The C. elegans hermaphrodite gonad is formed by migration of the two distal tip cells (DTC). These cells migrate away from the gonad primordium on the ventral side of the animal (one anterior and the other posterior), turn dorsally and migrate back toward the center of the animal on the dorsal surface resulting in two U-shaped gonad arms of proliferating germ cells (Kimble and White, 1981). We are interested in genes that affect this migratory path and performed a genome-wide RNA interference (RNAi) screen that revealed 99 genes required for DTC migration (Cram et al., 2006). In this screen, the RNAi-sensitized strain rrf-3(pk1426) was subjected to RNAi by feeding. In this approach, all tissues have the potential to be affected by RNAi; therefore DTC migration might be affected by RNAi knockdown in tissues other than the gonad. To overcome this, we took advantage of the lag-2p::GFP; rde-1(ne219); lag-2p::rde-1 strain of C. elegans (generated by Dr. Dana Byrd and Dr. Judith Kimble) that allows RNAi to have an effect mainly in the two DTCs.
We rescreened the 99 genes required for DTC migration with the lag-2p::GFP; rde-1(ne219); lag-2p::rde-1 strain to identify genes that act cell autonomously. We applied the same two step microscopy approach as Cram et al. (2006), first looking under light microscopy with a dissecting microscope and scoring for worms that show clear patches (intestinal displacement due to gonad abnormal migration or distension). In the second step, we characterized the migratory paths using DIC microscopy and put them into categories (types 1-3) based on gonad morphology. P0 and F1 generations were scored in both screens. The use of the rrf-3(pk1426) sensitized strain resulted in a higher percentage of animals with clear patches in both generations in the primary screen compared to the lag-2p::GFP; rde-1(ne219); lag-2p::rde-1 strain. Of those adults with clear patches that were analyzed in the secondary screen, the proportions of the three types of DTC migration defects were similar between the two strains. Therefore, whole animal and DTC-specific RNAi give similar DTC migration phenotypes but with different penetrance.
Knockdown of 28 of the 99 genes caused DTC migratory defects in 30% or more of the hermaphrodites. Another 31 genes affected DTC migration in 5-29% of adults. Thus, we conclude that these 59 genes have cell autonomous activities during DTC migration. No significant effect on DTC migration was observed in the remaining 40 genes. However, we cannot conclude that these do not act cell autonomously because lack of a defect could result from the less efficient knockdown in the DTC-specific strain. RNAi data are available at www.cellmigration.org/resource/discovery/schwarzbauer/schwarzbauer_dtc_auto.cgi.
References
Cram EJ, Shang H and Schwarzbauer JE. (2006). A systematic RNA interference screen reveals a cell migration gene network in C. elegans. J. Cell Sci. 119, 4811-4818. 
Kimble JE and White JG. (1981). On the control of germ cell development in Caenorhabditis elegans. Dev. Biol. 81, 208-219.
