Morphological observations on superficial medullary CO2--chemosensitive areas

Effect of estrogen on vagal afferent projections to the brainstem in the female

The effects of 17β-estradiol (E) on the distribution and density of brainstem projections of small or large diameter primary vagal afferents were investigated in Wistar rats using transganglionic transport of wheat germ agglutinin- (WGA; preferentially transported by non-myelinated afferent C-fibers; 2%), or cholera toxin B-subunit- (CTB, 5%; preferentially transported by large myelinated afferent A-fibers) conjugated horseradish peroxidase (HRP) in combination with the tetramethylbenzidine method in age matched ovariectomized (OVX) only or OVX and treated with E (OVX+E; 30 pg/ml plasma) females for 12 weeks. Additionally, these projections were compared to aged matched males. Unilateral microinjection of WGA-HRP into the nodose ganglion resulted in dense anterograde labeling bilaterally, with an ipsilateral predominance in several subnuclei of the nucleus of the solitary tract (NTS) and in area postrema that was greatest in OVX+E animals compared to OVX only and males. Moderately dense anterograde labeling was also observed in paratrigeminal nucleus (PAT) of the OVX+E animals. CTB-HRP produced less dense anterograde labeling in the NTS complex, but had a wider distribution within the brainstem including the area postrema, dorsal motor nucleus of the vagus, PAT, the nucleus ambiguus complex and ventrolateral medulla in all groups. The distribution of CTB-HRP anterograde labeling was densest in OVX+E, less dense in OVX only females and least dense in male rats. Little, if any, labeling was found within PAT in males using either WGA-or CTB-HRP. Taken together, these data suggest that small, non-myelinated (WGA-labeled) and large myelinated (CTB-labeled) diameter vagal afferents projecting to brainstem autonomic areas are differentially affected by circulating levels of estrogen. These effects of estrogen on connectivity may contribute to the sex differences observed in central autonomic mechanisms between gender, and in females with and without estrogen.

Cardiorespiratory and neural consequences of rats brought past their aerobic dive limit

The mammalian diving response is a dramatic autonomic adjustment to underwater submersion affecting heart rate, arterial blood pressure, and ventilation. The bradycardia is known to be modulated by the parasympathetic nervous system, arterial blood pressure is modulated via the sympathetic system, and still other circuits modulate the respiratory changes. In the present study, we investigate the submergence of rats brought past their aerobic dive limit, defined as the diving duration beyond which blood lactate concentration increases above resting levels. Hemodynamic measurements were made during underwater submergence with biotelemetric transmitters, and blood was drawn from cannulas previously implanted in the rats' carotid arteries. Such prolonged submersion induces radical changes in blood chemistry; mean arterial PCO(2) rose to 62.4 Torr, while mean arterial PO(2) and pH reached nadirs of 21.8 Torr and 7.18, respectively. Despite these radical changes in blood chemistry, the rats neither attempted to gasp nor breathe while underwater. Immunohistochemistry for Fos protein done on their brains revealed numerous Fos-positive profiles. Especially noteworthy were the large number of immunopositive profiles in loci where presumptive chemoreceptors are found. Despite the activation of these presumptive chemoreceptors, the rats did not attempt to breathe. Injections of biotinylated dextran amine were made into ventral parts of the medullary dorsal horn, where central fibers of the anterior ethmoidal nerve terminate. Labeled fibers coursed caudal, ventral, and medial from the injection to neurons on the ventral surface of the medulla, where numerous Fos-labeled profiles were seen in the rats brought past their aerobic dive limit. We propose that this projection inhibits the homeostatic chemoreceptor reflex, despite the gross activation of chemoreceptors.

Acid sensitivity and ultrastructure of the retrotrapezoid nucleus in Phox2b-EGFP transgenic mice

The retrotrapezoid nucleus (RTN) contains noncholinergic noncatecholaminergic glutamatergic neurons that express the transcription factor Phox2b (chemically coded or "cc" RTN neurons). These cells regulate breathing and may be central chemoreceptors. Here we explore their ultrastructure and their acid sensitivity by using two novel BAC eGFP-Phox2b transgenic mice (B/G, GENSAT JX99) in which, respectively, 36% and 100% of the cc RTN neurons express the transgene in complete or partial anatomical isolation from other populations of eGFP neurons. All but one of the eGFP-labeled RTN neurons recorded in these mice were acid activated in slices. These cells contained VGLUT2 mRNA, and 50% contained preprogalanin mRNA (determined by single-cell PCR in the B/G mouse). Two neuronal subgroups were revealed, which differed in discharge rate at pH 7.3 (type I approximately 2; type II approximately 4 Hz) and the degree of alkalization that silenced the cells (type I 7.4-7.6, type II 7.8-8.0). Medial to the RTN, C1 neurons recorded in a tyrosine hydroxylase-GFP mouse were pH insensitive between pH 6.9 and pH 7.5. Ultrastructural studies demonstrated that eGFP-labeled RTN neurons were surrounded by numerous capillaries and were often in direct contact with glial cells, pericytes, and the basement membrane of capillaries. Terminals contacting large proximal eGFP dendrites formed mainly symmetric, likely inhibitory, synapses. Terminals on more distal eGFP dendrites formed preferentially asymmetric, presumably excitatory, synapses. In sum, C1 cells are pH insensitive, whereas cc RTN neurons are uniformly acid sensitive. The RTN neurons receive inhibitory and excitatory synaptic inputs and may have unfettered biochemical interactions with glial cells and the local microvasculature.

Afferent and efferent connections of the rat retrotrapezoid nucleus

The rat retrotrapezoid nucleus (RTN) contains candidate central chemoreceptors that have extensive dendrites within the marginal layer (ML). This study describes the axonal projections of RTN neurons and their probable synaptic inputs. The ML showed a dense plexus of nerve terminals immunoreactive (ir) for markers of glutamatergic (vesicular glutamate transporters VGLUT1-3), gamma-aminobutyric acid (GABA)-ergic, adrenergic, serotonergic, cholinergic, and peptidergic transmission. The density of VGLUT3-ir terminals tracked the location of RTN chemoreceptors. The efferent and afferent projections of RTN were studied by placing small iontophoretic injections of anterograde (biotinylated dextran amine; BDA) and retrograde (cholera toxin B) tracers where RTN chemoreceptors have been previously recorded. BDA did not label the nearby C1 cells. BDA-ir varicosities were found in the solitary tract nucleus (NTS), all ventral respiratory column (VRC) subdivisions, A5 noradrenergic area, parabrachial complex, and spinal cord. In each target region, a large percentage of the BDA-ir varicosities was VGLUT2-ir (41-83%). Putative afferent input to RTN originated from spinal cord, caudal NTS, area postrema, VRC, dorsolateral pons, raphe nuclei, lateral hypothalamus, central amygdala, and insular cortex. The results suggest that 1) whether or not the ML is specialized for CO(2) sensing, its complex neuropil likely regulates the activity of RTN chemosensitive neurons; 2) the catecholaminergic, cholinergic, and serotonergic innervation of RTN represents a possible substrate for the known state-dependent control of RTN chemoreceptors; 3) VGLUT3-ir terminals are a probable marker of RTN; and 4) the chemosensitive neurons of RTN may provide a chemical drive to multiple respiratory outflows, insofar as RTN innervates the entire VRC.

Glutamatergic neuronal projections from the marginal layer of the rostral ventral medulla to the respiratory centers in rats

The marginal layer (ML) that lines the ventral surface of the medulla oblongata (VMS) contains neurons thought to contribute to central chemoreception, the process by which systemic hypercapnia activates respiration. The transmitters and connectivity of ML neurons are poorly known. The present study focuses on a group of nonserotonergic ML neurons, often located in close proximity to the entry point of penetrating blood vessels. These neurons (approximately 300/brain) contain vesicular glutamate transporter2 (VGLUT2) mRNA and are thus probably glutamatergic. They cluster below the caudal half of the facial motor nucleus, lateral to the serotonergic cells of the ML. The projections of serotonergic and nonserotonergic ML neurons were investigated by retrograde labeling with Fluoro-Gold. ML VGLUT2 mRNA-expressing neurons lack spinal projections and innervate the dorsolateral pons and the ipsilateral ventral respiratory column (VRC), most particularly, the region of the pre-Bötzinger complex and rVRG. The latter two regions receive a very small input from ML serotonergic neurons which, instead, heavily innervate the spinal cord. In conclusion, a small region of the VMS marginal layer contains glutamatergic neurons that innervate the main respiratory centers of the medulla oblongata and pons. These glutamatergic neurons are located in a chemosensitive region of the ML and their projections are consistent with a role in central chemoreception. The serotonergic neurons of the ML, though known to be activated by CO(2), probably do not contribute to central chemoreception, given that they innervate sympathetic efferents and project at best very lightly to the VRC.

Anatomical arrangement of hypercapnia-activated cells in the superficial ventral medulla of rats

The anatomical structure of central respiratory chemoreceptors in the superficial ventral medulla of rats was studied by using hypercapnia-induced c-fos labeling to identify cells directly stimulated by extracellular pH or PCO(2). The distribution of c-fos-positive cells was found to be predominantly perivascular to surface vessels. In the superficial ventral medullary midline, parapyramidal, and ventrolateral regions where c-fos-positive cells were concentrated, we found a common, characteristic, anatomical architecture. The medullary surface showed an indentation covered by a surface vessel, and the marginal glial layer was thickened. We classified c-fos-positive cells into two types. One (type I cell) was small, was located inside the marginal glial layer and close to the medullary surface, and surrounded fine vessels. The other (type II cell) was large and located dorsal to the marginal glial layer. c-fos Expression under synaptic blockade suggested that type I cells are intrinsically chemosensitive. The chemosensitivity of surface cells (possible type I cells) surrounding vessels was confirmed electrophysiologically in slice preparations. We suggest that this characteristic anatomical structure may be the central chemoreceptor.

Gap junctions in CO(2)-chemoreception and respiratory control

Recent evidence indicates that gap junctions play a more prominent role in normal functioning of the mammalian central nervous system (CNS) than was once believed. Accumulating evidence from both neonatal and adult rodents indicates that gap junctions participate in multiple aspects of respiratory control, including central CO(2) chemoreception, respiratory rhythmogenesis, and respiratory motoneuron output. This review provides an overview of gap junction neurobiology in the mammalian CNS and presents the anatomical and electrophysiological evidence for gap junctions in CO(2) chemoreception and respiratory control.

Localization of connexin26 and connexin32 in putative CO(2)-chemosensitive brainstem regions in rat

Recent studies have suggested that cell-to-cell coupling, which occurs via gap junctions, may play a role in CO(2) chemoreception. Here, we used immunoblot and immunohistochemical analyses to investigate the presence, distribution, and cellular localization of the gap junction proteins connexin26 (Cx26) and connexin32 (Cx32) in putative CO(2)-chemosensitive brainstem regions in both neonatal and adult rats. Immunoblot analyses revealed that both Cx subtypes were expressed in putative CO(2)-chemosensitive brainstem regions; however, regional differences in expression were observed. Immunohistochemical experiments confirmed Cx expression in each of the putative CO(2)-chemosensitive brainstem regions, and further demonstrated that Cx26 and Cx32 were found in neurons and Cx26 was also found in astrocytes in these regions. Thus, our findings suggest the potential for gap junctional communication in these regions in both neonatal and adult rats. We propose that the gap junction proteins Cx26 and Cx32, at least in part, form the neuroanatomical substrate for this gap junctional communication, which is hypothesized to play a role in central CO(2) chemoreception.

Cytoarchitecture of central chemoreceptors in the mammalian ventral medulla

We reviewed the previous reports on the fine anatomy of the mammalian ventral medulla with special attention to the cytoarchitecture of the superficial chemosensitive regions to summarize what is known, what is not yet known, and what should be studied in the future. We also reviewed studies on anatomical relationship between neurons and vessels, and morphological studies on dendrites of respiratory or chemosensitive neurons. When we compared the morphological reports on the ventral and dorsal putative chemosensitive regions, similarities were found as follows. Chemosensitive cells were often found not only near the ventral surface but near the dorsal surface of the brainstem. Dendritic projection towards the surface was a common characteristic in the ventral and dorsal chemosensitive neurons. Morphological abnormality in the brainstem of sudden infant death syndrome victims was also summarized. On the basis of the previous reports we discussed the perspective on the future study on central chemoreception. Among various unanswered questions in central chemosensitivity studies, physiological significance of surface cells and surface extending dendrites is the most important topic, and must be thoroughly investigated.

Decreased CSF pH at ventral brain stem induces widespread c-Fos immunoreactivity in rat brain neurons

Physiological evidence has indicated that central respiratory chemosensitivity may be ascribed to neurons located at the ventral medullary surface (VMS); however, in recent years, multiple sites have been proposed. Because c-Fos immunoreactivity is presumed to identify primary cells as well as second- and third-order cells that are activated by a particular stimulus, we hypothesized that activation of VMS cells using a known adequate respiratory stimulus, H(+), would induce production of c-Fos in cells that participate in the central pH-sensitive respiratory chemoreflex loop. In this study, stimulation of rostral and caudal VMS respiratory chemosensitive sites in chloralose-urethane-anesthetized rats with acidic (pH 7.2) mock cerebrospinal fluid induced c-Fos protein immunoreactivity in widespread brain sites, such as VMS, ventral pontine surface, retrotrapezoid, medial and lateral parabrachial, lateral reticular nuclei, cranial nerves VII and X nuclei, A(1) and C(1) areas, area postrema, locus coeruleus, and paragigantocellular nuclei. At the hypothalamus, the c-Fos reaction product was seen in the dorsomedial, lateral hypothalamic, supraoptic, and periventricular nuclei. These results suggest that 1) multiple c-Fos-positive brain stem and hypothalamic structures may represent part of a neuronal network responsive to cerebrospinal fluid pH changes at the VMS, and 2) VMS pH-sensitive neurons project to widespread regions in the brain stem and hypothalamus that include respiratory and cardiovascular control sites.

Carbonic anhydrase and CO2 chemoreception in the pulmonate snail Helix aspersa

We have studied the effects of carbonic anhydrase inhibition on the hypercapnic ventilatory response of the pulmonate snail, Helix aspersa, in an isolated brain-pneumostome preparation. We found that the cell permeant carbonic anhydrase inhibitor, acetazolamide (ACTZ), increased pneumostomal opening and ventilation during normocapnia (2-3% CO2) and decreased the rate of pneumostomal response to step changes in CO2 (4.5%), but did not change the steady-state ventilatory response to elevated CO2 (4.5%) compared to the inactive ACTZ analogue, N2-substituted 2-acetylamino-1,3,4-thiadiazole (Cl 13850). In contrast, the cell impermeant carbonic anhydrase inhibitor, quartenary ammonium sulfonilamide (QAS), had no effect on the pneumostomal response to CO2 compared to Cl 13850. Using Hansson's histochemical technique to stain for carbonic anhydrase activity, we identified a small number of neurons in the subesophageal ganglia that exhibited carbonic anhydrase activity. Some of these cells were in the region of CO2-sensitivity. In conclusion, carbonic anhydrase inhibition slows the ventilatory response to rapid changes in CO2, but does not affect the intrinsic ability of H. aspersa to respond to CO2. The ventilatory effects of carbonic anhydrase inhibition may be attributed to the intracellular actions of the carbonic anhydrase enzyme.

Candidate cell populations for respiratory chemosensitive fields in the human infant medulla

The histology and location of human respiratory chemosensitive fields are not known. In contrast, the physiology of respiratory chemosensory areas in the ventral medulla of cats has been studied extensively, and their anatomy has been partially described. Using basic principles of comparative cytoarchitecture and computer-aided reconstructions of serial-sectioned medullae, we describe the histology and three-dimensional distribution of putative respiratory chemosensors in the feline and human infant medulla. We found that ventrolateral neurons of the human nucleus conterminalis are homologous to neurons identified in the feline L chemosensitive field by Trouth and others, and that ventrolaterally situated neurons in the human arcuate nucleus correspond to neurons predominating in the feline S and M fields. In addition, there are foci of thickened marginal glia along the feline ventral medullary surface that colocalize with chemosensitive fields identified by physiologic studies reported by others; we identify similar foci in the infant medulla. Thickened marginal glia are intermixed with neuronal fibers, often adjacent to neurons of the feline chemosensitive fields and their human counterparts, suggesting that they constitute a chemosensory neuropil at the medullary surface. Computer-aided reconstructions provide insight into the three-dimensional topography of putative respiratory chemosensors and their relationships to other brainstem structures in ways not obvious in single or even multiple sections. This delineation of candidate human respiratory chemosensors is a first step toward their postmortem analysis in patients with central ventilatory control disorders where finding histological abnormalities in these sites would support their role in human ventilation.

Studies on the pyrexic effect of PGE1 injected into the region of the cochlear nuclei

In 30 of 33 unanesthetized rats, unilateral injection of PGE1 (100 ng in 1 microliter) into or near the cochlear nuclei (CN) produced a body temperature increase of at least 0.5 degrees C. Usually, the rise started within the first minute after injection. Bilateral destruction of the highly PGE1-sensitive anterior hypothalamic/preoptic region (AH/PO) eliminated the response. In approximately two-thirds of rats in which concentrated dye or 3H-PGE1 was injected into CN, there was evidence that a small portion of the injectate had rapid access to the AH/PO. In the remaining rats, the tracer material did not reach AH/PO tissue. In the rats involved in the tracer studies, there was no correlation between the magnitude of the response to PGE1 injected into CN and the extent of transport of tracer from CN to AH/PO. The results suggest that PGE1 injected into CN produced pyrexia through an action on tissue at or near the injection site and that neuronal pathways in AH/PO are necessary for the expression of the response.

Differential respiratory patterns induced by opioids applied to the ventral medullary and dorsal pontine surfaces of cats

The purpose of the present study was to make a functional dissection of the respiratory action of opioids, by their restricted application to the ventral surface of the medulla oblongata and to the rostro-dorsal surface of the pons in cats. The effects were compared to those induced after intracerebroventricular (i.c.v.) injection. Two mu-agonists, morphine and D-Ala2-Me-Phe4-Met(O)ol5-enkephalin (FK-33824), and the delta-agonist D-Ala2-D-Leu5-enkephalin (DADLE) were used. When applied to the ventral medullary surface, the opioids selectively depressed the generating mechanisms for tidal volume and the response to CO2, whereas the frequency was increased. The application to the rostral dorsal surface of the pons was followed by a selective depression of the respiratory frequency. By intracerebroventricular administration, the opioids depressed both the tidal volume and frequency generating mechanisms. The effects were always reversed by naloxone. The pontine structures were more sensitive to the action of the opioids than were the medullary centres. These findings suggest that the opioids can interact with different populations of respiratory neurones and that the respiratory output differs depending on the group of neurones selectively affected and the function they subserve in regulating respiratory activity.

Evidence for cholecystokinin-like immunoreactive neurons in the rat medulla oblongata which project to the spinal cord

The distribution of cholecystokinin-like immunoreactive (CCK-LI) neurons has been mapped in the rat medulla after local and intracerebroventricular colchicine injections. CCK-positive neurons were found in the nucleus raphe magnus, nucleus raphe pallidus, nucleus raphe obscurus, nucleus paragigantocellularis pars alpha, and a population of ventral medullary neurons. Combined retrograde tracing with the fluorescent dye True Blue and indirect immunofluorescence for visualizing CCK neurons suggested that there was a CCK-LI system originating in the medulla and projecting to the spinal cord. Additional double labelling experiments established that some of these CCK-LI containing neurons also contain 5-HT.

Ventral medullary relay neurones in the pathway from the defence areas of the cat and their effect on blood pressure

In cats anaesthetized with Althesin, the efferent descending pathway from the brain-stem defence areas has been traced through the medulla by identifying sites at which electrical stimulation evoked the characteristic pattern of the visceral alerting (defence) response. This response includes an increase in arterial blood pressure resulting from increased heart rate and cardiac output and vasoconstriction in renal and splanchnic beds, accompanied by active vasodilation in skeletal muscle. The efferent pathway runs as a narrow strip, about 3 mm from the mid line, ventral to the superior olive and the nucleus of the trapezoid body, extending caudally to the rostral portion of the inferior olive where it lies ventral to the facial nucleus. It was found to lie very close to the ventral medullary surface just rostral to and within the area at which bilateral topical application of glycine results in a profound fall in arterial blood pressure and cessation of respiration. On bilateral application of glycine to the sensitive area of the ventral medulla, the visceral alerting response evoked by stimulation in the defence areas of the amygdalo-hypothalamic complex, or the mid-brain central grey or tegmentum, was attenuated in parallel with the fall in arterial pressure, the vasoconstrictor responses being most strongly reduced. As soon as arterial blood pressure had fallen to its lowest level the visceral alerting response was virtually abolished. A small radio-frequency lesion made in the ventral medullary efferent pathway, in the rostral part of the 'glycine-sensitive area', had the same effect as that produced by unilateral application of glycine: it resulted in little respiratory or cardiovascular effect itself, but application of glycine to the contralateral area then produced the full effect otherwise seen only on bilateral application of glycine. It is suggested (1) that the effects of glycine result from blockade of a synaptic relay, close to the ventral surface of the medulla, in the efferent pathway from the defence areas to the preganglionic sympathetic neurones, and (2) that the neurones which receive an input from the alerting (defence) areas normally provide an essential, tonic excitatory drive to the sympathetic output and probably to respiration also. After sudden withdrawal of this drive, vasomotor tone and the normal level of arterial blood pressure are not maintained.

Substance P-immunoreactive and serotonin-containing neurones in the ventral brainstem of the cat

Substance P-like immunoreactive cell bodies have been described in the cat brainstem after local injections of colchicine. Substance P-containing neurons were found in the nucleus raphe pallidus, the nucleus raphe obscurus and at the caudal end of the nucleus raphe magnus. Two groups of substance P-positive (+) cells were also found in the ventrolateral medulla. Small substance P+ cells were present in a narrow band immediately ventral to the facial nucleus, and a second group of small and medium sized neurones was found in a strip ventrolateral to the rostral half of the inferior olive. Some of these cells lay in close proximity to the ventral surface of the brain. Double labelling studies indicated that within the raphe nuclei many substance P+ neurones also contained serotonin, while within ventrolateral areas of the medulla, the most rostral cell group stained only for substance P and the more caudal nucleus for both substance P and serotonin.