NR2 subunits and NMDA receptors on lamina II inhibitory and excitatory interneurons of the mouse dorsal horn

Abstract

Background: NMDA receptors expressed by spinal cord neurons in the superficial dorsal horn are involved in the development of chronic pain associated with inflammation and nerve injury. The superficial dorsal horn has a complex and still poorly understood circuitry that is mainly populated by inhibitory and excitatory interneurons. Little is known about how NMDA receptor subunit composition, and therefore pharmacology and voltage dependence, varies with neuronal cell type. NMDA receptors are typically composed of two NR1 subunits and two of four NR2 subunits, NR2A-2D. We took advantage of the differences in Mg2+ sensitivity of the NMDA receptor subtypes together with subtype preferring antagonists to identify the NR2 subunit composition of NMDA receptors expressed on lamina II inhibitory and excitatory interneurons. To distinguish between excitatory and inhibitory interneurons, we used transgenic mice expressing enhanced green fluorescent protein driven by the GAD67 promoter.

Results: Analysis of conductance ratio and selective antagonists showed that lamina II GABAergic interneurons express both the NR2A/B containing Mg2+ sensitive receptors and the NR2C/D containing NMDA receptors with less Mg2+ sensitivity. In contrast, excitatory lamina II interneurons express primarily NR2A/B containing receptors. Despite this clear difference in NMDA receptor subunit expression in the two neuronal populations, focally stimulated synaptic input is mediated exclusively by NR2A and 2B containing receptors in both neuronal populations.

Conclusions: Stronger expression of NMDA receptors with NR2C/D subunits by inhibitory interneurons compared to excitatory interneurons may provide a mechanism to selectively increase activity of inhibitory neurons during intense excitatory drive that can provide inhibitory feedback.

Figure 1

Analysis of the voltage dependence of NMDA receptors expressed in EGFP+ and EGFP- cells in the presence of 100 μM Mg²⁺. A, the current - voltage relationship recorded at the peak response to NMDA application in an EGFP+ and an EGFP- neuron. The maximal inward current (MIC) and voltage of maximal inward current (VMIC) are indicated by dashed lines. The conductance of NMDA receptors at -90 mV (g(-90 mV)) and conductance of maximal inward current (g(MIC)) are indicated by dotted lines. B, average g(-90 mV)/g(MIC) in EGFP+ and EGFP- interneurons (*P < 0.05, upper box) and the average VMIC in EGFP+ and EGFP- interneurons (*P < 0.05, lower box). Perforated lines indicate the conductance ratio of NR2A/B or NR2C/D receptor induced currents and VMIC derived from oocyte expression data of Kuner and Schoepfer [19].

Figure 2

Effect of NR2C/D preferring antagonist on NMDA induced current recorded from EGFP+ neurons. A, NMDA induced inward current in an EGFP+ dorsal horn neuron. Current responses to voltage ramps during NMDA application are cut off at the peak of the response to allow direct comparison of the NMDA response to that in B. Current - voltage relationship obtained at the peak of the NMDA response is plotted below with the g(-90 mV) and g(MIC) indicated by dotted lines and VMIC indicated by dashed lines. B, Current responses to voltage ramps during co-application of 15 μM NMDA and 25 μM UBP141, an NR2C/D preferring antagonist, to the same cell as in A. Current - voltage relationship from the peak response is again plotted below with the g(-90 mV) and g(MIC) indicated by dotted lines and VMIC indicated by dashed lines. C shows both individual and average g(-90 mV)/g(MIC) values in the absence and presence of UBP141 (*P < 0.05). D indicates individual and average VMIC values in the absence and presence of UBP141 (*P < 0.05). For C and D, perforated lines indicate the conductance ratio of NR2A/B or NR2C/D receptor induced currents and VMIC derived from oocyte expression data of Kuner and Schoepfer [19].

Figure 3

Effect of NR2A/B preferring antagonist on NMDA induced current recorded from EGFP+ neurons. A, NMDA induced inward current in an EGFP+ dorsal horn neuron. Current responses to voltage ramps during NMDA application are cut off at the peak of the response to allow direct comparison of the NMDA response to that in B. Current - voltage relationship obtained at the peak of the NMDA response is plotted below with the g(-90 mV) and g(MIC) indicated by dotted lines and VMIC indicated by dashed lines. B, current responses to voltage ramps during co-application of 15 μM NMDA and 100 nM EAB318, an NR2A/B selective antagonist, to the same cell as in A with the current -voltage relationship shown below. C, both individual and average g(-90 mV)/g(MIC) values in the absence and presence of EAB318 (*P < 0.05) are shown. D shows individual and average VMIC values in the absence and presence of EAB318 (*P < 0.05). For C and D, perforated lines indicate the conductance ratio of NR2A/B or NR2C/D receptor induced currents and VMIC derived from oocyte expression data of Kuner and Schoepfer [19].

Figure 4

No significant effect of NR2A/B or NR2C/D preferring antagonists on NMDA induced currents recorded from EGFP- neurons. The g(-90 mV)/g(MIC) values and VMIC are plotted for individual neurons and average values in the absence and presence of 25 μM UBP141 or 100 nM EAB318. Neither UBP141 or EAB318 significantly changed g(-90 mV)/g(MIC) and VMIC values (ns; no significance). Perforated lines indicate the conductance ratio of NR2A/B or NR2C/D receptor induced currents and VMIC derived from oocyte expression data of Kuner and Schoepfer [19].

Figure 5

Analysis of voltage dependent sensitivity of synaptic NMDA receptors to 100 μM Mg²⁺ in lamina II interneurons. A, Representative traces of focal stimulation evoked NMDA-EPSCs recorded at different holding potentials (-90 mV to +50 mV at 20 mV increment). B, Current - voltage relationship of NMDA EPSCs shown in A. The reversal potential for this example neuron is 0 mV. C shows that synaptic NMDA EPSCs from EGFP+ (n = 18) and EGFP- (n = 19) neurons have similar g(-90 mV)/g(MIC) that are comparable to the value of NMDA receptors with NR2A/B subunits. Perforated lines indicate the conductance ratio of NR2A/B or NR2C/D receptor induced currents derived from oocyte expression data.

Figure 6

Effect of NR2A and NR2B antagonists on synaptic NMDA receptors. A, Representative traces show the inhibiting effects of EAB-318 (200 nM), ifenprodil (3 μM) and CPP (200 nM) on NMDA EPSCs evoked by focal stimulation and recorded from 3 EGFP+ neurons. Cells were held at +50 mV while EPSCs were recorded. B, shows the summary of NMDA receptor pharmacology recorded from EGFP+ and EGFP- neurons. The percentages of inhibition by EAB318, ifenprodil, and CPP on synaptic NMDA EPSCs were not significantly different between EGFP+ and EGFP- neurons (ns; not significant). We were able to obtain washout data on several of these neurons. The washout of antagonists ranged from 25% to 100% recovery of the blocked portion of the original NMDA EPSC amplitude. For the EGFP+ neurons, we were able to hold the neuron under study long enough to show washout in 6/16 neurons exposed to EAB318, 2/11 to CPP and 1/8 to ifenprodil. For the EGFP- neurons, we recorded partial to full recovery from 5/20 neurons exposed to EAB318, 2/14 neurons exposed to CPP and 2/8 neurons exposed to ifenprodil.