CZ285023B6 - Vector comprising gene encoding gm-csf protein of primaries, host cell being transformed thereby and process for preparing the primary gm-csf protein by making use thereof - Google Patents

Abstract

Recombinant colony stimulating factor (CSF) protein is claimed. It is prepd by isolating CSF protein as expressed from eukaryotic or prokaryotic host cells into which has been transformed a vector coding for the CSF protein. Also claimed is a method for prepg and isolating a transformation vector contg. CSF/CDNa by (a) prepg. RNA from a cell that produces CSF, (b) prepg. polyadenylated mRNA from this, (c) prepg single stranded CDNA from mRNA, (d) converting the single stranded CDNA to double stranded cDNA, (e) inserting the double stranded cDNA into transformation vectors and transforming bacteria with the vector to form colonies (f) picking pools of 200-500 colonies and isolating plasmid DNA from each pool, (g) transfecting the plasmid DNA into suitable host cells for expressing CSF protein, (h) culturing the transfected cells and assaying the supernatant for CSF activity and selecting CSF positive pools and screening the colonies used to make the pool to identify a colony having CSF activity.

Description

(57)

Described is a vector comprising a gene encoding a primate GM-CSF protein having the amino acid sequence shown in Figure 1 after the arrow for CSF-Thr, CSF-Ile or CSF-G, or allelic or other functional equivalents thereof, retaining sl activity GMCSF in a human bone marrow assay in which at least one amino acid is added, replaced or removed. Further described is a host cell that is transformed with this vector, and a method for producing the above protein using the same, as well as a cDNA corresponding to the above gene.

A vector comprising a gene encoding a primate GM-CSF protein, a host cell transformed therewith, and a method of producing a primate GM-CSF protein using it

Technical field

The invention relates to a vector comprising a gene encoding a primate GM-CSF protein, a host cell transformed therewith, and a method of producing a primate GM-CSF protein using it, as well as a cDNA corresponding to the above gene.

BACKGROUND OF THE INVENTION

The different types of cells found in the blood can all be derived from the underlying cells, which can further differentiate. These basic cells have two functions: (1) they reproduce themselves to maintain a population of basic cells in the body, and (2) they produce additional cells that differentiate into different mature types of blood cells. A cell that differentiates in a certain way is called a progenitor. The progenitor for T lymphocytes, B lymphocytes, granulocytes, red blood cells, platelets and eosinifile cells, as well as immature progenitors that can give rise to different types of mature cells, have been experimentally studied both in vivo and in vitro, as described, for example, in Dexter. , TM 1983 J. Pathology 141,415-433. In vitro, it has been shown that the proliferation and / or differentiation of each type of progenitor depends on specific "factors" that can be derived from various sources. For example, red blood cell progenitors require the presence of a factor called erythropoietin. The factors needed for the survival, proliferation and differentiation of myeloid progenitors for the production of mature neutrophil granulcytes, monocytes and mature macrophages are called colony-forming factors (CSFs).

The efficacy of these factors has been studied extensively in mice. Most adult mouse organs produce a substance with CSF activity. Compounds having some activity of this type have been obtained from different tissues and in different ways, but these substances differ somewhat in their biochemical properties. For this reason, the structural relationship of the various factors remains unknown. In addition, CSF appears to be active in more than one stage of granulocyte and macrophage development, and again it is unknown whether a single factor is responsible for all types of efficacy or whether a different factor is required at each stage as described in more detail in Burgess, A. and Metcalf, D. 1980 Blood 56 947-957.

A substance with human CSF activity was obtained from the placenta, some fetal tissues, macrophages and stimulated T-cells. The T-cell line (Mo), which produces one or more CSF-active substances, was obtained from a patient with leukemia in which certain variants of T cells (Leukemia reticuloendothelliosis) were multiplied as described in Golde et al. 1978 Blood 52 1068-1072.

The ability of CSF to stimulate granulocytes and macrophages means that pharmaceutical compositions with this efficacy could be used clinically in those situations if it was necessary to increase the production of myeloid-like cells. Some patients with very high levels of normal granulocytes actually had tumors that produced too much CSF. In one case, after surgical removal of the tumor, the granulocyte count rapidly dropped to normal, suggesting that the number of circulating granulocytes could be controlled in this way. This view was also expressed in Hocking, W., Goodman, J., and Golde D, Blood 61,600 (1983). The CSF-containing pharmaceutical compositions can be used clinically in the treatment of pulmonary suppression caused by irradiation of tumors or in the chemical treatment of these tumors. In addition, pharmaceutical compositions containing CSF could be useful in

Treatment of severe infections, because CSF may increase the number of granulocytes and / or monocytes and / or activate these elements.

There are various types of known CSF activity, including granulocyte activity (G-CSF), activity on macrophages (M-CSF), activity on granulocytes and macrophages (GM-CSF), and activity on a larger number of these cells. The invention is particularly concerned with the production of GM-CSF. Proteins of this type have already been isolated from various animal sources, but the invention relates in particular to the production of CSF in mammals, particularly humans and monkeys.

Biological and biochemical studies of CSF-active substances have hitherto been hampered by the fact that, on the one hand, very little was available, particularly for human and / or other primate CSFs, and on the other hand, they were impure. Obviously, it would be desirable to accurately identify the protein or proteins that are responsible for CSF activity. Further, it would be desirable to obtain a stable source of CSF, preferably primates, and especially humans, so that these proteins can be readily obtained in larger quantities and with a purity sufficient to determine the biological and biochemical properties of these substances, to be used to treat.

Recently, a molecular cloning technique has been developed which allows cloning of the nucleotide sequence coding for proteins and thus obtaining a larger quantity of protein with appropriate choice of host and vector, as described in Maniatis, T. Molecular Cloning A Laboratory Manual Cold Spring Harbor Laboratory, Cold Spring Harbor, NY 1982. The protein may be isolated and purified by conventional means. The cloning methods used hitherto can generally be divided into three groups: (1) methods based on the knowledge of protein structure, for example amino acid sequence knowledge, (2) methods based on the identification of the protein expressed by expression using a cloned gene using an antibody specific for that protein; and (3) methods based on identifying RNA which can then be translated to obtain a protein encoded by said gene.

All of these methods are very difficult if a protein such as CSF is available in very small quantities. This means that if it is difficult to obtain a sufficient amount of purified protein, it is also difficult to determine the amino acid sequence of the protein or at least a portion of the sequence. Preferably, protein identification after expression by binding to an antibody is performed using a high titer monospecific polyclonal antiserum. However, this antiserum also cannot be obtained without a large amount of pure protein to be used as an antigen. The monoclonal antibody provides an alternative approach, but the desired antibody is also very difficult to obtain without a suitable antigen and, moreover, the monoclonal antibody may not react with the protein in the form in which the protein is expressed under the available host and vector systems. In addition, in order to translate RNA to obtain an identifiable protein, the RNA of interest is required to be present in the source of the RNA in sufficient quantity to separate it. The relative abundance of RNA that encodes for a particular protein usually goes hand in hand with sufficient protein, so it can be said that a protein occurring in a small amount is usually encoded by an mRNA that also occurs in a small amount.

The Mo cell line was used as a starting material for purification of human CSF as well as for identifying the corresponding mRNA types. However, even using this relatively good source of CSF, it is extremely difficult to isolate enough protein for structural studies.

SUMMARY OF THE INVENTION

The invention overcomes the problems of the prior art by providing a good protein source with CSF activity, using genetic engineering. For cloning cDNA encoding a protein with CSF activity, a new cloning method can be used, for which only a CSF activity assay is required. Suitable methods for measuring CSF are described below in Example 2.

-2 GB 285023 B6

Accordingly, the invention provides a vector comprising a gene encoding a primate GM-CSF protein having the amino acid sequence shown in Figure 1 after the arrow for CSF-Thr, CSF-Ile or CSF-G, or allelic or other functional equivalents thereof, retaining GM-CSF activity in a human bone marrow assay in which at least one amino acid is added, replaced or removed. The invention also provides a cDNA encoding a protein having CSF activity (i.e., CSF / cDNA), a microorganism or a cell line (host cell) transformed with a recombinant vector containing such CSF / cDNA, and a method of producing a GM-SCF primate protein by CSF / cDNA expression. wherein the above microorganism or cell line is cultured. Since this protein is produced by a cell of the invention, it is certain that it is a protein exhibiting CSF activity. A transformation vector containing CSF / cDNA can be prepared and isolated as follows:

prepare RNA from a cell that produces CSF;

polyadenyl mRNA is prepared from said RNA;

producing single stranded cDNA from said mRNA;

said single-stranded cDNA is introduced into a double-stranded cDNA;

the double-stranded cDNA is inserted into a transformation vector, and this vector transforms the bacteria to form colonies;

200-500 colonies were pooled and plasmid DNA was isolated from each colony group; plasmid DNA is transfected with appropriate host cells to express the CSF protein; the transfected cells are cultured and the supernatant is tested for CSF activity; and CSF positive samples are selected and colonies used to generate positive samples are examined to determine which colony exhibits CSF activity.

These CSF proteins are growth and differentiation hormones for myeloid system cells. They are intended, for example, for clinical use in the treatment of myolesuppression, particularly in the case of (symptomatic) granulocytopenia after chemical treatment of tumors or treatment of tumors by radiation.

Description of the drawings

Figure 1 shows the DNA sequence coding for the CSF protein. The complete DNA sequence encodes human CSF. The changes listed for the human sequence are differences that distinguish this sequence from the Gibbon monkey sequence. Also depicted is the amino acid sequence produced by the use of said DNA sequence.

Figure 2 schematically illustrates a method for producing plasmid pTPL from plasmid pAdD26SVpA (3).

Giant. 3 is a schematic continuation of FIG. 2 and illustrates a method for producing plasmids p91023 from plasmid pTPL.

Giant. 4 is a schematic continuation of FIG. 3 and shows plasmid p91023 (B).

The following is an explanation of some terms to facilitate understanding of the method of the invention. If the scope of the definition differs from the conventional scope in the art, this will be discussed below.

Amplification or multiplication means a process by which a cell produces a gene repeat with its own chromosomal DNA.

CSF is a substance with biological activity, which is further defined by methods of determination.

-3 CZ 285023 B6

The CSF protein is a primate protein that has CSF activity. For purposes of the invention, the term CSF as a protein is also used for modified CSF proteins, allelic variations of the CSF protein, and for this protein even when the rest of the MET is preceded by its chain.

Downstream means to the 3'-end of the nucleotide sequence.

A sequence supporting a reaction is a nucleotide sequence that can cause potentiation of the transcription of a gene independently of its position relative to that gene or of the orientation of the sequence.

A gene is a sequence of deoxyribonucleotides that encodes a given protein. For the purposes of the invention, the gene should not contain regions that are unnecessary for translation, for example, signal sequences, polyadenyl chains, promoters or support sequences.

Binding is a process in which a phosphodiester bond is formed between the 5 'end and the 3' end of two DNA strands. Binding can be accomplished by several known methods using enzymes including T4 ligase.

Orientation is the order of nucleotides in a DNA sequence. The inverted orientation of a DNA sequence is one in which the sequence of the 5'- and 3'-end of one sequence is reversed relative to that order in the other sequence. This storage is important for transcribing another DNA sequence in the DNA source, or for replicating vectors containing the sequence.

Transcription is the synthesis of RNA from the original DNA.

Transformation means altering the genotype of a cell by receiving the foreign DNA. In some cases, transformation can be demonstrated by altering the phenotype of the cell. Transformed cells are called transformants, cells before transformation are parent cells.

Translation is the synthesis of polypeptides from mRNA.

The efficacy of CSF can be derived from a wide variety of cellular sources, including peripheral blood mononuclear cells, lung and placental tissues, bone marrow, the urine of anemia patients, blood serum and normal and neoplastic T cell lines. and mononuclear phagocytes. One cell line that produces CSF is the Mo line. The CSF produced by this cell line is known as CSF for granulocytes and macrophages (GM-CSF), which is human CSF. One source of Gibbon monkey CSF is the T-cell line, which is designated UCDMLA-144. This cell line was deposited with the ATCC and can be obtained under deposit number.

In order to isolate the CSF clone according to the invention, it was necessary to use a new method which in particular requires the determination of CSF activity. First, it is necessary to identify a cell that produces CSF, such as T cells or other cells, as mentioned above. The mRNA of this cell is then isolated. Preferably, T lymphocytes are used. In this case, the membrane-bound mRNA containing the lymphokine m-RNA is separated from the free mRNA in the cells. This compartment enriches the isolated mRNA 5 to 10 times with the sequence required for lymphokine production, thereby reducing the effort involved in identifying the respective clone. Polyadenyl mRNA oligo dT cellulose chromatography is then prepared.

A cDNA group is prepared from the mRNA using a vector suitable for transfer to a host to express the desired protein with CSF activity. Then, cDNA was prepared using standard methods from the above mRNA. The hybrid RNA / cDNA is then converted to double stranded cDNA and the material can then be incorporated into a suitable vector.

-4 CZ 285023 B6

A suitable host and vector system for isolating a CSF clone is based on expression of the cDNA in a suitable vector for transformation. A suitable transformation vector may be a vector obtained by transiently incorporating DNA into mammalian cells according to the method of Mellon, P., Parker, Y. Gluzman, T. Maniatis 1981 Cell 27 279-288. In order to isolate the desired CSF transformants, it is not necessary that all population cells stably contain exogenous genes that can be used to express the desired CSF as a product. It is possible to temporarily incorporate exogenous genes into a subpopulation of cells so that the subpopulation produces the desired protein for at least several days. Since there is no need to use additional substances for labeling according to the invention, it may happen that exogenous DAN is lost in cell growth within one to two weeks. However, two to three days after transfer to suitable mammalian cells, the products can be detected and detected.

A host and vector system suitable for use herein is a monkey cell line CV-1 transformed with a defective SV40 DNA molecule as described in Gluzman, Y., Cell 23 175-182, 1981. Transformed monkey CV- 1 cells. 1, which contain a defective SV-40 DNA, are termed COS (CV-1, defective, SV40), these cells do not contain a full copy of the SV-40 genome, but produce a high level of large T antigen and SV40 DNA replicates. These cells also promote the replication of SV40, which is absent at the beginning of a portion, and of the bacterial plasmid containing SV40 according to Myers, R.M. and Tjian, R. 1980 PNA 77 6491-6495. Thus, this system provides a means for amplifying transferred exogenous DNA via SV40 to increase mRNA and protein levels expressed in the presence of exogenous DNA. However, similar systems may be used.

Vectors used to express CSF typically contain additional helper moieties such as transcription enhancers, promoters, iontrons, polyadenylated sites, 3'-non-coding regions, and translation activators, as will be explained in more detail below.

Vectors may contain transcription enhancers. Transcriptional enhancers differ from the promoter but have a similar effect with it. Their function at the cellular level is not yet well known, but their unique property is the ability to activate or potentiate transcription regardless of their position or orientation. Promoters must be upside down, whereas transcriptional enhancers may be upside down from the 5'-end of the promoter, within the gene as an intron, or downstream between the gene and the polyadenylated site or the 3 'end of the polyadenylated site. Reverse sequence promoters are not functional, but inverse transcription enhancers are effective. Transcriptional enhancers only affect promoters if they are present on the same DNA strand. Their role is explained in more detail in Khoury et al., Cell 33, 313-314 (1983).

Preferred transcriptional enhancers for use in mammalian cells can be obtained from animal viruses, for example, simian virus 40, polyoma virus, bovine papilloma virus, retrovirus or adenovirus. Ideally, the transcription enhancer is derived from a virus for which a host cell is permeable, i.e., a virus that normally infects cells of that host. The viral transcription enhancer is readily obtainable from viruses commonly available from collections. Transcriptional enhancer regions from some viruses, such as Rous sarcoma virus and simian virus 40, are well known, as described in See Luciw et al., Cell 33, 705-716 (1983). It is a routine molecular biology procedure to remove these regions according to known restriction enzyme treatment maps for individual viruses and, if necessary, modify these regions to incorporate the transcription enhancer into the vector as desired. Such procedures are described, for example, in Kaufman et al., J. Mol. Biol., 159, 601-621 (1982) and Mol. Cell Biol. 2 (11), 1304-1319 (1982). It is also possible to synthesize the amplifier from data known in the sequence. Virus amplifiers usually contain less than 150 base pairs and are small enough to be synthesized.

-5GB 285023 B6

Another part of the vector should be a polyadenylated sequence. This is a DNA sequence that is located downstream of the transferred region of the gene, usually just below the transcription stop site. Typically, adenine ribunucleotides are added to produce a polyadanir nucleotide terminus at the 3'-end of the mRNA. Polyadenylation is important for stabilizing mRNA against degradation in a cell, which can reduce the amount of mRNA and thus the amount of the resulting protein.

Eukaryotic polyadenyl termini are well known. The 5'-AAUAAA-3 'hexanucleotide is located 11 to 30 nucleotides from where the polyadenylation is beginning. DNA sequences containing a polyadenylated sequence can be obtained from viruses according to published reports. The polyadenylated sequence can be obtained, for example, from mouse beta-globin and simian virus 40, but polyadenylated sites isolated from viruses are more preferred. Since these sequences are known, they can be synthesized in vitro and coupled to vectors in a conventional manner.

Preferably, the sequence that separates the polyadenylated sequence from the translation stop codon is an untranslated DNA sequence, for example, a promoterless eukaroytic gene. Since the promoter is not present, the sequence is not expressed. This sequence should be sufficiently long, of the order of up to about 1000 bases from the translation termination codon to the polyadenylated sequence. This 3 'untranslated sequence usually results in an increase in the yield of the final product. The vector may end up to about 30 base pairs downstream of the polyadenylated sequence, but it is preferable to retain the 3'-sequences downstream of the polyadenylated sequence in their original form. These are sequences that typically have 200 to 600 base pairs downstream of the polyadenylated sequence.

The presence of introns in the untranslated portion of the vector may also increase the yields of the final product. These introns can also be obtained from sources other than the host cell or gene sources. For example, a hybrid intron can be used which is isolated from an adenovirus or immunoglobulin gene inserted downstream of an adenylovir transcriptional start site to increase the yield of the final product.

In a preferred embodiment of the method for cloning CSF and for constructing an expression vector, an activation gene for translation can be used. Translation activators are genes that encode for a protein or RNA that affect the translation of the desired mRNA. The best example is the adenovirus-associated gene (VAVAI), which transcribes and interacts with the sequence in the 5'-untranslated region of the adenovirus mRNA, as described in Thimmappaya et al., 1982 Cell 3,543. the activation of translation by VA RNA lies in the adenovirus mRNA leader region. This sequence originates from the genome of the adenovirus and is located near the 5 'end after transcription. VA RNA can activate translation of mRNA. Thus, cDNA cloning and expression vectors preferably contain this form of genes from the adenovirus VA gene.

These vectors can be synthesized in a known manner. Vector components such as translation enhancers, promoters and the like can be obtained from natural sources or synthesized as described above. In principle, if these components can be found in large quantities of DNA, for example as a component of viruses, or if these components can be readily synthesized, these components can be readily obtained by using restriction enzymes and thereby obtain large amounts of vectors are grown by growing the organism which is the source of the DNA, the DNA is digested with the appropriate endonuclease, the fragments are separated, the fragment containing the sequence of interest is identified and isolated. Typically, the transformation vector is obtained in small quantities and then ligated to a suitable synthesis vector that undergoes autonomous replication, such as a prokaryotic plasmid or phage. In most cases, plasmid pB322 can be used, as described in Kaufman et al., Supra.

Synthesis vectors are used to clone vectors for transformation in a conventional manner, for example by transfer to a prokaryotic organism, by replicating the synthesis vector to yield

A high number of copies followed by isolation of the vector by cell digestion and isolation of the vector from cell debris.

Vectors containing cDNA prepared from cells producing CSF activity are then transferred to E. coli and grown on plates in Petri dishes at approximately 2000 colonies per dish. The colonies are then transferred to a nitrocellulose filter and the filter is transferred to a new plate which is stored as a sample. After colonies are grown, these colonies are propagated and compared to the original colonies so that the filter sections can be identified with the corresponding portion of the plate used as the sample.

Each filter is cut into sections containing predetermined numbers of colonies, preferably about 200 to 500 colonies per section. Colonies from each region are transferred to a culture medium, for example L-broth, the bacteria are separated by centrifugation and plasmid DNA is isolated. Plasmid DNA from each section of filter paper is transferred to a suitable host to achieve protein expression. A preferred vector for synthesis is a mutant of plasmid pBR322 and an E. salt in which sequences that damage eukaryotic cells have been omitted, as described in Kaufman et al., Supra. Using this mutant, there is no need to destroy the remainder of the plasmid prior to transfection. After culturing the cells after transfection, the medium is examined for the presence of substances with CSF activity. A positive result indicates that there is a colony containing CSF / cDNA in the filter section under investigation.

In order to determine which of the CSF / cDNA content filter clones, it is necessary to grow each clone from the filter segment. The cultures are then embedded in a matrix and a mixture of each horizontal row and each vertical column of the matrix is prepared. DNA formulas are prepared from these mixtures and transfected into host cells for expression. The supernatants obtained in this manner are re-examined for CSF activity. Typically, one vertical column mixture and one horizontal row mixture have this efficiency. The common clone then contains CSF / cDNA. If the matrix contains more than one positive clone, more than one column and more than one row are positive. In this case, a small number of clones in which the presence of CSF / cDNA would be considered should be further screened.

The CSF / cDNA is isolated from the clone by restriction enzymes and the sequence can be detected in a known manner. It is understood that the described procedure can be used to obtain CSF / cDNA from any source. The complete sequence is shown in Figure 1 together with the predicted amino acid sequence of the translation product.

The DNA sequence coding for a protein with CSF activity, as shown in FIG. 1, can be modified in a conventional manner to yield an altered resulting CSF protein that still has the desired assay performance. For example, one, two, three, four or five amino acids may be replaced by other amino acids. Belgian Patent Specification No. 898,016 describes a typical method of replacing cysteine with, for example, serine.

The CSF / cDNA of the invention comprises a full length gene preceded by an ATG codon and a CSF / cDNA that encodes allelic variations of the CSF protein. One allele is shown in Figure 1. Another allele found in the present invention has a thymidine residue at position 365 instead of the cytosine residue shown in Figure 1. The CSF protein obtained according to the invention also includes 1-methionine derivative (Met-CSF) and allelic variations of the CSF protein. The complete CSF protein, as depicted in the sequence in Figure 1, starts with the sequence Ala-Pro-Ala-Arg ..., which is indicated by an arrow at nucleotide 66 at Figure 1. In the case of Met-CSF, this begins The allelic variation shown in Fig. 1 contains a Thr at position 100 (the residue starts at Ala after the arrow) and can be termed CSF (Thr). Another variation has an Ile residue at position 100 and is more easily termed CSF (Ile). For the sake of completeness, the abbreviation CSF-G used herein refers to the GM-CSF Gibbon.

-7EN 285023 B6

The purified CSF protein has a specific activity (activity) of at least 10 ⁷ units per mg protein, and preferably at least 4x10 ⁷ units per mg protein, in performing an activity assay on human bone marrow cells.

Host-vector CSF expression systems may be prokaryotic or eukaryotic, but due to the complexity of CSF, the mammalian expression system is more preferred. Expression is readily achieved by transforming prokaryotic or eukaryotic cells with a suitable vector for CSF. The DNA sequence obtained as described above can be expressed directly in mammalian cells under the control of a suitable promoter. Heterologous promoters well known in the art can be used. In order to achieve CSF expression in prokaryotic or yeast cells, a leader sequence (secretion sequence) must be removed. The codon position for the N-terminus of the full length CSF protein is shown in Figure 1. Once the desired CSF / cDNA clone is obtained, appropriate known means can be used to achieve expression of the CSF protein, for example by incorporation into the appropriate vector. selecting and transforming the cells and growing the transformed cells to achieve CSF expression. The CSF protein thus obtained may have a methionine residue at the N-terminus of the protein (i.e. Met-CSF). Full mature proteins produced by prokaryotic and eukaryotic cells will optionally have the same amino acid sequence, but eukaryotic cells may produce a product that will be somewhat differently glycosylated than the original natural product.

The CSF protein expressed in appropriate prokaryotic or eukaryotic cells can be isolated and purified by methods known in the art. Furthermore, a new purification method is described which makes it possible to obtain CSF protein from recombinant and natural sources in high purity and activity.

This method eliminates the problems previously associated with protein isolation and purification with CSF activity. The CSF protein obtained by this method has a specific activity of at least 1x10 ⁷ units per mg protein, preferably at least 2x10 ⁷ units per mg protein, especially 4x10 ⁷ units per mg protein, when tested on human bone marrow cells.

Working in this manner, the CSF protein is purified as follows: the protein is precipitated with ammonium sulfate added to saturation to 80% to give a pellet of CSF protein, which is resuspended in a buffer solution at pH 6-8 and applied to the top chromatography column, washing the column with sodium chloride buffer and collecting fractions with CSF activity. The active fractions were pooled, applied to the reverse phase C4 column, and the column was removed sequentially with increasing amounts of acetonitrile at a concentration of 0-90% to obtain the active fraction.

Description of cleaning in connection with the illustration

Figure 5 shows the analysis of purified CSF protein (SDS-PAGE).

The CSF protein to be purified as described above can be obtained from any of the natural sources described above as starting sources for genetic engineering. These are a variety of cellular sources including conditioned media, such as the Mo cell line or the USD MLA-144 gibbon cell line.

The CSF protein can also be produced using genetic engineering as described above.

Thus, the CSF protein can be purified from any source as described above. Preferably, the medium is concentrated by ultrafiltration to a protein concentration of at least 0.1 mg protein / ml. The protein is then precipitated by adding ammonium sulfate to a saturation of 80%. The resulting pellet is resuspended in an aqueous buffer solution of pH 6-8. Examples of suitable buffers include Tris-HCl, HEPES, sodium citrate, and the like.

-8 GB 285023 B6

The buffer solution is subjected to column chromatography. Suitable materials for use in this column are, for example, octylsepharose, DEAE-ultrogen, AcA44-ultragel, AcA-54 ultrogel, and the like. One or more of these materials in succession may be used to obtain greater purity of the resulting material.

Fractions were collected from each column and examined for CSF activity. The active fractions were combined and diluted with trifluoroacetic acid (TFA), heptafluorobutyric acid (HFBA), and the like, and the material was applied to the C4 column in reverse phase. The compound with CSF activity is then eluted using increasing amounts of acetonitrile from 0 to 90% in TFA or HFBA, preferably at a concentration of 0.10% to 0.15% by volume, depending on the acid used.

The fractions with CSF activity were analyzed on an SDS polyacrylamide gel by electrophoresis, using a 13.5% gel as described in Lammli, U. Nature 227, 680 (1970). Still other methods can be used to achieve a more homogeneous CSF protein. Purified CSF protein after fractionation by SDS-polyacrylamide gel electrophoresis yielded a heterogeneous CSF protein having a molecular weight in the range of 15,000 to 26,000 units. This apparent heterogeneity is due to the high glycosylation of the protein and is a common property of the glycoprotein. Fractionation of a less pure sample from Mo cells in the same way, protein analysis showed the presence of another protein with CSF activity, having a molecular weight of 28,000 to 30,000 units.

Substances with CSF activity bind and re-release from actylsepharose, DEAE ultrogel and reverse phase C4 column. Approximately 60% CSF activity binds to Noc-A sepharose (40% passes un captured) and can then be eluted with methyl mannoside.

Molecular weight analysis by recombinant CSF gel filtration in a low salt environment demonstrates that approximately 30% of the activity of a 19,000 molecular weight compound is homogeneous, but 70% of the material behaves as a dimer, eluting at a position that corresponds to the molecular When sodium chloride is added to the column at a concentration of 1M, all potency is eluted at a broad peak corresponding to a compound having a molecular weight of about 19,000 units.

The purified CSF protein is stable for at least 16 hours at 4 ° C and pH 7.4 in 4M guanidine hydrochloride, 10 mM EDTA, 10 mM 2-mercaproethanol and 30% (v / v) ethanol. The substance with CSF activity is also stable in 0.1% trifluoroacetic acid at pH 2.0 and 0.1% trifluoroacetic acid with 25% acetonitrile by volume.

As already mentioned, the CSF protein obtained by the method of the invention is intended for use in the treatment of marrow suppression, for example in (sympromatic) granulocytophenia, which may be caused by treatment of tumors chemically or by radiation. In addition, the CSF proteins produced by the method of the invention can be used to treat severe infections. For this use a dose of 200 to 1000 micrograms per day is used. Preferably, the CSF protein is administered intravenously together with a suitable pharmacological carrier. Examples of suitable carriers include sodium chloride solution and human serum albumin in sodium chloride solution.

In addition, the CSF protein produced by the process of the invention has other effects and other uses. For example, CSF in mice has been shown to activate neutrophils. It would be expected that the primate CSF obtained by the method of the invention would also activate neutrophils. Thus, the physiological functions of this protein are multifaceted. In the bone marrow, this substance can promote the development and differentiation of cells that provide defense, and activate new and existing cells in the periphery. In a local immunological response, CSF may cause the accumulation or absence of neutrophils

-9EN 285023 B6 in inflammation of the area. Inappropriate localization and / or activation of neutrophils may be one of the causes of the pathophysiology of various disorders of the immunological system, for example rheumatoid arthritis.

The invention will be further elucidated in conjunction with examples which are intended only to illuminate but not to limit the method of the invention and illustrate how the method of the invention can be practiced in practice.

Unless otherwise indicated, all temperature data in the exemplary portion of the application is in degrees Celsius.

Restriction endonucleases are used under conditions and in the manner delivered by their manufacturer. Binding reactions are performed according to Maniatise et al., Pages 245-6, using the buffer described on page 246 of this publication, and DNA concentrations in the range of 1-100 pg / ml at temperature 23 ° C for DNA with "blind ends" and 16 ° C for DNA with ends that easily bind to each other. Electrophoresis is performed on a 0.5-1.5% agarose gel containing 90 mM Trisborate and 10 mM EDTA. All radiolabeled DNA is 32P labeled.

The term "rapid prep" refers to the rapid production of small-scale bacteriophase or plasmid DNA, as described, for example, in Maniatise et al., Pp. 365-373.

Example 1

Stage 1 Mo cell lines

Mo cells (ATCC CRL 8066) are commonly grown in alpha medium with 6% carbon dioxide or Iscova 10% carbon dioxide containing 20% fetal calf serum (FCS), 2 mM glutamine, 100 units / ml straptomycin and 100 µg / ml penicillin. Cells should be re-vaccinated every 4-5 days. Cells are counted and seeded with Falcon T-175 flasks containing 100 to 150 ml of culture medium at a density of 3 to 4 x 10 ⁵ cells / ml. The number of cells is doubled in the presence of 20% FCS in 4 to 7 days. The growth rate is not constant and sometimes the cells can completely stop growing and then suddenly grow rapidly. Mo cells can also be grown in a serum-free environment. Cell survival is much better if cells do not wash when transferred from FCS to serum-free environments. The optimal density in a serum-free (SF) medium is 5 x 10 ⁵ cells / ml. The cells grow slightly, or at least retain the same number of cells for three days in a serum-free environment, then 20% FCS must be added for at least 4 days. This growth scheme (3 days SF), 4 days 20% FCS) can be repeated weekly if serum free environment is desired for several months without any apparent cell damage.

Tier 2 Determination of CSF effectiveness

A. Determination using bone marrow

Fresh bone marrow is obtained and bone fragments are removed, then the marrow is diluted 1: 1 with sterile sodium chloride phosphate buffered saline (PBS) at room temperature, and then the marrow is overlaid with Ficoll-Paque, approximately 30 ml BM-PBS and 6. ml Ficollu. The mixture is centrifuged at 1500 rpm for 40 minutes at room temperature. The fat and PBS layer is isolated and discarded. The turbid layer was then removed by pipette and washed twice with PBS, and then cells were counted and then plated in RPMI (GIBCO, RPMI 1640) with 10% Heat Inactivated FCS (HIFCS) for 3 hours to remove clinging cells.

- 10GB 285023 B6

The culture medium, which must be fresh, is prepared as follows:

% FCS,

0,3% agar, dissolved in water, cooled to 40 ° C,

2x Iscoves (volume ratio 1: 1 in agar),% P / S final concentration of 100 pg / ml straptomycin and 100 units / ml penicillin,

ΙΟΛμ α-thioglycerol in 2x Iscoves from a stock solution having a concentration of 10 ^-2 M, agar is cooled to 40 ° C and added to other ingredients in the water bath, the resulting mixture is cooled to 37-38 ° C and this temperature was maintained.

After 3 hours, the non-adherent cells are removed by pipette and the material is centrifuged and cells are counted again. Add 2 x 10 < 6 > cells / ml and maintain the medium in a water bath at 37-38 ° C. Then, a sample, such as the medium from the transfected cells, is usually added in an amount of usually 10 microliters to the first row of microtiter plate wells in duplicate. Then, 100 microliters of cell suspension is added to each 15 well and then an additional 50 microliters of cell suspension to each well of the first row. The material in the wells is thoroughly mixed and 50 microliters of solution from the first row is transferred to the next row and then passed through the entire plate to obtain a 1: 3 dilution step. The plate is wrapped in parafilm and incubated for 10-14 days at 10 % carbon dioxide at 37 ° C in a humid environment, counting 20 colonies.

The total number of colonies in each well is counted. Several wells without sample were used in each experiment to obtain a baseline colony count. This number is subtracted from the number of colonies found in each well containing the sample. One unit of CSF is 25 that will produce one colony above the base number of colonies per 10 ⁵ cells from human bone wood when 10 ⁵ cells per milliliter are used, if the CSF concentration is below the saturation level . This concentration is determined by diluting and comparing the number of colonies at different dilutions to determine the concentration just prior to the saturation level.

In this assay, colonies containing granulocytes, monocytes or both types of colonies are counted. The type of cells in the colonies is determined by colony harvesting and cell staining.

B. Assay using KG-1 cells

KG-1 cells, described in Blood, vol. 56, No. 3 (1980), grown in Iscova medium with 10% FCS, cells are inoculated twice a week using 2 x 10 ⁵ cells / ml at each passage. The cells are used for the assay at passages 30-35 only. The assay is the same as for bone wood cells except that KG-1 cells are grown in agar mix at 4 x 10 ³ cells / ml.

The number of colonies growing in each well is determined by subtracting the base number of colonies in the same manner as when using bone marrow cells. One unit / ml for these cells using CSF is that concentration of CSF that stimulates half the maximum amount of KG-1 cell colonies to grow (saturation). The maximum number of colonies can be achieved by adding CSF to several wells until saturation.

Stage 3 Construction of vector p91023 (B)

The vector used for the transformation was the vector pAdD26SVpA (3) described by Kaufman et al., Mol. Cell Biol. 2 (11), 1304-1319 (1982). This vector has the structure shown in Figure 2. This plasmid contains the cDNA gene for mouse dihydrofolate reductase (DHFR), which is under the transcriptional control of the adenovirus 2 (Ad2) major promoter. The sequence from the immunoglobulin gene that is present between the Ad2 promoter and the DHFR sequence is assigned to the adenovirus and alters the 5 'and 3' ends. The SV40 polyadenylation site is located downstream of the sequence,

Which is the code for DHFR. The sequence of this vector, which is derived from prokaryotic cells, is taken from pSVOd (Mellon, P., Parker, V., Gluzman, Y. and Maniatis, T., 1981, Cell 27. 279-288) and does not contain the sequences of zpBR322, which cause inhibition of replication in mammalian cells as described in Lusky, M., and Botchan, M, 1981, Nature 293, 79-81.

The above plasmid is converted to plasmid pCVSVL2 as shown in FIG. 2 and to plasmid pAdD26SVpA (3) (d) by deleting one of the two restriction endonuclease sites of PstI in the original plasmid. This method is accomplished by partial digestion with said enzyme to yield a subpopulation of linerized plasmids in which only one PstI site is cleaved, followed by treatment with a Klenow fragment which binds to the recirculation of the plasmids, which is then used to transform E. coli, and search for colonies having a cleaved PstI treatment site that is located near the 3 'end of the polyadenylated SV40 sequence.

The leader sequence of adenavir and the virus-derived genes (VA-genes) were inserted into plasmid pAdD26SVpA (3) (d) as shown in Figure 2. The vector was digested with PvuII to yield an open linear molecule which opens near the end 3 'of the first of the three elements forming the leader sequence. The plasmid pJAW 43, described in Zein et al., 1979, Cell 16 851, is digested with XhoI, the Klenow fragment is ligated, the material is digested with PvuII and the 140 base pair fragment containing the second element and part of the third element leading sequence was isolated by acrylamide gel electrophoresis (6% in trisborate buffer according to Maniatis et al. (1982), supra). The 140 base pair fragment was then ligated to plasmid pAdD26SVpA (3) (d) after digestion with PvuII. The product of this binding was used to transform E. coli to give this microorganism resistance to tetracycline, colonies were screened by the method of Grunstein and Hogness using a 32p-labeled sample by hydridation to a 140 base pair fragment. DNA was prepared from the positive hybridization colonies to determine whether the reconstructed PvuII site was located at the 5 'or 3' end of the 140 bp fragment inserted to demonstrate whether the hybridization was specific for the second or third element. Adenovirus leader sequence. The correct orientation of the site of action of said enzyme is found at the 5 'end of the 140 bp insertion sequence. The plasmid thus generated is designated pTPL and is shown in FIG. 2.

The SV40 Ava II D fragment containing the SV40 support sequence is obtained by digesting the SV40 DNA with Ava II, terminating with a Klenow fragment after digesting with Pol I, then linking the fragments with Xho 1, then digesting with the Xho 1 enzyme. the material is cleaved by Xho 1 at the cleavage site and the four longest fragments (D) are isolated by gel electrophoresis. The fragments thus obtained were ligated to the pTPL plasmid after digestion with Xho 1 to give plasmid pCVSVL2-TPL. The orientation of the SV40 D fragment in this plasmid is such that the SV40 promoter has the same orientation as the main adenovirus promoter.

In order to introduce the genes associated with the adenovirus virus (VA) into plasmid pCVSVL2-TPL, a plasmid pBR322 containing a fragment of type 2 Hind III B adenoviruses is first constructed, and the type 2 adenovirus DNA is digested with Hind III and the fragment B isolated. gel electrophoresis. The fragment thus obtained is then inserted into plasmid pBR322, which had been previously digested with Hind III. After transformation of E. coli to obtain ampicillin resistance, recombinant strains are screened to identify a strain having a Hind III B fragment, and the orientation of the fragment is determined by restriction enzyme digestion. Plasmid pBR322-Ad Hind III B contains a fragment of type 2 Hind III B adenoviruses in the orientation shown in FIG. 3.

As shown in Figure 3, the VA genes can be obtained from plasmid pBR322-Ad Hind III B by digestion with Hpa I, after addition of EcoR1 binding sequences followed by Eco RI digestion with 1.4 kb fragment isolation. The fragment, which is terminated after digestion with ExoR1, is then ligated to a site in the plasmid pTPL previously digested with the same.

- 12GB 285023 B6 enzyme. After transformation of E. coli HB101 and the selection of tetracycline-resistant clones, the colonies were subjected to a filter hybridization assay using a VA gene-specific DNA sample. DNA was then prepared from positive hybridization clones and subjected to restriction endonuclease characterization. The product is a plasmid called p91023.

The two EcoRI cleavage sites in plasmid p91023 are removed so that this plasmid is completely digested with EcoRI, yielding two DNA fragments, one containing about 7 Kb and the other about 1.3 Kb containing VA genes. The ends of both fragments are filled using the Klenow fragment and Poli, and then the two fragments, i.e. the 1.3 Kb and 7 Kb fragments, are recombined. Plasmid p91023 (A), which contains the VA genes, is similar to plasmid p91023, but VA genes are both inserted in the EcoRI site in both cases, as shown by Grundstein-Hogness analysis and restriction enzyme treatment.

The single PstI site in plasmid p91023 (A) was removed and replaced with an EcoRI site. Then, the plasmid p91023 (A) was completely digested with PstI, and then the Klenow fragment was ligated using Poli, and the helper sequences were ligated to create a EcoRI site in the plasmid p91023 (A) instead of the PstI site. The linear plasmid p91023 (A) with the aid sequences to create an EcoRI site was separated and completely digested with EcoRI, and the fragments so obtained were re-ligated. The plasmid p91033 (B) is isolated, identified and identified to be a plasmid-like structure of p91023 (A) except that the plasmid contains an EcoRI site instead of a PstI site.

Step 4 Preparation of a cDNA flask

Mo cells were cultured for 16-20 hours using PHA and PMA to stimulate lymphokin production. Cells are then plated at 5 x 10 ⁵ cells / ml in Iscova medium with 20% FCS, 0.3% v / v PHA and 5 ng / ml TPA. Cells are separated by centrifugation. The cells thus separated are resuspended in 20 ml ice-cooled hypotonic cell disruption buffer (RSB buffer: 0.01M Tris-HCl, pH 7.4, 0.01M KCl, 0.0015M MgCl ₂ , 1 µg / ml). cycloheximide, 50 units / ml RNAsin and 5 mM dithiothreitol). The cells were allowed to swell on ice for 5 minutes and then mechanically disrupted in a glass homogenizer of ten strokes of the tight-fitting piston. The homogenate is then centrifuged at 2000 rpm (Beckman J6 centrifuge) to centrifuge the nuclei and undisturbed cells. The supernatant is stored in ice and the cores are resuspended in 10 ml RSB and then centrifuged again at low speed. This second supernatant is combined with the first supernatant and the recovered material is centrifuged again at low speed to remove residual nuclei and undisturbed cells. The supernatant in this case is mixed by adding 2M KCl up to a concentration of 0.15M KCl and then centrifuged again for 30 minutes at 25,000 rpm (Beckman Sw 28 rotor) to obtain cell membranes as a sediment. The cell membrane sediment was carefully washed with cold RSB and then resuspended in 12 mL RSB containing 2M sucrose and 0.15M potassium chloride. A discontinuous gradient was then prepared in a centrifuge tube (Beckman / SW41) by overlaying 6 ml of the membrane solution in 2M sucrose to 2 ml RSB with 2.5M sucrose and 0.15M KCl. Fill the tubes with 2.5 ml RSB containing 1.3 M sucrose and 0.15 M KCl. These gradients are then centrifuged for 4 hours at 27,000 rpm (Beckman rotor SW41) at 4 ° C. The membrane layer, located at the interface between the concentration of 2.0 M and 1.3 M sucrose, was carefully removed with syringe and # 18 needle. The membrane fractions were mixed together and diluted with one volume of distilled water, then added. to a concentration of 0.5% Triton X-100 and to a concentration of 0.5% sodium deoxycholate, after which the material was extracted with an equal volume of phenol. The aqueous layer was then re-extracted with a 1: 1 mixture of phenol and chloroform, finally with an equal volume of chloroform. Then, the membrane bound RNA is precipitated by the addition of sodium chloride to a concentration of 0.25M and 2.5 volumes of cold ethanol, followed by incubation overnight at -20 ° C. Precipitated RNA is separated by centrifugation

Minutes at 4000 rpm (Beckman J-6 centrifuge) and then resuspended in 1 ml of distilled water. Approximately 1 mg of RNA is obtained from 2 x 10 ⁹ cells. An appropriate amount of mRNA is then isolated from the total amount of RNA by column chromatography containing 0.5 ml of oligo dT-cellulose. The RNA was heated to 70 ° C for 5 minutes, cooled rapidly in ice, and then diluted 5x at room temperature with binding buffer containing 0.5M LiCl, 0.01M Tris-HCl, pH 7.4 , 0.002M EDTA and 0.1% SDS. The RNA in this buffer is then passed through a column of oligo dT-cellulose at equilibrium in the same buffer at room temperature. Wash the column with 5 ml of binding buffer and then with 5 μΐ 0.15M LiCl, 0.01M Tris-HCl pH 7.4, 0.002M EDTA, and 0.1% SDS. Finally, mRNA is eluted using 2 ml of 0.01 M Tris-HCl pH 7.4, 0.002M EDTA and 0.1 SDS. Then, mRNA is precipitated by the addition of sodium chloride to a concentration of 0.25M and 2.5 volumes of ethanol and incubated overnight at -20 ° C. Precipitated mRNA was collected by centrifugation for 30 minutes at 30,000 rpm (Beckman SW55 rotor). The supernatant was carefully decanted and the mRNA pellet resuspended in 50 ml of water. This suspension is then adjusted to a concentration of 0.25M sodium chloride and then extracted with a 1: 1 mixture of phenol and chloroform, then three times with chloroform. Then, mRNA is precipitated by adding 2.5 volumes of ethanol. The mixture is frozen several times and thawed in a dry ice / ethanol bath and then centrifuged in an Eppendorf centrifuge for 15 minutes. Carefully decant the supernatant and resuspend the mRNA pellet in 20 μΐ of distilled water. The final yield is approximately 30 pg of mRNA.

The first strand cDNA was prepared using standard methods. To do this, dilute 10 µg of membrane mRNA with 100 µΐ of a cDNA formation containing 300 mmol Tris pH 8.4, 140 mmol KCl, 10 mmol MgCl ₂ , 10 mmol β-mercaptoethanol, 500 µΜ dATP, dGTP , dCTP and dTTP, 5 µg oligo-dT (phosphorylated at an average size of 12-18), 150 µCi of 32 P dCTP (400 Ci / mmol) and 20 RNAsin ribonuclease inhibitor units. The reaction is initiated by the addition of 100 units of reverse transcriptase incubated for 30 minutes at 42 ° C. Then, the reaction is stopped by adding EDTA to a concentration of 40 mmol and then the RNA is decomposed by incubation for 20 minutes at 65 ° C in 0.2 M sodium hydroxide. The base is then neutralized by the addition of 20 μΐ 2M Tris pH 7.4. The reaction mixture is then extracted with a mixture of phenol and chloroform and then back-extracted using 50 μΐ 10 mmol Tris pH 7.5, 1 mmol EDTA (TE) and the aqueous phases are combined. Then, the single stranded cDNA is converted to double stranded cDNA by incubating the mixture for 12 hours at 16 ° C with 40 units of Klenow fragment in the presence of DNA polymerase I in 100 μΐ reaction medium containing a solution of 50 mM potassium phosphate pH 7.4, 2.3 mmol DTT, 2-mercaptoethanol, 10 mmol MgCl ₂ , 150 µmol each of one of the four deoxynucleotide triphosphates and 25 pCi ³² P dCTP. The reaction is then stopped by extraction with a mixture of phenol and chloroform and the non-incorporated triphosphates are removed by passing the aqueous phase through a 1 ml Sephadex-G-50 column. The separated fractions were combined and precipitated with ethanol.

The cDNA pellet was washed with cold ethanol and then resuspended in 200 μΐ of 20MMO1 Tris pH 8.0, 1 mmol EDTA, 80 pmol S-adenosylmethionine and 300 units EcoRlmethylase for 60 minutes at 37 ° C. The reaction is stopped by extraction with a mixture of phenol and chloroform and the methylated cDNA is isolated by ethanol precipitation.

The cDNA pellet is washed with 70% ethanol and then resuspended in 200 μΐ SI buffer (Maniatis et al.) And the material is incubated with 200 units of S-nuclease at 30 ° C for 30 minutes. The reaction is then stopped by extraction with a mixture of phenol and chloroform and the cDNA is isolated by ethanol precipitation.

The end of the double stranded cDNA is filled by incubation in 100 μΐ 20 mmol Tris pH 7.4, 50 mmol sodium chloride, 10 mmol 2-mercaptoethanol and 500 pmol of all four deoxynucleotide triphosphates with 25 units of Klenow fragment at room temperature for 30 minutes. The reaction is stopped by extraction with a mixture of phenol and chloroform and the cDNA is isolated by ethanol precipitation.

- 14GB 285023 B6

The cDNA thus obtained is routinely treated in 50 μΐ buffer containing T4-ligase (Maniatis et al.) In the presence of 500 pMol of R1 binding chains with a sequence of pCGGAATTCCG (New England Biolabs) using 2000 units of T4-ligase overnight at 16 ° C. The reaction is stopped by incubation at 70 ° C for 20 minutes, then the reaction mixture is diluted to 300 microliters so that the final salt concentration is 0.1 mol NaCl, 10 mmol MgCl ₂ , 50 mmol Tris-HCl pH 7.4. Then, the cDNA was digested for 2 minutes at 37 ° C with 700 units of EcoR1. The reaction is stopped by extraction with a mixture of phenol and chloroform and the cDNA is isolated by ethanol precipitation. Resuspend the pellet in 50 μΐ TE and pass through a column containing 5 ml of C1-4B. The fractions obtained are combined and precipitated with ethanol. The precipitated cDNA was electrophoresed on a 1% agarose gel in Tris-acetate buffer in the presence of 1 µg / ml ethidium bromide. Then, 500-4000 base pairs cDNA is isolated from the gel using a standard procedure. The precipitated cDNA is extracted with a mixture of phenol and chloroform, precipitated with ethanol, and the pellet (after rinsing with ethanol) is resuspended in 50 μΐ TE. The final yield was 10 to 500 ng of cDNA.

Next, the preparation of the vector for expression of p91023 (B) will be described. 500 ng of the phosphatase treated EcoR1 digested vector were ligated with 100 ng of cDNA in 100 μΐ reaction medium (standard T4-ligase medium) overnight at 16 ° C. The reaction is stopped by extraction with a mixture of phenol and chloroform, and then the bound cDNA is separated by ethanol precipitation after addition of 5 µg / tRNA as carrier.

The ethanol precipitated DNA is washed with 70% ethanol and then resuspended in 100 μΐ TE. This DNA is then used in 4 μΐ aliquots to transform E. coli MC1061 (4 μΐ for 100 μΐ transformation). Each of the 25 transformed assays was spread on a 150 mm diameter Petri dish containing 1% agar, in an L-medium with 10 µg / ml tetracycline (Tet-plate), and the plates were incubated overnight at 37 ° C. Approximately 2000 colonies are developed on each plate, resulting in a total of approximately 50,000 colonies. Each of these colonies has a diameter of approximately 0.5 mm. Each of these colonies has a diameter of approximately 0.5 mm. Each of these colonies was transferred to a microcellulose disc of 137 mm diameter by carefully placing a dry filter on the agar surface, which was then carefully removed. All colonies remain on the filter, which is then deposited by the colonies up onto fresh Tet-plate. The colonies were allowed to grow for several hours and then resuspended by placing a freshly moistened filter exactly on the original filter, pressing the filters together, then separating and placing each filter on a fresh Tet-plate and incubating the plates overnight at temperature 37 ° C. Each copy shall be carefully marked so that the colony placement can be accurately compared to that of the original filter.

Step 5 Preparation of plasmid DNA mixture

Carefully cut each of the 25 copies of the filters into eighths with a scalpel, recording the orientation of each of the eight compared to the original filter. Colonies were removed from each section into 10 ml of L-broth. The bacteria were collected by centrifugation at 3000 rpm for 10 minutes (Beckman J-6 centrifuge) and then resuspended in 0.6 ml of 25% sucrose with 50 M Tris-HCl pH 8.5 and converted to protoplasts by addition 0.2 ml of 5 mg / ml lysozyme followed by incubation on ice for 5 to 10 minutes. The protoplasts are then incubated at room temperature for 10 minutes, 0.125 ml of 0.5 mol EDTA is added and the cells are disrupted by the addition of 0.12 ml of 10% SDS in 50 mmol Tris-DC1 pH 8.0. The disrupted material is mixed gently, incubated for 15 minutes at room temperature, and then protein and chromosome DNA is precipitated by the addition of 0.3 ml of 5M NaCl. After incubation on ice for 15 minutes, the material is centrifuged on an Eppendoff centrifuge for 30 minutes with cooling. The supernatant was carefully removed to give a viscous DNA and protein pellet which was diluted by adding 2.5 mL of water. The mixture was extracted with 1 mL of phenol, the layers were separated by centrifugation (10K for 10 minutes, Sorvall SS-34 rotor), and the aqueous layer was separated into a new tube. DNA was precipitated by the addition of 0.5 ml of 5M NaCl and 7.5 ml

Cold ethanol and the mixture was frozen several times in an ethanol-dry ice bath. The precipitated material was collected by centrifugation (10K, 15 minutes, Sorvall SS-34), resuspended in 0.3 ml of 0.3M sodium acetate and reprecipitated (in an Eppendorf tube) by addition of 1 ml of ethanol. After 10-15 minutes in a dry ice / ethanol bath, the precipitated DNA is separated by centrifugation (5 minutes in an Eppendorf centrifuge) and the resulting pellet is resuspended in 100 μΐ sterile TE (10 mmol Tris pH 8.1 mmol EDTA). . From a typical assay, 5-10 µg of plasmid DNA is obtained. Each sample contains DNA from 200 to 500 colonies of the original filter. A total of 200 DNA samples were prepared from 25 filters.

Step 6 Isolation of CSF clones

Each of the DNA samples of Step 5 was separately transferred to M6 monkey cells as described below.

M6 cells are commonly grown on modified Eagle's medium (Dulbecco, DME available from Gibco) containing 10% heat-inactivated fetal calf serum (HIFCS), and the cells are re-inoculated twice a week at a 1: 6 dilution at a ratio of 24 hours. 1: 6 are M6 cells ready for transfection. Twenty-four hours prior to transfection, 1.2 x 10 ⁸ M6 cells were plated (Cell Factory, Nunc) in 1.5 L of DME + 10% HIFCS. Immediately prior to transfection, plates were washed 2 times with 7 ml serum free DME (SF). The DNA was dissolved in 0.1 M Tris pH 7.3 and added to DME containing 2 mmol glutamine, 100 µg / ml streptomycin, 100 units / ml penicillin and 0.20 mg / ml DEAE-dextran, supplemented with per 4 ml of Tris-DNA solution 4 ml of medium containing dissolved DNA was added to a plate containing M6 COS cells and incubated for 12 hours.

After incubation, cells are washed once or twice with 7 ml SF DME. Then 5 ml of DME with 10% HIFCS, 100 units / ml penicillin, 100 µg / ml streptomycin, 2 mmol glutamine and 0.1 mmol chloroquine are added to the mixture and the cells are then incubated for 2 1/2 hours.

After incubation, cells are washed with SF DME and 10 ml DME + 10% HIFCS is added to the plates. After 30 hours, the medium is aspirated and 4 ml DME + 10% HIFCS is added per plate. Cells are harvested by centrifugation after 24 to 26 hours of further incubation.

The medium from each sample was examined for CSF activity using the KG-1 assay. For each CSF positive specimen, a clone must be found on the original filter that is responsible for this. For example, in the case of a positive transfection of CSF efficacy, all colonies of the original filter section on which this efficacy was found are removed. These are usually about 320 colonies, which are collected in 3 ml of L-medium with 10 µg / ml tetracycline. Cultures are grown overnight. 320 colonies were placed in an 18 x 18 matrix. Mixtures were prepared from the horizontal rows and vertical columns of the matrix (36 mixtures, the last horizontal row had only 14 clones). DNA samples are prepared from each mixture and used for transfection of cells (COS). The supernatants from the material thus obtained were tested for efficacy. Two positive tests were obtained, one in the vertical column and the other in the horizontal row. The culture common to both mixtures contained a clone with CSF activity.

A total of twelve individual clones were isolated from the culture and miniprep DNA was prepared from 10 ml cultures in L-medium as described above. Then, 10 μΐ DNA samples from these assays were digested with EcoRl and the resulting DNA fragments analyzed by agarose gel electrophoresis. Nine of the twelve clones had a common, incorporated sequence of approximately 750 base pairs. DNA from four of these clones and the remaining three clones were transferred to M6 COS cells as described above. Supernatants from the transfections thus performed were examined by the KG-1 assay as well as by the bone marrow cell assay for CSF activity. The four clones had 750 base pairs as a common fragment, incorporated in the correct sequence, so that M6 COS cells expressed at a high level,

As shown by the CSF efficacy in both assays, unlike the remaining three clones. This means that the code area for the CSF must be stored in an inbuilt sequence of 750 base pairs.

The DNA sequence coding for CSF was removed from the vector for transformation into a positive clone by digestion with EcoR1, and its sequence was determined using a standard method after subjecting the cloning fragments to the M13 vectors to demonstrate the sequence shown in Figure 1. Plasmid p91023 (B) - CSF, which was first shown to direct CSF expression in COS cells, was then designated pCSF-1. This plasmid was then deposited with the American Type Culture Collection in E. coli strain MC1061 under ATCC No. 39754 on July 2, 1984.

Step 7 CSF protein expression

M6 COS monkey cells transformed with p91023 (B) containing CSF / cDNA as isolated in step 6 were cultured as described in step 6 to produce the CSF protein, which is then found in the culture medium. .

To do this, 1 mg of this DNA (pCSF-1) was dissolved in 1 ml of 0.1 M Tris pH 7.3 and added to 600 ml of DME containing 2 mmol glutamine, 100 units / ml straptomycin, 100 µg / ml. ml of penicillin (P / S) and 0.25 mg / ml of DEAE Dextran with a molecular weight of 500,000 (Pharmacia). 600 ml of DEAE dextran DNA solution was added to M6 COS cells and incubated for 12 hours at 37 ° C. After incubation, cells are washed with 900 ml SF DME and then incubated for 2.5 hours in 600 ml DNA containing 0.1 mmol chloroquine, 10% HIFCS, 2 mmol glutamine and P / S. The chloroquine-containing medium is removed by suction, the cells are washed with SF DME and then 1500 ml of DME with 10% HIFCS is added. After 30 hours, the cells are washed with SF DME, replaced with 800 ml fresh SF DME, and the cells are incubated for 24 hours at 37 ° C for transfection. The medium is then aspirated and replaced with 800 ml of fresh SF DME medium. The cells are then incubated in this medium for 24 hours, after which the medium is removed. As soon as possible, the environmental samples are concentrated 20x by ultrafiltration under pressure using a 2.5 L Amicon chamber with a YM5 membrane that separates the compounds to a molecular weight of 5000.

Step 8 Purification of recombinant CSF

200 ml of concentrated medium from 4 l of the starting material of step 7 are saturated to 30% with ammonium sulfate by adding solid ammonium sulfate and the precipitated protein is collected by centrifugation. The supernatant was saturated to 80% with ammonium sulfate by the addition of additional solid ammonium sulfate and the precipitated protein separated by centrifugation. The pellet was resuspended in 5 ml of 20 mmol sodium citrate, pH 6.1, containing 1 M NaCl. Dissolve the dissolved protein onto a 1.6 x 100 cm column packed with Ultrogen AcA54 equilibrated in the same buffer. CDF activity is eluted from the column at a molecular weight of 19 units or approximately 90 ml of the initial eluate. It has been observed that when gel filtration is carried out at low ionic strength, the CSF agent elutes from the column at two positions with molecular weights of 19 and 38 per unit, possibly indicating dimer formation. The active fractions were pooled and 10% TFA was added to a concentration of 0.15%, and then the material was loaded onto a 0.46 x 25 cm Vydac C4 column at equilibrium in 0.1% TFA. The column is developed with a linear gradient of acetonitrile from 0 to 90% in 0.1% TFA at a rate of 1 ml per minute for a total volume of 340 ml. The CSF activity elutes between 39-43% acetonitrile, fractions 16-20. A sample of fraction 19 of 20 μΐ was analyzed by SDS polyacrylamide gel electrophoresis (13.5% gene according to Lammli, Nature 227, 680 (1970)). A single wide band was observed for a protein having a molecular weight of 18 to 26 units. A somewhat broader range for the molecular weight of CSF is a common property of glycoproteins and appears to reflect large but varying amounts of carbohydrate. The protein from fraction 19 was subjected to Edman degradation using a sequence determination apparatus. From approximately 20 pg of protein it was possible to determine

- 17GB 285023 B6 sequence of the first 16 amino acids as APARSPSPSTQPWEH. The high yield of this single sequence suggests that the CSF protein from fraction 19 was purified to homogeneity. The bioassay showed that fraction 19 contained 3x10 ⁷ units per absorbance unit Α ₂₈ ο. Since typical proteins in an aqueous solution have an extinction coefficient of 0.8 to 1.2 units absorbance A ₂ so per 1 mg protein, the specific efficacy of purified CSF is in the range of 1 x 10 ⁷ to 4x 10 ⁷ units / mg when used. tests with human bone marrow cells.

Example B

Cloning of CSF Gibbon

Step 1 Preparation of Gibbon T-cell mRNA

A Gibbon T cell line sample, designated UCD-MLA 144, was cultured for several weeks in RPMI 1640 (Gibco) and 20% fetal calf serum (FCS) until 1 x 10 ⁹ cells were obtained. These cells were processed to induce the production of high levels of CSF by activation for 24 hours with 10 ng / ml 12-O-tetradecanoylphorbol-13-acetate (TPA in RPNI 1640 with 1% FCS) and then harvested by centrifugation for 5 minutes at 1000 rpm, were washed with phosphate buffered saline (PBS) and then recovered by centrifugation.

Membrane-bound polysomes (MBP) mRNA was prepared from these cells as described in Example A for preparing RNA from Mo cells.

Step 2 Reaction with the first strand pg MBP mRNA from step 1 is diluted in the 50 µΐ cDNA preparation as described in step 4 of Example A and the reaction is started by adding reverse transcriptase. Incubate for 30 minutes at 42 ° C, then stop the reaction by adding EDTA to 50 mmol and dilute to 100 μ a with water. The mixture is then extracted first with a mixture of phenol and chloroform and then with chloroform. The cDNA / RNA hybrids were separated from the non-incorporated triphosphates by column chromatography containing 2 ml of Sepharose CL-4B. The separated fractions were combined and the hybrids precipitated with ethanol. Yield 570 mg.

Step 3 Reaction with second strand cDNA

The first strand cDNA pellet from step 2 is resuspended in 50 ml of water and the second strand synthesis is performed in a standard reaction mixture using E. coli polymerase I, E. coli ligase and ribonuclease H. The reaction mixture is incubated overnight at temperature 16 ° C and then an hour at 37 ° C. The reaction was quenched with EDTA and extracted with phenol / chloroform. The cDNA was then separated from the non-incorporated triphosphates by Sepharosa CL-4B column chromatography, the fractions obtained were pooled and the cDNA was isolated by ethanol precipitation.

Step 4 Preparation of recombinant cDNA

The cDNA pellet from step 3 is resuspended in 75 μΐ of water. Then add homopolymer ends to the ends of the cDNA by adding 10 μΐ of the cDNA solution to 25 μΐ of the standard terminal transferase reaction mixture and incubating the resulting mixture at 30 ° C for 5 minutes. The reaction is stopped by adding EDTA to a concentration of 40 mmol and then heat inactivated at 68 ° C for 10 minutes. Then, 10 ng of the cDNA so treated was combined with 50 ng of plasmid pBR322 with G (NEN) in 10 μΐ 10 mmol Tris pH 7.5, 1 mmol EDTA

285023 B6 and 100 mmol of sodium chloride. The resulting reaction mixture was incubated at 68 ° C for 10 minutes and then at 57 ° C for 2 hours.

Stage 5 Transformation of bacteria

The E. coli strain MC1061 is grown in L-broth, then cooled on ice, the cells separated by centrifugation and calcium chloride added to prepare the cells for transformation. Then, 5 μΐ of the solution from step 4 is incubated with 200 μΐ of the bacteria thus prepared. This transformation was repeated 15 times using all the amount of cDNA from the previous step and plated on 15 cm diameter plates with 1% L-broth agar containing 10 µg / ml tetracycline. Approximately 1000 colonies were grown on each plate.

Grade 6 Growing copies

The 1,000 colonies were transformed individually, transferred to fresh plates at 500 colonies per grid and grown overnight at 37 ° C. Then the colonies are separated from the plates by pressing the dry microcellulose filter onto the plate surface. Two copies are prepared from each base filter. The original filters were stored at 4 ° C and the copies were treated with alkali and then heat dried to prepare them for hybridization.

Step 7 Preparation of ³² P-labeled hybridization samples

The incorporated pCSF-1 cDNA sequence was isolated by digestion with EcoR1 followed by agarose gel electrophoresis using Tris acetate buffer and ethidium bromide. The band containing the cDNA fragment was excised from the gel and purified in the usual manner.

300 ng of cDNA fragment was then added to 1 μΙ 10x T4 DNA polymerase buffer (0.33 M Trisacetate pH 7.9, 0.66 M potassium acetate, 0.1 M magnesium acetate and 10 mmol dithiothreitol) with 3 T4 DNA polymerase units (New England Biolabs) and dilute to 10 μ vodou with water. After incubation for 5 to 10 minutes at 37 ° C, the mixture is mixed with 1 μΐ 10 x T4 buffer with DNA polymerase, 1 μΐ 2 mmol of dCTP, dTTP, dGTP and 10 μΐ ³² P dATP (10pCi / pl, 3000) are added. Ci / mmol) and 3 units of Tr DNA polymerase. The reaction mixture is then incubated for 20 minutes at 37 ° C, 1 μΐ 2 mmol dATP is added and the reaction mixture is incubated for 10 minutes at 37 ° C.

Non-incorporated triphosphates were separated from labeled cDNA by Sephadex G100 column chromatography. A second sample of synthetic oligonucleotide with sequence is then prepared

ATC TGG CTG CAC AG, which is complementary to the terminus of the CSF-coding region of the amino-containing portion. This oligonucleotide was labeled with 32p dATP at the 5'-end using a standard polynucleotide kinase reaction.

Step 8 Isolation of CSF cDNA Clones

In standard hybridization assays, 45 clones were shown to hybridize using pCSF-1 cDNA, designated T4. Of these clones, approximately 20 clones hybridized to a sample containing the labeled oligonucleotide. The coding region of one of these clones was subjected to sequence analysis to find that multiple base substitutions were found, some of which resulted in different amino acids in the protein being expressed. The demonstrated differences are shown in Figure 1 above the DNA sequence for the human CSF gene as cloned in Example A.

-19GB 285023 B6

Example C

CSF cloning from peripheral blood lymphocyte mRNA

Step 1 Preparation of peripheral blood lymphocyte mRNA

Peripheral blood lymphocytes were prepared from plasma by-products (provided by the Red Cross) by fractionation using a Ficoll-Hypaque gradient. It was obtained when grown in RMPI-1640 in the presence of 5% fetal calf serum, 0.17% phytohemagglutinide and 10ng / ml PMA density of 2 x 10 ⁶ cells / ml, totaling 6 x 10 ⁹ cells. The cells were separated by centrifugation for 5 minutes at 1000 rpm, then washed with sodium chloride containing phosphate buffer (PBS) and finally the cells were separated by centrifugation. Cytoplasmic RNA was then prepared by carefully disrupting the cells by suspending the cells in 50 ml of cold Triton-containing disruption buffer (buffer consisting of 140 mmol NaCl, 1.5 mmol MgCl ₂ , 10 mmol Tris pH 8.6). and 0.5% Triton X-100) with 10 mmol dithiothreitol (DTT) and 50 units / ml RNA sin (Bitec). The material was divided into two equal portions and each portion was layered on 10 ml of the same buffer containing 20% sucrose. The cell cores were removed by centrifugation at 4 ° C for 5 minutes at 400 rpm. The upper layer, consisting of the cytoplasmic extract, was carefully removed and sodium dodecyl sulfate (SDS) was added to a final concentration of 1%. Then, the solution was extracted twice with an equal volume of 1: 1 phenol / chloroform and RNA was precipitated by adding 2.5 volumes of cold ethanol. Precipitated RNA was collected by centrifugation at 4000 rpm for 15 minutes and then resuspended in a mixture of 0.01 M Tris pH 7.5, 1 mmol EDTA, 0.25 M NaCl (TE buffer with 0.25 M NaCl). ) and the material was reprecipitated by adding 2.5 volumes of cold ethanol. Finally, the RNA was separated by centrifugation and resuspended in 5 ml of water. The total yield was 7.5 mg.

MRNA was isolated from total cytoplasmic RNA by selection on oligo dT cellulose. Then 2.5 mg of all RNA was heated to 65 ° C for 5 minutes. Sodium chloride is added to a concentration of 0.5 M and the RNA is allowed to cool to room temperature and then passed through a column containing 1 ml of equilibrium oligo dT cellulose in TE + 0.5 M sodium chloride (binding buffer). Unbound RNA is separated by thoroughly washing the column with the same buffer. The bound mRNA is then eluted with 3 ml of water and precipitated by the addition of 0.2 ml of 4 M sodium chloride and 2.5 volumes of cold ethanol. Precipitated mRNA was collected by centrifugation at 25,000 rpm for 30 minutes. The resulting pellet of approximately 100 pg is resuspended in 50 μΐ of water.

Step 2 First strand cDNA reaction pg PBL mRNA is diluted in a 50 µm cDNA reaction mixture containing 100 mmol Tris pH 8.4, 140 mmol KCl, 10 mmol MgCl ₂ , 10 mmol 2-mercaptoethanol, 400 pmol dATP , dGTP, dCTP and dTTP, 5 µg oligo-dT with an average size of 12-18 and 25 µCi of ³² P dCTP (400 µCi / mmol) and 20 units of RNAsin ribonuclease inhibitor. The reaction was started by adding 60 units of reverse transcriptase at 37 ° C followed by incubation for 30 minutes at 42 ° C. The reaction was then stopped by adding ETDA to a concentration of 40 mmol followed by extraction with an equal volume of water saturated phenol. The phenol phase was subjected to back extraction with 50 µmol of TE buffer. The aqueous phases were pooled and cDNA / RNA hybrids were isolated from the non-incorporated triphosphates by passing the pooled aqueous phase through a column containing 5 ml Sepharose CL-4B (sigma) equilibrated with TE. The fractions that passed through the column were pooled, made up to 250 mmol sodium chloride, and the nucleic acids were precipitated by adding 2.5 volumes of cold ethanol. The hybrids were centrifuged for 30 minutes at 40,000 rpm. The resulting 2.5 µg cDNA deposit was resuspended in 50 µl water.

-20GB 285023 B6

Step 3 Second strand cDNA reaction

The second strand of cDNA was synthesized by the co-action of the enzyme DNA polymerase I, DNA ligase and RNAase Η from E. coli. 50 μΐ of the reaction mixture thus obtained contained 20 mmol of Tris at pH 8.0, 4 mmol of magnesium chloride, 1.2 mmol of EDTA, 25 μηιοί NAD, always 100 ppm of dATP, dGTP, dCTP and dTTP and 50 pCi of ³² P dCTP of 3000 Ci / mmol. The reaction is initiated by the addition of 3 units of DNA polymerase I, 0.5 units of DNA ligase, and 0.75 units of ribonuclease H followed by incubation for 18 hours at 16 ° C and then an hour at 37 ° C, after which the reaction is stopped by adding EDTA to concentration of 40 mmol and then extracted with an equal volume of phenol. The phenol phase is back extracted with 50 μΐ TE, the aqueous phases are combined and the cDNA is separated from the non-incorporated triphosphates by chromatography on a Sepharose CL ^ IB column as described above for the first strand. Depending on the incorporation of ³² P, the first strand of CDNA was quantitatively transformed into a double stranded form.

Step 4 Preparation of recombinant cDNA

At the end of the cDNA, homopolymic C-sequences were ligated by gently heating 400 ng of cDNA in a 50 μΐ reaction mixture containing 1 mmol of 2-mercaptoethanol, 1 mmol of cobalt chloride and 9 units of terminal deoxynucleotide tetrasferase for 5 minutes at 30 ° C. The reaction was then quenched by the addition of EDTA to a concentration of 40 mmol, followed by heating to 68 ° C for 10 minutes. then 200 ng of the obtained cDNA was bound to 500 ng of G-terminated pAT153 (Amersham) in 100 μΐ 10 mol Tris pH 7.5, 1 mmol ETDA and 100 mmol sodium chloride. Binding was performed at 57 ° C for 2 hours at 5 minutes pre-incubation at 68 ° C.

Stage 5 Transformation of bacteria

The product obtained by cDNA binding was used directly to transform E. coli strain MC1061. A fresh colony of bacteria was used to inoculate 50 ml of L-broth and was cultured for several hours until the optical density at 550 nm was 0.25. The cells were cooled in ice and then collected by centrifugation at 2000 rpm for 10 minutes. The resulting pellet was resuspended in 10 ml of cold 0.1 M calcium chloride and then left on ice for 10 minutes. The cells were centrifuged at 2000 rpm for 5 minutes and resuspended in 2.5 ml of 0.1 M calcium chloride. Then, 10 μΐ of the reaction mixture after cDNA binding was incubated with 200 μΐ of bacteria treated with calcium chloride first for 30 minutes on ice and then for 2 minutes at 37 ° C, then 0.8 ml of L-broth was added and the cells incubated 30 minutes at 37 ° C.

A total of 20 transformation reactions were performed using all the amount of bound cDNA. Each of the transformation mixtures was plated on 15 cm diameter plates containing L-broth containing 1% agar and 10 µg / ml tetracycline. Of the 20 transformations performed, 20 plates were prepared and incubated overnight at 37 ° C. Approximately 1500 colonies of bacteria were grown on each plate, representing a total of 30,000 clones.

Step 6 Making copies

The original colonies growing on each plate were transferred to 137 mm nitrocellulose filters so that the dry filter was pressed against the top of the colonies plate and then removed again. Two identical copies were then prepared from each original filter in a conventional manner so as to accurately mark the parts that match. Each original filter was carefully placed upside down on sterile filter paper (Whatman 3MM) on a glass plate. A pre-moistened nitrocellulose filter was then placed on the filter, the filter was covered with a second larger section of filter paper, and the filters were then pressed together by a second glass plate.

-21 GB 285023 B6

The filters were numbered and punctured at three asymmetrical locations so that they could be accurately refitted in the future. Then, the copy was separated from the original filter and was placed upside down on a new L-broth, agar, and tetracycline plate. Immediately, another clean filter was applied in the same way to obtain another copy. Then the original filter was returned to the agar plate and all plates were incubated at 37 ° C for several hours, after which time the colonies of bacteria had a diameter of approximately 1 mm. The original filters were stored at 4 ° C to prepare copies for hybridization as described above.

Step 7 Adjustment of filters for hybridization

Each copy of the original filter obtained in step 6 was deposited by the colonies upward on filter paper (Whatman 3 MM), which was moistened with 0.5 M sodium hydroxide and 1.5 M sodium chloride, for 7 minutes. The filters were then neutralized, moistened for 2 minutes with 1 M Tris pH 7.5 with 1.5 M sodium chloride, then neutralized again for 5 to 10 minutes and then placed on filters moistened with SSC buffer for 5 minutes (0.015 M). sodium citrate, 0.15 M sodium chloride, pH 7.4), then the filters are air dried and then heated to 80 ° C for 1 to 2 hours under vacuum.

Step 8 Isolation of CSF cDNA Clones

The duplicate filters were in duplicate contacted with the incorporated pCSF-1 cDNA after radiolabeling, the plasmid was prepared as described in Example B. 20 colonies were hybridized with the cDNA. 12 of these colonies were separated and grown overnight in L-broth for further analysis. The material obtained by Pst 1 digestion from these clones was somewhat heterogeneous, but three of these clones had almost the full length of the desired sequence. For one of these sequences, the entire sequence was analyzed in detail. The sequence that encoded the CSF in this clone was identical to the corresponding plasmid sequence, pCSF-1, which means that it contained a T at position 365-CSF (Ile).

Example D

Purification of CSF from Mo cell line

1 medium from the culture of the Mo cell line is incubated for 30 minutes at 55 ° C to inactivate the HTVL-II virus associated with this cell line. The medium is then concentrated by ultrafiltration under pressure using a Pellicon Casette with a PTGC (1/6 m ² ) membrane capable of separating the substance to a molecular weight of 10,000. The protein is then further concentrated by precipitation to 80% with ammonium sulfate. 800 mg of the resulting protein is resuspended in 100 ml of 20 mmol tris (hydroxymethyl) aminomethane hydrochloride (Tris-HCl), pH 7.4, and the material is then dialyzed against the same buffer a total of 3 times using 4 L of buffer each. The dialyzed protein is then loaded onto a 2.5 x 10 cm column containing DEAE (diethylaminoethyl) -ultrogen, equilibrated in the same buffer. The column is then washed with 800 ml of 20 mmol tris-HCl pH 4 and then with 800 ml of 20 mmol tris-HCl pH 7.4 containing 0.12 M sodium chloride, thereby eluting the substance with CSF activity. A series of 10 ml fractions were collected, the active fractions were combined. There are a total of three fractions which are concentrated to 1/6 of their volume, i.e. 5 ml, by ultrafiltration under pressure using an Amicon YM5 membrane which separates the compound to a molecular weight of 5000. The concentrated sample from the DEAE column is applied to the column o size 1.6 x 100 cm containing AcA44 ultrogel (acrylamidagarose ultrogel with fractionation up to 10 to 130 k molecular weight units) the column is in equilibrium in 20 mmol of N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid (HEPES), o pH 7.4 with 50 mmol sodium chloride and 0.01% polyethylene glycol (PEG-8000). The CSF effluent which was eluted from the column had a molecular weight of 30 k units. The active fractions were pooled and acid was added thereto

Trifluoroacetic acid (TFA) to a concentration of 0.15% by the addition of a 10% solution of the acid, and the mixture was applied to a 1 x 25 cm column containing Vydac C4 in reverse phase. The column was developed with a linear gradient of 0 to 90% acetonitrile in 0.1% v / v TFA at a rate of 4 mL / min to give a total of 1000 mL of eluate. The CSF is eluted at about 47% v / v acetonitrile. The active fractions are pooled and adjusted to a concentration of 0.05% by volume of heptafluorobutyric acid (HFBA) by addition of 0.15% HFBA and the resulting material is then applied to a 0.46 x 25 cm column containing equilibrium Vydac C _4. in HFBA at a concentration of 0,15% vol. The column is developed with a linear gradient of 0 to 90% v / v acetonitrile in 0.15% v / v HFBA at a rate of 1 mL / min to give a total of 340 mL of eluate. The substance with CSF activity elutes at a concentration of acetonitrile of approximately 53% by volume. Fractions 37 to 44 are effective, each having a volume of 1 ml. 0.15 ml of fraction 40 was concentrated to 1/4 volume using a SAVANT Speed Vac concentrator and then 40 μΐ 2x SDS buffer (0.125 M Tris-HCl pH 6.8, 4% SDS, 20% glycerol and 0.004) were added. % bromophenol blue). These samples were immersed for 2 minutes and then loaded onto a 13.5% SDS gel according to the method of Lammli, U. Nature 227, 680 (1970) as shown in Figure 2. Fraction 40 had an efficiency of 110,000 units / ml at use of bone marrow cells. This corresponds to approximately 3.0 x 10 ⁷ units per A ₂₈₀ absorbance unit. Since typical proteins have extinction coefficients in the range of 0.8 to 1.2 A λ units of ₂₈ µg, CSF purification has a specific efficiency in the range of 1 x 10 ⁷ to 4 x 10 ⁷ units / mg. A sample of 1 µg of purified GM-CSF was subjected to Edman degradation using an Applied Biosystems Gas Phase Microsequenator. The sequence between residues 3-5 was Ala Arg Ser.

Example E

By co-transforming and amplifying the CSF sequence in CHO cells, plasmid p91023 (B) -CSF was inserted into DUKX-B11 cells that do not contain DHFR CHO as described in Chasin and Urlaub PNAS 77, 4216 (1980) by protoplast fusion described in Sandri-Goldin et al., Mol. Cell. Bio.I., 743-752, 1981. Growth and maintenance of CHO cells has been described by Kaufman and Sharp, J. Mol.-Biol. 150, 601-621, 1981. For protoplast fusion, plasmid p91023 (B) ~ CSF-1 was incorporated into E. coli strain HB101 and the bacteria were grown in 50 ml m9 medium containing salts and 0.5% casein amino acids. 4% glucose, 0.012% magnesium sulfate, 5 µg / ml thiamine and 10 µg / ml tetracycline to an absorbance of 0.6 at 600 nm. Chloramphenicol was added to a concentration of 250 µg / ml and the culture was incubated at 37 ° C to propagate the plasmid copies. The cells were then centrifuged at 3000 rpm for 10 minutes at 4 ° C and then suspended in 2.5 ml of cooled 20a sucrose in 50 mmol Tris-HCl pH 8.0. Lysozyme was then added in 0.5 ml of a 5 mg / ml solution in 0.25 M Tris-HCl, pH 8.0, and the mixture was then stored on ice for 5 minutes. Then 1 ml of 0.25 M EDTA pH 8.0 is added and the mixture is placed on ice for a further 5 minutes, then 1.0 ml of 0.05 M Tris-HCl pH 8.0 is added and the suspension is incubated for 15 minutes. minutes at 37 ° C, after which time the bacteria are converted to protoplasts. The suspension is then slowly diluted with 20 ml of preheated medium containing 10% sucrose and 10 mmol of magnesium chloride and allowed to stand at 37 ° C for 15 minutes. A protoplast solution (approximately 10 ⁹ / ml) was then added to DHFR CHO-free DUKX-B11 cells in six wells at about 1 x 10 ⁶ cells / well in a ratio of about 1 to 2 x 10 ⁴ protoplasts / cell and protoplasts were added. Pellet together with cells by centrifugation for 8 minutes at 2000 rpm using an IEC model K centrifuge. After centrifugation, the supernatant is aspirated and 2 ml polyethylene glycol solution (50 g PEO-1450, Baker Chem. Co. in 50 ml medium) is added to each. recess of the above plate. The cells are centrifuged again at 2000 rpm for 90 seconds, the polyethylene glycol solution is removed, and the plates are rinsed 3 times with 4 ml of medium / well. The cells are then treated with trypsin, suspended in 10 ml of medium containing 10% fetal calf serum and centrifuged in a tube (conical) at 500 rpm in a conventional centrifuge. The pelleted cells from the three wells were pooled

And plated onto a 10 cm diameter plate, fresh plate containing 100 µg / ml kanamycin, thymidine, adenosine, deoxyadenosine, penicillin and streptomycin and 10% dialyzed fetal calf serum is added to each plate. Kanamycin is added to prevent the growth of any bacteria that would persist when converted to protoplasts.

After two days, the cells were transferred at a ratio of 1: 15 to α- with 10% dialyzed fetal calf serum, penicillin and straptomycin, but without nucleosides. A nearly selective nucleoside-free medium was then added to the cells after 4-5 days.

Colonies appeared after 10 to 12 days of cultivation in selective media. Two methotraxate (MTX) selection experiments and an amplification experiment were then performed. In a first series of experiments, individually independently cloned transformants were isolated according to DHFR expression, and each clone was grown under conditions where the copy number of foreign DNA was amplified, i.e. using increasing amounts of methotrexate. In a second series of experiments, a mixture of independent transformants according to DHFR expression was isolated, and this mixture was further cultured under conditions whereby foreign DNA was amplified, i.e., increasing methotrexate concentration. Individual clones were then isolated and analyzed for GM-CSF expression. Clones that showed the highest level of GM-CSF expression were further grown under conditions in which foreign DNA amplification occurred, i.e., increasing concentrations of methotrexate in the culture medium.

In one experiment, seven DHFR ⁺ transformants were pooled in a nucleoside-free environment. These cells were then grown at increasing MTX concentrations from 0.02 pmol through 0.1 and 0.5 up to 2.0 pmol. Examination of the efficacy of GM-CSF using KG-1 cells revealed 3000-12000 units / ml. These cells were then cloned at 0.5 and 2.0 pmol MTX. Clones that were obtained in 0.5 µM MTX (clones 010, D2 and B6) were sequentially selected for growth at 2.0 pmol MTX. In tests for GM-CSF activity in KG-1 cells, cloned cell lines produced 15,000 to 300,000 units / ml of this activity. The GM-CSF produced by the cells in the method of this example has the amino acid sequence shown for CSF-Thr in Fig. 1

Example F

Expression of GM-CSF in E. coli

GM-CSf expression was achieved in E. coli using the pTALC-195R vector, the material shown in Figure 6. The sequence coding for GM-CSF starts with a synthetic ATG sequence CCA CCA CCT CCT TCT CCA TCT CCA TCT ACT, which determines the initial eleven amino acid residues of complete GM-CSF. The remainder of the sequence coding for GM-CSF in pTALC185R is identical to the sequence in pCSF-1, nucleotides 97-447, followed by the TAR TAR TAG sequence. This is immediately followed by a termination sequence consisting of three parts and the linked p-UC-18 sequence. The tetracycline resistance gene from plasmid pBR322 was inserted into the CSF gene with the orientation opposite to that of the CSF 100 base gene downstream of the pUC-18 sequence. The tetracycline resistance gene carries its own promoter. Then, the β-lactamase gene is located in the direction opposite to the clockwise direction, followed by the sequence from pUC-18 (CoLE1) for the origin of replication.

The final plasmid structure before returning to the CSF sequence is the PL promoter. This promoter is essentially described in A. Skatzman and M. Rosenberg (in "Molecular cloning, a laboratory manual" (1982), Cold Spring Harbor Laboratory, p. 419). CSF expression is under the control of the PL promoter after heat induction in a suitable E. coli host strain.

-24GB 285023 B6

The strain used for all of these constructs was W3110 lac108, described by R. Brent and M. Ptashne PNAs 78 (1981) 4204-4208.

The λ DNA fragment (λ-nucleatides 34499 to 38214) was integrated into the chromosome W3110 lac10 at the lacZ site. Integration was performed using an integration vector composed of sequences of pBR325 carrying chloramphenicol and ampicillin resistance genes as well as an origin for replication from pBR322 as described in F. Bolivar Gene 4 (1978) 121-136. The λ DNA fragment is inserted into the lacZ gene, which itself is present in the plasmid as a fragment that extends into the BstEil site in the Lac to the TthillI site downstream of the lacZ.

Integration of λ DNA into the β chromazomal copy of lacZ was achieved by homologous recombination and lac ⁺ , ampicillin sensitive and chloramphenicol resistant colonies were obtained. Further recombination, which resulted in the removal of all additional plasmid sequences while leaving the integrated λ DNA fragment, occurred on lactose-containing MacConkey plates. The initial phenotype of lac ⁺ , amp ^s , cam ^R was changed after the second recombination to the phenotype lac ', amp ^s , cam ^s . The resulting strain was named GL400 and had the properties of X ^R at 30 ° C and X ^s at 42 ° C. This phenotype demonstrates the existence of a functional chromazomal copy of the C1 ⁸⁵⁷ allele.

The GL400 strain was applied to Ion by PL transduction from a lysate of strain SG20252 (lac Δ ul69, ara Δ 139 de Ion Δ 100: Tn10). Tn10 has been found in tests for Tet ^s on selective formulations as described in S. Maloy, W. Nunn, J. Bacteriol. 145 (1981) 1110-1112.

The resulting host strain was named GI413 (lacl ° L8, LacZ Δ. (Λ Cl, REX, N), Ion Δ 100).

pTALC-185R was transformed into GI413. The overnight culture of this strain was grown at 30 ° C in 5 ml induction medium containing 7 µg / ml tetracycline. The induction medium contains 1 liter of amino acid casein, g Na ₂ HPO ₄ .7H ₂ O, g KH ₂ PO ₄ ,

0.5 gNaCl, gNH ₄ Cl,% glycerol, mg vitamin B], mg CaCl ₂ .2H ₂ O,

0.2 g MgCl ₂ .6H ₂ O.

ml of this medium containing 7 pg / ml tetracycline is inoculated with 125 μΐ of the above culture and the culture is further cultured in a water bath at 30 ° C with shaking until the culture medium reaches an optical density of 0.5 at A550. The medium is then rapidly transferred to a 40 ° C water bath and shaken for a further 2 hours to achieve GM-CSF synthesis. The cells were separated and the CSF content of these cells was detected by SDS-polyacrylamide gel electrophoresis. Under these conditions, GM-CSF accumulates to make up approximately 5% of the cellular proteins.

-25GB 285023 B6

Example G

GM-CSF expression in Saccharomyces cerevisiae

A. Construction of vector

A plasmid was constructed that contained the uracin biosynthesis gene (URA3) as a selectable gene at the onset of 2U replication. The plasmid was derived from Ylp5 as described by Botstein et al., Gene 8, pp. 17-24 (1979), adding a fragment containing the origin of replication from the yeast plasmid.

B. Isolation of the glyceraldehyde phosphate gene

Dehydrogenase (GPDH)

Two GPDH genes were isolated from yeast as described in Holland and Holland, Joumal of Biological Chemistry 255, pp. 2596-2605 (1980). The oligonucleotide sample synthesized from the published sequence was used to isolate the GPDH gene from the yeast genome DNA plasmid array using standard methods. The plasmid containing the entire GAP491 gene had previously been deposited with the collection under ATCC No. 39777.

C. Preparation of the glyceraldehyde phosphate dehydrogenase promoter for expression of a heterologous gene

A plasmid was constructed in which the natural distance of the GPDH promoter from the beginning of the desired heterologous structural gel was maintained. This goal was achieved by incorporating a KpnI enzyme site immediately to the methionine codon at the beginning of the structural gene for GPDH. This region was then incorporated into a vector for the expression of YOp1 yeast.

D. Isolation of factor a genes

The factor a (pheromone) gene was also isolated from yeast and was described by Kurjan and Herskowitz Cell, Vol. 30, pp. 933-943 (1982). A sample of the oligonucleotide synthesized from this sequence was used to isolate the gene from the yeast genome DNA plasmid array in a known manner.

E. Preparation of a plasmid for CSF expression

The AJ14 expression vector shown in Fig. 1 was constructed in a conventional manner from the above components and the human CSF gene. In this vector, the natural CSF leader sequence was removed and the sequence coding for full CSF in the immediate vicinity of the factor sequence was inserted. The link between the GPDH promoter, the pre-pro sequence for factor a, and the sequence for full CSF is shown below and was confirmed by analyzing the sequence of dideoxynucleotide.

AAATAAACAAAATG.CGTTTTCCTTCA ..... AAA AGA GAG GCG

GCC CGC TCG ...

Claims (23)

PATENT CLAIMS A vector comprising a gene encoding a primate GM-CSF protein having the amino acid sequence shown in Figure 1 after the arrow for CSF-Thr, CSF-Ile or CSF-G, or allelic or other functional equivalents thereof, retaining activity GM-CSF in a human bone marrow assay in which at least one amino acid is added, replaced or removed. The vector of claim 1, wherein said gene encodes a protein having the amino acid sequence shown in Figure 1 after the arrow for CSF-Thr or CSF-Ile, or allelic or other functional equivalents thereof, as defined in claim 1. The vector of claim 1 or 2, wherein the functional equivalents of GM-CSF primates, when purified, have an activity of at least 1.10 ⁷ units per mg protein in the human bone marrow assay. The vector of claim 1 or 2, wherein said gene encodes a protein having the amino acid sequence shown in Figure 1 after the arrow for CSF-Thr. The vector of claim 1 or 2, wherein said gene encodes a protein having the amino acid sequence shown in Figure 1 after the arrow for CSF-Ile. The vector of claim 1, wherein said gene encodes a protein having the amino acid sequence shown in Figure 1 after the arrow for CSF-G. The vector of claim 4, 5 or 6, wherein the amino acid sequence of the protein encoded by said gene additionally comprises at its N-terminus a signal sequence preceding the arrow in Figure 1. The vector of claim 4, 5 or 6, wherein the amino acid sequence of the protein encoded by said gene additionally comprises a methionine residue at its N-terminus. The vector of claim 1, wherein said gene comprises the sequence shown in Figure 1 for CSFThr or CSF-Ile, beginning after the arrow and ending at the codon, corresponding to the amino acid at position 127. The vector of claim 1 which is plasmid p91023 (B) deposited under ATCC number 39754. A host cell that is transformed with a vector according to any one of claims 1 to 10. A host cell according to claim 11, transformed with a vector according to any one of claims 2, 3, 4, 5, 9 or 10. The host cell of claim 11 or 12, which is prokaryotic. The host cell of claim 13, which is Escherichia coli. The host cell of claim 11 or 12, which is eukaryotic. The host cell of claim 15, which is a yeast cell. -27GB 285023 B6 The host cell of claim 15, which is a mammalian cell. The host cell of claim 17, which is a Chinese hamster ovary cell. A method of producing a primate GM-CSF protein, comprising culturing a cell according to any one of claims 11 to 18 and isolating the GM-CSF protein. The method of producing a primate GM-CSF protein according to claim 19, characterized in that the cell of any one of claims 12 to 18 is cultured and the GMCSF protein is isolated. A cDNA corresponding to the gene of any one of claims 1 and 10. A cDNA according to claim 21 which corresponds to the gene shown in any one of claims 2, 3, 4, 5, 9 and 10. 23. cDNA of plasmid p91023 (B) which is deposited under ATCC number 39754.