α2δ-1 signaling in nucleus accumbens is necessary for cocaine-induced relapse

Abstract

Relapse to cocaine seeking is associated with potentiated excitatory synapses in nucleus accumbens. α2δ-1 is an auxiliary subunit of voltage-gated calcium channels that affects calcium-channel trafficking and kinetics, initiates extracellular signaling cascades, and promotes excitatory synaptogenesis. Previous data demonstrate that repeated exposure to alcohol, nicotine, methamphetamine, and morphine upregulates α2δ-1 in reward-related brain regions, but it was unclear whether this alteration generalized to cocaine. Here, we show that α2δ-1 protein was increased in nucleus accumbens after cocaine self-administration and extinction compared with saline controls. Furthermore, the endogenous ligand thrombospondin-1, responsible for the synaptogenic properties of the α2δ-1 receptor, was likewise elevated. Using whole-cell patch-clamp recordings of EPSCs in nucleus accumbens, we demonstrated that gabapentin, a specific α2δ-1 antagonist, preferentially reduced the amplitude and increased the paired-pulse ratio of EPSCs evoked by electrical stimulation in slices from cocaine-experienced rats compared with controls. In vivo, gabapentin microinjected in the nucleus accumbens core attenuated cocaine-primed but not cue-induced reinstatement. Importantly, gabapentin's effects on drug seeking were not due to a general depression of spontaneous or cocaine-induced locomotor activity. Moreover, gabapentin had no effect on reinstatement of sucrose seeking. These data indicate that α2δ-1 contributes specifically to cocaine-reinstated drug seeking, and identifies this protein as a target for the development of cocaine relapse medications. These results also inform ongoing discussion in the literature regarding efficacy of gabapentin as a candidate addiction therapy.

Keywords: cocaine self-administration; gabapentin; nucleus accumbens; relapse; thrombospondin; α2δ-1.

Figure 1.

Cocaine self-administration upregulates α₂δ-1. A, α₂δ-1 is increased after cocaine self-administration (C) compared with saline (S) in PSD subfractions and whole-cell lysates (left). α₂δ-1 was unchanged in the dorsal striatum at this time point (right). B, TSP-1 was elevated after extinction from cocaine self-administration compared with saline. C, ADAMTS-1 was unaltered by cocaine self-administration. Data shown as relative density to yoked-saline values, and N is shown in the bars.

Figure 2.

Gabapentin reduces evoked EPSC amplitude and increases PPR. A, Representative traces of evoked EPSC recorded from NAcore MSNs of yoked-saline (left) and cocaine-extinguished (right) rats before and after washing with 50 μm gabapentin. B, Gabapentin decreased EPSCs only in cocaine-extinguished rats. C, Gabapentin increased PPR only in cocaine-extinguished rats. *p < 0.05, comparing effect of GBP with baseline.

Figure 3.

Gabapentin attenuates cocaine-induced reinstatement. A, Active lever pressing during self-administration and extinction training corresponding to data in B. B, Microinjection of gabapentin into the NAcore dose-dependently reduced cocaine plus cue reinstatement compared with vehicle. EXT, Average active lever presses over the 2 d of extinction before each reinstatement trial. C, Histological verification of injection sites corresponding to data in B; numbers are millimeters rostral to bregma. D, Active lever pressing during self-administration and extinction training corresponding to data in E. E, Gabapentin (10 μg/side) significantly reduced cocaine-primed reinstatement (left) but had no impact on cue-induced reinstatement (right). F, Histological verification of injection sites corresponding to data in E. +p < 0.05 compared with extinction levels of active lever pressing using a Bonferroni's post hoc analysis. *p < 0.05, comparing aCSF to gabapentin.

Figure 4.

Gabapentin control experiments. A, 10 μg/side microinjection of gabapentin did not alter basal locomotor activity or cocaine-induced hyperactivity (10 mg/kg, i.p., cocaine injection at arrow). B, 33 μg/side GBP did not alter basal locomotor activity but blunted cocaine-induced hyperactivity. C, Active lever pressing during self-administration and extinction training for cocaine reinstatement in D. D, Microinjection of GBP (10 μg/side) in dorsal striatum has no effect on cocaine-induced reinstatement. E, Histological verification of injection placement in dorsal striatum corresponding to data in D; numbers refer to millimeters rostral to bregma. F, Active lever pressing during self-administration and extinction training for sucrose reinstatement in G. G, Microinjection of GBP (10 μg/side) in NAcore had no effect on sucrose pellet plus cue-primed reinstatement of sucrose seeking. H, Histological verification of injection placement in NAcore corresponding to data in G. *p < 0.05, comparing GBP with aCSF microinjection using a Bonferroni's post hoc analysis.