Address correspondence and reprint requests to Shoichi Ishiura, Department of Life Sciences, Graduate School of Arts and Sciences, University of Tokyo, 3-8-1 Komaba, Meguro-ku, Tokyo 153–8902, Japan. E-mail: email@example.com
The neurotransmitter dopamine plays an important role in the regulation of behavior in both vertebrates and invertebrates. In mammals, dopamine binds and activates two classes of dopamine receptors, D1-like and D2-like receptors. However, D2-like dopamine receptors in Caenorhabditis elegans have not yet been characterized. We have cloned a cDNA encoding a putative C. elegans D2-like dopamine receptor. The deduced amino acid sequence of the cloned cDNA shows higher sequence similarities to vertebrate D2-like dopamine receptors than to D1-like receptors. Two splice variants that differ in the length of their predicted third intracellular loops were identified. The receptor heterologously expressed in cultured cells showed high affinity binding to [125I]iodo-lysergic acid diethylamide. Dopamine showed the highest affinity for this receptor among several amine neurotransmitters tested. Activation of the heterologously expressed receptor led to the inhibition of cyclic AMP production, confirming that this receptor has the functional property of a D2-like receptor. We have also analyzed the expression pattern of this receptor and found that the receptor is expressed in several neurons including all the dopaminergic neurons in C. elegans.
Dopamine is a major neurotransmitter that modulates neuroendocrine, locomotor and emotional functions. The dopamine signal is transmitted through dopamine receptors on the cell surface. Agonists and antagonists of dopamine receptors are known to play roles in several neuroleptic disorders. In mammals, five dopamine receptors have been identified and all of them are G protein-coupled, seven transmembrane receptors (Missale et al. 1998; Vallone et al. 2000). According to their sequence similarities and pharmacological profiles, the receptors are divided into two classes, D1-like and D2-like receptors. D1-like receptors are capable of activating adenylyl cyclase upon stimulation by agonists, which leads to an increase in intracellular cyclic AMP (cAMP), whereas D2-like receptors inhibit adenylyl cyclase. Dopamine receptors have also been shown to have the ability to couple with other signaling pathways. However, the coupling of dopamine receptors to second messenger pathways has been studied predominantly in cultured cell lines transfected with receptor cDNAs. The heterologous expression system might not reproduce the in vivo receptor environment.
The nervous system of Caenorhabditis elegans consists of 302 neurons for which the synaptic connectivities have been determined (White et al. 1986). Since the complete genome, powerful genetics, and germline transformations are available, as well as a number of assays to quantify the behavior of the animals have been developed, C. elegans is a desirable model organism in which to study the molecular mechanisms of the nervous system.
Eight dopaminergic neurons (Sulston et al. 1975) and several genes that act in dopamine synthesis and signaling (Wintle and Van Tol 2001) exist in C. elegans. Exogenous dopamine inhibits locomotion and egg laying (Schafer and Kenyon 1995; Weinshenker et al. 1995), and mutant strains deficient in dopamine have defects in food sensation (Duerr et al. 1999; Sawin et al. 2000). These reports show that dopamine plays important roles in the control of behavior in C. elegans. The types of behavior controlled by dopamine are also reported to be regulated by G proteins (Mendel et al. 1995; Segalat et al. 1995; Brundage et al. 1996). These molecules might act downstream of dopamine and dopamine receptors. In the previous study we cloned the C. elegans dopamine receptor (CeDOP1), which shows high sequence similarity to vertebrate and invertebrate D1-like dopamine receptors (Suo et al. 2002). The characterization of a D2-like receptor in C. elegans is necessary to study in what signaling pathways D1-like and D2-like dopamine receptors are involved in vivo.
In this study, we performed cDNA cloning of the predicted gene K09G1.4, which shows high sequence similarity to mammalian D2-like receptors. Cloning the cDNA revealed the presence of two splice variants with different lengths of the intracellular loop, which is also found in mammalian D2 receptors (Dal Toso et al. 1989; Giros et al. 1989; Monsma et al. 1989). The receptor was expressed heterologously in mammalian cultured cells, and the pharmacological profile and its effect on adenylyl cyclase were determined to reveal that this receptor is the D2-like receptor of C. elegans (CeDOP2). We also analyzed the expression pattern of CeDOP2 in C. elegans and found that it is expressed in several neurons including all dopaminergic neurons.
Materials and methods
cDNA cloning of CeDOP2
We carried out BLAST searches against the protein sequences predicted from the C. elegans genome by Genefinder using mammalian dopamine receptor amino acid sequences. The predicted gene K09G1.4 showed high sequence homology to mammalian dopamine receptors. Therefore, we conducted the cDNA cloning of K09G1.4. Total RNA was isolated from wild type Bristol N2 strain of C. elegans with TriZOL reagent (Gibco BRL, Rockville, MD, USA) according to manufacturer's protocol. The oligonucleotides AS1 (AAACATCAGTTGACCCGA), AS2 (CACCCAAAAAACCCATGG), and SL1 (GGTTTAATTACCCAAGTTTGAG) were used for RT-PCR to clone K09G1.4 from the total RNA. AS1 and AS2 match the predicted 3′ untranslated region of K09G1.4, and SL1 matches the 5′-trans-spliced leader sequence found on the C. elegans mRNAs (Krause and Hirsh 1987). Reverse-transcription of the total RNA was performed with AS1 and the Thermoscript RT-PCR System (Gibco BRL). PCR was carried out in a total volume of 20 µL comprising LA buffer, 0.4 mm dNTPs, 0.5 U of LA-Taq polymerase (TAKARA BIO, Shiga, Japan), primers at 0.5 µm and the synthesized cDNA. The reaction was begun with a single denaturation step at 97°C for 3 min, followed by 30 cycles of 30 s at 97°C, 30 s at 59°C, and 3 min at 72°C. The Reactions were completed by one step at 72°C for 10 min The PCR products of 2.6 kbp and 2.2 kbp were gel purified and ligated into pGEM-T easy vector (Promega, Madison, WI, USA). The cDNA sequences of the receptors were determined using a Thermo Sequenase fluorescent labeled primer cycle sequencing kit (Amersham Pharmacia Biotech, Piscataway, NJ, USA) according to the manufacturer's instructions, and two splice variants (CeDOP2L and CeDOP2S) were identified. The cloned cDNA contained the 5′ untranslated region, the complete protein coding regions, and the partial 3′ untranslated region.
Expression of CeDOP2 in cultured cells
The sequence coding triple HA tags was inserted into the Not I site of pSecTagB to obtain pSecHA. The coding sequences of CeDOP2S and CeDOP2L were amplified by PCR using oligonucleotides F1 (GGTGATCCAGCTAGCCACCATGGAGGCCG) and R1 (TTGGGCCCTTAGACATGCGCCTGCTTG), which were designed to contain restriction enzyme sites. The PCR product was digested with Nhe I and Apa I, gel purified, and ligated into Xba I- and Apa I-digested pSecHA to obtain pHACeDOP2S and pHACeDOP2L. pHACeDOP2S and pHACeDOP2L were transiently transfected into exponentially growing COS-7 cells or CHO cells using FuGENE 6 (Roche, Indianapolis, IN, USA) according to manufacturer's instructions. For western blotting, transfected cells were washed with PBS and scraped from the culture dishes 40 h after transfection. The total proteins were extracted by sonication, loaded onto a 10% polyacrylamide gel, and then transferred onto a polyvinylidene difluoride (PVDF) membrane (Nihon Eido, Tokyo, Japan). The membrane was incubated with anti-HA antibody (12CA5, Roche) and visualized with a VECTASTAIN Elite ABC kit (Vector Laboratories, Burlingame, CA, USA) and a POD immunostain set (Wako, Osaka, Japan).
Radioligand binding assays
Radioligand binding assays were performed as described (Suo et al. 2002). In brief, 30 h after transfection, CeDOP2S or CeDOP2L expressing COS-7 cells were washed and collected in ice cold 50 mm Tris-HCl (pH 7.4). The cells were treated with a Polytron homogenizer (setting 8 for 10 s), and the homogenate was centrifuged at 40 000 × g for 30 min. The resulting pellet was resuspended and stored at −80°C. Protein concentration was determined using a DC Protein assay kit (Bio-Rad Laboratories, Hercules, CA, USA). The membranes (10–15 µg) were added to assay tubes containing 50 mm Tris-HCl (pH 7.4), 2 mm ascorbic acid and [125I]iodo-lysergic acid diethylamide ([125I]iodo-LSD, 2200 Ci/mmol; NEN Life Science Products, Boston, MA, USA) with or without competitors (all purchased from Sigma, St Louis, MO, USA), and incubated for 30 min at room temperature. Nonspecific binding was determined by coincubation with 100 µm dopamine. Assays were terminated by rapid dilution with ice-cold 50 mm Tris-HCl (pH 7.4), followed by filtration through a Whatman GF/B filter presoaked in 0.3% polyethylenimine. The filters were washed three times with 3 mL of 50 mm Tris-HCl (pH 7.4). The radioligand remaining on the filters was detected by a gamma counter (Aloka). For [3H]spiperone (113 Ci/mmol; Amersham Pharmacia Biotech, Buckinghamshire, UK) and [3H]SCH23390 (75.5 Ci/mmol; NEN Life Science Products) binding assays, the radioactivity remaining on the filters was measured by liquid scintillation counting. All data are representative of three independent experiments performed in duplicate. Analyses of the binding data were performed with the computer program Prism (GraphPad Software).
Measurement of cAMP formation
CHO cells grown on 60-mm culture dishes were transiently transfected with control vector, pHACeDOP2S, or pHACeDOP2L together with pCRE-Luc (Clonetech, Palo Alto, CA, USA). pCRE-Luc works as indicator of intracellular cAMP levels, and an increased cAMP level results in an elevation of luciferase activity (George et al. 1998). Ten hours after transfection, cells were trypsinized and re-seeded onto 96-well plates, and incubated for 14 h. Cells were then incubated with serum-free medium containing 1 µm forskolin, increasing concentrations of dopamine, and 10 µm antagonists for 3 h. Medium was removed and luciferase activity was measured using Luciferase Assay System (Promega). EC50 values were calculated with the computer program Prism (GraphPad Software).
Analysis of the expression pattern
The oligonucleotides F2 (CGAACCCCATCTGCAGCGGACCTAAAC) and R2 (CCATTGGATCCGCTAATGTTAGGCTGAC) were used to amplify a genomic DNA fragment from 3.5 kb upstream to exon 4 of the K09G1.4 gene from the C. elegans genomic DNA. The PCR fragment was inserted into pGEM-T easy vector (pGEM-CeDOP2). The Apa I and BamH I fragment from the YFP vector pPD136.64 was ligated into Apa I- and BamH I-digested pGEM-CeDOP2 to obtain CeDOP2-YFP. The oligonucleotides F3 (CGTGTTGTTAAGAACGTGCTTGATCG) and R3 (CTCATTGTCTAGAGTTTCCACGTGG) were used to amplify the genomic DNA fragment from 2.3 kb upstream to exon 1 of the cat-2 gene. The PCR fragment was inserted into the CFP vector pPD133.48 using the Hind III site and the Xba I site, which is contained in R3 (cat-2-NLS-CFP). pPD133.48 contains nuclear localization signals fused to CFP. CeDOP2-YFP and cat-2-CFP were coinjected into animals with pRF4 (Mello et al. 1991). The animals carrying extrachromosomal arrays were selected by the rol phenotype and were examined with a confocal laser microscope (LSM, Oberkochen, Germany, Zeiss).
Cloning of CeDOP2
The predicted protein K09G1.4 shows similarities to vertebrate dopamine D2-like receptors. To clone the cDNA for K09G1.4, RT-PCR was carried out using oligonucleotides matching the predicted 3′ untranslated region of the gene and an SL1 trans-spliced leader sequence. Two bands corresponding to 2.6 kb and 2.2 kb were observed after agarose gel electrophoresis of the PCR product. The PCR products were cloned and sequenced to reveal the presence of two splice variants. The cloned cDNAs contained complete protein coding regions. By comparison with the genome DNA sequence, we determined the exon–intron structure of K09G1.4 (Fig. 1a). The cloned sequences are roughly identical to the sequence predicted by GeneFinder, but several exon–intron junctions were not predicted correctly. An extra exon was inserted in CeDOP2L to produce an mRNA longer than CeDOP2S.
The hydropathy profile of the deduced amino acid sequence of CeDOP2 predicts seven putative transmembrane domains (TMs, Fig. 1b). Many characteristic features of G protein-coupled receptors were identified in CeDOP2, such as a conserved DRY sequence following the end of TM III, several consensus phosphorylation sites for protein kinases in the intracellular loops and the C-terminus, and a consensus N-glycosylation site in the N-terminus. CeDOP2 also contains amino acid residues thought to be important for the binding of dopamine, such as an aspartic acid residue in TM III (D118) and two serine residues in TM V (S203 and S206, Livingstone et al. 1992; Pollock et al. 1992). The splice variants of CeDOP2 differ in their third intracellular loops. CeDOP2L has a 135 amino acid insertion after position 267 (Fig. 1b). The deduced amino acid sequence of CeDOP2 shows high similarity to vertebrate and invertebrate dopamine receptors. A phylogenetic tree of CeDOP2 and human and invertebrate amine receptors shows that CeDOP2 is related to D2-like dopamine receptors (Fig. 2). Highest homology to CeDOP2 was seen with Drosophila melanogaster D2-like receptor DD2R (Hearn et al. 2002) and the identity was 51%. CeDOP2 shows higher homology with mammalian D2-like receptors (44% identical to human D2, 42% to D3 and 41% to D4) than with mammalian D1-like receptors (37% to human D1 and 37% to D5), which suggests that CeDOP2 is a D2-like dopamine receptor. Previously characterized invertebrate D1-like dopamine receptors, Apis mellifera AmDOP1 (Blenau et al. 1998), C. elegans CeDOP1 (Suo et al. 2002), Drosophila melanogaster DmDOP1 (Gotzes et al. 1994; Sugamori et al. 1995), and Drosophila melanogaster DAMB (Feng et al. 1996; Han et al. 1996) showed lower homologies than mammalian D2-like receptors (35, 37, 36, and 38%, respectively). The receptors of other biogenic amines also showed less identity.
Pharmacological profiles of CeDOP2
N-Terminally HA-tagged CeDOP2S and CeDOP2L were transiently expressed in COS-7 cells (Fig. 3). The membrane preparation from transfected cells was examined for its ability to bind [125I]iodo-LSD. Figure 4(a) shows the saturation binding curves and Scatchard plots for [125I]iodo-LSD binding to CeDOP2S (KD = 6.5 ± 2.1 nm, Bmax = 20.8 ± 4.9 pmol/mg protein) and CeDOP2L (KD = 6.6 ± 0.6 nm, Bmax = 14.9 ± 2.4 pmol/mg protein). COS-7 cells transfected with the control plasmid showed no specific binding (data not shown). CeDOP2S and CeDOP2L expressing cells were also examined for their ability to bind the D1-like dopamine receptor antagonist [3H]SCH23390 (10 nm) and the D2-like dopamine receptor antagonist [3H]spiperone (10 nm), but no specific binding was observed (data not shown). Several biogenic amine neurotransmitters were tested for their abilities to displace [125I]iodo-LSD binding to CeDOP2 splice variants (Fig. 4b,c). Dopamine was the most potent competitor among the drugs tested for both CeDOP2S and CeDOP2L with calculated dissociation constants (Ki) of 2.98 ± 0.21 µm and 2.24 ± 0.19 µm, respectively. In contrast, norepinephrine, serotonin, octopamine, and tyramine were 10–1000 times less potent competitors (Table 1). Several dopamine receptor antagonists were also tested for their ability to displace [125I]iodo-LSD binding (Table 1). The pharmacological profiles of CeDOP2S and CeDOP2L were essentially the same. Butaclamol showed the highest affinity to CeDOP2. SCH23390 and spiperone are known to exhibit high selectivities to mammalian D1-like receptor and mammalian D2-like receptor, respectively (Missale et al. 1998; Vallone et al. 2000). In contrast to vertebrate dopamine receptors, CeDOP2 showed similar and relatively low affinities for these two compounds. The rank order of affinity of the antagonists for AmDOP1, previously characterized Apis mellifera D1-like receptor, was chlorpromazine > spiperone > butaclamol > SCH23390 > haloperidol (Blenau et al. 1998), which was not similar to that of CeDOP2.
Table 1. Dissociation constants (Ki) for [125I]iodo-LSD binding to CeDOP2S and CeDOP2L
Dissociation constants (Ki, in µm) of biogenic amines and dopamine receptor antagonists in displacing [125I]iodo-LSD are listed. Kis were obtained using the Prism program and are expressed as mean ± SEM of three independent experiments performed in duplicates.
2.98 ± 0.21
2.24 ± 0.19
32.7 ± 2.0
33.5 ± 2.6
81.0 ± 13.4
106 ± 4
205 ± 70
152 ± 24
314 ± 72
312 ± 61
0.0350 ± 0.0082
0.0365 ± 0.0037
0.297 ± 0.013
0.325 ± 0.027
0.367 ± 0.053
0.431 ± 0.039
0.405 ± 0.011
0.408 ± 0.022
0.658 ± 0.031
0.884 ± 0.044
2.08 ± 0.25
2.25 ± 0.02
50.9 ± 9.1
51.2 ± 9.6
Effect of CeDOP2 on cAMP
D2-like dopamine receptors are known to inhibit adenylyl cyclase upon stimulation by agonists resulting in decreased cAMP levels. Since CeDOP2 shows high homology to mammalian D2-like receptors, we examined its ability to inhibit adenylyl cyclase. HA tagged CeDOP2S and HA tagged CeDOP2L were transiently expressed in CHO cells (Fig. 3), together with pCRE-Luc. Luciferase activities were elevated when cells were incubated with forskolin, which increases the intracellular cAMP levels. In both CeDOP2S and CeDOP2L expressing cells, dopamine attenuated the forskolin stimulated increases in luciferase activities in dose responsive manner (Fig. 5a). Dopamine had similar potencies on both splice variants and EC50s for CeDOP2S and CeDOP2L were 75 ±4.9 nm and 73 ± 8.8 nm, respectively. The same dopamine receptor antagonists used in the ligand binding experiments were examined for their ability to inhibit the dopamine-induced decrease in the luciferase activity. Butaclamol (10 µm) inhibited the effect of 0.1 µm dopamine both in CeDOP2S and CeDOP2L expressing cell (Fig. 5b), while other antagonists (clozapine, SCH23390, haloperidol, spiperone, chlorpromazine, and sulpride) did not (data not shown).
Expression pattern of CeDOP2
CeDOP2-YFP and cat-2-NLS-CFP vectors were coinjected into C. elegans. cat-2 encodes C. elegans tyrosine hydroxylase and it is reported that the expression of cat-2 is restricted to the dopaminergic neurons (Lints and Emmons 1999). There are 14 dopaminergic neurons in adult male (CEP, ADE, PDE, R5A, R7A, and R9A, from anterior to posterior, Fig. 6a). There are eight dopaminergic neurons in hermaphrodite which does not have ray neurons, R5A, R7A, and R9A. We observed the expression of CFP, which indicates the expression of cat-2, in eight neurons in a hermaphrodite and considered them to be CEP, ADE, and PDE neurons from the anterior to the posterior side of the body (Fig. 6b,f). The expression of YFP, which indicates the expression of CeDOP2, was observed in several neurons around the nerve ring and the posterior side of the body (Fig. 6c,g). The merged image indicates that expression of CFP is seen only in the YFP-expressing neurons (Fig. 6d,h). Since the nuclear localization signals were fused to CFP, the fluorescence of CFP seemed to be localized in the nuclei. YFP fluorescence is observed in the cytoplasm of the same cells. These results show that CeDOP2 is expressed in all the dopaminergic neurons and some other neurons. C. elegans were injected individually with CeDOP2-YFP or cat-2-NLS-CFP. The expression patterns in animals carrying CeDOP2-YFP or cat-2-NLS-CFP were not different from that of animals carrying both vectors (data not shown), which indicates that the colocalizations shown here are not artifacts resulting from the coinjection. The tail region of male animals injected with CeDOP2-YFP and cat-2-NLS-CFP were also examined. However, the fluorescence of CFP in adult male ray neurons was very strong and it was impossible to separate it from the fluorescence of YFP. Figure 7 shows an adult male carrying CeDOP2-YFP but not cat-2-NLS-CFP. The expression of YFP indicates that CeDOP2 is also expressed in the male tail region.
In this study, we cloned, characterized and analyzed the expression pattern of a C. elegans D2-like dopamine receptor (CeDOP2). Comparisons between the deduced amino acid sequence of CeDOP2 and vertebrate and invertebrate dopamine receptors showed that CeCOP2 resembles D2-like receptors rather than D1-like receptors. In addition, CeDOP2 has a relatively long third intracellular loop and a short C-terminal tail, which is typical of D2-like receptors. These features of CeDOP2 suggest that it is a D2-like receptor in C. elegans.
Radioligand binding experiments with [125I]iodo-LSD confirmed that CeDOP2 is a dopamine receptor of C. elegans. The binding experiments showed that both CeDOP2S and CeDOP2L expressed in COS-7 cells have higher affinities for dopamine than other biogenic amine neurotransmitters, which suggests that the endogenous ligand for this receptor is dopamine. CeDOP2, however, shows no specific binding to the D2-like receptor antagonist [3H]spiperone or the D1-like receptor antagonist [3H]SCH23390. Since invertebrate D1-like dopamine receptors are known to have a low affinity for [3H]SCH23390 (Sugamori et al. 1995; Suo et al. 2002), invertebrate receptors might, in general, have a low affinity for the subtype-specific ligands for mammalian receptors. Our competitive binding experiment showed that known dopamine receptor antagonists have the abilities to displace [125I]iodo-LSD, but the dissociation constants for CeDOP2 showed no apparent correlation with those for mammalian D1-like and D2-like receptors (Missale et al. 1998; Vallone et al. 2000) or Apis mellifera D1-like receptor AmDOP1 (Blenau et al. 1998).
cAMP assays showed that both splice variants have the abilities to inhibit adenylyl cyclase activity, presumably by coupling with Gi, just as known D2-like receptors do. Butaclamol, which showed the highest affinity for CeDOP2 in the ligand binding experiments, inhibited the effect of CeDOP2 on the inhibition of cAMP production. The other antagonists showed no inhibition, which might be due to their low affinities for CeDOP2. It is reported that butaclamol can also inhibit the Drosophila melanogaster D2-like receptor DD2R while spiperone or haloperidol does not (Hearn et al. 2002).
Characterization of both splice variants of CeDOP2 revealed that CeDOP2S and CeDOP2L have essentially identical pharmacological properties. This is comparable to the human D2 receptor splice variants. The human D2 receptor splice variants with different length of the third intracellular loop (Dal Toso et al. 1989; Giros et al. 1989; Monsma et al. 1989) exhibit similar pharmacological properties (Castro and Strange 1993; Malmberg et al. 1993). However, the human D2 receptor variants were reported to differ in their potencies on inhibition of intracellular cAMP (Dal Toso et al. 1989), while CeDOP2S and CeDOP2L showed similar potencies.
We have shown that CeDOP2 is expressed in several neurons, including the dopaminergic neurons, by comparing with the expression pattern with that of cat-2. This suggests that CeDOP2 might function as an autoreceptor in regulating the release of dopamine at the presynapse, as well as a heteroreceptor acting at the postsynapse. D2 receptors have been shown to function as autoreceptors in mammals (Mercuri et al. 1997; L'hirondel et al. 1998). It is reported that presynaptic dopamine transporter is regulated by D2-like dopamine receptor (Meiergerd et al. 1993; Cass and Gerhardt 1994). Dopamine transporter of C. elegans is shown to be expressed in dopaminergic neurons (Jayanthi et al. 1998; Nass et al. 2002). There could be a conserved mechanism for the regulation of presynaptic function in C. elegans and mammals. CeDOP2 contains a splice variation in the third intracellular loop. In mammal, it has been reported that D2L (the long variant) and D2S (the short variant) are differentially distributed (Khan et al. 1998), and that D2L acts mainly at postsynaptic sites and D2S functions as a presynaptic autoreceptor (Usiello et al. 2000; Wang et al. 2000). In this study, we have not determined whether CeDOP2 splice variants have different distributions and different functions. C. elegans, however, is suitable for studying the expression pattern of genes. Isolation of a mutant for CeDOP2 and its characterization would enable us to determine what behavior CeDOP2 regulates and what signaling molecules are involved in the dopamine receptor signaling in vivo, and the introduction of each splice variant to the mutant will reveal the functional differences between splice variants.
We thank Drs S. Okamoto and M. Ohmori for help with the cAMP measurement and Dr A. Fire for providing us with the vectors. We also thank Drs A. Kawaguchi and A. Iwamoto-Kihara for valuable discussions. S. S. is supported by a JSPS Research Fellowship for Young Scientists. This work is supported in part by a Grant-in-Aid for Scientific Research on Priority Areas (C) − Advanced Brain Science Project − from the Ministry of Education, Culture, Sports, Science and Technology, Japan.