[administrator]

Great question, semor! I think, in general, people tend to "forget" about the fact that 15N CSA in proteins is not uniform and can introduce significant errors in model-free results (e.g. S2). A common trend is just to use a CSA value from a widely cited, well respected paper about model-free analysis (like Backbone dynamics of Escherichia coli ribonuclease HI: correlations with structure and function in an active enzyme. Mandel AM, Akke M, Palmer AG 3rd. J Mol Biol. 1995 Feb 10;246(1):144-63.)


We could try to get opinions of NMR relaxation experts about this issue when they return from the Gordon conference.

Cheers,

Mark

Originally Posted by semor

Hi !

I have a quite general question about the value used for the CSA while studying protein dynamics of 15N-1H vectors with model-free approach.

In the literature, we mainly find two values for the CSA (-160 and -172 ppm).

There is, if I understand well, a link between the bond length and the CSA, but everyone seems to agree about using the same value of 1.02 A which should give rise to a mean S2 of 0.85 for secondary structure when combined to a CSA of -172 ppm. When using a CSA of -160 ppm, the mean S2 for secondary structure should slightly rise up from 0.85.

The manuals for several programs as relax (d'Auvergne) and Model-free (Palmer) suggest a value of -160 ppm for CSA (combined with a bond length of 1.02 A).

However, some recent papers (Biochemistry, 45, 11414-11424, 2006 ; JACS, 127, 3220-3229, 2005) suggest using a value of -172 ppm.

Also, what about the CSA variability from one vector to another (JACS, 128, 7855-7870, 2006) ?

What is the best (site specific or not) value to use and, most importantly, why ?

Thanks !


Sébastien