Site-Specific Integration from GenetiVision
Site-specific integration is a powerful new technique that has many important implications for Drosophila geneticists. This approach allows researchers to insert transgenes at single, precisely known sites in the genome. A full description of the technology of this system is available in these two publications, available online for free:
P[acman]: a BAC transgenic platform for targeted insertion of large DNA fragments in D. melanogaster.
Venken KJ, He Y, Hoskins RA, Bellen HJ.
Science. 2006 Dec 15;314(5806):1747-51. Epub 2006 Nov 30.
http://www.ncbi.nlm.nih.gov/pubmed/17138868
An optimized transgenesis system for Drosophila using germ-line-specific PhiC31 integrases.
Bischof J, Maeda RK, Hediger M, Karch F, Basler K.
Proc Natl Acad Sci U S A. 2007 Feb 27;104(9):3312-7.
http://www.ncbi.nlm.nih.gov/pubmed/17360644
There are three major benefits of using site-specific integration. First, it is possible to insert very large constructs into the genome. We have successfully recovered transgenics for DNAs up to 122 kb. This allows researchers to employ genomic rescue constructs and tagged genomic constructs for nearly all genes in the genome, something that is simply not possible using P-element mediated transgenesis.
Second, site-specific integration allows investigators the luxury of comparing multiple constructs that are inserted at precisely the same genomic location, eliminating position effects on expression levels from inserted transgenes. Thus, constructs differing by as little as one base pair can be directly and quantitatively compared.
Finally, another benefit of the PhiC31 system is that one needs to recover, balance and analyze only one insertion per construct since all insertions are in the same position. This is in stark contrast to P-element insertions, which are nearly random and require researchers to typically recover, balance and analyze 4-6 independent insertions per construct. Since you know in advance the precise location of the insertion, balancing is simple and only one or two insertions per construct need be studied.
One qualification for the above discussion is that there are two variations observed with PhiC31 integrase-mediated insertions. First, about 5% of insertions turn out to not be inserted at the intended docking site but are located somewhere else in the genome. Second, about 5% of insertions appear to turn a viable docking chromosome into a homozygous lethal chromosome. For these reasons, it is recommended that investigators balance at least two independent insertions with the PhiC31 system and test (by PCR) whether an insertion has actually occurred at the intended docking site. In addition, before deciding on a single transgenic to study, investigators should test if the transgenic chromosome is still homozygous viable. Since these events are relatively rare, isolating two independent insertions per construct is usually sufficient to ensure recovering the transgenic you want.