Why so many tags - MBP and GFP and poly-His?

Your 35kDA and 12kDa fragments evidently represent a single cleavage site. Does the sequence at this site in any way match the sequence specificity of the TEV protease? In theory you could of course, as you suggest, eliminate the cleavage by changing the amino acid at that position, but a) it seems a lot of extra effort; b) it might cause other problems with folding or the activity of your protein; c) maybe you could eliminate the problem more simply.  If it is the TEV protease that is the guilty party could you treat either with a smaller amount (also helping with your 30kDa band problem) or for a shorter time? If it is not the TEV protease then maybe a cocktail of inhibitors of other proteases might prevent this cleavage.  As for the contamination, maybe a different type of chromatography - hydrophobic?

Paul

Why so many tags - MBP and GFP and poly-His?

Your 35kDA and 12kDa fragments evidently represent a single cleavage site. Does the sequence at this site in any way match the sequence specificity of the TEV protease? In theory you could of course, as you suggest, eliminate the cleavage by changing the amino acid at that position, but a) it seems a lot of extra effort; b) it might cause other problems with folding or the activity of your protein; c) maybe you could eliminate the problem more simply.  If it is the TEV protease that is the guilty party could you treat either with a smaller amount (also helping with your 30kDa band problem) or for a shorter time? If it is not the TEV protease then maybe a cocktail of inhibitors of other proteases might prevent this cleavage.  As for the contamination, maybe a different type of chromatography - hydrophobic?

Paul

Hi Paul,

Thanks for your reply:)

MBP-to improve solubility, GFP- to be able to observe expression using microscopy and His- because I am using His purification as the capture method. 

No there isn't a sequence that is identical to TEV recognition site in my protein. Yes, you are right mutating residues would take more time and effort. I believe that the degradation took place after TEV protease digestion. I used PMSF and all purification steps were done quickly. 

Hydrophobic interaction chromatography might be an option, but our lab does not have the column unfortunately.

I rarely see MBP-GFP-HIS tags on the C-terminus. Have you tried just a strep tag or his tag at the N-terminus?

I understand you did Mass spec of the 30kDA TEV band from a gel from another prep, but not this one? You might want to check that here too. TEV should cut pretty specific.  Degradation of a part of a protein is pretty common and would point to structural flexibility or protease contamination from expression. 

Try purifying your protein with high protease inhibitors  (His with inhibitors, ion exchange without, not necessarily gel filtration but that might help too) before TEV cleavage and see if that fixes your problem.

Next thing I would do is limited proteolysis with your protein after the affinity column. Do you get the 12kDa  band as stable intermediate? in that case you should be able to separate the fragments with affinity chromatography, depending of the proteolytic stability of your C-terminal  tag.

Also, have you tried to just crystallize your prep? You might end up with crystals for either domain, or both. People also crystallize proteolytic digests a lot. One of my colleagues got a 0.75A structure out of a limited trypsin digest....

Either way, you should check if your protein is folded before you do anything with it (e.g. CD spectroscopy)

Stefan

Three points, Bahja:

1. If cleavage really happened after TEV treatment (What makes you think that?), then it is worth bearing in mind that PMSF will only inhibit one category of protease. You could try an -SH inhibitor for thiol proteases and EDTA for  metalloproteinases.

2. When you ran your size exclusion chromatography, did you get separation of fractions containing 47kDa, 35kDa, 30kDa and 12kDa material or did you see a single peak that  gave all 4 molecular weights on SDS? I ask because the second possibility would suggest that perhaps the intact protein is nicked at that one position in vivo in the expression system.

3. If you want to try hydrophobic chromatography and can't beg or borrow a column, it is worth remembering that you don't need an official hydrophobic column! If you load in concentrated ammonium sulphate then even DEAE cellulose becomes a hydrophobic support and your proteins will stick and be  gradually eluted by a descending gradient just as surely as if you had paid for Phenyl-Sepharose!

Paul

I agreee with Dr. Paul

Thanks for your reply Stefan:)

Yes, I did try to express my protein with a His tag at the N-terminus but the amount of expressed protein I got was very tiny! Not comparable at all with the MBP-GFP-His at C-terminus.

I also tried to set up crystallization trials with my prep., but luck wasn’t on my side! Not a single crystal unfortunately:(

Thanks again for your informative reply Paul:)

I think it will be easier to understand by looking at the attached SDS-PAGE gel.  Lane 2 represents the product after affinity chromatography. Lane 4 is TEV protease. Lane 6 is the post TEV digest reaction (so at this point you can see the band at 35 kDa which was not there before the digest reaction).

Size exclusion chromatography resulted in 1 peak (as shown in attachment); lane 9 represents the sample from the peak. Lane 10 represents the concentrated sample from Gel filtration peak.

Another thing which might be important to mention is that we know that the protein which I expressed is produced as pro-pre enzyme and the N-terminal part of the protein is cleaved in vivo to produce a mature enzyme. However, as I mentioned previously attempt to express the mature functional enzyme failed.

The  deeper we get into this the more puzzling it becomes and I have lots of questions:

1. Why does the pre-TEV digestion sample in Lane 5 look identical to the post-digestion sample in Lane 6? Obviously we don't know whether your zero-time sample was really zero-time, but it looks as if the 35kDa band appeared as soon as your sample saw TEV protease!

2. Since this protein is produced in a pre-pro form, is it a protease itself? Is it processed in vivo by another protease or by auto-cleavage?

3. What is the sequence of the half dozen residues either side of the apparent cleavage site that generates the 35 and 12kDa bands?

4. Is this a monomeric protein and what was the size exclusion range of the column you used?

5. Has anyone tried to unfold and refold this protein? (Might be a way of getting rid of a nicked portion - gel filtration before refolding).

6. Going back to the crystallisation attempt, a) what protein concentration did you use? Was it high enough?

b) did you try to define the appropriate concentration range of (e.g.) ammonium sulphate by first seeing what concentration will bring the protein out of solution in a crude test-tube experiment? If 40% brings it down, then setting up drops at 35% over solutions ranging from say 37% to 45% might stand a good chance.

Paul

Have you tried to have a second tag at the N-terminus and do another purification so that at the end you have only the full lenght protein?. If you have for example 6his-yourprotein-MBP-GFP and you start with amilose resin then TEV and then a nickel column, at the end you will end up only with a full lenght protein. I remember in the past I used a double tag (N- and C- terminus) and I got a full lenght product.

Many thanks Paul for getting back to me again, I will try to answer all your questions

1)      I took the sample for pre-TEV digest after adding the TEV to the protein+buffer+DTT, so maybe I was not quick enough or possibly TEV protease is very efficient in cleaving the fusion protein as soon as I added it to sample.

2)      The protein is processed by a serine protease which converts the inactive pro-enzyme to an active enzyme.

3)      Most probably the cut is between R & Q: SKRSFSR QSELIFD (However I am not 100% sure as the mass spec results of the small band was inconclusive.

4)      Yes the protein is a monomer, and the GF column I used was Superdex 75 10/300 GL.

5)      I am not sure if anyone has tried to refold the protein- but I have no experience at all in refolding (I haven’t tried to do it before).

6)      A. For crystallization, I set up 2 drops: one at 6mg/ml and the other at 10 mg/ml. Yes, I think it was high enough as many drops had precipitate in them.

B. No, I never tried to precipitate my protein using ammonium sulphate. Thanks for the suggestion. Is this technique usually used to determine the concentration of protein suitable for setting up crystals? I used the pre-crystallization kit from Hampton research to check whether the sample is concentrated enough for crystallization. I got heavy precipitate with one of the reagents, and so I decided to set up a less concentrated sample (6 mg/ml) as well.

Thanks for your suggestion Marco:) I believe it is a good idea but I will lose a lot of protein since there will be an extra purification step.

Your TEV protease should have a 6His-tag so after cleavage you just load the IMAC again but this time you collect the flow-through. Move directly to SEC. Unless of course your protein still binds IMAC without the his-tag. 

The cleavage might be auto-catalytic in which case you will just have to leave it until you think it is fully cleaved and run it over SEC again, in a buffer that doesn't promote the association of 12 kDa and 35 kDa. Or mutate residues in the approximate cleavage site until it stops. But start with the IMAC flow-through.