Brain stress systems in the amygdala and addiction

Abstract

Dysregulation of the brain emotional systems that mediate arousal and stress is a key component of the pathophysiology of drug addiction. Drug addiction is a chronically relapsing disorder characterized by a compulsion to seek and take drugs and the development of dependence and manifestation of a negative emotional state when the drug is removed. Activation of brain stress systems is hypothesized to be a key element of the negative emotional state produced by dependence that drives drug-seeking through negative reinforcement mechanisms. The focus of the present review is on the role of two key brain arousal/stress systems in the development of dependence. Emphasis is placed on the neuropharmacological actions of corticotropin-releasing factor (CRF) and norepinephrine in extrahypothalamic systems in the extended amygdala, including the central nucleus of the amygdala, bed nucleus of the stria terminalis, and a transition area in the shell of the nucleus accumbens. Compelling evidence argues that these brain stress systems, a heretofore largely neglected component of dependence and addiction, play a key role in engaging the transition to dependence and maintaining dependence once it is initiated. Understanding the role of the brain stress and anti-stress systems in addiction not only provides insight into the neurobiology of the "dark side" of addiction but also provides insight into the organization and function of basic brain emotional circuitry that guides motivated behavior.

Fig. 1

Schematic for the progression of alcohol dependence over time, illustrating the shift in underlying motivational mechanisms. From initial, positively reinforcing, pleasurable drug effects, the addictive process progresses over time to being maintained by negatively reinforcing relief from a negative emotional state. Data presented in this paper suggest that neuroadaptations encompassing a recruitment of extrahypothalamic CRF systems are key to this shift. [Modified with permission from Heilig and Koob, 2007.]

Fig. 2

Effects of a CRF antagonist on ethanol, nicotine, cocaine, and opioid motivational withdrawal. (A) Effect of intracerebroventricular administration of the CRF peptide antagonist α-helical CRF_9–41 in rats tested in the elevated plus maze after ethanol withdrawal. The black bar contains data from rats tested 8 h after withdrawal from control diet and 30 min after intracerebroventricular vehicle administration (control). The three hashed bars contain data from rats tested 8 h after withdrawal from ethanol diet and 30 min after intracerebroventricular α-helical CRF_9–41 administration (0, 5, and 25 μg). *p<0.05, difference from control. ^†p<0.05, difference from group receiving intracerebroventricular vehicle after ethanol withdrawal. [Taken with permission from Baldwin et al., 1991.] (B) Effect of intracerebroventricular administration of the CRF antagonist D-Phe CRF_12–41 on the anxiogenic-like effect following chronic cocaine administration. Rats received chronic cocaine (20 mg/kg, i.p., for 14 days) or saline (1 ml/kg, i.p.). Animals then were tested in the defensive burying paradigm 48 h after the last injection. D-Phe CRF_12–41 (0, 0.04, 0.2, and 1.0 μg/5 μl) was administered intracerebroventricularly immediately after the animal touched the electrified probe and received the shock and 5 min before the testing session. Each group contained 10–14 animals. Data represent the total duration of burying behavior (mean±SEM) expressed in seconds for all experimental groups. *p<0.05, compared with chronically saline-treated groups. **p<0.01, compared with cocaine/vehicle group. [Taken with permission from Basso et al., 1999.] (C) Effects of the CRF₁ small molecule antagoinst antalarmin on naloxone-precipitated place aversion conditioning in morphine-dependent rats. Antalarmin significantly reduced naloxone-precipitated place aversion conditioning in morphine-dependent rats. Within each dose group treatment, Wilcoxon signed ranks test (D vs. D0), *p<0.05; NS, no significant place preference or place aversion with the Wilcoxon signed ranks test; between-group comparison, Mann–Whitney test (Delta D), ^##p<0.01, compared with Morph-Nal 15 group.[Taken with permission from Stinus et al., 2005.] (D) Effects of a CRF₁ antagonist MPZP on precipitated nicotine withdrawal-induced anxiety-like behavior in nicotine-dependent rats. The CRF₁ antagonist blocked precipitated nicotine withdrawal-induced anxiety-like behavior in rats using the defensive burying test. Mecamylamine (1.5 mg/kg, i.p.) injection in nicotine-dependent rats increased the time spent burying (*p<0.05, compared with vehicle), an effect blocked by pretreatment with the CRF₁ antagonist (4 mg/kg, s.c., 1 h pretreatment). n=7–9 per group. ^#p<0.05, compared with mecamylamine. [Taken with permission from George et al., 2007.]

Fig. 3

(A) Effects of ethanol withdrawal on CRF-like immunoreactivity (CRF-L-IR) in the rat amygdala determined by microdialysis. Dialysate was collected over four 2 h periods regularly alternated with nonsampling 2 h periods. The four sampling periods corresponded to the basal collection (before removal of ethanol liquid diet), and 2–4 h, 6–8 h, and 10–12 h after withdrawal. Fractions were collected every 20 min. Data are expressed as mean±SEM (n=5 per group). Analysis of variance confirmed significant differences between the two groups over time (p<0.05). [Taken with permission from Merlo-Pich et al., 1995.] (B) Mean (±SEM) dialysate CRF concentrations collected from the central nucleus of the amygdala of rats during baseline, 12 h cocaine self-administration, and a subsequent 12 h withdrawal period (cocaine group, n=5). The control group consisted of rats with the same history of cocaine self-administration training and drug exposure but not given access to cocaine on the test day (n=6). Data are expressed as percentages of basal CRF concentrations. Dialysates were collected over 2 h periods alternating with 1 h nonsampling periods shown by the timeline at the top. During cocaine self-administration, dialysate CRF concentrations in the cocaine group were decreased by about 25% compared with control animals. In contrast, termination of access to cocaine significantly increased CRF efflux that began approximately 5 h post-cocaine and reached about 400% of presession baseline levels at the end of the withdrawal session. *p<0.05, **p<0.01, ***p<0.001, simple main effects after overall mixed-factorial analysis of variance. [Taken with permission from Richter and Weiss, 1999.] (C) Effects of cannabinoid CB₁ receptor antagonist SR 141716A (3 mg/kg) on CRF release from the central nucleus of the amygdala in rats pretreated for 14 days with the CB₁ receptor agonist HU-210 (100 mg/kg). Cannabinoid withdrawal induced by SR 141716A was associated with increased CRF release (*p<0.005, n=5–8). Vehicle injections did not alter CRF efflux (n=5–7). Data were standardized by transforming dialysate CRF concentrations into percentages of baseline values based on averages of the first four fractions. [Taken with permission from Rodriguez de Fonseca et al., 1997.] (D) Effects of morphine withdrawal on extracellular CRF in the central nucleus of the amygdala. Withdrawal was precipitated by administration of naltrexone (0.1 mg/kg) in rats prepared with chronic morphine pellet implants. [Taken with permission from Weiss et al., 2001.] (E) Effect of mecamylamine-precipitated (1.5 mg/kg, i.p.) nicotine withdrawal on extracellular levels of CRF-like immunoreactivity in the central nucleus of the amygdala measured by in vivo microdialysis in chronic nicotine pump-treated (nicotine-dependent, n=7) and chronic saline pump-treated (nondependent, n=6) rats. *p<0.05, compared with nondependent. [Taken with permission from George et al., 2007.]

Fig. 4

Effects of small molecule CRF₁ receptor antagonists on drug self-administration in dependent rats (A) Effect of small molecule CRF₁ receptor antagonist MPZP on operant self-administration of alcohol (g/kg) in dependent and nondependent rats. Testing was conducted when dependent animals were in acute withdrawal (6–8 h after removal from ethanol vapor chambers). Dependent animals self-administered significantly more alcohol than nondependent animals. MPZP significantly reduced alcohol self-administration only in dependent animals. MPZP had no effect on alcohol self-administration in nondependent animals. *p<0.05, compared with nondependent controls. ^#p<0.05, compared with vehicle (0 mg/kg MPZP). Data are expressed as mean±SEM (n=8 per vapor treatment group). [Taken with permission from Richardson et al., 2008.] (B) Effect of MPZP on nicotine self-administration during the active period in rats given extended access to nicotine (*p<0.05 vs. baseline, ^#p<0.05 vs. post abstinence vehicle treatment, n=8). [Taken with permission from George et al., 2007.] (C) Effect of MPZP on cocaine intake in short-access (ShA) and long-access (LgA) rats. MPZP dose-dependently reduced cocaine intake, achieving a maximal reduction of ∼20%, with a greater effect in LgA compared with ShA rats. A main effect of Access (*p<0.05), a main effect of MPZP dose (p<0.001), and a significant access×MPZP dose interaction (⁺p<0.05) were observed. Data are expressed as mean+SEM cocaine intake (mg/kg). [Taken with permission from Specio et al., 2008.] (D) The CRF, antagonist R121919 reduced total heroin responses in long-access rats. Mean±SEM of heroin self-administration responses during R121919 treatment in long-access rats (n=7). The 10 and 20 mg/kg doses were effective at significantly reducing heroin self-administration in long-access rats (p<0.05). [Taken with permission from Greenwell et al., 2009a.]

Fig. 5

Effects of the α₁ adrenergic receptor antagonist prazosin on drug self-administration in dependent rats. (A) Mean (± SEM) responses for ethanol during 30 min sessions in nondependent and ethanol-dependent animals following 0.0 and 1.5 mg/kg prazosin. Prazosin (1.5 mg/kg) attenuated ethanol self-administration in ethanol-dependent animals (***p<0.001), leaving nondependent self-administration intact. [Taken with permission from Walker et al., 2008.] (B) Effect of prazosin on the break-point for 0.5 mg/kg/injection of cocaine under a progressive-ratio schedule of reinforcement. Prazosin was intraperitoneally injected 10 min before a session. Data are expressed as the number of injections/session on the left axis and the ratio per injection on the right axis. Error bars represent SEM values. The upper panel represents data from long-access (LgA) rats, and the lower panel represents data from short-access (ShA) rats. *p<0.05, compared with vehicle treatment. ^#p <0.05, compared with ShA rats. [Taken with permission from Wee et al., 2008.] (C) Prazosin decreased first hour heroin responding in long-access (12 h) rats but not in short-access (1 h) rats. Prazosin at a dose of 2 mg/kg significantly (*p<0.05) reduced heroin intake (mean ± SEM) in the first hour in 12 h access rats (long-access, n = 7, left panel). No effect was observed in short-access rats (n=7, right panel) at any of the doses tested. *p<0.05, overall effect of dose and a specific effect at the 2 mg/kg dose compared with vehicle. [Taken with permission from Greenwell et al., 2009b.]

Fig. 6

Neurocircuitry in the central nucleus of the amygdala relating CRF and norepinephrine in motivational withdrawal. CRF is hypothesized not only to drive GABAergic interneurons that engage hypothalamic and midbrain emotional systems, but also to directly engage norepinephrine systems in the brain stem which in turn reciprocally activate CRF. Such interactions provide a neurocircuitry basis for the allostatic recruitment of brain stress systems in addiction. [Modified with permission from Koob, 2008.]