We have previously reported a form of learning in which worms starved for several hours in the presence of salt such as NaCl (food-/salt+ conditions) learn to avoid the salt, which is otherwise chemoattractive to worms. This dramatic change in the chemotaxis behavior is not induced under the food+/salt+ or food-/salt- conditions (Saeki, Yamamoto and Iino, 2001). Several mutants of
daf-2, which encodes an insulin/IGF-I receptor-like protein, and the
hx546 mutant of
age-1, which encodes a PI-3-kinase acting downstream of DAF-2, shows defects in this form of learning at 25C: they still shows positive chemotaxis after food-/salt+ conditioning. We found that INS-1, one of a multitude of insulin-like peptides predicted in C. elegans, is also involved in the plasticity of chemotaxis. The deletion mutant of
ins-1,
ins-1(
nr2091), shows a defect in this paradigm. This observation is intriguing given the results of Pierce et al. (2001) that this mutant shows no dauer phenotype nor affects life span, while overexpression of
ins-1 causes a weak Daf-c phenotype by itself and enhances the Daf-c phenotype of
daf-2 when combined with the
daf-2 mutations. A genomic clone for the
ins-1 gene rescued the learning defect when introduced into the
ins-1(
nr2091) mutant at a low dose. When introduced at a higher dose, however, the defect was not rescued. We are characterizing the overexpression phenotype in more detail. The
mg305ts mutant of
age-1 also shows a defect in learning at 25C. This learning defect of
age-1(
mg305) was rescued by neuronal expression of
age-1(+) from the pan-neuronal
unc-14 promoter. In contrast, expression of
age-1(+) in the body-wall muscle from the
unc-54 promoter or in the intestine from the
ges-1 promoter was not sufficient to rescue the defect. These results suggest that the insulin-like ligands act on neurons and affects plasticity of chemotaxis as well as dauer formation and longevity, possibly in a different way. We are in the process of further narrowing down the site of
age-1 action for the chemotaxis learning. We thank Cathy Wolkow for the cell-type restricted
age-1 strains, and Gary Ruvkun for the
ins-1(
nr2091) strain.