|Home | About | Journals | Submit | Contact Us | Français|
Oligodendrocyte precursor cells (OPCs) express NMDA receptors (NMDARs) and form synapses with glutamatergic neurons throughout the central nervous system (CNS). Although glutamate influences the proliferation and maturation of these progenitors in vitro, the role of NMDAR signaling in oligodendrogenesis and myelination in vivo is not known. Here, we investigated the consequences of genetically deleting the obligatory NMDAR subunit NR1 from OPCs and their oligodendrocyte progeny in the CNS of developing and mature mice. NMDAR-deficient OPCs proliferated normally, achieved appropriate densities in gray and white matter, and differentiated to form major white matter tracts without delay. OPCs also retained their characteristic physiological and morphological properties in the absence of NMDAR signaling, and were able to form synapses with glutamatergic axons. However, expression of calcium permeable AMPA receptors was enhanced in NMDAR-deficient OPCs. These results suggest that NMDAR signaling is not used to control OPC development, but to regulate AMPAR-dependent signaling with surrounding axons, pointing to additional functions for these ubiquitous glial cells.
Glia in the mammalian CNS express neurotransmitter receptors that may enable these cells to rapidly adapt to the changing needs of surrounding neurons. Indeed, glutamate and GABA receptor signaling in glia has been implicated in processes as diverse as synaptic plasticity, functional hyperemia, and cell differentiation (Belachew and Gallo, 2004; Perea et al., 2009; Attwell et al., 2010). Nevertheless, the consequences of this signaling in vivo are largely unknown. As many therapeutics used to treat neurological disorders and limit injury during stroke act on neurotransmitter receptors (Muller et al., 1995; Hoyte et al., 2004; Zarate et al., 2010), it is important to evaluate the roles of neurotransmitter signaling to glia in both physiological and pathological contexts.
The CNS contains an abundant class of progenitor cells that express the chondroitin sulfate proteoglycan NG2 and the alpha receptor for platelet derived growth factor (PDGFαR) (Nishiyama, 2007). These NG2+ glial cells are often referred to as oligodendrocyte precursor cells (OPCs), as they generate oligodendrocytes (OLs) during early postnatal development (Nishiyama, 2007). OPCs remain abundant in the adult CNS and retain the ability to differentiate into OLs (Rivers et al., 2008; Kang et al., 2010), and can regenerate OLs after they have been destroyed through chemical or autoimmune-mediated demyelination (Franklin and Ffrench-Constant, 2008; Tripathi et al., 2010). However, in chronic stages of disease in multiple sclerosis, axons remain demyelinated despite the presence of these progenitors near lesions (Chang et al., 2000), highlighting the need for a greater understanding of the factors that regulate OPC differentiation in vivo.
In all brain regions examined, OPCs express ionotropic glutamate receptors and form synapses with axons of glutamatergic neurons (Bergles et al., 2010), suggesting that OPC behavior may be influenced by neuronal activity (Barres and Raff, 1993). In support of this hypothesis, glutamate alters the proliferation, differentiation, and migration of OPCs in vitro (Gallo et al., 1996; Gudz et al., 2006). Both OPCs and OLs express functional NMDARs (Karadottir et al., 2005; Ziskin et al., 2007) that can induce intracellular Ca2+ transients (Micu et al., 2006), raising the possibility that signaling through these receptors modulates OPC physiology and controls myelination. Although NMDAR signaling influences the survival, migration, and differentiation of neuronal progenitors during development and shapes the physiological properties of mature neurons in the adult CNS (Komuro and Rakic, 1993; Ikonomidou et al., 1999; Malenka and Bear, 2004; Nacher and McEwen, 2006), the in vivo role of NMDAR signaling in OPCs has not been determined.
To evaluate the function of NMDAR signaling in OL lineage cells, we genetically deleted the NMDAR subunit NR1 from OPCs and their OL progeny in mice. Removal of NMDARs did not alter the survival of OPCs, their proliferation and migration, or their ability to develop into OLs during postnatal development. Moreover, myelination of major fiber tracts in the CNS proceeded normally, and these mice did not exhibit behaviors associated with myelin deficiency, indicating that NMDAR activation in OL lineage cells does not play a critical role in regulating oligodendrogenesis or myelination in vivo.
NR1flox/flox mice have been described previously (Fukushima et al., 2009). In these mice, exons 19–20 (putative transmembrane segment M4) of the gene for the NR1 subunit (Grin1) are flanked by LoxP sites. Olig1cre/+ mice were obtained from Dr. David Rowitch (Univeristy of California, San Francisco) and have been described previously (Lu et al., 2002); in these mice Cre recombinase was knocked into one allele of the Olig1 gene locus. PDGFαR-CreER BAC transgenic mice have been described previously (Kang et al., 2010).
For inducible ablation of NMDAR within a cohort of OPCs and their progeny, NR1flox/flox mice were bred to PDGFαR-CreER as well as Z/EG reporter mice (Novak et al., 2000) which allows visualization of cells in which Cre was active. For electrophysiology experiments, PDGFαR-CreER;NR1+/+;Z/EG (P-NR1+/+) and PDGFαR-CreER;NR1flox/flox;Z/EG (P-NR1fl/fl) mice were given a single 1mg intraperitoneal (IP) injection of 4-hydroxy-Tamoxifen (4HT) during the 3rd or 4th postnatal week and were analyzed 10–35 days post injection as indicated in the text. For analysis of cell morphology and differentiation capacity, P-NR1+/+ and P-NR1fl/fl mice were given a single 0.1mg subcutaneous injection of 4HT at P4 and were analyzed at P30. Some mice in these experiments were also given IP injections of BrdU (10mg/kg) twice daily for two days at P20.
For constitutive ablation of NMDAR within the oligodendrocyte lineage, Olig1cre/cre;NR1flox/+ mice were bred to Olig1+/+;NR1flox/+ mice to obtain both Olig1cre/+; NR1flox/flox (O-NR1fl/fl) and Olig1cre/+; NR1+/+ (O-NR1+/+) littermate offspring. To assess the extent and specificity of Cre activity, O-NR1+/+ and O-NR1fl/fl mice were bred to Z/EG reporter mice. For electrophysiology experiments, O-NR1+/+ and O-NR1fl/fl mice were bred to NG2-DsRed BAC transgenic mice (Ziskin et al., 2007; Zhu et al., 2008) to accurately identify OPCs in acute brain slices. For analysis of cell proliferation, O-NR1+/+ and O-NR1fl/fl mice aged P7 and P14 were given a single IP injection of Bromodeoxyuridine (BrdU, 300mg/kg) two hours before perfusion.
In all experiments, both male and female mice were used and the number of males and females in each analysis group was balanced. All experiments were carried out in strict accordance with protocols approved by the Animal Care and Use Committee at Johns Hopkins University.
Mice were anesthetized with isofluorane and decapitated; their brains were dissected into an ice-cold N-methyl-D-glucamine (NMDG)-based solution containing: 135 mM NMDG, 1 mM KCl, 1.2 mM KH2PO4, 20 mM choline bicarbonate, 10 mM glucose, 1.5 mM MgCl2, and 0.5 mM CaCl2 (pH 7.4, 310 mOsm). Coronal forebrain slices (250 μm thick) were prepared using a vibratome equipped with sapphire blade in ice cold NMDG-based cutting solution. Following sectioning, slices were transferred to artificial cerebral spinal fluid (ACSF) containing: 119 mM NaCl, 2.5 mM KCl, 2.5 mM CaCl2, 1.3 mM MgCl2, 1 mM NaH2PO4, 26.2 mM NaHCO3, and 11 mM glucose (290 mOsm) and were maintained at 37°C for 30 min, and at room temperature thereafter. Both NMDG solution and ACSF were bubbled continuously with 95% O2/5% CO2. All experiments were carried out at room temperature.
Oligodendrocyte precursor cells (OPCs), pre-oligodendrocytes (Pre-OLs), and oligodendrocytes (OLs) were visualized with an upright microscope (Zeiss Axioskop FS2) equipped with both DIC optics and filter sets for DsRed (HQ:TRITC 41002c, Chroma) and GFP (Brightline, GFP-ABasic-ZHE, Semrock). Whole-cell recordings from OPCs, Pre-OLs, and OLs were made under visual control using both GFP and DsRed fluorescence and IR-DIC as a guide. For hypertonic solution and UV-uncaging experiments, the electrode solution consisted of 100 mM CsCH3SO3H (cesium methanesulfonate), 20 mM tetraethylammonium (TEA) chloride, 20 mM HEPES, 1 mM MgCl2, 10 mM EGTA, 2 mM sodium ATP and 0.2 mM sodium GTP (pH 7.3, 295 mOsm). For experiments performed in current-clamp mode, CsMeS was replaced with KCH3SO3H (potassium methanesulfonate) in the electrode solution and TEA was omitted. Pipette resistance was 3.4 to 4.2 MΩ and recordings were made without series resistance compensation. Hypertonic solution (HS; normal ACSF containing 500 mM sucrose; 850–900 mOsm) was focally delivered through a glass pipette (resistance 1–1.5 MΩ) using a pressure application system (Toohey Company, Pressure System IIc) as described previously (De Biase et al., 2010). For glutamate uncaging experiments, MNI-L-glutamate (500 μM, Tocris) was locally perfused over the slice with a wide-bore glass pipette. Photolysis was accomplished with a 1 ms flash of UV light from a Stabilite 2017 argon laser (Spectra Physics) in a 100 μm circle centered on the cell body. D-serine (30 μM, Sigma Aldrich), a co-agonist at the NMDA receptor glycine binding site, was included in the ACSF during experiments aimed at detecting NMDAR-mediated currents. For AMPAR rectification experiments, spermine (Tocris, 100μM) was included in the recording pipette. The following agents were applied by addition to the superfusing ACSF: TTX (NaV antagonist; Ascent Scientific, 1 μM); gabazine (SR-95531; GABA receptor antagonist; Tocris, 5 μM); RS-CPP (NMDA receptor antagonist; Tocris, 5 μM or 20 μM for uncaging); 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo(f)quinoxaline (NBQX; competitive AMPA/kainate receptor antagonist; Tocris, 5 μM or 50 μM for uncaging); GYKI 53655 (AMPA selective, non-competitive antagonist; IVAX 100 μM); 7-Chlorokynurenic acid (NMDAR antagonist acting at the glycine site; Tocris, 50μM).
Responses were recorded using a MultiClamp 700A amplifier (Axon Instruments), filtered at 3 kHz, digitized at 50 kHz and recorded to disk using pClamp9.2 software (Axon Instruments). Data were analyzed offline using Clampfit (Axon Instruments), Origin (OriginLab Corp.) and Mini analysis (Synaptosoft Inc.) software. Membrane resistance and membrane capacitance were calculated from a 10 mV depolarizing step (holding potential −80 mV). The amplitude of voltage-gated sodium current (NaV) was calculated from a 70 mV depolarizing step (holding potential −80 mV) performed before and after application of TTX and within 2 minutes of initiating whole cell recording. Resting membrane potential was measured within 30 seconds of establishing whole cell recording (KMs-based internal solution). The response to HS was quantified by counting the number of evoked mEPSCs (> 5 pA amplitude, <1 ms rise as measured in MiniAnalysis software, within 15s of stimulus onset) during 2 successive puffs (separated by at least 1.5 min). For clarity, slow alterations in holding current observed during HS application were subtracted from baseline using PClamp software. For experiments in P-NR1+/+ and P-NR1fl/fl mice in which EGFP+ cells may be OPCs, Pre-OLs, or OLs, basic membrane properties were used to categorize the stage of cell differentiation as described previously (De Biase et al., 2010). Specifically, cells with NaV current and <35 pF capacitance were considered to be OPCs, cells +/−NaV current with >35 pF capacitance and >500 MΩ membrane resistance were considered Pre-OLs, and cells that lacked NaV currents with < 500 MΩ membrane resistance were considered OLs.
Mice were deeply anesthetized with sodium pentobarbital (100 mg/kg) and perfused transcardially with 4% paraformaldehyde in 0.1 M sodium phosphate buffer. Brain tissue was isolated and post-fixed in this solution for 6–18 h at 4°C, then washed in phosphate buffer, cryoprotected in 30% sucrose, and sectioned at 30–60 μm thickness on a cryostat (Microm). Free-floating sections were permeabilized/blocked with 0.3% Triton X-100 and 5% normal donkey serum in 0.1 M sodium phosphate buffer for 2 hrs at room temperature or 4°C overnight. Sections were incubated with primary antibodies prepared in permeabilizing/blocking solution for 4 hrs at room temperature or 4°C overnight. Sections were incubated with secondary antibodies in 5% normal donkey serum in 0.1 M sodium phosphate buffer for 2 hrs at room temperature. For immunolabeling with CC1 antibody, sections were treated with LAB solution (Polysciences) for 10–15 min prior to blocking. For visualization of cells with incorporated BrdU, sections were incubated with 2N HCl at 37°C for 30min, washed twice with 0.1M borate buffer for 5min, and were then permeabilized/immunostained as outlined above. Control sections incubated with secondary antibody alone did not result in labeling of cells. Primary antibodies used: Rabbit anti-NG2 (1:500; gift from Dr. Stallcup, Burnham Institute), Rabbit anti-GFP (1:500, gift from Dr. Huganir, Johns Hopkins), Chicken anti-GFP (1:500, Aves), Goat anti-GFP (1:500, Frontier Institute), Rabbit anti-PDGFαR (1:500; gift from Dr. Stallcup, Burnham Institute), Rat anti- PDGFαR (1:500, BD Pharmingen), Guinea Pig anti-GFAP (1:500, Advanced ImmunoChemical Inc.), Rabbit anti-GFAP (1:500, Sigma), Mouse anti-MBP (1:1000, Sternberger Monoclonals), ALDH1 (1:100, UC Davis/NeuroMab), Mouse anti-NeuN (1:500, Millipore), CC1 (1:50, Anti-APC/Ab-7, Calbiochem), Guinea pig anti-Olig2 (1:10,000, gift from Dr. Ben Novitch, UCLA), Rabbit anti-Olig2 (1:500, Millipore). Secondary antibodies (raised in donkey): Alexa 488- (Invitrogen), Cy2-, Cy3- or Cy5-conjugated secondary antibodies to rabbit, mouse, goat or guinea pig (1:500; Jackson ImmunoResearch).
For evaluation of MBP immunostaining, quantification of oligodendrocyte lineage cell density, and cell proliferation analysis, fluorescence images were collected on a AxioImager M1 microscope (Zeiss). For analysis of cell differentiation capacity in P-NR1+/+/P-NR1fl/fl mice, and extent of Cre expression in O-NR1+/+/O-NR1fl/fl mice, fluorescence images were acquired with a LSM 510 Meta confocal microscope (Zeiss) or a LSM 710 Meta confocal microscope (Zeiss). For morphological analysis of OPCs in P-NR1+/+/P-NR1fl/fl mice, stacks of confocal images (0.3μm z-interval) of EGFP+PDGFαR+ cortical OPCs were imported into Imaris software for 3D morphological reconstruction and quantification. For analysis of Cre expression in O-NR1+/+/O-NR1fl/fl mice (bred to Z/EG reporter), 6 O-NR1+/+ mice (P30-55) and 1 O-NR1fl/fl (P55) mouse were examined. To evaluate colocalization between BrdU and PDGFαR or Olig2, 3 O-NR1fl/fl and 3 O-NR1+/+ mice were examined at P7 and 6 O-NR1fl/fl and 5 O-NR1+/+ mice were examined at P14. In the differentiation capacity analysis, to evaluate colocalization between EGFP and PDGFαR or CC1, 6 P-NR1fl/fl and 4 P-NR1+/+ mice were examined. In all histological analysis, at least 4 brain sections from each mouse were analyzed.
O-NR1fl/fl and O-NR1+/+ mice age P24-26 and P55-58 were deeply anesthetized with isofluorane and perfused transcardially with ice cold PBS. Mouse cortical tissue was carefully microdissected and flash frozen in 2-methyl butane chilled with dry ice. Tissue samples were then homogenized in RIPA buffer (Boston BioProducts) using a hand-held homogenizer. Protein concentration was determined using Bio-Rad protein assay (Bio-Rad). For MBP and CNPase immunoblots, protein samples were denatured in Laemmli’s SDS sample buffer containing β-mercaptoethanol. For detection of MAG, protein samples were prepared in a non-denaturing Laemmli’s sample buffer (Boston BioProducts). 20 μg of each sample was separated on a 12% Tris HCl gel (Bio-Rad) and transferred to a PVDF membrane. Membranes were blocked in 0.1% Tween-20 TBS containing 5% non-fat milk for 1 hour prior to incubation with primary antibodies. Primary antibodies were diluted in the same solution as follows: anti-CNPase (1:200, clone 11-5B, Millipore), anti-MBP (1:1000, clone smi99, Covance), anti-MAG (1:300, clone 513, Millipore), and anti-β-actin (1:1000, clone 20–33, Sigma). Anti-mouse or anti-rabbit secondary antibodies conjugated to IRDye® 680 or 800 were incubated for 1 hour prior to scanning the membrane on a LI-COR Odyssey® imager. Images were analyzed using the Odyssey® infrared imaging system application software version 3.0. Protein expression was normalized to β-actin.
O-NR1fl/fl and O-NR1+/+ mice age P24-26 and P55-57 were anesthetized with pentobarbital and perfused transcardially with 4% paraformaldehyde/2.5% glutaraldehyde in 0.1 M phosphate buffer. Brain tissue was isolated and post-fixed in this solution for 4 hrs at 4°C, then washed in phosphate buffer. Brains were treated with 2% OsO4 for 1 hour. After washes in water, samples were incubated in 2% uranyl acetate for 30 min. After dehydration using 50%, 70%, 90%, and 100% ethanol and 100% propylene oxide, samples were embedded in Epon 812 resin (Ted Pella). Ultrathin sections were obtained using a Leica Ultracut UCT and stained with 2% uranyl acetate and lead citrate. Electron micrographs were taken with an H-7100 electron microscope (Hitachi). NIH Image J software was used to measure the diameter of axons and their myelin sheaths.
Data are expressed as mean ± SEM throughout, and statistical significance was determined using the Mann-Whitney test with a cutoff value of 0.05 and Kruskal-Wallis ANOVA with sequential Bonferroni correction for multiple comparisons.
NMDARs are formed through the heteromeric assembly of NR1 and NR2 or NR3 subunits. The NR1 subunit contains the binding site for glycine/D-serine and without this subunit functional NMDARs cannot be formed (Forrest et al., 1994). To determine if NMDAR expression within the OL lineage is required for OPC survival and OL generation in vivo, we bred mice in which exons 19–20 of the NR1 gene (Grin1) are flanked by loxP sites (NR1flox/flox mice) (Fukushima et al., 2009) to mice expressing inducible Cre recombinase (CreER™) under control of the PDGFαR promoter (PDGFαR-CreER mice) (Kang et al., 2010) as well as the Z/EG reporter (Novak et al., 2000) (Fig. 1A). In triple transgenic mice, injection of 4-hydroxy-tamoxifen (4HT) should activate Cre within a cohort of OPCs and induce EGFP expression and NR1 excision. Indeed, when control (PDGFαR-CreER;NR1+/+;Z/EG) or inducible knockout (PDGFαR-CreER;NR1flox/flox;Z/EG) mice were injected with 4HT (0.1mg) at postnatal day (P) 4 and examined at P30, EGFP+ cells were found throughout the forebrain (Fig. 1B); some of these cells remained PDGFαR+ OPCs (Fig. 1D), but many had differentiated into mature CC1+ oligodendrocytes (see Fig. 3A), consistent with previous results (Kang et al., 2010).
To confirm that Cre activity was sufficient to eliminate functional NMDARs within these cells, fluorescence-guided whole-cell recordings were made from EGFP+ OPCs in the corpus callosum of mice injected with 4HT (1mg) during the 3rd or 4th postnatal week and examined 10–15 days later. Local photolysis of caged glutamate (MNI-L-glutamate) elicited transient inward currents in all OPCs from both control (PDGFαR-CreER;NR1+/+;Z/EG, n = 15 cells) or inducible knockout mice (PDGFαR-CreER;NR1flox/flox;Z/EG, n = 14 cells) indicating that glutamate receptors were expressed. Although outward currents sensitive to the NMDAR antagonist CPP (20 μM) were observed in the majority of OPCs in control mice (PDGFαR-CreER;NR1+/+;Z/EG, 10/15 cells, n = 3 mice), CPP-sensitive currents were not observed in OPCs from inducible knockout mice (PDGFαR-CreER;NR1flox/flox;Z/EG, 0/14 cells, n = 3 mice) (Fig. 1C). These results indicate that this genetic strategy is effective at preventing the formation of functional NMDARs in OPCs. Control (PDGFαR-CreER;NR1+/+;Z/EG) and inducible knockout (PDGFαR-CreER;NR1flox/flox;Z/EG) mice will hereafter be referred to as “P-NR1+/+” and “P-NR1fl/fl”, respectively.
NMDAR activation has been shown to influence neurite outgrowth and spine morphogenesis in developing neurons (Ponimaskin et al., 2007). OPCs elaborate numerous, highly branched processes and make contact with surrounding axons (Butt et al., 1999; Dawson et al., 2003; Bergles et al., 2010), but the factors governing OPC process extension are unknown. To determine whether NMDAR signaling influences OPC morphology, P-NR1+/+ and P-NR1fl/fl mice were injected with 4HT (0.1mg) at P4 and examined at P30. Morphological reconstructions of isolated EGFP+ PDGFαR+ OPCs in the cortex (Fig. 1D,E) revealed no significant differences in total process length (P-NR1+/+,2233 ± 204 μm, n = 3 cells; P-NR1fl/fl, 2328 ± 285 μm, n = 3 cells; p = 1.0) or number of branch points (P-NR1+/+,334 ± 13, n = 3 cells; P-NR1fl/fl, 308 ± 24, n = 3 cells; p = 0.66) between OPCs in P-NR1+/+ and P-NR1fl/fl mice, suggesting that NMDAR activation does not play a prominent role in determining the extent of outgrowth or branching of OPC processes.
To determine whether NMDAR deletion resulted in changes in the membrane properties of OPCs, we recorded from OPCs in the corpus callosum of P-NR1+/+ and P-NR1fl/fl mice aged P40-45 (>19 days after 4HT administration). Callosal OPCs typically exhibit a small membrane capacitance (Cm, ~12–35pF), a negative resting membrane potential (Vm, ~ −90mV), and a moderate membrane resistance (Rm, ~100–1000MΩ) (Ziskin et al., 2007; De Biase et al., 2010). None of these membrane properties were significantly altered in NMDAR-deficient OPCs (P-NR1+/+ Cm, 21 ± 2pF, P-NR1fl/fl Cm, 23 ± 1pF, p = 0.06, n = 19–24 cells; P-NR1+/+ Vm, −87 ± 1mV, P-NR1fl/fl Vm, −86 ± 0.5mV, p = 0.29, n = 5–10 cells; P-NR1+/+ Rm, 503 ± 417MΩ, P-NR1fl/fl Rm, 286 ± 132MΩ, p = 0.43, n = 5–10 cells). Callosal OPCs also express voltage-gated sodium (NaV) channels and display complex voltage responses to injection of depolarizing current (Ziskin et al., 2007; De Biase et al., 2010). All recorded OPCs possessed NaV currents (P-NR1+/+ 29/29 cells, P-NR1fl/fl 15/15 cells) and NMDAR-deficient OPCs exhibited comparable responses to depolarization (Fig. 1F); however, there was a small but significant reduction in NaV current density in OPCs from P-NR1fl/fl mice (P-NR1fl/fl: 10 ± 1 pA/pF, P-NR1+/+: 17 ± 1 pA/pF, p = 0.002, n = 15–29 cells). Together, these data indicate that NMDAR signaling does not play a prominent role in regulating the membrane properties of OPCs.
Maturation of OPCs into myelinating OLs is accompanied by dramatic alterations in their morphological and physiological properties (De Biase et al., 2010; Kukley et al., 2010), changes that are thought to enable successful maturation (Gallo et al., 1996; Knutson et al., 1997). To evaluate whether these progressive changes in membrane properties are maintained in the absence of NMDARs, P-NR1+/+ and P-NR1fl/fl mice were injected with 4HT during the 3rd postnatal week and whole cell patch clamp recordings were performed from EGFP+ cells in the corpus callosum 24–26 days post injection. In accordance with previous results (Sontheimer et al., 1989; Ziskin et al., 2007; De Biase et al., 2010), OPCs, premyelinating oligodendrocytes (Pre-OLs), and OLs in control mice exhibited stereotyped responses to depolarization: OPCs displayed non-linear responses due to activation of voltage-activated conductances and intermediate Rm, Pre-OLs exhibited larger membrane potential shifts due to their higher Rm, and OLs had a low Rm and displayed a prolonged “tail current” following cessation of current injection (Fig. 2A,B). Pre-OLs and OLs from P-NR1+/+ mice also showed increased Cm and positive shifts in Vm relative to OPCs (Fig. 2C,D), as previously described (De Biase et al., 2010; Kukley et al., 2010). Simliar physiological changes were observed in Pre-OLs and OLs from P-NR1fl/fl mice (Fig. 2A-D), suggesting that NMDAR activation does not play an essential role in triggering the physiological changes that these cells undergo during maturation.
During early postnatal development, NMDAR signaling is thought to promote the survival of immature neurons (Ikonomidou et al., 1999; Adams et al., 2004) and play a role in guiding the differentiation of neural progenitors (Nacher and McEwen, 2006). To determine whether NMDAR activation similarly impacts the survival of OPCs or their capacity to generate OLs, P-NR1+/+ and P-NR1fl/fl mice were injected with 4HT (0.1mg) at P4 and examined at P30 (Fig. 3A). In accordance with the prominent generation of OLs in white matter at this age, the density of EGFP+ cells was significantly higher in corpus callosum (CC) than cortex (CC vs. Ctx, p = 0.0003), and a greater proportion of EGFP+ cells were immunopositive for CC1, a marker of mature OLs (CC vs. Ctx p = 0.0002) (Fig. 3B–C). However, neither the density of EGFP+ cells (Ctx, p = 0.46; CC, p = 0.75) (Fig. 3B) nor the percentage of EGFP+CC1+ cells (Ctx, p = 0.24; CC, p = 1.0) (Fig. 3C) varied between P-NR1+/+ (n = 4) and P-NR1fl/fl (n = 6) mice. Furthermore, when mice were injected with BrdU prior to analysis, the number of EGFP+BrdU+ cells in cortex and corpus callosum did not differ between P-NR1+/+ and P-NR1fl/fl mice (Ctx, p = 0.22; CC, p = 0.6). These data indicate that NMDAR signaling is not essential for OPC survival or differentiation. Furthermore, because NR1 excision does not occur in all OPCs using this inducible approach, NMDAR-deficient and NMDAR-containing OPCs coexist within the same tissue; therefore, these data also indicate that NMDAR-deficient OPCs are not at a competitive disadvantage, in terms of survival and maturation, compared to neighboring OPCs that engage in NMDAR signaling.
NMDAR-deficient OPCs in P-NR1fl/fl mice did not exhibit overt deficits; however, these experiments do not address whether NMDAR signaling in embryonic or early postnatal ages is involved in OPC development. Moreover, while comparable numbers of CC1+ OLs were generated from NMDAR-deficient OPCs in these mice, the capacity of these OLs to generate myelin is difficult to assess quantitatively using this approach. Therefore, we genetically deleted NR1 from the entire OL lineage by breeding NR1flox/flox mice to mice in which Cre recombinase was knocked into one allele of the Olig1 gene locus (Lu et al., 2002) (Olig1cre/+ mice) (Fig. 4A, inset). The transcription factor Olig1 is expressed embryonically by progenitors that give rise to the OL lineage (Ligon et al., 2006); thus, in Olig1cre/+; NR1flox/flox mice, Cre-mediated removal of NR1 should occur shortly after specification of OL lineage cells, preventing expression of functional NMDARs in all OPCs and OLs throughout life. Olig1cre/+; NR1flox/flox mice were bred to Z/EG reporter mice to determine the extent and specificity of Cre-mediated recombination (Fig. 4A). In both control (Olig1cre/+;NR1+/+;Z/EG) and constitutive NR1 knockout (Olig1cre/+; NR1flox/flox;Z/EG) mice at P30, EGFP was expressed by PDGFαR+, NG2+, and Olig2+ cells throughout the cortex and corpus callosum (Fig. 4B–D), confirming that Cre-mediated recombination occurred in the majority of OL lineage cells (Lu et al., 2002). In addition, EGFP was also expressed by many cortical astrocytes, as well as rare neurons (data not shown), as previously described (Samanta et al., 2007). Control (Olig1cre/+;NR1+/+) and NR1 knockout (Olig1cre/+; NR1flox/flox) mice in this constitutive NMDAR ablation approach are hereafter referred to as “O-NR1+/+” and “O-NR1fl/fl” mice, respectively.
To determine if NMDAR expression was prevented within the OL lineage, responses to glutamate were examined in callosal OPCs from O-NR1+/+ and O-NR1fl/fl mice. Both lines of mice were crossed to NG2-DsRed mice (Ziskin et al., 2007) to allow accurate, fluorescence-guided whole cell recording from OPCs in acute brain slices. Local photolysis of caged glutamate (MNI-L-glutamate) elicited transient inward currents in all OPCs from both O-NR1+/+ (n = 9 cells) and O-NR1fl/fl mice (n = 10 cells), indicating that glutamate receptors were expressed. Although outward currents sensitive to CPP (20 μM) were observed in the majority of OPCs in O-NR1+/+ mice (7/9 cells, n = 2 mice) (Fig. 4E), CPP-sensitive currents were not observed in OPCs from O-NR1fl/fl mice (0/10 cells, n = 3 mice), indicating that this genetic strategy is effective at preventing the formation of functional NMDARs in OL progenitors.
Despite the absence of NMDARs in OL lineage cells, O-NR1fl/fl mice were viable, survived into adulthood, and did not exhibit behaviors associated with hypomyelination, such as tremors, ataxia, or seizures (Filley, 2001). Moreover, these mice did not display gross anatomical abnormalities in either the forebrain or cerebellum. To determine whether NMDAR removal affected the viability of OL lineage cells, the distribution of PDGFαR+ OPCs and Olig2+ OL lineage cells was examined during early postnatal periods (P7-8, P13-14). In both the cortex and corpus callosum of O-NR1fl/fl mice, the density of OL lineage cells was unaltered (Fig. 5), suggesting that both the survival and migration of OPCs remains intact. However, OPCs are highly proliferative during development (Psachoulia et al., 2009) and can compensate for profound reductions in cell number to establish normal cell density and patterns of myelination (Kessaris et al., 2006). To determine whether a deficit in OPC or OL survival was masked by enhanced proliferation of these progenitors, BrdU incorporation by OL lineage cells was examined in early postnatal O-NR1+/+ and O-NR1fl/fl mice. OPC proliferation did not differ significantly different between the two genotypes (P7-8: Ctx, p = 0.66; CC, p = 1.0; P13-14: Ctx, p = 0.11; CC, p = 0.86) (Fig. 6), indicating that NMDAR activation is not critical for OPC survival.
To determine whether embryonic NMDAR deletion altered the membrane properties of OPCs, fluorescence-guided whole-cell recordings were made from DsRed+ OPCs in the corpus callosum, cortex, and hippocampus of O-NR1+/+ and O-NR1fl/fl mice (P40-45). Consistent with observations in P-NR1fl/fl mice, callosal OPCs from O-NR1fl/fl mice exhibited normal membrane capacitance (Cm), resting potential (Vm), and membrane resistance (Rm) (O-NR1+/+ Cm, 21 ± 1pF, O-NR1fl/fl Cm, 21 ± 1pF, p = 0.89, n = 32–44 cells; O-NR1+/+ Vm, −88 ± 1mV, O-NR1fl/fl Vm, −87 ± 1mV, p = 0.39, n = 6–11 cells; O-NR1+/+ Rm, 91 ± 11MΩ, O-NR1fl/fl Rm, 239 ± 93MΩ, p = 0.13, n = 6–11 cells). Furthermore, callosal OPCs from O-NR1fl/fl and O-NR1+/+ mice displayed similar responses to depolarization (data not shown) and all recorded OPCs possessed NaV currents (NR1+/+ 23/23 cells, O-NR1fl/fl 32/32 cells). There was a trend toward reduced NaV density in OPCs from O-NR1fl/fl, but unlike P- NR1fl/fl mice, this difference was not statistically significant (O-NR1fl/fl: 14 ± 1 pA/pF; O-NR1+/+: 16 ± 2 pA/pF, p = 0.36). Recordings from OPCs in the cortex and hippocampus in O-NR1fl/fl mice revealed that basic membrane properties were similarly preserved in gray matter OPCs that lacked NMDARs (data not shown). Together, these data confirm that NMDAR signaling does not play a prominent role in regulating the membrane properties of OPCs.
To determine if activation of OL lineage NMDARs plays a role in CNS myelination, we examined the progressive developmental changes in white matter tracts through myelin basic protein (MBP) immunolabeling. The onset of MBP expression in the forebrain was observed at P7-8 in both O-NR1+/+ and O-NR1fl/fl mice (Fig. 7A), and formation of major myelin tracts was indistinguishable between the two genotypes at P14 (Fig. 7B). Moreover, the extent of myelination in adults (P55-57) was not qualitatively different between the two genotypes (Fig. 7C). To obtain a quantitative measure of OL maturation, the abundance of major myelin proteins in cortex and subcortical white matter was compared by western blot (Fig. 8A). Although the levels of MBP, 2′, 3′-cyclic nucleotide 3′-phosphodiesterase (CNP), and myelin associated glycoprotein (MAG) increased with developmental age (P24-26 vs. P55-57: MBP, p = 0.005; CNP, p = 0.005; MAG, p = 0.008) (Fig. 8B), the abundance of these proteins did not differ significantly between O-NR1+/+ and O-NR1fl/fl mice (P24-26: MBP, p = 0.66; CNP, p = 1.0; MAG, p = 0.38; P55-58: MBP, p = 1.0; CNP, p = 0.66; MAG, p = 1.0), indicating that myelin protein expression occurs normally without NMDAR activation in OL lineage cells.
Rapid conduction of action potentials along large diameter axons depends on successful formation of compact myelin. Although myelin tracts appeared in appropriate regions and myelin protein expression was unaltered in O-NR1fl/fl mice, deficits in myelin compaction and myelin thickness at the level of individual nerve fibers could still have important functional consequences. To address whether myelin sheaths had normal structure when NMDAR signaling was disrupted, thin section electron micrographs were prepared from O-NR1+/+ and O-NR1fl/fl mice. Examination of the developing corpus callosum (P24-26) revealed unmyelinated axons, partially myelinated axons, and a subset of axons with multiple layers of compact myelin (Fig. 8C). In adult mice (P55-57), the number of callosal axons ensheathed by compact myelin increased considerably; however, the density of myelinated fibers was similar in both genotypes (p = 0.59). Myelin thickness is closely matched to the caliber of the underlying axon, with larger axons possessing thicker myelin sheaths (Sherman and Brophy, 2005). This relationship can be appreciated in plots of g-ratio (diameter of axon/diameter of axon+myelin) versus axon caliber (Fig. 8D), and the characteristics of myelinated axons can be assessed by examining the slope and variability (R2) of linear fits to population data. The variability of myelin thickness was greater early in development (P24-26 R2 vs. P55-57 R2, p = 0.04) (Fig. 8E), as expected due to ongoing myelination. However, there were no differences in variability or slope between O-NR1+/+ and O-NR1fl/fl mice at either age (P24-26: slope p = 0.67, R2 p = 0.67; P55-58: slope p = 0.86, R2 p = 1.0) (Fig. 8E,F), indicating that compact myelin was formed normally despite the prevention of NMDAR signaling within the OL lineage. Together these results indicate that activation of OL lineage NMDARs is not critical for successful CNS myelination.
Neither OPC development nor OL myelination were substantially altered by inducible or constitutive ablation of NR1 from OL lineage cells, suggesting that NMDARs serve other functions within these glial cells. In neurons, NMDAR activation plays a critical role in modulating density of AMPA receptors (AMPARs) found at synapses in both the developing and mature CNS (Malenka and Bear, 2004). Indeed, genetic ablation of NMDARs from subpopulations of neurons led to deficits in several forms of synaptic plasticity (Tsien et al., 1996; Engblom et al., 2008) and alterations in AMPAR-mediated synaptic responses (Adesnik et al., 2008; Engblom et al., 2008). OPCs in all brain regions form synapses with glutamatergic neurons, but little is known about the mechanisms regulating this form of neuron-glial communication.
To evaluate whether NMDAR activation plays a role in regulating neuron-OPC synaptic function, we measured the response of OPCs to focal application of hypertonic solution (HS, 500mM sucrose in ACSF), a manipulation that forces release of docked, primed synaptic vesicles at nerve terminals, and provides a measure of synaptic maturation and connectivity (De Biase et al., 2010). Responses to HS were investigated in callosal OPCs from mature (P40-45) mice, when synaptic connectivity has peaked and variability due to cell differentiation should be minimal (De Biase et al., 2010). In the presence of TTX (1μM), CPP (20μM), and gabazine (GABAR antagonist, 100μM), HS elicited miniature excitatory postsynaptic currents (mEPSCs) in all OPCs from both lines of control mice (Fig. 9A,B), consistent with previous results (O-NR1+/+, n = 11 cells; P-NR1+/+, n = 8 cells). HS similarly evoked mEPSCs in all OPCs from O-NR1fl/fl (n = 16 cells) and P-NR1fl/fl mice (n = 11 cells) (Fig. 9A,B), indicating that white matter neuron-OPC synapses are still formed in the absence of NMDARs. Synaptic connectivity was also preserved between neurons and OPCs in the CA1 region of hippocampus; 3/3 OPCs in O-NR1+/+ mice and 5/5 OPCs O-NR1fl/fl showed HS-elicited mEPSCs (data not shown). The degree of neuron-OPC synaptic connectivity was estimated by quantifying the number of mEPSCs observed during hypertonic challenge. Although there was a trend toward reduced synaptic connectivity in callosal OPCs lacking NMDARs (# of evoked mEPSCs: O-NR1fl/fl, 259 ± 68; P-NR1fl/fl, 249±43; O-NR1+/+, 330 ± 34;P-NR1+/+, 324 ± 53) (Fig. 9C), this difference was not statistically significant (p = 0.22, K-W ANOVA). These findings indicate that NMDAR expression by OPCs is not required to establish and maintain glutamatergic synapses with neurons. Furthermore, the amplitude of HS-evoked mEPSCs did not differ between OPCs from O-NR1+/+ and O-NR1fl/fl mice, indicating that postsynaptic AMPAR density was also preserved in the absence of OPC NMDARs (Fig. 9D).
Though synaptic responses are intact in NMDAR-deficient OPCs, these approaches do not reveal whether there are changes in the overall abundance of AMPARs. To determine whether NMDARs influence global surface AMPAR expression in OPCs, glutamate uncaging was used to examine AMPAR-mediated currents in OPCs in the corpus callosum of mature mice (P40-50). In the presence of tetrodotoxin (TTX, 1μM) and antagonists for NMDARs (CPP, 20μM; 7-chlorokynurenic acid, 100μM), glutamate uncaging elicited inward currents in all recorded OPCs from both O-NR1+/+ and O-NR1fl/fl mice. These inward currents were blocked by the AMPAR antagonists NBQX (20μM) and SR53655 (100μM, n = 19 cells) (Fig. 9E, inset). Although there was a trend toward larger AMPAR-mediated responses in OPCs from O-NR1fl/fl mice, neither the peak amplitude of AMPAR-mediated currents (O-NR1+/+, 310 ± 50pA, n = 17; O-NR1fl/fl, 406 ± 81pA, n = 14; p = 0.46), nor the density of surface AMPARs (O-NR1+/+, 11 ± 2pA/pF, n = 17; O-NR1fl/fl, 15 ± 2pA/pF, n = 14; p = 0.17) was significantly different from OPCs in O-NR1+/+ mice. Together these data suggest that surface AMPARs were expressed at normal levels in the absence of NMDARs.
AMPARs are formed from different combinations of the subunits GluR1-4 and those receptors which lack the edited GluR2 subunit are permeable to Ca2+ (Liu and Zukin, 2007). To determine whether AMPAR subunit composition in OPCs is influenced by NMDARs, we measured the current-voltage relationship of AMPAR-mediated currents in the presence of intracellular spermine (100 μM), a polyamine that reduces outward current flow through Ca2+ permeable AMPARs (cpAMPARs). Although the degree of inward rectification varied from cell to cell, OPCs from O-NR1fl/fl mice exhibited a greater rectification index (RI, peak −80mV/peak 40mV) than OPCs from O-NR1+/+ mice (O-NR1fl/fl, RI = 2.51 ± 0.1, n = 14; O-NR1+/+, RI = 1.98 ± 0.1, n = 17; p = 0.001) (Fig. 9E), indicating that there was an approximately 27% increase in surface cpAMPARs in NMDAR-deficient OPCs. These results suggest that NMDAR signaling provides negative feedback to control the density of cpAMPARs in this population of glial cells as has been suggested for neurons (Ho et al., 2007; Clem and Huganir, 2010; Wang and Gao, 2010).
Glial cells throughout the mammalian CNS express neurotransmitter receptors that are thought to regulate glial development, control homeostatic behaviors such as functional hyperemia, and enable neuromodulation through the release of gliotransmitters (Belachew and Gallo, 2004; Perea et al., 2009; Attwell et al., 2010). However, it has been difficult to establish the roles of these receptors in vivo using pharmacological approaches, as glia and neurons often express the same receptors. Here, NMDAR signaling was prevented in OL lineage cells by genetically deleting the obligatory NMDAR subunit NR1 from OPCs. OPCs lacking NMDARs exhibited normal rates of proliferation, achieved normal densities in gray and white matter, and matured into OLs to form the major white matter tracts of the CNS without appreciable delay. Moreover, the characteristic physiological and morphological properties of OPCs were preserved in the absence of NMDARs, and they continued to receive glutamatergic synaptic input. However, NMDAR deletion led to alterations in the subunit composition of AMPARs in OPCs. These results suggest that NMDAR signaling is used to modulate signaling with surrounding axons rather than control OPC development, raising the possibility that these ubiquitous glial cells perform additional functions in neural circuits.
Global inhibition of NMDARs through administration of MK-801 during early postnatal development induces widespread cell death in the cortex and hippocampus (Ikonomidou et al., 1999), and reduces the survival of migrating neuroblasts within neurogenic niches in the mature CNS (Platel et al., 2010), suggesting that the survival of immature neurons is dependent on NMDAR activation. Although NR1 knockout mice exhibit macroscopically normal CNS architecture at birth (Forrest et al., 1994), pronounced cell death is observed in the ventrobasal thalamus (Adams et al., 2004), where neurons are actively forming synapses and undergoing programmed cell death. OPCs are distributed throughout the brain at birth (Nishiyama, 2007) and formation of neuron-OPC synapses parallels neuronal synaptogenesis in the forebrain (De Biase et al., 2010). Although programmed cell death has not been reported for OPCs, apoptotic premyelinating oligodendrocytes (Pre-OLs) can be observed in the cortex between P7-P21 (Trapp et al., 1997), suggesting that NMDAR activation would most likely influence OPC survival during the second and third postnatal weeks. However, in P7 and P14 O-NR1fl/fl mice, the density of OL lineage cells in cortex and corpus callosum was normal (Fig. 5B,C), and no increases in OPC proliferation were observed (Fig. 6), which would be expected if OPC turnover was enhanced. When NMDARs were removed from OPCs shortly after birth (P4), the number of NMDAR-deficient (EGFP+) cells that survived until P30 also did not differ from control (Fig. 3B). Moreover, O-NR1fl/fl mice did not exhibit developmental delays in CNS myelination (Figs. 7–8), which would be expected if there were fewer progenitors from which to form OLs. It is possible that NMDAR ablation could affect the long term viability of OPCs or mature OLs in aged mice; however, 10 month-old O-NR1fl/fl mice did not exhibit tremors or motor abnormalities indicative of myelination deficits, suggesting that NMDAR activation is not required for OL survival in the ageing CNS.
Do these findings indicate that there are major differences in the survival requirements of immature neurons and glia? In chimeric mice containing NR1 null and wild type cells, NR1−/− cells often account for 50–90% of the neurons and glia in the CNS (Maskos and McKay, 2003), indicating that neuroblasts lacking functional NMDAR are capable of widespread maturation and integration into CNS tissue. Furthermore, reduced cell viability was not reported when NMDARs were ablated from neurons during embryonic or early postnatal periods (Iwasato et al., 2000; Engblom et al., 2008; Fukushima et al., 2009). Together, these data suggest that the dramatic apoptosis observed following pharmacological inhibition of NMDARs may be due to altered patterns of neural activity, rather than a direct effect of NMDAR blockade on cell survival (Homayoun and Moghaddam, 2007). Whether such abnormal neural activity also leads to enhanced apoptosis of OPCs has not been determined.
When NMDARs are removed from neurons during embryonic or early postnatal ages, excitatory synapses are still formed (Tsien et al., 1996; Adesnik et al., 2008; Engblom et al., 2008; Fukushima et al., 2009). However, in some cases, the amplitude of AMPAR currents and frequency of mEPSCs was enhanced (Adesnik et al., 2008; Engblom et al., 2008), indicating that NMDAR expression can influence both the number of synapses and the abundance of AMPARs at each synapse. Neuron-OPC synapses also formed in the absence of NMDARs, as spontaneous EPSCs (sEPSCs) and HS-evoked mEPSCs were detected in all recorded OPCs from both O-NR1fl/fl and P-NR1fl/fl mice. However, in NMDAR-deficient OPCs the frequency of sEPSCs (data not shown) and the number and amplitude of HS-evoked mEPSCs (Fig. 9C,D) were comparable to controls, suggesting that NMDAR activation in these glial cells is not closely linked to synapse formation or AMPAR density at these sites. The preservation of neuron-OPC synapses is unlikely to reflect developmental compensation or the absence of competition between neighboring OPCs for presynaptic input, because similar results were obtained in P-NR1fl/fl mice, in which NR1 was removed postnatally from only a subset of OPCs.
An unusual aspect of neuron-OPC synapses is the high prevalence of cpAMPARs, which support activity-dependent synaptic potentiation (Ge et al., 2006), but may also render these cells sensitive to excitotoxicity during ischemia (Back, 2006). The contribution of these receptors to synaptic currents in OPCs varies with development and among brain regions (Bergles et al., 2010), suggesting that their expression and incorporation at synapses may be modified by activity or maturation state. Glutamate uncaging revealed that a greater proportion of AMPARs in NMDAR-deficient OPCs were Ca2+ permeable (Fig. 9E). Although most mature pyramidal neurons do not express cpAMPARs, their expression can be elicited by plasticity-inducing protocols, such as whisker stimulation, fear conditioning, and cocaine administration (Clem and Barth, 2006; Engblom et al., 2008; Clem and Huganir, 2010) and subsequent NMDAR activation has been linked to internalization of these newly incorporated cpAMPAR (Clem and Huganir, 2010). Furthermore, expression of cpAMPARs in some interneurons and early postnatal pyramidal neurons appears to be correlated with low levels of NMDAR expression (Ho et al., 2007; Wang and Gao, 2010), suggesting that NMDARs exert a negative influence on cpAMPAR expression in neurons. The increased cpAMPAR expression by NMDAR-deficient OPCs may reflect the absence of this negative regulation pathway. Alternatively, increased cpAMPAR expression by these glial cells might represent an attempt to compensate for a reduction in Ca2+ influx through NMDARs. However, it is unclear whether cpAMPARs could substitute for NMDARs, as these receptors typically couple to distinct intracellular signaling and scaffolding proteins (Kim and Sheng, 2004).
The membrane properties of OPCs often vary between brain regions (Chittajallu et al., 2004; De Biase et al., 2010), suggesting that local environmental cues influence the particular complement and density of ion channels expressed by these progenitors. Characteristics such as membrane resistance and extent of synaptic communication can also vary over a wide range among OPCs within discrete brain regions (De Biase et al., 2010), and a subpopulation of OPCs that lack both NaV expression and synaptic input, and are resistant to hypoxic-ischemic injury in vitro, have been described (Karadottir et al., 2008). In the present study, we found that 22–33% of OPCs in the corpus callosum of young adult mice (P-NR1+/+ and O-NR1+/+) did not exhibit NMDAR-mediated currents (Fig. 1C, Fig. 4E), consistent with previous results (Ziskin et al., 2007). Together, these findings raise the possibility that OPCs consist of separate populations of cells adapted for distinct physiological functions. However, these data should be interpreted with caution, as they involved analysis of animals during early postnatal development when OPCs are continually differentiating into oligodendrocytes (Kang et al., 2010). Previous physiological studies (De Biase et al., 2010; Kukley et al., 2010) and transcriptional profiling datasets (Cahoy et al., 2008) indicate that OPCs rapidly alter their membrane properties and down regulate glutamate receptor expression when they begin to differentiate. If the sequence of gene expression changes associated with differentiation is subject to stochastic variation, cells defined as OPCs based on membrane properties and marker expression (e.g. NG2) may also include some cells in the initial stages of differentiation. In this scenario, OPCs lacking NMDARs should be encountered more frequently in regions of highest OPC differentiation. Consistent with this hypothesis, NMDAR-mediated currents were not observed in the majority of OPCs in the corpus callosum during the first two postnatal weeks, when extensive oligodendrogenesis occurs; whereas, at the same developmental stage in gray matter regions, where OPC differentiation is less prominent (Kang et al., 2010), nearly all OPCs exhibited NMDAR- mediated currents (L. De Biase and D.E. Bergles, unpublished observations). An alternative, but not mutually exclusive explanation, is that OPCs upregulate NMDARs during brain development in concert with the maturation of neural circuits. Additional methods to accurately establish the developmental state of OPCs will be required to determine which aspects of physiological diversity reflect heterogeneity among this population of progenitors, and which reflect changes arising from differentiation.
Although these studies indicate that NMDARs are not required for survival or maturation of OPCs, glutamate signaling to OPCs could regulate OL development by acting through other glutamate receptors. OPCs also express AMPARs that provide a route for Ca2+ influx, and AMPAR activation inhibits OPC differentiation in vitro (Gallo et al., 1996). However, evaluation of the role of these receptors using genetic manipulations is more challenging, as functional AMPARs can be formed through numerous subunit combinations and no one obligate subunit is required. It is also possible that signaling through other neurotransmitters, such as ATP, may influence OPC survival and maturation, as adenosine receptor activation enhances OPC differentiation and myelination in vitro (Fields and Burnstock, 2006).
If NMDARs do not play a role in OPC development or myelination, then why are they expressed by OL lineage cells? In addition to a potential role in regulating surface expression of cpAMPARs, NMDARs could allow OPCs to sense fluctuations in ambient glutamate levels and respond to extrasynaptic release or spillover from nearby neuronal synapses (Carter and Regehr, 2000; Matsui et al., 2005), as NMDARs have a particularly high affinity for glutamate. NMDAR signaling in OL lineage cells is profoundly enhanced during oxygen/glucose deprivation in vitro (Karadottir et al., 2005), raising the possibility that signaling through these receptors promotes the reactive changes exhibited by these cells following acute injury (Tan et al., 2005) and in neurodegenerative disease (Kang et al., 2010). As OPCs are found associated with astrocytes and microglia in glial scars (Tan et al., 2005), NMDARs may facilitate detection of local injury and enable these abundant cells to promote tissue repair in a variety of pathological conditions.
We thank D. Rowitch (University of California, San Francisco) for providing Olig1-Cre mice. W.B. Stallcup (Burnham Institute, San Diego) for antibodies to NG2, R. Huganir (Johns Hopkins University, Baltimore) for antibodies to GFP, B. Novitch (University of California, Los Angeles) for antibodies to Olig2, N. Ye for assistance with genotyping and mouse breeding, the IVAX Drug Research Institute for GYKI 53655, and A. Agarwal for advice regarding myelination assessments. This work was supported by the US National Institutes of Health grants NS051509, PAR-02-059, and MH084020 (D.E.B), The Maryland Stem Cell Research Fund Post Doctoral Fellowship Grant 2010-MSCRFF-0061-00 (E.G.P.), and The Kenneth and Claudia Silverman Family Foundation (P.A.C.).
Conflict of Interest: none