Create large deletion in bacteria - (Feb/27/2006 )

Hey guys,

I want to delete a large segement of DNA (up to 200 kb) from the bacterial chromosome of a Vibrio species. I was thinking of cloning 1kb fragments of DNA from either side of the area I want to delete and adding an antibiotic cassette in between. I thought I would use a sacB plasmid, that way I can select for a single cross-over using the antibiotic marker for selection, and sacB as a positive marker for the second cross-over. Does anybody think this may or may not work, or have a better approach? Thanks.

M.

has anyone heard of the Cre/Lox system for making large deletions?

sacB as a counterselection marker is problematic. I'd use something else. The pheS mutant version of Kast (1994) is pretty good. It's a point mutation in the wt pheS gene. You need to make sure the pH of the parachlorophenylalanine YEG medium is correct.

I have used the sacB approach successfully in Pseudomonas, but couldn't get it to work in Bacterodies. The cre/lox system is enzymatically driven (whereas the sacB system relies on passive recombination to achieve single and double cross over events), and thus might be more effective for your experiment.

But my main concern is the extent of the deletion you're trying to make.

What is it you're trying to remove that occupies 200 kb? If it's an operon, can you just insertionally mutagenize it by disrupting its promoter? In any event, were I to try such a large deletion, I would use flanks larger than one kb.

the area I want to delete is actually 120-kb...its not an operon but rather an area that accumulates mobile pieces of DNA. The entire area consists of ORFs that have been acquired by horizontal transfer and we would like to delete it to see if its important for adaptation and/or pathogenesis. I have heard about the problems with SacB and that you need to check at every step that the SacB is active as it tends to mutate. Hmmmm...I agree about having flanks larger than 1Kb. ....just not sure if I should try it. Thanks for your comments guys.