NMDA receptor antagonism blocks the cardiovascular responses to microinjection of trans-ACPD into the NTS of awake rats

The possible interaction of glutamatergic metabotropic agonists and N-methyl- d -aspartate (NMDA) receptors was investigated in the nucleus tractus solitarii (NTS) of awake rats. The cardiovascular responses to unilateral microinjection of trans -1-amino-1,3-cyclopentanediocarboxylic acid ( trans -ACPD; 250 pmol/50 nL) into the NTS ( n = 8) produced hypotension (−64 ± 4 mmHg) and bradycardic (−206 ± 11 bpm) responses, which were blocked by previous microinjection of 2-amino-5-phosphonovaleric acid (AP-5; 10 nmol/50 nL), a selective antagonist of NMDA ionotropic receptors, into the same site. Intravenous injection of methyl-atropine blocked both the bradycardic and hypotensive responses to microinjection of trans -ACPD into the NTS, indicating that the hypotension was secondary to the intense bradycardic response. The data also showed that the bradycardic and hypotensive responses to microinjection of an NMDA agonist (10 pmol/50 nL) into the NTS were not affected by previous microinjection of α-methyl-4-carboxyphenylglycine (MCPG; 5 nmol/50 nL), a non-selective antagonist of metabotropic receptors. The results showing that the cardiovascular responses to microinjection of trans -ACPD into the NTS were blocked by AP-5 indicate that the responses to metabotropic agonists in the NTS involves NMDA receptors.     

Electrophysiological evidence for N-methyl-D-aspartate excitatory amino acid receptors in the rat supraoptic nucleus in vitro.

The responses of neurons in slices of the rat supraoptic nucleus (SON) to afferent stimulation were recorded under current-clamp conditions. In magnesium (Mg2+)-free incubation medium, synaptic responses were prolonged and were partially antagonized by the non-competitive N-methyl-D-aspartate (NMDA) receptor antagonist (+)-5-methyl-10,11-dihydro-5H- dibenzo[a,d]cyclohepten-5,10-imine hydrogen maleate (MK-801). During blockade of non-NMDA excitatory amino acid (EAA) receptors, the synaptic responses in Mg(2+)-free medium were blocked by the competitive NMDA receptor antagonist 2-amino-5-phosphonopentanoic acid (AP5). The results of these experiments provide electrophysiological evidence for the existence of NMDA receptors in the rat SON.     

Evidence for gamma-aminobutyric acid receptor-mediated modulation of the aortic baroreceptor reflex in the nucleus tractus solitarii of the rat

Microinjections of the gamma-aminobutyric acid (GABA) receptor antagonist bicuculline into the medial area of the nucleus tractus solitarii of the rat enhanced the depressor and bradycardiac responses to aortic nerve stimulation whereas the glycine receptor antagonist strychnine did not affect them. The GABA receptor agonist muscimol and the GABA uptake inhibitor nipecotic acid reduced the responses to aortic nerve stimulation. These results provide evidence suggesting GABA receptor-mediated modulation of the aortic baroreceptor reflex in the nucleus tractus solitarii of the rat.     

NMDA and non-NMDA receptors contribute to synaptic transmission between the medial geniculate body and the lateral nucleus of the amygdala

We examined whether the NMDA class of excitatory amino acid receptors contribute to synaptic transmission in the pathway connecting the medial geniculate body (MGB) with the lateral nucleus of the amygdala (LA) using extracellular single unit recordings and microiontophoresis. Cells were identified in LA on the basis of responsivity to electrical stimulation of the MGB. For each cell, a level of current was found for the iontophoretic ejection of the NMDA antagonist AP5 that blocked responses elicited by iontophoresis of NMDA, but had no effect on responses elicited by AMPA. Iontophoresis of AP5 with this level of current blocked the excitatory response elicited by MGB stimulation in most cells tested. Microinfusion of AP5 (25, 50, or 100 M) also blocked the responses. Additional studies tested individual cells with both AP5 and the AMPA antagonist CNQX and showed that blockade of either NMDA or AMPA receptors interferes with synaptic transmission. Finally, iontophoretic ejection of either AP5 or CNQX blocked short-latency (<25 ms) responses elicited in LA by peripheral auditory stimulation. Together, these results suggest that the synaptic evocation of action potentials in the thalamo-amygdala pathway depends on both NMDA and non-NMDA receptors. We hypothesize that non-NMDA receptors are most likely required to depolarize the cell sufficiently to remove the blockade of NMDA channels by magnesium and NMDA receptors are required to further depolarize the membrane to the level required for action potential generation.     

Role of nucleus tractus solitarius 5-HT3 receptors in the defense reaction-induced inhibition of the aortic baroreflex in rats

Different stressful conditions elicit a typical behavior called the defense reaction. Our aim was to determine whether 5-HT3 receptors in the nucleus tractus solitarius (NTS) are involved in 1) the inhibition of the baroreflex bradycardia and 2) the rise in blood pressure, which are known to occur during the defense reaction. In urethane-anesthetized rats, the defense reaction was elicited by electrical stimulation of the dorsomedial nucleus of the hypothalamus (DMH) or the dorsal part of the periaqueductal gray (dPAG). Direct electrical stimulation of the aortic depressor nerve was used to trigger the typical baroreflex responses. Aortic stimulation at high (100-150 microA) and low (50-90 microA) intensity produced a decrease in heart rate of -39 to -44% (relative to baseline, Group 1 responses, n = 113) and -19 to -24% (Group 2 responses, n = 43), respectively. In spontaneously breathing rats, Group 1 and Group 2 bradycardiac responses were inhibited during DMH (-75 +/- 4% and -96 +/- 4%, n = 38 and n = 11, respectively), as well as dPAG (-81 +/- 3% and -95 +/- 4%, n = 36 and n = 10, respectively) stimulation. The aortic baroreflex bradycardia was hardly affected by DMH or dPAG stimulation when bicuculline (5 pmol), a specific GABAA receptor antagonist, had previously been microinjected into the NTS. Likewise, NTS microinjections of granisetron, a specific 5-HT3 receptor antagonist, prevented, in a dose-dependent manner, the baroreflex bradycardia inhibition. In addition, intra-NTS granisetron did not affect the rise in blood pressure induced by either site stimulation. These data show that 5-HT3 receptors in the NTS are involved in the GABAergic inhibition of the aortic baroreflex bradycardia, but not in the rise in blood pressure, occurring during the defense reaction elicited by DMH or dPAG stimulation.     

The involvement of N-methyl-d-aspartate (NMDA) and non-NMDA receptors in the responsiveness of rat spinal neurons with input from the chronically inflamed ankle

Unilateral adjuvant inflammation was induced at the rat ankle 2 or 20 days before an evaluation of the contribution of N-methyl-d-aspartate (NMDA) and non-NMDA receptors to the processing of nociceptive information by wide dynamic range neurons in the spinal cord. Microionophoretic application of either the NMDA receptor antagonists ketamine and DL-2-amino-5-phosphonovalerate (AP5) or the non-NMDA receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) reduced the responses to innocuous and noxious mechanical stimulation of the inflamed ankle. The pattern of these effects was comparable to that in rats with acute inflammation suggesting that non-NMDA and NMDA receptors are similarly involved in acute, prolonged acute and chronic inflammation-evoked activity.     

Vasopressin antagonist in nucleus tractus solitarius/vagal area reduces pressor and tachycardia responses to paraventricular nucleus stimulation in rats

In urethane-anesthetized rats, injections of 50 pmol of arginine-vasopressin (AVP) or thyrotropin-releasing hormone (TRH) into a lateral cerebral ventricle (i.c.v.) elicit short-latency increases in blood pressure. i.c.v. injection of 50 pmol of the AVP antagonist, d(CH2)5Tyr(Me)AVP, but not of the vehicle (artificial cerebrospinal fluid; a CSF), abolished the pressor action of i.c.v. AVP. The AVP antagonist did not antagonize the TRH-induced pressor responses. In another group of rats, a monopolar stainless-steel electrode was positioned stereotaxically in the paraventricular nucleus (PVN) and pressor responses were elicited by electrical stimulation of the PVN. Micro-injection of 1 nmol of the AVP antagonist, but not of aCSF alone, into the nucleus tractus solitarius/vagal area (NTS/VA), reduced PVN-stimulated pressor responses to 26 +/- 6% of control and stimulation-induced tachycardia to 37.3 +/- 9.0% of control. These studies indicate that the pressor and heart-rate responses to PVN stimulation may be mediated, in part, via AVP receptors in the NTS/VA.     

Evidence for the presence of GABAergic and glycine-like systems responsible for cardiovascular control in the nucleus tractus solitarii of the rat

Microinjections of γ-aminobutyric acid (GABA) and glycine, into the medial area of the nucleus tractus solitarii (NTS) of the rat, led to an increase in arterial pressure and heart rate. The GABA receptor antagonist bicuculline and the glycine receptor antagonist strychnine decreased both of these cardiovascular parameters whereas the GABA uptake inhibitor nipecotic acid produced hypertension. High K⁺ stimulation caused a calcium-dependent release of GABA and glycine from tissues in the area of the NTS. Our results suggest that GABA and glycine may modulate the cardiovascular control within the NTS.     

Contributions of glutamate receptors to the maintenance of mustard oil-induced hyperalgesia in spinalized rats

N-methyl-d-aspartate (NMDA) receptors, which are widely distributed throughout the central nervous system, appear to play a critical role in several types of plasticity and long-term potentiation. In the pain system, increased sensitivity to somatosensory stimuli, known as hyperalgesia and allodynia, can arise from tissue damage or excessive C-fiber nociceptor activation. Previously, NMDA, non-NMDA ionotropic, and metabotropic glutamate receptors have been proposed to contribute to the sustained hyperalgesia following tissue injury or nociceptor activation. Although non-NMDA receptors appear to mediate both hyperalgesia and normal (nonhyperalgesic) responses and behavior, NMDA receptors have been reported to participate only in hyperalgesic responses. In contrast, other studies have implicated NMDA receptors in both hyperalgesic and normal responses. The aim of this study was to critically compare the effects of the glutamate receptor antagonists ketamine and 2-amino-5-phosphonovaleric acid (APV; NMDA receptor antagonists), 6,7-dinitroquinoxaline-2,3-dione (DNQX; non-NMDA ionotropic receptor antagonist), and 2-amino-3-phosphonopropionic acid (AP3; metabotropic receptor antagonist) on intra-articular mustard oil-induced facilitation of flexion withdrawal reflexes in spinalized rats. Our results showed that, as expected from previous studies, ketamine, APV, and DNQX dose-dependently inhibited the flexion withdrawal reflex evoked by C-fiber electrical stimulation of the sciatic nerve. Surprisingly, however, ketamine, APV, and DNQX also inhibited flexion withdrawal reflexes in normal (nonhyperalgesic) rats with similar ED50s. In contrast, AP3 had no effect in either hyperalgesic or normal rats. These results demonstrate that NMDA and non-NMDA ionotropic, but not metabotropic, glutamate receptors contribute without preference to both facilitated and normal flexion withdrawal reflexes evoked by high-intensity electrical stimulation in the spinalized rat. Thus, the apparent preference of NMDA receptors for hyperalgesic states seen in some previous studies on nociception, as well as in other model systems, may have arisen from differences in experimental paradigm, such as the intensity of sensory stimulation or excitability of the spinal cord, coupled with the voltage dependency of the NMDA conductance. Received: 15 November / Accepted: 12 June 1997     

Developmental changes in NMDA receptor-mediated visual activity in the rat superior colliculus, and the effect of dark rearing

N-methyl-D-aspartate (NMDA) receptor-mediated activity is considered important for experience-dependent plasticity in the developing visual system. We investigated the influence of age and experience on the role of NMDA receptors in the visual transmission in the superficial grey layer of the superior colliculus (SGS) of the superior colliculus, where, in the adult, NMDA receptors mediate a substantial part of the visual response. In normally reared (postnatal day 14, P14, to adult) rats, visual responses were challenged with NMDA receptor-selective iontophoretic applications of the antagonist D-2-amino-5-phosphonovalerate (AP5). After eye opening (at P14), there was a significant increase in the number of neurones whose visual responses were reduced during AP5 ejection, which peaked at P22 (85%; n = 21), and then declined to adult levels (66%; n = 47) at P25. The mean reduction of the response (from control levels) by AP5 was similar at all ages (approximately 40%). Dark rearing had striking effects on the role of NMDA receptors in visual transmission, especially when comparisons were made between age-matched subjects greater than P25. In these subjects, AP5 ejection reduced the visual responses of all neurones studied. In addition, AP5 ejection caused a significantly larger reduction of visual responses in dark-reared rats (mean reduction 62 ± 4; n = 29) compared with age-matched controls (mean reduction 44 ± 8; n = 23). The D,L-alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) reduced the visual responses of every neurone studied and there were no age- or experience-dependent effects. We conclude that NMDA receptors, but not AMPA receptors, assume greater importance for visual transmission in the SGS of dark-reared rats. Received: 2 May 1997 / Accepted: 19 November 1997     

The area postrema and control of arterial blood pressure; Absence of hypertension after excision of the area postrema in rats

Summary Acute surgical excision of the area postrema (AP) in the rat failed to affect arterial blood pressure or heart rate.There was no effect on cardiovascular reflex responses during diving or on the heart rate responses to acute decreases or increases of blood pressure caused by bradykinin or angiotensin, respectively. Electrolytic lesions of the AP in acute experiments caused variable damage to the nucleus tractus solitarii (NTS). In these rats large variations in blood pressure occurred. Excision of the AP in a chronic experiment failed to change blood pressure, heart rate, water intake or plasma renin activity. In contrast, bilateral electrolytic lesions of the NTS at the level of the AP caused a severe acute hypertension and completely blocked cardiovascular reflex responses. Hypertension also existed in rats with NTS lesions studied for a longer period of time.These experiments failed to confirm the hypothesis that the AP exerts a tonic inhibitory control of basal blood pressure. Hypertension previously reported after ablation of the AP may be explained by damage to the NTS.     

DOPA cyclohexyl ester, a DOPA antagonist, blocks the depressor responses elicited by microinjections of nicotine into the nucleus tractus solitarii of rats

Nicotinic cholinergic receptors play a role in cardiovascular regulation in the lower brain stem. Herein, we present evidence that l-3,4-dihydroxyphenylalanine (DOPA), a putative neurotransmitter in the central nervous system, is involved in the depressor response to microinjection of nicotine into the nucleus tractus solitarii (NTS). Microinjection of nicotine into the medial area of the NTS led to decreases in arterial blood pressure and heart rate in anesthetized rats. Mecamylamine, a nicotinic receptor antagonist, microinjected into NTS, blocked the depressor and bradycardic responses to nicotine. Nicotine-induced depressor and bradycardic responses were blocked by DOPA cyclohexyl ester (DOPA CHE), an antagonist for DOPA. DOPA CHE did not modify the action of carbachol on excitatory postsynaptic potential in rat cortical slices. These results suggest that endogenous DOPA is involved in nicotine-induced depressor responses in the NTS of anesthetized rats.     

NMDA and AMPA receptors in the dorsal nucleus of the lateral lemniscus shape binaural responses in rat inferior colliculus

Binaural responses of single neurons in the rat's central nucleus of the inferior colliculus (ICC) were recorded before and after local injection of excitatory amino acid receptor antagonists (either 1,2, 3,4-tetrahydro-6-nitro-2,3-dioxo-benzo[f]quinoxaline-7-sulfonamide disodium [NBQX], (+/-)-3-(2-carboxypiperazin-4-yl)-propyl-1-phosphonic acid [CPP], 6-cyano-7-nitroquinoxaline-2,3-dione [CNQX], or (+/-)-2amino-5-phosphonovaleric acid [APV]) into the dorsal nucleus of the lateral lemniscus (DNLL). Responses were evoked by clicks delivered separately to the two ears at interaural time delays between -1.0 and +30 ms (positive values referring to ipsilateral leading contralateral click pairs). The neurons in our sample were excited by contralateral stimulation and inhibited by ipsilateral stimulation, and the probability of action potentials was reduced as the ipsilateral stimulus was advanced. Binaural inhibition resulted in response suppression that lasted up to 30 ms. Injection of excitatory amino acid antagonists into the DNLL contralateral to the recording site reduced the strength of binaural inhibition in the ICC. The alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor antagonist NBQX preferentially affected responses at small interaural time intervals (0-1.0 ms), whereas the N-methyl-D-aspartate (NMDA) antagonist CPP preferentially affected responses at longer intervals (1-30 ms). Both CNQX and APV produced a release from binaural inhibition, but neither drug was selective for specific intervals. The data support the idea that binaural inhibition in the rat ICC is influenced by both AMPA and NMDA receptor-mediated excitatory events in the contralateral DNLL. The results suggest that the AMPA receptors contribute selectively to the initial component of binaural inhibition and the NMDA receptors to a longer lasting component.     

Induction of stable long-term potentiation in the presence of the protein kinase C antagonist staurosporine.

We have studied the effects of staurosporine, an antagonist of the catalytic subunit of protein kinase C, on the mechanisms of long-term potentiation (LTP) in rat hippocampal slices maintained in vitro. Application of staurosporine did not affect pre-established LTP, but resulted in a decaying potentiation when high frequency stimulation was delivered in its presence. However, coactivation of two inputs to the same group of CA1 neurons during high frequency stimulation transformed the decaying potentiation into stable LTP. Staurosporine also reduced the NMDA receptor-mediated component of synaptic responses to burst stimulation. It is concluded that the PKC antagonist interferes with LTP induction, but not expression mechanisms.     

N-methyl-D-aspartate receptors in the amygdala are necessary for the acquisition and expression of conditioned defeat

Here, we describe a biologically relevant model called conditioned defeat that is used to examine behavioral responses to social defeat in Syrian hamsters. In this model experimental animals that are normally aggressive experience social defeat and consequently display high levels of submissive/defensive behavior even in response to non-threatening conspecifics. N-methyl-D-aspartate (NMDA) receptors within the amygdala play an important role in conditioned fear; therefore, the purpose of this study was to examine whether NMDA receptors within the amygdala are necessary for the acquisition and expression of conditioned defeat. Specifically, the present study examined whether bilateral infusions of the NMDA receptor antagonist DL-2-amino-5-phosphonopentanoic acid (AP5; 0.625, 1.25, 2.5, 5.0, 10.0 microg) into the amygdala would block the acquisition of conditioned defeat. Subsequently, we examined whether bilateral infusions of AP5 (0.625, 1.25, 2.5, 5.0 microg) into the amygdala prior to testing would block the expression of conditioned defeat. Infusions of AP5 into the amygdala immediately before the initial social defeat significantly reduced submissive/defensive behavior when hamsters were tested the following day with a non-aggressive intruder. Similarly, infusions of AP5 into the amygdala immediately before exposure to a non-aggressive intruder significantly attenuated the display of submissive/defensive behavior. These data demonstrate that NMDA receptors are necessary for both the acquisition and expression of conditioned defeat. We believe that conditioned defeat is a unique and valuable animal model with which to investigate the neurobiology of fear-related changes in social behavior.     

Effects of excitatory amino acid antagonists on synaptic responses of supraoptic neurons in slices of rat hypothalamus

1. Intracellular recordings from magnocellular neurons in the supraoptic nucleus (SON) were obtained from rat hypothalamic slices to determine the effects of specific transmitter antagonists on evoked postsynaptic potentials (PSPs), action potential after-discharge, and spontaneously occurring PSPs. 2. Broad-spectrum excitatory amino acid (EAA) antagonists, kynurenic acid (KYN) and gamma-d-glutamylglycine (DGG), significantly diminished or eliminated electrically evoked depolarizing PSPs and spike discharges. These compounds also greatly reduced the amplitude and frequency of spontaneous PSPs. 3. The specific N-methyl-D-aspartate (NMDA) receptor antagonist, DL-2-amino-5-phosphonopentanoic acid (AP5), did not significantly reduce these measures of synaptic activation under these experimental conditions. 4. The gamma-aminobutyric acid (GABA) antagonist, bicuculline methiodide (BIC), partially antagonized some PSPs when the cells were hyperpolarized (-75 to -80 mV) with steady injected currents; KYN antagonized BIC-resistant PSPs. 5. The involvement of a hypothetical cholinergic input to the SON in the responses to stimulation of the region dorsolateral to the SON was tested by bath application of nicotinic cholinergic antagonists, particularly d-tubocurarine (dTC). Nicotinic cholinergic antagonists, even after prolonged exposure to high concentrations, did not block the responses of SON cells to dorsolateral stimulation. 6. These findings strongly suggest that EAAs mediate fast excitatory synaptic responses of SON neurons to stimulation of cells and axons in the region dorsolateral to the SON. The blockade of almost all spontaneous EPSPs by broad-spectrum EAA antagonists likewise argues that EAAs are responsible for the majority of ongoing fast excitatory input. These responses appear to involve an interaction with kainate- and/or quisqualate-type EAA receptors.     

Contribution of AMPA and NMDA receptors to early and late phases of LTP in hippocampal slices

Alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) and N-methyl-D-aspartate (NMDA) receptor mediated responses were investigated in rat hippocampal slices under 4h of long-term potentiation (LTP) expression. A modified medium containing the NMDA receptor antagonist AP5 and low concentration of Mg(2+) was used to monitor isolated AMPA responses. NMDA components were determined from composite excitatory postsynaptic potentials (EPSPs) under brief (15-20 min) wash-out of AP5. LTP was induced in a medium with low concentration of AP5, resulting in an about two-fold larger increase of the AMPA component than of the NMDA component at both 1h and 4h after induction. Similar results were obtained if LTP was induced in "normal Mg(2+)" and the NMDA components were assessed at the end of experiment, from either composite or isolated NMDA EPSPs, with or without blockade of GABAergic inhibition. It is generally believed that LTP undergoes biochemical and/or structural conversions during the first few hours. Our study, however, shows constant expression of LTP, at least in terms of AMPA versus NMDA components, during this time. The data support the notion that LTP initiates as a predominant amplification of AMPA receptors and remains so for at least 4h.     

Modulatory effects of rat endothelin on central cardiovascular control in rats

To characterize the modulatory action of rat endothelin (endothelin-3 or ET-3) on the cardiovascular control by the central nervous system (CNS), ET-3 was injected into the cisterna magna of urethane-anesthetized and immobilized rats. An injection of 100 pmol of ET-3 caused immediate rises in arterial pressure (AP), renal nerve activity (RNA), and heart rate (HR). These variables subsequently decreased and, in 5-20 min, fell below the pre-injection level. Simultaneously, the arterial baroreceptor reflex was almost totally suppressed. Although RNA and HR subsequently returned to, or often exceeded, pre-injection levels in 20 to 60 min and reflex activity recovered, AP sometimes remained below control for at least 2 h. A similar pattern of changes was elicited in unanesthetized precollicular decerebrated rats. The responses to ET-3 were abolished by hexamethonium chloride, but were not conspicuously altered by arginine vasopressin antagonist or angiotensin II antagonist. The CNS sites responsible for ET-induced changes were subsequently searched. Topical application of ET-3 to the ventral surface of the medulla (VSM) caused the pattern of changes in AP, RNA, and HR similar to that following intracisternal injection. Microinjection of ET-3 into the nucl. tractus solitarius (NTS) increased AP and RNA, whereas intrathecal administration of it decreased them. We conclude that intracisternally administered ET-3 centrally modulates both tonic and reflex control of AP by the sympathetic nervous system and that the VSM appears to be primarily responsible for the modulation, although NTS and spinal cord may also be involved in it.     

Activation of N-methyl-D-aspartate receptors contributes to the EPSP at perforant path synapses in the rat dentate gyrus in vitro

The excitatory postsynaptic potential (EPSP) evoked in the granule cells of the rat dentate gyrus following low frequency stimulation of the perforant path has been investigated using intracellular recording. The EPSP was reduced by low microM concentrations of the non-N-methyl-D-aspartate (NMDA) receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX). A small CNQX-resistant component of the EPSP remained. This could be blocked by the NMDA receptor antagonist (+/-)-2-amino-5-phosphonovalerate, was enhanced in Mg2+-free medium and showed a potential-dependency characteristic of the activation of NMDA ionophores. These results demonstrate that NMDA receptors contribute to the EPSP in the granule cell.     

Excitatory amino acid-mediated components of synaptically evoked input from dorsal roots to deep dorsal horn neurons in the rat spinal cord slice

The participation of N-methyl-D-aspartate (NMDA) and non-NMDA receptors in the responses of deep dorsal horn neurons to single shock stimulation of dorsal roots was investigated using current- and voltage-clamp techniques. In the presence of Mg2+, superfusion of rat spinal slices with 6-cyano-2,3-dihydroxy-7-nitroquinoxaline (CNQX), a potent antagonist of non-NMDA receptors, reversibly blocks fast excitatory synaptic responses elicited by low-frequency stimulation of dorsal roots and to a greater extent the responses to quisqualate than to kainate or NMDA. The synaptic response elicited in a zero-Mg2+ medium is less sensitive to CNQX. The CNQX-resistant component is however abolished by D-APV, a selective antagonist of NMDA receptor. Under voltage-clamp, the excitatory postsynaptic currents also showed an initial fast (CNQX-sensitive) and a late slow (2-amino-5-phosphonovalerate (APV)-sensitive, Mg2+-sensitive) component, both of which had similar thresholds but differed in their latency, time-to-peak and duration. These results support the concept that both non-NMDA and NMDA receptor channels are present in a majority of deep dorsal horn neurons and could be simultaneously activated by transmitter released from stimulated primary afferents.