In Drosophila, anatomically discrete dopamine neurons that innervate distinct zones of the mushroom body (MB) assign opposing valence to odors during olfactory learning. Subsets of MB neurons have temporally unique roles in memory processing, but valence-related organization has not been demonstrated. We functionally subdivided the alphabeta neurons, revealing a value-specific role for the approximately 160 alphabeta core (alphabetac) neurons. Blocking neurotransmission from alphabeta surface (alphabetas) neurons revealed a requirement during retrieval of aversive and appetitive memory, whereas blocking alphabetac only impaired appetitive memory. The alphabetac were also required to express memory in a differential aversive paradigm demonstrating a role in relative valuation and approach behavior. Strikingly, both reinforcing dopamine neurons and efferent pathways differentially innervate alphabetac and alphabetas in the MB lobes. We propose that conditioned approach requires pooling synaptic outputs from across the alphabeta ensemble but only from the alphabetas for conditioned aversion.

BACKGROUND: Translation in axons is required for growth cone chemotropic responses to many guidance cues. Although locally synthesized proteins are beginning to be identified, how specific mRNAs are selected for translation remains unclear. Control of poly(A) tail length by cytoplasmic polyadenylation element (CPE) binding protein 1 (CPEB1) is a conserved mechanism for mRNA-specific translational regulation that could be involved in regulating translation in axons. RESULTS: We show that cytoplasmic polyadenylation is required in Xenopus retinal ganglion cell (RGC) growth cones for translation-dependent, but not translation-independent, chemotropic responses in vitro, and that inhibition of CPE binding through dominant-negative interference severely reduces axon outgrowth in vivo. CPEB1 mRNA transcripts are present at low levels in RGCs but, surprisingly, CPEB1 protein was not detected in eye or brain tissue, and CPEB1 loss-of-function does not affect chemotropic responses or pathfinding in vivo. UV cross-linking experiments suggest that CPE-binding proteins other than CPEB1 in the retina regulate retinal axon development. CONCLUSION: These results indicate that cytoplasmic polyadenylation and CPE-mediated translational regulation are involved in retinal axon development, but that CPEB1 may not be the key regulator of polyadenylation in the developing retina.