I guess it could be due to phosphorylation, try to incubate your sample with Lambda Protein Phosphatase (sigma, P9614) or (NEB, P0753S), in order to de phosphorylate to see if you get a single band

Hello Ali,

just a few suggestions:

1) posttranslational moddification(s)

2) Alternative translation start in frame with the correct one (when your FLAG is fused C-terminal)

3) electrophoresis problem (if other proteins also run unexpectedly as two bands

4) degradation of your protein (nice double band sounds more like a specific cleavage as done by e.g. caspases)

Hope you fin a solution

cheers

Michael

Another possibility is the appearance of shorter products due to unexpected end of translation.

I guess it could be due to phosphorylation, try to incubate your sample with Lambda Protein Phosphatase (sigma, P9614) or (NEB, P0753S), in order to de phosphorylate to see if you get a single band

I guess it would be protein modification (not necessarily phosphorylation) or this could indicate that the protein has been cleaved.

As the phosphorylation of a protein shifts the mobility to upwards, your second band might not a phosphorylated one. It might be a dephosphorylated from of it or other PTM's like proteolytic cleavage may yield in protein with lower mol.wt.

I haven't personally used FLAG tag but have been interested in it from my background with calcium-binding proteins. I remember one of the antibodies sold by Sigma had a calcium-dependency and all those negative charges in the FLAG tag are good for binding divalent cations (calcium, magnesium, zinc etc.). If you have 1-2 aspartate, glutamate, serine, threonine, asparagine or glutamine residues in the first five amino acids following the FLAG tag then you have something resembling a calcium-binding, EF-hand loop that could easily have micromolar affinities for calcium (and other divalent cations).

An easy way to check if you are changing the charge of your tagged protein through metal binding is to add EDTA to your sample and running buffers. 1 mM final concentration should be sufficient unless you know that you have been using larger concentrations of divalent cations to stabilize the protein. I think the probable answer lies elsewhere but this is quick and easy to check. Also, for other FLAG-tag users, EDTA might be a useful addition to ensure that you don't have this problem.

In addition to the above possibilities, some proteins do not denature fully or migrate at their expected mass in SDS-PAGE. Whenever we suspect an issue of that nature we run urea-SDS-PAGE gels, and that generally resolves both issues. Otherwise, some form of post-translational modification or processing is likely happening. If it is a problem of not fully denaturing the protein, you can demonstrate that very clearly by running a urea gradient (horizontally across gel from 0- 4M) gel. You'll have one size/pattern in 0M and a single band of the correct size at 4M. Also, one trivial possibility: how fresh is your reducing agent? If it's become oxidized you may have a mixture of reduced and non-reduced or partially reduced disulfides.

is the intensity relation of your two bands app 20/80? If the methanol concentration of your blotting buffer is not correct, the SDS gel will change size during blotting and you will find two protein bands transferred to the NC blotting paper. Mark your gelsize on the blotting paper for control, the size at the begining must be the same as at the end.

Thank you all for your answers. Michael, your suggestion inspired me to do a little digging. Turns out that the ORF supplied in my expression vector has had a number of amino acids added at the N-terminal (corresponding to the 5'UTR) so there are two predicted translation start sites, hence the multiple bands.

Interesting result Ali. In the decades since I created the Flag, I have learned that it almost always is telling you something about your protein. Artifacts are extremely rare. Good job, spotting the cause!

@Ali: Thats interesting. I will keep this in mind. Thanks for letting us know.

I'm also having a similar issue with a b-tubulin loading control protein in heart and kidney total protein lysate.

Attached is an image of the western blot i ran. I also wonder if the issue is phosphorylation of b-tubulin?