You could try the biotin tag described here:

http://www.mcponline.org/content/5/4/737.full.pdf

Hi there,

the fact that you are intending to purify under denaturing condition is limiting as most of tags are used for affinity purification with anti-tag conjugated column and the denaturing condition will affect antibody binding activity.

If there are cysteines in the sequence, you could reduce the protein and use a thiol-sepharose. Once a disulfide is formed between the column and protein it should be very stable if you can keep the pH < 6.

PNAS Vol. 72, No. 8, pp. 3029-3033, August 1975

Of course if you do not have any cysteines you would have to engineer them if you have an expression system. If you are in to making a specific purification construct, a related technique relies on the sequence -C-X-X-C- to bind to arsenic sepharose. Chem. Rev., 2013, 113 (10), pp 7769–7792

Tough but interesting problem. Good luck.

Thanks! : )

maybe...If you know the MW you can try run 1D SDS Gel and then electroelution (small amount of sample) or some size exclusion chromatography.. for sure the result depends how complex is your sample.

 @Carolina: Hej, the problem is, I would need to get quite some protein purified ... the problem if I purify it from inclusion bodies in native conditions is that I have to refold it at extremly low consentrations and it is really a very long and unefficient proces. If I clean it on native tags prior to refolding I have to denature it again after purification and refold it. 

@Kurt: The cysteines are really a good option, I read some papers before but not the one from 2013, so its really good.  Do you use it? how has it worked for you in comparison with other purification techniques?

@Maximilian: Thanks but I think I need a small tag ... 

Hi Karen

Yes, we have tried it in my group, and it does work. As we need  10s of mg quantities for structural studies, we often opted for other, less costly methods. Still, the SS bond is as stable as a covalent bond and the interaction does not require native structure. In fact, denaturing is preferred to guarantee access. The -C-X-X-C- binding to arsenic is reported to be quite specific. We just could achieve quite pure material with more conventional methods and consequently it did not get past the testing phases in our hands. 

Inclusion bodies tend to be a purification step of their own. Does your protein refold easily? Send me a PM if you wish to discuss it further. Good luck.

IEX works quite well with urea (not GuHCl).   I find that a good wash of inclusion bodies (IBs) can give >70% pure protein.  Try doing lysozyme lysis and then hit coli with ~1M urea and 1% TX-100. Try small scale first but this will not solubilize most inclusion bodies and lysis is beautifully efficient (adding a freeze/thaw helps too).  The IBs start out brown and chunky then wash down to snowy white paste.I then freeze down normalized IB pellets and do test purifications including IEX. Refolding is always more efficient if very pure in my experience, good luck!

Since you are denaturing your protein anyways, use reverse phase HPLC.  If necessary, a preperative column can be used for larger scale.  This has been applied in the past to purify proteins that have been denatured from inclusion bodies.  Look up publications that use KSI fusions (e.g. http://pubs.acs.org/doi/full/10.1021/bi800107a).  Agree that you should do the inclusion body purification as a first step.

@Karen

We have worked with a number of cysteine-containing proteins and looked into this as a possibility as the thiol linkage is so stable (at low pH). We have always been able to find more economical methods (thus far).

Why not His-tag? There are impressive reports of on-column refolding. As others above mention, inclusion bodies should get you well above 50%. Of course thiol-sepharose can also be used for column refolding, but not if you need to form native disulfides.

As long as we are thinking outside the box, avoiding inclusion bodies by pushing the protein into the periplasmic space is something that has worked very well for us (we work with smaller proteins). The added effort of making the fusion could pay back in that you may not have to refold. Here is an example of the technique. Biotechnology (N Y). 1993 Feb;11(2):187-93

Please report back on your progress.