From: Rose Disher <disherr@centocor.com>
  To  : rasmb@bbri.harvard.edu, Hilal@scripps.edu
  Date: Thu, 09 Jul 1998 09:36:41 -0400

Re: help with s values of membrane proteins in detergents

It isn't particularly surprising that you get different s values in
detergent for the native protein.  It may be that the detergent(s) you are
using are not optimal for completely denaturing the transmembrane domain
in question.  While CHAPS is attractive because of its chemical properties
and ease of removal, it often isn't the most effective detergent with
regards to the complete denaturation of integral membrane proteins.  You
may need to use something harsh, say SDS, to get the s value for the
completely denatured protein.  This won't be particularly physiologically
relevant, but it may help you determine why you are getting a
heterogeneous range of s values.  An oldie but goodie reference for the interaction of
detergents and membrane components with which you may or may not be
familiar is a book called Biological Membranes by J. B. C. Findlay and W.
H. Evans, published by IRL Press in 1987 as a part of their "A Practical
Approach" series.  Unfortunately, they don't cover analytical
ultracentrifugation.

By the way, you don't mention whether or not the deleted transmembrane
region is covalently modified.  If it is, heterogenetity in the
posttranslational modifications could be the culprit in your results.  In that case,
the SDS experiment I suggested should give you a heterogeneous range of s
values rather than a single value.

>>> "Hilal A. Lashuel" <Hilal@scripps.edu> 07/08 11:01 PM >>>
We would appreciate if anyone can help us address the situation outlined
below. 

We are currently characterizing an integral membrane protein by analytical
ultracentrifugation as well as other size determination 
methods (eg gel filtration) in detergent solutions (0.5% CHAPS).  We have
observed that deletion of the single predicted 
transmembrane domain (approx. 30 amino acids) drastically affects the
observed s values in sedimentation velocity.  The 
transmembrane domain containing protein exhibits an a heterogeneous range
of s values from 17-25 S, while the 
transmembrane deleted protein behaves as a single 11S species (both
proteins, however require the presence of detergents).  
These observations are confirmed over a 10 fold concentration range by
sedimentation velocity analysis, and a 100 fold 
concentration range on gel filtration (the proteins run as 810 kDa, for
the
transmembrane domain containing protein, and 480 
kDa for the transmembrane deleted protein).  Equilibrium analysis showed
the presence of a mixture of species for both 
proteins (however, the transmembrane deleted protein clearly showed a
single species by sedimentation velocity?).  We have 
observed aggregation of both proteins over time, which may explain the
equilibrium results. 

What are the potential pitfalls of comparing relative sizes of membrane
proteins in detergent solution by sedimentation velocity 
analysis? 
The data showing the difference in the oligomerization of both proteins
are
consistent and very reproducible from different 
patches of proteins. 

We have had trouble finding references for the use of sedimentation
velocity to evaluate s values of membrane proteins in the 
presence of detergents. 

Your help and cooperation is highly appreciated 

Hilal

Hilal Ahmed Al-ashuel
The Scripps Research Institute 
The Skaggs Institute for Chemical Biology 
Mail Drop MB 12, Room MB 28
10550 North Torrey Pines RD
La Jolla ,CA  92037
Tel 619-784-9607
Fax 619-784-9610
Email Hilal@scripps.edu