We apply a translational approach to development of novel pharmacological treatments for alcohol dependence. As a first step in this process, we model clinically important features of the disease in experimental animals. For this purpose, our laboratory has modeled in rats long-term changes of the brain that lead to increased voluntary ethanol intake. This has been achieved by subjecting animals to cycles of intoxication and withdrawal, in a process that mimics that which occurs in human alcohol-dependent subjects. This leads to life-long changes of increased voluntary ethanol consumption, and altered patterns of gene expression. Genetic factors that lead to elevated spontaneous ethanol intake are modeled, working with collaborating laboratories, using rat lines that have been genetically selected for high alcohol preference. The two lines on which work is currently done are the Finnish AA and the Italian msP rat. In addition, a number of mutant mouse lines are being screened to evaluate their use as complementary models. We also study negative affect and sensitivity to stress using these models since such factors are closely related to relapse in humans. The process of identifying novel candidate targets for treatment has traditionally been inductive, using hypotheses based on pre-existing knowledge of biological pathways. This remains a useful approach. A target system of interest identified through this approach is the neuropeptide Y (NPY) system. More recently, we have turned to functional genomics approaches for discovery of novel treatment targets. The basic strategy is to subject the models mentioned above to an analysis of differential gene expression in key brain areas that are thought to control ethanol self-administration and negative affect. Brain regions of particular interest are the amygdala complex, cingulate cortex and hippocampus. DNA microarray analysis is used as a primary screen, followed by in situ and/or quantitative (TaqMan) PCR confirmation. Recently, this analysis has established several differentially expressed genes/gene clusters that are either attractive candidate treatment targets or provide mechanistic insights that can be used for medications development: 1. Several glutamatergic genes, including the glutamate transporter gene GLAST, essential for removal of extracellular glutamate, have been found to be dysregulated in the cyclic intoxication model. 2. A cluster of endocannabinoid-related genes has been extensively analyzed and found to be dysregulated in several of the animal models. For instance, we found that in the genetically selected AA model, the enzyme that degrades endocannabinoids, fatty acid amidohydrolase (FAAH), has an abnormally low expression and activity in the prefrontal cortex leads to accumulation of endocannabinoids in this region and thus an increased drive on the cannabinoid receptors. 3. An over-expression of beta arrestin 2 (BARR2) has been found in the high preferring animals. We subsequently found that this is due to the presence of a unique genetic variant (haplotype) of this gene in the AA rats. BARR2 is crucial for regulation of signalling through a large number of G-protein coupled receptors, and signalling studies are ongoing to clarify the underlying mechanisms. 4. The endogenous stress-peptide Corticotropin-Releasing Hormone (CRH) and its R1 receptor have been found to be up-regulated in several regions of the msP rats, including the amygdala. Differential expression may indicate functional involvement, and is therefore an attractive tool for discovery. However, genes whose expression is found to be altered are, in fact, in most cases ?innocent bystanders.? Establishing a potential causal involvement is a separate challenge. Therefore, in the next step, candidate targets (either generated through analysis of differential gene expression or otherwise) are validated by pharmacological and/or molecular manipulations in behaving animals. Recently, the causal role of dysregulated FAAH expression has been validated in two ways: by showing that blockade of cannabinoid CB1 receptors in the prefrontal cortex reverses the high ethanol self-administration in high preferring AA rats; and that pharmacological inhibition of FAAH in normal rats leads to a behavioral phenotype of high preference, simulating that seen in the AA rats where FAAH expression is genetically low. The functional role of BARR2 has been validated using null-mutants for this gene, in which markedly reduced ethanol intake has been found together with other abnormalities indicating decreased ethanol reward. The CRH-R1 receptor has been validated by demonstrating that administration of the selective CRH-R1 antagonist, antalarmin, to msP rats reduces their self-administration of ethanol. Finally, the glutamate transporter GLAST is under validation using null-mutants for this gene. Validation of candidate targets in most cases relies on the availability of pharmacological tools. Even more importantly, this is obviously a necessity for continued clinical development. Development of small molecules that can target receptors or other gene products on our candidate list is a major challenge. Most of such development occurs in industry, while the willingness of established pharma to get involved in addictive disorders has traditionally been very low. Recently, however, we were able to complete a collaborative research and development agreement (CRADA) with Eli Lilly and Company, which gives us access to several unique tools and offers the prospect of developing compounds for some validated targets. The greatest challenge of translational medicine is to bring preclinically validated targets into clinical development. Perhaps the most important aspect of this is the degree to which full-scale clinical efficacy trials are resource demanding. With a growing list of preclinically validated targets, filters are essential that can help us devote resources to treatments with a high likelihood of success. We are currently developing several tools to achieve that goal. Functional brain imaging may provide functional measures that are affected by candidate treatments in ways predictive of clinical efficacy. We have developed and started recruitment for one clinical protocol in which central glutamate levels are measured in withdrawing alcoholics, and where the ability of acamprosate, to normalize elevated glutamate levels as previously demonstrated in preclinical work, is used to validate the imaging paradigm. In a second protocol, release of endogenous dopamine is measured in response to an i.v. ethanol challenge in healthy volunteers, as a measure of ethanol-induced reward. An experiment of nature, i.e., a genetic variant of the mu-opioid receptor that confers increased subjective effects of alcohol, is used to validate this model.
Showing the most recent 10 out of 39 publications
