Neural coding of reward and aversion: Functional organization of ventral tegmental area neuronal ensembles

This thesis investigates how distinct neuronal ensembles within the ventral tegmental area (VTA) encode rewarding and aversive experiences, offering insights into emotional processing and its relevance to psychiatric disorders. Grounded in foundational emotion theories, this work focuses on "valence", the hedonic value assigned to stimuli that guides approach or avoidance behavior, and explores how the brain translates this valuation into action. Traditionally, VTA neurons are categorized by neurotransmitter type (dopamine, GABA, glutamate), but recent findings suggest that ensembles (i.e. groups of co-active neurons) provide a more accurate representation of emotional states. This work adopts an ensemble-based framework to dissect VTA function. Using activity-dependent tagging techniques, such as TRAP2, FLiCRE, and tetOFF, this thesis captures and characterizes neuronal populations engaged during rewarding (DAMGO-induced) and aversive (social stress) experiences. These methods enabled the labeling of two distinct VTA ensembles: VTADAMGO, responsive to reward, and VTASTRESS, responsive to stress. Key findings include: • Identification of VTASTRESS ensemble: Approximately 11% of VTA neurons were activated by social stress, with greater intrinsic excitability than non-tagged neurons. These included a mix of dopaminergic, glutamatergic, and GABAergic cells. • Non-overlapping but intermingled ensembles: VTADAMGO and VTASTRESS ensembles were spatially co-located yet largely distinct, showing valence-specific responses in behavioral assays. • Behavioral relevance: Activation of VTADAMGO promoted approach and anxiolytic behaviors, while VTASTRESS drove avoidance and anxiety-like behaviors. Inhibition of these ensembles impaired the formation of valence-based associative learning. • Functional consistency across neurotransmitter types: Both ensembles maintained their valence-related functions regardless of neurotransmitter identity, underscoring the importance of ensemble organization over classical cell-type distinctions. • Circuit connectivity: Both ensembles projected to overlapping regions (e.g., nucleus accumbens, ventral pallidum), with the lateral habenula preferentially targeted by VTASTRESS. The findings support the neuronal ensemble model as a more dynamic and nuanced framework for understanding VTA-mediated emotional processing. Clinically, this has implications for disorders like depression and anxiety, where valence processing is disrupted. For example, hypoactivity in reward-related ensembles may underlie anhedonia, while hyperactivity in stress-related ensembles may contribute to heightened anxiety. Future research should explore the molecular identity, plasticity, and sex-specific differences in these ensembles. Integrating single-cell sequencing, imaging, and translational approaches may inform circuit-targeted therapies. In conclusion, this thesis reveals that the VTA contains distinct but interactive neuronal ensembles that encode reward and aversion, shaping emotional behavior through coordinated circuit activity. These insights advance our understanding of the brain's emotional architecture and offer a foundation for novel treatments targeting dysfunctional valence processing.