Stephen, many have made the point that TEV may be cleaving at a secondary site, which is a very reasonable possibility. Bear in mind, however, that proteases are not always perfectly well behaved, and it's possible you're getting cleavage by TEV at a site that won't be immediately obvious by a simple sequence search. Furthermore, you might even find a putative TEV site in your sequence that isn't cleaved because it's inaccessible. These kinds of puzzles are why it's so interesting to be a protein biochemist! You can get an approximate idea of where the cleavage site is located by considering the sizes of the cleavage products. Or if you have the resources, you can determine the cleavage site definitively with a quick MS-MS or N-terminal sequencing of the cleavage products. Once you know the site you'll be able to determine the specificity of the protease, which will in turn help you pick an appropriate inhibitor.

One other possibility is that you're getting some kind of autolytic cleavage. You didn't provide details about the conditions you're using for your work with this protein, but, for example, if you're exposing the protein to acidic conditions you could be getting autolytic cleavage at Asp-Pro bonds. Again, by mass spec you could determine the site and address that possibility definitively.

Use a fresh column first or use the same old one after an extensive washing with Urea/Guanidium-Cl. You could also use some EDTA (around 5 mM) with your cocktail inihibitor to chelate the divalent cations. During your column run, you could also use PMSF (1mM) in your wash and running buffer.

TEV cleaves after Q in the general sequence E-X-X-Y -X-Q-(G/S) , with E-N-L-Y-F-Q-(G/S) being the most used. Does your peptide contain the general sequence ?

It does not seem that it is the enzyme prep that causes the problem, as you re-purified it.

Ever run a check to see if there is similar tev cleavage sequence within your peptide sequence?

You can also try to elute MBP-peptide with maltose. This will tell you whether your peptide has been actually cleaved by e. coli endogenous protease, or it's just because of instability of your peptide (protein), throughout extraction and purification step.

First thing to check is if your peptide contains or resembles the recognition sequence for Tev. Second thing to check is whether or not some of the fusion peptide is being expressed in a truncated form or being cleaved after expression. If it is, try a plys S or E strain of BL21.

Dear Stephen

Which was the expression temperature in your experiment? I had the same problem in BL21(DE3), the protein was cut by cellular proteases, even in inclusion bodies when I made the expression at 37ºC. I changed my expression temperature to 24º C and I solved my problems. You can take a look to the article in the following link

http://rd.springer.com/article/10.1007%2Fs12033-012-9573-0

I hope it can be useful, in case that it was not your problem. please give us more details.

Best

I have similar issue with TEV but GFP-his, I decided to optimize the purification without cleavage, and remove the all tev gfp in my fusion construct, However I have not thought of checking the TEV recognition sequence, good point.

Maybe you have a secondary TEV site present in your protein of interest ?

In my own sequence I did find ENKAFQ followed by A instead of G or S I dont know if it is a potential secondary TEV recognition site although in the litterature i have not found it.

As many has suggested, you should check for internal sites. By the way, are you sure is cleavege by TEV??? I mean is the band absent when you analyse the total extract before the column? If it is present you could try milder expression conditions.

Are you sure about the size of the cleaved products? Are there three or two of them? What is the method to detect the product? As the "aggregate-prone peptide" it may form momomer and dimer because of aggregation (if you use PAAG-phoresis).

If your peptide sequence doesn't contain an obvious TEV site I would maybe start to blame co-purifying proteases. You may not see very much by Coomassie, but overnight cleavages leave a lot of time for those guys to mess up your protein. I guess I would think about ion exchange after your affinity purification, and after that bump up the TEV:peptide ratio to avoid overnight digestions if nothing is working.

I wonder about something a little further upstream - your expression system. The pMAL-p2x vector is for periplasmic expression of the MBP-fusion, but your cells BL21(DE3) and C43(DE3) are really designed for the pET vectors (they have the T7 promoter chromosomally expressed, hence DE3), and give great expression levels when using a pET vector, but can be less effective with other vectors. Another catch about the cells you're using is that they also express endogenous MBP, which will bind to your Amylose column and give you a very nice peak, but less fusion protein than you expect. So you have a double hit against you - non-optimal expression and competing MBP for your purification. We use the ER2507 strain from NEB to express any of our MBP-fusions; it works very well in our hands, and we get a lot of protein per prep. You may want to consider this in addition to the other suggestions here...the more protein you get upfront, the better (we're a crystallography lab, so we use a lot of protein!).

Stephen, many have made the point that TEV may be cleaving at a secondary site, which is a very reasonable possibility. Bear in mind, however, that proteases are not always perfectly well behaved, and it's possible you're getting cleavage by TEV at a site that won't be immediately obvious by a simple sequence search. Furthermore, you might even find a putative TEV site in your sequence that isn't cleaved because it's inaccessible. These kinds of puzzles are why it's so interesting to be a protein biochemist! You can get an approximate idea of where the cleavage site is located by considering the sizes of the cleavage products. Or if you have the resources, you can determine the cleavage site definitively with a quick MS-MS or N-terminal sequencing of the cleavage products. Once you know the site you'll be able to determine the specificity of the protease, which will in turn help you pick an appropriate inhibitor.

One other possibility is that you're getting some kind of autolytic cleavage. You didn't provide details about the conditions you're using for your work with this protein, but, for example, if you're exposing the protein to acidic conditions you could be getting autolytic cleavage at Asp-Pro bonds. Again, by mass spec you could determine the site and address that possibility definitively.

Have you already tried simply having the peptide synthesized?

not that you want to go back and do more cloning (although it might solve some problems here), but in general i highly recommend the SUMO protein fusion over MBP for the following (also very soluble):

1) smaller molecular weight (not discarding as much mass on cleavage!)

2) works great with T7-promoted pET systems

3) His-tagged (better purification than amylose)

4) most importantly for your problem- highly specific cleavage. SUMO protease does not recognize a sequence, but the entire tertiary structure of the SUMO domain, leading to virtually no random cleavage

5) on cleavage, there are no extra residues left on your protein/peptide

and no, i did not develop the technology, it's just worked very well for me on difficult to express proteins/polypeptides

Stephen,

when you say you get "the peptide in both its full-length form and a shortened peptide, cleaved at the same point in the middle of the peptide" you mean after you analize your reaction mixture on Coomassie stained gel (after cleaving with TEV)? Or? After second purification to remove cleaved MBP?

Because after cleavage with TEV it's possible to see MBP (that is cleaved off) on Coomassie gel as a band (it's around 47 kDa long), so maybe this is your second shorter peptide if this above is the case. How long is this shortened peptide?

First be sure that your protein of interest have a autocatalytic processing activity or not ?

Try to express the protein at temperature below 37 degree C.

I have some experience purifying peptides from very similar cell lines. Addition of a lypoil tag prior to the TEV site (i.e. : AFEFKLPDIGEGIHEGEIVKWFVKPGDEVNEDDVLCEVQNDKAVVEIPSPVKGKVLEILVPEGTVATVGQTLITLDAPGYENMTASSSGS ENLYFQG) of your peptide can help hugely in both peptide solubility and yield. Another C-terminal lypoil tag can be used to further increase stability, but I don't find it very useful personally. Furthermore, I have no issues with incorrect digestion with these tags with a TEV site.

Finally, peptides rarely run to their expected mas on an SDS-PAGE, so it is very much necessary to run a simple MS to ensure you have the correct mass after purification.

Thank you all for your advice! I ultimately ended up changing my construct completely. As some suggested, there was a significant amount of cleavage occurring in vivo, as well as additional cleavage in subsequent purification steps by small amounts of copurified protease. I tried further purification steps to remove the protease, such as ion exchange and HPLC, as well as adding additional inhibitor cocktails, but it did not appear to work. I think the peptide folds in the new construct in a way that it is not as accessible, and that has been working well. Thanks again!