Worm Breeder's Gazette 16(2): 23

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.

Optimizing the use of heat-shock sense + antisense constructs to obtain a heritable interfering element.

Morris F. Maduro , Joel H. Rothman

Neuroscience Res. Inst. and Dept. MCD Biology, UC Santa Barbara, Santa Barbara, CA 93106

We are using RNAi to study a redundant pair of genes, med-1 X and med-2 III, involved in mesendoderm (MS + E) specification.  Unlike many of the target genes used as standards for RNAi methods, the early zygotic med genes show only moderate interference, even by direct injection of concentrated dsRNA: at an optimized time between 7 and 9 hours after injection, ~50% of the F1 embryos display the mutant phenotype (see Figure 1).  P>P>

We have been exploring alternative methods to generate large numbers of interfered embryos at higher penetrance.  The Driscoll lab described a method for RNAi using hsp16-2 promoter-driven expression of transgenic inverted repeat (IR) or 'snapback' constructs (http://elegans.swmed.edu/wli/[wm99p11]/).  Several attempts to construct hs-IR plasmids for med-1 failed to yield recombinants, even using the E. coli SURE strain, in which IRs are stabilized. We therefore sought to test the effect of expressing both sense and antisense RNAs from an array.  The MED-1 ORF was cloned into both hsp16-2 and hsp16-41 promoter vectors in both sense and antisense orientations.  Transgenic worms carrying these four constructs on a single array were made using unc-119(+)as the selectable marker, which simplifies construction of doubles with rol-6(d)-basedGFP reporter arrays. P>P>

Heat-shock was performed on young transgenic adults for 3 hours at 33oC, and a time course was performed at 2-hour intervals to find the optimal window for generating Med embryos.  We found that the interval between 11 and 13 hours after heat-shock yielded >70% embryos with a Med phenotype.  We have since obtained similar results with an array bearing hs-sense + antisense plasmids for the redundant genes end-1 and end-3.  In this case, one of the cDNAs carried a frame-shift mutation. It is likely that introduction of such coding errors might be useful in mitigating possible effects of overexpressing the sense message.P>P>

Figure 1. Recovery of Med embryos by two methods.P>P>

Our results suggest that expression of sense + antisense constructs from an array works at least as well as direct injection of dsRNA, and as no special cloning approaches are needed, might be an easy alternative to the hs-IR approach.