[ccp4bb]: Summary: unknown cation at 2.6-3.0A bonding distances

["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         ***

On October 4th I've posted the following request:
>
> Could you please suggest any good site, database or so, in which I could find
> details of coordination (bond distances, coordination geometry, etc) for some
> light metal ions, or even better any single atom ions.
>
> I seem to have an ion [quite spherical density up to 6sigma, definitely much
> higher than water, refines to B=2.0 while average B for this structure is 26
> (from TLS refinement)] coordinated by 3 main-chain carbonyl oxygens of a
loop,
> and possibly also by 2 water molecules, at distances of 2.6-3.0A. The three
> carbonyl oxygens plus one water molecule oxygen form approx. a (relatively
> flat) pyramid with the unknown ion at the apex.
>

I would like to thank Jeanette Hobbs, Frank Vondelft, Deena Oren, David J.
Schuller and John Jenkins for their very quick and useful responses.

Jeanette, Deena, David suggested that what I see is possibly Ca++ or Na+ while
John suggested K+. I myself first thought that that is Na+ because I believed
that Na-phosphate was used as a precipitant. However, the refinement with Na+
was still not fully happy and Na+ B factor was still relatively low. There was
still a residual positive density at the Na+ and the density was at slightly
higher sigma level (and the B factor lower) than of the S atom of neighbour
Met. That led me to believe that the ion is 'heavier' than Na+.

Now, Deena suggested I read the paper on Na ions:
Nayal and DiCera, JMB 256, 228-234 (1996),
while John suggested to see the Marjorie Harding site
http://tanna.bch.ed.ac.uk/index.html and/or an article by Marjorie Harding,
Acta Cryst. D57, 401-411 (2001). This last paper is on Ca, Mg, Mn, Fe, Cu and
Zn ions from which I found that the metal-oxygen distances for these metals
seem to be shorter than those I observe.

Finally, I found the newest Marjorie Harding paper on Na and K ions interacting
with proteins: Acta Cryst (2002), D58, 872-874. From this paper it become very
clear to me that I see K+ ion which is coordinated by either 6 oxygen atoms (3
of the mainchain, others water mols, in 2.2A structure) in a distorted
octahedral geometry or by 7 oxygen atoms (in my newest structure, another
mutant, of higher resolution 1.95A, where I can see more waters).

Where could I have the K+ ion from? This problem was solved yesterday when the
author of crystallisation of these two mutants returned to the lab after few
weeks of absence. He simply have used K-phosphate for crystallisation, not
Na-phosphate...

Additionally, Frank Vondelft suggested to see the site
http://metallo.scripps.edu/
Frankly, I did not find this site that useful to me because what I was looking
for was a database which would just allow to choose K+, or other metal ion, and
get the typical metal-ligand distances, coordination geometry, etc., while I
couldn't do it (or I didn't know how) at that site. Nevertheless, the site can
be very usefull for other purposes.
Also, probably because the interactions between Na+ and K+ ions and the protein
are mainly of the ionic nature, these ions are usually not included in the
typical lists of metals in which the metals interact mainly by formation of
coordination bonds.

Aleks





David J. Schuller:
The very low B indicates that the atomic number is probably higher than
what you have in there now.

IS there any biochemical evidence? Atomic Absortion Spectroscopy, for
instance?

Did you add any metals as part of  the crystallization conditions?

I think Zn tends to be tetrahedral, so maybe Ca might be a good first
guess.

Anomalous scattering - most metals scatter anomalously, and the
wavelengths are characteristic.  For example, Zn has a peak at 1.28 A, so
if it doesn't show up in Cu rotating anode data (1.54 A) but does at
typical synchrotron wavelengths (~ 1.0 A) Zn is a good bet.  Ca would show
up at 1.54 and above.

I would recommend the U. Washington Anomalous Scattering periodic table;
http://www.bmsc.washington.edu/scatter/AS_index.html


-- 
**************************************************************************
* 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 4476 *
* University of Glasgow                              +44-(0)141-330 6180 *
* Glasgow G12 8QQ                                    +44-(0)141-330 6449 *
* Scotland, U.K.                   Web: http://www.chem.gla.ac.uk/~aleks *
**************************************************************************