Identification of a potent MAR element from the mouse genome and assessment of its activity in stable and transient transfections

https://doi.org/10.1016/j.jbiotec.2011.04.004Get rights and content

Abstract

Matrix attachment regions are DNA sequences found throughout eukaryotic genomes that are believed to define boundaries interfacing heterochromatin and euchromatin domains, thereby acting as epigenetic regulators. When included in expression vectors, MARs can improve and sustain transgene expression, and a search for more potent novel elements is therefore actively pursued to further improve recombinant protein production. Here we describe the isolation of new MARs from the mouse genome using a modified in silico analysis. One of these MARs was found to be a powerful activator of transgene expression in stable transfections. Interestingly, this MAR also increased GFP and/or immunoglobulin expression from some but not all expression vectors in transient transfections. This effect was attributed to the presence or absence of elements on the vector backbone, providing an explanation for earlier discrepancies as to the ability of this class of elements to affect transgene expression under such conditions.

Introduction

With the growing demand for increased production of clinical grade proteins, many of which can only be produced in mammalian cells, improvements in protein yields from mammalian cells and a decreased time to production are prime objectives. While improvements to vectors, cell culture conditions and downstream processing have greatly enhanced productivity over the last 20 years (Wurm, 2004), recombinant protein production in mammalian cell lines remains beset by clonal heterogeneity, instability, loss of gene expression and protein production over time, and it is therefore a lengthy process to isolate a suitably stable high producing clone. This makes recombinant protein production in mammalian cells an expensive and time-consuming process. In particular, these lengthy screening processes are not suited for protein production for in vitro assays, structural studies and initial in vivo studies, where relatively small amounts of a large number of different proteins are required.

Initial work on improving protein production focussed on improving cell growth medium and downstream processing. For instance, supplementing the growth medium with sodium butyrate or valproic acid, two agents that promote the loss of heterochromatin by inhibiting histone deacetylases, has resulted in improved protein yields, both in transient and stable transfections (Backliwal et al., 2008, Jiang and Sharfstein, 2008, Wulhfard et al., 2010). However, these substances have a general rather than specific effect on gene expression, which leads to toxic effects and cell death upon prolonged exposure, and their inclusion in the growth medium is undesirable for downstream processing. Nonetheless, the use of these substances has revealed that chromatin-remodelling processes are involved in transgene expression in stable cell lines (Schlake et al., 1994, Jiang and Sharfstein, 2008). In the last 20 years, DNA elements that can maintain an open chromatin transgene structure, but without the toxic effects linked to genome-wide changes in chromatin structure and the need of growth medium additives, have been sought. Several of these elements, termed matrix attachment region (MAR), anti-repressor (STAR), insulators (HS4), locus control regions (LCR) and upstream chromatin opening element (UCOE), have been used to improve transgene transcription and protein production in cultured cells and in vivo (Kwaks and Otte, 2006).

Recently, focus has shifted to improving cell line stability and decreasing the cumbersome evaluation of large numbers of clones (Chusainow et al., 2009, Pilbrough et al., 2009, Porter et al., 2010). In this regard, we have successfully used MAR elements to improve protein production in mammalian cells, including therapeutic proteins such as immunoglobulins (IgGs) (Girod et al., 2005, Girod et al., 2007, Varghese et al., 2008, Zahn-Zabal et al., 2001). Incorporating MAR elements into expression vectors was shown to decrease the number of silenced cells as well as clonal heterogeneity (Girod et al., 2005, Girod et al., 2007). One of these MARs was shown to be able to reverse epigenetic silencing and to increase the probability of switching to an active transcriptional state (Galbete et al., 2009). Moreover, time to production was decreased, as including MAR elements in expression vectors significantly reduces the number of clones that need to be screened (Girod et al., 2005, Varghese et al., 2008). Several MARs are known to increase protein production in mammalian cells including the human MARs from the β-globin and IFN-loci and the MAR from the chicken lysozyme locus (Girod et al., 2005, Girod et al., 2007, Kim et al., 2004, Kim et al., 2005, Zahn-Zabal et al., 2001). We also recently identified several new, more potent MARs from the human genome using an optimized in silico screening approach, and found that many of the new MAR thereby identified were significantly more potent than the previously known chicken lysozyme MAR (Girod et al., 2007). However, this study also indicated that the potency of MARs varies in terms of increasing protein production, and it opened the prospect that even more potent MAR elements may exist.

One unresolved question about the action of MARs is whether they may also improve transgene expression in transient transfections. One of the early defining criteria of MARs was that they are unable to increase transgene expression in transient transfection assays, distinguishing them from transcriptional enhancers (Bode et al., 2000). However, several reports indicated that MARs may also be able to increase transgene expression in transient transfections. Whether these discrepancies may be related to different features of distinct MARs (e.g. Klehr et al., 1991 vs. Kulkarni et al., 2004), or different vector backbones and/or reporter genes (e.g. Klehr et al., 1991 vs. Chancham et al., 2003), has remained unclear. In this article we describe the isolation of a new, more powerful MAR from the mouse genome, the use of this MAR element to improve production of a therapeutic protein in CHO cells, and we assess the ability of this MAR to increase IgG productivity in transient transfections.

Section snippets

In silico prediction of MAR elements

The whole mouse genome consisting of all mouse chromosome sequences corresponding to the NCBI m34 mouse assembly were compiled and analyzed with SMAR Scan I, a MAR predicting computer software developed in our laboratory (Girod et al., 2007). Low and high stringency screens were performed using either a threshold for the DNA bending criterion of 3.6° and a minimal window size of 300 bp, or a threshold of 4.2° and a minimal window size of 100 bp, respectively.

MAR cloning and MAR constructs, plasmids

Ten “super” MAR elements were selected

Identification of the S4 MAR as a potent activator of stable transgene expression

MARs have proven to be very potent elements for increasing transgene expression in stable transfections. As part of our search for even more potent MAR elements, functional in both human and non-human mammalian cells, a large scale bioinformatics analysis of the mouse genome was performed using a modified SMARScan program (Girod et al., 2007). However, its parameters were optimized to account for the murine genome G/C content, and they were tuned to identify only the most extreme, and

Discussion

Matrix attachment regions have proved to be very powerful elements for inclusion in vectors designed for stable recombinant protein production, allowing not only improved levels of protein production, but also improved expression stability and a reduction in the number of clones required for screening, and thus the time to production. High producing clones can also be observed occasionally without a MAR (see Fig. 4 as an example), as expected from the fact that vectors might integrate by chance

Conclusion

We observed that the S4 MAR can be used to obtain high transgene expression in stable transfections. MAR S4 was also required to achieve the highest transient mRNA and IgG levels from linearized vectors. However, we found that this effect of the MAR was abolished by the presence of selection and backbone sequences in the transient transfections, while a MAR-mediated increased expression was readily obtained in the presence of these selection markers in stable transfections. This finding may

Acknowledgements

This work was supported by grants from the Swiss Commission for Technology and Innovation, Selexis SA, and by the Etat de Vaud.

References (37)

  • M. Zahn-Zabal et al.

    Development of stable cell lines for production or regulated expression using matrix attachment regions

    J. Biotechnol.

    (2001)
  • G. Backliwal et al.

    Valproic acid: a viable alternative to sodium butyrate for enhancing protein expression in mammalian cell cultures

    Biotechnol. Bioeng.

    (2008)
  • J. Bode et al.

    Transcriptional augmentation: modulation of gene expression by scaffold/matrix-attached regions (S/MAR elements)

    Crit. Rev. Eukaryot. Gene Expr.

    (2000)
  • M. Buceta et al.

    Use of human MAR elements to improve retroviral vector production

    Gene Ther.

    (2011)
  • P. Chancham et al.

    Nucleic acid-matrix attachment recognition regions-as facilitators in plasmid transfer

    J. Drug Target.

    (2003)
  • J. Chusainow et al.

    A study of monoclonal antibody-producing CHO cell lines: what makes a stable high producer?

    Biotechnol. Bioeng.

    (2009)
  • H. Dorai et al.

    Correlation of heavy and light chain mRNA copy numbers to antibody productivity in mouse myeloma production cell lines

    Hybridoma

    (2006)
  • J.L. Galbete et al.

    MAR elements regulate the probability of epigenetic switching between active and inactive gene expression

    Mol. Biosyst.

    (2009)
  • Cited by (36)

    • Fusion with matrix attachment regions enhances expression of recombinant protein in human HT-1080 cells

      2020, Journal of Bioscience and Bioengineering
      Citation Excerpt :

      However, cMAR, c-gMAR and g-cMAR showed no the enhancing activity on transient expression. Harraghy et al. (23) pointed out that stable but not transient immunoglobulin expression increased by S4 MAR, but the mechanism underlying a failure in enhancing GOI in transient assays remains unclear. In contrast, some other reports found that MARs could increase transgene expression in transient transfections under certain conditions (17,19).

    • Therapeutic glycoprotein production in mammalian cells

      2017, Journal of Biotechnology
      Citation Excerpt :

      Inclusion of these various epigenetic elements not only can help to increase the expression level of biotherapeutics, but can also increase the number of clones that have integrated the transgene with a more defined copy number of transgene per cell, thus accelerating the selection process (Grandjean et al., 2011; Harraghy et al., 2015). It was also demonstrated that MAR and UCOE helped decrease the variability of expression between the different clones and contributed in maintaining an active transcriptional state for better transgene expression (Williams et al., 2005; Galbete et al., 2009; Harraghy et al., 2011). It has been shown that an UCOE can also prevent DNA methylation of the CMV promoter region, thus preventing gene expression silencing (Lindahl Allen and Antoniou, 2007; Zhang et al., 2010a).

    • A 914-bp promoter is sufficient to reproduce the endogenous prolyl oligopeptidase gene localization in the mouse placenta if not subject to position effect

      2013, Gene
      Citation Excerpt :

      After the treatment with TURBO DNase (Ambion, Austin, TX, USA), the first-strand cDNA was synthesized using a Superscript III reverse transcriptase (Invitrogen, Carlsbad, CA, USA). The primers used for PCR were 5′-GGAATCGATGCTGCTGATTA-3′ and 5′-CCATCCAGCTTTATGCCTTT-3′ for POP (Matsubara et al., 2010), 5′-AGCAAAGACCCCAACGAGAA-3′ and 5′-GGCGGCGGTCACGAA-3′ for EGFP (Harraghy et al., 2011), and 5′-GAGGACGGGATCCAAAAGC-3′ and 5′-CTGTGCAGCGTCCGAAAGT-3′ for arylhydrocarbon receptor-interacting protein (Aip, Frericks and Esser, 2008; Matsubara et al., 2010). All the data were normalized to Aip.

    View all citing articles on Scopus
    1

    These authors contributed equally to this work.

    2

    Present address: Selexis SA, Plan-les-Ouates, Switzerland.

    View full text