Specification of pharyngeal and rectal precursors requires
pha-4, a member of the winged-helix transcription factor family (1). This family of proteins includes the developmental regulators fork head in Drosophila and HNF-3 in mammals. Whereas the winged-helix DNA binding domain has been characterized extensively, relatively little is known about the roles of other parts of these proteins in transcriptional regulation. To elucidate the functions of regions outside the DNA binding domain, we have begun characterizing five different
pha-4 alleles. The PHA-4 protein contains two additional regions outside the DNA binding domain: an amino terminal region that has sequence similarity with FORK HEAD protein and a carboxyl terminal region that has sequence similarity with HNF-3 and Ce-DISTAL-LESS. Our characterization of five
pha-4 alleles indicates that the amino terminus is critical for
pha-4 function whereas the carboxyl terminus is dispensible for viability. A single missense mutation in the amino terminus results in a leucine to phenylalanine substitution that renders PHA-4 largely inactive. Antibody staining indicates that the protein is expressed normally in mutant embryos. Therefore, it is likely that this mutation marks a region within PHA-4 that is critical for activity. Analysis of phenotypes associated with four nonsense alleles indicates that the PHA-4 DNA binding domain, but not the carboxyl terminus is essential for viability. These alleles are predicted to generate successive truncations from the carboxyl terminus through the DNA binding domain. To analyze the phenotype associated with each of these alleles, we constructed
pha-4;
smg-3 double mutants. Mutations in
smg-3 disrupt the mRNA surveillance system that recognizes and degrades aberrant mRNAs, allowing accumulation of nonsense-bearing mRNAs and their protein products (2). We found that removing any portion of the DNA binding domain blocked the ability of
pha-4 to promote pharyngeal development. Surprisingly, embryos carrying PHA-4 proteins that lacked the entire carboxyl terminus (but retained the DNA binding domain) were viable in combination with an allele of
smg-3. These worms can be maintained as a homozygous strain although they are not as healthy as wild type worms. We are currently examining mutant worms for the presence of all pharyngeal cells. In summary, we have shown that the PHA-4 DNA binding domain but not the carboxyl terminus is essential for pharynx organogenesis. Furthermore, a point mutation delineates a site in the amino terminus required for PHA-4 activity. Our current goal is to identify protein(s) that interact with the amino and carboxyl termini of PHA-4. We are using two approaches to identify these factors: 1) a genetic screen for
pha-4 suppressors and 2) a yeast two-hybrid screen using the PHA-4 amino or carboxyl regions as bait. Thanks to Jim Priess for one of the
pha-4 alleles and Bob Barstead for the two-hybrid library. 1. Mango, S.E., Lambie, E.J. & Kimble, J. Development 120, 3019- 3031 (1994); Kalb, J.M., et al. Development 125, 2171-2180 (1997); Horner, M., et al. in press . 2. Hodgkin, J., Papp, A., Pulak, R., Ambros, V. & Anderson, P. Genetics 123, 301-313 (1989); Pulak, R. & Anderson, R.P. Genes Dev. 7, 1885-1897 (1993).