Distinct features of dopaminergic neuromodulation of motivation-dependent behavior in Drosophila
1 Institute of Physiology II, Faculty of Medicine, University of Bonn, Bonn, Germany
Animals exhibit remarkable flexibility in their behavior. Striving in a competitive environment demands optimum sensory and cognitive fitness that must be concatenated with motivation and internal state to generate ethologically appropriate behavior. To capture the underlying neuronal mechanism of state-dependent goal-directed behavior, we study olfactory foraging behavior in the food-deprived fly. Previously, using a single fly treadmill assay, our laboratory found that dopamine, a monoamine neuromodulator, plays a critical role in persistent state-dependent goal-directed behavior. Dopamine has been extensively studied pertaining its role in reinforcement learning, reward, motivation, and voluntary movement control which appears conserved across the species. However, the extent to which these distinct facets of information are multiplexed within the dopaminergic circuits and whether diverse subpopulations of dopaminergic neurons simultaneously or individually transmit this information to facilitate adjustable behavior remains unresolved. Our recent experiments using acute and temporally controlled perturbation of neuronal activity with optogenetics indicate that different dopaminergic neuron populations play distinct roles during goal-directed behavior in the fly brain. First, our experiments show that optogenetic silencing of a specific subpopulation of dopaminergic neurons located in the Protocerebral Posterior Lateral (PPL) impairs persistent odor tracking independent of the cue-expected outcome contingency as both constitutive silencing and short-term silencing at different timepoints during trial and inter-trial periods impairs tracking. Using specific driver lines, we found that a single PPL1 dopaminergic neuron, projecting to gamma1 compartment of mushroom body (PPL1-01) was responsible. Based on these data, we suggest that PPL1-01 accumulates state-dependent information (e.g. hunger, trial progression and number) throughout ongoing behavior independent of cue on- and offset. By contrast, optogenetic manipulation of the other cluster of dopaminergic neurons, Protocerebral Anterior Medial (PAM), affected odor tracking behavior exclusively when perturbation was temporally restricted to the odor offset but not during intertrial periods suggesting a role of PAM neurons in encoding temporal cue-expected outcome relationships. Moreover, inhibition of some PPL1 neurons, but interestingly not of the PPL1-01 or PAM neurons, acutely increased the ratio between turns and forward walk indicating a role some PPL1 neurons in movement modulation. Taken together, we propose a model wherein the PPL1-01 neuron accumulates state-dependent information over time, while PAM neurons encode the temporal relationship between cue and outcome. We are currently testing our model through in vivo imaging of dopaminergic neurons in behaving animals. Moreover, we are using the EM connectome to characterize the neural network relationship and possibly computation between PPL1 and PAM neurons in state-dependent goal-directed behavior.