His-tagged protein not pure - (Jun/04/2008 )

Hi there,

I am trying to optimize the Ni-NTA purification protocol of a C-terminal 6xHis-tagged protein.
The eluting protein is not pure and several unspecific bands appear on SDS gel.
Buffers used are:

resuspension buffer: 20 mM NaH2PO4/0.5 M NaCl
wash buffer 1: 20 mM NaH2PO4/0.5 M NaCl/20 mM Imidazole
wash buffer 2: 20 mM NaH2PO4/0.5 M NaCl/50 mM Imidazole
elution buffer: 20 mM NaH2PO4/0.5 M NaCl/300 mM Imidazole

I purify the protein with ÄKTA using a GE Healthcare HisTrap FF column.
The imidazole concentration in wash buffer 2 is already higher than recommended. How can I change the conditions to minimize unspecific binding to the column?

Thanks for your help,

chalet2

---

I assume that this is all at pH 7-8, and it is a bacterial lysate (which usually has a low background for me). And that these unspecific bands are seen by coomassie staining. If so, this is what I'd recommend.

Include 10mM BME is all buffers. And Protease inhibitors in your lysis buffer (I'm sure you do already)
Perform a low imidazole wash step with 2M NaCl and 1% NP40 (IGEPAL).

I also include 10%glycerol in all my buffers for easy freezing, but I don't know if it would affect nonspecific binding. I also use Tris rather than phosphate buffer, which could affect binding. I use NiNTA sepharose, not tightly packed like the hiTrap, so your problems might be due to clogging with large aggregates of protein/lipid/nucleic acid/etc. Not sure if your Load is a lysate or has been partially purified already.

If this doesn't work, you may just have to do a second chromatography step. Maybe just reloading NiNTa again.
dan

---

Mmh, yes, sample is high speed supernatant of a bacterial lysate and pH is 7.4. The sample contains protease inhibitors and I add 10 % glycerol before I freeze samples.
Do you think BME could inhibit the protein`s enzymatic function? I was rather cautious in using any additives.
Everyone in my lab works with E. coli BL21(DE3) for expression and they don`t have problems with impurities using the same buffers. I use Rosetta strain. Does this have any influence?

---

Not sure about Rosetta. I've been using DE3. I think a reducing agent should only help. I'd recommend DTT, but it's not good with NiNTA. But you can always dialyze it out. What's the enzyme?

---

I doubt that an one-step tag affinity chromatograpyhy is able to result in a pure protein; you may achieve a good enrichment;

but, the positive message is: you seem to have an Äkta system, and better try an additional purification step, f.i. gel filtration

---

Enzyme is a choline kinase. I think about adding a choline-sepharose column purification step.

---

Looks like a good drug target. You're probably right about avoiding reducing agents, as my brief scan of the literature showed no use of them in in vitro reactions.

For enzymatic reactions, you won't necessarily need 100% purity. Just so no endogenous (bacterial) proteins are competing for substrates of have the same enzymatic activity. I think gel filtration might be a good step, at least analytically, to see if the contaminants are actually associating with your protein and to show that your enzymatic activity peaks only with your protein peak. A choline column may concentrate bacterial choline-binders with your protein or may leach off choline, which would affect your enzyme assays.

good luck
d

---

I agree with the Bearer. Gel filtration should give you high-purity protein (our lab always uses it to prepare proteins and complexes for crystallization), unless of course the contaminants are close to the MWt of your enzyme...

Good luck.

---

another thing you can try is lower imidazole elution buffer.

the impurities may be binding to the ninta tighter than your enzyme (or looser but tight enough that it requires a higher concentration of imidazole than you are washing with to elute it).

you can try eluting with a gradient or by adding more, lower, elution steps and seeing where your enzyme elutes and where the contaminants come.

---