Muscarinic M4 Receptors on Cholinergic and Dopamine D1 Receptor-Expressing Neurons Have Opposing Functionality for Positive Reinforcement and Influence Impulsivity

Abstract

The neurotransmitter acetylcholine has been implicated in reward learning and drug addiction. However, the roles of the various cholinergic receptor subtypes on different neuron populations remain elusive. Here we study the function of muscarinic M4 receptors (M4Rs) in dopamine D1 receptor (D1R) expressing neurons and cholinergic neurons (expressing choline acetyltransferase; ChAT), during various reward-enforced behaviors and in a "waiting"-impulsivity test. We applied cell-type-specific gene deletions targeting M4Rs in D1RCre or ChATCre mice. Mice lacking M4Rs in D1R-neurons displayed greater cocaine seeking and drug-primed reinstatement than their littermate controls in a Pavlovian conditioned place preference (CPP) paradigm. Furthermore, the M4R-D1RCre mice initiated significantly more premature responses (PRs) in the 5-choice-serial-reaction-time-task (5CSRTT) than their littermate controls, indicating impaired waiting impulse control. In contrast, mice lacking M4Rs in cholinergic neurons did not acquire cocaine Pavlovian conditioning. The M4R-ChATCre mice were also unable to learn positive reinforcement to either natural reward or cocaine in an operant runway paradigm. Immediate early gene (IEG) expression (cFos and FosB) induced by repeated cocaine injections was significantly increased in the forebrain of M4R-D1RCre mice, whereas it remained normal in the M4R-ChATCre mice. Our study illustrates that muscarinic M4Rs on specific neural populations, either cholinergic or D1R-expressing, are pivotal for learning processes related to both natural reward and drugs of abuse, with opposing functionality. Furthermore, we found that neurons expressing both M4Rs and D1Rs are important for signaling impulse control.

Keywords: acetylcholine; addiction; cocaine; dopamine D1 receptor; impulsivity; muscarinic M4 receptor; reward learning.

Figure 1

Cocaine-induced locomotor responses are elevated in muscarinic M4 receptor (M4R)-D1RCre mice. M4R-D1RCre mice increased their locomotor response to repeated 15 mg/kg cocaine injections (i.p.) to a greater degree than wildtype littermate controls (A) (n = 11 WT, n = 15 M4R-D1R). Acute cocaine-induced locomotion was increased in M4R-D1RCre mice compared to controls (B). Locomotor sensitization (the ratio between day 1 and 5 or between day 1 and 21) remained unchanged (C,D), this was also the case for cocaine-induced locomotion after the sensitization protocol (E). In contrast, the locomotor response of M4R-ChATCre mice did not differ from WT mice at any time point (F–J) (n = 7 WT, n = 9 M4R-ChAT). *p < 0.05 repeated measure one-way analyses of variance (ANOVA) followed by Bonferroni’s post hoc test (A) or Student’s unpaired t-test (B), ^#p < 0.05 two-way ANOVA followed by Bonferroni’s post hoc test for genotype difference (A).

Figure 2

M4R-D1RCre and M4R-ChATCre mice exhibit opposite responses to Pavlovian conditioning, consistent with upregulated immediate early genes (IEGs) in forebrain of M4R-D1RCre mice after repeated cocaine exposure. (A) M4R-D1RCre mice displayed stronger cocaine-seeking compared to wildtype littermate control mice, whereas M4R-ChATCre mice failed to develop cocaine-induced conditioned place preference (CPP; n = 12 WT, n = 9 M4R-D1RCre, n = 8 M4R-ChATCre). (B) Deletion of M4Rs from dopamine D1 receptor (D1R)-expressing neurons significantly increased reinstatement in response to 5 mg/kg cocaine i.p. (n = 5 WT, n = 9 M4R-D1RCre). (C) Neither M4R-D1RCre nor M4R-ChATCre mice differed significantly from wildtype littermate control mice in a palatable food-enforced Pavlovian place conditioning paradigm (n = 12 WT, n = 8 M4R-D1RCre, n = 10 M4R-ChATCre). (D–H) Comparisons of IEG expression in the forebrain of saline controls (NaCl; n = 12), cocaine exposed controls (WT COC; n = 8) and cocaine exposed M4R-D1RCre (M4R-D1RCre COC; n = 8) mice. (D) cFos was significantly upregulated in M4R-D1RCre mice compared to saline control and cocaine control mice. (E) Both M4R-D1RCre and control mice displayed an elevation of FosB compared to saline control mice, which was significantly augmented in M4R-D1RCre mice. The IEGs Egr1 and Egr2 were significantly induced by cocaine in M4R-D1RCre mice, but only Egr1 was elevated in control mice that received cocaine (F,G). No changes were observed in the expression of Arc upon cocaine stimulation (H). (I–M) Comparisons of IEG expression in the forebrain of NaCl (n = 8), WT COC (n = 8) and M4R-ChATCre COC (n = 8). M4R-ChATCre mice lacked a significant induction of cFos, FosB and Egr1 post cocaine, whereas control cocaine mice displayed a significant upregulation of cFos in comparison to saline controls (I–K). Egr2 (L) was significantly elevated by cocaine independently of the genotype, while Arc (M) was only induced by cocaine in M4R-ChATCre mice. *p < 0.05, **p < 0.01, ***p < 0.001 Student’s paired t-test or one-way ANOVA followed by Bonferroni’s post hoc test (in the qPCR results this indicates differences relative to saline controls). ^#p < 0.05, ^##p < 0.01 (indicates qPCR differences between cocaine-groups) one-way ANOVA followed by Bonferroni’s post hoc test.

Figure 3

M4R-ChATCre mice are unable to learn operant responding to reward. (A) Both M4R-D1RCre and wildtype littermate control mice learned the operant runway paradigm for cocaine (0.3 mg/kg i.v.; n = 13 WT, n = 5 M4R-D1RCre, n = 8 M4R-ChATCre) and (B) palatable food (n = 10 WT, n = 6 M4R-D1RCre, n = 6 M4R-ChATCre). M4R-ChATCre mice were unable to acquire an operant response to neither palatable food, compared to M4R-D1RCre, nor cocaine, compared to both controls and M4R-D1RCre mice (A,B). *p < 0.05, **p < 0.01, ***p < 0.001 two-way ANOVA followed by Bonferroni’s post hoc test.

Figure 4

M4R-D1RCre mice exhibit impulsive behavior in the 5-choice-serial-reaction-time-task (5CSRTT). As illustrated in the flow-chart, each trial was initiated by a magazine nose-poke, followed by a 5-s Delay before visual stimulus (VS) display. Touching a window during the Delay was scored as Premature Response (PR) and resulted in Time-out (illumination of the chamber lamp for 5 s) followed by repeat of the same trial. Touches to the window displaying the VS were scored Correct (resulting in Reward), while touches to other windows were scored Incorrect (resulting in Time-out). If the mouse failed to respond during VS display, the trial entered a Limited Hold period, during which Correct or Incorrect touches were still applicable. Failure to respond during this time was scored as an Omission and resulted in Time-out. A 20-s inter-trial interval (ITI) followed each reward collection or Time-out. Summary of performance during the 5CSRTT showing mean values (n = 7 WT, n = 8 M4R-D1RCre) during three sequential baseline (BL) and impulsivity-testing (TEST) conditions. Parameters assessed were trials/session (A), PRs (as fold change from BL1; B), % accuracy (C), and % omissions (D). (A) During BL conditions M4R-D1RCre mice learn the 5CSRTT paradigm better than wildtype littermate control mice, as reflected in a significantly increased number of rewards earned per session. M4R-D1RCre mice showed a significantly increased number of PRs during Test 1 (B). In line with the increase in PRs, M4R-D1RCre mice were also less accurate during Test 1 (C). Omissions did not differ between the two genotypes (D). *p < 0.05, **p < 0.01 two-way ANOVA followed by Bonferroni’s post hoc test.