problem with ion exchange column

[tk7434]

I've been having issues with my anion exchange chromatography column on an FPLC that I was hoping someone would be able to help with. The column is a GE ResourceQ 1 mL. I'm using the column as a first step in purification of an enzyme, I've previously done the chromatography (two months ago) with very good results. I was just barely saturating the column with enough input material to generate a main peak of around 1400 mAU (so binding wasn't an issue originally). Since then, I noticed the capacity had decreased, and I was losing a lot of protein in flowthrough and only retaining enough to generate a peak of around 500 mAU. The buffers are the same formula but different batches, but I remade them and checked pH and tried again, still 500 mAU. So then I cleaned the column with GE's recommend procedure (2 M NaCl, 1 M NaOH, 2 M NaCl, H2O, buffer). Now, when I run just filtered deionized water through it the conductivity is around 0.18 mS/cm (it used to be around 0) and randomly will rise as a gaussian peak, sometimes up to 2 mS/cm. Does anyone here have any idea as to what could be happening, and how to fix it?

[John Cuthber]

OK, my experience is with HPLC of small organic molecules rather than protein purification but I think the idea is the same.

 

If you put clean solvent through the column and you get peaks coming off it then the column isn't clean.

I'd try the manufacturer's procedure for washing again and see if the peaks that come off afterwards are smaller. If so then the cleaning is working and you need to do it more. If not then you need a different clean up routine (or a new column).

 

However , there's still the question of where the stuff on the column came from. It might be from the solvent you used to wash the column so I'd make up fresh batches of everything first.

[CharonY]

Yeah, sounds like there is still stuff on it. Is the backpressure ok? Otherwise I would rinse it with several more column volumes of water and check whether baseline (I assume UV in your case?) is stable (as well as pH). If the contamination are precipitated proteins (as e.g. indicated by random UV signal) or lipids, additional wash steps might be necessary. This may include protease treatments or, in case of lipids and hydrophobic proteins, washes with compatible mild organic solvents (often ethanol/isopropanol).

[tk7434]

Thanks for your responses. I've since re-examined chromatograms and noticed that for some reason, newer batches of buffer had higher baseline conductivity despite following the same recipe (hence decrease in binding capacity). I'm currently troubleshooting this; I think the increase in the ionic strength was the result of re-using old filters which had embedded salt crystals. And for clarification, the random peaks I was referring to were, in general, not in the UV absorption channel but in conductivity measurement. After running the column through with water excessively (>120 CV) they seemed to have stopped appearing.

[CharonY]

Ah, in that case it really looks like salt contamination of sorts. If the buffer already has different conductivity it is a source of concern.