Search

The amygdala and bed nucleus of the stria terminalis (BNST) are thought to subserve distinct functions, with the former mediating rapid fear responses to discrete sensory cues and the latter longer "anxiety-like" states in response to diffuse environmental contingencies. However, these structures are reciprocally connected and their projection sites overlap extensively. To shed light on the significance of BNST-amygdala connections, we compared the antidromic response latencies of BNST and central amygdala (CE) neurons to brain stem stimulation. Whereas the frequency distribution of latencies was unimodal in BNST neurons (approximately 10-ms mode), that of CE neurons was bimodal (approximately 10- and approximately 30-ms modes). However, after stria terminalis (ST) lesions, only short-latency antidromic responses were observed, suggesting that CE axons with long conduction times course through the ST. Compared with the direct route, the ST greatly lengthens the path of CE axons to the brain stem, an apparently disadvantageous arrangement. Because BNST and CE share major excitatory basolateral amygdala (BL) inputs, lengthening the path of CE axons might allow synchronization of BNST and CE impulses to brain stem when activated by BL. To test this, we applied electrical BL stimuli and compared orthodromic response latencies in CE and BNST neurons. The latency difference between CE and BNST neurons to BL stimuli approximated that seen between the antidromic responses of BNST cells and CE neurons with long conduction times. These results point to a hitherto unsuspected level of temporal coordination between the inputs and outputs of CE and BNST neurons, supporting the idea of shared functions.

New data on the organization of the intra-amygdaloid circuit is reviewed, beginning with the basolateral (BL) complex, the main input station of the amygdala for sensory afferents, and concluding with the central (CE) nucleus, an important source of projections to brain-stem structures mediating fear responses. The BL complex is endowed with a highly divergent system of intrinsic glutamatergic connections. Yet, BL projection cells have unusually low firing rates. This apparent contradiction is explained by the presence of powerful inhibitory pressures in the BL amygdala: (1) interneurons that generate large-amplitude inhibitory synaptic potentials and (2) projection cells that express a Ca(2+)-dependent K(+) current that can be activated by subthreshold synaptic inputs. Likewise, excitatory projections from the BL amygdala to the CE nucleus are controlled by clusters of GABAergic neurons, termed the intercalated (ITC) cell masses. In response to BL inputs, ITC cells generate feedforward inhibition in CE neurons. However, ITC neurons exhibit properties that allow them to modify the amount of inhibition they generate depending on the distribution of BL activity in space and time. Indeed, ITC cell masses can inhibit each other via lateromedial connections. Moreover, they express an unusual K(+) conductance that modifies their response to BL inputs depending on their recent firing history. Thus, inhibitory mechanisms of the amygdala allow for flexible, context-dependent gating of BL impulses to the CE nucleus.

This study was performed to examine the hypothesis that thalamic-projecting neurons of mesopontine cholinergic nuclei display activity patterns that are compatible with their role in inducing and maintaining activation processes in thalamocortical systems during the states of waking (W) and rapid-eye-movement (REM) sleep associated with desynchronization of the electroencephalogram (EEG). A sample of 780 neurons located in the peribrachial (PB) area of the pedunculopontine tegmental nucleus and in the laterodorsal tegmental (LDT) nucleus were recorded extracellularly in unanesthetized, chronically implanted cats. Of those neurons, 82 were antidromically invaded from medial, intralaminar, and lateral thalamic nuclei: 570 were orthodromically driven at short latencies from various thalamic sites: and 45 of the latter elements are also part of the 82 cell group, as they were activated both antidromically and synaptically from the thalamus. There were no statistically significant differences between firing rates in the PB and LDT neuronal samples. Rate analyses in 2 distinct groups of PB/LDT neurons, with fast (greater than 10 Hz) and slow (less than 2 Hz) discharge rates in W, indicated that (1) the fast-discharging cell group had higher firing rates in W and REM sleep compared to EEG-synchronized sleep (S), the differences between all states being significant (p less than 0.0005); (2) the slow-discharging cell group increased firing rates from W to S and further to REM sleep, with significant difference between W and S (p less than 0.01), as well as between W or S and REM sleep (p less than 0.0005). Interspike interval histograms of PB and LDT neurons showed that 75% of them have tonic firing patterns, with virtually no high-frequency spike bursts in any state of the wake-sleep cycle. We found 22 PB cells that discharged rhythmic spike trains with recurring periods of 0.8-1 sec. Autocorrelograms revealed that this oscillatory behavior disappeared when their firing rate increased during REM sleep. Dynamic analyses of sequential firing rates throughout the waking-sleep cycle showed that none of the full-blown states of vigilance is associated with a uniform level of spontaneous firing rate. Signs of decreased discharge frequencies of mesopontine neurons appeared toward the end of quiet W, preceding by about 10-20 sec the most precocious signs of EEG synchronization heralding the sleep onset. During transition from S to W, rates of spontaneous discharges increased 20 sec before the onset of EEG desynchronization.(ABSTRACT TRUNCATED AT 400 WORDS)

This study describes the effects of brain-stem cholinergic laterodorsal tegmental (LDT) stimulation on the synaptic responsiveness of anterior thalamic (AT) neurons. A sample of AT cells, physiologically identified by their short-latency (less than 6.5 msec) response to mammillary body (MB) stimulation, was recorded in unanesthetized, chronically implanted cats and in urethane-anesthetized cats. In chronic experiments, LDT stimulation evoked a short-latency (10-20 msec) excitation in most AT cells. Moreover, brief LDT trains (3 shocks at 300 Hz, every 3 sec) enhanced the responsiveness of AT cells to both MB (orthodromic) and cortical (ortho- and antidromic) stimuli. This effect did not vary as a function of the interval between LDT conditioning and MB or cortical testing shocks, but as a function of the number of trials. The effects of LDT stimuli resisted reserpine treatment (0.75 mg/kg), suggesting that they were not dependent on the coactivation of monoaminergic fibers. Finally, LDT trains did not suppress inhibitory processes in AT neurons when conditioning-testing intervals were longer than 60 msec. Intracellular recordings performed in urethane-anesthetized cats revealed that LDT stimulation induced a short-latency depolarization which increased with membrane hyperpolarization and was associated with an increase in apparent membrane conductance. Usually, isolated LDT trains did not evoke lasting changes in membrane potential or conductance. However, when LDT trains were applied every 3 sec, they gradually decreased the apparent membrane conductance without altering the membrane potential. This conductance change had a time course similar to the LDT-induced potentiation of responsiveness observed in the chronic experiments. In some neurons, LDT conditioning trains also induced a marked increase in the probability of fast prepotentials being triggered by subthreshold MB or cortical orthodromic volleys. In order to distinguish the cumulative effects of repeated LDT trains from the possibly slow time course of LDT influences, we studied the effects of a unique 1 sec LDT train (at 30 Hz) on the synaptic responsiveness of AT cells recorded extracellularly in reserpine-treated, urethane-anesthetized animals. Such LDT trains induced a 2.9-fold increase in synaptic responsiveness, reaching its peak 40-50 sec after the LDT train and lasting up to 4 min. Iontophoretic application of the muscarinic blocker scopolamine blocked these long-lasting potentiating effects of LDT stimuli. Removal of cortical and basal forebrain inputs to the AT nuclear complex by appropriate transections did not abolish the potentiating effects of LDT trains.(ABSTRACT TRUNCATED AT 400 WORDS)

Several experimental results indicate that the peribrachial (PB) cholinergic area of the pedunculopontine nucleus is the final relay for the transfer of brainstem-generated pontogeniculo-occipital (PGO) waves to the thalamus. However, the mechanisms underlying the PGO-related activity of PB neurons remain unknown. In order to study these mechanisms, single unit recordings in the PB area were performed in reserpinized cats. Because PGO waves are closely related to rapid eye movements, our microelectrode explorations were also aimed to some structures of the preoculomotor network, namely, the superior colliculus (SC) and parts of the central tegmental field (FTC). We have found several classes of PGO-on cells in the PB area, most of them descharging 80 ms or less before the peak of PGO waves. These cell-classes comprised high-frequency bursting cells, slow-frequency bursting cells, and neurons discharging single spikes or doublets. Intracellular recordings showed that PGO-on single spikes arise from conventional excitatory postsynaptic potentials. Among PGO-related cells in structures outside the PB limits, it was found that most SC cells discharge during or after the PGO, whereas FTC cells increase their discharge rate several hundreds of ms before PGO waves, thus indicating that PGO waves are elaborated long before the activation of PB neurons. Massive retrograde labeling was found in FTC following horseradish peroxidase injections into the PB area. We suggest that long-lead FTC neurons provide an excitatory input to PGO-on PB neurons.

We combined the retrograde transport of wheat germ agglutinin conjugated with horseradish peroxidase with choline acetyltransferase immunohistochemistry to study the projections of cholinergic and non-cholinergic neurons of the upper brainstem core to rostral and caudal intralaminar thalamic nuclei, reticular thalamic complex and zona incerta in the cat. After wheat germ agglutinin-horseradish peroxidase injections in the rostral pole of the reticular thalamic nucleus, the distribution and amount of retrogradely labeled brainstem neurons were similar to those found after tracer injection in thalamic relay nuclei (see preceding paper). After wheat germ agglutinin-horseradish peroxidase injections in the caudal intralaminar centrum medianum-parafascicular complex, rostral intralaminar central lateral-paracentral wing, and zona incerta, the numbers of retrogradely labeled brainstem neurons were more than three times higher than those found after injections in thalamic relay nuclei. The larger numbers of horseradish peroxidase-positive brainstem reticular neurons after tracer injections in intralaminar or zona incerta injections results from a more substantial proportion of labeled neurons in the central tegmental field at rostral midbrain (perirubral) levels and in the ventromedial part of the pontine reticular formation, ipsi- and contralaterally to the injection site. Of all retrogradely labeled neurons in the caudal midbrain core at the level of the cholinergic peribrachial area and laterodorsal tegmental nucleus, 45-50% were also choline acetyltransferase-positive after the injections into central lateral-paracentral and reticular nuclei, while only 25% were also choline acetyltransferase-positive after the injection into the centrum medianum-parafascicular complex. These findings are discussed in the light of physiological evidence of brainstem cholinergic mechanisms involved in the blockade of synchronized oscillations and in activation processes of thalamocortical systems.

The projections of brainstem core neurons to relay and associational thalamic nuclei were studied in the cat and macaque monkey by combining the retrograde transport of wheat germ agglutinin conjugated with horseradish peroxidase with choline acetyltransferase immunohistochemistry. All major sensory (medial geniculate, lateral geniculate, ventrobasal), motor (ventroanterior, ventrolateral, ventromedial), associational (mediodorsal, pulvinar, lateral posterior) and limbic (anteromedial, anteroventral) thalamic nuclei of the cat were found to receive projections from cholinergic neurons located in the peribrachial area of the pedunculopontine nucleus and in the laterodorsal tegmental nucleus as well as from non-cholinergic neurons in the rostral (perirubral) part of the central tegmental mesencephalic field. Specific relay nuclei receive less than 10% of their brainstem afferents from non-cholinergic neurons located at rostral midbrain levels and receive 85-96% of their brainstem innervation from a region at midbrain-pontine junction where the cholinergic peribrachial area and laterodorsal tegmental nucleus are maximally developed. Of the total number of horseradish peroxidase-positive brainstem neurons seen after injections in various specific relay nuclei, the double-labeled (horseradish peroxidase + choline acetyltransferase) neurons represent approximately 70-85%. Three to eight times more numerous horseradish peroxidase-labeled brainstem cells were found after injections in associational (mediodorsal and pulvinar-lateral posterior complex) and diffusely cortically-projecting (ventromedial) thalamic nuclei of cat than after injections in specific relay nuclei. The striking retrograde cell labeling observed after injections in nuclei with associative functions and widespread cortical projections was due to massive afferentation from non-cholinergic parts of the midbrain and pontine reticular formation, on both ipsi- and contralateral sides. After wheat germ agglutinin-horseradish peroxidase injections in the associative pulvinar-lateral posterior complex and mediodorsal nucleus of Macaca sylvana, 45-50% of horseradish peroxidase-positive brainstem peribrachial neurons were also choline acetyltransferase-positive. While cells in the medial part of the cholinergic peribrachial area were found to project especially towards the pulvinar-lateral posterior nuclear complex in monkey, the retrograde cell labeling seen after the mediodorsal injection was mostly confined to the lateral part of both dorsal and ventral aspects of the peribrachial area.(ABSTRACT TRUNCATED AT 400 WORDS)

The projections of basal forebrain neurons to the thalamus and the brainstem were investigated in cats and primates by using retrograde transport techniques and choline acetyltransferase (ChAT) immunohistochemistry. In a first series of experiments, the lectin wheat germ-agglutinin conjugated with horseradish peroxidase (WGA-HRP) was injected into all major sensory, motor, intralaminar, and reticular (RE) thalamic nuclei of cats and into the mediodorsal (MD) and pulvinar-lateroposterior thalamic nuclei of macaque monkeys. In cats numerous neurons of the vertical and horizontal limbs of the diagonal band nucleus and the substantia innominata (SI), including its rostromedial portion termed the ventral pallidum (VP), were retrogradely labeled after WGA-HRP injections in the rostral pole of the RE complex, the MD, and anteroventral/anteromedial (AV/AM) thalamic nuclei. Fewer retrogradely labeled cells were observed in the same areas after injections in the ventromedial (VM) thalamic nucleus, and none or very few after other thalamic injections. After RE, MD, and AV/AM injections, 7-20% of all retrogradely labeled cells in the basal forebrain were also ChAT positive, while none of the retrogradely labeled neurons following VM injections displayed ChAT immunoreactivity. The basal forebrain projection to the MD nucleus was shown to arise principally from VP in both cats and macaque monkeys. In a second series of experiments performed in cats, injections of WGA-HRP in the brainstem peribrachial (PB) area comprising the pedunculopontine nucleus led to retrograde labeling of a moderate number of neurons in the lateral part of the VP, SI, and preoptic area (POA), only a few of which displayed ChAT immunoreactivity. In addition, a large number of retrogradely labeled cells were observed in the bed nuclei of the anterior commissure and stria terminalis after PB injections. In a third series of experiments, the use of the retrograde double-labeling method with fluorescent tracers in squirrel monkeys allowed us to identify a significant number of basal forebrain neurons sending axon collaterals to both the RE thalamic nucleus and PB brainstem area, while no double-labeled neurons were disclosed after injections confined to the ventral anterior/ventral lateral (VA/VL) thalamic nuclei and PB area or following injections in the cerebral cortex and PB area. Our findings reveal the existence of cholinergic and noncholinergic basal forebrain projections to the thalamus and the brainstem in both cats and macaque monkeys. We suggest that these projections may play a crucial role in the control of thalamic functions in mammals.

The projections of cholinergic and non-cholinergic neurons of the rostral brainstem reticular formation to the visual thalamic nuclei (dorsal lateral geniculate - LG, lateral posterior - LP, and perigeniculate - PG) were studied in cat by using the retrograde transport of horseradish peroxidase conjugated with wheat germ agglutinin (WGA-HRP) combined with choline acetyltransferase (ChAT) immunohistochemistry. After thalamic injections, less than 10% of all retrogradely labeled neurons in the upper brainstem reticular core were located at most rostral (perirubral) levels where there are virtually no cholinergic elements. Approximately 75-80% of all HRP-positive neurons in the reticular formation were found between stereotaxic planes anterior 1 and posterior 2, in the peribrachial (PB) area of the pedunculopontine nucleus and in the laterodorsal tegmental (LDT) nucleus. The brainstem afferents to LG and PG thalamic nuclei essentially derive from PB neurons, with a small contribution from LDT cells, whereas the LP thalamic nucleus receives massive inputs from both PB and LDT brainstem nuclei. Of all HRP-positive elements visualized in the PB nucleus after an LG or a PG injection, 87% and 73%, respectively, were also ChAT-positive. Of all HRP-positive elements in the PB and LDT nuclei after an LP injection, 82% and 92%, respectively, were also ChAT-positive. The numbers of labeled neurons in the contralateral brainstem reticular nuclei reach 30% to 50% of the numbers found in the ipsilateral reticular formation. These findings reveal the existence of a prominent cholinergic projection from the brainstem reticular formation to the visual thalamic nuclei. Such a chemospecific projection is probably involved in phasic and tonic events of activated behavioral states.

Following horseradish peroxidase injections in the brainstem peribrachial (PB) area, massive retrograde labeling was found in the posterior hypothalamic region. Single-unit recordings posterior hypothalamic neurons with antidromically identified projections to the PB area revealed that these neurons have higher firing rates in waking than in slow-wave sleep and dissimilar discharge patterns as compared with intralaminar thalamic neurons. The results are discussed in the context of reciprocal hypothalamo-brainstem circuits.