- Email: [email protected]
Electrophysiological characterization of N-methyl-d -aspartate receptors in rat dorsal root ganglia neurons
Pain 109 (2004) 443–452 www.elsevier.com/locate/pain
Electrophysiological characterization of N-methyl-D -aspartate receptors in rat dorsal root ganglia neurons Jichang Li, James A. McRoberts*, Jingjiang Nie, Helena S. Ennes, Emeran A. Mayer Division of Digestive Diseases, Department of Medicine, Center for Neurovisceral Sciences (CNS) and Women’s Health, University of California, Warren Hall, Room 14-103, 900 Veterans Avenue, Los Angeles, CA 90095, USA Received 11 October 2003; received in revised form 29 January 2004; accepted 23 February 2004
Abstract In the peripheral nervous system, N-methyl-D -aspartate receptors (NMDAR) expressed on the central and peripheral terminals of primary afferent neurons are involved in nociception. We used single cell imaging of intracellular calcium concentration ([Ca2þ]i) and patch clamp techniques to characterize the functional properties of NMDARs on adult rat dorsal root ganglia (DRG) neurons in primary culture and selectively on those innervating the distal colon. In Mg2þ-free extracellular solution, rapid perfusion of DRG neurons with 250 mM NMDA and 10 mM glycine caused a significant increase in [Ca2þ]i, and elicited inward currents in whole cell patch clamp recordings when the holding potential was 2 60 mV. Both effects were reversibly inhibited by 200 mM ketamine in a use-dependent manner. The EC50 values for NMDA and glycine were 64 and 1.9 mM with Hill slope coefficients of 1.4 and 1.3, respectively. At negative potentials, extracellular Mg2þ blocked currents in a concentration- and voltage-dependent manner. The IC50 for Mg2þ at a holding potential of 2100 mV was 2.0 mM. The NMDAR subtype-selective antagonist, ifenprodil, inhibited 94% of the NMDA and glycine-induced current with an IC50 of 2.6 mM. There was no evidence of multiple binding sites for ifenprodil. There was no significant difference in the NMDAR current density on DRG neurons that had innervated the colon, nor was there a difference in the EC50 for ifenprodil. These results demonstrate that functional NMDARs expressed by DRG neurons innervating both somatic and visceral tissues of adult rats are composed predominantly of NR2B subunits. q 2004 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. Keywords: Calcium imaging; Glutamate receptor channel; Glutamate receptor subtype; NMDA receptor; NMDA receptor antagonist; NMDA receptor channel; Primary afferent neuron; Whole cell patch clamp
1. Introduction N-methyl-D -aspartate receptors (NMDAR) are ionotropic glutamate receptors that require glycine as a co-agonist, display voltage-dependent inhibition by extracellular Mg2þ, and have high permeability to Ca2þ. These properties contribute to the unique role of NMDARs in neuronal plasticity (Dingledine et al., 1999; Yamakura and Shimoji, 1999). Molecular biological studies have revealed three NMDAR subunit families: NR1, NR2 and NR3. The NR1 family has one member with eight different splice variants, the NR2 family has four members derived from different genes, denoted NR2A – D, and the NR3 family has two members NR3A – B (Cull-Candy et al., 2001; Michaelis, 1998). All NMDARs appear to function as heteromeric assemblies composed of two NR1 subunits in combination * Corresponding author. Tel.: þ 1-310-825-4786; fax: þ1-310-794-2864. E-mail address: [email protected] (J.A. McRoberts).
with two or three NR2 and/or NR3 subunits (Hawkins et al., 1999; Premkumar and Auerbach, 1997; Schorge and Colquhoun, 2003). Studies with recombinant receptors have provided an understanding of how receptor properties are defined by individual NMDAR subunits. NR2 subunits influence the affinity of the receptors for glutamate, the time course of current decay, single-channel conductance, voltage-dependent block by extracellular Mg2þ and sensitivity to subunitselective blockers, such as ifenprodil (Cull-Candy et al., 2001; Michaelis, 1998). The presence of an NR3 subunit decreases Mg2þ sensitivity, ion channel conductance and permeability to Ca2þ (Matsuda et al., 2002; Nishi et al., 2001; Sasaki et al., 2002). In the peripheral nervous system, NMDARs participate in sensory transmission and nociception, particularly following inflammation. In the spinal cord, glutamanergic transmission between primary afferent terminals and dorsal
0304-3959/$20.00 q 2004 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.pain.2004.02.021
444
J. Li et al. / Pain 109 (2004) 443–452
horn neurons expressing NMDA and other ionotropic glutamate receptors is involved in wind-up and central sensitization to nociceptive stimuli (Herrero et al., 2000). In addition, NMDARs on the presynaptic terminals of primary afferents regulate the release of substance P, a key mediator involved in pain signaling, from unmyelinated C-fiber terminals through autocrine stimulation by glutamate (Liu et al., 1997; Marvizon et al., 1997). NMDARs are also expressed on the peripheral terminals of primary sensory afferents innervating both somatic and visceral tissues (Carlton et al., 1995; McRoberts et al., 2001). Activation of these receptors, either by endogenously released glutamate during injury or inflammation (Omote et al., 1998), or exogenously applied glutamate or NMDA (Lawand et al., 1997; Zhou et al., 1996), causes pain behavior that can be prevented by selective inhibitors of NMDARs. Together, these results strongly suggest a role for NMDAR on both the central and peripheral terminals of primary afferents in the transmission and encoding of painful stimuli. Sensory afferent fibers have their cell bodies either in nodose or dorsal root ganglia (DRG). Studies examining subunit mRNA and protein distribution in DRG indicated that NR1 and NR2B subunits are widely expressed, with selective expression of NR2D (Marvizon et al., 2002). However, there have been only a few brief reports on the functional properties of NMDARs in primary sensory neurons (Lovinger and Weight, 1988; Si and Li, 1999). In the present study, we used [Ca2þ]i imaging and whole cell patch clamp techniques to characterize functional properties of NMDARs on adult rat DRG neurons in primary culture.
2. Methods and materials 2.1. Cell isolation and primary culture All procedures were carried out in accordance with the USPHS Guide for Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee at UCLA. Adult Sprague –Dawley rats (150 –200 g) were used in these experiments. The rats were anesthetized by sodium pentobarbital (40 mg/kg, i.p.) and lumbosacral DRGs (level T12-S1) were collected. Each DRG was minced with scissors and placed in medium consisting of 5 ml Dulbecco’s Modified Eagle’s Medium (DMEM) (Sigma, St Louis, MO) containing 0.5 mg/ml of trypsin (Sigma, type III), 1 mg/ml of collagenase (Sigma, type IA) and 0.1 mg/ml of DNAase (Sigma, type III) and kept at 37 8C for 20– 30 min with agitation. Soybean trypsin inhibitor (Sigma, type III) was added to terminate cell dissociation. The cell suspension was centrifuged for 1 min at 1000 rev./min, and the cell pellet was resuspended in DMEM supplemented with 10% fetal bovine serum, 10% horse serum, 2 mM glutamine –penicillin – streptomycin mixture, 1 mg/ml DNAase, 5 ng/ml NGF (Sigma) and 200 mM ketamine. Neurons were plated on Matrigel-coated 15 mm
glass coverslips (Collaborative Research Co., Bedford, PA) or 35 mm culture dishes and kept at 37 8C in 5% CO2 incubator. Cells were studied after 24 –72 h in culture. 2.2. Retrograde labeling of colon-specific DRG neurons For some experiments, retrograde labeling was used to identify DRG neurons that had innervated the distal colon. Briefly, rats (150 –250 g) were anesthetized with sodium pentobarbital (Nembutal, Abbott Labs; 60 mg/kg i.p.). The colon was exposed by a midline abdominal incision and cholera toxin subunit B conjugated to Alexa Fluor 488 (Molecular Probes; 1 mg/ml) was injected at 5– 7 sites (7 –10 ml per site; total 50 ml/rat) into the muscle wall using a glass Hamilton syringe fitted with a 26 gauge needle. The colon was rinsed with 0.9% sodium chloride solution and each injection site carefully swabbed to ensure no leakage of dye. The abdomen was sutured closed and the animals used for culture preparation after 5 – 14 days. 2.3. Calcium imaging [Ca2þ]i in DRG neurons was determined by ratiometric imaging of the Ca2þ-sensitive fluorescent dye, Fura-2. Cells were loaded for 30 min at 37 8C with 5 mM Fura-2 acetoxymethyl ester (Molecular Probes; Eugene, OR) in PBS – Hepes buffer (138.0 mM NaCl, 2.67 mM KCl, 0.9 mM CaCl 2, 0.5 mM MgCl 2, 8.1 mM NaH 2PO 4, 1.47 mM KH2PO4, 10.0 mM Hepes and 10.0 mM glucose) then washed in buffer to remove remaining Fura-2 ester. Coverslips with attached cells were then mounted on a recording chamber and [Ca2þ]i in DRG neurons was monitored by an Attofluor Ratio Vision Digital Fluorescence Microscopy System (Atto Instruments, Rockville, MD). The cells were perfused with modified Hank’s Balanced Salt Solution (HBSS containing 0.5 mM CaCl2 and 10 mM Hepes, pH 7.4) at a flow rate of 2 ml/min. Drugs were delivered by fast perfusion onto neurons using computer-controlled gravity-fed multibarrel perfusion system. Fura-2 was successively excited by a xenon light source at 340 and 380 nm by means of two narrow beam bandpass filters selected by a computer controlled filter wheel. The emitted fluorescence was filtered through a 520 nm filter, captured with an intensified digital camera and analyzed with Attofluor software (Atto Instruments, Rockville, MD). The concentration of Ca2þ was calculated by comparing the ratio of fluorescence at 340 and 380 nm against a standard curve of known [Ca2þ]i. Data from regions of interest were displayed in real time and logged to hard disc. Data are presented as representative traces and were quantified by determining the difference between baseline and the transient peak [Ca2þ]i in response to perfusion of NMDA and glycine. Data are presented as mean ^ SEM. All experiments were performed at room temperature (20 – 23 8C).
J. Li et al. / Pain 109 (2004) 443–452
2.4. Electrophysiological recordings
3. Results
Whole cell patch clamp recordings were performed with an Axopatch 1D amplifier, Digidata 1200 analog-to-digital converter interface, and the pClamp 8.0 software (Axon Instruments, Foster City, CA). Currents were digitized at 1 kHz, filtered at 5 kHz, and stored in a personal computer for analysis. Pipettes were pulled from borosilicate capillary tubes (Warner Instruments, Hamden, CT) with a heatpolished tip resistance of 2.5 –5 MV when filled with pipette solution containing (in mM): 120 CsF, 10 CsCl, 10 EGTA, 10 Hepes, 1 MgCl2, 0.2 CaCl2, 10 TEA, 5 Na2ATP, 0.5 Na2GTP, pH 7.2 with CsOH, , 300 mOs/kg. Bath solution contained (in mM): 138 NaCl, 5 KCl, 0.5 CaCl2, 10 Hepes, 10 glucose, with 5 mM strychnine, 0.5 mM tetrodotoxin, 1 mM diethylenetriaminepentaacetic acid, pH 7.4 with NaOH, , 300 mOs/kg. Tight seals (1 –10 GV) from visualized small- and medium-sized cells (20 – 40 mm in diameter) were obtained by applying negative pressure. The membrane was disrupted with additional suction and the whole cell configuration was obtained. Series resistances were not compensated. Cells were voltage-clamped at 2 60 mV, unless indicated. Series resistance was monitored throughout the experiments. A 3 M KCl –agar bridge was inserted between the extracellular solution and the Ag – AgCl reference electrode.
3.1. NMDARs on DRG neurons are permeable to Ca2þ
445
Expression of NMDARs on DRG neurons in primary culture was first evaluated using the Ca2þ-sensitive dye Fura-2 and single cell fluorescence imaging. Fast perfusion of DRG neurons with 250 mM NMDA and 10 mM glycine for 5 s in a Mg2þ-free extracellular solution caused a significant increase in [Ca2þ]i 96 out of 173 neurons (55.5%). A representative example is shown in Fig. 1. The [Ca2þ]i rise elicited by NMDA and glycine could be repeatedly evoked with no decrease in peak response values. Prior exposure to the non-competitive NMDAR antagonist ketamine (200 mM) completely prevented NMDA and glycine-induced [Ca2þ]i responses. Application of NMDA at concentrations of 100 and 250 mM in the presence of 10 mM glycine, produced a dose-dependent increase of [Ca2þ]i level. Peak levels of [Ca2þ]i were 244 ^ 31 nM after exposure to 100 mM NMDA (n ¼ 18 cells) and 487 ^ 29 nM after exposure to 250 mM NMDA (n ¼ 29 cells).
2.5. Drug delivery NMDA and glycine were prepared as 100 mM stock solutions. Test solutions were prepared daily by diluting stock solutions to desired concentrations in extracellular solution. Ifenprodil (100 mM stock solution dissolved in dimethyl sulfoxide (DMSO)) was added to extracellular solution to desired concentration. DMSO was added such that all solutions contained equivalent amounts of vehicle. Gravity-fed multibarrel perfusion system (SF-77B Perfusion Fast-Step, Warner Instrument Corp. Hamden, CT) was controlled by a Labmaster board using the pClamp 8.0 (Axon Instruments, Foster City, CA) software. The tip of multibarrel was positioned 200 –300 mm from the neuron under study. When a test solution was not being applied, the neuron was continuously perfused with extracellular solution. Reagents were purchased from Sigma (St Louis, MO) except for trans1-aminocyclobutane-1,3-dicarboxylic acid (trans-ACBD) which was obtained from Tocris (Ellisville, MO). 2.6. Statistical analysis Paired or unpaired Student’s t-tests were used for statistical comparisons with P , 0:05 taken as statistically significant. All dose – responses were analyzed by nonlinear regression using GraphPad Prism software (version 4, GraphPad Software, Inc., San Diego, CA). The best fit of the data to two different equations was determined by F-test comparisons.
Fig. 1. NMDARs on DRG neurons are permeable to calcium. (A) Representative traces of the increase of [Ca2þ]i induced by repeated application of 250 mM NMDA combined with 10 mM glycine for 5 s showing almost equal responses. (B) Increase of [Ca2þ]i induced by application of NMDA combined with 10 mM glycine showed dose dependence with 50% greater responses to 250 mM NMDA ðn ¼ 18Þ compared to 100 mM NMDA ðn ¼ 29Þ:
446
J. Li et al. / Pain 109 (2004) 443–452
3.2. General properties of the currents induced by 250 mM NMDA combined with 10 mM glycine Small to medium size (, 40 mm diameter) DRG neurons were voltage clamped using the whole cell configuration of the patch clamp technique and held at 2 60 mV. In a nominally Mg2þ-free extracellular solution, rapid perfusion of 250 mM NMDA combined with 10 mM glycine induced an inward current in 68.6% (118 out of 172) of the DRG neurons tested. The inward currents induced by NMDA rapidly peaked and then slowly decayed as shown in Fig. 2. The decay rate fit first order, single exponential reaction kinetics with a time constant ðtÞ of 1.69 ^ 0.11 s. The offrate was much faster with a t value of 220 ^ 11 ms. Current responses showed minor rundown with repeated NMDA application (5.5% after 15 applications at 30-s intervals). The average amplitude of the currents was 233 ^ 87 pA ðn ¼ 118Þ: Current densities for peak responses were computed by dividing by cell capacitance (average 43.2 ^ 1.6 pF, n ¼ 118), giving an average of 5.4 ^ 1.8 pF/pA. As shown in Fig. 2, the currents induced by NMDA were completely inhibited by 200 mM ketamine, a non-competitive NMDAR antagonist. Inhibition of NMDA currents by ketamine showed use-dependence indicating that it blocked the channels in the open state. In five neurons, currents were progressively decreased in the presence of ketamine by 98.7 ^ 0.4% ðP , 0:05Þ after four sequential stimulations given 30 s apart (Fig. 2, inset). The effect of ketamine was fully reversible after three successive stimulations in the absence of ketamine (92.9 ^ 7.3% of control). The current – voltage ðI – VÞ relationship of NMDAinduced currents in the Mg2þ free extracellular solution was nearly linear between 2 100 and þ 40 mV (Fig. 3).
The average reversal potential ðErev Þ for the NMDAinduced currents was 1.49 ^ 1.19 mV ðn ¼ 8Þ; a value not significantly different from zero. This value is close to the equilibrium potential for a non-specific cation channel calculated from the given extra- and intracellular cation concentrations, indicating that NMDA activated nonspecific cation channels in the DRG neurons. 3.3. Affinity of NMDARs for glutamate and glycine site agonists For channel opening to be achieved, NMDAR channels require the binding of two different types of agonist: glutamate and glycine (Kleckner and Dingledine, 1988). To test the role of both agonists and their affinity to NMDARs in DRG neurons, we examined the response of DRG neurons to a series of concentrations of each agonist. With neurons held at 2 60 mV in Mg2þ-free extracellular solution and keeping glycine constant at 10 mM, neurons were exposed to a series of ascending concentrations of NMDA ranging from 1 to 2000 mM. As shown in Fig. 4A, the inward current elicited by NMDA increased significantly as NMDA concentration increased, reaching a maximum at 1000 mM. Dose– response relationship was sigmoidal with an EC50 value for NMDA of 63.8 mM and a Hill slope coefficient of 1.6 for steady state currents. Another glutamate site antagonist with higher affinity for NMDARs, trans-ACBD (Allan et al., 1990; Heppenstall and Fleetwood-Walker, 1997), gave an EC50 value of 5.32 mM with a Hill slope coefficient of 1.3. To examine affinity of glycine for NMDARs, a series of ascending glycine concentrations ranging from 0.05 to 50 mM was perfused onto neurons in the presence of 250 mM NMDA. As shown in Fig. 5, glycine was essential for activation of NMDARs.
Fig. 2. Use-dependent block of NMDARs by ketamine. Representative traces of inward currents elicited by application of 250 mM NMDA and 10 mM glycine for 5 s at holding potential of 260 mV. Ketamine (200 mM) completely inhibited the currents in use-dependent manner, with a maximal effect after four sequential additions of NMDA and glycine at 30 intervals. Currents totally recovered after washout of ketamine. The inset shows the average (^SEM) response of five neurons during the onset of ketamine inhibition (filled bars) compared to control responses in the absence of inhibitor (clear bars).
J. Li et al. / Pain 109 (2004) 443–452
447
Fig. 3. I – V relationship of NMDA currents in a nominally Mg2þ free extracellular solution. (A) Representative traces of NMDA-induced currents at different holding potentials. Currents appeared as inward currents at negative holding potentials, and changed to outward currents at positive holding potentials. (B) Plot of amplitude of peak currents elicited by 250 mM NMDA combined with 10 mM glycine as function of holding potential. I – V relationship of NMDA-induced current was nearly linear between 2100 and þ 40 mV. The average reversal potential, Erev ; was 1.49 ^ 1.19 mV ðn ¼ 8Þ:
NMDA and glycine activated current responses were inhibited in a voltage-dependent manner by extracellular
Mg2þ. The addition of 0.1 mM Mg2þ to the bath produced little change in the outward currents but markedly reduced the inward currents as shown in Fig. 6. The inhibitory effect of Mg2þ was also concentration dependent. Concentration– current relationships were established for concentrations of Mg2þ ranging from 0 mM (nominally Mg2þ-free control solution) to 500 mM. The dose-dependent effect of Mg2þ at different holding potentials was determined by expressing the current responses as a percentage of that determined in Mg2þ-free buffer. Non-linear regression gave IC50 values of 2.05, 6.94, and 12.2 mM at membrane potentials of 2 100, 2 80, and 2 60 mV, respectively.
Fig. 4. Dose dependence of NMDA and trans-ACBD in presence of 10 mM glycine. (A) Representative traces of currents induced by different concentrations of NMDA combined with 10 mM glycine at a holding potential of 260 mV in Mg2þ-solution. (B) Plot of normalized steady state current as function of NMDA concentration compared to trans-ACBD. In both cases, the dose–response curves were sigmoidal with Hill slope coefficients of 1.62 ^ 0.26 and 1.29 ^ 0.11, and EC50 values of 63.8 mM (95% CI: 53.5–76.0 mM) and 5.32 mM (95% CI: 4.63–6.12 mM).
Fig. 5. Affinity of glycine to NMDARs in presence of 250 mM NMDA. (A) Representative traces of currents induced by different concentrations of glycine combined with 250 mM NMDA at holding potential of 260 mV in Mg2þ-free solution. (B) Plot of the normalized steady state current as function of glycine concentration. The dose–response for glycine was sigmoidal with a Hill slope coefficient of 1.29 ^ 0.16 and an EC50 of 1.86 mM (95% CI: 1.40–2.46 mM).
A threshold concentration of 0.05 mM glycine induced a very small response even if in the presence of 250 mM NMDA. As glycine concentration increased, currents were significantly increased, reaching a maximum at 10 –20 mM. Curve fitting of dose-dependent effect of glycine on NMDA-inducible currents gave an EC50 of 1.9 mM with a Hill slope coefficient of 1.3 for steady state currents. 3.4. NMDA currents are highly sensitive to Mg2þ
448
J. Li et al. / Pain 109 (2004) 443–452
Fig. 6. Voltage-dependent inhibition of NMDA-induced currents by Mg2þ in DRG neurons. (A) Representative I – V curves for DRG neurons. Current–voltage relationships induced by perfusion of 250 mM NMDA in combination with 10 mM glycine were established at 0, 100 and 500 mM Mg2þ. Values were normalized to the current at 2100 mV in nominally Mg2þ-free solution. Mg2þ significantly blocked NMDA-induced currents at holding potentials more negative than 220 mV. There was no apparent effect at positive potentials. (B) Non-linear regression of inhibitory effect of Mg2þ on NMDA currents gave sigmoidal dose–response curves with IC50 values of 2.05, 6.94, and 12.2 mM at holding potentials of 2100, 280 and 260 mV, respectively. Values are mean ^ SEM of five determinations.
3.5. Sensitivity of NMDARs to NR2B selective blocker ifenprodil The sensitivity of NMDARs in DRG neurons to NR2B selective blocker ifenprodil was determined in the presence of 250 mM NMDA and 10 mM glycine in Mg2þ-free buffer. As shown in Fig. 7, ifenprodil dose-dependently inhibited NMDA-induced currents, with approximately 57 and 72% inhibition at ifenprodil concentrations of 3 and 10 mM, respectively ðn ¼ 5Þ: For further analysis of dose-dependent inhibitory effect of ifenprodil, up to five concentrations of ifenprodil were tested on each neuron. The data from six neurons was normalized and fit to one-site or two-site inhibition curves by non-linear regression (Fig. 7). Ifenprodil
dose –response curves for NMDARs in DRG neurons fit significantly better to the one-binding site model. The IC50 was 2.6 mM, with a maximal inhibitory effect of 94 ^ 4%. 3.6. Properties of NMDAR currents in DRG neurons innervating the distal colon DRG neurons that had innervated the distal colon were identified by retrograde label of fluorescently tagged cholera toxin subunit B. Of the 16 colon-specific neurons with small to medium cell diameters (20 –40 mm) studied, 12 (75%) responded to NMDA and glycine. Stimulating with 250 mM NMDA and 10 mM glycine, the average current in 11 retrograde labeled neurons was 252 ^ 67 pA with an
Fig. 7. Inhibition of NMDA-induced currents by ifenprodil in DRG neurons. (A) Representative traces of inward currents induced by 250 mM NMDA and 10 mM glycine before addition of ifenprodil, in presence of 3 mM ifenprodil, and after washout. Currents were inhibited by 56.6% by 3 mM ifenprodil, and almost completely recovered after washout. (B) The dose dependence of the effect was evaluated by normalizing the currents as the percent control response before addition of ifenprodil (n ¼ 3 – 5 for each concentration). Non-linear regression to a single site or to a two-site model gave a significantly better fit to the one-site model (2 site vs. 1 site: P ¼ 0:2; F test ¼ 1.72). The IC50 value was 2.57 mM (95% CI: 2.04– 3.24 mM) with a maximal inhibitory effect of 93.7 ^ 1.9%. The dose-dependent effect of ifenprodil was also evaluated in DRG neurons retrograde labeled from the distal colon. The IC50 value was 2.86 mM (95% CI: 2.13– 3.83 mM), with a maximal inhibitory effect of 92.7 ^ 2.8% for labeled neurons.
J. Li et al. / Pain 109 (2004) 443–452
average current density of 6.30 ^ 1.65 pA/pF. These values were not significantly different from neurons randomly selected from unlabeled preparations (233 ^ 87 pA and 5.39 ^ 1.80 pA/pF, respectively, n ¼ 118), nor were there any difference in the average capacitance of the neurons studied (40.8 ^ 1.1 pF, n ¼ 11 vs. 43.2 ^ 1.6 pF, n ¼ 118 for colon-specific and randomly selected neurons, respectively). In addition, the current kinetics in colon-specific neurons were similar with average t values for decay rate and off-rate of 1820 ^ 250, and 220 ^ 11 ms, respectively (compared with 1690 ^ 110 and 217 ^ 12 ms, respectively). NMDAR currents in colon-specific DRG neurons were also sensitive to ifenprodil and the dose-dependent effect of ifenprodil was not different from randomly studied neurons (Fig. 7). Non-linear regression analysis of the normalized data from five labeled neurons gave an IC50 of 2.9 mM with a maximal inhibitory effect of 93 ^ 5%.
4. Discussion The present study expands on our initial results that used [Ca2þ]i imaging to demonstrate functional expression of NMDAR by adult rat DRG neurons in primary culture (McRoberts et al., 2001) by employing whole cell patch clamp techniques for electrophysiological characterization of the kinetic properties of these receptors. We demonstrate the expression of NMDARs on DRG neurons with electrophysiological properties similar to those characterized in the CNS (Nagy et al., 2003; Priestley et al., 1995; Williams, 1993). The NMDAR in DRG neurons form channels that are Ca2þ permeable, require both glutamateand glycine-binding site agonists, display use-dependent inhibition by ketamine, and are inhibited by low concentrations of extracellular Mg2þ in a strongly voltagedependent manner. In addition, the NMDAR ion channels on DRG neurons are almost completely inhibited at relatively low doses by NR2B subunit selective inhibitor, ifenprodil. Identical observations were made with DRG neurons that had been retrograde labeled from the distal colon. These results are consistent with previous results demonstrating that the majority of NMDARs expressed on the cell body of DRG neurons from adult rats contain NR2B subunits (Marvizon et al., 2002). They also suggest that, at least on the cell body, DRG neurons innervating both somatic and visceral tissues express a functionally homogeneous population of NMDARs. Ca2þ permeability was demonstrated using Fura-2 and imaging of intracellular Ca2þ in individual neurons. DRG neurons in our preparation responded to co-application of NMDA and glycine, with a significant increase of [Ca2þ]i, which could be blocked by the reversible channel blocker, ketamine. In a previous study, we showed that the glutamate site competitive antagonist, AP-5, also blocked the response, and that D -serine could substitute for glycine and was absolutely required for NMDA-stimulated [Ca2þ]i increases
449
(McRoberts et al., 2001). Whole cell voltage clamp studies of DRG neurons showed NMDA and glycine stimulated inward currents that were blocked by 200 mM ketamine. Detailed analysis of the current – voltage relationship gave a reversal potential of near zero indicating that NMDARs are non-specific cation channels, consistent with the results from other groups (Kondoh et al., 2001; Ye et al., 2001). Native NMDARs are heteromeric complexes composed of four or five subunits (Laube et al., 1998; Mori and Mishina, 1995). Seven different genes encoding subunits of the NMDAR have been cloned, namely NR1, NR2A –D and NR3A – B in the rat (Das et al., 1998; Moriyoshi et al., 1991). In transfected cells, co-expression of the NR1 subunit with any member of the NR2 family will form functional NMDARs. However, each of the four NR2 subunits form receptors that differ in several functional properties, including voltage-dependent Mg2þ sensitivity, single channel conductance, and affinity for agonists, antagonists and modulators (Cull-Candy et al., 2001; Michaelis, 1998). With respect to Mg2þ sensitivity, recombinant NMDARs containing the NR2A or NR2B subunits are characterized by high sensitivity (IC50, 2 100 mV , 2 mM), whereas channels containing NR2C or NR2D exhibit lower sensitivity to Mg2þ block (IC50, 2 100 mV , 10– 15 mM) (Momiyama et al., 1996; Wyllie et al., 1998). NMDAR channels in DRG neurons had high sensitivity to Mg2þ block (IC50, 2 100 mV ¼ 2.05 mM), consistent with receptors composed predominantly of the NR2A or NR2B subunits. To distinguish between receptors composed of NR2A and NR2B, we used the subtype selective NMDAR antagonist, ifenprodil. In recombinant and native systems, ifenprodil inhibits NMDAR channels assembled from the NR1 –NR2B subunits with high affinity (IC50 , 0.5 mM), while it is much less potent for receptors assembled from NR1 –NR2A subunits (IC50 , 150 mM) (Williams, 1993). In DRG neurons, ifenprodil inhibited nearly all of the NMDA and glycine-induced current with an IC50 of 2.64 mM. There was no indication of multiple binding sites for ifenprodil’s inhibitory effect, indicting that the receptors expressed by DRG neurons are homogenous. Compared to recombinant NMDARs composed of NR1/NR2B subunits, NMDARs in DRG neurons are 5-fold less sensitive to ifenprodil. This discrepancy could be due to differences in post-translational modification of the receptors, for example by glycosylation, or it could be due to the modulatory influence of other subunits. We have shown that both the NR1 and NR2 subunits expressed by DRG have higher apparent molecular weights than subunits in the CNS, presumably due to glycosylation (Marvizon et al., 2002). Alternatively, native NMDARs can be composed of more than one type of NR2 subunit, yielding heterologous receptor assemblies with different ligand binding affinities (Chazot and Stephenson, 1997; Chazot et al., 1994; Misra et al., 2000; Wafford et al., 1993). Since small-medium size DRG express both NR2B and NR2D
450
J. Li et al. / Pain 109 (2004) 443–452
(Marvizon et al., 2002), the shift in ifenprodil sensitivity could be due to expression of heterologous NR1/NR2B/ NR2D receptors. However, co-expression of NR2A and NR2B along with NR1 leads to expression of three types of receptors: NR1/NR2A, NR1/NR2B and NR1/NR2A/NR2B. The dose dependence of ifenprodil in these cells is best fit to two sites with different affinities: a high affinity site identical to that observed in pure NR1/NR2B receptors, and a lower affinity site (54 mM) corresponding to inhibition of NR1/ NR2A/NR2B receptors (Kew et al., 1998). Since the dosedependent effect of ifenprodil in DRG neurons was best fit to a one-site model, expression of homogeneous NMDARs composed exclusively of NR1/NR2B/NR2D subunits is possible, but unlikely. The results of the current functional study complement previous studies examining the molecular expression of NMDAR subunits in DRG neurons. Studies examining subunit protein distribution have shown that NR1 subunits are widely expressed on the primary sensory neurons both in the cell body and the periphery (Sato et al., 1993). We have previously shown that mRNA encoding NR2B and NR2D subunits are predominately expressed in DRG (Marvizon et al., 2002). Immunocytochemical localization with antibodies recognizing NR2A/2B or NR2C/2D subunits indicated distinctly different intracellular distributions. NR1/NR2B NMDARs were present in the Golgi apparatus and occasionally on the plasma membrane of both smalland large-size DRG neurons. NMDARs containing the NR1/NR2D subunits were not co-localized with NR1/NR2B and were instead found throughout the cytoplasm, but only in small diameter neurons. In the present study, we focused on small to medium size (20 – 40 mm in diameter) DRG neurons that are likely to express both NR2B and NR2D. That we only observed functional expression of an apparently homogeneous NMDAR population that was sensitive to ifenprodil, is consistent with the immunocytochemical data showing cell surface expression of NR2B, but not NR2D. Our observations are also consistent with a previous report that found predominant expression of NR2B subunits on small-diameter primary afferent dorsal root ganglion neurons that give rise to the sciatic nerve (Ma and Hargreaves, 2000). We also characterized NMDARs on DRG neurons labeled retrogradely from the distal colon. We found that there was no difference in the ifenprodil sensitivity suggesting that NMDARs expressed in neurons innervating visceral tissues are not different from those innervating somatic tissues. NMDAR current density tended to be slightly larger in labeled neurons compared to randomly selected neurons, but the difference was not significant. Using Ca2þ imaging, we previously had shown that colon-selective neurons responded to co-application of NMDA and D -serine with a significantly larger increase in intracellular Ca2þ than randomly selected neurons (McRoberts et al., 2001). The discrepancy between this and the present study could be due to
differences in other parameters that affect neuronal excitability, Ca2þ permeability, or Ca2þ mobilization. Small to medium size DRG neurons generate fine myelinated Ad- and unmyelinated C-fibers that carry nociceptive information from the periphery to the spinal cord (Mense, 1990). In the spinal cord, NMDARs on dorsal horn neurons participate in central sensitization, as evidenced by their role in ‘wind-up’ of central responses to nociceptive stimuli (Davies and Lodge, 1987; Dickenson and Sullivan, 1987), and in long-term potentiation of synapses between primary afferent neurons and dorsal horn neurons (Liu and Sandku¨hler, 1995; Randic et al., 1993). NMDARs are also present on presynaptic terminals of primary afferents in the dorsal horn where they modulate the release of substance P (Liu et al., 1997; Marvizon et al., 1997), an important pro-inflammatory peptide mediating chronic and severe pain (Allen et al., 1997; Cao et al., 1998). In addition to expression of NMDARs on central terminals of DRG neurons, peripheral terminals of primary afferent nerves innervating visceral and somatic tissues also express NMDARs (Carlton et al., 1998; McRoberts et al., 2001). Pharmacological activation of peripheral NMDARs produces nociceptive behavior (Carlton et al., 1995; Lawand et al., 1997; Zhou et al., 1996), and peripheral injection of NMDA-R antagonists attenuates pain associated with neuropathic or inflammatory conditions (Davidson and Carlton, 1998; Warncke et al., 1997). Thus, NMDARs on both terminals of DRG neurons participate in pain responses. Our study demonstrates predominant functional expression of NR2B-containing NMDARs on the soma of DRG neurons. Assuming that these receptors are exported to the terminals of DRG neurons, the results support application of NR2B-selective receptor antagonists for pain management. Ifenprodil and related compounds are anti-nociceptive in a variety of preclinical pain models and have a much lower side-effect profile compared with other NMDAR antagonists (Chizh et al., 2001; Gurwitz and Weizman, 2002; Malmberg et al., 2003). Drugs targeting NR2B-containing NMDARs in pain-relevant structures, such as peripheral or central terminal of DRG neurons, could serve as a new class of medicine for managing chronic pain. Acknowledgements This study was supported by NIH grants R01 DK58173 and P50 DK64539-01 (EAM). We thank Teresa Olivas for her editorial contributions.
References Allan RD, Hanrahan JR, Hambley TW, Johnston GA, Mewett KN, Mitrovic AD. Synthesis and activity of a potent N-methyl-D -aspartic acid agonist,
J. Li et al. / Pain 109 (2004) 443–452 trans-1-aminocyclobutane-1,3-dicarboxylic acid, and related phosphonic and carboxylic acids. J Med Chem 1990;33:2905 –15. Allen BJ, Rogers SD, Ghilardi JR, Menning PM, Kuskowski MA, Basbaum AI, Simone DA, Mantyh PW. Noxious cutaneous thermal stimuli induce a graded release of endogenous substance P in the spinal cord: imaging peptide action in vivo. J Neurosci 1997;17:5921 –7. Cao YQ, Mantyh PW, Carlson EJ, Gillespie AM, Epstein CH, Basbaum AI. Primary afferent tachykinins are required to experience moderate to intense pain. Nature 1998;392:390–4. Carlton SM, Hargett GL, Coggeshall RE. Localization and activation of glutamate receptors in unmyelinated axons of rat glabrous skin. Neurosci Lett 1995;197:25– 8. Carlton SM, Zhou ST, Coggeshall RE. Evidence for the interaction of glutamate and NK1 receptors in the periphery. Brain Res 1998;790: 160–9. Chazot PL, Stephenson FA. Molecular dissection of native mammalian forebrain NMDA receptors containing the NR1 C2 exon: direct demonstration of NMDA receptors comprising NR1, NR2A, and NR2B subunits within the same complex. J Neurochem 1997;69: 2138–44. Chazot PL, Coleman SK, Cik M, Stephenson FA. Molecular characterization of N-methyl-D -aspartate receptors expressed in mammalian cells yields evidence for the coexistence of three subunit types within a discrete receptor molecule. J Biol Chem 1994;269:24403–9. Chizh BA, Headley PM, Tzschentke TM. NMDA receptor antagonists as analgesics: focus on the NR2B subtype. Trends Pharm Sci 2001;22: 636–42. Cull-Candy S, Brickley S, Farrant M. NMDA receptor subunits: diversity, development and disease. Curr Opin Neurobiol 2001;11:327 –35. Das S, Sasaki YF, Rothe T, Premkumar LS, Takasu M, Crandalli JE, Dikkes P, Conner DA, Rayudu PV, Cheung W, Vincent C, Lipton SA, Nakanish N. Increased NMDA current and spine density in mice lacking the NMDA receptor subunit NR3A. Nature 1998;393:377–81. Davidson EM, Carlton SM. Intraplantar injection of dextrorphan, ketamine or memantine attenuates formalin-induced behaviors. Brain Res 1998; 785:136–42. Davies SN, Lodge D. Evidence for involvement of N-methylaspartate receptors in wind-up of class 2 neurones in the dorsal horn of the rat. Brain Res 1987;424:402 –6. Dickenson AH, Sullivan AF. Evidence for a role of the NMDA receptor in the frequency dependent potentiation of deep rat dorsal horn nociceptive neurones following C fibre stimulation. Neuropharmacology 1987;26:1235– 8. Dingledine R, Borges K, Bowie D, Traynelis SF. The glutamate receptor ion channels. Pharmacol Rev 1999;51:7–62. Gurwitz D, Weizman A. The NR2B subunit of glutamate receptors as a potential target for relieving chronic pain: prospects and concerns. Drug Discov Today 2002;7:403–6. Hawkins LM, Chazot PL, Stephenson FA. Biochemical evidence for the coassociation of three N-methyl-D -aspartate (NMDA) R2 subunits in recombinant NMDA receptors. J Biol Chem 1999;274:27211–8. Heppenstall PA, Fleetwood-Walker SM. The glycine site of the NMDA receptor contributes to neurokinin1 receptor agonist facilitation of NMDA receptor agonist-evoked activity in rat dorsal horn neurons. Brain Res 1997;744:235 –45. Herrero JF, Laird JM, Lopez-Garcia JA. Wind-up of spinal cord neurones and pain sensation: much ado about something? Prog Neurobiol 2000; 61:169–203. Kew JN, Richards JG, Mutel V, Kemp JA. Developmental changes in NMDA receptor glycine affinity and ifenprodil sensitivity reveal three distinct populations of NMDA receptors in individual rat cortical neurons. J Neurosci 1998;18:1935–43. Kleckner NW, Dingledine R. Requirement for glycine in activation of NMDA-receptors expressed in Xenopus oocytes. Science 1988;241: 835–7.
451
Kondoh T, Nishizaki T, Aihara H, Tamaki N. NMDA-responsible, APVinsensitive receptor in cultured human astrocytes. Life Sci 2001;68: 1761–7. Laube B, Kuhse J, Betz H. Evidence for a tetrameric structure of recombinant NMDA receptors. J Neurosci 1998;18:2954–61. Lawand NB, Willis WD, Westlund KN. Excitatory amino acid receptor involvement in peripheral nociceptive transmission in rats. Eur J Pharmacol 1997;324:169 –77. Liu X-G, Sandku¨hler J. Long-term potentiation of C-fiber-evoked potentials in the rat spinal dorsal horn is prevented by spinal N-methyl-D -aspartic acid receptor blockage. Neurosci Lett 1995;191: 43– 6. Liu H, Mantyh PW, Basbaum AI. NMDA-receptor regulation of substance P release from primary afferent nociceptors. Nature 1997;386:721–4. Lovinger DM, Weight FF. Glutamate induces a depolarization of adult rat dorsal root ganglion neurons that is mediated predominantly by NMDA receptors. Neurosci Lett 1988;94:314– 20. Ma Q-P, Hargreaves RJ. Localization of N-methyl-aspartate NR2B subunits on primary sensory neurons that give rise to small-caliber sciatic nerve fibers in rats. Neuroscience 2000;101:699–707. Malmberg AB, Gilbert H, McCabe RT, Basbaum AI. Powerful antinociceptive effects of the cone snail venom-derived subtype-selective NMDA receptor antagonists conantokins G and T. Pain 2003;101: 109– 16. Marvizon JC, Martinez V, Grady EF, Bunnett NW, Mayer EA. Neurokinin 1 receptor internalization in spinal cord slices induced by dorsal root stimulation is mediated by NMDA receptors. J Neurosci 1997;17: 8129–36. Marvizon JC, McRoberts JA, Ennes HS, Song B, Wang X, Jinton L, Corneliussen B, Mayer EA. Two N-methyl-D -aspartate receptors in rat dorsal root ganglia with different subunit composition and localization. J Comp Neurol 2002;446:325 –41. Matsuda K, Kamiya Y, Matsuda S, Yuzaki M. Cloning and characterization of a novel NMDA receptor subunit NR3B: a dominant subunit that reduces calcium permeability. Mol Brain Res 2002;100:43–52. McRoberts JA, Coutinho SV, Marvizon JC, Grady EF, Tognetto M, Sengupta JN, Ennes HS, Chaban VV, Amadesi S, Creminon C, Lanthorn T, Geppetti P, Bunnett NW, Mayer EA. Role of peripheral N-methyl-D -aspartate (NMDA) receptors in visceral nociception in rats. Gastroenterology 2001;120:1737–48. Mense S. Structure–function relationships in identified afferent neurones. Anat Embryol (Berl) 1990;181:1–17. Michaelis EK. Molecular biology of glutamate receptors in the central nervous system and their role in excitotoxicity, oxidative stress and aging. Prog Neurobiol 1998;54:369 –415. Misra C, Brickley SG, Wyllie DJA, Cull-Candy SG. Slow deactivation kinetics of NMDA receptors containing NR1 and NR2D subunits in rat cerebellar Purkinje cells. J Physiol 2000;525:299–305. Momiyama A, Feldmeyer D, Cull-Candy SG. Identification of a native lowconductance NMDA channel with reduced sensitivity to Mg in rat central neurones. J Physiol 1996;494:479–92. Mori H, Mishina M. Structure and function of the NMDA receptor channel. Neuropharmacology 1995;34:1219–37. Moriyoshi K, Masu M, Ishii T, Shigemoto R, Mizuno N, Nakanishi S. Molecular cloning and characterization of the rat NMDA receptor. Nature 1991;353:31 –7. Nagy J, Kolok S, Dezso P, Boros A, Szombathelyi Z. Differential alterations in the expression of NMDA receptor subunits following chronic ethanol treatment in primary cultures of rat cortical and hippocampal neurones. Neurochem Int 2003;42:35–43. Nishi M, Hinds H, Lu HP, Kawata M, Hayashi Y. Motoneuron-specific expression of NR3B, a novel NMDA-type glutamate receptor subunit that works in a dominant-negative manner. J Neurosci 2001;21:185RC. Omote K, Kawamata T, Kawamata M, Namiki A. Formalin-induced release of excitatory amino acids in the skin of the rat hindpaw. Brain Res 1998; 787:161–4.
452
J. Li et al. / Pain 109 (2004) 443–452
Premkumar LS, Auerbach A. Stoichiometry of recombinant N-methyl-D aspartate receptor channels inferred from single-channel current patterns. J Gen Physiol 1997;110:485–502. Priestley T, Laughton P, Myers J, Le Bourdelles B, Kerby J, Whiting PJ. Pharmacological properties of recombinant human N-methyl-D -aspartate receptors comprising NR1a/NR2A and NR1a/NR2B subunit assemblies expressed in permanently transfected mouse fibroblast cells. Mol Pharmacol 1995;48:841–8. Randic M, Jiang MC, Cerne R. Long-term potentiation and long-term depression of primary afferent neurotransmission in the rat spinal cord. J Neurosci 1993;13:5228–41. Sasaki YF, Rothe T, Premkumar LS, Das S, Cui J, Talantova MV, Wong HK, Gong X, Chan SF, Zhang D, Nakanishi N, Sucher NJ, Lipton SA. Characterization and comparison of the NR3A subunit of the NMDA receptor in recombinant systems and primary cortical neurons. J Neurophysiol 2002;87:2052–63. Sato K, Kiyama H, Park HT, Tohyama M. AMPA, KA, and NMDA receptors are expressed in the rat DRG neurones. NeuroReport 1993;4: 1263–5. Schorge S, Colquhoun D. Studies of NMDA receptor function and stoichiometry with truncated and tandem subunits. J Neurosci 2003;23: 1151–8.
Si JQ, Li ZW. Inhibition by baclofen of NMDA-activated current in rat dorsal root ganglion neurons. Zhongguo Yao Li Xue Bao 1999;20: 324 –8. Wafford KA, Bain CJ, Le Bourdelles B, Whiting PJ, Kemp JA. Preferential co-assembly of recombinant NMDA receptors composed of three different subunits. NeuroReport 1993;4:1347– 9. Warncke T, Jorum E, Stubhaug A. Local treatment with the N-methylaspartate receptor antagonist ketamine, inhibit development of secondary hyperalgesia in man by a peripheral action. Neurosci Lett 1997;227:1–4. Williams K. Ifenprodil discriminates subtypes of the N-methyl-D -aspartate receptor: selectivity and mechanisms at recombinant heteromeric receptors. Mol Pharmacol 1993;44:851–9. Wyllie DJ, Behe P, Colquhoun D. Single-channel activations and concentration jumps: comparison of recombinant NR1a/NR2A and NR1a/NR2D NMDA receptors. J Physiol 1998;510:1–18. Yamakura T, Shimoji K. Subunit- and site-specific pharmacology of the NMDA receptor channel. Prog Neurobiol 1999;59:279 –98. Ye JH, Tao L, Zalcman SS. Interleukin-2 modulates N-methyl-aspartate receptors of native mesolimbic neurons. Brain Res 2001;894:241–8. Zhou S, Bonasera L, Carlton SM. Peripheral administration of NMDA, AMPA or KA results in pain behaviors in rats. NeuroReport 1996;7: 895 –900.
Electrophysiological characterization of N-methyl-D -aspartate receptors in rat dorsal root ganglia neurons Jichang Li, James A. McRoberts*, Jingjiang Nie, Helena S. Ennes, Emeran A. Mayer Division of Digestive Diseases, Department of Medicine, Center for Neurovisceral Sciences (CNS) and Women’s Health, University of California, Warren Hall, Room 14-103, 900 Veterans Avenue, Los Angeles, CA 90095, USA Received 11 October 2003; received in revised form 29 January 2004; accepted 23 February 2004
Abstract In the peripheral nervous system, N-methyl-D -aspartate receptors (NMDAR) expressed on the central and peripheral terminals of primary afferent neurons are involved in nociception. We used single cell imaging of intracellular calcium concentration ([Ca2þ]i) and patch clamp techniques to characterize the functional properties of NMDARs on adult rat dorsal root ganglia (DRG) neurons in primary culture and selectively on those innervating the distal colon. In Mg2þ-free extracellular solution, rapid perfusion of DRG neurons with 250 mM NMDA and 10 mM glycine caused a significant increase in [Ca2þ]i, and elicited inward currents in whole cell patch clamp recordings when the holding potential was 2 60 mV. Both effects were reversibly inhibited by 200 mM ketamine in a use-dependent manner. The EC50 values for NMDA and glycine were 64 and 1.9 mM with Hill slope coefficients of 1.4 and 1.3, respectively. At negative potentials, extracellular Mg2þ blocked currents in a concentration- and voltage-dependent manner. The IC50 for Mg2þ at a holding potential of 2100 mV was 2.0 mM. The NMDAR subtype-selective antagonist, ifenprodil, inhibited 94% of the NMDA and glycine-induced current with an IC50 of 2.6 mM. There was no evidence of multiple binding sites for ifenprodil. There was no significant difference in the NMDAR current density on DRG neurons that had innervated the colon, nor was there a difference in the EC50 for ifenprodil. These results demonstrate that functional NMDARs expressed by DRG neurons innervating both somatic and visceral tissues of adult rats are composed predominantly of NR2B subunits. q 2004 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. Keywords: Calcium imaging; Glutamate receptor channel; Glutamate receptor subtype; NMDA receptor; NMDA receptor antagonist; NMDA receptor channel; Primary afferent neuron; Whole cell patch clamp
1. Introduction N-methyl-D -aspartate receptors (NMDAR) are ionotropic glutamate receptors that require glycine as a co-agonist, display voltage-dependent inhibition by extracellular Mg2þ, and have high permeability to Ca2þ. These properties contribute to the unique role of NMDARs in neuronal plasticity (Dingledine et al., 1999; Yamakura and Shimoji, 1999). Molecular biological studies have revealed three NMDAR subunit families: NR1, NR2 and NR3. The NR1 family has one member with eight different splice variants, the NR2 family has four members derived from different genes, denoted NR2A – D, and the NR3 family has two members NR3A – B (Cull-Candy et al., 2001; Michaelis, 1998). All NMDARs appear to function as heteromeric assemblies composed of two NR1 subunits in combination * Corresponding author. Tel.: þ 1-310-825-4786; fax: þ1-310-794-2864. E-mail address: [email protected] (J.A. McRoberts).
with two or three NR2 and/or NR3 subunits (Hawkins et al., 1999; Premkumar and Auerbach, 1997; Schorge and Colquhoun, 2003). Studies with recombinant receptors have provided an understanding of how receptor properties are defined by individual NMDAR subunits. NR2 subunits influence the affinity of the receptors for glutamate, the time course of current decay, single-channel conductance, voltage-dependent block by extracellular Mg2þ and sensitivity to subunitselective blockers, such as ifenprodil (Cull-Candy et al., 2001; Michaelis, 1998). The presence of an NR3 subunit decreases Mg2þ sensitivity, ion channel conductance and permeability to Ca2þ (Matsuda et al., 2002; Nishi et al., 2001; Sasaki et al., 2002). In the peripheral nervous system, NMDARs participate in sensory transmission and nociception, particularly following inflammation. In the spinal cord, glutamanergic transmission between primary afferent terminals and dorsal
0304-3959/$20.00 q 2004 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.pain.2004.02.021
444
J. Li et al. / Pain 109 (2004) 443–452
horn neurons expressing NMDA and other ionotropic glutamate receptors is involved in wind-up and central sensitization to nociceptive stimuli (Herrero et al., 2000). In addition, NMDARs on the presynaptic terminals of primary afferents regulate the release of substance P, a key mediator involved in pain signaling, from unmyelinated C-fiber terminals through autocrine stimulation by glutamate (Liu et al., 1997; Marvizon et al., 1997). NMDARs are also expressed on the peripheral terminals of primary sensory afferents innervating both somatic and visceral tissues (Carlton et al., 1995; McRoberts et al., 2001). Activation of these receptors, either by endogenously released glutamate during injury or inflammation (Omote et al., 1998), or exogenously applied glutamate or NMDA (Lawand et al., 1997; Zhou et al., 1996), causes pain behavior that can be prevented by selective inhibitors of NMDARs. Together, these results strongly suggest a role for NMDAR on both the central and peripheral terminals of primary afferents in the transmission and encoding of painful stimuli. Sensory afferent fibers have their cell bodies either in nodose or dorsal root ganglia (DRG). Studies examining subunit mRNA and protein distribution in DRG indicated that NR1 and NR2B subunits are widely expressed, with selective expression of NR2D (Marvizon et al., 2002). However, there have been only a few brief reports on the functional properties of NMDARs in primary sensory neurons (Lovinger and Weight, 1988; Si and Li, 1999). In the present study, we used [Ca2þ]i imaging and whole cell patch clamp techniques to characterize functional properties of NMDARs on adult rat DRG neurons in primary culture.
2. Methods and materials 2.1. Cell isolation and primary culture All procedures were carried out in accordance with the USPHS Guide for Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee at UCLA. Adult Sprague –Dawley rats (150 –200 g) were used in these experiments. The rats were anesthetized by sodium pentobarbital (40 mg/kg, i.p.) and lumbosacral DRGs (level T12-S1) were collected. Each DRG was minced with scissors and placed in medium consisting of 5 ml Dulbecco’s Modified Eagle’s Medium (DMEM) (Sigma, St Louis, MO) containing 0.5 mg/ml of trypsin (Sigma, type III), 1 mg/ml of collagenase (Sigma, type IA) and 0.1 mg/ml of DNAase (Sigma, type III) and kept at 37 8C for 20– 30 min with agitation. Soybean trypsin inhibitor (Sigma, type III) was added to terminate cell dissociation. The cell suspension was centrifuged for 1 min at 1000 rev./min, and the cell pellet was resuspended in DMEM supplemented with 10% fetal bovine serum, 10% horse serum, 2 mM glutamine –penicillin – streptomycin mixture, 1 mg/ml DNAase, 5 ng/ml NGF (Sigma) and 200 mM ketamine. Neurons were plated on Matrigel-coated 15 mm
glass coverslips (Collaborative Research Co., Bedford, PA) or 35 mm culture dishes and kept at 37 8C in 5% CO2 incubator. Cells were studied after 24 –72 h in culture. 2.2. Retrograde labeling of colon-specific DRG neurons For some experiments, retrograde labeling was used to identify DRG neurons that had innervated the distal colon. Briefly, rats (150 –250 g) were anesthetized with sodium pentobarbital (Nembutal, Abbott Labs; 60 mg/kg i.p.). The colon was exposed by a midline abdominal incision and cholera toxin subunit B conjugated to Alexa Fluor 488 (Molecular Probes; 1 mg/ml) was injected at 5– 7 sites (7 –10 ml per site; total 50 ml/rat) into the muscle wall using a glass Hamilton syringe fitted with a 26 gauge needle. The colon was rinsed with 0.9% sodium chloride solution and each injection site carefully swabbed to ensure no leakage of dye. The abdomen was sutured closed and the animals used for culture preparation after 5 – 14 days. 2.3. Calcium imaging [Ca2þ]i in DRG neurons was determined by ratiometric imaging of the Ca2þ-sensitive fluorescent dye, Fura-2. Cells were loaded for 30 min at 37 8C with 5 mM Fura-2 acetoxymethyl ester (Molecular Probes; Eugene, OR) in PBS – Hepes buffer (138.0 mM NaCl, 2.67 mM KCl, 0.9 mM CaCl 2, 0.5 mM MgCl 2, 8.1 mM NaH 2PO 4, 1.47 mM KH2PO4, 10.0 mM Hepes and 10.0 mM glucose) then washed in buffer to remove remaining Fura-2 ester. Coverslips with attached cells were then mounted on a recording chamber and [Ca2þ]i in DRG neurons was monitored by an Attofluor Ratio Vision Digital Fluorescence Microscopy System (Atto Instruments, Rockville, MD). The cells were perfused with modified Hank’s Balanced Salt Solution (HBSS containing 0.5 mM CaCl2 and 10 mM Hepes, pH 7.4) at a flow rate of 2 ml/min. Drugs were delivered by fast perfusion onto neurons using computer-controlled gravity-fed multibarrel perfusion system. Fura-2 was successively excited by a xenon light source at 340 and 380 nm by means of two narrow beam bandpass filters selected by a computer controlled filter wheel. The emitted fluorescence was filtered through a 520 nm filter, captured with an intensified digital camera and analyzed with Attofluor software (Atto Instruments, Rockville, MD). The concentration of Ca2þ was calculated by comparing the ratio of fluorescence at 340 and 380 nm against a standard curve of known [Ca2þ]i. Data from regions of interest were displayed in real time and logged to hard disc. Data are presented as representative traces and were quantified by determining the difference between baseline and the transient peak [Ca2þ]i in response to perfusion of NMDA and glycine. Data are presented as mean ^ SEM. All experiments were performed at room temperature (20 – 23 8C).
J. Li et al. / Pain 109 (2004) 443–452
2.4. Electrophysiological recordings
3. Results
Whole cell patch clamp recordings were performed with an Axopatch 1D amplifier, Digidata 1200 analog-to-digital converter interface, and the pClamp 8.0 software (Axon Instruments, Foster City, CA). Currents were digitized at 1 kHz, filtered at 5 kHz, and stored in a personal computer for analysis. Pipettes were pulled from borosilicate capillary tubes (Warner Instruments, Hamden, CT) with a heatpolished tip resistance of 2.5 –5 MV when filled with pipette solution containing (in mM): 120 CsF, 10 CsCl, 10 EGTA, 10 Hepes, 1 MgCl2, 0.2 CaCl2, 10 TEA, 5 Na2ATP, 0.5 Na2GTP, pH 7.2 with CsOH, , 300 mOs/kg. Bath solution contained (in mM): 138 NaCl, 5 KCl, 0.5 CaCl2, 10 Hepes, 10 glucose, with 5 mM strychnine, 0.5 mM tetrodotoxin, 1 mM diethylenetriaminepentaacetic acid, pH 7.4 with NaOH, , 300 mOs/kg. Tight seals (1 –10 GV) from visualized small- and medium-sized cells (20 – 40 mm in diameter) were obtained by applying negative pressure. The membrane was disrupted with additional suction and the whole cell configuration was obtained. Series resistances were not compensated. Cells were voltage-clamped at 2 60 mV, unless indicated. Series resistance was monitored throughout the experiments. A 3 M KCl –agar bridge was inserted between the extracellular solution and the Ag – AgCl reference electrode.
3.1. NMDARs on DRG neurons are permeable to Ca2þ
445
Expression of NMDARs on DRG neurons in primary culture was first evaluated using the Ca2þ-sensitive dye Fura-2 and single cell fluorescence imaging. Fast perfusion of DRG neurons with 250 mM NMDA and 10 mM glycine for 5 s in a Mg2þ-free extracellular solution caused a significant increase in [Ca2þ]i 96 out of 173 neurons (55.5%). A representative example is shown in Fig. 1. The [Ca2þ]i rise elicited by NMDA and glycine could be repeatedly evoked with no decrease in peak response values. Prior exposure to the non-competitive NMDAR antagonist ketamine (200 mM) completely prevented NMDA and glycine-induced [Ca2þ]i responses. Application of NMDA at concentrations of 100 and 250 mM in the presence of 10 mM glycine, produced a dose-dependent increase of [Ca2þ]i level. Peak levels of [Ca2þ]i were 244 ^ 31 nM after exposure to 100 mM NMDA (n ¼ 18 cells) and 487 ^ 29 nM after exposure to 250 mM NMDA (n ¼ 29 cells).
2.5. Drug delivery NMDA and glycine were prepared as 100 mM stock solutions. Test solutions were prepared daily by diluting stock solutions to desired concentrations in extracellular solution. Ifenprodil (100 mM stock solution dissolved in dimethyl sulfoxide (DMSO)) was added to extracellular solution to desired concentration. DMSO was added such that all solutions contained equivalent amounts of vehicle. Gravity-fed multibarrel perfusion system (SF-77B Perfusion Fast-Step, Warner Instrument Corp. Hamden, CT) was controlled by a Labmaster board using the pClamp 8.0 (Axon Instruments, Foster City, CA) software. The tip of multibarrel was positioned 200 –300 mm from the neuron under study. When a test solution was not being applied, the neuron was continuously perfused with extracellular solution. Reagents were purchased from Sigma (St Louis, MO) except for trans1-aminocyclobutane-1,3-dicarboxylic acid (trans-ACBD) which was obtained from Tocris (Ellisville, MO). 2.6. Statistical analysis Paired or unpaired Student’s t-tests were used for statistical comparisons with P , 0:05 taken as statistically significant. All dose – responses were analyzed by nonlinear regression using GraphPad Prism software (version 4, GraphPad Software, Inc., San Diego, CA). The best fit of the data to two different equations was determined by F-test comparisons.
Fig. 1. NMDARs on DRG neurons are permeable to calcium. (A) Representative traces of the increase of [Ca2þ]i induced by repeated application of 250 mM NMDA combined with 10 mM glycine for 5 s showing almost equal responses. (B) Increase of [Ca2þ]i induced by application of NMDA combined with 10 mM glycine showed dose dependence with 50% greater responses to 250 mM NMDA ðn ¼ 18Þ compared to 100 mM NMDA ðn ¼ 29Þ:
446
J. Li et al. / Pain 109 (2004) 443–452
3.2. General properties of the currents induced by 250 mM NMDA combined with 10 mM glycine Small to medium size (, 40 mm diameter) DRG neurons were voltage clamped using the whole cell configuration of the patch clamp technique and held at 2 60 mV. In a nominally Mg2þ-free extracellular solution, rapid perfusion of 250 mM NMDA combined with 10 mM glycine induced an inward current in 68.6% (118 out of 172) of the DRG neurons tested. The inward currents induced by NMDA rapidly peaked and then slowly decayed as shown in Fig. 2. The decay rate fit first order, single exponential reaction kinetics with a time constant ðtÞ of 1.69 ^ 0.11 s. The offrate was much faster with a t value of 220 ^ 11 ms. Current responses showed minor rundown with repeated NMDA application (5.5% after 15 applications at 30-s intervals). The average amplitude of the currents was 233 ^ 87 pA ðn ¼ 118Þ: Current densities for peak responses were computed by dividing by cell capacitance (average 43.2 ^ 1.6 pF, n ¼ 118), giving an average of 5.4 ^ 1.8 pF/pA. As shown in Fig. 2, the currents induced by NMDA were completely inhibited by 200 mM ketamine, a non-competitive NMDAR antagonist. Inhibition of NMDA currents by ketamine showed use-dependence indicating that it blocked the channels in the open state. In five neurons, currents were progressively decreased in the presence of ketamine by 98.7 ^ 0.4% ðP , 0:05Þ after four sequential stimulations given 30 s apart (Fig. 2, inset). The effect of ketamine was fully reversible after three successive stimulations in the absence of ketamine (92.9 ^ 7.3% of control). The current – voltage ðI – VÞ relationship of NMDAinduced currents in the Mg2þ free extracellular solution was nearly linear between 2 100 and þ 40 mV (Fig. 3).
The average reversal potential ðErev Þ for the NMDAinduced currents was 1.49 ^ 1.19 mV ðn ¼ 8Þ; a value not significantly different from zero. This value is close to the equilibrium potential for a non-specific cation channel calculated from the given extra- and intracellular cation concentrations, indicating that NMDA activated nonspecific cation channels in the DRG neurons. 3.3. Affinity of NMDARs for glutamate and glycine site agonists For channel opening to be achieved, NMDAR channels require the binding of two different types of agonist: glutamate and glycine (Kleckner and Dingledine, 1988). To test the role of both agonists and their affinity to NMDARs in DRG neurons, we examined the response of DRG neurons to a series of concentrations of each agonist. With neurons held at 2 60 mV in Mg2þ-free extracellular solution and keeping glycine constant at 10 mM, neurons were exposed to a series of ascending concentrations of NMDA ranging from 1 to 2000 mM. As shown in Fig. 4A, the inward current elicited by NMDA increased significantly as NMDA concentration increased, reaching a maximum at 1000 mM. Dose– response relationship was sigmoidal with an EC50 value for NMDA of 63.8 mM and a Hill slope coefficient of 1.6 for steady state currents. Another glutamate site antagonist with higher affinity for NMDARs, trans-ACBD (Allan et al., 1990; Heppenstall and Fleetwood-Walker, 1997), gave an EC50 value of 5.32 mM with a Hill slope coefficient of 1.3. To examine affinity of glycine for NMDARs, a series of ascending glycine concentrations ranging from 0.05 to 50 mM was perfused onto neurons in the presence of 250 mM NMDA. As shown in Fig. 5, glycine was essential for activation of NMDARs.
Fig. 2. Use-dependent block of NMDARs by ketamine. Representative traces of inward currents elicited by application of 250 mM NMDA and 10 mM glycine for 5 s at holding potential of 260 mV. Ketamine (200 mM) completely inhibited the currents in use-dependent manner, with a maximal effect after four sequential additions of NMDA and glycine at 30 intervals. Currents totally recovered after washout of ketamine. The inset shows the average (^SEM) response of five neurons during the onset of ketamine inhibition (filled bars) compared to control responses in the absence of inhibitor (clear bars).
J. Li et al. / Pain 109 (2004) 443–452
447
Fig. 3. I – V relationship of NMDA currents in a nominally Mg2þ free extracellular solution. (A) Representative traces of NMDA-induced currents at different holding potentials. Currents appeared as inward currents at negative holding potentials, and changed to outward currents at positive holding potentials. (B) Plot of amplitude of peak currents elicited by 250 mM NMDA combined with 10 mM glycine as function of holding potential. I – V relationship of NMDA-induced current was nearly linear between 2100 and þ 40 mV. The average reversal potential, Erev ; was 1.49 ^ 1.19 mV ðn ¼ 8Þ:
NMDA and glycine activated current responses were inhibited in a voltage-dependent manner by extracellular
Mg2þ. The addition of 0.1 mM Mg2þ to the bath produced little change in the outward currents but markedly reduced the inward currents as shown in Fig. 6. The inhibitory effect of Mg2þ was also concentration dependent. Concentration– current relationships were established for concentrations of Mg2þ ranging from 0 mM (nominally Mg2þ-free control solution) to 500 mM. The dose-dependent effect of Mg2þ at different holding potentials was determined by expressing the current responses as a percentage of that determined in Mg2þ-free buffer. Non-linear regression gave IC50 values of 2.05, 6.94, and 12.2 mM at membrane potentials of 2 100, 2 80, and 2 60 mV, respectively.
Fig. 4. Dose dependence of NMDA and trans-ACBD in presence of 10 mM glycine. (A) Representative traces of currents induced by different concentrations of NMDA combined with 10 mM glycine at a holding potential of 260 mV in Mg2þ-solution. (B) Plot of normalized steady state current as function of NMDA concentration compared to trans-ACBD. In both cases, the dose–response curves were sigmoidal with Hill slope coefficients of 1.62 ^ 0.26 and 1.29 ^ 0.11, and EC50 values of 63.8 mM (95% CI: 53.5–76.0 mM) and 5.32 mM (95% CI: 4.63–6.12 mM).
Fig. 5. Affinity of glycine to NMDARs in presence of 250 mM NMDA. (A) Representative traces of currents induced by different concentrations of glycine combined with 250 mM NMDA at holding potential of 260 mV in Mg2þ-free solution. (B) Plot of the normalized steady state current as function of glycine concentration. The dose–response for glycine was sigmoidal with a Hill slope coefficient of 1.29 ^ 0.16 and an EC50 of 1.86 mM (95% CI: 1.40–2.46 mM).
A threshold concentration of 0.05 mM glycine induced a very small response even if in the presence of 250 mM NMDA. As glycine concentration increased, currents were significantly increased, reaching a maximum at 10 –20 mM. Curve fitting of dose-dependent effect of glycine on NMDA-inducible currents gave an EC50 of 1.9 mM with a Hill slope coefficient of 1.3 for steady state currents. 3.4. NMDA currents are highly sensitive to Mg2þ
448
J. Li et al. / Pain 109 (2004) 443–452
Fig. 6. Voltage-dependent inhibition of NMDA-induced currents by Mg2þ in DRG neurons. (A) Representative I – V curves for DRG neurons. Current–voltage relationships induced by perfusion of 250 mM NMDA in combination with 10 mM glycine were established at 0, 100 and 500 mM Mg2þ. Values were normalized to the current at 2100 mV in nominally Mg2þ-free solution. Mg2þ significantly blocked NMDA-induced currents at holding potentials more negative than 220 mV. There was no apparent effect at positive potentials. (B) Non-linear regression of inhibitory effect of Mg2þ on NMDA currents gave sigmoidal dose–response curves with IC50 values of 2.05, 6.94, and 12.2 mM at holding potentials of 2100, 280 and 260 mV, respectively. Values are mean ^ SEM of five determinations.
3.5. Sensitivity of NMDARs to NR2B selective blocker ifenprodil The sensitivity of NMDARs in DRG neurons to NR2B selective blocker ifenprodil was determined in the presence of 250 mM NMDA and 10 mM glycine in Mg2þ-free buffer. As shown in Fig. 7, ifenprodil dose-dependently inhibited NMDA-induced currents, with approximately 57 and 72% inhibition at ifenprodil concentrations of 3 and 10 mM, respectively ðn ¼ 5Þ: For further analysis of dose-dependent inhibitory effect of ifenprodil, up to five concentrations of ifenprodil were tested on each neuron. The data from six neurons was normalized and fit to one-site or two-site inhibition curves by non-linear regression (Fig. 7). Ifenprodil
dose –response curves for NMDARs in DRG neurons fit significantly better to the one-binding site model. The IC50 was 2.6 mM, with a maximal inhibitory effect of 94 ^ 4%. 3.6. Properties of NMDAR currents in DRG neurons innervating the distal colon DRG neurons that had innervated the distal colon were identified by retrograde label of fluorescently tagged cholera toxin subunit B. Of the 16 colon-specific neurons with small to medium cell diameters (20 –40 mm) studied, 12 (75%) responded to NMDA and glycine. Stimulating with 250 mM NMDA and 10 mM glycine, the average current in 11 retrograde labeled neurons was 252 ^ 67 pA with an
Fig. 7. Inhibition of NMDA-induced currents by ifenprodil in DRG neurons. (A) Representative traces of inward currents induced by 250 mM NMDA and 10 mM glycine before addition of ifenprodil, in presence of 3 mM ifenprodil, and after washout. Currents were inhibited by 56.6% by 3 mM ifenprodil, and almost completely recovered after washout. (B) The dose dependence of the effect was evaluated by normalizing the currents as the percent control response before addition of ifenprodil (n ¼ 3 – 5 for each concentration). Non-linear regression to a single site or to a two-site model gave a significantly better fit to the one-site model (2 site vs. 1 site: P ¼ 0:2; F test ¼ 1.72). The IC50 value was 2.57 mM (95% CI: 2.04– 3.24 mM) with a maximal inhibitory effect of 93.7 ^ 1.9%. The dose-dependent effect of ifenprodil was also evaluated in DRG neurons retrograde labeled from the distal colon. The IC50 value was 2.86 mM (95% CI: 2.13– 3.83 mM), with a maximal inhibitory effect of 92.7 ^ 2.8% for labeled neurons.
J. Li et al. / Pain 109 (2004) 443–452
average current density of 6.30 ^ 1.65 pA/pF. These values were not significantly different from neurons randomly selected from unlabeled preparations (233 ^ 87 pA and 5.39 ^ 1.80 pA/pF, respectively, n ¼ 118), nor were there any difference in the average capacitance of the neurons studied (40.8 ^ 1.1 pF, n ¼ 11 vs. 43.2 ^ 1.6 pF, n ¼ 118 for colon-specific and randomly selected neurons, respectively). In addition, the current kinetics in colon-specific neurons were similar with average t values for decay rate and off-rate of 1820 ^ 250, and 220 ^ 11 ms, respectively (compared with 1690 ^ 110 and 217 ^ 12 ms, respectively). NMDAR currents in colon-specific DRG neurons were also sensitive to ifenprodil and the dose-dependent effect of ifenprodil was not different from randomly studied neurons (Fig. 7). Non-linear regression analysis of the normalized data from five labeled neurons gave an IC50 of 2.9 mM with a maximal inhibitory effect of 93 ^ 5%.
4. Discussion The present study expands on our initial results that used [Ca2þ]i imaging to demonstrate functional expression of NMDAR by adult rat DRG neurons in primary culture (McRoberts et al., 2001) by employing whole cell patch clamp techniques for electrophysiological characterization of the kinetic properties of these receptors. We demonstrate the expression of NMDARs on DRG neurons with electrophysiological properties similar to those characterized in the CNS (Nagy et al., 2003; Priestley et al., 1995; Williams, 1993). The NMDAR in DRG neurons form channels that are Ca2þ permeable, require both glutamateand glycine-binding site agonists, display use-dependent inhibition by ketamine, and are inhibited by low concentrations of extracellular Mg2þ in a strongly voltagedependent manner. In addition, the NMDAR ion channels on DRG neurons are almost completely inhibited at relatively low doses by NR2B subunit selective inhibitor, ifenprodil. Identical observations were made with DRG neurons that had been retrograde labeled from the distal colon. These results are consistent with previous results demonstrating that the majority of NMDARs expressed on the cell body of DRG neurons from adult rats contain NR2B subunits (Marvizon et al., 2002). They also suggest that, at least on the cell body, DRG neurons innervating both somatic and visceral tissues express a functionally homogeneous population of NMDARs. Ca2þ permeability was demonstrated using Fura-2 and imaging of intracellular Ca2þ in individual neurons. DRG neurons in our preparation responded to co-application of NMDA and glycine, with a significant increase of [Ca2þ]i, which could be blocked by the reversible channel blocker, ketamine. In a previous study, we showed that the glutamate site competitive antagonist, AP-5, also blocked the response, and that D -serine could substitute for glycine and was absolutely required for NMDA-stimulated [Ca2þ]i increases
449
(McRoberts et al., 2001). Whole cell voltage clamp studies of DRG neurons showed NMDA and glycine stimulated inward currents that were blocked by 200 mM ketamine. Detailed analysis of the current – voltage relationship gave a reversal potential of near zero indicating that NMDARs are non-specific cation channels, consistent with the results from other groups (Kondoh et al., 2001; Ye et al., 2001). Native NMDARs are heteromeric complexes composed of four or five subunits (Laube et al., 1998; Mori and Mishina, 1995). Seven different genes encoding subunits of the NMDAR have been cloned, namely NR1, NR2A –D and NR3A – B in the rat (Das et al., 1998; Moriyoshi et al., 1991). In transfected cells, co-expression of the NR1 subunit with any member of the NR2 family will form functional NMDARs. However, each of the four NR2 subunits form receptors that differ in several functional properties, including voltage-dependent Mg2þ sensitivity, single channel conductance, and affinity for agonists, antagonists and modulators (Cull-Candy et al., 2001; Michaelis, 1998). With respect to Mg2þ sensitivity, recombinant NMDARs containing the NR2A or NR2B subunits are characterized by high sensitivity (IC50, 2 100 mV , 2 mM), whereas channels containing NR2C or NR2D exhibit lower sensitivity to Mg2þ block (IC50, 2 100 mV , 10– 15 mM) (Momiyama et al., 1996; Wyllie et al., 1998). NMDAR channels in DRG neurons had high sensitivity to Mg2þ block (IC50, 2 100 mV ¼ 2.05 mM), consistent with receptors composed predominantly of the NR2A or NR2B subunits. To distinguish between receptors composed of NR2A and NR2B, we used the subtype selective NMDAR antagonist, ifenprodil. In recombinant and native systems, ifenprodil inhibits NMDAR channels assembled from the NR1 –NR2B subunits with high affinity (IC50 , 0.5 mM), while it is much less potent for receptors assembled from NR1 –NR2A subunits (IC50 , 150 mM) (Williams, 1993). In DRG neurons, ifenprodil inhibited nearly all of the NMDA and glycine-induced current with an IC50 of 2.64 mM. There was no indication of multiple binding sites for ifenprodil’s inhibitory effect, indicting that the receptors expressed by DRG neurons are homogenous. Compared to recombinant NMDARs composed of NR1/NR2B subunits, NMDARs in DRG neurons are 5-fold less sensitive to ifenprodil. This discrepancy could be due to differences in post-translational modification of the receptors, for example by glycosylation, or it could be due to the modulatory influence of other subunits. We have shown that both the NR1 and NR2 subunits expressed by DRG have higher apparent molecular weights than subunits in the CNS, presumably due to glycosylation (Marvizon et al., 2002). Alternatively, native NMDARs can be composed of more than one type of NR2 subunit, yielding heterologous receptor assemblies with different ligand binding affinities (Chazot and Stephenson, 1997; Chazot et al., 1994; Misra et al., 2000; Wafford et al., 1993). Since small-medium size DRG express both NR2B and NR2D
450
J. Li et al. / Pain 109 (2004) 443–452
(Marvizon et al., 2002), the shift in ifenprodil sensitivity could be due to expression of heterologous NR1/NR2B/ NR2D receptors. However, co-expression of NR2A and NR2B along with NR1 leads to expression of three types of receptors: NR1/NR2A, NR1/NR2B and NR1/NR2A/NR2B. The dose dependence of ifenprodil in these cells is best fit to two sites with different affinities: a high affinity site identical to that observed in pure NR1/NR2B receptors, and a lower affinity site (54 mM) corresponding to inhibition of NR1/ NR2A/NR2B receptors (Kew et al., 1998). Since the dosedependent effect of ifenprodil in DRG neurons was best fit to a one-site model, expression of homogeneous NMDARs composed exclusively of NR1/NR2B/NR2D subunits is possible, but unlikely. The results of the current functional study complement previous studies examining the molecular expression of NMDAR subunits in DRG neurons. Studies examining subunit protein distribution have shown that NR1 subunits are widely expressed on the primary sensory neurons both in the cell body and the periphery (Sato et al., 1993). We have previously shown that mRNA encoding NR2B and NR2D subunits are predominately expressed in DRG (Marvizon et al., 2002). Immunocytochemical localization with antibodies recognizing NR2A/2B or NR2C/2D subunits indicated distinctly different intracellular distributions. NR1/NR2B NMDARs were present in the Golgi apparatus and occasionally on the plasma membrane of both smalland large-size DRG neurons. NMDARs containing the NR1/NR2D subunits were not co-localized with NR1/NR2B and were instead found throughout the cytoplasm, but only in small diameter neurons. In the present study, we focused on small to medium size (20 – 40 mm in diameter) DRG neurons that are likely to express both NR2B and NR2D. That we only observed functional expression of an apparently homogeneous NMDAR population that was sensitive to ifenprodil, is consistent with the immunocytochemical data showing cell surface expression of NR2B, but not NR2D. Our observations are also consistent with a previous report that found predominant expression of NR2B subunits on small-diameter primary afferent dorsal root ganglion neurons that give rise to the sciatic nerve (Ma and Hargreaves, 2000). We also characterized NMDARs on DRG neurons labeled retrogradely from the distal colon. We found that there was no difference in the ifenprodil sensitivity suggesting that NMDARs expressed in neurons innervating visceral tissues are not different from those innervating somatic tissues. NMDAR current density tended to be slightly larger in labeled neurons compared to randomly selected neurons, but the difference was not significant. Using Ca2þ imaging, we previously had shown that colon-selective neurons responded to co-application of NMDA and D -serine with a significantly larger increase in intracellular Ca2þ than randomly selected neurons (McRoberts et al., 2001). The discrepancy between this and the present study could be due to
differences in other parameters that affect neuronal excitability, Ca2þ permeability, or Ca2þ mobilization. Small to medium size DRG neurons generate fine myelinated Ad- and unmyelinated C-fibers that carry nociceptive information from the periphery to the spinal cord (Mense, 1990). In the spinal cord, NMDARs on dorsal horn neurons participate in central sensitization, as evidenced by their role in ‘wind-up’ of central responses to nociceptive stimuli (Davies and Lodge, 1987; Dickenson and Sullivan, 1987), and in long-term potentiation of synapses between primary afferent neurons and dorsal horn neurons (Liu and Sandku¨hler, 1995; Randic et al., 1993). NMDARs are also present on presynaptic terminals of primary afferents in the dorsal horn where they modulate the release of substance P (Liu et al., 1997; Marvizon et al., 1997), an important pro-inflammatory peptide mediating chronic and severe pain (Allen et al., 1997; Cao et al., 1998). In addition to expression of NMDARs on central terminals of DRG neurons, peripheral terminals of primary afferent nerves innervating visceral and somatic tissues also express NMDARs (Carlton et al., 1998; McRoberts et al., 2001). Pharmacological activation of peripheral NMDARs produces nociceptive behavior (Carlton et al., 1995; Lawand et al., 1997; Zhou et al., 1996), and peripheral injection of NMDA-R antagonists attenuates pain associated with neuropathic or inflammatory conditions (Davidson and Carlton, 1998; Warncke et al., 1997). Thus, NMDARs on both terminals of DRG neurons participate in pain responses. Our study demonstrates predominant functional expression of NR2B-containing NMDARs on the soma of DRG neurons. Assuming that these receptors are exported to the terminals of DRG neurons, the results support application of NR2B-selective receptor antagonists for pain management. Ifenprodil and related compounds are anti-nociceptive in a variety of preclinical pain models and have a much lower side-effect profile compared with other NMDAR antagonists (Chizh et al., 2001; Gurwitz and Weizman, 2002; Malmberg et al., 2003). Drugs targeting NR2B-containing NMDARs in pain-relevant structures, such as peripheral or central terminal of DRG neurons, could serve as a new class of medicine for managing chronic pain. Acknowledgements This study was supported by NIH grants R01 DK58173 and P50 DK64539-01 (EAM). We thank Teresa Olivas for her editorial contributions.
References Allan RD, Hanrahan JR, Hambley TW, Johnston GA, Mewett KN, Mitrovic AD. Synthesis and activity of a potent N-methyl-D -aspartic acid agonist,
J. Li et al. / Pain 109 (2004) 443–452 trans-1-aminocyclobutane-1,3-dicarboxylic acid, and related phosphonic and carboxylic acids. J Med Chem 1990;33:2905 –15. Allen BJ, Rogers SD, Ghilardi JR, Menning PM, Kuskowski MA, Basbaum AI, Simone DA, Mantyh PW. Noxious cutaneous thermal stimuli induce a graded release of endogenous substance P in the spinal cord: imaging peptide action in vivo. J Neurosci 1997;17:5921 –7. Cao YQ, Mantyh PW, Carlson EJ, Gillespie AM, Epstein CH, Basbaum AI. Primary afferent tachykinins are required to experience moderate to intense pain. Nature 1998;392:390–4. Carlton SM, Hargett GL, Coggeshall RE. Localization and activation of glutamate receptors in unmyelinated axons of rat glabrous skin. Neurosci Lett 1995;197:25– 8. Carlton SM, Zhou ST, Coggeshall RE. Evidence for the interaction of glutamate and NK1 receptors in the periphery. Brain Res 1998;790: 160–9. Chazot PL, Stephenson FA. Molecular dissection of native mammalian forebrain NMDA receptors containing the NR1 C2 exon: direct demonstration of NMDA receptors comprising NR1, NR2A, and NR2B subunits within the same complex. J Neurochem 1997;69: 2138–44. Chazot PL, Coleman SK, Cik M, Stephenson FA. Molecular characterization of N-methyl-D -aspartate receptors expressed in mammalian cells yields evidence for the coexistence of three subunit types within a discrete receptor molecule. J Biol Chem 1994;269:24403–9. Chizh BA, Headley PM, Tzschentke TM. NMDA receptor antagonists as analgesics: focus on the NR2B subtype. Trends Pharm Sci 2001;22: 636–42. Cull-Candy S, Brickley S, Farrant M. NMDA receptor subunits: diversity, development and disease. Curr Opin Neurobiol 2001;11:327 –35. Das S, Sasaki YF, Rothe T, Premkumar LS, Takasu M, Crandalli JE, Dikkes P, Conner DA, Rayudu PV, Cheung W, Vincent C, Lipton SA, Nakanish N. Increased NMDA current and spine density in mice lacking the NMDA receptor subunit NR3A. Nature 1998;393:377–81. Davidson EM, Carlton SM. Intraplantar injection of dextrorphan, ketamine or memantine attenuates formalin-induced behaviors. Brain Res 1998; 785:136–42. Davies SN, Lodge D. Evidence for involvement of N-methylaspartate receptors in wind-up of class 2 neurones in the dorsal horn of the rat. Brain Res 1987;424:402 –6. Dickenson AH, Sullivan AF. Evidence for a role of the NMDA receptor in the frequency dependent potentiation of deep rat dorsal horn nociceptive neurones following C fibre stimulation. Neuropharmacology 1987;26:1235– 8. Dingledine R, Borges K, Bowie D, Traynelis SF. The glutamate receptor ion channels. Pharmacol Rev 1999;51:7–62. Gurwitz D, Weizman A. The NR2B subunit of glutamate receptors as a potential target for relieving chronic pain: prospects and concerns. Drug Discov Today 2002;7:403–6. Hawkins LM, Chazot PL, Stephenson FA. Biochemical evidence for the coassociation of three N-methyl-D -aspartate (NMDA) R2 subunits in recombinant NMDA receptors. J Biol Chem 1999;274:27211–8. Heppenstall PA, Fleetwood-Walker SM. The glycine site of the NMDA receptor contributes to neurokinin1 receptor agonist facilitation of NMDA receptor agonist-evoked activity in rat dorsal horn neurons. Brain Res 1997;744:235 –45. Herrero JF, Laird JM, Lopez-Garcia JA. Wind-up of spinal cord neurones and pain sensation: much ado about something? Prog Neurobiol 2000; 61:169–203. Kew JN, Richards JG, Mutel V, Kemp JA. Developmental changes in NMDA receptor glycine affinity and ifenprodil sensitivity reveal three distinct populations of NMDA receptors in individual rat cortical neurons. J Neurosci 1998;18:1935–43. Kleckner NW, Dingledine R. Requirement for glycine in activation of NMDA-receptors expressed in Xenopus oocytes. Science 1988;241: 835–7.
451
Kondoh T, Nishizaki T, Aihara H, Tamaki N. NMDA-responsible, APVinsensitive receptor in cultured human astrocytes. Life Sci 2001;68: 1761–7. Laube B, Kuhse J, Betz H. Evidence for a tetrameric structure of recombinant NMDA receptors. J Neurosci 1998;18:2954–61. Lawand NB, Willis WD, Westlund KN. Excitatory amino acid receptor involvement in peripheral nociceptive transmission in rats. Eur J Pharmacol 1997;324:169 –77. Liu X-G, Sandku¨hler J. Long-term potentiation of C-fiber-evoked potentials in the rat spinal dorsal horn is prevented by spinal N-methyl-D -aspartic acid receptor blockage. Neurosci Lett 1995;191: 43– 6. Liu H, Mantyh PW, Basbaum AI. NMDA-receptor regulation of substance P release from primary afferent nociceptors. Nature 1997;386:721–4. Lovinger DM, Weight FF. Glutamate induces a depolarization of adult rat dorsal root ganglion neurons that is mediated predominantly by NMDA receptors. Neurosci Lett 1988;94:314– 20. Ma Q-P, Hargreaves RJ. Localization of N-methyl-aspartate NR2B subunits on primary sensory neurons that give rise to small-caliber sciatic nerve fibers in rats. Neuroscience 2000;101:699–707. Malmberg AB, Gilbert H, McCabe RT, Basbaum AI. Powerful antinociceptive effects of the cone snail venom-derived subtype-selective NMDA receptor antagonists conantokins G and T. Pain 2003;101: 109– 16. Marvizon JC, Martinez V, Grady EF, Bunnett NW, Mayer EA. Neurokinin 1 receptor internalization in spinal cord slices induced by dorsal root stimulation is mediated by NMDA receptors. J Neurosci 1997;17: 8129–36. Marvizon JC, McRoberts JA, Ennes HS, Song B, Wang X, Jinton L, Corneliussen B, Mayer EA. Two N-methyl-D -aspartate receptors in rat dorsal root ganglia with different subunit composition and localization. J Comp Neurol 2002;446:325 –41. Matsuda K, Kamiya Y, Matsuda S, Yuzaki M. Cloning and characterization of a novel NMDA receptor subunit NR3B: a dominant subunit that reduces calcium permeability. Mol Brain Res 2002;100:43–52. McRoberts JA, Coutinho SV, Marvizon JC, Grady EF, Tognetto M, Sengupta JN, Ennes HS, Chaban VV, Amadesi S, Creminon C, Lanthorn T, Geppetti P, Bunnett NW, Mayer EA. Role of peripheral N-methyl-D -aspartate (NMDA) receptors in visceral nociception in rats. Gastroenterology 2001;120:1737–48. Mense S. Structure–function relationships in identified afferent neurones. Anat Embryol (Berl) 1990;181:1–17. Michaelis EK. Molecular biology of glutamate receptors in the central nervous system and their role in excitotoxicity, oxidative stress and aging. Prog Neurobiol 1998;54:369 –415. Misra C, Brickley SG, Wyllie DJA, Cull-Candy SG. Slow deactivation kinetics of NMDA receptors containing NR1 and NR2D subunits in rat cerebellar Purkinje cells. J Physiol 2000;525:299–305. Momiyama A, Feldmeyer D, Cull-Candy SG. Identification of a native lowconductance NMDA channel with reduced sensitivity to Mg in rat central neurones. J Physiol 1996;494:479–92. Mori H, Mishina M. Structure and function of the NMDA receptor channel. Neuropharmacology 1995;34:1219–37. Moriyoshi K, Masu M, Ishii T, Shigemoto R, Mizuno N, Nakanishi S. Molecular cloning and characterization of the rat NMDA receptor. Nature 1991;353:31 –7. Nagy J, Kolok S, Dezso P, Boros A, Szombathelyi Z. Differential alterations in the expression of NMDA receptor subunits following chronic ethanol treatment in primary cultures of rat cortical and hippocampal neurones. Neurochem Int 2003;42:35–43. Nishi M, Hinds H, Lu HP, Kawata M, Hayashi Y. Motoneuron-specific expression of NR3B, a novel NMDA-type glutamate receptor subunit that works in a dominant-negative manner. J Neurosci 2001;21:185RC. Omote K, Kawamata T, Kawamata M, Namiki A. Formalin-induced release of excitatory amino acids in the skin of the rat hindpaw. Brain Res 1998; 787:161–4.
452
J. Li et al. / Pain 109 (2004) 443–452
Premkumar LS, Auerbach A. Stoichiometry of recombinant N-methyl-D aspartate receptor channels inferred from single-channel current patterns. J Gen Physiol 1997;110:485–502. Priestley T, Laughton P, Myers J, Le Bourdelles B, Kerby J, Whiting PJ. Pharmacological properties of recombinant human N-methyl-D -aspartate receptors comprising NR1a/NR2A and NR1a/NR2B subunit assemblies expressed in permanently transfected mouse fibroblast cells. Mol Pharmacol 1995;48:841–8. Randic M, Jiang MC, Cerne R. Long-term potentiation and long-term depression of primary afferent neurotransmission in the rat spinal cord. J Neurosci 1993;13:5228–41. Sasaki YF, Rothe T, Premkumar LS, Das S, Cui J, Talantova MV, Wong HK, Gong X, Chan SF, Zhang D, Nakanishi N, Sucher NJ, Lipton SA. Characterization and comparison of the NR3A subunit of the NMDA receptor in recombinant systems and primary cortical neurons. J Neurophysiol 2002;87:2052–63. Sato K, Kiyama H, Park HT, Tohyama M. AMPA, KA, and NMDA receptors are expressed in the rat DRG neurones. NeuroReport 1993;4: 1263–5. Schorge S, Colquhoun D. Studies of NMDA receptor function and stoichiometry with truncated and tandem subunits. J Neurosci 2003;23: 1151–8.
Si JQ, Li ZW. Inhibition by baclofen of NMDA-activated current in rat dorsal root ganglion neurons. Zhongguo Yao Li Xue Bao 1999;20: 324 –8. Wafford KA, Bain CJ, Le Bourdelles B, Whiting PJ, Kemp JA. Preferential co-assembly of recombinant NMDA receptors composed of three different subunits. NeuroReport 1993;4:1347– 9. Warncke T, Jorum E, Stubhaug A. Local treatment with the N-methylaspartate receptor antagonist ketamine, inhibit development of secondary hyperalgesia in man by a peripheral action. Neurosci Lett 1997;227:1–4. Williams K. Ifenprodil discriminates subtypes of the N-methyl-D -aspartate receptor: selectivity and mechanisms at recombinant heteromeric receptors. Mol Pharmacol 1993;44:851–9. Wyllie DJ, Behe P, Colquhoun D. Single-channel activations and concentration jumps: comparison of recombinant NR1a/NR2A and NR1a/NR2D NMDA receptors. J Physiol 1998;510:1–18. Yamakura T, Shimoji K. Subunit- and site-specific pharmacology of the NMDA receptor channel. Prog Neurobiol 1999;59:279 –98. Ye JH, Tao L, Zalcman SS. Interleukin-2 modulates N-methyl-aspartate receptors of native mesolimbic neurons. Brain Res 2001;894:241–8. Zhou S, Bonasera L, Carlton SM. Peripheral administration of NMDA, AMPA or KA results in pain behaviors in rats. NeuroReport 1996;7: 895 –900.