By contrast, aldicarb pretreatment had no effect on the amplitude of endogenous IPSCs recorded from either wild-type or rig-3 mutant muscles, suggesting that body muscle AUY 922 responses to GABA were unaltered ( Figure S3A). Taken together, these results

suggest that aldicarb enhances body muscle ACh responses in rig-3 mutants (but not in wild-type controls) and that this effect is specific for ACh responses. Increased ACh responses could be caused by altered expression or activity of nicotinic AChRs. C. elegans body muscles express two classes of nicotinic AChRs, homomeric ACR-16 receptors and heteropentameric αβ-type receptors that are sensitive to a synthetic agonist levamisole. Levamisole (Lev) receptors account for only 20% of the synaptic and ACh-activated currents in body muscles ( Francis et al., 2005 and Touroutine et al., 2005). After aldicarb treatment, ACR-16::GFP puncta fluorescence was significantly increased in rig-3 mutants (35%, p < 0.001), while levels in wild-type animals were unaltered ( Figure 4A). By contrast, aldicarb treatment had no effect on UNC-29::GFP Lev receptor fluorescence nor TSA HDAC manufacturer on UNC-49::GFP GABAA receptor fluorescence (consistent with the unaltered IPSC amplitudes) in both wild-type and rig-3 mutants ( Figure S4), indicating

that this effect was specific for ACR-16 receptors. This increase in ACR-16 fluorescence was fully rescued by a transgene expressing RIG-3 in cholinergic neurons ( Figure 4A).

Collectively, these results demonstrate that inactivation of rig-3 reveals an aldicarb-induced potentiation of synaptic transmission, which may result from increased synaptic abundance of ACR-16 receptors. Presynaptic RIG-3 could regulate postsynaptic receptors by either of two general mechanisms. RIG-3 could act in a spatially restricted manner, regulating ACR-16 levels in adjacent postsynaptic membranes. Alternatively, RIG-3 expressed in one neuron could regulate ACR-16 abundance at NMJs formed by neighboring neurons. To distinguish between these possibilities, we examined the effect of RIG-3 expression in the DA motor neurons. DA neurons have cell bodies in the ventral midline, they extend a dendritic process in the ventral cord (which receives synaptic input from interneurons), and an axonal process in the dorsal cord ADAMTS5 (which forms NMJs with dorsal body muscles) (Figure 4B). mCherry-tagged RIG-3 expressed in DA neurons was targeted to puncta in dorsal cord axons whereas little RIG-3 fluorescence was observed in the DA ventral cord processes (Figure 4B), consistent with presynaptic targeting of RIG-3 (Figure 2B). Transgenes expressing RIG-3 in DA neurons rescued the rig-3 ACR-16 fluorescence defect in the dorsal cord, but did not rescue the ACR-16 defect in the ventral cord ( Figures 4C and 4D) nor the rig-3 aldicarb paralysis defect ( Figure S4C).