Hi Ziguo,

Ni (II) can stripped by EDTA, although most of the protocols use 100 mM. NTA is slightly better chelator than IDA (the matrix used for HisTrap by GE Healthcare) . In the case of HisTrap, 1 mM EDTA is tolerated, hence it might lead to only partial elution in the case of NiNTA.

Another way of eluting would be to lower the pH below 7, say around 6-6.5.. In these conditions the Histidines will be mostly protonated and unable to chelate Ni, hence the protein will elute. In this case, provided the complex is not terribly pH sensitive, (most protein-protein interactions are mediated by patches of charges provided by D, E, K, R residues whose protonation state is not affected by working at pH=6.5) you should expect to release Histidines without need of imidazole.

Do consider if Zn is bound to Histidines only or Cysteines too...

In the worst case you can elute with EDTA 10 mM or more and dyalise EDTA out and Zn back in... reconstituting Zn chelating structure should assemble the complex back.

Hope this helps.

Marco

The concentration of EDTA that is (in)compatible with IMAC often depends on the supplier of the Ni-NTA resin. In my experience, IMAC appeared to be compatible with 1 mM EDTA and I required concentrations of at least 5 mM to see elution of bound proteins. However, you might want to consider to lower the pH of your elution buffer (e.g. to pH 5.5 or lower) and perform a pH elution of your complex if this is resistant to the altered pH (the acidic pH can be neutralized by adding an appropriate buffer in your fraction collection tubes). And why not try to use Zn++ for the IMAC and the elution? It sounds like this would make sense from a physiological point of view.

Good-luck.

My suggestion would be to try gradient elution with Imidazole, complex of interest may elute differently than other proteins. You could also try loading again on a Ni or Cobalt column to further purify/isolate the complex. Having a wash at 20 or 30 mM Imidazole also reduces non-specific binding of proteins to column.

I have used EDTA with Cobalt resin, and Cobalt was stripped off, along with protein ; dialysis was unable to remove Cobalt from the protein. Nickel columns may behave differently.

@Kris Pauwels I doubt that the bound zinc would be somewhere outside, so it doesn't probably matter much.

As of the question, you could also the the EDTA and immediately after elution add zinc in concentration slightly higher. However, striping the metal might lead to complex dissociation.

Why don't you elute all the proteins in the same time (for example at 200 mM Imidazole), then change buffer by dialysis to reconstitute the complex. If you still got heterogeneous complexes, you can concentrate you sample and run it through a size exclusion chromatography to purify an homogeneous complex (provided your complex concentration is enough higher than the Kd between the proteins in the complex).

Applying this strategy prevents you from having troubles with EDTA and your Zn...

Hope this will help

1mM EDTA might not be enough to release the Ni2+ (they come down with your protein!) and more might also chelate the Zn out of the complex. 150mM is not very much, I had to use up to 750mM Imidazole for strongly bound his-tags. Your complex might survive that but you need to test that, if you can.

I recommend reducing the numbers of histidines from the construct in that situation. as little as two histidine suffice to bind the Ni2+ matrix (that is how the e.coli contaminations stick to the matrix) but three to five can be enough to seperate your protein from the contaminations in an imidazole gradient. Also, the accessibility of the N- or C-terminus (wherever your tag is) affects the elution profile.

If your complex doesn't like imidazole nor EDTA you can use imidazole for washing only and go back to an elution buffer without imidazole, then spike in high concentrations of histidine to compete out the tags.

Thanks to all of you for the valuble advices.