Four groups of tyrosine hydroxylase-immunoreactive neurons in the ventrolateral medulla of rats, guinea-pigs and cats identified on the basis of chemistry, topography and morphology.

The data in the preceding paper [Halliday G. M. and McLachlan E. M. (1991) Neuroscience 43, 531-550] suggest that some neurons in the rostral ventrolateral medulla contain some catecholamine-synthesizing enzymes but may not produce catecholamines. The present study addresses this question directly by comparing the anatomical location and morphology of these neurons with those revealed by formaldehyde-induced fluorescence. Catecholamine-containing somata of rats and guinea-pigs have been demonstrated following FAGLU-perfusion in normal untreated animals, in animals pretreated with pargyline (a monoamine oxidase inhibitor), and in animals pretreated with colchicine (to block axoplasmic transport). The number and location of fluorescent somata in the ventrolateral medulla have been determined in serial coronal sections of tissue from the cervical spinal cord to the level of the facial nucleus. Catecholamine-fluorescent neurons at different levels of the ventrolateral medulla varied in their topography and sensitivity to pharmacological manipulation. However, the rostrocaudal distributions in rats and guinea-pigs were quantitatively remarkably similar implying that homologous groups of catecholamine-containing neurons exist. Comparison between these distributions and those of somata stained immunohistochemically for catecholamine-synthesizing enzymes and neuropeptide Y [Halliday G. M. and McLachlan E. M. (1991) Neuroscience 43, 531-550] revealed that the majority of fluorescent neurons in both species probably contain dopamine-beta-hydroxylase and neuropeptide Y as well as tyrosine hydroxylase. Those neurons lying just caudal to the facial nucleus immunoreactive for tyrosine hydroxylase and phenylethanolamine-N-methyltransferase but not dopamine-beta-hydroxylase and neuropeptide Y also lack catecholamine fluorescence. This rostral group of somata can be identified immunohistochemically in cats. The size and morphology of catecholamine-fluorescent neurons have been analysed in detail, and compared with the same features of the immunohistochemically stained neurons. Three morphological types of catecholamine-containing neurons could be distinguished in material prepared by both techniques from rats and guinea-pigs, and in immunohistochemical material from cats. Rostral tyrosine hydroxylase-positive neurons, which differed morphologically from these three types, were present in all three species. On the basis of anatomical location, neuronal morphology and chemical characteristics, four groups of tyrosine hydroxylase-immunoreactive neurons have been identified in the ventrolateral medulla of rats, guinea-pigs and cats. Only the caudal three of these four groups appear to synthesize catecholamine, probably noradrenaline. From published data it seems likely that these four groups of tyrosine hydroxylase-positive neurons have distinct projections and functions related to cardiovascular and respiratory control.     

Paraventricular neurosecretory neurons: synaptic inputs from the ventrolateral medulla in rats

Effects of electrical stimulation of the ventrolateral medulla on discharge activity of neurosecretory neurons in the paraventricular nucleus (PVN) were studied in male rats anesthetized with urethane-chloralose. Among 35 phasically firing neurosecretory neurons, stimulation of the lateral reticular nucleus and its vicinity produced excitation in 10 and inhibition in 2. The stimulation also enhanced the activity of 40% of the PVN neurosecretory neurons that fired continuously (n = 81); of these responsive neurons, half of the neurons tested (n = 12) were inhibited by i.v. administration of phenylephrine. The result suggests that both vasopressin- and oxytocin-secreting neurons in the PVN receive mainly excitatory synaptic inputs from the ventrolateral medulla.     

Hypotension activates neuropeptide Y-containing neurons in the rat medulla oblongata.

The present study was designed to determine whether neurons within cardiovascular control nuclei of the rat brainstem that become activated following a hypotensive insult also possess the capacity to utilize neuropeptide Y. Adult male Wistar-Kyoto rats were injected with glyceryl trinitrate (10 mg/kg, i.p.) or vehicle, and 4 h later anaesthetized (pentobarbitone, 60 mg/kg, i.p.) and transcardially perfused. The brains were removed and processed by standard two-colour peroxidase immunohistochemistry. Activated cells were determined by incubation with a primary antibody to Fos protein, which was followed by a second incubation with a primary antibody to neuropeptide Y for double labelling of Fos-positive cells. Compared to vehicle, glyceryl trinitrate-induced hypotension caused a marked induction of Fos protein in the caudal one-third of the nucleus tractus solitarius (bregma -14 to -13.3 mm), which tailed off rapidly in more rostral sections. Following hypotension, significant populations of activated cells were also observed in the rostral and caudal ventrolateral medulla. In the caudal nucleus tractus solitarius and the posterior part of the medial nucleus tractus solitarius, respectively, 15 of 104 and 40 of 120 Fos-positive cells exhibited cytoplasmic neuropeptide Y immunoreactivity following hypotension, compared to seven of 40 and 15 of 40 in vehicle-treated rats, indicating a significant (two- to three-fold) increase in double-labelled cells following systemic glyceryl trinitrate (P < 0.05, unpaired t-test). In contrast, in the anterior part of the medial nucleus tractus solitarius, the number of double-labelled cells did not change following hypotension. An increase in double-labelled cells was also observed in the rostral ventrolateral medulla (2.5-fold increase compared to vehicle) and caudal ventrolateral medulla (5.8-fold increase compared to vehicle) following hypotension. These data indicate that, in the rat, neuropeptide Y-containing neurons are involved in the central response to a hypotensive challenge. The primary regions where neuropeptide Y-containing neurons appear to be activated are the caudal one-third of the nucleus tractus solitarius and the caudal ventrolateral medulla/rostral ventrolateral medulla, which are key nuclei associated with the integration of the baroreceptor heart rate reflex and sympathetic vasomotor outflow.     

Late vagal inhibition in neurons of the ventrolateral medulla oblongata in the rat.

Stimulation of cervical vagal afferents elicits long-lasting inhibitory effects in a variety of neuronal populations, although little is known concerning the cellular mechanisms that are involved in these effects. In the present study, the electrophysiological characteristics of responses elicited by cumulative activation of vagal afferents were examined in neurons of the rostral ventrolateral medulla oblongata, which play an important role in the coordination of cardiovascular and other visceral activities. The study has focused on the late-onset, slow inhibitory component of vagal responses, which is likely to affect the temporal modulation of postsynaptic effects. Vagal stimulation elicited four distinct response patterns in intracellularly penetrated neurons (n = 78): excitation, inhibition, excitation-inhibition and inhibition-inhibition. The late inhibitory component was encountered in 43 (55%) of the cells, including five putative medullospinal neurons. It was due to a postsynaptic hyperpolarization which reversed at potentials more negative than -83 mV. The voltage dependency, as well as the average onset latency (93+/-3.0 ms), duration (270+/-16.5 ms) and amplitude (1.3+/-0.2 mV as measured at resting membrane potentials), of late inhibition were clearly different from those of the short-latency inhibitory response. The differences in the voltage dependency and time-course of the short-latency responses and the late inhibition indicate that they are mediated by different central relays. In the majority of neurons, late inhibition could be elicited by stimulating only myelinated vagal afferents. The magnitude of the response was, however, significantly enhanced in 63% of the examined cells when the intensity of stimulation was raised to recruit further myelinated and non-myelinated fibres. This indicates that late vagal inhibition is often elicited by a cumulative activation of convergent afferent inputs. The intracellularly labelled vagally responsive neurons were present at all rostrocaudal levels of the rostral ventrolateral medulla, with an accumulation in the region of the lateral paragigantocellular nucleus. Neurons that exhibited late vagal inhibition were dominant in the juxtafacial region of this nucleus. Due to its slow time-course, late vagal inhibition may contribute to a tonic modulation of the activity of neurons in the rostral ventrolateral medulla oblongata. It is proposed that late vagal inhibition plays an important role in the temporal integration of sensory inputs in neurons of the rostral ventrolateral medulla oblongata. The time-course and strength of this modulatory effect are related to the level of activity in those visceral sensory inputs that converge onto the inhibitory interneurons that mediate late inhibition to rostral ventrolateral medulla oblongata neurons.     

Microiontophoresis of flurazepam on inspiratory and postinspiratory neurons in the ventrolateral medulla of cats: an intracellular study in vivo

Effects of flurazepam on the periodic inhibitory postsynaptic potentials (IPSPs) and on the action of locally applied gamma-aminobutyric acid (GABA) were studied in bulbar respiratory neurons of decerebrate cats using concentric multibarrelled electrodes for intracellular recording and extracellular iontophoresis. Iontophoresis of flurazepam augmented spontaneous IPSPs and increased the hyperpolarization induced by GABA. Iontophoretic application of bicuculline suppressed the action of flurazepam. The reversal potential for spontaneous IPSPs was similar to that for the GABA-response. Intracellular Cl- injection shifted both the IPSP wave and the GABA response in a depolarizing direction. Flurazepam enhanced these depolarizing responses. These results suggest that GABA mediates the postsynaptic inhibition in bulbar respiratory neurons.     

Occurrence of neuropeptide Y (NPY)-like immunoreactivity in catecholamine neurons in the human medulla oblongata

We report here the coexistence of a neuropeptide and catecholamines in neurons of the human brain. Using indirect immunofluorescence histochemistry, combined with elution and restaining experiments, neurons in the medulla oblongata of man were demonstrated to contain both a neuropeptide Y-like peptide and the catecholamine synthesizing enzyme tyrosine hydroxylase.     

Parvalbumin in respiratory neurons of the ventrolateral medulla of the adult rat

A column of parvalbumin immunoreactive neurons is closely associated with the location of respiratory neurons in the ventrolateral medulla of the rat. The majority (66%) of bulbospinal neurons in the medullary ventral respiratory column (VRC) that were retrogradely labeled by tracer injections in the phrenic nucleus were also positive for parvalbumin. In contrast, only 18.8% of VRC neurons retrogradely labeled after a tracer injection in the VRC, also expressed parvalbumin. The average cross-sectional area of VRC neurons retrogradely labeled after VRC injections was 193.8 m² ± 6.6 SE. These were significantly smaller than VRC parvalbumin neurons (271.9 m² ± 12.3 SE). Parvalbumin neurons were found in the Bötzinger Complex, the rostral ventral respiratory group (VRG), and the caudal VRG, areas which all contribute to the bulbospinal projection. In contrast, parvalbumin neurons were sparse or absent in the preBötzinger Complex and in the vicinity of the retrotrapezoid nucleus, areas that have few bulbospinal projections. Parvalbumin was rarely colocalized within Neurokinin-1 receptor positive (NK1R) VRC neurons, which are found in the preBötzinger complex and in the anteroventral part of the rostral VRG. Parvalbumin neurons in the Bötzinger Complex and rostral VRG help define the rostrocaudal extent of these regions. The absence of parvalbumin neurons from the intervening preBötzinger complex also helps establish the boundaries of this region. Regional boundaries described in this manner are in good agreement with earlier physiological and anatomical studies. Taken together, the distributions of parvalbumin, NK1R and bulbospinal neurons suggest that the rostral VRG may be subdivided into distinct, anterodorsal, anteroventral, and posterior subdivisions.     

Glucose-sensitive neurons in the rat arcuate nucleus contain neuropeptide Y.

Glucose is known to regulate the activity of the hypothalamic feeding centers. Neuropeptide Y (NPY)-containing neurons in the hypothalamic arcuate nucleus (ARC) have been implicated in the stimulation of feeding. We examined the presence of glucose-sensitive neurons in the ARC and their coincidence with NPY-containing neurons. Cytosolic Ca2+ concentration ([Ca2+]i) in single ARC neurons isolated from rat hypothalamus was measured with fura-2 fluorescence imaging; the cells were then stained immunocytochemically with an anti-NPY antiserum. Lowering the glucose concentration from 10 to 1 mM increased [Ca2+]i in 36 out of 180 neurons (20%), the majority of which (34 neurons, 94%) were immunoreactive for NPY. In conclusion, the ARC contains glucose-sensitive NPY-containing neurons. The suggested role of these neurons is to transduce a reduction in the glucose concentration in the brain to the release of NPY and, subsequently, stimulation of feeding.     

Loss of putative GABAergic neurons in the ventrolateral medulla in multiple system atrophy

Study ObjectivesMultiple system atrophy (MSA) is associated with disturbances in cardiovascular, sleep and respiratory control. The lateral paragigantocellular nucleus (LPGi) in the ventrolateral medulla (VLM) contains GABAergic neurons that participate in control of rapid eye movement (REM) sleep and cardiovagal responses. We sought to determine whether there was loss of putative GABAergic neurons in the LPGi and adjacent regions in MSA.MethodsSections of the medulla were processed for GAD65/67 immunoreactivity in eight subjects with clinical and neuropathological diagnosis of MSA and in six control subjects. These putative GABAergic LPGi neurons were mapped based on their relationship to adjacent monoaminergic VLM groups.ResultsThere were markedly decreased numbers of GAD-immunoreactive neurons in the LPGi and adjacent VLM regions in MSA.ConclusionsThere is loss of GABAergic neurons in the VLM, including the LPGi in patients with MSA. Whereas these findings provide a possible mechanistic substrate, given the few cases included, further studies are necessary to determine whether they contribute to REM sleep-related cardiovagal and possibly respiratory dysregulation in MSA.     

Geniculo-geniculate projection of enkephalin and neuropeptide Y containing neurons in the intergeniculate leaflet of the thalamus in the rat

Geniculo-geniculate projections of immunoreactive neurons for gamma-aminobutyric acid (GABA), leucine-enkephalin (ENK), neuropeptide Y (NPY) and substance P (SP) in the intergeniculate leaflet (IGL) and ventral lateral geniculate nucleus (VLG) of the rat were examined by using a combination of retrograde tracing method and immunocytochemistry. After injection of the fast blue (FB) dye into the IGL and VLG, many ENK immunoreactive neurons, and some NPY immunoreactive neurons were labelled by FB dye in the contralateral IGL. However, GABA and SP immunoreactive neurons in the IGL and VLG were not labelled retrogradely by FB dye. A unilateral electrical lesion of the IGL and VLG caused a reduction of the ENK and NPY immunoreactive fibres in the contralateral IGL and VLG. These findings suggested that ENK is one major component of the neuroactive substances in the geniculo-geniculate projection, and NPY also contributes partly to this projection. Furthermore, the unilateral destruction of the IGL and VLG showed a marked loss of the NPY immunoreactive fibres and a slight loss of the ENK immunoreactive fibres in the bilateral suprachiasmatic nuclei.     

Sympathetic and metabolic mechanisms of the cerebrovasomotor function of the caudal ventrolateral medulla in rats.

We attempted to elucidate the cerebrovasomotor function of the caudal ventrolateral medulla. Sixty-one rats were anaesthetized, paralysed and artificially ventilated. The microsphere method was employed for the measurement of blood flow. Microinjection of an antagonist of excitatory amino acids, kynurenate (2 nmol), into functionally identified depressor sites within the caudal ventrolateral medulla produced arterial hypertension of about 140 mmHg. We found that the cerebral blood flow was substantially increased, but was maintained at the same level (17 rats) as that observed under phenylephrine-induced hypertension (26 rats). Bilateral severing of the cervical sympathetic trunks resulted in a further increase in blood flow in all brain regions studied (18 rats). The response was most significant in the cerebral parasagittal cortex (164 +/- 31% of baseline without, and 211 +/- 43% with sympathectomy; mean +/- S.D.; P < 0.001). The contributions of the cerebral metabolic mechanism to this flow increase under denervation was minimal, as evidenced by the observation of disproportionately smaller changes in cerebral metabolic rate for oxygen during any type of hypertension. We conclude that the cerebrovasomotor functions of the caudal ventrolateral medulla may operate to keep an equilibrium between simultaneously working tonic inhibitions against sympathetic vasoconstriction as well as against vasodilatation. This dual effect is mediated by excitatory amino acid receptors located within this particular brain area. The vasodilator mechanism may be of neurogenic origin. When the function of the brain area is suppressed, the subsequently disinhibited vasodilator mechanism dominates the cerebrovascular autoregulatory function.(ABSTRACT TRUNCATED AT 250 WORDS)     

The distribution of neuropeptide Y-like immunoreactive neurons in the human medulla oblongata

We have described the distribution of neuropeptide Y-like immunoreactive neurons in the medulla oblongata of the adult human. The majority of neuropeptide Y-like immunoreactive cells were found in four regions of the medulla: the ventrolateral reticular formation, the dorsomedial medulla, the secondary sensory nuclei and the rostral raphe nuclei. The morphology of neuropeptide Y-like immunoreactive cells varied in each of these regions. In the ventrolateral reticular formation, the labelled neurons were round and pigmented caudal to the obex but elongated and non-pigmented rostral to the obex; in the dorsomedial medulla, they were triangular and pigmented caudal to but not rostral to the obex; in the secondary sensory nuclei, they were multipolar, non-pigmented and significantly smaller than in the other areas; in the rostral raphe nuclei, they were bipolar and non-pigmented. Colocalization studies revealed that many neuropeptide Y-like immunoreactive cells also synthesize monoamines, consistent with conclusions based on a quantitative comparison of their distributions. Neuropeptide Y-like immunoreactivity was present in about 25% of presumed noradrenaline-synthesizing cells in the caudal ventrolateral medulla (corresponding to the A1 region); about 50% of adrenaline- and 70% of presumed serotonin-synthesizing cells in the rostral ventrolateral medulla (C1 and B2-3 regions); 90-100% of presumed noradrenaline-synthesizing cells in the dorsomedial medulla at and above the obex (A2 region); about 50% of adrenaline-synthesizing cells in the rostral dorsomedial medulla (C2 region); about 5% of presumed serotonin-synthesizing cells in the rostral raphe nuclei (B2-3 region). The largest of these groups was the presumed serotonin-synthesizing cells that contained neuropeptide Y-like immunoreactivity in the rostral ventrolateral medulla. This is the first report of such a cell group in the medulla of any mammal, and emphasizes the neuroanatomical differences between humans and other species.     

The distribution of dopamine-beta-hydroxylase, neuropeptide Y and galanin in locus coeruleus neurons

The locus coeruleus (LC) is composed of noradrenaline-producing neurons that project widely throughout the neuraxis. Subpopulations of LC neuron perikarya have been shown to contain neuropeptide Y (NPY) and galanin (GAL). In the major terminal fields of LC projections, the cerebral cortex, dorsal thalamus and cerebellar cortex, there are differing plexuses of dopamine-beta-hydroxylase (DBH), NPY and GAL immunoreactive axons. DBH immunoreactive plexuses are found in all areas which conform in appearance to previous demonstrations of noradrenaline localization by fluorescence histochemistry. In contrast, there are few NPY immunoreactive axons in thalamus and cerebellum, and the cortical plexus, while similar to the DBH immunoreactive plexus, is not affected by 6-hydroxydopamine treatment. Similarly, there are few GAL immunoreactive axons in either cerebral cortex, dorsal thalamus or cerebellar cortex. Transection of ascending LC axons results in accumulation of DBH but not NPY or GAL immunoreactivity proximal to the lesion. These observations indicate that NPY and GAL are distributed differently in LC neurons from noradrenaline and DBH.     

Target-related patterns of co-existence of neuropeptide Y, vasoactive intestinal peptide, enkephalin and substance P in cranial parasympathetic neurons innervating the facial skin and exocrine glands of guinea-pigs

The patterns of co-existence of neuropeptides in cranial autonomic neurons of guinea-pigs have been examined with quantitative double-labelling immunofluorescence and retrograde axonal tracing using Fast Blue. Within the sphenopalatine, otic, sublingual and submandibular ganglia, and a prominent intracranial ganglion associated with the glossopharyngeal nerve, most neurons contained immunoreactivity of vasoactive intestinal peptide, neuropeptide Y, enkephalin and substance P in combinations that were correlated with their projections. Hair follicles in the facial skin formed a major target of sphenopalatine ganglion cells. The combinations of peptides co-existing in these neurons depended upon the region of the skin where the follicles were located. The parotid gland was innervated by neurons with cell bodies in the otic ganglion or the intracranial ganglion. Most of these neurons contained immunoreactivity to all four peptides. The sublingual gland was innervated by local ganglion cells usually containing immunoreactivity to neuropeptide Y, vasoactive intestinal peptide and substance P. The submandibular gland was innervated by local ganglion cells containing enkephalin immunoreactivity and low levels of immunoreactivity to neuropeptide Y. Presumptive vasodilator neurons, containing immunoreactivity to vasoactive intestinal peptide but no other peptide examined here, comprised less than 10% of cranial autonomic ganglion cells. These results demonstrate that the patterns of co-existence of neuropeptides in cranial autonomic neurons show a high degree of target specificity. The discovery that hair follicles form a major parasympathetic target implies a broader range of actions of cranial autonomic neurons than has been suspected until now.     

Effects of gamma-aminobutyric acid on putative sympatho-excitatory neurons in the rat rostral ventrolateral medulla in vitro. Intracellular study

Neurons in the rat rostral ventrolateral medulla (RVLM) were electrophysiologically characterized and identified using an intracellular recording technique in vitro. The recorded neurons could be classified into three types: spontaneously active neurons with a regular pattern of action potential generation; spontaneously active neurons with an irregular pattern of discharge; and silent neurons. In regularly firing neurons during hyperpolarization below spike generation level there occurred: (a) a 'resetting' of regular pattern of firing; (b) the absence of underlying excitatory postsynaptic potentials; (c) an anomalous rectification that produced a decay in the hyperpolarization. In regularly firing neurons, gamma-aminobutyric acid (GABA) (2-5 microM) produced a reversible membrane hyperpolarization, reduction of frequency of discharge and a moderate decrease in membrane input resistance. These effects were completely blocked in the presence of the GABAa antagonists bicuculline (16 microM) or picrotoxin (50 microM). However, the superfusion of bicuculline (16 microM), or picrotoxin (50 microM) alone elicited depolarization, increase in firing rate and increase of membrane input resistance. This study has provided evidence for regularly firing neurons in the RVLM in vitro, with strikingly similar electrophysiological characteristics to a group of neurons described in vivo as tonic sympathoexcitatory. In vitro they are still modulated by gabaergic inputs acting predominantly upon GABAa receptors.     

Ultrastructural localization of extranuclear progestin receptors relative to C1 neurons in the rostral ventrolateral medulla

To better understand the role of progestins in the C1 area of the rostral ventrolateral medulla (RVLM), immunocytochemical localization of progestin receptors (PRs) was combined with tyrosine hydroxylase (TH) in single sections of RVLM from proestrus rat brains prepared for light and electron microscopy. By light microscopy, PR-immunoreactivity (-ir) was detected in a few nuclei that were interspersed between TH-labeled perikarya and dendrites. Electron microscopy revealed that PR-ir was in several extranuclear locations. The majority of PR-labeling was in non-TH immunoreactive axons (51+/-9%) near the plasma membrane. Additional dual labeling studies revealed that PR-immunoreactive axons could give rise to terminals containing the GABAergic marker GAD65. PR-ir also was found in non-neuronal processes (29+/-9%), some resembling astrocytes. Occasionally, PR-ir was in non-TH-labeled terminals (10+/-3%) affiliated with clusters of small synaptic vesicles, or in patches contained in the cytoplasm of dendrites (10+/-1%). These findings suggest that progestins can primarily modulate neurons in the C1 area of the RVLM by presynaptic mechanisms involving GABAergic transmission. Moreover, they suggest that PR activation may contribute to progestin's effects on arterial blood pressure during pregnancy as well as to sex differences in central cardiovascular regulation.     

Ultrastructural localization of extranuclear progestin receptors relative to C1 neurons in the rostral ventrolateral medulla

To better understand the role of progestins in the C1 area of the rostral ventrolateral medulla (RVLM), immunocytochemical localization of progestin receptors (PRs) was combined with tyrosine hydroxylase (TH) in single sections of RVLM from proestrus rat brains prepared for light and electron microscopy. By light microscopy, PR-immunoreactivity (-ir) was detected in a few nuclei that were interspersed between TH-labeled perikarya and dendrites. Electron microscopy revealed that PR-ir was in several extranuclear locations. The majority of PR-labeling was in non-TH immunoreactive axons (51 ± 9%) near the plasma membrane. Additional dual labeling studies revealed that PR-immunoreactive axons could give rise to terminals containing the GABAergic marker GAD65. PR-ir also was found in non-neuronal processes (29 ± 9%), some resembling astrocytes. Occasionally, PR-ir was in non-TH-labeled terminals (10 ± 3%) affiliated with clusters of small synaptic vesicles, or in patches contained in the cytoplasm of dendrites (10 ± 1%). These findings suggest that progestins can primarily modulate neurons in the C1 area of the RVLM by presynaptic mechanisms involving GABAergic transmission. Moreover, they suggest that PR activation may contribute to progestin's effects on arterial blood pressure during pregnancy as well as to sex differences in central cardiovascular regulation.