Formation of accumbens GluR2-lacking AMPA receptors mediates incubation of cocaine craving

Abstract

Relapse to cocaine use after prolonged abstinence is an important clinical problem. This relapse is often induced by exposure to cues associated with cocaine use. To account for the persistent propensity for relapse, it has been suggested that cue-induced cocaine craving increases over the first several weeks of abstinence and remains high for extended periods. We and others identified an analogous phenomenon in rats that was termed 'incubation of cocaine craving': time-dependent increases in cue-induced cocaine-seeking over the first months after withdrawal from self-administered cocaine. Cocaine-seeking requires the activation of glutamate projections that excite receptors for alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) in the nucleus accumbens. Here we show that the number of synaptic AMPA receptors in the accumbens is increased after prolonged withdrawal from cocaine self-administration by the addition of new AMPA receptors lacking glutamate receptor 2 (GluR2). Furthermore, we show that these new receptors mediate the incubation of cocaine craving. Our results indicate that GluR2-lacking AMPA receptors could be a new target for drug development for the treatment of cocaine addiction. We propose that after prolonged withdrawal from cocaine, increased numbers of synaptic AMPA receptors combined with the higher conductance of GluR2-lacking AMPA receptors causes increased reactivity of accumbens neurons to cocaine-related cues, leading to an intensification of drug craving and relapse.

Figure 1. Time-dependent increases in cue-induced cocaine-seeking (incubation of cocaine craving)

a, Experimental timeline. SA, self-administration. b, Training: the number of infusions (each paired with a 5-s light cue) during training are shown (means ± s.e.m.). Cocaine (0.5 mg kg⁻¹ per infusion) supported self-administration, as indicated by high nose-poking in the active hole. Responding in the inactive hole was very low (not shown). c, Drug-seeking tests: shown are the number (means and s.e.m.) of nose-pokes in the previously active hole (a measure of cocaine-seeking) and inactive hole during a 30-min test performed under extinction conditions (nose-pokes deliver cue but not cocaine). Cocaine-seeking increased on withdrawal day 45 (withdrawal day × hole interaction: F_1,12 = 14.9, P < 0.01) (n = 7 per group). Asterisk, significantly different (P < 0.05) from withdrawal day 1.

Figure 2. Accumbens GluR1 and GluR3 expression increase after withdrawal from cocaine self-administration

a–d, GluR1 increased markedly in cocaine-exposed rats on withdrawal day 45 (surface (a), intracellular (b) and total (c) GluR1; drug exposure × day interaction, F_1,56 = 9.9, F_1,56 = 9.2 and F_1,56 = 12.1, respectively; P < 0.01). d, Surface/intracellular ratio. e–h, GluR2 was unchanged except for a small increase in the surface/intracellular ratio in cocaine-exposed rats on withdrawal day 1 (drug exposure × day interaction, F_1,53 = 4.0; P < 0.01). i–l, Surface GluR3 and the surface/intracellular ratio increased after self-administration of cocaine (drug exposure, F_1,48 = 4.4 and F_1,48 = 3.9, respectively; P < 0.05). Data (means and s.e.m., n = 12–18 per group) are expressed as percentages of the values in the saline group on withdrawal day 1. Asterisk, significantly different (P < 0.05) from other conditions; dagger, significantly different (P < 0.05) from saline group on withdrawal day 1.

Figure 3. GluR2-lacking AMPA receptors are detected in accumbens neurons after prolonged withdrawal from self-administration of cocaine

a, Evoked EPSC recorded after 42–47 days of withdrawal from saline or cocaine self-administration. b, Current–voltage relationships for neurons shown in a. c, Rectification index (EPSC_{−70 mV}/EPSC_{+40 mV}; Supplementary Information) from 13 and 8 neurons recorded from four cocaine-exposed and three saline-exposed rats (t₁₉ = 3.47; asterisk, P < 0.01). Data are means and s.e.m. d, e, Naspm (200 μM, 5–10 min) decreased evoked EPSC amplitude in cocaine-exposed rats (e) (t₆ = 4.72; asterisk, P < 0.01, baseline versus Naspm, seven cells per group) but not in saline-exposed rats (d). f, The effect of Naspm illustrated as the evoked EPSC amplitude normalized to baseline (t₁₂ = 3.73; asterisk, P < 0.01). g, Representative traces illustrating the effect of Naspm after 10 min of bath application. h, Location of recordings. Numbers indicate distance rostral to Bregma (B) in mm.

Figure 4. Enhanced cue-induced cocaine-seeking after prolonged withdrawal from cocaine self-administration is inhibited by blockade of GluR2-lacking AMPA receptors

a, Cue-induced cocaine-seeking tests. Left: responses (means and s.e.m.) on the previously active or inactive levers after injections of Naspm or vehicle into the accumbens 15 min before extinction tests on withdrawal days 1 or 45 (n = 10–14 per group). Right: responses on previously active lever at each hour of test (Naspm dose × withdrawal day × session hour × lever interaction, F_1,45 = 4.6; asterisk, P < 0.05 compared with other groups). b, Injections of Naspm into the accumbens had no effect on stable self-administration of oral sucrose or intravenous cocaine. Results are numbers of oral sucrose deliveries (left; 0.75 ml per delivery, n = 10) or intravenous cocaine deliveries (right; 0.75 mg kg⁻¹ per infusion, n = 5) (means ± s.e.m.). c, Placement of cannulae, showing injector tips.