GPR3 is an immediate-early gene-like GPCR regulating CREB-dependent neuronal differentiation

TL;DR

GPR3 is a GPCR that acts like an immediate-early gene, helping regulate neuronal differentiation through CREB-dependent signaling.

Contribution

GPR3 is identified as an immediate-early gene-like GPCR that amplifies cAMP-CREB signaling during neuronal differentiation.

Findings

Gpr3 is rapidly induced by NGF and cAMP in PC12 cells with biphasic activation.

Early Gpr3 induction enhances Nr4a1–3 and Synapsin1 transcription via CREB signaling.

Gpr3 deletion reduces Nr4a1–3 and Syn1 expression in primary cortical neurons.

Abstract

GPR3 is a constitutively active Gs-coupled receptor whose transcriptional regulation during neuronal differentiation has remained unclear. Here, we identify Gpr3 as an immediate-early gene-like transcript rapidly induced by nerve growth factor (NGF) and cAMP signaling in PC12 cells, exhibiting biphasic activation captured by native elongating transcript-cap analysis of gene expression (NET-CAGE) at a core promoter ∼200 bp upstream of the transcription start site (TSS). Five cAMP response elements (CREs) within the 1-kb regulatory region cooperatively mediated stimulus-responsive transcription, with p-CREB enrichment selectively occurring at the proximal −34 CRE. Early Gpr3 induction enhanced delayed Nr4a1–3 expression and promoted Synapsin1 (Syn1) transcription through an Nr4a1-dependent mechanism. In primary cortical neurons, Gpr3 deletion diminished the developmental upregulation of Nr4a1–3 and Syn1 and reduced SYN1-positive vesicle density. These findings indicate GPR3 as an activity-dependent cAMP amplifier that couples early CREB activation to transcriptional programs governing NR4A signaling and presynaptic maturation during neuronal differentiation.

•GPR3 is a unique GPCR exhibiting immediate-early gene-like biphasic induction•Proximal CREs mediate rapid Gpr3 transcription through stimulus-dependent CREB binding•Early Gpr3 induction reinforces cAMP-CREB signaling to enhance Nr4a1–3 transcription•NR4A1 mediates GPR3-dependent Synapsin1 expression and presynaptic development

GPR3 is a unique GPCR exhibiting immediate-early gene-like biphasic induction

Proximal CREs mediate rapid Gpr3 transcription through stimulus-dependent CREB binding

Early Gpr3 induction reinforces cAMP-CREB signaling to enhance Nr4a1–3 transcription

NR4A1 mediates GPR3-dependent Synapsin1 expression and presynaptic development

Biological sciences; molecular neuroscience; systems neuroscience; cellular neuroscience

Figures

Immediate early *Gpr3* mRNA induction following various neuronal differentiation-related stimuli in PC12 cells(A–E) PC12 cells were induced to differentiate using serum deprivation (0.5% FBS) and KCl (50 mM; osmolarity-controlled), NGF (50 ng/mL), or forskolin (20 μM). The expression of *Gpr3* (A), *Nr4a1* (B), *Nr4a2* (C), *Nr4a3* (D), and *c-fos* (E) mRNAs was examined using RT-qPCR at 1–5 h after each cell-stimulating condition. Data represent relative mRNA expression (fold change vs. *β-actin*) and are presented as the mean ± SEM from three independent cultures (biological replicates).NGF,

NET-CAGE analysis during the induced differentiation of PC12 cellsNET-CAGE was performed at 0, 30 min, 2 h, and 24 h after stimulation by serum deprivation (0.5% FBS) and NGF (50 ng/mL) or a combination of forskolin (20 μM) and ionomycin (1 μg/mL) in PC12 cells. CAGE signals were aligned to the RGSC 6.0/rn6 rat genome assembly and visualized using the UCSC Genome Browser. The *Gpr3* gene is located on the minus strand. CAGE signal mapped to the plus strand are shown in blue, and those mapped to the minus strand are shown in red. NGF stimulation induced bidirectional CAGE activity within ∼200 b

The 5′ flanking region upstream of TSS contributes to *Gpr3* gene induction in response to neuronal differentiation-inducing stimuli in PC12 cells(A) Schematic of the 5′ flanking region in the rat *Gpr3* gene. Six putative CREs are located at −854, −507, −455, −412, −192, and −34 base pairs (bps) upstream of the TSS, and one potential AP-1 binding TRE is located at +2 bp. Red-lined areas indicate predicted core promoter regions near the TSS.(B) PC12 cells were transfected with reporter pGL-rGPR3WT plasmid, as detailed in STAR Methods. Twenty-four hours after transfection, the cells were stimul

CREB binding to the CRE-containing regions of the *Gpr3* promoter following NGF and forskolin stimulation in PC12 cells(A and B) PC12 cells were stimulated with NGF (50 ng/mL) or forskolin (20 μM), and the expression levels of *p*-CREB (ser133) and total CREB were evaluated by western blotting at 0, 15, 30, and 60 min after stimulation.(A) Representative immunoblots showing *p*-CREB (ser133) and total CREB protein expression. *p-*, phosphorylated; right margin indicates molecular mass standards (in kilodaltons, kDa).(B) Densitometric quantification of time-dependent changes in *p*-CREB (ser133

Induced *Gpr3* expression during neuronal differentiation modulates *Syn1–3* expression in PC12 cells(A–C) Effects of the suppression of *Gpr3* expression on *Syn* mRNA expression. PC12 cells were transfected with control or *Gpr3* siRNA. Twenty-four hours after transfection, differentiation was induced using serum deprivation (0.5% FBS) and NGF (50 ng/mL). *Syn1* (A), *Syn2* (B), and *Syn3* (C) mRNA expression was evaluated using RT-qPCR at 0, 24, 48, and 96 h after inducing differentiation. Data represent relative mRNA expression normalized to *β-actin* and are presented as the mean ± SEM fo

Developmental changes in *Gpr3*, *Syn1*, and *Nr4a* family gene expression in primary cortical neurons from wild-type and *Gpr3*^−/−^ mice(A–E) Primary cortical neurons were isolated from postnatal day 0–1 (P0–P1) wild-type (WT), and *Gpr3* knockout (*Gpr3*^−/−^) mice and cultured *in vitro* for up to 14 days (DIV0–DIV14). mRNA expression levels of *Gpr3* (A), *Syn1* (B), *Nr4a1* (C), *Nr4a2* (D), and *Nr4a3* (E) were measured by RT-qPCR at DIV0, DIV0.7 (16 h), DIV1, DIV2, DIV4, DIV7, and DIV14. Data represent relative expression normalized to *β-actin* and are presented as the mean ± SEM for

Reduced SYN1-positive puncta in cortical neurons from *Gpr3*^−/−^ mice *in vitro*(A) Staining for SYN1 and Bassoon in cortical neurons obtained from wild-type and *Gpr3*^−/−^ mice. Primary neurons were isolated from postnatal day 0–1 (P0–P1) and cultured for 14 days *in vitro* (DIV14). Neurons were fixed and immunostained with anti-SYN1 (green) and anti-Bassoon (red) antibodies. Representative images are shown.(B) For quantification, neurons were stained with anti-SYN1 (red) and DAPI (blue) at DIV7 and DIV14; representative images are shown.(C) SYN1-positive puncta were quantified within a reg

Full text

Introduction

Neuronal gene expression is modulated by neurotrophic factors and neuronal depolarization related to neuronal activity, both of which play pivotal roles in neuronal differentiation and adaptive plasticity in the central nervous system. Genes that respond to neuronal stimulation can be classified into two groups based on their expression timing: immediate early genes (IEGs) and late-response genes.1 IEGs, including transcription factors, secreted proteins, synaptic components, intracellular signaling molecules, and membrane proteins, are transiently induced within minutes following stimulation by factors such as nerve growth factor (NGF) and forskolin.1^,^2 Transcription factors such as basic leucine zipper proteins, zinc finger proteins, and nuclear receptors are rapidly induced by neuronal differentiation stimuli and subsequently enhance the transcription of downstream genes, playing essential roles in neuronal functions including synaptic plasticity and homeostasis.3^,^4 Moreover, dysregulation of IEGs has been associated with neurological disorders such as autism spectrum disorder and cognitive dysfunction. Thus, identifying previously uncharacterized IEGs in neurons and understanding their downstream effects is critical for elucidating the mechanisms of neuronal homeostasis and adaptation.

GPR3, a member of the class A rhodopsin-like GPCR family, is expressed in neurons,5^,^6^,^7 oocytes,8 adipocytes,9 and T cells.10 GPR3 can constitutively activate the Gαs protein without ligand binding, leading to elevated basal levels of intracellular cAMP.9^,^11 We have previously investigated the neuronal functions of GPR3 and revealed that this protein is upregulated during neuronal differentiation in cerebellar granule neurons, where it promotes neurite outgrowth and neuronal survival and differentiation.6^,^12^,^13 Additionally, we recently reported that GPR3 is rapidly upregulated following T cell stimulation, influencing the expression of nuclear receptor subfamily 4 group A member 2 (NR4A2) and modulating effector T cell differentiation.10 These findings suggest that GPR3 may play important roles in both neuronal and immune cell differentiation. However, the early-phase induction dynamics and transcriptional regulation of GPR3 following neuronal stimulation remain poorly understood.

The NGF-induced differentiation model of PC12 cells is widely used to study early responsive gene expression during neuronal differentiation.2 NGF stimulation promptly induces IEGs such as c-fos, JunB, and Egr.14 Of these, c-fos is rapidly induced by various stimuli including NGF, cAMP, protein kinase C, and membrane depolarization through the activation of serum response elements and cAMP response elements (CREs).1^,^2^,^15^,^16^,^17

NGF-induced Fos and Jun family proteins form the activator protein 1 (AP-1) transcription factor complex; this enhances the transcription of downstream genes, including members of the NR4A family, which are involved in neuronal differentiation.18 cAMP-elevating agents such as forskolin enhance CRE-binding protein (CREB) phosphorylation, which promotes NR4A1 transcription through CREs and subsequently upregulates Synapsin1 (Syn1) expression, thereby contributing to neuronal differentiation in PC12 cells.19^,^20

The Gpr3 gene contains several evolutionarily conserved CREB-binding motifs in its 5′ upstream region, suggesting that it may be regulated by activity-dependent transcription. Given that CREB and NR4A family members promote neuronal differentiation, GPR3 may act both as a transcriptional target and a regulatory component within this network. In this study, we used native elongating transcript-cap analysis of gene expression (NET-CAGE), a high-resolution technique that enables sensitive detection of dynamic transcriptional changes by capturing the 5′ cap structures of nascent RNA,21 to analyze the transcriptional dynamics of Gpr3 during NGF-induced neuronal differentiation. We also aimed to identify functional CREs within the 5′ regulatory region of Gpr3 that mediate transcriptional responses to NGF and forskolin. Furthermore, we examined whether GPR3 influences the expression of Nr4a1–3 and Syn1 during neuronal differentiation.

Results

Gpr3 is biphasically induced during NGF-mediated neuronal differentiation in PC12 cells

Gpr3 is constitutively expressed in various neuronal populations of the central nervous system,7 and its expression has been reported to increase during neuronal differentiation in cerebellar granule neurons.12 In addition, we recently demonstrated that Gpr3 mRNA is rapidly induced in T cells upon stimulation and contributes to effector T cell differentiation.10 These findings led us to hypothesize that Gpr3 may also be rapidly induced in neuronal cells during differentiation. To test this, we employed PC12 cells, a widely used rat pheochromocytoma cell line that serves as a well-established model for studying neuronal differentiation.22 Consistent with previous studies, PC12 cells extended neurites over a 48-h period in response to NGF stimulation under serum-deprived conditions (Figure 1A). To further evaluate neuronal differentiation induced by these stimuli, we examined the expression of neuronal marker proteins βIII-tubulin (Tub-βIII) and synapsin I (SYN1). Immunofluorescence analysis revealed increased Tub-βIII and SYN1 signals in NGF- and forskolin-treated cells at 48 h, whereas KCl stimulation produced weaker effects (Figure 1B). Consistent with these observations, western blot analysis demonstrated time-dependent increases in Tub-βIII and SYN1 protein expression up to 48 h following NGF and forskolin stimulation, while KCl induced only modest changes (Figures 1C–1E).Figure 1Gpr3 mRNA expressions after NGF-, Fsk-, and KCl-mediated neuronal differentiation in PC12 cells(A and B) PC12 cells were induced to differentiate using serum deprivation (0.5% FBS) and stimulation with NGF (50 ng/mL), Fsk (20 μM), or KCl (50 mM). (A) Representative images from PC12 cells stimulated with NGF, Fsk, or KCl at 8, 12, 24, and 48 h after induced differentiation. Scale bars are 50 μm. (B) To evaluate neuronal differentiation induced by various stimuli in PC12 cells, the expression of βIII-tubulin (Tub-βIII) and synapsin I (Syn1) was examined. Representative immunofluorescence images of PC12 cells cultured in normal serum-containing medium (15% FBS) or stimulated with NGF, Fsk, or KCl for 48 h are shown. Cells were double-stained with anti-Tub-βIII (red) and anti-SYN1 (green) antibodies. Scale bars are 50 μm.(C) Representative immunoblots showing Tub-βIII and SYN1 protein expression in PC12 cells at 0, 12, 24, and 48 h after stimulation with NGF, Fsk, or KCl. GAPDH is shown as an endogenous loading control. Right margin indicates molecular mass standards (in kilodaltons, kDa).(D and E) Densitometric quantification of time-dependent changes in Tub-βIII (D) and SYN1 (E) protein expression. Data were normalized to GAPDH and are expressed relative to the 0 h value. Data are presented as the mean ± SEM from five independent cultures (biological replicates). ∗∗∗, p < 0.001; ∗∗, p < 0.01 vs. 0 h for each condition.(F) The expression of Gpr3 mRNA was evaluated at 0, 2, 4, 6, 12, 24, 48, and 96 h after induced differentiation, using RT-qPCR. Data represent relative Gpr3 expression (fold change vs. β-actin) and are presented as the mean ± SEM from 4–6 independent cultures (biological replicates). ∗∗∗p < 0.001, ∗∗p < 0.01 and ∗p < 0.05 vs. 0 h.NGF, neuronal growth factor; Fsk, forskolin; FBS, fetal bovine serum; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.Source data for this figure are provided in Data S1.

Using the validated NGF-induced PC12 cell differentiation model, we next analyzed Gpr3 mRNA expression over time. Under normal culture conditions, Gpr3 expression remained largely unchanged throughout the time course. By contrast, under differentiation-inducing conditions, Gpr3 mRNA levels showed a transient upregulation as early as 2 h after NGF treatment, followed by a decline at 4 h, and then a second phase of upregulation that peaked at 24 h, with expression sustained for up to 96 h (Figure 1F).

Thus, Gpr3 exhibits a biphasic expression pattern in the neuronal differentiation model, with early transient and late sustained phases playing potentially distinct functional roles.

Gpr3 is induced early in response to various stimuli related to neuronal differentiation

In neurons, IEGs such as Fos, Jun, Arc, Egr, and Nr4a are rapidly induced in response to extracellular stimuli and play pivotal roles in neuronal functions, including differentiation and synaptic plasticity.1^,^4 To further investigate the early induction properties of Gpr3, we examined its expression following various known neuronal differentiation stimuli in PC12 cells. Under normal culture conditions with 15% fetal bovine serum (FBS), Gpr3 expression was nearly undetectable. However, serum deprivation (0.5% FBS) alone slightly induced Gpr3 expression (Figure 2A). Notably, a significant and transient induction of Gpr3 was observed as early as 1–2 h after serum deprivation combined with NGF stimulation, with expression levels declining by 3 h (Figure 2A). This immediate-early expression pattern was also strongly induced by forskolin, whereas KCl depolarization produced only a modest or minimal effect on Gpr3 induction (Figure 2A), which is consistent with the limited morphological differentiation observed under this condition (Figure 1A). To compare the induction kinetics with those of other IEGs, we measured Nr4a1–3 and c-fos expression under the same conditions. Nr4a1–3 exhibited a comparable transient expression pattern to Gpr3 (Figures 2B–2D), whereas c-fos expression peaked earlier, reaching maximal levels within 1 h (Figure 2E). Although Gpr3 displayed IEG-like rapid induction kinetics, the amplitude of its induction in our quantitative reverse-transcription PCR (RT-qPCR) assays was lower than that of classical IEGs such as c-fos and Nr4a1, indicating that Gpr3 behaves as a modest but reproducible early-response gene.Figure 2. Immediate early Gpr3 mRNA induction following various neuronal differentiation-related stimuli in PC12 cells(A–E) PC12 cells were induced to differentiate using serum deprivation (0.5% FBS) and KCl (50 mM; osmolarity-controlled), NGF (50 ng/mL), or forskolin (20 μM). The expression of Gpr3 (A), Nr4a1 (B), Nr4a2 (C), Nr4a3 (D), and c-fos (E) mRNAs was examined using RT-qPCR at 1–5 h after each cell-stimulating condition. Data represent relative mRNA expression (fold change vs. β-actin) and are presented as the mean ± SEM from three independent cultures (biological replicates).NGF, neuronal growth factor.

These findings suggest that Gpr3 possesses properties that are similar to IEGs and is rapidly induced by differentiation-related stimuli, such as NGF and forskolin, in PC12 cells.

Transcriptional regulatory sites related to early Gpr3 induction by NGF stimulation in PC12 cells

To investigate the transcriptional mechanisms underlying the early induction of Gpr3 by neuronal differentiation stimuli, we first examined the genomic context of the Gpr3 promoter. According to the NCBI gene database, a GC box is located near the transcription start site (TSS) of the Gpr3 gene (Figure S1). In addition, a TPA response element (TRE) and several CREs are present within a 1-kb upstream region of the TSS; these regulatory motifs are highly conserved among humans, mice, and rats (Figure S1). According to CAGE-based promoter annotations from the FANTOM5 project,23 a region of approximately 200 base pairs upstream of the TSS, flanked by bidirectional CAGE signals, has been identified as the core promoter region of Gpr3 in both human and mouse genomes (Figure S2). The assay for transposase-accessible chromatin with sequencing data, visualized in the UCSC Genome Browser24 and provided by the ENCODE project,25 revealed a well-defined open chromatin region spanning approximately 1 kb around the TSS mouse (mm10) genomes (Figure S2B). Histone modification data also exhibited a prominent H3K27ac peak—a marker of active enhancers and promoters—within the same region in human (hg38) and suggesting high transcriptional activity at this locus (Figure S2A). Given these findings, we hypothesized that Gpr3 expression is transcriptionally regulated within the 1-kb region upstream of the TSS. To test this, we used NET-CAGE, which enables the high-sensitivity detection of short-lived RNAs, such as enhancer RNAs and promoter upstream transcripts, by analyzing the 5′ cap structure of nascent, unprocessed RNAs during active transcription within the nucleus.21 We analyzed NET-CAGE signals at 30 min, 2 h, and 24 h following NGF stimulation in PC12 cells (Figure 3). In the unstimulated state, as observed in human and mouse, a region of approximately 200 base pairs flanked by bidirectional CAGE signals was detected upstream of the TSS of Gpr3, suggesting that this region corresponds to the core promoter region of the rat Gpr3 gene (Rnor_6.0/rn6). Upon NGF stimulation with serum deprivation, the number of CAGE signal counts increased as early as 30 min after stimulation (Figure 3). A comparison of gene expression before and after NGF stimulation revealed a significant increase in the count per million (CPM) at 30 min for known IEGs including Srf, Fos, Egr, Arc, and Jun (Figure S3). Of note, Gpr3 also showed approximately 10-fold induction at this time point, ranking 27th among all upregulated genes (Figure S3). No other GPCR showed a similar magnitude of induction (>10-fold), suggesting that Gpr3 may represent a unique GPCR with IEG-like transcriptional responsiveness to NGF in PC12 cells. The bidirectional NET-CAGE signals at both ends of the Gpr3 promoter showed a decreasing trend at 2 h after stimulation but increased again at 24 h (Figure 3), mirroring the biphasic expression pattern of Gpr3 mRNA (Figure 1B).Figure 3NET-CAGE analysis during the induced differentiation of PC12 cellsNET-CAGE was performed at 0, 30 min, 2 h, and 24 h after stimulation by serum deprivation (0.5% FBS) and NGF (50 ng/mL) or a combination of forskolin (20 μM) and ionomycin (1 μg/mL) in PC12 cells. CAGE signals were aligned to the RGSC 6.0/rn6 rat genome assembly and visualized using the UCSC Genome Browser. The Gpr3 gene is located on the minus strand. CAGE signal mapped to the plus strand are shown in blue, and those mapped to the minus strand are shown in red. NGF stimulation induced bidirectional CAGE activity within ∼200 bp upstream of the Gpr3 TSS.TSS, transcriptional start site.

Collectively, these findings suggest that Gpr3 induction upon NGF stimulation is regulated via promoter elements located between the TSS and ∼200 bp upstream.

CREB-dependent transcriptional control underlies Gpr3 induction by NGF and cAMP signaling

Within approximately 200-bp upstream of the Gpr3 TSS in the rat genome, one TRE (+2) and two CREs (−34 and −192) were identified (Figure 4A). In addition, multiple putative CREs were distributed within the 1-kb upstream region, with high sequence conservation between rodents and humans (Figures 4A and S1). Previous studies have implicated NGF-mediated CREB phosphorylation in promoting c-fos transcription in PC12 cells.17 Consistent with this, Gpr3 induction was more sensitive to forskolin stimulation than to NGF stimulation (Figure 2A), suggesting that cAMP-responsive transcription factors, such as phosphorylated CREB, may contribute to Gpr3 regulation. To test this hypothesis, we constructed a luciferase reporter plasmid (pGL-rGPR3WT) containing approximately 1 kb of the 5′ flanking region upstream of the rat Gpr3 TSS. In PC12 cells transfected with pGL-rGPR3WT, luciferase activity remained unchanged under standard culture conditions but was significantly upregulated following serum deprivation and NGF stimulation (Figure 4B). Moreover, forskolin, a potent activator of cAMP signaling, induced an even greater increase in luciferase activity. In contrast, calcium elevation via KCl or the calcium ionophore ionomycin had no effect on promoter activity on its own, but modest enhancement was observed when combined with forskolin (Figure 4B). Together, these findings suggest that NGF- and cAMP-mediated pathways cooperatively regulate Gpr3 transcription, whereas calcium signaling alone is insufficient to trigger expression but can potentiate cAMP-driven activation.Figure 4. The 5′ flanking region upstream of TSS contributes to Gpr3 gene induction in response to neuronal differentiation-inducing stimuli in PC12 cells(A) Schematic of the 5′ flanking region in the rat Gpr3 gene. Six putative CREs are located at −854, −507, −455, −412, −192, and −34 base pairs (bps) upstream of the TSS, and one potential AP-1 binding TRE is located at +2 bp. Red-lined areas indicate predicted core promoter regions near the TSS.(B) PC12 cells were transfected with reporter pGL-rGPR3WT plasmid, as detailed in STAR Methods. Twenty-four hours after transfection, the cells were stimulated with NGF (50 ng/mL), forskolin (20 μM), KCl (50 mM), ionomycin (1 μg/mL), forskolin + KCl, or forskolin + ionomycin under serum-deprived conditions (0.5% FBS). A luciferase assay was performed 4 h after stimulation. Data are presented as the mean ± SEM for each condition (three independent replicates). Each experiment was standardized based on the control group (15% FBS). ∗∗∗p < 0.001 vs. 15% FBS. Luciferase activity is expressed as arbitrary intensity units (au).(C and D) The contribution of individual CRE to Gpr3 induction was examined using deletion mutants. Constructs are schematically drawn (C) and were prepared as detailed in STAR Methods. The CRE site is drawn as a black-filled square in the promoter region. PC12 cells were transiently transfected with each reporter plasmid (D). Twenty-four hours after transfection, the cells were treated with NGF (50 ng/mL), forskolin (20 μM), or forskolin (20 μM) + ionomycin (1 μg/mL) in addition to serum deprivation (0.5% FBS). Data are presented as the mean ± SEM for each condition (three independent replicates). Each experiment was standardized using the wild-type plasmid in the control group, and statistical differences were analyzed accordingly. ∗∗∗p < 0.001, ∗∗p < 0.01 and ∗p < 0.05 vs. wild-type plasmid transfected group in each condition.(E–G) Effects of point mutations at −854, −455, −412, −192, and −34 CREs of the rat Gpr3 gene. The mutant promoter plasmids positioned at −854, −455, −412, −192, and −34 CREs are shown schematically, and mutant CREs are shown as open (white) boxes (E). Likewise, the cells were transfected with each reporter plasmid and treated with serum deprivation (0.5% FBS) and NGF (50 ng/mL) (F) or forskolin (20 μM) (G) 24 h after transfection. Data are presented as the mean ± SEM for each condition (four independent replicates). Each experiment was standardized using the total mutated plasmid, and the variation in each implementation was statistically analyzed. ∗∗∗p < 0.001 vs. total mutated plasmid transfected group, and ^###^p < 0.001 vs. wild-type plasmid transfected group, in each condition. Δ constructs represent deletion mutants in which the indicated CRE-containing regions were removed from the 1-kb promoter. mCRE constructs carry point mutations that disrupt the consensus CRE sequence at the specified positions (−854, −455, −412, −192, and −34). “Total mutated plasmid” refers to simultaneous mutation of all five CREs, whereas in single-site restored constructs, one intact CRE was reintroduced to assess its individual contribution.TSS, transcriptional start site; NGF, neuronal growth factor; CRE, cAMP response element.

In the rat Gpr3 locus, six highly conserved putative CREs (−854, −507, −455, −412, −192, and −34 relative to the TSS) were mapped within the 1-kb upstream region (Figures 4A and S1). To elucidate the functional contribution of these cis-regulatory elements to NGF- and cAMP-responsive transcription, we generated targeted deletion mutants and performed luciferase reporter assays following the stimulation of PC12 cells (Figure 4C). Deletion of individual CRE, except for the −507 site, significantly reduced luciferase activity in response to NGF, forskolin, or forskolin plus ionomycin stimulation, compared with the wild-type (WT) construct (Figure 4D). To further confirm the regulatory importance of these CREs, we generated a multi-site mutant carrying point mutations at five CREs (−854, −455, −412, −192, and −34) (Figure 4E). Mutation of all five sites markedly impaired promoter activity upon stimulation with NGF or forskolin, although a low level of residual activity remained (Figures 4F and 4G). Sequential restoration of individual WT CREs demonstrated that the reversion of any of the five sites partially, but significantly, rescued luciferase activity (Figures 4F and 4G). Among these, reintroduction of the promoter-proximal −34 CRE consistently produced the strongest rescue effect, indicating that this site plays a predominant role in stimulus-responsive transcription. No significant difference was observed between NGF- and forskolin-induced responses.

Following the demonstration that multiple CREs—particularly the promoter-proximal −34 site—are essential for stimulus-responsive Gpr3 transcription, we next examined whether CREB is activated and recruited to these elements in response to NGF or forskolin stimulation. Prior to chromatin immunoprecipitation (ChIP) analysis, we examined whether NGF or forskolin stimulation induces CREB activation in PC12 cells. Western blot analysis revealed that the p-CREB/CREB ratio was markedly and significantly increased as early as 15 min after NGF or forskolin stimulation, and this increase was sustained up to 60 min (Figures 5A and 5B). We then performed ChIP-qPCR analysis using anti-p-CREB (ser133) and total CREB antibodies to assess CREB binding to individual CREs within the Gpr3 promoter. Five primer-probe sets were designed to amplify genomic regions containing the −34, −192, −412/−455, −507, and −854 CREs (Figure 5C). Total CREB binding remained unchanged at all examined regions regardless of stimulation (Figures 5D–5H). In contrast, NGF or forskolin treatment selectively increased p-CREB binding at the promoter-proximal −34 CRE, with a modest increasing trend at the −192 CRE (Figures 5D and 5E). Among the upstream regions, including −412/−455, −507, and −854, no significant enrichment of p-CREB was detected under any stimulation condition (Figures 5F–5H).Figure 5CREB binding to the CRE-containing regions of the Gpr3 promoter following NGF and forskolin stimulation in PC12 cells(A and B) PC12 cells were stimulated with NGF (50 ng/mL) or forskolin (20 μM), and the expression levels of p-CREB (ser133) and total CREB were evaluated by western blotting at 0, 15, 30, and 60 min after stimulation.(A) Representative immunoblots showing p-CREB (ser133) and total CREB protein expression. p-, phosphorylated; right margin indicates molecular mass standards (in kilodaltons, kDa).(B) Densitometric quantification of time-dependent changes in p-CREB (ser133) and total CREB protein expression. For each time point, the p-CREB/CREB ratio was calculated and normalized to the value at 0 min. Data are presented as the mean ± SEM from six independent cultures (biological replicates). ∗∗∗p < 0.001, ∗∗p < 0.01, and ∗p < 0.05 vs. 0 min for each condition.(C–H) Chromatin immunoprecipitation (ChIP) assays were performed using anti-p-CREB (ser133) and total CREB antibodies to examine CREB binding to the CRE-containing regions of the rat Gpr3 promoter. PC12 cells were treated with NGF (50 ng/mL) or forskolin (20 μM) for the indicated times (0–60 min). Primer sets were designed to amplify five genomic regions (P1–P5) spanning the CRE-containing sequences within the 5′ flanking region of the Gpr3 gene. Data represent the mean ± SEM from five independent experiments and are normalized to 0 min in each condition. ∗∗∗p < 0.001 and ∗p < 0.05 vs. 0 min in each condition.CRE, cAMP response element; CREB, CRE-binding protein; NGF, nerve growth factor; p-CREB, phosphorylated CREB.Source data for this figure are provided in Data S1.

This pattern closely mirrored the NET-CAGE data, in which NGF and forskolin stimulation selectively increased transcription initiation signals near the TSS but not at upstream regions (Figure 3).

Together, these findings suggest that the clustered CREs within the proximal promoter region of Gpr3 cooperatively mediate its rapid transcriptional activation in response to NGF and cAMP signaling. While CREs positioned more distally upstream of the core promoter were inactive at the examined time points, they may possess latent enhancer activity that could contribute to Gpr3 transcription under different stimulation conditions or at later stages.

Gpr3 upregulation enhances Nr4a1–3 expression during neuronal differentiation in PC12 cells

We next explored whether early Gpr3 induction by NGF influences the expression of other neuronal genes. GPR3 constitutively activates Gs-type G proteins, leading to physiological elevations in intracellular cAMP levels.6^,^9^,^26 Recent studies have demonstrated that forskolin-mediated cAMP elevation promotes Nr4a1 expression through CREB phosphorylation, subsequently regulating downstream genes such as Syn1.20 The Nr4a family (Nr4a1–3) behaves as a group of IEGs that are transiently induced in response to neuronal stimulation and then re-expressed during later stages, displaying a biphasic expression pattern similar to that of Gpr3 (Figures 2B–2D). Notably, the untranslated regulatory regions of Nr4a1–3 also harbor multiple CRE motifs, reminiscent of GPR3 regulatory architecture.27 On the basis of these parallels, we hypothesized that Gpr3 upregulation may enhance the delayed-phase expression of Nr4a genes via cAMP signaling. To test this, we performed RT-qPCR analysis of Nr4a1–3 mRNA levels following Gpr3 knockdown in PC12 cells subjected to NGF-induced differentiation. Before assessing Nr4a expression, we confirmed that Gpr3 mRNA levels were significantly reduced by siRNA, with sustained suppression between 48 and 96 h (Figure S4A). In control siRNA-transfected cells, Nr4a1, Nr4a2, and Nr4a3 expression steadily increased over 96 h, whereas Gpr3 knockdown markedly suppressed this late-phase upregulation, particularly between 48 and 96 h post-induction (Figures 6A–6C). In contrast, Gpr3 upregulation under normal serum culture conditions resulted in a marked increase—approximately 1,000-fold compared with mock-transfected cells—confirming robust induction of Gpr3 (Figure S4B). Following Gpr3 upregulation, Nr4a1–3 mRNA levels robustly increased within 12 h, even in the absence of external differentiation stimuli (Figures 6D–6F).Figure 6Gpr3 induced during neuronal differentiation modulates Nr4a expression in PC12 cells(A–C) PC12 cells were transfected with control or Gpr3 siRNA. Twenty-four hours after transfection, differentiation was induced using serum deprivation (0.5% FBS) and NGF (50 ng/mL); Nr4a1 (A), Nr4a2 (B), and Nr4a3 (C) mRNA expression was evaluated at 0, 24, 48, and 96 h after differentiation.(D–F) PC12 cells were transfected with pc-mAGFL or pc-GPR3mAGFL. Nr4a1 (D), Nr4a2 (E), and Nr4a3 (F) mRNA expression was evaluated at 6, 12, 24, and 48 h after transfection. Data represent relative mRNA expression normalized to β-actin (A–F) and are presented as the mean ± SEM for each condition (three independent replicates). ∗∗∗p < 0.001, ∗∗p < 0.01, and ∗p < 0.05 vs. control at each time point.(G) GPR3-mediated Nr4a1 induction was analyzed using a luciferase-based promoter assay. Four putative CREB binding sites (CREs) at −242, −222, −78, and −49 bp upstream of the rat Nr4a1 gene are shown as black-filled squares. Reporter plasmids with CRE deletions or CRE point mutations (shown as white plaid boxes) were prepared as detailed in STAR Methods. PC12 cells were co-transfected with each reporter plasmid and either pc-mAGFL (mock) or pc-GPR3mAGFL plasmid vector. Twenty-four hours after transfection, the luciferase reporter assay was performed. For the positive control, mock-transfected cells were treated with forskolin (10 μM) 4 h prior to the assay. Data are presented as the mean ± SEM for each condition (three independent replicates). Each experiment was normalized to the wild-type plasmid in the mock-transfected group, and statistical comparisons were performed for each condition. ^‡‡‡^p < 0.001 vs. wild-type reporter plasmid + mock-transfected group; ∗∗∗p < 0.001 vs. wild-type reporter plasmid + pc-GPR3mAGFL-transfected group; ^###^p < 0.001 vs. wild-type reporter plasmid + mock-transfected group treated with forskolin.

Collectively, these findings suggest that NGF-mediated Gpr3 induction plays a role in enhancing the delayed-phase expression of the Nr4a gene family.

GPR3 enhances Nr4a1 transcriptional activity via a cAMP-CREB-dependent mechanism

Given that Gpr3 upregulation elevates intracellular cAMP levels and activates CREB and that Nr4a1 transcription is driven by CREs within its promoter, we next investigated whether GPR3 enhances Nr4a1 transcriptional activity via a cAMP-CREB-dependent pathway. Previous studies have reported that forskolin stimulation increases Nr4a1 expression through CREB phosphorylation.20 To specifically assess the role of GPR3, we constructed a luciferase reporter plasmid (pGL-rNR4A1WT) containing approximately 500 bp of the rat Nr4a1 5′ flanking region and generated deletion mutants that lacked individual CRE. Consistent with previous reports, forskolin treatment significantly increased luciferase activity driven by the Nr4a1 promoter in PC12 cells (Figure 6G). Similarly, transfection with a GPR3-expressing vector significantly enhanced Nr4a1 promoter activity compared with mock-transfected controls (Figure 6G). Notably, GPR3-dependent promoter activation was markedly reduced by sequential deletion of the −242, −222, and −78 CREs, and the −49 CRE site alone was insufficient to support promoter activity. Furthermore, point mutations introduced individually at the −242, −222, or −78 CRE substantially impaired the ability of GPR3 to induce Nr4a1 promoter activity (Figure 6G).

Together, these findings demonstrate that NGF-induced Gpr3 expression upregulates Nr4a1 transcription through a cAMP-CREB-dependent mechanism that targets specific CREs within their proximal promoter.

GPR3 modulates Syn1 expression via an NR4A1-dependent pathway during neuronal differentiation

Synapsins (Syn1–3) are key endogenous substrates of cAMP-dependent protein kinase (PKA) in presynaptic terminals and play essential roles in synaptic plasticity and neurotransmitter release. Previous studies have shown that forskolin-induced cAMP elevation upregulates Syn1 expression via an NR4A1-dependent mechanism in PC12 cells.19^,^20 As NR4A1 acts as a major transcriptional effector downstream of GPR3, we examined whether GPR3 similarly regulates Syn1 expression through an NR4A1-dependent pathway during neuronal differentiation. Following NGF-induced differentiation, expression levels of Syn1, Syn2, and Syn3 progressively increased over 96 h in control siRNA-transfected PC12 cells (Figures 7A–7C). Gpr3 knockdown significantly suppressed the induction of Syn1 and Syn3 at 96 h, while Syn2 expression showed a downward trend that did not reach statistical significance (Figures 7A–7C). To directly assess whether this regulation is mediated via NR4A transcription factors, we examined the effects of Nr4a isoform-specific knockdowns in the context of Gpr3 overexpression. Ectopic GPR3 expression robustly enhanced Syn1 mRNA levels within 24 h; however, this upregulation was abolished by co-transfection with Nr4a1-specific siRNA, but not with Nr4a2-or Nr4a3-specific siRNA (Figure 7D). To validate this mechanism in primary neurons, we further performed Gpr3 upregulation together with Nr4a knockdown in rat cerebellar granule neurons (CGNs). Consistent with the results obtained in PC12 cells, Gpr3-induced Syn1 upregulation in CGNs was markedly suppressed by Nr4a1 knockdown, whereas Nr4a2/3 knockdown had little or no effect (Figure 7E).Figure 7. Induced Gpr3 expression during neuronal differentiation modulates Syn1–3 expression in PC12 cells(A–C) Effects of the suppression of Gpr3 expression on Syn mRNA expression. PC12 cells were transfected with control or Gpr3 siRNA. Twenty-four hours after transfection, differentiation was induced using serum deprivation (0.5% FBS) and NGF (50 ng/mL). Syn1 (A), Syn2 (B), and Syn3 (C) mRNA expression was evaluated using RT-qPCR at 0, 24, 48, and 96 h after inducing differentiation. Data represent relative mRNA expression normalized to β-actin and are presented as the mean ± SEM for each condition (three independent replicates). ∗∗∗p < 0.001, ∗∗p < 0.01, and ∗p < 0.05 vs. control at each time point.(D) Effects of the suppression of Nr4a expression on GPR3-mediated Syn1 mRNA expression. PC12 cells were transfected with Nr4a1 siRNA, Nr4a2 siRNA, or Nr4a3 siRNA together with pc-GPR3mAGFL plasmids. For the control, cells were co-transfected with control siRNA and pc-mAGFL (mock). Syn1 mRNA expression was evaluated using RT-qPCR 24 h after transfection. Data represent relative mRNA expression normalized to β-actin and are presented as the mean ± SEM for each condition (three biological replicates). ∗∗∗p < 0.001 for each indicated group.(E) Primary cerebellar granule neurons (CGNs) were transfected with pc-GPR3mAGFL together with control siRNA, Nr4a1 siRNA, or Nr4a2/3 siRNAs. Syn1 mRNA expression was evaluated using RT-qPCR 24 h after transfection. Gpr3 upregulation significantly increased Syn1 expression, and this effect was abolished by Nr4a1 siRNA, whereas Nr4a2/3 siRNAs showed no significant suppression. Data represent relative mRNA expression normalized to β-actin and are presented as the mean ± SEM for each condition (three biological replicates). ∗∗p < 0.01 and ∗p < 0.05 for each indicated group.

Together, these data indicate that Gpr3 enhances Syn1 expression through a cAMP-CREB-NR4A1 signaling axis, linking early GPR3 induction to transcriptional programs that underlie synaptic remodeling during neuronal differentiation.

GPR3 deficiency suppresses the developmental upregulation of Syn1 and Nr4a family genes in primary cortical neurons

To further investigate whether GPR3 influences synaptic and transcriptional programs during neuronal maturation, we analyzed the temporal expression profiles of Syn1 and Nr4a family genes in primary cortical neurons prepared from WT and Gpr3 knockout (Gpr3^−/−^) mice. Neurons were cultured for up to 14 days in vitro (DIV), and mRNA levels were assessed at multiple time points. In WT neurons, Gpr3 mRNA expression gradually increased throughout the 14-day culture period (Figure 8A). Syn1 mRNA expression progressively increased between DIV7 and DIV14 in WT neurons, whereas this late-phase upregulation was significantly reduced in Gpr3^−/−^ neurons (Figure 8B). Nr4a1 mRNA exhibited a similar pattern, showing increased expression up to DIV7 followed by sustained levels through DIV14 in WT neurons, while GPR3^−/−^ neurons displayed a marked reduction in Nr4a1 expression between DIV7 and DIV14 (Figure 8C). Nr4a2 mRNA also progressively increased up to DIV7 in WT neurons, but its induction was significantly suppressed at earlier developmental stages (DIV4–DIV7) in Gpr3^−/−^ neurons (Figure 8D). Nr4a3 mRNA displayed a transient increase at DIV1 in WT neurons, and this early induction was markedly attenuated in Gpr3^−/−^ neurons (Figure 8E).Figure 8. Developmental changes in Gpr3, Syn1, and Nr4a family gene expression in primary cortical neurons from wild-type and Gpr3^−/−^ mice(A–E) Primary cortical neurons were isolated from postnatal day 0–1 (P0–P1) wild-type (WT), and Gpr3 knockout (Gpr3^−/−^) mice and cultured in vitro for up to 14 days (DIV0–DIV14). mRNA expression levels of Gpr3 (A), Syn1 (B), Nr4a1 (C), Nr4a2 (D), and Nr4a3 (E) were measured by RT-qPCR at DIV0, DIV0.7 (16 h), DIV1, DIV2, DIV4, DIV7, and DIV14. Data represent relative expression normalized to β-actin and are presented as the mean ± SEM for each condition (three biological replicates). ∗∗∗p < 0.001, ∗∗p < 0.01, and ∗p < 0.05 vs. WT neurons at each time point.

Together, these results demonstrate that the transcript levels of Syn-1 and Nr4a1 are strongly affected in GPR3^−/−^ primary cortical neurons during neuronal differentiation (DIV0–14). Moreover, the distinct temporal alterations observed for Nr4a2 and Nr4a3 indicate that GPR3 regulates multiple Nr4a family members in a phase-specific manner, suggesting roles beyond the control of Syn1 expression.

Loss of GPR3 reduces SYN1-positive vesicle density during cortical neuron differentiation in mice

We previously showed that GPR3 is broadly expressed in neurons across multiple brain regions in mice, including the cortex, hippocampus, and cerebellum.5^,^7^,^28 To determine whether GPR3 regulates Syn1 expression during neuronal differentiation, we cultured primary cortical neurons from WT and Gpr3 knockout (Gpr3^−/−^) mice. SYN1-positive vesicles were observed along neurites in both genotypes and co-localized with the presynaptic marker Bassoon (Figure 9A). However, quantitative analysis revealed that Gpr3^−/−^ neurons exhibited a significant reduction in SYN1-positive vesicle density compared with WT controls at both 7 and 14 DIVs (Figures 9B and 9C). This decrease in vesicle density is consistent with the reduced Syn1 mRNA expression observed in Gpr3^−/−^ cortical neurons (Figure 8), suggesting that GPR3 contributes to the transcriptional upregulation and subsequent accumulation of presynaptic vesicle components during neuronal differentiation.Figure 9. Reduced SYN1-positive puncta in cortical neurons from Gpr3^−/−^ mice in vitro(A) Staining for SYN1 and Bassoon in cortical neurons obtained from wild-type and Gpr3^−/−^ mice. Primary neurons were isolated from postnatal day 0–1 (P0–P1) and cultured for 14 days in vitro (DIV14). Neurons were fixed and immunostained with anti-SYN1 (green) and anti-Bassoon (red) antibodies. Representative images are shown.(B) For quantification, neurons were stained with anti-SYN1 (red) and DAPI (blue) at DIV7 and DIV14; representative images are shown.(C) SYN1-positive puncta were quantified within a region 10–25 μm from the nucleus. Quantification was performed using neurons from wild-type (n = 6) and Gpr3^−/−^ (n = 9) mice at DIV7 (left), and from wild-type (n = 11) and Gpr3^−/−^ (n = 9) mice at DIV14 (right) (C). Data are presented as the mean ± SEM for each condition. ∗p < 0.05 for each indicated group.

Discussion

In the present study, we identified GPR3 as a rare GPCR that exhibits IEG-like biphasic expression in response to NGF or forskolin stimulation in PC12 cells. NGF-induced GPR3 expression regulated Nr4a1–3 and Syn1 in a cAMP-CREB-dependent manner during late neuronal differentiation.

Classical IEGs, such as Fos, Jun, and the Nr4a family, are rapidly and transiently induced by neuronal activity and are essential for orchestrating transcriptional cascades that drive neuronal differentiation, synaptic remodeling, and adaptive plasticity.1^,^4 In the present study, we used NET-CAGE—a quantitative transcriptomic approach that is capable of capturing nascent RNA and precisely mapping promoter and enhancer regions21—to selectively profile immediate-early transcriptional responses in PC12 cells following NGF stimulation. Consistent with previous reports, classical neuronal IEGs such as Srf, Fos, Jun, Nr4a, Arc, and Egr were robustly induced within 30 min (Figure S3). Strikingly, Gpr3 was among the top-ranked IEGs exhibiting immediate-early gene-like induction, with levels of rapid induction similar to these canonical IEGs. (Figures 3 and S3). Importantly, no other GPCRs exhibited comparable immediate-early induction under these conditions, underscoring the unique transcriptional responsiveness of Gpr3. Moreover, our previous findings show that Gpr3 expression is acutely upregulated in T cells following immune stimulation, indicating that GPR3 displays IEG-like properties both in the neuronal and non-neuronal contexts.

We identified the Gpr3 proximal promoter (TSS to −200 bp) with two conserved CREs as essential for rapid induction by NGF or forskolin. This region is highly conserved among rats, mice, and humans, and it is annotated as the Gpr3 promoter in the FANTOM5 database. Luciferase assays showed that five CREs in the 1-kb upstream region cooperatively drive Gpr3 transcription, but NGF-induced p-CREB binding was restricted to the core promoter, with no activation at distal CREs observed in NET-CAGE or ChIP analyses. One possibility is that luciferase assays may overestimate the contribution of distal CREs because reporter plasmids lack chromatin structure. ChromHMM (Chromatin state discovery and characterization; https://compbio.mit.edu/ChromHMM/), which predicts chromatin states based on histone modifications, classifies this region of the human Gpr3 locus as a weak poised enhancer. Although the present study examined only the early phase after stimulation (0–60 min), we cannot exclude the possibility that distal CREs may become responsive during later phases, and this will require further investigation. Collectively, these findings suggest that this region may act as an enhancer under specific cellular contexts or stimulatory conditions that induce more robust Gpr3 expression.

In PC12 cells, approximately 6,300 genes have been identified to harbor CREB-binding regions near their exons.29 However, only a small subset of these genes are transcriptionally activated following CREB phosphorylation.30 In response to NGF stimulation, CREB is phosphorylated at ser133 through Ras-dependent activation of ribosomal S6 kinases (RSK),31 which subsequently recruits transcriptional coactivators such as CREB-binding protein/p300, promotes histone acetylation (notably H3K27ac), and enhances the transcription of CRE-containing genes such as c-fos.30^,^32 CREB phosphorylation can also be mediated by alternative pathways to RSK, including PKA activation via cAMP elevation33 and calcium/calmodulin-dependent protein kinase type IV activation via calcium influx.34 In the present study, Gpr3 transcription was robustly induced not only by NGF but also by forskolin, an adenylyl cyclase activator that elevates intracellular cAMP levels (Figures 2A, 3, and 4B–4F); this indicates that Gpr3 expression is, at least in part, regulated in a p-CREB-dependent manner. By contrast, calcium influx induced by either KCl depolarization or ionomycin alone did not induce Gpr3 transcription; however, when combined with NGF or forskolin, both calcium stimuli further enhanced Gpr3 expression (Figures 2A, 3, 4B, and 4D). This synergistic effect suggests the involvement of calcium-dependent coactivators such as CREB-regulated transcription coactivators, which enhance p-CREB-mediated transcriptional activation in response to calcium signaling.35 Moreover, several IEGs and neuropeptides, including Nr4a1, are cooperatively regulated by cAMP and calcium as integrated second messengers.1^,^20^,^27 Given that CREB phosphorylation can be triggered through multiple upstream pathways, Gpr3 transcription may plausibly be dynamically induced by stimuli other than NGF or forskolin, depending on cellular contexts and the convergence of signaling cascades. Thus, GPR3 may serve as a highly flexible transcriptional target within activity-dependent gene networks, finely tuned by diverse extracellular signals.

While our data indicate that Gpr3 transcription is primarily regulated by CREB in response to cAMP signaling, additional transcription factors may also be involved. In particular, the Gpr3 promoter harbors a TRE sequence in addition to CRE (Figure S1), suggesting a possible contribution of AP-1 factors. The AP-1 complex, typically composed of c-Fos and c-Jun dimers, binds TREs to activate early genes such as Egr1–4 and Arc.36 Supporting this possibility, siRNA-mediated knockdown of c-Fos and c-Jun partially reduced NGF-induced Gpr3 expression (data not shown). However, the relative contribution of AP-1 to Gpr3 regulation, compared to that of p-CREB, remains to be elucidated.

NGF stimulation rapidly induced Gpr3 expression within 1–2 h, followed by a decline at 3–4 h. Expression rose again at 24 h and remained elevated up to 96 h, showing a biphasic pattern (Figure 1B). NET-CAGE analysis also revealed a similar biphasic expression profile (Figure 3). A comparable biphasic pattern has been reported for Nr4a1–3 following NGF stimulation in neuronal models (Maruoka et al., 2020; Parra-Damas et al., 2017). Similar to Gpr3, Nr4a1–3 harbor multiple CREs and exhibit rapid but transient induction upon CREB phosphorylation, which is subsequently suppressed by negative transcriptional regulators such as myocyte enhancer factor 2 and inducible cAMP early repressor (Herdegen and Leah, 1998; Deisseroth et al., 1996). Secondary signals, including the production of brain-derived neurotrophic factor and sustained CREB activation, contribute to the delayed second wave of Nr4a expression (Sheng et al., 1991; West et al., 2001). Although the precise mechanism underlying the biphasic induction of Gpr3 remains unclear, it may involve transcriptional repressors such as myocyte enhancer factor 2 or inducible cAMP early repressor, similar to those reported for Nr4a genes. Additionally, early CREB-dependent activation of Gpr3 transcription may secondarily amplify the cAMP–CREB signaling pathway itself. Notably, no new CAGE signals or distal enhancer activity were detected up to 24 h post-NGF stimulation, suggesting that the biphasic profile is primarily regulated at the promoter level through transcription factor dynamics, rather than distal regulatory elements.

Here, we identified GPR3 as an activity-dependent regulator of NR4A family gene expression during neuronal differentiation. The Nr4a promoters contain multiple CREs,37 suggesting shared regulatory mechanisms with Gpr3. Consistent with prior studies,19^,^20 forskolin stimulation upregulated Nr4a expression through cAMP-mediated CREB activation. Importantly, NGF-induced Gpr3 expression also contributed to late-phase Nr4a1–3 upregulation, likely via sustained CREB phosphorylation driven by GPR3-mediated cAMP elevation. Furthermore, obtained data from primary cortical neuron demonstrate that Gpr3 deficiency suppresses physiological Nr4a1 upregulation during neuronal maturation, indicating that the GPR3–NR4A axis operates not only in PC12 cells but also in native developing neurons. These findings suggest that GPR3 acts as a key activity-dependent regulator that links Gs-cAMP signaling to temporally controlled transcriptional programs essential for neuronal development.

Moreover, NGF-induced GPR3 expression promoted Syn1 expression in a Nr4a1-dependent manner. Although the Syn1 promoter contains a CRE site that is activated by cAMP-PKA-mediated CREB phosphorylation,38^,^39 CREB-dependent Syn1 induction is typically transient, decaying within several hours. By contrast, we observed a sustained increase in Syn1–3 mRNA levels up to 96 h after NGF stimulation, and this late-phase upregulation was significantly attenuated by Gpr3 knockdown. Primary cortical neurons from Gpr3^−/−^ mice also exhibited impaired Syn1 upregulation, supporting the conserved role of GPR3 in Syn1 regulation across cell types. This finding is consistent with those of previous studies showing that forskolin-induced Nr4a1 activation promotes Syn1 expression through the cAMP–CREB axis (Maruoka et al., 2020), and suggests that in the late phase, GPR3-mediated upregulation of Syn1 specifically depends on NR4A1. The precise mechanisms by which NR4A1 enhances Syn1 transcription remain unclear and warrant further investigation.

In line with these findings, primary cortical neurons derived from Gpr3-knockout mice also exhibited a slight but statistically significant reduction in the number of SYN1-positive puncta after 1–2 weeks in culture (Figure 9). Synapsins are key presynaptic proteins essential for synaptic vesicle dynamics and neurite outgrowth.38^,^40^,^41 Beyond transcriptional control, cAMP promotes synapsin phosphorylation via PKA, modulating neurotransmitter release and synaptic plasticity.38^,^39 Despite the pivotal role of synapsins in neuronal development, endogenous mechanisms that elevate cAMP during differentiation remain poorly understood. Our results suggest that GPR3, as a ligand-independent activator of Gαs and cAMP,9^,^12^,^26 coordinates CREB-dependent transcription of targets like Nr4a and Syn1 to support neuronal differentiation. Moreover, cAMP elevation is known to enhance axon regeneration after injury,42 and we recently showed that GPR3 promotes this process by enhancing CREB phosphorylation in retinal ganglion cells.43 These findings collectively suggest that GPR3 contributes to synaptic plasticity and neuronal regeneration by modulating cAMP-CREB signaling.

Although Gpr3 knockdown or deletion led to a clear reduction in Nr4a1–3 expression with distinct temporal profiles, the decrease in SYN1-positive puncta observed in Gpr3^−/−^ cortical neurons was relatively modest. One plausible explanation is functional redundancy among NR4A family members. In lymphoid systems, NR4A1, NR4A2, and NR4A3 compensate for one another, and only combined deletion of all three leads to complete loss of regulatory T cell development (Sekiya et al., 2013; Safe et al., 2016). By analogy, residual NR4A2/3 activity may be sufficient to sustain Syn1 expression, even when NR4A1 induction is impaired. Moreover, NR4A receptors are regulated by multiple upstream signaling pathways beyond GPR3, including cAMP-PKA, calcium-CaMK, and MAPK-dependent pathways, which may provide alternative routes for maintaining NR4A activity.

Thus, in Gpr3^−/−^ neurons, compensatory transcriptional or signaling mechanisms might mitigate the impact of reduced GPR3 on synaptic gene expression and presynaptic maturation. Future studies clarifying isoform-specific and redundant NR4A functions in neurons will be required to define how these compensatory mechanisms shape presynaptic development under GPR3-deficient conditions.

Limitations of the study

This study has several limitations. First, although we demonstrate that GPR3 exhibits immediate-early gene-like induction and regulates NR4A-synapsin signaling during neuronal differentiation, most mechanistic analyses were performed in PC12 cells, which might not have fully captured all aspects of neuronal differentiation in vivo. Second, while loss-of-function and gain-of-function approaches support a functional role for GPR3 in regulating Nr4a family genes and Syn1 expression, the contribution of additional upstream signaling pathways and transcription factors, such as AP-1 and other calcium-dependent regulators, was not comprehensively examined. Third, although reduced SYN1-positive vesicle density was observed in Gpr3-deficient primary cortical neurons, the functional consequences on synaptic transmission or network activity were not directly assessed. Finally, potential redundancy among NR4A family members may partially compensate for GPR3 deficiency, which could mask the full extent of GPR3-dependent regulation. Future studies using in vivo models and functional synaptic analyses will be required to further define the physiological and pathological relevance of GPR3-mediated transcriptional programs.

Resource availability

Lead contact

Further information and requests for resources, reagents, and analytical protocols should be directed to and will be fulfilled by the lead contact, Shigeru Tanaka (tanakamd@hiroshima-u.ac.jp).

Materials availability

This study did not generate new unique reagents. All plasmids, reporter constructs, and siRNAs used in this work, including pc-GPR3mAGFL and pc-mAGFL expression vectors, Gpr3 promoter luciferase reporters, Nr4a1 promoter reporter constructs, siRNAs targeting Gpr3, Nr4a1–3, and c-fos are available from the lead contact without restriction.

Data and code availability

• •NET-CAGE data: Native elongating transcript-cap analysis of gene expression (NET-CAGE) datasets generated during this study have been deposited in the NCBI Sequence Read Archive (SRA): PRJNA1264035 and are publicly available.
• •Other data: All RT-qPCR datasets, luciferase assay raw values, and immunocytochemical quantifications supporting the findings of this study will be provided by the lead contact upon reasonable request.
• •Code availability: This study did not generate custom code. NET-CAGE analysis used standard pipelines (STAR, Paraclu, CAGEr, and DESeq2), fully described in STAR Methods.

Acknowledgments

This work was supported by the 10.13039/501100001691Japan Society for the Promotion of Science (JSPS) KAKENHI: Grant-in-Aid for Scientific Research (JP21K07274) to S.T., Grant-in-Aid for Start-up (JP23K19402) to H.S., and Grant-in-Aid for Young Scientists (JP25K18967) to H.S. We thank Kenta Sasaki, Kouta Narai, Taishin Mine, and Mao Wakahara for their contributions to preliminary experiments conducted as part of a laboratory course. We also thank the 10.13039/501100024233Natural Science Center for Basic Research and Development and 10.13039/501100024233Natural Science Center for Basic Research and Development, Hiroshima University for providing access to their facilities.

Author contributions

Conceptualization, S.T.; methodology, S.T.; data curation, S.T., F.I., H.S., and K.H.; formal analysis, S.T., F.I., and H.S.; investigation, S.T., F.I., and H.S.; validation, K.H. and I.H.; visualization, S.T.; writing – original draft, S.T. and F.I.; writing – review & editing, S.T., K.H., I.H., and N.S.; supervision, N.S.; project administration, S.T.; funding acquisition, S.T., H.S., and N.S.

Declaration of interests

The authors declare no competing interests.

STAR★Methods

Key resources table

REAGENT or RESOURCESOURCEIDENTIFIERAntibodiesRabbit monoclonal anti-Synapsin ICell Signining TechnologyCat# 5297; RRID: AB_2616578Mouse monoclonal anti–βIII-tubulin (Tub-βIII)Sigma–AldrichCat# T8578; RRID:AB_1841228Mouse monoclonal anti-Bassoon (SAP7F407)Enzo Life SciencesCat# ADI-VAM-PS003; RRID:AB_1659573Rabbit monoclonal anti–phospho-CREB (Ser133)Merck/MilliporeCat# 06-519Rabbit monoclonal anti-CREB (For Western blotting)Cell Signaling TechnologyCat# 9197; RRID:AB_331277Rabbit anti-CREB (ChIP-grade)MerckCat# 06-863Rabbit monoclonal anti-GAPDHCell Signining TechnologyCat# 2118; RRID:AB_561053Normal Rabbit IgGCell Signaling TechnologyCat# 2729; RRID:AB_1031062Alexa Fluor 488–goat anti-rabbit IgGThermo Fisher ScientificCat# A-11008; RRID:AB_143165Alexa Fluor 568–goat anti-mouse IgGThermo Fisher ScientificCat# A-11031; RRID:AB_144696Biological samplesPrimary cortical neurons (WT and Gpr3^−/−^; P0–P1)This studyN/AChemicals, peptides, and recombinant proteinsNGF (β-nerve growth factor)Thermo Fisher ScientificCat# 13257-019ForskolinNacalai TesqueCat# 160-00991IonomycinCalbiochemCat# 407950PMASigma–AldrichCat# P8139KClSigma–AldrichCat# P9333PolyethylenimineNacalai TesqueCat# 129-04214Cultrex Poly-L-LysineR&D SystemsCat# 3438-200-01Triton X-100Sigma–AldrichCat# X100RIPA bufferNacalai TesqueCat# 16488-14Neurobasal™-A MediumThermo Fisher/GibcoCat# 10888022B-27™ Supplement (50X)Thermo Fisher/GibcoCat# 17504044Opti-MEMThermo Fisher/GibcoCat# 31985070GlutaMAXThermo Fisher/GibcoCat# 35050061penicillin–streptomycinFujifilm WakoCat# 164-25251RPMI-1640Fujifilm WakoCat# 189-02025PBS(−)MedicagoCat# 09-2051-100FBSThermo Fisher/GibcoCat# A5256701LOT: B2672835RPCritical commercial assaysRNeasy Mini KitQiagenCat# 74104QuantiTect Reverse Transcription KitQiagenCat# 205313PrimeTime qPCR AssaysIDTVariousDual-Luciferase Reporter Assay SystemPromegaCat# E1910SimpleChIP Plus Enzymatic Chromatin IP KitCell Signaling TechnologyCat# 9005Micrococcal NucleaseCell Signaling TechnologyCat# 10011ChIP-Grade Protein G Magnetic BeadsCell Signaling TechnologyCat# 9006QIAquick PCR Purification KitQiagenCat# 28104Papain Dissociation SystemWorthington BiochemicalCat# LK003150EzFastBlotATTOCat# AE-1465EzWest Lumi PlusATTOCat# WSE-7120EzReprobeATTOCat# WSE-7240Deposited dataNET-CAGE sequencing dataNCBI Sequence Read Archive (SRA)NCBI SRA: PRJNA1264035Experimental models: Cell linesPC12 cellsRIKEN Cell BankCat# RCB0009Experimental models: Organisms/strainsGpr3 knockout miceDeltagen–Wistar rat (P7)Japan SLCN/AOligonucleotidesqPCR primersIDTPrimeTime AssaysPromoter deletion primersSigma–AldrichVariousCRE-mutant primersSigma–AldrichVariousGpr3 siRNASigma–AldrichSASI_Rn01_00067651Nr4a1 siRNASigma–AldrichSASI_Rn01_00042824Nr4a2 siRNASigma–AldrichSASI_Rn01_00055889Nr4a3 siRNASigma–AldrichSASI_Rn01_00120166Control siRNASigma–AldrichSIC001Recombinant DNApC-GPR3mAGFLPreviously describedN/ApC-mAGFLPreviously describedN/ApMAX-EGFPLonzaCat# VPA-1001Gpr3 promoter luciferase constructsThis studyVariousNr4a1 promoter luciferase constructsThis studyVariousSoftware and algorithmsPrism 9GraphPadhttps://www.graphpad.com/STAR aligner v2.7.9aDobin et al.44https://github.com/alexdobin/STARCAGErBioconductorhttps://bioconductor.org/packages/CAGErParacluFrith et al.45https://gitlab.com/mcfrith/paracluDESeq2Bioconductorhttps://bioconductor.org/packages/DESeq2Leica STELLARIS confocal microscopeLeica Microsystems–ImageJ (Fiji)NIHRRID:SCR_002285ABI PRISM 7500 Sequence Detection SystemApplied Biosystems–OtherP60/P100 dishesThermo Fisher Scientific–coverslipsMatsunami–PVDF membrane (Immobilon-P)MilliporeCat# IPVH00010Semi-dry blotter (PoweredBLOT 2M)ATTOCat# WSE-4045Ez-Capture MG (Chemiluminescence Imaging System)ATTOCat# AE-9300

Experimental model and study participant details

PC12 cell line

Rat pheochromocytoma PC12 cells (European Collection of Cell Cultures; originally obtained from RIKEN Cell Bank) were used as a model for neuronal differentiation.

Cells were maintained in RPMI-1640 medium supplemented with 15% fetal bovine serum (FBS) and 1% penicillin–streptomycin, and incubated at 37°C in a humidified 5% CO_2_ atmosphere. For differentiation assays, cells were plated onto polyethylenimine-coated dishes and switched to low-serum medium (0.5% FBS) prior to stimulation. PC12 cells are not derived from human subjects; therefore, sex, ancestry, and ethnicity are not applicable.

Animals

GPR3-knockout mice were generated from cryopreserved fertilized oocytes obtained from the Mutant Mouse Resource & Research Centers (Bar Harbor, ME) and rederived at the Institute of Laboratory Animal Science, Hiroshima University, under license from Deltagen (San Mateo, CA). All animal experiments were approved by the Animal Care and Use Committee of Hiroshima University (approval numbers A23–117 and A23-144-2). Both male and female pups at postnatal day 0–1 were used for primary cortical neuron cultures. Sex-dependent biological differences are not at this neonatal stage. No exclusion criteria were applied. All procedures were performed in compliance with institutional and national ethical guidelines.

Mouse strains

Primary cortical neurons were prepared from newborn (P0–P1) C57BL/6J mice carrying either wild-type or Gpr3 knockout (Gpr3^−/−^) alleles. Gpr3^−/−^ mice were generated by conventional gene targeting as previously described and maintained on a C57BL/6J background. Breeding pairs were heterozygous (Gpr3^+/−^ × Gpr3^+/−^), permitting isolation of WT and KO littermates for each experiment. Both male and female pups were used, and because neurons were isolated at P0–P1, no sex-related biological differences were expected or observed.

Primary cortical neuron culture

Cerebral cortices were isolated from postnatal day 0–1 (P0–P1) pups obtained from heterozygous Gpr3^+/−^ × Gpr3^+/−^ matings. Each pup was processed separately, and genotyping was performed retrospectively to distinguish wild-type and Gpr3^−/−^ littermates. Cortical tissues were enzymatically dissociated using the Worthington Papain Dissociation System, followed by gentle trituration. Dissociated neurons were resuspended in Neurobasal A medium supplemented with 2% B-27, 2 mM GlutaMAX, and 1% penicillin–streptomycin. Cells were plated at a density of 5 × 10^5^ cells/cm^2^ onto Cultrex poly-L-lysine (R&D Systems, Minneapolis, MN, USA) –coated coverslips (Matsunami, Osaka, Japan) or poly-L-lysine–coated 24-well culture dishes (BD Falcon). Neurons were maintained in Neurobasal-A medium supplemented with 2% B-27, 1% GlutaMAX, and antibiotics. Cultures were used for analysis between 7 and 14 days in vitro (DIV7–14) to assess Synapsin1 puncta formation and gene expression during neuronal maturation. Cells from WT and Gpr3^−/−^ pups were processed in parallel for each experiment to ensure comparability.

Primary cerebellar granule neuron (CGN) isolation

Cerebellar granule neurons were isolated from postnatal day 7 (P7) Wistar rat pups as previously described (Miyagi et al., PLoS One). Briefly, whole cerebella were dissected and enzymatically dissociated using the Worthington Papain Dissociation System, followed by Percoll density-gradient centrifugation (35%/60%). After centrifugation at 2,000 × g for 10 min, the CGN-enriched fraction was collected from the interface between the two layers, washed once with PBS, and resuspended in serum-free Opti-MEM for subsequent transfection experiments. Sex was not determined for cerebellar granule neurons, as cells were isolated at postnatal day 7, a developmental stage at which sex-dependent biological differences are not expected.

Ethical compliance

All procedures involving animals were conducted in accordance with the guidelines of the Hiroshima University Animal Care and Experimentation Committee, and adhered to the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Method details

Culture medium, chemicals, inhibitors, siRNAs, and plasmid vectors

RPMI 1640 medium and penicillin–streptomycin were purchased from Fujifilm Wako (Tokyo, Japan). Phosphate-buffered saline (PBS, without Ca^2+^/Mg^2+^) was obtained from Medicago (Uppsala, Sweden). Opti-MEM, Neurobasal A medium, B27 supplement, and GlutaMAX were from Thermo Fisher Scientific (Waltham, MA, USA). Fetal calf serum (FBS) was from Biological Industries (Cromwell, CT, USA). Ionomycin was obtained from Calbiochem (La Jolla, CA, USA). Phorbol 12-myristate 13-acetate (PMA) and potassium chloride (KCl) were from Sigma–Aldrich (St. Louis, MO, USA). Human β-nerve growth factor (NGF) was purchased from Thermo Fisher Scientific. Forskolin and polyethylenimine (for plate coating) were obtained from Nacalai Tesque (Kyoto, Japan). P100 and P60 culture dishes were from Thermo Fisher Scientific, and 6-well, 24-well, and 96-well plates were from Corning (Corning, NY, USA). Predesigned siRNAs targeting rat Gpr3, c-fos, Nr4a1, Nr4a2, Nr4a3, as well as Control siRNA, were purchased from Sigma–Aldrich. The FLAG-tagged GFP fusion vector expressing GPR3 (pc-GPR3mAGFL) and the control vector (pc-mAGFL) were described previously.13

Cell culture, induced neuronal differentiation, and image taken

PC12 cells were cultured in RPMI 1640 medium supplemented with 15% fetal bovine serum (FBS) and 1% penicillin–streptomycin in a humidified 5% CO_2_ incubator (PHCbi, Tokyo, Japan) at 37°C. For differentiation induction, cells were seeded onto P60 culture dishes (Thermo Fisher Scientific) pre-coated with 0.01% polyethylenimine at a density of 2 × 10^6^ cells per well. On the following day, cells were washed twice with phosphate-buffered saline (PBS, without Ca^2+^/Mg^2+^) and then cultured in Opti-MEM containing 0.5% FBS and 50 ng/mL nerve growth factor (NGF). In addition to NGF, PC12 differentiation was also induced under serum-deprived conditions (0.5% FBS) supplemented with forskolin (10 μM), potassium chloride (KCl; 50 mM), or ionomycin (1 μg/mL), respectively. To prevent the increase in medium osmolarity caused by KCl addition, the KCl stimulation medium was prepared by diluting Opti-MEM with sterile water to 74% of its original concentration (i.e., 26% water), thereby maintaining the original osmolarity of Opti-MEM (approximately 300 mOsm/L). Cell morphology was monitored at each time point after differentiation induction using differential interference contrast (DIC) imaging with a confocal microscope (STELLARIS 5; Leica Microsystems, Wetzlar, Germany).

For immunostaining, cells cultured on coverslips were fixed 48 h after stimulation with 4% paraformaldehyde (PFA) in PBS for 15 min at room temperature. After washing with PBS, cells were permeabilized with 0.1% Triton X-100 and blocked with PBS containing 3% normal goat serum for 1 h at room temperature. Cells were then incubated overnight at 4°C with mouse monoclonal anti-βIII-tubulin (Tub-βIII) antibody (1:200) and rabbit polyclonal anti-synapsin I (SYN1) antibody (1:200). After washing, cells were incubated for 1 h at room temperature with Alexa Fluor–conjugated secondary antibodies (1:400). Fluorescence images were acquired using a confocal laser scanning microscope.

Immunoblotting

PC12 cells were cultured under serum-deprived conditions (0.5% FBS) and stimulated as indicated. Protein samples (10 μg) were supplemented with 20 mM dithiothreitol, boiled at 98°C for 5 min, separated by Tris–glycine SDS–polyacrylamide gel electrophoresis (10% gel), and transferred onto polyvinylidene difluoride (PVDF) membranes (Millipore, Bedford, MA) using Semi-dry Blotter (WSE-4045 PoweredBLOT 2M; ATTO, Tokyo, Japan) with EzFastBlot (ATTO). Membranes were blocked with phosphate-buffered saline (PBS) containing 0.1% Tween 20 and 3% skim milk (Bio-Rad Laboratories, Inc., Hercules, CA) for 1 h at room temperature, and then incubated with primary antibodies overnight at 4°C. After washing three times with PBS containing 0.1% Tween 20, membranes were incubated with horseradish peroxidase–conjugated anti-mouse or anti-rabbit IgG secondary antibodies (1:10,000; Thermo Fisher Scientific) for 1 h at room temperature. Immunoreactive bands were detected using a chemiluminescence detection kit (EzWest Lumi Plus; ATTO) and quantified using a chemiluminescence image analyzer (Ez-Capture MG; ATTO).

For analysis of neuronal differentiation markers, PC12 cells were treated with NGF (50 ng/mL), forskolin (20 μM), or KCl (50 mM) for the indicated times (0–48 h). Cells were lysed in RIPA buffer supplemented with protease inhibitor cocktails (Nacalai Tesque). Membranes were probed with antibodies against βIII-tubulin (Tub-βIII; 1:1,000), synapsin I (SYN1; 1:1,000), and GAPDH (1:10,000). The same membranes were sequentially stripped using stripping buffer (EzReprobe; ATTO) and reprobed with the indicated antibodies. Band intensities were quantified by densitometric analysis, normalized to GAPDH, and expressed relative to the 0 h value for each condition. Quantitative data are presented as the mean ± SEM from five independent biological replicates.

For analysis of CREB activation, PC12 cells were stimulated with NGF (50 ng/mL) or forskolin (20 μM) for short time periods (0, 15, 30, and 60 min). Cells were lysed using a lysis buffer containing 50 mM Tris (pH 8.0), 150 mM NaCl, and 1% Triton X-100, supplemented with protease and phosphatase inhibitor cocktails. Membranes were probed with antibodies against phospho-CREB (Ser133) and total CREB. Band intensities were quantified from the same membranes, and the p-CREB/CREB ratio was calculated for each time point. Ratios were normalized to the value at 0 min. Quantitative data are presented as the mean ± SEM from three independent biological replicates.

Real-time RT-PCR analysis

Total RNA was isolated from cultured cells using the RNeasy Mini Kit (Qiagen, Venlo, Netherlands). RNA concentration and purity were assessed using a spectrophotometer (NanoDrop One; Thermo Fisher Scientific, Waltham, MA, USA). One microgram of total RNA was reverse transcribed into cDNA using the QuantiTect Reverse Transcription Kit (Qiagen). For PC12 cell experiments, primers and TaqMan probes specific for rat Gpr3, Nr4a1, Nr4a2, Nr4a3, Syn1, Syn2, Syn3, and β-actin were obtained from the PrimeTime Standard qPCR Assays (Integrated DNA Technologies, Coralville, IA, USA). Real-time quantitative PCR was performed using the ABI PRISM 7500 Sequence Detection System (Applied Biosystems, Foster City, CA, USA) and analyzed using the ΔCt method according to the manufacturer’s instructions. For primary cortical neuron experiments, total RNA was extracted from mouse neurons at the indicated days in vitro (DIV0–DIV14) using the same procedure as described above for PC12 cells. Gene expression was quantified using mouse-specific TaqMan probes for Gpr3, Nr4a1, Nr4a2, Nr4a3, Syn1, and β-actin. All reactions and cycling conditions were identical to those used for PC12 cells.

NET-CAGE analysis

CAGE (Cap Analysis of Gene Expression) library construction, sequencing, read mapping, and expression quantification were outsourced to DNAFORM (Yokohama, Kanagawa, Japan). The quality of total RNA was assessed using a Bioanalyzer (Agilent Technologies), and only samples with an RNA integrity number (RIN) greater than 7.0 were used for analysis. Complementary DNA (cDNA) was synthesized from total RNA using random primers. The ribose diols in the 5′ cap structure of RNA molecules were selectively oxidized and biotinylated, and RNA/cDNA hybrids were enriched using streptavidin-coated magnetic beads (cap-trapping). Following digestion of RNA with RNase ONE/H and ligation of adaptors to both ends of the cDNA, double-stranded cDNA libraries (CAGE libraries) were constructed. The libraries were sequenced on an Illumina NextSeq 500 platform using 75-base single-end reads. The obtained CAGE tags were mapped to the rat genome (Rnor_6.0) using the STAR aligner (version 2.7.9a). Tag counts were clustered using the CAGEr R package (Haberle et al., Nucleic Acids Res., 2015) together with the Paraclu algorithm (Frith et al., Genome Res., 2007) under default parameters. Clusters with counts per million (CPM) less than 0.2 were excluded. Differential gene expression analysis was performed using the DESeq2 package (version 1.20.0). Raw sequencing data are available from the Sequence Read Archive (SRA; The National Center for Biotechnology Information): PRJNA1264035.

Plasmid DNA and siRNA transfection

For DNA electroporation, 5 × 10^6^ cells were washed twice with PBS (without Ca^2+^/Mg^2+^) by centrifugation at 800 rpm for 5 min. The cell pellet was resuspended in 100 μL of Opti-MEM (Thermo Fisher Scientific). For plasmid DNA electroporation, 3 μg of plasmid DNA was added to the cell suspension. For RNAi electroporation, 1 μg of siRNA and 2 μg of a GFP-expressing plasmid (pMAX-EGFP; Basel, Switzerland) were co-transfected to enable visualization. The mixture was transferred to a 2 mm gap cuvette (Nepa Gene, Chiba, Japan). Electroporation was performed using the NEPA21 Super Electroporator (Nepa Gene) with the following parameters: Poring pulse: 150 V, 3.0 ms pulse length, 50 ms interval, 2 pulses, 10% decay rate. Transfer pulse: 20 V, 50 ms pulse length, 50 ms interval, 5 pulses, 40% decay rate. Following electroporation, cells were immediately transferred to 3 mL of DMEM containing 15% FBS and plated onto 0.01% polyethylenimine-coated 6-well plates at a density of 1 × 10^5^ cells/cm^2^. Twenty-four hours after plating, cells were induced to differentiate by replacing the medium as previously described and subsequently used for further experiments.

Luciferase-based promoter assays

A 5′ flanking region of the rat Gpr3 gene was amplified by PCR from rat genomic DNA and inserted into the pGL4 luciferase-based reporter plasmid vector (Promega, Madison, WI, USA) at Nhe I and Hind III sites. Deletion constructs of the rat Gpr3 promoter region (−854, −507, −455, −412) were constructed using KOD-Plus Mutagenesis Kit (TOYOBO) according to the manufacturer’s protocol. Promoter fragments were amplified by PCR using the following primer pairs (all sequences in lowercase):

PCR primers for full-length (WT) Gpr3 promoter construct

forward: gcccacataagtgacgtcacccagct

reverse: gcgaagcttcagccttccgccggccc

Deletion construct primers

Δ854 construct

• forward: ggcctacacctctctatcgtcaacgtccc
• reverse: gcgaagcttcagccttccgccggccc

Δ507–854 construct

• forward: gtcaccacgctcatatcgtcacagcca
• reverse: gcgaagcttcagccttccgccggccc

Δ455–507 construct

• forward: gtgctcctgtcaccgtcacccca
• reverse: gcgaagcttcagccttccgccggccc

Δ412–455 construct

• forward: gggcgccgcgtcaaggacag
• reverse: gcgaagcttcagccttccgccggccc

Site-directed mutagenesis of CRE within the Gpr3 –854, −455, −412, −192, and −34 promoter regions was also achieved using the KOD-Plus Mutagenesis Kit with the following primer sets: For constructs requiring multiple mutations, sequential site-directed mutagenesis was performed by using additional primer sets on the previously mutated plasmids.

CRE-mutagenesis primers

mCRE(854) construct

• forward: tgcccagctccttatttcaacc
• reverse: acgtcacttatgtgggccca

mCRE(455) construct

• forward: tgcagccacccacctccctgct
• reverse: gtcaccacgctcatatcgtcacagcca

mCRE(412) construct

• forward: tgccccagggccctatcagagg
• reverse: acggtgacaggagcacgagg

mCRE(192) construct

• forward: tgaggacagagagagccggctc
• reverse: acgcggcgccctctgcgggt

mCRE(34) construct

• forward: tgcccagctccttatttcaacc
• reverse: acgtctgaagcccagcagac

For luciferase-based promoter assays of Nr4a1, the 5′ flanking region of the rat Nr4a1 gene was amplified by PCR from rat genomic DNA and inserted into the pGL4 luciferase-based reporter plasmid vector. Deletion constructs (−242, −222, −78, −49) and CRE-mutated promoter constructs were generated using the KOD-Plus Mutagenesis Kit (TOYOBO) following the methods and primer sets described previously (ref. 20). The nucleotide sequences of all deletion and mutant constructs were confirmed by outsourcing sequence analysis.

Chromatin immunoprecipitation (ChIP) assay

Chromatin immunoprecipitation assays were performed using the SimpleChIP Plus Enzymatic Chromatin IP Kit (Magnetic Beads) (Cell Signaling Technology, #9005) according to the manufacturer’s instructions. In brief, 4 × 10^6^ PC12 cells were cultured in a 100-mm dish. Cells were crosslinked with 1% formaldehyde for 10 min at room temperature. Chromatin digestion was carried out enzymatically using Micrococcal Nuclease (Cell Signaling Technology, #10011). The nuclease volume was optimized (0.5 μL per IP) to generate chromatin fragments of approximately 150–900 bp, which represented an appropriate size range for subsequent immunoprecipitation.

PC12 cells were stimulated under five different conditions (control, NGF (50 ng/mL) for 30 or 60 min, or forskolin (20 μM) for 30 or 60 min), and chromatin derived from 5 × 10^6^ cells was incubated with an anti–phospho-CREB (Ser133) antibody (Merck, #06–519), an anti-total CREB antibody (Merck, #06–863), or normal rabbit IgG (Cell Signaling Technology, included in the kit) as a negative control. Immunoprecipitated chromatin complexes were captured and purified using ChIP-Grade Protein G Magnetic Beads (Cell Signaling Technology, #9006) followed by DNA purification with the columns included in the kit. Purified DNA was analyzed by real-time quantitative PCR. The following primer and probe sets (Integrated DNA Technologies, Coralville, IA, USA) were used to amplify genomic regions containing the −34, −192, −412/−455, −507, and −854 CRE sites within the 5′ flanking region of the rat Gpr3 gene (all sequences shown in lowercase):

Purified DNA was analyzed by real-time quantitative PCR. The P1–P5 primer and FAM-BHQ probe sets (Integrated DNA Technologies, Coralville, IA, USA) were designed to amplify the genomic regions containing the −34, −192, −412/−455, −507, and −854 CRE sites, respectively, within the 5′ flanking region of the rat Gpr3 gene (all sequences shown in lowercase):

P1 primer set (−34 CRE region)

• sense: cggtttggtgacgtctgaa
• antisense: cgcgaggcctcattactatg
• FAM-BHQ probe: agcagaccaatggacgctgc

P2 primer set (−192 CRE region)

• sense: cggctctctctgtccttga
• antisense: tcccaatcccttgacctctt
• FAM-BHQ probe: tttaaggctctcccactagcgtcc

P3 primer set (−412/−455 CRE region)

• sense: gcagcgaggacgctatt
• antisense: gtaagttgctgtcaggacca
• FAM-BHQ probe: acgtcacccagctccttatttcaacc

P4 primer set (−507 CRE region)

• sense: ctccaagatcctgtcaccac
• antisense: gacaggctttgcctctgata
• FAM-BHQ probe: catatcgtcacagccacccacctc

P5 primer set (−854 CRE region)

• sense: tcccaatcccttgacctctt
• antisense: cggctctctctgtccttga
• FAM-BHQ probe: tttaaggctctcccactagcgtcc

Transfection of cerebellar granule neurons (CGNs)

Plasmid and siRNA transfection into primary CGNs was performed using a NEPA21 Super Electroporator (Nepa Gene, Chiba, Japan) according to the manufacturer’s instructions. Briefly, 5 × 10^6^ cells were counted using a TC20 automated cell counter (Bio-Rad, Hercules, CA), resuspended in 100 μL of serum-free Neurobasal-A medium, and mixed with 30 μg of plasmid DNA or 2 μg of siRNA. The cell–nucleic acid mixture was transferred to a 2-mm-gap electroporation cuvette (Nepa Gene).

Electroporation was carried out using the following parameters: poring pulse—175 V, 2.5 ms pulse length, 50 ms pulse interval, 4 pulses, 10% decay rate; transfer pulse—20 V, 50 ms pulse length, 50 ms pulse interval, 5 pulses, 40% decay rate. Immediately after electroporation, cells were recovered in 600 μL of DMEM containing 10% FBS and plated onto poly-D-lysine–coated P60 dishes (100 μg/mL) at a density of 5 × 10^4^ cells/cm^2^ (equivalent to 1 × 10^6^ cells per P60 dish). Cells were collected at the indicated time points for subsequent RT-PCR analysis.

Primary mouse cortical neuron culture and immunostaining

Primary cortical neurons were isolated from postnatal day 0–1 pups obtained by mating male and female GPR3 heterozygous knockout mice. Each pup was processed individually, and cortical tissues were dissociated using the Worthington Papain Dissociation System (Worthington Biochemical, Lakewood, NJ), as previously described.13 Dissociated neurons were resuspended in Neurobasal A medium supplemented with 2% B-27, 2 mM GlutaMAX, and 1% penicillin–streptomycin. Cells were plated at a density of 5 × 10^5^ cells/cm^2^ onto Cultrex poly-L-lysine (R&D Systems, Minneapolis, MN, USA) –coated coverslips (Matsunami, Osaka, Japan) and cultured in a humidified incubator at 37°C with 5% CO_2_ for 7 to 14 days in vitro (DIV).

For immunostaining, neurons were fixed at DIV7 and DIV14 with 4% paraformaldehyde (PFA) in PBS for 15 min at room temperature. After washing, cells were permeabilized with 0.1% Triton X-100 and blocked with 3% normal goat serum in PBS for 1 h. Cells were then incubated overnight at 4°C with rabbit monoclonal anti-SYN1 antibody (1:200; Cell Signaling Technology, #5297) and/or mouse monoclonal anti-Bassoon antibody (1:200; Enzo, clone SAP7F407). For secondary detection, Alexa Fluor 488– or 568–conjugated goat anti-rabbit IgG (1:400; Thermo Fisher) and Alexa Fluor 568–conjugated goat anti-mouse IgG (1:400; Thermo Fisher) were used. Nuclei were counterstained with DAPI (10 ng/mL; Molecular Probes, D1306).

Images were acquired using a Leica STELLARIS 5 confocal microscope (Leica Microsystems, Wetzlar, Germany). SYN1-positive puncta were quantified using Leica LAS X software. A circular region of interest (ROI) with a radius of 25 μm was drawn around each DAPI-stained nucleus, excluding the central 10 μm region, and puncta within the 10–25 μm annular zone were counted. Image analysis was performed under blinded conditions with respect to genotype.

Quantification and statistical analysis

Statistical analyses were performed using Prism 9 software (GraphPad Software, San Diego, CA, USA). Data are presented as the mean ± standard error of the mean (SEM). Statistical significance was determined using unpaired Student’s t test for comparisons between two groups, or one-way analysis of variance (ANOVA) followed by Tukey’s multiple comparisons test for comparisons among multiple groups. A p-value less than 0.05 was considered statistically significant.

All quantitative data in this study represent biological replicates, defined as independently prepared PC12 cultures or independently prepared primary neuronal samples. When multiple measurements were obtained from the same culture (e.g., qRT-PCR wells or luciferase wells), these were treated as technical measurements and averaged to yield a single value for each biological replicate. Exact n values and definitions are provided in the corresponding figure legends.