Re: [ccp4bb]:SUMMARY- unusually high solvent content

["JoaoMDias" <jmdias@dq.fct.unl.pt>]

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

Dear All,
I want to thank you for all the answers (about 30, sorry it is impossible to
quote you all, but many thanks) to the question about "unusually high
solvent content in protein crystals". I also thank the people that even sent
me reprints. Here follows my contribution as a short summary.

Question:
"According to Matthews, (Mathews, B. W. (1968). J. Mol. Biol. 33, 491-497.),
the Matthew's coefficient of protein crystals is usually comprehended
between 1.7-3.5 A3/Da.
Are there many proteins with a much higher solvent content that the Matthews
limits?
Do you have experience with protein crystals with high solvent contents?
Could you point me some references to published works, or reviews about
protein crystal structures with unusual VM (above 70%)?"


Summary:
As the number of PDB deposited structures increases exponentially, it is
also observed the increase in the number of proteins which present solvent
contents (VM) which deviate from the average.

Kleywegt pointed the PDB report
http://www.cmbi.kun.nl/gv/pdbreport/pdbreport/revindex.html

A search through the PDB deposited coordinates, analysed with WHATIF (Hooft
et al., (1996), Nature, 381, 272; check the) states that 487 present a low
VM, 11898 present an acceptable VM, and 1061 structures have a high VM
(comprised between 4.0-6.9 E3/Da) and that 296 have a very high VM (above
7.0 E3/Da).
It should be taken into account that some of these entries reflect errors in
terms of molecular weight, space group, cell constants, Z values, NCS
molecules.


A short example of recently deposited structures, which present unusually
high solvent contents, are compiled in the following list:
Protein name; PDB code; VM(E3/Da); Solvent Content (%); Diffraction limit
(E); Reference
Apolipoprotein H; 1QUB; 8.5; 86; 2.70; Bouma et al. (1999), EMBO J., 18,
5166-5174
Apolipoprotein H; 1C1Z; 8.9; 86; 2.87; Schwarzenbacher et al. (1999) EMBO J,
18, 6228-6239
Serine chemotaxis receptor; 1QU7; 7.4; 83; 2.60; Kim et al. (1999), Nature,
400, 787-792
Acyltransferase; 1B5S; 11.5; 89; 4.40; Izard et al. (1999), PNAS, 96,
1240-1245
Proline peptidase; 1AZ9; 4.5; 72; 2.00; Wilce et al. (1998), PNAS, 95, 3472
Duck carboxypeptidase ; 1QMU; 4.6; 73; 2.70; Gomis-R|th et al. (1999), EMBO
J., 18, 21, 5817-5826
Apolipoprotein AI; 1AV1; 5.8; 72; 3.00; Borhani et al. (1997), PNAS,
94,12291-12296
Colicin Ia; 1CII; 5.9; 78; 3.00; Wiener, et al., (1997), Nature, 385,
461-464
Green fluorescent protein; 2EMD; 4.2; 71; 2.00-1.60; Palm et al., (1997),
Nat. Struct. Biol., 4, 361
Chloroamphenicol phosphotransferase complexed ATP; 1QHX; 8.8; 86; 2.50;
Izard & Ellis, (2000), EMBO J., 19, 2690-2700

I can not forget to mention a particular answer by Jim Naismith:
"Dear Joao,
Yes we did one dTDP-glucose complex of RmlB (its in the PDB with the data)
comes out in Structure next year.
As I said to my student there are either three of four subunits in this asu,
no way does a xtal diffract to 1.9A with a Matthew No of 4.93 (two
monomers). 'Go back and run AMoRE properly.' After several attempts with my
encouragement 'I'll eat my hat' etc, he of course proved me wrong. Yet more
evidence for the dictum that supervisors have a tendency to confuse fact and
their opinion..."

As a final conclusion I leave you with Shakespeare:
"There are more things in heaven and earth, Horatio,
than are dreamt of in your philosophy."

Best regards,
Joao Miguel