The amygdala plays a crucial role in processing emotions, motivation, and memory, and its dysfunction is implicated in various neuropsychiatric disorders, including addiction. Repeated drug use strengthens drug-associated memories and reinforces drug-seeking behavior within the amygdala. During withdrawal, the amygdala contributes to negative emotional states, increasing the drug's incentive value. While the amygdala's behavioral functions and connectivity are well-established, the specific roles of its diverse neuronal and non-neuronal cell populations in addiction remain unclear. Previous single-cell studies, mostly using inbred strains and focusing on acute drug effects, lacked the resolution to fully capture the complex molecular changes associated with long-lasting addiction. This study aimed to address these limitations by using snRNA-seq and snATAC-seq in outbred heterogeneous stock (HS) rats, known for their high genetic variation and phenotypic diversity, after prolonged cocaine exposure and abstinence.

Existing research highlights the amygdala's involvement in addiction, with repeated drug use leading to the formation of drug-associated memories and the reinforcement of drug-seeking behavior. Withdrawal symptoms mediated by the amygdala further enhance drug-seeking. However, the specific contributions of different neuronal and glial cell types within the amygdala to addiction remain poorly understood. While single-cell technologies have advanced our understanding of brain cell diversity, their application to addiction research, particularly using outbred models and focusing on long-term effects, has been limited. This study aimed to build upon previous research by employing single-nucleus omics approaches in a genetically diverse rat model to investigate the molecular underpinnings of addiction-like behaviors after extended cocaine exposure and a period of abstinence.

The study utilized outbred HS rats, known for their high genetic diversity and phenotypic variability, to model cocaine addiction-like behaviors. Rats underwent extended-access intravenous self-administration (IVSA) of cocaine, followed by a 4-week abstinence period. An addiction index (AI) was calculated for each rat based on three measures: escalation of cocaine intake, motivation (progressive ratio test), and compulsive-like behavior (drug taking despite footshock). Rats were then classified into high AI and low AI groups. Single-nucleus RNA sequencing (snRNA-seq) and single-nucleus ATAC sequencing (snATAC-seq) were performed on amygdala tissue from these rats, along with naive controls. Data analysis involved cell type identification and annotation using marker genes, differential gene expression analysis (negative binomial test), gene set enrichment analysis (GSEA), and differential chromatin accessibility analysis (MACS2 and chromVAR). Electrophysiology experiments using CeA slices were conducted to measure GABAergic transmission before and after treatment with pBBG, a glyoxalase 1 inhibitor. Cue-induced cocaine-seeking behavior was assessed to evaluate the effects of pBBG on relapse-like behaviors. Finally, partitioned heritability analysis was conducted to assess the relevance of the rat data to human addiction traits.

The study identified several key findings: 1. Outbred rats displayed clear individual differences in cocaine addiction-like traits, with high AI rats exhibiting escalation of intake, heightened motivation, and compulsive-like behavior. 2. snRNA-seq and snATAC-seq revealed a comprehensive atlas of amygdala cell types in both normal and cocaine-addicted states. 3. Differential gene expression analysis revealed significant changes in energy metabolism pathways across multiple cell types in high AI rats, particularly enrichment of oxidative phosphorylation and glycolysis genes. 4. High AI rats showed increased GABAergic transmission in the central amygdala (CeA), which was reversed by pharmacological inhibition of glyoxalase 1 (GLO1) using pBBG. 5. This GLO1 inhibition also reduced relapse-like behaviors in high AI rats. 6. snATAC-seq analysis showed cell-type-specific differences in chromatin accessibility between high and low AI rats, with excitatory neurons showing increased accessibility in high AI rats and inhibitory neurons showing increased accessibility in low AI rats. 7. Chromatin accessibility differences were associated with differences in the activity of several transcription factors (TFs) involved in addiction. 8. Linkage disequilibrium score regression analysis showed that the regulatory elements identified in rats are relevant to human addiction-related traits.

This study provides a comprehensive single-cell atlas of the rat amygdala under normal and cocaine addiction-like conditions. The findings highlight the importance of cellular energetics and GABAergic signaling in the long-term effects of cocaine exposure. The reversal of addiction-like behaviors and altered GABAergic transmission by GLO1 inhibition suggests a novel therapeutic target for cocaine addiction. The observation that both genetic factors and cocaine exposure likely contribute to the observed molecular differences warrants further investigation using larger datasets and more sophisticated methods. The cell-type specificity of these findings emphasizes the intricate molecular mechanisms underlying addiction vulnerability and resilience.

This study provides a valuable resource for understanding cocaine addiction. The findings highlight the role of energy metabolism, GABAergic signaling, and cell-type-specific chromatin accessibility in addiction-like behaviors and identify GLO1 as a potential therapeutic target. Future research should focus on dissecting the interplay between genetics and cocaine exposure, investigating the specific roles of identified TFs, and further characterizing the cell-type-specific effects of GLO1 inhibition.

The study primarily used male rats, potentially limiting the generalizability of findings to females. The predictive models for gene expression based on genetic variation had limitations due to the use of whole-brain data and a relatively small training cohort. The pharmacological inhibition experiments were not cell-type specific, limiting mechanistic conclusions. Further research with larger sample sizes and more refined methodologies is needed to address these limitations.