From: Arthur Rowe <Arthur.Rowe@nottingham.ac.uk>
  To  : rasmb@bbri.harvard.edu
  Date: Tue, 4 May 1999 16:38:53 GMT0BST

Why determine D ?

Dear RASMBers

The multiple responses to the questions originated by Lumelle and Yarabe raise an 
interesting point - namely, why do we actually want to estimate values for D ? It is 
not, taken by itself, the most totally interesting parameter so far as interpreting the 
molecular properties of a system is concerned, nor does it actually give you any 
information for a well-defined system for which M (via MS/sequence) and s are 
already known. Indeed, for the latter case it will usually be a LESS accurate way (as 
compared to s/M) of estimating the friction, and hence probing shape parameters.

Nonetheless, as various people have brought out, there is value in seeing if a 
putatively *simple* (i.e. monodisperse) system is really what it seems to be, 
especially if M is not pre-known.  The major value lies in being able, by fairly 
simple means, to check via s + (estimated D) that an M value known accurately 
either *a priori* or from SE is duly echoed back. This is quite a critical test of 
monodispersity.  The use of sedimentation velocity data means no extra 
experiments, whether time derivative or other/LAMM approaches are used. 
Especially perhaps in the former case (or if dn/dr data is the basic input), then the 
treatment can be extended to a small number of additional components. 

To give an idea of the level of precision attainable, Neil Errington and I (MSS 
submitted) have shown that one can resolve just 3% of native conformer in the 
presence of 97% recombinant TMADH, and that a range of conformers of myosin S1 
in the presence of effectors can be monitored for constancy and homogeneity of 
monomeric mass to within a few % by this approach (Progress in Colloid & 
Polymer Science, in press).  

The question of DLS has been brought up. We do have and use DLS equipment 
here, but it is important to realise that DLS is very poor indeed at resolving 
multiple species whose mass difference is less than 2:1 (we view it as a triumph if 
we pick up a some BSA dimer by DLS, whereas a g(s*) profile makes the detection 
and quantitation of the latter an easy job). DLS is a great tool for screening for 
aggregates in pre-crystallisation conditions, the point made about the nice 
temperature control attainable is a good one, and clearly one more estimate of D can 
do no harm in any study.  I am sure DLS also has application to characterising large 
polymeric systems, as Steve Harding points out.  Nonetheless, even a fairly 
hard-nosed referee ought to be convinced by evidence from sedimentation work 
alone, combining SE and SV.  

Sorry - this seems to have gone on to some length !

Best wishes to all

Arthur Rowe
*************************************************
Professor Arthur J Rowe
NCMH Business Centre
University of Nottingham
School of Biological Sciences
Sutton Bonington
Leicestershire LE12 5RD   UK

phone/voicemail					+44 (0)115 951 6156
phone/telefax							    +44 (0)115 951 6157
email														            	arthur.rowe@nottingham.ac.uk
www.nottingham.ac.uk/ncmh/business
*************************************************