Decreased Uptake and Release of D-Aspartate in the Guinea Pig Spinal Cord After Partial Cordotomy

This study attempts to determine if L-glutamate and/or L-aspartate may be transmitters of neural tracts descending from the brain to the spinal cord. The uptake and electrically evoked release of D-[3H]aspartate, a putative marker for L-glutamate and L-aspartate, were measured in the cervical enlargement of the guinea pig spinal cord. These activities were compared using unlesioned animals and others with a lesion on the right side of the spinal cord. Partial cordotomy (segment C5) produced a heavy loss of descending fibers, a small loss of primary sensory fibers, and a depression of the uptake and the Ca2+ -dependent, electrically evoked release of D-aspartate ipsilateral and caudal to the lesion. Contralaterally, there was a moderate loss of corticospinal fibers, some loss of other descending axons, and a depression of D-aspartate release. Dorsal rhizotomy (segments C4-T1) produced a heavy loss of primary sensory fibers ipsilateral to the lesion. Ipsilaterally, but not contralaterally, the uptake and release of D-aspartate were depressed. Degeneration after partial cordotomy in combination with dorsal rhizotomy was assumed to be the sum of that produced by each lesion separately. This combined lesion depressed D-aspartate uptake ipsilaterally and depressed D-aspartate release on both sides of the cervical enlargement. None of the lesions altered the uptake and the evoked release of [3H]GABA. These findings support the hypothesis that the synaptic endings of one or more neural tracts descending from the brain to the spinal cord mediate the uptake and release of D-aspartate and, therefore, may use L-glutamate or L-aspartate as a transmitter.     

Distribution and origin of bombesin,substance P and somatostatin in cat spinal cord

Bombesin (BN), substance P-(SP) and somatostatin (SRIF) were measured in individual laminae of the cervical, thoracic and lumbar (L) spinal cord of control cats, and in the L6 segment of cats receiving a spinal hemisection (L2) or deafferentation via dorsal rhizotomy at L6, 7, S1. The interlaminar distribution of BN, SP, and SRIF was remarkably similar. Highest concentrations were found in the superficial dorsal horn, and progressively less was found proceeding ventrally. Some intersegmental variations in peptide concentration within a single lamina were found. Dorsal rhizotomy caused a significant decline in BN, SP and SRIF in lamina I-III, therefore all three peptides appear to be contained in dorsal root ganglion cells. Evidence is presented for the existence of ascending BN and SP projections originating in lamina I-III and VII, for a descending SRIF pathway terminating in lamina VIII, and for an ascending BN path in lamina VIII. Dorsal root afferents to lamina VIII influence levels of BN, SP and SRIF.     

Decreased Release of D-Aspartate in the Guinea Pig Spinal Cord After Lesions of the Red Nucleus

This study attempts to determine if fibers that project from the guinea pig red nucleus to the spinal cord use L-glutamate and/or L-aspartate as transmitters. Unilateral injections of kainic acid were placed stereotaxically in the red nucleus to destroy the cells of origin of the rubrospinal tract. Six days after the injection, Nissl-stained sections through the lesion site showed that the majority of neurons in the red nucleus ipsilateral to the kainic acid injection were destroyed. In addition, the lesioned area included parts of the surrounding midbrain reticular formation. Silver-impregnated, transverse sections of the cervical spinal cord revealed the presence of degenerating fibers contralaterally in laminae IV-VII of the gray matter. Ipsilaterally, very sparse degeneration was evident in laminae VII and VIII of the gray matter. Two to six days after surgery, the electrically evoked, Ca2(+)-dependent release of both D-[3H]aspartate, a marker for glutamatergic/aspartatergic neurons, and gamma-amino[14C]-butyric acid ([14C]GABA) was measured in dissected quadrants of the spinal cervical enlargement. Lesions centered on the red nucleus depressed the release of D-[3H]aspartate by 25-45% in dorsal and ventral quadrants of the cervical enlargement contralaterally. The release of [14C]GABA was depressed by 27% in contralateral ventral quadrants. To assess the contribution of rubro- versus reticulospinal fibers to the deficits in amino acid release, unilateral injections of kainic acid were placed stereotaxically in the midbrain reticular formation lateral to the red nucleus. Nissl-stained sections through the midbrain revealed the presence of extensive neuronal loss in the midbrain and rostral pontine reticular formation, whereas neurons in the red nucleus remained undamaged. In the spinal cord, degenerating axons were present ipsilaterally in laminae VII and VIII of the gray matter. Some fiber degeneration was also evident contralaterally in laminae V and VI of the gray matter. This lesion did not affect the release of either D-[3H]aspartate or [14C]GABA in the spinal cord. The substantial decrements in D-[3H]aspartate release following red nucleus lesions suggests that the synaptic endings of rubrospinal fibers mediate the release of D-[3H]aspartate in the spinal cord. Therefore, these fibers may be glutamatergic and/or aspartatergic. Because other evidence suggests that rubrospinal neurons are probably not GABAergic, the depression of [14C]GABA release probably reflects changes in the activity of spinal interneurons following the loss of rubrospinal input.     

Avian pancreatic polypeptide (APP) immunoreactive neurons in the spinal cord and spinal trigeminal nucleus

Using indirect immunofluorescence technique, avian pancreatic polypeptide (APP) immunoreactive cell bodies and fibres have been observed in the superficial laminae of the dorsal horn of the spinal cord and of the spinal trigeminal nucleus. Fibres were also seen in the ventral horns, in low numbers at the cervical and thoracic levels and in high numbers at the lower lumbar and upper sacral levels. Neither total cord transection, nor dorsal rhizotomy, nor capsaicin treatment seemed to affect the APP systems described above. The present findings suggest that an APP-like peptide may be involved in processing of sensory information at the level of the first relay station.     

Ultrastructural characteristics of dendrites of spinal neurons

A qualitative and quantitative analysis of outgrowths from the surface membrane of dendrites in regions of the gray matter of the spinal cord where different long descending tracts terminate was undertaken by electron microscopy. These outgrowths of dendrites were found to be most simply organized in the medial part of the ventral horn, where synaptic endings of fibers of the medial lemniscus are concentrated. In the dorsal regions of gray matter, where fibers of the lateral systems terminate, the outgrowths have a more complex structural organization, expressed as changes in size and shape and also as the appearance of elements of a weakly developed spinous apparatus in the matrix. These outgrowths were described as spine-like. A statistically significant difference was found between the size of dendritic outgrowths in the lateral part of Rexed's lamina V and the lateral part of laminae VI–VII, and also the medial part of lamina VIII.A. A. Bogomolets Institute of Physiology, Academy of Sciences of the Ukrainian SSR, Kiev. Translated from Neirofiziologiya, Vol. 10, No. 3, pp. 271–277, May–June, 1978.     

Experimental morphological study of primary afferent terminals at the base of the dorsal horn of the cat spinal cord

The distribution and ultrastructure of primary afferent terminals in the gray matter of the cervical and lumbar regions of the cat spinal cord were studied by the experimental degeneration method of Fink and Heimer. Most preterminals of primary afferents were shown to be concentrated in the region of the intermediate nucleus of Cajal (central part of Rexed's laminae VI–VII), in the substantial gelatinosa (laminae II–III), and in the nucleus proprius of the dorsal horn (central and medial parts of lamina IV). Fewer are found in the region of the motor nuclei. The number of degenerating axon terminals in the lateral parts of laminae IV and V differed: 31.5 and 0.4% respectively of all axon terminals. Many terminals of primary afferents in lamina IV contribute to the formation of glomerular structures in which they exist as terminals of S-type forming axo-axonal connections with other terminals. These results are in agreement with electrophysiological data to show that interneurons in different parts of the base of the dorsal horn differ significantly in the relative numbers of synaptic inputs formed by peripheral afferents and descending systems.A. A. Bogomolets Institute of Physiology, Academy of Sciences of the Ukrainian SSR, Kiev. Translated from Neirofiziologiya, Vol. 5, No. 4, pp. 406–414, July–August, 1973.     

Distribution of some enzymes associated with the metabolism of glutamate, aspartate, gamma-aminobutyrate and glutamine in cat spinal cord

The activities of glutamine synthetase, glutaminase, glutamate decarboxylase, GABA aminotransferase, glutamate dehydrogenase, and aspartate aminotransferase were measured in four areas of the cat spinal cord and in dorsal and ventral roots. Five of the six enzymes showed identical distribution patterns; i.e. the activities in the dorsal and ventral gray matter were equal and those of dorsal and ventral white matter were equal. No statistical differences in the mean enzyme activities in the dorsal and ventral roots were found. Glutamate decarboxylase was the only enzyme which had a different pattern. The enzyme activity in dorsal gray was twice that of ventral gray; the same pattern as the GABA concentration in both these areas. The glutamine synthetase activities in the cord areas and roots correlated with the glutamine distribution reported earlier. Thus, the distribution of glutamine (not a transmitter) and GABA (questionable transmitter) in gray matter are dictated by their synthesizing enzymes, whereas the distribution of glutamate and aspartate (likely transmitter suspects) cannot be explained on the basis of enzyme activities. Therefore, the enzyme activities may be related to the amino acid levels primarily in metabolic compartments, whereas the excess of certain amino acids in specific areas of the cord and roots may be related to functional compartments accumulated for use in synaptic transmission.     

Endomorphin-2 is an endogenous opioid in primary sensory afferent fibers

Evidence is presented that the recently discovered endogenous mu-selective agonist, endomorphin-2, is localized in primary sensory afferents. Endomorphin-2-like immunoreactivity was found to be colocalized in a subset of substance P- and mu opiate receptor-containing fibers in the superficial laminae of the spinal cord and spinal trigeminal nucleus. Disruption of primary sensory afferents by mechanical (deafferentation by dorsal rhizotomy) or chemical (exposure to the primary afferent neurotoxin, capsaicin) methods virtually abolished endomorphin-2-like immunoreactivity in the dorsal horn. These results indicate that endomorphin-2 is present in primary afferent fibers where it can serve as the endogenous ligand for pre- and postsynaptic mu receptors and as a major modulator of pain perception.     

D-Aspartate Uptake and Release in the Guinea Pig Spinal Cord After Partial Ablation of the Cerebral Cortex

This study attempts to determine if L-glutamate and L-aspartate may be transmitters of the guinea pig corticospinal tract. Unilateral ablations were made of the frontal and parietal neocortex which destroyed most of the motor and somatosensory areas in the right cerebral hemisphere. In lesioned animals, transverse sections of the cervical enlargement of the spinal cord (segments C6--T1) were stained to reveal degenerating fibers. Degeneration of axons first appeared 4 days after surgery, reached a maximum on the seventh day, and began to wane by the ninth day. The most prominent loss of axons appeared deep in the dorsal funiculus and in laminae IV-IX of the gray matter contralateral to the cortical lesion. Ipsilaterally, there was very sparse degeneration of fibers in the dorsal and ventral funiculi and in the spinal gray matter. The uptake and release of D-[3H]aspartate, a putative nonmetabolizable marker for L-glutamate and L-aspartate, were measured in dissected quadrants of the cervical enlargement taken from intact and lesioned animals. The uptake and the electrically evoked, Ca2+-dependent release of D-[3H]aspartate were depressed by 29-35% at 4 and 7 days after surgery, but only in tissue that was contralateral to the cortical ablation. The lesion had no effect on the uptake and release of exogenous gamma-[14C]aminobutyric acid, which were measured as indices of the postlesion integrity of neurons in the spinal gray matter. These findings suggest that the synaptic endings of corticospinal fibers probably mediate the uptake and release of D-[3H]aspartate and, therefore, may use L-glutamate and/or L-aspartate as a transmitter.     

Hrp-labeled sources of descending propriospinal pathways in cats

Unilateral injections of horseradish peroxidase into the cat spinal cord at different segmental levels revealed a laminar distribution of spinal interneurons that are sources of ipsilateral and contralateral propriospinal pathways of different lengths. The majority of the long pathways connecting cervical and lumbar segments are formed by neurons located in the central quadrants (laminae VII and VIII) bilaterally; a few such neurons also are present in the marginal layer and in lateral zones at the base of the dorsal horn (ipsilaterally). The zones containing numerous propriospinal neurons forming short (extending over a few segments) connections were more extensive. In the lumbar portion neurons which were sources of short uncrossed pathways tended to be concentrated in the lateral areas of the base of the dorsal horn, intermediate zone, and ventral horn, whereas sources of crossed pathways were concentrated in the ventromedial zones of gray matter. In the cervical portion "short" propriospinal neurons forming both ipsilateral and contralateral projections were concentrated in the lateral zones of gray matter. Neurons of the marginal layer and substantia gelatinosa and neurons of intermediolateral sympathetic nuclei also were sources of descending propriospinal pathways. Some propriospinal axons were intermediate in length. The distribution of neurons with axons of this kind largely coincided with the distribution of neurons that were sources of long propriospinal pathways. The connection between the spatial distribution of different groups of propriospinal neurons and the organization of the synaptic inputs into them, and also correlation between the morphological and functional characteristics of these neurons are discussed.     

Decreased Uptake and Release of D-Aspartate in the Guinea Pig Spinal Cord After Dorsal Root Section

This study attempts to determine if L-glutamate and/or L-aspartate may be transmitters of dorsal sensory neurons. The uptake and the electrically evoked release of D-[3H]aspartate, a putative marker for L-glutamate and L-aspartate, were measured in the cervical enlargement (segments C4-T1) of the guinea pig spinal cord before and after cutting dorsal roots C5-T1 on the right side. The uptake and the release of gamma-aminobutyric acid (GABA) also were measured as indices of the integrity of GABAergic neurons in the spinal cord. The cervical enlargement was excised and divided into left and right halves, then into dorsal and ventral quadrants. Quadrants from unlesioned animals took up D-aspartate and GABA, achieving concentrations in the tissues which were 14-25 times that in the medium. Subsequently, electrical stimulation evoked a Ca2+-dependent release of D-aspartate and of GABA. The uptake and release of D-aspartate and GABA were similar in tissues taken from intact and sham-operated animals. However, dorsal rhizotomy, without damage to dorsal radicular or spinal blood vessels, depressed the uptake (by 22-29%) and the release (by 50%) of D-aspartate only in quadrants ipsilateral to the lesion. The uptake and the release of GABA were unchanged. In transverse sections of the cervical enlargement, stained to reveal degenerating fibers, by far the heaviest loss of axons occurred in the cuneate fasciculus and in the gray matter ipsilateral to the cut dorsal roots. These findings suggest that the synaptic endings of dorsal sensory neurons probably mediate the uptake and the release of D-aspartate and, therefore, may use L-glutamate or L-aspartate as a transmitter. When spinal blood vessels were damaged during dorsal rhizotomy, the deficits in D-aspartate uptake and release were larger than those in the absence of vascular damage and were accompanied by deficits in GABA uptake and release. These findings imply that vascular damage results in the loss of intraspinal neurons, some of which probably mediate the uptake and release of D-aspartate and, therefore, may use L-glutamate and/or L-aspartate as a transmitter.     

Lamina VII and VIII neurons of the S2 segment bilaterally projecting to the C6 segment of the spinal cord in the cat

Intracellular and extracellular recordings of antidromic action potentials were applied to invetigate neurons of the S2 segment projecting to the C6 segment of the cat spinal cord. The cell bodies were located in laminae VII and VIII of the gray matter while axons ascended in lateral funiculi. Thirty-two out of the total 45 neurons were found to project to the C6 segment bilaterally, seven ipsilaterally and six contralaterally. The axonal conduction velocities were in the 42–96 m/s range and in some neurons were significantly lower in distal parts of axons, supposing that some neurons may give off collateral branches to various segments of the spinal cord. It is discussed if the investigated neurons form a part of the propriospinal system or if their cervical projections are only collaterals of long tracts ascending to supraspinal levels. The organisation of the presented connections between spinal enlargements indicates their contribution in complex mechanisms of co-ordination of movements of the limbs.     

Transient Spinal Cord Ischemia in Rat: The Time Course of Spinal FOS Protein Expression and the Effect of Intraischemic Hypothermia (27°C)

1. In the present study, we characterize the time course of spinal FOS protein expression after transient noninjurious (6-min) or injurious (12-min) spinal ischemia induced by inflation of a balloon catheter placed into the descending thoracic aorta. In addition, this work examined the effects of spinal hypothermia on FOS expression induced either by ischemia or by potassium-evoked depolarization (intrathecal KCl).2. Short-lasting (6-min) spinal ischemia evoked a transient FOS protein expression. The peak expression was seen 2 hr after reperfusion in all laminar levels in lumbosacral segments. At 4 hr of reperfusion, more selective FOS expression in spinal interneurons localized in the central part of laminae V–VII was seen. At 24 hr no significant increase in FOS protein was detected.3. After 12 min of ischemia and 2 hr of reflow, nonspecific FOS expression was seen in both white and gray matter, predominantly in nonneuronal elements. Intrathecal KCl-induced FOS expression in spinal neurons in the dorsal horn and in the intermediate zone. Spinal hypothermia (27°C) significantly suppressed FOS expression after 6 or 12 min of ischemia but not after KCl-evoked depolarization.4. Data from the present study show that an injurious (but not noninjurious) interval of spinal ischemia evokes spinal FOS protein expression in glial cells 2 hr after reflow. The lack of neuronal FOS expression corresponds with extensive neuronal degeneration seen in this region 24 hr after reflow. Noninjurious (6-min) ischemia induced a transient, but typically neuronal FOS expression. The significant blocking effect of hypothermia (27°C) on the FOS induction after ischemia but not after potassium-evoked depolarization also suggests that simple neuronal depolarization is a key trigger in FOS induction.     

Myocardial ischemia induces the release of substance P from cardiac afferent neurons in rat thoracic spinal cord

Antibody-coated microprobes were inserted into the thoracic (T3-4) spinal cord in urethane-anesthetized Sprague-Dawley rats to detect the differences in the release of immunoreactive substance P-like (irSP) substances in response to differential activation of cardiac nociceptive sensory neurons (CNAN). CNAN were stimulated either by intrapericardial infusion of an inflammatory ischemic exudate solution (IES) containing algogenic substances (i.e., 10 mM each of adenosine, bradykinin, prostaglandin E2, and 5-hydroxytryptamine), or by transient occlusion of the left anterior descending coronary artery (CoAO). There was widespread basal release of irSP from the thoracic spinal cord. Stimulation of the CNAN by IES did not alter the pattern of release of irSP. Conversely, CoAO augmented the release of irSP from T3-4 spinal segments from laminae I-VII. This CoAO-induced irSP release was eliminated after thoracic dorsal rhizotomy. These results indicate that heterogeneous activation of cardiac afferents, as with focal coronary artery occlusion, represents an optimum input for activation of the cardiac neuronal hierarchy and for the resultant perception of angina. Excessive stimulation of cardiac nociceptive afferent neurons elicited during regional coronary artery occlusion involves the release of SP in the thoracic spinal cord and suggests that local spinal cord release of SP may be involved in the neural signaling of angina.     

The vestibular nuclei in the domestic hen (Gallus domesticus). VII.--Afferents from the spinal cord

Lesions of different parts of the spinal cord at different levels in the hen have been made and the resulting degeneration in the vestibular complex has been studied in silver impregnated sections. Spinovestibular fibres originate from cervical as well as lumbosacral levels of the cord and run in the dorsal part of the lateral funiculus. The spinovestibular fibres from all levels of the spinal cord terminate ipsilaterally in the nucleus Deiters ventralis, the nucleus Deiters dorsalis, the medial nucleus and rostrally in the descending nucleus. The spinovestibular fibres terminating in the above nuclei are few in number while spinovestibular fibres terminating bilaterally in the caudal part of the descending nucleus are much more abundant. In a few cases HRP injections in the vestibular complex resulted in labelled cells in upper cervical segments of the spinal cord localized in lamina VII. The findings are discussed in the light of data concerning the spinovestibular pathway in mammals.     

GLUTAMATE AND RELATED AMINO ACIDS IN CAT SPINAL ROOTS, DORSAL ROOT GANGLIA AND PERIPHERAL NERVES

Of the free amino acids found in extracts of cat spinal roots, dorsal root ganglia and peripheral nerves, only glutamate was present in disproportionately high concentrations in those parts of the dorsal roots between ganglia and spinal cord. This distribution suggests that the high dorsal root levels of glutamate may result from synthesis in dorsal root ganglia and subsequent transport towards the spinal cord. Four excitant amino acids were detected in the extracts: aspartate, cysteate, cysteine sulphinate and glutamate. The unique regional distribution of glutamate is consistent with the proposed role of this amino acid as an excitatory transmitter at the terminals of primary afferent fibres.     

Ascending and descending projections of the lateral vestibular nucleus in the rat

The tracer neurobiotin was injected into the lateral vestibular nucleus in rat and the efferent fiber connections of the nucleus were studied. The labeled fibers reached the diencephalon rostrally and the sacral segments of the spinal cord caudally. In the diencephalon, the ventral posteromedial and the gustatory nuclei received the most numerous labeled fibers. In the mesencephalon, the inferior colliculus, the interstitial nucleus of Cajal, the nucleus of Darkschewitch, the periaqueductal gray matter and the red nucleus received large numbers of labeled fibers. In the rhombencephalon, commissural and internuclear connections originated from the lateral vestibular nucleus to all other vestibular nuclei. The medioventral (motor) part of the reticular formation was richly supplied, whereas fewer fibers were seen in the lateral (vegetative) part. In the spinal cord, the descending fibers were densely packed in the anterior funiculus and in the ventral part of the lateral funiculus. Collaterals invaded the entire gray matter from lamina IX up to lamina III; the fibers and terminals were most numerous in laminae VII and VIII. Collateral projections were rich in the cervical and lumbosacral segments, whereas they were relatively poor in the thoracic segments of the spinal cord. It was concluded that the fiber projection in the rostral direction was primarily aimed at sensory-motor centers; in the rhombencephalon and spinal cord, fibers projected onto structures subserving various motor functions.     

Interferon-γ receptors are expressed at synapses in the rat superficial dorsal horn and lateral spinal nucleus

Summary Interferon-γ can facilitate the spinal nociceptive flexor reflex and elicit neuropathic pain-related behavior in rats and mice. Immunoreactivity for the interferon-γ receptor (IFN-γR) occurs in the superficial layers of the dorsal horn and the lateral spinal nucleus in the rat and mouse spinal cord, as well as in subsets of neurons in the dorsal root ganglia. The aim of the present study was to examine the cellular localization and origin of the IFN-γR in the spinal cord. As viewed by confocal microscopy, the immunopositivity for the IFN-γR was co-localized with that of the presynaptic marker synaptophysin and with neuronal nitric oxide synthase in the lateral spinal nucleus, whereas only a minor overlap with these molecules was observed in laminae I and II of the dorsal horn. There was no co-localization of the IFN-γR with markers for astrocytes and microglial cells. Ultrastructurally, the IFN-γR was found predominantly in axon terminals in the lateral spinal nucleus but also at postsynaptic sites in dendrites in laminae I and II. The IFN-γR expressed in neurons in dorsal root ganglia was transported in axons both centrally and peripherally. Hemisection of the spinal cord caused no reduction in immunolabelling of the IFN-γR in the dorsal horn or the lateral spinal nucleus. Since rhizotomy does not effect the immunolabelling in the lateral spinal nucleus, our observation indicates that the presynaptic receptors in this nucleus are derived from intrinsic neurons. The localization of the IFN-γR in the spinal cord differed from that of the AMPA glutamate receptor subunits 2 and 3 and the substance P receptor (NK1). Our results, showing localization of IFN-γR to pre- and postsynaptic sites in the dorsal horn and lateral spinal nucleus indicate that IFN-γ can modulate nociception at the spinal cord level.     

大鼠脑内代谢型谷氨酸受体5亚型(mGluR5)定位分布的免疫细胞化学研究

本研究用免疫细胞化学技术观察了大鼠脑内参与兴奋性突触传递的代谢型谷氨酸受体5亚型(mGluR5)的精确定位分布.mGluR5阳性浓染的神经元胞体和纤维密集地分布于大脑皮质浅层、嗅球、伏核、尾壳核、前脑基底部、隔区、苍白球、腹侧苍白球、海马CA1和CA2区、下丘中央核、被盖背侧核和三叉神经脊束核尾侧亚核浅层;淡染而稀疏的mGluR5阳性神经元胞体和纤维见于屏状核、终纹床核、杏仁中央核、丘脑部分核团、上丘浅灰质层、外侧丘系背侧核和延髓中央灰质.     

High-frequency, but not low-frequency, transcutaneous electrical nerve stimulation reduces aspartate and glutamate release in the spinal cord dorsal horn

Transcutaneous electrical nerve stimulation (TENS) is a commonly utilized non-pharmacological treatment for pain. Studies show that low- and high-frequency TENS utilize opioid, serotonin and/or muscarinic receptors in the spinal cord to reduce hyperalgesia induced by joint inflammation in rats. As there is an increase in glutamate and aspartate levels in the spinal cord after joint inflammation, and opioids reduce glutamate and aspartate release, we hypothesized that TENS reduces release of glutamate and aspartate in animals with joint inflammation by activation of opioid receptors. Using microdialysis and HPLC with fluorescence detection, we examined the release pattern of glutamate and aspartate in the dorsal horn in response to either low-frequency (4 Hz) or high-frequency (100 Hz) TENS. We examined the effects of TENS on glutamate and aspartate release in animals with and without joint inflammation. High-frequency, but not low-frequency, TENS significantly reduced spinal glutamate and aspartate in animals with joint inflammation compared with levels in those without joint inflammation. The reduced release of glutamate and aspartate by high-frequency TENS was prevented by spinal blockade of delta-opioid receptors with naltrindole. Thus, we conclude that high-frequency TENS activates delta-opioid receptors consequently reducing the increased release of glutamate and aspartate in the spinal cord.