Protocol Online logo
Top : New Forum Archives (2009-): : Cell Biology

uncertain about stable overexpression - discrepancy between different antibodies (Apr/20/2009 )

Hi all,

The following explanation is a bit longer but I could need some help here.
I currently try to figure out what is going on in my stably transfected cells and could need some advice, ideas... ,... ,.

I tried to overexpress my gene of interest in fusion with a FLAG-tag (C-terminus).
When I started screening surviving cells with my FLAG-tag antibody they show the overexpression compared to respective control cells using confocal microscopy and western blot. Huurraayy... .
However, I can't detect the overexpression with the regular antibody which is working very nice in transient transfections (but is not the most specific one - that is also why I used a flag-tag in the beginning).
I can also see that, compared to the transient transfections my selected clones do express my protein much less (using the flag-tag antibody), which is quite usual as far as I know.
So to put this to an end, I was thinking of analysing the mRNA expression to get an idea about how well the overexpression is.
I started the whole overexpression approach to complement some data I have from siRNA studies; the thing is I do not see a phenotype so far in the overexpressing clones (which is crap obviously and I do lack a good explanation for that...). Currently I am afraid that the overexpression is not efficient enough to cause a phenotype, or the overexpression does not cause a phenotype at all - but how to proof that??

The second thing I observe on a western here is that my antibody detects two bands for the flag (one is too high, the other one at the predicted size) and I am thinking about checking potential modifications of my protein including phosphorylation, glycosylation whatever...,. Since I have not the slightest idea where to start I planned to use 'global' inhibitors for considerable modifications - it would help a lot if somebody could recommend some.

Thanks a lot in adavance for any shared thought.
Best regards


I presume your stable and transient transfections are both FLAG tagged? If they are not, is the epitope for the antibody near the FLAG tag? There could be some steric hindrance or a mis-folding of your target protein that is preventing the antibody from binding.


Hi Bob1,

thanks for your reply and ideas.
Yes, the transient and stable transfections are done with the same construct. Indeed, I first tested the transfection transiently and then started selecting the cells. Also the regular antibody, which is showing high expression in the transient tranfection (though transfection efficiancy is rather low) reveals two strong bands on the gel (the ones I also detect with the flag-antibody). So for the transient situation it looks similar between the antibodies.

I rather assume that the overexpression is difficult to show with the regular antibody but is there since I can show the expression of the flag. Question remains if this sounds convincing to you?

Best regards


There is a possibility that the recombination event for the stable tranafection happened in the middle of your coding sequence, so the flag is still expressed, but the protein isn't. I would try doing a southern to determine if you have had the plasmid containing sequence insert at all. Did you transfect linear or circular plasmid for the stable transfection - if circular, the stable transfection rate will be much much lower and the recombination could take place anywhere in the plasmid, so you need some way of determining the suitability of the insertion. A DNA extraction followed by RNase treatment and then PCR using coding sequence spanning primers may work (as you would for cDNA cloning). This is an absence of a band indicates that "something is wrong" type assay, you would have to be sure of your PCR and include a positive control (plasmid containing your gene) for this to be definitive.

How did you select your cells? Did you seed them at a low density, select, and allow the remaining cells to grow into colonies, pick each separately and grow from there, or did you just select normally transfected cells and do a poly-clonal type thing? Single colony selection has a few issues of its own, in part due to the issues above, but is better in the long run as you have a single population of cells with the insert(s) in defined position(s) rather than a poly-clonal population with multiple clones with different levels of insertion and different levels of expression.

Are you keeping the cells under selection pressure? Even with "stable" transfection, removal of selective pressure can allow the cells to drop the inserts, especially as the form of the genes is quite different to eukaryotic structures - there is some sort of nuclease that removes unusual sequence, or so I read somewhere once.


Hi Bob1,

yes, cells remain under selection except when I seed them for an experiment.
I transfected a circular plasmid and isolated colonies using cloning rings (which is a kind of clonal selection I'd say).
I doubt that the insertion is a problem since the plasmid is designed in such a way that my gene of interest and the selection marker are expressed as a fusion mRNA (selection marker is downstream of my gene).
So if I have a foggy insertion the cells should also loose their resistance.
On a western blot I also see the expression of the full lenght product. I find it just odd that it is not possible for me to show the overexpression with my regular antibody but with the flag-antibody.
I guess if I show some lightcycler results and proof the overexpression of my gene it should be fine.
In my opinion (because this is what I see also with the flag anitbody) the stable transfected cells expresses my gene much less compared to the transient transfected ones and this is why I do not detect it with the regular antibody.

Thanks again for your reply.


In my opinion (because this is what I see also with the flag anitbody) the stable transfected cells expresses my gene much less compared to the transient transfected ones and this is why I do not detect it with the regular antibody.

Damn, why didn't I think of that earlier!

The transient cells will have excess plasmid floating around and being expressed, whereas the stable should only have a few copies at most inserted (usually concatemerised), so that could well be the explanation.