Neuronal Per Arnt Sim (PAS) Domain Protein 4 (NPAS4) Regulates Neurite Outgrowth and Phosphorylation of Synapsin I*

Background: NPAS4 is involved in memory formation, but its roles in neuronal function remain to be fully elucidated. Results: NPAS4 increased CDK5-dependent phosphorylation of synapsin I that was associated with neurite elongation. Conclusion: NPAS4 induced CDK5-dependent phosphorylation of synapsin I to facilitate neurite outgrowth. Significance: NPAS4 plays an important role in the structural and functional plasticity of neurons. Neuronal Per Arnt Sim domain protein 4 (NPAS4), a brain-specific basic helix-loop-helix transcription factor, has recently been shown to regulate the development of the GABAergic inhibitory synapses and transcription program for contextual memory formation in the hippocampus. We previously reported that chronic social isolation or restriction stress in mice resulted in an impairment in memory and emotional behavior, which was associated with a decrease in Npas4 mRNA levels. In this study, we investigated the role of NPAS4 in neuronal function in vitro and in vivo. Differentiation medium-induced neurite outgrowth was inhibited in Npas4 knockdown Neuro2a cells, whereas overexpression of NPAS4 accelerated the neurite outgrowth in Neuro2a cells. Furthermore, depolarization-induced neurite outgrowth was abolished in Npas4 KO hippocampal neurons. NPAS4 overexpression increased cyclin-dependent kinase 5 (CDK5)-dependent synapsin I phosphorylation in Neuro2a cells and primary cultured hippocampal neurons. A CDK5 inhibitor, roscovitine, inhibited the neurite outgrowth and the increase in phosphorylated synapsin I (p-SYN I) levels in Npas4-overexpressed Neuro2a cells. Interaction of NPAS4 with promoters of Cdk5 and NeuN genes was demonstrated by a chromatin immunoprecipitation assay. In an in vivo study, pentylenetetrazole-induced convulsions in mice resulted in an increase in NPAS4 and p-SYN I levels in the prefrontal cortex of wild-type mice, although no changes in p-SYN I levels were observed in Npas4 knock-out mice. These results suggest that NPAS4 plays an important role in the structural and functional plasticity of neurons.

Neuronal Per Arnt Sim domain protein 4 (NPAS4), a brainspecific basic helix-loop-helix transcription factor, has recently been shown to regulate the development of the GABAergic inhibitory synapses and transcription program for contextual memory formation in the hippocampus. We previously reported that chronic social isolation or restriction stress in mice resulted in an impairment in memory and emotional behavior, which was associated with a decrease in Npas4 mRNA levels. In this study, we investigated the role of NPAS4 in neuronal function in vitro and in vivo. Differentiation medium-induced neurite outgrowth was inhibited in Npas4 knockdown Neuro2a cells, whereas overexpression of NPAS4 accelerated the neurite outgrowth in Neuro2a cells. Furthermore, depolarization-induced neurite outgrowth was abolished in Npas4 KO hippocampal neurons. NPAS4 overexpression increased cyclin-dependent kinase 5 (CDK5)-dependent synapsin I phosphorylation in Neuro2a cells and primary cultured hippocampal neurons. A CDK5 inhibitor, roscovitine, inhibited the neurite outgrowth and the increase in phosphorylated synapsin I (p-SYN I) levels in Npas4-overexpressed Neuro2a cells. Interaction of NPAS4 with promoters of Cdk5 and NeuN genes was demonstrated by a chromatin immunoprecipitation assay. In an in vivo study, pentylenetetrazole-induced convulsions in mice resulted in an increase in NPAS4 and p-SYN I levels in the prefrontal cortex of wild-type mice, although no changes in p-SYN I levels were observed in Npas4 knock-out mice. These results suggest that NPAS4 plays an important role in the structural and functional plasticity of neurons.
The basic helix-loop-helix family has been studied for its important roles in various physiological and pathophysiological processes, including as transcription factors, in the regulation of neuronal differentiation and even in psychiatric disorders (1)(2)(3)(4)(5)(6). Neuronal Per Arnt Sim domain protein (NPAS) is a member of the basic helix-loop-helix family, which is categorized as Npas1, Npas2, Npas3, and Npas4 (7)(8)(9). Npas3 has been identified as a candidate gene associated with schizophrenia, learning disabilities, and bipolar disorder (5,10), and it is known to play a role in hippocampal neurogenesis (1,11). NPAS4 is a neuron-specific transcription factor (12), and the Npas4 expression level in the hippocampus is regulated by cerebral ischemic insults, the AMPA receptor agonist and kainic acid (13)(14)(15). Lin et al. (13) have reported that NPAS4 regulates the development of GABAergic inhibitory synapses in an activity-dependent manner and suggested its homeostatic role in the balance between excitatory and inhibitory neuronal activities in the brain. We previously reported that reduced Npas4 mRNA levels may contribute to impairments in adult neurogenesis in the hippocampus, memory and emotional behaviors induced by social isolation or restriction stress (16,17). Recent studies (18,19) revealed that NPAS4 is important in memory formation and consolidation. Synaptic remodeling is considered to be a way for neurons to adapt cellular and neuronal circuits to environmental changes (20), and neurite arborization and rewiring may contribute to the neuronal plasticity in the brain (21)(22)(23)(24). In this study, to investigate a possible role for NPAS4 in structural and functional plasticity of neurons, we examined the effect of knockdown or overexpression of NPAS4 on neurite outgrowth in Neuro2a cells. Neurite outgrowth in Npas4 knock-out primary cultured hippocampal neurons was also investigated. Then, we investigated the underlying mechanism by which NPAS4 regulates neurite outgrowth. We focused on the phosphorylation of a synaptic vesicle-associated protein, Syn I, 3 via cyclin-dependent kinase 5 (CDK5). Finally, we conducted an in vivo study to see if the NPAS4-induced CDK5/ SYN I pathway could be operated under physiological and pathophysiological conditions using a pentylenetetrazole (PTZ)-induced epilepsy model in mice.
Quantitative Real Time RT-PCR-Complementary DNA was synthesized from total isolated RNA using a SuperScriptIII first-strand synthesis system for RT-PCR (Invitrogen). Levels of mRNA expression were quantified by using a 7300 real time PCR System (Applied Biosystems, Foster City, CA). The quantitative real time RT-PCR was performed in a volume of 25 l with 1 g of cDNA and 500 nM primers in a Power SYBR Green Master Mix (Applied Biosystems). The primers used were as follows: 5Ј-AGCATTCCAGGCTCATCTGAA-3Ј (forward) and 5Ј-GGCGAAGTAAGTCTTGGTAGGATT-3Ј (reverse) for Npas4; 5Ј-TGTCAAGCTCATTTCCTGGTATGA-3Ј (forward) and 5Ј-CTTACTCCTTGGAGGCCATGTAG-3Ј (reverse) for glyceraldehyde-3-phosphate dehydrogenase for GAPDH; and 5Ј-CGATGCCCTGAGGCTCTTT-3Ј (forward) and 5Ј-TGGATGCCACAGGATTCCA-3Ј (reverse) for ␤-Actin used as internal controls.
NPAS4 Overexpression and Neurite Length Measurement-On the basis of the GenBank TM sequence, Mus musculus Npas4 mRNA (BC129861) was chosen to construct the overexpression vector. Briefly, PCR was carried out using mouse cDNA with primers, which introduced a NotI site with deletion of the stop codon at the 3Ј end. The PCR product was subcloned into the pCR-XL-TOPO (Invitrogen) and sequenced. The cloned vector was digested using PstI and NotI endonuclease (Takara, Otsu, Shiga, Japan), and the fragment was inserted at the PstI and NotI site of pCMV/myc/nuc vector (Invitrogen). Neuro2a cells (2 ϫ 10 4 cells/ml) were transfected with the Npas4/ pCMV/myc/nuc vector (1 g, Npas4 vector) or pCMV/myc/ nuc vector (mock vector) using FuGENE 6 (6 l, Roche Diagnostics). Seventy two hours later, the medium was replaced with DM, and the cells were cultured for 12 h. Neurite length was measured as described above.
CDK5 Immunoblot and Activity Assay-After transfection of Npas4 or mock expression vector in Neuro2a cells, the transfected cells were enriched by G418 (1.2 g/ml, Calbiochem, Merck KGaA) selection. The whole cell lysates were collected, and an immunoblot assay was performed with anti-Cdk5 (rabbit, 1:1,000, Santa Cruz Biotechnology) antibody. CDK5 activity assay was performed with ADP-Glo TM kinase assay kit according to manufacturer's instructions (Promega, Madison, WI).
Primary Hippocampal Neuron Culture and the Measurement of p-SYN I and Neurite Outgrowth-Hippocampal neurons were dissociated and prepared from E16 to E17 mouse embryos as described previously (26). To measure p-SYN I levels, ICR mice were killed on gestational day 16 or 17, and embryo hippocampi were digested with 0.25% trypsin and 0.01% DNase for 10 min at 37°C and mechanically dissociated by gentle pipetting. Neurons were plated at 5 ϫ 10 4 cells on poly-L-lysine-coated coverslips (12 mm, BD BioCoat, Bedford, MA) and cultured in DMEM (D6546, 10% fetal bovine serum, 2 mM L-glutamine, antibiotics/antimycotics) on day in vitro (DIV) 0. DMEM was replaced with glial conditioned medium (nerve cell and microglial cell culture system, SUMITOMO BAKELITE, Tokyo, Japan) with Ara-C (5 M, Sigma) and antibiotics/antimycotics. The Npas4 overexpression vector was transfected on DIV 4. Briefly, the medium was removed and kept at 37°C in a humidified atmosphere with 5% CO 2 . Npas4/ pCMV/myc/nuc vector (1 g) was transfected with Opti-MEM I (100 l, Invitrogen) and Lipofectamine 2000 (5 l, Invitrogen) in a new medium (500 l, without antibiotics/antimycotics), and the transfection medium was replaced with old medium 4 h later. Immunocytochemistry was performed on DIV 5, and the p-SYN I fluorescence intensity was measured as described above. To measure neurite outgrowth, hippocampal neurons were dissociated and prepared from E16 to E17 mouse embryos of Npas4 knock-out (KO) or wild-type (WT) littermates as described above. Npas4 KO mice were kindly donated by Dr. Greenberg (13). The longest Tau-positive axonal length was measured at DIV 3. MAP2-positive dendrites length was measured at DIV 9, after treatment of KCl (16 mM) for 48 h (DIV 7-9) according to Wayman et al. (27). The measurement of neurite length and Sholl analysis were performed using a microscope and a visual image analysis system (Neurolucida, MBF Bioscience, Williston, VT).
Pentylenetetrazole-induced Convulsion Model of Epilepsy-Npas4 KO mice or WT littermates at 6 -8 weeks old were housed 4 -5 per cage under standard conditions (23 Ϯ 1°C, 50 Ϯ 5% humidity), with a 12-h light/dark cycle. Food and water were available ad libitum. The animals were handled in accordance with the guidelines established by the Institutional Animal Care and Use Committee of Nagoya University. Npas4 KO and WT mice were administered intraperitoneally with PTZ (Sigma) at a dose of 60 mg/kg, and the intensity of convulsions induced by PTZ was scored according to previous studies (28,29). After seizure testing, mice were sacrificed. The prefrontal cortex was isolated, and an immunoblot assay was performed as described above (including p35 detection with p35 antibodies, rabbit, 1:1,000, Santa Cruz Biotechnology).
Statistical Analysis-Differences between the two groups were analyzed by Student's t test. One-way analysis of variance (ANOVA) followed by Dunnett's test or two-and three-way ANOVA followed by Bonferroni's test was used for multigroup comparisons. To compare neurite length, differences between two groups were analyzed by the Kolmogorov-Smirnov test, and differences among more than two groups were analyzed by one-way ANOVA followed by Dunnett's test and confirmed by the Kolmogorov-Smirnov test. The relationship between CDK5 and p-SYN I expression was analyzed by linear regression. The comparison of regression lines was performed by parallel line analysis.

Effect of Differentiation Medium on Neurite Outgrowth, NPAS4, and p-SYN I Expression in Neuro2a
Cells-Neuronal differentiation of Neuro2a cells can be induced by treatment with DM leading to morphological changes with longer neurites. When Neuro2a cells were treated with DM for 48 h, neurite outgrowth in Neuro2a cells was accompanied by an increase in p-SYN I levels (Fig. 1, A-C) with no change in total SYN I levels. Next, we investigated whether neurite outgrowth of Neuro2a cells and the increase in p-SYN I levels induced by DM were associated with changes in Npas4 mRNA levels. We measured Npas4 mRNA levels in DM-treated Neuro2a cells using quantitative RT-PCR. DM significantly increased the mRNA levels of Npas4 in Neuro2a cells (Fig. 1D).
Effect of Npas4 Knockdown by siRNA Treatment on Neurite Outgrowth in Neuro2a Cells-To study the causal relationship between the DM-induced increase in NPAS4 expression and the neurite outgrowth in Neuro2a cells, we next examined whether knockdown of Npas4 expression by siRNA treatment suppressed neurite outgrowth induced by DM treatment. The siRNA or scrambled siRNA of Npas4 with a pZsGreen1-N1 vector, which encodes green fluorescent protein (GFP), was transfected in Neuro2a cells. Seventy two hours after the transfection, cells were treated with DM for 24 h, and the TUJ1positive neurite length (25) was quantified. GFP-positive cells were regarded as siRNA-or scrambled RNA-transfected cells, whereasGFP-negative cells were considered to be nontransfected cells. The siRNA transfection significantly suppressed the neurite outgrowth induced by DM, whereas scrambled siRNA did not affect the neurite length (Fig. 2).
Effect of NPAS4 Overexpression on Neurite Outgrowth in Neuro2a Cells-To investigate the role of NPAS4 in neurite outgrowth in Neuro2a cells further, we examined whether overexpression of NPAS4 facilitated neurite outgrowth induced by DM treatment. Seventy two hours after the transfection, cells were treated with DM for 12 h, and the neurite length was quantified. At this time point, neurite outgrowth induced by DM treatment in control cells was minimum (Fig. 3) compared with the length at 24 h (Fig. 2). Myc-positive cells were regarded as Npas4 vector-or mock vector-transfected cells, whereas Myc-negative cells were considered to be nontransfected cells. Overexpression of NPAS4 significantly increased the neurite outgrowth induced by DM, whereas the mock vector had no effect on the neurite length in Neuro2a cells (Fig. 3).
Effect of NPAS4 Overexpression on NeuN, p-SYN I, and Tuj1 Expression in Neuro2a Cells-To clarify the role of NPAS4 in neuronal maturation, the expression levels of NeuN, p-SYN I, and TUJ1 were investigated in NPAS4-overexpressed Neuro2a cells. Seventy two hours after the transfection, cells were harvested, and immunoblot analysis was performed. Overexpression of NPAS4 increased NeuN and p-SYN I expression levels, whereas the mock vector had no effect in Neuro2a cells (Fig. 4,   A and B). In addition, overexpression of NPAS4 significantly increased the neurite marker TUJ1 expression levels induced by DM (Fig. 4C).
Involvement of CDK5 in p-SYN I Expression in NPAS4-overexpressed Neuro2a Cells-The anti-p-Syn I antibodies used in this study recognize phosphorylated Ser 553 of SYN I, which is known to be a substrate of CDK5 (30,31). Therefore, we measured CDK5 expression levels and its role for p-SYN I expression levels in NPAS4-overexpressed cells. Immunocytochemical analysis was performed 24 and 72 h after transfection with the Npas4 overexpression vector in Neuro2a cells. Overexpression of NPAS4 significantly increased CDK5 and p-SYN I expression levels in Neuro2a cells, whereas mock transfection had no effect (Fig. 5, A and B). Regression lines (72 h) were analyzed. There was a positive correlation between p-SYN I and CDK5 expression levels in the control (y ϭ 0.691x ϩ 0.308, R 2 ϭ 0.288, p Ͻ 0.001) and NPAS4-overexpressed Neuro2a cells (y ϭ 1.159x ϩ 2.427, R 2 ϭ 0.156, p Ͻ 0.001), and the slope for NPAS4-overexpressed Neuro2a cells was significantly greater than that for the control (p Ͻ 0.05). However, no correlation was observed in mock-transfected Neuro2a cells (y ϭ Ϫ0.092x ϩ 2.731, R 2 ϭ 0.000) (Fig. 5C).
Furthermore, we measured the CDK5 expression levels and kinase activity in NPAS4-overexpressed Neuro2a cells. Overex-  pression of NPAS4 increased CDK5 expression and activity, whereas the mock vector had no effect in Neuro2a cells (Fig. 6,  A and B). To explore the causal relationship, NPAS4-overexpressed Neuro2a cells (72 h after the transfection) were treated with DM for 12 h, during which a CDK5 inhibitor, roscovitine (30 M, Calbiochem), or DMSO was added to the culture medium (total 84 h), and p-SYN I expression levels and neurite length were measured. Roscovitine significantly reduced the p-SYN I expression (Fig. 6, C and D) and neurite outgrowth (Fig. 7) in Neuro2a cells induced by NPAS4 overexpression.
Phosphorylated SYN I Expression and Neurite Outgrowth in Hippocampal Neurons-To confirm NPAS4-induced phosphorylation of SYN I in hippocampal neurons, p-SYN I expression levels were investigated in NPAS4-overexpressed hip-   pocampal neurons. Twenty four hours after the transfection, immunocytochemical analysis was performed. Overexpression of NPAS4 increased p-SYN I levels in hippocampal neurons, whereas the mock vector had no effect (Fig. 8). We compared neurite length between WT and Npas4 KO hippocampal neurons. Npas4 KO hippocampal neurons showed an impaired MAP2-positive dendritic (Fig. 9, A and B) and Tau-positive axonal outgrowth (Fig. 9, D and E) and a reduction in the number of intersections of dendrites (Sholl analysis, Fig. 9C) compared with WT hippocampal neurons. Furthermore, KCl treatment increased dendritic length and complexity in WT hippocampal neurons but not in Npas4 KO hippocampal neurons (Fig. 9, A-C).
Binding of NPAS4 to Promoter Regions of CDK5 and NeuN Genes in Neuro2a Cells and Primary Cultured Neurons-We used a ChIP assay to confirm the binding of NPAS4 to Cdk5 and NeuN promoters. An Npas4 promoter was used as a positive control (13). The ChIP assay revealed that NPAS4 binds to promoter regions of Cdk5 and NeuN in DM-treated Neuro2a cells (Fig. 10A). In primary cultured neurons, a ChIP assay was conducted with or without KCl stimulation because the treatment was reported to induce NPAS4 in cultured neurons (13). As shown in Fig. 10C, we confirmed the increase in NPAS4 protein levels after KCl stimulation in primary hippocampal neurons. Only under such conditions was the binding of NPAS4 to the promoters of Npas4 and Cdk5 proven by the ChIP assay (Fig.  10B). Furthermore, NPAS4 bound to a promoter of p35, a coactivator of CDK5, in Neuro2a cells as well as in primary cultured neurons (Fig. 10).

Effect of PTZ Administration on NPAS4 and p-SYN I Expression in the Prefrontal Cortex of Npas4 KO and WT Mice-Fi-
nally, we investigated if p-SYN I expression was affected by in vivo PTZ treatment in WT and Npas4 KO mice because it has been reported that PTZ-induced convulsions in mice lead to an increase in the Npas4 mRNA levels in the hippocampus (32). First, we measured NPAS4, p-SYN I, SYN I, p35 and CDK5 expression levels 2, 6, 1, and 24 h after PTZ administration in the prefrontal cortex of WT mice (Fig. 11). PTZ-induced NPAS4 and p-SYN I were most highly expressed 2 h after PTZ administration. We next compared the effects of PTZ on NPAS4 and p-SYN I levels in WT and KO mice 2 h after treatment. PTZ administration increased Npas4 levels in the prefrontal cortex of WT mice but not in Npas4 KO mice (Fig. 12). Similarly, p-SYN I levels were increased by PTZ administration  in WT mice, whereas the same treatment had no effect on p-SYN I levels in Npas4 KO mice (Fig. 12). There was no difference in the intensity score of convulsions induced by PTZ treatment between Npas4 KO mice (convulsion score 4.5 Ϯ 0.5) and WT mice (3.9 Ϯ 0.13), suggesting that the changes in p-SYN I levels after PTZ treatment in Npas4 KO mice are not due to the reduced excitability of the mutant mouse brains.

DISCUSSION
NPAS4 is a brain-specific neuronal transcription factor that may respond to various excitatory stimuli and has a role in GABAergic neuronal synapse development (13)(14)(15). A recent study demonstrated that the activity-dependent expression of NPAS4 regulates a transcriptional program in the CA3 hippocampus required for contextual memory formation (18).
In this study, we showed that DM induced an increase in neurite outgrowth with an elevated Npas4 mRNA level. DMinduced neurite outgrowth was inhibited in Npas4 knockdown Neuro2a cells, whereas overexpression of NPAS4 accelerated the neurite outgrowth induced by DM. Furthermore, we demonstrated that axonal and dendritic outgrowths were impaired in Npas4 KO hippocampal neurons. Although it was reported that NPAS4 was expressed after maturation in primary cultured neurons (13), our data indicate that this transcription factor is expressed even in immature neurons at DIV 3 (Fig.  10C). Under such culture conditions, an impaired axonal outgrowth was evident in Npas4 KO neurons compared with WT neurons. Lin et al. (13) showed that Npas4-RNAi transfection at DIV 6 did not decrease dendritic complexity at a later stage (DIV [25][26]. We compared the dendritic length of WT and conventional KO mice hippocampal neurons at DIV 9, and we found that the dendritic length of Npas4 KO neurons was significantly shorter than that of WT neurons. Furthermore, KClinduced dendritic outgrowth, which was evident in WT neurons, was abolished in Npas4 KO hippocampal neurons. These results suggest that the role of NPAS4 in dendritic outgrowth varies depending on neuronal development stages in an activity-dependent manner. Activity-dependent neurite outgrowth in cultured hippocampal neurons is reported to be mediated via activation by CaM-dependent protein kinase of the membrane-associated ␥ isoform of CaMKI (27), whereas the DM-induced neurite outgrowth in Neuro2a cells is associated with inhibition of GSK-3␤ (33). In our study, NPAS4 overexpression elevated CDK5 and p-SYN I expression levels and enhanced neurite outgrowth. The CDK5 inhibitor roscovitine reduced p-SYN I expression, as well as neurite outgrowth induced by NPAS4 overexpression. SYN I has been characterized as one of the major phosphoproteins in nerve terminals and is thought to be involved in the  regulation of neurotransmitter release (34,35). Both phosphorylated and nonphosphorylated Syn I regulate neuronal development (36 -38). The anti-p-Syn I antibodies used in this study recognize phosphorylated Ser 553 of SYN I, which is known to be phosphorylated by CDK5 (30,31). Although previous DNA microarrays and ChIP-Seq studies did not reveal that CDK5 or CDK5-coactivator p35 is a target of NPAS4 (13,46), CDK5 is essential for neurite outgrowth (39 -41) and has roles in the physiological regulation of cytoskeletal components during neurotransmitter release via phosphorylation of SYN I (30). We demonstrated that NPAS4 overexpression increased CDK5 protein and the kinase activity in Neuro2a cells. In addition, NPAS4 bound to promoter regions of CDK5 as well as p35 in KCl-activated primary cultured neurons and DM-treated Neuro2a cells. Accordingly, it is suggested that NPAS4 may directly regulate transcription of CDK5 and consequently increase the protein expression with kinase activity, which is associated with neurite outgrowth.
We also demonstrated that PTZ-induced phosphorylation of SYN I in vivo was absent in NPAS4 KO mice, suggesting that NPAS4 has a crucial role in SYN I phosphorylation in vivo. However, there might be concern that the NPAS4/CDK5/SYN I signaling in neurite outgrowth described here may operate only in vitro cell culture systems because the in vivo study showed that phosphorylation of SYN I was induced by PTZ without an increase in the levels of CDK5 or p35. In fact, CDK5 expression was increased at 24 h, but not 2 h, after PTZ treatment, whereas an increase in p-SYN I and NPAS4 was observed at 2 h, but not 24 h, after PTZ treatment (Fig. 11). It is therefore likely that unidentified downstream signaling mechanisms other than CDK5 of NPAS4 may operate in the regulation of SYN I phosphorylation in neurons. To address this issue, further experiments with other animal models of epileptogenesis (42) are necessary in Npas4 KO mice.
A useful mature neuronal marker, NeuN, was induced in NPAS4-overexpressed Neuro2a cells, whereas NeuN was hardly detected in nontransfected control cells, as demonstrated in previous reports (43,44). Although NeuN was not identified as a target of NPAS4 in previous studies (13,45), NeuN has been identified as a transcriptional factor, FOX-3,  and appears to be a nervous system-specific nuclear regulatory molecule and a splicing regulator (46,47). In our study, a ChIP assay showed that Npas4 binds the promoter of NeuN in Neuro2a cells. Accordingly, NPAS4 may have indirect effects through the regulation of other transcriptional factors (18). Taking these findings together, it is suggested that NPAS4 contributes to neurite development and functional maturation of neurons through the transcriptional regulation of neurotransmission and neurodevelopment-associated molecules.