I Know the E3 ligase for my protein and i am intrested to find the site of ubiquitination. In its vicinity 7-8 lysine are resent at regular interval. Does anyone have any idea?
In your protein mutate lysines to other aminoacids eg arginine, overexpress your E3 ligase, the protein mutants alongside unmodified protein ideally with small tag so in next step you could immunoprecipitate the protein from cell lysates and look at banding pattern for the proteins using western blotting and anti-ubiquitin antibody - ubiquitination smear or ladder characteristic for polyubiquitination should disappear for the mutant lacking key lysine. It's not ideal as pull down may co-precipitate other proteins interacting with your overexpressed one and these may be ubiquitinated too. However as an initial look its fine and can give you quick answer.
If this ubiquitination targets your protein for degradation and there is an easy way to test/measure the activity of your protein (not only enzymatic activity) it is also a possibility (beside the WB smear) to use proteasome inhibitor (e.g. MG132) and test its effect on the downstream activity of your wt vs. lysine-mutant proteins.
Mass spec would be a good way of nailing down which regions of the protein may be involved, and also as suggested above lysine mutations to arginine may also help in functional assays. But there are numerous different outcomes; you may find that one lysine is really important; or you may find a number of them are required together; or you may find that it doesn't matter which one is there, as long as there is one - from this you have to make single K->R mutants, and all K->R to make "lysineless" protein; and you should also try knocking each individual K back into the lysineless protein.
I've attached a review I wrote on some of this, especially with the view to other possible sites - it may be you do all these things and even the lysineless protein is ubiquitylated: in which case it could be the N-terminal amino group; or even cysteines, serines, or threonines. This of course may not be the case, but there are some easy controls to check this isn't happening and I hope the review and some of our papers are useful to help with that.
Article Non-canonical ubiquitylation: Mechanisms and consequences