Hypocretin Neurotransmission Within the Central Amygdala Mediates Escalated Cocaine Self-administration and Stress-Induced Reinstatement in Rats

Abstract

Background: Cocaine addiction is characterized by patterns of compulsive drug-taking, including preoccupation with obtaining cocaine and loss of control over drug intake. The lateral hypothalamic hypocretin/orexin (HCRT) system has been implicated in drug-taking and the reinstatement of drug-seeking. Evidence suggests that HCRT may drive drug-seeking through activation of specific brain regions implicated in stress system dysfunction, including the central amygdala (CeA). The role of HCRT in the persistence of compulsive-like cocaine-taking has yet to be fully elucidated.

Methods: Systemic and intra-CeA microinfusions of the HCRT-receptor 1 antagonist, SB-334867, were administered to rats allowed either short (1 hour; ShA) or long (6 hours; LgA) access to cocaine self-administration. Animals were tested for fixed and progressive ratio responding for cocaine and stress-induced reinstatement of drug-seeking. In addition, using electrophysiological techniques on in vitro slices, we investigated gamma-aminobutyric acidergic (GABAergic) neurotransmission in the medial CeA and the sensitivity of GABAergic synapses to modulation of the HCRT system in ShA or LgA rats.

Results: We found systemic administration of SB-334867 (0, 7.5, 15, 30 mg/kg) dose dependently decreased cocaine intake specifically in LgA rats but not in ShA rats. Microinjections of SB-334867 (20 nmol) bilaterally into the CeA significantly reduced cocaine intake in LgA rats. We also observed a significant attenuation of yohimbine-induced reinstatement of cocaine-seeking after intra-CeA SB-334867 (10 nmol) administration. Finally, electrophysiological data indicated enhanced GABAergic neurotransmission within the medial CeA in LgA rats, which was blocked with SB-334867 (10 μmol/L).

Conclusions: These findings suggest that HCRT neurotransmission within the CeA is implicated in compulsive-like cocaine-seeking.

Keywords: Central amygdala; Cocaine; Drug dependence; Hypocretin/orexin; Intravenous self-administration; Reinstatement.

Figure 1

Systemic HCRT-R1 antagonist, SB-334867, dose-dependently reduces fixed ratio (FR) cocaine self-administration in long access (LgA) but not short access (ShA) rats. The bars represent mean number (+ SEM) of cocaine infusions under an FR1 schedule of reinforcement. Administration of SB-334867 (0, 7.5, 15, or 30 mg/kg, i.p.) significantly decreases cocaine intake in LgA (n = 9), but not ShA (n = 10), rats during the first hour (A) and for the total 6 hour session (B; LgA only) of cocaine self-administration at the 30 mg/kg dose. *p < 0.05 and **p < 0.01 versus vehicle (VEH).

Figure 2

Systemic administration of the HCRT-R1 antagonist, SB-334867, reduces cocaine self-administration in both short access (ShA) and long access (LgA) rats. The bars represent mean number (+ SEM) of cocaine infusions under a progressive ratio (PR; left axis) schedule of reinforcement, which correspond to the ratio breakpoint (right axis). Administration of SB-334867 (30 mg/kg; 30SB) significantly decreased cocaine self-administration in LgA (n = 9) and ShA (n = 7) rats. **p < 0.01 versus vehicle (VEH).

Figure 3

Intra-CeA administration of the HCRT-R1 antagonist, SB-334867, reduces cocaine self-administration in long access rats under a fixed ratio (FR1) schedule of reinforcement. The bars represent mean number (+ SEM) of cocaine infusions during the first hour of a six-hour cocaine self-administration session. Administration of SB-334867 (0, 10, or 20 nmol SB) bilaterally into the CeA significantly decreases cocaine intake in LgA (n = 9) at the highest dose tested (20 nmol SB) versus vehicle (VEH). *p < 0.05 versus vehicle (VEH).

Figure 4

Intra-CeA administration of the HCRT-R1 antagonist, SB-334867, blocks yohimbine-induced reinstatement of cocaine seeking in rats with a history of long access (LgA) to cocaine self-administration. The bars represent mean number (+ SEM) of cocaine infusions during the six-hour self-administration session. Rats were allowed to self-administer cocaine during escalation (ESC) and then underwent extinction (EXT) training with a saline self-administration substitution. Rats showed a significant reduction in FR1 responding following EXT. LgA rats were then pretreated with either vehicle (VEH) or SB-334867 (10 nmol; SB) bilaterally into the CeA before receiving a systemic injection of yohimbine (2 mg/kg; YOH). LgA rats showed a significant increase in FR1 responding following VEH + YOH treatments. In the same LgA rats, YOH-induced reinstatement of drug-seeking was attenuated with pretreatment of intra-CeA SB. Order of VEH and SB intra-CeA injections was counterbalanced to reduce order effects. ^^p < 0.01 versus ESC; **p < 0.01 versus EXT, #p < 0.05 versus VEH+YOH.

Figure 5

Spontaneous GABA transmission is elevated in CeA neurons from rats allowed long access (LgA) to cocaine self-administration. A. Representative traces of spontaneous inhibitory postsynaptic currents (sIPSCs) in CeA neurons from naïve (top), short access (ShA, middle), and LgA rats (bottom). B. Average sIPSC frequency (upper panel) and sIPSC amplitude (lower panel) in CeA neurons from naïve (n = 27), ShA (n = 27), and LgA (n = 30) rats. C. Average miniature IPSC (mIPSC) frequency (left panel) and sIPSC amplitude (right panel) in CeA neurons from naïve (n = 15), ShA (n = 13), and LgA (n = 13) rats. *p< 0.05 versus Naïve and ShA.

Figure 6

Stimulation of HCRT-R1 enhances GABAergic neurotransmission in the CeA of naïve but not short access (ShA) or long access (LgA) rats. A. Representative traces of sIPSCs in CeA neurons from naïve (left), ShA (middle), and LgA (right) before and during superfusion of the peptide ligand HCRT-1 (1 μM). B. Average change in sIPSC frequency (left) and sIPSC amplitude (right) in CeA neurons from naïve (n = 6), ShA (n = 6), and LgA (n = 6) rats following superfusion of HCRT-1. C. Average mIPSC frequency (left) and sIPSC amplitude (right) in CeA neurons from naïve (n = 8), ShA (n = 6), and LgA (n = 6) rats following superfusion of HCRT-1. *p < 0.05 by one-sample t-test, #p < 0.05 versus ShA and LgA.

Figure 7

Blockade of HCRT-R1 reduces enhanced GABAergic neurotransmission in the CeA of long access (LgA) rats. A. Representative traces of sIPSCs in CeA neurons from naïve (left), short access (ShA, middle), and LgA rats (right) before and during superfusion of the HCRT-R1 antagonist, SB-334867 (SB; 10 μM). B. Average change in sIPSC frequency (left) and sIPSC amplitude (right) in CeA neurons from naïve (n = 5), ShA (n = 8), and LgA (n = 8) rats following superfusion of SB. C. The correlation between the basal sIPSC frequency and the change in sIPSC frequency with SB in CeA neurons from LgA (n = 8) rats. D. Average mIPSC frequency (left) and sIPSC amplitude (right) in CeA neurons from naïve (n = 6), ShA (n = 6), and LgA (n = 6) rats following superfusion of SB. E. The correlation between the basal mIPSC frequency and the change in mIPSC frequency with SB in CeA neurons from LgA (n = 6) rats. *p< 0.05 by one-sample t-test, #p < 0.05 versus Naive.