HK1086030A

HK1086030A - Expression vector, methods for the production of heterologous gene products, and selection method for high-producing recombinant cells

Info

Publication number: HK1086030A
Application number: HK06105903.9A
Authority: HK
Inventors: Barbara Enenkel; Jürgen FIEDER; Ralf Otto; Stefanos Grammatikos
Original assignee: 贝林格尔英格海姆法玛两合公司
Priority date: 2002-11-29
Filing date: 2003-11-25
Publication date: 2006-09-08

Description

Expression vector, method for producing heterologous gene product, and method for selecting high-yield recombinant cell

Technical Field

The present invention relates to methods for selecting high producing recombinant cells, methods for producing heterologous gene products and expression vectors, and host cells transfected therewith and useful in such methods.

Background of the invention

Mammalian cells are the preferred host cells for the production of complex biopharmaceutical proteins because their post-translational modifications are not only functionally but also pharmacokinetic compatible with humans. Commercially relevant cell types are hybridomas, myelomas, CHO (chinese hamster ovary) cells and BHK (baby hamster kidney) cells. The cultivation of host cells is increasingly carried out under serum-free and protein-free production conditions. The reasons for this are the associated reduced costs, reduced interference in the purification of the recombinant protein and a reduced potential for the introduction of pathogens (e.g. prions, viruses). The use of CHO cells as host cells has become more widespread because this cell is adapted to growth in suspension in serum and protein free medium and is also recognized and accepted by regulatory authorities as a safe production cell.

To produce a stable mammalian cell line expressing a heterologous gene of interest, the heterologous gene is introduced into the desired cell line, typically by transfection, along with a selectable marker gene, such as neomycin phosphotransferase. The heterologous gene and the selectable marker gene may be expressed simultaneously or together from a single vector, or from separate vectors co-transfected. 2 to 3 days after transfection, cells are transferred to a medium containing a selection agent, for example G418 in the case of the neomycin phosphotransferase gene, and cultured for several weeks under such selection conditions. Emerging resistant cells can then be isolated and investigated for expression of the desired gene product. Due to random and non-directional integration within the host cell genome, a population of cells is obtained that exhibit disparate expression rates of heterologous genes. These may also be non-expressing cells which express the selectable marker but do not express the gene of interest. In order to identify cell clones exhibiting very high expression of the heterologous gene of interest, a large number of clones must therefore be examined and tested, which results in high time, labor and cost expenditure.

Gene amplification is a very popular phenomenon in animal cell culture for the production of recombinant biopharmaceutical proteins. Gene amplification significantly improves the otherwise relatively low yields of many mammalian cell lines. One widely used amplification technique is a dihydrofolate reductase (DHFR) -based gene amplification system, often used in DHFR-deficient Chinese Hamster Ovary (CHO) cells, where DHFR-deficient CHO cells, such as CHO-DUKX (ATCC CRL-9096) or CHO-DG44(Urlaubet al 1983), are transfected with an appropriate vector system encoding DHFR and a protein of interest. Transfectants were then selected in glycine-, hypoxanthine-and thymidine-free medium. Amplification and thus establishment of high-producing cell lines is achieved by the gradual addition of Methotrexate (MTX), an inhibitor of dihydrofolate reductase (Kaufmanet al 1982; US 4,656,134). Subsequent screening of the resulting high producing cells, also subject to randomization and on a probabilistic basis, is therefore a highly labor intensive and time consuming screening procedure.

Various methods have been developed for better and more rapid monitoring of gene transformation and expression. This involves first the use of a reporter molecule such as chloramphenicol-acetyltransferase, luciferase, beta-galactosidase, or the use of a fusion protein containing the coding region for beta-galactosidase or luciferase. However, these corresponding reporter assays have the disadvantage that the cells have to be fixed or lysed and have to be incubated with exogenously added substances and cofactors. Therefore, it is not possible to further culture the analyzed cells at all. A novel approach is based on the co-expression of the E.coli enzyme β -galactosidase, which, although it allows sorting of live cells by means of FACS apparatus (Nolan et al, 1988), requires a hypotonic pretreatment in order to load the fluorogenic substance into the cells. This activity must also be inhibited prior to FACS-based sorting.

Identification of cells expressing heterologous genes is facilitated by the introduction of Green Fluorescent Protein (GFP) from Aequorea victoria (Aequorea victoria) and the development of GFP mutants therefrom as reporter molecules. Co-expression of GFP allows for immediate analysis in vivo and selection of transfectants based on their fluorescence without the need for additional substances or co-factors. The use of GFP as a reporter to monitor gene transfer has been described in various publications. In U.S. Pat. Nos. 5,491,084 and 6,146,826, Chalfie et al describe a method for selecting cells expressing a protein of interest. The method comprises co-transfecting a cell with a DNA molecule comprising a coding sequence for a protein of interest, and a second DNA molecule encoding a GFP-gene. Cells expressing GFP were then selected. Gubin et al (1997) investigated the stability of GFP expression in CHO cells in the absence of selective growth conditions. Cells were transfected with plasmids containing both GFP and neomycin phosphotransferase. Mosser et al (1997) used a plasmid containing a dicistronic expression cassette encoding GFP and a gene of interest (also known as a gene of interest) to identify and select cells expressing inducible products. The target gene is under the control of a regulatable promoter. The coupling of GFP and target gene expression is achieved using a viral IRES (internal ribosome entry site) element, resulting in the expression of bicistronic mRNA encoding GFP and the protein of interest. The plasmid used here does not contain any selectable marker gene itself. Thus, the second plasmid is introduced in the co-transfection or in the subsequent transfection. In contrast, Levenson et al (1998) use retroviral vectors with a bicistronic expression cassette, in which the gene of interest can be cloned before the IRES sequence. The sequence following the IRES sequence encodes instead a selectable marker gene, either a marker conferring resistance to G418, puromycin, hygromycin B, histidinol D or phleomycin, or GFP.

Vectors containing IRES elements from the picornaviridae family placed between the product gene and the selectable marker gene have been described (Pelletier et al, 1988; Jang et al, 1989; Davies et al, 1992).

GFP has also been successfully fused to resistance marker genes. For example, Bennett et al (1998) describe GFP/reomycin (zeomycin) fusion proteins. The dual function selectable marker is successfully used to identify and select transfected mammalian cells. In contrast Primig et al (1998) used a fusion protein of GFP and neomycin phosphotransferase to study its enhancer.

In the publication of Meng et al (2000) and in International patent application WO 01/04306, an expression system is used in which a gene of interest is expressed together with amplifiable selectable marker genes DHFR and GFP obtained from a single vector to select and identify cells that highly express recombinant proteins. These three genes are mixed in one transcription unit, or divided in two units. This spatial and transcriptional linkage of all three genes in a single expression vector allows for an increased probability of co-amplification under selection pressure and thus identification and selection of highly productive clones. The best clones expressing the protein of interest in the order of at most 3-4.5 pg/cell-day were isolated by using DHFR selectable marker amplified with recombinant selection reagents and FACS sorting on the basis of GFP. Here, experiments were carried out using adherent cells and in serum-containing medium, i.e. using well-known cells and conditions which are substantially stronger and which are characterized by a higher basal yield.

Summary of The Invention

It was therefore the task of the present invention to develop a selection system for recombinant cells with improved yield, which meets the following requirements:

(1) reducing the time spent in developing high-yielding cells to produce biopharmaceutical proteins while reducing development costs;

(2) selecting high producing cells at low capacity cost and at high output;

(3) the use of "fermentatively robust" high producing cells, which show less damage to growth, e.g. at increased methotrexate concentrations;

(4) transfecting, selecting and culturing suspension-adapted cells, preferably in serum-free medium;

(5) reducing the required gene amplification steps.

It is a further object of the present invention to provide expression vectors and host cells transfected therewith, which can be used in such a clonal selection system, as well as methods for producing heterologous gene products using such host cells.

According to one aspect of the invention, this task is solved by means of an expression vector for a gene encoding a protein of interest (hereinafter also referred to as "gene of interest") functionally linked to a hamster ubiquitin/S27 a promoter and to a gene encoding a fluorescent protein.

The expression vector preferably also contains an amplifiable selectable marker gene, such as the dihydrofolate reductase (DHFR) gene. Preferred expression vectors also contain other regulatory elements, such as enhancers, which are functionally linked to the promoter. Furthermore, the expression vector preferably also contains an Internal Ribosome Entry Site (IRES) which allows bicistronic expression of the gene encoding the fluorescent protein and the gene of interest.

The invention also relates to a basic vector which has a multiple cloning site for replacing a gene of interest in order to incorporate one of these genes, that is to say a sequence region having multiple restriction endonuclease cleavage sites.

Another aspect of the invention concerns host cells which have been transfected with one of the mentioned expression vectors. These are eukaryotic host cells, preferably mammalian cells, with rodent cells, such as hamster cells, and in particular CHO cells or BHK cells being particularly preferred.

In another aspect of the invention, a method of producing a heterologous gene product is provided, wherein a host cell transfected with an expression vector according to the invention is cultured under conditions permitting expression of the gene product, and the gene product is isolated from the culture or medium.

In a particular embodiment of the invention, the host cell is transfected, preferably co-transfected, with an expression vector according to the invention and additionally with one or more vectors carrying genes encoding one or more further proteins of interest.

In this connection, the invention provides a method of preparing a heterodimeric protein for utilization, wherein a host cell of this type, which has been co-transfected with expression vectors encoding different subunits of the heterodimeric protein, is cultured under conditions permitting expression of the heterodimeric protein, and the heterodimeric protein is isolated from the culture or medium. One particular application of such methods is in the production of antibodies and subunits thereof.

Another aspect of the invention relates to a method for selecting host cells expressing a protein of interest, wherein a population of host cells which have been transfected with an expression vector according to the invention is cultured under conditions allowing the expression of the protein of interest and a fluorescent protein, and the cell or cells showing the highest expression rate of the fluorescent protein is identified and selected. Selection is preferably performed using a fluorescence-activated cell sorter (FACS).

Surprisingly, it has been found that with a system prepared according to the invention it is possible to isolate in a minimum time a cell mass expressing an average specific yield of more than 15 picograms (single chain protein) or 10 picograms (humanized antibody) of recombinant protein per cell-day without the need for a gene amplification step. By a single DHFR-based gene amplification step, the specific productivity can be increased to more than 30 picograms per cell per day. The productivity thus achieved in the cell mass is higher by a factor of 8 to 10 than the maximum yield of the best cell clones disclosed hitherto.

Surprisingly, there is also a good correlation between the expression of the protein of interest and the expression of the fluorescent protein. This is even suitable for co-transfection, if, as in the case of antibody expression, both immunoglobulin chains are expressed separately by their own vectors and rely on their coupling to the fluorescent protein for transcription in FACS sorting, selection can be made for expression of only one chain. Nevertheless, the high expression rate of the fluorescent protein does not have any negative effect on the growth and viability of the cells. In addition, the development time for selecting high producing cells can be reduced by at least half compared to the conventional stepwise gene amplification strategy, resulting in a significant reduction in development efficiency and cost.

Drawings

FIG. 1 shows a comparison of the expression levels obtained with recombinant cell clones in which the heterologous gene product is expressed under the control of the CMV promoter, or under the control of the hamster ubiquitin/S27 a promoter. The two promoters are functionally linked to the CMV enhancer, the stop sequence, the BGH poly a, and are identical in each case. In the CMV¹In the case of (3), the expression vector is based on pcDNA3- (Invitrogen, Kalsruhe, DE), in CMV²It is a pBluescript-based expression vector (Stratagene, La Jolla, CA, US), while in the case of CHO it is a pAD-CMV-based expression vector (Werner et al, 1998). All expression vectors contain the amplifiable selectable marker dihydrofolate reductase (DHFR) by expression of lysosomal enzymes, and use aminopterin (MTX) to increase expression of heterologous genes by subsequent amplification steps. To express both chains of the antibody (Ab), co-transfection was performed with a second vector containing a neomycin-resistance gene as a selectable marker. The resulting titer or specific yield is provided relative to expression based on the CMV promoter, which is set to 1 (CMV)¹Is an enzyme, CMV²Is an antibody).

FIG. 2 shows a schematic preparation of the basic vector for expression of recombinant proteins in CHO-DG44 cells "P/E" is a combination of the CMV enhancer and the hamster ubiquitin/S27 a promoter, only "P" representing the promoter element and "T" being the transcriptional stop signal necessary for polyadenylation of the transcribed mRNA. The position and direction of transcription initiation within each transcription unit is indicated by arrows. For cloning of heterologous genes, a sequence region with multiple cleavage sites for endonuclease restriction enzymes is inserted after the promoter element (multiple cloning site-mcs). The amplifiable selectable marker dihydrofolate reductase is abbreviated as "dhfr" and the selectable marker neomycin phosphotransferase is abbreviated as "neo". The encephalomyocarditis virus-derived IRES element, which serves as an internal ribosome entry site within the dicistronic transcription unit, allows for the subsequent translation of the green fluorescent protein "GFP".

FIG. 3 shows a schematic preparation of eukaryotic expression vectors, each encoding a biopharmaceutical protein, and used to transfect CHO-DG44 cells. "P/E" is a combination of CMV enhancer and hamster-ubiquitin/S27 a promoter, only "P" represents the promoter element, and "T" is the transcriptional stop signal, which is necessary for polyadenylation of transcribed mRNA. The position and direction of transcription initiation within each transcription unit is indicated by arrows. The amplifiable selectable marker dihydrofolate reductase is abbreviated as "dhfr" and the selectable marker neomycin phosphotransferase is abbreviated as "neo". The encephalomyocarditis virus-derived IRES element, which serves as an internal ribosome entry site within the dicistronic transcription unit, allows for the subsequent translation of the green fluorescent protein "GFP". "sICAM" encodes a soluble intracellular adhesion molecule (US5,412,216), whereas "F19 HC" and "F19 LC" encode the heavy or light chain, respectively, of the humanized antibody F19 (EP 953639).

FIG. 4 shows the correlation between sICAM yield and GFP fluorescence, using cell pool ZB1 as an example. This collection of cells was obtained from transfection with the vector pBIDG-sICAM, where the therapeutic proteins sICAM and GFP were expressed together by a bicistronic transcription unit. The pools were subjected to serial GFP-based FACS sorting. After each sorting step (selection), the concentrations of sICAMs in the cell culture supernatants of the pools were determined by ELISA and the specific yields per cell per day (micrograms/cell day) were calculated. Each data point represents the average of at least three culture passages. A total of six picks were performed.

FIG. 5 shows the isolation of high expressing sICAM cells by GFP-based FACS sorting, using cell pool ZB1 as an example. This cell pool was obtained from transfection with the vector pBIDG-sICAM, where the therapeutic proteins sICAM and GFP were expressed together by a bicistronic transcription unit. Pools were subjected to serial GFP-based FACS sorting. After each selection, the concentration of sICAMs in the cell culture supernatants of the pools was determined by ELISA and the specific productivity per cell per day (micrograms/cell day) was calculated. Each data point represents the average of at least three culture passages. A total of six picks were performed.

FIG. 6 shows the increase in sICAM yield achieved by combining GFP-based selection with an MTX amplification step, using cell pool ZB1 as an example. This cell pool was obtained from transfection with the vector pBIDG-sICAM and subjected to sequential GFP-based FACS sorting. DHFR-mediated gene amplification was performed after the fourth selection or the sixth selection by adding Methotrexate (MTX) (5nM, 50nM, 500nM or 2. mu.M MTX) to the medium. The concentrations of sICAMs in the cell culture supernatants of the pools were determined by ELISA and the specific productivity per cell per day (micrograms/cell day) was calculated. Each data point represents the average of at least three culture passages.

Fig. 7 shows the survival pattern of cell aggregates following the addition of different high doses of methotrexate to the medium. Continuous GFP-based FACS sorting was performed on the cell pool ZB1 obtained by transfection with the vector pBIDG-sICAM (FIG. 3). DHFR-mediated gene amplification was performed after the fourth selection or the sixth selection by adding Methotrexate (MTX) to the medium. During the selection phase, cell number and survival were determined by staining with trypan blue and the culture (dic) was monitored over a number of days.

FIG. 8 shows the correlation between antibody yield (monoclonal antibody F19) and GFP fluorescence, using the cell pool ZB1 as an example. This cell pool was obtained from transfection using vector combinations pBIDG-F19HC and pBIN-F19LC (FIG. 3). The pools were subjected to serial GFP-based FACS sorting. After each selection, the concentration of antibody F19 in the cell culture supernatant of the pool was determined by ELISA and the specific yield per cell per day (micrograms/cell day) was calculated. Each data point represents the average of at least three culture passages. A total of six picks were performed.

FIG. 9 shows the isolation of a pool of highly expressed monoclonal antibody F19 cells by GFP-based selection using FACS, using cell pool ZB1 as an example. The cell pool obtained by transfection with the vectors pBIDG-F19HC and pBIN-F19LC (FIG. 3) was subjected to sequential GFP-based FACS sorting. After each selection, the concentration of antibody F19 in the cell culture supernatant of the pool was determined by ELISA and the specific yield per cell per day (micrograms/cell day) was calculated. Each data point represents the average of at least three culture passages.

Detailed description of the invention and preferred embodiments

The expression vector according to the present invention contains a gene encoding a protein of interest ("gene of interest") functionally linked to the hamster ubiquitin/S27 a promoter and to the gene encoding the fluorescent protein. The expression vector preferably also contains an amplifiable selectable marker gene.

Hamster ubiquitin/S27 a promoter

The hamster ubiquitin/S27 a promoter is a potent homologous promoter, described in WO 97/15664. Such promoters have preferably at least one of the following characteristics: a GC-rich sequence region, Sp1 binding site, polypyrimidine element, lacking a TATA box. Particularly preferred are promoters with an Sp1 binding site but lacking a TATA box. Furthermore preferred are constitutively active promoters, and in particular promoters which are equally active in serum-containing, low-serum and serum-free culture conditions. In another embodiment, it is an inducible promoter, in particular a promoter activated by removal of serum.

A particularly advantageous embodiment is a promoter having the nucleotide sequence contained in FIG. 5 of WO 97/15664. Particularly preferred in this case are promoters which contain the sequence from positions-161 to-45 in FIG. 5.

The promoters used in the examples of the present specification contain the sequences of the sequences having SEQ ID NOs: 1 from position 1923 to 2406. This sequence corresponds to the fragment-372 to +111 from FIG. 5 of WO 97/15664 and represents a preferred promoter, i.e.a preferred promoter should comprise this sequence region. Another suitable promoter fragment contains the sequence from positions 2134 to 2406 (corresponding to-161 to +111 in FIG. 5 of WO 97/15664). The promoter containing only the sequence at positions 2251 to 2406 is no longer functional (corresponding to positions-45 to +111 in FIG. 5 of WO 97/15664). It is possible to extend the promoter sequence in the 5' direction starting from position 2143.

It is also possible to use functional subfragments of the complete hamster ubiquitin/S27 a promoter sequence, as well as functional mutant variants of the complete sequence or subfragments thereof, which have been modified, for example, by substitutions, insertions or deletions. The corresponding subfragments, mutants or variants are also referred to hereinafter as "modified promoters".

Modified promoters, preferably in combination with other regulatory elements, preferably having transcriptional activity, which correspond to the sequences shown in SEQ ID NO: 1 from nucleotide sequence positions 1923 to 2406 of the promoter fragment WO 97/15664 from-372 to +111 of figure 5). Modified promoters were found to be useful in the sense of the present invention if they have transcriptional activity in a comparative reporter assay of at least 50%, preferably at least 80%, more preferably at least 90%, and even more preferably at least 100% of the 1923 to 2406 fragments (-372 to +111 fragments). Particularly preferred are modified promoters which are specific for the wild-type sequence of the hamster ubiquitin/S27 a promoter SEQ ID NO: 1 has a minimum sequence homology of at least 80%, preferably at least 85%, preferably at least 90%, even more preferably at least 95%, and especially preferably at least 97%, and has a corresponding promoter activity in a comparative reporter assay.

In a corresponding comparative reporter assay, the promoter fragment to be tested comprises reference sequences cloned in front of respective promoterless reporter genes encoding, for example, luciferase, secreted alkaline phosphatase or Green Fluorescent Protein (GFP). This construct (promoter sequence + reporter gene) is subsequently introduced into test cells, for example CHO-DG44, by transfection, and the reporter gene expression induced by the respective promoter fragment is determined by measuring the protein content of the reporter gene. Corresponding tests are also exemplified in, for example, Ausubel et al, Current Protocols in Molecular Biology, 1994, update.

Promoter sequences and modified promoters of the hamster ubiquitin/S27 a promoter, which may also include, for example, the 5' untranslated region or a fragment selected therefrom, and the coding region of the ubiquitin/S27 a gene or a fragment selected therefrom, can be used by the skilled worker from a knowledge of the sequences described in WO 97/15664, for example in Sambrook et al, 1989; obtained by various standard methods described in Ausubel et al, 1994. For example, one can start with the sequence described in WO 97/15664, select an appropriate fragment, and chemically synthesize an oligonucleotide probe containing the sequence of the fragment. For example, such probes may be used, for example, to clone the ubiquitin/S27 a gene or its 5' untranslated region or other fragment from a hamster genome library by hybridization. Using the reporter gene assay described above, one skilled in the art would be able to identify fragments having promoter activity without undue effort and for the purposes of the present invention. The 5' untranslated region or its specific fragment can be easily obtained by PCR amplification using the corresponding primer from genomic DNA or genomic library. Fragments of the 5' untranslated region can also be obtained from larger DNA fragments by restriction exonuclease III digestion. Such DNA molecules may also be synthesized chemically, or produced by ligation from chemically synthesized fragments.

Deletion, insertion and substitution mutants can be produced by "site-specific mutagenesis" and/or "PCR-based mutagenesis techniques". Comparable methods are mentioned, for example, in chapter 36.1 of Lottspeich and Zorbas 1998 and in further references.

By cross-hybridization using probes derived from the 5' untranslated region of the hamster ubiquitin/S27 a gene or from the S27a portion of the hamster ubiquitin/S27 a gene, it is also possible to identify and isolate appropriate promoter sequences from other, preferably mammalian, corresponding homologous genes. Corresponding techniques are described in e.g. Lottspeich and Zorbas 1998 chapter 23. For the purposes of the present invention, "homology" is a gene which is "homologous" if its nucleotide sequence exhibits at least 70%, preferably at least 80%, preferably at least 90%, more preferably at least 95%, and particularly preferably at least 97% identity to the nucleotide sequence of the gene.

Target gene

The gene of interest contained in the expression vector according to the present invention includes a nucleotide sequence of any length encoding a product of interest. The gene product or "product of interest" is typically a protein, polypeptide, peptide, or fragment or derivative thereof. However, it may also be RNA or antisense RNA. The gene of interest may exist in its full-length, shortened form, as a fusion gene or as a tagged gene. It may be genomic DNA or, preferably, cDNA or a corresponding fragment or fusion. The gene of interest may be a native gene sequence, or may be mutated or otherwise modified. Such modifications include codon optimization and humanization to accommodate established host cells. The gene of interest may encode a secreted, cytoplasmic, nuclear, membrane-bound or cell surface-bound polypeptide.

The term "nucleotide sequence" or "nucleic acid sequence" refers to oligonucleotides, nucleotides, polynucleotides and fragments thereof, as well as DNA or RNA of genomic or synthetic origin, which exist in single or double strands and may represent the coding or non-coding strand of a gene. Nucleic acid sequences can be modified using standard techniques, such as site-specific mutagenesis or PCR-mediated mutagenesis (e.g., as described in Sambrook et al, 1989 or Ausubel et al, 1994).

"encoding" means the property or property of a nucleotide of a particular sequence in a nucleic acid, e.g., a gene in a chromosome or mPNA, as a substrate in a biological process for the synthesis of other polymers and macromolecules, e.g., rRNA, tRNA, mRNA, other RNA molecules, cDNA or polypeptides. Thus, a gene encodes a protein if the desired protein is produced in a cell or other biological system by transcription and subsequent translation of mRNA. Both the coding strand, which is identical in nucleotide sequence to the mRNA sequence and is normally also provided in sequence databases, such as EMBL or GenBank, and the non-coding strand of the gene or cDNA as a transcription matrix, can be referred to as being used to encode a product or protein. Nucleic acids encoding proteins also include nucleic acids having a different nucleotide sequence order based on the degenerate genetic code, but which result in the same amino acid sequence of the protein. The nucleic acid sequence encoding the protein may also contain introns.

The term "cDNA" refers to deoxyribonucleic acid, which is prepared by reverse transcription and synthesis of a second DNA strand from mRNA or other RNA produced by a gene. If the cDNA is present as a double-stranded DNA molecule, it contains both coding and non-coding strands.

The term "intron" refers to a non-coding nucleotide sequence of any length. They occur naturally in many eukaryotic genes and are removed from previously transcribed mRNA precursors by a process known as splicing. This requires precise excision of the intron at the 5 'and 3' ends and proper ligation of the resulting mRNA ends to allow for the production of a mature processed mRNA with the correct reading frame for successful protein synthesis. Many splice donor and splice acceptor sites involved in this splicing process, i.e., sequences located directly adjacent to exon-intron or intron-exon, have been characterized. For a general overview, see Ohshima et al, 1987.

Protein of interest

Proteins/polypeptides of biopharmaceutical importance include, for example, but are not limited to, antibodies, enzymes, cytokines, lymphokines, adhesion molecules, receptors, and derivatives or fragments thereof. In general, all polypeptides that are agonists or antagonists and/or have therapeutic or diagnostic utility are of value.

The term "polypeptide" is used with respect to amino acid sequences or proteins, and means a polymer of amino acids of any length. The term also includes proteins that are modified post-translationally by reactions such as glycosylation, phosphorylation, acetylation, or protein processing. The structure of a polypeptide can be modified, for example, by substituting, deleting or inserting amino acids, and fusing with other proteins, while still retaining their biological activity.

Examples of therapeutic proteins are insulin, insulin-like growth factors, human growth hormone (hGH) and other growth factors, tissue plasminogen activator (tPA), Erythropoietin (EPO), cytokines, e.g., Interleukins (IL), such as IL-1, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-10, IL-11, IL-12, IL-13, IL-14, IL-15, IL-16, IL-17, IL-18, Interferon (IFN) - α, - β, - γ, - ω or- τ, Tumor Necrosis Factor (TNF), such as TNF- α, β or- γ, TRAIL, G-CSF, GM-CSF, M-CSF, MCP-1 and VEGF. Further examples are monoclonal, polyclonal, multispecific and single-chain antibodies, and fragments thereof, such as, for example, Fab ', F (ab')₂Fc and Fc' fragments, light (L) and heavy (H) immunoglobulin chains, and constant, variable or hypervariable regions thereof, as well as Fv and Fd fragments (Chamov et al, 1999). The antibody may be of human or non-human origin. Humanized and chimeric antibodies are also possibleIn (1).

Fab fragments (antigen binding fragments ═ Fab) consist of two chain variable regions, which are held together by adjacent constant regions. They may be produced, for example, by treatment with proteases, such as papain, from conventional antibodies or by DNA cloning. Other antibody fragments are F (ab')₂Fragments which can be produced by proteolytic digestion with pepsin.

It is also possible to prepare, by gene cloning, truncated antibody fragments consisting only of the variable regions of the heavy (VH) and light (VL). It is called Fv fragment (variable fragment-fragment of the variable part). Because in these Fv fragments it is not possible to covalently bind via the cysteine groups of the constant chains, Fv fragments are generally stabilized by some other means. For this purpose, the variable regions of the heavy and light chains are generally joined together by short peptide fragments of about 10 to 30 amino acids, particularly preferably 15 amino acids. This results in a single polypeptide chain in which VH and VL are linked together by a peptide linker. Such antibody fragments are also referred to as single chain Fv fragments (scFv). Examples of scFv antibodies are known and described, for example, in Huston et al, 1988.

In recent years, various strategies for the production of multimeric scFv derivatives have been developed. The intent is to produce recombinant antibodies with improved pharmacokinetic properties, increasing the avidity of the bound antibody. To achieve multimerization of scFv fragments (Multimerizing), they can be produced as fusion proteins with multimerization domains. The multimerization domain may be, for example, the CH3 region of IgG or a helix ("coiled coil"), such as a leucine-zipper domain. In another strategy, interactions between the VH and VL regions of scFv fragments are used to carry out multimerization (e.g., minibifunctional antibodies (diabodies), minitrifunctional antibodies (tribodies), and minipentafunctional antibodies (pentadies)).

The term "minibifunctional antibody" is used by the person skilled in the art to denote a bivalent homodimeric scFv derivative. Peptide linkers are shortened to 5-10 amino acids in scFv molecules, resulting in homodimers by overlapping VH/VL chains. The minibifunctional antibody may also be stabilized by inserting a disulfide bridge. Examples of minibifunctional antibodies can be found, for example, in Perisic et al, 1994.

The skilled person uses "miniantibodies" (minibodies), representing bivalent homodimeric scFv derivatives. It consists of a fusion protein which contains, as dimerization domain, the CH3 region of an immunoglobulin, preferably of IgG, particularly preferably of IgG 1. This is linked to the scFv fragment by a hinge region, which also belongs to IgG, and a linker region. Examples of such miniantibodies are described by Hu et al, 1996.

The skilled person uses "minitrifunctional antibodies", representing trivalent homotrimeric scFv derivatives (Kortt et al, 1997). Direct fusion of VH-VL does not require the use of linker sequences, resulting in trimer formation.

Fragments, which are referred to by the person skilled in the art as miniantibodies (mini antibodies), have a di-, tri-or tetravalent structure and are also derivatives of scFv fragments. Multimerization is achieved by the "coiled-coil" structure of a di-, tri-or tetramer (Pack et al, 1993 and 1995; Lovejoy et al, 1993).

Gene encoding fluorescent protein

The expression vector according to the present invention contains a gene encoding a fluorescent protein, is functionally linked to a gene of interest, and is under the control of a hamster ubiquitin/S27 a promoter, a modified hamster ubiquitin/S27 a promoter, or a homolog thereof.

The fluorescent protein may be, for example, a green, cyan, blue, yellow, or other colored fluorescent protein. One particular example is the Green Fluorescent Protein (GFP) obtained from aequorea victoria or jellyfish reniformis (Renilla reniformis), and mutants developed therefrom, see e.g. Bennet et al, 1998; chalfie et al, 1994; WO 01/04306 and the references cited therein.

Other fluorescent proteins and the genes encoding them are described in WO 00/34318, WO 00/34326, WO 00/34526 and WO 01/27150, incorporated herein by reference. These fluorescent proteins are fluorophores from non-bioluminescent organisms of the species coralloides (Anthozoa), such as, for example, mahalanobis (Anemonia majano), corallinum (Clavularia sp.), buttonhallium (Zoanthus sp.) I, buttonhallium II, sea anemone (Discosomastiata), Cornus edodes (Discosoma sp.) "Red", Cornus edodes "Green", Cornus edodes "magenta", and sea anemone rugosa (Anemonia sulcata).

In addition to wild-type proteins, the fluorescent proteins used according to the invention also comprise mutants and variants, which are produced naturally or by genetic techniques, fragments, derivatives thereof or variants which have been fused, for example, with other proteins or peptides. Mutations used, for example, may alter the excitation or emission spectrum, chromophore formation, extinction coefficient or stability of the protein. In addition, codon optimization can be used to improve expression in mammals or other species. According to the invention, it is also possible to use a fluorescent protein fused to a selectable marker, preferably an amplifiable selectable marker, such as dihydrofolate reductase (DHFR).

The fluorescence emitted by the fluorescent protein allows the protein to be detected, for example, by a flow cytometer with Fluorescence Activated Cell Sorter (FACS) or by fluorescence microscopy.

Other regulating elements

The hamster ubiquitin/S27 a promoter can be functionally associated with other regulatory sequences in order to increase/modulate transcriptional activity in the expression cassette.

For example, a promoter may be functionally linked to an enhancer sequence in order to increase transcriptional activity. To this end, one or more enhancers and/or several copies of an enhancer sequence may be used, such as the CMV or SV40 enhancer.

The term "enhancer" refers to a polynucleotide sequence that acts in cis position on the activity of a promoter and thus stimulates transcription of a gene functionally linked to the promoter. Unlike promoters, enhancers are not position and orientation dependent, and thus can be placed before or after the transcription unit, within an intron, or even within the coding region. Enhancers can be located in close proximity to the transcription unit and at a considerable distance from the promoter. Physical and functional overlap with the promoter is also possible. Those skilled in the art know that many enhancers from various sources (and deposited in databases, such as GenBank, for example SV40 enhancer, CMV enhancer, polyoma enhancer, adenovirus enhancer) are available as separate elements or as elements cloned into polynucleotide sequences (for example deposited at ATCC, available from commercial and industrial sources). Many promoter sequences also contain enhancer sequences, such as the frequently used CMV promoter. The human CMV enhancer is one of the strongest enhancers identified to date. An example of an inducible enhancer is the metallothionein enhancer, which can be stimulated by glucocorticoids or heavy metals.

Other possible modifications are, for example, the introduction of multiple Sp1 binding sites. The promoter sequence may also be associated with regulatory sequences which allow the control/regulation of the transcriptional activity. Thus, promoters can be made repressible/inducible. For example, by linking it to sequences of binding sites of positively-or negatively-regulated transcription factors. The transcription factor Sp1 mentioned above, for example, has a positive effect on the transcription activity. Another example is the binding site of the activator protein AP-1, which can act positively and negatively on transcription. AP-1 activity can be controlled by a wide variety of factors, such as growth factors, cytokines, and serum (Faisst et al, 1992, and references therein). It is also possible to increase the transcription efficiency by altering the promoter sequence through a mutation (substitution, insertion or deletion) of one, two, three or more bases and then determining whether this increases the promoter activity in a reporter gene assay.

Basically, additional regulatory elements include promoters other than the hamster ubiquitin/S27 a promoter, enhancers, termination and polyadenylation signals, and other expression control elements. Inducible and constitutive regulatory sequences are known for various cell types. "transcriptional regulatory elements" typically include a promoter upstream of the gene sequence to be expressed, transcription initiation and termination sites, and polyadenylation signals.

The term "promoter" denotes a polynucleotide sequence which allows and controls the transcription of a gene or sequence to which it is functionally linked. The promoter contains a recognition sequence for RNA polymerase binding, and a transcription initiation site (transcription initiation site). In order to express the desired sequence in certain cell types or host cells, an appropriate functional promoter must be selected. Those skilled in the art are familiar with a variety of promoters from a variety of sources, including constitutive, inducible, and repressible promoters. They are deposited in databases, such as GenBank, and are available as individual elements, or as elements cloned into polynucleotide sequences obtained from commercial or industrial sources. In inducible promoters, the activity of the promoter may be decreased or increased in response to a signal. An example of an inducible promoter is the tetracycline (tet) promoter. The promoter contains the tetracycline operator sequence (tetO), which is inducible by the tetracycline-regulated transactivator protein (tTA). When tetracycline is present, the binding of tTA to tetO is inhibited. Examples of other inducible promoters are jun, fos, metallothionein and heat shock promoters (see also Sambrook et al, 1989; Gossen et al, 1994). Promoters particularly suitable for high expression in eukaryotes are, for example, the SV40 early promoter, the adenovirus major late promoter, the mouse metallothionein-I promoter, the long terminal repeat of the Rous sarcoma virus, and the early promoter of the human cytomegalovirus. Examples of other heterologous mammalian promoters are actin, immunoglobulin or heat shock promoters.

The term "transcription start site" means a nucleic acid in a construct that corresponds to the first nucleic acid to be incorporated into the main transcript, i.e., the mRNA precursor. The transcription initiation site may overlap with the promoter sequence.

The term "transcription termination site" means a nucleotide sequence which is normally at the 3' end of the gene of interest or gene segment to be transcribed and which causes termination of transcription by RNA polymerase.

A "polyadenylation signal" is a signal sequence which causes cleavage at a specific site at the 3 'end of eukaryotic mRNA and incorporates approximately 100-200 adenine nucleotides (poly A tail) at the cleaved 3' end after transcription. The polyadenylation signal comprises the sequence AATAAA, located approximately 10-30 nucleotides upstream of the cleavage site, and the sequence located downstream. Various polyadenylation elements are known, such as tk polyadenylation, SV40 late and early polyadenylation, or BGH polyadenylation (described in e.g. US5,122,458).

In a preferred embodiment of the invention, each transcription unit has a promoter or promoter/enhancer element, a gene of interest and/or a marker gene, and a transcription termination element. In another preferred embodiment, the transcription unit further comprises a translation regulatory unit.

"translational regulatory elements" include, for each polypeptide to be expressed, the translation initiation site (AUG), the stop codon, and the polyadenylation signal. For optimal expression, it may be advisable to remove, add or alter the 5 '-and/or 3' -untranslated regions of the nucleic acid sequence to be expressed, in order to exclude any additional translation initiation codons that may be inappropriate, or other sequences that may affect the level of transcription or expression. To facilitate expression, a ribosome consensus binding site may be inserted separately immediately upstream of the start codon. To produce a secreted polypeptide, the gene of interest typically contains a signal sequence that encodes a signal precursor peptide, which transports the polypeptide that has been synthesized to and through the ER membrane. The signal sequence is usually, but not always, located at the amino terminus of the secretory protein and can be cleaved by a signal peptidase after the protein has been inserted into the ER membrane. A gene sequence typically, but not necessarily, contains its own signal sequence. If the native signal sequence is not present, a heterologous signal sequence can be introduced in a known manner. Many signal sequences of this species are known to those skilled in the art and are deposited in sequence databases, such as GenBank and EMBL.

A particularly important regulatory element according to the present invention is the Internal Ribosome Entry Site (IRES). The IRES element comprises a sequence that functionally activates translation initiation independent of the upstream 5' -terminal methylguanidine (guanosinium) CAP (CAP structure) of the gene and allows translation of two cistrons from a single transcript (open reading frame) in animal cells. The IRES element provides an independent ribosome entry site for translation of the open reading frame immediately downstream. In contrast to bacterial mRNA, which may be polycistronic, i.e., it may encode more different polypeptides or products, most mRNA from animal cells are monocistronic and encode only one protein or product by translation of the mRNA one after the other. In the case of polycistronic transcripts in eukaryotic cells, translation will start from the upstream closest translation start site and will be stopped by the first stop codon, followed by release of the transcript from the ribosome. Thus, only the first polypeptide or product encoded by the mRNA will be produced during translation. In contrast, polycistronic transcripts having an IRES element functionally linked to a second or subsequent open reading frame in the transcript, allowing for subsequent translation of the open reading frame coding construct located downstream thereof, to produce two or more polypeptides or products encoded by the same transcript in eukaryotic cells.

IRES elements can be of various lengths and of various origins, and can originate, for example, from encephalomyocarditis virus (EMCV) or other picornaviruses. Various IRES sequences and their use in constructing vectors are described in the literature, see e.g., Pelletier et al, 1988; jang et al, 1989; davies et al, 1992; adam et al, 1991; morgan et al, 1992; sugimoto et al, 1994; ramesh et al, 1996; mosser et al, 1997.

The gene sequence located downstream is functionally linked to the 3' end of the IRES element, i.e. the separation distance is chosen such that the expression of the gene is not or only slightly affected, or is sufficiently expressed for the intended purpose. The optimal permissive distance for sufficient expression between the IRES element and the start codon of the gene downstream thereof can be determined by simple experimentation by varying the spacing and determining the rate of expression as a function of the spacing using reporter gene measurements.

By means of said measurements it is possible to obtain optimal expression cassettes which are valuable for the expression of heterologous gene products. Accordingly, expression cassettes obtained by one or more of these measurements are also a further object of the invention.

Amplifiable selectable marker gene

Preferred vectors according to the invention additionally contain an amplifiable selectable marker gene which allows amplification of the amplifiable marker gene and preferably co-amplification of a transcription unit consisting of the hamster ubiquitin/S27 a gene, the gene of interest and the gene for the fluorescent protein. In this regard, host cells transfected with the corresponding expression vector are cultured in the presence of an appropriate selection agent such that only those host cells having many gene copies of at least the amplifiable selectable marker gene are capable of replication (being replicated). This is preferably achieved by stepwise culturing of the cells in the presence of increasing amounts of the selection agent.

Amplifiable selectable marker genes typically encode enzymes required for eukaryotic cells to grow under certain culture conditions. For example, the amplifiable selectable marker gene may encode dihydrofolate reductase (DHFR). In this case, if the host cell transfected therewith is cultured in the presence of the selection agent Methotrexate (MTX), the gene is amplified.

Table 1 below provides examples of other amplifiable selectable marker genes and associated selection agents that may be used in accordance with the present invention, and is described in Kaufman, "methods of enzymology", 185: 537-.

Table 1: amplifiable selectable marker gene

Amplifiable selectable marker gene	Deposit number	Selection agent
Amplifiable selectable marker gene	Deposit number	Selection agent	Dihydrofolate reductase	M19869 (hamster) E00236 (mouse)	Methotrexate (MTX)
Metallothionein	D10551 (hamster) M13003 (human) M11794 (rat)	Cadmium (Cd)	Dihydrofolate reductase	M19869 (hamster) E00236 (mouse)	Methotrexate (MTX)
Metallothionein	D10551 (hamster) M13003 (human) M11794 (rat)	Cadmium (Cd)	CAD (carbamyl phosphate synthase: aspartate transcarbamylase: dihydroorotase)	M23652 (hamster) D78586 (human)	N-phosphoric acid acetyl-L-aspartic acid
Adenosine deaminase	K02567 (human) M10319 (mouse)	Xyl-A-or adenosine, 2' -deoxysynemetic		M23652 (hamster) D78586 (human)	N-phosphoric acid acetyl-L-aspartic acid
Adenosine deaminase	K02567 (human) M10319 (mouse)	Xyl-A-or adenosine, 2' -deoxysynemetic	AMP (adenylate) -deaminase	D12775 (human) J02811 (rat)	Adenine, azaserine, synbiotics
UMP-synthetase	J03626 (human)	6-azauridine, pyrazole furan	AMP (adenylate) -deaminase	D12775 (human) J02811 (rat)	Adenine, azaserine, synbiotics
UMP-synthetase	J03626 (human)	6-azauridine, pyrazole furan	IMP 5' -dehydrogenase	J04209 (hamster) J04208 (human) M33934 (mouse)	Mycophenolic acid
Xanthine-guanine-phosphoribosyl transferase	X00221 (Escherichia coli)	Mycophenolic acid with limited xanthines	IMP 5' -dehydrogenase	J04209 (hamster) J04208 (human) M33934 (mouse)	Mycophenolic acid
Xanthine-guanine-phosphoribosyl transferase	X00221 (Escherichia coli)	Mycophenolic acid with limited xanthines	Mutated HGPRT enzymes or mutated thymidine-kinases	J00060 (hamster) M13542, K02581 (human)	Hypoxanthine, aminopterin and thymidine (HAT)

	J00423, M68489 (mouse) M63983 (rat) M36160 (herpes virus)
	J00423, M68489 (mouse) M63983 (rat) M36160 (herpes virus)		Thymidylate-synthetase	D00596 (human) M13019 (mouse) L12138 (rat)	5-Fluorodeoxyuridine
P-glycoprotein 170(MDR1)	AF016535 (human) J03398 (mouse)	Several kinds of medicinesFor example doxorubicin, vincristine, colchicine	Thymidylate-synthetase	D00596 (human) M13019 (mouse) L12138 (rat)	5-Fluorodeoxyuridine
P-glycoprotein 170(MDR1)	AF016535 (human) J03398 (mouse)		Ribonucleotide reductase	M124223, K02927 (mouse)	Aphidicolin
Glutamine-synthetase	AF150961 (hamster) U09114, M60803 (mouse) M29579 (rat)	Methionine sulfoxide amine (Methioninseulfoximin) (MSX)	Ribonucleotide reductase	M124223, K02927 (mouse)	Aphidicolin
Glutamine-synthetase	AF150961 (hamster) U09114, M60803 (mouse) M29579 (rat)	Methionine sulfoxide amine (Methioninseulfoximin) (MSX)	Asparagine-synthetase	M27838 (hamster) M27396 (human) U38940 (mouse) U07202 (rat)	Beta-aspartyl hydroxamic acid, Albizidine, 5' azacytidine
Argininosuccinate synthetase	X01630 (human) M31690 (mouse) M26198 (cattle)	Canavalia gladiata nitric acid	Asparagine-synthetase	M27838 (hamster) M27396 (human) U38940 (mouse) U07202 (rat)	Beta-aspartyl hydroxamic acid, Albizidine, 5' azacytidine
Argininosuccinate synthetase	X01630 (human) M31690 (mouse) M26198 (cattle)	Canavalia gladiata nitric acid	Ornithine-decarboxylase	M34158 (human) J03733 (mouse) M16982 (rat)	Alpha-difluoromethyl ornithine
HMG-CoA-reductase	L00183, M12705 (hamster) M11058 (human)	Compactin	Ornithine-decarboxylase	M34158 (human) J03733 (mouse) M16982 (rat)	Alpha-difluoromethyl ornithine
HMG-CoA-reductase	L00183, M12705 (hamster) M11058 (human)	Compactin	N-acetyl glucosaminyl-transferase	M55621 (human)	Tunicamycin

Sumoyl-tRNA-synthetases	M63180 (human)	Borrelidin
Sumoyl-tRNA-synthetases	M63180 (human)	Borrelidin	Na⁺K⁺-ATPase	J05096 (human) M14511 (rat)	Ouabain

According to the invention, the amplifiable selectable marker gene is preferably a gene which codes for a polypeptide having the function of DHFR, for example DHFR or a fusion protein from a fluorescent protein and DHFR. DHFR is essential for purine biosynthesis. Cells lacking the DHFR gene cannot grow in purine-deficient media. DHFR is therefore a useful selectable marker for selecting and amplifying genes in cells cultured in purine-free medium. The selection medium used with the DHFR gene is Methotrexate (MTX). Accordingly, the present invention includes a method for producing a high-producing recombinant host cell comprising the steps of: (i) transfecting a host cell with a gene encoding at least one protein of interest, a fluorescent protein and DHFR; (ii) (ii) culturing the cells under conditions permitting expression of the various genes, and (iii) amplifying the co-integrated genes by culturing the cells in the presence of a selective agent, such as methotrexate, that permits amplification of at least one amplifiable selectable marker gene. It is preferred to culture transfected cells in hypoxanthine/thymidine free medium in the absence of serum, with increasing concentrations of MTX added. In the first amplification step, the concentration of MTX is preferably at least 200nM, and in a more preferred embodiment at least 500nM, and can be increased stepwise up to 1. mu.M. In individual cases, concentrations in excess of 1 μ M may be used.

Mammalian cells, preferably mouse myeloma and hamster cells, are preferred host cells for use of DHFR as an amplifiable selectable marker. The cell lines CHO-DUKX (ATCC CRL-9096) and CHO-DG44(Urlaub et al, 1983) are particularly preferred since they have no DHFR activity per se as a result of the mutation. In order to be able to use DHFR-dependent amplification in other cell types with their own endogenous DHFR activity, it is possible to use mutated DHFR genes encoding proteins with reduced sensitivity to methotrexate during transfection (Simonson et al, 1983; Wigler et al, 1980; Haber et al, 1982).

Preparation of an expression vector according to the invention

In principle, the expression vectors according to the invention can be prepared by conventional methods customary to those skilled in the art, as described, for example, by Sambrook et al (1989). Functional components of the vector are also described, such as appropriate promoters (in addition to the hamster ubiquitin/S27 a promoter), enhancers, stop and polyadenylation signals, antibiotic resistance genes, selectable markers, origins of replication and splice signals. They can be prepared using conventional cloning vectors, such as plasmids, phages, phagemids, cosmids or viral vectors, such as baculovirus, retrovirus, adenovirus, adeno-associated virus and herpes simplex virus, and artificial chromosomes/minichromosomes. Eukaryotic expression vectors, which also typically contain prokaryotic sequences such as origins of replication and antibiotic resistance genes, allow the vector to replicate and select in bacteria. Many eukaryotic expression vectors containing multiple cloning sites for introducing polynucleotide sequences are known, and some are available from various companies, such as Stratagene, La Jolla, CA, USA; invitrogen, Carlsbad, CA, USA; promega, Madison, Wis., USA or BD Biosciences Clontech, Palo Alto, Calif., USA.

The hamster ubiquitin/S27 a promoter, the gene of interest, the gene encoding the fluorescent protein, preferably also an amplifiable selectable marker gene, for example dihydrofolate reductase, and optionally additional regulatory elements, such as an Internal Ribosome Entry Site (IRES), enhancer or polyadenylation signal, are introduced into the expression vector in one of the ways familiar to the person skilled in the art. The expression vector according to the present invention contains at a minimum the ubiquitin/S27 a promoter, the gene of interest, and the gene encoding the fluorescent protein. The expression vector preferably also contains an amplifiable selectable marker gene. According to the present invention, also modified ubiquitin/S27 a promoters, such as the modified ubiquitin/S27 a promoter described in the present invention, can be used. Particularly preferred are expression vectors wherein the ubiquitin promoter, the gene of interest and the gene encoding the fluorescent protein are functionally linked together or in functional linkage.

Within the context of this specification, the term "functionally linked" or "functionally linked" means two or more nucleic acid sequences or partial sequences which are positioned such that they perform their intended function. For example, a promoter/enhancer is functionally linked to a coding gene sequence if it can control or regulate the transcription of the linked gene sequence in an antegrade position. Typically, but not necessarily, functionally linked DNA sequences are in close proximity and, if two coding gene sequences are linked, or in the case of a secretory signal sequence, in the same reading frame. Although the functionally linked promoter is usually located upstream of the coding gene sequence, it does not necessarily have to be in close proximity thereto. Enhancers also need not be in close proximity, so long as they facilitate transcription of the gene sequence. For this purpose, they can be upstream and downstream of the gene sequence, optionally at a distance therefrom. The polyadenylation site is functionally linked to the gene sequence and, if it is located 3' to the gene sequence, allows transcription via the coding sequence to the polyadenylation signal. The ligation may be performed according to conventional recombination methods, for example by PCR techniques, by ligation at appropriate restriction cleavage sites, or by splicing. If no suitable cleavage sites are available, synthetic oligonucleotide linkers or adapters may be used in a known manner. According to the invention, it is preferred that the functional linkage is not via an intron sequence.

In one of the embodiments described, the ubiquitin/S27 a promoter or modified form thereof, the gene of interest, and the gene encoding the fluorescent protein are functionally linked together. This means, for example, that the gene of interest and the gene encoding the fluorescent protein are expressed starting from the same ubiquitin/S27 a promoter or a modified form thereof. In a particularly preferred embodiment, this functional linkage is performed via an IRES element, allowing the synthesis of a bicistronic mRNA from two genes. The expression vectors according to the invention may additionally contain enhancer elements which function functionally with respect to one or more promoters. Particularly preferred are expression vectors in which the ubiquitin/S27 a promoter or modified form thereof is linked to an enhancer element, such as the SV40 enhancer or CMV enhancer element.

Basically, the expression of a gene in an expression vector can be carried out starting from one or more transcription units. A transcription unit is defined as a region containing one or more genes to be transcribed. The genes within a transcription unit are functionally linked to each other such that all genes within such unit are under the transcriptional control of the same promoter or promoter/enhancer. The result of the transcriptional ligation of the gene is that more than one protein or product can be transcribed from a single transcriptional unit and expressed thereby. Each transcription unit may contain the regulatory elements necessary for the transcription and translation of the gene sequences contained therein. Each transcription unit may contain the same or different regulatory elements. Functional linkage of genes in a transcription unit can be performed using IRES elements or introns.

The expression vector may contain a single transcription unit for expressing the gene of interest, the fluorescent protein, and the amplifiable selectable marker gene. Alternatively, the genes may be arranged in two or more transcription units. Various combinations of genes in the transcription unit are possible here. In another embodiment of the invention, more than one expression vector consisting of one, two or more transcription units can be introduced into the host cell by co-transfection, or in sequential transfections in any order. Any combination of regulatory elements and genes on each vector may be selected so long as sufficient expression of the transcription unit is ensured. If desired, additional regulatory elements and genes, such as additional genes of interest or selectable markers, can be placed on the expression vector.

Thus, the expression vector according to the invention may contain, in one or in two separate transcription units, a gene encoding a fluorescent protein and an amplifiable selectable marker gene. Each transcription unit can transcribe and express one or more gene products. If two genes are contained in a single transcription unit, they are under the control of the same promoter or promoter/enhancer, and an IRES is preferably used to ensure functional linkage of all components. If the gene encoding the fluorescent protein and the amplifiable selectable marker gene are contained in two separate transcription units, they may be under the control of the same or different promoters/enhancers. However, for the selectable marker gene, it is preferred to use its native or weaker heterologous promoter, such as the SV40 early promoter, and preferably no enhancer. In the context of the present invention, expression vectors with two separate transcription units are preferred. One (bicistronic) transcription unit contains the gene of interest and the gene encoding the fluorescent protein, while the other transcription unit contains the amplifiable selectable marker gene. Preferably, each transcription unit is restricted at the 3' end by a sequence encoding a poly A signal, preferably tk poly A, BGH poly A or SV40 poly A.

Also preferred according to the invention are those vectors in which a desired gene is replaced by a multiple cloning site which allows cloning of the desired gene via a recognition sequence for a restriction endonuclease. From the prior art, numerous recognition sequences for all kinds of restriction endonucleases are known, as well as the restriction endonucleases belonging thereto. It is preferable to use a sequence consisting of at least six nucleotides as a recognition sequence. A list of suitable recognition sequences can be found, for example, in Sambrook et al, 1989.

Host cell

For transfection with the expression vectors according to the invention, eukaryotic host cells, preferably mammalian cells, and more particularly rodent cells, such as mouse, rat and hamster cell lines, are used. Transfection of corresponding cells with the expression vectors according to the invention results in transformed, genetically modified, recombinant or transgenic cells, which are also an object of the invention.

Preferred host cells within the scope of the present invention are hamster cells, such as BHK21, BHK TK^-CHO, CHO-K1, CHO-DUKX B1 and CHO-DG44 cells, or derivatives/progeny of these cell lines. Particularly preferred are CHO-DG44, CHO-DUKX, CHO-K1 and BHK21 cells, in particular CHO-DG44 and CHO-DUKX cells. Mouse myeloma cells are also suitable, with NS0 and Sp2/0 cells being preferred, as well as derivatives/progeny of these cell lines.

Examples of hamster and mouse cells that can be used according to the invention are provided in table 2 below. However, derivatives and progeny of these cells, other mammalian cells, including but not limited to human, mouse, rat, simian, rodent cell lines, or eukaryotic cells, including but not limited to yeast, insect, and plant cells, may also be used as host cells for the production of biopharmaceutical proteins.

Table 2: hamster and mouse producer cell lines

Cell lines	Accession number
Cell lines	Accession number	NS0	ECASS No. 85110503
Sp2/0-Ag14	ATCC CRL-1581	NS0	ECASS No. 85110503
Sp2/0-Ag14	ATCC CRL-1581	BHK21	ATCC CCL-10
BHKTK^-	ECACC No. 85011423	BHK21	ATCC CCL-10
BHKTK^-	ECACC No. 85011423	HaK	ATCC CCL-15

2254-62.2 (BHK-21-derivatives)	ATCC CRL-8544
2254-62.2 (BHK-21-derivatives)	ATCC CRL-8544	CHO	ECACC No. 8505302
CHO-K1	ATCC CCL-61	CHO	ECACC No. 8505302
CHO-K1	ATCC CCL-61	CHO-DUKX(＝CHOduk^-，CHO/dhfr^-)	ATCC CRL-9096
CHO-DUKX B1	ATCC CRL-9010	CHO-DUKX(＝CHOduk^-，CHO/dhfr^-)	ATCC CRL-9096
CHO-DUKX B1	ATCC CRL-9010	CHO-DG44	Urlaub et al; cell 33[2 ]]，405-412，1983
CHO Pro-5	ATCC CRL-1781	CHO-DG44	Urlaub et al; cell 33[2 ]]，405-412，1983
CHO Pro-5	ATCC CRL-1781	V79	ATCC CCC-93
B14AF28-G3	ATCC CCL-14	V79	ATCC CCC-93
B14AF28-G3	ATCC CCL-14	CHL	ECACC No. 87111906

Transfection of eukaryotic host cells with a polynucleotide or one of the expression vectors according to the invention is carried out by conventional methods (Sambrook et al, 1989; Ausubel et al, 1994). Suitable methods for transfection include, for example, lipoid-mediated transfection, calcium phosphate co-precipitation, electroporation, polycation (e.g., DEAE dextran) -mediated transfection, protoplast fusion, microinjection, and viral infection. According to the invention, stable transfection is preferably carried out, wherein the construct is integrated into the genome of the host cell, or into the artificial chromosome/minichromosome, or is contained in the host cell in episomal form in a stable manner. Transfection methods which provide the best transfection frequency and expression of the heterologous gene in the respective host cell are preferred here. By definition, each sequence or each gene inserted into a host cell is referred to as a "heterologous sequence" or "heterologous gene" with respect to the host cell. Even if the sequence to be introduced or the gene to be introduced is identical to the endogenous sequence or endogenous gene of the host cell, it is applicable. For example, the hamster actin gene introduced into hamster host cells is also a heterologous gene by definition.

In the recombinant preparation of heterodimeric proteins, such as monoclonal antibodies (mAbs), transfection of appropriate host cells can in principle be carried out by two different methods. Monoclonal antibodies of this class are composed of many subunits, heavy and light chains. The genes encoding these subunits can be placed in separate or polycistronic transcription units on a single plasmid, which is then used to transfect host cells. This attempts to ensure stoichiometric expression of the gene after integration into the host cell genome. However, in the case of independent transcription units, it must be ensured thereby that the mrnas encoding the different proteins exhibit the same stability, as well as the transcription and translation efficiency. In the second case, the expression of the gene in the polycistronic transcription unit is carried out by a single promoter, and only one transcript is formed. By using IRES elements, correct and efficient internal translation initiation of genes in the second and subsequent cistrons is obtained. However, the expression rate of these cistrons is lower than that of the first cistrons, which is substantially more efficient than IRES-dependent translation initiation by so-called "cap" -dependent translation initiation of the pre-initiation complex. In order to achieve true equimolar expression of the cistrons, additional intercistronic elements may for example be introduced which together with the IRES element lead to a uniform expression rate (WO 94/05785).

A further and preferred possibility for the simultaneous preparation of a plurality of heterologous proteins according to the invention is co-transfection, in which genes are integrated in separate expression vectors. This has the advantage that certain ratios of genes and gene products can be adjusted to each other, thereby balancing the differences in mRNA stability, as well as in transcription and translation efficiency. In addition, the expression vectors are more stable because of their smaller size and easier to manipulate on cloning and transfection.

Thus, in a particular embodiment of the invention, the host cell is additionally transfected, preferably co-transfected, with one or more vectors having a gene encoding one or more other proteins of interest. The other vector or vectors used for co-transfection encode, for example, another protein or proteins of interest, and at least one other selectable marker, such as neomycin phosphotransferase, under the control of the same promoter/enhancer combination.

According to the invention, adaptation and culture of the host cells is preferably established in the absence of serum, optionally in a medium free of animal proteins/peptides. Examples of commercially available media include Ham ' S F12(Sigma, Deisenhofen, DE), RPMI-1640(Sigma), Dulbecco ' S modified Eagle ' S Medium (DMEM; Sigma) minimal essential Medium (MEM; Sigma), eskoff (Iscove ' S) modified Dulbecco ' S Medium (IMDM; Sigma), CD-CHO (Invitrogen. Carlsbad, CA, USA), CHO-S-SFMII (Invitrogen), serum-free CHO-medium (Sigma), and protein-free CHO-medium (Sigma). Each of these media is optionally supplemented with various compounds, such as hormones and/or other growth factors (e.g., insulin, transferrin, epidermal growth factor, insulin-like growth factor), salts (e.g., sodium chloride, calcium, magnesium, phosphate), buffered solutions (e.g., HEPES), nucleosides (e.g., adenosine, thymidine), glutamine, glucose or other equivalent nutrients, antibiotics, and/or trace elements. Although a serum-free medium is preferred according to the invention, it is also possible to use a medium which has been mixed with a suitable amount of serum to culture the host cells. To select genetically modified cells expressing one or more selectable marker genes, one or more selection agents are added to the medium.

The term "selective agent" means a substance that interferes with the growth and survival of a host cell lacking the selectable marker gene in question. For example, in order to select for the presence of an expressed antibiotic resistance gene, such as neomycin phosphotransferase, it is preferred to use the antibiotic Geneticin (G418) as a medium supplement. If the gene used is an amplifiable selectable marker gene (see Table 1), the selection agent may also be one that induces amplification of the selectable marker gene. For example, methotrexate is a selection medium suitable for use in the amplification of the DHFR gene. Examples of other amplification-inducing selection agents are listed in table 1.

A "selectable marker gene" is a gene which allows the specific selection of cells containing the gene by adding the corresponding selection agent to the medium. More specifically, antibiotic resistance genes can be used as positive selection markers. Only cells which have been transformed with this gene are able to grow in the presence of the corresponding antibiotic and are therefore selected. In contrast, untransformed cells are unable to grow or survive under these selection conditions. There are positive, negative and bifunctional selectable markers. Positive selection markers allow for selection and enrichment of transformed cells by resistance to a selection agent, or by compensating for metabolic or catabolic defects in the host cell. Conversely, cells that have received the selectable marker gene can be selectively cleared by a negative selectable marker. An example of this is the thymidine kinase gene of herpes simplex virus, whose expression in cells results in the elimination of the cell by the simultaneous addition of acyclovir (acyclovir) and 9- [1, 3-dihydroxy-2-propoxymethyl ] guanine (ganciclovir). Selectable markers, including amplifiable selectable markers, including genetically modified mutants and variants, fragments, functional equivalents, derivatives, homologs, and fusions to other proteins or peptides, are used in the invention so long as the selectable marker retains its selectable properties. Such derivatives exhibit considerable homology in the amino acid sequence, in regions or domains which are considered to be selective. The literature describes a number of selectable marker genes, including bifunctional (positive/negative) markers (see, e.g., WO 92/08796 and WO 94/28143). Examples of selectable markers often used in eukaryotic cells include the genes for Aminoglycoside Phosphotransferase (APH), hygromycin phosphotransferase (HYG), dihydrofolate reductase (DHFR), Thymidine Kinase (TK), glutamine synthetase, asparagine synthetase, and genes that mediate resistance to neomycin (G418), puromycin, histidinol D, bleomycin, phleomycin, and lipoxin (zeocin).

It is also possible to select transformed cells by Fluorescence Activated Cell Sorting (FACS). In this regard, transformed cells may be selected, for example, using bacterial beta-galactosidase, cell surface markers, or fluorescent proteins, such as Green Fluorescent Protein (GFP) and variants thereof from Victoria and nephroleph or other species, red fluorescent protein and proteins that fluoresce in other colors, and variants thereof from non-bioluminescent organisms such as Corallocta shiitake, Helianthus rugosa, Corallocta yunnanensis, and Helichrysum nikosum.

In the present invention, it is preferred to use the DHFR gene for selecting genetically modified (recombinant) host cells as an amplifiable selectable marker gene. This marker is particularly suitable for selection and subsequent expansion when using DHFR-negative basal cells, such as CHO-DG44 or CHO-DUKX, since these cells do not express endogenous DHFR and therefore do not grow in purine-free medium. As a result, the DHFR gene herein can be used as a dominant selection marker and transformed cells can be selected in media that does not contain hypoxanthine/thymidine. To obtain DHFR mediated gene amplification, Methotrexate (MXT) was used. By adding MTX, the growth characteristics are decisively influenced. Substantial deterioration in fermentation robustness of the cells was observed with increasing MTX concentration and amplification steps. However, surprisingly it has been found that with the clonal selection system according to the present invention, recombinant host cells that exhibit much more robust behavior are increased in the presence of high concentrations of MTX (see figure 7). Thus, host cells that have been identified and selected using fluorescence-activated cell sorting (FACS) can be cultured and expanded in the presence of 500nM MTX, preferably 1. mu.M MTX, resulting in a significant increase in yield.

Thus, a method for selecting high-producing host cells is considered to be particularly according to the invention if host cells transfected with an expression vector according to the invention and expressing at least the gene of interest, the fluorescent protein and the DHFR gene are selected by FACS sorting and subjected to a gene amplification step in the presence of at least 500nM, preferably 1. mu.M MTX.

The term "fermentation robustness" means the growth characteristics of the cells, such as maintaining a certain growth rate, robustness to "Upscaling" (the larger size of the bioreactor), and achieving high cell counts and viability in germ line maintenance in order to meet industrial passage rates when scaled up.

Expression of

The term expression relates to the transcription and/or translation of a heterologous gene sequence in a host cell. The rate of expression is generally determined based on the amount of the corresponding mRNA present in the host cell, or based on the amount of gene product produced that is encoded by the gene of interest. The amount of mRNA produced by transcription of the selected nucleotide sequence is determined, for example, by northern blot hybridization, RNase-RNA-protection, in situ hybridization of cellular RNA or by PCR (Sambrook et al, 1989; Ausubel et al, 1994). The protein encoded by the selected nucleotide sequence may also be determined by various methods, such as ELISA, Western blotting, radioimmunoassay, immunoprecipitation, detection of the biological activity of the protein, or by immunostaining of the protein followed by FACS analysis (Sambrook et al, 1989; Ausubel et al, 1994).

By "high expression level (rate)", "high expression", "increased expression" or "high yield" is meant a sustained and sufficient amount of high expression or synthesis of a heterologous sequence, e.g., a gene encoding a therapeutic protein, introduced into a host cell. If a cell according to the invention is cultured by a method according to the invention described herein and if the cell produces at least more than about 5 picograms of the desired gene product per day (5 picograms/day/cell), increased or high expression, or high expression levels or rates, or high yields occur. Increased or high expression, or high expression level or rate, or high yield occurs if the cell according to the invention produces at least more than about 10 picograms of the desired gene product per day (10 picograms/day/cell, then also occurs), or high yield, particularly if the cell according to the invention produces at least more than about 15 picograms of the desired gene product per day (15 picograms/day/cell), or high expression level or rate, or high yield, particularly if the cell according to the invention produces at least more than about 20 picograms of the desired gene product per day (20 picograms/day/cell), then increased or high expression, or high expression level or rate, or high yield, particularly if the cell according to the invention produces at least more than about 30 picograms of the desired gene product per day (30 picograms/day/cell), increased or high expression, or high expression level or rate, or high yield occurs.

High or increased expression, high yield or high expression levels or rates in the sense of the present invention can be achieved in various ways. For example, through co-expression of a gene of interest with an amplifiable selectable marker gene, it is possible to select and identify cells that express heterologous genes to a high degree. The amplifiable selectable marker allows not only the selection of stably transfected host cells, but also the gene amplification of a heterologous gene of interest. Additional copies of the nucleic acid may be integrated into the genome of the host cell, into additional artificial/minichromosomes, or into polynucleotides positioned in episomal forms. This can be mixed with FACS-supported selection of recombinant host cells that contain one or more fluorescent proteins (e.g., GFP) or cell surface markers as additional selectable markers. Other methods for obtaining increased expression, which can also use combinations of different methods, are based, for example, on the use of (artificial) transcription factors, natural or synthetic preparations which up-regulate the expression of endogenous or heterologous genes, on improving the stability (half-life) of mRNA or protein, on improving the initiation of translation of mRNA, on increasing gene dosage by using episomal plasmids based on the use of viral sequences as replication origins, for example sequences of SV40, polyoma virus, adenovirus, EBV or BPV (Hemann et al, 1994), or on DNA-catena-based in vitro amplification systems (Monaco et al, 1996).

According to the invention is the coupled transcription of a gene of interest and a gene encoding a fluorescent protein. The resulting bicistronic mRNA expresses a protein of interest and a fluorescent protein. Based on the coupled expression of the protein of interest and the fluorescent protein, it is possible to easily select and isolate high-producing recombinant host cells according to the present invention by the expressed fluorescent protein, for example, by sorting using a Fluorescence Activated Cell Sorter (FACS).

The selection of recombinant host cells that exhibit high viability and increase the rate of expression of the desired gene product is a multi-stage process. Host cells transfected with an expression vector according to the invention, or optionally co-transfected with other vectors such as co-transfected, have been studied at least for the expression of a gene encoding a fluorescent protein and coupled to a gene of interest, in order to identify and select the cells/cell populations exhibiting the highest expression rate of the fluorescent protein. Preferably, only cells belonging to 10% of the cells having the highest fluorescent protein expression rate are selected and further cultured. In practice, this means that the brightest 10% of the fluorescing cells are sorted out and further cultured. Accordingly, the brightest 5%, preferably the brightest 3% or even the brightest 1% of the fluorescing cells in the cell mixture may also be selected and replicated. In a particularly preferred embodiment, only the brightest 0.5% or brightest 0.1% fluorescing cells are selected and replicated.

For this purpose, the cells which have been transfected before with the expression vectors according to the invention are cultured in a selection medium which optionally contains a selection agent which can amplify the specificity of the selection marker. A gradual increase in the concentration of the selective agent may be used, with a gradual increase in the selective pressure applied.

The screening step may be performed on a collection of cells, or a previously sorted collection/clone of cells may be used. One or more, preferably two or more, and especially three or more sorting steps may be performed, while between individual sorting steps the cells may be cultured and replicated for a specific period of time, e.g. about 2 weeks in case of pooling.

Optionally, the host cell may be subjected to one or more gene amplification steps in order to increase at least the copy number of the gene of interest and the amplifiable selectable marker gene. The process of stepwise amplifying genes using methotrexate is as described, for example, in U.S. Pat. No. 5,179,017. According to the invention, high yields can be achieved without being limited to increased gene copies. But are nevertheless characterized by high efficiency cloning, increased stability and fermentation robustness. It is thus possible to reduce the number of gene amplification steps required and, for example, to carry out only a single gene amplification.

Thus, the method of selecting cells according to the invention comprises the following steps:

(i) transforming a suitable host cell with at least the vector according to the invention, wherein the DNA of the expression vector is preferably stably incorporated into the host cell genome, or into an artificial chromosome/minichromosome;

(ii) culturing the transformed cells under conditions that allow expression of the gene of interest and the fluorescent protein;

(iii) culturing the cells in the presence of at least one selection agent, while only those cells capable of growing in the presence of said selection agent are replicated;

(iv) selecting cells exhibiting the highest fluorescent protein expression rate from the cell mixture, wherein the cells are selected and detected by a Fluorescence Activated Cell Sorter (FACS);

(v) the cells selected for the highest fluorescent protein expression rate were cultured.

Optionally repeating steps ii) -v) one or more times with the cells obtained according to step v). In addition, the transformed cells may also optionally be subjected to one or more gene amplification steps, wherein culturing them in the presence of a selection agent results in the amplification of the amplifiable selectable marker gene. This step can be performed using cells that have not been selected, and also using cells that have been selected one or more times in advance.

The invention also relates to a method in which correspondingly sorted cells are replicated and used for producing the desired encoded gene product. The selected high producing cells are preferably cultured in a serum free medium, and preferably in a suspension medium, under conditions that allow expression of the gene of interest. The protein of interest is preferably obtained from the cell culture medium in the form of a secreted gene product. However, by expressing proteins without secretion signals, it is also possible to isolate the gene product from the cell lysate. In order to obtain a pure, homogeneous product, which is substantially free of other recombinant proteins and host cell proteins, conventional purification steps are performed. First, cells and cell debris are typically removed from the culture medium or lysate. The desired cell product can then be freed from contaminating soluble proteins, polypeptides and nucleic acids by, for example, fractionation on immunoaffinity and ion exchange columns, ethanol precipitation, reverse phase HPLC or chromatography on sephadex, silica gel or cation exchange resins, such as DEAE. Methods for purifying heterologous proteins expressed by recombinant cells are known to those skilled in the art and are described in the literature, for example, Harris et al (1995) and Scopes (1988).

In the following, the invention is explained in more detail by means of non-limiting embodiments.

Examples

Abbreviations

AP: alkaline phosphatase

bp: base pairing

CHO: chinese hamster ovary

DHFR: dihydrofolate reductase

ELISA: enzyme linked immunosorbent assay

FACS: fluorescence activation cell sorter

FAP: fibroblast activation protein

GFP: green fluorescent protein

HBSS: hanks' balanced salt solution

HT: hypoxanthine/thymidine

HRPO: horseradish peroxidase

IRES: internal ribosome entry site

kb: kilobases

MAb: monoclonal antibodies

MTX: methotrexate (MTX)

And (3) PCR: polymerase chain reaction

sICAM: soluble intracellular adhesion molecules

Method

1. Cell culture and transfection

At 37 deg.C, in a humid environment and 5% CO₂In cell culture flasks, the cells CHO-DG44/DHFR-/- (Urlaub et al, 1983) were permanently cultured in serum-free CHO-S-SFMII medium (Invitrogen GmbH, Karlsruhe, DE) supplemented with hypoxanthine and thymidine in the form of suspension cells. Cell count and viability were determined using a CASY1 cell counter (Schaerfe System, DE), or by trypan blue staining, followed by 1-3X 10⁵Inoculation at a concentration of one ml and passage every 2-3 days.

CHO-DG44 was transfected using Lipofectamine Plus reagent (Invitrogen GmbH). For each transfection mixture, a total of 1. mu.g of plasmid DNA, 4. mu.l of Lipofectamine and 6. mu.l of Plus reagent were mixed together and added in a volume of 200. mu.l to 6X 10 in 0.8 ml of HT-supplemented CHO-S-SFMII medium according to the manufacturer' S instructions⁵In exponentially growing CHO-DG44 cells. After 3 hours of incubation at 37 ℃ in a cell incubator, 2 ml of CHO-S-SFMII medium supplemented with HT was added. For DNFR-based selection of stably transfected CHO-DG44 cells, 2 days after transfection, cells were transferred to CHO-S-SFMII medium without hypoxanthine and thymidine addition, and the medium was changed every 3 to 4 days. In the case of co-transfection, in which one expression vector contains DHFR and the other expression vector contains the neomycin-phosphotransferase selection marker, in the selection based on DHFR and neomycin phosphotransferase, G418(Invitrogen) was also added to the medium at a concentration of 400 μ G/ml.

DHFR-based gene amplification of the integrated heterologous gene was obtained by adding 5-2000nM concentration of the selection agent MTX (Sigma, Deisenhofen, DE) to the CHO-S-SFMII medium without HT.

2. Expression vector

To analyze expression, eukaryotic expression vectors are used, which are based on the pAD-CMV vector (Werner et al, 1998) and which mediate constitutive expression of heterologous genes by means of a combination of the CMV enhancer/hamster ubiquitin/S27 a promoter (WO 97/15664). The basic vector pBID, on the other hand, contains the DHFR minigene, which serves as an amplifiable selectable marker (see, for example, EP 0393438), and the DHFR minigene has been replaced in the vector pBIN by a neomycin resistance gene (see FIG. 2). For this purpose, the selection marker neomycin phosphotransferase, including the SV40 early promoter and the TK-polyadenylation signal, was isolated from the commercially available plasmid pBK-CMV (Stratagene, La Jolla, USA) as the 1640 base pair Bsu36I fragment. After the reaction to fill the ends of the fragment with Klenow-DNA-polymerase, the fragment was ligated with the 3750 base pair Bsu36I/StuI fragment of the vector pBID, which was also treated with Klenow-DNA-polymerase.

In the bicistronic basic vector pBIDG (FIG. 2), the IRES-GFP gene region was isolated from the vector pIRES2-EGFP (Clontech, Palo Alto, Calif., USA) and the vector pBID was introduced under the control of the CMV enhancer/promoter to allow retention of the multiple cloning site between the promoter region and the IRES element. The following procedure was used. In the PCR mutagenesis, in which the plasmid pIRES2-EGFP is used as template, the HindIII cleavage site AAGCTT in the IRES sequence is converted into the sequence ATGCTT by using mutagenic primers on the one hand and is thus excluded. On the other hand, the XbaI cleavage site is inserted by a primer complementary to the 5 'end of the IRES sequence, or the SpeI cleavage site is introduced by a primer complementary to the 3' end of the GFP sequence. The resulting PCR fragment, containing the complete IRES and GFP sequences, was digested with XbaI and SpeI and cloned into a single XbaI cleavage site 3' of the multiple cloning site of the vector pBID.

From pAD-sICAM, the human sICAM gene was isolated as a HindIII/SalI fragment (Werner et al, 1998) and cloned into the corresponding cleavage site of the vector pBIDG, resulting in the vector pBIDG-sICAM (FIG. 3).

For the expression of the monoclonal humanized F19 antibody, the heavy chain was isolated from plasmid pGlD105F19HC (NAGENESEQ: AAZ32786) in the form of a 1.5kb NaeI/HindIII fragment and cloned into the vector pBIDG digested with EcoRI (filled in with Klenow-DNA-polymerase) and HindIII, resulting in the vector pBIDG-F19HC (FIG. 3). On the other hand, the light chain was isolated from plasmid pKN100F19LC (NAGENESEQ: AAZ32784) in the form of a 1.3kb HindIII/EcoRI fragment and cloned into the corresponding cleavage site of the vector pBIN, yielding vector pBIN-F19LC (FIG. 3).

3.FACS

Flow cytometric analysis and sorting was performed using a Coulter Epics Altra apparatus. The FACS mounted helium-argon laser has an excitation wavelength of 488 nanometers. The fluorescence intensity is absorbed at a wavelength corresponding to the fluorescent protein and processed by the attached software Coulter Expo 32. The selection was done at 8000-. The suspended cells were centrifuged (at 180 Xg for 5 min) and adjusted to 1-1.5X 10 in HBSS⁷Cell concentration per ml. Cells are then selected based on their fluorescent protein signals. The cells are placed in test tubes which already contain the culture medium, centrifuged and, depending on the number of cells selected, inoculated into the corresponding culture vessel.

4.ELISA

SICAM titers in supernatants of stably transfected CHO-DG44 cells were quantified by ELISA using standard procedures (Ausubel et al, 1994, update) using two SICAM-specific monoclonal antibodies developed in mice (e.g. as described in US5,284,931 and US5,475,091). One of the two antibodies was an HRPO-conjugated antibody. Purified SICAM protein was used as standard.

The F19 monoclonal antibody in the supernatant of stably transfected CHO-DG44 cells was quantified by ELISA using goat anti-human IgG Fc fragment (Dianova, Hamburg, DE) on the one hand and AP-conjugated goat anti-human kappa light chain antibody (Sigma) on the other hand, according to standard procedures (Ausubel et al, 1994, update). Purified F19 antibody was used as a standard.

By the formula picogram/((C)_t-C_o)t/In(C_t-C_o) Calculated yield (microgram/cell/day), where C_oAnd C_tCell counts at the time of inoculation or harvest, respectively, and t is the culture time.

Example 1: comparison of CMV with hamster-ubiquitin/S27 a-promoter Activity

To compare the activity of the hamster-ubiquitin/S27 a-promoter with the activity of the CMV promoter, which is often used in eukaryotic expression vectors, CHO-DG44 cells were transfected with various recombinant vectors. Under the control of the CMV-promoter, or under the control of the hamster ubiquitin/S27 a promoter, heterologous gene products, on the one hand lysosomal enzymes, and on the other hand IgGl-antibodies, are expressed. Both promoters are functionally linked to the CMV-enhancer. BGH polyadenylation was used as a stop signal for the heterologous gene. The expression vector containing the CMV promoter was modified pcDNA3 ("CMV¹", Invitrogen) or pBluescript vector (" CMV²", Stratagene) and additionally encodes the amplifiable selectable marker dihydrofolate reductase. On the other hand, expression vectors with hamster promoters are based on the pAD-CVM vector (Werner et al, 1998). To express the heavy and light chains of the antibody, a second vector containing a neomycin resistance gene as a selectable marker was co-transfected. However, the CMV enhancer may also be replaced by the SV40 enhancer.

Cell clones were selected and isolated in HT-free medium by limiting dilution in 96-well plates after transfection (in case of co-transfection 400. mu.g/ml G418 was added). The cell clones with the highest yield for the recombinant protein were subjected to a stepwise DHFR-based gene amplification by stepwise increasing the methotrexate concentration from 5nM, via 50nM, 500nM to 2. mu.M, together with dilution cloning in each case. At each amplification stage, approximately 20 to 30 clones with the highest yield were selected.

Generally, hamster promoters were found to have higher efficiency. The yields or titers obtained for the expression of the lysosomal enzyme, as well as the expression of the antibody, are 2 to 5 times higher than those obtained with cells in which the heterologous gene is expressed under the control of the CMV promoter. FIG. 1 shows, by way of example, the relative titer and the relative specific yield of the best cell clones, setting the expression of the respective heterologous gene based on the CMV promoter at 1 (CMV)¹Is a lysosomal enzyme, CMV²Is an antibody).

Example 2: isolation of cells highly expressing sICAM by GFP-based FACS sorting

Soluble forms of the intercellular adhesion molecule ICAM1, sICAM, are potential therapeutic agents for the cold because it competes with the ICAM receptor for binding to rhinoviruses, and this approach can reduce, or even prevent, rhinovirus interaction with the ICAM receptor, a prerequisite for its entry into the cell and subsequent infection (Bella et al, 1999; Marlin et al, 1990).

sICAM was chosen as an example for the expression of single-chain proteins (480 amino acids) in CHO cells. For this purpose, CHO-DG44 was transfected with pBIDG-sICAM (FIG. 3). Additional expression of GFP in pBIDG-sICAM transfected cells makes it possible to use FACS-based selection strategies. The therapeutic proteins sICAM and GFP are expressed concomitantly by the bicistronic transcription unit, and DHFR is expressed by separate transcription units. From 2 to 3 weeks after the first selection in CHO-S-SFMII medium without HT, 5% of the cells with the highest GFP fluorescence were selected. After approximately 2 weeks of culture, 5% of the cells with the highest GFP fluorescence were isolated again. This successive picking was performed a total of 6 times. A good correlation between sICAM yield and GFP fluorescence was shown (figure 4). By FACS-supported selection alone, cell pools with high specific yields of up to 16 picograms/cell/day were isolated in a minimum time without any MTX amplification step (FIG. 5). By combining GFP-based selection with a single subsequent MTX amplification step, it is even possible to increase the yield to more than 30 picograms/cell/day (FIG. 6). These yields were achieved with amplification of pools after the fourth selection with 500nM MTX and also with amplification of pools after the sixth selection with 2. mu.M MTX. In contrast to stepwise amplification, which usually starts with very low MTX concentrations in the range of 5-20nM MTX, higher MTX concentrations must be used to start with in order to achieve amplification. Thus, neither 5 or 50nM MTX was added to cells from the fourth selection, nor 500nM MTX was added to cells from the sixth selection, resulting in a significant increase in productivity (FIG. 6). It is clear that the level of DHFR in the starting set is already so high that complete inhibition of DHFR can only be achieved with high doses of MTX. Furthermore, despite the high initial dose of MTX, the survival of the pre-selected cell collection during selection was much improved, i.e. a high viability cell population was obtained in a shorter time than the traditional stepwise gene amplification strategy (fig. 7).

Example 3: isolation of cells highly expressing monoclonal antibody F19 by GFP-based FACS sorting

In co-transfection, CHO-DG44 cells were transfected with the plasmid combination pBIDG-F19HC and pBIN-F19LC (FIG. 3). The expressed humanized antibody F19, directed against the surface molecule FAP, was synthesized by reactive stromal fibroblasts (see also patent EP 0953639). In the vector configuration used, the two protein chains of the antibody expressed by each of its own vectors are expressed, which additionally encode, in separate transcription units, either the DHFR or the neomycin phosphotransferase selectable marker. In addition, another selection marker, GFP, was included in the vector pBIDG-F19 HC. By co-transfecting CHO-DG44 with the vector pBIDG-F19HC/pBIN-F19LC by transcriptional ligation of GFP and heavy chain expression by IRES elements, cells highly expressing antibody F19 can be isolated in a short time by merely selecting cells having a high GFP content using continuous FACS sorting. For this, after selection of the transfected cell pool in the HT-free CHO-S-SFMII medium supplemented with 400. mu.g/ml G418 for the first 2-to 3-weeks, 5% of the cells with the highest GFP fluorescence were selected by FACS. This selection was performed a total of up to 6 times with an incubation period of approximately 2 weeks between each selection. Surprisingly, a good correlation was shown between F19 yield and GRP fluorescence (fig. 8), although both protein chains were expressed from their own vectors, and in GFP-based FACS sorting, it was only possible to select for expression of the heavy chain, as it was transcribed together with GFP. The yield was increased to 10 picograms/cell/day (FIG. 9) and further increased to an average of 37 picograms/cell/day by adding 1000nM MTX to the selection medium starting from the fifth selected cell pool by a single subsequent MTX amplification step. Comparative data can also be obtained by functionally linking the hamster promoter to the SV40 enhancer in place of the CMV enhancer. Also, the conventional strategy, which typically includes four amplification stages, can reduce the development time for selecting highly productive cells by half to about 120 days, with a significant reduction in development efficacy and cost, as compared to conventional stepwise gene amplification strategies.

Reference to the literature

Adam，M.A.et al.，J Virol 1991，65，4985-4990

Ausubel，F.M.et al.，Current Protocols in molecular biology.New York：Greene Publishing

Associates and Wiley-Interscience.1994(updated)

Bella，J.et al.，J.Struct.Biol.1999，128，69-74

Bennett，R.P.et al.，BioTechniques 1998，24，478-482

Chalfie，M.et al.，Science 1994，263，802-805

Chamov，S.M.et al.，Antibody Fusion Proteins，Wiley-Liss Inc.，1999

Davies，M.V.et al.，J Virol 1992，66，1924-1932

Faisst，S.et al.，Nucleic Acids Research 1992，20，3-26

Gossen，M.et al.，Curr Opi Biotech 1994，5，516-520

Gubin，A.N.et al.，Biochem Biophys Res Commun 1997，236，347-350

Haber，D.A.et al.，Somatic Cell Genetics 1982，8，499-508

Harris et al.，Protein Purification：A Practical Approach，Pickwood and Hames，eds.，IRL Press，1995

Hemann，C.et al.，DNA Cell Biol 1994，13(4)，437-445

Huston，C.et al.，Proc Natl AcadSci USA 1988，85(16)，5879-5883

Hu，S.et al.，Cancer Res.1996，56(13)，3055-3061

Jang，S.K.et al.，J Virol 1989，63，1651-1660

Kaufman，R.J.et al.，JMol Biol 1982，159，601-621

Kaufman，R.J.，Methods in Enzymology 1990，185，537-566

Kortt，A.A.et al.，Protein Engineering 1997，10(4)，423-433

Levenson，V.V.et al.，Human Gene Therapy 1998，9，1233-1236

Lottspeich F.and Zorbas H.eds.，Bioanalytic，Spektrum Akad.Verl.，1998

Lovejoy，B.et al.，Science 1993，259，1288-1293

Marlin，S.D.et al.，Nature 1990，344，70-77

Meng，Y.G.et al.，Gene 2000，242，201-207

Monaco，L.et al.，Gene 1996，180，145-150

Morgan，R.A.et al.，Nucleic Acids Research 1992，20，1293-1299

Mosser，D.D.et al.，Bio Techniques 1997，22，150-161

Nolan，G.P.et al.，Proc Natl Acad Sci USA 1988，85，2603-2607

Ohshima，Y.et al.，J Mol Biol 1987，195，247-259

Pack，P.et al.，Biotechnology 1993，11，1271-1277

Pack，P.et al.，J Mol Biol 1995，246(11)，28-34

Pelletier，J.et al.，Nature 1988，334，320-325

Perisic，O.et al.，Structure 1994，2，1217-1226

Primig，M.et al.，Gene 1998，215，181-189

Ramesh，N.et al.，Nucleic Acids Research 1996，24，2697-2700

Sambrook，J.，Fritsch，E.F.& Maniatis，T.，Molecular Cloning：A Laboratory Manual Cold

Spring Harbor Laboratory，Cold Spring Harbor，New York，1989

Scopes，R.，Protein Purification，Springer Verlag，1988

Simonson，C.C.et al.，Proc Natl Acad Sci USA 1983，80，2495-2499

Sugimoto et al.，Biotechnology 1994，12，694-698

Urlaub，G.et al.，Cell 1983，33，405-412

Werner，R.G.et al.，Arzneim.-Forsch./Drug.Res.1998，48，870-880

Wigler，M.et al.，Proc Natl Acad Sci USA 1980，77，3567-3570

Sequence listing

<110> Beilingger Engium Framema Bihe (BOEHRINGER INGELHEIM PHARMA GmbH & Co. KG)

<120> expression vector, method for producing heterologous gene product, and method for selecting high-producing recombinant cell

<130>Case1-1412

<140>

<141>

<160>1

<170>PatentIn Ver.2.1

<210>1

<211>2406

<212>DNA

<213> Chinese hamster (cricetulus griseus)

<300>

<310>PCT/EP/96/04631

<311>1996-10-24

<312>1997-05-01

<400>1

gatctccagg acagccatgg ctattacaca gagaaaccct gtctggaaaa acaaaaaatt 60

agtgtccatg tgtaaatgtg tggagtatgc ttgtcatgcc acatacagag gtagagggca 120

gtttatggga gtcagttcct attcttcctt tatgggggac ctggggactg aactcaggtc 180

atcaggcttg gcagaaagtg cattagctca cggagcctta tcattggcga aagctctctc 240

aagtagaaaa tcaatgtgtt tgctcatagt gcaatcatta tgtttcgaga ggggaagggt 300

acaatcgttg gggcatgtgt ggtcacatct gaatagcagt agctccctag gagaattcca 360

agttctttgg tggtgtatca atgcccttaa aggggtcaac aacttttttt ccctctgaca 420

aaactatctt cttatgtcct tgtccctcat atttgaagta ttttattctt tgcagtgttg 480

aatatcaatt ctagcacctc agacatgtta ggtaagtacc ctacaactca ggttaactaa 540

tttaatttaa ctaatttaac cccaacactt tttctttgtt tatccacatt tgtggagtgt 600

gtgtgtgtgt gtgtgtgtgt gtgtgtgtgt gtgtgtgtgt gtgtgtgtgt gtgtgtgtgc 660

gcgcgcgcgc gcgctcggat cattctacct tttgtttaaa aaatgttagt ccaggggtgg 720

ggtgcactgt gaaagtctga gggtaacttg ctggggtcag ttctttccac tataggacag 780

aactccaggt gtcaactctt tactgacaga accatccaaa tagccctatc taattttagt 840

tttttattta tttatttttt gtttttcgag acagggtttc tctgtggctt tggaggctgt 900

cctggaacta gctcttgtag accaggctgg tctcgaactc agagatccac ctgcctctgc 960

ctcctgagtg ctgggattaa aggcatgcgc caccaacgct tggctctacc taattttaaa 1020

agagattgtg tgtcacaagg gtgtcatgtc gccctgcaac cacccccccc ccaaaaaaaa 1080

aaaaaaaaaa acttcactga agctgaagca cgatgatttg gttactctgg ctggccaatg 1140

agctctaggg agtctcctgt caaacagaat ctcaacaggc gcagcagtct tttttaaagt 1200

ggggttacaa cacaggtttt tgcatatcag gcattttatc taagctattt cccagccaaa 1260

aatgtgtatt ttggaggcag cagagctaat agattaaaat gagggaagag cccacacagg 1320

ttattaggaa gataagcatc ttctttatat aaaacaaaac caaaccaaac tggaggaggt 1380

ctacctttag ggatggaaga aaagacattt agagggtgca atagaaaggg cactgagttt 1440

gtgaggtgga ggactgggag agggcgcaac cgctttaact gtcctgtttt gcctattttt 1500

tggggacagc acatgttcct atttttccca ggatgggcaa tctccacgtc caaacttgcg 1560

gtcgaggact acagtcattt tgcaggtttc cttactgtat ggcttttaaa acgtgcaaag 1620

gtgaccatta accgtttcac gctgggaggg cacgtgcggc tcagatgctt cctctgactg 1680

agggccagga gggggctaca cggaagaggc cacacccgca cttgggaaga ctcgatttgg 1740

gcttcagctg gctgagacgc cccagcaggc tcctcggcta caccttcagc cccgaatgcc 1800

ttccggccca taacccttcc cttctaggca tttccggcga ggacccaccc tcgcgccaaa 1860

cattcggccc catcccccgg tcctcacctg aatctctaac tctgactcca gagtttagag 1920

actataacca gatagcccgg atgtgtggaa ctgcatcttg ggacgagtag ttttagcaaa 1980

aagaaagcga cgaaaaacta caattcccag acagacttgt gttacctctc ttctcatgct 2040

aaacaagccc cctttaaagg aaagcccctc ttagtcgcat cgactgtgta agaaaggcgt 2100

ttgaaacatt ttaatgttgg gcacaccgtt tcgaggaccg aaatgagaaa gagcataggg 2160

aaacggagcg cccgagctag tctggcactg cgttagacag ccgcggtcgt tgcagcgggc 2220

aggcacttgc gtggacgcct aaggggcggg tctttcggcc gggaagcccc gttggtccgc 2280

gcggctcttc ctttccgatc cgccatccgt ggtgagtgtg tgctgcgggc tgccgctccg 2340

gcttggggct tcccgcgtcg ctctcaccct ggtcggcggc tctaatccgt ctcttttcga 2400

atgtag 2406

Claims

1. An expression vector comprising a gene encoding a protein of interest functionally linked to a hamster ubiquitin/S27 a-promoter and a gene encoding a fluorescent protein.

2. An expression vector according to claim 1, characterized in that said vector contains an amplifiable selectable marker gene.

3. The expression vector according to claim 1 or 2, characterized in that said vector comprises one or more enhancers functionally linked to said one or more promoters.

4. An expression vector according to any one of claims 1 to 3, characterized in that the vector additionally contains an Internal Ribosome Entry Site (IRES) which allows bicistronic expression of the gene encoding the fluorescent protein and the gene encoding the protein/product of interest.

5. The expression vector according to any one of claims 1 to 4, characterized in that the gene encoding a fluorescent protein and the amplifiable selectable marker gene are located in one or two separate transcription units.

6. The expression vector according to any one of claims 1 to 5, characterized in that the functional linkage is not a linkage via an intron sequence.

7. The expression vector according to any one of claims 1 to 6, characterized in that the amplifiable selectable marker gene encodes dihydrofolate reductase (DHFR) or a fusion protein of a fluorescent protein and DHFR.

8. The expression vector according to any one of claims 3 to 7, characterized in that the enhancer is the CMV-or SV 40-enhancer.

9. The expression vector according to any one of claims 1 to 8, characterized in that the vector additionally comprises at least one polyadenylation signal.

10. An expression vector according to any one of claims 1 to 9, characterized in that said vector comprises a multiple cloning site required for incorporation of a gene encoding a protein of interest in place of a gene encoding a protein/product of interest.

11. A eukaryotic host cell which has been transfected with an expression vector according to any one of claims 1-10.

12. Host cell according to claim 11, characterized in that the host cell is a mammalian cell, preferably a CHO cell.

13. Host cell according to claim 11 or 12, characterized in that the host cell has additionally been transfected with one or more vectors carrying genes encoding one or more further proteins/products of interest, and at least one further selectable marker.

14. A method for producing a heterologous gene product, wherein a host cell according to claims 11-13 is cultured under conditions allowing expression of said gene product, and said gene product is isolated from the culture or culture medium.

15. A method for the production of a protein/product in heteromeric form, wherein a host cell according to claim 13, which has been co-transfected with a plurality of expression vectors encoding different subunits of the protein/product in heteromeric form, is cultured under conditions permitting expression of said protein/product in heteromeric form, and the protein/product in heteromeric form is isolated from the culture or culture medium.

16. Method according to claim 15, characterized in that the protein in heteromeric form is an antibody.

17. Method according to any one of claims 14 to 16, characterized in that said host cell is subjected to one or more gene amplification steps.

18. Method according to claim 17, characterized in that the amplifiable selectable marker is DHFR and the amplification agent is methotrexate.

19. The method according to claim 18, characterized in that only one step of amplifying the gene using methotrexate is performed on the host cell.

20. Method according to any one of claims 14 to 19, characterized in that the method is carried out in serum-free medium.

21. Method according to any one of claims 14 to 20, characterized in that the host cell is cultured in suspension culture.

22. A method of screening for host cells expressing a protein/product of interest, wherein a population of host cells according to any one of claims 11-13 is cultured under conditions that allow expression of the protein/product of interest and a fluorescent protein, and one or more cells showing the highest level of expression of the fluorescent protein are identified and/or selected.

23. The method according to claim 22, characterized in that the selection is performed using a Fluorescence Activated Cell Sorter (FACS).

24. Method according to claim 22 or 23, characterized in that said selected host cell is subjected to one or more additional gene amplification steps.

25. The method according to claim 24, characterized in that the amplifiable selectable marker is DHFR and the amplification agent is methotrexate.