Analysis of whole genome sequencing and plasma metabolomics unveil genetic determinants and clinical implications for human health.

The birth of a new memory involves the synchronous engagement of a heterogeneous population of neurons, promoting molecular changes that strengthen the connectivity of a neuronal ensemble in the brain. These neurons can be excitatory or inhibitory, depending on the specialized molecular signatures and electrophysiological properties they carry. The molecular composition of a neuron enables it to quickly respond to converging afferent stimuli and average these signals into one final output and may be fine-tuned in response to salient activity. However, there is a lack of understanding regarding the molecular dynamics occurring on short timescales in different neurons in vivo, especially in the context of memory formation. Most studies interrogate the molecular profile of neurons at long time points (e.g., days to weeks) after memory acquisition, when memory formation has already completed, but do not examine how the process starts and develops. Furthermore, these studies rely on measurements of the transcriptome at the single-cell level, which generates sparse data enriched in highly-expressed transcripts, and transcript levels do not automatically correspond to protein abundance. On the other hand, studying the dynamic changes in the proteome is challenging due to low technical sensitivity, especially when studying small neuronal subpopulations. At the intersection between transcriptome and proteome, the translatome, or the full set of translating mRNAs, represents an alternative approach to dissect functional changes occurring in neurons during consolidation. The translatome is especially attractive given that long-term memory consolidation is characterized by transient enhancement of translation, whose disruption modifies memory formation. This reinforces translational mechanisms as potential therapeutic targets for memory-related disorders. The translatome at early stages of memory consolidation has been examined previously, but the lack of cell type specificity of these studies precludes the determination of differential mRNA programs triggered in discrete neuron populations. Furthermore, without cell type specificity, glial cell contamination serves as a significant confound. On the other hand, when the translatome was studied in a neuron type-specific manner, a weaker threat memory paradigm was utilized, resulting in smaller molecular changes. Given the critical role of de novo translation during memory consolidation, there is an imminent need to identify the neuron type-specific molecular changes that underlie stable memory formation. Here, we leverage translating ribosome affinity purification (TRAP) combined with RNA-sequencing to examine threat conditioning-induced alterations in the translatome of three important neuron types of the dorsal hippocampus (dHPC) for long-term memory consolidation: excitatory neurons (), and inhibitory neurons ( or ). We find that these neurons rapidly recruit specialized mRNAs to ribosomes (henceforth referred to as ibosome-ssociated mRNAs - RA-mRNAs). We further demonstrate a role for the translation initiation modulator GADD34 in translatome dynamics and long-term memory. Finally, we find that modifications of the translatome during consolidation are a function of elements embedded in the mRNA sequence, particularly AU-rich elements (AREs), and disrupting a candidate ARE-targeting RNA binding protein leads to memory deficits. Altogether, this work provides the scientific community with a new resource and extensive hypothesis-generating database that uses cellular and systems-level analyses aiming to unravel the molecular basis of long-term memory consolidation.