Say I transduce a nucleic acid sequence using a viral vector that encodes a protein having at least one disulfide linkage. For simplicity, let’s assume that there are only two cysteines in the protein and the side chains of these cysteines are close together when the protein is folded with the cysteines unlinked (reduced/thiol). Also assume that the cell being transduced is a eukaryotic cell (mouse or human). What is the chance that this disulfide will be linked (oxidized) when expressed by the cell? Does this happen pretty reliably? What might it depend on?
I am not an expert on expression of genes from viral vectors, but as I requested the poster to clarify his question I feel an obligation to provide at least a partial answer.
The first question to address is whether the mRNA for the protein of interest is expressed in the same way as normal cell proteins. In the case of the lentovirus vectors which are integrated into the genome there is no reason to think otherwise.
In this case the factors determining the formation of disulphide bonds would be the same as for any similar normal cell protein translated from mRNA.
Only secreted proteins have disulphide bonds.
In eukaryotes these protein are synthesised on ribosomes on the endoplasmic reticulum, which obviously must be present in the cell in which one is trying to express the protein (i.e. it must be capable of secreting proteins). One assumes that if one is restoring a function found in normal cells this will be so.
Oxidation of sulphydryl groups occurs in the intercisternal space before secretion. It involves a number of proteins and there are reduction and chaperone systems to deal with inappropriate oxidation, which occurs to a greater or lesser extent. A recent freely-accessible account of this topic can be found in a Cold Spring Harbor perspective by Neil Bulleid in 2012.
The (small?) proportion of incorrectly oxidised proteins that are not recycled are presumably removed by the cellular mechanisms that degrade aberrant proteins.
It would not seem possible to answer in a literal manner the question what the numerical ‘chance’ is of correct oxidation, but this would not seem to be pertinent. The question of interest is whether the protein would be oxidised to the correct native structure with the same efficiency as in normal circumstances. There seems no reason to think it would not.