Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Epigenetic reprogramming of cortical neurons through alteration of dopaminergic circuits


Alterations of the dopaminergic system are associated with the cognitive and functional dysfunctions that characterize complex neuropsychiatric disorders. We modeled a dysfunctional dopaminergic system using mice with targeted ablation of dopamine (DA) D2 autoreceptors in mesencephalic dopaminergic neurons. Loss of D2 autoreceptors abolishes D2-mediated control of DA synthesis and release. Here, we show that this mutation leads to a profound alteration of the genomic landscape of neurons receiving dopaminergic afferents at distal sites, specifically in the prefrontal cortex. Indeed, we observed a remarkable downregulation of gene expression in this area of ~2000 genes, which involves a widespread increase in the histone repressive mark H3K9me2/3. This reprogramming process is coupled to psychotic-like behaviors in the mutant mice. Importantly, chronic treatment with a DA agonist can revert the genomic phenotype. Thus, cortical neurons undergo a profound epigenetic reprogramming in response to dysfunctional D2 autoreceptor signaling leading to altered DA levels, a process that may underlie a number of neuropsychiatric disorders.


Epigenetic reprogramming is at the heart of cellular differentiation.1 Critical control by chromatin remodeling occurs also in fully differentiated neurons upon pharmacologic and behavioral challenges.2 However, how defective neuronal circuits impinge on the specificity and extent of epigenetic programs has not been fully elucidated. Dysfunctions of dopamine (DA) transmission have been causally linked to cognitive and functional impairments of a number of neuropsychiatric disorders, such as schizophrenia. Indeed, the DA hypothesis of schizophrenia is one of the most credited hypotheses for the genesis of this disorder, based also on the therapeutic benefits of DA antagonists in patients.3, 4, 5, 6, 7, 8, 9, 10, 11

DA release is strictly regulated by convergent signaling from different neurotransmitters12 on dopaminergic neurons, as well as by proteins involved in the presynaptic control of this function. The DA D2 receptor (D2R) is responsible for inhibiting the synthesis and release of DA from dopaminergic neurons, acting as an autoreceptor.13,14 Recently, we generated site-specific knockout mice, lacking D2R selectively from dopaminergic neurons (hereafter referred as DA-D2RKO).15 These mice allow addressing in vivo the consequences of D2 autoreceptor deficiency. As expected, these mice are characterized by loss of inhibition of DA synthesis and altered DA release; electrochemical analyses showed diminished DA release after single or multiple pulses.15 Nevertheless, when DA-D2RKO mice are given cocaine, which blocks the DA transporter inhibiting DA reuptake, these mice have a much heightened motor response to the drug indicating increased levels of DA in the striatum. Thus, absence of D2 autoreceptors eliminates the D2R-mediated cell-autonomous control of DA levels both in normal conditions and in response to stimuli.15

In the present study, we used these mice to model a dysfunctional DA system where to analyze the consequences of altered DA levels on areas receiving dopaminergic fibers at the molecular and behavioral levels. With this purpose, we focused on areas involved in neuropsychiatric disorders, the prefrontal cortex (PFC) and striatum. Our study reveals that dysfunctional control of DA synthesis and release results in an unexpected reprogramming of gene expression in the PFC linked to epigenetic modifications of chromatin accompanied by cognitive and behavioral deficits. These studies highlight the critical role of proper DA signaling in the cortex and suggest that, in some cases, neuropsychiatric disorders might arise from dysfunctional neurotransmission.

Materials and methods

Detailed descriptions are found in Supplementary Methods (SM).

Animals and treatments

DA-D2RKO15 and wild-type (WT) littermates were housed under standard conditions (12 h light/dark cycles) with food and water ad libitum. All experiments were performed using 8- to 12-week-old mice (unless otherwise specified) in accordance to the Institutional Animal Care and Use Committee of the University of California Irvine, National Institute of Health and institutional guidelines. Clozapine (Sigma, St Louis, MO, USA) was dissolved in 0.1N HCl and then diluted in 0.9% NaCl (adjusted to pH 7), and 3 mg kg−1 were administered (i.p.) for 21 days. Quinpirole (Sigma) in 0.9% NaCl was injected (i.p.) at 0.2 mg kg−1 for 15 days. Amphetamine (Sigma) in 0.9% NaCl was administered (i.p.) at 1 and 3 mg kg−1.

Brains were quickly extracted, placed on ice and tissue punches for RNA, mass spectrometry (MS) and western analyses were obtained from 1 mm slices of the PFC (bregma 3.5–2.5 mm) and striatum (bregma 1.5–0.5 mm) using a coronal brain matrix (Harvard apparatus for adult mice ~30 g). Nissl staining of brain sections are described in SM.

RNA preparation and microarray experiments

Total RNA was extracted using mini RNeasy Kit (Qiagen, Valencia, CA, USA). cDNAs were obtained from 100 ng of total RNA per sample (GeneChip cDNA synthesis Kit; Affymetrix, Santa Clara, CA, USA) and hybridized to Mogene (Affymetrix 'MoGene 1.0 ST') by DNA Microarray Core Facility (UCI). Four (PFC) and three (ventral striatum, vStr) microarrays per genotype were used. Normalization and analysis were performed using Partek Genomics Suite software (St Louis, MO, USA), Genecodis software (Madrid, Spain) and DAVID (Frederick, MD, USA) (see SM).


Total RNA was extracted with TRIzol and cDNAs prepared using M-MLV Kit (Invitrogen, Carlsbad, CA, USA). In all, 25–50 ng cDNA per sample was used and PCR was performed with iQSYBR Green Supermix (Bio-Rad, Hercules, CA, USA) using the following settings: 95 °C for 10 min, 40 cycles at 95 °C for 15 s and 60 °C for 45 s. See SM for primer sequences.

Western blot analyses

Frozen tissue punches of the relevant regions were homogenized and boiled in 1% sodium dodecyl sulfate. For clozapine and quinpirole treatments, tissues were obtained 24 h after the last injection. Extracts (50 μg) were separated on sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto membranes. Antibodies used were as follows: anti-H3K9me2 (1:5000) and anti-H3K9me3 (1:1000) from Active Motif (Carlsbad, CA, USA); anti-H3K4me3 (1:1000) and anti-H3 (1:1000) from Abcam (Cambridge, MA, USA); anti-GAPDH (1:20,000) from Millipore (Billerica, MA, USA); anti-Akt1 (1:1000) from Cell Signaling (Danvers, MA, USA); and anti-NR4A2 (1:1000) from Antibody Verify (Las Vegas, NV, USA). Secondary anti-rabbit and/or anti-mouse antibodies (1:5000) were from Millipore. Western blots were revealed using ECL Plus (GE Healthcare, Pittsburgh, PA, USA or Westdura, Pierce, Rockford, IL, USA). Quantifications were performed using the NIH ImageJ (version 1.42q) software (Bethesda, MD, USA).

MS analyses

PFCs and striatal punches from WT and DA-D2RKO mice were used for MS analyses of histone modifications (n=4/5 per genotype per area) performed as described previously16 (see SM).

Immunohistochemical analyses

Immunostainings were performed as described in SM on 10 μm coronal brain sections using rabbit anti-H3K9me3 (1:500; Active Motif), mouse anti-GAD67 (1:500; Abcam) and mouse anti-NeuN (1:500; Millipore) antibodies, followed by anti-rabbit CY3 (1:500; Jackson Laboratories, Sacramento, CA, USA) or anti-mouse Alexa-488 (1:800; Jackson Laboratories) or anti-mouse Alexa-546 (1:500; Jackson Laboratories) antibodies. Nuclei were counterstained with Draq5 (1:500; Enzo Life Science, Farmingdale, NY, USA). Images were taken at the confocal (DMRE; SP5 Leica, Buffalo Grove, IL, USA) and quantified using LAS AF (Leica) in three regions of interest of 145 × 145 μm2 per image (in Figure 2f n=16–18 images per mouse, 3 mice per genotype; in Figure 3f n=14 images per mouse, 3 mice per genotype). Data represent ratios of pixel fluorescence/Draq5 per regions of interest.

Chromatin immunoprecipitation

Chromatin immunoprecipitation (ChIP) experiments were performed as described previously.17 PFC and Str punches from six mice per genotype were pooled and DNA sheered by sonication (five times, 30 s (40% power) every 30 s). Five microliters of anti-H3K9me3 or anti-IgG antibodies per sample were used. Recovered DNA was subjected to quantitative reverse transcriptase-PCR (qRT-PCR) (98 °C for 3 min, 40 cycles at 98 °C for 30 s and 60 °C for 60 s). See SM for primer sequences.

Behavioral experiments

Acoustic startle reflex and prepulse inhibition

Acoustic startle reflex and prepulse inhibition (PPI) were performed as described previously18,19 on independent mouse groups (see SM).

Motor activity

Mice were habituated for 2 h to a novel home cage, and then administered either saline or amphetamine at 1 or 3 mg kg−1 (i.p.), and the activity was recorded for 1h. Mice injected with either saline or clozapine (3 mg kg−1) were analyzed in the open field (a white square box of 30 × 30 cm2; 70 lx) and their motor activity analyzed for 15 min (Supplementary Figure 5) at 24 h after the last injection. The activity was recorded by a video-tracking system (Viewpoint, Lyon, France).

Radial maze

The 8-arm radial maze test was performed to investigate both spatial and working memory exactly as described.20 Time to complete the task, working memory errors (re-entry into a previously visited arm) and reference memory errors (entry into an arm without bait) were recorded and scored for 15 days after 6 days of acquisition.

Statistical analyses

Statistical analyses for microarrays were performed as described in Supplementary Methods section. For western blot, immunohistochemical, qRT-PCR and ChIP experiments, statistical comparisons were performed using Prism 3.0 (GraphPad, San Diego, CA, USA) or SPSS 19 softwares (IBM, Armonk, NY, USA) and consisted of t-tests, one-way, two-way or two-way with repeated-measures analyses of variance as indicated, followed by appropriate post hoc analyses. For behavioral experiments, acoustic startle reflex, PPI and radial maze values were corrected by the Mauchly’s test if sphericity was violated.


Distinct gene expression profiles in the PFC and striatum of DA-D2RKO mice

To elucidate the impact of a defective DA signaling on gene expression, we performed unbiased microarray profiling on two areas targeted by the DA mesolimbic pathway, the PFC and vStr. Strikingly, analysis of the PFC of DA-D2RKO revealed a robust, widespread downregulation of gene expression (Figure 1a) (1930 genes (Supplementary Table 1), P<0.05, 8.6% of the total transcripts, of which 1809 genes annotated) as compared with WT (Figure 1b), whereas the expression of only two genes (<0.009% of transcripts) was upregulated (Supplementary Table 1). By contrast, analyses of the vStr revealed a total of 132 genes differentially expressed between DA-D2RKO and WT mice (Figure 1b, Supplementary Figure 1a and Supplementary Table 2). Among these, only 20 genes were downregulated (Supplementary Figure 1b) in DA-D2RKO versus WT vStr, with a negligible overlap between downregulated genes expressed in both areas (Supplementary Figure 1b). The expression of 112 genes was instead upregulated in the vStr (P<0.05) (Supplementary Table 2). Gene ontology analyses of genes differentially expressed in DA-D2RKO PFC revealed that a preponderant number is involved in the regulation of transcription (240 genes; P=1.8E−47); importantly, among these genes, 15% participate in chromatin remodeling (Figure 1c and Supplementary Table 3). In contrast, gene ontology analyses of the DA-D2RKO vStr identified 15 differentially expressed genes (P=0.042) involved in the regulation of transcription, but none in chromatin remodeling (Supplementary Figure 1c and Supplementary Table 4). Pathway comparisons between PFC and vStr of DA-D2RKO mice showed virtual absence of common alterations in specific pathways (Supplementary Figure 2 and Supplementary Tables 3). These results demonstrate that ablation of D2 autoreceptors in DAergic neurons generates site-specific modifications of gene expression in the areas targeted by the mesolimbic pathway.

Figure 1

Widespread downregulation of cortical genes expression in DA-D2RKO mice. (a) Representative heat map of statistically significant (P<0.05) differentially expressed genes between DA-D2RKO and WT PFCs. (b) Percentage of genes differentially expressed relative to WT in the vStr and PFC of DA-D2RKO. (c) Gene ontology (GO) analysis of the DA-D2RKO PFC transcriptome. Pie chart (left) of the most significantly enriched biological processes; the pie chart on the right specifies the function of genes contained in the most enriched function: regulation of transcription (P=1.8E−47).

PowerPoint slide

Extensive epigenetic reprogramming in the PFC of DA-D2RKO mice

Transcript downregulation in the PFC was confirmed for a number of randomly selected genes (Supplementary Table 5). To dissect the molecular mechanisms underlying such widespread effects on gene expression, we analyzed posttranslational histone modifications.1,21, 22, 23 H3K9me2/3 (ref. 24) is a well-characterized repressive mark, with dimethylated (H3K9me2) or trimethylated (H3K9me3) modifications enriched in heterochromatic portions of the genome.25 Western blot analyses of PFC and vStr revealed a significant increase of both H3K9me2 and H3K9me3, especially in the PFC of DA-D2RKO mice as compared with WT littermates (Figures 2a and b), but not in the striatum (Figures 2a and c); H3K4me3 and H3K27me3 were unmodified in both regions (Figures 2a and c). To unequivocally confirm these results, we performed quantitative MS using extracts from the PFC and striatum of DA-D2RKO and WT mice. Interestingly, while the striatum of DA-D2RKO mice showed only a slight increase of H3K9me2/3 as compared with WT samples in MS analyses, we observed a robust increase of H3K9me2/3 specific to the PFC of DA-D2RKO mice as compared with WT PFCs or to the striatum of WT and mutant mice (Figure 2d and Supplementary Figure 3).

Figure 2

Epigenetic modifications in the PFC of DA-D2RKO mice. (a) Representative western blot analyses of histone H3K9, K4 and K27 modifications (as indicated) in PFC and striatal extracts from WT (lanes 1 and 3) and DA-D2RKO (lanes 2 and 4) mice. (b–d, f) WT: white bars; DA-D2RKO: black bars. Values are mean±s.e.m. (b and c) Quantifications of experiments in (a) (WT values arbitrarily set to 1) show a specific increase of H3K9me2 and me3 in the PFC of mutants (Student’s t-test: **P<0.01) (n=3 WT/5 DA-D2RKO per area). Loaded quantities were normalized to H3 total (H3tot) and GAPDH; (d) box-plot representation of tandem MS/MS analysis of H3K9me2/3 in the PFC and striatum extracts of both genotypes (n=4 WT/5 DA-D2RKO per area). (e) Representative images of Nissl staining of WT PFC (left panel; layers are indicated) and of H3K9me3 immunofluorescence of the same region in WT and DA-D2RKO, as indicated. Inset: magnifications of layer V. Scale bars: 100μm; inset: 20 μm. (f) Quantifications of experiments in (e) (WT intensity set as 1) showed significantly increased H3K9me3 staining in mutant PFCs (Student’s t-test: **P<0.01) (n=3 WT/3 DA-D2RKO).

PowerPoint slide

Immunofluorescence analyses using H3K9me3 antibodies supported these findings showing a 20% increase in the overall intensity of H3K9me3 labeling in PFC brain sections of DA-D2RKO mice (Figures 2e and f), with a more sustained increase in deep layers (V and VI) (Figure 2e (inset) and Supplementary Figure 4a). Double immunofluorescence analyses using the neuronal markers NeuN (Supplementary Figure 4b) or GAD67 to label interneurons (Supplementary Figure 4c) showed that the increased H3K9me3 staining is localized in NeuN+ neurons of DA-D2RKO PFC. Importantly, similar neuronal density was found in the PFC of WT and DA-D2RKO mice (Supplementary Figure 4d).

Silencing histone modifications associated with genes involved in psychiatric disorders

Analyses of functions of genes downregulated in the PFC (Ingenuity IPA software, Redwood City, CA, USA) revealed the prominent presence of 48 genes previously involved in neurologic/psychiatric disorders (P<10−13) in the DA-D2RKO mice, including 36 genes directly implicated in schizophrenia (Supplementary Table 6). Among these, we selected two genes, Nr4a2 (with highest differential expression from WT; 1.5-fold) and Akt1 (with the lowest differential expression from WT; 1.2-fold) to link the changes in gene expression to chromatin remodeling. These genes were also favored among others because reduced expression of both has been associated with DA-related disorders.26,27 Their expression profile (together with that of other randomly selected genes; Supplementary Table 5) confirmed the microarray data, with a decrease of mRNA expression for both genes in the DA-D2RKO PFC as compared with WT (Figure 3a). In contrast to the PFC, comparison of Nr4a2 and Akt1 expression patterns in the striatum of DA-D2RKO and WT mice showed increased Nr4a2 expression (Figure 3b), but no difference for Akt1 expression in mutant mice. ChIP17 experiments using H3K9me3 antibodies demonstrated a remarkable enrichment of this mark on both Nr4a2 and Akt1 promoters in the PFC of DA-D2RKO mice as compared with WT littermates (Figure 3c). In the striatum, H3K9me3 ChIPs showed either reduced (Nr4a2) or similar enrichments (Akt1) of their promoters in DA-D2RKO as compared with WT samples (Figure 3d), mirroring mRNA expression (Figure 3b). A corresponding reduction in the protein levels of NR4A2 and AKT1 was found in PFC protein extracts from DA-D2RKO mice (Figure 3e). Immunofluorescence analyses showed that increased H3K9me3 in the PFC is not present at birth, but it appears postnatally, becomes significant by the 4th week after birth and persists into adulthood (Figure 3f).

Figure 3

Silencing of Nr4A2 and Akt1 genes in the PFC of DA-D2RKO mice is associated with H3K9me3. (af) WT: white bars; DA-D2RKO: black bars. Values are means±s.e.m. (a) qRT-PCR of Nr4a2 and Akt1 mRNA expression in the PFC and (b) striatum (n=4 WT/6 DA-D2RKO per area) (WT value set as 1) (Student’s t-test: *P<0.05; **P<0.01). (c) Quantitative H3K9me3 ChIP on Nr4a2 (n=4 per genotype ), Akt1 (n=3 per genotype) and Myoglobin G promoters in the PFC and (d) striatum extracts (Student’s t-test; *P<0.05) (e) Western blot analyses (left) and quantifications (right) of NR4A2 (n=4 WT/5 DA-D2RKO), AKT1 (n=6 WT/7 DA-D2RKO) and GAPDH protein levels in WT (lane 1) and DA-D2RKO mice (lane 2) expressed as fold changes from WT (Student’s t-test: *P<0.05; **P<0.01). (f) Quantification of H3K9me3 immunostaining of sections from WT and DA-D2RKO PFCs during postnatal life, as indicated. A significant genotype × age interaction is noted (F(5,30)=2.48 P=0.0337) (n=3 mice per genotype).

PowerPoint slide

Psychotic-like behaviors in DA-D2RKO mice

The PFC-specific widespread epigenetic reprogramming and the newly acquired gene expression profile of the mutant mice prompted us to test DA-D2RKO mice in paradigms associated with psychotic-like behaviors.28 Hyperactivity in a novel environment and heightened response to the motor effect of amphetamines have been used to model psychotic behaviors,28,29 DA-D2RKO mice show both phenotypes; indeed, they are hyperactive in a novel environment,15 and display an increased motor reactivity to amphetamine administration (Supplementary Figure 5a). Interestingly, hyperactivity in a novel environment can be reversed by the administration of clozapine,30 an atypical antipsychotic (Supplementary Figure 5b).

We then evaluated the sensorimotor gating of DA-D2RKO mice by testing the startle reflex and PPI in comparison with WT mice. PPI is normally impaired in animal models of psychosis.29 DA-D2RKO mice exposed to pulses of different intensities showed a significantly amplified startle reflex to high pulse intensities as compared with WT controls (Figure 4a). Importantly, they also showed a significant reduction of PPI as compared with WT mice to the lowest startle-eliciting pulse (Figure 4b).

Figure 4

Behavioral alterations of DA-D2RKO mice. (ae) WT: white bars; DA-D2RKO: black bars. Values are mean±s.e.m. (a) Startle reflex in WT and DA-D2RKO mice to pulses of increasing intensity (decibels: dB) indicates significant genotypexpulse intensity interaction (F(1.19, 52.24)=4.027, P=0.043), genotype effect (F(1,46)=8.534, P=0.005) and pulse intensity effect (F(1.19, 52.24)=33.432, P=0.000) (Bonferroni’s: *P<0.05). (b) PPI experiments indicate significant genotypexPPI interaction (F(2, 53)= 3.409, P=0.037), and prepulse intensity effect (F(2, 53)= 143.555, P=0.000) (Bonferroni’s: **P<0.01). (ce) WT and DA-D2RKO mice were tested in the radial maze for 15 days; results were grouped into five blocks of three sessions each. Analyses of radial maze results indicate significant differences between WT and DA-D2RKO mice in (c) working memory errors (number of re-entries into previous visited arms), genotypexsession interaction (F(3.43, 106.33)=2.828, P=0.035) and (d) in the time to complete the task genotypexsession interactions (F(3.07, 95.31)=3.339, P=0.022), genotype effect (F(1,33)=8.243, P=0.007) and trial effect (F(3.07, 95.31)=20.670, P=0.000) (Bonferroni’s: *P<0.05, **P<0.01). (e) Analyses of reference memory errors in the radial maze (entrance in an unbaited arm) revealed the absence of significant interactions (genotypextrials: F(4, 124)= 2.306, P=0.062; genotype effect: F(1,33)=0.613, P=0.440).

PowerPoint slide

PFC-dependent disorders are characterized by deficits of working memory.31,32 Thus, DA-D2RKO and WT mice were tested in the radial maze.20 Statistical analyses showed that DA-D2RKO mice had an impaired working memory (Figures 4c and d) but a normal reference memory in this test as compared with WT littermates (Figure 4e). Altogether, these studies show that subtle modifications of DA signaling (ablation of D2 autoreceptors) cause PFC-specific cellular and behavioral impairments.

Quinpirole, a D2R agonist, reverts epigenetic silencing

Our previous studies using DA-D2RKO mice showed a dual regulation of DA release orchestrated by D2 autoreceptors and by postsynaptic D2R-mediated inhibitory feedbacks on DA neurons.15 We thus analyzed whether chronic postsynaptic D2R stimulation might reverse the cellular phenotype of DA-D2RKO mice. Accordingly, DA-D2RKO mice were treated daily for 2 weeks with the D2R agonist quinpirole. Interestingly, in these conditions, we observed a reversal of the PFC H3K9me3 to WT levels (Figure 5a); in agreement with these data, analyses performed at the mRNA level by qRT-PCR showed that quinpirole re-established the expression levels of selected differentially expressed genes (Supplementary Table 5) in DA-D2RKO PFCs (Supplementary Figure 6) to WT levels. Concomitant to the effect of quinpirole on H3K9me3 levels in DA-D2RKO PFC, we found restored expression of AKT1 and NR4A2 proteins to WT levels (Figures 5c and d). Conversely, chronic clozapine was unable to reverse H3K9me3 (Figure 5b) or to restore protein expression for AKT1 and NR4A2 (Supplementary Figure 7). These results suggest that the PFC phenotype of DA-D2RKO mice is dependent of altered DA-mediated regulation of cortical neurons. In this respect, dysfunctional DA signaling has been associated with human neuropsychiatric disorders.3,33

Figure 5

Stimulation of D2Rs reverses the molecular phenotype of DA-D2RKO PFCs. (a, b and d) WT: white bars; DA-D2RKO: black bars. Values are expressed as mean±s.e.m. Quantifications and representative of western blot analyses of H3K9me3 levels in PFC extracts of WT and DA-D2RKO mice chronically treated with saline (Sal) or (a) quinpirole (Quin) or (b) clozapine (Cloz). A significant effect of quinpirole was noted in DA-D2RKO vs WT mice, genotypextreatment interaction (F(1,9)=11.31; P=0.0083) and treatment effect (F(1,9)=6.75; P=0.0289) (Bonferroni’s: *P<0.05, ##P<0.01). Conversely, clozapine revealed no effect of the treatment (F(1,14)=0.61, P=0.4464) but a significant genotype difference (F(1,14)=9.68; P=0.0067) (Student’s t-test: *P<0.05). (c) Representative western blot analyses of AKT1 and NR4A2 levels in the PFC of WT and DA-D2RKO mice chronically treated with quinpirole and (d) quantifications of experiments as shown in (c). Data show that the observed decrease of AKT1 and NR4A2 protein levels in the PFC of DA-D2RKO mice can be rescued upon chronic quinpirole administration to WT levels for AKT1 genotypextreatment interaction (F(1,19)=5.78; P=0.0266) and treatment effect (F(1,19)=5.09; P=0.0360) (Bonferroni’s: **P<0.01, ##P<0.01), and for NR4A2, genotype (F(1,19)=4.42; P=0.0490) and treatment effect (F(1,19)=11.33; P=0.0032) (Student’s t-test: **P<0.01, ##P<0.01).

PowerPoint slide


Widespread epigenetic changes have been proposed to underlie alterations in gene expression associated with addiction and complex neuropsychiatric disorders, including schizophrenia.2,22,23 In this article, we present novel findings showing that altered control of DA synthesis and release due to ablation of D2 autoreceptors from dopaminergic neurons during development leads to major changes in the PFC, an area targeted by DA neurons of the mesolimbic pathway. Interestingly, we report a massive downregulation of gene expression affecting ~2000 genes in DA-D2RKO PFC; this effect is paralleled by increased levels of the repressive histone mark H3K9me2/3. Gene ontology analyses revealed that a large number of downregulated PFC genes are involved in transcriptional regulation and chromatin remodeling, including the demethylases KDM4b and KDM4c that specifically target H3K9me3 (refs 25, 34) (Supplementary Table 5), as well as genes involved in protein phosphorylation, transport and metabolic processes (Supplementary Table 3). Future studies will address whether reduced expression of these proteins is the leading event of the phenotype observed in DA-D2RKO mice. We show that H3K9me repressive chromatin mark in the PFC of DA-D2RKO mice directly targets at least two genes linked to schizophrenia, Akt1 27 and NR4A2.26 Thus, we establish a direct link between DAergic signaling, chromatin remodeling and the expression of genes implicated in psychiatric disorders. Importantly, we show that exposing mice to chronic treatment with quinpirole reverses histone methylation to normal levels, as reflected by western blot analyses of this modification on the PFC of DA-D2RKO-treated mice as well as on the expression of Akt1 and NR4A2. Our findings might contribute to understanding the limited efficacy of antipsychotics in the treatment of schizophrenia.3,4,9

Indeed, it is tempting to speculate that the use of D2R antagonists, while appropriate to treat symptoms associated with heightened DA signaling in the striatum, are inefficient at improving PFC-linked functions, likely linked to lowered DA stimulation.

The strikingly different expression profile in the vStr of DA-D2RKO mice, where virtually neither transcriptional silencing nor increase in H3K9me3 are present, stresses the specificity of the widespread reprogramming occurring in PFC neurons and underscores the differences between striatal and cortical circuits. Thus, loss of D2 autoreceptors from DAergic neurons generates profound epigenetic modifications in brain areas targeted by DAergic fibers at distal sites. It is tempting to speculate that a similar reprogramming might operate in human psychiatric disorders.


  1. 1

    Jaenisch R, Bird A . Epigenetic regulation of gene expression: how the genome integrates intrinsic and environmental signals. Nat Genet 2003; 33 (Suppl): 245–254.

    CAS  Article  Google Scholar 

  2. 2

    Borrelli E, Nestler EJ, Allis CD, Sassone-Corsi P . Decoding the epigenetic language of neuronal plasticity. Neuron 2008; 60: 961–974.

    CAS  Article  Google Scholar 

  3. 3

    Howes OD, Kapur S . The dopamine hypothesis of schizophrenia: version III—the final common pathway. Schizophr Bull 2009; 35: 549–562.

    Article  Google Scholar 

  4. 4

    Kuepper R, Skinbjerg M, Abi-Dargham A . The dopamine dysfunction in schizophrenia revisited: new insights into topography and course. Handb Exp Pharmacol 2012; 212: 1–26.

    CAS  Article  Google Scholar 

  5. 5

    Lyon GJ, Abi-Dargham A, Moore H, Lieberman JA, Javitch JA, Sulzer D . Presynaptic regulation of dopamine transmission in schizophrenia. Schizophr Bull 2011; 37: 108–117.

    Article  Google Scholar 

  6. 6

    Miyake N, Thompson J, Skinbjerg M, Abi-Dargham A . Presynaptic dopamine in schizophrenia. CNS Neurosci Ther 2011; 17: 104–109.

    CAS  Article  Google Scholar 

  7. 7

    Heinz A, Romero B, Gallinat J, Juckel G, Weinberger DR . Molecular brain imaging and the neurobiology and genetics of schizophrenia. Pharmacopsychiatry 2003; 36 (Suppl 3): S152–S157.

    CAS  Google Scholar 

  8. 8

    Howes O, Bose S, Turkheimer F, Valli I, Egerton A, Stahl D, et al. Progressive increase in striatal dopamine synthesis capacity as patients develop psychosis: a PET study. Mol Psychiatry 2011; 16: 885–886.

    CAS  Article  Google Scholar 

  9. 9

    Howes OD, Egerton A, Allan V, McGuire P, Stokes P, Kapur S . Mechanisms underlying psychosis and antipsychotic treatment response in schizophrenia: insights from PET and SPECT imaging. Curr Pharm Des 2009; 15: 2550–2559.

    CAS  Article  Google Scholar 

  10. 10

    Bertolino A, Knable MB, Saunders RC, Callicott JH, Kolachana B, Mattay VS, et al. The relationship between dorsolateral prefrontal N-acetylaspartate measures and striatal dopamine activity in schizophrenia. Biol Psychiatry 1999; 45: 660–667.

    CAS  Article  Google Scholar 

  11. 11

    Meyer-Lindenberg A, Miletich RS, Kohn PD, Esposito G, Carson RE, Quarantelli M, et al. Reduced prefrontal activity predicts exaggerated striatal dopaminergic function in schizophrenia. Nat Neurosci 2002; 5: 267–271.

    CAS  Article  Google Scholar 

  12. 12

    Rice ME, Patel JC, Cragg SJ . Dopamine release in the basal ganglia. Neuroscience 2011; 198: 112–137.

    CAS  Article  Google Scholar 

  13. 13

    Rouge-Pont F, Usiello A, Benoit-Marand M, Gonon F, Piazza PV, Borrelli E . Changes in extracellular dopamine induced by morphine and cocaine: crucial control by D2 receptors. J Neurosci 2002; 22: 3293–3301.

    CAS  Article  Google Scholar 

  14. 14

    Benoit-Marand M, Borrelli E, Gonon F . Inhibition of dopamine release via presynaptic D2 receptors: time course and functional characteristics in vivo. J Neurosci 2001; 21: 9134–9141.

    CAS  Article  Google Scholar 

  15. 15

    Anzalone A, Lizardi-Ortiz JE, Ramos M, De Mei C, Hopf FW, Iaccarino C, et al. Dual control of dopamine synthesis and release by presynaptic and postsynaptic dopamine D2 receptors. J Neurosci 2012; 32: 9023–9034.

    CAS  Article  Google Scholar 

  16. 16

    Jasencakova Z, Scharf AN, Ask K, Corpet A, Imhof A, Almouzni G, et al. Replication stress interferes with histone recycling and predeposition marking of new histones. Mol Cell 2010; 37: 736–743.

    CAS  Article  Google Scholar 

  17. 17

    Kumar A, Choi KH, Renthal W, Tsankova NM, Theobald DE, Truong HT, et al. Chromatin remodeling is a key mechanism underlying cocaine-induced plasticity in striatum. Neuron 2005; 48: 303–314.

    CAS  Article  Google Scholar 

  18. 18

    Paylor R, Crawley JN . Inbred strain differences in prepulse inhibition of the mouse startle response. Psychopharmacology 1997; 132: 169–180.

    CAS  Article  Google Scholar 

  19. 19

    Errico F, Rossi S, Napolitano F, Catuogno V, Topo E, Fisone G, et al. D-aspartate prevents corticostriatal long-term depression and attenuates schizophrenia-like symptoms induced by amphetamine and MK-801. J Neurosci 2008; 28: 10404–10414.

    CAS  Article  Google Scholar 

  20. 20

    Wood MA, Kaplan MP, Brensinger CM, Guo W, Abel T . Ubiquitin C-terminal hydrolase L3 (Uchl3) is involved in working memory. Hippocampus 2005; 15: 610–621.

    CAS  Article  Google Scholar 

  21. 21

    Maze I, Noh KM, Allis CD . Histone regulation in the CNS: basic principles of epigenetic plasticity. Neuropsychopharmacology 2013; 38: 3–22.

    CAS  Article  Google Scholar 

  22. 22

    Akbarian S, Huang HS . Epigenetic regulation in human brain—focus on histone lysine methylation. Biol Psychiatry 2009; 65: 198–203.

    CAS  Article  Google Scholar 

  23. 23

    Russo SJ, Nestler EJ . The brain reward circuitry in mood disorders. Nat Rev Neurosci 2013; 14: 609–625.

    CAS  Article  Google Scholar 

  24. 24

    Barski A, Cuddapah S, Cui K, Roh TY, Schones DE, Wang Z, et al. High-resolution profiling of histone methylations in the human genome. Cell 2007; 129: 823–837.

    CAS  Article  Google Scholar 

  25. 25

    Fodor BD, Kubicek S, Yonezawa M, O'Sullivan RJ, Sengupta R, Perez-Burgos L, et al. Jmjd2b antagonizes H3K9 trimethylation at pericentric heterochromatin in mammalian cells. Genes Dev 2006; 20: 1557–1562.

    CAS  Article  Google Scholar 

  26. 26

    Rojas P, Joodmardi E, Hong Y, Perlmann T, Ogren SO . Adult mice with reduced Nurr1 expression: an animal model for schizophrenia. Mol Psychiatry 2007; 12: 756–766.

    CAS  Article  Google Scholar 

  27. 27

    Emamian ES, Hall D, Birnbaum MJ, Karayiorgou M, Gogos JA . Convergent evidence for impaired AKT1-GSK3beta signaling in schizophrenia. Nat Genet 2004; 36: 131–137.

    CAS  Article  Google Scholar 

  28. 28

    Arguello PA, Gogos JA . Modeling madness in mice: one piece at a time. Neuron 2006; 52: 179–196.

    CAS  Article  Google Scholar 

  29. 29

    Jones CA, Watson DJ, Fone KC . Animal models of schizophrenia. Br J Pharmacol 2011; 164: 1162–1194.

    CAS  Article  Google Scholar 

  30. 30

    Seeman P . All roads to schizophrenia lead to dopamine supersensitivity and elevated dopamine D2(high) receptors. CNS Neurosci Ther 2011; 17: 118–132.

    CAS  Article  Google Scholar 

  31. 31

    Yoon T, Okada J, Jung MW, Kim JJ . Prefrontal cortex and hippocampus subserve different components of working memory in rats. Learn Mem 2008; 15: 97–105.

    Article  Google Scholar 

  32. 32

    Papaleo F, Yang F, Garcia S, Chen J, Lu B, Crawley JN, et al. Dysbindin-1 modulates prefrontal cortical activity and schizophrenia-like behaviors via dopamine/D2 pathways. Mol Psychiatry 2012; 17: 85–98.

    CAS  Article  Google Scholar 

  33. 33

    Goto Y, Grace AA . The dopamine system and the pathophysiology of schizophrenia: a basic science perspective. Int Rev Neurobiol 2007; 78: 41–68.

    CAS  Article  Google Scholar 

  34. 34

    Whetstine JR, Nottke A, Lan F, Huarte M, Smolikov S, Chen Z, et al. Reversal of histone lysine trimethylation by the JMJD2 family of histone demethylases. Cell 2006; 125: 467–481.

    CAS  Article  Google Scholar 

Download references


We thank Drs F Torri and C De Mei for assistance and interest in the initial phase of this study; Dr P Sassone-Corsi for critical discussions and reading of the manuscript. This work was supported by NIH Grant DA024689 and INSERM-44790 (to EB).

Author information



Corresponding author

Correspondence to E Borrelli.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies the paper on the Molecular Psychiatry website

Supplementary information

PowerPoint slides

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Brami-Cherrier, K., Anzalone, A., Ramos, M. et al. Epigenetic reprogramming of cortical neurons through alteration of dopaminergic circuits. Mol Psychiatry 19, 1193–1200 (2014).

Download citation

Further reading


Quick links