Testing nucleotide-protein interaction - (Jun/17/2013 )

Hi everybody

I work with an intrinsically disordered protein with unknown function and no close homologs. Psi Blasts show some low degree of homology to ATPsaes, but with e-values above cutoff. Motif searches predict an incomplete Walker A motif. If so, only the GKS motif is conserved, but not the P-loop.
The protein forms a stable complex with Ca-calmodulin.

I tried atpase assays, but there was no activity. Upon addition of Mg-ATP and Mg-GTP there is an increased precipitation of the complex with increasing GTP concentrations, but not with ATP (I checked the pH of both solutions).

Running a native gel with lower GTP/ATP concentrations does not result in a visible gel shift. I tried several other nucleotides, but without result.

Could you give me some ideas how to test the function/nucleotide binding abilities of this protein? Is there a change that this protein is a nucleotide binding protein at all?

Thanks

you could try equilibrium dialysis with labelled diphosphonucleotides to determine nucleotide binding ability.

You could try measuring binding enthalpies by ITC.

Also, NMR is a popular method to study intrinsically disordered proteins. Get a 2D spectrum with and without NTPs and see if the spectrum changes.

You could also try circular dichroism. If NTP's bind, then you might expect some local folding of your protein, which can be detected by CD.

If you want to see a band shift in native page, you will likely need the NTPs in your gel and gel running buffer as the off rate of the NTP could be quick enough that the NTP dissociated shortly after entering the gel. Similarly, you could do gel-filtration with and without NTP's to see if there is any change in migration, which would be indicative of either oligomerization or significant conformation changes.

You could simultaneously label the N- and C-termini with CFP and YFP and see if you get a FRET signal upon addition of NTP's, which could occur if NTP binding stimulates some folding.

If you have a tryptophan in your protein, you could measure intrisic fluorescence in the presence and absence of NTPs. This would work best if you only have one trp in the seuqnece. If you have multiple trp's, it still might work as is, or you could mutate all but one of the trp's to Phe (or something similar) and then measure intrinsic fluorescence.

There are lots of possibilities really. Sometimes you just have to be a little creative to find a way to test it with the facilities you have.