Use of eat-5 as a selection marker for transgenics

eat-5 encodes a Caenorhabditis elegans innexin, a component of gap junctions between the muscle cells in the pharynx. The eat-5 null worms grow normally on an easy-to-eat strain of E. coli, HB101, but arrest as L1s on a harder-to-eat strain, DA837, a derivative of OP50. This difference in growth on two E. coli strains makes eat-5 a convenient co-transformation marker. This is particularly useful for situations when rol-6 is not good because the movement phenotype is studied. The gene of interest is co-injected into the eat-5(ad1402) null strain DA1402 with the genomic fragment containing the eat-5 gene, and the injected worms are transferred on the DA837 E. coli. Only transgenic F1s will grow. (It is important to use the null allele ad1402--mutants carrying the weak allele ad464 can grow on DA837.) Once the transgenic strains are isolated, they can be maintained on DA837, so a population of adult transgenic worms is easily obtained. If non-transgenic sibs are needed for controls, a transgenic worm can be transferred on HB101, so that non-transgenics can grow. In this case, however, a second marker, such as GFP, is needed to identify transgenics.

We tested this idea by co-injecting three constructs into DA1402: a constitutively active glr-1 receptor under its own promoter, glr1::glr1(Ala->Thr), which is known to cause an increased reversal frequency phenotype, myo2::GFP (pPD118.33), to have an additional label for the transgenics, and a genomic eat5::eat5 fragment generated by PCR. Several dozens transgenic lines were established on DA837. Because all adult worms on DA837 are transgenics, it was very easy to see the increased reversal frequency phenotype by looking at the population of worms.