Re: [ccp4bb]: Ions in the internal protein cavities

["Aleks Roszak" <aleks@chem.gla.ac.uk>]

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

Hi Petr,

Sorry for responding to your request so late but I have read it only today. I
am not sure how helpful it will be for you but I worked on a case of a
dodecameric enzyme (dodecamer formed by four trimers, overall tetrahedral NCS
symmetry P23) in which I observed both positive and negative non-metal ions on
the NCS-trifold axis. Both of these ions you have actually mentioned as
possibly present in your structure:

1) Phosphate (or sulfate) ions from the crystallisation conditions were found
on the NCS-trifold in a relatively nice tetrahedral density and they are being
H-bonded to 3 histidines and 3 water molecules trigonally positioned around the
axis. Distances from the central P or S atoms to the histidine N atoms are in
the range of 3.7-3.8A. That does not seem to resemble your case.

2) The TRIS protonated cation was also located on this NCS-trifold as being
H-bonded to 3 glutamate and to 3 threonine sidechain oxygens. The distances
from the central carbon atom to glutamate carboxyl oxygens are 3.75 and 4.33A
which is kind of similar to your case. In my case however (at 1.8A resolution),
the density is not spherical but forms kind of propeller with three CC-C-OH
arms positioned in trygonal manner around the N-CC bond which is oriented along
the NCS-trifold [where the CC = the central carbon atom]. Perhaps at 3A
resolution this could look somewhat more spherical, especially if the ion is
disordered on the axis, i.e. is present in two opposite orientations, but I am
not sure. If the N atom is ignored though (as disordered) then the density is
relatively flat and trigonal (propeller part). In our case TRIS was also used
as a buffer during prep, and sometimes not used in crystallisation at all, so
we assumed that the TRIS ion was present on the 3-fold axis prior to
crystallisation (this enzyme is known to be active not as a monomer but as a
trimer or dodecamer).

Please let me know if you could fit a TRIS ion in your density. I can provide
you with the TRIS coordinates if needed.

Aleks




On May 31, Petr Leiman wrote:
> Subject: [ccp4bb]: Ions in the internal protein cavities
> ***  For details on how to be removed from this list visit the  ***
> ***          CCP4 home page http://www.ccp4.ac.uk         ***
>
> Hello,
>
> I have a fairly spherical piece of density on a crystallographic 3-fold
> axis to which 3 symmetry related Asps are pointed. The piece is pretty
> big. Its volume is slightly larger than that corresponding to S in the
> nearby Mets if contoured at the same contour level. The height of the
> density peak in that piece is aging slightly (about 1 sigma) greater
> than any protein density around it. Apart from 3 Asps, the piece is
> surrounded by hydrophobic residues.  The distance from the center of the
> piece to the nearby Asp is about 4.0 A. The distance between the O's of
> the corresponding Asps is 6.9 A.  I have several data sets and this
> piece of density present in all of them. The piece is there when I
> calculate the maps using the heavy atom phases as well when I use the
> model phases. Because the piece is surrounded by the hydrophobics I
> think that the material of the piece is there from the moment when the
> protein get folded and this material did not come from the purification
> or crystallization solutions.
>
> The distances are to long for any metal I can think of. And the density
> is not high enough. I had a suggestion that the the piece is just
> water(s) artificially enhanced by the crystallographic 3-fold but I do
> not buy it. Too long distances for 1 water molecule. If I put 3 they
> _have_ to be covalently bound then, because they would be something like
> 1.9 A apart.
>
> I have only 1 candidate now - a phosphate. Has anyone ever seen the
> phosphate buried inside the protein molecule and coordinated by 3 Asps?
> Any other suggestions? If anyone knows a positively charged molecule
> with a triangular or tetrahedral geometry that is present in the cell it
> would be of great help.
>
> Petr
>
> P.S. Sorry for such a long letter.
> P.S. # 2: The resolution of my maps is 3 A.
>
>
>-- End of excerpt from Petr Leiman



-- 
**************************************************************************
* Aleksander W. Roszak, Ph.D.               E-mail: aleks@chem.gla.ac.uk *
* Protein Crystallography                   Fax:     +44-(0)141-330 4888 *
* Department of Chemistry                   Tel:     +44-(0)141-330 6175 *
* University of Glasgow                              +44-(0)141-330 6180 *
* Glasgow G12 8QQ                                    +44-(0)141-330 4476 *
* Scotland, U.K.                   Web: http://www.chem.gla.ac.uk/~aleks *
**************************************************************************