Worm Breeder's Gazette 11(2): 27
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.
We have begun an analysis of homologous recombination following microinjection of DNA into the gonadal syncytium. Following their introduction into C. elegans. DNA molecules do appear to recombine within shared homologous regions (D. Stinchcomb and J. Shaw, and A. Fire). In order to investigate the frequency of these reactions, we constructed plasmids carrying either 5' or 3' deletions of the rol-6( su1006) semi-dominant right roller gene. When injected independently these partial genes fail to produce roller transformants. However when two deleted rol-6 genes sharing only 600bp of homology within the gene were coinjected at 100 g/ml each, they yielded an average of 20 F1 rollers per injected animal, a frequency approaching that observed for injection of the intact rol-6 gene. When the shared region was reduced to 290bp the efficiency dropped to less than one F1 roller per injected animal. Southern analysis of DNA isolated from transgenic lines indicates that a substantial amount (~5-10%) of the injected DNA has undergone recombination within the shared region. The high frequency of homologous interaction between plasmids coinjected into the syncytial gonad suggests that even much rarer events such as recombination between an injected sequence and its chromosomal homolog might be detected with this technique. Recombination with homologous chromosomal sequences has been observed on a few occasions following injection of DNA into C. elegans oocytes (Boverman et al, wbg vol. 11 #1, and A. Fire personal communication). We have begun looking for such events following syncytial injection of single stranded oligonucleotides. We have chosen single stranded oligos for three reasons: they are not expected to produce a transformed phenotype without interacting with chromosomal DNA, they can be introduced in large molar quantities and finally, they have been shown to be efficient substrates for the gene conversion of yeast and mammalian genes (Moerschell, R.P. et al, PNAS USA 83:5587-5591 and Campbell, C.R., et al, Science, in press). In our initial experiment we synthesized a 50mer containing the single nucleotide change found in rol-6(su1006). This nucleotide change also causes an allele specific restriction site polymorphism eliminating an AatII site. From ~500 animals injected with 0.6mg/ml oligo, we obtained a single F1 roller which proved to be heterozygous for both the roller mutation and the predicted AatII polymorphism. Although the previous experiment required only ~9 hours of injection time, over 100 hours were spent scoring for transformants. We are investigating genetic systems in which a selection for the gene conversion event will reduce time spent staring at worms. Along these lines we have begun injecting a wild type oligo in an effort to correct a recessive ts-lethal mutation in emb-9(g34) animals. Due to a gonadal defect and a semi-dominant phenotype associated with emb-9, we believe that it is necessary to wait until the F2 to impose selection, and so far we have not observed rescue among the progeny of ~200 injected animals. Strategies involving F1 selection are clearly preferable. In an alternative approach we have been coinjecting the rol-6 plasmid along with oligos designed to cause recessive mutations in lin- 29 (in collaboration with Ann Rougvie) or unc-54. The dominant right roller phenotype associated with the plasmid is used to identify F1 animals exposed to the injected DNA. The progeny of rollers are then scored for individuals expressing the recessive mutant phenotype. Using this strategy we have looked at approximately 1500 F1 rollers but have so far failed to detect the recessive mutations predicted by the oligo mediated gene conversion events. At this stage we do not know how many F1 transformants we might need to look at in order to see the desired events, perhaps more than ten thousand. Clearly, while conducting the experiments described above, we would like to be more confident that we have chosen optimal conditions for preparation and injection of the oligo. A major factor controlling the efficiency of gene conversion must be the relative concentration of the oligo to its target site. To maximize this ratio we have simply been using the highest concentration of oligo which does not appear to reduce the brood size of injected animals. (This has generally proven to be not greater than 1mg/ml). In order to shift this ratio further we are experimenting with the coinjection of oligo along with plasmid borne target sequences. Under these conditions it is reasonable to expect a much higher frequency of gene conversion events. Therefore we plan to use coinjection to identify how concentration, length, and method of preparation affect the efficiency of oligos as substrates for gene conversion in worms. Ultimately we hope to develop a reverse genetic approach for C. elegans which utilizes single stranded oligos as site directed mutagens.