Should I use IRES or 2A (and which 2A) in my polycistronic expression cassette?
When designing a gene expression vector to co-express multiple ORFs under the control of a single promoter, you can choose to place multiple ORFs behind the promoter, separated by linkers such as the internal ribosome entry site (IRES) or the 2A family peptides. For either type of linker, multiple proteins will be produced from a single mRNA transcript.
The most commonly used IRES element, and also the one used by VectorBuilder, is derived from the encephalomyocarditis virus (EMCV). It functions by acting as an additional ribosome recruitment site, allowing translation initiation to occur at an internal region of the mRNA in addition to the primary translation initiation site.
The 2A peptides are short (~18-25 aa) peptides derived from viruses. They are often called “self-cleaving” peptides, which will produce multiple proteins from the same transcript. 2A peptides do not entirely “self-cleave,” as they function by making the ribosome skip the synthesis of the glycine and proline peptide bond at the C-terminal end of the 2A element, causing separation between the end of the 2A sequence and downstream peptide. As a result, the upstream protein will have a few extra 2A residues added to its C terminus while the downstream protein will have an extra proline added to its N terminus. There are four commonly used 2A peptides, P2A, T2A, E2A and F2A, that are derived from four different viruses.
Pros and cons of the two types of linkers
The key advantage of IRES over 2A is that it does not affect the protein sequence of either the upstream or the downstream ORF, which is not the case for 2A.
The main disadvantage of IRES is that the ORF downstream of the IRES is expressed at much lower levels (typically 10-20%) as compared to the upstream ORF in the polycistron. This is why when people express fluorescent proteins after IRES, they often have trouble getting detectable fluorescence signal, especially for in vivo applications where there is a limited number of transgenes per cell. IRES elements can also be problematic due to their size (>500 bp), which increases the difficulty of vector construction and virus packaging.
The key advantage of 2A over IRES is that the downstream ORF can be expressed at a level comparable to (or just moderately less than) the upstream ORF.
The disadvantage of 2A is the possibility of undesired biological effects of the additional peptide residues left behind on either the upstream or the downstream ORF. Additionally, 2A self-cleavage is not 100% efficient, and the efficiency can be strongly influenced by the sequence context of the upstream and downstream ORF. As such, a significant fraction of the translation product from the polycistron could be fusion protein that has failed to self-cleave, which could be a concern in some applications.
If your experiments do not require high-level expression of the second ORF of a bicistron (e.g. a drug-selection marker), then IRES should be sufficient. However, if there are more than two ORFs in the polycistron, or if predictable or equal amount of the multiple co-expressed ORFs is important, we suggest using 2A peptides.
Comparison of different 2A peptides
Of the four commonly used 2A peptides (P2A, T2A, E2A and F2A), P2A is shown to generally have the highest cleavage efficiency (close to 100% in some cases). T2A comes next, followed by E2A and F2A. The cleavage efficiency of F2A is only about 50%. We generally recommend using P2A or T2A in polycistrons.
|Biotechniques. 41:283 (2006)||Translational efficiency of EMCV IRES in bicistronic vectors|
|PLoS One. 6:e18556 (2011)||Cleavage efficiency of 2A linkers in human cell lines, zebrafish and mice|