The authors have declared that no competing interests exist.
Conceived and designed the experiments: SKP SL YK YUP SAL HL HS. Analyzed the data: SL YK YUP SAL HL HS. Contributed reagents/materials/analysis tools: SL JJ YUP. Wrote the paper: SL SKP. Coordinated the study: SKP.
Chromatin remodeling through histone modifications has emerged as a key mechanism in the pathophysiology of psychiatric disorders. Valproate (VPA), a first-line medication for bipolar disorder, is known to have histone deacetylase (HDAC) inhibitor activity, but the relationship between its efficacy as a mood stabilizer and HDAC inhibitory activity is unclear. Here we provide evidence that prostate apoptosis response-4 (Par-4), an intracellular binding partner of dopamine D2 receptors (DRD2), plays a role in mediating the effectiveness of VPA. We found that chronic VPA treatment enhanced the expression of Par-4 in cultured neurons and adult mouse brains. This Par-4 induction phenomenon occurred at the transcriptional level and was correlated with an increase in histone H3 and H4 acetylation of the Par-4 promoter regions. Furthermore, chronic VPA treatment potentiated the suppression of the cAMP signaling cascade upon dopamine stimulation, which was blocked by sulpiride treatment. These results indicate that VPA potentiates DRD2 activity by enhancing Par-4 expression via a chromatin remodeling mechanism.
Bipolar disorder, also known as manic depression, is a severe mental illness affecting about 1% of the population and has multifaceted potential causes
One of the most intriguing recent developments in the field is the emergence of chromatin remodeling events associated with the pathophysiology of psychiatric disorders
Valproate (VPA) is a short-chain fatty acid with HDAC inhibitor activity that is widely prescribed as one of the first-line medications for epilepsy and bipolar disorders
In this study, we provide evidence that VPA engages in dopamine signaling pathways through the induction of Par-4, an intracellular modulator of DRD2 activity, by mediating chromatin remodeling of the Par-4 promoter region
Par-4 has been proposed as a positive modulator of intracellular DRD2 signaling, which is thought to be associated with normal mood-associated behavior in experimental animals
Time-dependent induction of Par-4 protein in primary cultured neurons. Cultured hippocampal and striatal neurons (DIV 7) were treated with 1 mM VPA. Par-4 protein levels relative to α-tubulin were analyzed by western blot (a) and subject to densitometric analysis (b). Error bars are mean ± SEM values (n = 4; *p<0.05, **p<0.01, ***p<0.001; One-way ANOVA with Bonferroni post hoc test). A. Dose-dependent increase of Par-4 proteins in mouse primary cultured neurons. Cultured primary neurons (DIV 7) were incubated with VPA at doses indicated for 24 hrs, and the Par-4 expression levels were analyzed as in (a). Par-4 levels relative to α-tubulin were analyzed (b). Error bars are mean ± SEM (n = 4; *p<0.05, **p<0.01, ***p<0.001; One-way ANOVA with Bonferroni post hoc test). B. Increased Par-4 mRNA after VPA treatment. Cultured hippocampal neurons (DIV 7) were treated with 1 mM VPA. Fold inductions of Par-4 mRNA relative to GAPDH were assessed by quantitative real-time PCR. Error bars represent mean ± SEM (n = 4; *p<0.05, **p<0.01, ***p<0.001; One-way ANOVA with Bonferroni post hoc test).
Because VPA is known to mediate HDAC inhibition, we tested whether the transcriptional induction of Par-4 is caused by changes in HDAC activities. To accomplish this, we first examined if the Par-4 gene is affected by known HDAC inhibitors, such as sodium butyrate (SB) and trichostatin A (TSA). Cultured hippocampal and striatal neurons (DIV 7) were incubated with 1 mM SB and 100 nM TSA for 30 hrs, and Par-4 protein levels were examined. Chronic treatment with HDAC inhibitors effectively increased the Par-4 protein levels, indicating that the Par-4 gene is sensitive to HDAC inhibition. We also tested if other mood stabilizers including carbamazepine (CBZ), lamotrigine (LTG), and LiCl have a similar effect on Par-4 induction. After 48 hrs of treatment, CBZ, LTG, and LiCl did not significantly increase Par-4 protein levels in cultured hippocampal neurons (
A. Increased Par-4 expression upon chronic treatment of HDAC inhibitors in primary cultured neurons. Cultured neurons (DIV 7) were treated with 1 mM VPA. Sodium butyrate (SB, 1 mM) and trichostatin A (TSA, 100 nM) were treated for 30 hrs, and Par-4 protein levels were analyzed by western blot. B. Effect of common mood stabilizers. Carbamazepine, lamotrigine, and lithium chloride were tested in addition to VPA. Hippocampal neurons (DIV 7) were incubated with the indicated drugs for 48 hrs, and Par-4 protein levels were analyzed by western blot. CBZ, carbamazepine and LTG, lamotrigine. C. Involvement of HDAC4 and HDAC5 in Par-4 induction. Knock-down efficiency of shRNA constructs targeting HDACs was measured by semi-quantitative RT-PCR (a). Increase in Par-4 protein after HDAC knock-down. CAD cells were transfected with shRNA constructs followed by 5 days of differentiation. Cell lysates were analyzed by western blot to detect changes in Par-4 protein levels after knock-down of specific HDACs (b). Fold changes of Par-4 protein levels relative to the control sample (c). Error bars represent mean ± SEM. (n = 3; *p<0.05, **p<0.01, ***p<0.001; Student's t-test).
We next attempted to scrutinize which HDACs are primarily involved in chronic VPA-mediated upregulation of Par-4. For this, we generated shRNA constructs for specific HDACs and confirmed effective knock-down of HDACs using semi-quantitative RT-PCR in differentiated CAD cells, which showed robust VPA-dependent Par-4 induction (
To further investigate the link between VPA and HDAC inhibition, we measured the levels of acetylated histones H3 (ac-H3) and H4 (ac-H4) in the Par-4 promoter region using cultured hippocampal neurons at DIV 7. Using chromatin immunoprecipitation (ChIP) assays with anti-ac-H3 and ac-H4 antibodies, a significant increase in ac-H3 and ac-H4 levels were observed at the Par-4 promoter region after 24 hrs of VPA treatment (
A. Increased acetylation of histone H3 and H4 at the Par-4 promoter region in a time-dependent manner. Hippocampal neurons (DIV 7) were incubated with 1 mM VPA for 24 hrs and subjected to chromatin immunoprecipitation (ChIP) assay (a). Fold changes in ac-H3 and ac-H4 at the Par-4 promoter regions are shown in (b) relative to control. Error bars are mean ± SEM (n = 4; *p<0.05, **p<0.01, ***p<0.001; One-way ANOVA with Bonferroni post hoc test). B. Dose-dependent increases in ac-H3 and ac-H4 after chronic VPA treatment. Hippocampal neurons (DIV 7) were treated with increasing concentrations of VPA for 24 hrs and subject to ChIP assay (a). Relative fold changes in ac-H3 and ac-H4 at the Par-4 promoter regions were normalized to GAPDH levels (b). Error bars are mean ± SEM (n = 4; *p<0.05, **p<0.01, ***p<0.001; One-way ANOVA with Bonferroni post hoc test).
To assess whether these histone modifications occur in physiological conditions, adult mice (C57BL/6, 9 weeks, male) were injected with 100 mg/kg VPA intraperitoneally, as shown in
A. Experimental scheme of daily intraperitoneal VPA injection. ChIP assays were performed using hippocampal lysates. The acute treatment group was injected with saline for 6 days and VPA on the last day. Mice were sacrificed at the indicated time points after the last injection. The chronic treatment group was injected with 100 mg/kg of VPA for 7 days and sacrificed at the indicated time points after the last injection. B. Dynamic changes in histone modifications at the Par-4 promoter region in the brain. Mice (C57BL/6, male, 9 weeks of age) were intraperitoneally injected with saline or VPA (100 mg/kg per day) as indicated in (A). Ac-H3 and ac-H4 levels were normalized to GAPDH, and relative fold changes are shown compared to the CTL group. Error bars are mean ± SEM (n = 3; *p<0.05, **p<0.01, ***p<0.001; One-way ANOVA with Bonferroni post hoc test). C. Induction of Par-4 protein levels by chronic VPA treatment in the adult mouse brain. Par-4 protein levels were measured by western blot from hippocampal and striatal extracts.
Accumulating evidence shows that dopamine signaling may play important roles in the expression of genes that regulate normal mood
A. Altered cAMP-response profile by VPA treatment in cultured striatal neurons. Cultured striatal neurons (DIV 7) were treated with increasing concentrations of dopamine (DA) for 30 min in the presence or absence of 1 mM VPA (24 hrs). cAMP levels in the cell lysates were measured using cAMP enzyme-immunoassays. Error bars are mean ± SEM. (n = 4, *p<0.05, **p<0.01, ***p<0.001; One-way ANOVA with Bonferroni post hoc test). B. VPA-mediated cAMP response is abolished by sulpiride. Cultured striatal neurons (DIV 7) were incubated with VPA (1 mM) for 24 hrs followed by sulpiride (3 µM) and DA (10 µM) treatment. Sulpiride was pretreated for 30 min prior to DA treatment. cAMP levels from the cell lysates were measured using cAMP enzyme-immunoassays. Data are mean ± SEM values. (n = 4; *p<0.05, **p<0.01, ***p<0.001; One-way ANOVA with Bonferroni post hoc test).
Because DRD2 activity has been implicated in normal mood control, we tested if the effect of VPA on dopamine signaling could be attributed to enhanced DRD2 activity using sulpiride, a specific DRD2 antagonist. Cultured striatal neurons (DIV 7) were incubated with 1 mM VPA for 24 hrs. The cells were then pretreated with 3 µM sulpiride for 30 min followed by 10 µM DA for 30 min. Sulpiride decreased the extent of cAMP reduction in VPA-treated neurons, indicating that DRD2 activity was enhanced by chronic VPA treatment (
In this study, we provide evidence that chronic treatment of the common mood stabilizer, VPA, elicits robust induction of Par-4, which significantly impacts DRD2-mediated signaling. Furthermore, the induction mechanism involves dynamic chromatin remodeling in the Par-4 promoter regions and indicates that part of the mood stabilizing efficacy of VPA is mediated by changes in DRD2-mediated signaling processes.
Evidence for a link between the dopamine system and mood disorders has been consistently reported. For example, anhedonia and amotivation commonly seen in mood disorder patients can be reasonably explained by deficits in dopamine circuits
The clinical effects of mood stabilizers and antidepressants are known to take a relatively long time to be achieved, which suggests the involvement of long lasting changes in the cellular environment. Chromatin remodeling has been proposed as a plausible mechanism to explain the delayed but enduring effect of mood stabilizers. In particular, VPA is known to harbor HDAC inhibitory activity that exerts an enduring impact on the chromatin state. In this study, we demonstrated that chronic VPA-associated alterations of DRD2 signaling may result from changes in chromatin states, which enhance the transcription of the Par-4 gene. It is especially noteworthy that Par-4 mRNA and protein accumulation did not increase until after several hours of treatment, which suggests a potential mechanism to explain the delayed clinical effect of VPA in human patients.
Lithium, another first line medication for bipolar disorder, has overlapping but distinct mechanisms with VPA as a mood stabilizer. For example, lithium mediates GSK-3β inhibition, which may be linked to neuroprotective effects against glutamate-induced toxicity
This protocol was approved by the Pohang University of Science and Technology Institutional Animal Care and Use Committee (Approval ID: LIFE 020). This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health, and all efforts were made to minimize suffering.
Hippocampal and striatal regions were dissected from E14.5 embryos in Hank's balanced salt solution (HBSS, Invitrogen) without calcium or magnesium. The dissociated cells were diluted in plating media (Neurobasal for hippocampal neurons and MEM for striatal neurons (Invitrogen) supplemented with 1 M HEPES, pH 7.4, 10% glutamine, and 10% horse serum) and plated into 12-well or 24-well plates coated with poly-D-lysine with laminin. Two to four hours after the plating, the media were replaced with the appropriate culture media supplemented with 0.02% B27, 10% glutamine, and antibiotics.
Cultured neurons at DIV 7 were subjected to the drug treatment as indicated. VPA, sodium butyrate, trichostatin A, lithium chloride, carbamazepine, and lamotrigine were purchased from Sigma. Adult mice (C57BL/6, male, 9 weeks of age) were housed in a 12-h light/dark cycle, and intraperitoneally injected with saline or VPA (100 mg/kg) daily for 7 days. Mice were sacrificed at the indicated time points after the last injection of either VPA (100 mg/kg per day) or saline, and the brains were immediately dissected and processed for experimental procedures. For the cAMP assay, striatal neurons (DIV 7) were pretreated with 1 mM VPA for 24 hrs. After the VPA treatment, an increasing dose of dopamine (Sigma) was added for 30 min. Sulpiride (3 µM, Tocris) was pretreated for 30 min ahead of the dopamine treatment.
Media were replaced with fresh culture media (MEM supplemented with B27, glutamine and antibiotics) 1 day before drug treatment in striatal neurons cultured in 12-well plates. The cells were lysed in 400 µl of 0.1 M HCl solution for 20 min with gentle shaking and spun in microcentrifuge tubes. The cAMP content in the supernatant was measured using the cAMP-Enzyme Immunoassay Kit (Sapphire Bioscience).
Hippocampal tissues and cultured primary neurons were lysed in the buffer containing 10 mM Tris-HCl, pH 8.1, 1% SDS, 0.5% Triton X-100, 10 mM EDTA, 0.5 mM EGTA and protease inhibitor cocktail (Roche). After sonication (amplitude 30, 15 sec, total 10 times, Vibra-Cells, Sonics & Materials, Inc), samples were subjected to ChIP assays immediately using the Upstate Biotechnology ChIP kit (Upstate Biotechnology) with slight modifications from previous reports
Total RNA was extracted using Tri-Solution (Bioscience Technology) from cultured neurons (DIV 7) or differentiated CAD cells. For reverse transcription, 1 µg RNA was used to synthesize first-strand cDNA using the SuperScript III system (Invitrogen). The PCR condition was one cycle of 95°C for 2 min; 21–32 cycles of 95°C for 1 min, 52°C or 56°C for 40 sec, and 72°C for 40 sec; and one cycle of 72°C for 7 min followed by 8°C.
For real-time PCR, 3 µL of 1/5-diluted cDNA and specific primers were mixed in a final volume of 20 µL in the SYBR premix Ex Taq (Takara). Quantitative real-time PCR was performed using the StepOnePlus real-time PCR system (Applied Biosystems). Primer sequences for PCR were;
mPar-4 Forward:
HDAC1 Forward:
HDAC2 Forward:
HDAC4 Forward:
HDAC5 Forward:
GAPDH Forward:
Cultured neurons were lysed in 2X sampling buffer or 1X ELB lysis buffer (250 mM NaCl, 50 mM Tris-HCl, pH 8.0, 5 mM EDTA, 0.1% NP40, and a protease inhibitor cocktail). Protein concentrations were determined by the Bradford method, and 15–50 µg of protein was subjected to SDS-PAGE followed by Western blot analysis. The antibodies used were anti-Par-4 (sc-1807, 1∶2000, Santa Cruz Biotechnology), anti-α-Tubulin (sc-32293, 1∶2000, Santa Cruz Biotechnology), anti-GAPDH (sc-32233, 1∶2000, Santa Cruz Biotechnology), anti-acetylated H3 K9, and K14 (#06-599, 1∶2000, Millipore), anti-acetylated H4 K5, K8, K12 and K16 (#06-866, 1∶2000, Millipore), and anti-mouse and anti-rabbit HRP-labeled secondary antibodies (1∶15000; Thermo Scientific and KPL, respectively).
Central adrenergic tyrosine hydroxylase expressing (CATH) a-differentiated (CAD) cells were transfected using Lipofectamine 2000 (Invitrogen) followed by differentiation, as described previously