The reproductive/dauer developmental decision is controlled by environmental cues, such as food, pheromone and temperature. A TGF-beta-related pathway promotes reproductive over dauer development. Mutations in
daf-7 (a TGF-beta-like ligand),
daf-1 and
daf-4 (type I and type II receptor kinases), and
daf-8 and
daf-14 (Smad transcription factors) induce dauer at restrictive temperature regardless of environmental cues. The Daf-c phenotype of these mutants is suppressed by mutations in
daf-3 (a Smad) or
daf-5 . Epistasis analysis suggests that
daf-5 and
daf-3 are either antagonized by this TGF-beta-related pathway or acting in a parallel pathway. Our cloning of
daf-5 showed that
daf-5 encodes a novel protein with weak similarity to chromatin remodeling proteins such as Sno/Ski and ACF-1. Vertebrate Sno has been recently shown to be capable of acting as an inhibitor of TGF-beta pathways by interacting with Smad proteins. However, because of the low similarity, the functional relationship of DAF-5 to Sno/Ski is uncertain. We searched for a better Sno/Ski homolog in C. elegans and found nothing; however, we found convincing Sno/Ski homologs in the nematodes Brugia malayi and Globodera rostochiensis , and in Drosophila . Thus, either one group of nematodes has lost Sno/Ski, or
daf-5 is a Sno/Ski that is so highly diverged as to be almost unrecognizable. The genetic function of
daf-5 in dauer formation is different from that suggested for Sno/Ski by the vertebrate tissue culture studies in that
daf-5 is not simply acting to block signaling by the receptors. We sequenced twenty
daf-5 alleles and identified four missense mutations, one inframe deletion, seven nonsense mutations, one frame-shift, and two splice site mutations. Three alleles had no mutation in the coding region. Five out of fifteen alleles affected amino acid residues 158-173 (of a 627 amino acid residue protein). This hotspot is in a region that shows no significant homology to other proteins. We examined
daf-5 expression with
daf-5 :: GFP constructs. One contains 6.5Kb of upstream sequence and an intact
daf-5 coding region, which has GFP inserted in the first exon. This construct rescued
daf-5 . The other construct has GFP fused to the C-terminal end of the first exon of
daf-5 with a longer (13Kb) upstream region. The rescuing construct shows a relatively strong expression in ganglia in the head and tail and in the anterior pharynx. A small number of animals show weak expressions in the hypodermis, muscles, intestine, and distal tip cells. The non-rescuing GFP construct is more strongly expressed, and shows more consistent expression in the hypodermis, muscles, intestine, and distal tip cells; still, its expression is strongest in the head and tail ganglia. This observation is consistent with expression and functional data from the Riddle and Thomas labs that indicate that the TGF-beta receptors function in the nervous system. We are testing whether daf- 5 functions in the nervous system by expressing
daf-5 with nervous-system-specific and other promoters to see where
daf-5 must be expressed to rescue a
daf-5 mutant.
daf-5 and
daf-3 have been shown to interact with each other in the yeast-two-hybrid system (see abstract by Hu, Tewari, Ruvkun and Vidal). We are going to perform immuno-precipitation from extracts of C. elegans to see if
daf-5 and
daf-3 physically interact during different developmental stages and to determine if the interaction is controlled by TGF-beta signaling.