Frontiers | GABAB receptor ligands for the treatment of alcohol use disorder: preclinical and clinical evidence | Neuropharmacology
GABAB receptor ligands for the treatment of alcohol use disorder: preclinical and clinical evidence
Roberta Agabio1 and Giancarlo Colombo2*
1Department of Biomedical Sciences, University of Cagliari, Monserrato, Italy
2Section of Cagliari, Neuroscience Institute, National Research Council of Italy, Monserrato, Italy
The present paper summarizes the preclinical and clinical studies conducted to define the “anti-alcohol” pharmacological profile of the prototypic GABAB receptor agonist, baclofen, and its therapeutic potential for treatment of alcohol use disorder (AUD). Numerous studies have reported baclofen-induced suppression of alcohol drinking (including relapse- and binge-like drinking) and alcohol reinforcing, motivational, stimulating, and rewarding properties in rodents and monkeys. The majority of clinical surveys conducted to date—including case reports, retrospective chart reviews, and randomized placebo-controlled studies—suggest the ability of baclofen to suppress alcohol consumption, craving for alcohol, and alcohol withdrawal symptomatology in alcohol-dependent patients. The recent identification of a positive allosteric modulatory binding site, together with the synthesis of in vivo effective ligands, represents a novel, and likely more favorable, option for pharmacological manipulations of the GABAB receptor. Accordingly, data collected to date suggest that positive allosteric modulators of the GABAB receptor reproduce several “anti-alcohol” effects of baclofen and display a higher therapeutic index (with larger separation—in terms of doses—between “anti-alcohol” effects and sedation).
Introduction
Alcohol abuse and dependence are severe mental disorders characterized by bouts of compulsive and uncontrolled alcohol consumption, and an inability to cut down drinking despite the knowledge of its negative consequences (American Psychiatric Association, 2000). These frequent disorders are related to chronic medical conditions, car crashes, domestic violence, fetal alcohol syndrome, neuropsychological impairment, economic costs, loss of productivity, and psychiatric comorbidity (see Rehm et al., 2009; Nutt et al., 2010). Up to the penultimate edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR) (American Psychiatric Association, 2000), alcohol abuse and dependence were classified as independent alcohol use disorders (AUDs); in the last DSM edition (DSM-5), they have been joined into a single mental disorder, alcohol use disorder, or AUD (American Psychiatric Association, 2013). Diagnostic criteria for AUD include those listed for the diagnosis of dependence and abuse, together with a new criterion, a subjective experience of wanting to consume alcohol, also defined “craving” (Tiffany and Wray, 2012).
The aims of AUD treatment consist in achieving abstinence, or reducing alcohol consumption, reducing the frequency and severity of relapses, and improving psychological and social functioning. The process involved in acquiring and maintaining abstinence from alcohol may require a first phase known as “detoxification” aimed at decreasing withdrawal symptoms (when present), and a second “rehabilitation” phase, aimed at maintaining the motivation to abstain, developing an alcohol-free lifestyle, and reducing the risk of relapse. Treatment to achieve these aims includes psychological and pharmacological therapies. Psychological and social interventions (including Alcoholics Anonymous and various counseling approaches) are usually considered core treatment options (see Schuckit, 2009). However, these approaches have demonstrated only a moderate success rate. Pharmacotherapy should be used to increase the efficacy of non-pharmacological approaches. Unfortunately, the few available AUD medications display only moderate efficacy, with several limitations to their clinical use, including poor compliance and even abuse liability (see Chick and Nutt, 2012; see Franck and Jayaram-Lindstr?m, 2013).
Over the last 15 years, multiple lines of experimental and clinical evidence have suggested a role for the GABAB receptor in the control of several alcohol-related behaviors (for reviews on the GABAB receptor, see Bettler et al., 2004; Couve et al., 2004; Emson, 2007). Pharmacological activation of the orthosteric binding site of the GABAB receptor has repeatedly been reported to suppress alcohol drinking and alcohol withdrawal syndrome (AWS) in rodents and human alcoholics, as well as alcohol reinforcing and motivational properties in rodents and non-human primates and craving for alcohol in human alcoholics. These data have led to the development of the prototypic GABAB receptor agonist, baclofen (used for more than 50 years in the treatment of spasticity), as a promising, novel pharmacotherapy for AUD.
More recently, identification of a binding site of positive allosteric modulation (topographically distinct from the orthosteric binding site) has provided a new strategy for pharmacological manipulation of the GABAB neurotransmission. Activation of this binding site—by a class of agents known as positive allosteric modulators of the GABAB receptor (GABAB PAMs)—augments the affinity of the GABAB receptor for GABA and agonists and synergistically potentiates their effects (see Froestl, 2010). GABAB PAMs are devoid of substantial intrinsic agonistic activity in the absence of GABA; they do not perturb receptor signaling on their own, but potentiate the effect of GABA only in those synapses where and when endogenous GABA has been released (see Froestl, 2010). Accordingly, GABAB PAMs are expected to produce fewer side effects and lower tolerance. Rodent data confirm that GABAB PAMs reproduced several in vivo effects of baclofen, displaying a notably larger separation between the “desired” pharmacological effects (e.g.: anxiolysis, “anti-addictive” effects) and “unwanted,” or adverse, effects (e.g., sedation, motor-incoordination). For these reasons, GABAB PAMs currently represent a major step forward in the pharmacology of the GABAB receptor.
The present paper is aimed at reviewing the lines of preclinical and clinical evidence featuring baclofen and GABAB PAMs as potentially effective pharmacotherapies for AUD.
Preclinical Data
Baclofen
The vast majority of data presently available have been collected with baclofen; some of these results have subsequently been reproduced with additional GABAB receptor agonists, including CGP44532 and SKF97541 (also known as CGP35024).
Acquisition of alcohol drinking
A first set of studies investigated the effect of acute or repeated administration of baclofen on alcohol intake in rats or mice exposed to the conventional, homecage 2-bottle “alcohol vs. water” choice regimen. Under this experimental procedure, animals are exposed to the choice between a relatively low concentrated alcohol solution (usually 10%, v/v) and water and freely allowed to consume alcohol. This procedure —largely used because of its relative simplicity—provides information on the mere consumption of alcohol. It possesses however considerable predictive validity, especially when applied to rats selectively bred for high alcohol preference and consumption.
When repeatedly (once daily for 10 consecutive days) administered over the initial period of exposure to alcohol, baclofen (1–3 mg/kg, i.p.) has been reported to completely block the acquisition of alcohol drinking behavior in selectively bred, Sardinian alcohol-preferring (sP) rats (Colombo et al., 2002). Acquisition of alcohol drinking occurred only once treatment with baclofen had been interrupted. These data suggest that activation of the GABAB receptor effectively blocked discovery and experience of those psychopharmacological effects of alcohol that sustain alcohol drinking behavior in sP rats. Reduction in alcohol intake was associated to a compensatory increase in daily water intake, so that total daily fluid intake remained unchanged; these data suggest the selectivity of baclofen effect on alcohol intake and lead to reasonably exclude that the suppressing effect of baclofen on alcohol intake was secondary to possible motor-incapacitating or sedative effects (that would disrupt the normal rates of drinking).
These data have subsequently been replicated with CGP44532: its repeated administration (0.03–1 mg/kg, i.p., once daily for 10 consecutive days) completely blocked acquisition of alcohol drinking in sP rats (Colombo et al., 2002).
Maintenance of alcohol drinking
More numerous studies have investigated the effect of baclofen on alcohol intake in alcohol-experienced rats (i.e., rats in which the consumption of pharmacologically relevant doses of alcohol was already well consolidated before the start of treatment with baclofen; these alcohol-experienced rats are thought to model the “maintenance” or “active drinking” phase of human alcohol dependence). Acute or repeated (once daily for 4–14 consecutive days) administration of doses of baclofen ranging between 1 and 10 mg/kg (i.p.) suppressed, in a dose-related fashion, alcohol intake in selectively bred sP (Colombo et al., 2000) and University of Chile bibulous (UChB) (Quintanilla et al., 2008) rats, unselected Long Evans (Daoust et al., 1987; see however Smith et al., 1999) and Wistar (Stromberg, 2004) rats, and C57BL/6N (Peters et al., 2013) and Swiss (Villas Boas et al., 2012) mice. These results suggest the ability of baclofen to decrease alcohol intake in rats and mice displaying consolidated and relatively high levels of alcohol consumption.
Alcohol deprivation effect
Additional studies used the free-choice regimen to investigate the effect of treatment with baclofen in a rodent model of alcohol relapse named “alcohol deprivation effect” (ADE). ADE is defined as the temporary increase in voluntary alcohol intake—often doubling the regular intake—occurring after a period of forced abstinence, or deprivation, from alcohol (see Martin-Fardon and Weiss, 2013). An initial study (Colombo et al., 2003a) employed the 2-bottle “alcohol (10%, v/v) vs water” choice regimen and found that acute administration of baclofen (1-3 mg/kg, i.p.) to sP rats previously deprived of alcohol for 7 consecutive days resulted in the complete suppression of the extra amount of alcohol consumed during the first hour of re-access (the time period during which ADE is maximal in sP rats). In a subsequent study (Colombo et al., 2006), sP rats were exposed to a 4-bottle “alcohol vs. water” choice regimen with multiple alcohol concentrations (10, 20, and 30%, v/v). Acute injection of baclofen (1 mg/kg, i.p.) at the end of the 14-day alcohol-deprivation period resulted in the complete suppression of both aspects of the augmented demand for alcohol: (a) increase in the amount of alcohol consumed; (b) shift in preference for the highest concentrated alcohol solution (that would result in a more rapid absorption of alcohol and faster perception of its psychopharmacological effects). Notably, in both studies water and food intake, as well as spontaneous locomotor activity, were not altered by baclofen treatment, suggesting that its suppressing effect was selective for alcohol and occurred at doses devoid of any sedative effect. Similar data have been collected with CGP44532: its acute administration (0.03–1 mg/kg, i.p.) selectively blocked indeed ADE in sP rats re-exposed to alcohol after a 14-day period of alcohol deprivation (Carai et al., 2005). These data suggest a role for the GABAB receptor in the neural substrate controlling relapse-like drinking in rats.
Binge-like alcohol drinking
Baclofen has also been found to suppress alcohol intake in two mouse models of binge drinking. The first study (Moore and Boehm, 2009) used the experimental model Drinking in the Dark (DID), based on the exposure of alcohol-consuming C57BL/6J mice to daily drinking sessions of 2 h (3 h into the dark phase of the light-dark cycle) with a single bottle containing 20% (v/v) alcohol. Acute microinjection of baclofen (0.01 and 0.02 μg) into the anterior, but not posterior, portion of the ventral tegmental area (VTA) resulted in a marked reduction of alcohol intake. Baclofen microinjection did not alter the intake of water or sweetened water, demonstrating the drug selectivity on alcohol intake. The second study (Tanchuck et al., 2011) used the experimental model Scheduled High Alcohol Consumption, in which Withdrawal Seizure Control mice were exposed to 30-min drinking sessions every third day with 5% (v/v) alcohol. Again, acute administration of baclofen (2.5–5 mg/kg, i.p.) markedly reduced alcohol intake.
Operant self-administration of alcohol
Several studies have also tested the capacity of baclofen to suppress operant, oral alcohol self-administration in rats. Under these procedures, alcohol is made available to rats via completion of a behavioral response (usually, responding on a lever for a given number of times), to allow the reinforcing and motivational properties of alcohol to be assessed (in addition to its mere consumption). Operant procedures differ from the alcohol-drinking procedures in that alcohol is not “freely” available and a specific amount of “work” is required to access alcohol. Baclofen has been tested under the three most commonly used procedures of alcohol self-administration: (a) fixed ratio (FR) schedule of reinforcement, in which the response requirement (RR; i.e., the “cost” of each alcohol presentation in terms of number of responses on the lever) is predetermined and kept fixed throughout the session (providing a measure of both alcohol consumption and reinforcing properties of alcohol); (b) within-session progressive ratio (PR) schedule of reinforcement, in which—over the same single session—RR is progressively increased after the delivery of each reinforcer, and the lowest ratio not completed (named breakpoint) is taken as measure of motivational properties of alcohol; (c) within-session extinction responding (ER), in which lever-responding is never reinforced: the highest number of lever-responses (named ER) is taken as measure of motivational properties of alcohol.
The studies using the FR (varying from FR1 to FR4) schedule of reinforcement have reported that acute (often using a Latin square design) and repeated (3 consecutive daily self-administration sessions) treatment with baclofen (0.5–5.6 mg/kg, i.p., in rats; 1–17 mg/kg, i.p., in mice) reduced both number of lever-responses for alcohol and amount of self-administered alcohol in (a) selectively bred alcohol-preferring sP (Maccioni et al., 2005, 2012), Indiana P (Liang et al., 2006; Maccioni et al., 2012), and Alko Alcohol (AA) (Maccioni et al., 2012) rats, (b) unselected Long-Evans (Anstrom et al., 2003; Janak and Gill, 2003) and Wistar (Petry, 1997; Walker and Koob, 2007; Dean et al., 2011) rats, and (c) C57BL/6J mice (Besheer et al., 2004). In close agreement with this large set of data on baclofen, acute treatment with SKF97541 (0.1–1 mg/kg, i.p.) reduced operant alcohol self-administration in C57BL/6J mice (Besheer et al., 2004).
Similar data were also collected in the PR and ER experiments: treatment with baclofen (1–3 mg/kg; i.p; administered using a Latin square design) (a) reduced the breakpoint for alcohol in sP (Maccioni et al., 2008b, 2012), P (Maccioni et al., 2012), AA (Maccioni et al., 2012) and Wistar (Walker and Koob, 2007) rats as well as (b) virtually completely suppressed ER in sP rats (Colombo et al., 2003b).
Taken together, data from alcohol self-administration studies suggest that treatment with non-sedative doses of baclofen effectively reduced—beside alcohol consumption—the reinforcing and motivational properties of alcohol. Since operant procedures of alcohol self-administration in rodents possess predictive validity for human craving for alcohol (see Markou et al., 1993), these data bear translational relevance. In terms of translation “from lab bench to bedside,” it may be worth noting that baclofen resulted to be more potent and effective in those rats (either selectively bred or made physically dependent on alcohol) seeking and taking larger amounts of alcohol (Walker and Koob, 2007; Maccioni et al., 2012). This view shares some similarities with clinical data. Indeed, as stated below, the hypothesis that baclofen may be more effective in patients affected by severe forms of AUD has been advanced to reconcile the discrepancy observed among some clinical surveys: surveys recruiting patients affected by severe alcohol dependence provided stronger evidence on the suppressing effect of baclofen on alcohol craving and consumption than those studies including individuals affected by moderate AUD.
When investigated, the selectivity of the reducing effect of baclofen on alcohol self-administration was relatively modest, as treatment with baclofen tended to also reduce self-administration of and ER for sucrose or saccharin solutions or regular food pellets (used as alternative, non-drug reinforcers) with magnitude comparable to that observed in the “alcohol” experiments (Anstrom et al., 2003; Colombo et al., 2003b; Janak and Gill, 2003; Maccioni et al., 2005, 2008b, 2012).
Reinstatement of alcohol-seeking behavior
Operant procedures of alcohol self-administration can be profitably used also for studies of reinstatement of alcohol-seeking behavior. In this procedure, previously extinguished, unreinforced lever-responding for alcohol is resumed—or reinstated—by alcohol-related stimuli, alcohol itself, stressors, or drugs such as nicotine and cannabinoids. Reinstatement of alcohol-seeking behavior represents an experimental model of loss of control over alcohol and relapse episodes occurring in alcohol-dependent patients (see Martin-Fardon and Weiss, 2013). In agreement with the results of the ADE experiments (see above), acute treatment with baclofen (3 mg/kg, i.p.) markedly reduced lever-responding in sP rats initially trained to lever-respond for alcohol under an FR4 schedule of reinforcement, then subjected to an ER phase, and finally to a single reinstatement session in which lever-responding was effectively reinstated by presentation of an alcohol-associated stimulus complex (Maccioni et al., 2008a).
GABAB receptor ligands for the treatment of alcohol use disorder: preclinical and clinical evidence
Roberta Agabio1 and Giancarlo Colombo2*
1Department of Biomedical Sciences, University of Cagliari, Monserrato, Italy
2Section of Cagliari, Neuroscience Institute, National Research Council of Italy, Monserrato, Italy
The present paper summarizes the preclinical and clinical studies conducted to define the “anti-alcohol” pharmacological profile of the prototypic GABAB receptor agonist, baclofen, and its therapeutic potential for treatment of alcohol use disorder (AUD). Numerous studies have reported baclofen-induced suppression of alcohol drinking (including relapse- and binge-like drinking) and alcohol reinforcing, motivational, stimulating, and rewarding properties in rodents and monkeys. The majority of clinical surveys conducted to date—including case reports, retrospective chart reviews, and randomized placebo-controlled studies—suggest the ability of baclofen to suppress alcohol consumption, craving for alcohol, and alcohol withdrawal symptomatology in alcohol-dependent patients. The recent identification of a positive allosteric modulatory binding site, together with the synthesis of in vivo effective ligands, represents a novel, and likely more favorable, option for pharmacological manipulations of the GABAB receptor. Accordingly, data collected to date suggest that positive allosteric modulators of the GABAB receptor reproduce several “anti-alcohol” effects of baclofen and display a higher therapeutic index (with larger separation—in terms of doses—between “anti-alcohol” effects and sedation).
Introduction
Alcohol abuse and dependence are severe mental disorders characterized by bouts of compulsive and uncontrolled alcohol consumption, and an inability to cut down drinking despite the knowledge of its negative consequences (American Psychiatric Association, 2000). These frequent disorders are related to chronic medical conditions, car crashes, domestic violence, fetal alcohol syndrome, neuropsychological impairment, economic costs, loss of productivity, and psychiatric comorbidity (see Rehm et al., 2009; Nutt et al., 2010). Up to the penultimate edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR) (American Psychiatric Association, 2000), alcohol abuse and dependence were classified as independent alcohol use disorders (AUDs); in the last DSM edition (DSM-5), they have been joined into a single mental disorder, alcohol use disorder, or AUD (American Psychiatric Association, 2013). Diagnostic criteria for AUD include those listed for the diagnosis of dependence and abuse, together with a new criterion, a subjective experience of wanting to consume alcohol, also defined “craving” (Tiffany and Wray, 2012).
The aims of AUD treatment consist in achieving abstinence, or reducing alcohol consumption, reducing the frequency and severity of relapses, and improving psychological and social functioning. The process involved in acquiring and maintaining abstinence from alcohol may require a first phase known as “detoxification” aimed at decreasing withdrawal symptoms (when present), and a second “rehabilitation” phase, aimed at maintaining the motivation to abstain, developing an alcohol-free lifestyle, and reducing the risk of relapse. Treatment to achieve these aims includes psychological and pharmacological therapies. Psychological and social interventions (including Alcoholics Anonymous and various counseling approaches) are usually considered core treatment options (see Schuckit, 2009). However, these approaches have demonstrated only a moderate success rate. Pharmacotherapy should be used to increase the efficacy of non-pharmacological approaches. Unfortunately, the few available AUD medications display only moderate efficacy, with several limitations to their clinical use, including poor compliance and even abuse liability (see Chick and Nutt, 2012; see Franck and Jayaram-Lindstr?m, 2013).
Over the last 15 years, multiple lines of experimental and clinical evidence have suggested a role for the GABAB receptor in the control of several alcohol-related behaviors (for reviews on the GABAB receptor, see Bettler et al., 2004; Couve et al., 2004; Emson, 2007). Pharmacological activation of the orthosteric binding site of the GABAB receptor has repeatedly been reported to suppress alcohol drinking and alcohol withdrawal syndrome (AWS) in rodents and human alcoholics, as well as alcohol reinforcing and motivational properties in rodents and non-human primates and craving for alcohol in human alcoholics. These data have led to the development of the prototypic GABAB receptor agonist, baclofen (used for more than 50 years in the treatment of spasticity), as a promising, novel pharmacotherapy for AUD.
More recently, identification of a binding site of positive allosteric modulation (topographically distinct from the orthosteric binding site) has provided a new strategy for pharmacological manipulation of the GABAB neurotransmission. Activation of this binding site—by a class of agents known as positive allosteric modulators of the GABAB receptor (GABAB PAMs)—augments the affinity of the GABAB receptor for GABA and agonists and synergistically potentiates their effects (see Froestl, 2010). GABAB PAMs are devoid of substantial intrinsic agonistic activity in the absence of GABA; they do not perturb receptor signaling on their own, but potentiate the effect of GABA only in those synapses where and when endogenous GABA has been released (see Froestl, 2010). Accordingly, GABAB PAMs are expected to produce fewer side effects and lower tolerance. Rodent data confirm that GABAB PAMs reproduced several in vivo effects of baclofen, displaying a notably larger separation between the “desired” pharmacological effects (e.g.: anxiolysis, “anti-addictive” effects) and “unwanted,” or adverse, effects (e.g., sedation, motor-incoordination). For these reasons, GABAB PAMs currently represent a major step forward in the pharmacology of the GABAB receptor.
The present paper is aimed at reviewing the lines of preclinical and clinical evidence featuring baclofen and GABAB PAMs as potentially effective pharmacotherapies for AUD.
Preclinical Data
Baclofen
The vast majority of data presently available have been collected with baclofen; some of these results have subsequently been reproduced with additional GABAB receptor agonists, including CGP44532 and SKF97541 (also known as CGP35024).
Acquisition of alcohol drinking
A first set of studies investigated the effect of acute or repeated administration of baclofen on alcohol intake in rats or mice exposed to the conventional, homecage 2-bottle “alcohol vs. water” choice regimen. Under this experimental procedure, animals are exposed to the choice between a relatively low concentrated alcohol solution (usually 10%, v/v) and water and freely allowed to consume alcohol. This procedure —largely used because of its relative simplicity—provides information on the mere consumption of alcohol. It possesses however considerable predictive validity, especially when applied to rats selectively bred for high alcohol preference and consumption.
When repeatedly (once daily for 10 consecutive days) administered over the initial period of exposure to alcohol, baclofen (1–3 mg/kg, i.p.) has been reported to completely block the acquisition of alcohol drinking behavior in selectively bred, Sardinian alcohol-preferring (sP) rats (Colombo et al., 2002). Acquisition of alcohol drinking occurred only once treatment with baclofen had been interrupted. These data suggest that activation of the GABAB receptor effectively blocked discovery and experience of those psychopharmacological effects of alcohol that sustain alcohol drinking behavior in sP rats. Reduction in alcohol intake was associated to a compensatory increase in daily water intake, so that total daily fluid intake remained unchanged; these data suggest the selectivity of baclofen effect on alcohol intake and lead to reasonably exclude that the suppressing effect of baclofen on alcohol intake was secondary to possible motor-incapacitating or sedative effects (that would disrupt the normal rates of drinking).
These data have subsequently been replicated with CGP44532: its repeated administration (0.03–1 mg/kg, i.p., once daily for 10 consecutive days) completely blocked acquisition of alcohol drinking in sP rats (Colombo et al., 2002).
Maintenance of alcohol drinking
More numerous studies have investigated the effect of baclofen on alcohol intake in alcohol-experienced rats (i.e., rats in which the consumption of pharmacologically relevant doses of alcohol was already well consolidated before the start of treatment with baclofen; these alcohol-experienced rats are thought to model the “maintenance” or “active drinking” phase of human alcohol dependence). Acute or repeated (once daily for 4–14 consecutive days) administration of doses of baclofen ranging between 1 and 10 mg/kg (i.p.) suppressed, in a dose-related fashion, alcohol intake in selectively bred sP (Colombo et al., 2000) and University of Chile bibulous (UChB) (Quintanilla et al., 2008) rats, unselected Long Evans (Daoust et al., 1987; see however Smith et al., 1999) and Wistar (Stromberg, 2004) rats, and C57BL/6N (Peters et al., 2013) and Swiss (Villas Boas et al., 2012) mice. These results suggest the ability of baclofen to decrease alcohol intake in rats and mice displaying consolidated and relatively high levels of alcohol consumption.
Alcohol deprivation effect
Additional studies used the free-choice regimen to investigate the effect of treatment with baclofen in a rodent model of alcohol relapse named “alcohol deprivation effect” (ADE). ADE is defined as the temporary increase in voluntary alcohol intake—often doubling the regular intake—occurring after a period of forced abstinence, or deprivation, from alcohol (see Martin-Fardon and Weiss, 2013). An initial study (Colombo et al., 2003a) employed the 2-bottle “alcohol (10%, v/v) vs water” choice regimen and found that acute administration of baclofen (1-3 mg/kg, i.p.) to sP rats previously deprived of alcohol for 7 consecutive days resulted in the complete suppression of the extra amount of alcohol consumed during the first hour of re-access (the time period during which ADE is maximal in sP rats). In a subsequent study (Colombo et al., 2006), sP rats were exposed to a 4-bottle “alcohol vs. water” choice regimen with multiple alcohol concentrations (10, 20, and 30%, v/v). Acute injection of baclofen (1 mg/kg, i.p.) at the end of the 14-day alcohol-deprivation period resulted in the complete suppression of both aspects of the augmented demand for alcohol: (a) increase in the amount of alcohol consumed; (b) shift in preference for the highest concentrated alcohol solution (that would result in a more rapid absorption of alcohol and faster perception of its psychopharmacological effects). Notably, in both studies water and food intake, as well as spontaneous locomotor activity, were not altered by baclofen treatment, suggesting that its suppressing effect was selective for alcohol and occurred at doses devoid of any sedative effect. Similar data have been collected with CGP44532: its acute administration (0.03–1 mg/kg, i.p.) selectively blocked indeed ADE in sP rats re-exposed to alcohol after a 14-day period of alcohol deprivation (Carai et al., 2005). These data suggest a role for the GABAB receptor in the neural substrate controlling relapse-like drinking in rats.
Binge-like alcohol drinking
Baclofen has also been found to suppress alcohol intake in two mouse models of binge drinking. The first study (Moore and Boehm, 2009) used the experimental model Drinking in the Dark (DID), based on the exposure of alcohol-consuming C57BL/6J mice to daily drinking sessions of 2 h (3 h into the dark phase of the light-dark cycle) with a single bottle containing 20% (v/v) alcohol. Acute microinjection of baclofen (0.01 and 0.02 μg) into the anterior, but not posterior, portion of the ventral tegmental area (VTA) resulted in a marked reduction of alcohol intake. Baclofen microinjection did not alter the intake of water or sweetened water, demonstrating the drug selectivity on alcohol intake. The second study (Tanchuck et al., 2011) used the experimental model Scheduled High Alcohol Consumption, in which Withdrawal Seizure Control mice were exposed to 30-min drinking sessions every third day with 5% (v/v) alcohol. Again, acute administration of baclofen (2.5–5 mg/kg, i.p.) markedly reduced alcohol intake.
Operant self-administration of alcohol
Several studies have also tested the capacity of baclofen to suppress operant, oral alcohol self-administration in rats. Under these procedures, alcohol is made available to rats via completion of a behavioral response (usually, responding on a lever for a given number of times), to allow the reinforcing and motivational properties of alcohol to be assessed (in addition to its mere consumption). Operant procedures differ from the alcohol-drinking procedures in that alcohol is not “freely” available and a specific amount of “work” is required to access alcohol. Baclofen has been tested under the three most commonly used procedures of alcohol self-administration: (a) fixed ratio (FR) schedule of reinforcement, in which the response requirement (RR; i.e., the “cost” of each alcohol presentation in terms of number of responses on the lever) is predetermined and kept fixed throughout the session (providing a measure of both alcohol consumption and reinforcing properties of alcohol); (b) within-session progressive ratio (PR) schedule of reinforcement, in which—over the same single session—RR is progressively increased after the delivery of each reinforcer, and the lowest ratio not completed (named breakpoint) is taken as measure of motivational properties of alcohol; (c) within-session extinction responding (ER), in which lever-responding is never reinforced: the highest number of lever-responses (named ER) is taken as measure of motivational properties of alcohol.
The studies using the FR (varying from FR1 to FR4) schedule of reinforcement have reported that acute (often using a Latin square design) and repeated (3 consecutive daily self-administration sessions) treatment with baclofen (0.5–5.6 mg/kg, i.p., in rats; 1–17 mg/kg, i.p., in mice) reduced both number of lever-responses for alcohol and amount of self-administered alcohol in (a) selectively bred alcohol-preferring sP (Maccioni et al., 2005, 2012), Indiana P (Liang et al., 2006; Maccioni et al., 2012), and Alko Alcohol (AA) (Maccioni et al., 2012) rats, (b) unselected Long-Evans (Anstrom et al., 2003; Janak and Gill, 2003) and Wistar (Petry, 1997; Walker and Koob, 2007; Dean et al., 2011) rats, and (c) C57BL/6J mice (Besheer et al., 2004). In close agreement with this large set of data on baclofen, acute treatment with SKF97541 (0.1–1 mg/kg, i.p.) reduced operant alcohol self-administration in C57BL/6J mice (Besheer et al., 2004).
Similar data were also collected in the PR and ER experiments: treatment with baclofen (1–3 mg/kg; i.p; administered using a Latin square design) (a) reduced the breakpoint for alcohol in sP (Maccioni et al., 2008b, 2012), P (Maccioni et al., 2012), AA (Maccioni et al., 2012) and Wistar (Walker and Koob, 2007) rats as well as (b) virtually completely suppressed ER in sP rats (Colombo et al., 2003b).
Taken together, data from alcohol self-administration studies suggest that treatment with non-sedative doses of baclofen effectively reduced—beside alcohol consumption—the reinforcing and motivational properties of alcohol. Since operant procedures of alcohol self-administration in rodents possess predictive validity for human craving for alcohol (see Markou et al., 1993), these data bear translational relevance. In terms of translation “from lab bench to bedside,” it may be worth noting that baclofen resulted to be more potent and effective in those rats (either selectively bred or made physically dependent on alcohol) seeking and taking larger amounts of alcohol (Walker and Koob, 2007; Maccioni et al., 2012). This view shares some similarities with clinical data. Indeed, as stated below, the hypothesis that baclofen may be more effective in patients affected by severe forms of AUD has been advanced to reconcile the discrepancy observed among some clinical surveys: surveys recruiting patients affected by severe alcohol dependence provided stronger evidence on the suppressing effect of baclofen on alcohol craving and consumption than those studies including individuals affected by moderate AUD.
When investigated, the selectivity of the reducing effect of baclofen on alcohol self-administration was relatively modest, as treatment with baclofen tended to also reduce self-administration of and ER for sucrose or saccharin solutions or regular food pellets (used as alternative, non-drug reinforcers) with magnitude comparable to that observed in the “alcohol” experiments (Anstrom et al., 2003; Colombo et al., 2003b; Janak and Gill, 2003; Maccioni et al., 2005, 2008b, 2012).
Reinstatement of alcohol-seeking behavior
Operant procedures of alcohol self-administration can be profitably used also for studies of reinstatement of alcohol-seeking behavior. In this procedure, previously extinguished, unreinforced lever-responding for alcohol is resumed—or reinstated—by alcohol-related stimuli, alcohol itself, stressors, or drugs such as nicotine and cannabinoids. Reinstatement of alcohol-seeking behavior represents an experimental model of loss of control over alcohol and relapse episodes occurring in alcohol-dependent patients (see Martin-Fardon and Weiss, 2013). In agreement with the results of the ADE experiments (see above), acute treatment with baclofen (3 mg/kg, i.p.) markedly reduced lever-responding in sP rats initially trained to lever-respond for alcohol under an FR4 schedule of reinforcement, then subjected to an ER phase, and finally to a single reinstatement session in which lever-responding was effectively reinstated by presentation of an alcohol-associated stimulus complex (Maccioni et al., 2008a).
Comment