Consistent Production of Genetically Stable Mammalian Cell Lines

Published on: 
BioPharm International, BioPharm International-05-01-2012, Volume 25, Issue 5

The author describes expression technology that produces cell lines with high genetic stability.

The author describes expression technology that produces cell lines with high genetic stability. This technology can be used in multiple cell types using a single or multiple gene constructs.

Genetic and expression stability are important metrics that should be used to evaluate any cell line development method. A cell line that is genetically stable throughout the biopharmaceutical manufacturing process is essential for any development program and is required for regulatory adherence. The definition of genetic stability can vary slightly depending on application; however, typically, genetic stability is confirmation that the transgene DNA and subsequent mRNA sequence along with the number of transgene copies do not change over the length of time required to perform a biopharmaceutical manufacturing run. The starting point for genetic stability testing is usually a vial of cells from a research cell bank, master cell bank or working cell bank. These cells are thawed and cultured for multiple generations. The minimum length of the standard stability study should be based on maximum number of cell doublings that will occur in the biomanufacturing program. Additional generations can be added to the study to allow for future changes to the manufacturing process and to be certain that the number of generations chosen is easily above the maximum doublings possible in the manufacturing run.

The majority of cell line development technologies that result in consistently high biopharmaceutical production generate cell lines that contain multiple copies of the transgene of interest. Most of these methodologies are based on transfection of a DNA construct into the desired cell line. The transfected DNA constructs contain a selectable marker, and cell lines that survive when placed under the selection pressure usually contain the transgene of interest inserted into their genome. Cell lines produced via transfection of DNA most often contain multiple copy transgene inserts at a single genetic locus. These multi-copy insertion sites are normally found with the transgenes lined up in a "head to tail" manner (1). Cell lines containing head to tail gene arrays tend to have problems with genetic stability. During cell mitosis, homologous recombination may occur at the locus between different sequences within the array. Recombination can cause a reduction in the number of gene copies at the site and a subsequent decrease in the quantity of mRNA and protein produced. These problems can be accentuated when some form of gene amplification is performed on the cell line to increase transgene copy number. Amplification methods used during cell line development typically force the duplication of the head to tail gene array at that same locus in the host cell genome. The resulting larger array becomes more unstable and more prone to homologous recombination or other replication errors during cell division. An example of this phenomenon is an antibody-expressing cell line that had been produced using an amplification method. A significant reduction in the amount of light chain mRNA over extended culture was observed (2). In addition, without continual selective pressure, dramatic decreases in the number of transgene copies and a subsequent decrease in mRNA levels were recorded. Major stability differences between clones produced by amplification methodology have been observed (3, 4).

Biopharmaceutical development companies have continually focused on shortening the timeline to get product into clinical trials. A main roadblock to a shorter timeline is the need to perform a genetic stability study. Even a minimal study (i. e., 40 generations) takes 3 to 4 months to complete. This timeframe represents a large portion of the cell line development timeline. However, for cell lines containing multiple gene copies at a single genetic locus, The study must be performed on numerous cell clones in order to select stable clones to move forward in the development process. The ability to eliminate stability studies from the critical development path can provide a significant time savings to drug developers.


GPEx is a unique and versatile system designed to insert genes of interest into a wide variety of mammalian host cells. The GPEx method is based on the use of replication defective retroviral vectors (retrovectors) to actively insert the desired genes into the genome of dividing cells. The majority of the components from GPEx retrovectors are derived from Moloney Murine Leukemia virus (MLV). The Vesicular Stomatitis Virus G Protein (VSV-G) is used as an envelope for the retrovector particle. These particles stably insert single copies of the transgene at multiple sites in the chromatin of dividing cells. Retrovectors deliver genes coded as RNA that, after entering the cell, are reverse transcribed to DNA and integrated stably into the genome of the host cell (see Figure 1). Two enzymes, reverse transcriptase and integrase, provided transiently in the vector particle, perform this function. These integrated genes are maintained through subsequent cell divisions as if they were endogenous cellular genes. By controlling the number of retrovector particles accessing the cell, multiple gene insertion can be achieved without traditional amplification steps.

Figure 1


Each transgene copy is inserted at a unique site in the chromosome

GPEx gene insertions occur at unique locations in the cell genome, with a single copy of the gene being inserted at each independent site. Unlike most other methods of transgene insertion that are undefined "passive" processes, each insertion by a retrovector is an "active" process that is modulated by the MLV integrase enzyme (5). This method prevents head-to-tail arrays at a single site in the host cell genome.


The technology inserts genes into a wide variety of cell lines

The method uses VSV-G as an envelope on the retrovector particles. This envelope protein allows the retrovectors to insert genes into all mammalian cells, in addition to numerous other cell types, because of its ability to bind to various membrane phospholipids and glycolipids (6–8).

Transgene inserts target open or active regions of the cell genome

The MLV based retrovectors have been shown to preferentially insert into or around the transcription start point of genes (9, 10). This preference for transcriptionally active regions of the genome allows for higher, more consistent levels of expression per copy of the gene inserted compared to other methods of gene insertion.

Amplification without antibiotic selection or use of toxic compounds

Because of the extremely high gene insertion efficiency of the GPEx process, no selectable markers (e.g. neomycin, blasticidin, hygromycin, or puromycin resistance genes) are needed for cell line generation. This has a number of advantages over other cell line development methods, including reduced costs for culturing cells, no additional taxing of the cells, and reduced time for clonal cell line selection. The high transduction efficiency and the ability to do repeat cell transductions generate high copy number cell lines, thereby eliminating the need to amplify gene copy number by adding toxic compounds, such as methotrexate or methionine sulphoximine. To increase copy number, retransduction of a cell line is performed yielding clonal lines with copy numbers ranging from 25–250.

Figure 2


The data presented in this article are from a number of GPEx generated cell lines producing various products. Retrovectors were produced based on methods described by Burns et al., 1993 (7). A master cell bank of the GPEx–CHO cell line was the parent cell line for all data outlined in this article. Cell lines expressing recombinant proteins and antibodies were produced as shown in Figures 2 and 3. Retrovector transductions were performed at a multiplicity of infection of at least 1000 retrovector particles per CHO cell. To generate recombinant protein expressing cell lines, the parent GPEx–CHO cells were transduced three times with the genetically engineered retrovector. For generation of antibody producing cell lines, an initial transduction of GPEx–CHO cells was performed using a retrovector containing the light chain gene. The light chain containing pool of cells was then transduced with a retrovector containing the heavy chain gene. Upon completion of both transductions, the resulting pool of cells was then transduced a second time with light chain retrovector and two additional times with heavy chain retrovector for a total of five transduction cycles.

Table I

Single cell clones were isolated from the final recombinant protein or antibody producing cell pool using limited dilution cloning. Approximately 300–500 clonal lines were screened for protein production levels and various protein specific characteristics. The top 20 clones were selected based on protein production in 96-well plates. Fed-batch (generic conditions) protein production and specific productivity results from triplicate T150-flasks were then used to narrow the number of clones to the top 3–5 candidates. These clones were subsequently moved into shake flask upstream process development for production analysis using a matrix of different fed-batch culture conditions to select the master cell bank candidate clone and the upstream process base conditions for further development. Antibody producing candidate clones in these generic fed-batch conditions typically produce titers of 1–2 g/L and have specific productivities of 25–70 picograms/cell/day (p/c/d). As shown in Table I, after the initial upstream development screening the titers improve to a range from 2–5 g/L, and specific productivities of 40–100 p/c/d are achieved.

Figure 3

To analyze genetic stability, DNA or RNA was isolated from cells in log growth phase. A quantitative real-time PCR based assay was used to estimate the number of gene copies inserted in the cell lines. A DNA sequence present in each transgene insert was used as the target sequence to estimate the total number of insertions. The β 1,4 galactosyltransferase-1 gene is used as an endogenous marker gene to control for the amount of genomic DNA in each reaction. The gene index is calculated by subtracting the transgene assay threshold cycle from the control assay threshold cycle. A similar quantitative real-time PCR assay along with a reverse transcription step was used to determine the level of heavy chain and light chain mRNA being expressed. Either a portion of the constant region of the heavy chain or light chain was used to determine heavy and light chain mRNA levels respectively. The glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene was used as a control mRNA for the CHO cell lines. Isolated RNA was reverse transcribed and three quantitative real-time PCR (heavy chain, light chain, GAPDH control) reactions were run in triplicate. Similar to the assays performed with genomic DNA, a transgene mRNA index value was calculated by subtracting the sample threshold cycle number for the either the heavy chain or light chain assay from the control GAPDH assay threshold cycle value.

Figure 4


Because of the genetic stability of GPEx generated cell lines, stability studies do not need to be performed as part of the cell line selection process. Cell lines selected to master cell bank are chosen strictly based on protein expression and protein quality. Once the master cell bank is produced, confirmation of cell line genetic stability is performed on the bank. Results for 17 GPEx cell banks each producing a different protein are shown in Figure 4. The cell bank for each cell line was designated as generation 0 for the purpose of the study. Cells were continuously cultured by serial passage from generation 0 to approximately generation 40 or 60 depending on the study protocol. At the end of the experiment, samples of cells from the generation 0 cell bank and from the generation 40 or 60 culture were used for DNA and RNA isolation. Real-time PCR analysis of the DNA showed no significant difference between the number of transgenes at generation 0 and generation 40 or 60 for any of the 17 banks. To examine the stability of mRNA expression, cell banks producing seven different antibodies were tested at generation 0 and 60. Heavy and light chain mRNA levels were estimated individually at the two time points for each of the cell lines. No significant differences in heavy chain or light chain mRNA levels over the extended culture were observed (see Figure 5).

Figure 5


The GPEx method of gene insertion consistently produces genetically stable cell lines. The fact the gene inserts are single copy make the method inherently more stable than any technique that results in more than one copy of the transgene inserted at a single chromosomal location. This innate genetic stability of the cell lines allows biopharmaceutical development timelines to be shortened 3–5 months by removing cell line stability screening from the critical path.

GREGORY T. BLECK, PHD, Research and Development Platform Lead- Biologics, Catalent Pharma Solutions, Middleton WI,


1. B.J. Pomerantz et al., Mol. Cell. Biol. 3 (9), 1680–1685 (1983).

2. K. Strutzenberger et al., J. Biotech 69, 215–226 (1999).

3. N.S. Kim et al., Biotech. Bioeng. 60 (6), 679–688 (1998).

4. L.M. Barnes et al., Biotech. Bioeng. 85 (2), 115–121 (2004).

5. M.D. Andrake and A.M. Skalka, J. Biol. Chem. 271 (33), 19633–19636 (1996).

6 R. Schlegel et al., Cell 32, 639–646 (1983).

7. J.C. Burns et al., Proc. Natl. Acad. Sci. USA 90, 8033–8037 (1993).

8. J.K. Yee, T. Friedmann, and T.C. Burns, Methods Cell Biol. 43, 99–112 (1994).

9. X. Wu et al., Sci. 300, 1749–1751 (2003).

10. R.S. Mitchell et al., PLOS Biology 2 (8), 1127–1137 (2004).