Strange Column Conditioning on UPLC Protein method

[tristanewalters]

Hello all-

Without getting into a bunch of specifics, I wanted to see if people could come up with a reason(s) that we have this issue with conditioning a column for a UPLC method. The API we are monitoring is a protein. Samples and standards contain a complex matrix of about 8 amino acids and a small amount (0.01%)of polysorbate 20 in water. The samples and standards are diluted in that same matrix. The issue that we are seeing is that we have to do anywhere from 25-100 injections of sample or standard before the peak of interest's peak area finally stops growing. Why do you guys think we are seeing this issue?

Here are some method specifics:
Column- Acquity BEH C18 2.1x50 mm 1.7 mcm Part # 186002350
Mobile Phase A- 0.1% TFA in water
Mobile Phase B- 0.1% TFA in ACN
Injection volume- 10 mcL of solutions containing anywhere from about 15-40 mcg/mL of the protein

What do you guys think?

[HPLCaddict]

It sounds as if there are "hungry" surfaces in your system that needs to get saturated before peak areas are stable. These surfaces can be in the column, but they might also be anywhere else in the UPLC. Stainless steel can be quite hungry...
You might try if fewer high-volume sample injections can speed up the process. There are special-purpose "bio" systems out there, which avoid stainless steel components in the fluid path. Titanium and PEEK are possible alternatives.

[Tomtotom]

Hello all-

Without getting into a bunch of specifics, I wanted to see if people could come up with a reason(s) that we have this issue with conditioning a column for a UPLC method. The API we are monitoring is a protein. Samples and standards contain a complex matrix of about 8 amino acids and a small amount (0.01%)of polysorbate 20 in water. The samples and standards are diluted in that same matrix. The issue that we are seeing is that we have to do anywhere from 25-100 injections of sample or standard before the peak of interest's peak area finally stops growing. Why do you guys think we are seeing this issue?

Here are some method specifics:
Column- Acquity BEH C18 2.1x50 mm 1.7 mcm Part # 186002350
Mobile Phase A- 0.1% TFA in water
Mobile Phase B- 0.1% TFA in ACN
Injection volume- 10 mcL of solutions containing anywhere from about 15-40 mcg/mL of the protein

What do you guys think?

Do you have a gradient method? Which detector do you use? In case of a UV-detector it may also be that if you have a well folded protein that denaturates, formerly not accessible UV-absorbing groups are presented resulting in an increasing peak area. You can test that by denaturating the protein before injecting it. Another thought that came to my mind is that 10µl sample is a huge amount of sample for such a small column. I get problems with 3µl sample on a 150mm column. Well, i use core shell material, so maybe that makes a difference. Nevertheless, if your sample amount is to much you are maybe facing carry over effects or a change of the column chemistry by applying a new "sample-phase" onto your column material.

[Andy Alpert]

Try passivating your metal surfaces by eluting the column overnight, at a low flow rate, with 40 mM EDTA.2Na. In the morning, flush out the EDTA.2Na with water. This treatment does two things:
1) It temporarily eliminates any "hot" sites in the metal columns, flow path, etc. that might absorb your analytes;
2) It renders stainless steel resistant to attack by halide ions for 3-4 weeks (not that that's a concern in your case).

[tristanewalters]

It sounds as if there are "hungry" surfaces in your system that needs to get saturated before peak areas are stable. These surfaces can be in the column, but they might also be anywhere else in the UPLC. Stainless steel can be quite hungry...
You might try if fewer high-volume sample injections can speed up the process. There are special-purpose "bio" systems out there, which avoid stainless steel components in the fluid path. Titanium and PEEK are possible alternatives.

Interesting thought. I may try switching systems to another UPLC after our column conditioning on the initial system to see if we have to do a bit of reconditioning. That could show that it is maybe some stainless steel interactions. I will also check our development reports for any mention of stainless steel sensitivity or problems with cleaning. The column conditioning isn't really a problem because it is only a 4 minute injection, I'm just trying to look for answers.

I have also wondered if this could perhaps be related to our diluent with the amino acids and polysorbate 20. One of the scientists here seemed to think it may have something to do with the polysorbate 20.

Below is a chromatogram of our first five injections of the standard. Maybe this can help you guys visualize it.

[tom jupille]

That looks like almost a "textbook case". Many years ago, we used to make a couple of *big* injections of BSA on any new column in order to tie up protein-eating active sites.

[tristanewalters]

That looks like almost a "textbook case". Many years ago, we used to make a couple of *big* injections of BSA on any new column in order to tie up protein-eating active sites.

Thanks for the input. What do you think that peak that looks like is coming out in the void is? It continues to get smaller and smaller and is more or less gone by injection #25. I initially thought that until the column/system chemistry was "equilibrated" that it might be our protein shooting out, but the peak areas don't really add up.

[DJ]

I've seen this with all kinds of columns for various analysis-- sugars, inorganic ions, and of course, proteins. It is not always consistent-- some columns need equilibration over-night with periodic injections of standards, others work well right out of the box. It's a mystery.

[tom jupille]

Sorry, I lost track of the thread for a while.

I'll echo DJ's comment. "t0 noise" consists of unretained garbage plus slight, pseudo-random fluctuations in mobile phase composition as the system comes back to equilibrium from the shock of the injection. My take on it is that there is so much going on that it's essentially chaotic. Most people handle t0 noise the old fashioned way: by ignoring it.