Metal won't strip from IMAC

From: Katherine Sippel
Date: 13 January 2012 00:01

Hi all,

I've run into a bit of a protein purification conundrum and wondered if anyone had encountered a similar situation. I've exercised all of my google-fu and can't find anything. It's a fairly straightforward setup; His-tagged protein and Talon Co2+ resin, load lysate, wash with 5 mM imidazole, elute with 150 mM imidazole. There is protein in the elution fractions as would be expected. The strangeness occurs when I try to regenerate the column. Using the standard protocol of 25 mM MES, 100 mM NaCl pH 5 doesn't change the color of the resin back to light pink the way it should with a regenerated column. I try stripping with the suggested 0.2M EDTA, still pink, 0.5M EDTA, still pink, 8 M urea plus 4% CHAPS and then EDTA, still pink, 1 M NaOH then EDTA, still pink. I've checked the resin using a Western (with a really specific monoclonal Ab) and it seems that my protein has somehow irreversibly bound to the column and is preventing the metal from releasing the sepharose. I've even tried competing the protein off with excess Co2+ and Mg2+ (the endogenous divalent bound cation).

Clearly the solution is swapping to a Ni column, but this is really bugging me now. Has anyone run into this problem with IMAC before?

Background: The protein does bind divalent cations (Mg and Mn) with low affinity (~1 mM) and has a ridiculous number of cysteines (10 in 416 residues total). There is 1 mM BME and 1 mM MgCl2 in all of the buffers. 

Thanks,

Katherine

----------
From: Michael Thompson

Katherine,

You are not alone. I have inadvertently destroyed a GE HisTrap column with high concentrations of proteins that contain many exposed cysteines. In my case the Co2+ resin turned a very dark purplish-brown and the protein appeared to have crashed out on the column. I didn't try to strip it, because I figured it was done for anyway, so I can't tell you any more about the problem. Here's how I explained it to myself (whether or not this is actually right I'm not 100% sure, but it makes sense in my head). The columns I was using have a maximum concentration of 5mM for DTT and 10mM for B-mercaptoethanol. So that seems like the column can handle 10mM thiol groups. If you have a protein with many cysteines and it is very highly concentrated (as was the case for me) then you are adding considerably more thiol groups to the solution. This abundance of thiols reduces the metal on the column, and disaster ensues. For me, repeating the same prep with less DTT (3mM vs. 5mM) in the buffer fixed the issue. If you are concerned about your protein oxidizing at lower concentrations of DTT or BME, the other alternative is to switch to TCEP. The IMAC columns can tolerate higher concentrations of TCEP, and it is a far superior reducing agent (more stable, more reductive, etc.)...but also a lot more expensive (although you can get away with using much less because it works so much better).

HTH,

Mike

--

Hi,

 

This is a fairly familiar issue. A hydrophobic protein with thiols often contaminates IMAC columns and this kind of contamination can be difficult to remove completely. In the past some of the methods that worked included trypsin or papain treatment, followed by massive amounts of chelating agents. I would like to note, however, that the cost of such 'rejuvenation therapy' easily approaches the cost of 5-ml resin batch. In general with IMAC it is almost always easier to perform the first step in batch, since it takes less time for larger samples. So unless your column is very large and expensive it sounds cheaper to get a new bit of resin...

Good luck!

 

Artem

----------
From: Prof. Dr. Arne Skerra


Hi Katherine,

I recommend Zn-IDA Sepharose (Chelating Sepharose Fast Flow, GE Healthcare), which we have been using successfully for more than 20 years, since the early days of IMAC:

Skerra et al. (1991) The functional expression of antibody Fv fragments in Escherichia coli: improved vectors and a generally applicable purification technique. (Nature) Biotechnology 9, 273-8.

This metal/chelate combination has exquisite selectivity for the His6-tag, at least if operated with an imidazole concentration gradient. Importantly, Zn(II) typically forms reversible sulfide complexes and it is not redox-active (in contrast with Co, Cu, Ni)!

In deviation of our old protocol I would just recommend to use a concentrated ZnSO4 stock solution (instead of ZnCl2), which is less prone to hydrolysis upon longer storage.

Good luck,
Arne




----------
From: Nadir T. Mrabet


Have run into a similar problem.
Cleared the background color by running 2M NaOH together with 0.2M EDTA.
Better replace BMT with TCEP (1 mM).
Also keep in mind that adsorption is pH dependent, that is the higher the pH, the  better is adsorption.
Many proteins adsorb irreversibly above pH 7.0.
If you reduce the pH, say to 5.2-5.5, not only you make adsorption less stronger (hence, column capacity may drop down), but you will at the same time prevent cysteine oxidation.
You can also increase [imidazole] in the equilibration buffer to reduce adsorption.
HTH,

Nadir

Pr. Nadir T. Mrabet