From: David Scott <djs17@york.ac.uk>
  To  : rasmb <rasmb@alpha.bbri.org>
  Date: Fri, 24 Aug 2001 13:17:25 +0100

Responses to inquiry on membrane proteins...

Hello all

These are the list of answers I received in response to my inquiry about
performing analytical ultracentrifuge experiments on detergent
solubilised membrane proteins.

Thanks to Jo Butler, Holger Strauss, Tom Laue, John Harlan (top marks
for both detail and useful references), and Christine Ebel.

Hope these will be of help to others as well.

regards

Dave Scott.



XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

ORIGINAL EMAIL:

Hello all
I wish to do AUC on a (possibly) tetrameric membrance protein. The catch
is that the detergent has to be at a conc around the CMC so there will
be
lots of micelles there.

Can AUC give any Mwt or shape info. in these circumstances? I seem to
remember previous emails on the subject of density matching, or am I in
a
no-win situation?

Thanks in advance

Dave Scott.

Dr. David J. Scott
York Structural Biology Laboratory
Department of Chemistry
University of York
YORK. North Yorkshire. Y010 5DD
Phone: +44 1904 432590
Fax:   +44 1904 410519
Email: djs17@york.ac.uk
Home Page: www-users.york.ac.uk/~djs17

PLEASE NOTE: FROM 1ST SEPT NEW ADDRESS: DEPT. BIOCHEMISTRY, SCHOOL OF
MEDICAL SCIENCES, UNIVERSITY OF BRISTOL, UNIVERSITY WALK, BRISTOL. BS8
1QD. UNITED KINGDOM.

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

JO BUTLERS RESPONSE:

Dear Dave,

Not a no-win situation at all.  We regularly look at the molecular
weights 
of membrane proteins, to determine the aggregation.  The catch is that
one 
really needs to measure the apparent partial specific volume (or, more 
precisely, the density increment) at "dialysis equilibrium" with the 
buffer.  This can be achieved either by the necessary dialysis or, 
sometimes more simply, by gel permeation chromatography to adjust the 
protein/detergent micelles into the detergent buffer, provided that
these 
micelles can be separated from the detergent alone micelles.
With the Paar densitimeter one only needs 1ml sample, although this
should 
be at a reasonably high concentration (5 - 10 mg/ml for choice).  I gave
a 
talk on using this technique at the UK AUC meeting a few years ago,
which 
was published in the report on the meeting (Butler, P. J. G. (1998). Use
of 
density increment in assessing protein aggregation under unusual 
conditions. Biochem. Soc. Trans. 26, 749-753).

This method has the advantage of not requiring a special detergent or 
detergent mix, but being applicable to the detergent in which your
membrane 
protein is most stable, or is isolated.

Good luck,

Jo

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

HOLGER STRAUSS'S RESPONSE

Hi Dave,

although i didn't do the technique myself, it's well established. You
might consult 

A. Lustig et al, BBA 1464 (2000) 199-206 and 
A. Lustig et al. BBA 1115 (1991) 89-95 

and references therein for a start,
i found
these papers quite nice for a start (as i'm not some witty old user of
the
centrifuge...).

Hope that helps, holger

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - 

Holger Strauss 

Forschungsinstitut fuer Molekulare Pharmakologie (FMP)
Robert-Roessle Strasse 10

13125 Berlin/Germany

Tel: +49 (0)30 94793 - 223 (office)
                     - 316 (lab)

Fax: +49 (0)30 94793 - 169 


XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

TOM LAUE'S RESPONSE:

Hi Dave,
You can certainly do sedimentation with detergent present, though
interpreting the results requires a bit of thought. What detergent are
you
planning on using? Does it absorb at 280 nm? Is it dialyzable (i.e. is
the
cmc high enough to allow dialysis in a reasonable time frame?).
If the detergent you are using allows absorbance measurements, then the
absorbance system is the way to go. Sedimentation equilibrium with
detergent
can be conducted as described by Tanford and Reynolds, or Tanford and
McCaslin, or Edelstein and Schachman. These measurements will provide Mw
and
an estimate of the extent of detergent binding (if the vbar of the
detergent
differs from protein sufficiently). Sedimentation velocity will provide
a
measure of the size of the protein-detergent complex, but not the
protein
alone.
Let me know if I can help.
Best wishes,
Tom

University of New Hampshire
Rudman 379
46 College Rd.
Durham, NH 03824-3544


XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX


JOHN HARLAN'S RESPONSE

Dave-

The presence of micelles in your solution should not give you much
problem in
data collection, at least for absorbance data.  The problem lies in the
amount
of detergent bound in your protein-detergent aggregate.  This amount is
often
substantial (values in the range of 0.3gm/gm to 0.7gm/gm are not
unusual,
Jesper Moller and Marc leMaire have made some study of this).  If you
can get
at this number  and you know the vbar of the detergent, you can
calculate an
apparent vbar for the complex.  You can then use normal buffers for your
sedimentation experiments to get molecular weight of the complex, factor
in
the amount of detergent bound, and then determine the aggregation state
of
your protein.  One thing to keep in mind, however, this approach assumes
that
the ration of detergent to protein (gm detergent / gm protein) remains
constant as the aggregation state of the protein changes.  If you have
multiple aggregation states present and this assumption is not true,
your
estimate of the vbar of the complex may not be accurate.

An alternative approach is to use heavy water to increase the density of
your
buffer to match that of the detergent used (the density matching you
mentioned
below).  Relevant reference include: Tanford and Reynolds, (1976),
Biochim.
Biophys.  Acta, 457, 133-170; Flemming et al., (1997), J. Mol. Biol.,
272,
266-275; and Kochendoefer et al., (1999), Biochem., 38, 11905-11913. 
Karen
Flemming has written a Technical Application Bulletin on this subject
for
Beckman (available on their web-page, look for note #A-1855A).   When
matched,
the detergent becomes neutrally buoyant and is no longer a factor in the
buoyant molecular weight.  The molecular weight of the protein aggregate
(oligomer) can then be calculated directly.  The density of this buffer
can
also help you estimate the vbar of the detergent as well.

Bear in mind it is not unusual for the aggregation state of your
membrane
protein to change, depending on the detergent conditions you use.  Also,
what
you get in detergent solution may not be the active form in the
membrane.  The
trick is to find the conditions that give you the relevant oligomeric
state.

Good luck.

John Harlan



XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

CHRISTINE EBEL'S RESPONSE:

I do not know if you had a response for Karen Flemming. In case, she 
used -see JMB, (1997) 272, 266-275- C8E5 as a detergent, which has 
nearly the same density as the buffer and thus is not visible in 
equilibrium sedimentation. Best whishes
Christine EBEL


XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX