Carryover of protein sample on reverse phase

[MGagne]

I'm doing quantification of a purified protein on Agilent POROSHELL 300SB-C3 with a gradient from 20% to 95% acetonitrile with 0,1% TFA. Protein is eluted at the end of the gradient. The problem is that we see significant carryover that lasts for at least 2 blank injections.
We tried isopropanol instead of acetonitrile, we made runs without injection to make sure that it's not an injector problem and we tried a Luna cyano column from Phenomenex to decrease binding strength. The carryover problem is still there...
I'm about to try decreasing TFA concentration in order to decrease binding strength... I dont know if it may impact.
so, do anybody has experience with sticky protein?

Thank you
Martin

[Mark Tracy]

I once got ovalbumin stuck on a column, and it was never the same after that. Some proteins have an extremely narrow band between enough MeCN to elute it and enough to precipitate it. Some protein then gets denatured and precipitated. You just have to run gradients until it is all gone. The perfect isocratic concentration is simply too hard to hit.

[james little]

I saw a talk at ASMS 2 years ago about using monolithic columns for peptide separation. They said they had a lot of luck and a lot less problems with carryover.

Found some references by searching using google and monolithic columns for protein separation as a topic.

I don't do peptides myself, so no personal experience, just impressed by the talk..

[tom jupille]

I saw a talk at ASMS 2 years ago about using monolithic columns for peptide separation. They said they had a lot of luck and a lot less problems with carryover.

That makes sense because monoliths generally have a significantly lower phase ratio compared with packed columns, and are thus less retentive. I would imagine that gives a somewhat bigger ACN concentration range between desorption and precipitation.

[Uwe Neue]

Ovalbumin is known to create carry over. This has been attributed to folding/unflodling issues. It changes with the details of the chromatographic conditions and the protein.

Contrary to proteins, peptides do not suffer from such issues. If there is carry-over with peptides, it is essentially the same problem as any other sample.

[HW Mueller]

There is quite a conceptual problem with carryover that does not stem from a mechanically moving part like a Rheodyne or syringe. If a protein sticks partially to the frit or column then I can see that one can get a peak from that if one jolts some (or all) of it loose via sudden solvent changes of a gradient or possibly an injection. Gradual deposition and redissolving will cause only a rising or lowering baseline or even a steady state of a raised baseline. Am I missing something?

The best thing to do is to find media in which the protein is quite "happy" (well solvated). Unfortunatly, for those who have to deal with analyses of serum/plasma this is often not possible as there is always a protein which is unhappy with the given situation. On the other hand, if one has a huge problems with a highly purified protein it may well be that it is not as pure as thought.

[Mark Tracy]

Voice of experience-- ovalbumin behaves the way we have described it. Find a theory that accounts for the observation, don't deny the observation from theory

[HW Mueller]

Mark, if you are referring to me, I didn´t deny anything, I just stated that there is a conceptual problem, unless the points I mentioned are responsible (or maybe some that I missed). Now it could be helpful to know the exact conditions under which ovalbumin caused carryover. Normally one can isolate the area which causes it.
How could anybody dare to advance a theory of how your ovalbumin carried over with no information other than that it did this?
Are you saying that nobody in this world ever chromatographed ovalbumin without carryover?

[Uwe Neue]

Hans: something to read: Benedek, Dong, Karger, J. Chrom. 317 (1984), 227

[ym3142]

any one can summary what is in

Benedek, Dong, Karger, J. Chrom. 317 (1984), 227

it is not easy for me to access this journal.

[anupama]

Try to inject blank(solvent) with gradient or you can used some washing method with gradient to avoid carryover issue.

[ym3142]

anupama, what is

inject blank(solvent) with gradient

? and

some washing method with gradient

?

[HW Mueller]

OK, got hold of that article, Benedek, et al, J Chrom 317...... Havn´t had time to read it thouroughly, but it´s about getting two peaks with some proteins under some conditions (RP, etc.): One peak being that of the normal protein, the other, later peak, that of a denatured version. Now in the language I used above, these were very unhappy proteins and the article is also about how to prevent this (as mentioned: keep them well solvated = happy, modify the hydrophobic surface, etc.). These were the days long before I started working with proteins, sort of "teething" problems. Nevertheless, this may be the mechanism I missed in my portrayal of carryover. To get it you have to use a mobile phase in a gradient such that some of the protein is moved out in peak shape, but some of it precipitates on the stationary phase. Now as the gradient changes, there has to come a mobile phase composition which immediatly dissolves this precipitated protein so as to move it out in a peak. Any slow (in relation to the chromatography) redissolving mechanisms would create a smear. Again, clearly: To prevent this one has to choose conditions, denatured or not, which does not split the protein´s characteristics. Any account on handling proteins helps on this.
(I thought of carryover as something that comes out in a second [blank] injection, though).