Inductive cell interactions determine the developmental fate of certain blastomeres of the C elegans embryo. For example, the interaction of the ABa and EMS blastomeres, or their descendants, is required for the production of pharyngeal cells by descendants of the ABa blastomere (1). We are interested in identifying the molecules that mediate this early inductive cell interaction in nematode embryos. One player has already been identified as the product of the
glp-1 gene, which also is required to mediate an inductive event during the development of the germ line (2, 3). Animals that are homozygous for a
glp-1 mutation (that does not affect the germ line) cannot make viable progeny. The embryos of such animals arrest during embryogenesis with only a partial pharynx - they produce the posterior half of the pharynx which is generated by descendants of the EMS blastomere, but lack the anterior half, normally produced by ABa descendants. Molecular analysis of
glp-1 has revealed that it is a member of the Notch family of transmembrane proteins (4). We have found a candidate for a second gene product required to mediate the EMS-ABa interaction. This gene,
aph-1 (for anterior pharynx defective), is defined by two alleles (
zu123 and
z147 )that result in a matemal-effect embryonic lethality indistinguishable from that of
glp-1 . Embryos from
aph-1 homozygous mothers arrest in development and lack an anterior pharynx. The hypodermal defect (extra lateral hypodermal cells) observed in
glp-1 embryos also is observed in
aph-1 embryos. The
zu147 auele of
aph-1 has a weaker effect than
zu123 ,and allows for more complete morphogenesis and a very low sunrival rate. The striking similarity between the embryonic phenotypes of
aph-1 and
glp-1 mutant embryos suggests that these genes are involved in the same event in embryonic development. However, two lines of evidence lead us to believe that the
aph-1 and
glp-1 gene products have very distinct functions. First, in contrast to
glp-1 ,
aph-1 does not appear to be involved in cell interactions which occur post-embryonically during germ line development. The phenotype of
aph-1 (
zu147)/nDf25is indistinguishable from that of
aph-1 (
zu147)/aph-1 (
zu123),and
aph-1 (
zu123)/nDf25,and shows only the maternal effect lethality described above. We have failed to observe any effect on the germ line, and have found normal brood sizes for
aph-1 (
zu123)or
aph-1 (
zu147 )homozygous animals. Second,
aph-1 does not appear to have a zygotic role in late embryogenesis, as
glp-1 does. Lambie and Kimble have shown that in a
lin-12 background,
glp-1 has an essential zygotic role:
lin-12 ,
glp-1 double mutants die soon after hatching (4). We have constructed the
aph-1 (
zu123);
lin-12 (
ng41 ),doublemutant, and find viable adults. Thus,
aph-1 and
glp-1 may be involved in mediating the same inducdve cell interaction event during early emb~yogenesis, but their specific functions appear to be distinct. We are currently trying to better define the role of
aph-1 in development, and are attempting to clone the
aph-1 gene. By genetic mapping, it is located on the right arm of chromosome I, between
unc-29 and
lin-11 .We have localized the
aph-1 locus within about seven cosmids by RFLP analysis, and are now attempting cosmid-rescue.