Re: [DIYbio] Re: E. coli chromossome integration

Homologous recombination, which uses lambda red, isn't the same as integration, which uses lambda integrase (yea it uses some homology, but is a different mechanism). When I say integration, by the way, I mean integration into the genome using a integrase because technically both are integration into the genome. Homologous recombination, using lambda red, will only transform your DNA into a defined location using overlaps.The integrase only integrates into a *specific site* and is far more efficient for normal strains. I have never heard of a zinc finger integrase, I'm not quite sure they exist. In fact, one of the objectives of some directed evolution projects (PACE) is to get new specificity for these integrases to use in human cells, which would be great for gene therapy. 

E coli has very little "free space" in their intergenetic regions, bacteria in general are fairly compressed. The only genes they really don't need are prophages or transposons which are just invasive elements.

E coli, without lambda red, won't homologously integrate things into their genome (efficiently). In order to do that, you must first transform lambda red (pKD46), then create competent cells of that strain and transform your linearized DNA, then select for the integration, then cure pKD46 from the strain. Sometimes the selection is CRISPR, sometimes its just KanR, sometimes it's just selecting against something that's already integrated. It can even be zinc fingers, if you wish. CRISPR is getting pretty popular lately, the lab I work at has begun to use it for integrations even in yeast. I agree 100% with Matt though, lacZ would be a way better choice than a prophage if you're going into a strain that allows it.

With an integrase, you bypass the homologous recombination. This avoids the requirement for lambda red, which can be useful. All you would do there is transform and you're set (or remove your marker with FLP). In addition, it is not restricted by restricted by size ( http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2847246/ ), so it's a good choice if your gene size is >2500 base pairs.