There are a number of ways to do it depending on the sites of your vector. My suggestion would be to employ an EGFP vector with a 3'-mcs site and a unique site (U) immediately 5'- of the EGFP cdr. Design primers for the cdr of your gene of interest (GOI). Both primers should have a U site 5'- extension. Furthermore, the reverse primer should replace the stop codon with a specific restriction site (X), which is also present in the mcs (5'- of the stop codon). Correct cloning using a dephosphorylated vector fragment should generate U-GOI-X-U-EGFP-X-stop. The EGFP can be excised with X, leaving only a short non-specific tag at the CT og your gene. Your GOI can be excised with U. Some people prefer adding a (GGGS)5 polypeptide linker between X and U.

That said, you can never rule interference with any tag no matter how small. To solution is usually to generate two independent constructs, one with a CT and one with an NT tag. Best of luck with the cloning.

I need a C - tag because on the N-terminus I got the mitochondria targeting sequence in my protein. Nevertheless C- terminus it also important in that protein that is why I think the linker composition may be important. You suggest to add 5* glycine as the linker. Is it ok to add so many of the same codons nearby? ( is it possible that this may prone polymerase to sliding and possibly cause nucleotide insertion during replication and out of frame mutation ?) . In side directed mutagenesis in most cases alanine is used as the substitution of essential amino-acids in the mutated protein. Can anyone tell something more about the amino - acid composition of the linker between GFP- and your protein of interest. Is it best to add only glycine in the linker?

Linkers are usually composed of small and flexible residues like glycine, alanine and serine so that the adjacent protein domains are free to move relative to one another. Serine helps solubility also. The length of the linker has to be chosen so that the two adjacent domains do not sterically interfere with one another. This depens of course of what you want to do with your construct. And these aas have different codons (4-6 each ) you can use.

There is no "Golden standard", best is to try a few different lengths.

2 links :

- Robinson CR, Sauer RT. Optimizing the stability of single-chain proteins by

linker length and composition mutagenesis. Proc Natl Acad Sci U S A. 1998 May

26;95(11):5929-34. PubMed PMID: 9600894; PubMed Central PMCID: PMC34497.

- http://partsregistry.org/Protein_domains/Linker

Thank you dr Hubert - this article will be very helpful for me. At first glance I see that it is in general better when the linker is rather longer than shorter and there is both glycine and serine and alanine present in the sequence.

I have previously employed successfully the oligo linker below between two "tethered" proteins (transcription factors). The linker codes for GGGSGGGSGGGSGGGSGGGS or (GGGS)5. The nucleotide sequence was

5'-cccgggtcgaccctccgccggaccctccgccgctcccacctccagaacccccgccggatccaccccccactagtgcggccgcc-3'

For details refer to my publication: J Biol Chem. 2007 Nov 30;282(48):34623-33. Epub 2007 Sep 24. PMID: 17893140