HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Free A corrigendum has been published Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (9) Citing Articles via Google Scholar Google Scholar Articles by Haxhiu, M. A. Articles by Ferguson, D. G. Search for Related Content PubMed PubMed Citation Articles by Haxhiu, M. A. Articles by Ferguson, D. G.

INVITED REVIEW

Brain stem excitatory and inhibitory signaling pathways regulating bronchoconstrictive responses

Specialized Neuroscience Research Program, Departments of ¹Physiology and Biophysics and ²Pharmacology, Howard University College of Medicine, Washington, District of Columbia; and Departments of ³Pediatric and ⁴Anatomy, Case Western Reserve University, Cleveland, Ohio

	ABSTRACT

TOP ABSTRACT GENERAL CONSIDERATIONS OF THE... VAGAL PREGANGLIONIC NEURONS... PATHWAYS AND NEUROTRANSMITTERS... GLUTAMATERGIC PATHWAYS... CENTRAL GABAERGIC CONTROL OF... PARACRINE-AUTOCRINE REGULATION... FUNCTIONAL RELEVANCE OF THE... ALTERATIONS IN CENTRAL CONTROL... CONCLUSIONS AND FUTURE STUDIES GRANTS REFERENCES

 
This review summarizes recent work on two basic processes of central nervous system (CNS) control of cholinergic outflow to the airways: 1) transmission of bronchoconstrictive signals from the airways to the airway-related vagal preganglionic neurons (AVPNs) and 2) regulation of AVPN responses to excitatory inputs by central GABAergic inhibitory pathways. In addition, the autocrine-paracrine modulation of AVPNs is briefly discussed. CNS influences on the tracheobronchopulmonary system are transmitted via AVPNs, whose discharge depends on the balance between excitatory and inhibitory impulses that they receive. Alterations in this equilibrium may lead to dramatic functional changes. Recent findings indicate that excitatory signals arising from bronchopulmonary afferents and/or the peripheral chemosensory system activate second-order neurons within the nucleus of the solitary tract (NTS), via a glutamate-AMPA signaling pathway. These neurons, using the same neurotransmitter-receptor unit, transmit information to the AVPNs, which in turn convey the central command to airway effector organs: smooth muscle, submucosal secretory glands, and the vasculature, through intramural ganglionic neurons. The strength and duration of reflex-induced bronchoconstriction is modulated by GABAergic-inhibitory inputs and autocrine-paracrine controlling mechanisms. Downregulation of GABAergic inhibitory influences may result in a shift from inhibitory to excitatory drive that may lead to increased excitability of AVPNs, heightened airway responsiveness, and sustained narrowing of the airways. Hence a better understanding of these normal and altered central neural circuits and mechanisms could potentially improve the design of therapeutic interventions and the treatment of airway obstructive diseases.

airways; airway reflex responses; autonomic control; nucleus of the solitary tract; glutamatergic pathways: glutamate; AMPA receptors; NMDA receptors; GABAergic microcircuitry; GAD; GABA_A receptors; synaptic transmission; volume transmission; vagal preganglionic neurons

Airway bronchoconstrictive responses initiated by inhaled agents or psychogenic factors suggest the existence of bidirectional central nervous system (CNS)-airway communication that serves to protect the respiratory system. Repeated exposure to environmental pollutants (like ozone, cigarette smoke, allergens), through parallel or convergent inflow (68), may modulate afferent airway sensory pathways (38, 108, 175, 200–202, 216), setting the stage for airway hyperresponsiveness. Under these conditions, the excitability of airway-related vagal preganglionic neurons (AVPNs) is increased and weak stimuli may trigger responses that last for hours, suggesting that the CNS is involved in causing airway constrictive changes.

A better understanding of the role that CNS pathways play in regulating airway functions in normal and diseased states undoubtedly will provide novel therapeutic approaches in the treatment of diseases whose clinical, biochemical, and pharmacological features indicate a pathophysiological link with the CNS. Therefore, this review is focused on the central determinants of airway control and adds to several recent excellent editorials and reviews on sensory neuroplasticity, autonomic regulation of airway functions, and associated pharmacological implications (42, 43, 55, 111, 113, 137, 176, 201–203, 213, 214).

In this article, we briefly discuss recent studies that clarify two of the basic processes in the central control of the vagal preganglionic neurons that provide cholinergic outflow to the airways: 1) glutamatergic signaling pathways involved in transmission and processing of bronchoconstrictive signals from airway sensory receptors, via the nucleus of the solitary tract (NTS) to AVPNs and 2) GABAergic inhibitory network that downregulates excitability of the AVPNs, consequently decreasing cholinergic outflow to the tracheobronchial system. An imbalance between excitatory and inhibitory inputs to AVPNs could be of considerable importance.

	GENERAL CONSIDERATIONS OF THE NEURAL CONTROL OF THE AIRWAYS

TOP ABSTRACT GENERAL CONSIDERATIONS OF THE... VAGAL PREGANGLIONIC NEURONS... PATHWAYS AND NEUROTRANSMITTERS... GLUTAMATERGIC PATHWAYS... CENTRAL GABAERGIC CONTROL OF... PARACRINE-AUTOCRINE REGULATION... FUNCTIONAL RELEVANCE OF THE... ALTERATIONS IN CENTRAL CONTROL... CONCLUSIONS AND FUTURE STUDIES GRANTS REFERENCES

 
CNS control of airway functions (Fig. 1, level 4) involves integrated networks along the neural axis that funnel information to tracheobronchopulmonary effector units via the AVPNs in the medulla oblongata. The AVPNs are the final common pathway from the brain to the airways and transmit signals to the intrinsic tracheobronchial ganglia that are part of the network for automatic feedback control (level 3). Each ganglion, located in close proximity to effector systems, possesses a relatively large number of neurons (11, 40, 54, 131, 155) that can be considered as an expanded parasympathetic, preganglionic efferent motor system (level 2). However, intrinsic airway neurons contain neurochemicals other than ACh, including vasoactive intestinal peptide (VIP) and neuronal nitric oxide synthase [nNOS, an enzyme involved in generation of nitric oxide (NO); 54, 219] that are considered to mediate nonadrenergic neuronal airway smooth muscle relaxation (21, 56, 78).

View larger version (14K): [in this window] [in a new window]   Fig. 1. General scheme illustrating the organization of autonomic parasympathetic control of airway functions. Central nervous system (CNS) cell groups (level 4) regulate the activity of airway-related vagal preganglionic neurons (AVPNs; level 3). Axons of the AVPNs, as the final common pathway out of the medulla oblongata, synapse on intrinsic ganglionic neurons within airway walls (level 2). These ganglia give rise to postganglionic fibers that control the function of specific effector targets (i.e., airway smooth muscle, mucous glands, and blood vessels). Sensory feedback for these systems occurs via sensory fibers originating from sensory ganglia (nodose and jugular ganglionic neurons). These fibers innervate sensory receptors and transmit information from the airways to the CNS. They modulate the activity of AVPNs through central multisynaptic pathways and they may affect function of effector organs via 2 ill-defined local networks that include axon reflex responses (level 1) and sensory innervation of intrinsic airway ganglia (level 2).

However, it is not clear whether individual AVPNs provide parallel innervation to airway smooth muscle, submucosal glands, and local blood vessels. A relative simultaneity of airway effector cell responses to stimulation of bronchoconstrictive vagal afferent fibers, such as reflex elevation of smooth muscle tone, increase in submucosal gland secretion, and changes in blood flow, is consistent with the assumption that the same cell bodies of AVPNs that cause smooth muscle constriction also elicit activation of smooth muscle glands or changes in blood flow supplying regional airway structures. Alternatively, it is also possible that groups of functionally selective AVPNs may exert a highly coordinated control over multiple airway functions by central mechanisms that synchronize their output to individual effectors. This assumption is analogous to evidence obtained relevant to coordination of multiple cardiac functions by vagal preganglionic neurons, where functionally distinct, but closely integrated, vagal preganglionic neurons in the nucleus ambiguus (NA) mediate the coordinated control of cardiac rate, atrioventricular conduction, and left ventricular contractility (65, 133, 134). Recent preliminary data in ferrets suggest that presumptive gap junctions between identified AVPNs (Blinder K, Karibi-Ikiriko A, Massari VJ, Haxhiu MA, unpublished data) may serve to synchronize their electrical activity in the rostral nucleus ambiguus (rNA). Alterations in central mechanisms that mediate such synchronization could cause regional differences in airway reflex responses, expressed as inhomogeneity in ventilation. For example, in an asthma attack or induced bronchoconstriction, some branches of the airway may totally close while others remain normal (181). Future studies should address the question of the central mechanisms and origin of synchronization and asynchrony of bronchoconstrictive reflex responses.

An extensive network of vagal afferent fibers of sensory ganglia innervates the bronchopulmonary sensory receptors that are specialized for detecting changes in chemical, mechanical, or thermal stimuli. The bipolar airway vagal afferent neurons are located in the nodose and jugular ganglia and participate in reflex events. Furthermore, the afferent nerve endings (i.e., C fibers) are also believed to be responsible for mediating local axon reflexes (level 1) and neurogenic inflammation via release of neuropeptides such as substance P (141), which can be modulated by alterations in activity of neutral endopeptidases (156). In addition, endogenous substance P facilitates synaptic transmission in airway parasympathetic ganglia (34, 154), similar to the effect observed with cyclooxygenase activation and the release of prostaglandins in antigen-induced lung injury (112). The central fibers of these sensory neurons ascend in the vagus nerve and enter the brain stem through the solitary tract (26, 88, 89, 114, 123), making their first synapse with the NTS second-order neurons that are required for full expression of the pulmonary C fiber reflex (26) and bronchoconstrictive airway reflex responses (94).

In the NTS, the afferent vagal terminals responsible for transmitting information from the airway bronchoconstrictive receptors form a distinct wiring diagram with specific second-order neurons that project to AVPNs (74, 88, 89, 167). However, a variety of inputs from afferent receptors is transmitted to the NTS, including those from cough receptors that are activated by stimuli that may also elicit reflex bronchoconstriction, submucosal gland secretion, and increased blood flow. Cough receptors are preferentially located within the larynx, trachea, and large intrapulmonary airways (25, 35, 213). As is the case with vagal efferents, it is not clear whether the same or different vagal afferent neurons transmit different components of airway defensive reflex responses.

More recently, however, it is reported that a subpopulation of myelinated, polymodal A-fibers that arise from the nodose ganglia respond to punctate mechanical stimulation and acid but are unresponsive to capsaicin, bradykinin, histamine, smooth muscle contraction, longitudinal or transverse stretching of the airways, or distension. Unlike these afferent neurons, considered as the putative cough receptors, the majority of capsaicin-sensitive afferents (both A- and C-fibers) innervating the rostral trachea and larynx have their cell bodies in the jugular ganglia and project to the airways via the superior laryngeal nerves (35). These findings suggest the possibility of presynaptic or postsynaptic interactions of afferent nerves originating from cough and non-cough receptor neurons at the NTS level. (25, 35). For example, in conscious guinea pigs, C fiber-dependent cough may require coactivation of bronchoconstrictive airway afferent nerves for the full expression of the cough reflex.

Bronchoconstrictive inputs can be modulated by increasing or decreasing the afferent signals from slowly adapting receptors to NTS second-order neurons, the first site of synaptic contact of primary bronchopulmonary slowly adapting receptors (49). Activation of these receptors causes a reflex airway smooth muscle relaxation (212). Similarly, peripheral chemoreceptors and baroreceptors acting centrally can readily affect cholinergic outflow to the airways. Although stimulation of the carotid bodies reflexly elicits bronchoconstriction (157) and submucosal gland secretion (50) and facilitates bronchoconstrictive responses (205), the activation of baroreceptors leads to opposite changes (157). In this article, signaling mechanisms involved in these and other possible reflex interactions that can centrally influence bronchoconstrictive responses are not considered.

Recently, using conventional and transneuronal labeling techniques and ultrastructural, molecular, and physiological approaches, we identified the central excitatory (82–84, 91, 93) as well as inhibitory pathways and neurotransmitters (82, 92, 94, 96) that regulate the excitability and firing rate of AVPNs. These processes occur via both wiring (synaptic) and volume (nonsynaptic) transmission. We hypothesize that downregulation of central inhibitory influences upon AVPNs result in a shift from inhibitory to excitatory transmission, leading to a hyperexcitable state of the AVPNs and, consequently, cholinergic hyperresponsiveness that can predispose to and worsen bronchial asthma (13, 57).

	VAGAL PREGANGLIONIC NEURONS INNERVATING THE AIRWAYS

TOP ABSTRACT GENERAL CONSIDERATIONS OF THE... VAGAL PREGANGLIONIC NEURONS... PATHWAYS AND NEUROTRANSMITTERS... GLUTAMATERGIC PATHWAYS... CENTRAL GABAERGIC CONTROL OF... PARACRINE-AUTOCRINE REGULATION... FUNCTIONAL RELEVANCE OF THE... ALTERATIONS IN CENTRAL CONTROL... CONCLUSIONS AND FUTURE STUDIES GRANTS REFERENCES

 
Vagal preganglionic neurons that generate cholinergic outflow to the airways can be viewed as the central integrators of multiple excitatory and inhibitory inputs that connect the brain with the bronchopulmonary effector system. The critical circuits that regulate these processes include both central glutamatergic and GABAergic pathways controlling cholinergic outflow to the airways (Fig. 2).

View larger version (23K): [in this window] [in a new window]   Fig. 2. Central glutamatergic excitatory and GABAergic inhibitory pathways regulating cholinergic outflow to the airways. In this oversimplified schematic illustration are presented 2 main neural pathways that determine the activity of AVPNs. Excitatory signaling pathway uses glutamate (Glu) as a neurotransmitter in conveying bronchoconstrictive signals from airway sensory receptors to the nucleus of the solitary tract (NTS) and from the NTS to the AVPNs. Inhibitory projections employ GABA as a signaling molecule to downregulate excitability of the AVPNs and cholinergic outflow to the tracheobronchial system transmitted through ACh release. These 2 signaling pathways are emphasized because they comprise the central theme of this review.

Studies using retrograde tracer techniques showed that in mammals (Fig. 3), the cholinergic preganglionic motor neurons innervating the airways arise from the rNA and from the rostral portion of the dorsal motor nucleus of the vagus (DMV). Furthermore, the findings imply that the majority of AVPNs are involved in the innervation of multiple airway segments, thereby assuring the symmetry and simultaneity of bronchomotor responses (20, 74, 80, 88, 99, 114, 139, 167, 168, 204). In addition, some preganglionic neurons may provide direct innervation to the airway tissues and lung parenchyma, without the interposition of intrinsic neurons (74, 168). This somewhat atypical organization of parasympathetic circuits has been demonstrated in other effector systems, for example, ciliary muscle, which receives dual parasympathetic innervation directly from preganglionic neurons and via ganglionic cells (210), indicating the possible functional importance of direct projections that bypass intrinsic cholinergic ganglia, an assumption that was not supported by recent physiological experiments (119).

View larger version (103K): [in this window] [in a new window]   Fig. 3. Location of AVPNs innervating airways of the ferret. A: schematic presentation of a coronal section of the medulla oblongata showing that in ferrets, as in other mammals, the preganglionic motor neurons innervating the airways arise from the rostral nucleus ambiguus (rNA) and from the rostral portion of the dorsal motor nucleus of the vagus (DMV). MeV, medical vestibular nucleus; LV, lateral vestibular nucleus. B: confocal photomicrographs showing retrogradely labeled AVPNs within the rNA 5 days after cholera toxin -subunit (CT-b) injections into the lung. C: same section immunostained for choline acetyl transferase (ChAT). D: higher power of the M image of CT-b and ChAT traits of the region marked by a quadrangle in A. Arrow indicates a representative neuron that is immunolabeled with CT-b and ChAT. *ChAT-positive neuron that does not contain CT-b. Scale bar = 50 µm for B and C and 35 µm for D. Data are from Ref. 119.

Ultrastructural Characteristics of AVPNs

It has been suggested that the viscerotropic representation, ultrastructure, and synaptology of the different divisions of the NA innervating the alimentary system may be associated with specific physiological functions (22, 104, 183). To define ultrastructural characteristics of AVPNs, recently, brain stems from ferrets, in which cholera toxin -subunit (CT-b) conjugated to horseradish peroxidase had been used as a retrograde cell body tracer, were examined. Electron microscopy was employed to determine ultrastructural characteristics of these tracheal AVPNs (135). Retrogradely labeled AVPNs in the rNA were readily detectable in the electron microscope. The cell bodies of labeled AVPNs were observed to be 32 ± 1 x 23.0 ± 1.3 µm (means ± SE) in size, with abundant cytoplasm and intracellular organelles. The cell bodies had round uninvaginated nuclei that occasionally contained a prominent nucleolus and displayed both somatic and dendritic spines (Fig. 4). Somatosomatic appositions or somatodendritic appositions without intervening glial processes and dendritic bundling of the tracheal AVPNs were not observed. The axons of these neurons were seldom labeled (Fig. 5).

View larger version (159K): [in this window] [in a new window]   Fig. 4. Electron microscope image of AVPN in the rNA. This neuron (N) is retrogradely labeled from the trachea and is identified ultrastructurally by the presence of tetramethylbenzidine tungstate (TMB) crystalline reaction product (large arrows). Note the round nucleus and a prominent nucleolus. Area inside the box is enlarged in the inset and shows an example of a substance P-immunoreactive nerve terminal (T), indicated by amorphous diaminobenzidine reaction product, forming an asymmetric axosomatic synapse with the perikaryon. Calibration bars = 2 µm in the larger panel and 500 nm in the inset. Modified from Ref. 135.

View larger version (168K): [in this window] [in a new window]   Fig. 5. Somatic, dendritic, and axonic profiles of retrogradely labeled AVPNs. A: this perikaryon contains abundant rough endoplasmic reticulum (RER) and somatic spines forms axosomatic synaptic contacts (small arrows) with SP-ir terminal. B: a dendrite illustrating dendritic spines forming axodendritic synapses (small arrows) with SP-ir terminals. C: axons of AVPNs are myelinated (M). Unlabeled myelinated axons (m) are indicated for comparison. Calibration bar = 500 nm for all panels. Modified from Ref. 135.

In summary, at the ultrastructural level, AVPNs innervating the extrathoracic trachea are clearly distinguished from pharyngeal, laryngeal, or esophageal motoneurons in other subdivisions of the NA. These data are consistent with the hypothesis that differences in the ultrastructure and synaptology of the different divisions of the NA may be associated with specific physiological functions (104, 133, 134, 183).

Chemical Profile of AVPNs

Studies using a double-labeling method that combines immunolocalization of the retrograde tracer CT-b and immunohistochemistry for choline acetyl transferase (ChAT), the enzyme catalyzing the biosynthesis of ACh, indicate that in ferrets, AVPNs innervating the trachea and the intrapulmonary airways are cholinergic in nature (Fig. 3) and use ACh as a neurotransmitter to convey signals to the airway motor systems (119). These neurochemical findings are in agreement with the results of physiological studies showing that chemical stimulation of the vagal preganglionic neurons (80, 82–84) or efferent fibers of the vagus nerve originating from AVPNs produces pronounced contraction of the airway smooth muscle and lung parenchyma that is solely mediated via cholinergic parasympathetic nerves and mechanisms (120, 132, 143, 166). Furthermore, reflexly (42, 43, 171, 218) and centrally (86, 87) induced increases in airway secretion are mediated mainly via cholinergic mechanisms.

Virtually all vagal preganglionic neurons innervating the trachea and intrapulmonary airways coexpress VIP, but not NOS, indicating that ACh and VIP are coexisting messenger molecules in AVPNs (119). In addition, recent studies demonstrated that brain-derived neurotrophic factor (BDNF) mRNA and its protein transcript are expressed by a subpopulation of AVPNs, suggesting that active BDNF synthesis occurs in the cells within the rNA that innervate the airways (Zaidi SI, Jafri A, Doggett T, Haxhiu MA, unpublished data). The possible role of the ACh, VIP, and BDNF-tyrosine kinase (Trk) B signaling pathways in the autocrine/paracrine regulation of AVPNs responses to afferent inputs is briefly mentioned at the end of the article.

CNS Innervation of AVPNs

Activity of AVPNs depends on afferent inputs, although they possess an ability to express synchronous electrical oscillations, unveiled by stimulation of NMDA receptors (83) or blockade of GABA_A receptors (150). Recently, it has been shown using conventional and transneuronal labeling techniques that the innervation of vagal preganglionic neurons regulating parasympathetic outflow to the airways arises from cell groups located in the brain stem and from several higher brain regions. By comparing the CNS inputs to the vagal preganglionic neurons that innervate intrapulmonary airways with those controlling extrathoracic trachea, common patterns of innervation are seen (74, 88, 167). AVPNs receive inputs from cell groups located in the ventral aspect of the medulla oblongata; NTS; pons; ventrolateral part of periaqueductal gray (PAG) cell group; dorsal, lateral, and paraventricular hypothalamus; and the central nucleus of amygdala (74). Hence, the parasympathetic preganglionic neurons innervating the airways are controlled by networks of brain stem and suprapontine cell groups that lie in regions known to be involved in the central control of autonomic functions (129). Recent studies suggest that some respiratory and airway responses could be produced by single neurons capable of affecting multiple neural pathways, rather than by a complex set of heterogeneous cells regulating individual systems (90, 98).

The AVPNs provide the final common pathway for vagal control of the airways. Although we emphasized the importance of medullary circuits that control the reflex output of these AVPNs, functions of the airways are also powerfully regulated by information from higher CNS centers that are relayed caudally to the AVPNs. These forebrain circuits are critical for expression of behavioral state control and the emotional dimensions of airway disorders. For example, the prefrontal cortex innervates the interconnected amygdaloid complex, the central nucleus of which project subsequently to multiple targets regulating autonomic functions, including the dorsomedial and ventrolateral PAG, the NTS, and the NA (74, 172).

The projections from the amygdala to the PAG are of particular importance because the PAG neurons coordinate functions of multiple visceral organs involved in responses to stress. Active coping responses (fight or flight reactions) are evoked by activation of either the dorsolateral or lateral columns of the PAG; whereas passive coping strategies (e.g., quiescence, immobility, decreased responsiveness to the environment) are usually elicited by activation of the ventrolateral part of the PAG (118). The phenotypic traits of some patients experiencing severe or near-fatal asthma, such as psychosocial barriers, blunted respiratory, and other sensory responses that cannot be explained by different medication regimens (16), resemble the expression of passive coping response seen after stimulation of the ventrolateral PAG region in animals (118).

In humans, little is known about CNS innervation of AVPNs. More recently, in conscious subjects, the CNS sites subserving the experience of loaded breathing and air hunger were studied using the positron emission tomography (PET) and functional nuclear magnetic resonance imaging (fMRI; 71, 77, 165). For example, compared with the unloaded control condition, a moderate inspiratory resistive load is associated with an increased fMRI signal intensity in discrete brain regions, including areas of the dorsal pons that correspond to the locus ceruleus (LC) and parabrachial nucleus (PBN). In animals, stimulation of the LC noradrenergic cell group (97) and activation of PBN (153) induces centrally mediated airway smooth muscle relaxation, a compensatory response that tends to decrease the resistive load. In addition, activity associated with perceived intensity of respiratory discomfort induced by loaded breathing was found within the right posterior cingulate cortex and amygdaloid complex, demonstrating laterality to the challenge (165). Of particular interest are recent findings in obstructive sleep apnea (OSA) showing lower signals in medullary and midbrain areas (77), sites that innervate AVPNs (74, 88). A decrease in activity of these sites may lead to withdrawal of inhibitory inputs to AVPNs and facilitation of cholinergic outflow to the airways (92, 95, 98), suggesting that an association of OSA and bronchoconstriction during sleep could be expected more often than it is recognized.

	PATHWAYS AND NEUROTRANSMITTERS INVOLVED IN CONVEYING BRONCHOCONSTRICTIVE INFORMATION FROM THE AIRWAYS TO THE NTS

TOP ABSTRACT GENERAL CONSIDERATIONS OF THE... VAGAL PREGANGLIONIC NEURONS... PATHWAYS AND NEUROTRANSMITTERS... GLUTAMATERGIC PATHWAYS... CENTRAL GABAERGIC CONTROL OF... PARACRINE-AUTOCRINE REGULATION... FUNCTIONAL RELEVANCE OF THE... ALTERATIONS IN CENTRAL CONTROL... CONCLUSIONS AND FUTURE STUDIES GRANTS REFERENCES

 
Bronchopulmonary sensory receptors, innervated by small diameter (A) myelinated vagal afferents known as rapidly adapting receptors (RARs), and nonmyelinated C fibers (C fiber receptors), are characterized by their well-defined sensitivity to chemical stimuli in the airways, most of which affect both RARs and C fiber receptors (42). The net airway motor response results from the interaction of sensory signals within the NTS, including those from cough and slowly adapting pulmonary receptors (25, 35, 212, 213).

Recently, the NTS second-order neurons activated by afferent inputs from the airways were identified using the expression of the nuclear protein (c-Fos), encoded by the protooncogene c-fos. The c-fos gene, a member of the immediate early genes, and its product, c-Fos, are expressed in restricted numbers of CNS neurons in response to strong stimuli applied to the airway receptors. Thus c-Fos is considered to be an inducible and high-resolution marker of neurons activated by sensory inputs and extracellular stimuli (63).

The advantage of using c-fos expression as a functionally oriented anatomical approach is that the activity of a large number of cells can readily be identified under conditions that do not affect reflex responses such as anesthesia. Using this approach, we found that after stimulation of pulmonary and bronchial C fibers by capsaicin or stimulation of rapidly adapting receptors and bronchial C fiber receptors by histamine aerosol (41), a significant number of NTS neurons within the commissural, medial, and the ventrolateral NTS subregions expressed c-Fos protein (63, 94), indicating that they had been activated by stimulation of pulmonary C fiber and rapidly adapting receptors (26, 94). In ferrets, the labeling pattern correlated well with studies that defined retrograde tracings of NTS subregions in which afferent fibers from the airways terminate (89). The same subregions contained cell body labeling after pseudorabies virus (PRV) injections into the tracheal wall or into the most distal airways (74, 88, 167), suggesting that the NTS activated neurons are those that transmit information from the airways to the AVPNs. Multiple neurotransmitters, including ACh and/or glutamate, are expressed by the primary sensory neurons that may participate in sensory transmission (17, 99, 160).

Cholinergic Transmission

It was predicted that afferent inputs could be conveyed to the NTS by ACh acting on nicotinic ACh receptors. This postulate was based on findings that cholinergic neurons are present in nodose ganglia (160, 163), whereas ¹²⁵I-labeled -bungarotoxin binding sites, indicating the presence of nicotinic ACh receptors (nAChRs) that mediate nicotinic cholinergic transmission, are observed within the NTS (9). Furthermore, nodose ganglionectomy decreases ChAT activity and causes a reduction in nAChR binding sites in NTS (99). The dissociated rat NTS cells were observed to express nicotinic but not muscarinic acetylcholine receptors (199). In addition, a subset of second-order NTS neurons expresses ChAT in perikarya and dendrites (6).

Experiments demonstrated that activation of airway sensory receptors (RARs and C fiber receptors) induced c-Fos expression in a subset of NTS neurons that also expressed the ₃-subtype nicotinic ACh receptor (nAChR). Furthermore, activation of nAChR within the commissural NTS subnucleus by nicotine increased cholinergic outflow to the airways. These effects were diminished by prior administration of hexamethonium (nAChR antagonist) within the commissural NTS cell group. However, hexamethonium had no significant effects on airway reflex constrictions induced by lung deflation. These findings indicated that endogenously released ACh within the NTS and activation of nAChR are not required for transmission of bronchoconstrictive stimuli from the airways to the NTS (63), suggesting involvement of other excitatory molecules such as glutamate.

Glutamatergic Transmission

L-Glutamate, a naturally occurring excitatory amino acid, is a main sensory neurotransmitter (47) mediating communication between neurons within neuronal networks coordinating motor outputs (for review and references see 72, 209). It is present in vagal afferents in the NTS (179, 180, 195, 206) and is required for transmission of fast signals from sensory nerve endings to second-order neurons (5, 94, 211), including baroreceptor inputs (125, 126, 197).

Recently, we studied the role of glutamate in conveying bronchoconstrictive inputs from the airways to the NTS and from the NTS to the AVPNs (91, 93, 94). Glutamate release within the NTS after stimulation of airway sensory receptors was also examined. These studies indicated that stimulating airway sensory receptors increased the spontaneous efflux of glutamate in the commissural NTS subnucleus (94), the site where airway afferent fibers terminate in ferrets (89), and that the release was correlated with airway smooth muscle contraction (94). Typical HPLC chromatograms of microdialysates were collected from the commissural NTS subnucleus in a control state, during repeated excitation of bronchopulmonary sensory receptors, and in the poststimulation period. After airway sensory stimulation, glutamate concentration increased significantly from 42 + 11 to 109 + 17 pg/µl (average of 3 trials for each ferret), and pressure in a bypassed tracheal segment (Ptseg) increased from 7.4 + 1.4 to 26.1 + 4.6 cmH₂O (P < 0.01). In the poststimulation recovery period, L-glutamate and Ptseg returned to control levels. Removal of the vagal efferent and afferent innervations of the tracheobronchial tree eliminated the reflex increase in glutamate release induced by lung deflation (Fig. 6).

View larger version (19K): [in this window] [in a new window]   Fig. 6. Reflex-induced glutamate release within the NTS. Top: typical HPLC chromatograms of microdialysates collected from the commissural subnucleus of the NTS in a control state (baseline), during repeated excitation of bronchopulmonary sensory receptors (stimulation), and in a poststimulation period (recovery). Middle: average concentrations (means ± SE) of L-glutamate, expressed in pg/µl, under 3 different experimental conditions. Filled bars: animals with intact innervation of the airways and lungs (vagus intact; n = 5). Open bars: ferrets after afferent and efferent denervation of the airways and lungs (vagotomized and superior laryngeal nerves cut; n = 3). Bottom: tracheal smooth muscle tone measured as pressure in a bypassed tracheal segment (Ptseg) before, during, and after stimulation in ferrets with intact innervation of the airways. *P < 0.05. Modified from Ref. 94.

Over the last two decades, extensive research has revealed a host of glutamate receptor subtypes belonging to the ionotropic (iGluR) and metabotropic glutamate receptor (mGluR) classes, which subserve excitatory synaptic transmission and neurotransmitter release.

iGluRs.   iGluRs are ligand-gated ion channels consisting of three known functional receptor subtypes: the AMPA, kainate, and NMDA receptors, which respond to glutamate and glutamate analogs by the opening of a cation channel (4, 10, 27, 44, 138). Quisqualate, AMPA, glutamate, and kainate are potent agonists of the AMPA receptor subunits (GluR1-GluR4) when applied alone or in combination. The quinoxalinediones, such as 6,7-dinitroquinoxaline-2, dione (DNQX) and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), are the most potent antagonists of the AMPA receptors. This class of antagonists also blocks the kainate receptors but does not block NMDA receptors. Hence, it can be used to differentiate AMPA receptor-mediated effects from those of the NMDA signaling pathway.

The kainate receptors, composed of five subunits, GluR5-GluR7 and KA1 and KA2, contribute to excitatory postsynaptic currents in many regions of the central and peripheral nervous systems, including the hippocampus, cortex, spinal cord, and retina. In some cases, postsynaptic kainate receptors are codistributed with AMPA and NMDA receptors, but there are also synapses where transmission is mediated exclusively by postsynaptic kainate receptors. Recent analysis of knockout mice lacking one or more of the subunits that contribute to kainite receptors, as well as studies with subunit-selective agonists and antagonists, have revealed the important roles that kainate receptors play in short- and long-term synaptic plasticity (27, 106). Because AMPA receptors are more abundant than kainate receptors and exhibit faster signaling kinetics than NMDA receptors (70, 169), the AMPA receptors are uniquely suited in mediating cardiopulmonary reflexes (5, 69, 211). Furthermore, synaptic activation of AMPA receptors may elicit not only postsynaptic excitation but also presynaptic inhibition of GABAergic transmission. By suppressing the inhibitory inputs, the activation of AMPA receptors could facilitate bronchoconstrictive inputs to the NTS second-order neurons and from these neurons to the AVPNs. This possibility is supported by findings showing that stimulation of AMPA receptors induces presynaptic inhibition at cerebellar GABAergic synapses (182) but increases ACh release in the hippocampus of young and aged rats (177).

Experiments to determine whether specific AMPA receptors are expressed by the activated NTS neurons were performed in ferrets by using a double-staining technique and confocal microcopy (94). Employing GluR2 mouse monoclonal antibody, it was observed that this AMPA receptor subtype is present on the subpopulation of the NTS neurons activated by sensory inputs from the airways (Fig. 7). In addition, the role of the glutamate-AMPA receptor signaling pathway in transmission of bronchoconstrictive signals from the airways to the NTS was studied using physiological experiments. Blockade of AMPA receptors by bilateral microinjections of CNQX into the commissural subnucleus elicited a significant decrease in reflex tracheal smooth muscle response, markedly slowed the rate of rise of tracheal tone, decreased the peak response, and enhanced the decline of the response after cessation of lung deflation (Fig. 7, bottom). Therefore, bronchoconstrictive inputs from the airways to NTS neurons are transmitted primarily by a glutamate-AMPA receptor signaling pathway.

View larger version (22K): [in this window] [in a new window]   Fig. 7. Coexpression of c-Fos and GluR2 AMPA receptor subunit in second-order nTS neurons after stimulation of excitatory bronchopulmonary receptors. Top: confocal microscope image of the commissural region of the NTS. c-Fos staining (red) was observed only within nuclei. These were visualized as round or oval structures, depending on their orientation within the plane of the section. GluR2 receptor staining (green) was observed to be punctate and uniformly distributed on the membrane of the perikarya of neurons with unstained nuclei (GluR2) and on neurons with c-Fos-stained nuclei (c-Fos and GluR2). Middle: numbers of neurons within the commissural region of the NTS that express c-Fos, GluR2, and c-Fos-positive neurons that coexpress GluR2. Bottom: superimposed tracings from a paralyzed ferret ventilated with oxygen. In a control period, lung deflation induced an increase in tracheal tone, expressed by elevation of Ptseg that was not affected by bilateral microinjection of vehicle into the commissural subnucleus of the NTS (A). Administration of 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) into the same region in a dose-dependent fashion decreased the response (B: 1 nmol; C: 4 nmol/site CNQX). Modified from Ref. 94.

Therefore, it seems that the involvement of AMPA receptors in airway reflex responses is determined by their kinetics, which is much faster than those of the NMDA or kainate receptors. Hence the AMPA receptors are more appropriate for transmission of fast signals, whereas NMDA and kainate receptors may contribute to primary visceral afferent transmission in the NTS, mediating tonic influences.

G protein-coupled metabotropic glutamate receptors.   G protein-coupled metabotropic glutamate receptors are expressed in autonomic cell groups of the medulla oblongata (79), including NTS second-order neurons (136, 158, 162). Coactivation of iGluRs and G protein-coupled metabotropic glutamate receptors (mGluRs) may affect glutamatergic transmission through regulation of neurotransmitter release. On the basis of amino acid sequence homology, pharmacology, and signal transduction mechanisms, these subtypes have been classified into three groups (103, 136, 170, 196). Group I mGluRs consist of mGluR1 and mGluR5 that are found mainly on postsynaptic terminals and are positively coupled to phospholipase C. Group II mGluRs (mGluR2 and mGluR3) are observed on both pre- and postsynaptic terminals, whereas group III mGluRs (mGluR4, mGluR6, mGluR7, and mGluR8) are predominantly located in or near presynaptic zones. In general, group I mGluRs increase neuronal excitability, whereas group II and group III modulate frequency dependence of synaptic vesicle exocytosis, inhibiting signal transmission (33, 36, 128, 158, 161, 164, 184).

It is accepted that released glutamate acts locally on postsynaptic receptors and is cleared from the synaptic cleft within a few milliseconds by diffusion and by specific reuptake mechanisms. This rapid clearance restricts the spread of neurotransmitter and, combined with the low affinities of many ionotropic receptors, ensures that synaptic transmission occurs in a point-to-point fashion. However, when glutamate release is enhanced at synapses, the concentration increases and glutamate escapes the synaptic cleft and, via volume transmission, may activate presynaptic inhibitory mGluRs. Hence, at higher frequency stimulation, when the released amount of glutamate exceeds the uptake quanta, presynaptic mGluRs of groups II and III become activated, leading to a rapid inhibition of neurotransmitter release (58, 184) and, consequently, presynaptic depression of synaptic transmission (33) that might have neuroprotective effects (61).

In the rat, activation of mGluRs affects frequency dependence response and depresses vagal and aortic baroreceptor signal transmission in the NTS (66, 127, 128, 144) via activation of one or more phosphoprotein phosphatases such as phosphoprotein phosphatase 2 and/or calcineurin (67). Furthermore, recent findings, determining presynaptic vs. postsynaptic effects, indicate that glutamate released at the first central baroreceptor synapses cannot only regulate its own signaling but can further shape signal transmission by suppressing GABA release via activation of heterosynaptic group II mGluRs (39). The results highlight the complexity of mGluRs functions in modulation of reflex responses within the NTS and suggest the importance of differentiation of presynaptic vs. postsynaptic effects of drugs tested.

In ferrets, recent neuroanatomical and physiological studies showed that subtype 1 of the group I mGluRs is rarely expressed by activated NTS neurons after stimulation of bronchopulmonary receptors. Blockade of the group I mGluRs using a specific antagonist had no significant inhibitory effects on airway reflex responses, when administered into the NTS. Furthermore, blockade of groups II/III mGluRs within the NTS had no significant enhancing effect on reflex airway smooth muscle contraction (Ferguson DG, Haxhiu MA, unpublished data). However, these results can be considered only as preliminary, because microinjection of the drug into the rNA, as in any other circumscribed region, is not very selective. Namely, administered antagonist of group II or III mGluRs may have effects on presynaptic as well as on postsynaptic sites. Future studies, separating presynaptic from postsynaptic influences, are needed to define whether under pathological conditions, such as animal model of bronchial asthma and experimentally induced non-allergic airway hyperreactivity, upregulation of group I and/or downregulation of group II/III of mGluRs may facilitate reflex airway constriction and hyperresponsiveness.

In summary, neuroanatomical and physiological studies further highlight the key role of the glutamate-AMPA receptor signaling pathways in transmitting bronchoconstrictive inputs from the airways to the NTS, where signals are processed, modulated, and relayed to the vagal parasympathetic neurons innervating the airways.

	GLUTAMATERGIC PATHWAYS TRANSMITTING BRONCHOCONSTRICTIVE INPUTS FROM THE NTS TO THE AVPNS

TOP ABSTRACT GENERAL CONSIDERATIONS OF THE... VAGAL PREGANGLIONIC NEURONS... PATHWAYS AND NEUROTRANSMITTERS... GLUTAMATERGIC PATHWAYS... CENTRAL GABAERGIC CONTROL OF... PARACRINE-AUTOCRINE REGULATION... FUNCTIONAL RELEVANCE OF THE... ALTERATIONS IN CENTRAL CONTROL... CONCLUSIONS AND FUTURE STUDIES GRANTS REFERENCES

 
Until recently, signaling mechanisms involved in transmitting bronchoconstrictive signals from the NTS to the AVPNs were not established. However, it was assumed that the glutamate-AMPA receptor signaling pathway may play an important role. This assumption was based on findings that a majority of NTS neurons contain glutamate (121) and information from these neurons is transmitted to AVPNs mainly via hard-wired synaptic pathways (74, 88, 167).

Recent work has shown that the excitatory glutamatergic neurons express different types of vesicular transporter proteins, named VGLUT1, VGLUT2, and VGLUT3 (102, 115). VGLUT2, a protein that localizes to synaptic vesicles, functions as an important vesicular glutamate transporter (19, 64, 102, 194). Our unpublished studies in ferrets, using a double-labeling technique, demonstrate heavy glutamatergic innervation of the AVPNs (Acquah S, Kc P, Massari VJ, Haxhiu MA, unpublished results), forming the morphological basis for the powerful influence of excitatory amino acids on the activity of AVPNs (Fig. 8). Effects of glutamate on AVPNs are mediated through different subtypes of glutamate receptors.

View larger version (34K): [in this window] [in a new window]   Fig. 8. Glutamatergic axon terminals innervating identified AVPNs. A: confocal microscopic image of AVPNs immunolabeled for CT-b with a green fluorescein-conjugated secondary antibody (green) and glutamate transporter 2 (VGLUT2) with Texas Red-conjugated secondary antibody (red). B: higher magnification image of neurons identified in A (quadrangle), illustrating VGLUT2-like immunoreactive puncta targeting AVPNs and their proximal dendrites. Scale bar = 20 µm for A and 10 µm for B.

Neurochemical studies indicate the presence of AMPA, NMDA, and kainate receptors in vagal preganglionic neurons (45). Neurophysiological investigations have shown that iGluRs play an important role in transmission of excitatory inputs to the AVPNs regulating airway smooth muscle tone. For example, NMDA receptors are expressed by AVPNs and, when activated with a specific NMDA receptor agonist (NMDA), cause airway constriction that is blocked by selective NMDA antagonists (83). The possibility that NMDA receptors may also play a role in reflex airway constriction has been studied by inducing blockade of NMDA receptors with 2-aminophosphonovalerate. Topical application or microinjection of this NMDA antagonist within the rNA, the area in which the AVPNs are located, only slightly affected reflex changes in tracheal tone. However, administration of selective antagonists for the AMPA/kainite subtype of glutamate receptor to the same site caused a dose-dependent decrease in reflex response of tracheal tone induced by 1) lung deflation, 2) stimulation of laryngeal cold receptors, and 3) activation of peripheral or central chemoreceptors (91). These reflexes are known to cause centrally mediated increases in cholinergic tone (42, 51, 52). Inhibitory effects of AMPA receptor blockade were potentiated by prior administration of an NMDA receptor antagonist. Thus an increase in central cholinergic outflow to the airways by a variety of reflex excitatory inputs is mediated mainly via glutamate-AMPA receptors that in turn activate the NMDA receptor signaling pathway.

Physiological studies indicate that increases in airway blood flow and submucosal gland secretion are integral components of pulmonary defensive reflex responses (42, 43, 171, 218). Neural regulation of the bronchial vasculature differs from that of the general systemic circulation in that vasodilator reflexes play a major part in determining blood flow (43). These reflexes originate in the upper or lower airways, in carotid chemoreceptors, or in cardiac chemosensitive nerves. Those arising from the lower airways are the most potent and may increase bronchial blood flow several-fold and cause swelling of the airway mucosa. In addition, neuropeptides released from C fiber terminals provide a local mechanism for vasodilation independent of central reflex control. This so-called axon reflex plays a major role in bronchial vasodilation in rodents but makes only a small contribution in larger animals. In dogs, a centrally mediated vagal reflex vasodilator pathway appears more important. Cooling the cervical vagus nerves eliminates afferent and efferent vagal pathways but preserves local mechanisms (including axon reflex pathways). This treatment diminishes more than two-thirds of the reflex vasodilatation; thus a large portion of the vasodilatation in dogs, whether caused by activation of sensory C fiber receptors and/or RARs, is due to centrally mediated neural reflexes that include afferent and efferent pathways in the vagus nerves (171). In addition, stimulation of bronchial and pulmonary C fibers or RARs evokes a reflex increase in secretion by tracheal submucosal glands. The responses are abolished on interruption of the afferent and efferent transmissions by cutting the vagus nerves or cooling them to 0°C (42, 218). Furthermore, focal cooling of the rostral ventrolateral medulla between 20°and 15°C significantly decreases the secretion rates produced by capsaicin-induced stimulation of pulmonary C fiber receptors and by mechanical stimulation of the carina and larynx (87).

The involvement of glutamate and glutamate receptors in the transmission of excitatory inputs from the airway sensory receptors to the NTS and from this site to the AVPNs was studied (91). Stimulation of airway sensory fibers by lung deflation induced reflex increases of tracheal blood flow (Fig. 9) and submucosal gland secretion (Fig. 10). These responses were diminished by prior administration of an AMPA/kainate receptor antagonist, CNQX, into the fourth ventricle, or microinjection of AMPA/kainate receptor blockers into the external formation of the rNA, where the AVPNs are located. These findings indicate that the transmission of excitatory inputs from the NTS to the AVPNs is mediated mainly via the release of glutamate and activation of the AMPA/kainate subtype of glutamate receptors. Therefore, it is likely that the airway sensory stimulation-evoked airway smooth muscle contraction, vasodilation, and hypersecretion are mediated mainly via cholinergic mechanisms, using the same, glutamate-AMPA signaling pathway, that in turn activate NMDA receptors, as summarized in Fig. 11.

View larger version (47K): [in this window] [in a new window]   Fig. 9. Glutamatergic control of reflex-induced changes in airway blood flow. Top: schematic presentation of the extrathoracic tracheal segment that can be used for simultaneous recording of submucosal blood flow, gland secretion, and airway smooth muscle tone. Tracheal segment preparation used to measure trachealis smooth muscle tension was developed by Brown et al. (29), whereas video camera method of measuring mucous secretion was developed by Davis et al. (50). Bottom: tracings from decerebrate, paralyzed, and mechanically ventilated dog after administration of vehicle into the IV ventricle. Vehicle had no effects on resting tracheal circulation. Lung deflation induced an increase in submucosal blood flow (t; left) and slightly elevated systemic blood pressure (BP; mmHg). Administration of CNQX, indicated by the arrow, tended to decrease the basal blood flow and abolished the response to lung deflation (right). (l/s), air flow. Modified from Ref. 93.

View larger version (151K): [in this window] [in a new window]   Fig. 10. Glutamatergic control of reflex changes in submucosal secretion. An example of the effect of CNQX administration into the IV ventricle on secretory response of tracheal submucosal glands to lung deflation. Prior administration of CNQX abolished the response of submucosal glands to lung deflation. A and C: tracheal epithelium 1 min after spraying with tantalum (an inert metal dust that prevents spread of secreted fluid), baselines. B and D: tracheal epithelium 1 min after lung deflation after vehicle (B) or after CNQX administration into the IV ventricle (D). Similar effects were observed when CNQX was bilaterally microinjected into rNA. Modified from Ref. 93.

View larger version (33K): [in this window] [in a new window]   Fig. 11. Summary diagram of glutamatergic signal transmission pathways mediating airway constrictive reflex responses. Glutamate-AMPA receptor signaling pathways play a major role in transmission of excitatory sensory information from the airways (sensor, afferent nerve, sensory neuron) to the second-order NTS neurons and consequently from these neurons (terminal of presynaptic neuron) to the AVPNs. Activation of AMPA receptors via increase in Na+ influx elicits excitatory postsynaptic potential (EPSP) that in turn activates NMDA receptors leading to Ca2+ entry. Signals from the stimulated AVPNs are transmitted to the airways using ACh as a neurotransmitter.

In ferrets, mGluRs belonging to group I are expressed by AVPNs (Ferguson DG, Haxhiu MA, unpublished data). Their role in glutamatergic transmission of bronchoconstrictive inputs was investigated using a selective antagonist. Blockade of these receptors within the rNA region had no significant effect on airway basal smooth muscle tone or on reflex bronchoconstriction. Similarly, inhibition of group II/III mGluRs had no demonstrable facilitatory effect on reflex increases in unit activity of the AVPNs (Kc P, Haxhiu MA, unpublished data), suggesting that the physiological relevance of the excitatory and/or inhibitory mGluRs is, at best, limited. However, these findings do not exclude the possibility that these receptors may operate at higher frequency afferent inputs, enhancing glutamatergic responses (group I mGluRs), or suppressing excessive signal transmission at AVPN synapses (groups II and III mGluRs), consequently modulating the transmission of bronchoconstrictive inputs from the NTS to the AVPNs within the rNA.

	CENTRAL GABAERGIC CONTROL OF CHOLINERGIC OUTFLOW TO THE AIRWAYS

TOP ABSTRACT GENERAL CONSIDERATIONS OF THE... VAGAL PREGANGLIONIC NEURONS... PATHWAYS AND NEUROTRANSMITTERS... GLUTAMATERGIC PATHWAYS... CENTRAL GABAERGIC CONTROL OF... PARACRINE-AUTOCRINE REGULATION... FUNCTIONAL RELEVANCE OF THE... ALTERATIONS IN CENTRAL CONTROL... CONCLUSIONS AND FUTURE STUDIES GRANTS REFERENCES

 
Processing of central afferent excitatory signals by the AVPNs and conveying integrated output to the airways are highly dependent on synaptic GABAergic inhibitory inputs (152). In general, GABA release and activation of postsynaptic GABA_A receptors play an important role in controlling neuronal excitability in adult mammalians. This occurs by increasing membrane conductance to chloride ions, thereby causing a potent inhibition of neuronal activity that in turn diminishes the depolarizing effects of excitatory synaptic signals (208). The action of released GABA within or in close vicinity to the synaptic cleft is terminated by its rapid uptake into surrounding neurons and astrocytes by selective GABA transporters (53, 81, 186).

GABAergic Microcircuitry Regulating AVPNs

GABA, the main inhibitory neurotransmitter in mammalian brain, is synthesized mainly via decarboxylation of glutamic acid by two specialized enzymes, the glutamic acid decarboxylases (GADs), designated GAD65 and GAD67 according to their molecular masses (65 and 67 kDa). These enzymes differ in their affinities for their cofactor pyridoxal-5'-phosphate and are encoded by two different genes located on separate chromosomes (32, 60). Both enzymes are often colocalized in the same GABAergic neurons but sometimes differ in their subcellular distribution. GAD67, although also detected in axon terminals, is thought to be preferentially localized to cell bodies, whereas GAD65 tends to be associated with synaptic vesicles in nerve terminals (62, 100). However, more detailed studies revealed a more complex situation, showing that both GAD65 and GAD67 provide important reserve pools of GABA for the regulation of inhibitory neurotransmission, and a deficiency of either isoform of GAD can have significant physiological sequelae. For example, GAD67-deficient animals are born with a cleft palate and die within the first day of life, apparently from respiratory failure (8). By contrast, GAD65-deficient mice appear normal at birth, but the pyridoxal-5'-phosphate-inducible apoenzyme reservoir is significantly decreased, leading to an increased susceptibility to epileptigenic stimuli (7). Hence both isoforms of GAD appear to be physiologically important in the dynamic regulation of neural network excitability.

Recent ultrastructural studies have shown that retrogradely labeled AVPNs receive a significant GABAergic innervation (Fig. 12). Out of a pooled total of 3,161 synaptic contacts with retrogradely labeled somatic and dendritic profiles, 20.2% are GAD-immunoreactive (IR), forming significantly more axosomatic symmetric synaptic specializations than axodendritic synapses (P < 0.02). These ultrastructural findings indicate that central GABAergic modulation of cholinergic outflow is mediated in large part via classically defined inhibitory axosomatic synapses and to a lesser degree through axodendritic synaptic transmission. A dense population of GABAergic synaptic contacts on AVPNs provides a morphological basis for the potent physiological effects of GABA on the excitability of AVPNs (150).

View larger version (128K): [in this window] [in a new window]   Fig. 12. GABAergic innervation of AVPNs. A: a GABA immunoreactive nerve terminal (T) containing a large dense-core vesicle as well as multiple small pleomorphic vesicles forms an axodendritic synapse (small arrowhead) with the proximal dendrite (D) of a retrogradely labeled (large arrowheads) tracheal vagal preganglionic neuron. Unlabeled nerve terminal (t) forms an asymmetric axodendritic synapse (small arrowheads) with a dendrite (D). B: in the same field, note that a GABA immunoreactive nerve terminal (T) forms a symmetric axodendritic synapse (small arrowheads) with an unlabeled dendrite (d). Calibration bar equals 200 nm. Modified from Ref. 150.

GABA Levels Within the rNA

Recently, baseline levels of GABA within the rNA region have been measured by microdialysis and HPLC. After equilibration of the dialysis probe, the average measured concentration of GABA was 29.3 ± 7.1 pg/20 µl during a steady-state condition. After stimulation of the ventrolateral region of the vl PAG, GABA concentration significantly increased (61.4 ± 22.1 pg/20 µl; Fig. 13, top). Although the increase in GABA levels after chemical stimulation is of an exocytotic origin, the basal levels of extracellular GABA could be derived in part from nonexocytotic source, as in other brain regions (53, 198).

View larger version (51K): [in this window] [in a new window]   Fig. 13. GABA release within rNA and GABAA receptors expressed by AVPNs. Top: GABA levels (means ± SE; n = 6) obtained from microdialysates collected from the rNA region in a control state (Baseline) and during repeated excitation of ventrolateral periaqueductal gray (vl PAG) neurons (PAG stim). *P < 0.05. Bottom: example of a confocal microscope image of GABAA 2-subunit receptor expression by the AVPNs innervating the extrathoracic trachea. A: CT-b-labeled AVPNs (red). B: in the same region, punctate GABAA receptor-specific staining (green) is observed that is likely associated with the cell membranes, because it surrounds empty spaces (N) in which neuron cell bodies lie. There were also profiles of GABAA receptor-specific staining in between the neuron cell bodies (*). These presumably identify receptors located in presynaptic terminals that contact processes of the motor neurons. C: in double-labeling studies it is clearly observed that the CT-b-labeled neurons contain GABAA 2-subunit (green) expressed as punctate green staining on the membrane of the perikaryon (arrows) as well as on dendrites. Bar = 50 µm (A), 30 µm (B), 15 µm (C). Modified from Ref. 96.

GABA Receptor-Mediated Signaling

GABA receptors are categorized in two distinct types: the ionotropic GABA_A and GABA_C and the metabotropic GABA_B receptors.

GABA_A receptors.   Fast GABA-mediated synaptic inhibitory neurotransmission requires that GABA_A receptors are expressed and assembled at appropriate postsynaptic sites facing GABA-releasing nerve terminals (124, 130, 149). These chloride ion channel-associated, ligand-gated receptors are heteropentamers that can be assembled from a number of subunit class