay; in Fig. 6B the concentration was determined by absorbance at 280 nm, in Fig. 2C the concentration was determined by RIA.

DETAILED DESCRIPTION OF THE INVENTION In accordance with the present invention, analogs of erythropoietin are prepared.

As used herein, the term "human erythropoietin analogue" refers to an erythropoietin having one or more changes in the amino acid sequence of human erythropoietin, resulting in an increase in sialic acid attachment sites. Analogs are produced by site-specific mut-35 genesis and contain additions of amino acid residues, • j · ·.

8 106563 deletions or substitutions that alter sites for Gly cosylation. Such analogs have a higher number of carbohydrate chains than human erythropoietin.

Analogs leading to increased biological activity are constructed by increasing the sialic acid content of the erythropoietin molecule. Analogs with higher levels of sialic acid than those found in human erythropoietin are generated by the addition of glycosylation sites that do not interfere with the secondary or tertiary structure required for biological activity. Preferably, the human erythropoietin analogue has 1, 2 or 3 additional sites for N-glycosylation or O-glycosylation. For example, leucine at position 69 is replaced by asparagine to provide the Asn-Leu-Ser sequence, which provides a fourth site for N-glycosylation. Such a change can usually provide up to four additional sialic acids per molecule. Examples of other changes that create additional N- or O-glycosylation sites include alterations of alanines at position 125 and 127 to asparagine and serine respectively, alanine at position 125 to threonine and alanines at positions 124 and 125 to proline and threonine, respectively. As will be understood by those skilled in the art, the present invention encompasses many other human erythropoietin analogs that have additional sites for glycosylation.

Erythropoietin isoforms, their preparation, activity and assay for activity are described in detail in the parent application FI 912829 of this publication.

30 Isoelectric focusing (IEF) separates proteins based on charge. Set in a pH gradient and obtained by an electric field, the proteins move to and remain in a position where they have no net charge. This is the isoelectric point (pl) of the protein. Each discrete band detected in IEF 35 represents molecules having ♦ - ≥ 9 106563 given p1, and therefore the same total charge, termed isoform. The term "erythropoietin isoform" as used herein refers to erythropoietin preparations having one p1 and the same amino acid sequence.

In a preferred embodiment, the erythropoietin is the product of expression of exogenous DNA transfected into a non-human eukaryotic host cell, meaning that in the preferred embodiment, the erythropoietin is "recombinant erythropoietin". The recombinant erythro-10 poietin is preferably produced according to the methods described in U.S. Patent 4,703,008 to Lin et al., Which is incorporated herein by reference. The recombinant erythropoietin is preferably purified according to the general procedures described in U.S. Patent No. 4,667,016 to Lai et al., Ex. 15, which is incorporated herein by reference, or alternatively, the DEAE-agarose chromatography is replaced by Q-Sepharose chromatography. . In the Q-Sepharose column modification, 55 mM NaCl replaces 25 mM NaCl in buffer used to bring the column to neutral pH-20, and 140 mM NaCl replaces 75 mM NaCl in buffer used to elute the erythropoietin from the column. This material, when analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, moves as a single species (it is a band). When purified erythropoietin, 25, is introduced into IEFr, a large number of bands are visible, indicating that various charge forms of glycoprotein are present.

Within the framework of the embodiment of the invention disclosed in the parent application, it has been found that various isoforms of recombinant erythropoietin having the amino acid sequence of human erythropoietin derived from urine correspond to erythropoietin molecules having 1 to 14 in vivo, which is related to the amount of sialic acids contained in the isoform.

The term "erythropoietin" as used herein includes naturally occurring erythropoietin, urinary human erythropoietin, as well as non-naturally occurring polypeptides having sufficiently similar amino acid sequence and glycosylation to naturally occurring erythropoietin to exert its biological properties in vivo. increase in revenue.

Thus, there is a relationship between the relative specific activity of erythropoietin in vivo and the amount of sialic acid residues per molecule of erythropoietin in isoforms 5-11 (each isoform is named here based on the number of sialic acids per molecule of erythropoietin). Isoforms 11-14 have approximately the same relative activity in vivo. Isoforms 5-14 are analyzed by bioassay of exhypoxic polysythmic mice in vivo, and the amount of each isoform present is determined by the Bradford protein assay, absorbance at 280 nm, or radio-20 immunoassay (RIA) for erythropoietin. RIA assays (Egrie et al., Immunobiology 172, 213 (1986)), expressed in units / ml, are divided by 212,770 units / mg of erythropoietin polypeptide, the average specific activity of purified erythropoietin as determined by 25 RIAs, to give isolated isoforms or isoforms mg erythropoietin polypeptide / ml. As shown in the Reference Examples, the relative in vivo specific activities increase gradually from isoform 5 to isoform 11 (see Table 2 in Table 30).

The in vivo specific activities referred to herein are assays for relative in vivo specific activities and not absolute in vivo specific activities. For the purposes of the present invention, specific activities are used to compare the relative activities of the isoforms determined using the same assay using the same conditions comprising the same internal standard, the same animal type, the same assay used to calculate the specific activity. and the same assay for protein content. It is not intended that any of the in vivo specific activity values reported for any isoform would represent an intrinsic or absolute value for this isoform.

One embodiment of the invention relates to mammalian (e.g. Chinese hamster ovary, CHO) host cells, which preferably synthesize erythropoietin isoforms having a greater than specific amount, e.g., greater than 15 sialic acids per molecule. Erythropoietin mimicryl molecules have N-linked or O-linked oligosaccharide structures which may limit the sialic acid content of the molecule. For example, tetra-antennal (quaternary) N-linked oligosaccharides most commonly provide four possible sites for attachment of sialic acid, while bi- and tri-antennal oligosaccharide chains that can replace the tetra-antenna forms at the asparagine binding sites generally have no more than two or three attachments. O-linked oligosaccharide 25s generally provide two sites for sialic acid binding. Thus, erythropoietin molecules may be provided with a total of 14 sialic acid residues, provided that all three N-linked oligosaccharides are tetra-antennae. Mammalian cell cultures are screened for 30 cells which preferably add tetra-antennal chains to recombinant erythropoietin, thereby maximizing the number of sialic acid binding sites.

N-linked oligosaccharides of urinary erythropoietin contain sialic acid in both α-2,3 and α-2,6 bonds to 35 galactose (Takeuchi et al., J. Biol. Chem. 263, 3657-12106563 (1988)). Usually, sialic acid in the α-2,3 bond is linked to galactose in the α-1,6-branch of mannose and sialic acid in the α-2,6 bond is linked to galactose in the α-1,3 branch of mannose. The enzymes that increase these sialic acid-5 sons (β-galactoside α-2,3-sialyltransferase and β-galactoside α-2,6-sialyltransferase) are most effective in increasing sialic acid mannose α-1,6 and mannose α-1, respectively. , 3-branches.

Dihydrofolate reductase deficient (DHFR) Chinese-10 hamster ovary (CHO) cells are commonly used as host cells for the production of recombinant glycoproteins, including recombinant erythropoietin. These cells do not express the enzyme β-galactosidase α-2,6-sialyltransferase and therefore do not add sialic acid to the α-2,6-linkage to the N-linked oligosaccharides of the glycoproteins produced by these cells. (Mutsaers et al. Eur. J. Biochem. 156: 651 (1986); Takeuchi et al.

J. Chromatogr. 400, 207 (1987). Thus, recombinant erythropoietin produced in CHO cells lacks a 2,6-bond of sialic acid to galactose (Sasaki et al. (1987), direct; Takeuchi et al. (1987), direct). In another embodiment of the present invention, the erythropoietin used to produce isoforms is made in CHO cells transfected with a β-galactoside α-2,6-sialyltransferase functional gene to produce sialic acid in the α-2,6 bond to galactose. See Lee et al.

J. Biol. Chem. 264, 13848 (1989), incorporated herein by reference, for publication of techniques for the generation of modified CHO cells or other mammalian host cells.

Uses of the invention include pharmaceutical compositions comprising a therapeutically effective amount of a specific isoform or mixture of isoforms in combination with a suitable diluent, excipient and / or carrier for use in erythropoietin therapy. As used herein, "therapeutically effective amount" means the amount of 13,106,663 which provides a therapeutic effect in a particular condition and regimen of administration. Administration of the erythropoietin isoforms is preferably via parenteral routes. The specific route chosen depends on the condition being treated. Administration of erythro-5 polyethylene isoforms is preferably made as part of a formulation comprising a suitable carrier such as human serum albumin, a suitable diluent such as buffered saline, and / or a suitable excipient. The required dosage is in amounts suitable to increase the hematocrit of the patient and will vary depending on the severity of the condition being treated, the method of administration and the like.

The following examples are provided to illustrate the invention more fully, but are not to be construed as limiting the scope of the invention. The erythropoietin standard used in the in vivo bioassays described in the Examples is a recombinant erythropoietin standard which was standardized against a partially purified urinary erythropoietin standard. Thus, only relative in vivo specific activities are determined. In vivo specific activities are expressed in "units / ml", "units / mg" and "units / A280" (unit = U), not "IU / ml", "IU / mg" and "IU / A280" , because the erythropoietin standard used does not directly correlate with any existing international standard.

25 Reference Example 1

Isolation of recombinant erythropoietin isoforms

Recombinant erythropoietin is produced as described in Lin, supra. Recombinant erythropoietin used as starting material for the first and third isolates is purified according to Lain et al. patent, supra. according to the method described in Example 2. The starting material for the second and fifth isoform isolations is purified according to Lain et al. straight according to modification 35 by Q-Sepharose chromatography. These preparations contain a mixture of * 14 106563 isoforms of recombinant erythropoietin having the same sequence as urinary human erythropoietin and contain predominantly isoforms 9-14. The starting material for the fourth isoform preparation is 5 material eluted by Lain et al. anion exchange column of Example 2 with 5 mM acetic acid / L. mM glycine / 6 M urea wash. This fraction contains isoforms having 9 or less sialic acid and is further purified by gel filtration chromatography as described by Lain et al. Example 2 10 describes a pre-use isoelectric focusing process. In the sixth isoform preparation a purified recombinant erythropoietin preparation having 4 to 13 sialic acid residues was used as starting material. This material was purified according to Law et al. Example 2 except modification of the ion exchange column (elution of recombinant erythropoietin with sodium chloride gradient at pH 8.4 and absence of acetic acid / urea wash) which results in retention of most of the starting isoforms in the column.

Six different preparations of individual isoforms are obtained by isoelectric focusing in a granulated gel bed (Ultrodex, LKB), essentially as disclosed in LKB Application Note 198. Phar malite (Pharmacia) 2.5-5 ampholytes are used and si-25 gel filling contains 5M urea.

In the first preparation, about 20 mg of recombinant erythropoietin in 6.8 ml of 20 mM sodium citrate / 100 mM sodium chloride, pH 7.0, is applied to the gel and concentrated at 8 watts for about 16 hours. After isoelectric focusing, the isoform bands in the gel are visualized by: a paper contact copy of the gel layer. The copy is developed and fixed by soaking 3 times (about 10 minutes each time at room temperature) in fixative solution (40% methanol / 10% acetic acid / 10% TCA / 3.5% sulfosalicylic acid-35 po) which is changed once (about 10 minutes). ) • 1S 106563 in 40% methanol / 10% acetic acid solution (30-60 ° C), stained for 15 minutes at 60 ° C in 0.125% Coomassie-Blue R-250/40% methanol / 10% acetic acid solution, and rinsed with color 7, 5% methanol / 10% acetic acid to visualize discrete isoforms. The isoform-containing region of the gel gel layer (about 50% of the resin) is removed, water (about 16 ml) is added and the slurry is poured into a 14 x 62 cm container and evaporated to about 40 g net weight. This preparation is focused a second time and a contact copy of the gel layer 10 is made as before. The gel region containing each of the six distinct isoforms is removed from the gel layer.

To elute the isoforms from the gel, a solution containing 10 mM Tris-HCl, pH 7.0 / 5 mM Chaps is added to each of 15 isoforms to form a slurry. The slurries are applied to small columns and washed with Tris-Chaps buffer. The flow-through fractions are collected and applied separately to small columns (open column structure) containing Vydac C4 reverse phase resin equilibrated with 20% ethanol / 10 mM Tris-HCl, pH 7.0. The columns are developed stepwise with 20% ethanol / 10 mM Tris-HCl, pH 7.0, 35% ethanol / 10 mM Tris-HCl, pH 7.0 and 65% ethanol / 10 mM Tris-HCl, pH 7.0. Fraction eluting at 65%

ethanol / 10 mM Tris is diluted 1: 1 with 10 mM

25 Tris / HCl, pH 7.0 and concentrated, and then the buffer is changed to 10 mM Tris-HCl, pH 7.0 using a Centricon-10 (Amicon) microconcentrator. The analytical isoelectric focusing of this preparation is performed essentially as described in LKB Technical Specification 250, using Servalyte 3-5 ampholines (Serva) on a 5 M urea containing polyacrylamide gel.

In another formulation, about 26 mg of recombinant tierythropoietin in 6.5 mL of deionized water is added to the gel and focused at 2.5 watts for 35 minutes and 10 wa-35 accounts for about 17 hours. The bands of the focus protein, ie 106563 that appear in the gel layer, are removed in 11 separate pools. Each pool is made up to 7.5 mL with deionized water and 20 mL of each resulting pooled supernatant is subjected to isoelectric focusing as described above. To each pool is added 5 mL of 1.5 M Tris-HCl, pH 8.8, and each slurry is placed on a small column and the liquid phase is allowed to drain. The resin is washed with about three volumes of 0.5 M Tris-HCl, pH 7, and the rinsing solution is combined with the flow-through solution. The eluents are concentrated and the buffer exchanged with 20 mM sodium citrate / 100 mM sodium chloride, pH 7.0 using Amicon disposable ultrafiltration devices with a 10,000 Dalton molecular weight cutoff filter. The concentrated solutions (about 0.5 mL) are then passed through a 0.22 micron cutoff in a cellulose acetate filter. Based on analytical isoelectric focusing, the five pools are found to contain mainly single isoforms 10, 11, 12, 13 and 14.

In a third method of preparation, about 30 mg of recombinant 20 nant-erythropoietin in 21.8 ml of distilled water is placed in a gel and concentrated at 2 watts for 25 minutes, 10 watts for 20 hours and 15 watts for 15 minutes. Protein bands corresponding to individual isoforms are visually detected and removed from the gel layer. Distilled water is added to the gel isolated isoforms to provide a slurry and the resulting supernatants are analyzed by analytical isoelectric focusing. An equal volume of Tris-HCl, pH 7.2, is added to each slurry, the suspensions are applied to separate small columns, and the liquid phase is allowed to flow through the column to elute isoforms. Each effluent fraction is concentrated and the buffer exchanged with 20 mM sodium citrate / 100 mM sodium chloride, pH 7.0 using Amicon disposable ultrafiltration devices with a 10,000 Dalton molecular-weight cutoff filter. The analytical isoelectric • 106563 17 focusing gel showed that the resulting pools mainly contained single isoforms 9, 10, 11, 12, 13 and 14.

In the fourth isoform preparation, erythropoietin containing isoforms 3-9 (prepared as described above) was used as starting material. Prior to the ready isoelectric focusing performed essentially as described in Preparations 1-3 above, the ampholytes (Pharmalyte 2.5-5) were pre-fractionated in a Rotofor (Bio-Rad, Richmond, CA) liquid phase isoelectric focusing unit to provide starting material for lower a better ampholytic region for isoelectric points. Pre-fractionation was performed by mixing 6.7 ml of Pharmalyte 2.5-5 with 15 g of urea and adding purified water to bring the volume to 50 ml. This mixture was fractionated in a Rotofor at 10 watts at 1 ° C for 5 M using 0.1 M phosphoric acid as the anolyte and 0.1 M sodium hydroxide respectively as the catholyte. Ampholytic fractions having defined pHs between 4.5 and about 6 were used for flat-bed isoelectric focusing.

Amphytes were removed from the isoforms using a Centrifuge Luther (Amicon, Danvers, MA) and a 10,000 MW cutoff Centricon (Amicon) using the following parameters: 0.18 Tris buffer, pH 8.8, 100 volts, 25-30 mA, 25 hours . The isoforms were then subjected to buffer exchange with 0.1 M sodium chloride by gel filtration using Sepha-dex G-25 (Pharmacia). The analytical isoelectric focusing of the five obtained pools indicated that they contained isoforms 4, 5, 6, 7 and 8. Isoform 4 passed through several bands, indicating that it may have been somewhat cleaved.

«

The fifth isoform preparation method was modified by adding a pre-focusing step to a flat-bed isoelectric focusing procedure. In this modification, the protein was not added to the ampholytic / urea / gel mixture prior to electrophoresis, 18,106,663, but was added to the isoelectric focusing apparatus after forming the pH gradient of the gel layer. After 75 minutes of pre-focusing (1500 volts), the gel layer band from 2.25 to 4.25 cm from the cathode was removed, mixed with the erythropoietin solution and added back to the gel layer. After isoelectric focusing, isoforms 10, 11, 12, 13 and 14 were eluted from the gel layer and separated from the ampholytes by ultrafiltration using Centricon-10 (Amicon).

Pre-focusing modification was performed to make the ultraviolet absorbance characteristics of the isoform preparations more similar to those of the recombinant erythropoietin used as starting material. This improvement in spectral characteristics can be seen at the absorbance ratios of the isolated isoforms at 280 and 260 nm. The average absorbance ratio at 280 nm compared to 260 nm (A280 / A260) for isoforms (unfocused) of Preparations 2 and 3 is 1.36 ± 0.11, while the average ratio of A280 / A260 for Preparations 5 and 6 for iso-20 forms (pre-focused) is 1.68 ± 0.20. When isoform 14 is omitted, the average A280 / A260 ratios are 1.39 ± 0.11 and 1.74 + 0.09 for isoforms 2 and 3 and 5 and 6 respectively. (Isoform 14 may have the most atypical spectrum because it is present in very small amounts and is therefore more susceptible to interference by minor ampholytic component contamination or because it is mainly an electrode during the flatbed isoelectric focusing method). Mean A280 / A260 Ratio According to Lain et al. the recombinant erythropoietin prepared according to Example 2 (modified as previously described using Q-Sepharose as an anion exchange resin) has 1.91 ± 0.04.

As described above, the starting material for isoform preparation 6 was a recombinant erythropoietin preparation 35 comprising isoforms 4-13. The ampholytes were pre-focused at 19 106563

Rotofor as in the fourth manufacturing method. Ampholytic fractions having defined pHs between 3.7 and 4.8 were used for flat-bed isoelectric focusing. Flat bed was pre-focused as in Preparations 5 and 5, isoforms 9, 10, 11, 12 and 13 were obtained after ultrafiltration (Centricon-10) to remove carrier ampholytes.

Reference Example 2

Sialic acid content of recombinant erythropoietin isoforms

According to the methods described in Example 1 for the isoforms and for the erythropoietin purified by Lain et al., Supra (mixture of isoforms 9-14), the buffer is exchanged with 0.10-0.15 M sodium chloride and analyzed for the sialic acid content by Jourdian et al.

J. Biol. Chem. 246, 430 (1971). Sialic acid residues are digested from glycoproteins by hydrolysis with 0.35 M sulfuric acid at 80 ° C for 30 minutes and the solutions are neutralized with sodium hydroxide before analysis. To evaluate the amount of erythropoietin protein present, a Bradford protein assay (Bradford, Anal. Biochem. 72, 248 (1976)) is performed using recombinant erythropoietin having the amino acid sequence of human erythropoietin, assay reagents and the microarray method. -Rad delivers. The results, expressed in moles of sialic acid per mole of erythropoietin, are shown in Table 1. The isoforms are named according to the amount of sialic acids per molecule and the series is the least acidic (isoform 9) to the most acidic (isoform 13). The amounts of isoform 14 are insufficient to accurately measure the amount of sialic acid. The sialic acid content of this isoform has been deduced from its migration in IEF gels compared to other isoforms. The sialic acid content of isoforms 5-8 (Preparation 20 1 0 6 5 63 4) has not been determined, but it has also been deduced from their transport in IEF gels.

Table 1 5 moles of erythropoietin mole sialic acid / isoform moles erythropoietin isoform 13 12.9 ± 0.5 10 isoform 12 11.8 ± 0.2 isoform 11 11.0 ± 0.2 isoform 10 9.8 ± 0.3 isoform 9 8.9 + 0.6 isoform mixture 11.3 ± 0.2 15 (9 - 14)

Reference Example 3

Activity of the recombinant erythropoietin isoforms Isolated as described in Example 1 is analyzed for the amount of recombinant erythropoietin present by absorbance at 280 nm, Bradford protein assay and RIA for erythropoietin. Bioanalysis of exhypoxic polycythaemic mice (Cotes et al., Nature 191, 1065 (1961)) is used to determine relative biological activity in vivo. Quantification of the amount of erythropoietin protein present by radioimmunoassay for erythropoietin yielded results of greater relative specific activity in vivo for certain isoforms, due to the apparent reduced immunoreactivity of the high sialic acid -35 for the most viscous isoforms. Mice bioassay assays, 2i 106563 expressed in units / ml, have been subdivided into corresponding protein concentrations to obtain specific activities in vivo expressed in units / mg erythropoietin polypeptide. These specific activities are shown in Table 5 2.

In Table 2, "n" is the number of individual isoform preparations that contribute to the specific activity value. In most cases, many assays were performed in vivo for each isoform preparation. The same results in vivo affect the estimates of specific activity in all three columns, units / mg erythropoietin polypeptide were determined by absorbance at 280 nm, radioimmunoassays and Bradford protein assay results. 15 standard purified recombinant erythropoietin containing isoforms 9-14 were used in the Bradford protein assay. The "n" may be less in the calculations made using the Bradford protein assay because some preparations were no longer available when the Bradford assay was performed.

20 Lain et al. supra-purified erythropoietin containing isoforms 9 -14 was used as a standard in RIA assays and assays in vivo.

Relative specific activities expressed. 25 units / mg erythropoietin polypeptide can be converted to units / A280 by multiplying 0.807 mg erythropoietin polypeptide / A280. The conversion factor is derived by multiplying the extinction coefficient of erythropoietin (1.345 mg / A280) by the amount of erythropoietin glycoprotein protein (about 60% by weight, Davis et al. Biochemistry 26, 2633 (1987)) to obtain mg erythropoietin A2 ( it is 1.345 mg erythropoietin polypeptide / A280 x 0.60 mg polypeptide / mg erythropoietin = 0.807 mg polypeptide / A280). In addition, specific activities expressed per unit mg mg erythro-35 polyethylene polypeptide may be multiplied by a factor of 0.60 mg • * 22 106563 polypeptides per mg erythropoietin glycoprotein to obtain specific activities expressed per unit mg mg erythropoietin glycoprotein.

23 1 0 6 5 6 3 d cMiDLn «rnc \ i« -i <-it-i. o o o o o o o £ o o o o o o

J CTiCNLOr ~ LOLOOOO

"Π" X inocsitMT-taiin ^ rvo tr in sr * r lo m to ^ X ro m £ + l + l + l + l + l + l + l + l + l 73 o o o o o o X ooooooooo

M r'-Cjr '- ^ J'iJ'LOOOO

—- ^^ «. ^^^ OCOiO

&> <! 10 r ~ r ~ m m ro ^^

Sm vonoomr ^ MMtMin

^ p4 nncMcsiMMVo ^ rrH

D -—____ j CNILnLTl'CrOfMi — ΙΜιΉ o o o 'o o

Ό O O O O O O O

• Η Γ- ~ ιΛ [^ <ΟιΛΟΟΟΟ • P .. .. .. .. ^ r ~ o o o

Cij r ~ eri <—I r ~ - hn tn Q) ηιη'τΌΓονο ^ ^ di sr * T LO Μ> <+ Ι + Ι + Ι + Ι + Ι Ή + ι + ι + ι + ι ο Ο Ο ο ο ϋ ^ ooooooooo o cor ^^ rcomoooo ρ οο «. ν «. ,. . * ιο ιο n ro £ (Ν m 00 CO ιΟ Ο \ rf <οιοιουοη-ιοοοοο ρ '- (NjCsiCvICMi-tO ^ r-ini-i

(~ j CM * T CO i — I «—1 T — 1 f-H

w __! _ o

M

M

3 H -rl | o o o 3 PO— ^ ooooo «S -HMmOVDVDMO o o o H + J Q,>. H - - - OL o o o d, -p - O LO M * oooor-

(DC-HfOfOLnsrM

O .r4 Li fO lT) * H

> ιΌ: rö -H +1 +1 +1 +1 1

rH U: r0 ^ Λ ^ _ _ +1 +1 -H

OOpOOOOO

q.llj.5 O O O O O O O O O

^ 73 Ssrvoojr-o ooo

- - - - - (NCMVO

? 2. [1cnr-u -> <Noo CmmOOOr-COCO ROCK

N O {M {M H LO sr M

• M

| g srrOfs | (-HO <7'QOr ~ LO Q Li r —i rH «Ή» "H rH

en o h m 24 106563

The results of Table 2 are also shown graphically in Figures 6A, 6B and 6C. These results indicate that the relative activity of erythropoietin in vivo is increased as a function of sialic acid up to isoform # 11. Isoforms 11-14 have substantially the same relative bioactivity in vivo. (This is most evident when the concentration of isoform 14 is expressed using Bradford assay values. The Bradford value may be more accurate due to the generally small amounts obtained for isoform 14, and the consequent difficulty in determining A 4, and the apparent reduced reactivity of very negative forms as previously mentioned). The higher relative specific activity of the erythropoietin isoforms containing more sialic acid in vivo is evidently due to the longer half-life of these forms in the circulation. Isoforms 9 and 13 were labeled with radioactive iodine (125 I) and their clearance in rats was determined. The circulating half-life was clearly longer for isoform 13 than for isoform 9.

2 0 Reference example. 4

Selection of Recombinant Erythropoietin Isoform Mixtures by Q-Sepharose Chromatography Adjusted cellular media for recombinant erythropoietin production Lin et al., Direct. is concentrated and diafiltered against 10 mM Tris, pH 7.2. Protein concentration is determined by the Bradford Microprotein Assay using bovine serum albumin as a standard. 19.6 ml of a solution containing 40 mg of total protein is made up to 20 μΜ ^ βϊ Cu-30 SO4, filtered through a 0.45 micron cutoff filter and loaded with 4 ml of a layered volume (1.05 cm high x 2.2 cm diameter) ) column packed with Q-Sepharose Fast Flow (Pharmacia) equilibrated with 10 mM Tris pH 6.8-7.0 at 4 ° C. After the sample has been removed, column 35 is washed with two column volumes of the same buffer. The flow rate is approximately 1 ml / min. Six separate columns are prepared using this method to assay specific erythropoietin isoform mixtures.

The columns are washed with 6-9 column volumes of low pH 5 buffer containing: Column # 1, 150 mM acetic acid, 1 mM glycine, 20 μΜ CuSO4, 6 M urea adjusted to pH 4.7 with NaOH; Column # 2, 200 mM acetic acid, 1 mM glycine, 20 μΜ CuSO4, 6 M urea, adjusted to pH 4.7 with NaOH; column # 3, 250 mM

10 acetic acid, 1 mM glycine, 20 μΜ CuSO4, 6 M urea, adjusted to pH 4.7 with NaOH; Column # 4, 300 mM acetic acid, 1 mM glycine, 20 μΜ CuSO4, 6 M urea, adjusted to pH 4.7; column # 5, 150 mM acetic acid, 1 mM Glycine, 20 μΜ CuSO4, 6 M urea; Column # 6, 300 mM etiquic acid, 1 mM glycine, 20 μΜ CuSO4, 6 M urea. The pH of the columns is raised to about seven by washing each with 8-11 columns

10 mM Tris-HCl, 55 mM NaCl, 20 μΜ Cu-SO4, pH 7. Eluted erythropoietin isoform mixtures are eluted from the columns by washing with 10 mM Tris-HCl, 140 mM

20 NaCl, 20 μΜ CuSO4, pH 7.0.

The eluted isoform pools from each column are concentrated and the solvent is changed to water using an Amicon-Centricon-10 microconcentrator. The erythropoietin-nisoform mixture represents a cell culture medium which, on a dose-slide on a Q-Sepharose column as described above, is washed with 5 mM acetic acid, 1 mM glycine, 20 μΜ CuSO 4, 6 M urea, and eluted using the column above. . This eluted mixture of isoforms is further purified according to the methods described by Law et al., Supra, before analytical isoelectric focusing.

«106563

Reference Example 5

Fractionation of recombinant erythropoietin isoforms using a low pH gradient in Q-Sepharose

In another approach, the erythropoietin isoforms 5 are resolved using a decreasing pH gradient and increasing the ionic strength. The concentrated diafiltered erythropoietin-containing medium is loaded on a Q-Sepharose column at a ratio of about 40 mg total protein / ml gel. The column is then washed with about two column volumes of 10 mM Tris-HCl, pH 7.0 and then about 10 column volumes contaminated with 2 mM acetic acid / 1 mM glycine / 20 μΜ CuSO 4/6 M urea (pH 4.8). proteins and erythropoietin isoforms containing less than 7 sialic acid residues. Isoforms containing about 8-12 sialic acids are eluted from the column using a gradient starting from about 2 mM acetic acid in 6 M urea / l mM glycine / 20 μΜ in CuSO 4 and ending in 40 mM acetic acid / 6 M urea / l mM glycine / 20 μΜ CuSO4 (pH about 4). The total gradient volume is about 40 columns to 20 volumes, and fractions of about one column volume are each collected in wells containing Tris buffer in an amount sufficient to adjust the pH to 6-8.5 to avoid exposing the collected fractions to low pH. Samples of the fractions are subjected to 25 analytical isoelectric focusing to monitor separation. The isoforms 12-14, which remain bound to the column at the end of the gradient, are eluted by washing with buffer containing 10 mM Tris-HCl, 140 mM NaCl, 20 μS CuSO 4 (pH 7). Isoforms (separated during gradient or eluted with sodium chloride solution) are released from contaminating proteins by reverse phase chromatography followed by gel filtration chromatography such as Lain et al. as described in Example 2.

27 1 0 6 5 6 3

Example

Human Erythropoietin Analogs Containing Extra Glycosylation Sites A) Construction of Human Erythropoietin Analogs 5 The positions of the existing and putative carbohydrate attachment sites in the erythropoietin amino acid sequence are shown in Figure 1, and the method to create these additional glycosylation sites is shown in Figure 2.

The following oligonucleotide primers were synthesized for use in in vitro mutagenesis: [A3n4, Ser6] EPO: 5 'CGCCCACCAAACCTCAGCTGTGACAGCCGA 3 * [Asn *, Ser11) EPO: 5.' ATCTGTACAACCGAAGCCTGGAGAGGT 3 '[Asr> 69] EPO: 5' GGGCCTGGCCAACCTGTCGGAAG 3 '15 [Asnl24j EPO: 5' TCCCCTCCAGATAATGCCTCAGCTGC 3 '[Asnl25f Serl27] .EPOG: 5' C 'C' C 'C' AGGCCTGCAGGAATGGGAGCAGATGACCAGGTG 3 '[Thr * 25] EPO: 5' TCCAGATGCG & CCTCAGCTGCTC 3 '2Q [Prol24, Thrl25) Ep0: 5' CCTCCAGATCCGACCTCAGCTGC 3 '

The underlined codons indicate mismatched regions where the amino acids shown in parentheses replace wild type amino acids.

[Asn4, Ser6] EPO was constructed to insert the N-glycosylation site into position Asn 4. [Asn9, Ser11] EPO was constructed to insert N-glycosylation site into Asn 9. [Asn69] EPO

was constructed to insert an N-glycosylation site at Asn 69. [Asn125, Ser127] EPO was constructed to insert an N-glycosylation site at Asn125. [Thr125] EPO and [Pro124,300 Thr125] EPO were constructed to insert the O-glycosylation site into Thr125.

The following oligonucleotide primers are synthesized for use in in vitro mutagenesis: 2β 106563 [Asn69, Thr ^ l] EPO: 5 'GGGCCTGGCCM? Epo: 5 ′ CAGATGCGAACTCAACGGCTCCAC 3 ′ [Asnl25, Thrl27, Thrl31] EPO: 5′ATGCGAACTCAACGGCTCCACTCACAACAATCACT 3 ′ 10 [Prol24, Asnl25, Serl27] EPO: 5 ′ CCAGATCCAAATTCATCTGTccGCTCCACCATAC1AT 3 '[Thr125, Thr126] Epo: 5' OCAGATGCGACAACAGCTGCTCCA 3 '[Prol24, Thr125, Thr126, Thrl31] EPO: 20 From the [Pro124, Thr125] EPO cDNA, the oligonucleotide dialect 5'AGATCCGTACCACG is used to produce Thr125, Thr126] EPO. The oligonucleotide primer 51TGCTCCACTCACAACAATCACTG31 is then used to produce [Pro124, Thr125, Thr126, Thr131] EPO.

The [Asn69, Thr71] EPO and [Ser68, Asn69, Thr71] EPO constants are prepared to add the N-glycosylation site to Asn69 and to promote N-glycosylation at that position. [Asn125, Thr127] EPO, [Asn125, Thr127, Thr131] EPO, [Pro124, Asn125, Ser127] EPO and [Pro124, Asn125, Thr127] EPO are constructed to insert the N-glycosylation site at Asn125 and to add glycosylation at this site. [Thr125, Thr126] EPO and [Pro124, Thr125, Thr126, Ser131] EPO are constructed to insert the O-glycosylation site into Thr125 and to increase glycosylation at that position.

The source of erythropoietin DNA for in vitro mutagenesis was plasmid Hul3, a cDNA clone of human erythropoietin 29 106563 in pUC 8 (Law et al., Proc. Natl. Acad. Sci. 83, 6920 (1986)). Plasmid DNA obtained from Hul3 was digested with restriction enzymes BstEII and BglII, the resulting DNA fragments were subjected to agarose gel electrophoresis and the 8105 bp (ep) erythropoietin DNA fragment was isolated from the gel using the GeneClean ™ kit (BIO 101, Inc.). Plasmid pBRgHuEPO contains the erythropoietin-genomic gene as a BamH1 fragment incorporated into the pBR322 derivative as described in Lin, supra. pBRgHuEPO was also digested with BstEII and BglII and a 6517 bp vector fragment was obtained. Ligation of the two fragments produces IGT1. To construct pEC-1, pDSVL (described in Lin's patent, supra, and shown in Figure 5B) was digested with BamHI and an isolated BamHI fragment of 2.8 kb (kb) from IGT1 containing the erythropoietin-ethene cDNA.

To produce single-stranded DNA for in vitro mutagenesis, pEC-1 was digested with BamHI and BglII and the 820 bp erythropoietin cDNA fragment was isolated. It was ligated into the BamHI site of ml3mpl8 to obtain ml3-EC-1. Single-stranded DNA was obtained from supernatant of E. coli strain RZ1032 infected with ml3-EC-1, such as Kunkel et al., Methods in Enzymol. 154, 367 (1987) and Messing, Methods in Enzymol. 101, 20 (1983). For in vitro mutagenesis, approximately 1 µg of single-stranded DNA and 0.2 pmol of one of the synthetic primers described above was mixed with 6 ml of buffer (250 mM Tris, pH 7.8, 50 mM MgCl 2 and 50 mM dithiothreitol). . To incorporate the primer into the template, the reaction volume 30 was adjusted to 10 µl with water, the mixture was heated to 65 ° C for 5 minutes and then allowed to cool to room temperature. For the extension reaction, dTTP, dATP, dGTP, dCTP, and ATP were each added in 2.5 mL (10 µM each), followed by 1.1 µl (1 unit) of E. coli DNA. 35 polymerase (Klenow fragment) and 1 / mu / 1 (1 unit) 30 1 0 6 5 6 3 T4 DNA ligase. The mixture was then incubated overnight at 14 ° C and used to transform E. coli JM109 (Yanish-Perron et al., Gene 33, 103 (1985) 9) as described (Messing, supra).

To identify mutant clones by differential hybridization, media plaques were transferred to Gene Screen filters (New England Nuclear). The filters were dried under a heat lamp and incubated for one hour in 6x SSC containing 1% SDS at 60 ° C. For Hybri-10 disassembly, the above oligonucleotide primer (8 pmoles) was labeled with T4 polynucleotide kinase and γ32ρ-labeled ATP and incubated with filters overnight in 6x SSC, 0.5% SDS and 100 mg / ml salmon sperm DNA at 37 ° C for the [Asn124] mutation, 55 at 15 ° C for the [Asn4, Ser6] mutation, 65 ° C for the [Thr125] mutation and [Pro124, Thr125] mutation and at 70 for [Asn9 Ser11] and [Asn163 Ser165] mutations. The next day, the filters were washed three times with 6x SSC at room temperature and subjected to autoradiography. If necessary, the filters were then washed with 6x SSC at rising temperatures until little or no hybridization was observed in plaques containing the wild-type erythropoietin cDNA sequence. Clones that gave positive hybridization signals at these conditions were identified and re-transfected into JM109 to isolate the Pure Clone. Dideoxy chain termination sequence analysis indicated that mutations to Asparagine, Serine, Threonine and Proline were present.

30 Double-stranded ml3 EC-1 DNAs containing [Asn4,

Ser6], [Asn9, Ser11], [Asn69], [Asn124], [Asn125], [Ser127], [ASn163, Ser165], [Thr125] and [Pro124, Thr125] were obtained from cells transfected with JM109 by the cooking method. (Holmes et al., Anal. Biochem. 117, 193 (1981)).

35 DNAs were digested with BstEII and XhoII and 810 bp erythropoietin DNA fragments were isolated. pEC-1 was digested with BstEII followed by partial digestion with BglII and the 5 'ends of the resulting fragments dephosphorylated with alkaline phosphatase of bacterial origin in 10 mM Tris pH 8, 60 ° C for 60 minutes. The 7 kb vector fragment lacking the 810 kb BstEII-BglII fragment was isolated and ligated into the erythropoietin fragment Ila. The plasmids produced (named pEC-X, where X represents a specific mutation) contain human erythropoietin having modified amino acid residues at the tagged positions.

CDNA clones and analogs of the human erythropoietin sequence corresponding to [Asn4, Ser6] -, [Asn9, Ser11] -, [Asn69] -, [Asn124] -, [Asn125, Ser127] -, [Asn163, Ser165] -, 15 [ Thr125] and [Pro124, Thr125] erythropoietin cDNA clones were transferred to COS-1 cells (ATCC No. CRL-1650) by electroporation. COS-1 cells were harvested from semisolid plates, washed with medium (Dulbecco's modified basic nutrient medium containing 5% fetal calf serum and 1% 20 L-glutamine / penicillin / streptomycin (Irvine Scien Tific)) and resuspended in 4 x 10 6 cells / ml . One ml of cells was transferred to an electroporation cuvette (Bio-Rad) and electroporated with Bio-Rad Gene Pulser ™ at 25 µl Farad and 1600 volts 100 to 200 µg carrier DNA and 25 and 2 to 20 µg plasmid DNA encoding the erythropoietin analogue. in the presence of. Cells that had undergone electroporation were plated at 2 x 10 6 cells per 60 mm tissue culture dish in 5 ml medium. Two to four hours after plating, the medium was changed to 5 ml of fresh medium. Treated 30 media were harvested 5 days after electroporation.

• l> 1 32 106563 B) Determination of activity of erythropoietin analogs RIAs were performed according to Egrien et al. supra. The biological activity of the erythropoietin analogs in vivo was determined by bioassay using exhypoxic polycythaemic mice (Cotes et al., Direct).

Erythropoietin activity in vitro was determined by analysis based on erythroid colony formation as described by Iscove et al. J. Cell. Phy-10 sol. 83, 309-320 (1974), with modifications. Mononuclear cells of the human bone marrow were partially purified on a Ficoll-paque mattress and washed in Iscove's medium prior to plating to remove adherent cells. The medium contained 0.9 methylcellulose and did not contain 15 bovine serum albumin at all. Erythroid colonies are scored after 8 to 10 days of culture.

According to Part A, the erythropoietin analogs transfected and expressed in COS cells were analyzed by raw ROS for COS cell supernatants and analysis based on the formation of erythroid colonies. The activity of the human erythropoietin sequence in vitro is proportional to the RIA activity as determined by the above assays. Analogues [Asn69] EPO, [Asn125, Ser127] EPO, [Thr125] EPO and [Pro124, Thr125] EPO showed activity in vitro that is proportional to RIA activity and provided evidence of additional carbohydrate chains (as determined in section C). These analogs are further analyzed by transfecting the cDNA clone encoding the erythropoietin analog into CHO cells, purifying the erythropoietin analog 30, and determining the biological activity of the purified analog in vivo.

C) Western blot analysis

As described in Part A, supernatant -35 of COS cells transfected with cDNA of the erythropoietin analog containing 5-20 units was immunoprecipitated with rabbit anti-erythropoietin poly-clonal antibody for more than 33 10 6 5 6 3 nights. 20-80 / mu / l of a 1: 1 protein A-Sepharosea in phosphate buffered saline (PBS) was added to immunoprecipitation and incubated for 1 hour at room temperature. The samples were centrifuged, washed with PBS, and if necessary, the pellet was treated with N-glycanase to remove N-linked carbohydrate chains. Samples were analyzed by 15% SDS polyacrylamide gel electrophoresis, transferred to nitrocellulose 10 and subjected to Western analysis as described (Burnette et al., Anal. Biochem. 112, 195-203 (1981), Elliot et al., Gene 79, 167-180). (1989)) using a mixture of murine anti-erythropoietin monoclonal antibodies. One such antibody, 9G8A, is described in Elliot et al. (1989) Blood 74, Supp. 1, A. 1228.

Analysis of supernatants of [Asn69] EPO and [Asn125, Ser127] EPO cDNA transfected with cDNA showed an increase in protein size compared to the human erythropoietin sequence. This increased size indicates an excess of 20 N-linked carbohydrate chains (Figure 3). Treatment of supernatants of COS cells transfected with [Thr125] EPO and [Pro124, Thr125] EPO cDNA with N-glycanase showed increased protein size compared to human erythropoietin sequence. This increased size is indicative of 25 additional O-linked carbohydrate chains (Figure 4).

D) Isolation of the erythropoietin analogue isoforms

The erythropoietin analogue [Thr125] EPO was constructed as described in Part A. The 810 bp erythropoietin cDNA-30 fragment containing the [Thr125] mutation was isolated by digesting the plasmid containing [Thr125] mutation with pEC BstE-II and BglII and inserting the fragment into pDECA, a derivative of pDSa2. pDSa2 is generally described in U.S. Patent Application No. 501,904, which is incorporated herein by reference. pDECA was deduced from pDSa2 in the following steps: 34 106563 1) The HindIII site of pDSa2 was removed by digestion of pDSa2 DNA with HindIII, treatment of the HindIII binding ends with E. coli DNA polymerase (Klenow fragment) and dNTPs and re-ligation of the blunt end. vector.

The resulting plasmid was pDSa2AH.

2) pDSa2AH was digested with SalI and a synthetic oligonucleotide with the SV40 linker and the SalI linker attached to the 3 'end of the linker was inserted. The synthetic oligonucleotides had the following sequence: 10

5 'TCGAGGAACTGAAAAACCAGAAAGTTAACTGGTAAGTTTAGT

CTTTTTGTCTTTTATTTCAGGTCCCGGATCCGGTGGTGGTGCAAATCA 15 AAGAACTGCTCCTCAGTGGATGTTGCCTTTACTTCTAGGCCTGTACGG

AAGTGTTACTTCTGCTCTAAAAGCTGCTGCAACAAGCTGGTCGACC 3 '

The resulting plasmid was the pDSa2AH insert.

3) The pDSa2AH supplement was digested with SalI and blunt-ended by treatment of the sticky ends with T4 DNA polymerase and dNTPs. The 820 bp BamHI-BGIIII cDNA fragment of human erythropoietin was blunt-ended by the same method and ligated into the plasmid. The resulting plasmid was pDEC.

4) pDEC was digested with KpnI and PvuII and blunt-ended by treatment of the sticky ends with mung bean nuclease. The plasmid was ligated to remove the excised Kpn1-PvuII fragment to give plasmid pDECA. Plasmid pDECA containing [Thr125] erythropoietin cDNA was transfected into DHFR-deficient CHO cells.

770 ml of the treated medium for CHO cells was concentrated using a 10,000 dalton molecular weight cutoff membrane and diafiltered against 10 mM Tris-HCl, pH 8.6 to a final volume of 34 ml. A volume of 17 ml of concentrate 35 106563 was applied to a Q-Sepharose high-flow column (5 ml fill volume), equilibrated with the same buffer, and eluted with a linear gradient of 0-250 mM NaCl in 10 mM Tris-HCl, pH 8.6. The same amounts of column fractions, either untreated or digested with N-glycanase, were analyzed by SDS-PAGE or IEF and the pools (designated 2, 3 and 4) were made on the basis of the isoform and / or carbohydrate composition of the fractions. Each pool was added to a Vydac C4 column (214TPB 2030; 1 cm diameter; 1.8-2.5 mL 10 fill volume); 0.34 mL / min) and washed with two column volumes of 20% ethanol in 10 mM Tris-HCl, pH 7.0. The columns were eluted with linear gradients of 20-94% ethanol, 10 mM Tris, pH 7.0. Pools were made, dissolved in 10 mM Tris-HCl, pH 7.0, and added to Q-Sepharose-no-15 main flow columns. After washing with 10 mM Tris-HCl pH 7.0, the samples were eluted with 20 mM sodium citrate, 250 mM NaCl, pH 7.0. Purified [Thr125] pools were analyzed by IEF and are shown in Figure 5. The EPO analog has been analyzed for biological activity in vivo. as described above (Cotes et al., supra).

• «• ·

Claims (11)

36 1 0 6 5 6 3 A process for the preparation of a human erythropoietin analogue having an amino acid sequence different from amino acid sequence 1-165 or 1-166 of human erythropoietin shown in Figure 1, which adds at least one additional site for glycosylation of the polypeptides, wherein the additional linkage is -10 that a) the human erythropoietin amino acid sequence is subjected to one or more alterations to form an analog with at least one new glycosylation site, and b) the analog is expressed in a host cell where the carbohydrate chain is attached to a new glycosylation site. Process according to claim 1, characterized in that an analogue is prepared wherein the additional site is for an N-linked carbohydrate chain. A process according to claim 1, characterized in that an analogue is prepared wherein the additional site is for an O-linked carbohydrate chain. A method according to claim 1, characterized in that the amino acid sequence changes are additions, deletions or substitutions of amino acid groups. A method according to claim 2, characterized in that the additional site is a replacement at position 69 of the human erythropoietin amino acid sequence. The method of claim 3, characterized in that the additional site is a substitution at position 125 of the amino acid sequence of human erythropoietin. The method of claim 1, wherein the additional carbohydrate chain provides a site for attachment of sialic acid. 106563 A method according to any one of claims 1 to 7, characterized in that the analogue is produced by expression of an exogenous DNA sequence. The process of claim 1, wherein the analog is Asn69 EPO, Asn69, Thr71 EPO, Ser68, Asn69, Thr71 EPO, Thr125 EPO, or 10 Pro124, Thr125 EPO. A DNA sequence encoding a human erythropoietin analogue according to any one of claims 1-9. An eukaryotic host cell transfected with the DNA sequence of claim 10, wherein the host cell expresses a human erythropoietin analogue. »*«. 38 106563