[RASMB] Interpreting SV and SE data

[John Philo]

Eric, I agree with Walt Stafford's points with regard to what would be
necessary to fully characterize the association scheme and thermodynamics.
But it sounds to me like all you really want to do is to show that the
trimer you see in the crystal is at least a major species present in
solution. 

It sounds like in velocity you have a range of concentrations over which a
c(s) analysis shows a peak at a constant position of 2.6 S, but the peak
area goes up and down with concentration. That constant S value suggests
this peak is predominantly a single species and might be your putative
trimer. Given that you have a crystal structure you should be able to
calculate the expected sedimentation coefficient of that trimer and see if
that matches the 2.6 S peak. That is, even for an interacting boundary we
sometimes can have conditions where the species are pure enough that the
sedimentation coefficients are meaningful and can potentially be used to
identify the state of oligomerization. But as Walt said, trying to identify
stoichiometry based on a c(M) distribution is highly dangerous.

I am not clear though what you mean by this "aggregate" forming at high
concentrations. If this is a reversible species then it may be a normal part
of the assembly pathway under non-crystallizing solution conditions. At a
minimum it would have to be considered in analyzing sedimentation
equilibrium experiments at higher concentrations, but if this higher
oligomer forms reversibly in substantial amounts then it may not be clear
whether the trimer is the relevant species under physiological conditions.

John


-----Original Message-----
From: rasmb-bounces at rasmb.bbri.org [mailto:rasmb-bounces at rasmb.bbri.org] On
Behalf Of Walter Stafford
Sent: Thursday, May 29, 2008 11:43 AM
To: Eric Salgado
Cc: rasmb at server1.bbri.org
Subject: Re: [RASMB] Interpreting SV and SE data


Hi Eric,

    I don't think c(MW) is an appropriate way to analyze an interacting
system. In the case of a self-associating system the boundary is called a
reaction boundary and is not composed in individual peaks that can be
resolved by any method.  Molecular weight values obtained from c(MW) would
not have any meaning in that case. So I would be very suspicious of a trimer
molecular weight value obtained that way.

 The boundary must be treated using one of the methods for whole boundary
fitting to solutions of the Lamm equation for an interacting system. And the
fit must be done globally over the widest range of concentrations possible
to get reliable estimates of the stoichiometry and equilibrium constants.
You could use either SEDPHAT or SEDANAL for global fitting to the various
models you want to try.  SEDANAL has a Model Editor that allows you to
specify a very wide variety of arbitrary association schemes.

SE is a good way to verify the SV results but in your case it sounds like
you need to go to higher concentrations to populate the larger species in
order to resolve the ambiguity between monomer-dimer and monomer
-dimer-trimer. It is well known also that a monomer-dimer-tetramer system
can masquerade as a monomer-trimer system if the concentration range is not
high enough. Solubility issues may prevent you from increasing the
concentration enough, though.


Walter Stafford

Eric Salgado wrote:
> Hello all,
>   I have a system for which I've done SV experiments at multiple 
> concentrations that, on  increasing the concentration, reveals a peak 
> that starts at an s value of about 2, shifts to a plateau between 2.4 
> and 2.6, then steadily increases at 2.6. I should point out that all 
> of these peaks are very broad.
>   The c(MW) analysis describes a monomer at low concentrations; dimer 
> at intermediate concentrations;  something between  dimer and trimer 
> at high concentrations;  and finally, as  an aggregate  starts to form 
> at very high concentrations, a trimer peak is resolved. This trimer 
> peak correlates to our crystal structure, but it appears as if, 
> because of the apparently low association constant and fast kinetics 
> involved with the trimer formation, this fact is masked in the SV data.
>   In order to hopefully rid of this ambiguity, I performed SE 
> experiments in the concentration regime that did not form an aggregate 
> (and therefore apparently only a small amount of trimer), but find 
> that, as far as a species analysis goes, both a model with 
> monomer-dimer or monomer-dimer-trimer present are statistically equal.
> That being said, the statistics are lower for the later model, while 
> the fit and residuals for both are equally good.
>   With all of this in hand, how can I determine the best SE model to 
> use, and can I say with certainty that the trimer exists in solution 
> from these data?
>
> I know this is a lot, but I would appreciate any help I can get  with 
> this problem.
>
> Thank you in advance,
>
> Eric Salgado
> _______________________________________________
> RASMB mailing list
> RASMB at rasmb.bbri.org
> http://rasmb.bbri.org/mailman/listinfo/rasmb
>

--
Walter F. Stafford, Ph.D.
Senior Scientist
Analytical Ultracentrifugation Research Laboratory Boston Biomedical
Research Institute
64 Grove Street
Watertown, MA 02472-2829
-------------------------
mailto:stafford at bbri.org
direct dial:   617-658-7808
receptionist:  617-658-7700
-----
Boston Biomedical Research Institute... 
Today's Research for Tomorrow's Health.	
      Please visit us at http://www.bbri.org/



_______________________________________________
RASMB mailing list
RASMB at rasmb.bbri.org
http://rasmb.bbri.org/mailman/listinfo/rasmb