Worm Breeder's Gazette 11(5): 67

These abstracts should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.

Of Rollers and Twitchers

Andrew Fire and Susan White Harrison

Craig Mello has in the last few gazettes introduced the use of rol-6 
as a coselection marker and some highly efficient and user-friendly 
techniques for creating transformed lines.  We have been using these 
techniques, and have several observations that should be useful to 
other transformers.
Getting a few good roller 
Craig's technique of high volume distal injections yields large 
numbers of F1 animals that express the injected DNA (this is seen both 
with rol-6 and with lacZ fusions).  Cloning of all of these animals is 
very tedious.  To make a set of rol-6 transformed lines from a given 
construct, we inject 9-12 animals, placing three injected animals each 
on 3-4 medium size (6cm) plates.  After a week the plates are starved, 
with mostly F2 progeny.  From each initial plate three large chunks 
are placed onto a large seeded plate (9cm).  Larval rollers can easily 
be spotted after a day or so, we generally pick 6-8 for each initial 
plate, and of these about 90% yield transformed lines.  Different F2-
F3 clones from the same initial plate need not be independent, so we 
only analyze one clone from each initial plate, generally the one with 
the highest transmission frequency.  This procedure yields 3-4 
independent high transmission lines for each injection set.
When mixtures of the unc-22 antisense plasmid pPD10.46 and a second 
plasmid are injected at high volume into the distal arm, 3-10 
twitchers are seen per injected adult.  Unlike the roller injections, 
a relatively high fraction of these animals yield transformed lines: 
between 40% and 80%.  This suggests that pPD10.46 does not cause 
twitching except when present at high copy number, and thus that 
twitching is a more reliable indicator of array formation in the F1 
than F1 rolling or  -gal staining.
Any marker used for co-transfomation experiments has the potential 
to affect the expression of co-transformed DNAs.  Although these 
effects do not preclude the use of a marker, they should be taken into 
account when interpreting results (particularly unexpected results).  
We find that both pPD10.46 and the rol-6 plasmid pRF4 can have 
distinct effects on the expression of other constructs in mixed arrays.

When cotransformed with the 'twitcher' plasmid pPD10.46, some 
constructs (e.g.  unc-6::lacZ and glp-1::lacZ fusions) tend to 
ectopically express in body wall muscle .  This probably results from 
cis enhancement by the unc-54 enhancer and promoter used to drive 
antisense RNA production.
In co-transformation experiments with pRF4 we have seen both 
negative and positive effects on expression of co-injected lacZ 
fusions.  Mixed arrays with pRF4 and a normally inactive glp-1::lacZ 
fusion express in an uncharacterized but reproducible collection of 
cells on the surface of the embryo.  Mixed arrays with pRF4 and an unc-
6::lacZ fusion have abnormally low levels of postembryonic expression 
outside of the pharynx.  Similar negative effects have been seen with 
some myosin and hlh-1 fusions.  The pRF4 effects are more noticeable 
when the co-transformed plasmid has a minimum of 5' flanking DNA.  The 
protection from context effects afforded by the extra 5' flanking DNA 
may be nonspecific.  This argues for caution in interpreting putative 
negative regulators uncovered during deletion analyses.