Immunofluorescence - (Oct/27/2005 )

Hi all,

I have a question for those of you that use/know about IF.

If I stain my protein by IF, it is nuclear....I'd say 99%.
If I do biochemical fractionation (nuclear/cytosol) (controls work perfect) then I see a significant amount of my protein in the cytosol and nucleus.


Why do I see cytosolic protein by biochemical fractionation and not by IF?
Any explanations?

Thanks

Dapo

Hi dapo,

What are your controls?

Are you sure that your nucleus vs cytosol fractionation is correct? I also trap my protein in nucleus and then seperate nucleus from cytosol by frctionation. My protein is still in nucleus by IF. I have never done biochemical fractination.

Hi, thanks for replying.
My fractioantion works (tubulin for Cyto) (p84 (abcam) for nucleus)...no cross contamination. And my protein is in both.

Looks like your fractionation is correct. Well then I will suspect antibodies against your protein. Once I also got weired results, IF data gave me perfect results, so i tried If without my protein, and guess what? I was getting nuclear signal without my protein. Apparently those "so call specific antibodies" wwere giving nonspecific signal.

[Hey,

one thing I fotgot to mention is that IF is very qualitative data. You can quatify with certain sofisticated software, which I do. Mostly it is guess work, We think it is 90%, but it could be 70% as well. In that case you get values in biochemical fractionation. Since I am using quatification software I can tell you it is really a big guess work.

dapo, does your protein have a different form when it is present in the cytosol? perhaps it forms a complex in the cytosol with another protein, and the epitope is no longer recognized by your antibody?

That was certainly one posibility I was considering. But I wasn't sure if that was some great excuse to adjust my data, do you know what I mean? Are there examples of that?

Soni
I understand your point
What program do you use? is it available?

I have tried couple of programs. One is Image-Pro from cybernetics and other one is by Cellomics (its not a software,but a adavanced machine which will measure fluroscence in nucleus and cytoplasm. Unfortunately both of them are preety expensive.

Also aimikins has raised a very good point. If your protein's epitope is masked in cystol, it won't be recognised by your antibody.
I will do a quick experiment to rule out this possibily. I will do a western with your fractionated nuclear extract and cystolic extract with same antobody to check if they can rtecognised cystolic form.

Thanks Soni
Regarding your suggestion, this is how I find out about the presence of my protein in the cytosol. in an SDS PAGE I will detect it, right?

Thanks