Oligodendrocyte ablation triggers central pain independently of innate or adaptive immune responses in mice.

Mechanisms underlying central neuropathic pain are poorly understood. Although glial dysfunction has been functionally linked with neuropathic pain, very little is known about modulation of pain by oligodendrocytes. Here we report that genetic ablation of oligodendrocytes rapidly triggers a pattern of sensory changes that closely resemble central neuropathic pain, which are manifest before overt demyelination. Primary oligodendrocyte loss is not associated with autoreactive T- and B-cell infiltration in the spinal cord and neither activation of microglia nor reactive astrogliosis contribute functionally to central pain evoked by ablation of oligodendrocytes. Instead, light and electron microscopic analyses reveal axonal pathology in the spinal dorsal horn and spinothalamic tract concurrent with the induction and maintenance of nociceptive hypersensitivity. These data reveal a role for oligodendrocytes in modulating pain and suggest that perturbation of oligodendrocyte functions that maintain axonal integrity can lead to central neuropathic pain independent of immune contributions.

Localization of soluble guanylyl cyclase in the superficial dorsal horn.

Nitric oxide (NO) has been implicated in pain processing at the spinal level, but the mechanisms mediating its effects remain unclear. In the present work, we studied the organization of the major downstream effector of NO, soluble guanylyl cyclase (sGC), in the superficial dorsal horn of rat. Almost all neurokinin 1 (NK1) receptor-positive neurons in lamina I (a major source of ascending projections) were strongly immunopositive for sGC. Many local circuit neurons in laminae I-II also stained for sGC, but less intensely. Numerous fibers, presumably of unmyelinated primary afferent (C fiber) origin, stained for calcitonin gene-related peptide or isolectin B4, but none of these was immunopositive for sGC. These data, along with immunoelectron microscopy results, imply that unmyelinated primary afferent fibers terminating in the superficial dorsal horn lack sGC. Double labeling showed that neuronal nitric oxide synthase (nNOS) seldom colocalized with sGC, but nNOS-positive structures were frequently closely apposed to sGC-positive structures, suggesting that in the superficial dorsal horn NO acts mainly in a paracrine manner. Our data suggest that the NK1 receptor-positive projection neurons in lamina I are a major target of NO released in superficial dorsal horn. NO may also influence local circuit neurons, but it does not act on unmyelinated primary afferent terminals via sGC.

Effect of protein kinase C blockade on phosphorylation of NR1 in dorsal horn and spinothalamic tract cells caused by intradermal capsaicin injection in rats.

We have previously reported that protein kinase A (PKA) is involved in the phosphorylation of NR1 subunits of N-methyl-d-aspartate (NMDA) receptors in dorsal horn neurons after intradermal injection of capsaicin (CAP). To see if protein kinase C (PKC) also participates in the phosphorylation of NR1, we used electron microscopic techniques to determine further where the phosphorylated NR1 subunits (pNR1) are expressed in the spinothalamic tract (STT) cells and immunohistochemistry to examine whether a PKC inhibitor, chelerythrine chloride, blocks the enhanced phosphorylation of NR1 on serine 896. The pNR1 subunits were in the soma and dendrites of STT cells and in presynaptic endings. Western blots showed that pretreatment with the PKC inhibitor caused a decrease in CAP-induced phosphorylation of NR1 protein. In immunofluorescence staining, the number of pNR1-like immunoreactive neurons was significantly decreased on the side ipsilateral to the injection when chelerythrine chloride was administered intrathecally before CAP injection. In addition, when STT cells were labeled by microinjection of the retrograde tracer, fluorogold (FG), into the thalamus, we found that the proportion of p-NR1-LI STT cells was markedly reduced after PKC inhibition. Combined with our previous findings, these results strongly suggest that NR1 subunits in spinal dorsal horn neurons are phosphorylated following CAP injection, and this phosphorylation is catalyzed by PKC, as well as by PKA.

Synaptology of trigemino- and spinothalamic lamina I terminations in the posterior ventral medial nucleus of the macaque.

We used the electron microscope to examine lamina I trigemino- and spinothalamic (TSTT) terminations in the posterior part of the ventral medial nucleus (VMpo) of the macaque thalamus. Lamina I terminations were identified by anterograde labeling with biotinylated dextran, and 109 boutons on 38 terminal fibers were closely studied in series of ultrathin sections. Five unlabeled terminal boutons of similar appearance were also examined in detail. Three-dimensional, volume-rendered computer models were reconstructed from complete series of serial sections for 29 boutons on 10 labeled terminal fibers and one unlabeled terminal fiber. In addition, postembedding immunogold staining for GABA was obtained in alternate sections through 23 boutons. Lamina I TSTT terminations in VMpo generally have several large boutons (mean length = 2.16 microm, mean width = 1.29 microm) that are densely packed with vesicles and make asymmetric synaptic contacts on low-order dendrites of VMpo neurons (mean diameter 1.45 microm). They are closely associated with GABAergic presynaptic dendrites (PSDs), and nearly all form classic triadic arrangements (28 of 29 reconstructed boutons). Consecutive boutons on individual terminal fibers make multiple contacts with a single postsynaptic dendrite and can show evidence of progressive complexity. Dendritic appendages that enwrap and invaginate the terminal bouton constitute additional anatomic evidence for secure, high-fidelity synaptic transfer. These observations provide direct ultrastructural evidence supporting the hypothesis that VMpo is a lamina I TSTT thalamocortical relay nucleus in primates that subserves pain, temperature, itch, and other sensations related to the physiological condition of the body.

Ultrastructural localization of glutamate receptor subunits (NMDAR1, AMPA GluR1 and GluR2/3) and spinothalamic tract cells.

The associations of glutamate receptor subunits (NMDAR1, AMPA GluR1 and GluR2/3) and spinothalamic tract neurons in the rat lumbar spinal cord dorsal horn were investigated. Staining for NMDAR1 and AMPA GluR1 and GluR2/3 receptor subunits was observed throughout the spinothalamic tract soma and dendrites, particularly in association with the rough endoplasmic reticulum and some postsynaptic membrane sites. Immunostaining for NMDAR1 and AMPA GluR2/3 was also noted in presynaptic membrane sites. Localization of both NMDA and AMPA glutamate receptor subunits in association with spinothalamic tract neurons provides anatomical evidence in support of the various interactions reported for glutamate receptors in nociception. Presynaptic localization of the AMPA GluR2/3 receptor subunit suggests that spinothalamic tract cells may also be affected presynaptically by AMPA glutamate receptor interactions.

Fine structural organization of spinothalamic and trigeminothalamic lamina I terminations in the nucleus submedius of the cat.

We examined lamina I trigemino- and spinothalamic tract (TSTT) terminals labeled with Phaseolus vulgaris leucoagglutinin in the nucleus submedius (Sm), a nociceptive relay in the cat's thalamus. Volume-rendered (three-dimensional) reconstructions of ten lamina I TSTT terminals identified with light and electron microscopy were built from serial ultrathin sections by computer, which enabled the overall structures of the terminal complexes to be characterized in detail. Two fundamentally different terminations were observed: compact clusters of numerous boutons, which predominate in the dense focus of a lamina I terminal field in the Sm, and boutons-of-passage, which are present throughout the terminal field and predominate in its periphery. Reconstructions of cluster terminations reveal that all boutons of each cluster make synaptic contact with protrusions and branch points on a single dendrite and involve presynaptic dendrites (PSDs) in triadic arrangements, providing a basis for the secure relay of sensory information. In contrast, reconstructions show that boutons-of-passage are generally characterized by simple contacts with PSDs, indicating an ascending inhibitory lamina I influence. These different synaptic arrangements are consistent with physiological evidence indicating that the morphologically distinct nociceptive-specific and thermoreceptive-(cold)-specific lamina I TSTT neurons terminate differently within the Sm. Thus, a suitable structural substrate exists in the cat's Sm for the inhibitory effect of cold on nociception, a behavioral and physiological phenomenon of fundamental significance. We conclude that the Sm is more than a simple relay for nociception, and that it may be an integrative comparator of ascending modality-selective information that arrives from neurons in lamina I.

Morphology of thalamocortical neurons projecting to the primary somatosensory cortex and their relationship to spinothalamic terminals in the squirrel monkey.

This study examined the morphology of thalamocortical neurons projecting to the primary somatosensory cortex (SI; hand region of areas 3a, 3b, 1, and 2) and their relationship to the spinothalamic (STT) terminals in the squirrel monkey. Retrogradely labeled thalamocortical neurons were intracellularly filled with Lucifer yellow (LY), and the STT terminals were anterogradely labeled with biotinylated dextran. Both filled neurons and labeled terminals were differentially visualized in the same field by a dual immunocytochemical staining method. SI-projecting neurons appeared at the light level to be in contact with STT terminal boutons in the ventroposterior lateral (VPL), ventroposterior inferior (VPI), and centrolateral (CL) nuclei and the posterior complex (PO). The analyses of the neuronal morphology revealed that somatic and dendritic morphologies of SI-projecting neurons in these thalamic nuclei, as well as in the anterior pulvinlar (Pulo), centromedial (CM), and ventrolateral (VL) nuclei, were generally comparable with some exceptions: VL neurons had the largest soma sizes, the most primary dendrites, and the longest total dendritic length among all neurons studied; VPI neurons had the smallest soma sizes; VPL SI-projecting neurons were different from those in VPI in their soma sizes, shape factors, and orientations; in VPL the cells projecting to the superficial layers of SI were smaller than those projecting to the deeper layers, but in VPI the two groups of neurons were similar in soma sizes. In general, the SI-projecting neurons in VPL, VPI, and CL were similar in their dendritic morphologies and branching patterns, and varied from those in Pulo, PO, CM, and VL.

Direct catecholaminergic innervation of primate spinothalamic tract neurons.

Catecholaminergic axonal varicosities identified by immunocytochemical staining for dopamine-beta-hydroxylase were observed at the light microscopic level apposing the somata of retrogradely labeled spinothalamic tract neurons in the monkey spinal cord. Three retrogradely labeled and two intracellularly labeled spinothalamic neurons were serially sectioned and examined at selected intervals at the electron microscopic level. Electron microscopic study revealed that axonal boutons directly contacted the somata and/or dendrites of lamina I, IV, and V spinothalamic tract neurons. All of the profiles apposing one of the retrogradely labeled lamina I spinothalamic tract neurons were categorized from eight planes of section spaced at 1-micron intervals. Of the 305 profiles counted that were adjacent to this soma, 17 (5.6%) stained positively for dopamine-beta-hydroxylase. Of these 17 appositions, three were followed in serial sections to confirm that they had synaptic thickenings and alignment of vesicles along the membrane contacting the spinothalamic tract soma. Catecholaminergic boutons were observed apposing the somata and dendrites of intracellularly filled STT cells characterized as high threshold and wide dynamic range neurons. These observations clearly indicate a direct innervation of spinothalamic tract neurons by catecholaminergic neurons, providing anatomical data to support previous physiological findings demonstrating that catecholamines modulate nociceptive transmission.

Serotonergic innervation of the lateral cervical nucleus: an immunohistochemical study in cats and monkeys (Aotus trivirgatus).

A serotonergic input to the lateral cervical nucleus of cats and monkeys (Aotus trivirgatus) was demonstrated with immunohistochemical methods. In both species, the lateral cervical nucleus was found to contain a network of serotonin-immunoreactive fibers. However, the density of labeled fibers was greater in the monkeys than in the cats. Most labeled fibers were thin and had irregularly spaced varicosities. Electron microscopic examination showed that labeled varicosities were in apposition with dendrites, neuronal somata and unlabeled terminals, but synapses were rare. The results demonstrate that the lateral cervical nucleus receives a serotonergic innervation, as is the case with other somatosensory relay structures such as the spinal dorsal horn and the dorsal column nuclei. The presence of a serotonergic innervation suggests that the transmission of somatosensory information through the lateral cervical nucleus is modulated by a descending pathway. However, its effect on the response properties of neurons in the lateral cervical nucleus is unknown.

Ultrastructural analysis of axosomatic contacts on functionally identified primate spinothalamic tract neurons.

The morphology and frequency of axosomatic contacts on three functionally identified primate spinothalamic tract (STT) cells were analyzed at the electron microscopic level. The STT cells analyzed were wide-dynamic-range neurons responsive to activation of low- and high-threshold cutaneous afferents innervating the foot. The somas were located in the lateral border of lamina V; the dendritic trees were oriented dorsally and were very extensive. Numerous spinelike appendages were observed emanating from two of the cell bodies. Terminal types contacting the cell bodies were categorized at several different layers through each neuron. Six morphologically different terminal types were established following analysis of serial sections. Profiles classified as round (R) terminals containing round clear vesicles and zero or one dense-core vesicle made up over 50% of the total population in contact with the STT somas. Profiles containing round clear vesicles and two to four small-diameter dense-core vesicles (D1 category) made up approximately 10% of the population in contact with each soma. Flat (F) terminals with oblong or flattened clear vesicles made up approximately 8% of the population. The remaining three categories (D2, L1, and L2) distinguished by the number and size of the dense-core vesicles made up a small percentage of the total population in contact with the cell bodies. The distribution of terminal types on the soma proper versus somatic spines was also determined for one cell. The proportions of the six terminal types contacting the soma of these cells were very similar, although the physiological characteristics of each cell were different. However, the relative proportions of terminal types on these three lamina V STT cell bodies were different from those previously reported contacting somata in lamina V, suggesting that there may be a unique innervation of STT cells that differentiates them from other cell types in lamina V.

The differential synaptic organization of the spinal and lemniscal projections to the ventrobasal complex of the rat thalamus. Evidence for convergence of the two systems upon single thalamic neurons.

The synaptic organization of terminals originating either from the spinal cord (spinothalamic) or from the dorsal column nuclei (lemniscal) was investigated in the ventrobasal complex of the rat thalamus. Wheatgerm agglutinin conjugated to horseradish peroxidase was used as an anterogradely transported axonal tracer, using benzidine dihydrochloride as a chromogen for the identification by electron microscopy of spinal and lemniscal projections to the ventrobasal thalamus. A double anterograde tract tracing strategy, based labeling by wheatgerm agglutinin conjugated to horseradish peroxidase of spinal terminals and simultaneous visualization of lemniscal terminals identified by Wallerian degeneration induced by lesion of the neurons of origin in the dorsal column nuclei, was used to compare the postsynaptic elements contacted by the two pathways and to look for a possible convergence of the two pathways onto single thalamic neurons. Spinal and lemniscal terminals are large (2-2.5 microns mean average diameter) terminals containing several mitochondria and numerous rounded vesicles. A quantitative analysis of the mean average diameters of the terminals revealed that one could not differentiate between synapses formed by the two pathways on a morphological basis. Terminals of the two pathways make asymmetrical contacts (Gray type I) with dendrites of varying diameter, dendritic protrusions, and cell somata. A quantitative analysis of the least diameter of the postsynaptic elements demonstrates projections of the two systems to different, partially overlapping regions of thalamic neurons. Lemniscal terminals originating from the dorsal column nuclei frequently contact cell somata; axosomatic spinothalamic contacts are uncommon. In addition, lemniscal projections tend to contact more proximal dendrites than do spinal projections, and this differential synaptic organization is statistically significant. From a functional point of view, this differential synaptic organization might indicate that lemniscal inputs have greater influence than spinal inputs in affecting the activity of thalamic neurons. Labeled spinothalamic terminals contact the same dendritic profile as do degenerating lemniscal terminals in about 10% of single sections. Because the present study did not include a complete reconstruction of ventrobasal complex neurons of the thalamus or even regions of dendritic arbors, the degree of convergence is likely to be significantly underestimated. These findings indicate that the anatomical basis exists for an interaction between nociceptive and non-nociceptive somesthetic systems at the level of single ventrobasal neurons of the thalamus of the rat.

Fine structure of the spinothalamic projections to the central lateral nucleus of the rat thalamus.

The fine structure of labelled spinothalamic terminals in the central lateral nucleus has been studied in the rat following injection of wheat germ agglutinin-horseradish peroxidase into the spinal cord. Myelinated axons gave rise to the labelled terminals, which were large profiles which contained round vesicles, numerous mitochondria, and formed asymmetrical contacts with large dendrites or dendritic protrusions. These profiles are similar to those described in other somatosensory thalamic nuclei, and in many other nuclei of the thalamus.

Immunocytochemistry of enkephalin and serotonin distribution in restricted zones of the rostral trigeminal spinal subnuclei: comparisons with subnucleus caudalis.

The spinal trigeminal subnucleus caudalis processes nociceptive input from the head. However, physiological and behavioral studies in monkeys and humans indicate that painful stimuli from the central face and oral cavity also project through trigeminal nuclei rostral to the spinal subnucleus caudalis. Both enkephalin (ENK) and serotonin (5-HT) are present in rostral trigeminal nuclei and these regions receive inputs from the raphe complex. Thus, it appears that elements of pain-modulating circuitry proposed by Basbaum and Fields (Annu. Rev. Neurosci., 7:309-338, 1984) for the spinal and medullary dorsal horn may also exist in this region. In order to begin an exploration of this circuitry, the present study combines the techniques of retrograde transport of HRP from the ventral posteromedial thalamic nucleus (VPM) of the cat's thalamus to label trigeminothalamic relay cells. Secondarily, immunocytochemical techniques are employed to define the distribution patterns of ENK and 5-HT cells and terminals in relationship to both labeled and nonlabeled neurons in each of the subnuclei of the spinal trigeminal nucleus. Trigeminothalamic relay cells were observed in laminae I and II, the magnocellular region, and the interstitial nucleus (IN) of subnucleus caudalis (Vc). ENK was found in axodendritic and axosomatic terminals, together with a population of small fusiform neurons in all these same areas except the magnocellular region. ENK axosomatic contacts innervated approximately 30% of labeled relay cells, chiefly in lamina I and the IN, or small unlabeled neurons in the same area. Serotonin activity occurred principally in lamina I and the IN and was confined almost exclusively to axodendritic terminals. Examination of subnucleus interpolaris (Vi) revealed relay cells distributed throughout the length of the nucleus and increasing in numbers at rostral levels. A rostral extension of the IN was found just ventrolateral to the main body of Vi and contained numerous labeled cells. The distribution of ENK activity was restricted to the ventral part of Vi and the IN and occurred in axodendritic and axosomatic terminals. These latter elements innervated 30-40% of labeled relay cells in Vi, particularly those located in the IN. Cells containing ENK generally resembled the fusiform cells found in Vc and were distributed in ventral Vi and the IN. Some ENK cells were larger, displayed several dendrites, and occurred only in the ventral Vi. Serotonin within Vi and Vc was confined principally to axodendritic terminals.(ABSTRACT TRUNCATED AT 400 WORDS)

Ultrastructural analysis of the terminals of various somatosensory pathways in the ventrobasal complex of the rat thalamus: an electron-microscopic study using wheatgerm agglutinin conjugated to horseradish peroxidase as an axonal tracer.

We used wheatgerm agglutinin conjugated to horseradish peroxidase (WGA:HRP) as an anterograde tracer to label the terminals of the lemniscal, spinothalamic, and trigeminothalamic pathways in the ventrobasal complex of the rat thalamus (VB). The use of benzidine dihydrochloride (BDHC) as the chromogen allowed us to view the labeled profiles with the electron microscope and permitted us to compare the morphology of the terminals from the various pathways. We found that all the labeled somatosensory pathways terminate in the VB in the form of large terminals that contain round synaptic vesicles and make numerous asymmetrical synaptic contacts, usually with dendritic protrusions and proximal dendrites. The present results demonstrate that pathways conveying noxious and non-noxious somatosensory information terminate upon thalamic neurons with synaptic terminals having similar morphological features.

Volume composition of the lateral cervical nucleus in the cat. I. A stereological and electron microscopical study of normal and deafferentated animals.

The neuropil of the lateral cervical nucleus (LCN) has been studied by means of ultrastructural stereological methods. Estimates of the fractional volumes of axons, glial cells (three types), extracellular space, boutons, bouton mitochondria, dendrites, dendritic mitochondria, neuronal somata and blood vessels were calculated for four normal cats and eight cats which had been subjected to spinal deafferentation 2, 4, 9, and 14 days before perfusion. 14 days after operation the fractional volume of boutons and dendrites in the nucleus was considerably lower on the operated side, corresponding to a reduction of 67% and 64% respectively. The bouton reduction was higher than could have been expected from findings in earlier qualitative studies. The reduction in dendritic volume, which was already present after 4 days, had not been detected in earlier studies of the LCN. Dark dendrites were found both in normal animals and on the normal and operated sides in operated animals. The variation from animal to animal was great but the figures for the deafferented sides were significantly higher. The general value of the stereological technique in ultrastructural neuroanatomical research is discussed and it is concluded that quantitative methods are more sensive to small and gradual changes and should give a better estimation of transneuronal effects and of the amount of degeneration than purely qualitative methods.