Using Mammalian Competent Cells For Evaluation Of IRES Sequences

Genlantis, a division of Gene Therapy Systems

10190 Telesis Court
San Diego, CA, 92121
Website: http://www.genlantis.com

Using Mammalian Competent Cells For Evaluation Of IRES Sequences

by Tara Dobson, Marie Callahan, Anne Willis, Michael Lee, and Alan Greener

Figure 1. Structure of the IRES Cloning Vector, pRF.

Expression of multiple recombinant proteins in mammalian cells from a single mRNA is technically more difficult than expression in bacterial cells due to the different nature by which translational initiation occurs. In bacteria, a single mRNA molecule may encode multiple open reading frames with every gene product initiating from a ribosome-binding site preceding the start codon. In mammalian cells, most mRNA molecules encode a single protein as a specific structure (CAP), must be formed at the 5' end of the mRNA, and translation initiates once, usually at the closest ribosome-binding site encountered by the ribosome complex.⁽¹⁾ However, there are specific sequences that enable translation to initiate from mRNA molecules that do not contain a CAP structure; they will allow for polycistronic messages to be converted into multiple proteins. These 5' untranslated regions, called internal ribosome entry sequences (IRES⁽²⁾), were first observed in viruses (EMCV^(2,3) but have since been described from eukaryotic genomes.⁽⁴⁾ One feature about gene expression directed from an IRES sequence is that the level of protein produced is usually far lower than if a conventional (capped) mRNA is utilized (For a recent review, see 5).

We have developed a bicistronic expression system as a means to evaluate the relative (and absolute) efficiencies of IRES sequences. A vector was constructed containing the Renilla luciferase gene expressed from the 5' end of the mRNA, followed by an IRES insertion region and the gene encoding luciferase from Photinus (see Figure 1). The Photinus luciferase gene will only be expressed if an IRES is positioned between the two genes. Since IRES-directed expression is significantly lower than the upstream expression levels, the ability to accurately detect low levels of expression (compared with background) becomes crucial.

In this study, we compared conventionally passaged HeLa S3 cells with a new format of transfection-ready frozen competent mammalian cells (MaxPAK HeLa S3 cells, Genlantis, Inc., San Diego, CA). The reason we chose to compare expression levels in these two formats was because the frozen competent cells can be transfected at a much higher cell density per well without affecting transfection efficiency per cell. Since we expected the absolute levels of the IRES-directed Photinus luciferase to be very low, background luciferase levels obtained with standard concentration of HeLa S3 cells made detecting IRES expression-directed signals problematic. Our results demonstrate that the ability to transfect cells at higher density results in enhanced signal strength and unequivocal identification of IRES sequences in our bicistronic vector.

Materials and methods

Transfection experiments were performed following the recommended protocol. Regular passaged HeLa S3 cells (passage 11) were plated at 4 × 10⁵ cells/well in a 12 well plate-overnight for attachment prior to transfection. On the following day 1.6 × 10⁶ MaxPAK HeLa S3 cells/well (4× passaged cell plating density) were thawed and plated in the same 12-well plate. Three hours after plating the MaxPAK cells, 1µg of the bicistronic plasmid DNA (containing both Renilla and Photinus luciferase genes) was transfected using Fugene 6 (Roche Applied Science, Indianapolis,IN) at a ratio of 3:2 into both passaged and MaxPAK HeLa S3 cell lines. For the negative control, a plasmid DNA containing a non-IRES sequence (β-globin) inserted between the luciferase genes was utilized. The transfected cells were incubated at 37 C/5% CO₂ for 24 hours then lysed with Passive Lysis Buffer (Promega Corp., Madison, WI). Luciferase activities were quantified using a kit from Promega. Analysis of IRES activity was a comparison of the expression of the upstream (cap-dependent) Renilla luciferase gene with the downstream Photinus luciferase (see Results and Discussion). All experiments were done in triplicate and the results averaged.

Results and discussion

In order to evaluate the relative efficacy of different IRES sequences, a bicistronic vector was constructed containing the Renilla luciferase gene transcribed from the SV40 promoter followed by a cloning site and the Photinus luciferase gene (see Figure 1). Since downstream expression of the Photinus gene requires that an IRES sequence be present, different 5' untranslated regions can be inserted and only those sequences that result in Photinus luciferase expression are characterized as containing an IRES. Furthermore, by comparing the relative activities of the two luciferase genes, the IRES sequences themselves can be quantified. Both the Renilla and Photinus luciferase activities are readily determined by transient transfection of the bicistronic vector into HeLa S3 cells and their relative activities extrapolated to IRES activity. As a positive control, the IRES from EMCV⁽²⁾ was inserted upstream of the Photinus luciferase gene; for a negative control, a segment of β-globin that does not contain an IRES was inserted. For this study, any segment that directs Photinus luciferase activity to be two-fold greater than the β-globin control insert is considered an IRES candidate.

Limiting factors when assaying gene products that are poorly expressed are the transfection efficiency of the DNA into the host cells and the number of transfected cells that are necessary to detect the protein being produced. MaxPak frozen mammalian competent cells have been designed to address the cell number limitation; 4 - 16 × more cells can be successfully transfected per tissue culture surface without a decrease in transfection efficiency per cell. This is not the case with standard passaged cells, as increasing the cell number will result in a decrease in both transfection efficiency and overall gene expression.

In this work, 8-fold more frozen cells were transfected than passaged (see Materials and Methods). In Table 1, the numbers of luciferase light units are indicated for the passaged and frozen cells. We observed an overall increase in absolute luciferase levels for the MaxPAK cells - this increase ranges from 7.6 × - 50 ×. If transfection efficiency were linearly related to cell number, we would have expected all results to be approximately 8 ×. However, the majority of the assays indicated greater than an 8-fold difference, suggesting an increase in either the percentage of the frozen cells that are transfected or the number of plasmids molecules transfecting each MaxPAK cell is greater.

When expression of Photinus enzyme directed from the EMCV IRES vector was monitored and compared with the Renilla for the passaged cells, the relative luciferase ratio was approximately 2.3 fold greater than the β-globin control vector, fulfilling the 2 × target for IRES definition. For the MaxPAK frozen cells, the ratio was 2.4. However, the standard deviation of these assays indicated that assigning the EMCV as an IRES for the passaged cells was equivocal. This was not the case with the frozen cells, as the reproducibility is much greater and the standard deviation lower. Thus with the MaxPAK cells, validation of the well-studied EMCV IRES was unequivocal. When the unknown and very active IRES was inserted, the ratio compared with the β-globin was 10-15 for both formats, but again, the standard deviation was much lower for the frozen competent cells. This reproducibility, observed as a lower standard deviation when the assay was performed multiple times, indicated that the MaxPAK format was more reliable when expression levels are low. The overall increase in absolute luciferase levels (for both the CAP-dependent Renilla luciferase and the IRES-dependent Photinus luciferase) is related to the number of cells used; however, in some instances the luciferase expression was even more pronounced, suggesting an increase in either the percentage of cells being transfected or the number of plasmids molecules transfecting each cell.

In summary, expression of the upstream (Renilla) luciferase increased 8-50 fold when using MaxPak HeLa S3 cells compared with the standard passaged cells. This increase was most likely a combination of the increased cell number that can be used with the MaxPak cells and an increase in either the percentage of cells that are transfected or the number of plasmids that enter each cell. The physiological state of the rapidly frozen and immediately transfected MaxPAK cells is not completely understood. Nevertheless, the ability to transfect MaxPak cells at higher plating density without losing efficiency per cell (which does not work for conventionally passaged cells) enabled us to better analyze IRES-directed expression.

About the authors

Tara Dobson and Anne Willis are with the 1 Department of Biochemistry and Molecular Genetics, University of Colorado Health Sciences Center in Denver. Marie Callahan, Michael Lee and Alan Greener are with Genlantis, Inc., in San Diego, CA.

References

1. Kozak, M. Regulation of translation via mRNA structure in prokaryotes and eukaryotes. Gene 361:13-37 (2005).

2. Jang, S.K., Krausslich, H.G., Nicklin, M.J., Duke, G.M., Palmenberg, A.C., and Wimmer, E. A segment of the 5' nontranslated region encephalomyocarditis virus RNA directs internal entry of ribosomes during in vitro translation. J. Virol. 62:2636-2643 (1988).

3. Jang, S.K., Davies, M.V., Kaufman, R.J., and Wimmer, E. Initiation of protein synthesis by internal entry of ribosomes into the 5' nontranslated region of encephalomyocarditis virus RNA in vivo. J. Virol. 63: 1651-1660 (1989).

4. Hellen, C.U., and Sarnow, P. Internal ribosome entry sites in eukaryotic mRNA molecules. Genes Dev. 15:1593-1612 (2001).

5. Baird, S.D., Lewis, S.M., Turcotte, M., and Holcik, M. A search for structurally similar cellular internal ribosome entry sites. Nucl. Acids. Res. 35:4664-4677 (2007).