[Drosophila] is there transposase repressor construct?

[Kevin Cook]

Hi Olaf--

Using a repressor construct in crosses to make deletions from 
trans-heterozygous P elements is a reasonable idea. I considered it when 
we were using the method to make deletions for the Bloomington Stock 
Center collection (described in Parks et al. (2004) Nature Genetics 36: 
288-292), but decided it wasn’t worth it for our screens. Let me see if 
I can explain my thinking.

First, it’s impossible to predict the exact positions of breakpoints of 
deletions made by the Hybrid Element Insertion (HEI) mechanism in the 
presence of trans-heterozygous P insertions. One breakpoint should 
correspond to the insertion site of one of the starting insertions. The 
other breakpoint will usually be in the vicinity of the insertion site 
of the second P element, but you can’t predict exactly where it will 
be--it could be proximal or distal to the second insertion or within the 
transposon. No one knows why the hybrid element usually inserts near one 
of the starting insertions. It’s probably a physical constraint on the 
chromatids involved, but that’s just speculation. For an in-depth look 
at the HEI mechanism, you should work your way through the following 
three papers. It’s straightforward to see how their results apply to 
trans-heterozygous P insertions.

Gray, Y.H., Tanaka, M.M., Sved, J.A. (1996). P-element-induced 
recombination in Drosophila melanogaster: hybrid element insertion. 
Genetics 144(4): 1601--1610.
Preston, C.R., Engels, W.R. (1996). P-element-induced male recombination 
and gene conversion in Drosophila. Genetics 144(4): 1611--1622.
Preston, C.R., Sved, J.A., Engels, W.R. (1996). Flanking duplications 
and deletions associated with P-induced male recombination in 
Drosophila. Genetics 144(4): 1623--1638.

We were generating very large deletions. A local hop in the intermediate 
generation preceding the Hybrid Element Insertion event giving rise to 
the deletion wouldn’t change the predicted size of the deletion by very 
much. Indeed, most of our deletions had the expected breakpoints at the 
level of polytene cytology. A few kb of doubt about the positions of the 
endpoints probably doesn’t matter too much for a deletion in the 
megabase size range. I figured that anyone wanting to know the exact 
genomic coordinates of the deletion breakpoints could characterize them 
by sequencing off the ends of the P element retained on the chromosome 
following the HEI deletion event.

I understand that it’s a different situation when you’re trying to 
generate small deletions. A few kb difference between the size of the 
deletion based on the insertion sites of the starting P elements and the 
actual deletion size can make a big difference if you’re trying to 
delete specific genes. Using a repressor construct MIGHT make sense, but 
I think you still need to think about whether it’s worth the effort. 
Basically, you’re playing a numbers game. Not every P element hops in 
the presence of transposase in every generation. The probability of P 
transposition in the intermediate generation of the screen is not that 
high. It can happen, but the odds are in favor of it not transposing. 
You may save yourself work by doing the screens as we described in Parks 
et al. and then characterizing the breakpoints in the resulting 
deletions. You can keep the ones with desirable breakpoints and discard 
the others.

It would be interesting to know if the inclusion of a repressor 
construct would prevent transposition in the intermediate generation of 
the screen, so I’m not discouraging you from trying. I’m just saying 
that it’s not absolutely necessary to get what you want.

If you want a single gene deletion, you may be able to get it more 
easily from a conventional male recombination screen employing a single 
P insertion as described in the Preston, Sved and Engels article cited 
above. Those screens are easier and generate a lot of deletions of less 
than ~10 kb. I don’t think screens using trans-heterozygous P insertions 
would be my first choice for recovering small deletions. Of course, the 
FLP-FRT system would be my first choice, but I’m guessing that 
appropriate insertions aren’t available for the gene you’re interested 
in if you’re thinking about HEI screens. You could also check to see if 
appropriate insertions are available to use hobo transposition to make 
deletions using P{wHy} insertions (see 
http://flystocks.bio.indiana.edu/Browse/insertions/PwHy-top.htm).

I hope this helps a bit,
Kevin

-- 
Kevin Cook, Ph.D.               
Bloomington Drosophila Stock Center
Department of Biology           
Indiana University              
1001 E. Third St.               
Bloomington, IN  47405-7005   

kercook from indiana.edu
812-856-1213 (office), 812-855-2577 (fax)
http://flystocks.bio.indiana.edu