In our screens for maternal effect lethal mutants that fail to differentiate intestine, we identified 4 mutants in which the embryos arrest as masses of many differentiated cells, have normal early cleavage patterns, localize P granules normally, but fail to produce gut granules. (WBG 10(2) p. 56, 1988). These mutants may identify genes that are important for specification of the intestinal cells in C. elegans development. We have found that all four of these mutants exhibit a specific alteration in the E cell lineage: Ea and Ep divide prematurely and fail to move to the interior of the embryo to begin the process of gastrulation. The daughter cells remain at the surface of the embryo. The division of Ea and Ep in the mutants occurs 6-12 minutes earlier than in wild type, and immediately follows the division of MSa and MSp. This E lineage-defective phenotype is the earliest defect that we have identified for the mutants. This early phenotype suggests that the absence of differentiated intestinal cells in late stage embryos may result from a defect in the determination of the E lineage. We hope to ascertain the fates of the E daughter cells in the mutants and to learn whether other cell lineages are affected. The mutant embryos produce body wall myosin, indicating that differentiation of body wall muscle cells, at least, is not blocked by these mutations. The four mutations,
it92 V,
it93 I.
it103 II and
it125 II, while giving very similar phenotypes, each identify a different gene. This was somewhat surprising to us, since the mutations were identified from screens of over 23,000 haploid genomes, in which we were able to identify 7 new
par-4 alleles and 6
par-1 alleles. The rare mutations that give an E lineage-defective phenotype may be unusual alleles of their respective genes, or may identify genes that are unusually small or rarely mutated. We are presently mapping the mutations and doing complementation tests with lethals in these regions to distinguish between these possibilities. The fact that we obtained only one mutation in each locus also suggests that we have not saturated the genome for mutations of this kind. In fact, Jocelyn Shaw and coworkers have followed our screening procedure and identified additional mutants with different map locations that give similar E lineage alterations (C. elegans Meeting Abstracts, p. 231,1989). One mutation is allelic with
it92 (J. Shaw pers. communication). We are collaborating with the Shaw lab on further genetic and phenotypic analyses of E lineage-defective mutations. We hope that our studies of these mutations will provide insights into how the embryonic intestinal cell lineage pattern and determination of cell type is controlled.