Hi there,

The answer to question 1 depends on the strategy you use... Do you intend to clone into a vector providing the initiation codon and/or the stop codon and using restriction enzymes? If yes, no need to add further aa than what you want to clone. But the thing you have to care about is to respect the reading frame defined by the initiation codon and the termination codon. The primers should contain the suitable restriction site upstream the sequence homologous to what you intend to clone. Furthermore an extra 3 nucleotide at the 5'end of the primers is necessary to ensure the proper cleavage activity of REs.

About question 2: you have disuphilde bonds in the domains and glycosylation sites so for an efficient folding and glycosylation you should consider eukaryotic expression system with targeting to ER and secretion pathway (Pichia pastoris for instance).

I agree with Dominique Liger

If the 3D structure of the protein is known, it would be good to have a look . If the sequences immediately  flanking your domain of interest form connecting loops not tightly packaged as part of the other domains (as the schematic representation suggests) it could be good to add several of these residues at each end of your recombinant domain to facilitate proper folding in a natural context.

The protein has several disulfide bonds, then you could produce the domains  as  secreted proteins in E.coli (periplasmic expression), yeast or perhaps even mammalian cells, by adding a suitable signal peptide sequence at the N-terminus. In any of these cases the disulfide bonds can be formed. On the other hand, if you want to take advantage of high level cytoplasmic E.coli expression, probably you will get inclusion bodies that have to be refolded in vitro.  Which expression system are you planning to use?

The first thing I would do is to search in the literature how this protein or similar ones have been previously expressed by others. This can give you important clues about the best expression system.

Thank you all for your answers!

I plan on cloning bikunin and each of its kunitz domain in a pcDNA3.1 + plasmid that already contain a signal peptids at the N-term side and two tags at the C-term (c-myc, Flag) and express it in HEK cells. In parallel I will insert the same constructs with different restriction site in a pTrcHis B plasmid and then express it in E.coli.

I'm a bit drowning in my plasmid design right now... I have this problem that in my pTrcHis B plasmid my two restriction sites are separated by a number of nucleotide that is not a multiple of 3. As I have a his tag at the N-term the reading frame is imposed by the plasmide structure so to get back in frame I must either add some nucleotide and thus have an extra amino acid to my sequence (argine or serine) or remove some nucleotide and lose one aa from the sequence (Only possible with the full bikunin which has an Alanine at its N-term, the kunitz domains start with cysteins so I can't lose that...).

So I have two more questions :

-Would it be better to had a serine or an arginine to the N-term of my sequences? Bikunin is translated with alpha-1-macroglobulin as a a1m-Bik propeptide, they are separated by a RVRR sequence that get cleaved by an unknown enzyme. So technically arginine is the natural amino acid you would find directly at the N-term side of Bikunin.

-In general if the choice is given to lose an aa or to add an extra one, is there any general rule of thumb to pick one option over the other?

I don't think there are general rules, it depends on your protein and your purposes.

 For the full Bikunin, either removing the first Ala or adding a Ser should be ok, unless the N-terminus is functionally important for the interactions you are trying to study. Is this the case? Even when the protein is translated with an Arg in this position, its not present in the mature protein, and changing the net charge could have consequences...that's why I wouldn't use it if there are other options.

For the domains,  of course you cannot remove the first Cys. You could either  add a Ser or add some more aminoacids from the domain-flanking regions as I told you. I would prefer the second variant.

Something I would do for sure is looking the possible sequences with a program designed to predict cut efficiency by signal peptidases in the selected host (for the secreted proteins). There are free programs in the web to do that. As the inserted residues will be between the signal peptide and your sequence of interest, they can directly influence the function of signal peptidase.  So, I would test in silico whether the signal peptidase can cut in the correct position with good efficiency. If so, you can go ahead.

Hi again,

The distance between the 2 cloning sitesin the plasmid is of no importance as it will be removed and replaced by your DNA of interest. The essential issue is for you to respect the reading frame defined by the initiation codon of the plasmid in your upstream primer and the one defined by the stop codon in the downstream primer (if no stop codon is introduced in it ).

For aa introduction due to the constraint of respecting the reading frame I think that serine would be more neutral than arginine...

The last part of my comment was not complete. I have corrected it.  The text should be

As the inserted residues will be between the signal peptide and your sequence of interest, they can directly influence the function of signal peptidase.  So, I would test in silico whether the signal peptidase can cut in the correct position with good efficiency. If so, you can go ahead.