Naltrexone aversion and treatment efficacy are greatest in humans and rats that actively consume high levels of alcohol Jennifer M. Mitchell a,b,c, ⁎, Lindsey J. Bergren c , Katherine S. Chen c , Michael C. Rowbotham a,c , Howard L. Fields a,b,c a Department of Neurology, Box 0114, University of California at San Francisco, San Francisco, CA 94143, USA b Wheeler Center for the Neurobiology of Addiction, Box 0453, University of California at San Francisco, San Francisco, CA 94143, USA c Ernest Gallo Clinic and Research Center, University of California at San Francisco, 5858 Horton Street, Suite 200, Emeryville, CA 94608, USA abstract article info Article history: Received 10 July 2008 Revised 10 September 2008 Accepted 12 September 2008 Available online 1 October 2008 Keywords: Alcoholism Revia Abuse Treatment Midbrain Dysphoria Compliance The opioid antagonist naltrexone is the standard pharmacotherapy for alcoholism, although compliance is often low. The mechanism by which naltrexone reduces drinking is yet unclear. Here we show that in active alcoholics the magnitude of naltrexone treatment efficacy is correlated with the level of naltrexone-induced aversive side effects. This correlation is not observed when subjects are sober, but emerges following alcohol administration, when subjects are intoxicated. In contrast, there is no correlation following placebo administration. To clarify these results, naltrexone was administered to ethanol-consuming rats prior to quantification of naltrexone aversion. Ethanol consumption preceding naltrexone treatment was again correlated with subsequent naltrexone-induced aversion, and this aversion correlated with subsequent decrease in ethanol consumption. In contrast, when naltrexone was given to ethanol-free rats, aversion was not predictive of ethanol consumption. We conclude that naltrexone treatment efficacy is greater during active ethanol consumption and may be partly due to aversive side effects. © 2008 Elsevier Inc. All rights reserved. Introduction Although FDA approved for over a decade, naltrexone (Revia) has failed to gain widespread acceptance as a treatment for alcohol abuse. Doctors hesitate to prescribe naltrexone because of perceived ineffec- tiveness, expense, and unpleasant side effects (Mark et al., 2003), and because of low patient compliance (Kranzler et al., 2000; Bouza et al. 2004). Despite these limitations, naltrexone is an effective treatment for alcohol abuse. It reduces the number of heavy drinking days (Monti et al., 2001; Balldin et al., 2003; Gueorguieva et al., 2007; Rosner et al., 2008), craving for alcohol (Davidson et al., 1999; Chick et al., 2000), and may inhibit relapse to drinking (Anton et al., 1999; Guardia et al., 2002; Latt et al., 2002). Therefore, the fact that its use may be limited by aversive side effects suggests that further investigation into the relationship between naltrexone-induced side effects and treatment efficacy could lead to improved therapy for alcoholism. The ventral tegmental area (VTA) is a brain region crucial to the rewarding effects of ethanol. Alcohol preferring rats will self- administer ethanol directly into the VTA (McBride et al., 1993), while the microinjection of dopamine D2 antagonists decreases ethanol consumption (Nowak et al., 2000). Furthermore, opioids can produce conditioned place preference when microinjected into the VTA (Bals-Kubik et al., 1993; Olmstead and Franklin, 1997; Terashvili et al., 2004) and an opioid action in the VTA contributes to expression of ethanol place preference in mice (Bechtholt and Cunningham, 2005). In human alcoholics, the VTA is activated following exposure to alcohol related olfactory cues (Kareken et al., 2004) suggesting an involvement of this area in the motivation to consume ethanol. By injecting naltrexone into the VTA of ethanol consuming rats, it is possible to study whether naltrexone efficacy and aversion share a common neural substrate. Alcoholics vary in their level of ethanol consumption upon entry into treatment, making it challenging to determine how recent experience with ethanol contributes to naltrexone effectiveness (Killeen et al., 2004). Place conditioning provides a way to address this issue in animals. The benefit of using a place conditioning paradigm with rats is that alcohol exposure can be tightly controlled and examined in relation to naltrexone administration. These results can then be compared to those obtained in human alcoholics undergoing naltrexone treatment. There is evidence that ethanol consumption leads to the release of an endogenous opioid, which promotes further ethanol consumption (Reid and Hunter, 1984; Nylander et al., 1994; Marinelli et al., 2005). Since naltrexone blocks opioid receptors, it likely reduces ethanol consumption by blocking the action of this endogenously released opioid. However, when given after even brief exposure to exogenous Neurobiology of Disease 33 (2009) 72–80 ⁎ Corresponding author. Ernest Gallo Clinic and Research Center, 5858 Horton Street, Suite 200, Emeryville, CA 94608, USA. Fax: +1510 985 3101. E-mail address: jennifer.mitchell@ucsf.edu (J.M. Mitchell). Available online on ScienceDirect (www.sciencedirect.com). 0969-9961/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.nbd.2008.09.018 Contents lists available at ScienceDirect Neurobiology of Disease journal homepage: www.elsevier.com/locate/ynbdi