[Drosophila] Re: PEP transposon jumpout mutagenesis

Adelaide atc12 at mole.bio.cam.ac.uk
Thu Aug 11 16:10:04 EST 2005

Dear Man,

In the early days of using P-insertion lethal mutations, one
of the important things to show was that the P insert caused
the mutant phenotype;  one way to show this was to hop the P
back out and (with luck) recover a wild-type allele.  By
present thinking, such events are much more likely to have
come from double-strand break repair off the homologue than
directly from precise excision, but either way:  w[-] (or
ry[-]) are not expected to be (particularly) enriched for
*imprecise* excision events.

What has reproducibly worked for me is to treat an
imprecise-excision "mutagenesis" exactly like any other form
of F2 mutagenesis:  test all of the progeny, regardless of
eye color.  And, like say an X-ray mutagenesis, expect to
test a lot of F1 flies, a couple of thousand at least. 
Exactly what I've gotten has depended on the specific
insert, there do seem to be local effects;  but one specific
ry[+] hop gave exactly the same proportion of imprecise
events among the ry[+] as among the ry.  For another, all
the imprecise events going distal were white;  all the ones
going proximal still had the w[tr] color of the original insertion.

Other details:  the delta2-3 transposase promotes events
throughout development so can give clusters of the same
event -- if you are concerned about knowing whether two
different F[1] "hits" came from the same premeiotic event,
then set up your parental Pinsert/delta2-3 males as single
males (by anything/Balancer females) -- and keep track of
which F1 progeny came from which parental male.

It is important to test late-eclosing progeny as well as
early.  My experience has always been that the positives
come from the later setups;  this makes sense, since they
are a more extreme mutation than the progenitor (see below),
and even though heterozygous they will have *some* (however
slight) effect on viability/developmental rate.  (I haven't
ever been able to force myself to throw away untested those
first-eclosing progeny, however.)  So:  set up your hopping
crosses, single male by three females, and let them lay for
say 5 days;  then transfer the parents to fresh food and
give them another 5 days.  Then discard the parents. 
Assuming normal fertility, 10 such crosses will give you as
many F1 flies to test as you can handle.  Collect virgin and
male/Balancer progeny from these parental vials and set up
the test generation as soon as you have say a hundred;  but
keep collecting flies and setting up until the vials cease
to yield adults.  (Yes, of course set up the females too. 
Means half as much work expanding tester stock etc.)  Then
plan to score the test result as soon as its eclosion
begins;  >90% of no-events will present as such immediately;
 those tests can be counted and discarded, leaving the real
positives and a few slow developers for further effort. 
Scoring early also lets you detect mild hits (delay in
eclosion) that may have interesting phenotypes on further
study.  Anything you want to keep, collect virgins and males
for stock from the test vial;  it is *not* necessary to make
a stock before testing!

I am assuming you are starting from a homozygous viable
insertion and have available a deficiency or lethal allele
of the gene in question:  this is the simplest case, test
for gain of lethality on the originally viable insert
chromosome, but with modifications any kind of insertion can
be used.

This is all, of course, assuming your insertion is a simple
P insertion;  P{EP} should be fine, and that you've already
succeeded in getting whites is reassuring here!

Good luck,

More information about the Dros mailing list