Purifying protein with Neutravidin binding peptide tag - Protein purification (Mar/31/2009 )

Hello everyone,

In order to purify a protein of interest we cloned inframe a peptide tag that has high affinity for Neutravidin beads. The peptide tag was cloned such that it does not interfere with the protein function. Sequencing results and assays specific to protein, was carried out to verify that, and the results were positive. However, upon attempting to purify the tagged protein with Neutravidin beads, purchased from Thermo Scientific, the results obtained were negative.

The procedure for testing protein binding was very simple, we added calculated amounts of beads (based upon the binding affinity) necessary to bind secreted proteins in defined medium. As a negative control, we used regular agarose beads, to another alloquot of secreted proteins. The proteins + beads were incubated for an hour at 4C, after which they were assayed for the protein. Both control and neutravidin beads showed enzyme activity in the medium, contrary to what we had expected i.e. protein bound to the neutravidin beads.

What could be the reason for this? As mentioned above, the sequencing results reveal the tag to be inframe with the rest of the DNA, also, the protein is functioning and getting secreted into the medium. If anyone of you have faced similar issues, could you come up with possible explanation and what needs to be done? I really find this mind boggling, I could be that the tag is workin fine, but the beads sent to us was from a bad batch. Please, any suggestion would be useful.

Thanks,

rumi84

have you determined that the beads actually bound your protein (put the beads on a gel)?

despite the beads affinity for your protein you may not have bound as much of your protein as you calculated (ie not enough to clear the medium).

if they did bind the protein then you could try using more beads (or less medium).

if not, then check binding conditions and adjust as necessary.

rumi84 on Mar 31 2009, 08:21 PM said:

Sounds to me that you may be experiencing one (or both) of the two major "snags with tags".

You have shown that (1) your tag sequence has been cloned in-frame and (2) that your protein is expressed and (3) that the recombinant protein can be detected by specific assays. But are you sure that your neutravidin tag (AviD tag?) is actually present on your expressed protein?

Protease activity could be removing your tag during expression. Can you detect the tag using western blotting?
If you can, then an alternative explanation is that your tag peptide may be partially masked by your protein. This would lower its affinity for neutravidin to a point where you may lose all binding. If this is the case you may need to add a spacer (3 or more amino acids) between the tag and protein or change the terminus used for tagging.

Good luck
Hope this helps

klinmed on Apr 8 2009, 01:01 AM said:

Thank you very much for your suggestion. Prior to cloning the tag into the protein of interest, we did necessary structuraly analysis. The tag was inserted into the N'terminus, because crystal structure of the protein reveals that it does not form any secondary/tertiary structure with the rest of protein. The tag was inserted after the region of signal peptide cleavage site, and we have confirmed this by using SignalP 3.0 server, so in theory the tag should be there, and since it is inserted a region that has no interaction whatsoever with the rest of protein, should not get masked, although I have done western blot using High Sensitve Neutravidin HRP, and the results were negative, so I can't say for sure. Don't know where to go from here, any suggestions will be useful, thanks again.

rumi 84

rumi84 on Apr 9 2009, 09:43 PM said:

Sometimes, expression of a recombinant protein can be a real pain in the.... (insert you own words).

You seem to have done all the "right" things. Predictive programs though do only "predict". I am a little concerned that you are trying to purify via a tag that you cannot find in your expressed protein by western blot. Is it there or not?

You do not mention what system you are using (prok, insect, yeast etc). I hope you are not trying to bind the expressed protein to neutravidin beads from crude culture medium. All undefined culture media contain huge amounts of free biotin that will block your neutravidin beads.

I work with tags including stepII, his, myc, nus.... but not the one you describe. But I think it may be a cyclic peptide that is disulfide constrained? Do you have any reducing agent in your extraction (or sample) buffer that could reduce this bond? I think that an intact disulfide is necessary for binding to neutravidin.

You mention that the tag is downstream of a signal peptide. Sometimes cleavage of these peptides is not "accurate". Would a few amino acids on the N-terminus of your tag affect neutravidin binding?

I am so sorry that I cannot be of more help.

klinmed on Apr 11 2009, 09:13 AM said:

Thanks for your response. Even questions can keep one in check. All observation was done using defined medium, which contained no serum (we use horse serum in the crude medium) but contains other essentials including, porcine insulin, fibroblast, progesterone, selenium. We are using eukaryotic system, all transfection are done in HEK-293 cells.
You are absolutely right about intact disulfide bonds being crucial for binding to neutravidin, however, as mentioned before we have not tested crude extracts on the beads, so that should not be a factor of interference between protein bead interaction. Western blot using neutravidin-HRP did not show any bands, so I am concerned about whether the tag is there or not, and if there maybe partially cleaved.
It could be an issue with signal peptide cleavage, it could be that part of the tag is getting cleaved. Well, molecular biology is always a challenge. Thanks again.

Just one correction, I meant to say Fetal Bovine Serum is used in crude medium and not Horse Serum.

rumi84 on Apr 13 2009, 08:30 AM said: