Functional Identification of NR2 Subunits Contributing to NMDA Receptors on Substance P Receptor-Expressing Dorsal Horn Neurons

Abstract

NMDA receptors are important elements in pain signaling in the spinal cord dorsal horn. They are heterotetramers typically composed of two NR1 and two of four NR2 subunits: NR2A-2D. Mice lacking specific NR2 subunits show deficits in pain transmission yet subunit location in the spinal cord remains unclear. We have combined electrophysiological and pharmacological approaches to investigate the composition of functional NMDA receptors expressed by lamina I, substance P receptor-expressing (NK1R+) neurons, as well as NK1R- neurons. Under low Mg²⁺ conditions (100 μM), the conductance of NMDA receptors at −90 mV (g(−90 mV)) with NR2A or NR2B subunits (NR2A/B) is low compared to conductance measured at the membrane potential where the inward current is maximal or maximal inward current (MIC) (ratio of ~0.07 calculated from Kuner and Schoepfer, 1996). For NR2C or NR2D subunits (NR2C/D), the ratio is higher (ratio ~0.4). NK1R+ and NK1R- neurons express NMDA receptors that give ratios ~0.28 and 0.16, respectively, suggesting both types of subunits are present in both populations of neurons, with NK1R+ neurons expressing a higher percentage of NR2C/D type NMDA receptors. This was confirmed using EAB318, an NR2A/B preferring antagonist, and UBP141, a mildly selective NR2C/D antagonist to increase and decrease the g(−90 mV)/g(MIC) ratios in both subpopulations of neurons.

Background

NMDA receptors in the spinal cord dorsal horn are key elements in the initiation of changes in synaptic strength [1] and pain hypersensitivity [2,3]. These receptors consist of two obligatory NR1 subunits and two NR2 subunits, of which there are four types encoded by distinct genes: NR2A, NR2B, NR2C and NR2D [4]. The incorporation of different NR2 subunits has a major impact on the functional properties of the NMDA receptor, critically influencing agonist and antagonist affinity, receptor deactivation kinetics, channel conductance and interactions with intracellular proteins [3]. Additionally, NMDA receptors with NR2A or NR2B show higher Mg²⁺ sensitivity at negative membrane potentials than those with NR2C or NR2D [5,6].

Involvement of NMDA receptors in dorsal horn function has been demonstrated through experiments interfering with expression of different NMDA receptor subunits. Knockdown of the NR1 subunit of NMDA receptors to eliminate functional NMDA receptors in the spinal cord reduces hyperalgesia and allodynia in a number of animal models but does not alter acute pain responses [7 –9]. NR2A knockout mice show some reduced forms of hypersensitivities [10 –12]. However, these NR2A knockouts display normal acute pain responses [12], formalin-induced hyperalgesia [13] and nerve ligation or injury-induced allodynia [14,15]. NR2B knockout mice do not survive postnatally [16,17], therefore NR2B specific antagonists have been used to study the role of this protein in pain hypersensitivity. Intrathecal administration of NR2B antagonists blocks or decreases PGE2 or NMDA induced allodynia [11] as well as capsaicin-induced hyperalgesia [18]. NR2D knockout mice fail to develop nerve ligation [12], PGE2 [19] or PGF2alpha-induced allodynia [11,20]. Overall, these data suggest that different NR2 subunits are involved in dorsal horn circuits important for the development of hyperalgesia or allodynia but their specific functions remain unresolved.

Lamina I of the spinal cord is a critical site for nociceptive processing, receiving abundant monosynaptic input from nociceptors. The main output neurons of lamina I, the substance P receptor-expressing (NK1R+) projection neurons, are essential in mediating pain hypersensitivity [21,22]. NK1R+ neurons express NMDA receptors [23,24] but little is known about the subtypes of NMDA receptors they express.

In this paper, we have taken advantage of the different magnesium sensitivities and pharmacology of NMDA receptors with different NR2 subunit composition to identify functionally expressed NMDA receptors on NK1R+ and NK1R- dorsal horn neurons in lamina I.

Methods

Transverse slice preparation

Lumbar spinal cords were obtained from rats of postnatal day 14 (P14) to P19. The animals were first anesthetized with isoflurane and then decapitated. All experiments were conducted with the approval of the Columbia University Institutional Animal Care and Use Committee and in accord with the Guide for the Care and Use of Laboratory Animals. The spinal cords were excised and placed in ice-cold oxygenated high Mg²⁺ Krebs solution (95% O₂/5% CO₂ saturated Krebs solution, in mM: NaCl 125 or sucrose 250, KCl 2.5, NaHCO₃ 26, NaH₂PO₄ 1.25, glucose 25, MgCl₂ 6, CaCl₂ 1.5, pH 7.4) plus 1 mM kynurenic acid. After removal of the dura mater and arachnoid membranes, all ventral roots were cut close to the cord and the spinal cord embedded in low melting agarose (Invitrogen Life Technologies) for slicing. Transverse slices (350–450 μm) with attached dorsal roots were obtained using a Leica VT1000S vibrating blade microtome. Slices were then incubated in oxygenated high Mg²⁺Krebs solution (no sucrose included) at 36°C for 1 hour before recording.

Recording from pre-identified NK1R+ and NK1R- lamina I neurons

The labeling of NK1R+ dorsal horn neurons with fluorescent dye has been described elsewhere [25,26]. In brief, spinal cord slices were incubated in high Mg²⁺ Krebs solution containing 20 – 40 nM tetramethylrhodamine-conjugated substance P (TMR-substance P) for 20 – 30 minutes at room temperature following 1 hour of recovery at 36°C. After unbound substance P was washed away for at least 20 minutes in an incubation chamber containing oxygenated high Mg²⁺ Krebs solution, slices were transferred to a submersion style chamber for recording. NK1R+ neurons were identified as expressing NK1R by clear, intense labeling with TMR-substance P. NK1R- neurons were chosen as lamina I neurons showing no evidence of TMR-substance P staining.

Recording solutions

Intracellular solution used for most of these experiments had the following composition (in mM): Cs-methylsulfonate 130, Na-methylsulfonate 10, EGTA 10, CaCl₂ 1, HEPES 10, QX-314·Cl 5, Mg²⁺-ATP 2, pH adjusted to 7.2 with CsOH, osmolarity adjusted to 290 with sucrose. For some experiments in which intracellular Ca²⁺ needed to be strongly chelated, BAPTA intracellular solution was used. It was composed of (in mM): Cs-methylsulfonate 50, Na-methylsulfonate 10, BAPTA·Cs 40, CaCl₂ 4, HEPES 10, QX-314·Cl or QX-222·Cl 5, Mg²⁺-ATP 2, TEA·Cl 10, pH adjusted to 7.2 with CsOH, osmolarity about 310. Junction potentials were measured empirically and corrected in the bath before GOhm seal formation for each cell.

Modified Krebs solutions were used for the extracellular bath. To prevent possible neurotoxicity associated with Ca²⁺ influx through activated NMDA receptors, we replaced 95–98% of the extracellular Ca²⁺ with 3 mM Ba²⁺. The barium Krebs comprised: NaCl 125, KCl 2.5, NaH₂PO₄1.25, NaHCO₃ 26, glucose 25, MgCl₂ 0.1, CaCl₂ 0.04–0.1, BaCl₂ 3 and pH 7.4. TTX (0.5 μM), SR95531 (5–10 μM) and strychnine (1 μM) were included in the extra-cellular solutions to eliminate action potential generation and involvement of inhibitory circuits.

Analysis of NMDA induced membrane currents

To obtain the current-voltage relationships of NMDA receptor-mediated currents, NMDA (15 μM) was superfused onto pre-identified, lamina I neurons for 2–3 minutes following several minutes of baseline, whole-cell recording. Triangle voltage ramp commands (the ramp up and ramp down were 0.9 sec duration each) were applied continuously at low frequency (0.05 Hz). Digital sampling frequency was 10 KHz. NMDA applications were repeated 2–3 times before NMDA co-application with antagonists. The data for the first NMDA application were not included for analysis due to changing baseline conditions. Current responses to triangle voltage ramps before and after recovery from NMDA application were averaged as a control current then subtracted from each triangle ramp made during NMDA induced currents. The resulting NMDA current ramps were plotted as a function of membrane potential and further analyzed. To minimize noise for measuring the following parameters, NMDA current ramps were subjected to a rolling average procedure over a 100 msec time frame.

For each voltage ramp during NMDA applications, the membrane current at −90 mV (I(−90 mV)), the maximal inward current (MIC), and the membrane potential for the MIC (VMIC) were measured (Figure 1E). The current measured at −90 mV holding potential was determined as I(−90 mV). The MIC was initially determined as the most negative current value in the rolling average. The VMIC was then determined as the voltage corresponding to the MIC. Because each ramp had an up and a down phase, each parameter from a ramp current had a pair of values and they were averaged for following analysis. The conductance at −90 mV and MIC (g(−90 mV) and g(MIC) respectively) as well as conductance ratio were then calculated based on the formulae:

\begin{array}{l} g (- 90 mV) = - I (- 90 mV) / 90 mV \\ g (MIC) = - MIC / VMIC \\ g (- 90 mV) / g (MIC) = I (- 90 mV) * VMIC / (90 * MIC) \end{array}

Figure 1

Parameters representing the voltage sensitivity of current flow through NMDA receptors were determined following NMDA application onto pre-identified dorsal horn neurons. (A) NK1R+ neuron in the superficial dorsal horn was selectively labeled by TMR-substance P in a transverse slice (40 × objective). (B) IR-DIC image showing the NK1R+ neuron patched with a pipette. (C) A representative trace shows an NMDA-evoked inward current with 100 μM Mg²⁺ in the bath. (D) Current responses to voltage ramps are shown at expanded time base. (E) The voltage dependence of NMDA receptor-mediated currents derived from NMDA evoked currents. (F) The same as in Figure E. The conductance of NMDA receptors at −90 mV (g(−90 mV)) and MIC (g(MIC)) are illustrated.

To compare NMDA receptor g(−90 mV)/g(MIC) ratios under different pharmacological conditions, we averaged three ratio values calculated for each NMDA application near the peak NMDA response at −70 mV. The ratios under different pharmacological conditions or represented by different neuron populations were then compared.

Only cells starting with reversible NMDA induced membrane currents, in which the difference between g(−90 mV)/g(MIC) ratios during wash-in and wash-out of NMDA was less than 0.15, were included for analysis. Cells with high membrane holding current (> −100 pA) were discarded.

Materials

SR 95531 hydrobromide and QX-222 · Cl were purchased from Tocris Cookson (Bristol, UK). QX-314 · Cl was purchased from Sigma-Aldrich or Alomone labs (Jerusalem, Israel). Low melting point agarose and TMR-substance P were purchased from Invitrogen Corp. Some TMR-substance P was synthesized and purchased from AnaSpec, Inc. Strychnine was obtained from Sigma-Aldrich. EAB318 was provided by Wyeth Research. EAB318 has IC50 of 20, 80 and 3500 nM for NMDA receptors with NR2A, NR2B and NR2C respectively [27]. UBP141 was synthesized as described [28]. The K_i of UBP141 for NMDA receptors with NR2A – NR2D are 14, 19, 4 and 2.7 μM respectively [28].

Results

The l-V relationship of NMDA currents induced by superfusion of NMDA onto dorsal horn neurons

We investigated the total population of functional NMDA receptors expressed by different classes of lamina I neurons. NMDA was bath-applied onto spinal cord slices to activate all functional NMDA receptors. To identify the type of NMDA receptors expressed by pre-identified lamina I neurons in the spinal cord dorsal horn, we took advantage of the differential sensitivity to Mg²⁺ inherent in NMDA receptors composed of different NR2 subunits. NMDA receptors containing NR2A or NR2B subunits show more negative slope conductance at negative membrane potentials than those containing NR2C or NR2D [6]. Because the measurable difference in Mg²⁺ sensitivity is enhanced when extracellular Mg²⁺ concentration is low, 100 μM extracellular Mg²⁺ was used throughout these experiments. SR95532 (10 μM), strychnine (1 μM) and TTX (0.5–1 μ;M) were always included to eliminate the inhibitory currents and action potential triggered responses. Most of the Ca²⁺ in the Krebs was replaced with Ba²⁺ to diminish evoked neurotransmitter release and Ca²⁺ dependent currents in the cells.

NK1R+ and NK1R- dorsal horn neurons were visually identified for whole-cell recording following incubation of spinal cord slices in TMR-substance P [25,26] as shown in Figure 1A and Fig 1B, in which fluorescence and IR-DIC images are shown respectively. Bath application of NMDA (15 μM) to these neurons generated inward currents at a holding potential of −70 mV as shown in Figure 1C. To determine the voltage-dependent Mg²⁺ sensitivity of these receptors, triangle voltage ramp commands from −100 mV to + 50 mV and back to − 100 mV were applied before, during and after NMDA application at 0.05 Hz (Figure 1C and Fig 1D). After subtraction of control current, the resulting ramps of NMDA receptor-mediated current were plotted as a function of command voltage as shown in Figure 1E. The voltage sensitivity of the NMDA current generated by the ascending ramp command is similar to that generated by the descending command (Figure 1E). In addition, the current-voltage relationships of these NMDA induced currents had an average reversal potential of − 2.0 ± 1.0 mV (n = 15), close to the predicted NMDA receptor reversal potential. The pair of NMDA current responses obtained from each triangle voltage command were used to determine the current at −90 mV (I(−90 mV)), the maximal inward current (MIC) and the membrane potential at which MIC occurs (VMIC) as illustrated in Figure 1E (see Methods). The membrane conductance at −90 mV (g(−90 mV)) and MIC (g(MIC)) were then calculated as shown in Figure 1F.

The voltage sensitivity of NMDA currents depends predominantly on the voltage dependence of Mg²⁺ block of the receptors [29,30]. The voltage sensitivity of the agonist activated NMDA receptors in our experiments was quantified by dividing g(−90 mV) with g(MIC). The ratio was then compared to the value derived from heterologous expression data using specific NR1 and NR2 subunit combinations. From such data we calculated that NMDA receptors containing NR1/NR2A or NR1/NR2B show g(−90 mV)/g(MIC) ratios of about 0.07 and that their VMIC is between −37 and −40 mV. Conversely, NMDA receptors containing NR1/NR2C or NR1/NR2D have ratios around 0.39 and VMICs around −52 to −57 mV (extracted from Kuner et al. [6]). Thus, for example, lower g(−90 mV)/g(MIC) ratios near 0.07 and less negative VMICs indicate expression of NR2A/B-containing NMDA receptors with high Mg²⁺ sensitivity.

In Figure 2A, the control response of an NK1R+ neuron to 15 μM NMDA is shown. The ramp response recorded when the agonist response to NMDA was at its greatest is plotted at the bottom. The value of the g(−90 mV)/g(MIC) ratio is 0.2. This is intermediate between the values for receptors that include NR2A/B and NR2C/D subunits suggesting that NMDA receptors with both types of NR2 subunits are present on this NK1R+ lamina I neuron. The VMIC value (−48 mV in this example) is also intermediate between the VMIC values of NR2A/B and NR2C/D subunits, supporting the interpretation that the NMDA receptors expressed by this NK1R+ neuron are heterogeneous in NR2 subtype expression.

Figure 2

The voltage-dependence of NMDA receptor-mediated currents is a good indicator of NMDA receptor subtype expression.A-C show current responses to NMDA bath applied to the same NK1R+ neuron as in Figure 1. (A) NMDA-evoked current response is the same as in Figure 1. (B) Co-application of NMDA and UBP141, an NR2C/D preferring antagonist, induced a smaller inward current. The I-V relationship shows a more pronounced negative slope conductance. (C) Co-application of NMDA and EAB318, an NR2A/B preferring antagonist, induced NMDA receptor-mediated current with less negative slope conductance.

Although most of the extracellular Ca²⁺ was replaced with Ba²⁺ in these experiments, it was still possible that the remaining bath Ca²⁺ or Ca²⁺ released from endoplasmic reticulum (ER) was sufficient to trigger activation of other currents, altering g(−90 mV)/g(MIC) ratio and VMIC values. In 9 of 15 cells recorded, intracellular solution containing 40 mM BAPTA was used to fully suppress accumulation of intracellular Ca²⁺ associated with NMDA receptor activation. There was no significant difference between the g(−90 mV)/g(MIC) values when recording with BAPTA or EGTA intracellular solutions. Thus the data from these two groups were pooled.

Pharmacological test of NR2 subunit confirms the expression of NR2A/B and NR2C/D type NMDA receptors

Next we used pharmacology to confirm that g(−90 mV)/g(MIC) ratio and VMIC are good indicators of NMDA receptor subtypes expressed by dorsal horn neurons. EAB318 (100 – 200 nM) and UBP141 (20–30 μM) block different NMDA receptor subtypes: UBP141 [28] is a mildly selective NR2C/D preferring antagonist while EAB318 is a NR2A/B selective blocker [27]. If both categories of NR2 subtypes are present, EAB318 should make the NMDA evoked current more NR2C/D like and UBP141 should make the current more NR2A/B like. As predicted, when NMDA was co-applied with UBP141, the current-voltage relationship had a smaller g(−90 mV)/g(MIC) ratio and less negative VMIC (Figure 2A and Fig 2B). This suggests that in the presence of UBP141, a higher proportion of NR2A/B type NMDA receptors dominate the current. When NMDA was co-applied with EAB318, the current voltage relationship shifted to a larger g(−90 mV)/g(MIC) ratio and a more negative VMIC, suggesting that a higher proportion of NR2C/D type NMDA receptors were revealed (Figure 2C).

To ensure that the shift of g(−90 mV)/g(MIC) ratio and VMIC were genuinely associated with selective block of a subpopulation of NMDA receptors and not simply caused by errors associated with measuring smaller amplitude NMDA evoked currents, the relationship between g(−90 mV)/g(MIC) and NMDA evoked current amplitude was plotted as in Figure 3A. As the NMDA plus antagonists washed onto the dorsal horn neuron under study, the impact of the two antagonists on g(−90 mV)/g(MIC) were different. UBP141 depressed the amplitude of NMDA evoked currents and g(−90 mV)/g(MIC) values. EAB318 also depressed the amplitude of NMDA evoked current but caused a large shift to higher g(−90 mV)/g(MIC) values. In the two situations, UBP141 and EAB318 depressed the peak amplitudes of NMDA induced currents from −306 ± 48 to −130 ± 222 pA (n = 15, p < 0.01 for paired t-test) and −108 ± 16 pA (n = 15, p < 0.01 for paired t-test) respectively. Figure 3B shows the individual and mean g(−90 mV)/g(MIC) ratios determined when NMDA evoked currents were maximal under different drug conditions. UBP141 significantly decreased the g(−90 mV)/g(MIC) ratio from 0.23 ± 0.03 to 0.15 ± 0.01 (n = 15, p < 0.01 for paired t-test) while EAB318 significantly increased the ratio from 0.23 ± 0.03 to 0.36 ± 0.05 (n = 15, p < 0.01 for paired t-test) (see Methods). The upper and lower broken horizontal lines represent the g(−90 mV)/g(MIC) for pure NR2A/B and NR2C/D-containing NMDA receptors respectively as calculated using the data of Kuner and Schoepfer [6]. Comparing our data to these benchmarks shows that the two antagonists are pushing the g(−90 mV)/g(MIC) values in the directions predicted from heterologous expression data, assuming both types of subunits are present in the neuron tested.

Figure 3

The NMDA g(−90 mV)/g(MIC) ratio is not a function of the amplitude of NMDA-evoked membrane current.(A) The g(−90 mV)/g(MIC) ratio was calculated for each test ramp throughout NMDA applications in the absence and presence of antagonists and plotted as a function of NMDA-evoked current amplitude. (B) Summary of the antagonist effects on the g(−90 mV)/g(MIC) ratio. (C) The VMIC was calculated for each test ramp throughout NMDA applications in the absence and presence of antagonists and plotted as a function of NMDA-evoked current amplitude for the same data as (A). (D) Summary of the antagonist effects on the VMIC (n = 15). (E) The same data as in (B) but grouped according to the sequences of drug application.

We also analyzed VMICs under different drug conditions. Figure 3C, calculated from the same data as Figure 3A, shows the action of the two antagonists on VMIC. As expected, UBP141 caused the VMIC to become somewhat more positive or more NR2A/B like, while EAB318 caused the VMIC to become more negative or more NR2C/D like. Figure 3D is the summary showing that UBP141 significantly shifted the mean VMIC, measured when the currents evoked by NMDA were maximal (see Methods), from −50 ± 2 mV to −45 ± 2 mV (n = 15, p < 0.01 for paired t-test) while EAB318 significantly changed the mean VMIC to −53 ± 2 mV (n = 15, p < 0.05 for paired t-test).

To rule out the possibility that the sequence of antagonist co-application with NMDA may have some effects on the g(−90 mV)/g(MIC) ratio, we grouped the experiments according to the order of drug application. In 9 of 15 cells tested, UBP141 was co-applied with NMDA before EAB318 co-application with NMDA. In 6 of 15 cells tested, UBP141 was co-applied after EAB318. UBP141 significantly decreased the g(−90 mV)/g(MIC) from 0.23 ± 0.04 to 0.13 ± 0.02 (n = 9, p < 0.01 for paired t-test) and EAB318 significantly increased the g(−90 mV)/g(MIC) from 0.23 ± 0.04 to 0.37 ± 0.07 (n = 9, p < 0.01 for paired t-test) when UBP141 was applied first (Figure 3E left side). Similarly, EAB318 significantly increased the g(−90 mV)/g(MIC) from 0.23 ± 0.04 to 0.33 ± 0.06 (n = 6, p < 0.05 for paired t-test) and UBP141 significantly decreased the g(−90 mV)/g(MIC) from 0.23 ± 0.04 to 0.17 ± 0.02 (n = 6, p < 0.05 for paired t-test) when EAB318 was applied earlier (Figure 3E right side).

Comparison of NMDA receptor types between NK1R+ and NK1R- neurons

Having verified that these two approaches, Mg²⁺ sensitivity and pharmacological blockade, allowed us to distinguish different NMDA receptor subtypes, we asked if NK1+ and NK1– neurons differed in the proportions of NR2A/B and NR2C/D containing receptors that they expressed. Indeed, although co-application of NMDA with either UBP141 or EAB318 significantly changed the g(−90 mV)/g(MIC) ratios, the ratio changes were not the same for all neurons tested. Figure 4A1 shows the UBP141-induced change in the g(−90 mV)/g(MIC) ratio in individual lamina I neurons pre-identified as either NK1R+ or NK1R- neurons. The g(−90 mV)/g(MIC) ratios of most neurons showed high sensitivity to UBP141, suggesting that most dorsal horn neurons express some NMDA receptors that include NR2C/D subunits. On average, NK1R+ neurons had a significantly higher g(−90 mV)/g(MIC) ratio than NK1R- neurons (0.26 ± 0.03, n = 11 v.s. 0.16 ± 0.02, n = 4, p < 0.05 for unpaired t-test), indicating that in these neurons, a higher percentage of NMDA receptors include NR2C/D subunits than in NK1R- neurons (Figure 4A2).

Figure 4

NK1R+ neurons express higher proportion of NMDA receptors with NR2C/D subunits than do NK1R- neurons. (A1) Individual NK1R+ and NK1R- neurons showed different degree of g(−90 mV)/g(MIC) ratio decrease following exposure to UBP141. (A2) Overall, NK1R+ neurons show significantly higher g(−90 mV)/g(MIC) ratio than NK1R- neurons (p < 0.05). (B1) Individual NK1R+ and NK1R- neurons showed increased g(−90 mV)/g(MIC) ratio when NMDA was applied in the presence of EAB318. (B2) In summary, EAB318 significantly increased g(-90)/g(MIC) ratios in both NK1R+ NK1R- lamina 1 neurons.

To confirm this observation, we also observed the effect of EAB318 on the g(−90 mV)/g(MIC) ratio of individual neurons in Figure 4B1. EAB318 increased the g(−90 mV)/g(MIC) ratio in most, but not all dorsal horn neurons tested, again suggesting that most of them expressed both high and low Mg²⁺ sensitivity, consistent with the data of UBP141. On average, EAB318 caused the g(−90 mV)/g(MIC) ratio measured from NK1R+ neurons to become more NR2C/D like than from NK1R- neurons (Figure 4B2), consistent with the interpretation that NK1R+ neurons express a higher percentage of NMDA receptors with NR2C/D subunits.

Discussion

We have identified NMDA receptor subtypes expressed by two populations of dorsal horn neurons; NK1R+ and NK1R- lamina I neurons. Based on our experiments, both highly Mg²⁺ sensitive (NR2A/B) and poorly Mg²⁺ sensitive (NR2C/D) NMDA receptors are expressed by NK1R+ neurons. NR2C/D subunits are less strongly expressed by NK1R- lamina I neurons and therefore the NR2A/B receptor subtypes dominate more strongly there.

Ratio assay confirmed by pharmacology

The main approach for identification of NR2 subunit expression by different neurons in our study was to establish and apply a ratio assay for NMDA receptor-mediated currents recorded in 100 μM Mg²⁺. After identifying conditions to minimize the change of the g(−90 mV)/g(MIC) ratio during activation of NMDA receptors, including recording in the presence of Ba²⁺ and using low concentrations of agonist, it was possible to repeatedly measure the g(−90 mV)/g(MIC) throughout the duration of NMDA application with minimal variation in the ratio in many of the neurons tested. The ratio values observed, particularly in NK1R+ neurons, indicated that functional receptors composed of NR2A/B and NR2C/D subunits are present. The reversible shift of the ratios to larger values in the presence of the NR2A/B antagonist, EAB318, and to smaller values in the presence of the NR2C/D preferring antagonist, UBP141, confirm this interpretation. In addition, the action of these drugs in shifting the measured g(−90 mV)/g(MIC) ratio in the predicted direction strongly supports the use of this ratio assay to identify natively expressed NR2 subunits.

NR2A/B and NR2C/D subunits expressed by subpopulations of lamina I neurons

While published evidence suggests expression of both NR2A/B and NR2C/D subunit types in the dorsal horn generally, our data, collected on subpopulations of lamina I neurons, show cell specific differences. Previous reports indicate that NMDA receptors with NR2A and NR2B subunits are expressed in superficial dorsal horn based on in situ hybridization [31 –33], single cell PCR [34] and immunostaining [14,35,36] studies. Our observations show strong evidence of NR2A and/or NR2B expression in both NK1R+ and NK1R- lamina I neurons. Earlier studies suggest that NR2D mRNA is expressed by many and NR2C mRNA by few dorsal horn neurons [34]. In addition, more NR2D mRNA is expressed in adult dorsal horn and embryonic spinal cord than NR2C mRNA [5,37,38]. Further support for the presence of NR2D is that NMDA receptors with NR2D-like single channel conductance have been reported for lamina II neurons in rat dorsal horn [39,40]. Based on our experiments, we have found that NK1R+ neurons express NR2C/D subunits more strongly than the NK1R- neurons. While it remains uncertain which NMDA receptors with low Mg²⁺ sensitivity are expressed by these lamina I neurons, NR2D is the best candidate.

NK1R+ lamina I neurons represent a comparatively uniform population of neurons that are predominantly projection neurons [21]. The NK1R- neuron population is heterogeneous, including inhibitory and excitatory interneurons as well as a small population of NK1R- projection neurons [41]. Within the NK1R- population of neurons, some of the variability of NR2 subunit identity may represent different receptor configurations on different subpopulations of dorsal horn neurons.

At the whole cell level, particularly for NK1R+ neurons, we have evidence that NMDA receptors with NR2C/D subunits are present. NMDA receptors with these less Mg²⁺ sensitive NR2 subunits could be expressed at synapses, extrasynaptically or both. Momiyama (2000) has suggested an extrasynaptic localization of NR2D containing NMDA receptors by lamina II neurons in the dorsal horn. Because of their higher binding affinity with glutamate, these extrasynaptic receptors may be more sensitive to ambient glutamate levels in the extracellular space that could accumulate due to glial release [42,43], spill over associated with high amounts of activity, and to injury [44]. Activation of these receptors would be expected to have a potent impact on neuronal cell function due to their lowered Mg²⁺ sensitivity, prolonged time over which they open following glutamate binding, and lack of desensitization [5,6,45].

Other factors that could influence NMDA receptor conductance ratio

One concern with our approach to NR2 subunit identification is the possibility that changes in membrane currents secondary to NMDA receptor activation will alter the g(−90 mV)/g(MIC). It is because of this concern that we have recorded in low Ca²⁺ solution with added Ba²⁺ and limited our analysis to those neurons showing no change in g(−90 mV)/g(MIC) while NMDA washes on and off the spinal cord slices. Even more importantly, we have used pharmacological tools as an independent test of subunit composition under these carefully controlled drug application conditions. In some of the neurons excluded from these studies, NMDA-induced currents showed strongly increased g(−90 mV)/g(MIC) ratios during wash-out of NMDA (data not shown). The underlying mechanism for this is not clear. For the data that met the criteria for our study, we have confirmed identification of subunit composition by the use of NMDA receptor specific compounds to alter conductance ratio in predictable ways. The opposing effects of EAB318 and UBP141 on g(−90 mV)/g(MIC) supports our interpretation of conductance ratio in terms of subunit composition.

Significance

We have taken advantage of the Mg²⁺ sensitivity of NMDA receptors to identify NMDA receptors of different NR2 subunits in identified subpopulations of lamina I neuorns and confirmed this with pharmacology. We show that individual neurons express NMDA receptors with different NR2 subunits at different ratios. When comparing identified populations of lamina I neurons, NK1R+ neurons express a higher mean ratio of NR2C/D type NMDA receptors compared with NK1R- neurons. NR2D has been suggested to have a role in the development of allodynia or hyperalgesia in several different pain models [12] and lamina I, NK1R+ neurons are importantly involved in the expression of allodynia [46]. In this context, it is possible that these receptors may contribute to development of NR2D-dependent allodynia.

Competing interests

The authors declare that they have no competing interests.

Authors' contributions

CKT, EJK and ABM conceived and designed the experiments. CKT and ABM analyzed the data and wrote the manuscript. CKT carried out the experiments. EJK provided the essential compounds UBP141 and EAB318. All authors read and approved the final manuscript.

Footnotes

Acknowledgements

This work was supported by a National Institutes of Health grant NS 029797 and Wyeth Research.

References

Ikeda

Heinke

Ruscheweyh

Sandkuhler

: Synaptic plasticity in spinal lamina I projection neurons that mediate hyperalgesia. Science 2003, 299:1237–1240.

Petrenko

Yamakura

Baba

Shimoji

: The role of N-methyl-D-aspartate (NMDA) receptors in pain: a review. Anesth Analg 2003, 97:1108–1116.

Cull-Candy

Leszkiewicz

: Role of distinct NMDA receptor subtypes at central synapses. Sci STKE 2004:re16.

Nakanishi

: Molecular diversity of glutamate receptors and implications for brain function. Science 1992, 258:597–603.

Monyer

Burnashev

Laurie

Sakmann

Seeburg

: Developmental and regional expression in the rat brain and functional properties of four NMDA receptors. Neuron 1994, 12:529–540.

Kuner

Schoepfer

: Multiple structural elements determine subunit specificity of Mg2+ block in NMDA receptor channels. J Neurosci 1996, 16:3549–3558.

South

Kohno

Kaspar

Hegarty

Vissel

Drake

Ohata

Jenab

Sailer

Malkmus

Masuyama

Horner

Bogulavsky

Gage

Yaksh

Woolf

Heinemann

Inturrisi

: A conditional deletion of the NR1 subunit of the NMDA receptor in adult spinal cord dorsal horn reduces NMDA currents and injury-induced pain. Journal of Neuroscience 2003, 23:5031–5040.

Shimoyama

Davis

Monaghan

Inturrisi

: An antisense oligonucleotide to the N-methyl-D-aspartate (NMDA) subunit NMDAR1 attenuates NMDA-induced nociception, hyperalgesia, and morphine tolerance. Journal of Pharmacology & Experimental Therapeutics 2005, 312:834–840.

Garraway

Inturrisi

: Design and Evaluation of Small Interfering RNAs That Target Expression of the N-Methyl-D-aspartate Receptor NR1 Subunit Gene in the Spinal Cord Dorsal Horn. J Pharmacol Exp Ther 2007, 322:982–988.

10.

Minami

Okuda-Ashitaka

Mori

Sakimura

Watanabe

Mishina

Ito

: Characterization of nociceptin/orphanin FQ-induced pain responses in conscious mice: neonatal capsaicin treatment and N-methyl-D-aspartate receptor GluRepsilon subunit knockout mice. Neuroscience 2000, 97:133–142.

11.

Minami

Matsumura

Okuda-Ashitaka

Shimamoto

Sakimura

Mishina

Mori

Ito

: Characterization of the glutamatergic system for induction and maintenance of allodynia. Brain Res 2001, 895:178–185.

12.

Hizue

Pang

Yokoyama

: Involvement of N-methyl-D-aspartate-type glutamate receptor epsilon1 and epsilon4 subunits in tonic inflammatory pain and neuropathic pain. Neuroreport 2005, 16:1667–1670.

13.

Petrenko

Yamakura

Baba

Sakimura

: Unaltered pain-related behavior in mice lacking NMDA receptor GluRepsilon 1 subunit. Neurosci Res 2003, 46:199–204.

14.

Boyce

Wyatt

Webb

O'Donnell

Mason

Rigby

Sirinathsinghji

Hill

Rupniak

: Selective NMDA NR2B antagonists induce antinociception without motor dysfunction: correlation with restricted localisation of NR2B subunit in dorsal horn. Neuropharmacology 1999, 38:611–623.

15.

Abe

Matsumura

Katano

Mabuchi

Takagi

Yamamoto

Hattori

Yagi

Watanabe

Nakazawa

Yamamoto

Mishina

Nakai

Ito

: Fyn kinase-mediated phosphorylation of NMDA receptor NR2B subunit at Tyr1472 is essential for maintenance of neuropathic pain. Eur J Neurosci 2005, 22:1445–1454.

16.

Kutsuwada

Sakimura

Manabe

Takayama

Katakura

Kushiya

Natsume

Watanabe

Inoue

Yagi

Aizawa

Arakawa

Takahashi

Nakamura

Mori

Mishina

: Impairment of suckling response, trigeminal neuronal pattern formation, and hippocampal LTD in NMDA receptor epsilon 2 subunit mutant mice. Neuron 1996, 16:333–344.

17.

Sprengel

Suchanek

Amico

Brusa

Burnashev

Rozov

Hvalby

Jensen

Paulsen

Andersen

Kim

Thompson

Sun

Webster

Grant

Eilers

Konnerth

McNamara

Seeburg

: Importance of the intracellular domain of NR2 subunits for NMDA receptor function in vivo. Cell 1998, 92:279–289.

18.

Taniguchi

Shinjo

Mizutani

Shimada

Ishikawa

Menniti

Nagahisa

: Antinociceptive activity of CP-101,606, an NMDA receptor NR2B subunit antagonist. Br J Pharmacol 1997, 122:809–812.

19.

Minami

Sugatani

Sakimura

Abe

Mishina

Ito

: Absence of prostaglandin E2-induced hyperalgesia in NMDA receptor epsilon subunit knockout mice. Br J Pharmacol 1997, 120:1522–1526.

20.

Minami

Okuda-Ashitaka

Hori

Sakuma

Sugimoto

Sakimura

Mishina

Ito

: Involvement of primary afferent C-fibres in touch-evoked pain (allodynia) induced by prostaglandin E2. European Journal of Neuroscience 1999, 11:1849–1856.

21.

Todd

McGill

Shehab

: Neurokinin 1 receptor expression by neurons in laminae I, III and IV of the rat spinal dorsal horn that project to the brainstem. Eur J Neurosci 2000, 12:689–700.

22.

Hunt

Mantyh

: The molecular dynamics of pain control. Nat Rev Neurosci 2001, 2:83–91.

23.

Chapman

Buritova

Honore

Besson

: Physiological contributions of neurokinin 1 receptor activation, and interactions with NMDA receptors, to inflammatory-evoked spinal c-Fos expression. J Neurophysiol 1996, 76:1817–1827.

24.

Aicher

Sharma

Cheng

Pickel

: The N-methyl-D-aspartate (NMDA) receptor is postsynaptic to substance P-containing axon terminals in the rat superficial dorsal horn. Brain Res 1997, 772:71–81.

25.

Labrakakis

MacDermott

: Neurokinin receptor 1-expressing spinal cord neurons in lamina I and III/IV of postnatal rats receive inputs from capsaicin sensitive fibers. Neurosci Lett 2003, 352:121–124.

26.

Tong

MacDermott

: Both Ca2+-permeable and -impermeable AMPA receptors contribute to primary synaptic drive onto rat dorsal horn neurons. J Physiol 2006, 575:133–144.

27.

Sun

Chiu

Kowal

Simon

Smeyne

Zukin

Olney

Baudy

Lin

: EAA-090 (2-[8,9-dioxo-2,6-diazabicyclo [5.2.0]non-1(7)-en2-yl]ethylphosphonic acid) and EAB-318 (R-alpha-amino-5-chloro-1–(phosphonomethyl)-1H-benzimidazole-2-propanoic acid hydrochloride). Journal of Pharmacology & Experimental Therapeutics 2004, 310:563–570.

28.

Morley

Tse

Feng

Miller

Monaghan

Jane

: Synthesis and pharmacology of N1-substituted piperazine-2,3-dicarboxylic acid derivatives acting as NMDA receptor antagonists. Journal of Medicinal Chemistry 2005, 48:2627–2637.

29.

Nowak

Bregestovski

Ascher

Herbet

Prochiantz

: Magnesium gates glutamate-activated channels in mouse central neurones. Nature 1984, 307:462–465.

30.

Mayer

Westbrook

Guthrie

: Voltage-dependent block by Mg2+ of NMDA responses in spinal cord neurones. Nature 1984, 309:261–263.

31.

Luque

Bleuel

Malherbe

Richards

: Alternatively spliced isoforms of the N-methyl-D-aspartate receptor subunit 1 are differentially distributed within the rat spinal cord. Neuroscience 1994, 63:629–635.

32.

Watanabe

Mishina

Inoue

: Distinct spatiotemporal distributions of the N-methyl-D-aspartate receptor channel subunit mRNAs in the mouse cervical cord. Journal of Comparative Neurology 1994, 345:314–319.

33.

Shibata

Watanabe

Ichikawa

Inoue

Koyanagi

: Different expressions of alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid and N-methyl-D-aspartate receptor subunit mRNAs between visceromotor and somatomotor neurons of the rat lumbosacral spinal cord. Journal of Comparative Neurology 1999, 404:172–182.

34.

Karlsson

Sjodin

Angeby Moller

Johansson

Wikstrom

Nasstrom

: Glutamate-induced currents reveal three functionally distinct NMDA receptor populations in rat dorsal horn – effects of peripheral nerve lesion and inflammation. Neuroscience 2002, 112:861–868.

35.

Yung

: Localization of glutamate receptors in dorsal horn of rat spinal cord. Neuroreport 1998, 9:1639–1644.

36.

Nagy

Watanabe

Fukaya

Todd

: Synaptic distribution of the NR1, NR2A and NR2B subunits of the N-methyl-d-aspartate receptor in the rat lumbar spinal cord revealed with an antigen-unmasking technique. European Journal of Neuroscience 2004, 20:3301–3312.

37.

Tolle

Berthele

Zieglgansberger

Seeburg

Wisden

: The differential expression of 16 NMDA and non-NMDA receptor subunits in the rat spinal cord and in periaqueductal gray. Journal of Neuroscience 1993, 13:5009–5028.

38.

Akazawa

Shigemoto

Bessho

Nakanishi

Mizuno

: Differential expression of five N-methyl-D-aspartate receptor subunit mRNAs in the cerebellum of developing and adult rats. J Comp Neurol 1994, 347:150–160.

39.

Momiyama

: Distinct synaptic and extrasynaptic NMDA receptors identified in dorsal horn neurones of the adult rat spinal cord. J Physiol 2000:621–628.

40.

Green

Gibb

: Characterization of the single-channel properties of NMDA receptors in laminae I and II of the dorsal horn of neonatal rat spinal cord. European Journal of Neuroscience 2001, 14:1590–1602.

41.

Laing

Todd

Heizmann

Schmidt

: Subpopulations of GABAergic neurons in laminae I-III of rat spinal dorsal horn defined by coexistence with classical transmitters, peptides, nitric oxide synthase or parvalbumin. Neuroscience 1994, 61:123–132.

42.

Erreger

Geballe

Kristensen

Chen

Hansen

Lee

Yuan

Lyuboslavsky

Micale

Jorgensen

Clausen

Wyllie

Snyder

Traynelis

: Subunit-specific agonist activity at NR2A-, NR2B-, NR2C-, and NR2D-containing N-methyl-D-aspartate glutamate receptors. Mol Pharmacol 2007, 72:907–920.

43.

Lee

Mannaioni

Yuan

Woo

Gingrich

Traynelis

: Astrocytic control of synaptic NMDA receptors. J Physiol 2007, 581:1057–1081.

44.

Park

Velumian

Fehlings

: The role of excitotoxicity in secondary mechanisms of spinal cord injury: a review with an emphasis on the implications for white matter degeneration. J Neurotrauma 2004, 21:754–774.

45.

Wyllie

Behe

Colquhoun

: Single-channel activations and concentration jumps: comparison of recombinant NR1a/NR2A and NR1a/NR2D NMDA receptors. J Physiol 1998, 510:1–18.

46.

Nichols

Allen

Rogers

Ghilardi

Honore

Luger

Finke

Lappi

Simone

Mantyh

: Transmission of chronic nociception by spinal neurons expressing the substance P receptor. Science 1999, 286:1558–1561.