for almost 20 years (all EMBL alumni, bonjour;-), like everybody else, I'm an happy user of TEV for almost all my projects. As we know the site ENLYFQ (G) in N terminal of your POI (Protein Of Interest) leave us with a G after cleavage (if you put a G...).
For some of our projects we have to put the fusion used for expression/detection in C terminal of the POI. If we use TEV, we still have POI-ENLYFQ, usually we live with that but some "Ayatollah" would love to have a clean sheet (no or a minimum of AA in C terminal) after the removal of the fusion. In that case what would you do?
See you in HD.
The nearest we have got to what you are looking for is using CarboxypeptidaseA on a c-terminal ..KHHHHHH tag where the enzyme is supposed to stop at the charged lysine to leave only POI-K. The biggest problem with this is that it is not straightforward, it normally requires a bit of optimization for each protein and occasionally the enzyme doesn't stop at the lysine. Having said that if anyone can optimize it it is you!
Perhaps new proteases could be a new challenge for a P4EU workgroup?
Another option might be the Intein system (or IMPACT as it is called by NEB: https://www.neb.com/products/e6901-impact-kit). You can make a construct of your gene fused to the intein-chitin binding domain part. Expression yields protein-Intein-CBD, which you can purify by affinity chromatography on chitin beads. Your protein can be removed by incubation with reducing agent (DTT), eluting native protein with chemical modification at the C-terminus.
I tried it ages ago; did not work then. Now I am making a protein for a custumor who uses the cleaved product for chemical ligation and that works...
Hello guys (Sabine and Peggy, usually quick answerers are probably on vacation :-),
Before sending this mail I have to admit that I didn't look much on recent publications, I'm happy to see that this is not a trivial and idiot concern still today. We could work on this together in the future. What would be crucial would be to have access to good MS people in the case of the CarboxypeptidaseA that was mentioned by Nick. For Venomics I actually challenged the TEV with 2 more proteases and clearly TEV was better for my application (I will never publish the other studies as we stopped in between when we saw that TEV was better and stopped loosing time with the 2 other options, but we could discuss it during informal P4EU meetings, that's why they are for :-)).
We could clearly set up something to work on the C teminal proteases. I have heard Dave talking at PEPTALK about his protease, my feelings at the time (that was before the paper came out I guess), was that this was a start but it was far from the perfect answer to my problems. I have just read the paper, clearly more work is needed to get the perfect candidate.
This being said, I started with a TEV clone yielding 0,5 mg/L at its best (1995 or so I guess, Arie, do you know when Gunther Stier/Hans van der Zandt did this clone?) and before that people were refolding TEV, we have now a TEV clone yielding 100 mg/L (thanks to Helena and colleagues)...The way that Dave produced its enzyme could be much better in the light of my recent work with DsbC done on VENOMICS, I may contact him to see where he is with this matter and what are his thoughts to go further.
The Intein system would be a clever way but like everything else once you add a tag this can have an impact on the solubility, function... And if I want to add various tag to enhance solubility or detection that complexify the story.
Clearly, I'm not dogmatic and not asking too much:-), I just want to have a protease that I can produce at 100 mg/L, 100 % functional even with some detergents, that recognizes with high specificity/selectivity a small sequence that will probably not interfere, a protease that I can use with any tag in C terminal of my choice, cleaving it off clean of extra AA in a native buffer, in short, just the C terminal TEV!
Another possibility is to leave the C-terminus and use an n-terminal SUMO fusion and SUMO protease. This will give you a native peptide sequence and has worked really well for us for peptides and SIL peptides, even ones that normally aggregate quite readily.
The only problem will be that you will not be able to use your Gateway cloning method as you will disrupt the SUMO/SUMO protease site structure. You would have to switch to LIC, SLIC, In-Fusion or Gibson cloning to generate the fusion.
I also, a long time ago, tried the IMPACT system (unsuccessfully) but the addition of DTT may kill off the activity of your toxins if they contain multiple di-sulfide bridges. Depends on your toxins.
back from vacation my late reply. Apart waiting for D. Waughs efforts (at his last presentation, acitivity was still low) I have used vectors published and obtained by Aimee Shen (Plos One) 2009 | Volume 4 | Issue 12 | e8119
The principle is a C terminal Cysteine protease domain as autoprocessing enzyme tag, that is induced by InsP6 addition. So it can't be used for constructs expressed in mammalian or insect cells, but it works for E.coli.- We have purified nanobody fusions via that tag.