Worm Breeder's Gazette 9(2): 88
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.
In earlier work (Herman, 1984), I showed that an unc-3 phenotype is produced when cells derived from AB.p lack the unc-3(+) gene and that a semi-unc-3 phenotype is produced when cells derived from either AB. pl or AB.pr lack unc-3(+). I suggested, on the basis of this evidence, that the prime candidates for the cell-specific action of the unc-3 gene were the dorsal and ventral cord motor neurons (or some subset of these neurons), since all but one of the 75 cord motor neurons of the adult hermaphrodite descend, in about equal measures, from AB.pl and AB.pr. At the last C. elegans Meeting, John White reported on the neuroanatomy of unc-3. He reported that the anatomy of the nerve ring of unc-3 animals seemed to be normal; this result seems consistent with the phenotype of unc-3 animals, because although they are unable to propagate the dorsoventral bends along the body necessary for movement, they seem able to move their heads normally and the motor neurons driving head movement make their synapses in the ring. In addition, I think it is of interest that many of these motor neurons descend from AB.a rather than AB.p. John also reported that the neuroanatomy of the cord was so deranged that it was difficult to distinguish specific neurons and he suggested that perhaps the focus of action of unc-3(+) might be in the surrounding hypodermis rather than the neurons themselves. I've been testing this suggestion by identifying semi-unc-3 animals derived from unc-3 d unc-3 sup-10 gotes. Ten of the 22 embryonically-generated cord motor neurons descend from AB.plp and 11 descend from AB.prp (one descends from AB.a). The post-embryonically derived cord motor neurons descend from AB.pla and AB.pra. Virtually all of the relevant hypodermal nuclei, however, descend from AB.pla and AB.pra only; the only hypodermal nuclei descending from AB.plp and AB.prp are situated in positions posterior to the anus. The question I have been asking, therefore, is whether loss of unc-3(+) by AB.plp or AB.prp results in a semi-unc-3 phenotype. I am working on the assumption that osm-1 behaves cell autonomously with respect to its effect of FITC staining of certain amphid and phasmid sensory neurons. Thus, loss of osm-1(+) on mnDp7 from the unc- 3 gote at AB.plp is signified by lack of staining by PHAL (which descends from AB.plpp) and ASHL (which descends from AB.plpa); similarly, loss at AB.prp is signified by lack of staining by PHAR (which descends from AB.prpp) and ASHR (which descends from AB.prpa). Although I'm not finished with these experiments, a significant fraction of the semi-unc-3 animals show precisely these patterns of non-staining (with all other neurons staining) as if duplication loss at AB.plp or AB.prp does indeed result in the semi-unc-3 phenotype; it therefore seems unlikely that the focus of action of unc-3 is in (or at least limited to) the hypodermis. I note that mnDp7 carries Marty Chalfie's dominant mutation mec-4(e1611), which promotes death of the touch cells; duplication loss at either AB.plp or AB.prp does not rescue the posterior touch cells, PLML and PLMR, which descend from AB.pla and AB. pra, respectively. Finally, I've been using the unc-3 sup-20 imals because duplication loss at AB.plp or AB.prp leads to quite different consequences for expected FITC staining patterns depending on whether the focus of action of daf-6 is sockets, sheaths or both. The results should also confirm expectations for the cell-specific actions of unc- 3 and mec-4(e1611). (The presence of sup-10 in this strain is simply a consequence of how the strain was constructed and is irrelevant to the mosaic analysis.) The results, which are still very fragmentary, so far support the notion that daf-6 action is limited to sheath cells.