Figure 4.

Rescue of Genetically Abolished eCB-LTD and Reward-seeking Behavior

(A) Average time-courses of mean EPSC (represented as percentage of the basal value) showing that in NAc slices prepared from D1^miRmGluR5 mice, low-frequency stimulation (10min 10 Hz, indicated by arrow) induced LTD in the presence of the MAGL inhibitor, JZL184 (1μM, red triangles, n = 5 mice) but not in untreated slices (red circles, n = 16 mice). Adjacent to the timecourse, individual experiments (red symbols) and group average (black symbols) before (baseline) and after LTD induction are shown. LTD is absent in untreated slices (on the left). In contrast, LTD was present in slices pretreated with JZL184 (on the right). Error bars indicate SEM, n = individual mouse, ∗p < 0.05, Mann-Whitney U-test.

(B) Effects of MAGL inhibition on cue-induced reinstatement of saccharin-seeking behavior. Increasing 2-AG levels by administration of the MAGL inhibitor JZL 184 (16 mg/kg, i.p.; n = 15), produces a complete rescue of the reinstatement response in the D1miRmGluR5 mice.

(C–F) (C) Intra-accumbal administration of JZL 184 into the NAc core of D1^miRmGluR5 mice resulted in a dose-dependent rescue of the reinstatement response. Behaviorally, pharmacological blockade of eCB-LTD by cocaine (n = 6–7) or the CB1 agonist CP55,940 (n = 7) selectively abolished saccharin-seeking behavior triggered by the conditioned cues in wild-type mice (D, E), whereas it did not affect saccharin self-administration (F).

(G) Accumbal D1 eCB-LTD is the physiological substrate for reward-seeking behavior. Graphic summary (G) showing the mechanism of reward seeking driven by accumbal D1-eCB-LTD. (Left) Presentation of the conditioned cue stimulates prefrontal cortico-accumbal glutamatergic neurons, glutamate is released into the synaptic space, and stimulation of mGluR5 of D1-containing MSNs triggers the synthesis and release of 2-AG through the DAGL pathway (blue arrows). Released 2-AG retrogradely activates CB1Rs in the presynaptic neuron (blue arrow), which in turn results in LTD, and thus a persistent reduction of synaptic neurotransmission (glutamate release, red arrow). This reduction on synaptic transmission might induce a “craving” state and participate to the seeking response. (Right) In D1^miRmGluR5, presentation of the cue activates mGluR5-dependent perisynaptic signaling machinery (the synthesis and release of 2-AG through the diacylglycerol pathway) only in D2 neurons, not in D1 neurons. Without this D1-dependent LTD, seeking behavior is not triggered. Blue arrows indicate activation, and red arrows indicate inhibition.

All data represent mean ± SEM. Two-way ANOVA, (∗) p < 0.05 and (∗∗) p < 0.01 vs. vehicle treatment, respectively.