WO2002099100A2

WO2002099100A2 - Method of production of a protein in cells which inducibly express the cell cycle inhibitor protein, p21

Info

Publication number: WO2002099100A2
Application number: PCT/EP2002/006054
Authority: WO
Inventors: Mohamed Al-Rubeai; John Shuttleworth
Original assignee: Lonza Biologics Plc
Priority date: 2001-06-01
Filing date: 2002-06-03
Publication date: 2002-12-12
Also published as: AU2002317773A1; AU2002317773A8; WO2002099100A3; GB0113318D0

Abstract

A method for expression of a recombinant protein in a producer cell line comprises the step of co-expressing recombinant p21 or another cell cycle inhibitor protein.

Description

Method for production of a protein

The present invention relates to the field of protein expression and a method of expressing a recombinant protein from a producer cell line or of an antibody from a hybridoma cell line. It further relates to suitable producer or hybridoma cell lines.

The production of therapeutic proteins at an industrial scale aims at obtaining maximized yields of the desired protein, which is either secreted into the medium or is entrapped in the producer cells. The higher the yield, the more easy and cost-efficient downstream processing will be. However, high level expression of protein is a complex problem only partially depending on promotor strength of the recombinant product gene itself. Protein degradation, metabolic status of the cell, occupancy of the cellular translation apparatus or unwanted negative events such as untimely apoptosis of producer cells are decreasing yield, too.

Kim et al. (Biotechnology and Bioengineering, Vol. 71, p.184-193, 2000) describe the overexpression of apoptosis-inhibiting bcl-2 in antibody-producing CHO cells in the presence of butyrate in order to overcome the cytotoxic effect of butyrate. As a disadvantage, this system does not allow to steer high-level expression since the butyrate is a constitutive component of the medium and thus requires steady expression of the second transgene, bcl-2. Thus biomass production and timing of the production phase are affected.

Fussenegger et al. (Biotechnology and Bioengineering, Vol. 65, p. 144-150, 1999) describe enhancement of heterologous protein production in anchorage-dependent CHO cells by using a plasmid-borne multicistronic expression unit endocing a product gene and p27 ^{K l} , a specific cell cycle inhibitor exerting a cytostatic effect when expressed. Both genes are under control of a single tetracycline-repressible promoter PhCMV*-l in a multicistronic construct.

Fussenegger et al. (Biotechnology and Bioengineering, Vol. 55, p. 927.-939, 1997) describe a cytostatic process for enhanced recombinant protein production in anchorage- dependent CHO cells based on transient over-expression of genes encoding p27, p21 or p53(175P). Again both the cytostatic and the product gene are under control of a single tetracycline-repressible viral promoter PhCMV*-l in a plasmid-borne dicistronic expression cassette. Tetracycline-repression of PhCMV*-l relies on expression of the construct in a specific CHO cell line XMK1-9 that is expressing a tetracycline regulatable transactivator (tTA).

The general disadvantage of the system of Fussenegger et al. is the requirement to suppress gene expression up to the production phase and thus to remove or lower the concentration of the repressing agent tetracycline in a controlled way at this point of time. For an industrial process, an antibiotic is an agent far too costly. However, choice of the tetracycline-repressible promoter PhCMV*-l (Gossen et al., 1992, Tight control of gene expression, Proc. Natl. Acad. Sci. USA 89, p. 5547-5551, 1992) is said to be of utmost importance by reasoning that most viral promoters are most active at S-phase. Only the latter is said to be highly active in Gl -phase as required. In view of the need of a cell line engineered for tTA expression and the tretracycline repression entailed by choice of promoter, the system of Fussenegger is not yet optimal for devising a production process at an industrial scale. It is very inflexible in that the expression of both the regulatory protein and the producer protein are tightly linked such as that for high-level expression of the product protein, the cell cycle inhibitor has to be expressed in equal amounts thus limiting the enhancing effect of coexpression on productivity. The system has only been shown to work in anchorage dependent cells; p21 co-expression in this system was observed to promote detachment of CHO cells from substrate, in contrast to p27 expression.

Fussenegger et al. (Nature Biotechnology, 16, p. 468-472, 1998) describe a more effective method of arresting the cell cycle of anchorage-dependent CHO cells for enhancing the production of a recombinant protein by overexpressing specifically p21 gene product. This was achieved by tetracycline regulated, tricistronic co-expression of p21 and the enhancer binding protein CCAAT and a recombinant product protein. As a disadvantage, this system requires in addition to tTA expression, multiple regulatory proteins to be co-expressed and thus to decrease the part of the cellular protein synthesis machinery available to product protein synthesis. It is an object of the present invention to avoid the disadvantages of the prior art and to devise another method for protein expression of a product protein such as an antibody or a recombinant protein. Another object of the present invention are producer cells having such expression capacity. These objects are achieved by co-expressing a cell cycle inhibitor protein inducibly in a producer or host cell line.

According to the present invention, the method of enhancing production of a product protein, preferably of a recombinant product protein, in a producer cell line comprises the step of co-expressing a recombinant cell cycle inhibitor protein or a functional mutein thereof with said recombinant protein in the producer cell line, wherein expression of the cell cycle inhibitor is under control of an inducible promoter. Preferably, the cell cycle inhibitor is p21 protein, more preferably, it is p21 protein that is expressed in the absence of a further recombinant inducer and/or stabiliser of p21 expression, most preferably, it is p21 protein that is expressed in the absence of recombinant CC AAT/enhancer-binding protein alpha which has such specific enhancing effect on cellular p21 activity.

A cell cycle inhibitor protein according to the present invention has an inhibitory effect on cell cycle progression of eukaryotic cells by means of inhibiting at least one mammalian cyclin dependent kinase (cdk's), a large family of protein kinases which function to regulate the cell cycle (Morgan et al., 1995, Principles of cdk regulation. Nature 374, 131 ff). Cell cycle inhibition is understood as arresting cell cycle, thus effectively synchronizing a population of cells. Such activity can be tested as described in Xiong et al., Nature 366:701 ff, 1993. Preferably, it is a mammalian cell cycle inhibitor protein. For instance, inhibition of proliferation can be tested by incorporation rate of ³H-thymidine. A cell cycle inhibitor protein according to the present invention further is devoid of regularly triggering cell death or apoptosis in addition to its cytostatic effect. It is well established that inhibition of cell cycle progression and initiation of apoptotis are separate signalling pathways. Wild-type p53 is therefore excluded by from the definition of a cell cycle inhibitor protein according to the present invention. Muteins of p53 (such as e.g. described in Crook et al., Cell 79:817 ff) that are devoid of the apoptosis promoting acitivity or functional protein domains are within the scope of the present invention. For the sake of clarity, Bcl-2 protein and variants thereof is further exempted from the definition as restraining entry of resting cells into the cell cycle but not arresting proliferating cells.

p21 (CIP1, WAF1, SDH) is such an inhibitor of cdk/cyclin complexes (Xiong, Y. et al., 1993, p21 is a universal inhibitor of cyclin kinases, Nature 366: 701-704; Zhang, H. et al., 1993, Proliferating cell nuclear antigen and p21 are components of multiple cell cycle kinase complexes, Mol. Biol. of the Cell 4: 897-906). p21 has a N-terminal region of high homology to p27, another cell cycle inhibitor. Unlike p27 (Toyoshima et al., 1994, Cell 78:67-74) which is transiently kept inactive by binding of a masking' protein factor, the latter being removed upon contact-inhibition of animal cells, p21 is primarily regulated at the level of transcription. p21 expression is decoupled from p27 expression in mammalian cell cycle (Wong et al., 2001, Differential expression of pl6/p21/p27 .. and their relationships to cell proliferation, J. Pathol. 194:35-42; ) and is directly involved in control of DNA replication (Waga et al., The p21 inhibitor of cyclin-dependent kinases controls DNA replication by interaction with PCNA, Nature 369, p. 574-578, 1994). Further inhibitors are known to date (e.g. pl6, pl9: Chan et al., Mol. Cell. Biol. 15:2682; pl8: Guan et al., Genes & Develop. 8:2929), some of them having specific inhibitory effect only towards specific Cdk kinase subtypes. Preferably, the cell cycle inhibitor protein according to the present invention is a universal inhibitor of Gl cyclin-Cdk protein kinase activity in mammalin cells, though possibly to varying degree. More preferably, the cell cycle inhibitor is p21 or p27 protein or a variant thereof. Such variants are called muteins and are referred to im more detail below. Most preferably, the cell cycle inhibitor is p21 protein or a variant thereof that allows to achieve the objects of the present invention.

Surprisingly, e.g. p21 conferred stable and quantitative cell cycle arrest in neoplastically transformed stable suspension cell lines. Such tumor or anchorage-independent cells are devoid of contact inhibition and are strictly proliferation competent. Further, cell viability was not affected by stable cell cycle arrest conferred by e.g. p21 though individual clones might irregularly show loss of viability. Further surprising the expression of p21 was accompanied by a substantial 4 to 5-fold increase in expression of a desired recombinant protein, e.g. a chimeric antibody, due to enhanced single cell productivity. This despite the fact that co-expression of the cytostatic gene product consumed part of the cellular protein synthesis capacity. The effect lasted as long as p21 was kept being induced. It was even more astonishing that the effect persisted even in very high density culture after extended times of cell culture, i.e. after adaption to stationary phase growth.

A possible embodiment of the invention is described in more detail in the figures. What is shown is:

Fig. 1 A. Effect of ectopic p21 expression on cell proliferation in batch culture of NSO cells. Fig. IB. Effect of ectopic p21 expression on cell viebility in batch culture of NSO cells.

Fig. 2 Western-blot analysis of ectopic p21 expression in NSO cells.

Fig. 3 Cell cycle analysis of NSO cells induced for p21 expression.

Fig. 4 MAb production rate for proliferating and arrested NSO cells.

Fig. 5 Effect of ectopic p21 expression in high-density perfusion culture Fig. 6 Mab production rate in perfusion culture

Fig. 7 Single cell productivity in perfusion culture

The the product protein is that is sought to be produced and harvested in high amount from cell culture. It may be any protein of interest, e.g. therapeutic proteins such as interleukins or enzymes, e.g. enzyme inhibitors or antibodies or fragments thereof (a fab fragment for instance). The preferably recombinant product protein may include a signal sequence allowing secretion of the polypeptide from the host producer cell. In a preferred embodiment of the present invention, the recombinant product protein is a secreted protein, in particular an antibody. More preferably, the recombinant protein is an antibody or engineered antibody or a fragment thereof, most preferably it is an Immunoglobulin G (IgG) antibody.

p21 according to the present invention is any naturally occuring eukaryotic, preferably mammalian, p21 protein. p21 can be e.g. human p21 as is described in the nucleotide and protein database of the National Center for Biotechnology Information, Bethesda, U:S.A., under accession number XM_011458 and XP Ol 1458.2, respectively. Human p21 is in its native, translated form a 164 amino acid protein of approximately 18 kDa. p21 according to the present invention can also be any cytostatically functional (as defined in Xiong et al., Wong et al., Zhang et al, ibd.) homologue or variant of a mammalian p21 protein. Preferably, any functional homologue of p21 according to the present invention has a homology of 70%, more preferably a homology of 85%, most preferably a homology of 95% with human p21 at the DNA sequence level. This definition of functional equivalent sequence homologues or variants applies to the cell cycle inhibitors of the present invention in general.

Variants according to the present invention of known, naturally occuring cell cycle inhibitor proteins can be obtained from any naturally occuring p21 gene sequence or other cell cycle inhibitor sequence by methods well-known in the art. Such mutated protein is a mutein. Suitable muteins can be scanning (triplett) mutants, point mutants, insertion or deletion mutants or truncated versions of a naturally occuring cell cycle inhibitor amino acid or gene sequence, respectively. Methods for creating muteins (Sambrook et al., Molecular cloning, Cold Spring Harbor Laboratory Press 1989) usually comprise manipulation of the gene sequence, e.g. by error prone PCR (Biotechnol. Bioeng. 2000, 70:446), directed mutagenesis of individual residues, mutation scanning with codon tripletts, DNA shuffling comprising controlled fragmentation and PCR reassembly (Kolkman et al., Nat. Biotechnol. 2001, 19:423), and so hence and so forth, which manipulation is expediently followed by screening for cytostatic functionality in an enzyme or cellular assay e.g. according to Xiong et al. 1993, Nature 366, p.701-704.

'Recombinant' according to the present invention means a protein expressed from at least one exogenous copy of the corresponding cell cycle inhibitor gene in a cell line that has originally been introduced into said cell line by any technique of genetic engineering. As in the case of e.g. p21, the genome of mammalian producer cells will usually harbor naturally occuring copies of p21, that is the respective species homologue of p21 of that given producer cell line. Any further expressable copy of a p21 gene introduced by manipulation gives rise to recombinant protein by definition. This includes replacing the endogenous gene promoter by a stronger gene promoter directing expression of the original p21 coding sequence at its native gene locus. Any manipulation of the native p21 locus leading to enhanced expression is therefor comprised in the present definiton. More conventionally, newly introduced additional copies of a recombinant gene can e.g. be integrated in the genome or can be carried on an episomal element. Both stable or transient expression can be employed. Any known expression vector technology, whether being circular or linear, whether comprising one or several vectors and/or promoting co-expression of further auxiliary proteins directing the expression of the desired 'recombinant' protein can be used according to the present invention.

Preferably, at least the cycle inhibitor protein such as p21 protein is transiently expressed by means of an episomal vector element. More preferably, such an episomal expression vector allows for high copy number expression of the inhibitor protein. In the context of the present invention, such a high copy number epression vector is a vector that is present in a cell in an average copy number of at least 7 copies.

Co-expression of two proteins means that the two proteins are expressed from their coding DNA sequences in or from the same cell at a time. In the context of the present invention, co-expression refers to timewise parallel expression of recombinant inhibitor protein, preferably p21 protein, and the recombinant product protein.

According to the present invention, the cycle inhibitor such as p21 protein is expressed under control of an inducible promoter. An inducible promoter according to the present invention is a promoter capable of directing transcription of the e.g. p21 gene in a eukaryotic cell that is either constitutively repressed in the absence of an 'inductor' molecule or is not recruiting RNA polymerase in the absence of an 'inductor' molecule.

Such an 'inductor' molecule may be understood as a triggering agent. Preferably, the inductor is a small organic molecule of a molecular weight up to 500 g/mol, more preferably this definition applies with the exception of antibiotics such as streptogramin, tetracyclin and the like.

In a preferred embodiment of the present invention, the inducible promoter is an operator- regulated promoter. An example of such a permanently repressed, operator-regulated promotor is for instance the lacSwitch^® system (Stratagene Corp., described in US 4'833'080 and US 5 589 392) which is a preferred embodiment of the present invention. The lacSwitch system employs a retroviral promoter, the Rous-Sarcoma- Virus LTR- Promotor as given in SEQ ID No. 2, which is immediately followed downstream by multiple lac-Repressor binding sites forming an operator sequence. Lac-repressor is constitutively expressed from a second expression vector harboring another strong, tissue or cell type independent constantly active promoter; repression of the RSV-LTR-Promoter is released by addition of an inductor such as allolactose or a thiogalactoside, preferably isopropyl-thiogalactosid (IPTG), as is well-known in the art. Due to the bacterial origin of the lac repressor, no further interference with genomic regulation of the host cells occurs. Another example for a suitable repressor protein (tagged with a suitable nuclear localization sequence) to create an operator-regulated eukaryotic promoter is lexA repressor or streptogramin inducible promoter (PipOn, Fussenegger et al, Nature Biotechnology, 18, 1203-1208, 2000). In principle, any such bacterial operator and corresponding repressor protein might be used to create a system similiar to the LacSwitch system. Preferably, an inducer according to the present invention that is binding to the repressor protein and thus isabolishing operator mediated repression, is not an antibiotic. Permanent repression of a eukaryotic promotor by repression with lac repressor protein and a suitable operator sequence comprising lac repressor binding sites is more preferred for p21 or any other cell cycle inhibitor co-expression according to the present invention. Antibiotics are expensive and may given rise to unwanted cellular response due to some toxicitiy. They may inadvertently trigger apoptosis in cytostatically arrested cells. An example of suitable, lac repressor operated p21 -expression cassette comprising the RSV- LTR-Promoter is given in SEQ ID No.l and is a most preferred embodiment of the present invention. The expression cassette of SEQ ID No.l is further described in the examples.

Any strong, constitutively (constantly) expressed promoter is suited for construction of such an operator regulated inducible promoter. Examples of suitable promoters are viral promoters or tissue specific promoters that are active only in certain cell types. A wealth of examples of such tissue specific promoters can be found in US5 589 392. Preferably, such a promoter is a viral promoter such as early and late promoters of the SV40 virus, the immediate early promoter of the human cytomegalovirus (hCMV), the tymidine kinase promoter of Herpes Simplex virus or the RS V-LTR. More preferably, the viral promoter is a promoter that is strongly expressed in Gl -phase thus efficiently driving expression of a functional cell cycle inhibitor as p21, examples being hCMV (Greenaway et al., Gene, 18, 355-360, 1982) or RSV-LTR. Most preferably, the promoter is RSV-LTR as given in SEQ ID No.2 or any functional analogue thereof with at least 95% DNA sequence homology; the RSV-LTR promoter is particularly preferred in conjunction with an myeloma, more specifically in conjunction with an NSO cell line as producer cell line. An expression cassette harboring p21 gene sequence or a variant thereof which sequence is transcribed from RSV-LTR is particularly preferred embodiment of the present invention.

An example of an inducible promotor that is not recruiting RNA polymerase in the absence of inductor is the metallothionein promoter derived from mouse which is induced by addition of heavy metal salts to the medium. An inducible promoter of this type usually comprises enhancer elements stirring activity of the RNA polymerase binding site of the core promoter.

It is also possible to express not only the cell cycle inhibitor protein, but also the recombinant protein from an inducible promoter. It is also possible to express both from the same promotor, as an at least dicistronic expression unit. Such expression unit suitably would further comprise internal ribosome entry sites (IRES). It is also possible to express the recombinant protein transiently or, in the contrary, stably integrated.

It is also possible to direct expression of the recombinant protein, or an antibody in naturally secreting lymphoblastoid cell lines such as hybridoma or EBV transformed lymphoblasts, from a second, independent promoter that can be e.g. constitutive or inducible. Preferably, expression of the recombinant product protein is driven from a second, independent promoter or operon. This allows, optionally, for constitutive expression of the recombinant product protein from a second promoter different from the first one driving p21 expression or induction with a second inducing agent, respectively , thus allowing for different induction kinetics of the recombinant protein. More preferably, such promoter is a constitutive promoter, e.g. a viral promoter or a cell-type specific promoter such as an immunoglobulin promoter as already been described in detail in the context of construction of operator regulated promoters. Expediently, such a promoter should be a strong promoter depending on the producer cell line used. Preferably, such an promoter is a viral promoter such as early and late promoters of the SV40 virus, the immediate early promoter of the human cytomegalovirus (hCMV), adenoviral promoters, early and late promoters of any of the polyoma viruses or papova viruses, the thymidine kinase promoter of Herpes Simplex virus or the RSV-LTR. More preferably, the viral promoter is a promoter that is strongly expressed in Gl -phase thus efficiently driving expression of a functional cell cycle inhibitor such as p21, examples being hCMV (Greenaway et al., Gene, 18, 355-360, 1982) or RSV-LTR. Most preferably, it is RSV- LTR, in particular in conjunction with an myeloma, more specifically an NSO cell line, as producer cell line.

The producer cell line may be any stable, anchorage-independent producer or host cell line such as e.g. hybridoma, EBV immortalized lymphoblasts and myeloma cells such as e.g. the NSO cell line. Preferably, it is a mammalian cell or cell line. ,cell line' refers to stably established cell lines, thus excluding primary cell types. Anchorage independent cells according to the present invention are growing freely suspended in culture medium and to do not rely on substrate or cell-cell contact for growth, thus also excluding semi- independent cell types growing as cellular aggregates. According to the present invention, such anchorage-independent cell lines are to be construed as tumor or artificially immortalized cell lines. Myeloma cells truly are plasmacytoma cells, that is B-cell tumors. Hybridoma cells, including non-secreting hybridoma cells such as SP2/0, are created by cell fusion of a primary cell line with an immortal tumor cell line, namely e.g. a myeloma or pre-existing hybridoma cell line. The latter gives rise to so-called trioma cells,which is to be comprised by the term ,hybridoma cell' in the notion of the present invention. Active or artificial immortalization of cells may be further achieved by viral transforming agents, e.g. SV 40 or Eppstein Barr Virus (EBV) which usually work for certain cell types only, or suitable transforming agents such as e.g. overexpressed oncogenes or suitable combinations of overexpressed oncogenes such as bcl-2 and c-myc. Preferably, a transforming agent is at least one gene product, including antisense RNA expression, used for genetic manipulation of a cell line.

Suitable media and culture methods for mammalian cell lines are well-known in the art, as described for instance in US 5 633 162. Usually, a medium will comprise 1-10 g/1 Glucose, without being limited to this range. Suitable media are for instance Roswell Park Memorial Institute (RPMI) 1640 medium (Morre, G., The journal of the American Medical Association, 199, p.519 f. 1967), L-15 medium (Leibovitz, A. et al., Amer. J. of Hygiene, 78, lp.173 ff, 1963), Dulbecco's modified Eagle's medium (DMEM), Eagle's minimal essential medium (MEM), Ham's F12 medium (Ham, R. et al., Proc. Natl. Acad. Sc.53, p288 ff. 1965) or Isocves' modified DMEM lacking albumin, transferrin and lecithin (Iscoves et al., J. Exp. med. 1, p. 923 ff., 1978). Individual preferences of certain cell types, e.g. lymphoid cells, for certain media are well-known in the art. For instance, RPMI medium is typically used for lymphoid cell types. It is also possible to grow cells according to the present invention under glucose- or glutamin-limiting conditions or both (US 6180 401; Ljunggren, J. et al., Biotech. Letters, 12, p.705-710, 1990).

In a further preferred embodiment, the cell culture medium is free of growth factors, that is free of fetal serum or addition of essentially pure protein growth factors that are triggering signal transduction and cell cycle progression, respectively. Such medium may still comprise transferrin for iron uptake or albumin for rendering lipids in the culture medium accessible to the cells. In such medium, the cytostatic effect of e.g. p21 overexpression is most effective, althought the absence of growth factors might have been expected to render cells more liable to apoptotic death.

Preferably, the cell culture medium is a high-glucose medium comprising at least 2 g/1, preferably 4 g/1, Glucose and is preferably applied at least after induction of p21 expression, i.e. in the production phase. An example for such a medium is high-glucose DMEM, as described for instance in the experimental section.

In a preferred embodiment, the producer cell line according to the present invention is a lymphoid cell line, more preferably it is selected from the group comprising hybridoma cells, lymphoblasts immortalized by EBV or other transforming gene product or myeloma cells, most preferably it is a nonsecreting myeloma cell line and in particular NSO myeloma cell line such as e.g. cell line ECACC No. 85110503 and derivatives thereof, freely available from the European Collection of Cell Cultures (ECACC), Centre for Applied Microbiology & Research, Salisbury, Wiltshire SP4 0JG, United Kingdom. ,Lymphoblast' designate actively antibody or antibody fragment secreting B-cells including fully differentiated plasma cells. NSO cells are specifically of advantage if used with the Glutamine synthetase (GS) expression system (Bebbington et al., 1992, High-level expression of a recombinant antibody from myeloma cells using a glutamine synthetase gene as an amplifiable selctable marker, Bio/Technology 10:169-175; Cockett et al., 1990, High level expression of tissue inhibitor of metalloproteinases in Chinese Hamster Ovary (CHO) cells using Glutamine synthetase gene amplification, Bio/Technology 8: 662-667). NSO cells are renowned for their high productivity. Surprisingly, the yield enhancing effect according to the present invention can even be observed in this cell type; furthermore, despite the proliferation competence of those neoplastic cells, which is usually due to a plurality of abnormally regulated genes as compared to normal cells, cell cycle arrest can be surprisingly reliably achieved by induced expression of a cell cycle inhibitor protein, in particular p21.

In a further preferred embodiment, the producer cell line is a transfected NSO cell line which is expressing glutamine synthetase (GS). The GS-system is one of only two systems that are of particular importance for the production of therapeutic proteins. In comparison to the dihydrofolate reductase (DHFR) system, the GS system, and in particular the GS system used in combination with NSO cells offers a large time advantage during development because highly productive cell lines can often be created from the initial tranfection thus avoiding the need for multiple rounds of selection in the presence of increasing concentrations of selective agent in order to achieve gene amplification (Brown et al., 1992, Process development for the production of recombinant antibodies using the glutamine synthetase (GS) system, Cytotechnology 9:231-236). NSO cells are phenotypically deficient in Glutamine-synthetase. Therefore the NSO cell line which was derived from a mouse tumour cell line (Galfre, G. and Milstein,, C, Methods in Enzymol. 73, 3-75, 1981) is frequently the cell line of choice used in combination with the GS system at an industrial scale.

Suitable producer cell lines according to the above said, capable of expressing p21 protein or an other recombinant cell cycle inhibitor protein, are a further object of the present invention. The description of suitable cell lines in the foregoing applies likewise to this object.

Expression of antibodies or antibody fragments according to the present invention, which antibodies are produced from a hybridoma or EBV transformed lymphoblastoid cell line and whose expression is driven by their natural immunoglobulin promoter, is a further object of the present invention. Accordingly, both a respective method and suitable hybridoma or EBV transformed lymphoblastoid cell lines that are secreting antibodies or fragments thereof or can be stimulated to produce antibodies from their natural promoters and which cell lines are capable of expressing recombinant p21 -protein or a functional mutein thereof are objects of the present invention. The relevant parts of the foregoing description apply likewise to this object of the present invention, in particular those describing immortalization of cell lines by other transforming agents equivalent to EBV. EBV is specifically suited for transforming B-cells.

In a further preferred embodiment of the invention, expression of the cell cycle inhibitor protein is induced at a cell density of about 10⁵ cells/ml, preferably at about 10⁶ cells/ml or above. Such culture condition is called high-density culture. Such culture is at higher risk of being limited by sufficient tranfer of oxygen and supply of nutrients. Surprisingly, the yield enhancing effect of recombinant cell cycle inhibitor co-expression persists under those conditions and is not jeopardized e.g. apoptosis or adaption to stationary phase. High-density culture usually requires suitable bioreactors, as is well-known in the art, as well as fortified culture media e.g. as described in GB 2251 249 or feed regimens. More preferably, expression of the cell cycle inhibitor is induced at the above said high cell densities in a perfusion reactor system. Perfusion bioreactors are well-known in the art and allow for extended culture periods due to a constant supply of fresh medium passing throught a culture compartment. Product protein commonly is constantly removed, though this not being a limiting condition in the present context. Most preferably, high density culture according to the present invention in a perfusion reactor system is accompanied by further co-expression of recombinant bcl-2 or another apoptosis inhibitor protein, bcl-2 according to the present invention can be any naturally occuring eukaryotic, preferably mammalian, bcl-2 protein that can act as a suppressor of apoptosis (Tsujimoto, Y. et al., Analysis of the structure, transcripts and protein products of bcl-2, the gene involved in human follicular lymphoma, PNAS 83, 5214-5218, 1986; Hockenbery, D. et al., bcl-2 is an inner mitochondrial membrane protein that blocks programmed cell death, Nature 348, 6299, 1990). By definition, this includes anti-apoptotically functional, natural isoforms of bcl-2 (e.g. bcl-XL) as well as variants or muteins.

According to the present invention, it is also possible that the cell culture medium further comprises components known to enhance the expression level of antibodies in hybridoma cell culture or of other recombinantly expressed proteins in cell culture such as alkanoic acids. Preferably, butyric acid or a salt thereof is added to the cell culture medium, more preferably it is added in an amount of from 1 mM to 20 mM.

Examples

Example 1

The effect of inducible over-expression of ectopic p21^cπ>1 in anchorage-independent NSO 6A1 myeloma cells was investigated. The parental cell line NSO 6A1 was supplied by LONZA Biologies (Slough, UK) and had been previously engineered to constitutively express chimeric B72.3 IgG4 antibody, using the GS system (Bebbington et al., 1992).

Copies of p21 cDNA were introduced into 6A1 cells using the LacSwitch inducible system II (Stratagene Corp.,LaJolla, CA/U.S.A.; described in patent US 5 589 392). The LacSwitch system II comprises a first plasmid pCMVlacI encoding a modified bacterial lac repressor harboring a nulear localization sequence that is under control of a Cytomegalo virus promoter, and further comprises a second plasmid pOPRSVl into whose multiple cloning site the p21 coding sequence was engineered. An 855bp EcoRI fragment containing the mouse _p2r^AF1/cπ>1 cDNA (Huppi, K. et al., 1994, Oncogene 9, 3017-3020) was subcloned blunt-ended into the Notl site of the pOPRSVl Lacswitch- vector. In the vector construct, p21 expression is driven by Rous-Sarcoma-Virus (RSV)-LTR promoter that is followed, immediately downstream and prior to the p21 coding sequence, by interspersed multiple ideal lac repressor binding sites, so called operator sequences. The entire sequence of the expression cassette of pOPRSVl /p21 comprising the LTR promoter, the operator (engineered within the SV 40 small t antigen intron sequence), the p21 coding sequence and the thymidine kinase (TK) polyadenylation site is given in table 1 below.

Table 1 RSV-LTR Promotor 1-563

SV40 small t intron 564 - 1051

P21 1052 - 1907

TK Poly (A) 1908- 2245

1 GGTGGCAGTA CAATCTGCTC TGATGCCGCA TAGTTAAGCC AGTATCTGCT

51 CCCTGCTTGT GTGTTGGAGG TCGCTGAGTA GTGCGCGAGC AAAATTTAAG 101 CTACAACAAG GCAAGGCTTG ACCGACAATT GCATGAAGAA TCTGCTTAGG

151 GTTAGGCGTT TTGCGCTGCT TCGCGATGTA CGGGCCAGAT ATACGCGTAT

201 CTGAGGGGAC TAGGGTGTGT TTAGGCGAAA AGCGGGGCTT CGGTTGTACG

251 CGGTTAGGAG TCCCCTCAGG ATATAGTAGT TTCGCTTTTG CATAGGGAGG

301 GGGAAATGTA GTCTTATGCA ATACTCTTGT AGTCTTGCAA CATGGTAACG 351 ATGAGTTAGC AACATGCCTT ACAAGGAGAG AAAAAGCACC GTGCATGCCG

401 ATTGGTGGAA GTAAGGTGGT ACGATCGTGC CTTATTAGGA AGGCAACAGA

451 CGGGTCTGAC ATGGATTGGA CGAACCACTG AATTCCGCAT TGCAGAGATA

501 TTGTATTTAA GTGCCTAGCT CGATACAATA AACGCCATTT GACCATTCAC

551 CACATTGGTG TGCACCAGAT CTAAGCTTGG ACAAACTACC TACAGAGATT 601 TAAAGCTCTA AGGTAAATAT AAAATTTACT AGGTTGTGGA ATTGTGAGCG

651 CTCACAATTC CACAGTCGAC CCTAGGTTGT GGAATTGTGA GCGCTCACAA

701 TTCCACAGTC GACCCTAGGT TGTGGAATTG TGAGCGCTCA CAATTCCACA

751 GTCGACCCTA GTGTATAATG TGTTAAACTA CGGATCCGTC TCCCATTAGG

801 CCTACAATGG TGAGACAAGT AGCCAACAGG GAAGGGTTGC AAATATCATT 851 TGGGCACACC TATGATAATA TTGATGAAGC AGACAGTATT CAGCAAGTAA

901 CTGAGAGGTG GGAAGCTCAA AGCCAAAGTC CTAATGTGCA GTCAGGTGAA

951 TTTATTGAAA AATTTGAGGC TCCTGGTGGT GCAAATCAAA GAACTGCTCC

1001 TCAGGGATCC TAATTGTTTG TGTATTTTAG ATTCCAACCA AGCTTGCGGC

1051 CCTGGGGTAA ACAGGACGGT GACTCCTACT TCTGTGGACA TCACCCGTGA 1101 CCTTGGGGTG CAGGGCTGGC TGAACTCAAC ACCCACCTTA GTCTCATGGT

1151 GTGGTGGAAA AGCACCTGCA AGACCAGAGG GAGCCTGAAG ACTGTGATGG

1201 GGTAGTTTCC ATAGTGACCC GGGTCCTTCT TGTGTTTCAG CCACAGCGAC

1251 CATGTCCAAT CCTGGTGATG TCCGACCTGT TCCGCACAGG AGCAAAGTGT 1301 GCCGTTGTCT CTTCGGTCCC GTGGACAGTG AGCAGTTGCG CCGTGATTGC

1351 GATGCGCTCA TGGCGGGCTG TCTCCAGGAG GCCCGAGAAC GGTGGAACTT

1401 TGACTTCGTC ACGGAGACGC CGCTGGAGGG CAACTTCGTC TGGGAGCGCG

1451 TTCGGAGCCT AGGGCTGCCC AAGGTCTACC TGAGCCCTGG GTCCCGCAGC 1501 CGTGACGACC TGGGAGGGGA CAAGAGGCCC AGTACTTCCT CTGCCCTGCT

1551 GCAGGGGCCA GCTCCGGAGG ACCACGTGGC CTTGTCGCTG TCTTGCACTC

1601 TGGTGTCTGA GCGGCCTGAA GATTCCCCGG GTGGGCCCGG AACATCTCAG

1651 GGCCGAAAAC GGAGGCAGAC CAGCCTGACA GATTTCTATC ACTCCAAGCG

1701 CAGATTGGTC TTCTGCAAGA GAAAACCCTG AAGTGCCCAC GGGAGCCCCG 1751 CCCTCTTCTG CTGTGGGTCA GGAGGCCTCT TCCCCATCTT CGGCCTTAGC

1801 CCTCACTCTG TGTGTCTTAA TTATTATTTG TGTTTTAATT TAAACGTCTC

1851 CTGATATACG CTGCCTGCCC TCTCCCAGTC TCCAAACTTA AAGTTATTTA

1901 AAAAAACGGC CGCAATTCTT AAGAACTGAA ACACGGAAGG AGACAATACC

1951 GGAAGGAACC CGCGCTATGA CGGCAATAAA AAGACAGAAT AAAACGCACG 2001 GTGTTGGGTC GTTTGTTCAT AAACGCGGGG TTCGGTCCCA GGGCTGGCAC

2051 TCTGTCGATA CCCCACCGAG ACCCCATTGG GGCCAATACG CCCGCGTTTC

2101 TTCCTTTTCC CCACCCCACC CCCCAAGTTC GGGTGAAGGC CCAGGGCTCG

2151 CAGCCAACGT CGGGGCGGCA AGCCCTGCCA TAGCCACGGG CCCCGTGGGT

2201 TAGGGACGGG GTCCCCCATG GGGAATGGTT TATGGTTCGT GGGGG

Cells were maintained in high glucose DMEM medium (Gibco, Paisley, UK) supplemented with 10 % fetal calf serum (FCS, First link Ltd., UK), ImM sodium pyruvate (Gibco) and 10 μM methylamine sulphoximine (MSX, Sigma UK). After electroporation at 300 V and at a capacitance of 450 μF (BioRad Gene pulser), pCMVlacI transfectants were selected in a medium containing 400 μg/ml Hygromycin B (Sigma). After having chosen one clone E9-10D showing constitutive high level expression of Lac repressor protein, from this clone pOPRSVl /p21 transfectants were selected in 400 μg/ml Geneticin 418 mediated lipofection with DOTAP (Roche UK). Expression of p21 protein was induced by adding 5mM isopropyl-^β-D-thio-galactoside (IPTG, Gibco) into culture medium. From approximately 100 clones, 12 clones displaying IPTG inducible effects on proliferation and p21 expression were obtained. Four of these clones, namely 4-9F, 1-4A, 1-3D and 2-4 G, were chosen for further study. Fig. 1 depicts the effect of ectopic p21 expression on cell proliferation and viability in batch culture. Cell cultures were incubated in the absense and presence of IPTG for 6 days. Clone 4-9F, 1-3D, 1-4A and 2-4G cultures were set up at a density of approximately 2 x 10⁵ cells/ml and incubated for 6 days, either without IPTG (•) or with 5 mM IPTG (o). Viable cells were counted at 24 hours intervals by Trypan blue exclusion. In the absence of IPTG these clones proliferated to a maximum cell density of approximately 1.4 x 10⁶ cells/ml (Fig. 1 A). Induction of p21 expression caused complete or partial inhibition of cell proliferation in the cultures. Over-expression of ectopic p21 had very little effect on cell viability. Fig. 1 B shows the viability of p21 induced cell lines following the addition of IPTG. A particularly pronounced drop of viability of 4-9F cultures was also observed for this clone in the absence of IPTG (data not shown). In the other clonal cultures, cell viability remained relatively high, dropping from approx. 90 % to <80% over the the period of 6 days in the presence of IPTG. Thus the effect of p21 is cytostatic, not cytotoxic.

Fig. 2 shows Western-Blot analysis of ectopically expressed p21 in NSO 4-9F and 1-4A cells. Western Blots were carried out on total protein from approximately 2x10⁵ cells lysed in SDS sample buffer separated by 12 % SDS-PAGE and electroblotted to nitrocellular membrane. Membranes were probed with anti-p21 antibody (1:1000 of mouse monoclonal F5, Santa Cruz Biotechnology Inc., UK) in TBST/2% dried skimmed milk. Blots were developed using horse radish conjugated antibodies and ECL detection (Amersham). Expression after 24 hours and 48 hours with (+) and without (-) addition of IPTG are shown (upper panel). For normalization with regard to total protein, the level of alpha- tubulin was assessed in each sample by reprobing the same blot with monoclonal anti- alpha-tubulin antibody (lower panel). For comparison, parental cell line 6A1 is shown incubated without IPTG for 24 hours (upper panel). Expression of ectopic p21 was found to be tightly regulated, and could not be detected in non-induced cells. The expression levels of other proteins were unaffected by IPTG treatment.

Fig. 3 shows the cell cycle analysis of p21 induced cells. 4-9F and 1-4 A cultures were grown for 72 hours in the presence of 5mM IPTG. For cell cycle analysis, 1 x 10⁶ cells from batch cultures were fixed with 70%ice-cold ethanol and stored at -20°C until use. Cells were stained with propidium iodide (PI, Sigma UK) after an incubation with 1 ml RNase (50 μg ml^"1, Sigma) for 20 min. at 37°C. The DNA content of cells was measured using a Flow Cytometer (Coulter EPICS Elite analyser), and cell cycle distribution was calculated using Multicycle software program (Phoenix Flow Systems). The distribution of cells in each phase of the cell cycle is indicated for (A) 4-9F at 0 hours, (B) 4-9F cells at 72 hours. The left shoulder of the Gl peak represents apoptotic cells; (C)1-4A at 0 hours and (D), 1-4 A at 72 hours.

Fig. 4 shows the effect of over-expression of p21 protein on chimeric B72.3 IgG4 antibody production. The clones 4-9F and 1-4A constitutively produce chimeric B72.3 IgG4 antibody. The cell lines were incubated in the absence and presence of IPTG and supernatants harvested at intervals. The concentration of the human-mouse chimeric Mab in cell culture supernatant harvested at 72 h was assayed by enzyme linked immunosorbent assay (ELISA). Cell culture supernatants were applied to ELISA plates, which were coated with monoclonal anti-human IgG (Fc specific, Sigma UK). Peroxidase-conjugated anti- human kappa light chain antibody (Sigma) was used as a secondary antibody. O- phenylendiamine dihydrochloride (OPD, Sigma) was used as a substrate and the absorbance was detected at 492 nm. The antibody concentration of the standard was analysed by HPLC and then the concentration in each curve was calculated from a standard curve. The productivity per cell was calculated at 24, 48 and 72 hours; the amounts of antibody produced are expressed in pg per cell and day for proliferating and p21 -arrested (IPTG induced) cells. In 4-9F cells productivity significantly increased by >4fold at 72 hours after the addition of IPTG (Fig.4), whereas 1-4A cells showed an increase of approx. 2-fold. The over-expression of p21 protein and ensuing cell cylce arrest enhanced productivity in both clones.

Fig. 5-7: Cell line NSO 6A1-49F was constructed essentially as described above to express chimeric antibody (cB72.3 IgG4) using GS selectable marker and to inducibly express p21^cπ>1 using the Lacswitch system. The cell line was then further transfected with the expression vector pEF(bcl-2-Y28A)-2pGKpuro using the Lipofectin reagent and stable transfectants were selected in the presence of puromycin (Huang et al., 1997, The anti- apoptosis function of bcl-2 can be genetically separated from its inhibitory effect on cell cycle entry, EMBO J., 16, 4628-4638). Thus the cell line produces a chimeric antibody (cB72.3 IgG4) and contains the glutamine synthetase (GS) marker gene plus bcl-2 and p21^clpl genes. Cells were maintained in DMEM F12 w/o glutamine supplemented with 10% FCS, 10 μM MSX, 400 μg/ml hygromycin B, 4 μg/ml puromycin and 400 μg/ml geneticin 418.

Expression of p21^cπ>1 protein was induced by adding 5 mM IPTG into culture medium. A standard 2 L glass perfusion bioreactor with 1 L working volume equipped with a spin filter was used. The dissolved oxygen was controlled at 50 % air saturation. The pH and the temperature were maintained at 6.9 and 37°C respectively. The perfusion rate was fixed at 0.5 d^"1 from day 3 onward. Viable cell density was determined by Trypan Blue exclusion and verified by dual staining of cells with acridine orange and propidium iodine for distinguishing viable from both necrotic and apoptotic cells. Fig. 5 shows the effect of ectopic p21 expression on NS06A1-49F proliferation in perfusion system with and without supplementation of IPTG. Fig. 6 shows antibody production during perfusion culture with and without IPTG. Fig. 7 shows specific antibody production rate. IPTG induction of cell culture at day 19 both increased product titer and specific productivity consistingly about 4-fold. End of IPTG exposure precisely abolished enhanced secretion.

Claims

Claims:

1. Method for production of a protein, preferably a recombinant protein, in a mammalian anchorage-independent producer cell line, comprising the step of co-expressing with said protein in the producer cell line a recombinant cell cycle inhibitor protein, preferably p21, or a functional mutein thereof, wherein the recombinant cell cycle inhibitor protein is expressed from an inducible promoter.

2. Method according to claim 1, characterized in that the cell cycle inhibitor, preferably p21, is transiently expressed from an episomal vector element.

3. Method according to claim 2, characterized in that the inducible promotor is a viral promoter or a functional derivative thereof under control of the lac repressor.

4. Method according to claim 1 or 2, characterized in that the p21 and the recombinant protein are expressed from different operons.

5. Method according to claim 4, characterized in that the recombinant protein is expressed from a second promotor and that the second promoter is a constitutively expressed promotor in said producer cell line.

6. Method according to claim 1, characterized in that the medium comprises butyric acid or a salt thereof.

7. Method according to claim 2, characterized in that expression of p21 is induced at a cell density of 10⁵ cells/ml or above.

8. Anchorage independent producer cell line capable of co-expressing a product protein, preferably a recombinant protein, and a recombinant cell cycle inhibitor protein or a functional mutein thereof, wherein the cell cycle inhibitor protein is expressed from an inducible promoter.

9. Cell line according to claim 8, characterized in that the recombinant protein and the cell cycle inhibitor, preferably p21 protein, or the mutein are expressed from different operons.

10. Cell line according to claim 8 or 9, characterized in that the product protein is a recombinant protein and that the product protein is a secreted protein.

11. Cell line according to claim 8-10, characterized in that the cell line is a lymphoid cell line, preferably a myeloma cell line.

12. Cell line according to claims 11, characterized in that the cell line is a NSO cell line, preferably a NSO GS-cell line.

13. Immortalized lymphoblastoid cell line capable of expressing a recombinant cell cycle inhibitor, preferably p21 protein, or a functional mutein thereof, said B-lymphoid cell line secreting an antibody or a fragment thereof.

14. Cell line according to claim 13, characterized in that the cell line is a hybridoma cell line generated by cell fusion or an EBV-immortalized lymphoblastoid cell line.

15. Cell line according to claims 13 or 14, characterized in that the recombinant cell cycle inhibitor is under control of an inducible promotor.

16. Method for production of an antibody or an antibody fragment from a hybridoma host cell line or EBV immortalized lymphoblastoid B-cell line, comprising the step of co- expressing recombinant cell cycle inhibitor, preferably recombinant p21 protein, or a functional mutein thereof together with the antibody secreted by said host cell line.