SE546587C2 - Modified ctcf-bindning sequence and expression vector comprising the same - Google Patents

Abstract

The invention provides a nucleic acid sequence comprising a modified CTCF-binding sequence represented by:TGCAGTACCTCCCTN1N2N3CCAGCAGGN4GGCAN5N6AGN7GAAN8GGTGAACTGGAGT(SEQ ID NO: 10)whereinN1, N2, N3, and N5 each independently is C or G;N4 is T or G;N6 and N7 each independently is C or T; andN8 is T or A.Each of N1 to Ns represents a nucleic acid substitution at a corresponding locus on the CTCF-binding site (SEQ ID NO: 1) of the wild-type chicken Hypersensitivity Site 4 (cHS4) insulator. Provided are also expression vectors and recombinant cells incorporating the modified CTCF-binding sequence, and an associated method of producing at least two polypeptides.

Description

TECHNICAL FIELD The present invention relates to expression vector constructs useful in protein production in mammalian host cells. More specifically, the present invention relates to nucleic acid sequences and expression vector constructs comprising spacer elements that can be placed between two or more expression cassettes within an expression vector. The invention also relates to cells and cell lines comprising such constructs, and to a method of recombinantly producing a protein of interest.

BACKGROUND Plasmids are small circular extra-chromosomal DNA ranging in size from 1 to over 200 kbp and found occurring naturally in several bacteria. A single bacterial cell may contain several different plasmids and hundreds of copies of each of those plasmids residing within it. The most useful property of a plasmid is its ability to replicate autonomously and be stably maintained within a bacterial cell line. This property of plasmids makes them highly suitable as a tool for transferring foreign genetic material into bacterial host cells. A plasmid usually comprises several genetic elements such as origin of replication, replication initiation gene and antibiotic resistance genes. Smaller plasmids generally rely on the replication machinery of the bacterial host cell whereas larger plasmids may carry their own specific replication genes.

Genetícally engineered artificial plasmid vectors are one ofthe most commonly used vectors for introducing genes of interest into host cells. Such engineered plasmids are widely used as cloning vectors and are designed to have genetic elements that allow for the insertion of genes of interest within them. When bacterial host cells are transformed by such a plasmid vector, replication of the plasmid vector begins within the cell, resulting in an increasing copy number of the plasmid and thereby an increase of the number of the inserted gene of interest. Transient expression of said genes from plasmid can be done in the bacterial host cells but this usually generates low quantities of protein.

A better approach to expressing genes of interest is by using expression vectors. Expression vectors can be circular DNA or linear DNA fragments that have been engineered to contain the sequence(s) of the genes of interest in the form of expression cassettes. An expression vector typically also contains several other design features or components which are engineered in the expression vector sequence such as regulatory genes that encode promoters and enhancers for effective transcription of the gene(s) of interest to produce stable messenger RNA (mRNA) which can then be translated into a protein in mammalian host cells. However, it is often seen that replication of such expression vectors in mammalian cells is unsatisfactory.

One way to overcome the above limitation and improve the expression of the genes of interest is to integrate the expression vector construct within the genome of the mammalian host cells when the host cells are transfected with said expression vector. This could be done by including restriction endonuclease binding sites into the design of the expression vector to facilitate integration of the expression vector within the genome of the mammalian host cells.

Several expression vectors have been developed over time with the aim of further improving protein expression by incorporating different promoter elements and other elements like insulators and spacers in the vector construct. insulators are a class of DNA sequence elements that possess a common ability to protect genes from undesirable signals emanating from their surrounding environment. There are two ways in which insulators protect an expressing gene from its surroundings. The first way is by blocking the action of a distal enhancer on a promoter. However, enhancer blocking only occurs if the insulator is situated between the enhancer and the promoter. Such activity can prevent an enhancer from activating expression of an adjacent gene from which it is blocked, while leaving it free to stimulate expression of genes located on its unblocked side. The second way in which insulators protect genes is by acting as "barriers" that prevent the extension of nearby condensed chromatin domain into a transcriptionally active region that might otherwise silence the gene expression. Some insulators can act both as enhancer blockers and as barriers. For example, in the H54 (Hypersensitivity Site 4, HS4) insulator found at the 5' end of the chicken ß- globin locus (referred to hereinafter as cHS4), the two activities can occur together but are separable. The cHS4 insulator marks the border between the active euchromatin in the chicken ß- globin locus and the upstream heterochromatin region that is highly condensed and inactive. lncorporation of cHS4 insulator, WPRE, and SAR elements in vector constructs for improving protein expression in eukaryotic cells has been demonstrated (Ali Ramezani et. al., "Performance and safety- enhanced lentiviral vectors containing the human interferon-ß scaffold attachment region and the chicken ß-globin insulator", Blood 2003 (101:4717-4724)).

The enhancer blocking activity and the barrier activity of cHS4 insulator within the 1.2kb full-length of cHS4 insulator sequence have been mapped to a 250 bp "core" element as shown in Figure 1A. The core element ofthe cHS4 insulator contains five protein binding sites/footprints (Fl-FV) for three different insulator proteins: CTCF (Fll), VEZF1(FI, Flll, and FV), and USF1/USF2 (FIV). The CTCF-binding site or the footprint ll (Fll) is necessary and sufficient for enhancer blocking activity but can be deleted from cHS4 insulator sequence without affecting the barrier activity. The four remaining protein binding sites are all essential for barrier activity (Fl, Flll, FIV and FV) but dispensable for enhancer blocking activity. For isolation between expression cassettes and recruitment of transcription factors, use of two sequential copies of the core element, called the "double core", as shown in Figure 1B has been known. Qin et. al. demonstrated use ofa strong promoter EF1 and spacers between expression cassettes in order to isolate gene expression (Qin JY, Zhang L, Clift KL, Hulur I, Xiang AP, et al. (2010) Systematic Comparison of Constitutive Promoters and the Doxycycline-lnducible Promoter. PLoS ONE 5(5): e10611).

An example of a commercially available plasmid expression vector is the pcDNATM 3.1(+) vector from ThermoFisher Scientific as shown in Figure 2, that has been used for several decades for protein expression in mammalian cells. Vector pcDNATM 3.1(+) is a simple monocistronic expression vector and contains an E. coli propagation cassette, a selection marker cassette, and an expression cassette for the protein of interest. Although this vector has been used to establish high-titer clones, a successful result depends on extensive clone screening and the vector design is not practical for production of more than one polypeptide which is sometimes required, for example when the protein is a monoclonal antibody (mAb) which requires two polypeptide chains to be expressed namely the heavy chain (HC) and the light chain (LC) of the mAb.

Other known expression vectors suffer from low expression due to low integration efficiency. Scientists in the field have been trying to find ways to improve vector constructs so that a cost efficient and effective cell line development process can be established which produces stable clones with high expression levels.

Thus, there is a need of improved expression vector constructs.

SUMMARY OF THE INVENTION lt is an object of the invention to overcome or at least partly alleviate drawbacks of the prior art.Accordingly, it is an object of the invention to provide improved nucleic acid sequences and expressions vectors that are useful in the production of recombinant proteins in mammalian cell lines.

This and other objects are achieved by a nucleic acid sequence represented by: TGCAGTACCTCCCTNlNzNgCCAGCAGGN4GGCAN5N6AGNyGAANgGGTGAACTGGAGT (SEQ ID NO: 10) wherein N1, NZ, N3, and NS each independently is C or G; N4 is T or G; NG and Ny each independently is C or T; and NgisTorA.

SEQ ID NO: 10 represents a CCCTC bínding factor (CTCF) bínding sequence and is modified in relation to the CTCF-bínding sequence of the wild type chicken Hypersensitivity Site 4 (cHS4) insulator. Each of Nl to Ng in SEQ ID NO: 10 represents a nucleic acid substitution at a corresponding locus on the CTCF bínding site (SEQ ID NO: 1) of the wild type cHS4 insulator. Hereinafter, SEQ ID NO: 10 is referred to as a "modified CTCF-bínding sequence". Preferably, the modified CTCF-bínding sequence may be selected from SEQ ID NO: 2 and SEQ ID NO: The nucleic acid sequence may be an isolated nucleic acid sequence.

The nucleic acid sequence may comprise a modified cHS4 core element including said modified CTCF- binding DNA sequence, which is also referred to as a modified footprint II (FII) of the cHS4 core element. The modified cHS4 core element may, in addition to said modified CTCF-bínding DNA sequence, contain footprints I, III, IV and V, which provide bínding sites for VEZF1(FI, FIII, and FV), and for USF1/USF2 (FIV).

In embodiments, the nucleic acid sequence may comprise two HS4 core elements, each comprising a CTCF-bínding sequence, at least one of which being a modified CTCF-bínding sequence in accordance with the present disclosure (SEQ ID NO: 10).

The present invention offers the potential to improve expression of polypeptides from multiple expression cassettes provided in the same expression vector. In relation to the known cHS4 core element, the present invention provides the additional benefit that it allows some variability in the nucleotide sequence of the CTCF-binding motif. For example, either the modified CTCF-binding DNA fragment A (SEQ ID NO: 2) or the modified CTCF-binding DNA fragment B (SEQ ID NO: 3) may be used. This variability allows integration of multiple copies of a CTCF-binding DNA fragment, such as in the form of a modified cHS4 double core element, with less risk for genome instability due to integration of multiple copies of identical sequences.

In a HS4 double core element at least one of said HS4 core elements may thus comprise a CTCF- binding sequence selected from SEQ ID NO: 2 and SEQ ID NO: 3. Optionally, in a HS4 double core element one of said HS4 core elements may be represented by SEQ ID NO: 4, which corresponds to the wild type cHS4 core element. For example, in a HS4 double core element, one of the HS4 core elements may comprise a CTCF-binding sequence represented by SEQ ID NO: 1 and one of said HS4 core elements may comprise a CTCF-binding sequence selected from SEQ ID NO: 2 and SEQ ID NO: 3. Alternatively, one of said HS4 core elements may comprise a modified CTCF-binding sequence according to SEQ ID NO: 2 and the other of said HS4 core elements may comprise a modified CTCF- binding sequence according to SEQ ID NO: Thus, in a HS4 double core element at least one of the HS4 core elements may contain at least one CTCF-binding motif according to either of SEQ ID NO 2 and 3. For example a double core element may have a sequence according to SEQ ID NO: 5, SEQ ID NO: 6 or SEQ ID NO: 7 (described in more detail hereinbelow). Alternatively a double core element may have a sequence that is based on any one of SEQ ID NOs: 4-7, but where one ofthe CTCF-binding motifs has been replaced with a different CTCF-binding motif, such that at least one of the two CTCF-binding motifs is a modified CTCF-binding sequence of the present disclosure. For example, SEQ ID NO: 4 may be modified to contain one CTCF- binding sequence according to SEQ ID NO: 2 or 3, while preserving one wild-type CTCF-binding motif (SEQ ID NO: 1). As another example, SEQ ID NO: 5 may be modified to contain one CTCF-binding motif according to SEQ ID NO: 1 or 3, while preserving one CTCF-binding fragment A (SEQ ID NO: 2). As yet another example, one of SEQ ID NOs: 6 or 7 may be modified to contain one CTCF-binding motif according to SEQ ID NO: 1 or 2, while preserving one CTCF-binding fragment B (SEO ID NO: 3).

The nucleic acid sequence may further comprise a spacer sequence of at least 100 nucleotides in length located downstream of a modified cHS4 core element. In the case of a HS4 double core element, a spacer sequence of at least 100 nucleotides in length may be located downstream of the two HS4 core elements, i.e. downstream of the double core element, in the 5' to 3' direction. The spacer sequence may have a length of at least 200 nucleotides, and optionally up to 500 nucleotides, such as up to 300 nucleotides. The additional spacer sequence may be a non-coding sequence or a non-functional sequence, and may be a random sequence. The additional spacer sequence should preferably not contain any sequence motifs known to interact with a DNA binding protein.

The present invention further provides an expression vector comprising the nucleic acid sequence as described herein, useful for expression of at least two non-identical polypeptides. ln particular, the expression vector may comprise: a first expression cassette comprising a first coding sequence encoding a first protein of interest, a second expression cassette comprising a second coding sequence encoding a second protein of interest, and the nucleic acid sequence as described above arranged between said first expression cassette and said second expression cassette. The nucleic acid sequence comprising the CTCF-binding motif thus forms a spacer sequence. The expression vector may comprise further expression cassettes or sequences useful for vector propagation, genomic integration, gene expression, gene product processing, selection and/or purification. For example, the expression vector may comprise a third expression cassette comprising a selection marker. The expression vector may further comprise recombinase sites for integration into the genome of a host cell. The expression vector may be provided as a circular vector, or it may be linear or linearized.

The first and second proteins of interest typically form part of the same protein complex, which may be a therapeutic protein complex, such as an antibody or an antibody fragment formed of two separate polypeptide chains. Thus, the first protein ofinterest may be a first polypeptide chain of an antibody or antibody fragment, and the second protein of interest may be a second polypeptide chain of an antibody or antibody fragment. As a straightforward example, the first protein of interest may represent the heavy chain, and the second protein of interest may represent the light chain of an antibody, such as a monoclonal antibody. However, given the multitude of antibody types and fragments, other variants are also conceivable. Thus, for example, the first polypeptide chain may comprise a variable domain of an antibody, such as ofan antibody heavy chain, and the second polypeptide chain may comprise another variable domain of an antibody, such as of an antibody light chain.

The first expression cassette and/or the second expression cassette may comprise a eukaryotic promoter sequence, such as a CMV, EFlalpha or SV40 PGK promoter. Preferably first expression cassette and/or the second expression cassette may comprise a CMV promoter.

The present invention further provides a cell, a cell line, or a cell culture, comprising the expression vector described herein. The cell may be an in vitro growing cell. The cell may be a mammalian cell, such as a Chinese Hamster Ovary (CHO) cell, transfected with the expression vector. The expression vector may be integrated, preferably stably integrated, into the cell genome.

The present invention further provides a method of recombinantly producing two polypeptides, comprising i) transfecting mammalian cells, such as CHO cells, with an expression vector comprising expression cassettes encoding first and second proteins of interest as described herein, and ii) culturing the transfected cells in cell culture media under conditions enabling expression ofthe first and second proteins of interest. The method may further comprise purifying the polypeptides from the cell culture, according to known methods. The two polypeptides may be as described above, for instance two polypeptide chains of a monoclonal antibody or of an antibody fragment.

As used herein, the terms "peptide" and "polypeptide" are used synonymously herein to refer to compounds formed of sequences of amino acids, without restriction as to size. "Protein" may be used to refer to the larger compounds of this class.

The term "wild type" or "wt" refers to the typical naturally occurring form.

Preferred aspects ofthe present disclosure are described below in the detailed description and in the dependent claims. lt is noted that the invention relates to all possible combinations of features recited in the claims.

Use of the verb "to comprise" and its conjugations does not exclude the presence of elements or steps other than those stated. The article "a" or "an" preceding an element does not exclude the presence ofa plurality of such elements. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

BRIEF DESCRIPTION OF THE DRAWINGS This and other aspects of the present invention will now be described in more detail, with reference to the appended drawings, in which: FIG. 1A (see Background) shows a schematic representation of the core element of the wild type chicken Hypersensitivity Site 4 (cHS4) insulator.

FIG. 1B (see Background) shows a schematic representation of the cHS4 double core formed by placing two core elements of the wild type cHS4 insulator one after the other.

FIG. 2 (see Background) shows a map of the known monocistronic plasmid expression vector pcDNATM3.

FIG. 3 shows a map of the parental/base bicistronic modular expression vector ofthe present invention in circular configuration for random integration.

FIG. 4 shows the parental/base bicistronic modular expression vector of Fig. 3 in Iinear configuration.

FIG. 5 shows a map of the parental/base bicistronic modular expression vector ofthe present invention in circular configuration for site directed integration.

FIG. 6 shows the parental/base bicistronic modular expression vector of Fig. 5 in Iinear configuration.

FIG. 7 shows the flow cytometry data pertaining to cells transfected with expression vector pGEO308 pertaining to experiment 1. The left panel shows a forward scatter to side scatter plot used to gate viable single CHO cells (N-gate). The right panel shows a mTagBFP2 to eGFP plot used to calculate % expression of both eGFP and mTagBFP2 (P-gate) from CHO cells in gate N.

FIG. 8 shows a plot with the percentage of expressing recombinant CHO cells from gate P pertaining to experiment 1. The x-axis shows functional Spacer A-D with different length of additional spacer X.

"C" represents a control with only a 100bp spacer X (no double core element).

FIG. 9 shows the flow cytometry data pertaining to cells transfected with expression vector pGEO359 pertaining to experiment 2. The left panel shows a forward scatter to side scatter plot used to gate viable single CHO cells (F-gate). The right panel shows a mTagBFP2 to eGFP plot used to calculate % expression of both eGFP and mTagBFP2 (M-gate) from CHO cells in gate F.

FIG. 10 shows a plot of the percentage of expressing recombinant CHO cells from gate M pertaining to experiment 2. The x-axis shows functional Spacer A-C with different length of additional spacer.

FIG. 11 shows the flow cytometry plots used for calculation of mTagBFPZ and eGFP expression for experiment 3. Left panel shows a forward scatter to side scatter plot used to gate viable cells (A- gate). The middle panel shows a side scatter to mRasberry plot used to gate mRasberry positive cells corresponding to cells having undergone integration at the LP (B-gate). The right panel shows a mTagBFP2 to eGFP plot used to calculate mean expression values based on the fluorescence signals of the main focused population (C-gate).

FIG. 12 shows the normalized expression of functional spacer A with two different 200 bp additional spacers (a and ß) and with only 200 bp spacer ß without the double core element. Error bar = +/-standard deviations.

As illustrated in the figures, some features may be exaggerated for illustrative purposes and, thus, are provided to illustrate the general structures of embodiments of the present invention.

DETAILED DESCRIPTION Several experiments were conducted to construct a modular expression vector for enhanced expression of polypeptides of interest where different components of said expression vector, such as expression cassettes, promoters and selection markers could be easily replaced or exchanged to reconfigure said expression vector for production of a specific polypeptide of interest. Although the improvement in expression is exemplified mostly by using random integration (RI) vectors, the findings are also applicable for site directed integration (SDI).

Figure 2 shows a map of the known monocistronic plasmid expression vector pcDNATM3.1 for expression in a variety of mammalian cell lines. This vector was used as the starting point for designing the modular bicistronic expression vector ofthe present invention.

Figure 3 shows a map of the parental/base bicistronic modular expression vector for random integration in circular configuration. The bicistronic expression vector was designed for expression of two polypeptides. As shown in Figure 3, the parental/base expression vector comprises a standard E. coli propagation cassette as in Figure 2 (not shown in picture). The selection marker is represented as expression cassette 1. Expression cassette 2 comprises an eGFP encoding gene and expression cassette 3 comprises an mTagBFP2 encoding gene. In these examples, the two polypeptides expressed are eGFP and mTagBFPZ. As further shown in Figure 3, expression cassettes 2 and 3 have spacer elements between them to isolate gene expression from said cassettes to allow improved expression of the polypeptides of interest. Figure 3 also shows promoters 2 and 3 to improve expression from the expression cassettes 2 and 3 respectively. Figure 4 shows the same parental/base bicistronic modular expression vector in linear configuration.

Figure 5 shows a map of the parental/base bicistronic modular expression vector for site directed integration in circular configuration. The bicistronic expression vector was designed for expression of two polypeptides. Similar to the expression vector of Fig. 3, the parental/base expression vector comprises a standard E. coli propagation cassette as in Figure 2 (not shown in picture). The selection marker gene is represented as expression cassette 1 and have no promotor, but instead an upstream attB2 recombination site. Expression cassette 2 comprises a Fc-eGFP encoding gene and expression cassette 3 comprises a mTagBFP2 encoding gene. When the vector is integrated into the unique landing pad integrated in the CHO genome using PhiC31 recombinase, the selection marker will be activated by a promotor from the landing pad. In these examples, the two polypeptides expressed are eGFP and mTagBFPZ. As further shown in Figure 3, expression cassettes 2 and 3 have spacer elements between them to isolate gene expression from said cassettes to allow improved expression of the polypeptides of interest. Figure 3 also shows promoters 2 and 3 to improve expression from the expression cassettes 2 and 3 respectively. FIG. 6 shows the same parental/base bicistronic modular expression vector in linear configuration.

To enhance protein expression further, several different spacers were developed for placement between expression cassettes 2 and 3 and tested to evaluate their effect on gene expression. It was observed that a longer (500 bp or more) spacer element, for example, a cHS4 double core element, improved gene expression. It was also observed that protein expression improved when the wild type CTCF binding site (SEQ ID NO: 1) present in the wild type cHS4 double core element (SEQ ID NO: 4) was modified.

Various modified versions of the CTCF binding site of the wild type cHS4 double core element were developed and evaluated, including for example, a modified CTCF-binding DNA fragment A (SEQ ID NO: 2) and a modified CTCF-binding DNA fragment B (SEQ ID NO: 3). The respective nucleic acid sequences of the wild type and the modified CTCF-binding motifs or fragments are shown in Table la.Table la.

CTCF-binding Nucleic acid sequence SEQ ID motif CTCF binding GTAATTACGTCCCTCCCCCGCTAGGGGGCAGCAGCGA SEQ ID NO: 1 site (Fll) of wt GCCGCCCGGGGCTCC cHS4 (prior art) modified CTCF TGCAGTACCTCCCTCCCCCAGCAGGTGGCAGCAGTGA SEQ ID NO: 2 binding DNA ATGGTGAACTGGAGT fragment A modified CTCF TGCAGTACCTCCCTGGGCCAGCAGGGGGCACTAGCGA SEQ ID NO: 3 binding DNA AAGGTGAACTGGAGT fragment B The sequence identity between SEQ ID NO: 1, SEQ ID NO: 2 and SEQ ID NO: 3 was calculated with Geneious Prime software (Biomatters Ltd) and is shown in Table lb: Table lb: Sequence identity between CTCF-binding sequences.

SEQ ID NO: 1 (wt) SEQ ID NO: 2 SEQ ID NO: 3 SEQ ID NO: 1 (wt) - 65 % 60 % SEQ ID NO: 2 65 % - 85 % SEQ ID NO: 3 60 % 85 % - Several modified versions of the entire cHS4 double core element, comprising the above-described modified versions of the CTCF binding sites, were developed and evaluated for gene expression.

These modified double core elements were used in spacers A-C as outlined below.

SpacerA (SEQ /D NO: 5): ACGGGGACAGCCCCCCCCCAAAGCCCCCAGGGATIGCAGTACCTCCCTCCCCCAGCAGGTGGCAGCAGTGAAT GGTGAACTGGAGTGCTCCGGTCCGGCGCTCCCCCCGCATCCCCGAGCCGGCAGCGTGCGGGGACAGCCCGGG CACGGGGAAGGTGGCACGGGATCGCITTCCTCTGAACGCTICTCGCTG CTCTTTGAG CCTGCAGACACCTGGG GGGATACGGGGAAAAATGTITAGGCTGAAAGAGAGATTIAGAATGACAGGCACGGGGACAGCCCCCCCCCA AAG CCCCCAGGGATTG CAGTACCTCCCTCCCCCAG CAGGTGGCAG CAGTGAATG GTGAACTGGAGTGCTCCG GTCCGGCGCTCCCCCCGCATCCCCGAGCCGGCAGCGTGCGGGGACAGCCCGGGCACGGGGAAG GTGGCACG GGATCGCTTTCCTCTG AACGCTFCTCG CTG CTCTTTGAGCCTG CAGACACCTGGGGGGATACGGGGAAAAATG TTTAGGCTGAAAGAGAGATITAGAATG ACA The modified cHS4 double core element A (SEQ ID NO: 5) included the modified CTCF binding DNA fragment A (SEQ ID NO: 2) as its CTCF binding site. The modified cHS4 double core element A isreferred to as "spacer A" when used as a spacer element between two expression cassettes (here expression cassettes 2 and 3).

Spacer B (SEQ ID NO: 6): AAGGTGAACTGGAGTG CTCCGGTCCGGCG CTCCCCCCG CATCCCCGAGCCGGCAGCGTGCGGGGACAGCCCG GGCACGGGGAAGGTGGCACGGGATCGCTITCCTCTG AACGCTTCTCGCTG CTCTITG AGCCTG CAGACACCTG GGGGGATACGGGGAAAAATGTITAGGCTGAAAGAGAGATITAGAATGACAGGCACGGGGACAGCCCCCCCC CAAAG CCCCCAGGGATTGCAGTACCTCCCTGGGCCAG CAGGGGGCACTAGCG AAAGGTGAACTGGAGTGCTC CGGTCCGGCGCTCCCCCCG CATCCCCGAGCCGGCAGCGTGCGGGGACAG CCCGGGCACGGGGAAGGTGG CA CGGG ATCGCTFTCCTCTGAACGCITCTCG CTG CTCTITG AG CCTG CAG ACACCTGGGGGGATACGGGGAAAAA TGTITAGGCTGAAAGAGAGATTTAGAATG ACA The modified cHS4 double core element B (SEQ ID NO: 6) included the modified CTCF binding DNA fragment B (SEQ ID NO: 3) as its CTCF binding site. The modified cHS4 double core element B is referred to as "spacer B" when used as a spacer element between two expression cassettes (here expression cassettes 2 and 3).

Spacer C (SEO ID NO: 7): CAATAGACAG CCCCCCCCCAAATTG G CCATTAG CTGCAGTACCTCCCTGG G CCAG CAG G G G G CACTAG CG AAA G GTG AACTGGAG TCATATTATTCATCG CTCCCCCCG CATCCCCGATG GTTATATAG CATAAATCAATATFG G CT ATTG G CCATTG G CACGG G ATCG CTTTTG CATACGTTG TATCTATATCATAATAGGTACATTTATAC CT G G G GG G ATACGG G GAAAAATTG G CTCATG TCCAATATGACCCGAACTGAAG ATATG CAATAG ACAG CCCCCCCCCAAAT TG G CCATFAG CTG CAGTACCTCCCTG G G CCAG CAG G G G G CACTAG CGAAAG GTG AACTG G AG TCATATTATTC ATCG CTCCCCCCG CATCCCCGATG GTTATATAG CATAAATCAATATFG G CT ATTG G CCATTG G CACG G GATCG C 'ITTTG CATACGTFGTATCTATATCATAATAGGTACATTTATACCTG G GG GG ATACG G G GAAAAATTG G CTCATG TCCAATATGACCCGAACTG AAG ATAT The modified cHS4 double core element C (SEQ ID NO: 7) included the modified CTCF binding DNA fragment B (SEQ ID NO: 3) as its CTCF binding site. Furthermore, the DNA sequences in between the footprints Fl, Fll, Flll, FIV and FV, as well as the DNA sequence in between the two core elements, were different in relation to the Spacer B (SEQ ID NO: 6). The modified cHS4 double core element C is referred to as "spacer C" when used as a spacer element between two expression cassettes (here expression cassettes 2 and 3).

Furthermore, the wild type cHS4 double core element (SEQ ID NO: 4) is referred to as "spacer D" when used as spacer element between two expression cassettes (here expression cassettes 2 and 3).

Spacer D/Wi/d type cHS4 double core element (SEQ /D NO: 4): ACGGGGACAGCCCCCCCCCAAAG CCCCCAGGGATGTAATTACGTCCCTCCCCCGCTAGGGGGCAGCAGCGAG CCGCCCGGGGCTCCGCTCCGGTCCGGCGCTCCCCCCGCATCCCCGAGCCGGCAGCGTGCGGGGACAGCCCGGGCACGGGGAAGGTGGCACGGGATCGCTTT CCTCTGAACGCTTCTCG CTG CT CTITGAGCCTG CAGACACCTGG GGGGATACGGGGAAAAAGCTATAGGCTGAAAGAGAGATITAGAATGACAGGCACGGGGACAG CCCCCCCCC AAAGCCCCCAGGGATGTAATTACGTCCCTCCCCCGCTAGGGGGCAGCAGCGAGCCGCCCGGGGCTCCGCTCC GGTCCGGCG CTCCCCCCGCATCCCCGAGCCGGCAGCGTGCGGGGACAGCCCGGGCACGGGGAAGGTGGCAC GGGATCGCTITCCTCTG AACG CTTCTCGCTG CT CTTTG AGCCTG CAGACACCTGGGGGGATACGGGGAAAAAG CTATAGGCTGAAAGAGAGATTTAGAATG ACA The total length of the spacer elements between the expression cassettes 2 and 3 was also evaluated, and it was found that attaching an additional spacer sequence downstream of the modified cHS4 double core element led to even higher gene expression from the expression cassettes 2 and 3. This combination ofthe additional spacer element, herein referred to as "spacer X" or "spacer element X", and a modified cHS4 double core element (Spacer A, B or C) is referred to as a functional spacer element when used as a spacer between the expression cassettes 2 and Several different additional spacer elements X were constructed and evaluated, comprising, for example, nucleic acid sequences differing in the order of the nucleotides and nucleic acid sequences of different lengths such as 100 bp, 200 bp, 300 bp, 400 and 500 bp, in combination with the modified cHS4 double core elements as described above to evaluate their effect on gene expression.

Table 2 below shows the various plasmid expression vectors that were constructed based on the parental/base expression vector design as shown in Figure 3 and 5. The expression cassette 2 was configured to encode for eGFP or Fc-eGFP protein with the human cytomegalovirus (hCMV) or chimeric CMV promotor as promotor 2. The expression cassette 3 was configured to encode for mTagBFP2 protein with the mouse cytomegalovirus (mCMV) promotor as promoter 3. A spacer as described above based on the cHS4 double core element was arranged between expression vectors 2 and 3. The selection marker was eighter glutamine synthetase (SM1), TagRFP-T (SM2) or mRaspberry (SM3). For the additional spacer X sequences, stretches of 100 bp, 200 bp, 300 bp or 500 bp were used. Furthermore, to study the potential effect of different sequences, two different 200 bp SeqUenCeS Were COmpafed.

Additional spacer sequence variant ot (SEQ ID NO: 8): ACCCTTGTCCTATCAAATCTCCTCGTFFCG CGG CAG GACACCTG CTTCGAACG CGACCTTTAG CAGTG CGTG CG ATFACCG ACGG CCTCGTTCACTG AG GACG CCTG G CG GCACAACG CCATCGAGCAATTACATAAAGTG CACAG C CACTG CG G G GAATCGTTCAG G GATTGTTCG GTCACAG G G CG GAATCGTATTACAdditional spacer sequence variant ß (SEQ ID NO: 9): ACCCTTGTCCAAATTAAATCAG CTGTATACTCTTCCAGAAAACCGAAGAGAAGATG CTACACAAATCTCAAGGC CGACATAAAAGTFCTITGGAAAATATGAG AAGTTG CTAG G CATGATGGCATCTTCCTITAATCACACTTGGAÅA ACCAATÅGCAAAAG ACTTCAGTTCCTG CCAAACTGACCTACAGATTCTGGAC Spacer D refers to the wild type cHS4 double core element (SEQ ID NO: 4).

Table 2. Expression Spacer CTCF binding site Additional Selection Promoter 2 vector between Spacer X, marker construct cassettes 2&3 size pGE0173 No No 100 SM1 hCMV pGE0296 Spacer A Fragment A (SEQ ID NO:2) 0 SM1 hCMV pGE0297 Spacer A Fragment A (SEQ ID NO:2) 100 SM1 hCMV pGE0298 Spacer A Fragment A (SEQ ID NO:2) 500 SM1 hCMV pGE0299 Spacer C Fragment B (SEQ ID NO:3) 0 SM1 hCMV pGE03OO Spacer C Fragment B (SEQ ID NO:3) 100 SM1 hCMV pGE0301 Spacer C Fragment B (SEQ ID NO:3) 500 SM1 hCMV pGE0302 Spacer B Fragment B (SEQ ID NO:3) 0 SM1 hCMV pGE0303 Spacer B Fragment B (SEQ ID NO:3) 100 SM1 hCMV pGE0307 Spacer D Wild type CTCF binding 0 SM1 hCMV fragment (SEQ ID NO:1) pGE0308 Spacer D Wild type CTCF binding 100 SM1 hCMV fragment (SEQ ID NO:1) pGE0309 Spacer D Wild type CTCF binding 500 SM1 hCMV fragment (SEQ ID NO:1) pGE0341 Spacer A Fragment A (SEQ ID NO:2) 100 SM2 Chimeric CMV pG E0355 Spacer A Fragment A (SEQ ID NO:2) 200 (oi) SM2 Chimeric CMV pGEO356 Spacer A Fragment A (SEQ ID NO:2) 300 SM2 Chimeric CMV pG E0357 Spacer A Fragment A (SEQ ID NO:2) 500 SM2 Chimeric CMV pGE0358 Spacer B Fragment B (SEQ ID NO:3) 100 SM2 Chimeric CMV pGE0359 Spacer B Fragment B (SEQ ID NO:3) 200 (oi) SM2 Chimeric CMV pGE0360 Spacer B Fragment B (SEQ ID NO:3) 300 SM2 Chimeric CMV pGE0361 Spacer B Fragment B (SEQ ID NO:3) 500 SM2 Chimeric CMV pGE0362 Spacer C Fragment B (SEQ ID NO:3) 100 SM2 Chimeric CMV pGE0363 Spacer C Fragment B (SEQ ID NO:3) 200 (oi) SM2 Chimeric CMV pGE0364 Spacer C Fragment B (SEQ ID NO:3) 300 SM2 Chimeric CMV pGE0365 Spacer C Fragment B (SEQ ID NO:3) 500 SM2 Chimeric CMV pUP0062 Spacer A Fragment A (SEQ ID NO:2) 200 (a) SM3 hCMV pUP0071 SpacerA FragmentA (SEQ ID NO:2) 200 (ß) SM3 hCMV pUP0073 No No 200 (ß) SM3 hCMV EXAMPLES Abbreviations FACS Fluorescence Activated Cell Sorter CHO Chinese Hamster Ovary mAb Monoclonal antibody BFP Blue fluorescent protein GFP Green fluorescent protein MSX Methionine sulfoximine CMW Cytomegalovirus DWP Deep well plates Exampleln this experiment, plasmid expression vectors pGEO173, pGEO296-pGE0303 and pGE0307-pGEO309 having the general structure as described in connection with Fig. 3 and in Table 2 were linearized and transfected using electroporation into Chinese Hamster Ovary (CHO) cells for random integration within the genome of said CHO cells. The transfected CHO cells were allowed to grow with 25 uM of methionine sulfoximine (MSX). After 2 weeks, cell samples were analysed by flow cytometry for determination of the levels of eGFP and mTagBFP2 expression.

Figure 7 shows the flow cytometry data pertaining to cells transfected with expression vector pGEO308 pertaining to experiment 1. The left panel shows selected viable single CHO cells (N-gate) for evaluation and the right panel shows the expression of mTagBFP2 and eGFP from N-gate where % CHO cells with expression of both eGFP and mTagBFP2 (P-gate) can be calculated. All expression vectors in experiment 1 were evaluated in the same way as pGEO308 as described above.

The expression result from spacers of experiment 1 can be seen in Figure 8 where the percentage of expressing recombinant CHO cells from gate P are shown for the different spacers. The results show that when no double core element is present ("C") between expression cassette 2 and 3 around 0.1 % of the CHO cells express from both cassettes 2 and 3, as compared with 0.2 % in average when a functional spacer including a double core element is present. Functional spacer A with additional 100 bp Spacer X sequence has the most improved expression, with 0.42 % of the CHO cells expressing in gate P. ln conclusion, the functional spacers (which include double core elements) result in an improved expression as compared to no functional spacer. Furthermore, Spacer A with an additional spacer of 100 bp shows a substantial improvement compared to Spacer D (wild-type double core).Exampleln this experiment, plasmid expression vectors pGE0341, pGE0355-pGE0365 having the general structure as described in connection with Fig. 3 were linearized and transfected using electroporation into mammalían CHO cells for random integration within the genome of said CHO cells. The transfected CHO cells were allowed to grow. After 2 weeks, cell samples were analysed by flow cytometry for integration of vector with TagRFP-T and for determination of the levels of eGFP and mTagBFP2 expression.

Figure 9 shows a plot of flow cytometry data pertaining to cells transfected with expression vector pGE0359 pertaining to experiment 2. The left panel shows selected viable single CHO cells (F-gate) for evaluation and the right panel shows the expression of mTagBFP2 and eGFP from F-gate where % CHO cells with both expression of eGFP and mTagBFP2 (M-gate) can be calculated. All expression vectors in experiment 2 were evaluated in the same way as pGE0359 as described above.

The expression result of spacers from experiment 2 can be seen in Figure 10 where the percentage of expressing recombinant CHO cells from gate M are shown for the different spacers. The results show good expression with 0.2-0.3 % in average for Spacers A and B with original sequence between footprints (Fl-FV). Spacer C with additional 200 bp spacer X shows the best expression with 0.49 % of the CHO cells expressing in gate M. As a group, the Spacer C variants provides a generally improved expression compared with Spacers A or B with a chimeric CMV promotor for expression cassette 2. Among the individual spacer elements, Spacer B with additional 200 bp spacer shows the best improvement.

ExampleThe impact of different additional spacer sequences was evaluated using two different 200 bp additional spacers sequences for the functional spacer separating the expression cassettes 2 and 3, additional spacer variant a (SEQ ID NO: 8) and additional spacer variant ß (SEQ ID NO: 9). ln this experiment site directed integration (SDI) were used with plasmid expression vectors pUP0062, pUP0071 and pUP0073 together with PhiC31 recombinase. The expression vectors were transfected with electroporation into CHO cells for integration at a landing pad (LP) sequence previously inserted in the CHO cell genome. For each spacer sequence element, duplicate transfections were performed. After 1 week of growth, cell samples were analysed by flow cytometry for integration of vector with mRasberry and for determination of the levels of Fc-eGFP and mTagBFP2 expression.Figure 11 shows a plot of the flow cytometry data pertaining to experiment 3. Mean mTagBFPZ and eGFP signals were calculated based on the sub-population of viable and mRasberry-positive cells as shown in Figure 11. Data were then normalized to the donor plasmid (pUP0073) giving the lowest expression level. As shown in Figure 11, the left panel shows a forward scatter to side scatter plot used to gate viable cells (A-gate). The middle panel shows a side scatter to mRasberry plot used to gate mRasberry positive cells corresponding to cells having undergone integration at the landing pad (B-gate). The right panel shows a mTagBFPZ to eGFP plot used to calculate the mean expression values based on the fluorescence signals of the main focused population (C-gate). Next, the mean value and standard deviation for each spacer element were calculated based on the duplicate data. The resulting data is plotted in Figure 12. As can be seen in this Figure, Experiment 3 shows that the Spacer A with a 200 bp additional spacer has clearly improved expression compared to only a 200 bp spacer (i.e. lacking a cHS4 double core element) and that Spacer A works well with 200 bp additional spacers of different sequences.

Based on the above experiments 1, 2 and 3, it was concluded that Spacers A, B and C have the potential to significantly improve gene expression, and that the specific nucleotide sequence of the additional spacer element X was inconsequential. However, it was seen that the length of the additional spacer had an impact on the gene expression from the expression cassettes 2 and 3, as demonstrated by running experiment 3 where two different sequences of 200 bp length were used as the additional spacer elements encoded by nucleic acid sequences SEQ ID NO: 8 and SEQ ID NO: ExampleThis example demonstrates successful production of monoclonal antibody (Herceptin) by expression of light and heavy chain expression cassettes separated using functional spacer elements according to embodiments of the present invention.

Example 4A. Cornnarison of expression vectors with different functional spacers ln this experiment, five different expression plasmids differing only in the design of the functional spacer were used to generate minipools with expression vector integrated by random integration (RI) mechanisms. The expression vectors were constructed according to FIG. 3, with the following changes: a) A Glutamine Synthetase (GS) gene driven by a mPGK promoter were used as selection marker (Expression Cassette 1). b) A Herceptin HC gene driven by a chimeric CMV promoter was used as Expression Cassettec) A Herceptin LC gene driven by a mCMV promoter was used as Expression Cassette The functional spacer design of the expression vectors used are summarized in Table Table 3: Functional spacer designs used in antibody expression vectors.

Expression vector Description pGEO349 Spacer A + 100 bp Spacer X. pGE0352 Spacer A + 500 bp Spacer X. pGE0353 Spacer C + 100 bp Spacer X. pGE0354 Spacer C + 500 bp Spacer X.

The different expression vectors (Table 3) were linearised and used to transfect CHO-Kl cells by a lipofectamine based method. Two days after transfection, CHO cells were sorted into 96-well plates for static culture using FACS. For each well in the plates, 2000 cells were sorted into 100 ul growth medium supplemented with 25 pIVI of methionine sulfoximine (MSX). Cells were then grown by static culture in the presence of 25 pM MSX for a total of 26 days. During this static culture period only cells having integrated the expression vector (and hence the GS gene) can divide and multiply in the presence of 25 uM MSX and the absence of L-glutamine. With a seeding density of 2000 cells per well, typically 0-5 individual cells per well can grow out. Following static culture, wells with growing cells were transferred to a 96 Deep Well Plate containing 550 ul culture medium supplemented with 37.5 uM MSX. Plates were incubated in conditions for suspension culture until stable growth was detected. New Deep Well Plates were then seeded and used to perform an 8-days batch culture. The supernatant from the batch culture plates was then used to measure Herceptin titer using a commercial antibody titer kit (ValitaCell/Beckman Coulter).

The measured antibody titers for the two minipools with the best expression for each expression vector design are summarized in Table Table 4: Measured titers for the two best performing minipools for each expression vector.

Minipool identity Measured titer (g/ L) pGEO349_D4 1.23 pGEo349_A11 0.75 pGEo352_Ds 0.74 pGEo3s2_c2 0.64 pGEo353_F11 0.51 pGEo353_B4 0.30 pGEo354_F9 1.08 pGEo354_D2 0.As shown in Table 4, both expression vectors containing Spacer A (pGE0349, pGEO352) and expression vectors containing spacer C (pGE0353, pGE0354) resulted in desirably high antibody titers (around or above 1 g/L).

Example 4B. Comparison to reference vector The pGE0349 expression vector was compared to a reference expression vector (pGE0164) containing Spacer D as a functional spacer. The reference vector contained no additional Spacer X.

Table 5: Comparison of expression vectors pGE0164 and pGE Expression Selection Cassette 2 Cassette 3 Functional spacer vector marker pGE0164 GS driven by Herceptin HC driven Herceptin LC driven Spacer D (no mPGK by hEF1-alpha by hEF1-alpha additional Spacer X) pGE0349 GS driven by Herceptin HC driven Herceptin LC driven Spacer A + 100 bp mPGK by chimeric CMV by mCMV Spacer X.

The two expression vectors (Table 5) were linearised and used to transfect CHO-Kl cells by a lipofectamine based method. Three days after transfection, CHO cells were sorted into two 96-well plates for static culture for each expression vector using FACS. For each well in the plates, 5000 cells were sorted into 100 ul growth medium supplemented with 25 uM MSX. Cells were then grown by static culture in the presence of 25 uM MSX for a total of 25 days for pGE0164 and 28 days for pGE0349. During this static culture period only cells having integrated the expression vector (and hence the GS gene) can divide and multiply in the presence of 25 uM MSX and the absence of L- glutamine. With a seeding density of 5000 cells per well, typically 3-10 individual cells per well can grow out. Following static culture, wells with growing cells were transferred to a 96 Deep Well Plate containing 550 ul culture medium supplemented with 37.5 ulvl MSX. Plates were incubated in conditions for suspension culture until stable growth were detected. New Deep Well Plates were then seeded and used to perform an 11-days batch culture. The supernatant from the batch culture plates was then used to measure Herceptin titer using a commercial antibody titer kit. Table 6 shows the number of expressing minipools and the median titer for each expression vector.

Table 6: Median titer of expressing clones using either pGE0164 or pGE Expression vector No. of expressing minipools Median titer (g/L) pGE0164 150 0.53 pGE0349 50 0.From this study it was concluded that the expression vector with the inventive functional spacer element produced antibody titers at least comparable to those ofthe expression vector including the Spacer D (wild type cHS4 double core).

Example 4C. Fu/I cell line development ln this experiment a modified expression vector (pGE0419) was used to perform a full cell line development campaign, including assessment of producer clones using fed-batch cultures in shake flasks.

The expression vector pGE0419 contained the following modifications compared to the pGE0349 vector of Examples 4A and 4B: First, the GS gene used as a selection marker was modified by mutation of amino acid number 299 from arginine to Glycine (R299G). Secondly, the promoter driving the Herceptin HC was changed from a chimeric CMV to a mCMV sequence. Lastly, the additional spacer contained 200 bp instead of 100 bp.

Transfection and minipool generation were performed as described above for Example 4B. However, in this experiment the best performing minipools were selected for further expansion and then used in single cell cloning using FACS. During single cell cloning, single cells were seeded into each well of 96-well static culture plates containing 100 ul cloning media supplemented with 25 uM MSX. Following outgrowth in static culture conditions, cells were transferred to suspension culture in deep well plates as previously described. Following titer assessment after an 11-days batch culture, the 24 best performing producer clones were selected for further assessment in a 24 deep well plate fed- batch assessment. The 14 best performing producer clones from this stage were further expanded and evaluated in fed-batch cultures in shake-flasks. Cell density, cell viability and antibody titers were measured throughout the cultures using commercial equipment. Cultures were terminated as cell viability dropped below 20% (typically after 18-21 days). Results are summarized in TableTable 7: Final fed-batch titers for the 14 best performing producer clones generated using pGE Clone ID 24 DWP Titer CS 6,94 Bl 6,18 D4 5,49 D1 540 C4 4,20 C6 4,14 B5 3,51 D5 3,45 C3 3,00 A2 217 A5 2,03 B2 1,48 A4 1,40 B4 0,Example 4C thus demonstrates successful establishment of a CHO cell line capable of producing a desired antibody at very high levels. ln total, the Examples 1-4 indicate that the spacer elements disclosed herein offer the potential to provide improved cell lines for high expression of a protein of interest, in particular where two proteins are expressed simultaneously, such as is the case of mAbs and other antibody fragments or variants incorporating both light chain and heavy chain components.

Claims (22)

1. A nucleic acid sequence comprising a modified CTCF-binding sequence represented by: TGCAGTACCTCCCTN1N2N3CCAGCAGGN4GGCAN5N6AGNyGAANgGGTGAACTGGAGT (SEQ ID NO: 10) wherein N1, NZ, Ng, and Ng each independently is C or G; N4 is T or G; NG and Nv each independently is C or T; and NgisTorA.

2. The nucleic acid sequence according to claim 1, wherein the modified CTCF-binding sequence is selected from SEQ ID NO: 2 and SEQ ID NO:

3. The nucleic acid sequence according to claim 1 or 2, comprising a modified chicken Hypersensítivity Site 4 (cHS4) core element including said modified CTCF-binding DNA sequence.

4. The nucleic acid sequence according to claim 1 or 2, comprising two Hypersensítivity Site 4 (H54) core elements, each comprising a CTCF-binding sequence, at least one of which being said modified CTCF-binding sequence.

5. The nucleic acid sequence according to claim 4, wherein the CTCF-binding sequence of one of said HS4 core elements is represented by SEQ ID NO:

6. The nucleic acid sequence according to claim 4, comprising a sequence selected from SEQ ID NO: 5-

7. The nucleic acid sequence according to claim 4, wherein one of said HS4 core elements comprises a modified CTCF-binding sequence according to SEQ ID NO: 2 and one of said HS4 core elements comprises a modified CTCF-binding sequence according to SEQ ID NO:

8. The nucleic acid sequence according to claim 3, comprising a spacer sequence of at least 100 nucleotides in length located downstream of said modified cHS4 core element.

9. The nucleic acid sequence according to any one of the claims 4 to 7, comprising a spacer sequence of at least 100 nucleotides in length located downstream of both ofthe two HS4 core elements.

10. The nucleic acid sequence according to claim 8 or 9, wherein the spacer sequence has a length of at least 200 nucleotides, and optionally up to 500 nucleotides, such as up tonucleotides.

11. An expression vector comprising the nucleic acid sequence according to any one of the preceding claims.

12. The expression vector according to claim 11, comprising - a first expression cassette comprising a first coding sequence encoding a first protein of interest, - a second expression cassette comprising a second coding sequence encoding a second protein of interest, and -the nucleic acid sequence according to any one of claims 1 to 10 arranged between said first expression cassette and said second expression cassette.

13. The expression vector according to claim 12, wherein the first protein of interest is a first polypeptide chain of an antibody or antibody fragment, and the second protein of interest is a second polypeptide chain of an antibody or antibody fragment.

14. The expression vector according to claim 13, wherein the first polypeptide chain comprises a variable domain of an antibody heavy chain, and wherein the second polypeptide chain comprises a variable domain of an antibody light chain.

15. The expression vector according to any one of claims 12 to 14, wherein the first expression cassette and/or the second expression cassette comprises a eukaryotic promoter sequence, such as a CMV, EFlalpha or SV40 PGK promoter, and preferably a CMV promoter.

16. The expression vector according to any one of claims 11 to 15, which is a circular vector.

17. A cell comprising the expression vector according to any one ofthe claims 11 to

18. The cell according to claim 17, wherein the expression vector is integrated into the cell genome. 219. The cell according to claim 17 or 18, wherein the cell is a mammalian cell, such as a

19. Chinese Hamster Ovary (CHO) cell, transfected with the expression vector.

20. A method of recombinantly producing at least two polypeptides, comprising - transfecting mammalian cells, such as CHO cells, with an expression vector according to any one of the claims 12 to 16 comprising expression cassettes encoding first and second proteins of interest, and - culturing the transfected cells in cell culture media under conditions enabling expression of the first and second proteins of interest.

21. The method of claim 20, wherein said two polypeptides are two polypeptide chains ofa monoclonal antibody or an antibody fragment.

22. The method of claim 20 or 21, further comprising purifying the polypeptides.