[RASMB] concentration range for non-ideal behavior

[John Sumida]

Dear RASMB,

 

David and all who have responded, thank you for your comments. I think in the aggregate of comments received I have my answer and more questions.

 

In the range of concentrations measured 0.2 – 4.9mgs/ml the slope of the experimental s with concentration is negative but only slightly so (slope=-0.033 s/c, where c is the mass concentration g/cc).  The extrapolated s at 0mgs/ml is 6.88, given the information received in these posts I think that at 4.2 mgs/ml I ought to expect a ~4.4% decrease in s due solely to excluded volume effects.  If so the expected s-exp at 4.9mgs/ml ought to be 6.58. What is actually observed is a value of 6.72s.  So I think what David and others have suggested is probably accurate, namely that self-association is offsetting the non-ideality here.  

 

This raises some questions perhaps that would be good to test.  I can’t shut off self-association but perhaps one could slow it down by reducing the temperature from 20C to 4C? I think excluded volume is only weakly dependent on temperature and self-association more strongly dependent on temperature?  

 

Interestingly, the sw vs conc relationship changes as a  function of buffer, whereas the slope of relationship is only weakly negative in HEPES (0.01M HEPES, 150mM NaCl, pH7.5) in PBS (0.137M NaCl) it is strongly positive.

 

Once again thank you all for your thoughts and suggestions.

 

Best regards,

John Sumida

University of Washington

 

From: RASMB [mailto:rasmb-bounces at list.rasmb.org] On Behalf Of David Hayes
Sent: Wednesday, January 21, 2015 6:25 PM
To: RASMB
Subject: Re: [RASMB] concentration range for non-ideal behavior

 

Hi John,

 

The opposite of non-ideality is reality.  All real molecules take up space, so all real molecules sediment with some excluded volume non-ideality.  Charge is another source of non-ideality:  the ideal model of a molecule, besides being a geometric point with no volume, also has no attractive or repulsive forces.

The type of non-ideality that is not avoidable is excluded volume.  So it is not exactly mg/ml that determines whether or not the non-ideality is significant, but the excluded volume.  For most proteins it is reasonable to see little effect from concentration alone up to 1-2 mg/ml.  But, put a molecule in a salt free buffer where charge is not screened (Actin) or take a molecule that has a really big carbohydrate attached (Pegasys) and nonideality can be evident at 0.5 mg/ml and below.

 

In sedimentation velocity, the other non-ideality is the back flow.  When a real molecule moves down the cell, real buffer has to move up the cell to fill in the otherwise to be empty spot where the molecule was.  This hydrodynamic non-ideality is also always there, but usually negligible below 1-2 mg/ml.

 

Did you do a concentration series of your antibody?  If so, it is expected that around 5 mg/ml; you would have measurable concentration dependence of the sedimentation coefficient.  If your peak moves the same S at 0.5 and 5 mg/ml I would suspect self association is adding S and offsetting the concentration dependence of S.  Static light scattering is helpful to use as an orthogonal method when self association and non-ideality balance out in AUC-SV.

 

David

 


  

  _____  

From: John Sumida <jpsumida at u.washington.edu>
To: rasmb at rasmb.org 
Sent: Tuesday, January 20, 2015 9:00 PM
Subject: [RASMB] concentration range for non-ideal behavior

 

Dear RASMB,

 

My recollection is that concentrations in excess of 1.0 mg/ml should be avoided due to the possibility of non-ideal interactions that would complicate the analysis of SV and SE measurements.  However in two recent cases (in one case a mAb was being studied, and in another a globular viral capsid protein), we exceeded this value by a factor of five and failed to observe any evidence of non-ideal behavior. In both cases, it was possible to fit both SV and SE data without assuming any non-ideal interactions when the experiment was performed in PBS or HEPES.

 

I would be curious to hear if it is the experience of the community that non-ideality is commonly to be expected at concentrations >1.0mg/ml or whether the concentration cutoff for non-ideality is less of an actual cutoff being more a function of the particular properties of the molecule being studied.

 

Best regards

John Sumida

University of Washington


_______________________________________________
RASMB mailing list
RASMB at list.rasmb.org
http://list.rasmb.org/listinfo.cgi/rasmb-rasmb.org



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.rasmb.org/pipermail/rasmb-rasmb.org/attachments/20150122/0c28d7e2/attachment.htm>