How efficient is the PhiC31 system?

The earliest published methods for using the PhiC31 integrase to generate transgenic flies was Groth et al. (2004). In this method, the authors tested 4 genomic attP landing sites (attP1-4) for integration while supplying the integrase as capped mRNA. Using this method, the authors reported transformation frequencies of up to about 50% (% of fertile offspring yielding transgenic offspring). In our first attempts to use this system, our frequencies came nowhere near these frequencies. We know from other sources as well, that this frequency was very hard to reproduce, although, to be fair, it is very possible that we are simply not as good at injecting as the authors of that paper (said very honestly). Thus, we set out to improve our integration frequencies and, especially, we wanted to circumvent the coinjection of integrase mRNA, which is clearly a hurdle to a convenient integration system. One of the main ways we have been able to improve our efficiencies is through the use of germline-specific integrase sources. We have made a number of fly lines that express integrase only in the germline, some of which can be found here.

A second thing that we found influences efficiency is the position of the integration site in the genome. There can easily be a more than two-fold difference between the best and the worst landing sites.

Given our improvements to the technique and using typical landing platforms, we generally achieve between 30-50% integration frequencies using the PhiC31 system. However, that being said, the best integrase sources combined with the best landing platforms (homozygous for each) can yield integration frequencies closing on 70%! On our list of tested landing platforms, we have indicated their integration efficiency using specific integrase sources.