Gene expression profiling in animal models of alcoholism

Abstract

Genetic and environmental factors in alcoholism interact at the level of the transcriptome to encode the phenotypic traits of this complex clinical syndrome. Because of this, the prospect of simultaneous, genome wide, high-throughput analysis of gene expression in key brain areas potentially offers a novel strategy to identify new molecular treatment targets in this disease. Here, the evolution of experimental expression profiling approaches to models of alcoholism is described, starting with differential display (DD), progressing to the first generation of Affymetrix expression microarrays and then to present high-density arrays; and from the use of original Affymetrix software to more advanced analysis of the probe signal, and different statistical approaches to creating candidate gene lists. Gene expression profiling methods were applied to animal models of genetic susceptibility, and alcohol-induced neuroadaptation.

To identify new susceptibility genes, we analyzed expression profiles in genetically selected alcohol-preferring AA rats. Initial data were obtained by the RT-PCR based DD methodology, and two novel transcripts that were differentially expressed in cortical tissue of AA and ANA rats were identified: the signal transduction enzyme diacylglycerol kinase iota and a ribosomal protein. A reanalysis of the AA model with the microarray methodology was performed, where we attempted to identify genes that are differentially expressed in regions controlling alcohol related behaviors. Analysis of regional expression correctly identified several genes known to be preferentially expressed within the Nc. Accumbens, e.g. dopamine D1 and D2, adenosine A2a receptors, and substance P. Analysis of line-specific expression revealed 48 differentially expressed genes between AA and ANA rats. Among these, elevated hippocampal neuropeptide Y (NPY) was found in ANA rats in agreement with previous studies.

A robust down-regulation of beta-arrestin was found in several brain regions of the AA rat, providing a possible mechanism underlying the increased responsiveness in dopamine and opiate systems observed in this line. A cluster of MAPkinases indicating altered signal transduction was up-regulated within the Nc. Accumbens of the AA line, and is of particular functional interest. Results from an extended microarray study of the AA medial prefrontal cortex (mPFC) indicated higher expression of the rat glutathione-S-transferase alpha 4 (Gsta4). Quantitative RT-PCR detected increased transcript levels of Gsta4 in the AA mPFC, Nc. accumbens, amygdala, and hippocampus. These results contradicted earlier studies on alcohol preferring (iNP) and non-preferring (iNP) rats reporting lower Gsta4 transcript levels in the iP rat, and we concluded that Gsta4 expression is separated from the inherited high drinking behavioral trait in these rat lines bred for high ethanol consumption.

To identify genes involved in neuroadaptive processes underlying dependence, we established a model based on prolonged exposure of the brain to repeated cycles of intoxication and withdrawal using inhalation of ethanol vapor. This leads to a marked and long-lasting increase in voluntary ethanol intake. Exposure-induced intake is antagonized by acamprosate, a compound clinically effective in human alcoholism. Initial expression analysis revealed a set of long-term up-regulated transcripts in this model. These include members of pathways previously implicated in alcohol dependence as well as pathways not previously known to be involved in this disorder (e.g. members of the mitogen-activated protein kinase pathway).

As expression data accumulate, mining the expression database for patterns of overlap will offer an increasingly powerful second level of analysis, identification of potential treatment targets, and understanding interactions between susceptibility and neuroadaptive mechanisms.