@Michele Sandre..I had used the P2A system to express GFP and luciferase (GFP-P2A-luc) driven by CMV promoter. It cleaves  linked protein segments at P2A cleavage site GGA(Gly) CCT(Pro). Thus, two cleaved protein segments are formed. For simplicity one protein that is upstream of P2A cleavage site and another protein downstream of P2A cleavage site. FYI: Add a GSG motif (GGA AGC GGA) prior to P2A SEQUENCE for enhanced cleavage. 

Second question: Inorder to detect the cleavage efficiency, you are supposed to do Western blotting. Un-cleaved protein will have higher molecular weight than cleaved protein (lower mol weight) and thus finding the ratio between the two will give you the percentage cleavage. Mathematically; Cleaved/ Total protein (Cleaved + uncleared)X 100 will give you cleavage %.

I have also attached a manuscript for your reference. Let me know if you need further information.

I asked the same question myself a while ago, and just to make sure decided to try both strategies in parallel in a knock-in approach. I hope to get the results from both in 3 months or so, that would be a direct experimental answer to the question.

We have generated a few mice using the ires sequence, and my opinion about their performance are mixed. It is very hard to generalize how well ires works as there is considerable tissue-specific difference and variation between the loci they are placed in. In the worst example, in a GFPiresCreER allele, we obtained absolutely no Cre expression albeit bright GFP expression. Even the sequence placed after ires might influence its activity in vivo. It worked well in more cases than it failed though.

And lets not forget that it is a transgenic that you are making, probably with multi-copy integration. This should help increase the expression level and thus you have little to worry there.

For the fusion part, in viral systems we have very good experience with the 2A peptide with little fusion. All the data is in vitro. So far there is few (1? I only remember the re-programming mosue with 4 factors) examples of the 2A peptide used in vivo, and those probably did not care much about the fusion proteins.

Well not really an answer, but good luck!

You might interested in the following paper:

http://www.biomedcentral.com/1741-7007/6/40/

Good luck !

Hi Onur,

Thank you very much for answering. I actually plan to make a knock-in transgenic, so I will have a single copy of my transgene in a locus. I'm curious to know what would be the result from your experiment that directly compares IRES and 2A performance in a knock-in transgenic. This would be incredibly useful for transgenic construct design in the future.

Your point on IRES unpredictable and unreliable performance is echoed by many researchers. In the case that you didn't get any Cre expression in your GFPiresCreER, isn't that sort of expected? Expression of sequence placed downstream of IRES is known to be attenuated to the point where you can't detect any expression at all in some cases.

Anyways, thank you again for your answer, and please update your findings if you don't mind sharing what you discover.

Cheers!

Hi Michael, thanks for the link to the paper. I've already read that paper before I posted the question. It does look promising, but I just wanted to know if anyone else can confirm that. Also, it would have been great if they were to include a construct using IRES as their positive control.

It would be nice also to have a direct comparison between the 2 sequences especially in a knock-in transgenic approach (the experiments that Onur is doing currently would answer the question).

Thanks again for your comment.

Here is another reference in PLOS|one

http://www.plosone.org/article/info:doi/10.1371/journal.pone.0027321

My question regarding this construct is related to transient transfections rather than using it to make virus for creating transgenic animals. Does anyone have experience with using it for transmembrane proteins like N-Cadherin. We require a fluorescent label and would like to get away from our current fusion protein constructs because we worry about interference with function. In our hands IRES has proved to be difficult with the full length N-Cadherin.

Hi guys,

Big success with ires, nothing from the 2A peptide! I dont know what the difference is, but the targetting sites were identical. This is hte 2A peptide I used: GAGGGCAGAGGAAGTCTTCTAACATGCGGTGACGTGGAGGAGAATCCCGGCCCT

The peptide should be fine, I assume it is the expression level that caused the problem. I wouldnt give up hope on 2A, but first do a proper test if you want to replace ires with 2A.

Ires looks good in multiple tissues, but is patchy and has a proximal to distal gradient in the intestine (we occasionally see these with the ires). Other organs loks fine.

I hope this was not too late.

Hi,

Of ccourse 2A peptide outperforma IRES in protein coexpression. In dual reporters construct in which I put EGFP downstream of FMDV 2A.  The fluorescence intensity is the same with  monocistron (EGFP). The IRESs(Ev71 or HCV) were uncomparable(10%) with that, although I did not try with EMCV IRES, which tends to be higher than other IRESs.   

The question is 2A cleavage is not 100% percentage. I had done a lot with FMDV 2A, the cleavage efficiency is related to many aspects, such as the protein you chose, the status of cell. et al. One of the strange things is that the cleavage percentage tends to be larger with time going. For example, in dual reporters constructs. the cleavage% was ~60% 16h post-tranfection, while it could reach ~90% 36h post-transfection.

I do not know if it helps. Good luck!

Hi,

In my case 2A peptide was way more reliable compare to IRES. Moreover, we had three proteins divided with 2A and it worked very well.

Hi,

Do you worry about the extra amino acids on the C-terminus of protein you are interested in?

Thanks!

I do not worry about these extras since I have tags on C-terminal of the protein (V5, His-tag, Myc etc). We checked activity, localization and FRET, everything work as well as control (three separate plasmids).

I have used the viral 2A sequence:

GAGGGCAGAGGAAGTCTTCTAACATGCGGTGACGTGGAGGAGAATCCCGGCCCT

This is what Onur Basak (above) has used. Cleavage efficiency is very poor in RPE1 cells 16hrs post transfection. I did not assay any other timepoint, although I will now do so.

@Kia Wee Tan and @Onur Basak, this peptide is not P2A but T2A. It was shown to be less efficient than P2A. Additionally, putting GSG motif before P2A can improve the "cleavage' efficiency ( I will add reference later).

@Michele Sandre..I had used the P2A system to express GFP and luciferase (GFP-P2A-luc) driven by CMV promoter. It cleaves  linked protein segments at P2A cleavage site GGA(Gly) CCT(Pro). Thus, two cleaved protein segments are formed. For simplicity one protein that is upstream of P2A cleavage site and another protein downstream of P2A cleavage site. FYI: Add a GSG motif (GGA AGC GGA) prior to P2A SEQUENCE for enhanced cleavage. 

Second question: Inorder to detect the cleavage efficiency, you are supposed to do Western blotting. Un-cleaved protein will have higher molecular weight than cleaved protein (lower mol weight) and thus finding the ratio between the two will give you the percentage cleavage. Mathematically; Cleaved/ Total protein (Cleaved + uncleared)X 100 will give you cleavage %.

I have also attached a manuscript for your reference. Let me know if you need further information.

Hi,

I wonder if someone can help me! We are using a viral vector with T2A for expression of 2 fluorescent proteins and we intend to change one of them by our protein of interest. We decided to test it first in HeLa cells by transient transfection although it's a viral vector.  We saw some of the downstream protein, very little though (6%), but nothing of the upstream one, which is strange right!? Now we don't know if the problem is from the T2A or if beacuse it's a viral vector it has to integrate in the genome to express properly. Has anyone ever tried to transfect a viral vector with this type of system?

Thanks! 

@Joana Carmelo

I have transfected lentiviral vectors containing a T2A-P2A tandem sequence into RPE1 hTERT cells. Imaging-wise, the construct functioned properly and both proteins were expressed. No western blot was done, but in most cases, the proteins were properly cleaved and targeted to the correct compartments.

I used the pCDH line of lentiviral vectors.

@Kia Wee Tan tank you for sharing your case. I guess we'll try to transfect a different cell line and see if we get a different result.

@Joana Carmelo

You can definitely get expression of viral vectors (lenti, and AAV in my hands) with transient transfection, though keep in mind that particularly with a lentiviral vector, the promoter for your gene of interest must compete with the viral promoter in your plasmid. 

Beyond that, it sounds like your issue may be that you're just not getting very much protein produced at all. So you may want to think about the promoter that you're using. If you're seeing protein 2, but not protein 1, consider whether protein 2 is considerably brighter than protein 1. As has been mentioned in this post, you may want to look at the flanking regions of your T2A sequence. Otherwise, the remaining cleavage peptides themselves can potentially interfere with protein performance. To my awareness, best practices here are to add an n-term flexible linker with a c-term rigid linker (see, e.g., http://dx.doi.org/10.1371/journal.pone.0018556.g006). Finally, I'd think about the transfection method you are using. For HeLa cells, I'd go with a PEI type of transfection. There are multiple kits for this type of transfection, though I have found Effectene from Qiagen works well for heterologous cell lines including HeLa.

You may also want to go with a P2A linker instead of the T2A, as this linker has been shown to produce higher cleavage efficiencies in many contexts.

Best of luck!

@Siddik Sarkar, is the P2A system to express GFP and luciferase (GFP-P2A-luc) driven by CMV promoter plasmid available on addgene? I am looking to acquire such a vector and can't seem to find one. Any info you can offer would be helpful. Thank you.

Hi all,

I want to use p2A to express 2 genes, 1st one an ion channel fused to GFP (~1.7kb) and 2nd one a receptor (~0.4 kb) in AAV and use in vivo.

Any idea if this could potentially work or is it too risky?

I am mostly afraid that the receptor won't be trafficked properly to the membrane because it's 2nd.

Then, if I switch them, the ion channel is too big and might not be efficiently translated?

Any advice?

Dear Nefeli

I hope that this answer is not coming too late. As you already appreciate the short "self-cleaving" 2A peptide strategy is aimed at making two or more separate proteins in the same cell at roughly equimolar levels. And this seems to work pretty well across a range of reports in the literature. The problem is that which often does not get reported in the literature, i.e. the constructs that don't work. The answer very often is to make two or more constructs and check out which works best in a ubiquitous cell culture system, e.g. NIH 3T3 fibroblasts. See for example Mort et al., Cell Cycle. 2014;13(17):2681-96. doi: 10.4161/15384101.2015.945381 in which they have used a CAG promoter cassette reporter construct to express two fluorescent reports at different times in the cell cycle in mice from the Rosa 26 'safe harbour' insertion site. Alternative versions of the promoter reporter construct were first tested in cell culture.

In answer to your specific question, although your ion channel-GFP fusion is potentially quite big it ought to be translated reasonably efficiently in either position, though one will only truly know when the alternatives are tested. More of a worry is to ensure that the construct contains an appropriate extrinsic or intrinsic 'signal' sequence to direct the fusion protein to the membrane, although in the case of ion channels, these sequences are usually intrinsic and therefore not so much of a problem. I would be concerned however, about (i) not expressing too much of the ion channel to ensure that the host cell is not killed by excessive depolarisation and (ii) whether the ion channel-GFP fusion will assemble correctly in the membrane. When GFP is coupled to a secretory protein or fronted up by an N-terminal signal sequence (https://link.springer.com/article/10.1007/s12257-013-0333-1) it can pass through the cell membrane; however, it can inhibit appropriate folding necessary to insert a protein into the cell membrane. Presumably you are giving yourself the best chance of it working by fusing GFP to the C-terminus of your ion channel. If a direct fusion does not work it can sometimes be helpful to include a short b=neutral peptide linker between your protein of interest and GFP to facilitate correct folding. Can I suggest that you check the following threads for further information and up-score any relevant answers:

https://www.researchgate.net/post/What_if_I_add_another_cDNA_sequence_after_the_stop_code_will_it_express_protein?view=58a1891ddc332d99af2b2041

https://www.researchgate.net/post/Can_the_GFP_fusion_protein_expressed_in_Cos7_cells_be_secreted_into_the_medium

Good luck and I hope that helps!

Steve Morley

Hi! Here I have some protocols for your reference: