[RASMB] Equilibrium Data Analysis with WinNonlin

[John Correia]

If you get resolution of the monomer & the dimer then that's the answer
- they are not in equilibrium - see example in ref below with a smad4
construct - if the boundary is not conc dependent then its not an
interacting system.  TCEP worked for our construct and converted it all
to pure monomer.

J.J. CORREIA, B.M. Chacko, S.S. Lam and K. Lin. (2001)  "Sedimentation
Studies Reveal a Direct Role of Phosphorylation in Smad3:Smad4 Homo- and
Hetero-Trimerization." Biochemistry, 40, 1473-1482.

In an equil experiment every load with give a different Kd with the Kd
decreasing with increasing loading concentration and thus global fits
will all fail - try separate K's as an option to see this.   This is
classic nonsuperposition of the MW vs Conc.  But if as you say single
channel fits to a 1-2 model don't work then maybe its more complicated
than nonreversible 1-2.  5s dimer is small if S1 = 4 s - dimer should
approach 6.35 s?




-------------------------------------------------------------------
 Dr. John J. "Jack" Correia
 Department of Biochemistry
 University of Mississippi Medical Center
 2500 North State Street
 Jackson, MS  39216
 (601) 984-1522                                 
 fax (601) 984-1501                             
 email address: jcorreia at biochem.umsmed.edu     
 homepage location: http://biochemistry.umc.edu/correia.html  
 dept homepage location:    http://biochemistry.umc.edu
-------------------------------------------------------------------
 
 


>>> "Blasie, Cheryl [CNTUS]" <CBlasie at CNTUS.JNJ.COM> 10/14/02 01:28PM
>>>
Good afternoon, everyone.
 
        I have a system that appears to have two species in it when
the
velocity data is modeled with SedFit, one at 4.0 S (monomer) and the
other
at 5.5 S (dimer).  Velocity data collected at different concentrations
do
not seem to indicate a reaction boundary.  Based on the peak heights
and
areas from the c(S) distribution, the dimer is 14 times higher than
the
monomer.
        Recently, I have been trying to fit equilibrium data with
Nonlin to
confirm the molecular weight of the two species and determine the
equilibrium constant.  I have a large range of concentrations (1.0
mg/ml to
0.015 mg/ml) and many rotor speeds (4K to 40K).  I have tried
different
scans in the fitting routine, but I can't seem to fit the data without
nonrandom residuals.  Some of the residuals indicate nonideality, some
indicate reversible association with n=2, and some indicate
aggregation.
These residuals are all when trying to fit to a 1-2 system.  I will add
that
even fitting single data sets (3 concentrations, 1 speed) show similar
residual patterns as for the global fitting when fitting to a monomer
system; single sets do not fit to a 1-2 system.  
        Some of my parameters:  sigma is calculated and held constant.
DeltaY is set to 0, then allowed to float after first "fit", LnA starts
off
at 3, and I set LnK2 to -1, N2 floats. 
        I realize I may have left out pertinent info, so please let me
know
if more is needed.  I am wondering if I am missing something with the
program or the concentrations or the speeds... Any advice would be
appreciated, as I have struggled with this problem way too long.  Thank
you
very much.
 
Sincerely,
 
Cheryl
 
 
******************************
Cheryl A. Blasie, Ph.D.
Centocor
200 Great Valley Parkway
Malvern, PA 19355
 
Mail Stop M-1-6
Phone: 610-889-4447
FAX:  610-889-4771
******************************