Widely distributed GABA-mediated afferent inhibition processes within the ventrobasal thalamus of rat and their possible relevance to pathological pain states and somatotopic plasticity

Summary We have recently described extensive inhibitory interactions between inputs to the ventroposterolateral (VPL) (Roberts and Wells 1990, 1991) and ventropos-teromedial (VPM) (Salt 1989) portions of the ventrobasal nucleus of the thalamus (VB). We wished to determine whether (i) the inhibition observed in the VPL was operating at the thalamic level, (ii) was dependant on GABA receptors, (iii) was demonstrable on neurons of the ventro-posteromedial nucleus of the thalamus (VPM) and (iv) was operant on test responses evoked by natural stimuli. Conditioning stimulation of sciatic nerve afferents caused inhibition of air jet evoked test responses of single VB neurons in urethane-anaesthetized rats. Both VPM and VPL neurons were subject to inhibition by conditioning stimulation of hindlimb afferents, indicating the widespread nature of the inhibitory process. This inhibition was reduced by the iontophoretic application of SR95531, a GABA_A receptor antagonist. We conclude that there is a widely distributed inhibitory system operating in the somatic thalamus which involves both the medial and lateral portions of the nucleus and is, at least in part, mediated by GABA_A receptors. The possible involvement of inhibitory processes and intrinsic membrane properties of thalamic neurones in the somatotopic plasticity of the sensory thalamus following deafferentation and in deaf-ferentation pain is discussed.     

Laminated cytoplasmic bodies and annulate lamellae in the cat ventrobasal and posterior thalamus

Large and small laminated cytoplasmic bodies are reported in neurons and dendrites of the ventrobasal and posterior (PO) nuclear groups of the thalamus of the cat. The bodies are more frequently seen in dendritic profiles than in nerve cell bodies. They differ in size, as well as in number and complexity of orientation of the constituent tubules. Their topographic relationship to endoplasmic reticulum, synapses and adhesion plaques is noted and their possible evolution is discussed. A single collection of annulate lamellae is described in the perinuclear soma of one neuron.     

Electrophysiological properties of lemniscal afferents in rat after kainic acid lesions in the ventrobasal thalamus

Kainic acid (KA) has been largely used as a neurotoxin, and its axon-sparing effect being repeatedly emphasized, on the basis of anatomical and biochemical data. The present study examines this 'axon-sparing' effect from an electrophysiological point of view and demonstrates that lemniscal fibers retain the capacity to convey somesthetic information 5-60 days after an injection of KA in the ventrobasal complex of the thalamus depriving these afferent fibers of their target cells.     

Some aspects of the synaptic circuitry underlying inhibition in the ventrobasal thalamus

Summary We describe here, and review, the ultrastructural features and synaptic relationships of flat-vesicle containing, presumptively inhibitory presynaptic elements in the glomerular and extraglomerular neuropils of the thalamic ventrobasal (VB) nucleus in monkey, cat and rat. This account is based on EM study of normal material, LM and EM immunocytochemistry for GABA, anterograde tracing with HRP and EM of physiologically characterized interneurons intracellularly injected with HRP. It emerges clearly from this study that attempts to categorize flat-vesicle containing terminals in thalamic tissue as either F-boutons (axon terminals with flattened synaptic vesicles and Gray type II synaptic specializations) or P-boutons (dendritic appendages of interneurons with flattened vesicles) by examining only single sections are likely to produce unreliable results. In many cases it is only by studying serial sections that such profiles can be unambiguously identified. Within glomeruli the P-boutons participate in triplet (triadic) synapses which are thought to mediate rapid feed forward inhibition of projection cells, and serial synaptic arrays involving other P-boutons. Since P-boutons from more than one interneuron are present in individual VB glomeruli, P-bouton to P-bouton synapses may mediate disinhibition of interneurons. We show that dendritic shafts of interneurons make and receive synaptic contacts and that in the monkey, at least, reciprocal synaptic contacts between shafts or between a shaft and a P-bouton are not uncommon. Finally, we confirm that in the rat VB there are insignificant numbers of P-boutons or cells with the morphological and transmitter characteristics of interneurons and we suggest that comparative electrophysiological studies of inhibitory events in rat VB versus those in cat or monkey VB during transmission of somatosensory information might help to clarify the roles of thalamic intrinsic neurons.     

Nitric oxide synthase-containing projections to the ventrobasal thalamus in the rat

Microiontophoretic studies of thalamic neurons suggests that nitric oxide (NO) plays an important role in mediating somatosensory transmission. The thalamus contains few nitric oxide synthase (NOS)-immunoreactive neurons; thus, the major source of thalamic NO is presumably from NOS-positive axons of extrathalamic origin. The cells of origin of these putative NOS-containing pathways to the ventrobasal thalamus were investigated in rats by combining retrograde tracing with immunocytochemistry for NOS. The location and morphology of double-labeled neurons was compared with that of single-labeled neurons. The most significant sources of NOS-containing afferents to the thalamus were found to be the pedunculopontine (PPN) and laterodorsal tegmental (LDT) nuclei. NOS-immunoreactive neurons in these cholinergic nuclei project bilaterally to the thalamus, most strongly ipsilaterally. The thalamus appears to be a major target of PPN, since even selective thalamic injections result in retrograde labeling of at least one third of its NOS-immunoreactive neurons. A significant number of NOS-negative neurons in both the PPN and LDT also project to the thalamus. Minor sources of NOS-containing thalamic afferents include the lateral hypothalamus, the dorsal, median and pontine raphe nuclei, the parabrachial nuclei, and the pontomedullary reticular formation. In all these structures, NOS-negative thalamopetal neurons greatly outnumber the NOS-positive ones. Ascending sensory pathways to the thalamus, including those from the sensory trigeminal nuclei, the dorsal column nuclei, and the spinal cord, as well as the auditory and vestibular centers, arise exclusively from NOS-negative neurons. The major NOS-positive projections are implicated in affective and alerting systems, supporting that NO may act to modulate attentiveness in thalamic relay nuclei.     

Somatotopic distribution of trigeminal nociceptive neurons in ventrobasal complex of cat thalamus

Recordings were made from single thalamic units in the urethan-chloralose anesthetized cat. Altogether 2,905 trigeminal single units having a receptive field in the contralateral trigeminal integument were isolated from the somatosensory part of nucleus ventralis posteromedialis, or VPM proper. Each isolated unit was tested for responses to a series of mechanical stimuli. The stimuli included brushing the skin, touch, pressure, noxious pinch, and pinpricks. The majority of VPM proper units responded with the greatest discharge frequency to gentle mechanical stimulation: either hair movement or light pressure to the trigeminal integument, but 341 units were identified as trigeminal nociceptive units. They were partitioned into two functionally defined subclasses, nociceptive specific (NS) and wide dynamic range (WDR) units, but not intermingled with low-threshold mechanoreceptive (LTM) units. Both NS and WDR units were found at or near the margin of the VPM proper but not outside this nucleus. This marginal area was referred to as the shell region of the VPM proper. A total of 248 NS units was found within the shell region of the caudal third of the VPM proper. This part was called the NS zone. These units were somatotopically organized. In the rostral part of the NS zone, ophthalmic NS units having a receptive field in the contralateral ophthalmic division were located dorsolaterally, maxillary NS units occurred dorsomedially, and mandibular NS units were found ventromedially. In the caudal part of the NS zone, maxillary NS units were encountered in the dorsal shell region, whereas mandibular NS units were found in the ventromedial shell region. Ophthalmic NS units were not found in this part of the NS zone. Altogether 93 WDR units were encountered in the shell region of the VPM proper. They were confined to a narrow band approximately 300 micron wide just rostral to the NS zone. These units were somatotopically organized. Ophthalmic WDR units having a low-threshold center of the receptive field in the contralateral ophthalmic division were located dorsolaterally, maxillary WDR units were located dorsomedially, and mandibular WDR units were located ventromedially. The majority of maxillary as well as mandibular WDR units were activated by electrical stimulation of the contralateral maxillary and/or mandibular canine tooth pulp afferents. Both NS and WDR zones of the VPM proper extended into the shell region of the nucleus ventralis posterolateralis (VPL).(ABSTRACT TRUNCATED AT 400 WORDS)     

Neurons in ventrobasal region of cat thalamus selectively responsive to noxious mechanical stimulation

1. A survey was made of neurons located in the ventral posterior lateral nucleus of the cat thalamus and its immediate vicinity for elements with specifically nociceptive properties. 2. Pipette microelectrodes filled with a dye solution were used to obtain extracellular recordings of unitary activity in 34 animals anesthetized with chloralose. 3. The great majority of the over 1,000 different single units responding to sciatic nerve stimulation noted in this series of experiments could also be excited by innocuous mechanical stimulation of skin or subcutaneous tissues. An infrequent but consistently noted group of units excited by A-alpha beta delta sciatic nerve volleys did not respond to innocuous mechanical manipulation or A-alpha beta sciatic nerve volleys; they were excited only by either noxious levels of mechanical stimulation or when volleys included the activity of more slowly conducting myelinated fibers. The latencies of such "high-threshold" units to sciatic volleys were longer than those of the other units. 4. Histologically identified recording sites marked by dye were recovered for 17 high-threshold units. Twelve of the 17 could be excited by noxious manipulations of restricted parts of the contralateral hindlimb. Nine of the 12 had cutaneous receptive fields, whereas 3 responded only to stimulation of subcutaneous tissues. None of the 17 high-threshold units evidenced additional discharges that could be correlated with the C-fiber component of sciatic nerve volleys. 5. The high-threshold units typically exhibited a low level of irregular background activity, which increased on repeated noxious stimulation of the peripheral receptive fields. Tactile units of the same or adjacent penetrations usually had a much greater degree of ongoing activity, often marked by bursts at a relatively high frequency. 6. The recording sites for the 17 high-threshold neurons were located dorsal and ventrolateral to the core of the ventrobasal nuclei and were not found in the midst of the low-threshold, cutaneous, mechanoreceptive population. During vertical stereotaxic penetrations, high-threshold units were noted dorsal or ventral to the location of ventrobasal tactile units in a pattern consistent with the core's somatotopic arrangement. 7. These results support the concept that the cat ventrolateral thalamus receives a small but distinct selectively nociceptive projection. The nociceptive neurons appear to be located in a shell that surrounds the main tactile projection to the ventral posterior lateral nucleus and that retains at least part of the topographic arrangement characteristic of the tactile core. Presumably, this projection is part of an organization identifying and localizing noxious stimulation.     

Immunocytochemical study of serotoninergic and noradrenergic innervation of the ventrobasal complex of the rat thalamus

Whereas the anatomy and function of monoaminergic afferents to the spinal cord areas involved in somesthesia and pain have been widely studied, little is known about the monoaminergic innervation of the primary somatosensory thalamic relay nucleus. The present study demonstrates immunocytochemically at both the light and the electron microscopic levels the presence of noradrenergic and serotoninergic fibers in the ventrobasal complex of the rat thalamus (VB). Despite the presence of numerous immunoreactive varicosities, synaptic differentiation was not observed at the level of apposition of membranes between monoaminergic afferents and VB neuronal profiles. The hypothesis of a non-synaptic modulation of VB neuronal activity by monoaminergic afferents is discussed.     

Neurones in the ventrobasal complex of the rat thalamus responding to scrotal skin temperature changes

1. In rats the scrotal temperature was raised or lowered with a water-perfused thermode while micro-electrode recordings were made of unit activity in the ventrobasal complex of the thalamus. The electrodes were aimed at the region where evoked responses had been found by electrical stimulation of the scrotum. Recording sites were marked by iontophoresis of dye from the micro-electrode.2. Changes in firing rate of thalamic neurones were only found in the scrotal temperature range of 31-40 degrees C. Within this range, 72% of the 123 cells tested were excited or suppressed by skin warming. At temperatures above or below this range, activity was not affected. Most of the cells responded just to temperature and only 7% were also excited by touch.3. Raising temperature in the range 31-40 degrees C caused 82% of the thermally responding cells to increase their firing rate and 18% to decrease their rate. Individual neurones showed a sudden and maintained change in their activity for scrotal temperature increases of only 2, 1 or even 0.5 degrees C. Mean firing rates changed by factors of about 8 or more with these temperature increases and further warming did not change the rate. These step-like changes in firing rate were found at different points over the whole skin temperature range of 31-40 degrees C, but most were between 33 and 38 degrees C.4. For a given neurone the step-like change in activity occurred once its critical temperature was reached, irrespective of whether this was achieved by a step increase of skin temperature over 1-2 sec or by a slow ramp increase lasting several minutes.5. It is not possible to say whether the skin warm receptors, cold receptors or both were responsible for these thalamic responses, but the results do show that incoming thermal information is considerably processed when it reaches the thalamic level.     

Acetylcholinesterase in the ventrobasal thalamus: transience and patterning during ontogenesis

In this study, maturational alterations in acetylcholinesterase-dependent staining of the thalamic ventrobasal complex of rat and mouse were examined. The study was undertaken to address the question of whether this nucleus exhibits transient acetylcholinesterase positivity during its development and whether the enzyme is likely to be synthesized by its immature intrinsic neurons. Also, the patterning due to acetylcholinesterase staining of cells and fibers, and the developmental changes in these patterns, have not been described in earlier work. In contrast to surrounding thalamic nuclei, the ventrobasal complex is acetylcholinesterase-positive at birth. In rat, acetylcholinesterase staining of the ventrobasal thalamus is still more intense than in adjacent nuclei at the end of the first week postnatally. Virtually all somata in the nucleus are filled with dense reaction product at this time. Ultrastructurally, reaction product is associated with the granular endoplasmic reticulum. At this stage, there is a marked difference in intensity of staining between the medial and lateral subdivisions of the nucleus, and patterned clustering of somata within each subdivision is readily appreciated in acetylcholinesterase-stained material. In the second postnatal week, intrinsic acetylcholinesterase activity is progressively lost. By the end of the third postnatal week, the nucleus is quite pale except for one area. In the posterior portion of the lateral subdivision, adjacent to the nucleus reticularis, interconnecting bundles of acetylcholinesterase-positive fibers enter the nucleus. They course medially in the lateral subdivision and break up into a plexus of fine fibers. The development of acetylcholinesterase-dependent staining patterns in the mouse is quite similar, except that histochemically detectable levels of enzyme are substantially lower in the neonatal period. It is concluded that the ventrobasal complex can be distinguished from other thalamic nuclei in regard to earlier onset and/or transience of acetylcholinesterase staining. Ultrastructural observations suggest that virtually all immature ventrobasal neurons are synthesizing acetylcholinesterase. It is suggested that the transient staining for enzyme is due primarily to alteration in synthesis and/or turnover in neurons of the ventrobasal complex. In addition, the acetylcholinesterase staining reveals a patterning of fibers and cells that also undergoes developmental alteration. Evidence is discussed suggesting that axons in the barrels of somatosensory cortex (SmI) are derived from these transiently acetylcholinesterase-positive somata. Consequently, the loss of acetylcholinesterase fiber staining in the barrels, during the third postnatal week (noted previously), may be related to a decrease in synthesis of enzyme in the neuronal somata of the ventrobasal complex.     

Modulatory effects of serotonin on excitatory amino acid responses and sensory synaptic transmission in the ventrobasal thalamus

Excitatory amino acid receptors are thought to mediate sensory input to the ventrobasal thalamus. There is evidence for a brainstem serotonergic projection to the ventrobasal thalamus which may have a modulatory role. The possibility that serotonin may selectively modulate responses to excitatory amino acid receptor agonists, and its effects on sensory synaptic transmission has been examined in the rat ventrobasal thalamus in vivo. Iontophoretic ejection of serotonin at low currents produced a marked facilitation of responses to excitatory amino acids. In contrast, excitatory responses to cholinomimetic agonists were attenuated. Synaptic transmission was concomitantly enhanced or unchanged in these circumstances. Higher serotonin ejection currents reversed the facilitation, or inhibited excitatory amino acid responses and synaptic transmission. It is concluded that serotonin can modulate responses to excitatory amino acids relatively selectively and that synaptic transmission of somatosensory information through the ventrobasal thalamus may be susceptible to brainstem serotonergic modulation.     

Prefrontal modulation of tactile responses in the ventrobasal thalamus of rats

Prefrontal cortex (PFC) has been implicated in modulation of sensory information processing in somatosensory cortex. However, it remains unclear whether or not PFC regulates sensory information in thalamus. In the present study, the effect of PFC stimulation on tactile responses of neurons in the ventrobasal thalamus (VB) of the rat was investigated by single-unit recording. PFC stimulation significantly enhanced the signal-noise ratio (tactile responses/background activities) in 16 out of 66 VB neurons (24.2%) that had receptive fields in fore or hind limbs. Such changes can be classified into three different categories: (1) PFC stimulation not only increased the tactile responses, but also suppressed the background activities of neurons (six neurons, 9.1%); (2) PFC stimulation only increased the tactile responses of neurons (five neurons, 7.6%); (3) PFC stimulation only suppressed the background activities of neurons (five neurons, 7.6%). Our results suggest that PFC also modulates somatosensory information at the thalamic level.     

Neurons in the dorsal column nuclei of the rat emit a moderate projection to the ipsilateral ventrobasal thalamus

The dorsal column nuclei (DCN; gracile and cuneate nuclei) give rise to the medial lemniscus, the fibre system that provides an organised somatosensory input to the thalamus. Unlike the spinothalamic and trigeminothalamic tracts that project, also to the ipsilateral thalamus, the medial lemniscus system is believed to be entirely crossed. We demonstrate that DCN emit a small number of axons that reach the ipsilateral thalamus. As retrograde fluorescent neuronal tracer Fluoro-gold was stereotaxically injected in the ventrobasal thalamus of nine young adult Wistar rats. The injection foci were voluminous and encroached upon adjacent nuclei, but the periphery of the injection halo never spilled over to the contralateral thalamus. All sections of the contralateral gracile and cuneate nuclei and the midline nucleus of Bischoff contained abundant retrogradely labelled neurons. The comparison with the Nissl-stained parallel sections suggests that approximately 70–80% of the DCN neurons project to the contralateral thalamus. Counting of retrogradely labelled neurons in two cases revealed 4,809 and 4,222 neurons in the contralateral and 265 and 214 in the ipsilateral DCN, respectively. Thus, although less prominent than the ipsilateral spinothalamic tract, the lemniscal system also emits an ipsilateral projection that accounts for about 5% of the neuronal population in DCN that innervates the ventrobasal thalamus.     

Interaction between peripheral afferent activity and presynaptic inhibition of ia afferents in the cat

It has been demonstrated in man that the H-reflex is more depressed by presynaptic inhibition than the stretch reflex. Here we investigated this finding further in the alpha-chloralose-anesthetized cat. Soleus monosynaptic reflexes were evoked by electrical stimulation of the tibial nerve or by stretch of the triceps surae muscle. Conditioning stimulation of the posterior biceps and semitendinosus nerve (PBSt) produced a significantly stronger depression of the electrically than the mechanically evoked reflexes. The depression of the reflexes has been shown to be caused by presynaptic inhibition of triceps surae Ia afferents. We investigated the hypothesis that repetitive activation of peripheral afferents may reduce their sensitivity to presynaptic inhibition. In triceps surae motoneurones, we measured the effect of presynaptic inhibition on excitatory postsynaptic potentials (EPSPs) produced by repetitive activation of the peripheral afferents or by fast and slow muscle stretch. EPSPs evoked by single electrical stimulation of the tibial nerve or by fast muscle stretch were significantly depressed by PBSt stimulation. However, the last EPSP in a series of EPSPs evoked by a train of electrical stimuli (5-6 shocks, 150-200 Hz) was significantly less depressed by the conditioning stimulation than the first EPSP. In addition, the last part of the long-lasting EPSPs evoked by a slow muscle stretch was also less depressed than the first part. A single EPSP evoked by stimulation of the medial gastrocnemius nerve was less depressed when preceded by a train of stimuli applied to the same nerve than when the same train of stimuli was applied to a synergistic nerve. The decreased sensitivity of the test EPSP to presynaptic inhibition was maximal when it was evoked within 20 ms after the train of EPSPs. It was not observed at intervals longer than 30 ms. These findings suggest that afferent activity may decrease the efficiency of presynaptic inhibition. We propose that the described interaction between afferent nerve activity and presynaptic inhibition may partly explain why electrically and mechanically evoked reflexes are differently sensitive to presynaptic inhibition.