Blockade of melanocortin transmission inhibits cocaine reward

Abstract

Melanocortins and the melanocortin-4 receptor (MC4-R) are enriched in the nucleus accumbens, a brain region that has been implicated in the rewarding action of cocaine and other drugs of abuse. In the present study we use a number of rat behavioral models to show that infusion of a melanocortin peptide antagonist into the nucleus accumbens blocks the reinforcing, incentive motivational, and locomotor sensitizing effects of cocaine. We also show that locomotor responses to repeated cocaine exposure are completely blocked in MC4-R null mutant mice and reduced in Agouti mice that overexpress an endogenous inhibitor of melanocortins in the brain. The results also demonstrate that cocaine administration increases the expression of MC4-R in the nucleus accumbens and striatum, and that MC4-R is co-localized with prodynorphin in medium spiny neurons in the nucleus accumbens. Together, these findings indicate that the behavioral actions of cocaine are dependent on activation of MC4-R, and suggest that upregulation of this receptor by drug exposure may contribute to sensitization of these behavioral responses. Modulation of cocaine reward is a novel action of the melanocortin-MC4-R system and could be targeted for the development of new medications for cocaine addiction.

Fig. 1

ocaine self-administration is blocked by intra-accumbens infusion of a melanocortin antagonist. The effect of direct infusions of SHU-9119 into the nucleus accumbens on lever pressing for cocaine was determined. (A) Rats were trained to lever press for cocaine (0.50 mg/kg/injection) and tested with different doses of cocaine on test day 1 as indicated. Subjects were then tested in the presence of a single dose of SHU-9119 (0.25 μg/side) on test day 2 at the same dose of cocaine from the previous day. The results are expressed as the number of lever presses and are the means ± SEM (n = 3 for the 0.125 dose of cocaine and n = 6 animals for each of the two higher doses of cocaine, 0.25 and 0.50). Repeated-measures anova shows a significant interaction between treatment groups (vehicle vs. SHU-9119) and dose of cocaine (F_2,12 = 13.99). Post-hoc analysis demonstrates that there are significant differences at the two lower doses of cocaine (*P < 0.01). (B) The results are expressed as the total amount of cocaine self-administered at each dose in the absence or presence of SHU-9119.

Fig. 2

Cocaine-induced place conditioning is blocked by intra-accumbens infusion of SHU-9119. Rats received intra-accumbens infusions of SHU-9119 (1 μg/0.5 μL) or vehicle according to three different schedules. (A) SHU-9119 was infused each day 30 min before cocaine administration, but not on the test day. (B) SHU-9119 was infused on the test day only. (C) SHU-9119 was infused alone in the absence of cocaine. The results are presented as time spent in the drug-paired side minus time spent in the saline-paired side, and are the mean ± SEM of eight—nine animals per group (A and B) or six—seven per group (C). *P < 0.05 compared with saline-infused control (anova and Fisher’s post-hoc test).

Fig. 3

Cocaine-enhanced conditioned reward is blocked by intra-accumbens infusion of SHU-9119. Rats were trained to associate a compound stimulus (tone + light) with water reward for 15 days. Responding for the conditioned reinforcer (CR lever) relative to the no CR stimulus (NCR lever) was measured after (A) cocaine (15 mg/kg) or (B) saline injections. Intra-accumbens SHU-9119 significantly blocked cocaine-enhanced responding for CR compared with the saline infusions. After saline, animals receiving intra-accumbens infusions of SHU-9119 (1 μg/0.5 μL) also made significantly fewer CR responses than subjects given intra-nucleus accumbens saline infusions. Bars represent the mean number of lever responses ± SEM, n = 6,7 for (A) and n = 14 for (B). *P < 0.01 SHU-9119 compared with saline-infused controls (ANOVA and Fisher’s post-hoc test).

Fig. 4

Cocaine-induced locomotor activation is blocked by intra-accumbens infusion of SHU-9119. Rats received bilateral intra-nucleus accumbens infusions of SHU-9119 (1 μg/0.5 μL) or saline 30 min before administration of cocaine (15 mg/kg) for four consecutive days. Locomotor activity is shown 10 min prior to cocaine and for 10-min intervals after cocaine administration as indicated. The locomotor activity counts on day 1 (D1) and day 4 (D4) of cocaine administration are shown. The results are presented as mean activity counts ± SEM, n = 13 for saline + cocaine, and n = 11 for SHU-9119 + cocaine. Post-hoc comparisons revealed significant reductions in the SHU-9119 animals compared with saline animals at 10, 20, 40, 50 and 60 min (P < 0.05), and a trend at the 30 min time point after cocaine on day 4. No differences were observed between the groups on day 1 or before the cocaine infusions on day 1 or 4. P < 0.05 compared with vehicle-infused control (ANOVA and Fisher’s post-hoc test).

Fig. 5

Cocaine-induced locomotor activity is blocked in Agouti and MC4-R null mutant mice. (A, B) Agouti mice and C57BL/6 control mice were given cocaine (10 mg/kg) or saline on days 1–6, and levels of locomotor activity were measured. Day 0 reflects locomotor activity after 3 days of habituation to the chambers. Cocaine-induced locomotor activation was significantly reduced in the Agouti mice (A), but the response to saline administration (B) was not significantly different between Agouti and C57BL/6 mice. The results are presented as locomotor activity counts and are the mean ± SEM. The numbers of animals per group were: C57BL/6 + cocaine = 8; Agouti + cocaine = 9; C57BL/6 + saline = 6; Agouti + saline = 5. (C, D) MC4-R wild-type (+/+), heterozygous (+/–) and homozygous null mutant (–/–) littermates were also tested for cocaine-induced locomotor behavior. The response to cocaine on each day was significantly decreased in MC4-R –/– mice (C). (D) The response to either cocaine or saline treatment on day 5 only, demonstrating a significant, but intermediate, reduction in the MC4-R +/– mice, as well as complete inhibition of the cocaine response in the MC4-R –/– mice. The locomotor activity counts in response to cocaine in homozygous null mutant mice demonstrated no significant difference from locomotor activity counts in wild-type littermates receiving saline in all the days tested (data not shown). In addition, (D) demonstrates that there is no significant difference in baseline locomotor activity, in the absence of cocaine, between the MC4-R +/+, +/– and –/– littermates. The results shown are the first 10 min after cocaine administration and are presented as the mean locomotor activity counts ± SEM. The numbers of animals per group for cocaine treatment were: +/+ = 5; +/– = 7; and –/– = 10; and for saline treatment: +/+ = 4; +/- =7; and –/– = 9. *P < 0.05 compared with the corresponding control (ANOVA and Fisher’s post-hoc test).

Fig. 6

Regulation of MC4-R mRNA by administration of cocaine. (A, B) Rats were given cocaine (15 mg/kg, twice daily for 14 days) and MC4-R mRNA was determined by RNase protection analysis. MC4-R mRNA was measured in hypothalamus (HYPO), hippocampus (HP), septum (SEPT), striatum (STR), and cerebral cortex (CTX). (A) Representative autoradiograms for each of the brain regions examined. (B) MC4-R mRNA was quantified by densitometry and the results are expressed as percent of control (saline). Each bar represents the mean ± SEM, n = 6. *P < 0.05 or **P < 0.005, compared with control (Student’s t-test).

Fig. 7

Co-localization of MC4-R in prodynorphin-expressing neurons. (A) In situ hybridization analysis of MC4-R mRNA using a ³⁵S-labeled riboprobe demonstrates enrichment of MC4-R expression in the striatum, including the dorsal striatum (DS) and nucleus accumbens (NAc). Co-localization of ³⁵S-labeled MC4-R riboprobe and a digoxigenin-labeled prodynorphin riboprobe (B) in the nucleus accumbens was conducted according to standard procedures. Representative micrographs demonstrate ³⁵S-labeled MC4-R grains over (B) prodynorphin-expressing cells in the nucleus accumbens. (C) The number of MC4-R-positive cells that also express either prodynorphin or proenkepalin was determined. Co-localization of MC4-R with proenkephalin was also determined (not shown). A total of 600 MC4-R-positive cells was identified for each double-labeling condition (100 cells per rat, n = 6). The results are expressed as percent of MC4-R-positive cells that are also positive for either prodynorphin or proenkephalin.