Ablation of vagal preganglionic neurons innervating the extra-thoracic trachea affects ventilatory responses to hypercapnia and hypoxia.

This study tested the hypothesis that during hypercapnia or hypoxia, airway-related vagal preganglionic neurons (AVPNs) of the nucleus ambiguus (NA) release acetylcholine (ACh), which in a paracrine fashion, activates ACh receptors expressed by inspiratory rhythm generating cells. AVPNs in the NA were ablated by injecting a saporin- (SA) cholera toxin b subunit (CTb-SA) conjugate into the extra-thoracic trachea (n=6). Control animals were injected with free CTb (n=6). In CTb treated rats, baseline ventilation and ventilatory responses to hypercapnia (5 and 12% CO(2) in O(2)) or hypoxia (8% O(2) in N(2)) were similar (p>0.05) prior to and 5 days after injection. CTb-SA injected rats maintained rhythmic breathing patterns 5 days post injection, however, tachypneic responses to hypercapnia or hypoxia were significantly reduced. The number of choline acetyltransferase (ChAT) immunoreactive cells in the NA was much lower (p<0.05) in CTb-SA rats as compared to animals receiving CTb only. These results suggest that AVPNs participate in the respiratory frequency response to hypercapnia or hypoxia.

Orexin stimulates breathing via medullary and spinal pathways.

A central neuronal network that regulates respiration may include hypothalamic neurons that produce orexin, a peptide that influences sleep and arousal. In these experiments, we investigated 1) projections of orexin-containing neurons to the pre-Botzinger region of the rostral ventrolateral medulla that regulates rhythmic breathing and to phrenic motoneurons that innervate the diaphragm; 2) the presence of orexin A receptors in the pre-Botzinger region and in phrenic motoneurons; and 3) physiological effects of orexin administered into the pre-Botzinger region and phrenic nuclei at the C3-C4 levels. We found orexin-containing fibers within the pre-Botzinger complex. However, only 0.5% of orexin-containing neurons projected to the pre-Botzinger region, whereas 2.9% of orexin-containing neurons innervated the phrenic nucleus. Neurons of the pre-Botzinger region and phrenic nucleus stained for orexin receptors, and activation of orexin receptors by microperfusion of orexin in either site produced a dose-dependent, significant (P <0.05) increase in diaphragm electromyographic activity. These data indicate that orexin regulates respiratory activity and may have a role in the pathophysiology of sleep-related respiratory disorders.

Sleep networks and the anatomic and physiologic connections with respiratory control.

A central neuronal network regulates airway functions from the nares to the bronchioles and is an integral component of a regulatory system for brain control of breathing and airway patency during wakefulness and sleep. This network, components of which include sleep generating sites and monoaminergic neurons in particular, is characterized by reciprocal interconnections, parallel organization, and state-dependent activity patterns, which can be influenced by both genes and environment. Sleep generating neurons are interconnected with the monoaminergic containing cells to the extent that sleep-related changes in upper and lower airway patency could be due to inhibitory influences of sleep-activated neurons on serotonergic and noradrenergic producing cells. Neurochemical studies and physiologic experiments show that serotonergic and noradrenergic producing cells can make parallel pathways, directly innervating the hypoglossal motor cells regulating upper airway dilating muscles, and vagal preganglionic neurons providing cholinergic outflow to the airways. Activation of serotonergic and noradrenergic cell groups preferentially increases activity of the genioglossus muscle, but diminishes cholinergic outflow to the airways. Hence, inhibition of monoaminergic neurons during sleep may lead to a decrease in upper airway dilating forces and an elevation of cholinergic outflow to the airways. Qualitatively different responses of hypoglossal and airway-related vagal preganglionic neurons (AVPNs) occur in response to endogenously released serotonin or norepinephrine and could be related to its simultaneous action on different serotonin or norepinephrine receptor subtypes. Dysfunction of monaminergic cell groups during sleep may predispose to upper airway occlusion as well as bronchoconstriction. Pharmacological corrections of alterations of these transmitter specific converging systems might be an avenue for treatment of sleep related airway disorders such as sleep apnea and worsening of asthma.

Paraventricular vasopressin-containing neurons project to brain stem and spinal cord respiratory-related sites.

We studied in the rat projections of vasopressin-containing neurons of the paraventricular nucleus (PVN) to phrenic nuclei and to the pre-Botzinger complex (pre-BotC). In addition, we determined vasopressin receptor expression within the pre-BotC and the physiological effects of vasopressin on respiratory drive and arterial blood pressure when injected into the pre-BotC. Retrograde tracing with cholera toxin B subunit (CT-b) showed that a subpopulation of vasopressin-containing PVN neurons project to phrenic nuclei and the pre-BotC. The latter region, identified by expression of neurokinin-1 receptors, contained a subpopulation of neurons that were immunoreactive for the vasopressin type 1 receptor (V(1)R). Microinjection of vasopressin in the pre-BotC (0.2 nmol/200 nl) significantly increased diaphragm electromyographic activity and frequency discharge (P<0.05). In addition, vasopressin increased blood pressure and heart rate (P<0.05). These data indicate that PVN vasopressin-containing neurons innervate respiratory-related regions of the medulla oblongata and spinal cord and when vasopressin is released at these sites, it may increase respiratory drive via activation of the distinct V(1)R.

CO(2)-induced c-Fos expression in brainstem preprotachykinin mRNA containing neurons.

Tachykinin peptides are found in brainstem regions involved in central chemoreception and they may play a modulatory role in ventilatory response to hypercapnia. We determined whether tachykinin peptide containing neurons are activated by CO(2) by combining in situ hybridization and immunohistochemistry (IHH). Experiments were performed in 21-day-old rats exposed to 12% CO(2) for 1 h. c-Fos expression was identified by IHH on free floating sections (40 microm) that were mounted and then hybridized with anti-sense 35S labeled ribonucleotide probe of the rat preprotachykinin A (PPT-A) gene. Sections were analyzed for expression of the PPT-A gene, c-Fos protein and colocalization of PPT-A gene with c-Fos protein. Within the chemosensory region of the nucleus tractus solitarius (nTS), 19% of c-Fos positive cells expressed PPT-A mRNA after hypercapnic loading. In medullary raphe nuclei, 64% of c-Fos positive cells expressed the PPT-A gene after exposure to CO(2), while 21% of c-Fos labeled neurons in parapyramidal nuclei also expressed PPT-A mRNA. These results indicate that a subpopulation of CO(2) activated neurons within the nTS and in the parapyramidal and midline regions of the ventral aspect of the medulla oblongata express the PPT-A gene, suggesting that these are substance P- or neurokinin A-containing neurons. Furthermore, these peptides may play a role in modulation of respiratory and cardiovascular responses to changes in CO(2)/H(+) content of the extracellular fluid.

CO(2)-induced c-Fos expression in hypothalamic vasopressin containing neurons.

Following exposure of anesthetized and unanesthetized rats to hypercapnic stress, arginine vasopressin (AVP)-containing neurons of the hypothalamic paraventricular (PVN) and supraoptic (SON) nuclei were examined for expression of the c-fos gene encoded protein (c-Fos). In addition, we determined whether AVP-containing PVN neurons activated by hypercapnia project to phrenic nuclei. In adult control rats, only scant c-Fos-like immunoreactive neurons were observed within the hypothalamic nuclei. A marked increase in c-Fos positive cells was induced after 2 h of breathing a gas mixture with elevated CO(2) (5% CO(2), 21% O(2) and 74% N(2), or 1 h following breathing of 12% CO(2,) 21% O(2,) and 67% N(2)). Colocalization studies of AVP and c-Fos protein revealed that in the PVN, 75% of AVP-containing cells expressed c-Fos immunoreactivity. c-Fos and AVP were coexpressed in 60% of SON neurons in anesthetized rats. In addition, retrograde labeling studies with cholera toxin b subunit (CTb) revealed that a subpopulation of PVN cells (15%) that project to phrenic nuclei are activated by hypercapnia, as indicated by c-Fos expression. These results indicate that (i) PVN and SON AVP-containing neurons are part of the neuronal networks that react to hypercapnic exposure; and (ii) a subset of CO(2) reactive PVN cells innervate phrenic nuclei.