Elevated blood pressure in transgenic mice with brain-specific expression of human angiotensinogen driven by the glial fibrillary acidic protein promoter.

In addition to the circulatory renin (REN)-angiotensin system (RAS), a tissue RAS having an important role in cardiovascular function also exists in the central nervous system. In the brain, angiotensinogen (AGT) is expressed in astrocytes and in some neurons important to cardiovascular control, but its functional role remains undefined. We generated a transgenic mouse encoding the human AGT (hAGT) gene under the control of the human glial fibrillary acidic protein (GFAP) promoter to experimentally dissect the role of brain versus systemically derived AGT. This promoter targets expression of transgene products to astrocytes, the most abundant cell type expressing AGT in brain. All transgenic lines exhibited hAGT mRNA expression in brain, with variable expression in other tissues. In one line examined in detail, transgene expression was high in brain and low in tissues outside the central nervous system, and the level of plasma hAGT was not elevated over baseline. In the brain, hAGT protein was mainly localized in astrocytes, but was present in neurons in the subfornical organ. Intracerebroventricular (ICV) injection of human REN (hREN) in conscious unrestrained mice elicited a pressor response, which was abolished by ICV preinjection of losartan. Double-transgenic mice expressing the hREN gene and the GFAP-hAGT transgene exhibited a 15-mm Hg increase in blood pressure and an increased preference for salt. Blood pressure in the hREN/GFAP-hAGT mice was lowered after ICV, but not intravenous losartan. These studies suggest that AGT synthesis in the brain has an important role in the regulation of blood pressure and electrolyte balance.

In vitro studies of the neuroform microstent using transparent human intracranial arteries.

For better understanding of relevant morphology and mechanics, direct visualization of a Neuroform microstent (NFM) within an actual human intracranial artery is essential. Twelve NFM were deployed into 8 various segments of formaldehyde-fixed cadaver intracranial arteries. The arteries were then dehydrated and cleared in methyl salicylate to create transparency. The morphology of NFM was studied by digital macro-photography with a back illumination system. The possible limitations and important findings of the NFM were discussed.

The blood supply of the common peroneal nerve in the popliteal fossa.

We investigated the blood supply of the common peroneal nerve. Dissection of 19 lower limbs, including six with intra-vascular injection of latex, allowed gross and microscopic measurements to be made of the blood supply of the common peroneal nerve in the popliteal fossa. This showed that a long segment of the nerve in the vicinity of the fibular neck contained only a few intraneural vessels of fine calibre. By contrast, the tibial nerve received an abundant supply from a constant series of vessels arising directly from the popliteal and posterior tibial arteries. The susceptibility of the common peroneal nerve to injury from a variety of causes and its lack of response to operative treatment may be explained by the tenuous nature of its intrinsic blood supply.

Mechanisms of cardiovascular responses to glycine injected into the dorsal vagal motor nucleus in rat.

Microinjection of glycine into the dorsal vagal motor nucleus of anesthetized rats elicits increases in arterial pressure and heart rate. In the nucleus tractus solitarii, where cardiovascular responses to injection of glycine may be mediated through release of acetylcholine, there is a dense concentration of glycinergic nerve terminals and glycine receptors. In this study, using immunohistochemical methods, we show that glycine terminals and receptors are present in caudal dorsal vagal motor nucleus, although the concentration of both terminal elements is less than in adjacent nucleus tractus solitarii. Responses to glycine microinjected into the dorsal vagal motor nucleus are blocked by the muscarinic antagonist atropine microinjected at the same site; but, unlike responses to glycine in the nucleus tractus solitarii, responses to glycine in the dorsal vagal motor nucleus are not prolonged by physostigmine. These data support the possibility that endogenous glycine may play a role as a transmitter in the dorsal vagal motor nucleus. Responses to glycine may be mediated through actions at muscarinic receptors but not through acetylcholine itself.

Morphology of peptide-immunoreactive neurons in the rat central nucleus of the amygdala.

The morphological characteristics of peptide-immunoreactive neurons in the rat central nucleus of the amygdala were examined. Observations were compared with details of neuron morphology available from Golgi-stained tissue to determine whether peptide immunoreactivity was associated with specific cell types in the central nucleus. The lateral subdivision (CL) of the central nucleus contained mainly medium-sized, densely spiny neurons. Larger, pyramiform spiny neurons; medium-sized, sparsely spinous neurons; and small, aspinous cells were also present in CL. Somatostatin-, neurotensin-, corticotropin-releasing factor (CRF)-, and enkephalin-immunoreactive neurons in CL were characterized as the medium spiny and larger, pyramiform types. No obvious morphological differences were evident among medium spiny neurons containing different peptides. In the medial subdivision, substance P, neurotensin, somatostatin, and CRF were present within pyramiform, sparsely spinous neurons with long dendrites. Galanin immunoreactivity in the medial subdivision was associated with moderately spiny, pyramiform neurons and a larger, aspinous, polygonal neuron. The ventral subdivision of the central nucleus contained neurons similar to those found in the adjacent medial and lateral subdivisions. In addition, this subdivision contained a characteristic ovoid neuron with long, sparsely spinous dendrites. Vasoactive intestinal polypeptide (VIP) and neurotensin appeared to be present within this cell type. In the lateral capsular subdivision, neurotensin and enkephalin were present in cells resembling the medium spiny neurons characteristic of this part of the central nucleus. Numbers of spindle-shaped, biopolar somatostatin, and VIP neurons were identified in the medial, lateral, and ventral subdivisions. The present results provide evidence for a heterogeneous morphology of peptide-immunoreactive neurons in the rat central nucleus that are distributed across cytoarchitectonic boundaries. Except for substance P, neuropeptides in the central nucleus appear to be expressed by a variety of neurons rather than morphologically characteristic types of cell.

Intrinsic GABAergic neurons in the rat central extended amygdala.

The present study examined the distribution, morphology, and connections of gamma-aminobutyric acid-immunoreactive (GABA-IR) neurons in the three principal components of the central extended amygdala: the central amygdaloid nucleus, the bed nucleus of the stria terminalis (BNST) and the sublenticular substantia innominata. In the central nucleus, large numbers of GABA-IR neurons were identified in the lateral, lateral capsular, and ventral subdivisions, though in the medial subdivision, GABA-IR neurons were only present at very caudal levels. Combined immunocytochemistry-Golgi impregnation revealed that GABA-IR neurons in the lateral central nucleus were medium-sized spiny neurons that were morphologically similar to GABAergic neurons in the striatum. Injections of horseradish peroxidase into the bed nucleus of the stria terminalis labeled a major proportion of the GABA-IR neurons in the central nucleus. In the bed nucleus, the majority of GABA-IR neurons were located in the anterolateral subdivision, ventral part of the posterolateral subdivision and the parastrial subdivision. GABA-IR neurons in the anterolateral bed nucleus were of the typical medium-sized spiny type. Injections of horseradish peroxidase into the central nucleus labeled a few GABA-IR neurons in the posterior part of the anterolateral bed nucleus. GABA-IR neurons were identified in the sublenticular substantia innominata and medial shell of the nucleus accumbens and contributed to the continuum of GABA-IR extending from the central nucleus to the bed nucleus. Injections of horseradish peroxidase (HRP) into the central nucleus, but not the BNST, labeled a few GABA-IR neurons in the substantia innominata. The data point to GABA-IR neurons being a characteristic feature of the central extended amygdala and that GABA-IR neurons participate in the long intrinsic connections linking the major components of this structure. Since lesions of the stria terminalis and basolateral amygdaloid nucleus failed to deplete GABA-IR terminals in the central nucleus, the role of GABA in local and short intrinsic connections in the central extended amygdala is discussed. Further, physiological findings implicating the intrinsic GABAergic system of the central extended amygdala in the tonic inhibition of brainstem efferents are reviewed.

Relationship of thalamic basal forebrain projection neurons to the peptidergic innervation of the midline thalamus.

To better understand the input-output organization of the midline thalamus, we compared the distribution of its peptidergic and monoaminergic afferents, which were visualized by using immunocytochemistry, with the distribution of neurons projecting to different basal forebrain structures, which were mapped using retrograde fluorescent tracers. Serotonin and most of the peptides were found throughout paraventricular thalamic nucleus (PV) and in other midline and intralaminar nuclei (type 1 pattern). Neuropeptide Y, alpha MSH and the catecholamine synthetic enzymes were largely restricted to dorsolateral PV (type 2 pattern). Vasopressin was found in dorsomedial PV and intermediodorsal nucleus in a pattern complementary to the type 2 distribution (type 3 pattern). Neurons projecting to accumbens core were present in paraventricular, intermediodorsal, and other midline nuclei. Neurons projecting to accumbens shell and to central amygdaloid nucleus were found in dorsal PV. The peptidergic zones were only loosely correlated with the distribution of different classes of projection neurons. The type 2 pattern overlapped best with neurons projecting to accumbens shell, and to a lesser extent to central amygdaloid nucleus, while the type 3 pattern overlapped best with neurons projecting to core of accumbens. This partial overlap suggests that some brainstem and hypothalamic nuclei preferentially affect different basal forebrain targets through the midline thalamus, and may allow, for example, information about stress to specifically influence accumbens shell and central amygdaloid nucleus. Nevertheless, most of the peptidergic afferents (type 1 pattern) to midline thalamus cover neurons projecting throughout the basal forebrain, which suggests that all of these neurons receive a variety of brainstem and hypothalamic inputs.

Cortical, thalamic, and amygdaloid projections of rat temporal cortex.

The cortical, thalamic, and amygdaloid connections of the rodent temporal cortices were investigated by using the anterograde transport of iontophoretically injected biocytin. Injections into area Te1 labeled axons and terminals in the ventral regions of the dorsal and ventral subnuclei of the medial geniculate complex, area Te3, the rostrodorsal part of area Te2, and the ventrocaudal caudate putamen. No amygdaloid labeling was observed. Thalamic projections from Te2 targeted the lateral posterior nucleus, the dorsal part of the dorsal subnucleus of the medial geniculate complex, and the peripeduncular nucleus. Corticocortical projections mainly terminated in the dorsal perirhinal cortex, but moderately dense projections were observed in medial and lateral peristriate cortex, and only light projections were observed to Te1 and Te3. Projections to these isocortical regions terminated in layers I and VI. Amygdaloid projections targeted the ventromedial subdivision of the lateral nucleus and the adjacent part of the anterior basolateral nucleus. Area Te3 was observed to project to the ventrolateral parts of the dorsal and ventral subnuclei of the medial geniculate complex, the dorsal perirhinal cortex, rostral Te2, and Te1. In the amygdala, labeled fibers and terminals were concentrated in the dorsolateral subdivision of the lateral nucleus. These data confirm that areas Te1 and Te3 are hierarchically organized cortical areas connected with auditory relay nuclei in the thalamus. Area Te2, in contrast, appears to be weakly connected with Te1 and Te3 but is heavily connected with the peristriate cortex and tectorecipient thalamic nuclei. Te2 appears to be a visually related cortical area. The data also indicate that projections from Te2 and Te3 target different subregions of the lateral nucleus and that Te2, but not Te3, projects to the basolateral nucleus.

Perirhinal cortex projections to the amygdaloid complex and hippocampal formation in the rat.

The differential efferent projections of the perirhinal cortex were traced by using anterograde and retrograde tracing techniques. The dorsal bank cortex (area 36) projected lightly to the lateral entorhinal cortex and more strongly to the lateral, posterolateral cortical, and posterior basomedial amygdaloid nuclei and amygdalostriatal transition zone. The ventral bank (dorsolateral entorhinal cortex) projected to the lateral entorhinal cortex, dorsal subiculum, and subfield CA1 and mainly targeted the basolateral amygdaloid nucleus. Corticocortical projections from the dorsal and ventral banks targeted different cortical areas. The fundus of the rhinal sulcus (area 35) projected to both lateral and medial entorhinal cortices, ventral subiculum, lateral and basolateral nuclei, and amygdalostriatal transition zone. Corticocortical projections targeted areas projected to by both dorsal and ventral banks and also by second somatosensory area, first temporal cortical area, and striate cortex. Neurons projecting to the lateral nucleus were distributed in all layers of the dorsal bank, wheras those projecting to CA1 and subiculum were found in superfical layers (mostly layer III) of the ventral bank. Projections to the basolateral nucleus arose from superfical layers (mostly layer II) of the fundus and deep layers of the ventral bank. Furthermore, projections to the amygdala mostly arose from rostral levels, whereas hippocampal projections primarily originated caudally. The rat perirhinal cortex is heterogeneous in its efferent connectivity, and distinct projections arise from the dorsal and ventral banks and fundus of the rhinal sulcus. The widespread cortical connectivity of the fundus suggests that only this part of the perirhinal cortex is similar to area 35 of the primate brain.

Cortical, thalamic, and amygdaloid connections of the anterior and posterior insular cortices.

Cortical, thalamic, and amygdaloid projections of the rat anterior and posterior insular cortices were examined using the anterograde transport of biocytin. Granular and dysgranular posterior insular areas between bregma and 2 mm anterior to bregma projected to the gustatory thalamic nucleus. Granular cortex projected to the subjacent dysgranular cortex which in turn projected to the agranular (all layers) and granular cortices (layers I and VI). Both granular and dysgranular posterior areas projected heavily to the dysgranular anterior insular cortex. Agranular posterior insular cortex projected to medial mediodorsal nucleus, agranular anterior insular and infralimbic cortices as well as granular and dysgranular posterior insula. No projections to the amygdala were observed from posterior granular cortex, although dysgranular cortex projected to the lateral central nucleus, dorsolateral lateral nucleus, and posterior basolateral nucleus. Agranular projections were similar, although they included medial and lateral central nucleus and the ventral lateral nucleus. Dysgranular anterior insular cortex projected to lateral agranular frontal cortex and granular and dysgranular posterior insular regions. Agranular anterior insular cortex projected to the dysgranular anterior and prelimbic cortices. Anterior insuloamygdaloid projections targeted the rostral lateral and anterior basolateral nuclei with sparse projections to the rostral central nucleus. The data suggest that the anterior insula is an interface between the posterior insular cortex and motor cortex and is connected with motor-related amygdala regions. Amygdaloid projections from the posterior insular cortex appear to be organized in a feedforward parallel fashion targeting all levels of the intraamygdaloid connections linking the lateral, basolateral, and central nuclei.

Cascade projections from somatosensory cortex to the rat basolateral amygdala via the parietal insular cortex.

The pathways by which somatosensory information could be relayed from the cortex to the amygdaloid complex were investigated by using the anterograde axonal transport of biocytin following cortical microinjections. Injections of biocytin into head and limb areas of secondary somatosensory cortex (S2) produced heavy labeling of fibers and terminals in granular and dysgranular parietal insular cortex from bregma to 3.8 mm behind bregma but only extremely sparse labeling in the lateral and basolateral amygdaloid nuclei. Biocytin injections into granular parietal insular cortex produced a heavy labeling of the subjacent dysgranular parietal insular cortex, but only sparse labeling in the basolateral amygdala. Biocytin injections into dysgranular parietal insular cortex resulted in heavy labeling of the subjacent agranular parietal insular cortex and strong labeling of fibers and terminals in the dorsal part of lateral nucleus, with moderate labeling of fibers in the anterior and posterior basolateral nuclei, and the central nucleus. Injections into S2 labeled the ventroposterior medial, ventroposterior lateral and posterior thalamic nuclei; injections in rostral granular and dysgranular parietal insular cortex labeled the ventral posterior and parvicellular part of ventroposterior lateral thalamic nuclei; and injections in middle to caudal dysgranular parietal insular cortex labeled only the posterior nucleus. These results suggest that whereas somatosensory cortex projects only very sparsely to the amygdala, somatosensory-related inputs to the amygdala arise in the dysgranular parietal insular cortex. The association of dysgranular parietal insular cortex with the posterior thalamus suggests it may relay nociceptive information to the amygdala.

The distribution of Timm's stain in the nonsulphide-perfused human hippocampal formation.

A detailed description is given of the distribution of Timm's staining in the human hippocampal formation. In sections cut transverse to the long axis of the hippocampal formation the same pattern of staining is seen throughout almost all of the structure, i.e., through all the main body except the most caudal region and, following the medial bend made by the long axis, through most of the pes hippocampi. The staining typically filled almost the whole of the hilus of the dentate gyrus and the whole depth of the adjacent stratum pyramidale and stratum lucidum of subfield CA3. Rostrally, staining was traced into the tip of the uncus. Caudally, staining was traced to the inferior surface of the splenium at the apex of the caudal taper of the hippocampus. No staining was found in the gyrus fasciolaris, indusium griseum, or anterior hippocampal rudiment. Except in the medial part of the pes hippocampi, the Timm's staining did not reach the proximal border of the subfield CA1. In some sections of the main body of the hippocampus a narrow, infragranular stain-free zone was observed in the hilus. In the dentate gyrus of three older subjects, but not in those of three younger subjects, supragranular staining was found. It is argued that the Timm's staining described in this study is specific to the hippocampal mossy fibres. The distribution of the staining is discussed in relation to the boundaries of subfields CA3 and CA2 as delineated by other authors.

Evidence for a GABAergic interface between cortical afferents and brainstem projection neurons in the rat central extended amygdala.

The synaptic circuitry of the intrinsic GABAergic system of the central extended amygdala (CEA) in relation to efferent neurons and cortical afferents was examined in the present study. Neurons in the CEA projecting to the dorsal vagal complex and the parabrachial complex were identified by the retrograde transport of wheat germ agglutinin-horseradish peroxidase (WGA-HRP). Postembedding GABA-immunocytochemistry revealed that GABA-immunoreactive (GABA-IR) terminals formed largely symmetrical synaptic contacts with the perikarya and proximal dendritic processes of almost all WGA-HRP-labeled neurons in the CEA. To determine the relationship between cortical afferents and CEA GABAergic neurons, WGA-HRP was used to anterogradely label afferents from the insular cortex in combination with postembedding immunogold detection of GABA. Cortical afferents formed asymmetrical synaptic contacts predominantly on small dendrites and dendritic spines. Many of the dendrites postsynaptic to cortical terminals in the central nucleus were immunoreactive for GABA although only relatively few spines were GABA-IR. Combining pre-embedding GAD-immunocytochemistry with cortical lesions resulted in approximately 40% of degenerating terminals of insular cortical origin in the central nucleus in contact with small, GAD-IR dendrites and spines. The present results demonstrate that the neurons providing the major CEA outputs to the brainstem receive an extensive GABAergic innervation, strongly supporting our proposal that CEA efferent neurons are under strong tonic inhibition by intrinsic GABAergic neurons. Further, our finding that the major cortical input to the central nucleus preferentially innervates intrinsic GABAergic neurons suggests that these neurons in the CEA may serve as an interface between the principal inputs and outputs of this forebrain region.

Neuronal architecture in the rat central nucleus of the amygdala: a cytological, hodological, and immunocytochemical study.

The organization of neurons in the rat central nucleus of the amygdala (CNA) has been examined by using Nissl stain and immunocytochemical and retrograde tracing techniques. Four main subdivisions were identified on the basis of quantitative analyses of Nissl-stained material: medial (CM), lateral (CL), lateral capsular (CLC), and ventral (CV). An intermediate subdivision (CI), previously described by McDonald ('82), was apparent only in animals that had HRP-WGA injected into the bed nucleus of the stria terminalis. Large populations of neurotensin-, corticotropin-releasing factor (CRF)-, and enkephalin-immunoreactive neurons were present within the lateral divisions (mainly CL), although they were also seen within CM. Somatostatin-immunoreactive neurons were distributed mainly within CL and CM. Within CL, neurotensin- and enkephalin-immunoreactive neurons were more numerous laterally whereas CRF- and somatostatin-immunoreactive neurons were more numerous medially. Substance P-immunoreactive neurons were almost exclusively confined to CM. Only a few cholecystokinin- and vasoactive-polypeptide-immunoreactive neurons were seen in the CNA, and they were observed within CL, CV, and CM. The majority of neurons projecting to the dorsal medulla, hypothalamus, and ventral tegmental area were located within CM, although a significant number of cells were also seen within CL. Efferent projections to the bed nucleus of the stria terminalis were found to arise from neurons located within all subdivisions of the CNA. Thus, the distributional patterns of peptidergic and efferent neurons were not confined to individual cytoarchitectonically- defined subdivisions of the CNA. Rather, the results suggest overlapping medial to the lateral trends. Comparisons with the results of previous studies indicate that peptidergic and afferent terminal distribution patterns are more restricted to individual cytoarchitectonically defined subregions of the CNA. These observations suggest that the detailed cytoarchitecture of the CNA more likely reflects the functional integration of afferents rather than the organization of the CNA output neurons.

Collateralization of the amygdaloid projections of the rat prelimbic and infralimbic cortices.

Previous studies indicate that the distribution of corticoamygdaloid neurons in the rat prelimbic (PL) and infralimbic (IL) cortices overlaps with the distribution of neurons projecting to the contralateral medial prefrontal cortex (MPC), insular cortex, mediodorsal thalamus, and dorsal medulla. In view of the poorly differentiated cytoarchitecture of PL and IL, and their designation as cortical regions transitional between the allocortex and isocortex, the present study sought to determine whether several cortical and subcortical projections from these areas arise as collaterals of corticoamygdaloid neurons. Injections of the fluorescent dyes Fast Blue (FB) or bisbenzimide (BB) were made into the amygdaloid complex and the following areas: agranular and granular insular cortices; mediodorsal thalamic nucleus (MD); nucleus tractus solitarii/dorsal medulla (NTS); contralateral amygdaloid complex; and ipsilateral and contralateral MPC. Neurons projecting to the ipsilateral amygdaloid complex were located mainly in layers II and V with fewer cells in layer III. Concomitant injections into the insular cortex, MD, and NTS labeled populations of neurons arranged in laminae that partially overlapped with, but were essentially separate from, corticoamygdaloid neurons. Projections to the insular cortex arose from layers II and V; those to MD arose from layers V and VI. Corticobulbar projections from IL originated from neurons arranged in a thin lamina in the deep part of layer V. Very few neurons projecting to both the amygdaloid complex and any of these areas were observed. Bilateral injections of FB and BB into the amygdaloid complex producted very few double-labeled cells in PL and IL. Further, in layer V, ipsilaterally projecting corticoamygdaloid neurons tended to be located more deeply than contralaterally projecting neurons. Combined injections of BB and FB into the amygdaloid complex and the contralateral (but not ipsilateral) MPC resulted in significant numbers of double-labeled neurons in layers II, III, and V of PL and IL. Control injections of fluorescent dyes into the cerebrospinal fluid labeled few neurons in the superficial layers of PL and IL and a combined injection into the amygdaloid complex (FB) and subarachnoid space (BB) resulted in a very small number of double-labeled cells in layer II only. The results suggest that a significant proportion of neurons in PL and IL projecting to the amygdaloid complex issue collaterals innervating the contralateral MPC. Evidence is discussed that suggests that the interhemispheric collaterals of MPC corticoamygdaloid neurons may serve to correlate the amygdaloid outputs of the MPC bilaterally.

Two neurotropic viruses, herpes simplex virus type 1 and mouse hepatitis virus, spread along different neural pathways from the main olfactory bulb.

Several neurotropic viruses enter the brain after peripheral inoculation and spread transneuronally along pathways known to be connected to the initial site of entry. In this study, the pathways utilized by two such viruses, herpes simplex virus type 1 and mouse hepatitis virus strain JHM, were compared using in situ hybridization following inoculation into either the nasal cavity or the main olfactory bulb of the mouse. The results indicate that both viruses spread to infect a unique and only partially overlapping set of connections of the main olfactory bulb. Both quantitative and qualitative differences were observed in the patterns of infection of known primary and secondary main olfactory bulb connections. Using immunohistochemistry for tyrosine hydroxylase combined with in situ hybridization, it was shown that only herpes simplex virus infected noradrenergic neurons in the locus coeruleus. In contrast, both viruses infected dopaminergic neurons in the ventral tegmental area, although mouse hepatitis virus produced a more widespread infection in the A10 group, as well as infecting A8 and A9. The results suggest that differential virus uptake in specific neurotransmitter systems contributes to the pattern of viral spread, although other factors, such as differences in access to particular synapses on infected cells and differences in the distribution of the cellular receptor for the two viruses, are also likely to be important. The data show that neural tracing with different viruses may define unique neural pathways from a site of inoculation. The data also demonstrate that two viruses can enter the brain via the olfactory system and localize to different structures, suggesting that neurological diseases involving disparate regions of the brain could be caused by different viruses, even if entry occurred at a common site.

Glycine-containing terminals in the rat dorsal vagal complex.

The present study examined the distribution of glycine and glycine-receptors in the dorsal vagal complex using pre-embedding immunocytochemistry. Glycine-immunoreactive terminals were present in moderate densities in the medial, intermediate, interstitial, commissural and ventrolateral subnuclei of the nucleus tractus solitarii. The dorsolateral nucleus tractus solitarii and the dorsal vagal motor nucleus contained only very few, scattered glycine-containing terminals. Glycine terminals appeared to be concentrated in regions of the dorsal vagal complex receiving primary vagal afferents, though previous studies have suggested that glycine is not present in these afferents. A conspicuously high concentration of glycine terminals was observed in the medial nucleus tractus solitarii where a population of cholinergic neurons has been identified previously. Ultrastructurally glycine immunoreactivity was principally associated with terminals containing flattened, pleomorphic vesicles and forming symmetrical synaptic contacts, mostly with dendrites. Glycine receptor immunoreactivity was present throughout the dorsal vagal complex with little evidence of subnuclear localization. With electron-microscopic examination, glycine receptor immunoreactivity was associated with dendritic membranes and was associated presynaptically with terminals containing flattened pleomorphic vesicles. Overall, the present data provide evidence consistent with a neurotransmitter role for glycine in the dorsal vagal complex. The presence of glycine in regions of the dorsal vagal complex receiving vagal afferents suggests a prominent role for this neurotransmitter in autonomic regulation.

Glycine receptor (gephyrin) immunoreactivity is present on cholinergic neurons in the dorsal vagal complex.

We previously demonstrated that microinjection of exogenous glycine into the nucleus tractus solitarii of anesthetized rats elicits responses that are qualitatively like those elicited by microinjection of acetylcholine at the same site. The responses to glycine, like those to acetylcholine, are blocked by administration of a muscarinic receptor antagonist and prolonged by administration of an acetylcholinesterase inhibitor. Furthermore, glycine leads to release of acetylcholine from the nucleus tractus solitarii and surrounding dorsal vagal complex. An anatomical framework for interactions between glycinergic and cholinergic neurons was established by studies that identified glycine terminals and receptors in the dorsal vagal complex. The current study investigated the relationship between glycine receptors and neuronal elements that were immunoreactive for choline acetyltransferase in the dorsal vagal complex. Neurons that were immunoreactive for choline acetyltransferase were located in the dorsal motor nucleus of the vagus, hypoglossal nucleus and nucleus ambiguus, and stained cells were also present in medial, intermediate, and ventrolateral subnuclei of the nucleus tractus solitarii. We found that glycine receptors, immunolabeled with an antibody to gephyrin, were present on cholinergic dendrites in the nucleus tractus solitarii. Gephyrin immunoreactivity was also present on dendrites that did not stain for choline acetyltransferase. These data further support the contribution of cholinergic neurons in mediating cardiovascular responses to glycine in the nucleus tractus solitarii.

Direct evidence for nitric oxide synthase in vagal afferents to the nucleus tractus solitarii.

The anatomical relationship between vagal afferents and brain nitric oxide synthase containing terminals in the nucleus tractus solitarii was studied by means of anterograde tracing combined with immunocytochemistry and immuno-electron microscopy. Biotinylated dextran amine was injected into the nodose ganglion with a glass micropipette. Four to eight days following the injection, regions of the nucleus tractus solitarii containing biotinylated dextran amine-labelled vagal afferents and those containing nitric oxide synthase-immunopositive terminals were congruent. Many neurons exhibiting nitric oxide synthase immunoreactivity were found within the biotinylated dextran amine-containing terminal field. However dense labeling of terminals with biotinylated dextran amine precluded determination if the terminals were nitric oxide synthase-immunoreactive. Therefore, we combined degeneration of vagal afferents after removal of one nodose ganglion with nitric oxide synthase immuno-electron microscopy. Axon terminals that possessed characteristic vesicle clusters and were partially or completely engulfed by glial processes were identified as degenerating vagal afferents. Degenerating axon terminals comprised 38% of the total axon terminals in the nucleus tractus solitarii in a sample of sections; and of the degenerating axon terminals, 67% were nitric oxide synthase-immunoreactive. Nitric oxide synthase immunoreactivity was present in 41% of the non-degenerating axon terminals. Prominent staining of dendrites for nitric oxide synthase immunoreactivity indicated that much of the nitric oxide synthase in the nucleus tractus solitarii is not derived from peripheral afferents. Of the total number of dendritic profiles sampled, half were nitric oxide synthase-immunoreactive. Our data support the hypothesis that nitric oxide or nitric oxide donors may be present in primary vagal afferents that terminate in the nucleus tractus solitarii. While this study confirms that vagal afferents contain brain nitric oxide synthase, it demonstrates for the first time that the majority of nitric oxide synthase immunoreactivity in the nucleus tractus solitarii is found in intrinsic structures in the nucleus. In addition, our data show that second or higher order neurons in the nucleus tractus solitarii may be nitroxidergic and receive both nitroxidergic and non-nitroxidergic vagal input.

Increased expression of nitric oxide synthase in the gracile nucleus of aged rats.

Aging is associated with disturbances in autonomic cardiovascular control. The purpose of this study was to test the hypothesis that changes in nitric oxide occur with aging in brainstem nuclei involved in autonomic cardiovascular control. The principal and unexpected finding in this study was that NADPH-diaphorase reactivity, which is considered a marker of neuronal nitric oxide synthase activity, was decidedly increased in the neuronal bodies of the gracile nucleus but decreased in the axons and axon terminals in old compared with young rats. In situ hybridization also revealed that nitric oxide synthase gene expression was enhanced predominantly in the gracile nucleus neurons of aged rats. The differences between the young and old rats were most dramatically evident in the gracile nucleus, but not evident in other brainstem nuclei. The significance of this finding as it might relate to autonomic or other specific neural dysfunction with aging is not evident at this time.