Proteins and liquid chromatography

Philip J. Wyatt phil at
Fri Jan 6 13:49:41 EST 1995

In article <3ejjqo$75s at> elarson at  
(larson eric) writes:
> phil at (Philip J. Wyatt) writes:
> >We are a small company that manufactures multi-angle laser light  
> >scattering detectors (MALLS) used in combination with  
> >separated samples to determine their absolute molecular weights and  
> >and the distributions of these quantities.  
> >	1)  What do you think of the chromatographic separation process as  
> >a tool for characterizing protein samples (assuming that an absolute  
> >determination may result)? 
> >===============
> The only technique that seems to yield definitive results is  
> gel filtration.  By definitive I mean results you can talk about results  
> reasonably solid evidence that there is an aggregation difference.  Even  
> the absence of exact stoichiometries (i.e. eluted mol. wt./subunit mol.
> wt.) stating that there is a difference is reasonable.
> I have one protein under study that seems to elute from ion exchange  
> as a split peak.  Subject the eluted samples to a molecular weight  
> gel and both elute at the same point.  Re-elute these samples from the  
> wt. sizing gel on ion exchange and both now elute at the same salt
> concentration.  I believe the separation afforded by the first ion  
> column is real, but cannot definitively determine "why" the protein  
> on the column.  I suspect an aggregation phenomenon -- something an  
> mol. wt. detector could determine (we looked at these several years ago  
> they couldn't correctly assess the  mol. wt. of a test protein we had --
> might be time to see if the years have been good to the technique --  
> send me some literature. :-)
> >	2)  If you use such techniques, what type do you use?  (E. g.  
> >reverse phase, size exclusion, ion exchange, capillary electrophoresis)
> Of your list, size exclusion and ion exchange are the most popular (at
> least with our lab.)  Reverse phase and capillary electrophoresis
> techniques almost always use conditions that are far too harsh for most
> proteins.  It would be foolish to try and assess aggregation state of a
> protein after it had been subjected to trifluoroacetic acid, phosphoric
> acid, organic solvents, etc.  Reverse phase and capillary seem suited to
> purifying a polypeptide chain (as opposed to purifying a native  
> >	3)  Do you think the buffer and/or the mobile phase can (will)  
> >affect  the composition of your sample?  (E. g. might you detect  
> >aggregates that were not present in your sample?  Could some of the  
> >stick to the columns during the separation process itself?)
> I would say yes.  With two proteins we currently work with, aggregation,
> whether part of the native "operational mode" of the protein, or caused  
> buffer conditions is observed (one proteins aggregation state depends on 
> the presence or absence of one substrate.)
> >	4)  If a column structure could be developed  that would permit  
> >the use of any type of mobile phase without affecting the separation  
> >process itself (i. e. no sticking and no false aggregation), would that  
> >make such chromatographic techniques more attractive?  (We do NOT  
> >manufacture columns!)
> This would be a neat trick if you could do it.  Your description pretty
> much describes the goal of a mol. wt. sizing column.  A priori, this  
> not seem to apply to the ion exchange, reverse phase, affinity
> chromatography, capillary, etc..
> It sounds like your company is focussing on devices for on-line
> identification of mol.wt. from chromatographic resin (be they ion  
> mol. wt. sizing, or whatever.)  This is useful.  However, many of the
> aforementioned chromatographic techniques are being used in the hopes  
> differential aggregation (in our case it would protein-protein  
> between a regulatory protein and its substrate) can be analyzed.  Of  
> as much utility would be direct measurement of the mol. wt. distribution
> within a single (stirred?) vessel versus time (i.e. add substrate  
> check mol. wt., add regulatory protein, check mol. wt.)  Evidence of
> changes, even if not definitive in terms of mol. wt., would be very  
> -- 
> Eric Larson                  | University of Illinois at  
> USDA/Agronomy                | 190 PABL; 1201 W. Gregory; Urbana, IL  
> elarson at     | Voice 217.244.3079  Fax 217.244.4419
> Fidonet: 1:233/4.1           | My opinions are my own, but correct :-) 

Let me comment on your insightful remarks referring to my questions 1  
through 4.

1) Regarding the split peak you observe on elution from ion exchange,  
putting in line a light scattering detector like our miniDAWN would show  
immediately the molecular weights and (if the protein is large enough) the  
rms radius of each slice and, of course, each peak without any need for  
calibration.  (Dr. Charles Zukoski of the Dept. of Chemical Engineering  
has a large 18 detector DAWN system that will work equally well.  You  
might want to give him a call and discuss this with him at 333 7379.)   
We've seen similar differences between GPC and reverse phase separations,  
yet whenever we make the light scattering measurements, the results are  
clarified.  Nevertheless, we may see the SAME sample separating  
differently with different injections on the SAME GPC column.  We often  
attribute this to sample attaching to the column and eluting with a later  
sample, often with some weird aggregation (based on LS MW determination)  

2) There is another method that appears quite promising that we shall be  
working with during this next few months.  At present it is not user  
friendly, was developed by another company, and adds another $20,000 to an  
already expensive LS detector, but we hope to have better news to report  
soon.  We'll keep you posted.

3) If the separation technique mentioned briefly in 2), above, were to  
obviate these concerns, would anyone want to spend this additional amount  
of money to achieve it?

4) See above regarding the separation techniques.  As for the ablity to  
monitor MW and size with time, a brute force demonstration of this (using  
LS) was presented by Wilson and Benight in J. Bio. Chem. vol 265, p 7351  
(1990).  With our newer instruments, this type of measurement is very  
easily done.

Thanks so much for your comments.  If you need any more information from  
us, please don't hesitate to ask.  We'll keep you informed as to the new  
separation experiments.

Phil Wyatt

More information about the Proteins mailing list