D1 receptor hypersensitivity in mice with low striatal D2 receptors facilitates select cocaine behaviors

Abstract

Vulnerability for cocaine abuse in humans is associated with low dopamine D2 receptor (D2R) availability in the striatum. The mechanisms driving this vulnerability are poorly understood. In this study, we found that downregulating D2R expression selectively in striatal indirect-pathway neurons triggers a multitude of changes in D1 receptor (D1R)-expressing direct-pathway neurons, which comprise the other main subpopulation of striatal projection neurons. These changes include a leftward shift in the dose-response to a D1-like agonist that indicates a behavioral D1R hypersensitivity, a shift from PKA to ERK intracellular signaling cascades upon D1R activation, and a reduction in the density of bridging collaterals from D1R-expressing neurons to pallidal areas. We hypothesize that the D1R hypersensitivity underlies abuse vulnerability by facilitating the behavioral responses to repeated cocaine, such as locomotor sensitization and drug self-administration. We found evidence that littermate control mice develop D1R hypersensitivity after they are sensitized to cocaine. Indeed, D1-like agonist and cocaine cross-sensitize in control littermates and this effect was potentiated in mice lacking striatal D2Rs from indirect-pathway neurons. To our surprise, mice with low striatal D2Rs acquired cocaine self-administration similarly to littermate controls and showed no significant change in motivation to take cocaine but lower seeking. These findings indicate that downregulation of striatal D2Rs triggers D1R hypersensitivity to facilitate cocaine locomotor sensitization, which by itself was not associated with greater cocaine taking or seeking under the conditions tested.

Fig. 1

Cross-sensitization between cocaine and a D1-like agonist suggests a common mechanism. a Experimental timeline. Mice were treated with saline- (gray) or SKF-81297 (blue) for 5 days and all mice were challenged with a cocaine injection 14 days later. b Horizontal locomotor activity during habituation (h), two saline sessions (s1, s2) and during 5 daily sessions of treatment with saline (black) or SKF-81297 (blue). Posthoc t-test: Saline vs. SKF-81297: *p < 0.05. c Horizontal locomotor activity on cocaine challenge day before and after cocaine injection (red box) for mice pretreated with saline (black) or SKF-81297 (blue). d Normalized total counts during 1-h post cocaine challenge injection as a function of pre-challenge baseline for each mouse. Unpaired t-test, *p < 0.05. e Experimental timeline. Cocaine-naive (left) or cocaine-pretreated (right, red box) mice received escalating SKF-81297 doses (blue box). f Horizontal locomotor activity before and after saline (black) or SKF 81297 (blue, 5 mg/kg) administration for mice with no previous cocaine experience (cocaine naive, left plot) or mice pretreated with cocaine that did not show cocaine sensitization (middle plot) and those that did show cocaine locomotor sensitization (right plot, “sensitized”). g Locomotor activity to all doses of SKF-81297 tested for each group: cocaine naïve (black), cocaine-treated “non-sensitized” (red open), and cocaine-treated “sensitized” mice (red filled). Data is normalized to same-day, pre-injection baseline. Posthoc t-tests: * p < 0.05

Fig. 2

Targeted deletion of D2Rs from striatal medium spiny neurons induces behavioral hypersensitivity to D1Rs. a, b Experimental timeline. Cocaine-naive (a) or cocaine-pretreated (b) mice received escalating SKF-81297 doses (blue box). c–e Horizontal locomotor activity in response to SKF-81297 injection normalized to pre-injection baseline for cocaine-naive (a) and cocaine pretreated non-sensitized (d) and cocaine pretreated sensitized (e) Drd2^loxP/loxP (filled) and iMSN-Drd2KO (open). Posthoc t-tests: Drd2^loxP/loxP vs. iMSN-Drd2KO: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. f Correlation between sensitization score (ratio of activity on cocaine day 5/cocaine day 1; C5/C1) and the acute locomotor response to SKF-81297 (5 mg/kg) for individual Drd2^loxP/loxP (filled) and iMSN-Drd2KO (open) mice (R = 0.42, p < 0.05). g Mean ± SEM of sensitization score and SKF-81297-induced locomotion (5 mg/kg) for Drd2^loxP/loxP (open) and iMSN-Drd2KO (filled)

Fig. 3

Shift in D1R signaling in direct-pathway MSNs from the NAc in mice lacking striatal D2Rs in iMSNs. a, b Western blot analysis from NAC tissue samples showing phosphorylated GluA1 at PKA-dependent serine 845 residue (pGluA1) and total protein (a) and phosphorylated ERK1/2 (pERK1/2) and total protein (b). Top, representative images of two samples of each treatment. Bottom, densitometry values for phospho-protein normalized to total protein after saline (black) or SKF-81297 (blue) in littermate Drd2^loxP/loxP (filled) and iMSN-Drd2KO mice (open). pGluA1: Main effect of Drug: ****p < 0.0001. pERK1/2: Posthoc t-tests, SKF-81297 vs. saline: ***p < 0.001; SKF-treated Drd2^loxP/loxP vs. iMSN-Drd2KO: *p < 0.05. c SKF-81297-induced phosphorylation of GluA1 (filled) and ERK1/2 (striped) in littermate Drd2^loxP/loxP (left) and iMSN-Drd2KO mice (right). Normalized levels of pGluA1 and pERK1/2 in NAc samples from SKF-81297 treated mice are expressed as a percentage of average levels in NAc samples of saline-treated mice. Unpaired t-tests: pGluA1 vs. pERK1/2: *p < 0.05; **p < 0.01. d Confocal images of Drd1-expressing neurons (red) and immunohistochemistry labeling for pERK (green) in NAc of control Drd2^loxP/loxP (top) and iMSN-Drd2KO (bottom) following SKF-81297 (5 mg/kg) injection. e Density of pERK-positive D1R-containing neurons in Drd2^loxP/loxP (filled) and iMSN-Drd2KO (open) mice after saline (black) and SKF-81297 (blue) administration. Posthoc t-test: SKF-treated Drd2^loxP/loxP vs. iMSN-Drd2KO: **p < 0.01. f Cumulative histogram showing the frequency distribution of pERK immunostaining intensity in D1R-containing neurons after saline (black) and SKF-81297 (blue) for control Drd2^loxP/loxP (solid) and iMSN-Drd2KO mice (dashed) relative to threshold (gray dashed line). g Density of D1R-containing neurons in the NAc of Drd2^loxP/loxP and iMSN-Drd2KO after saline or SKF-81297

Fig. 4

Partial reduction of striatal D2Rs is sufficient to trigger D1R hypersensitivity and structural alterations to striatopallidal neurons. a Schematic of the experimental approach showing ChR2 expression in iMSNs and whole-cell recording in neighboring dMSNs in NAc core. b Representative traces of optogenetic-evoked inhibitory post-synaptic currents (oIPSC) recorded at baseline (bsln), in quinpirole (1 µM), and sulpiride (1 µM) in slices from Adora2aCre (top) or iMSN-Drd2HET (bottom) mice. c Time course of oIPSC amplitude normalized to baseline before quinpirole in Adora2aCre (black) or iMSN-Drd2HET (gray). d Summary of the quinpirole and sulpiride effects in control Adora2aCre and iMSN-Drd2HET mice. Posthoc t-tests: Baseline vs. Drug: *p < 0.05, **p < 0.01, ****p < 0.0001; Drd2^loxP/wt vs. iMSN-Drd2HET at quinpirole: **p < 0.01. e Left, Timeline of behavioral experiment (h, habituation, s, saline). Right, Horizontal locomotor activity in response to escalating doses of SKF-81297, shown as a percent of same-day pre-injection baseline, in cocaine-naïve Drd2^loxP/wt control (dark blue), iMSN-Drd2HET (light blue), and iMSN-Drd2KO mice (white). 2-way ANOVA: Main effect of Genotype: **p < 0.01; 2-way ANOVA: Main effect of SKF-81297: ****p < 0.0001. f–h Left, representative images of parasagittal brain sections showing red-fluorescence from D1R-containing neurons in Drd1-tdTomato mice and overlay of the region used for quantification in the striatum (DS or NAC) and the projection areas (SNr, GP, VP). Middle plots represent the mean ± SEM of the fluorescence intensity (in arbitrary units, a.u.) in the striatal and projection regions for Drd2^loxP/wt (black) and iMSN-Drd2HET mice (gray). Right plots represent the mean ± SEM of the normalized fluorescence intensity of the projection regions as a function of the striatal region for Drd2^loxP/wt (black) and iMSN-Drd2HET mice (gray). f Unpaired t-test: ns = not significant. g, h Unpaired t-tests: Drd2^loxP/wt vs. iMSN-Drd2HET: *p < 0.05

Fig. 5

Reduction in striatal D2Rs does not enhance intravenous cocaine self-administration in mice. a Operant chamber for intravenous cocaine self-administration experiment. b Experimental timeline. c, d Rate of responding on active (red) and inactive (black) lever (c) and rate of earned cocaine infusions (d) over 10 days of acquisition of cocaine self-administration (FR1, 1 mg/kg/infusion, ≤ 30 mg/kg cocaine or 6 h/day) for Drd2^loxP/wt (dark) and iMSN-Drd2HET (light) mice. 2-way ANOVA: Main effect of Day: ****p < 0.0001. e, f Rate of active lever responding (e) and total cocaine intake (f) as a function of the cocaine unit dose delivered during two dose-response sessions (FR1, escalating unit-dose every hour) for Drd2^loxP/wt (dark) and iMSN-Drd2HET (light) mice. 2-way ANOVA: Main effect of Cocaine Dose: ****p < 0.0001; Posthoc t-test: Drd2^loxP/wt vs. iMSN-Drd2HET. g, h Breakpoint values (number of lever presses emitted to receive the last cocaine infusion) reached (g) and total cocaine intake (h) during sessions with a progressive ratio schedule of reinforcement as a function of the cocaine unit dose delivered for Drd2^loxP/wt (dark) and iMSN-Drd2HET (light) mice. 2-way ANOVA: Main effect of Cocaine Dose: ****p < 0.0001. i Cocaine seeking under extinction conditions was measured as rate of active lever responding over increasing duration of cocaine abstinence for Drd2^loxP/wt (dark) and iMSN-Drd2HET (light) mice and compared to mean rate over the last 4 days of acquisition (red). 2-way ANOVA: Main effect of Genotype: *p < 0.05; Posthoc t-tests: Acquisition vs. Day 1 for both genotypes: ****p < 0.0001; Acquisition vs. Day 14 for both genotypes: **p < 0.01