[ccp4bb]: summary of protein crystallization with His Tags

[Narayanan Manoj <nm77@cornell.edu>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

The question:

Dear all,
>            I am trying to crystallize a protein with a 10 His tag. I
>would like to know of any experience with a successful crystallization.
>Is a 6 His tag better than a 10 His tag or is it best to have no His tag
>at all?
>many thanks in advance
>manoj

Responses so far:

1)
My experience with one protein was the following:

We made expression clones with and without his-tag (residues
GAPAAAHHHHHH).
The tagged protein could be purified easily with Ni-column followed by 
hydrophobic interaction chromatography. It also crystallised but
diffracted 
anisotropically to 3 A in best direction and 4 A in others. This protein
was 
however very useful for biochemical studies because we could bind it to 
Ni-surfaces, columns etc.
The non-tagged protein took longer to purify (5 columns), and to
crystallise 
- crystals appeared after 6 months. But - they diffracted to 1.3 A and
we 
solved the structure.
I tried to solve the structure of the tagged form but was unsuccessful,
I 
think the crystals have some strange kind of twinning or disorder.
So, I would recommend to express both, if you are lucky the his-tagged
form 
will crystallise, while waiting for the crystals to grow I would also
purify 
the non-tagged form and set this up also.

Mark van Raaij (markvanraaij@hotmail.com)


2)
We have a protein that crystallizes with and without a 6-His tag.  When
the 
tag is there, it participates in crystal packing. This gives us a
different 
unit cell but the diffraction is no better or worse the the tag-free 
crystal form.
We haven't tried 10-his tags.  A student whose committee I sit on had
some 
trouble with them dimerizing, but I don't know any more details than
that.
         Phoebe


3)
I work on a protein that crystallises with the His-tag cut off
(the pProEx-Ht[a-c] vectors from Lifetechnologies come with a 6xHis-tag
and a Tev cleavage site), but not with the His tag left on. As far as I
know, His-tags are believed to be unstructured and flexible, i.e., they
inhibit crystallisation. Therefore, the shorter the better, but this
must
not always be true. The charged tag may as well positively alter your
proteins behaviour.
Tim


4)
I think it is a bit more differentiated. If everything that is
unstructured
and flexible would inhibit crystallization, we would not have any
structures
in the pdb that don't show defined N (40%+ or so) or C termini (20%?).

One could argue that if it *prevents necessary crystal contacts from
being
formed*
it is bad. Unfortunately, we lack predictive power wrt crystal contacts,
even given a
structure model.

A tag also could prevent proper folding if the termini are buried. This
appears statistically to be more likely for C termini than N. Given a
model, you could look at that. Novel structure, up the creek.

Anekdotal case examples exist either way, we had a 68 residue tag on a
(large)
protease and it diffracted to 1.8A, others report even the vicinity of
his
tag
plasmid containing vial in the same room as crystals jinxes theirs.

All of the above translates into we have no clue, because we seldom hear
about
negatives, which we would need to establish proper statistics. So maybe
after
unbiased Structural genomics data bases have been filled, we may be able
to
give a
likelihood of reduced/increase success of his vs non-intagged
constructs.

A (probably outdated) collection wrt His tag successes can be found

http://www-structure.llnl.gov/internal/his/histag.htm

The ease of purification is perhaps worth it, also there are some
systems
(R-TEV) that cleave the histag off and leave only one residue of the
TEV site on the nterminal but I am no expert.

Bernhard Rupp



5)
It is probable better to remove
a stretch of continuous positive residues
which can be difficult to stabilize
in protein-protein interaction.
And with the His-tag you add probably
also a linker before the protein sequence.
Both are most likely to be disordered
and to prevent crystallization.

But if there is some trouble to cut it
you can always try crystallization
without removing it:
we succeeded without removing the his-tag
to crystallize
a 24 kDa protein with 6 His and a linker of 15 residues
(disordered)
and
a 22 kDa with 6 His tag and a short 2 residues linker
(partially ordered in one crystal form).

good luck
louis RENAULT

6)
We had good success by removing the his tag (an enzyme from Thiamine
biosynthetic
pathway). I think, 6 his should be better than 10.

regards,
Mathews

7)
I'm interested in any answers you get, plese share them.

D


Thanks again to everyone who replied.

manoj

Narayanan Manoj,
372, Baker Lab,
Cornell University,
NY 14850