The canonical weak
unc-20 mutation,
e112 , causes a characteristic coiling uncoordinated defect, as well as defects in axon outgrowth and pathfinding in several neurons, and increased sensitivity to serotonin. Strong
unc-20 mutations, recovered as part of a smg- dependent screen performed by Phil Anderson's lab, cause developmental arrest in late embryogenesis or at hatching. Given the pleiotropies of
unc-20 mutations, the gene may be involved in a central cellular process, either one specific to the nervous system or a general process. Genetic mapping of
unc-20 places it between
imo-1 and
fax-1 on XL. This region is covered by genomic cosmid clone C15C7, however we have been unable to obtain stable transformants with this clone, preventing us from determining the location of
unc-20 by transformation rescue experiments. Cosmids to the left and right of C15C7 do not rescue
unc-20 and a large deletion just to the right of C15C7 complements
unc-20 . Therefore, we have focused our attention on the predicted genes on cosmid C15C7. We amplified genomic DNA from
unc-20(
e112) and
unc-20(
r977) homozygous animals from regions of C15C7 that correspond to predicted genes C15C7.3, C15C7.4 and C15C7.5; all sequences were the same as wild-type. RNAi experiments with a cDNA clone corresponding to C15C7.5 (
yk64b6; obtained from Y. Kohara) did not generate any obvious visible phenotype. Therefore, we conclude that these three genes are unlikely to encode
unc-20. The two remaining candidates, C15C7.1 and C15C7.2, appear to be transcribed together in an operon. Both predicted genes encode products that are predicted to be localized to trans-Golgi. C15C7.1 encodes a protein that is similar to syntaxin-6, a SNARE that is thought to function in trans - Golgi vesicle trafficking. C15C7.2 encodes a protein that includes a coiled-coil region and a C-terminal GRIP domain. The GRIP domain is associated with vertebrate proteins that are localized to trans-Golgi. One can imagine mechanisms by which defects in vesicle trafficking and/or secretion may result in all known
unc-20 phenotypes. For example, axon pathfinding and extension defects may be the result of difficulties in supplying new membrane to growth cones or the result of compromised transport of receptor molecules. RNAi experiments with a cDNA clone corresponding to C15C7.2 (
yk121h6) causes weak embryonic lethality, but does not cause the Unc phenotype characteristic of
unc-20 putative hypomorphs. We are currently amplifying DNA corresponding to both of these predicted genes from
unc-20 mutants; come to our poster for the outcome. We are also embarking on an electron microscopic study of
unc-20 mutants. We will evaluate global nervous system phenotypes, such as integrity of axon bundles, as well as subcellular Golgi and vesicle anatomy.