Bioinformatics Analyses Reveal Age‐Specific Neuroimmune Modulation as a Target for Treatment of High Ethanol Drinking

Alcoholism is a complex trait involving the interaction of numerous biological and environmental factors. In the past several decades, researchers have convincingly elucidated that, for a disease of such complexity as alcoholism, no single molecular target underlies a particular associated phenotype. Instead, many interconnected molecular and cellular targets of small effect size underlie alcohol use disorders (AUD) and many remain poorly characterized. This multi-factorial nature of alcoholism has limited the currently available treatment options and resulted in their disparate efficacy across AUD subtypes. Global gene expression profiling has proven to be a valuable tool to predict the molecular components leading to predisposition to high drinking (Edenberg et al., 2005, Tabakoff et al., 2003, Mulligan et al., 2008, Mulligan et al., 2006), and to uncover ethanol-induced transcriptome changes in vivo and in vitro (Lewohl et al., 2000, Daniels and Buck, 2002, Mulligan et al., 2011). In silico analysis of gene expression data coupled with the use of bioinformatics programs has detected alcohol-related loci, and functional networks (Daniels and Buck, 2002, Kerns and Miles, 2008 and others to numerous to list). Genomic data, including the use of in silico bioinformatics analyses from our laboratories, has led to the identification of a new neuroimmune-targeted pharmacotherapy for the treatment of high alcohol consumption (Blednov et al., 2012). The purpose of our study was two-fold. First, we sought to determine whether a commonly used, high drinking, isogenic F1 mouse, FVB/NJ × C57BL/6J, would show age and gender differences in binge drinking. Second, a translational approach that included in silico bioinformatics analysis of brain gene expression was used to identify and test targets for pharmacotherapeutic treatment of high alcohol consumption. The Drinking-In-Dark (DID) paradigm of voluntary ethanol consumption was used to best model binge drinking (Rhodes et al., 2005) in C57BL/6J (B6) and its F1 hybrid FVB/NJ × C57BL/6J (F1) mice, which are well-characterized mouse models (Blednov et al., 2005b). Age of an individual at the time of onset of alcohol consumption is an important risk factor that affects alcohol-related problems later on in life (Grant and Dawson, 1997, Brown and Tapert, 2004). Age-differential responses to alcohol are confounding factors in the efficacy of various treatment modalities (Brown and D'Amico, 2001). Hence, to find age-appropriate medication, we tested both adolescent and adult F1 and B6 mice for binge ethanol consumption. Sex/gender differences in AUDs is an active research area with recent studies having shown that females that drink have a higher risk of developing alcohol-associated medical problems (Medina et al., 2008, Squeglia et al., 2012, Key et al., 2006, Urbano-Marquez et al., 1995). To determine important gender-related differences in alcohol consumption, both males and females were tested using the DID paradigm. The need for better therapies led us to test three sequential hypotheses: 1) Age and sex/gender influence alcohol consumption. 2) Alcohol-mediated brain gene expression shows age-specificity. 3) Age-divergent, neuroimmune function modulates commensurate binge drinking. Based on a convergence of literature suggesting that age and gender are important factors to consider when developing a translational approach (Greenfield et al., 2010, Dawes and Johnson, 2004), we tested the first general hypothesis that both influence binge alcohol consumption. After detecting a developmental difference in drinking only in male animals, we generated our second hypothesis that brain gene expression would show age and alcohol specific changes. Microarray hybridization, followed by in silico functional analyses of the transcriptome revealed age-divergent over-represented pathways related to neuroimmune function. Numerous studies have shown that ethanol mediates its effects, in part, through mis-regulation of the neuroimmune system, leading to neuroinflammation and neurodegeneration (Davis and Syapin, 2005, Sullivan and Zahr, 2008, Cippitelli et al., 2010, Crews and Nixon, 2009). The role of the neuroimmune system had recently been implicated in regulating ethanol consumption through its interaction with the neurotransmitter system regulating the reward pathways of the brain (Crews, 2011), yet it is not completely understood. Activation of neuroimmune pathways leads to the release of cytokines, chemokines and other mediators, affecting both neuronal and non-neuronal cells. Moreover, we have previously shown that alteration in neuroimmune networks by deletion (knock out) of components caused a diminution in ethanol consumption in mice (Blednov et al., 2005a, Blednov et al., 2012). Our third broad hypothesis was generated based on our bioinformatics pathway analyses and is thus: that neuroimmune physiology plays a role in regulating ethanol consumption in an age-divergent manner. To test our hypothesis, we used minocycline, a pharmacological modulator of neuroimmune pathways identified in our bioinformatics analyses (Kobayashi et al., 2013, Yang et al., 2007). We measured ethanol intake in adolescent and adult F1 mice following minocycline or saline pretreatment using the DID paradigm. A parallel study was completed in C57BL/6J mice to further validate the effects of minocycline. To investigate any action of minocycline on ethanol elimination, we tested the pharmacokinetic profile of ethanol by measuring blood ethanol concentrations over time after a single dose of ethanol following minocycline pretreatment. Results from our studies indicate an age-dependent modulation of ethanol drinking behavior by minocycline for both males and females. Our testing of sequential hypotheses, including the terminal one generated based on the output of in silico bioinformatics analyses has led to an important proof of concept: that bioinformatics analysis of brain gene expression can, and has, led to age-specific translational outcomes for the pharmacotherapeutic treatment of high alcohol consumption.