Dear  Austin

Interesting one!  Did you 'pull-down' them?

Regards

Prash

Not yet. I've got some antibody for the target protein, but it's been set aside for doing a immunofluorescence assay against the plasmid with the GFP cut out (To ensure the C-terminal-bound version of the fusion protein isn't localizing differently due to the presence of the GFP).

I'm not very familiar with IP assays besides using premade nickel-His tag columns; Is it possible to prepare custom microbeads bound to our own antibodies?

I think that could be possible, but you could preferentially opt for DNA aptamers instead as a target.

 In our lab we had an interesting case- we fused GFP to a small protein (11kDa), and the GFP caused it to alter its cellular localization totally...

Good one, Yishay :-)

Austin,

Can I assume these are over-expressing lines? Transient or permanent expression? Is there any chance that the C-terminal portion of your protein might be proteolytically cleaved or have a transmembrane domain?  Could you tell us anything about the functional domains of your protein: localization/interaction/degradation domains that could be resulting in something kooky with the GFP present?

Cheers,

Nick

Thanks Nick,

Yes, these are over-expressing lines. The protein of interest is a bacterial ankyrin, and as far as I am aware (I am new to the lab, so I'm not previously familiar with ankyrin proteins) the primary functionality is distant from the C and N terminals. Right now we're seeing localization of the ankyrin-GFP to, we believe, the Golgi, which is in line with our predictions, and the control GFP without ankyrin is showing nonspecific cytoplasmic fluorescence. It's just our GFP-ankyrin fusion protein that's showing no expression whatsoever, even compared to just the nonspecific GFP.

Given that the GFP is upstream of the ankyrin, even if the inserted ankyrin was out of frame (Which it's not, as best as we can tell with sequencing) the GFP should still (in theory) be folding correctly and at least displaying nonlocalized cytoplasmic fluorescence. The reason I suspect that the issue may be due to folding is that the ankyrin we're working with is over 3.5 kilobases compared to the 700bp for GFP; My thought is that if the protein begins folding in a 5' to 3' manner, for our Ankyrin-GFP fusion protein the larger ankyrin would fold first, and the GFP would be relatively insignificant compared to the rest of the length for folding, while the GFP-Ankyrin protein might encounter issued with the GFP barrel structure formation due to steric hindrance from the longer ankyrin peptide tail. However, this is the first time I've worked with translation-level fusion proteins, as before I've only done work with binding markers to proteins, so I'm not sure what issues to keep a watchful eye out for when working with fusion proteins.

Austin, I suspect the problem is folding, as you've surmised.  If I recall correctly, ankyrin has a repeat structure.  Repeat sequences can be a mess to fold.  Proteins don't necessarily fold in a strictly N- to C-terminal sequence, and if any part of ankyrin interacts with the gfp it may be interfering with formation of proper beta-barrel structure.  It is certainly possible to add a fusion partner to the C-terminal end of gfp.  We've used a gfp-bsd fusion to transform Babesia bovis to blasticidin resistance and have nicely glowing parasites.  Have a look at your transformants by western blot with anti-GFP or anti-ankyrin and make sure you're getting full-length protein expressed.  The ankyrin sequences may be destabilizing the mRNA, or if it is a folding problem you may find that the protein is very rapidly degraded and you don't get functional protein for that reason.  If it is a folding issue, a short, unstructured spacer sequence between gfp and ankyrin might help.  Good luck.

What David said.

Maybe you have signal peptide sequence N-terminal of your protein that is necessary for localization to golgi? 

If this is the case that could interfere with right protein placement and folding.

I suggest you to run your protein on SIignal P software to predict the presence of signal sequence

Thank you David. I will look into making a spacer sequence, as that seems like it will be the best solution to folding issues.

Elena, I will look into the Signal P software to see if we have recognizable signal sequence. Currently, though, we're seeing no fluorescent expression of any kind, even nonspecific cytoplasmic fluorescence like what the GFP plasmid without ankyrin displays, which is what I would expect if just the ankyrin was misfolding and not localizing.