A phosphatase cascade by which rewarding stimuli control nucleosomal response

Abstract

Dopamine orchestrates motor behaviour and reward-driven learning. Perturbations of dopamine signalling have been implicated in several neurological and psychiatric disorders, and in drug addiction. The actions of dopamine are mediated in part by the regulation of gene expression in the striatum, through mechanisms that are not fully understood. Here we show that drugs of abuse, as well as food reinforcement learning, promote the nuclear accumulation of 32-kDa dopamine-regulated and cyclic-AMP-regulated phosphoprotein (DARPP-32). This accumulation is mediated through a signalling cascade involving dopamine D1 receptors, cAMP-dependent activation of protein phosphatase-2A, dephosphorylation of DARPP-32 at Ser 97 and inhibition of its nuclear export. The nuclear accumulation of DARPP-32, a potent inhibitor of protein phosphatase-1, increases the phosphorylation of histone H3, an important component of nucleosomal response. Mutation of Ser 97 profoundly alters behavioural effects of drugs of abuse and decreases motivation for food, underlining the functional importance of this signalling cascade.

Figure 1. Drugs of abuse and food self-administration induce accumulation of DARPP-32 in nuclei of mouse striatal neurons through a D1R/cAMP-mediated mechanism

(a) DARPP-32 (green) and phospho-Thr-34-DARPP-32 (P-Thr34-DARPP-32, red) immunofluorescence in dorsal striatum (DStr), 15 min after i.p. injection of saline (Sal) or D-amphetamine (10 mg/kg, Amph), single confocal sections. (b) Effects of D-amphetamine (Student t-test), cocaine (20 mg/kg, time-course, F_(6,42)=26.6, p<0.01), and morphine (Mor, 5 mg/kg s.c., 15 min, Student t-test). (c) Effects of reinforcement learning 1-hour session in which mice were placed in the self-administration chamber without food (No food, NF), learned to nose poke for food pellets (Active, A), or received pellets when the active animal nose poked (Yoked, Y). DStr: F_(2,14)=15.38, p<0.001; core: F_(2,14)=14.45, p<0.001; shell: F_(2,14)=17.93, p<0.001). Means±SEM, 4-8 mice per group. One-way ANOVA, Bonferroni test (unless indicated), *p<0.05, **p<0.01, ***p<0.001. Bar: 10 μm.

Figure 2. DARPP-32 undergoes a continuous cyto-nuclear shuttling

(a) Leptomycin B (LMB, 10 ng/ml) effect on DARPP-32 localization in striatal neurons in culture. (b) DARPP-32 putative NES and major phosphorylation sites. (c) Mutagenesis of hydrophobic residues in the NES increases D32-GFP nuclear accumulation:. F_(4,21)=35.0, p<0.001; ***p<0.001 vs. wild type. (d) Stimulation of D1R has no effect on the localization of an unrelated protein with an NES (CaMKIα-GFP) or when DARPP-32 contains an additional NES (D32-NES^proh-GFP): Neurons transfected with D1R and GFP-chimeras treated with vehicle or SKF81297 (10 μM, 15 min): F_(5,22)=35.2, p<0.001; ***p<0.001 SKF vs. vehicle; ^ooop<0.001 mutant vs. wild type. Data are means ± SEM, n=3-6, one-way ANOVA, Bonferroni test.

Figure 3. Phosphorylation of Ser-97 controls intracellular localization of DARPP-32

(a) S97 mutation alters localization of DARPP-32 (first panel): Striatal neurons transfected with D1R and D32-GFP wild type or Ser-97-Ala (S97A), -Asp (S97D), or -Glu (S97E), treated for 15 min with vehicle (−) or 10 μM SKF81297 (+); F_(7,42)=16.6, p<0.001; ***p<0.001 SKF vs. basal; ^oop<0.01, ^ooop<0.001, mutant vs. wild type. Inhibition of CK2 by TBB promotes nuclear accumulation of DARPP-32 (second panel): Treatment with vehicle or 10 μM SKF81297 in the absence or presence of TBB (50 μM, 45 min before SKF); F_(3,16)=14.9, p<0.001; ***p<0.001 SKF vs. vehicle; ^ooop<0.001 TBB vs. vehicle. Mutation of Ser-97 to glutamate (S97E) prevents the effects of TBB (third panel): F_(3,12)=14.6, p<0.001; ***p<0.001 TBB vs. vehicle. Okadaic acid prevents D1R-induced nuclear accumulation of DARPP-32 (fourth panel): Treatment with vehicle or SKF81297, in the absence or presence of okadaic acid (OA, 500 nM, 45 min before SKF81297); F_{(3, 21)}=38.9, p<0.001; ***p<0.001 SKF vs. vehicle; ^ooop<0.001 OA vs. vehicle. (b) Stimulation of D1R (10 μM SKF81297) induces phosphorylation of Thr-34 and dephosphorylation of Ser-97 in mouse striatal slices: Phosphorylation of Thr-34 and Ser-97 measured by immunoblotting, normalized to untreated slices (n=5-7); **p<0.01. (c) Forskolin (10 μM) induces phosphorylation of Thr-34 and dephosphorylation of Ser-97 in slices (n=5-7): **p<0.01. (d) Okadaic acid (OA 1 μM) prevents dephosphorylation of Ser-97 induced by forskolin (Forsk) in slices: P-Thr-34-DARPP-32: F_(3,24)=16.3, p<0.001, vs. control ***p<0.001; vs. OA ∇∇p<0.01. P-Ser-97-DARPP-32: F_(3,26)=24.7, p<0.001, vs. control ***p<0.001. (e) B56δ promotes the dephosphorylation of DARPP-32 Ser-97 in response to forskolin: HEK293 cells transfected with DARPP-32 and vector, B56δ or Bα PP2A subunit, incubated with vehicle or 10 μM forskolin (10 min); DARPP-32-phospho-Ser-97 immunoblotting (n=3, Student t-test *p<0.01). Statistics: means ± SEM, one-way ANOVA and Bonferroni test (unless indicated).

Figure 4. Mutation of Ser-97 alters behavioral responses to drugs of abuse and motivation for food reward

(a) DARPP-32 immunoreactivity in the dorsal striatum of wild type (WT) or Ser-97-Ala mutant mice (S97A) injected with saline (S) or cocaine (C, 20 mg/kg, 10 min): Single confocal sections; Bar 10 μm; (n=3-8); dorsal striatum: genotype-cocaine interaction F(1,19)=11.5, p<0.01; genotype effect, F_(1,19)=5.58, p<0.05; cocaine effect, F_(1,19)=3.81, NS; **p<0.01 Sal vs. Coc; ^oop<0.01 S97A vs. wild type; core: genotype-cocaine interaction, F_(1,8)=24.0, p<0.01; genotype effect, F_(1,8)=6.5, p<0.05; cocaine effect, F_(1,8)=1.9, NS; **p<0.01 Sal vs. Coc; ^oop<0.01 S97A vs. wild type; shell, genotype-cocaine interaction F_(1,10)=12.0, p<0.01; genotype effect, F_{(1, 10)}=1.49, NS; cocaine effect, F_(1,10)=5.00, p<0.05; *p<0.05 Sal vs. Coc; ^op<0.05 S97A vs. wild type. (b) Sensitization of locomotor response to cocaine (20 mg/kg) and morphine (5 mg/kg) is decreased in S97A mutant mice: Mice received two drug injections (on days 1 and 7). Locomotor activity was recorded in a circular maze for 1 h (cocaine) or 3 h (morphine) and sensitization expressed as the ratio of response to the 2^nd injection over response to the 1st injection. Student t-test: cocaine: **p<0.01 (n=16); morphine: *p<0.01 (n=8). (c) Conditioned place preference to cocaine was prevented in S97A mice: Scores are differences between times spent in the cocaine-paired compartment after and before conditioning (n=7-8): genotype-treatment interaction, F_(1,26)=3.16, NS; genotype effect, F_(1,26)=3.76, NS; treatment effect, F_{(1, 26)}=14.21, p<0.01; **p<0.01 vs. saline. (d) Mice trained to nose poke for food reward were subjected to progressive ratio (PR) schedule and the breaking point determined as the number of pokes at which mice stopped working: Student t-test : *p<0.05 (n=7-8). Statistics: means ± SEM, two-way ANOVA and Bonferroni test (unless indicated).

Figure 5. Role of nuclear DARPP-32 in the phosphorylation of histone H3 Ser-10 in striatal neurons

(a) Cocaine-induced phosphorylation of H3 Ser-10 is abolished in Thr-34-Ala and Ser-97-Ala mutant mice: Phospho-Ser-10-H3 immunofluorescence (P-H3) in dorsal striatum of wild type (WT), T34A-DARPPP-32 (T34A) or S97A-DARRP-32 (S97A) mutant mice treated with saline (Sal) or cocaine (20 mg/kg, 30 min). (b) D1R stimulation increases Ser-10 H3 phosphorylation in DARPP-32-transfected neurons: Striatal neurons transfected with D32-GFP and D1R, treated with vehicle or SKF81297 (10 μM, 15 min). Confocal section of GFP fluorescence and phospho-Ser-10-H3 (P-H3), acetyl-Lys-14-H3 (Ac-H3) and phosphoSer-10-acetyl-Lys-14-H3 (P/Ac-H3) immunolabeling. Bars: 5 μm. (c) H3 phosphorylation requires DARPP-32 Thr-34 and nuclear accumulation: Striatal neurons transfected with wild type (WT) or mutant D32-GFP: T34A (unable to inhibit PP1), or S97E and D32-NES^proh-GFP, treated with SKF81297. Arrows: median values; Kruskal-Wallis and Dunn's tests. (d) Working model of DARPP-32 regulation of nucleosomal response in striatal neurons.