Neurochemistry and neural circuitry in the dorsal horn.
Explore the source record for details and available documents.
Biomedical subjects
Publications and source records attributed to G J Bennett.
Explore the source record for details and available documents.
Spinomesencephalic tract neurons in the cat spinal cord were retrogradely labeled following injections of wheatgerm agglutinin conjugated to horseradish peroxidase into the region of the midbrain parabrachial area. Labeled cell bodies were concentrated in lamina I, bilaterally. A more scattered distribution was observed in lamina V and deeper laminae. Bilateral lesions of the dorsolateral funiculus (DLF) at thoracic levels eliminated labeling of lamina I neurons below the lesions, but had no effect on the labeling of the neurons in deeper laminae. Injections of colchicine into the spinal white matter caused the label to accumulate intra-axonally and revealed labeled axons bilaterally in the DLF and ipsilaterally in the ventrolateral and ventral funiculi.
We have observed a population of lamina I neurons in the cat that has projections to the parabrachial area (parabrachial and cuneiform nuclei). A subpopulation of these neurons also projects to the contralateral thalamus. The majority of projecting cells responded exclusively to noxious stimuli, a few wide-dynamic-range neurons were also observed. Conduction velocities for antidromic activation from the midbrain ranged from 1 to 18 m/s. We stained 14 cells intracellularly with horseradish peroxidase. These findings suggest that a major nociceptive projection pathway originating in lamina I and terminating in the parabrachial area exists in the cat.
Antidromic search stimuli were delivered to cervical (C2-C3) dorsal and dorsolateral funiculi that were dissected apart from one another and from the rest of the spinal cord. Fifty-six neurons were antidromically identified in the dorsal horn of the lumbosacral enlargement. Of these neurons, 23 were activated antidromically from both the dorsal columns and the ipsilateral dorsolateral funiculus. The neurons were found at depths corresponding to laminae III and IV. About half of the neurons responded only to innocuous, tactile stimuli whereas the other half responded to both innocuous and noxious stimuli. The existence of neurons with branched axons ascending both the dorsal and dorsolateral funiculi raises the possibility that the dorsal column postsynaptic and spinocervical tracts may not be completely independent projections.
Dorsal column postsynaptic (DCPS) neurons in the dorsal horn of the spinal cord have been identified by antidromic stimulation and intracellularly recorded in anesthetized cats. In about one-half of the cells, the antidromic stimulus evokes only an antidromic potential. This potential sometimes has an atypical shape due to penetration-induced depolarization. Atypical potentials are converted into typical antidromic spike potentials after intracellular injection of hyperpolarizing current. In the other one-half, the antidromic potentials are followed by postsynaptic activity. Intracellular analysis indicates that this postsynaptic activity is generated largely by activation of the intraspinal collaterals of A-beta primary afferents that are ascending the dorsal columns. Part of this postsynaptic activity has been produced monosynaptically. Polysnaptic responses are also evident; these are thought to be initiated by A-beta axons, desending axons from neurons in the dorsal column nuclei, or the local collaterals of the DCPS axons themselves. The results indicate that the DCPS system is nonlemniscal in nature and may be involved in those pain modulation systems that are activated by A-beta afferents.
Dorsal column postsynaptic ( DCPS ) spinomedullary neurons from the cat's lumbosacral enlargement were identified by antidromic stimulation of the cervical dorsal columns and stained intracellularly with horseradish peroxidase. The cell bodies were located in laminae III-IV. Their dendritic arbors were elongated rostrocaudally but narrow mediolaterally. On the average, the arbors were X 5 longer than they were wide. Most of the neurons had nearly all of their dendrites in laminae III-IV and some of the neurons had, in addition, a considerable amount of dendritic surface area in lamina V. Only one neuron had more than a very small amount of dendritic surface area dorsal to lamina III. Seven of the neurons had unmyelinated axon collaterals that arborized extensively and issued varicosity-bearing terminal branches in laminae III-V, both within and beneath their dendritic territories. All of the neurons were excited by myelinated, low-threshold mechanoreceptors. Since the rostrocaudally elongated and mediolaterally narrow dendritic arbors of DCPS neurons are in register with the laminae III-IV terminal distributions of myelinated, low-threshold mechanoreceptors, it is probable that this excitation arises from a monosynaptic and topographically discrete innervation. About one-half of the DCPS neurons were also excited by noxious stimuli. It is probable that this excitation is accomplished by a polysynaptic pathway since DCPS dendritic arbors and nociceptor terminal distributions are largely or completely separate.
Dorsal column postsynaptic (DCPS) spinomedullary neurons in the dorsal horn of spinal segments L6-S1 of adult cats anesthetized with sodium pentobarbital were identified by antidromic stimulation of cervical dorsal columns that were dissected free of, and electrically isolated from, the rest of the spinal cord. The neurons were categorized with respect to natural stimulation of their cutaneous receptive fields. An equal number of low-threshold mechanoreceptive and wide-dynamic-range neurons were found. No DCPS neurons could be classified as nociceptive-specific. All neurons received input from low-threshold mechanoreceptors with myelinated axons. There was no evidence that any neurons received monosynaptic input from unmyelinated, primary afferent fibers. The average conduction velocity of the antidromic responses was 45.7 m/s. Nearly half of the DCPS cells showed an antidromic spike followed by synaptically driven responses that were probably evoked by antidromic invasion into the intraspinal collaterals of A-beta primary afferent fibers that ascended the dorsal columns. Intracellularly recorded synaptic responses of DCPS neurons to dorsal column and receptive field stimulation usually consisted of an EPSP with overriding spike potentials followed by a prolonged IPSP whose amplitude decreased markedly as the stimulus frequency was increased in the range of 5 to 30 Hz. The results indicate that DCPS neurons constitute a projection system capable of signaling innocuous and tissue-damaging mechanical stimuli. The DCPS projection may play a role in the modulation of touch and pain perception.
Dorsal column postsynaptic neurons in the lumbosacral enlargements of cats and a monkey were retrogradely labeled by placing horseradish peroxidase on their severed axons in the thoracic dorsal columns. After visualizing the retrogradely-labeled neurons, the tissue was immunocytochemically stained with an antiserum directed against serotonin. Immunoreactive axonal varicosities contacted the perikarya and proximal dendrites of every retrogradely-labeled neuron examined in cat (mean 61 contacts/cell) and nearly every neuron in the monkey (mean 18 contacts/cell). Electron microscopy showed that the immunoreactive axonal varicosities contained pleomorphic (round to oval) agranular vesicles and formed symmetrical synapses on retrogradely-labeled neurons. It is concluded that dorsal column postsynaptic neurons are innervated directly by the brain stem's descending, serotoninergic system(s).
Explore the source record for details and available documents.
Dorsal column postsynaptic (DCPS) neurons in the lumbosacral enlargements of cats and macaque monkeys were retrogradely labeled by placing HRP on their severed axons within the dorsal columns. The enlargements of both species contained 800-1,100 labeled DCPS neurons. The DCPS projection is thus as large as the feline spinocervical tract. It is very probable that most of these neurons project to the dorsal column nuclei and thereby constitute one of the major sources of somatosensory input to the brain. In the cat, DCPS neurons were concentrated in a band centered in lamina IV that swept down through laminae V-VI along the medial border of the dorsal horn. A second concentration of labeled cells was found in dorsomedial lamina VII. In the monkey, DCPS neurons were concentrated in a relatively broader band in laminae III-IV, and scattered cells were consistently observed throughout laminae V-VII and X and in the dorsolateral white matter. The prominence of the monkey's DCPS projection suggests that humans also have such a projection.
Neurons in lamina II of the lumbar spinal cords of colchicine-pretreated cats were stained immunocytochemically for enkephalin. Two morphological types were found. The most common type had the light microscopic characteristics of stalked cell. The other type was found in the deep part of the lamina and had the light and electron microscopic characteristics of the lamina IIb islet cell.
The morphology an location of physiologically characterized neurons in layer I of the spinal cord dorsal horn were revealed by the intracellular deposition of horseradish peroxidase (HRP). This material showed that the axons of some layer I neurons issue varicosity-bearing collaterals that generate arbors that overlap the neurons' dendritic territories in layer I.
Explore the source record for details and available documents.
Neurons in Rexed's layer II were physiologically characterized with natural and electrical stimuli applied to their cutaneous receptive fields. The neurons were then intracellularly stained with horseradish peroxidase. Three general patterns of physiological responses were found. Nociceptive specific neurons did not respond to gentle mechanical stimulation. Most responded exclusively to tissue-damaging stimuli. Some also responded to moderately heavy pressure, but these responded to noxious stimuli with an increased discharge frequency. Wide dynamic range neurons responded to both gentle mechanical stimulation and to tissue-damaging stimulation. Low-threshold mechanoreceptive neurons responded only to gentle mechanical stimulation. Some of the low-threshold mechanoreceptive neurons were innervated by primary afferents with unmyelinated axons. Excepting those low-threshold mechanoreceptive neurons with input from unmyelinated afferents, the patterns of primary afferents innervation of layer II neurons were similar to the patterns of innervation that have been found for neurons in layers I and IV-V. All but 2 of the 22 neurons that we found were recognized as being of two general morphological types. Stalked cells had their perikarya situated along the superficial border of layer II. Most of their dendrites traveled ventrally while spreading out rostrocaudally. This gave their dendritic arbors a fan-like shape. Stalked cell axons arborized largely in layer I. Islet cell perikarya were found throughout layer II. Most of their dendrites traveled rostrocaudally. Their dendritic arbors were shaped like cylinders with their long axes parallel to the long axis of the spinal cord. Islet cell axons arborized in the immediate vicinity of their dendritic territories, within layer II. Stalked cells and those islet cells whose dendritic arbors were largely contained within the superficial one-third of layer II (layer IIa) were either nociceptive specific or wide dynamic range neurons. The islet cells whose dendritic arbors were largely within the deeper two-thirds of layer II (layer IIb) were all low-threshold mechanoreceptive neurons. These observations suggest that layers IIa and IIb have different functional roles and that stalked cells and islet cells are separate and distinct components of the neural circuitry of the superficial dorsal horn.
Single cell evoked activity was recorded from spinal cord interneurons in rats prepared with microinjection cannulae or stimulating electrodes in the periaqueductal central gray matter (PAG). Morphine microinjections (4-16 microgram) inhibited the response evoked by a noxious stimulus in 55% of the wide dynamic range neurons tested. Microinjections of etorphine (0.25-0.5 microgram) inhibited 82% of the nociceptive neurons tested. Neither drug inhibited neurons which responded only to innocuous mechanical stimulation. The inhibition of wide dynamic range neurons produced by narcotic microinjection was antagonized by naloxone (1 mg/kg, i.p.) in 7 of 11 cases. Control experiments indicated that the effects obtained with microinjections could not be attributed to the drugs' diffusion to the spinal cord. Focal electrical stimulation of the PAG inhibited the responses to noxious stimuli of 60% of wide dynamic range neurons but was without effect on the responses of neurons that were activated only by innocuous stimuli. These experiments directly demonstrate that narcotic analgesics restricted to an intracerebral site of action activate a neural system which preferentially inhibits the responses of spinal cord wide dynamic range neurons to noxious stimuli. The system has a specificity for nociceptive input since non-nociceptive neurons were unaffected. Directly comparable results were produced by electrical stimulation of the PAG, supporting the concept that stimulation and narcotics modulate the transmission of nociceptive information by similar mechanisms.
Explore the source record for details and available documents.
These experiments characterized the analgesia resulting from exposure to certain noxious and/or stressful manipulations. Rats exposed either to electric grid shock (0.35-2.0 mA for 10-30 sec) or to 5 min of presumably non-painful centrifugal rotation (about 7.0 transverse g's) were analgesic as measured by tail-flick, hot plate and responses to applications of a calibrated paw pinch or alligator clip. Analgesia produced by shock (SA) or centrifugal rotation (RA) persisted after termination of these manipulations. Neither SA nor RA were attended by generalized sensory, attentional or motoric deficits. Intraperitoneal injection of hypertonic saline also increased tail-flick latencies. Exposure to brief ether anesthesia or horizontal oscillation, both of which have been reported to increase ACTH secretion (a commonly used indicator of stress), did not produce analgesia as measured by the tail-flick test. The use of classical conditioning procedures to pair shock with environmental stimuli resulted in increased tail-flick latencies. The narcotic antagonist naloxone (1 mg/kg, i.p.) did not reduce the tail-flick inhibition produced by shock, rotation, hypertonic saline or classical conditioning. Chlordiazepoxide (5 mg/kg, i.p.) also failed to antagonize the increased tail-flick latencies produced by shock or conditioning. Tail-flick inhibition produced by shock or rotation was markedly reduced by complete spinal cord transection at thoracic levels. These results suggest that: (1) the selective modulation of nociceptive input at the level of the spinal cord can be mediated by a supraspinal system or systems physiologically distinct from those involved in analgesia produced by the administration of opiates; (2) non-narcotic modulation of nociceptive input occurring within the spinal cord can be learned by exposure to classical conditioning procedures; and (3) noxious stimuli are sufficient but not necessary to produce a non-narcotic analgesia; stress alone, however, is not always sufficient to produce this analgesia.
These studies examined the effects of bilateral lesions of the dorsolateral funiculus (DLF) of the rat spinal cord on the inhibition of a nociceptive reflex produced either by a systemic injection of 4 mg/kg of morphine or by a 20 sec exposure to 1.0 mA of transcutaneous electric shock. Reflex inhibition was quantified and analgesia inferred by use of a modified version of the D'Amour-Smith tail-flick test. Lesions which included only the DLF reduced morphine-produced analgesia (MA) by 73% but had no effect on shock-produced analgesia (SA) observed in the same rats. Baseline tail-flick latencies of this group were not affected by the lesions. Control lesions restricted to the dorsal columns attenuated neither MA nor SA. Lesions which included both the dorsal columns and DLF did not affect SA and produced no greater reduction in MA than lesions of the DLF alone. Previous work indicates that both MA and SA result, at least in part, from supraspinal activity. The current data indicate that: (1) supraspinal modulation participating in two different types of analgesic induction involves separate descending spinal pathways and (2) the maximal expression of analgesia produced by administration of narcotics requires the integrity of a supraspinal neural system projecting in the DLF.