Dynamic expression patterns of G protein-regulated inducer of neurite outgrowth 1 (GRIN1) and its colocalization with Gαo implicate significant roles of Gαo-GRIN1 signaling in nervous system

Differentiating neuronal cells widely express GRIN1 mRNA both in the central and peripheral nervous system during mouse embryogenesis

In order to comprehensively determine temporal and spatial expression patterns of GRIN1 during mouse embryogenesis, we applied in situ hybridization (ISH) technique that allows sensitive and reliable detection of mRNAs. Since hybridization probes may not deeply penetrate into thick samples, we performed ISH in whole-mount preparations for samples up to the embryonic day (E)10.5: for samples older than E10.5, we made transverse slices of embryos/brains with ∼500 μm thickness to maximize efficiency of probe hybridization and washings. Specificity of ISH was evaluated by parallel application of sense and antisense riboprobes. Our GRIN1 sense probes did not yield any detectable staining throughout the experiments (data not shown). An exception was the case when both sense and antisense probes were trapped within the embryonic brain vesicles, producing non-specific staining in the whole-mount preparations. In such cases, we identified specific signals by comparing sense to antisense signals. Under these conditions, we were able to obtain reproducible ISH results as follows:

Expression of GRIN1 mRNA was first detected at E9.5 within a confined region of the central nervous system (CNS) as well as subgroups of cells in the peripheral nervous system (PNS; Fig. 1C), while no intense expression was found at E8.5-9.0 (Fig. 1A and B). In the CNS, the earliest expression emerged within the ventral domain of the spinal cord at E9.5 (Fig. 1C), and this GRIN1 mRNA positive area gradually spread into the dorso-lateral region of the spinal cord by E10.5 (Fig. 1D). Transverse slices through upper body of the E10.5 embryo demonstrated that the signal was localized within the ventral horn (Fig. 1F). At E12.5, GRIN1 mRNA was detected in the whole area of the spinal cord except for the ventricular zone (Fig. 2C and D). This expression pattern continued during embryonic stages (Fig. 2F, 3H and 3P). Within the brain, clear GRIN1 signals were visible in the restricted developmental units such as the ventral midbrain and rhombic lip at E10.5 (Fig. 1D and E), whereas nonspecific background staining was accumulated in the brain ventricles from E9.5 onward in whole-mount preparations (Fig. 1C). Intense signals were also detectable in the diencephalon and telencephalon at E11.5 (Fig. 2A). While the dorsal portion of the telencephalon (i.e. the future cerebral cortex) had no GRIN1 signal until E12.5 (Fig. 2A and C), many of GRIN1 positive cells emerged progressively to fill the area at E14.5 (Fig. 2E). The appearance of GRIN1 mRNA in the olfactory bulb was observed at E12.5 (Fig. 2D), and the expression was maintained until adult stage (Fig. 2F, 3A, 3I, 4A and 5A). Dynamic features of GRIN1 expression profiles in the CNS well correlated to patterns of neuronal migrations and/or maturations. For instance, GRIN1 expression in the spinal cord started from the ventral region and this signal expanded dorsally to eventually cover entire domain, which well reflected patterns of neuronal maturations in the spinal cord (Nornes and Das, 1972). Also, the developmental changes of GRIN1 positive domains in the telencephalon appear to mirror the migration pattern of interneurons in telencephalic vesicles, where interneurons originate in the basal domain and migrate dorsally to fill the cerebral cortex at around E.12.5 (Anderson et al., 1997).

In the PNS, the expression of GRIN1 mRNA was first detected at E9.5 within the cranial ganglia, such as the trigeminal, facio-acoustic, glossopharyngeal and vagal ganglia, and within the dorsal root ganglia at the trunk level (Fig. 1C). The GRIN1 signals increased in all these regions at E10.5 (Fig. 1D) and these signals were maintained until E14.5 (Fig. 2A-F). The staining signal was also observed in the sympathetic ganglia at E11.5 (data not shown) and the expression was sustained until E14.5 (Fig. 2D and F). In the neural retina, the moderate level of expression was detected at E12.5 (Fig. 2D) and E14.5 (Fig. 2F), whereas no expression was observed in other sensory organs, tongue, olfactory epithelium and primordium of follicle of vibrissa, during embryonic stages tested (data not shown). It should be noted that positive signals were localized primarily in the differentiating neurons, while emigrating/migrating neural crest cells did not show any GRIN1 expression both in the cranial and trunk regions.

Most importantly, no GRIN1 expression was observed in the tissue/organs outside of the nervous system. In addition, both in the CNS and PNS, the expression was restricted in the differentiating cell populations at all developmental stages tested: cells in the ventricular zone and neural crest generating zone (i.e. roof plate) did not show any GRIN1 expressions during mouse embryogenesis.

GRIN1 mRNA expression is restricted to differentiating fields in the mouse developing brain

To further examine the expression profiles of GRIN1 mRNA during brain development, we prepared ∼500 μm slices of E14.5 to adult brains and subjected them to ISH. Cells expressing GRIN1 mRNA were found in almost entire differentiating fields in the embryonic CNS (Fig. 3), while they again became restricted to specific regions from the postnatal to adult stages (Fig. 4 and 5). The ISH results are detailed below for each brain subdivision.

A GRIN1 specific antibody labels GRIN1 mRNA positive cell populations in the adult mouse brain that substantially coexpressed Gαo protein

In order to determine sites of actions for GRIN1 protein, we next performed immunohistochemical staining on 30 μm sections of adult mouse brain by utilizing an anti-GRIN1 antibody against the oligopeptides containing GRIN1 specific sequences. We have previously confirmed the specificity of this GRIN1 antibody for immunostaining through the assay of competition with the immunogenic peptides or GRIN1 protein (Chen et al., 1999). Also, pre-immune antiserum yielded no staining (data not shown). The specificity of the antibody was further demonstrated by Western blot analysis (Fig. 6): the antibody reacts specifically with an apparent molecular weight of about 110 kDa in the mouse adult brain extract, while no significant signal was detected from the liver extract. Simultaneously, we examined colocalization of GRIN1 with Gαo using an anti-Gαo antibody. We also performed ISH on 14 μm sections of adult mouse brain to rigorously compare the distribution of GRIN1 mRNA and its protein.

First of all, we found that the overall labeling pattern of immunostaining with our anti-GRINl antibody in mouse adult brain was similar to that obtained in our ISH analysis. For instance, immunohistochemical labeling of the olfactory bulb, piriform cortex, amygdala, hippocampus, cerebrum and cerebellum by the GRIN1 specific antibody was noted with prominent intensity (Fig. 7G-L), which was consistent with results from ISH analysis (Fig. 5A-G and 7A-F). Within these GRIN1 positive areas, conspicuous immunostaining for Gαo was detected (Fig. 7M-R), and considerable overlaps between GRIN1 and Gαo staining were observed at various regions of the adult brain (Fig. 7S-X). The followings are summaries of detailed expression profiles by each area.

GRIN1 protein is specifically expressed in neurons and localized in the dendrite and axon with Gαo protein

Our ISH results demonstrated that GRIN1 mRNA was expressed in neuron-enriched regions of the developing and adult brains. To ensure the type of cells expressing GRIN1, we prepared monolayer cultures from the cerebellum of mouse P7 brain. The cerebellar cells were then analyzed by double immunostaining using antibodies to GRIN1 and MAP2 (neuronal marker) or GFAP (glial marker). As shown in Fig. 9A and 9B, GRIN1 immunoreactivities were preferentially found in MAP2-immunoreactive cells but not in GFAP-immunoreactive cells. These results demonstrate that GRIN1 is a neuron-specific protein.

Our immunoreactive mapping of GRIN1 protein on brain sections in Fig. 7 and 8 suggested that the protein might localize at the axons and/or dendrites of neuronal cells. In order to further examine the distribution of GRIN1 protein in neuronal cells, we doubly immunostained primary culture from the cerebellum with antibodies to GRIN1 and marker proteins, such as MAP2 for dendrites, neurofilament for axons, or synaptophysin for synapses. Double immunostaining of 10 days-cultured cells with typical neurites clearly demonstrated colocalization of GRIN1 with MAP2 and neurofilament, and partially with synaptophysin (Fig. 9B, C and D). We also found that GRIN1 and Gαo shared the same subcellular distribution in these cerebellar cells (Fig. 9E). Similar analyses were performed using the hippocampal cultures with essentially the same results (data not shown). These observations clearly indicate that GRIN1 protein is specifically expressed in neurons and arranged along the dendrites, axon, and synapses with its interacting partner, Gαo.

Spatiotemporal expression pattern of GRIN1 well correlated with differentiation waves of neuronal cells, implicating roles of GRIN1 in neuronal differentiation

Spatiotemporal patterns of neurogenesis in the rat nervous system have been characterized extensively (for review, see Bayer and Altman, 2004). Given that rat development is generally ∼1.5 day behind the mouse, GRIN1 expression profiles determined in the mouse brain well reflect patterns of neurogenesis mapped during rat brain development. For instance, in the rat hippocampal formation, neurogenesis begins at E16 with the generation of pyramidal cells. The granule cells in the dentate gyrus are noted for their late time of origin and most granule cells are generated after birth in rat. Likewise, GRIN1 mRNA expression was first detected in the mouse hippocampal formation at E14.5, and the distinct expression in the mouse dentate gyrus was observed at P2 and adult stages. GRIN1 expression in the mouse nervous system tended to be highest in the developing fields at stage coincident with the peak of neurogenesis.

In addition to the spatial and temporal features of GRIN1 expression patterns, following observations supported that GRIN1 is preferentially expressed along the postmitotic yet immature neurons. First, earliest hybridization signals were detectable in the restricted developmental units such as the cranial ganglia and ventral spinal cord at E9.5 when first neurons start to migrate and differentiate (Easter et al., 1993; Mastick and Easter, 1996). Second, the proliferative ventricular zones exhibited no expression throughout the neuroaxis at all developmental stages tested. Third, in the adult brain, prominent ISH signals were detected in the olfactory bulb, piriform cortex, and dentate gyrus where neuronal differentiation takes place (Pekcec et al., 2006; for review, see Abrous et al., 2005). These observations indicated that GRIN1 probably may not play any roles in developmental decisions of neuroblast cells, since it only appeared in the differentiating field of the nervous system. Instead, GRIN1 is likely to function in signal transduction pathways involved in neuronal differentiation.

Similar expression profiles between GRIN1 and Gαo suggests physiological significance of Gαo-GRIN1 signaling in neuronal cells

Our previous reports have shown that GRIN1 is specifically expressed in the brain and is enriched in the nerve growth cones (Chen et al., 1999). In this study, we demonstrated that mouse GRIN1 transcript appeared in the maturing postmitotic neurons from E9.5 and was maintained until the adult stage. Among members of Gαi/o subfamily, only Gαo appeared to harbor similar expression profiles to GRIN1 (Strathmann et al., 1990; Strittmatter et al., 1990; Schmidt et al., 1994; Morishita et al., 1999). We also confirmed that GRIN1 protein is colocalized with Gαo protein in the section of the adult brain and that GRIN1 and Gαo share the dendritic, axonal and synaptic localization in the primarily cultured neurons. In addition to the in vivo and in vitro similarity of expression profiles between GRIN1 and Gαo, coexpression of GRIN1 with a constitutively active mutant of Gαo promotes neurite extension in a neuronal cell line (Chen et al., 1999; Nakata et al., 2005), supporting the notion that Gαo-GRIN1 interaction plays physiologically significant roles in the nervous system.

We have previously shown that GRIN1 also interact with Gαi members in vitro (Chen et al., 1999). Our data do not exclude interactions of GRIN1 with Gαi members in vivo. For instance, in the main olfactory bulb where Gαo exhibits only weak expression, GRIN1 may function with Gαi subunits.

Possible functions of Gαi/o-GRIN1 signaling in the mouse developing nervous system

Landmarks of neuronal development include controlled birth of immature neurons from neuronal precursors, their directed migrations from the birthplaces to final positions, neurite outgrowth and guidance of the motile growth cone towards postsynaptic partners, and finally the generation and/or maintenance of synapses along functional neuronal circuits. The enrichment of GRIN1 in the developing fields of the nervous system along with its time course of expression in the elaborated brain organizations suggested that GRIN1 could be more or less involved in these processes of neural development.

Although the results need to be confirmed by double staining with markers for migratory neurons, our ISH analysis suggested that GRIN1 was expressed in a migratory population of neurons: GRIN1 mRNA expression in the rhombic lip and posterior precerebellar extramural migration implicates that the GRIN1 might be involved in dorsoventral migration from the caudal rhombic lip to the precerebellar nuclei of the brainstem (Wang et al., 2005). Furthermore, abundant GRIN1 expression was observed in the olfactory bulb and dentate gyrus of the adult brain, where migrations and incorporations of immature neurons were continuously processed. Importantly, Gαi/o-coupled receptor signaling, in which GRIN1 might participate, is known to regulate directed neuronal migrations. GABA_B receptors, for example, harbor a critical modulatory role in the migration of cortical interneurons and affect the length of their leading process (López-Bendito et al., 2003). SDF-1/CXCR-4 signalings can also control migration of the hippocampal, cerebellar and cortical neurons (Bagri et al., 2002; Zhu et al., 2002; Stumm et al., 2003). Taken together, it is likely that Gαi/o-GRIN1 signaling has regulatory function in directed migration of immature neurons.

Several lines of evidence further argue for a specific role of Gαi/o signaling in regulating neurite outgrowth and axon guidance. Go is a major G protein subclass accumulated in the growth cone membrane (Strittmatter et al., 1990). It was also shown that the transfection of PC12 cells with constitutively active Gαo mutant doubles the neurite length and increase the number of neurites (Strittmatter et al., 1994). N-CAM- and N-cadherin-induced neurite outgrowth from PC12 cells can be further blocked by pertussis toxin, a specific inhibitor for Gαi/o (Doherty et al., 1991). Gαo also modulates neurite outgrowth by regulating the signaling pathway initiated by cyclic AMP in F11 neuroblastoma cells (Ghil et al., 2000). Many neurotransmitters have also been reported to control neurite growth and axonal guidance: D2 dopamine, serotonin 1B, CB1 cannabinoid, and CXCR4 receptors regulates neurite growth (Reinoso et al., 1996; Lotto et al., 1999; Jordan et al., 2005); CXCR4 and GABA_B receptors guide nerve growth cones (Xiang et al., 2002); and CXCR4 and 5-HT_1B/1D receptors modulates axonal responsiveness to guidance cues (Chalasani et al., 2003; Bonnin et al., 2007). A common feature among these receptors is that they couple and function through the Gαi/o pathway. These findings, together with our observations that GRIN1 protein was localized along the dendrites, axons and synapses, and that Gαo-GRIN1 signaling pathway could regulate neurite growth/extension (Chen et al., 1999; Nakata et al., 2005), lead us to conclude that Gαi/o-GRIN1 interaction may influence cellular differentiation through GPCR-Gαi/o-mediated signaling for neurite growth and nerve growth cone guidance, which would bring effects on neuronal circuit formation/maintenance in the developing and adult brains. Such a hypothesis will be further examined by generating GRIN1 gene knockout mice or using RNAi approach for GRIN1 in the future studies.

Animals

All animal experiments in this study conform to Japanese governmental guidelines and have been approved by the Animal Care and Use Committee of the National Institute of Neuroscience, Japan. ICR (CD-1) mice (Charles River, Japan) were mated, and embryos at various stages were collected. The day of appearance of the vaginal plug was taken as E0 and the day of birth was regarded as P0. C57BL/6 mice (Charles River, Japan) were used for immunohistochemistry and primary culture of cerebellar cells.

Antibodies

For GRIN1 protein detection, a polyclonal antibody against GRIN1 specific peptide corresponding to amino acid residues 550 to 567 (Chen et al. 1999) was used. Anti-GFAP (glial fibrillary acidic protein) monoclonal antibodies (clone G-A-5, Sigma, MO, USA), anti-MAP2 (microtubule-associated protein 2) monoclonal antibodies (clone HM-2, Sigma, MO, USA), anti-neurofilament monoclonal antibodies (clone 2H3, Developmental Studies Hybridoma Bank, IA, USA), anti-synaptophysin monoclonal antibodies (clone SVP-38, Sigma, MO, USA) and anti-Gαo monoclonal antibodies (clone 2A.3, BIOMOL, PA, USA) were applied to mark distinct cell types and cellular compartments.

A Plasmid Construct

A 3.6kb mouse GRIN1 cDNA fragment was released from the pCMV5-GRIN1 expression vector (Chen et al., 1999) by using KpnI and HindIII restriction enzymes. The 3.6kb fragment was then cloned into KpnI-HindIII sites of pBluescriptII KS(-) (Stratagene, CA, USA), yielded pBS-GRIN1.

Synthesis of Hybridization Probes

Sense or antisense GRIN1 probes labeled with digoxigenin (Dig)-11-UTP (Roche Diagnostic GmbH, Penzberg, Germany) were prepared from pBS-GRIN1. For whole-mount in situ hybridization of GRIN1, pBS-GRIN1 linearized by Asp718 was transcribed with T7 RNA polymerase for antisense probes (Invitrogen, CA, USA) and that linearized by SpeI was transcribed with T3 RNA polymerase for sense probes (Roche Diagnostic GmbH, Penzberg, Germany). These probes cover the full-length coding region of GRIN1. For section in situ hybridization of GRIN1, pBS-GRIN1 was digested at the XhoI site, which cleaved 3 prime untranslated region of GRIN1 cDNA, generating a template for the antisense RNA probes. T7 RNA polymerase was then used to transcribe the riboprobes only covering the 956bp 3 prime untranslated region of GRIN1 gene, whose DNA sequences are specific for GRIN1 among the mouse homologues. These RNA probes were purified by G-50 Sephadex column (Roche Diagnostic GmbH, Penzberg, Germany), and stored at -80°C.

Whole-Mount In Situ Hybridization

Embryos were dissected out to remove extraembryonic membranes in Tyrode’s solution and fixed overnight in 4% paraformaldehyde (PFA) in PBS at 4°C. Fixed embryos were washed with phosphate-buffered saline (PBS) containing 0.1% Tween 20 (PBT) and bleached with 5:1 methanol/30% hydrogen peroxide for 6 hours at room temperature. Bleached embryos were dehydrated into 100% methanol and stored at -20°C. For sliced samples, we fixed embryos and brains overnight in PBS containing 4% PFA at 4°C and bleached them for 6 hours at room temperature in 5:1 methanol/30% hydrogen peroxide. After rehydration in a graded series of methanol/PBT, they were embedded in 2% agarose/PBS, and sectioned into 500-μm slices by microslicer (D.S.K. DTK-3000, Kyoto, Japan). Slices were dehydrated into 100% methanol and stored at -20°C. Whole-mount in situ hybridization was performed as described previously (Conlon et al., 1992; Inoue et al., 1997; Inoue et al., 1998).

Western Blot Analysis

The liver and brain were dissected out from a 10 weeks-old C57BL/6 female mouse. Approximately 0.6 mg of tissue was sonicated in 5 volume of RIPA buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS). Debris was removed by centrifugation at 20,000 x g for 30 min at 4°C. Supernatants of tissue extracts were subjected to protein analysis using a DC Protein Assay (BioRad, CA, USA). 10 μg protein (per lane) of tissue extracts were separated on 10% SDS-polyacrylamide gel, and transferred onto nitrocellulose membrane. Non-specific binding sites were blocked with 5% skim milk in PBT for 30 min at room temperature. The blot was incubated overnight at 4°C with anti-GRIN1 antibody (at 1:2,000 dilution with 5% skim milk in PBT). After washing with PBT, horseradish-peroxidase-conjugated anti-rabbit IgG was applied as the secondary antibody (at 1:5,000 dilution with 5% skim milk in PBT) for 45 min at room temperature. Finally, the blot was incubated with ECL plus Western blotting detection reagents (Amersham Pharmacia Biotech, IL, USA) and chemiluminescence was detected with an imager (LAS-1000; Fuji Film, Tokyo, Japan).

Section In Situ Hybridization

Section in situ hybridization was performed as described previously (Funatsu et al., 2004).

Primary Culture of the Cerebellar Cells

The cerebella were dissected out from 7-day-old mouse pups and minced into 1 mm slices in dissection buffer (10 mM HEPES, 137 mM NaCl, 5.3 mM KCl, 33 mM glucose, 45 mM sucrose). Then, the minced cerebella were incubated in dissection buffer containing 20 U/ml papain (Worthington, NJ, USA) for 20 min at 37°C. The tissues were washed three times with neuronal medium (Minimum Essential Medium (Invitrogen, CA, USA), 6 mM glucose, 2 mM L-glutamine, 0.1% Serum Extender (Becton Dickinson, MA, USA), 10% horse serum (Invitrogen, CA, USA), 20 mM KCl), and dissociated by trituration with a fire-polished Pasteur pipette. The cerebellar cells (500 cells/mm²) were seeded onto glass coverslips coated with poly-L-lysine (Sigma, MO, USA) and collagen (Vitrogen, CA, USA), and were cultured in neuronal medium. After 3 days in culture, 100 nM 5-Fluoro-2′ deoxyuridine (Sigma, MO, USA) and 250 nM uridine (Sigma, MO, USA) were added in cultured medium for suppression of glial cells proliferation, and cells were incubated for 24 hours. After 6 days, the cells were used for the experiments.

Immunohistochemistry

Under anesthesia, adult C57BL/6 mice were perfused through the ascending aorta with 4% PFA in PBS. The whole brains were removed, postfixed in 4% PFA in PBS overnight, and immersed in 30% sucrose in PBS for 48 hours at 4°C. To prepare cryosections, brains were placed in OCT compound and snap-frozen on dry ice. Cryosections (30∼40 μm in thickness) were prepared and free-floating immunohistochemistry were performed. The sections were treated with 0.5% Triton X-100 in PBS for 30 min, and blocked with 5% skim milk in PBS for 2 hours. After the blocking, sections were incubated with 5% skim milk/PBS containing anti-GRIN1 antibody (1:300) and anti-Gαo antibody (1:200) for 48 hours at 4°C. After washing, the sections were incubated with Alexa 594 goat anti-rabbit IgG antibody (1:1000) and Alexa 488 goat anti-mouse IgG antibody (1:500) in PBS containing 5% skim milk at 4°C overnight. After serial washings, sections were mounted onto adhesive slides and coverslipped with PermaFluor Mountant Medium (Shandon Immunon, PA, USA).

For immunohistochemistry on primarily cultured cells, the cerebellar cells at 10 days in vitro (DIV) were fixed with 4% PFA in PBS at 4°C for 20 min. Cells were permeabilized with 0.5% Triton X-100 in PBS for 5 min at room temperature. Cells were then incubated with 5% skim milk in PBS for 1 hour at room temperature, followed by incubation for 48 hours at 4°C in 5% skim milk/PBS containing anti-GRIN1 antibody (1:300) plus antibodies against marker proteins such as anti-GFAP antibody (1:750), anti-MAP2 antibody (1:500), anti-neurofilament antibody (1:300), anti-synaptophysin antibody (1:500) or anti-Gαo antibody (1:300). After washing, cells were incubated with Cy5 goat anti-rabbit IgG antibody (1:1000) and Alexa 488 goat anti-mouse IgG antibody (1:500) for 90 min at room temperature. After immunoreaction, cells were mounted with PermaFluor Mountant Medium. Fluorescent images were captured using fluorescence microscopes (Axiophot2, Carl Zeiss Jena Gmbh, Jena, Germany; DM5000B, Leica, Wetzlar, Germany) equipped with CCD cameras (ORCA-ER, Hamamatsu, Japan; DFC300FX, Leica, Wetzlar, Germany) or a confocal laser scanning microscope (TCS SP2, Leica, Wetzlar, Germany).

Identification of GRIN1 Positive Regions

For anatomical identification of GRIN1 positive regions, we referred to atlases by Paxinos and Franklin (2001), Kaufmann (1992), and Altman and Bayer (1995).