Stress, Alcohol Craving and Relapse Risk: Mechanisms and Viable Treatment Targets

Stress and Relapse to Drug Seeking: An Update (Y Shaham)

In 1995 we reported that exposing rats to mild intermittent footshock stress reinstates heroin seeking after extinction of the drug-reinforced behavior (Shaham and Stewart, 1995). This finding led to a series of studies by us and other investigators on the behavioral and neuronal mechanisms involved in stress-induced relapse to drug seeking, as measured in an operant reinstatement model (Ghitza et al., 2006; Le et al., 2000; Shaham et al., 2000; Wang et al., 2005). In this lecture I summarized the main findings from these and related studies. My lecture covered four topics:

• (1) The generality of the phenomenon of footshock stress-induced reinstatement of heroin seeking in rats to other stressors, other procedures (CPP), other drug and non-drug reinforcers, and other species.

• (2) The effect of different pharmacological agents on stress-induced reinstatement of drug seeking.

• (3) The neuroanatomical sites involved in stress-induced reinstatement of drug seeking.

• (4) The application of the reinstatement model to the study of relapse to palatable food seeking.

Role and Treatment Target Potential of Group I and II Metabotropic Glutamate Receptors in Stress-Induced Drug-Seeking (F Weiss)

Stress is a major factor contributing to relapse in abstinent drug and alcohol abuse patients. Growing evidence supports a role of metabotropic glutamate receptors (mGluRs) in stress-associated drug use and drug-seeking. Moreover, neuroadaptation in the function of these receptors that develops during chronic drug exposure may play a significant role in the development of dependence and propensity to relapse (Aujla et al., 2008; Sidhpura et al., 2007; Sidhpura et al., 2008). To advance understanding of the role and treatment target potential of mGluRs in drug addiction, the effects of a selective mGlu₅ antagonist (MTEP) and a selective mGlu_2/3 agonist (LY379268) were examined in tests of drug self-administration, reinstatement, and anxiety. LY379268 (0.3–3 mg/kg) as well as MTEP (0.3–3 mg/kg) dose-dependently reversed the effects of footshock stress on reinstatement of both cocaine and ethanol-seeking behavior (Sidhpura et al., 2007; Sidhpura et al., 2008; Zhao et al., 2006). Importantly, in the case of ethanol-seeking, rats with a history of ethanol dependence, tested after three weeks of abstinence, were more sensitive to the “anti-reinstatement” effects of LY379268 than MTEP and, conversely, non-dependent rats were more sensitive to the effects of MTEP than LY379268 (Hao et al., 2008; Sidhpura et al., 2007; Sidhpura et al., 2008; Zhao et al., 2007). In an animal model of anxiety (i.e., the “shock-probe” or “defensive burying” test), LY379268 (0.3–3 mg/kg) dose-dependently reduced anxiety-like behavior in rats with a history of escalated cocaine self-administration. In contrast, in rats without a history of cocaine escalation and in drug-naïve controls, this mGlu_2/3 agonist reduced anxiety-like behavior only at the highest dose (Aujla et al., 2008). These finding confirm that Group I and II mGluRs represent a promising treatment target for stress-induced drug-seeking and relapse as well as general anxiety associated with protracted drug withdrawal. Additionally, these findings suggest that chronic ethanol and cocaine intoxication neuroadaptively alters the function of Group I and II mGluRs. These changes may be of significance for understanding the neurobiological basis of drug dependence and, in particular, vulnerability to stress-induced relapse associated with a history of drug dependence. Supported by AA10531 and DA07348.

Stress-Induced Reinstatement of Alcohol Seeking: Studies with the α2 Noradrenergic Receptor Antagonist Yohimbine (AD Le)

We have found that systemic injections of yohimbine (an alpha 2 adrenoceptor antagonist that induces stress and anxiety like responses in humans and laboratory animals) potently reinstate alcohol seeking in drug-free rats (Le et al., 2005). Unlike exposure to intermittent footshock stress, the original stressor found to reinstate alcohol seeking, yohimbine also increases ongoing alcohol self-administration. In an initial pharmacological study, we found that the CRF1 receptor antagonist antalarmin attenuated yohimbine-induced increases in alcohol self-administration and yohimbine-induced reinstatement (Marinelli et al., 2007). Although yohimbine is a prototypical α2 adrenoceptor antagonist, it also acts at other receptor types and subtypes, including 5-HT1a, α1 and D2-like receptors (Millan et al., 2000). Results from our recent studies with idazoxan (another prototypical α2 adrenoceptor antagonist) or RS-79948 (a highly selective α2 adrenoceptor antagonist) demonstrated that these compounds do not mimic the effects of yohimbine on alcohol self-administration or reinstatement. On the other hand, systemic injections of the 5-HT1a receptor antagonist WAY 100,635 decreased yohimbine-induced increases in alcohol self-administration and yohimbine-induced reinstatement. Clonidine, the widely used alpha-2-agonist did not significantly affect Yohimbine-induced increases in alcohol self-administration but did attenuate reinstatement (Le et al., 2009, in press). The results of our studies suggest that alpha-2- antagonists properties of yohimbine may play a role in the reinstatement of alcohol seeking, but not self-administration. On the other hand, yohimbine's actions on 5-HT1A receptors contribute to its effect on both alcohol self-administration and reinstatement.

A Human Laboratory Model of Risk Factors for Relapse in Alcohol Dependence and for Predicting Medication Efficacy (BJ Mason)

The protracted abstinence phase in alcohol dependence involves a state of heightened vulnerability to relapse following acute withdrawal that has been linked to prolonged activation of the brain stress arousal systems, and a heightened reactivity to internal cues, e.g. negative affect, and external cues, e.g. the sight and smell of alcohol, that are common precipitants of relapse in dependent subjects. We developed a human laboratory model of such risk factors for relapse to provide an early phase screen for potential medications for protracted abstinence, as an alternative to human laboratory methods involving alcohol administration (Mason et al., 2008).

This human lab model involves initially classifying non treatment-seeking paid volunteers with alcohol dependence as cue reactive if their ratings of craving severity to in vivo alcohol cues exceed that to water cues by at least 50%. Cue reactors were randomly assigned to one week of double-blind gabapentin 1200 mg a day or matched placebo. At the end of the medication phase, subjects' subjective and physiological reactivity to pairs of affective (negative, neutral, positive) and beverage (alcohol, water) cues were measured to determine if gabapentin modified reactivity to these laboratory analogues of risk factors for relapse.

Subjects were 33 cue reactors (21% female, mean age of 39.7 years). Using mixed-effects modeling, gabapentin significantly reduced the effect of alcohol beverage exposure (relative to water) on ratings of both subjective strength of craving and also intention to drink. Gabapentin also significantly attenuated subjective arousal rating for both positive and negative affect-inducing image exposures, relative to neutral images.

This highly standardized laboratory assessment of reactivity to alcohol and affective cues offers a screen for potential anti-drinking relapse medications as an alternative to methods involving alcohol administration that violate the condition of protracted abstinence. Using this model, gabapentin was found to suppress craving to alcohol cues and arousal to affective cues often associated with relapse (Mason et al., 2009).