Over the past year or so, we've been working on simplifying the
cloning steps required for the construction of donor vectors suitable
for RMCE. Our reasoning was that if short 40 bp attB sequences would
suffice for integration, they could be added as tails on PCR primers
when amplifying a sequence of interest; simply cloning the resulting
PCR fragment would create a de factor donor vector. In our hands, this
works quite well, and we hope that others will be able to involve a
similar scheme to simplify their own experiments. Our manuscript
describing this approach is now available as an epub ahead of print here.
We have been working on moving lines carrying RMCE targets
into the Bloomington stock collection, and one has now been made
available (BL25091, the P[attP.w+.attP] insert at 89B). Of course, all
lines are still available from us upon request, and hopefully others
will soon join the Bloomington collection as well.
We have deposited sequence and annotation info for the RMCE
target vector P[attP.w+.attP] (formerly known as pUASTP2.1) into
Genbank. Accession number is EU761203.
At the suggestion of the good folks at Bloomington, we have decided to give a new name to the vector formerly known as pUASTP2.1. The former name was largely based on history, relating to the first RMCE target vector that we made, pUASTP2. The new name is intended to be more descriptive of the relevant structural elements, namely a miniwhite gene flanked by attP sites and borne on a P element (as described elsewhere, TATA and UAS sequences have been removed in this version of the target). We have updated the vectors page to reflect this change, and hope to have lines carrying these targets available from the Bloomington stock center soon. (Thanks to Kevin Cook for suggesting the name change!)
Just to make sure everyone is aware of a publication by Johannes Bischof and the Basler/Karch labs describing Drosophila lines expressing the phiC31 integrase in the germline. The use of these flies avoids the tricky task of preparing integrase RNA to co-inject with donor plasmids; instead, the integrase can be introduced by crossing females expressing the integrase to males carrying target sites, and injecting into the progeny embryos. They report nice rates (30-50%) of integration using their landing platforms containing single attP sites. I have the stocks in but have not yet tested integration using our RMCE targets - I'll post a note when I have some data. Information on ordering these flies is available at their website.
Andrey and Art from the Kuroda lab alerted us to the fact that piB-GFP contains a small piece of sequence derived from a 3' P element end, which immediately flanks the 3' UTR (SV40 tail) of the GFP gene (this came along for the ride in our cloning strategy). The 100 bp sequence extends from 2906-3004 in the piB-GFP vector sequence. Popping out GFP with BamHI removes this region, while cutting with SalI or HindIII does not, meaning that following an RMCE event, this 100bp sequence will be present both in the cassette and in the P element end of the target site. We don't expect this to have an effect on gene expression, but this sequence should be avoided when designing primers to confirm insertions.
I have been having difficulty with some PCR reactions that span the phiC31 att sites, likely because they are quite GC-rich. I've been having most success using the hot start Phusion polymerase and the GC buffer that comes with it - it may be worth a try if you are having similar difficulties.
In related news, it seems that the att sites in our cassettes can misbehave in PCR reactions. Art Alekseyenko and Andrey Gortchakov from Mitzi Kuroda's lab have been seeing strange things in their PCR reactions, most notably that reactions spanning the entire cassette (i.e. using primers in genomic DNA that flank the target-containing P element) produce a bizarre product that seems to skip over the cassette from one attP site to the other, producing an internally-truncated product. This appears to be a PCR artifact: thus far, all RMCE-derived insertions tested have appeared normal by Southern analysis, and Art and Andrey tested their lines by inverse PCR and by staining of polytene chromosomes to show that the inserted cassettes are where they are supposed to be.
Just back from Fly Meeting in Philadelphila, where Bob Levis gave an update on the Gene Disruption Project. The main story was that they are moving away from P elements and PiggyBac transposons, and are starting to use Minos elements for genome-wide screening. Excitingly, Bob also mentioned that they are creating a version of the Minos vector carrying a phiC31-based RMCE target cassette, meaning that there will soon be target lines throughout the genome. No word on the exact timeline - we will keep you posted when we hear of their availability.
As promised, we are making available a new target vector, pUASTP2.1, in which the UAS and TATA sequences flanking the target cassette have been removed. For researchers wishing to insert genes of interest under control of UAS, or for analyses sensitive to the presence of a flanking promoter, these lines may be preferable to the original pUASTP2 targets (we have not specifically tested whether the UAS and TATA flanking insertions in the original targets can be used to drive expression). A description of the vector and sequence information are available on the vector page, and insertion lines are described on the target page.
In June, 2006, we published a method for Drosophila transgenesis that targets constructs to predetermined genomic sites using the phiC31 integrase system in conjunction with Recombinase Mediated Cassette Exchange (RMCE). This website is intended as a resource for researchers who wish to use phiC31-mediated RMCE in their studies. The site is a work in progress - we hope you find it useful, and we welcome feedback.