[RASMB] Why is my interference data different from the absorbance data

Mon Sep 7 00:13:22 PDT 2009

Dear Paul
Just to reinforce what Arthur has said.  Although the raw data from interference optics is intrinsically more precise, it is frustratingly more difficult to interpret compared with absorption optics as the profiles you observe represent concentration at a particular radial position relative to another radial position.  So to get absolute concentration (expressed in interference fringe units) you need some method of evaluating the meniscus concentration ("Ja" or "Ca") or other reference position concentration to transform your relative concentration set.  Flotation of this as an offset doesnt always work I'm afraid. There is a strong literature from the 60's and 70's devoted to this which has largely been forgotten, notably articles by Creeth and Pain in Progress in Biophysics and Molecular Biology  (1967) and Teller (1973) in Methods in Enzymology, and describe methods of obtaining Ja or Ca which avoid flotation ..... although unfortunately it is still difficult to obtain. We also did some work on this in the 80's, although with the focus on highly heterogeneous systems such as mucins.  Errors in this will lead to a systematic error in the results, although this can be minimized by having a decent fringe increment between meniscus and base, at least 2-3 fringes   My guess is the flotation procedure you are using is underestimating the reference concentration, even with concentration of mass constraints.  Although the raw data is less precise (absorbance of non-sedimenting species still has to be corrected for) absorption optics gives a direct measure of concentration (with the Lambert-Beer limit) - for the case you describe I would trust your absorbance results more. 
Best wishes
Steve Harding

Prof. Stephen Harding 
National Centre for Macromolecular Hydrodynamics
University of Nottingham 
Sutton Bonington 
LE12 5RD, UK 
http://www.nottingham.ac.uk/ncmh
<http://www.nottingham.ac.uk/-sczsteve>  
Tel: +44(0) 115 951 6148
Mob: +44(0) 78110 90635
steve.harding at nottingham.ac.uk <mailto:steve.harding at nottingham.ac.uk>  

________________________________

From: rasmb-bounces at rasmb.bbri.org on behalf of Paul Leonard
Sent: Mon 07/09/2009 04:01
To: RASMB Mailing List
Subject: Re: [RASMB] Why is my interference data different from the absorbance data

Dear Professor Rowe

Thank-you for your prompt and thorough response.  I really appreciate your
suggestions.  I'll set up multiple loading concentrations and try again.
Whilst this will help me determine whether my protein is in equilibrium
between monomer or dimer or not I still have concern that the curvature
observed with the absorbance optics should match the interference optics.
You are looking at the exact same protein distribution after all.

Should I have used exactly the same volume in the buffer reference channel
as I used in the protein channel? Would having 10 microlitres more
solution in the buffer reference channel mean that the buffer subtraction
is wrong and perhaps interference is more sensitive to buffer component
concentration gradients than absorbance measurements at 250nm?

The experiment setup was as follows.

40 microM protein dissolved in 20 mM Tris pH 7.0, 100 mM NaCl, 5 mM MgCl2. or
40 microM protein dissolved in 20 mM Tris pH 7.0, 100 mM NaCl, 5 mM MgCl2,
5.3 mM BeCl2, 33 mM NaF (the major beryllium and fluoride species present
at these concentrations is BeF3- although there will be some BeF2, BeF42-
and other species present).

The original buffer used in the preparative size exclusion chromatography
column was 1.08x buffer so that water (71.3 microlitres per mL) or
BeCl2/NaF solutions could be added to the protein solution or buffer blank
solution prior to the experiment to give the concentrations described
above.

I used a 2 channel Epon centrepiece containing 170 uL of protein solution
or 180 uL of buffer.
The cells were 'aged' by multiple cycles of 42k and 3k running with 200
microlitres of water inside.  Then the water blanks were recorded for each
rotor speed.  The water was removed and buffer used to wash out the cells
3 times.  The buffer was removed and then the sample loaded as described
above.

I ran the centrifuge for 48 hours at each rotor speed collecting a scan
every 4 hours.  Rotor speeds were 12k, 20k and 26k r.p.m.  Temperature was
20oC.
The AUC was set to record absorbance scans at 250nm and 280nm as well as
interference scans.

I checked that the last three scans for each speed overlaid perfectly so i
am confident the sample reached equilibrium at each rotor speed.

SDS PAGE analysis did not reveal any degradation of the protein placed
inside the centrifuge.  There were no contaminating bands on the coomassie
stained gel.

For data analysis

I used sedfit initially to subtract the water blank scans and then
transferred the curves into Sedphat so that I could load the three
different rotor speed curves into a single window and use a global fit for
all three curves. I tried fitting the data both with and without the mass
conservation restriction applied.  The absorbance scans (250nm scans were
used as 280 absorbance exceeded 1.5 abs units).  The partial specific
volume of the protein and solvent density were calculated using SEDNTERP.

Thanks again for your suggestions.  I really appreciate you sharing your
expertise with a novice like myself.

kind regards

Paul

> Hi Paul
>
> What software are you using, I wonder? And how are you doing the
experiments? It is a curious fact that getting molecular weight values
by SE using interference optics is not actually the easiest of things to
do. You have to float the baseline, which is a parameter rather much
correlated with the sigma parameter!
>
> On the basis of the old curve-fitter's saying - "fix what you know,
float what you don't" - I would normally myself treat sigma as
> something given to me by SEDNTERP and therefore fixed, leaving the
interaction parameter to be floated. A single SE run will not
> distinguish between dimer being present in equilibrium and dimer being
present as a mixture, for that you need several runs with c varied.
Analysed globally or not, this will sort it out.
>
> Regards to you and all
>
> Arthur
>
>
> On Sep 6, 2009, at 17:52, Paul Leonard wrote:
>
>> Dear all,
>> I have been running both sedimentation velocity and sedimentation
equilibrium AUC experiments on a relatively simple protein system. The
sedimentation velocity experiments at multiple protein
>> concentrations
>> show that the protein is just a monomer unless BeF3- is present (to mimic
>> phosphorylation) and then all of the protein becomes dimeric.
>> However if I run sedimentation equilibrium experiments at multiple speeds
>> I find that if I use the absorbance optics it is in agreement with the
sedimentation velocity result - protein is just a monomer in the
absence
>> of BeF3- but fits as a dimer if BeF3- is present.
>> However if I look at the interference scans for the same cells (for
sedimentation equilibrium AUC) the protein looks considerably bigger -
i.e. it would appear that it is in equilibrium between monomer and
dimer
>> in the absence of BeF3-
>> I have also record analytical size exclusion chromatography profiles for
>> concentrations of the protein all the way from 10uM upto 500 uM and if
BeF3- is absent I cannot see any evidence of dimer formation.  If I add
BeF3- the protein is all dimer.
>> Therefore I am pretty certain that I should only be seeing monomer in the
>> absence of BeF3- and dimer if BeF3- is present so I'd like to know why my
>> interference data makes the protein look larger than it should.  Is there
>> some correction that I need to apply.  I have recorded water blanks at
each speed and it makes virtually no difference if I subtract these
scans
>> (Not enough of a correction to account for the additional curvature in the
>> scans).
>> Could it be that the radial calibration is wrong?  Is something wrong with
>> the interference optics?  Any explanation would be greatly appreciated.
I fear that if we cannot get reliable results for this simple case
using
>> this equipment for more complex systems will become impossible (at least
>> for sedimentation equilibrium experiments).
>> Thank-you for any help you can provide.
>> Paul
>> _______________________________________________
>> RASMB mailing list
>> RASMB at rasmb.bbri.org
>> http://rasmb.bbri.org/mailman/listinfo/rasmb
> ************************************************************************
*******
> Arthur J Rowe
> Professor of Biomolecular Technology / Director NCMH Business Centre
School of Biosciences
> University of Nottingham
> Sutton Bonington
> Leics LE12 5RD
>
> TEL:  0115 9516156
> FAX:  0115 0516157
> ************************************************************************
*******
>
>

Department of Biochemistry
UMDNJ-Robert Wood Johnson Medical School
Center for Advanced Biotechnology and Medicine
679 Hoes Lane
Piscataway, New Jersey 08854-5627

Phone: (732) 235-4206
Email: leonard at cabm.rutgers.edu

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

This message has been checked for viruses but the contents of an attachment
may still contain software viruses, which could damage your computer system:
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.rasmb.org/pipermail/rasmb-rasmb.org/attachments/20090907/8713bda4/attachment.htm>