A screen for muhtions that disrupt CAN cell migrations. Wayne Forrester and Gian Garriga. Department of Molecular and Cell Biology, University of California, Berkeley, CA 94720. The canal associated neurons (CANs), a pair of bilaterally symmetric neurons in C. elegans, are born in the head and migrate to the middle of the animal during embryogenesis. The CANs are bipolar neurons, extending one anterior and one posterior axon. As a first step toward understanding the control of CAN migrations, we have taken a genetic approach to identify genes required for CAN migration. Two phenotypes are associated with defects in CAN cells. Laser ablation of both CANs during the first larval stage causes a distinctive clear (Clr) phenotype, followed by death (1). Clr animals appear unusually transparent and have thin, dark intestines. Mutants in which the CANs fail to migrate to their proper position develop a withered tail (Wit) phenotype (2). To identify genes required for CAN migration or function, we screened for Clr or Wit animals among F2 progeny from approximately 1000 individual F1 animals of mutagenized parents. We identified 31 mutants with Wit-like phenotypes and 10 with Clr phenotypes. The Clr phenotype resembled either the phenotype of clr-l mutants (2 mutations) or the rod-like larval lethal phenotype of Ras pathway mutants (8 mutations). To examine CAN morphology and position directly, an integrated array containing
ceh-73 sequences fused to
unc-76 and the jellyfish green fluorescent protein (GFP) was used(3, 4). This reporter, which produced GFP in the CANs, as well as several cells in the head and tail, was crossed into mutants identified in the screen. In most mutants, the CANs appeared normal. The remaining mutants fell into three classes. Class 1 (2 mutants-both appear to be alleles of clr-l ) mutants appear to have CAN pathfinding defects; the posterior axon does not always follow its normal pathway. Class 2 (2 mutants-
gm58 and
gm71 ) mutants appear defective in CAN differentiation; CANs are not detected in these rod-like animals.
gm58 and
gm71 may be allelic; they have identical phenotypes and map between lon-l and
unc-32. Class 3 (1
epi-1 mutants and l
unc-73 mutant) mutants had CAN cell migration and axon pathfinding defects; CAN cell bodies are misplaced anteriorly and the axons do not follow their normal pathways. The screen described here was successful in identifying mutations that affect CAN migrations. However, a large number of mutations that apparently did not affect CANs also were identified. In addition, mutations affecting the CANs that did not cause Wit or Clr phenotypes were missed in this screen. For these reasons, we are now screening directly for misplaced CANs using the GFP reporter to visualize CANs. References: 1. J. E. Sulston, J. A. Hodgkin, WBG 5#1, 19; E. Jorgenson, (pers comm); our unpublished observation. 2. J. Manser, W. B. Wood, Dev Genet 11, 49*4 (1990). 3. Generously provided by J. Zallen and C. Bargmann. 4. M. Chalfie, Y. Tu, G. Euskirchen, W. W. Ward, D. C. Prasher, Science 263, 802-805 (1994).