Satellite glia cells in dorsal root ganglia express functional NMDA receptors

Neuroscience 240 (2013) 135–146

SATELLITE GLIA CELLS IN DORSAL ROOT GANGLIA EXPRESS FUNCTIONAL NMDA RECEPTORS C. CASTILLO, a M. NORCINI, b L. A. MARTIN HERNANDEZ, b G. CORREA, a T. J. J. BLANCK b,c AND E. RECIO-PINTO b,d*

between neurons and SGCs. In summary we demonstrated for the first time that SGCs express functional NMDAr. Ó 2013 IBRO. Published by Elsevier Ltd. All rights reserved.

a

Unidad de Neurociencias, Instituto de Estudios Avanzados IDEA, Apartado 17606, Caracas 1015-A, Venezuela

Key words: N-methyl-D-aspartate, satellite glia, MK-801, AP-5, cytoplasmic calcium, sensory neurons.

b

Department of Anesthesiology, New York University Langone Medical Center, New York, NY 10016, USA

c Department of Physiology and Neuroscience, New York University Langone Medical Center, New York, NY 10016, USA d

INTRODUCTION

Department of Pharmacology, New York University Langone Medical Center, New York, NY 10016, USA

There is growing interest in identifying the receptors and molecules involved in the interactions between satellite glia cells (SGCs) and between SGCs and the neuronal somata of primary sensory neurons within the dorsal root ganglia (DRG); in part because those interactions are believed to contribute to the processing of afferent information (Hanani, 2005; Takeda et al., 2009; Gu et al., 2010). Functional N-methyl-D-aspartate receptors (NMDAr) have been detected in glial cells in the brain cortex (Luque and Richards, 1995; Conti et al., 1996; Schipke et al., 2001), hippocampus (Gottlieb and Matute, 1997; Krebs et al., 2003), and spinal cord (Ziak et al., 1998). The NMDAr activity on those glia appears to contribute to the interactions between glia and neurons during physiological and pathological conditions (Karadottir et al., 2005; Salter and Fern, 2005; Lalo et al., 2006; Micu et al., 2006). To our knowledge NMDAr have not been previously described in SGCs. SGCs closely surround the somata of most primary sensory neurons, isolating individual neuronal somata and closely regulating their microenvironment (Hanani et al., 2002; Pannese et al., 2003; Hanani, 2005). NMDAr are expressed on the peripheral and central terminals as well as in the soma of primary afferent sensory (DRG) neurons (Li et al., 2004). In the brain, NMDAr are involved in numerous functions including synaptic plasticity, learning and memory (Morris, 1989; Abumaria et al., 2011). In DRG sensory neurons, NMDAr contribute to the transmission of strong and/or prolonged painful stimuli but not to non-painful sensory signaling (Petrenko et al., 2003). NMDAr are ionotropic glutamate receptors that allow the entry of Ca2+ into the cell which can function as a second messenger in different signaling pathways as well as promote the release of intracellular Ca2+ and of neuromodulators. Glutamate is released from both the central (Jessell et al., 1986; Schneider and Perl, 1988; Yoshimura and Jessell, 1990) and peripheral (Bledsoe et al., 1980; Omote et al., 1998; deGroot et al., 2000; Lawand et al., 2000) nerve terminals of primary sensory neurons.

Abstract—Satellite glia cells (SGCs), within the dorsal root ganglia (DRG), surround the somata of most sensory neurons. SGCs have been shown to interact with sensory neurons and appear to be involved in the processing of afferent information. We found that in rat DRG various N-methyl-D-aspartate receptor (NMDAr) subunits were expressed in SGCs in intact ganglia and in vitro. In culture, when SGCs were exposed to brief pulses of NMDA they evoked transient increases in cytoplasmic calcium that were inhibited by specific NMDA blockers (MK-801, AP5) while they were Mg2+ insensitive indicating that SGCs express functional NMDAr. The percentage of NMDA responsive SGCs was similar in mixed- (SGCs plus neurons) and SGC-enriched cultures. The pattern of the magnitude changes of the NMDA-evoked response was similar in SGCs and DRG neurons when they were in close proximity, suggesting that the NMDA response of SGCs and DRG neurons is modulated by their interactions. Treating the cultures with nerve growth factor, and/or prostaglandin E2 did not alter the percentage of SGCs that responded to NMDA. Since glutamate appears to be released within the DRG, the detection of functional NMDAr in SGCs suggests that their NMDAr activity could contribute to the interactions

*Correspondence to: E. Recio-Pinto, Department of Anesthesiology, New York University (NYU) Langone Medical Center, 560 First Avenue, RR605, New York, NY 10016, USA. Tel: +1-212-263-6136; fax: +1-212-263-6139. E-mail addresses: [email protected] (C. Castillo), Monica. [email protected] (M. Norcini), [email protected] (L.A. Martin Hernandez), [email protected] (G. Correa), [email protected] (T. J. J. Blanck), recioe02@nyumc. org (E. Recio-Pinto). Abbreviations: BP, blocking peptide; BSA, bovine serum albumin; [Ca2+]cyt, cytoplasmic calcium; DRG, dorsal root ganglia; EDTA, ethylenediamine tetraacetic acid; HEPES, 4-(2-hydroxyethyl)-1piperazineethanesulfonic acid; NDS, normal donkey serum; NGF, nerve growth factor; NGS, normal goat serum; NHS, normal horse serum; NMDA, N-methyl-D-aspartate; NMDAr, N-methyl-D-aspartate receptors; PBS, phosphate-buffered saline; PBST, Triton-X-100 in PBS; PFA, paraformaldehyde; PGE2, prostaglandin 2; RT, room temperature; SGCs, satellite glia cells; SP, substance P.

0306-4522/12 $36.00 Ó 2013 IBRO. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.neuroscience.2013.02.031 135

136

C. Castillo et al. / Neuroscience 240 (2013) 135–146

There are also indications that glutamate can be released from the soma of primary sensory neurons (Brumovsky et al., 2007; Gu et al., 2010) and from SGCs (Jasmin et al., 2010). Hence glutamate release within the DRG will affect the neuronal somata excitability and potentially also modulate the interactions between SGCs and neurons. In this study, we investigate for the first time whether SGCs express NMDAr both in intact ganglia and in vitro.

EXPERIMENTAL PROCEDURES Cultures Adult male Sprague–Dawley rats (250 g) were used following the guidelines approved by the New York University Langone Medical Center Institutional Animal Care and Use Committee. Rats were anesthetized, perfused with cold artificial cerebrospinal fluid (through the aorta), decapitated and their DRG isolated, dissociated and cultured as previously described (Castillo et al., 2011). The ganglia roots were cut as close as possible to the ganglia to reduce the contribution of Schwann cells. Serum-free culture medium was used to reduce fibroblast replication which consisted of Neurobasal A medium (Gibco– Invitrogen, Carlsbad, CA, USA) supplemented with 2% B27, 0.5 mM Glutamax, penicillin (100 U/mL), streptomycin (100 lg/mL) and 10 mM HEPES, pH 7.4. These cultures will be referred to as ‘‘mixed-cultures’’ since they contained a large number of neurons and SGCs (see Results). Following cell attachment, culture medium was added (2 mL to the 25-mm coverslips that were used for functional studies, and 0.5 mL to the 12-mm diameter coverslips that were used for immunostaining). Prostaglandin 2 (PGE2) (1 lM) and nerve growth factor (NGF) (100 nM) were added after the cells were attached. PGE2 was added three times daily and the cells were fed every day. PGE2 analog 9-deoxy-9-methylene-16 Prostaglandin E2 was from Cayman Chem Co. (Ann Arbor, MI, USA), and NGF was from Upstate Biotechnology Inc. (Lake Placid, NY, USA). SGC-enriched cultures were prepared from 2-day-old mixedcultures as previously described (Villa et al., 2010). Briefly, the medium was removed and the cells were detached by incubating them with 1 mL of 0.25% Trypsin–EDTA solution (Sigma, St. Louis, MO, USA) for 15 min at 37 °C. The cell suspension was collected in a Eppendorff 1.5 mL tube, centrifuged (1200 rmp, 100g, 5 min, room temperature (RT)), resuspended in fresh NB medium and replated on uncoated glass coverslips. This procedure takes advantage of the ability of SGCs to strongly adhere to uncoated surfaces while the neurons cannot (Capuano et al., 2007, 2009). The SGC-enriched cultures were used at 2–5 days following enrichment. The few neurons that remained in the SGC-enriched cultures were rapidly removed by perfusion during the calcium imaging experiments. Hippocampal cultures were prepared as previously described (Brewer, 1995) and used at 6–7 days.

Calcium imaging We used 48–72-h cultures to allow the cells to recuperate from the mechanical and enzymatic treatment used for their isolation. Cells were loaded with the fluorescent Ca2+ indicator Fura-2 AM (5 lM) (Molecular Probes, Eugene, OR, USA) and cytoplasmic Ca2+ ([Ca2+]cyt) changes were measured as previously described at 32–35 °C (Corrales et al., 2003; Sutachan et al., 2006; Castillo et al., 2011). Coverslips were mounted on an open perfusion chamber (250 lL volume, type RC-21BRFS, or type RC-25F; Warner Instruments, Hamden, CT, USA). Cells were perfused (250 lL/min) with a

Mg2+-free-HEPES buffer ((in mM) 140 NaCl, 5 KCl, 5 NaHCO3, 10 HEPES, 2 CaCl2 and 10 glucose, pH 7.4) for 30 min prior to any stimulation. Drugs NMDA (Sigma), MK-801 and AP-5 (Tocris Biosciences, Bristol, UK), were prepared in the Mg2+-free-HEPES buffer and added directly on top of the cells by using the puffer valve-link system (Automate Scientific Inc., San Francisco, CA, USA) or the Octaflow focal perfusion system (ALA Scientific, Farmingdale, NY, USA). The solution containing NMDA (100 lM) also contained glycine (10 lM). Measurements were similarly done by using three systems: the PTI (Photon Technology International Inc., Lawrenceville, NJ, USA), the ATTO (Atto Instruments, Rockville, MD, USA) and the Ionoptix (Milton, MA, USA) systems. The excitation wavelengths for Fura-2 (340 nm and 380 nm) were alternately (every 0.6 s – ATTO, 2.4 s – PTI; or 0.9 s – Ionoptix) generated by passing a transmitted light source thorough separate filters (340- and 380-nm filters (ATTO, Ionoptix)); or by using a highspeed multiwavelength illuminator (PTI). The emitted fluorescence from individual cells was filtered with a fluorescence barrier filter (475 nm – ATTO; 515 nm – PTI, 510 nm – Ionoptix) and collected in a camera (Intensified CCD camera – ATTO; CS-HQ2 1394 – PTI; Myocam Camera – Ionoptix). The relative changes in [Ca2+]cyt are given by the ratio of the emission of Fura-2 at the barrier filter generated by the alternate excitation at 340 nm and 380 nm (ratio (340/380)). Either at the beginning or at the end of the experiment cells were depolarized with a pulse of K+ (50 mM) and/or stimulated with ATP (100 lM) to check that the neurons and SGCs that were being recorded from were healthy as determined by their capacity to display a depolarization- and/or ATP-evoked increase in [Ca2+]cyt (regardless of their capacity to respond to NMDA).

Immunocytochemistry Cells attached to coverslips were fixed, permeabilized and blocked as previously described (Castillo et al., 2011). Then coverslips were incubated (overnight, 4 °C) with primary antibodies (anti-): goat polyclonal anti-NMDAe2 (NR2B) #sc-1469 (1:1000), goat polyclonal anti-NMDAf1 (NR1) #sc-1467 (1:1000) (Santa Cruz Biotechnology, CA, USA), or with mouse monoclonal anti-S-100 protein #MAB079-1 (1:250; Chemicon International Inc., MA, USA). Then coverslips were incubated (1 h, RT) with the respective secondary antibody the donkey anti-goat Alexa Fluor-546 (1:1500) or the donkey antimouse Alexa Fluor-546 (1:1000) (Invitrogen Molecular Probes). Cells were counterstained with the nuclear stain Hoechst 33342 (Sigma–Aldrich) (10 lg/mL in phosphate-buffered saline (PBS), 15 min, RT). Negative controls were incubated in PBS containing 2% donkey serum without the primary antibody. DRG sections were obtained and immunostained as follows. Anesthetized animals underwent transcardiac perfusion through the aorta first with 10% sucrose (500 mL) and then with 4% paraformaldehyde (PFA, 500 mL) (ice cold). The lumbar DRG with 2–3 mm of their attached roots were collected and post-fixed over night (4% PFA, 4 °C). The ganglia were then placed in 30% sucrose (in PBS) and stored for a week at 4 °C (for cryoprotection). Then the ganglia were placed into individual tissue molds (Peel-A-Way Disposable Embedding Molds, Polyscience Inc.) filled up with Tissue-Tek OCT (Sakura Finetek Inc., Torrance, CA, USA), frozen (dry ice) and stored at 80 °C. Sections (12–20-lm thick) were cut using a cryostat (Leica Microsystem; Model cat# CM3050-S). The 10% sucrose solution (in PBS) contained Heparin (10 U/mL; Hospira Inc., Lake Forest, IL, USA). The 4% (w/v) PFA solution was prepared by mixing 40 g PFA (Acros, NJ, USA) with 450 mL ddH20 (pH 7.4 with NaOH), and then adding 500 mL 2-phosphate buffer and 50 mL 10-normal saline. The sections were placed on glass slides (Fischer Scientific) and a wall was made around them with PAP-PEN (Scientific Device

C. Castillo et al. / Neuroscience 240 (2013) 135–146 Laboratory, Des Plaines, IL, USA). Slices were rinsed four times for 5 min each, with PBS at RT. Then they were incubated for 60–90 min at RT in blocking/permeabilized solution (4% normal donkey serum (NDS) or normal goat serum (NGS) or 20% normal horse serum (NHS) with 1% bovine serum albumin (BSA) and 0.4% Triton-X-100 in PBS (PBST)). Then sections were incubated overnight (4 °C, 1% BSA, and 1% of either NGS, NDS or NHS, in 0.4% PBST) with the corresponding primary antibodies: goat polyclonal anti-NMDAe2 (1:100); goat polyclonal anti-NMDAf1 (1:100); mouse monoclonal anti-glutamine synthetase clone GS-6, #MAB302 (1:200; Millipore); mouse monoclonal anti-GFPA #MAB34020 (5 lL/ mL, Chemicon); or with mouse monoclonal anti-S-100 protein (1:100 – Fig. 1B; 1:400 – Fig. 1C). When using the anti-NMDAr subunits, the day after, the sections were incubated with the primary antibody for an additional 1 h at RT. Afterward the sections were rinsed five times for 5 min each with 0.25% BSA in 0.02% PBST; and incubated (90 min, RT) with the respective secondary antibody: donkey anti-goat Alexa Fluor-456 (1:1000), goat anti-mouse Alexa Fluor-456 (1:1000), or donkey anti-mouse Alexa Fluor-488 (1:1000). The secondary antibodies were diluted in 1% of either NDS, NGS or NHS, 1% BSA in 0.02% PBST. Sections were washed three times with 0.1% PBST and two times with PBS, for 5 min each. Slices were stained with Nissl #N21480 (1:100; Invitrogen Molecular Probes) (20 min, RT) according to the manufacturer’s protocol and with the nuclear stain Hoechst 33342 (Sigma–Aldrich) (10 lg/mL in PBS, 20 min, RT). Negative controls were incubated in PBS containing 1% of either NDS, NGS or NHS, and the primary antibody was omitted or replaced with IgG goat or IgG mouse (Jackson ImmunoResearch). The IgG working solutions were prepared to the same concentration as the corresponding primary antibodies working solutions. An additional negative control was used for the NR1 and NR2B antibodies that consisted in the simultaneous incubation with the primary antibody and their corresponding blocking peptide (BP). Prior to their addition, the primary antibodies and corresponding BPs were premixed (for 2 h at RT with agitation); the BP concentration was 1 lg/0.1 mL (NR1 BP cat# sc-1467-P; NR2B BP cat# sc-1469-P, Santa Cruz Biotechnology). Coverslips were mounted using Aqua Poly/ Mount (Polysciences Inc., Bayonne, NJ, USA). Images were captured (and analyzed) with a Zeiss Axiovert 200 (Germany) inverted microscope equipped with fluorescence and Nomarski optics, objective 40; ApoTome (for optical sections), using an Axiocam camera (Zeiss) and Axiovision Software version 4.6.3 (Carl Zeiss Imaging Systems, Thornwood, NY, USA). Confocal images were captured (and analyzed) with a Zeiss LSM 700 laser confocal microscope; Objective: Plan-Apochromat 63/1.40 Oil DIC M27 and Zen Lite Software version 2011 (Carl Zeiss MicroImaging GmbH).

RESULTS Detection of satellite glia cells in DRG with Nissl staining One of our goals was to determine whether NMDAr were expressed in SGCs in intact ganglia. In intact ganglia, however, SGCs are closely located and in most cases they almost entirely surround the soma of DRG neurons (Pannese, 1981; Hanani et al., 2002; Hanani, 2005). Moreover NMDAr are known to be expressed in DRG neurons (Marvizon et al., 2002; Willcockson and Valtschanoff, 2008). Therefore, we first tested in thin DRG sections various neuronal and glial markers to determine a way to identify regions in which both of these cell types were not overlapping. We found that the use of Nissl staining provided us with a means to

137

locate and identify regions of neuronal somata and SGCs that did not overlap with each other in thin DRG sections. Nissl binds to negatively charged nucleic acids such as RNA and DNA, and has mostly been used as a neuronal marker, and as expected it labeled the DRG neurons (Fig. 1A, thick arrows). We found that Nissl also labeled very strongly the nucleus and to a lower extent the soma of non-neuronal cells including the SGCs surrounding the neuronal soma (Fig. 1A, B, green). The strong Nissl labeling in these non-neuronal cells completely overlapped with the nuclear Hoechst staining (Fig. 1A, B, blue). The Nissl staining therefore allowed us to identify SGCs that did not overlap with neurons (Fig. 1A, thin arrows). In intact ganglia the non-neuronal cells that surround the neuronal soma were labeled with glial markers including the anti-glial fibrillary acidic protein (GFAP) (Fig. 1A, red) and with the anti-S100 protein (Fig. 1B, red) confirming that they were SGCs as previously reported (Stefansson et al., 1982; Jessen et al., 1984; Nascimento et al., 2008). Visualization of thin optical sections (3–4 lm) showed that the Nissl staining remained highly detectable in the nuclei of the SGCs while the level of staining in the neuronal somata was reduced as was expected due to the much lower neuronal volume (Fig. 1C, D, green). In these optical sections the SGCs can be easily identify with Nissl labeling (green) and their somas were also labeled with either the anti-S100 protein (Fig. 1C, red) or with another glial marker, the anti-glutamine synthetase that has also been previously shown to label SGCs (Weick et al., 2003; Hanani, 2005) (Fig. 1D, red). These images also show that the nuclei of Schwann cells (identified by their location within the DRG, mostly at their nerve roots) were also labeled with Nissl but they were negative for the anti-glutamine synthetase and only a few of them were positive for the anti-S100 protein (Fig. 1C, D, dashed arrows). Hence, within the DRG, the anti-S100 protein and anti-glutamine synthetase labeled almost exclusively the SGCs (based on their location surrounding the neuronal somata). Satellite glia cells in DRG ganglia express the NR1 and NR2B subunits of the NMDA receptor The NR1 (Fig. 2A) and NR2B (Fig. 2D, G) subunits of the NMDAr are expressed in the DRG neuronal somata in intact ganglia, as previously reported (Marvizon et al., 2002; Willcockson and Valtschanoff, 2008). In order to ascertain whether SGCs also expressed these NMDAr subunits, we selected SGCs whose soma was mostly not overlapping with the neuronal soma. This was done by using Nissl staining which labels the SGC’s soma to a lesser extent than the neuronal soma. In the two enlarged regions of Fig. 2A (second and third rows), five of the seven SGCs, that do not appear to completely overlap with the neuron, were labeled with anti-NR1 (arrows) the other two did not labeled with anti-NR1 (open arrows). Similarly some (arrows) but not all (open arrows) of the SGCs labeled with the anti-NR2B (Fig. 2D, G). The selected SGCs are in very close contact with the neuronal soma as can be seen in the light images (Fig. 2A, D, right panels). Colabeling with

138

C. Castillo et al. / Neuroscience 240 (2013) 135–146

Fig. 1. Identification of SGCs using Nissl staining and glial markers in DRG sections. DRG sections were immunostained with either anti-GFAP (A), anti-S100 protein (B, C), or anti-glutamine synthetase (D) (red). The slices were then stained with Nissl (green) (A–D) and with the nuclear stain Hoechst (blue) (A, B). Some of the neurons (thick arrows), SGCs (thin arrows) and Schwann cells (dashed arrows) are indicated. In (D) there is a view of the surface of one of the neuronal somata covered with SGCs (thick open arrow). Regular pictures (A, B); optical sections (4 lm thick, ApoTome) (top: C, D) and corresponding bright field images (bottom: C, D). Magnification 40 objective. Scale bars = 20 lm. Dashed yellow lines separate regions containing neuronal somas from those containing nerve roots.

anti-glutamine synthetase, confirmed that some but not all of the SGCs expressed NR1 (Fig. 2H) and NR2 (Fig. 2I)

subunits. These data support that some of the SGCs identified either by their close location to the neuronal

C. Castillo et al. / Neuroscience 240 (2013) 135–146

139

Fig. 2. In intact ganglia SGCs express the NR1 and NR2B subunits of the NMDAr. (A–G) DRG sections were immunostained with either anti-NMDAn1 (NR1 subunit) (A–C) or anti-NMDAe2 (NR2B subunit) (D–G) (red); and stained with Nissl (green). Individual and merged images are shown. (A, D) The regions within the white dashed boxes, are shown at a higher magnification at the bottom (A, D) or to the right (G) of the image. Two negative controls are shown: (B, E); replacing the primary antibody with goat IgG; and (C, F) incubation of the primary antibody with their corresponding blocking peptide (+BP). (H, I) DRG sections were immunostained with anti-glutamine synthetase (green) and with either anti-NMDAn1 (H) or anti NMDAe2 (I) (red). The nuclei stained with Hoechst (blue). Images are optical sections taken either with, confocal (A–C, H, I, left: 4 lm thick; I, right: 1.5 lm thick) or with ApoTome (G: 4 lm thick). (D–F) Regular images and the merged image also include the bright field image. Right panels in (A) and (D) of the enlarged regions show the corresponding bright field pictures. In all the images some of the SGCs that were positive (solid arrows) or negative (open arrows) for a given NMDAr subunit are marked. Magnification objective 40 (D–G) or 63 (A–C, H, I). Scale bars = 20 lm.

somata (Fig. 2A, D and G) or by labeling with anti-glutamine sythetase (Fig. 2H, I) express the NR1

and NR2B subunits of the NMDAr in intact ganglia. In addition we found that these NMDA receptor subunits

140

C. Castillo et al. / Neuroscience 240 (2013) 135–146

were also expressed in SGCs in cultures of dissociated DRG (Fig. 3A, B). As previously published, labeling with anti-NR2B was high and spotty at the perinuclear regions in the neuronal somata (Marvizon et al., 2002). We found that in SGCs labeling with anti-NR2B was also spotty throughout their somata (Fig. 3A). In the dissociated DRG cultures many of the neurons were either not surrounded or partially surrounded by SGCs (Fig. 3C, D, solid arrows) while some of the neurons were completely surrounded by SGCs (Fig. 3D, right open arrow). Moreover, most of the non-neuronal cells in these cultures were labeled with the anti-S100 protein, indicating that most of them were SGCs. Satellite glia cells express functional NMDA receptors In order to examine whether the SGCs expressed functional NMDAr we measured their NMDA-evoked [Ca2+]cyt responses in dissociated DRG cultures by using the cytoplasmic Ca2+ indicator Fura-2. While doing these experiments it was noticed that the emitted fluorescence following excitation at 380 nm was usually higher in SGCs than in neurons (Fig. 3E, F). The larger Fura-2 signal in SGCs as compared to neurons that we observed in dissociated DRG cultures has been previously observed in intact murine trigeminal ganglia (Weick et al., 2003). Hence, prior to exposing the cells to NMDA, the cells were excited with 380 nm, and these images together with the light images were used to determine neuronal and SGC regions that did not overlap with each other (e.g. Fig. 4A). NMDA (100 lM) pulses were applied for 4 s every 2–5 min for about an hour. SGCs displayed transient increases in [Ca2+]cyt when exposed to NMDA (Fig. 4B–D, top traces). It has been reported that brief exposure to substance P (SP) (Wu et al., 2004; Castillo et al., 2011), as well as the simultaneous treatment with NGF and PGE2 (Castillo et al., 2011) enhances the NMDA response of sensory neurons. Therefore, we decided to investigate whether these treatments also affected the NMDA response in SGCs. As we previously reported in sensory neurons (Castillo et al., 2011) in the absence of SP prepulse, sensory neurons became unresponsive to repetitive NMDA applications (Fig. 4B, bottom trace), while in SGCs the response to NMDA was usually maintained over time (Fig. 4B, top trace). Fig. 4C, D shows NMDAevoked responses of individual SGCs (top traces) and neurons (bottom traces) when cultured in the presence of NGF without or with PGE2, respectively. The proportion of NMDA-responsive neurons was higher in the presence of NGF plus PGE2 than in the presence of only NGF, consistent with our previous findings (Castillo et al., 2011). In contrast the proportion of NMDAresponsive SGCs was similar in both conditions. In fact the proportion of NMDA-responsive SGCs was similar whether they were cultured in the absence or presence of NGF, PGE2 or NGF plus PGE2 (Fig. 4C, open bars). This is in contrast with our previous observation in which the percentage of NMDA-responsive neurons increased from about 15% to 60% when they were cultured with NGF plus PGE2 (Fig. 4E, filled bars taken from Castillo et al., 2011).

We attempted to do a concentration response for the NMDA-evoked transient increase in [Ca2+]cyt. We found that when cells were initially exposed to low NMDA levels (1–10 lM) they did not demonstrate a clear response (Fig. 5A). However following exposure to high NMDA levels (100–200 lM), subsequent exposure to low NMDA levels (0.1–10 lM) evoked clear increases in [Ca2+]cyt (Fig. 5A–C). This behavior was similar in SGCs and DRG neurons (Fig. 5A–C). The magnitude of the NMDA-evoked response was directly proportional in DRG neurons (black traces) and in those SGCs that were in physical contact with the neurons (blue and green traces) (Fig. 5A–C); but not in SGCs that were closely located but not in physical contact (red trace) with the DRG neuron (Fig. 5C). Hence the magnitude of the NMDA response of SGCs and DRG neurons appears to be modulated by their interactions. The percentage of SGCs that responded to NMDA was similar in mixed- and SGC-enriched cultures (Fig. 5F). The lack of a clear concentration response for the NMDA-evoked [Ca2+]cyt increase in both SGCs and DRG neurons (Fig. 5E); reflects both the change in the magnitude of the NMDA-evoked response upon multiple exposures to the same NMDA concentration (Fig. 4); and the dependence of the magnitude of the response to low NMDA levels on previous stimulations with high NMDA levels (Fig. 5A–D). The underlying causes for these changes in the NMDA response were not investigated, but most likely involve alterations of intracellular Ca2+ stores and/or priming of NMDAr following repetitive NMDA exposures. It has been reported that NMDAr from CNS glia are insensitive to block by Mg2+ (Ziak et al., 1998; Karadottir et al., 2005; Lalo et al., 2006); we found that 1.5 mM of Mg2+ did not block the NMDA-evoked transient increase in [Ca2+]cyt in SGCs. The magnitude of the normalized NMDA-evoked [Ca2+]cyt transient following exposure to Mg2+ (control NMDA-evoked response/NMDA-evoked response in the presence of Mg2+) was 1.64 ± 0.35 mean ± SEM, n = 7 SGCs. To confirm that the observed NMDA-evoked response was initiated by the opening of NMDAr we used two specific NMDA receptor blockers, MK-801 and AP-5 (Parks et al., 1991). In SGCs, MK-801 (10 lM) (Fig. 5G) and AP5 (50 lM) (Fig. 5H) blocked the NMDA-evoked responses and in both cases the block was slowly reversible. Under similar conditions MK-801 also blocked the NMDA-evoked response in DRG sensory neurons (Fig. 5I) and hippocampal neurons (Fig. 5J). Following, removal of MK-801, the recovery of the NMDA response appears to be fastest in DRG neurons, and within the period measured it did not occur in hippocampal neurons. Hence, the observed NMDAevoked [Ca2+]cyt increases in SGCs and sensory neurons are initiated by the opening of NMDAr.

DISCUSSION In this study we demonstrated for the first time that SGCs express functional NMDAr. Our findings also indicate that the magnitude of the NMDA response of SGCs and DRG

C. Castillo et al. / Neuroscience 240 (2013) 135–146

141

Fig. 3. Identification of SGCs in dissociated DRG serum-free cultures. (A, B) Images of cultures that were immunostained with either anti-NMDAn1 or anti NMDAe2 (red) and the nuclei stained with Hoechst (blue). Arrows point to some of the SGCs. The (B) regions of two SGCs are shown at a higher magnification. Right panels show their corresponding bright field images overlapped with the one showing the nuclear staining. Magnification 40 objective. Scale bars = 20 lm. (C, D) Images of cultures that were immunostained with anti-S100 protein (red) and the nuclei stained with Hoechst (blue). Three neurons are indicated (arrows) one of them was completely surrounded with SGCs (D, right open arrow). Magnification 40 objective. Top: The fluorescence images are overlapped with the bright field images. Scale bars = 20 lm. (E, F) Images of live cells. Bright field image (left) and image of their corresponding emitted fluorescence when excited at 380 nm (right). Magnification 60 objective. Thin arrows point to SGCs that were visualized in the bright field images and that had a strong emitted fluorescence. Thin broken arrow shows a SGC, which was on top of a neuron that was visualized in the bright field image but did not emit a strong fluorescence at the plane of focus. (D) Red dotted box in bright field image corresponds to the area shown in the fluorescence image. All images are from 48 h cultures. Scale bars = 20 lm (A–E) and 30 lm (F).

142

C. Castillo et al. / Neuroscience 240 (2013) 135–146

Fig. 4. NMDA-evoked increases in cytoplasmic calcium in SGCs. (A) An image of live cells. Bright field image (left) and picture of the corresponding emitted fluorescence when excited at 380 nm (right). Magnification 40 objective. Scale bars = 20 lm. For selecting neuronal and SGC regions (dotted circles) that were not overlapping with each other, we used both the light images and the emitted fluorescence images. (B) NMDA-evoked transient increases in [Ca2+]cyt in a SGC and a neuron, that were not exposed to a SP prepulse. (C, D) NMDA-evoked [Ca2+]cyt changes of individual SGCs (top) and neuronal somas (bottom); when the cells were cultured in the presence of NGF (C) or NGF plus PGE2 (D). In (C) and (D), cells were exposed to a 5-s SP prepulse (0.3 lM) prior to starting the NMDA pulses. (B–D) The relative changes in [Ca2+]cyt (D[Ca2+]cyt) are given by the Dratio of the emission of Fura-2 at the barrier filter generated by the alternate excitation at 340 nm and 380 nm (Dratio (340/380)). The images and traces shown in (A)–(D) were collected with the Atto system. (E). Average of the percentage of SGCs that responded to NMDA with a transient increase in [Ca2+]cyt, when cells were grown in the absence (control) or presence of NGF (100 nM) and/or PGE2 (1 lM). Data collected with the three systems were used. For SGCs no significant difference was found between the groups. For neurons #P < 0.05 between the group and the untreated group. One-way ANOVA, and Dunnett’s multiple test (GraphPad Prism 5.0 Software, San Diego, CA, USA). The data for sensory neurons were taken from Castillo et al. (2011); and presented as a mean between experiments.

neurons is modulated by the interactions between these cells. NMDAr are ionotropic glutamate receptors, hence in vivo they are activated by glutamate. There are several indications that glutamate can be released within the DRG. When outside-out patches of cells that are overexpressing glutamatergic receptors are placed on the surface of the somata of a DRG neuron these patches display glutamatergic excitatory currents following an electrical stimulation of the somata (Gu et al., 2010). In addition, the presence of glutamate receptors in the neuronal somata (Marvizon et al., 2002; Willcockson and Valtschanoff, 2008), the presence of vesicular glutamate transporters that are associated with the somatic plasma membrane (Brumovsky et al., 2007), the presence of all the proteins necessary for the uptake and recycling of glutamate in SGCs (Jasmin et al., 2010), also support that glutamate could be released within the DRG. SGCs appear to also contribute to glutamate recycling since they express both the glutamate-aspartate transporter and glutamine synthetase (Hanani, 2005; Ohara et al., 2009). Reducing glutamine synthetase expression in SGCs in the trigeminal ganglion produces analgesia (formalin test), presumably by decreasing the source of glutamine that is being produced and released by SGCs and subsequently used by neurons to produce glutamate (Jasmin et al., 2010). Therefore, the activity of NMDAr not only in the neuronal somata but also in SGCs could potentially modulate the excitability of neuronal somata. Additional studies are needed to investigate this possibility as well as other potential physiological roles of NMDAr in SGCs. Previously labeling of NMDAr subunits in DRG was ascribed to the neuronal somata (Marvizon et al., 2002; Willcockson and Valtschanoff,

2008). However, in those studies no attempt was done to identify SGCs (i.e. nuclear or SGCs labeling), and since SGCs are very closely located to the surface of the neuronal somata, it is possible that the previously reported NMDAr subunit labeling was from both the neuronal somata and their surrounding SGCs. It has been reported that over time in culture the expression of some SGCs markers can be altered. Although glutamine synthetase remains detectable in long term cultures (Capuano et al., 2009); it has also been reported that the expression of glutamine synthetase decreases while the expression of the purinergic receptor P2X7 is maintained, indicating that SGCs may be undergoing phenotypic changes over time in culture (Belzer et al., 2010). We do not know whether the level of NMDAr expression in SGCs was altered in culture; however, NMDAr expression was detected in SGCs both in intact ganglia and in culture supporting that NMDAr in SGCs are physiological relevant. SGCs have been previously postulated to interact with neurons within the DRG. One of the postulated functions of SGCs is to regulate the ionic homeostasis of the small extracellular space that surrounds the neuronal somata within the DRG (Hanani, 2005; Tang et al., 2010). SGCs could prevent the accumulation of extracellular K+ (released by neurons when excited) since a decrease of the activity of the inwardly rectifying K+ channel, Kir4.1, in SGCs leads to neuronal hyperexcitability and heightened nociception (Hanani, 2005; Vit et al., 2008; Tang et al., 2010). SGCs have been postulated to also be actively involved in the processing of afferent information since nerve stimulation not only leads to a [Ca2+]cyt increase in the neuron but also in the surrounding SGCs; and

C. Castillo et al. / Neuroscience 240 (2013) 135–146

143

Fig. 5. NMDA-evoked increases in cytoplasmic calcium in SGCs and neurons are sensitive to MK-801. (A–C) Increases in [Ca2+]cyt in SGCs and neurons evoked by different levels of NMDA. When SGCs were in contact with a neuron, the NMDA response of the SGCs (blue and green traces) resembled that of the neurons (black traces). Red trace is the NMDA-evoked response of a SGC that was close to (20 lm) but not in contact with the neuron. (D) Increases in [Ca2+]cyt in SGCs evoked by different levels of NMDA in SGC-enriched cultures. (A–D) The time scale = 5 min (top left lines); KCl 50 mM, and ATP 100 lM; and [Ca2+]cyt is given by the Dratio of the emission of Fura-2 at the barrier filter generated by the alternate excitation at 340 nm and 380 nm (ratio (340/380)). (E) Magnitude of the NMDA-evoked transient increase in [Ca2+]cyt in a SGCs and DRG neurons at various NMDA levels. Cells were exposed to 3–4 concentrations of NMDA. For a given cell all of the responses (area of the transient increase in [Ca2+]cyt) at a given concentration were averaged; and normalized to that obtained at 10 lM. #P < 0.05, one-way ANOVA, and Dunnett’s multiple test. (F) The % of NMDA responsive SGCs at various NMDA levels in mixed- and SGC-enriched cultures. n = number of experiments. No significant difference was found between the groups in either mixed- or SGC-enriched cultures (one-way ANOVA). (G–I) MK-801 blocked the NMDA-evoked increases in [Ca2+]cyt in SGCs (G), DRG neurons (H) and in hippocampal neurons (I). One group of cells (Group 1) was exposed to consecutive pulses of only NMDA. A second group of cells (Group 2) were exposed to consecutive pulses of NMDA first in the absence (open bars), then in the presence of MK-801 (10 lM, black bars) and following the removal of MK-801 (open bars). Consecutive pulses of NMDA (10 lM for SGCs, 100 lM for DRG neurons and hippocampal neurons) were applied every 3–5 min. For each cell the NMDA-evoked responses (area of the transient increase in [Ca2+]cyt) were normalized to the first one. (E, G–I) The relative changes in [Ca2+]cyt (D[Ca2+]cyt) are given by the Dratio (340/ 380). All the data in this figure was collected using the Ionoptix system. #P < 0.05, +P < 0.01, ⁄P < 0.001 between group and the group prior to adding MK-801. One-way ANOVA, and Dunnett’s multiple test.

144

C. Castillo et al. / Neuroscience 240 (2013) 135–146

mechanical stimulation of SGCs result in [Ca2+]cyt waves that spread not only to the neighboring SGCs but also to the neighboring neuron (Zhang et al., 2007; Gu et al., 2010; Suadicani et al., 2010). Hence, it is possible that NMDAr activity in SGCs is contributing to the interactions between SGCs and neurons within the DRG. Functional NMDAr have been detected previously in other glia such as in astrocytes in the brain cortex (Luque and Richards, 1995; Conti et al., 1996, 1999; Standaert et al., 1999; Schipke et al., 2001, 2002); astrocytes and microglia in the hippocampus (Gottlieb and Matute, 1997; Krebs et al., 2003), and in astrocytes and oligodendrocytes in the spinal cord (Ziak et al., 1998). Glutamate released from neurons activate both ionotropic (including NMDAr) and metabotropic receptors on cortical astroglial cells leading to an increase in intracellular calcium (Dani et al., 1992; Porter and McCarthy, 1996; Pasti et al., 2001) which in turn evokes glutamate release from astroglia (Parpura and Haydon, 2000; Pasti et al., 2001). Activation of NMDAr in SGCs could lead to an increase in the release of glutamate and other substances from SGCs and by doing so enhances excitability of the neuronal somata. NMDAr in SGCs were found to be insensitive to block by extracellular Mg2+ as previously reported for various CNS glia (Ziak et al., 1998; Karadottir et al., 2005; Lalo et al., 2006). The lack of Mg2+ block indicates that NMDAr in SGCs can be functional under physiological conditions. This is important since glia cells do not undergo the strong depolarization that is required for removing the voltage-dependent Mg2+ block displayed by neuronal NMDAr (Mayer et al., 1984; Nowak et al., 1984; Kirson et al., 1999; Li et al., 2006). Two NMDA blockers, MK-801 and AP5 blocked the NMDA-evoked responses in SGCs, supporting that the NMDA-evoked [Ca2+]cyt increase in SGCs is initiated by the opening of NMDAr. In SGCs, upon removal of MK-801 about 40– 50% of the NMDA response recovered following two NMDA pulses; in contrast the NMDA response shows no recovery in hippocampal neurons. It has been reported, that in hippocampal neurons the block of the NMDA response with MK-801 displays no recovery not even after 30 min following MK-801 removal (Tovar and Westbrook, 2002). In contrast, in neocortical neurons the block with MK-801 displayed 40% recovery at 5 min and if the cells were depolarized the NMDA response was fully recovered within a minute following MK-801 removal (Huettner and Bean, 1988). Depolarization also promoted recovery from MK-801 block of the NMDA responses in hippocampal neurons (Atasoy et al., 2008). In our study SGCs and DRG neurons display a recovery from MK-801 block that resembles more closely that of neocortical neurons than those of hippocampal neurons. Presently the underlying basis for the difference in recovery rates following the removal of MK-801 observed on different cell types is not known. Such differences could reflect differences in the amount of NMDAr subunits that are available for recycling, or differences in subunits expressed or their processing. In primary sensory neurons brief exposure to SP, or the simultaneous treatment with NGF and PGE2

enhances their NMDA response (Wu et al., 2004; Castillo et al., 2011); which may in turn contribute to their known action in enhancing the electrical activity of these neurons (England et al., 1996; Nicol et al., 1997; Carlton et al., 1998; Abdulla et al., 2001; Kasai and Mizumura, 2001a,b; Kitamura et al., 2005). In this study we found that similar exposure to these factors (SP, NGF, PGE2) did not affect the NMDA response in SGCs. Hence, although the magnitude of the NMDA response of SGCs and DRG neurons appears to be modulated by the interactions between these cells; the NMDA response is differentially regulated in SGCs and DRG neurons by these factors. In summary we demonstrated for the first time that SGCs express functional NMDAr. NMDAr were detected in SGCs both in intact dorsal root ganglia and in vitro. The activation of the NMDAr leads to increases in [Ca2+]cyt in SGCs. The NMDA response is differentially regulated in SGCs and DRG neurons by SP and NGF plus PGE2; while the magnitude of the NMDA response of SGCs and DRG neurons appears to be modulated by the interactions between these cells.

Acknowledgments—Special thanks to Lisbeth Garcia, Jorge Nun˜ez and Dr. Jin Zhang, for technical assistance.

REFERENCES Abdulla FA, Stebbing MJ, Smith PA (2001) Effects of substance P on excitability and ionic currents of normal and axotomized rat dorsal root ganglion neurons. Eur J Neurosci 13:545–552. Abumaria N, Yin B, Zhang L, Li XY, Chen T, Descalzi G, Zhao L, Ahn M, Luo L, Ran C, Zhuo M, Liu G (2011) Effects of elevation of brain magnesium on fear conditioning, fear extinction, and synaptic plasticity in the infralimbic prefrontal cortex and lateral amygdala. J Neurosci 31:14871–14881. Atasoy D, Ertunc M, Moulder KL, Blackwell J, Chung C, Su J, Kavalali ET (2008) Spontaneous and evoked glutamate release activates two populations of NMDA receptors with limited overlap. J Neurosci 28:10151–10166. Belzer V, Shraer N, Hanani M (2010) Phenotypic changes in satellite glial cells in cultured trigeminal ganglia. Neuron Glia Biol 6:237–243. Bledsoe Jr SC, Bobbin RP, Thalmann R, Thalmann I (1980) Stimulusinduced release of endogenous amino acids from skins containing the lateral-line organ in Xenopus laevis. Exp Brain Res 40:97–101. Brewer GJ (1995) Serum-free B27/neurobasal medium supports differentiated growth of neurons from the striatum, substantia nigra, septum, cerebral cortex, cerebellum, and dentate gyrus. J Neurosci Res 42:674–683. Brumovsky P, Watanabe M, Hokfelt T (2007) Expression of the vesicular glutamate transporters-1 and -2 in adult mouse dorsal root ganglia and spinal cord and their regulation by nerve injury. Neuroscience 147:469–490. Capuano A, De Corato A, Lisi L, Tringali G, Navarra P, Dello Russo C (2009) Proinflammatory-activated trigeminal satellite cells promote neuronal sensitization: relevance for migraine pathology. Mol Pain 5:43. Capuano A, Curro D, Dello Russo C, Tringali G, Pozzoli G, Di Trapani G, Navarra P (2007) Nociceptin (1–13)NH2 inhibits stimulated calcitonin-gene-related-peptide release from primary cultures of rat trigeminal ganglia neurones. Cephalalgia 27:868–876.

C. Castillo et al. / Neuroscience 240 (2013) 135–146 Carlton SM, Zhou S, Coggeshall RE (1998) Evidence for the interaction of glutamate and NK1 receptors in the periphery. Brain Res 790:160–169. Castillo C, Norcini M, Baquero-Buitrago J, Levacic D, Medina R, Montoya-Gacharna JV, Blanck TJ, Dubois M, Recio-Pinto E (2011) The N-methyl-D-aspartate-evoked cytoplasmic calcium increase in adult rat dorsal root ganglion neuronal somata was potentiated by substance P pretreatment in a protein kinase Cdependent manner. Neuroscience 177:308–320. Conti F, DeBiasi S, Minelli A, Melone M (1996) Expression of NR1 and NR2A/B subunits of the NMDA receptor in cortical astrocytes. Glia 17:254–258. Conti F, Barbaresi P, Melone M, Ducati A (1999) Neuronal and glial localization of NR1 and NR2A/B subunits of the NMDA receptor in the human cerebral cortex. Cereb Cortex 9:110–120. Corrales A, Xu F, Garavito-Aguilar Z, Blanck TJ, Recio-Pinto E (2003) Isoflurane reduction of carbachol-evoked cytoplasmic calcium transients is dependent on caffeine-sensitive calcium stores. Anesthesiology 99:882–888. Dani JW, Chernjavsky A, Smith SJ (1992) Neuronal activity triggers calcium waves in hippocampal astrocyte networks. Neuron 8:429–440. deGroot J, Zhou S, Carlton SM (2000) Peripheral glutamate release in the hindpaw following low and high intensity sciatic stimulation. Neuroreport 11:497–502. England S, Bevan S, Docherty RJ (1996) PGE2 modulates the tetrodotoxin-resistant sodium current in neonatal rat dorsal root ganglion neurones via the cyclic AMP-protein kinase A cascade. J Physiol 495(Pt. 2):429–440. Gottlieb M, Matute C (1997) Expression of ionotropic glutamate receptor subunits in glial cells of the hippocampal CA1 area following transient forebrain ischemia. J Cereb Blood Flow Metab 17:290–300. Gu Y, Chen Y, Zhang X, Li GW, Wang C, Huang LY (2010) Neuronal soma–satellite glial cell interactions in sensory ganglia and the participation of purinergic receptors. Neuron Glia Biol 6:53–62. Hanani M (2005) Satellite glial cells in sensory ganglia: from form to function. Brain Res Brain Res Rev 48:457–476. Hanani M, Huang TY, Cherkas PS, Ledda M, Pannese E (2002) Glial cell plasticity in sensory ganglia induced by nerve damage. Neuroscience 114:279–283. Huettner JE, Bean BP (1988) Block of N-methyl-D-aspartateactivated current by the anticonvulsant MK-801: Selective binding to open channels. Proc Natl Acad Sci USA 85:1307–1311. Jasmin L, Vit JP, Bhargava A, Ohara PT (2010) Can satellite glial cells be therapeutic targets for pain control? Neuron Glia Biol 6:63–71. Jessell TM, Yoshioka K, Jahr CE (1986) Amino acid receptormediated transmission at primary afferent synapses in rat spinal cord. J Exp Biol 124:239–258. Jessen KR, Thorpe R, Mirsky R (1984) Molecular identity, distribution and heterogeneity of glial fibrillary acidic protein: an immunoblotting and immunohistochemical study of Schwann cells, satellite cells, enteric glia and astrocytes. J Neurocytol 13:187–200. Karadottir R, Cavelier P, Bergersen LH, Attwell D (2005) NMDA receptors are expressed in oligodendrocytes and activated in ischaemia. Nature 438:1162–1166. Kasai M, Mizumura K (2001a) Increase in spontaneous action potentials and sensitivity in response to norepinephrine in dorsal root ganglion neurons of adjuvant inflamed rats. Neurosci Res 39:109–113. Kasai M, Mizumura K (2001b) Effects of PGE(2) on neurons from rat dorsal root ganglia in intact and adjuvant-inflamed rats: role of NGF on PGE(2)-induced depolarization. Neurosci Res 41:345–353. Kirson ED, Schirra C, Konnerth A, Yaari Y (1999) Early postnatal switch in magnesium sensitivity of NMDA receptors in rat CA1 pyramidal cells. J Physiol 521(Pt. 1):99–111.

145

Kitamura N, Konno A, Kuwahara T, Komagiri Y (2005) Nerve growth factor-induced hyperexcitability of rat sensory neuron in culture. Biomed Res 26:123–130. Krebs C, Fernandes HB, Sheldon C, Raymond LA, Baimbridge KG (2003) Functional NMDA receptor subtype 2B is expressed in astrocytes after ischemia in vivo and anoxia in vitro. J Neurosci 23:3364–3372. Lalo U, Pankratov Y, Kirchhoff F, North RA, Verkhratsky A (2006) NMDA receptors mediate neuron-to-glia signaling in mouse cortical astrocytes. J Neurosci 26:2673–2683. Lawand NB, McNearney T, Westlund KN (2000) Amino acid release into the knee joint: key role in nociception and inflammation. Pain 86:69–74. Li J, McRoberts JA, Nie J, Ennes HS, Mayer EA (2004) Electrophysiological characterization of N-methyl-D-aspartate receptors in rat dorsal root ganglia neurons. Pain 109:443–452. Li J, McRoberts JA, Ennes HS, Trevisani M, Nicoletti P, Mittal Y, Mayer EA (2006) Experimental colitis modulates the functional properties of NMDA receptors in dorsal root ganglia neurons. Am J Physiol Gastrointest Liver Physiol 291:G219–G228. Luque JM, Richards JG (1995) Expression of NMDA 2B receptor subunit mRNA in Bergmann glia. Glia 13:228–232. Marvizon JC, McRoberts JA, Ennes HS, Song B, Wang X, Jinton L, Corneliussen B, Mayer EA (2002) Two N-methyl-D-aspartate receptors in rat dorsal root ganglia with different subunit composition and localization. J Comp Neurol 446:325–341. Mayer ML, Westbrook GL, Guthrie PB (1984) Voltage-dependent block by Mg2+ of NMDA responses in spinal cord neurones. Nature 309:261–263. Micu I, Jiang Q, Coderre E, Ridsdale A, Zhang L, Woulfe J, Yin X, Trapp BD, McRory JE, Rehak R, Zamponi GW, Wang W, Stys PK (2006) NMDA receptors mediate calcium accumulation in myelin during chemical ischaemia. Nature 439:988–992. Morris RG (1989) Synaptic plasticity and learning: selective impairment of learning rats and blockade of long-term potentiation in vivo by the N-methyl-D-aspartate receptor antagonist AP5. J Neurosci 9:3040–3057. Nascimento RS, Santiago MF, Marques SA, Allodi S, Martinez AM (2008) Diversity among satellite glial cells in dorsal root ganglia of the rat. Braz J Med Biol Res 41:1011–1017. Nicol GD, Vasko MR, Evans AR (1997) Prostaglandins suppress an outward potassium current in embryonic rat sensory neurons. J Neurophysiol 77:167–176. Nowak L, Bregestovski P, Ascher P, Herbet A, Prochiantz A (1984) Magnesium gates glutamate-activated channels in mouse central neurones. Nature 307:462–465. Ohara PT, Vit JP, Bhargava A, Romero M, Sundberg C, Charles AC, Jasmin L (2009) Gliopathic pain: when satellite glial cells go bad. Neuroscientist 15:450–463. Omote K, Kawamata T, Kawamata M, Namiki A (1998) Formalininduced release of excitatory amino acids in the skin of the rat hindpaw. Brain Res 787:161–164. Pannese E (1981) The satellite cells of the sensory ganglia. Adv Anat Embryol Cell Biol 65:1–111. Pannese E, Ledda M, Cherkas PS, Huang TY, Hanani M (2003) Satellite cell reactions to axon injury of sensory ganglion neurons: increase in number of gap junctions and formation of bridges connecting previously separate perineuronal sheaths. Anat Embryol (Berl) 206:337–347. Parks TN, Artman LD, Alasti N, Nemeth EF (1991) Modulation of Nmethyl-D-aspartate receptor-mediated increases in cytosolic calcium in cultured rat cerebellar granule cells. Brain Res 552:13–22. Parpura V, Haydon PG (2000) Physiological astrocytic calcium levels stimulate glutamate release to modulate adjacent neurons. Proc Natl Acad Sci U S A 97:8629–8634. Pasti L, Zonta M, Pozzan T, Vicini S, Carmignoto G (2001) Cytosolic calcium oscillations in astrocytes may regulate exocytotic release of glutamate. J Neurosci 21:477–484.

146

C. Castillo et al. / Neuroscience 240 (2013) 135–146

Petrenko AB, Yamakura T, Baba H, Shimoji K (2003) The role of Nmethyl-D-aspartate (NMDA) receptors in pain: a review. Anesth Analg 97:1108–1116. Porter JT, McCarthy KD (1996) Hippocampal astrocytes in situ respond to glutamate released from synaptic terminals. J Neurosci 16:5073–5081. Salter MG, Fern R (2005) NMDA receptors are expressed in developing oligodendrocyte processes and mediate injury. Nature 438:1167–1171. Schipke CG, Boucsein C, Ohlemeyer C, Kirchhoff F, Kettenmann H (2002) Astrocyte Ca2+ waves trigger responses in microglial cells in brain slices. FASEB J 16:255–257. Schipke CG, Ohlemeyer C, Matyash M, Nolte C, Kettenmann H, Kirchhoff F (2001) Astrocytes of the mouse neocortex express functional N-methyl-D-aspartate receptors. FASEB J 15:1270–1272. Schneider SP, Perl ER (1988) Comparison of primary afferent and glutamate excitation of neurons in the mammalian spinal dorsal horn. J Neurosci 8:2062–2073. Standaert DG, Friberg IK, Landwehrmeyer GB, Young AB, Penney Jr JB (1999) Expression of NMDA glutamate receptor subunit mRNAs in neurochemically identified projection and interneurons in the striatum of the rat. Brain Res Mol Brain Res 64:11–23. Stefansson K, Wollmann RL, Moore BW (1982) Distribution of S-100 protein outside the central nervous system. Brain Res 234:309–317. Suadicani SO, Cherkas PS, Zuckerman J, Smith DN, Spray DC, Hanani M (2010) Bidirectional calcium signaling between satellite glial cells and neurons in cultured mouse trigeminal ganglia. Neuron Glia Biol 6:43–51. Sutachan JJ, Montoya GJ, Xu F, Chen D, Blanck TJ, Recio-Pinto E (2006) Pluronic F-127 affects the regulation of cytoplasmic Ca2+ in neuronal cells. Brain Res 1068:131–137.

Takeda M, Takahashi M, Matsumoto S (2009) Contribution of the activation of satellite glia in sensory ganglia to pathological pain. Neurosci Biobehav Rev 33:784–792. Tang X, Schmidt TM, Perez-Leighton CE, Kofuji P (2010) Kir4.1 is responsible for the native inward potassium conductance of satellite glial cells in sensory ganglia. Neuroscience 166:397–407. Tovar KR, Westbrook GL (2002) Mobile NMDA receptors at hippocampal synapses. Neuron 34:255–264. Villa G, Fumagalli M, Verderio C, Abbracchio MP, Ceruti S (2010) Expression and contribution of satellite glial cells purinoceptors to pain transmission in sensory ganglia: an update. Neuron Glia Biol 6:31–42. Vit JP, Ohara PT, Bhargava A, Kelley K, Jasmin L (2008) Silencing the Kir4.1 potassium channel subunit in satellite glial cells of the rat trigeminal ganglion results in pain-like behavior in the absence of nerve injury. J Neurosci 28:4161–4171. Weick M, Cherkas PS, Hartig W, Pannicke T, Uckermann O, Bringmann A, Tal M, Reichenbach A, Hanani M (2003) P2 receptors in satellite glial cells in trigeminal ganglia of mice. Neuroscience 120:969–977. Willcockson H, Valtschanoff J (2008) AMPA and NMDA glutamate receptors are found in both peptidergic and non-peptidergic primary afferent neurons in the rat. Cell Tissue Res 334:17–23. Wu ZZ, Guan BC, Li ZW, Yang Q, Liu CJ, Chen JG (2004) Sustained potentiation by substance P of NMDA-activated current in rat primary sensory neurons. Brain Res 1010:117–126. Yoshimura M, Jessell T (1990) Amino acid-mediated EPSPs at primary afferent synapses with substantia gelatinosa neurones in the rat spinal cord. J Physiol 430:315–335. Zhang X, Chen Y, Wang C, Huang LY (2007) Neuronal somatic ATP release triggers neuron–satellite glial cell communication in dorsal root ganglia. Proc Natl Acad Sci U S A 104:9864–9869. Ziak D, Chvatal A, Sykova E (1998) Glutamate-, kainate- and NMDAevoked membrane currents in identified glial cells in rat spinal cord slice. Physiol Res 47:365–375.

(Accepted 13 February 2013) (Available online 26 February 2013)