MeCP2 in the nucleus accumbens contributes to neural and behavioral responses to psychostimulants - PubMed (original) (raw)

. 2010 Sep;13(9):1128-36.

doi: 10.1038/nn.2614. Epub 2010 Aug 15.

Affiliations

MeCP2 in the nucleus accumbens contributes to neural and behavioral responses to psychostimulants

Jie V Deng et al. Nat Neurosci. 2010 Sep.

Abstract

MeCP2 is a methyl DNA-binding transcriptional regulator that contributes to the development and function of CNS synapses; however, the requirement for MeCP2 in stimulus-regulated behavioral plasticity is not fully understood. Here we show that acute viral manipulation of MeCP2 expression in the nucleus accumbens (NAc) bidirectionally modulates amphetamine (AMPH)-induced conditioned place preference. Mecp2 hypomorphic mutant mice have more NAc GABAergic synapses and show deficient AMPH-induced structural plasticity of NAc dendritic spines. Furthermore, these mice show deficient plasticity of striatal immediate early gene inducibility after repeated AMPH administration. Notably, psychostimulants induce phosphorylation of MeCP2 at Ser421, a site that regulates MeCP2's function as a repressor. Phosphorylation is selectively induced in GABAergic interneurons of the NAc, and its extent strongly predicts the degree of behavioral sensitization. These data reveal new roles for MeCP2 both in mesolimbocortical circuit development and in the regulation of psychostimulant-induced behaviors.

PubMed Disclaimer

Figures

Figure 1

Figure 1. Local lentiviral regulation of MeCP2 in the NAc alters AMPH-induced behaviors

a–f, Lentiviral shRNA knockdown of MeCP2. a, GFP expression in NAc 20 days post-injection. b–d, enlarged area in panel a. Open arrowheads indicate the absence of MeCP2 expression in GFP-expressing cells. e, 7-9 days after infection with an shRNA targeting MeCP2 (n=12) or a control scrambled shRNA (SCR, n=14), mice were administered 3mg/kg AMPH and placed into the open field for 5 consecutive days. RMANOVA showed a significant effect of treatment (_F_1,24=8.95, _p_=0.006) and days (_F_4,96=23.4, p<0.0001). * - _p_s<0.05, pair-wise comparisons between groups. f, Seven days after bilateral intra-NAc stereotactic injection of scrambled (SCR) or MeCP2 shRNA virus, C57BL/6 mice (n=7-8/group) received six alternate pairings of AMPH (1mg/kg or 3mg/kg) or Veh in the CPP apparatus as described. ANOVA revealed a significant dose by treatment interaction (_F_1,28=5.6, _p_=0.026; * - _p_=0.018, shRNA-MeCP2 compared to shRNA-SCR at 1mg/kg AMPH). g–k, Lentiviral overexpression of MeCP2 in the NAc. g, GFP expression 20 days post-injection; h–j, enlarged area in panel g. Filled arrowheads indicate the enhanced expression of MeCP2 in GFP-expressing cells. k, 7-9 days after infection with lenti-GFP (n=8) or lenti-MeCP2 (n=8), mice were tested for CPP with 3 pairings of 3mg/kg AMPH as described. RMANOVA revealed a significant chamber-preference by treatment interaction (lenti-GFP compared with lenti-MeCP2) (_F_1,13=5.47, _p_=0.036); * - _p_=0.01, time in AMPH chamber compared to that in Veh chamber within lenti-GFP treated mice. Scale bars indicate 30μm. Results displayed as mean ± s.e.m.

Figure 2

Figure 2. _Mecp2_308/y mutant mice show altered AMPH-induced behaviors

a, AMPH-induced open field locomotion in Mecp2+/y (WT) and _Mecp2_308/y (MUT) littermates. (_F_7,53=20.03, p < 0.001; * - p<0.001, WT compared to MUT for 3mg/kg; # - _p_=0.019, WT compared to MUT for 2mg/kg). b, Open field locomotor activity induced by repeated 2mg/kg AMPH. By RMANOVA the within subject effect of days and the day by genotype interaction were not significant. However, the between subjects effect of genotype was significant (_F_1, 14 =14.96, p<0.002; * - _p_s<0.05, MUT compared to WT). By RMANOVA within each genotype, the days effect was significant for WT (_F_4,28=5.115, p<0.003), but not for MUT (_F_4,28=0.609; _p_=0.659). c, CPP in _Mecp2_308 mice (n=9/genotype for 1mg/kg; n=12/genotype for 3mg/kg). For WT, RMANOVA found a significant preference for the AMPH-paired chamber at 3mg/kg (chamber by dose: _F_1,18=4.29, _p_=0.05); * - p < 0.001, Veh- compared to AMPH-paired chamber). In MUT, there was no significant main effect of dose (_F_1,20=1.45, _p_=0.24) or chamber preference (_F_1,20=1.05, _p_= 0,32) and the chamber by dose interaction was not significant (_F_1,20=0.50, _p_=0.49). d, Sucrose preference test in Mecp2+/y (WT) and _Mecp2_308/y (MUT) littermates (n=9 mice/group). For sucrose, RMANOVA showed significant main effects of concentration (_F_3,48=19.75, p<0.0001) and genotype (_F_1,16=12.88, _p_=0.002), and a significant concentration by genotype interaction (_F_3,14=2.78, _p_=0.051). * - _p_s<0.05, WT compared to MUT at 0.25% and 0.5%. _p_=0.079, WT compared to MUT at 1.0%. For quinine, RMANOVA revealed a main effect of concentration (_F_2,34=12.52, _p_=0.001), but no genotype effect. Results displayed as mean ± s.e.m.

Figure 3

Figure 3. _Mecp2_308/y MUT mice show altered GABAergic synaptic densities in the NAc

a-f, Immunofluorescent labeling of presynaptic markers for GABAergic (GAD65 and VGAT) and glutamatergic (VGLUT1) synapses in coronal sections through the NAc from Mecp2+/y WT and _Mecp2_308/y MUT mice (n=4-5/group). Scale bar indicates 10μm. g–l, Quantitation of overall intensities of synaptic protein expression and numbers of synaptic punctae. g,j, Increased expression of GAD65 in MUT compared with WT littermates: intensity (_F_1,8=14.18, * - _p_=0.007), number of punctae (_F_1,8=10.36, * - _p_=0.015). h,k, VGAT immunofluorescence shows increased expression in MUT versus WT: intensity (_F_1,8=8.33, * - _p_=0.023), number of punctae (_F_1,8=8.33, * - _p_=0.023). i,l, VGLUT1 shows no significant change in expression by genotype. Results displayed as mean ± s.e.m.

Figure 4

Figure 4. Chronic treatment with AMPH fails to increase dendritic spine density in Mecp2 MUT mice

MeCP2 MUT mice and their WT littermates were treated with Veh or AMPH (3mg/kg; i.p.) for 21 days. One day after the last injection, mice were perfused transcardially, and the brains were processed for Golgi-Cox staining. a. Representative images of dendrites from WT littermates treated with Veh or AMPH. Scale bar indicates 5μm. b. Quantitation of overall spine density per 10μm. Two-way ANOVA indicates a significant overall treatment effect (_F_3,14=8.33, _p_=0.015) and a significant genotype by treatment interaction (_F_3,14=4.85, _p_=0.05); * - _p_=0.01 WT treated with Veh compared to WT treated with AMPH. Results displayed as mean ± s.e.m.

Figure 5

Figure 5. Mecp2 mutant mice show altered induction of Fos/Jun family IEGs

a–d, c-Fos expression in the NAc of Mecp2+/y (WT) and _Mecp2_308/y (MUT) mice that received repeated Veh (Veh:AMPH) or AMPH (AMPH:AMPH) treatments and AMPH at challenge. Scale bar indicates 50μm. e, Quantitation of c-Fos expression as shown in panels a–d. ANOVA indicated a treatment by genotype effect (_F_3,26=4.36, _p_=0.048; * - _p_=0.006, Veh:AMPH-WT compared to AMPH:AMPH-WT; # - _p_=0.05, AMPH:AMPH-WT compared to AMPH:AMPH-MUT). c-Fos expression in MUT mice is not different between the Veh:AMPH and AMPH:AMPH conditions (_p_=0.90). f–i, FosB and JunB following acute AMPH challenge. Scale bar indicates 45 μm**. j–k**, Quantitations of FosB and JunB expression. j, In WT mice, FosB expression is reduced after repeated AMPH treatment (ANOVA; treatment by genotype: _F_3,23=11.01, _p_=0.003; * - p<0.001, Veh:AMPH-WT compared to AMPH:AMPH-WT). In MUT mice, the single injection of AMPH at challenge did not induce FosB expression as compared to WT animals ((x002C6) - _p_=0.001, Veh:AMPH-WT compared to Veh:AMPH-MUT), and there was no difference between the Veh:AMPH-MUT and AMPH:AMPH-MUT groups (_p_=0.96). k, Repeated injection of AMPH upregulates JunB expression in WT mice (ANOVA; treatment by genotype: _F_3,21 = 10.929, _p_=0.004; * - _p_=0.01, Veh:AMPH-WT compared to AMPH:AMPH-WT). For MUT animals acute AMPH injection at challenge caused JunB expression to be significantly elevated compared with similarly-treated WT mice (ˆ - p<0.001, Veh:AMPH-WT compared to Veh:AMPH-MUT); however, there was no statistical difference in JunB expression between the Veh:AMPH-MUT and AMPH:AMPH-MUT groups (_p_=0.47). Error bars indicate mean ± s.e.m.

Figure 6

Figure 6. Psychostimulants induce pMeCP2 in the NAc

a-g, Double immunofluorescence of the NAc with antibodies against pMeCP2 and total MeCP2, 2 h after Veh or 3 mg/kg AMPH. ac - anterior commissure. Scale bar indicates 50 μm. e–g, High magnification enlargement of box shown in c. e, overlay. Scale bar indicates 15 μm; * - co-immunolabeled cells. h, Timecourse of pMeCP2 immunofluorescence within the NAc following 3 mg/kg AMPH (n=8/group). (_F_6,48=9.04, p<0.0001); * - _p_=0.018, Veh compared to 0.5 h, and p<0.0001, compared to 2 h. i, Timecourse of total MeCP2 immunofluorescence in the NAc following 3 mg/kg AMPH (n=4-5 mice/time-point). (_F_6,22=1.28, _p_=0.31). j–n, Stimulation of dopamine D1-class receptors induces phosphorylation of MeCP2. J–m, pMeCP2 immunofluorescence after Veh (j; Veh), 3 mg/kg AMPH (k; AMPH), 5 mg/kg SKF81297 (l; SKF), or 0.25 mg/kg of SCH22390 15 min prior to AMPH injection (m; SCH+AMPH). Scale bar indicates 45 μm. n, Quantitation of pMeCP2 immunofluorescence intensity (n=6 for Veh- and AMPH-treated groups, n=9 for other groups). (_F_3,25=16.0, p<0.001). * - _p_=0.001, Veh compared to AMPH, and p<0.0001, compared to SKF; # - _p_=0.026, SCH + AMPH compared to AMPH, and _p_=0.001, compared to SKF. o, Integrated intensity of pMeCP2 immunofluorescence following Veh, 1 or 3 mg/kg AMPH, or 10 mg/kg cocaine. Open circles represent values for individual animals, and the bar represents the group mean (n=5-6/group). (_F_3,19=12.45, p<0.0001). * - p<0.0001, Veh compared to 3mg/kg AMPH, and _p_=0.042, compared to 10 mg/kg cocaine. Results displayed as mean ± s.e.m.

Figure 7

Figure 7. AMPH selectively induces pMeCP2 in fast spiking GABAergic interneurons in the NAc

_Drd1a_-tdTomato (a), _Drd2_-eGFP (b), _Lhx6_-GFP (c), or C57BL/6 mice (d–g) were injected with 3 mg/kg AMPH, euthanized 2 h later, brain sections were cut, and immunolabeled for pMeCP2. pMeCP2 staining was compared to tdTomato or eGFP fluorescence in the BAC transgenic mice (a–c), or to immunostaining for parvalbumin (d), glutamic acid decarboxylase 67 (GAD67)(e), somatostatin (f), or choline acetyltransferase (ChAT) (g). Hoechst nuclear dye (blue). Filled arrowheads indicate cells co-immunolabeled with two fluorophores; open arrowheads indicate singly labeled cells; * indicates cells shown at high magnification after 3-D deconvolution. The scale bars indicate 15μM.

Figure 8

Figure 8. Increases in pMeCP2 correlate with behavioral sensitization following repeated AMPH administration

a, Experimental scheme. Arrows represent injections. n=9-10/group. b, Cumulative distance traveled over 1 h following challenge with Veh or AMPH. ANOVA revealed a significant main effect of treatment [_F_3,35=206.3, p<0.001]. * - _p_s<0.0001, Veh:Veh or AMPH:Veh compared to Veh:AMPH or AMPH:AMPH groups; # - _p_=0.025, Veh:AMPH compared to AMPH:AMPH. c-f, pMeCP2 immunofluorescence in the NAc 2 h after challenge injections of Veh (c,d) or AMPH (e,f) from mice that received repeated injections of Veh (c,e) or AMPH (d,f). Scale bar indicates 30 μm. g, Quantitation of pMeCP2 immunofluorescence intensity in the NAc. ANOVA indicated a significant effect of treatment (_F_3,27=8.3, p<0.001). * - _p_=0.011, Veh:Veh compared to Veh:AMPH, and _p_=0.017, Veh:Veh compared to AMPH:AMPH; # - _p_=0.007, AMPH:Veh compared to Veh:AMPH, and _p_=0.013, compared to AMPH:AMPH. h–k, Pearson product correlations of pMeCP2 immunofluorescence intensity in the NAc and cumulative locomotor activity from individual animals. h,i, pMeCP2 expression does not correlate with locomotor activity in mice that received Veh injection (Veh:Veh and AMPH:Veh) at challenge, r=0.17, _p_=0.26 (Veh:Veh) and r=0,55, _p_=0.096 (AMPH:Veh), n=9 and 10. j, pMeCP2 expression does not correlate with locomotor activity in mice that received a single AMPH injection (Veh:AMPH) at challenge, r= -0.073, _p_= 0.80, n=15. k, pMeCP2 expression strongly correlates with locomotor activity in animals that received repeated AMPH injections and AMPH at challenge (AMPH:AMPH), r=0.61, _p_=0.0087, n=17. Error bars indicate mean ± s.e.m.

Comment in

Similar articles

Cited by

References

    1. Hyman SE, Malenka RC, Nestler EJ. Neural Mechanisms of Addiction: The Role of Reward-Related Learning and Memory. Annu Rev Neurosci. 2006 - PubMed
    1. Robinson TE, Kolb B. Persistent structural modifications in nucleus accumbens and prefrontal cortex neurons produced by previous experience with amphetamine. J Neurosci. 1997;17:8491–7. - PMC - PubMed
    1. Huang YH, et al. In vivo cocaine experience generates silent synapses. Neuron. 2009;63:40–7. - PMC - PubMed
    1. Thomas MJ, Beurrier C, Bonci A, Malenka RC. Long-term depression in the nucleus accumbens: a neural correlate of behavioral sensitization to cocaine. Nat Neurosci. 2001;4:1217–23. - PubMed
    1. Nestler EJ. Molecular basis of long-term plasticity underlying addiction. Nature Reviews Neuroscience. 2001;2:119–128. - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources