I have to choose an expression host with encoded rare codons for expressing pET vector insert. There are multiple choices of E.coli derived strains available from Novagen. The family of Origami strains carry the trxB and gor mutations for enhanced disulfide bond formation. I am wondering is there any method to analyse presence or required formation of disulfide bond in the final expressed protein. (Target protein is prokaryotic in origin)
You can't tell from one amino acid sequence alone. If the origin of the sequence is a cytoplasmic protein, the cysteins are unlikely to form disulfides, if it is a secreted protein, they are likely to form disulfides. If the cysteine positions are highly conserved between orthologs in different species, they are likely to either form disulfide bonds or be directly involved in function, if they show low conservation, they probably are not. If you can build a homology model, you can assess the relative position of the Cys residues: if they are positioned to form a disulfide bond, they probably are, if they are out-of-reach of each other, they probably won't.
Dear Aniket
How many cisteines contain your protein?
As Annemarie told you..only a good model (to built it you need to check if there is some similar protein with solced structure by performing a last search of your protein in the PDB) can show to you the position og the cysteines. Of course the nature if the protein can help you to make hypotesis and the relative position of the cysteines (ex a cxxc motif can be a thioredoxib and a cxxxc a copper binding protein)
bacterial and citoplamic--> no disulphides ( you can you standard bl21(de3):
bacterial (periplasmic or surface exposed) --> disulphides--> secretion in periplasm using ompa or pelb leader or origami ( or similar) strain
similar consideration are valid also for mammalian proteins
in case yoy have many cisteines (>8 ) i think that you have to evaluate a mammalian expression system because it is an hard work for e.coli. it may succeed but you need to be very luck and smart.
good luck
Manuele
Hi Aniket,
Please, check this link, maybe you can find something interesting to you.
https://omictools.com/protein-disulfide-bonding-prediction-category
The task is easier if you have a structure, you can visually check results by downloading the structure.
In the case with just a primary sequence, I would recommend to run jobs on different servers and then cross-compare results. If they will be consistent and support each other the result will be more trustable. Which still would be statistical approach. But at least something than nothing.
George.
And I forgot to mention, that this question was discussed earlier...
Check this link
https://www.researchgate.net/post/Is_there_any_software_for_predicting_disulphide_bonds_in_a_protein
Sorry about asking redundant question although the answers I got from Doctors Annemarie Honegger , Manuele Martinelli and George Minasov are excellently informative and beneficial providing much more comprehensive understanding about this problem along with specifically addressing this question. Thanks a lot for helping me out!
- Badges
- Science topic
- Similar topics
- Computer Science
- Research Topics