Projection-defined amygdala neurons encode emotional memory through distinct transcriptomic programs

The amygdala is crucial for emotional memory, but the molecular programs that enable its distinct circuits to encode different memory features remain unclear. Combining single-cell RNA sequencing with retrograde tracing, we investigated transcriptomic dynamics of amygdala neurons projecting to nucleus accumbens (NAc) or auditory/temporal association cortex (AuC/TeA) during formation and long-term retention of fear and reward memories. We found that memory engages pathway-specific transcriptional programs of synaptic genes determined by valence, temporal stage, and projection target. Notably, the AMY?AuC/TeA pathway, previously considered a fear circuit, showed robust modulation during reward memory retention, and chemogenetic silencing confirmed its essential role in reward memory. Projection-defined neurons are transcriptionally and functionally heterogeneous: some subpopulations encode a single feature of emotional memory, such as an AMY?NAc subpopulation for reward memory formation and retention, and an AMY?AuC/TeA subpopulation for retention of fear and reward memories. Most subpopulations, however, encode multiple features via largely non-overlapping gene programs. We further identified Dcn as a selective marker for an AMY?NAc subpopulation preferentially recruited during fear memory formation. Our findings reveal a principle of projection-, valence-, and time-dependent transcriptional programming, demonstrate how subpopulations multiplex memory features via dynamic gene networks, and provide a comprehensive resource for dissecting amygdala function. Overall design: We utilized stereotaxic viral injections to retrogradely label Amygdala (AMY) neurons projecting to the Nucleus Accumbens (AMY?NAc) and the Auditory Cortex/Temporal Association Cortex (AMY?AuC/TeA). Labeled mice were then trained using fear and reward conditioning paradigms (CS-US pairing), and the amygdala was dissected for scRNA-seq analysis at 2 hours or 5 days post-conditioning, representing the phases of memory formation and retention, respectively. A control group was subjected only to the conditioned stimulus (CS, sound only, no US) and sampled 2 hours post-exposure.