From: jack correia <jcorreia@fiona.umsmed.edu>
  To  : rasmb@bbri.harvard.edu
  Date: Fri, 24 Apr 1998 10:10:47 -0500

g(s), gaussians, etc.

RASMBers,

	In response to my last comments Peter Jeffrey replied:

"If you are dealing with a reversible equilibrium why not use sedimentation
equilibrium and be done with it?"


	And my short answer to him (the long version follows below) is, I agree!


	If the system were stable, I agree that sed equil is the best way to go -
for the work I have done on some kinesin constructs ( Correia et al Biochem
34, 4898 1995) it was the only way to get the stoichiometries and the K's.
The weight average values are consistent with the answer but we never could
have done it that way, at least not with absorbance optics.  But we have
found with other constructs (see Rosenfeld et al JBC 271, 30212 1996) that
irreversible aggregation intercedes and definitive work is not possible.
Even for a monomeric construct in that work I had to use Svedberg or
gaussian fitting of the g(s) to verify the monomer state.  Short column
might help but it is generally not as good at low concentrations as the 3
mm columns are.  (I know Dave Yphantis has been running them at higher
speeds so I may be wrong about that for interference - we plan to do some
trials to see what the limitation are, but my interests are answers for the
systems I am working on right now!)  For the case of tight affinities or
stoichiometric binding as described in that email request it is possible
that neither method works.  The advantage of Walter's gaussian method for
antibody interactions is in getting the mass and thus the stoichiometry,
not K, as well as the distribution of species formed.  Tom's new
fluorescence system will help indicating that fluorescence is historically
the better way to study high affinity, low concentration interactions.

	For all the work we do on tubulin vinca interactions (Lobert et al Biochem
34, 8050, 1995; Biochem 35, 6806 1996; Biophys J 72, 416 1997; etc)  there
is no question - sed equil will not work due to denaturation problems
coupled with the size of the polymers & thus the times to equilibrium.
This is not to say we don't try to measure MW.  (Occasionally it has worked
as long as you are content with an extrapolated MW.)  Even in sed vel runs
we see irreversible junk at higher temperatures & we have not figured out
how to stop it!  It really happens in the few hours of a velocity run.  The
problem then of course is the set of assumptions you must make about f/fo
and the correct Si values.  There is no unique set of answers.  We
currently make the simplest assumption and do relative comparisons of the
K's inferred.  Even the people predicting S values from crystal structures
have difficulty getting unique answers and use hydration almost as a fudge
factor.  (This is not criticism but my, possibly naive, impression of how
the parameters couple!)  This has always been true of scattering methods as
well, although used in combination, as I believe Olwyn Byron is doing with
neutron scattering and sed vel, the answers may be more robust.  But
introduce fexibility and polymorphism (ie. tubulin) and I am not sure how
many methods you need to make the analytical problem solvable.  

	Which brings you to Preston Hensley's famous slide showing the global set
of methods and their appropriate regimes.  I think Preston might say 'Cool!
Go global and ride the bull market!'

	So I agree that hydrodynamic methods have their limitations and problems.
We are constantly testing our answers and assumptions and looking for
better methods.  But there simply are times when it is necesary.  Plus why
not push the technique as hard as possible to see what we can resolve?  


To all of this Peter asked another question which I have no answer to - Do
any of you?  Please let us both know the reference(s)!

	"...it occurred to me that I have not seen any account of studies
utilizing information about stoichiometries, equilibrium constants, species
present etc in equilibria which were gained from sed eq studies being
plugged into velocity boundary analysis to extract the sed coeffs and,
thereby, shape information.  Do you know of any such studies and how
successful this strategy might be?"



-------------------------------------------------|
| Dr. John J. "Jack" Correia			 |
| Department of Biochemistry                     |
| University of Mississippi Medical Center       |
| 2500 North State St.                           |
| Jackson, MS 39126                              |
| (601) 984-1522                                 |
| fax (601) 984-1501                             |
| email: jcorreia@fiona.umsmed.edu               |
| homepage:					 |
| http://fiona.umsmed.edu/~biochem/correia.html  |
-------------------------------------------------|