The C. elegans Synaptic Signaling Network is composed of at least three interacting G a signaling pathways: G a o (GOA-1), G a q (EGL-30), and G a s (GSA-1). Genetic epistasis studies of the G a q and G a s pathways suggest that these two major pathways converge downstream of the messenger molecule diacylglycerol (DAG), which is produced by the G a q pathway. 1 Knocking out either the G a q or the G a s pathway results in strong paralysis that is the result of functional (non-developmental) neuronal defects. Curiously, however, only the G a q pathway, which is the core, obligatory synaptic vesicle priming pathway, appears required for neurotransmitter release: knocking out the neuronal G a s pathway, which is known to be required for learning and memory in other organisms, does not significantly alter overall levels of neurotransmitter release. 1 Our observations of neuronal G a s pathway nulls in different genetic backgrounds, combined with our finding that the G a q and G a s pathways converge, led us to hypothesize that the neuronal G a s pathway may transduce positional information onto the downstream part of the G a q pathway to activate optimal synapses for specific behaviors. To investigate the source(s) of this hypothesized positional information, we sought mutants with phenotypes similar to mutants that lack a neuronal G a s pathway. Our search led us to UNC-31, the C. elegans homolog of CAPS, which is required for the fusion of one or more classes of secretory vesicles. Like G a s pathway nulls, unc-31
mutants are strongly paralyzed, and yet they exhibit only slightly reduced levels of neurotransmitter release. Using a heat-shock inducible promoter, we found that the paralysis of unc-31
nulls is the result of functional (non-developmental) defects and that expression of an unc-31
cDNA in the ventral cord cholinergic neurons of an unc-31
null restores locomotion rate to wild type levels. To investigate whether UNC-31 provides input for the G a q pathway, the G a s pathway, or both, we undertook an extensive genetic epistasis analysis, in which we analyzed double mutants containing an unc-31
null mutation in combination with gain- and loss-of-function mutations in the G a q and G a s pathways. Our results suggest that the major function of UNC-31 (CAPS) - mediated secretion is to activate the neuronal G a s pathway. We propose that one or more classes of molecules released by UNC-31 activates the neuronal G a s pathway, which in turn acts on downstream components of the core G a q priming pathway to activate optimal synapses for locomotion. 1 Reynolds, N.K., Schade, M. A., and K. G. Miller, submitted.