Regulation of Brain DNA Methylation Factors and of the Orexinergic System by Cocaine and Food Self-Administration.

Inhibitors of DNA methylation and orexin type-1 receptor antagonists modulate the neurobiological effects driving drugs of abuse and natural reinforcers by activating common brain structures of the mesolimbic reward system. In this study, we applied a self-administration paradigm to assess the involvement of factors regulating DNA methylation processes and satiety or appetite signals. These factors include Dnmts and Tets, miR-212/132, orexins, and orx-R1 genes. The study focused on dopamine projection areas such as the prefrontal cortex (PFCx) and caudate putamen (CPu) and in the hypothalamus (HP) that is interconnected with the reward system. Striking changes were observed in response to both reinforcers, but differed depending on contingent and non-contingent delivery. Expression also differed in the PFCx and the CPu. Cocaine and food induced opposite effects on Dnmt3a expression in both brain structures, whereas they repressed both miRs to a different extent, without affecting their primary transcript in the CPu. Unexpectedly, orexin mRNAs were found in the CPu, suggesting a transport from their transcription site in the HP. The orexin receptor1 gene was found to be induced by cocaine in the PFCx, consistent with a regulation by DNA methylation. Global levels of 5-methylcytosines in the PFCx were not significantly altered by cocaine, suggesting that it is rather their distribution that contributes to long-lasting behaviors. Together, our data demonstrate that DNA methylation regulating factors are differentially altered by cocaine and food. At the molecular level, they support the idea that neural circuits activated by both reinforcers do not completely overlap.

Cocaine represses protein phosphatase-1Cß through DNA methylation and Methyl-CpG Binding Protein-2 recruitment in adult rat brain.

Repeated cocaine exposure induces epigenetic factors such as DNA methyl-binding proteins, indicating that resulting changes in gene expression are mediated by alterations in brain DNA methylation. While the activity of protein phosphatase type-1 (PP1) is involved in cocaine effects and in brain plasticity, the expression of the PP1Cß catalytic subunit gene was identified here as modulated by cocaine. Its expression was induced together with that of PP1Cγ in the brain of Methyl-CpG Binding Protein-2 (Mecp2) mutant mice, whereas PP1Cα expression was not affected, illustrating a different regulation of PP1C isoforms. Repeated cocaine administration was found to increase DNA methylation at the PP1Cß gene together with its binding to Mecp2 in rat caudate putamen, establishing a link between two genes involved in cocaine-related effects and in learning and memory processes. Cocaine also increased DNMT3 expression, resulting in PP1Cß repression that did not occur in the presence of DNMT inhibitor. Cocaine-induced PP1Cß repression was observed in several brain structures, as evaluated by RT-qPCR, immunohistochemistry and Western blot, but did not occur after a single cocaine injection. Our data demonstrate that PP1Cß is a direct MeCP2-target gene in vivo. They suggest that its repression may participate to behavioral adaptations triggered by the drug.

The double-H maze test, a novel, simple, water-escape memory task: acquisition, recall of recent and remote memory, and effects of systemic muscarinic or NMDA receptor blockade during training.

To explore spatial cognition in rodents, research uses maze tasks, which differ in complexity, number of goals and pathways, behavioural flexibility, memory duration, but also in the experimenter's control over the strategy developed to reach a goal (e.g., allocentric vs. egocentric). This study aimed at validating a novel spatial memory test: the double-H maze test. The transparent device made of an alley with two opposite arms at each extremity and two in its centre is flooded. An escape platform is submerged in one arm. For experiments 1-3, rats were released in unpredictable sequences from one of both central arms to favour an allocentric approach of the task. Experiment 1 (3 trials/day over 6 days) demonstrated classical learning curves and evidence for recent and nondegraded remote memory performance. Experiment 2 (2 days, 3 trials/day) showed a dose-dependent alteration of task acquisition/consolidation by muscarinic or NMDA receptor blockade; these drug effects vanished with sustained training (experiment 3; 4 days, 3 trials/day). Experiment 4 oriented rats towards a procedural (egocentric) approach of the task. Memory was tested in a misleading probe trial. Most rats immediately switched from response learning-based to place learning-based behaviour, but only when their initial view on environmental cues markedly differed between training and probe trials. Because this simple task enables the formation of a relatively stable memory trace, it could be particularly adapted to study consolidation processes at a system level or/and the interplay between procedural and declarative-like memory systems.