Structure-function relationships in identified afferent neurones

Summary The review deals with structure-function relationships in primary afferent and spinal cord neurones that were intracellularly injected with a marker substance (mostly HRP) after physiological identification. At the level of dorsal root ganglion (DRG) cells, there is a significant correlation between soma size and conduction velocity (or diameter) of the afferent fibre for most subpopulations of DRG cells, but the scatter of data is considerable, so that the size of a DRG cell soma cannot be predicted from the diameter of its axon or vice versa. The spinal terminations of primary afferent fibres are the best example of a relationship between structure and function, since most of the afferent units possess characteristic patterns of spinal arborization, e.g. the flame-shaped arbors of hair follicle afferents in lamina III of the dorsal horn, or the projection of nociceptive afferents onto lamina I. The morphological features of spinal cord neurones can be used only to a limited extent for functional identification. Thus, many SCT neurones can be recognized by their triangular dendritic tree and STT cells in lamina VII/VIII by their dendritic projection into the white matter. It is still not possible, however, to distinguish a nociceptive STT cell from a low-threshold mechanoreceptive one on the basis of morphological criteria.     

Retrograde changes in Clarke's column following neonatal hemicerebellectomy in the rat

Long-Evans black-hooded rat pups underwent hemicerebellectomies at various postnatal ages. The animals were killed at 28 days of age and thoracolumbar segments of the spinal cord were removed and processed according to Golgi and Nissl techniques. Hemicerebellectomy at two and three days postnatally produced, in addition to cellular hypotrophy in Clarke's column, dendritic alterations which included a decrease in the number of primary dendrites and a reduction of the remaining dendrites to truncated stubs. Hemicerebellectomy at 21 days postnatally had little effect on either cell size or dendritic arborization. The role of the target nucleus in maintaining cell growth and dendritic arborization during critical developmental periods is discussed.     

Synaptic action of the fastigiobulbar impulses upon neurones in the medullary reticular formation and vestibular nuclei

Summary In anaesthetized cats, the fastigial nucleus of cerebellum was stimulated with electric pulse currents, and the effects thereby induced were investigated by recording intracellularly from cells in the medullary reticular formation, the nucleus of Deiters and the descending vestibular nucleus. The early effect commonly seen in these cells was initiation of excitatory postsynaptic potentials (EPSPs) with monosynaptic latencies from both sides of the fastigial nuclei. These EPSPs appeared to be produced in part by a kind of axon reflex through cerebellar afferent fibres, but a certain portion of them was ascribable to the crossed fastigiobulbar axons, as they were influenced by stimulation of the cerebellar cortex in the manner to be expected from the previous study on cerebellar nuclei. These EPSPs were followed by a sequence of a prolonged disfacilitatory hyperpolarization and a late facilitatory depolarization, which apparently reflected the inhibition and disinhibition, respectively, produced in fastigial neurones via Purkinje cell axons of the corticonuclear projection. Either EPSPs or IPSPs were also induced in both reticular and vestibular neurones through polysynaptic pathways in which the fastigiobulbar projection might have been involved.     

A pharmacological study of group I muscle afferent terminals and synaptic excitation in the intermediate nucleus and Clarke's column of the cat spinal cord

Summary When administered microelectrophoretically GABA and piperidine-4-sulphonic acid depolarized the central terminations of muscle group Ia and Ib afferent fibres in the lumbar intermediate nucleus and Clarke's column of cats anaesthetised with pentobarbitone sodium. Both this depolarization, and primary afferent depolarization, generated by impulses in other primary afferent fibres which produce prolonged bicuculline-sensitive inhibition of the firing of group I afferent fibre-excited interneurones in the intermediate nucleus and cells in Clarke's column, are reduced by microelectrophoretic bicuculline methochloride. Systemically administered (±)-baclofen hydrochloride (maximum dose 8 mg kg^–1) depressed the monosynaptic excitation of Clarke's column neurones by impulses in muscle and cutaneous afferent fibres. Microelectrophoretically administered (–)-baclofen reduced the bicuculline-sensitive primary afferent depolarization of group I terminations without, however, reducing the depolarizing action of GABA or piperidine-4-sulphonic acid. The depression by (–)-baclofen of the group I monosynaptic excitation of intermediate nucleus neurones is not reduced by concentrations of bicuculline methochloride adequate to suppress prolonged inhibition of these neurones     

The response of some sympathetic neurones to volleys in various afferent nerves

1. Responses of sympathetic neurones to various afferent inputs have been analysed both in anaesthetized cats with intact neuraxis and in spinal cats.2. In anaesthetized cats electrical stimulation of low threshold group III afferent fibres in skin and muscle nerves inhibited sympathetic neurones and gave depressor responses. The silent period following a sympathetic reflex discharge is most likely due to inhibition brought on by a particular subgroup within this afferent group.3. High threshold group III and group IV afferent fibres excited sympathetic neurones and elicited pressor responses.4. Sympathetic reflex arcs could be temporally facilitated during the somatic afferent induced inhibition, by group IV, and to a lesser extent by group III volleys.5. Section of the spinal cord shortened the time course and lessened the degree of group III inhibition suggesting that suprasegmental pathways are involved in the long-lasting depression following a reflex.6. Baroreceptor afferent stimulation inhibited group III- and group IV- evoked reflexes even with some temporal facilitation in the pathway.7. The special properties of the sympathetic reflex arcs and the relation of the results to other work on B.P. reflexes are discussed. It is suggested that the group III and IV muscle afferents have a chemoreceptor function and are responsible for mediation of the pressor reflex during muscle exercise.     

Responses of cortical neurones to stimulation of the visual afferent radiations

Summary Extracellular spikes of visual cortical neurones in unanesthetized cats were recorded. The latency after electric stimuli of the optic radiation to the onset of firing of cells with concentric fields, simple receptive fields and complex receptive fields was measured.Simple receptive field neurones had response latencies averaging 0.93 msec longer than neurones with concentric fields. The majority of the latter represented responses of geniculate axones terminating in the visual cortex and a few were action potentials of first-order cortical cells. Cells with complex receptive fields had longer latencies than simple field neurones and often showed prolonged cessation of firing following electrical stimulation.Many neurones showed a period of firing arrest after radiation shock. The duration and variability of this period also varied according to receptive field type.The data were consistent with the hypothesis that neurones with simple receptive fields receive the initial contact of incoming geniculocalcarine afferent fibres. Complex receptive field neurones appear to derive their input from other cortical cells rather than from the direct geniculocortical connections.A preliminary report of this research was presented to the German Physiological Society and an abstract published in Pflügers Arch. ges. Physiol. 294, 56 (1967).     

Inhibition of cuneate neurones: its afferent source and influence on dynamically sensitive `tactile'' neurones

1. Responses were recorded in decereberate, unanaesthetized cats from individual cuneate neurones in order to determine firstly, the afferent sources of inhibition on cuneate neurones and secondly, the influence of afferent-induced inhibition on those response features of dynamically sensitive tactile neurones which determine their capacity to code information about parameters of tactile stimuli.

2. For all cuneate neurones which displayed afferent-induced inhibition from areas surrounding or within their excitatory receptive field (71% of the sample) it was consistently found that 300 Hz vibration at low amplitudes (< 25-50 μm) which selectively engages Pacinian corpuscles was an effective source of inhibition. In contrast, steady indentation which activates slowly adapting tactile afferents was quite ineffective, as was low frequency vibration (30 Hz) at amplitudes of < 50-100 μm. The latter stimulus can be used to engage rapidly adapting receptors either within glabrous skin (presumed to be Meissners corpuscles) or in association with hair follicles. It is concluded that afferents from Pacinian corpuscles are the dominant or exclusive source of afferent-induced inhibition of cuneate neurones.

3. For dynamically sensitive neurones responsive to low frequency cutaneous vibration (30 Hz) there was a reduction in the slope of stimulus—response relations with afferent-induced inhibition, but no expansion of the range of stimulus amplitudes over which the neurone responded.

4. The influence of afferent-induced inhibition on the phase-locking of impulse activity to a cutaneous vibratory wave form was examined by constructing post-stimulus time histograms and cycle histograms. Measures of dispersion of impulse activity around the preferred point of firing in the vibratory waveform indicated that the capacity of individual cuneate neurones to code information about the frequency of the cutaneous vibration was not systematically changed in the presence of afferent-induced inhibition.

     

Laminar differences in development of afferent innervation to striate cortex neurones in kittens

Summary We studied 587 cells in the striate cortex of 47 kittens, ranging in age postnatally from the 8th day to the 20th week, to explore differences in rates of functional maturation of cells and afferents among cortical layers. For all cells studied we determined spontaneous activity level, visual responsiveness, responsiveness to electrical stimulation of the afferent pathways and histologically reconstructed their laminar localization. At the 2nd–3rd week postnatally, single shocks applied to the dorsal lateral geniculate nucleus (LGN) elicited relatively inconsistent responses with long latencies in most of the cells in all layers, except in layer V where responses were consistent. In layers IVab, V, and VIu (the upper half of layer VI) the LGN-latency decreased very rapidly in the 4th week, while in layer IVc such a decrease occurred in the 5th week. In layers II+III and VI (the lower half of layer VI) a less substantial decrease in latency occurred during the 4th and 5th week. At 2 weeks, nearly half of layer V cells had orientation-specific visual responses or spontaneous activity, but such cells were rare in other layers. The proportions of the specific or spontaneously active cells matured successively in the order of layers V IVab and VIu IVc II+III and VI. We conclude that the maturation of functional properties of cortical cells may occur successively in the above-mentioned laminar order, and that cell size and connectivity with afferents may be factors influencing the rate of functional maturation of cortical cells. The intracortical synaptic delay was estimated for each of the cells driven mono-synaptically from the LGN. The average delay decreased most rapidly during the 4th and 5th week. Conduction velocities of afferents innervating the mono-synaptic cells in layers IVab and IVc were calculated. The velocity of the former afferents increased very quickly and reached a value suggesting myelination at 4 weeks, while that of the latter afferents at 5 weeks. Since each type of LGN cell is known to project segregatedly to layers IVab and IVc, respectively, this suggests that Y-cells of LGN may mature earlier than X-cells.Supported by a grant from the Ministry of Education of Japan     

Synaptic actions of individual vestibular neurones on cat neck motoneurones

1. Unitary synaptic potentials evoked by the activity of single vestibulocollic neurones were recorded by means of spike-triggered signal averaging in neck extensor motoneurones of decerebrate cats. Properties of the vestibulocollic neurones which produced the potentials were examined.2. Vestibulocollic neurones were first identified as projecting to the C3 grey matter by antidromic microstimulation within the C3 extensor motoneurone pool. The spontaneous or glutamate-driven activity of the vestibulocollic neurones was then used to trigger the averaging computer. In this way ten inhibitory and two excitatory neurones were identified (20% of neurones tested).3. Action potentials in local branches of vestibulocollic neurones were usually recorded in the vicinity of motoneurones. Mean orthodromic conduction time from the foot of the extracellular spike, recorded in the vestibular nuclei, that triggered the averager was 0.72 msec. Mean synaptic delay was 0.4 msec.4. I.p.s.p.s had a mean time to peak of 0.81 msec and were readily reversed by injection of hyperpolarizing current. These data, together with the shape indices of i.p.s.p.s indicate that they are generated proximally on motoneurones.5. All vestibulocollic neurones making synapses with motoneurones were monosynaptically driven by stimulation of the ipsilateral vestibular nerve. Four out of seven tested were inhibited by stimulation of the contralateral vestibular nerve (commissural inhibition).6. Two excitatory neurones were located in Deiters' nucleus or on the Deiters'-descending border. Inhibitory neurones were found relatively medially in the vestibular complex in the medial, descending and Deiters' nuclei.7. Vestibulocollic neurones acting on motoneurones were tested for axon branching to more caudal levels of the spinal cord with electrodes placed at C5-7. Both of the excitatory and two out of nine inhibitory neurones branched.     

Alterations in Clarke's column secondary to granule cell degeneration in the neurological mutant mice,weaver and staggerer

In the neurological mutants weaver and staggerer, the granule cells either do not successfully migrate into the granular (weaver) or, having migrated, die during the second postnatal week (staggerer). We wished to determine if the resulting agranular cortex might produce retrograde transneuronal changes in the spinal cord similar to the changes observed after neonatal hemicerebellectomy. Golgi analysis of Clarke's column indicates that neurons in weaver are affected; neurons in staggerer are not. These observations support the view that neuronal maturation requires the presence of a target nucleus during ‘critical’ postnatal growth periods.     

Synaptic organization of afferent projections to the supramammillary nucleus of the rat

We studied the fine structure of afferent terminals from the preoptic area, the nucleus of the diagonal band of Broca, the infralimbic cortex and the laterodorsal tegmental nucleus within the supramammillary nucleus (SUM) using the anterograde tracing method of horse-radish peroxidase conjugated with wheat germ agglutinin (WGA-HRP). Injection of WGA-HRP into the preoptic area permitted ultrastructural recognition of many anterogradely labeled terminals in the SUM. Almost all labeled terminals (99%) contained clear round synaptic vesicles and formed asymmetric synaptic contacts (Gray's type I). About 86% of labeled terminals from the nucleus of the diagonal band were asymmetric (Gray's type I), whereas 14% contained pleomorphic synaptic vesicles and formed symmetric synaptic contacts (Gray's type II). Almost all labeled terminals from the infralimbic cortex were located in the ventral part of the SUM, and 95% of labeled terminals were Gray's type I. The majority of labeled terminals (90%) from the laterodorsal tegmental nucleus were Gray's type I, and the remaining (10%) were Gray's type II. The percentage of labeled terminals with dense-cored vesicles was very high in terminals from the preoptic area (70%), and low in terminals from the infralimbic cortex (19%). Labeled terminals in all cases contacted mainly intermediate-sized dendrites (0.5–1.0 m diameter). All cases had only a few labeled axosomatic terminals. The cases of injections into the preoptic area and the diagonal band nucleus had some reciprocal connections at the ultrastructural level.