Investigating cocaine- and abstinence-induced effects on astrocyte gene expression in the nucleus accumbens

Abstract

In recent years, astrocytes have been increasingly implicated in cellular mechanisms of substance use disorders (SUD). Astrocytes are structurally altered following exposure to drugs of abuse; specifically, astrocytes within the nucleus accumbens (NAc) exhibit significantly decreased surface area, volume, and synaptic colocalization after operant self-administration of cocaine and extinction or protracted abstinence (45 days). However, the mechanisms that elicit these morphological modifications are unknown. The current study aims to elucidate the molecular modifications that lead to observed astrocyte structural changes in rats across cocaine abstinence using astrocyte-specific RiboTag and RNA-seq, as an unbiased, comprehensive approach to identify genes whose transcription or translation change within NAc astrocytes following cocaine self-administration and extended abstinence. Using this method, our data reveal cellular processes including cholesterol biosynthesis that are altered specifically by cocaine self-administration and abstinence, suggesting that astrocyte involvement in these processes is changed in cocaine-abstinent rats. Overall, the results of this study provide insight into astrocyte functional adaptations that occur due to cocaine exposure or during cocaine withdrawal, which may pinpoint further mechanisms that contribute to cocaine-seeking behavior.

Keywords: Abstinence; Astrocyte; Cocaine; Nucleus Accumbens; RNA Sequencing; RiboTag.

Figure 1:. Validation of astrocyte-specific expression of RiboTag (AAV5-GfaABC1D-Rp/22-HA) and RNASeq.

A) Experimental timeline. B) Virally transduced NAc slices stained for anti-HA to detect Rpl22 virus expression within astrocytes; GFAP and NeuN were used to detect astrocytes and neurons, respectively. Rpl22-transduced astrocytes are labeled in white in Merge image (magnification 20X; scale bar: 30 μm). C) Colocalization of GFAP + Rpl22 expression in top panel and colocalization of NeuN + Rpl22 expression in bottom panel (magnification 20X; scale bar: 30 μm). D) Confirmation of Rpl22 AAV specificity in NAc astrocytes of naïve male rats using RNASeq to compare markers of other cell types (neurons, oligodendrocytes, microglia, and endothelial cells).

Figure 2:. Experiment I. Cocaine-administering rats displayed significantly increased drug-seeking behavior than saline controls; differential gene expression detected within experimental group comparisons.

A) Timeline for behavior and immunoprecipitation experiments. B) Active lever presses for both saline and cocaine groups shown across self-administration. C) Number of infusions shown for each treatment group throughout self-administration. (p-value < 0.05). Volcano plots showing number of significant differentially expressed genes (DEGs) in experimental comparisons including cocaine vs. saline at withdrawal day (WD1) (D), cocaine vs. saline at WD45 (E), saline WD45 vs. saline WD1 (F), cocaine WD45 vs. cocaine WD1 (G) (nonsignificant: gray, significant downregulated DEGs: blue, and significantly upregulated DEGs: red). No DEGs were detected between saline and cocaine groups at WD45 (FDR p-value cutoff of < 0.05). (H) Pathway enrichment analysis displaying most significant cellular pathways between saline- and cocaine-administering rats at withdrawal day 1. Downregulated pathways shown in blue and upregulated pathways shown in red (adjusted p-value < 0.05). (I) Venn diagram showing the total number of significant DEGs detected within saline and cocaine groups across time (WD45 vs. WD1).

Figure 3:. Experiment II. Cocaine-administering rats received greater reinforcers than saline controls.

A) Experimental timeline for behavior, immunoprecipitation, and RNASeq. Rats were euthanized in a synchronized manner to align withdrawal days 1 and 45. B) Quantification of active lever presses for saline and cocaine groups across self-administration. C) Infusion counts shown for saline and cocaine groups throughout self-administration. (p < 0.05). Differential gene expression in nucleus accumbens gene expression after saline or cocaine operant self-administration and 1 or 45 day(s) of home cage abstinence. Volcano plots displaying the total number of significant differentially expressed genes (DEGs) in experimental group comparisons including cocaine vs. saline at withdrawal day (WD1) (D), cocaine vs. saline at WD45 (E), saline WD45 vs. saline WD1 (F), cocaine WD45 vs. cocaine WD1 (G) (nonsignificant: gray, significant downregulated DEGs: blue, and significantly upregulated DEGs: red) (FDR p-value cutoff of ≤ 0.05). Pathway enrichment analysis displaying most significant cellular pathways between saline-administering (H) and cocaine-administering (I) rats across abstinence (WD45 vs. WD1). Downregulated pathways are shown in blue and upregulated pathways shown in red (adjusted p-value < 0.05). (J) Venn diagram showing the total number of significant DEGs detected within saline and cocaine groups across time (WD45 vs. WD1).