Atypical binding of 6His to Ni-NTA (Was: reversible protein staining)

Dima Klenchin klenchin at
Mon Dec 4 10:14:41 EST 2000

"Dr. Duncan Clark" <Duncan at> wrote:
:In article <90946s$hqq$2 at>, the eminent Dima Klenchin
:at UW wrote
:>"My" protein binds very poorly and to elute even at 10 mM
:>imidazole. No problems - I just load without any, wash extensively
:>with 5 mM, briefly with 10 mM, elute with bump of 100 mM 
:Are you using Ni or a Co charged column? For some proteins we found that
:Co is way better than Ni for binding etc.
:Maybe worth a go.

I used Ni only so far. Definitely worth a try. One of these days...

I also remember your (?) post that Ni/Co-IDA for some proteins
gives lower contaminants level than NTA and was meaning
to try this too. 

What I really don't understand is _why_ it binds so poorly.
It has myc and 6his tags at C-term. Anti myc Ab IP it very well - 
thus (because myc peptide has no structure) both tags must
be exposed well. 

Moreover, another version of it that is 6His N-terminally tagged,
displays practically the same binding properties (elutes way,
way too early for a normal 6His). 

Even more strangely, the binding has abnormal pH dependence:
it _increases_ with decreasing pH. That is, at pH 8.0, there is
essentially no binding at 5 mM imidazole, while at 7.0 at least
90% binds happily. With every other 6his-tagged protein I worked
with, the direction of pH effect is opposite. 

I think it is really, really unlikely that this protein happened to
have tags on both ends magically buried and tend to think there
is something in this protein that repulses it from Ni-NTA complexes 
(thus k(on) very low and Kd lower correspondingly). But I don't 
have a clue what this might be. This is 145k, very acidic protein, 
heavily phosphorylated to top it off. 

        - Dima

More information about the Proteins mailing list