Neural regulation of airway smooth muscle tone

Neural regulation of airway smooth muscle tone

Respiration Physiology 125 (2001) 113– 127 www.elsevier.com/locate/resphysiol Neural regulation of airway smooth muscle tone Brendan J. Canning a,*, ...

325KB Sizes 0 Downloads 8 Views

Respiration Physiology 125 (2001) 113– 127 www.elsevier.com/locate/resphysiol

Neural regulation of airway smooth muscle tone Brendan J. Canning a,*, Axel Fischer b a

The Johns Hopkins Asthma and Allergy Center, 5501 Hopkins Bay6iew Circle, Baltimore, MD 21224, USA b Allergy Research Group, Charite´, Biomedical Research Center, Berlin, Germany Accepted 2 October 2000

Abstract Airway smooth muscle is innervated by sympathetic and parasympathetic nerves. When activated, airway nerves can markedly constrict bronchi either in vivo or in vitro, or can completely dilate a precontracted airway. The nervous system therefore plays a primary role in regulating airway caliber and its dysfunction is likely to contribute to the pathogenesis of airways diseases. The predominant contractile innervation of airway smooth muscle is parasympathetic and cholinergic in nature, while the primary relaxant innervation of the airways is comprised of noncholinergic (nitric oxide synthase- and vasoactive intestinal peptide-containing) parasympathetic nerves. These parasympathetic nerves are anatomically and physiologically distinct from one another and differentially regulated by reflexes. Sympathetic-adrenergic nerves play little if any role in directly regulating smooth muscle tone in the human airways. Activation of airway afferent nerves (rapidly adapting receptors, C-fibers) can evoke increases in airway smooth muscle parasympathetic nerve activity, or decreases in parasympathetic nerve activity (through activation of slowly adapting receptors). Extrapulmonary afferents can also modulate nerve mediated regulation of airway smooth muscle tone. In guinea pigs and rats, peripheral activation of tachykinin-containing airway afferent nerves evokes bronchospasm via release of substance P and neurokinin A. This effect of airway afferent nerve activation appears to be unique to guinea pigs and rats. The actions and interactions between the components of airway innervation are discussed. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Airways, smooth muscle; Bronchi, spasm, dilatation; Mammals, humans; Mediators, NO, tachykinins, VIP; Muscle, smooth, sympathetic, parasympathetic innervation; Nerves, sympathetic, parasympathetic

1. Introduction Pioneering studies carried out prior to and just after the beginning of the 20th century established a powerful hypothetical model of autonomic * Corresponding author. Tel.: +1-410-5502156, fax: +1410-5502130. E-mail address: [email protected] (B.J. Canning).

control of airway smooth muscle: Parasympathetic, cholinergic pathways with preganglionic fibers carried by the vagus nerves mediate bronchospasm via activation of cholinergic ganglion neurons intrinsic to the airways, while adrenergic nerve fibers emanating from thoracic and superior cervical sympathetic ganglia mediate bronchodilatation (see Dixon and Brodie, 1903; Dixon and Ransom, 1912). Based on this model it was fur-

0034-5687/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 0 3 4 - 5 6 8 7 ( 0 0 ) 0 0 2 0 8 - 5

114

B.J. Canning, A. Fischer / Respiration Physiology 125 (2001) 113–127

ther hypothesized that airway caliber was determined by the balance of these opposing influences, and that pathological conditions such as asthma produced an imbalance in autonomic control, leading to bronchospasm and airways hyperresponsiveness (reviewed in Macklin, 1929; Widdicombe, 1963; Barnes, 1986). The utility of the model of airway smooth muscle innervation described above should not be underestimated. In some species, this hypothetical arrangement of the nervous system has withstood nearly a century of research without refutation (Widdicombe, 1963; Cabezas et al., 1971). Likewise, in virtually all other species, this scheme has had reasonably accurate predictive value, an assertion that is supported by the fact that it produced two major classes of therapeutics for pulmonary disease (Goldie et al., 1990; Gross, 1997). Nevertheless, with the discovery of nonadrenergic, noncholinergic (NANC) nerves innervating airway smooth muscle, it became necessary to reassess the validity of some aspects of this model of airway neural control (Barnes, 1986; Diamond and Altiere, 1988; Canning and Undem, 1994). This review summarizes our current understanding of airway smooth muscle innervation.

nerves and the pulmonary branches of the vagus nerves. The right and left vagus nerves and right and left sympathetic ganglia contribute equally to the innervation of the trachea. Parasympathetic ganglia in the right and left lobes of the lung are innervated bilaterally, but are preferentially innervated by the ipsilateral vagi. Contralateral projections of postganglionic airway sympathetic nerves have not been found (see section by Myers in this issue for a detailed description of the intrinsic innervation of the airways).

2.2. Neurochemistry of ner6e fibers inner6ating airway smooth muscle Although total nerve fiber density in the smooth muscle of the airways remains relatively constant from the extrapulmonary airways to the bronchioles (Ward et al., 1995), the neurochemistry of these nerve fibers changes dramatically (Fig. 1 and Fig. 2). Tyrosine hydroxylase, for example, a histochemical marker for bronchodilating adrenergic nerves, is found preferentially in nerve fibers of the extrapulmonary airways, being virtually absent from nerve fibers in the small bronchi and bronchioles (Kummer et al., 1992; Ind, 1994). Similarly, vasoactive intestinal peptide (VIP) and NO synthase (NOS), histochemical

2. Morphology of airway smooth muscle nerves

2.1. Extrinsic sources of airway efferent ner6es The extrinsic efferent nerves regulating airway smooth muscle tone are derived from two primary sources: Preganglionic parasympathetic fibers carried by the vagus nerves, and postganglionic sympathetic fibers emanating from thoracic and cervical sympathetic ganglia (Kummer et al., 1992; Canning and Undem, 1994). Preganglionic parasympathetic fibers reach the airways via three routes: The tracheal ganglia receive preganglionic input from the recurrent and superior laryngeal nerves, while the bronchial ganglia receive preganglionic input from the pulmonary branches of the vagus nerves. Postganglionic sympathetic nerves project to the airways through diffuse, loosely grouped pathways and reach end organs in the lungs and airways via the recurrent laryngeal

Fig. 1. Specific labeling of nerve fibers innervating human airway smooth muscle with an antibody to protein gene product 9.5 (PGP 9.5). More than 15 neurotransmitters with distinct pharmacological properties have been localized to nerve fibers innervating airway smooth muscle (see Table 1; magnification: 400 ×; calibration bar: 20 mm).

B.J. Canning, A. Fischer / Respiration Physiology 125 (2001) 113–127

115

Fig. 2. Nerve fiber density in airway smooth muscle. Panel A: Total (using PGP 9.5-IR) and NO synthase-containing (NOS-IR) nerve fiber density in human airway smooth muscle. Total nerve fiber density remains relatively constant from the central to peripheral airways, whereas NOS-IR nerve fiber density decreases markedly in the peripheral airways (reprinted with permission from Ward et al., 1995). Similar results have been reported elsewhere (Fischer and Hoffmann, 1996). Panel B: Cholinergic nerve fiber density in guinea pig airway smooth muscle as measured by choline acetyltransferase (ChAT). Cholinergic nerve fiber density remains relatively constant throughout the airway tree (modified with permission from Canning and Fischer, 1997).

markers for noncholinergic parasympathetic nerves (as well as some postganglionic sympathetic nerves innervating the airways), are readily localized to nerve fibers innervating smooth muscle of the large bronchi and trachea, yet are absent in or minimally expressed by nerves innervating airway smooth muscle in the peripheral airways (Kummer et al., 1992; Ward et al., 1995; Fischer and Hoffmann, 1996). By contrast, nerve fibers with either choline acetyltransferase, a specific marker for cholinergic nerves, or substance P, can be found at high densities in smooth muscle throughout the airways (Kummer et al., 1992; Canning and Fischer, 1997). Functional studies confirm the heterogeneity of nerve fibers innervating central and peripheral airways (Cabezas et al., 1971; Diamond and O’Donnell, 1980; Diamond and Altiere, 1988; Takahashi et al., 1995; Ward et al., 1995; Canning and Fischer, 1997). Nerve fibers containing different neurotransmitters are intermingled in the smooth muscle layers throughout the airways. The close association between nerve fiber subtypes provides an anatomical basis for nerve – nerve interactions in the airways (Table 1).

3. Functional effects of nerves on airway smooth muscle With the notable exception of rodents, all mammalian species studied to date have both contractile and relaxant innervation of airway smooth muscle. Contractile innervation of mammalian airways is almost exclusively cholinergic, whereas relaxant innervation is both adrenergic and nonadrenergic in nature (Fig. 3).

3.1. Contractile inner6ation of airway smooth muscle 3.1.1. Parasympathetic-cholinergic ner6es Postganglionic, parasympathetic-cholinergic nerves innervate the airways (Barnes, 1986; Canning and Undem, 1994). When activated, these nerves are capable of obliterating the lumen of small bronchi and bronchioles, and markedly increasing airways resistance in larger, cartilaginous airways (Dixon and Brodie, 1903; Cabezas et al., 1971; Canning and Undem, 1994). In vitro, activation of these nerves can evoke near maximal contractions of isolated airway smooth muscle preparations (Canning and Undem, 1993). It is clear, therefore, that airway choliner-

116

B.J. Canning, A. Fischer / Respiration Physiology 125 (2001) 113–127

gic nerves play a pre-eminent role in regulating airway caliber. Agents such as atropine and ipratropium bromide prevent bronchospasm evoked by exogenously administered acetylcholine, or broncho constriction mediated by activation of postganglionic parasympathetic cholinergic nerves (Gross, 1997). The capacity of these antimuscarinics to prevent acetylcholine-mediated bronchospasm is due to their ability to antagonize the actions of the neurotransmitter at smooth muscle M3 receptors (Barnes, 1999). Interestingly, M3 receptors represent only a small percentage (10 –30%) of the muscarinic receptors expressed by airway smooth muscle cells, the vast majority of which are of the M2 subtype coupled to regulatory Gi proteins. When activated, M2 receptors can partially antagonize the ability of relaxant agonists to evoke bronchodilatation through activation of adenylate cyclase. M2 receptor antagonists are, however, without effect on bronchospasm evoked by exogenously

administered acetylcholine, but potentiate vagally-mediated bronchospasm by blocking prejunctional inhibitory M2 autoreceptors (Fryer and Maclagan, 1984; Barnes, 1999). Consistent with the classic descriptions by Hillarp, few if any specializations are apparent at the autonomic neuromuscular junctions of the airways (Canning and Undem, 1994). Accordingly, it is not known what percentage of muscarinic receptors on the airway smooth muscle are ‘innervated’, and further, what relative percentage of the innervated receptors are of the M2 or M3 subtypes. Based on nerve fiber density studies (Canning and Fischer, 1997), only a small percentage of the muscarinic receptors expressed by the airway smooth muscle may be innervated by cholinergic nerves. It also seems likely that the innervated receptors are exposed to high local concentrations of acetylcholine released from the postganglionic parasympathetic nerves. Consistent with this hypothesis, atropine is less effective at antagonizing bronchospasm evoked by vagus

Table 1 Neurotransmitters in nerve fibers innervating airway smooth musclea Neurotransmitters Bronchoconstrictors Acetylcholine (ACh) Tachykininsc

Origin

Cotransmitters

Parasympatheticb

Blocking drugs

Ipratropium bromide, atropine

Afferent nerve endings, parasympathetice

CGRPd, dynorphin

CP 99994, SR 48968

Bronchodilators Noradrenaline (NA) VIPf

Sympathetic

neuropeptide Y (NPY)

Propranolol

Parasympathetic, sympathetic

Selective blockers unavailable

Nitric Oxide (NO)g

Parasympathetich

NO, NPY, galanin, Met-enkaphalin VIP, NPYi

a

L-NNA, ODQ, carboxy-PTIO hydroquinone

See text for references. Most common cotransmitters listed. Exceptions and subtle species differences have been noted. Choline acetyltransferase also found in perikarya of subsets of sympathetic, vagal afferent neurons innervating guinea pig airways. c Substance P, neurokinin A. d Calcitonin gene-related peptide. e Substance P found in airway ganglia of ferrets, cats and rats, not humans or guinea pigs. f Vasoactive intestinal peptide (VIP) and related peptides (PACAP, PHI, PHM). Also found in some airway afferent nerve endings. g Synthesized from arginine by NO synthase. h NO synthase found in some afferent, sympathetic nerve endings innervating guinea pig airways. i CGRP, NPY, dynorphin, galanin and Met-enkephalin have little or no direct effects on airway smooth muscle. b

B.J. Canning, A. Fischer / Respiration Physiology 125 (2001) 113–127

117

contractile responses are minimal or nonexistent. The density of catecholaminergic nerve fibers, as well as the density of a-adrenoceptors on the smooth muscle, contribute to the wide range of contractile responses evoked by activation of sympathetic adrenergic nerves (Partanen et al., 1982; Barnes, 1986; Goldie et al., 1990; Ind, 1994). Both a1 and a2 adrenoceptors are expressed by airway smooth muscle and both, when activated, produce bronchospasm (Barnes, 1986; Goldie et al., 1990). When no adrenoceptors are antagonized during nerve stimulation, the effects of b-adrenoceptor activation by noradrenaline overwhelms the effects of a-adrenoceptor activation, thus producing bronchodilatation.

Fig. 3. Effects of vagal and sympathetic nerve stimulation on airways resistance and compliance in cats before and after stable bronchospasm evoked by serotonin (5-HT) infusion. Effects of atropine and propranolol on nerve-mediated responses are also illustrated. Cat airways possess cholinergic contractile innervation, adrenergic relaxant innervation, and noncholinergic, parasympathetic relaxant innervation (reprinted with permission from Diamond and O’Donnell, ‘A nonadrenergic inhibitory pathway to feline airways’, Science, v208, pp 185– 188. Copyright 1980 American Association for the Advancement of Science). Similar results have been reported in guinea pigs (Yip et al., 1981).

nerve stimulation than bronchospasm of a similar magnitude evoked by exogenously administered acetylcholine (Sheppard et al., 1983).

3.1.2. Sympathetic-adrenergic ner6es Agonists of a1 and/or a2 adrenoceptors evoke contractions of airway smooth muscle in several species (Barnes, 1986; Goldie et al., 1990; Ind, 1994). When postjunctional b-adrenoceptors are blocked and sympathetic nerves are then activated, contractions of airway smooth muscle in vivo or in vitro can also be evoked. These contractile responses mediated by sympathetic nerve activation can be substantial, while in some species, notably humans, guinea pigs, and rats, the

3.1.3. Tachykinin-containing ner6es The tachykinins substance P and neurokinin A, are potent, direct mediators of airway smooth muscle contraction in most mammalian species (notable exceptions being mice and rats, whose airways either dilate in an epithelium-dependent manner in response to tachykinins or capsaicin, or contract secondary to tachykininmediated airway mast cell activation and serotonin release; Manzini, 1992; Joos et al., 1997). Both neurokinin1 (NK1) and NK2 receptors are expressed by airway smooth muscle cells, and both receptors, when activated, can mediate bronchospasm (the relative contribution of these receptor subtypes to contractions of the airways varies considerably amongst species). NK3 receptors are not expressed by airway smooth muscle, and neurokinin B, an NK3-selective agonist, is not expressed by airway nerves. Potent and highly selective tachykinin receptor antagonists have been developed (Canning, 1997). In guinea pigs, vagus nerve stimulation, electrical field stimulation (EFS) of isolated bronchial preparations, or airway challenge with the vanilloid capsaicin evokes marked and longlasting bronchospasm and airway smooth muscle contractions that are unaffected by ganglionic blockade, and only partially prevented by atropine (Lundberg et al., 1983; Martin et al., 1992). These noncholinergic contractile responses are abolished by NK1 and NK2 receptor antago-

118

B.J. Canning, A. Fischer / Respiration Physiology 125 (2001) 113–127

nists, prevented by desensitization to capsaicin, and mediated by tachykinins (see section by Barnes in this issue for a detailed description of airway axon reflexes). Immunohistochemical studies of airway innervation have indentified three potential neuronal sources of tachykinins in the uninflamed/ undiseased airways: Capsaicin-sensitive afferent Cfibers derived from the vagus nerves, afferent fibers emanating from thoracic dorsal root ganglia, and in some species, postganglionic parasympathetic nerves (Kummer et al., 1992; Dey et al., 1996). The guinea pig and rat are unique in their ability to produce tachykinin-dependent bronchospasm as a consequence of peripheral airway nerve activation (Lundberg et al., 1983; Joos et al., 1997). These contractions are mediated by capsaicin-sensitive nerve endings activated peripherally by stimulants such as capsaicin, or antidromically, by electrical stimulation of the vagus nerves. In no species has a function (i.e. bronchospasm, mucus secretion, vasodilatation) for the substance P found in postganglionic parasympathetic nerves been identified. Thus, ganglionic blockade with hexamethonium, for example, is without effect on responses known to be mediated by peripherally released tachykinins, and hexamethonium-sensitive, vagally-mediated responses in the airways described thus are unaffected by tachykinin receptor antagonists, and not mimicked by exogenously administered tachykinins (Canning and Undem, 1994). Capsaicin evokes bronchospasm in humans, dogs, cats and other species by initiating a cholinergic reflex (Ichinose et al., 1987; Coleridge et al., 1989; Canning and Undem, 1994). The inability of capsaicin or vagus nerve stimulation to evoke tachykinin-dependent contractions in the airways of most species is likely due to low density of tachykinin-containing nerve fibers innervating the smooth muscle (Fischer and Hoffmann, 1996; Canning, 1997; Ellis et al., 1997). Calcitonin gene-related peptide (CGRP) is colocalized to airway afferent nerves with tachykinins (Barnes, 1986; Kummer et al., 1992). CGRP is a potent vasodilator with minimal activity in airway smooth muscle.

4. Relaxant innervation of airway smooth muscle

4.1. Adrenergic-sympathetic ner6es Noradrenergic sympathetic nerves are found in the airways of all species (Barnes, 1986; Ind, 1994; Diamond and O’Donnell, 1980). Activation of these nerves can evoke dilatation of preconstricted airways. The dilating effects of sympathetic nerve stimulation are effectively abolished by the nonselective b-adrenoceptor antagonist propranolol, thereby confirming the presence of sympathetic-adrenergic innervation of airway smooth muscle. In several species, notably humans, there is little functional evidence for adrenergic relaxant innervation of airway smooth muscle (Barnes, 1986; Diamond and Altiere, 1988). That these nerves are present but dysfunctional when studied in vitro seems unlikely, as pulmonary arteries taken from the same lungs exhibit a-adrenergic contractile responses to EFS. Similarly, it is unlikely that b-adrenoceptors on the muscle are dysfunctional or poorly expressed, as exogenously administered b-agonists are effective bronchodilators in vivo and potent and effective relaxants of isolated airways. Rather, it seems likely that the lack of measurable adrenergic nerve-mediated relaxant responses in the airways of some species is due to a paucity of adrenergic nerve fibers innervating the airway smooth muscle (Partanen et al., 1982; Barnes, 1986; Ind, 1994). Relaxations of airway smooth muscle evoked by adrenergic agonists are mediated primarily through activation of b2 adrenoceptors (Goldie et al., 1990). Evidence for both b1 and b3 receptors on airway smooth muscle has also been published. The nature of the adrenoreceptors mediating sympathetic nerve-evoked relaxations of the airways has not been clearly defined. Noradrenaline is colocalized to airway sympathetic nerves with neuropeptide Y (Kummer et al., 1992). This potent vasoconstrictor has little if any direct effects on airway smooth muscle, but may modulate (inhibit) noradrenaline release via prejunctional mechanisms. Similarly, activation of prejunctional a2 adrenoceptors on sym-

B.J. Canning, A. Fischer / Respiration Physiology 125 (2001) 113–127

pathetic-adrenergic nerve terminals might also inhibit noradrenaline release from sympatheticadrenergic nerves innervating the airways.

4.2. Noncholinergic-parasympathetic ner6es Following precontraction of airway smooth muscle and blockade of muscarinic receptors, vagus nerve stimulation evokes hexamethonium-sensitive bronchodilatation and/or smooth muscle relaxation in both cats (Diamond and O’Donnell, 1980; Lama et al., 1988) and guinea pigs (Yip et al., 1981; Canning and Undem, 1993). These relaxant responses are unaffected by b-adrenoceptor antagonists and are similar in kinetic and pharmacological properties to the nonadrenergic relaxant responses evoked in vitro by EFS of feline and guinea pig airway smooth muscle preparations (Li and Rand, 1991; Takahashi et al., 1995). Similar relaxant responses can be evoked by EFS of airways from most mammalian species (Diamond and Altiere, 1988; Canning and Undem, 1994). Nonadrenergic, noncholinergic (NANC) relaxations are thought to be mediated by a combination of vasoactive intestinal peptide (VIP and related peptides) and nitric oxide (NO, formed from arginine by neuronal NO synthase (NOS)) (Li and Rand, 1991; Takahashi et al., 1995). Consistent with this hypothesis, VIP, VIP-related peptides and NOS can be localized (often colocalized) to nerve fibers innervating airway smooth muscle and in parasympathetic ganglia innervating the airways (Kummer et al., 1992; Canning and Undem, 1994; Ward et al., 1995; Dey et al., 1996; Fischer and Hoffmann, 1996; Fischer et al., 1996; Table 1). Contrary to previous assertions, it is highly unlikely that the neurotransmitters mediating NANC relaxations are coreleased with acetylcholine from post-ganglionic parasympathetic nerves. Rather, compelling evidence in guinea pigs (Canning and Undem, 1993; Canning and Fischer, 1997), cats (Ichinose et al., 1987; Lama et al., 1988) and ferrets (Dey et al., 1996), and suggestive evidence in humans (Barnes and Belvisi, 1993; Fischer and Hoffmann, 1996; Fischer et al., 1996), indicates that noncholinergic parasympathetic nerves form a distinct autonomic pathway innervating the airways.

119

The relative contribution of VIP, NO, and other putative neurotransmitters to nerve-mediated relaxant responses of airway smooth muscle is not clearly established. Based primarily on the effects of NO synthase inhibitors, it has been suggested that NO may be the exclusive mediator of NANC relaxations of airway smooth muscle in humans and in other species (Barnes and Belvisi, 1993; Ward et al., 1995). There are, however, several lines of evidence that contradict this assertion (Hobbs et al., 1991; Takahashi et al., 1995). It is not even clear that NO must necessarily leave the nerve terminal to play a role in nonadrenergic relaxant responses (Grider and Jin, 1993; Modin et al., 1994). Such an effect would be more consistent with a neuromodulatory role, and not the role of a neurotransmitter. VIP and the structurally related peptide PACAP are 100 –1000 times more potent than NO as airway relaxant agonists, and both of these neuropeptides are colocalized with NO synthase in airway nerves (Dey et al., 1996; Fischer and Hoffmann, 1996). The tools available to study the pharmacology of VIP and PACAP are woefully inadequate. Resolution of this issue awaits the development of potent and selective antagonists of airway smooth muscle VIP/PACAP receptors.

5. Baseline tone and airway smooth muscle reflexes

5.1. Baseline tone Vagotomy, vagal cooling, administration of atropine or ganglionic blockade with hexamethonium evokes bronchodilatation in conscious or anesthetized, freely breathing or ventilated animals and humans (Canning and Undem, 1994). Conversely, following atropine and propranolol administration and production of a stable increase in smooth muscle tone, vagotomy or ganglionic blockade evokes bronchospasm (Clerici et al. 1989; Fig. 4). These observations are consistent with the hypothesis that airway parasympathetic nerves are tonically active during tidal breathing, producing a stable, readily reversible baseline obstruction of the airways that likely reflects the opposing influences of contractile and relaxant

120

B.J. Canning, A. Fischer / Respiration Physiology 125 (2001) 113–127

airway parasympathetic nerves. Electrophysiological recordings from both preganglionic parasympathetic nerve fibers and postganglionic parasympathetic ganglion neurons confirm the existence of a persistant outflow of parasympathetic nerve activity to the airways (Widdicombe, 1966; Mitchell et al., 1987). Baseline parasympathetic nerve activity can be correlated to respiratory drive. Mitchell and colleagues (1985) hypothesized that airway parasympathetic nerve activity is controlled centrally by respiratory rhythm generators in the brainstem, and modulated peripherally, by afferent input arising from the airways and elsewhere (Fig. 5). Other studies carried out in cats (Jammes and Mei, 1979) and guinea pigs (Kesler and Canning, 1999) provide evidence that ongoing airway parasympathetic nerve activity is absolutely dependent upon ongoing afferent nerve activity arising from the intrapulmonary airways and lungs. The available data seem to favor the latter scenario, whereby the very presence of any baseline tone whatsoever is absolutely dependent upon peripheral afferent input. It is likely, however, that both central and peripheral mechanisms ulti-

mately determine the degree of baseline autonomic nerve activity in the airways. The close association between respiratory drive and airway smooth muscle tone suggests that smooth muscle activity is neurally regulated to facilitate efficient gas exchange. Experimental evidence is consistent with this hypothesis (Widdicombe, 1963).

5.2. Baseline and reflex-mediated alterations in sympathetic ner6e acti6ity Baseline and reflex-mediated alterations in airway sympathetic nerve activity are poorly characterized. The difficulty of discriminating hormonal from neuronal effects mediated by catecholamines contributes to the poor understanding of these airway nerves (Ind, 1994). Electrophysiological recordings from postganglionic sympathetic nerves projecting to the airways provide evidence of a tonic sympathetic input to airway smooth muscle (Widdicombe, 1966). The effects of badrenoceptor antagonists on pulmonary function and airways reactivity supports the hypothesis that airway sympathetic nerves are tonically active (Ind, 1994).

Fig. 4. Effects of vagotomy on bronchospasm evoked by serotonin infusion in young and adult anesthetized guinea pigs pretreated with atropine, phentolamine and propranolol. Vagotomy augmented bronchospasm mediated by serotonin, an effect that was prevented by pretreatment with ganglionic blockers (reprinted with permission from Clerici et al., 1989).

B.J. Canning, A. Fischer / Respiration Physiology 125 (2001) 113–127

121

Fig. 5. The effects of stimulating carotid chemoreceptors with cyanide (CN) on airway smooth muscle tone (Tr tension) and phrenic nerve activity (Phr IF) in an anesthetized, mechanically ventilated cat. Also illustrated is the effect of sustained lung inflation (PEEP) on airway smooth muscle tone and phrenic nerve activity. Note inspiratory rhythm to airway smooth muscle tone and transpulmonary pressure (TPP). Reprinted with permission from Mitchell et al. (1985).

5.3. Reflexes initiating alterations in airway smooth muscle tone Reflex-mediated alterations in airway smooth muscle tone are evoked by stimulation of airway afferent nerves. Such reflexes can also be evoked by stimulation of extrapulmonary afferent nerves. The existence of opposing parasympathetic pathways innervating airway smooth muscle suggests that bronchospasm could be reflexly mediated by an increase in cholinergic nerve activity, or a decrease in tonic inhibitory noncholinergic nerve activity. Conversely, bronchodilatation can be induced by a withdrawal of ongoing cholinergic nerve activity, or an increase in noncholinergic nerve activity. Sympathetic and axon reflex-mediated alterations in airway smooth muscle tone might also influence airway caliber. With the exception of reflex-mediated decreases in noncholinergic parasympathetic nerve activity, all of the above reflex effects in the airways have been described (Coleridge et al., 1989; Widdicombe and Wells, 1994).

5.3.1. Rapidly adapting receptors Pulmonary rapidly adapting receptors monitor alterations in lung compliance and when activated, evoke reflex bronchospasm (Dixon et al., 1979; Pack and Delaney, 1983; Coleridge et al., 1989; Widdicombe and Wells, 1994). Consequently, virtually any constricting autacoid or agonist that acts directly upon the smooth muscle, particularly the smooth muscle of the peripheral airways, evokes a coincident reflex bronchospasm (Fig. 6 and Fig. 7). This includes exogenously administered muscarinic agonists (Wagner and Jacoby, 1999). The observation that bronchospasm evokes reflex bronchospasm has been misinterpreted as evidence that baseline cholinergic tone may be regulated in a positive feedback loop, with cholinergic nerve-mediated contractions evoking further parasympathetic nerve mediated reflex bronchospasm. This assertion is probably incorrect, however, as rapidly adapting receptors respond to the dynamic aspects of bronchospasm and/or alterations in lung compliance, not the static tonus or lung stiffness produced by ongoing parasympathetic nerve activity (Pack and Delaney, 1983).

122

B.J. Canning, A. Fischer / Respiration Physiology 125 (2001) 113–127

Activation of rapidly adapting receptors also evokes reflex-mediated alterations in noncholinergic parasympathetic nerve activity (Inoue et al., 1989). This coincident activation of noncholinergic parasympathetic nerves blunts the magnitude of reflex activation of the cholinergic pathways (Kesler and Canning, 2000). It is not known whether rapidly adapting receptor activation can independently activate these two parasympathetic pathways.

5.3.2. Slowly-adapting stretch receptors Activation of slowly adapting receptors, with positive end-expiratory pressure, for example, decreases airway cholinergic tone (Mitchell et al., 1985; Widdicombe and Wells, 1994; Kesler and Canning, 1999). This reflex directly opposes reflexes evoked by rapidly adapting receptor activation (Figs. 5–7). The opposing influence of slowly adapting receptors may be particularly important in reflex studies using constant volume ventilation. Thus, as rapidly adapting receptors are activated by smooth muscle contraction in response to exogenously administered bronchoconstrictors, airway dead space decreases, thereby increasing peripheral airway and alveolar distension for any given volume of lung inflation. This in turn pro-

longs activation of slowly adapting receptors, which inhibit or blunt further reflex bronchospasm. In this way, airway smooth muscle tone, lung volume and pulmonary stretch receptors interact in a closed feedback loop (Richardson et al., 1984). Slowly adapting stretch receptor effects on noncholinergic parasympathetic nerve activity are poorly characterized. In one study, however, lung hyperinflation did not increase noncholinergic nerve activity (Inoue et al., 1989).

5.3.3. Unmyelinated C-fibers Capsaicin and the inflammatory peptide bradykinin are without marked effects on isolated airway smooth muscle. Both, however, evoke profound increases in parasympathetic nerve activity and thus bronchial smooth muscle tone in large part through direct activation of airway C-fibers (Coleridge et al., 1989; Widdicombe and Wells, 1994). Similar responses can be obtained following challenge with other C-fiber stimulants such as citric acid, hypertonic saline and sulfur dioxide. Activation of airway C-fibers also evokes noncholinergic nerve-mediated bronchodilatation (Ichinose et al., 1987; Kesler and Canning, 2000; Figs. 6 and 7).

Fig. 6. Representative traces of reflex bronchospasm in anesthetized, paralyzed, mechanically-ventilated (fixed volume), guinea pigs evoked by (A) activation of airway rapidly adapting receptors with histamine, or (B) activation of nociceptive-type airway afferent nerves by bradykinin. Tracheal perfusate contained the histamine H1 receptor antagonist pyrilamine (3 mM) and the bradykinin B2 receptor antagonist FR 173657 (0.3 mM), thereby preventing any direct effects of the intravenously administered autacoids on tracheal smooth muscle. All tracheal contractile responses evoked by histamine and bradykinin in these studies were abolished by atropine or prevented by prior vagotomy (n= 3–5; see Fig. 7). Both histamine and bradykinin also evoke a coincident bronchodilatation mediated by airway parasympathetic noncholinergic nerves upon intravenous administration (see text for details). Each trace is representative of six similar experiments.

B.J. Canning, A. Fischer / Respiration Physiology 125 (2001) 113–127

123

Fig. 7. Mean data for reflex bronchospasm and increases in pulmonary insufflation pressure (PIP) in anesthetized, paralyzed, mechanically-ventilated (fixed volume), guinea pigs evoked by (A) histamine, or (B) bradykinin. Panel A: Mean data for the effects of histamine (0.5, 1, 2, 5, 10 and 20 mg/kg iv) on peak PIP and peak tracheal cholinergic tone (expressed as a percentage of the maximum contraction evoked by administering barium chloride to the tracheal perfusate; see Kesler and Canning (1999) for details of the experimental design). Histamine-mediated reflex tracheal contractions were correlated with increases in PIP 5 50% of baseline PIP (r 2 =0.64). When PIP exceeded 50% of baseline, however, further increases in reflex tracheal tension were not evoked, perhaps due to coincident activation of an inhibitory reflex mediated by slowly adapting receptors (see Kesler and Canning, 1999). Panel B: Mean data for the effects of bradykinin on peak PIP and peak tracheal cholinergic tone (expressed as a percentage of the maximum contraction evoked by administering barium chloride to the tracheal perfusate; see Kesler and Canning (1999) for details of the experimental design). Bradykinin, delivered via aerosol or intravenously, evoked similar or larger reflex-mediated increases in tracheal tension than that evoked by histamine despite having only marginal effects on PIP. Each point or column represents the mean9 sem of 3 – 6 experiments.

Airway C-fibers are quiescent in normal lungs whereas slowly and rapidly adapting receptors are sporadically activated during the respiratory cycle (rapidly adapting receptors are far less active than slowly adapting receptors; Pack and Delaney, 1983; Coleridge et al., 1989). The ongoing activity of other airway afferent subtypes makes it possible that reflex effects on smooth muscle tone initiated by C-fiber activation are due in part to synergistic effects with rapidly adapting receptor inputs in the brainstem. Conversely, in guinea pigs and rats, C-fiber-mediated bronchospasm via axon reflex and peripheral release of tachykinins might also coincidentally activate rapidly adapting receptors secondary to tachykinin-mediated plasma leakage and/or bronchospasm.

5.3.4. Effects of extrapulmonary afferent ner6es and chemoreceptors on autonomic regulation of airway smooth muscle tone Activation of extrapulmonary afferent nerves can alter airway smooth muscle autonomic nerve activity (Coleridge et al., 1989; Canning and Undem, 1994; Widdicombe and Wells, 1994). Activation of striated muscle afferents in the limbs

or diaphragm, perhaps similar to the effects of strenuous exercise, evokes dilatation of the airways through inhibition of cholinergic tone. Conversely, activation of esophageal afferents or afferents in the upper airway can evoke reflex cholinergic bronchospasm (Widdicombe and Wells, 1994; Canning, 1999). Such reflexes may contribute to the pulmonary consequences of upper airways diseases (i.e. rhinitis), or perhaps asthma precipitated by gastroesophageal reflux disease (GERD). Central and peripheral chemoreceptors can also regulate airway smooth muscle tone (Nadel and Widdicombe, 1962; Widdicombe, 1963; Mitchell et al., 1987; Ichinose et al., 1987) (Fig. 5). Interestingly, the effects of chemoreceptor stimulation seem to be selective for the cholinergic-parasympathetic nerves, as noncholinergic parasympathetic nerve activity is unaffected by chemoreceptor stimulation (Ichinose et al., 1987). This indicates that distinct parasympathetic pathways mediate bronchospasm and bronchodilatation. It is, however, the only stimulus identifed that differentially activates these distinct pathways.

124

B.J. Canning, A. Fischer / Respiration Physiology 125 (2001) 113–127

6. Neural regulation of airway smooth muscle tone in disease There is compelling physiological evidence to support the hypothesis that dysfunction or dysregulation of the autonomic innervation of airway smooth muscle contributes to the pathogenesis of asthma and chronic obstructive pulmonary disease (COPD). These heterogeneous diseases are characterized by airways obstruction, airways hyperresponsiveness, wheezing, coughing, dyspnea and the sensation of breathlessness. Each of these symptoms has been directly or indirectly linked to alterations in airway nerve function.

6.1. Airways obstruction Airways obstruction is elevated in asthma and COPD and can be substantially and rapidly reduced by anticholinergics such as atropine and ipratropium bromide (Molfino et al., 1993; Gross, 1997). The effects of these agents approximates the effects of direct acting beta agonists, indicating that a large proportion of the airways obstruction in asthma and COPD is due to excessive constriction mediated by airway parasympathetic nerves. The mechanism by which airways disease processes produce such marked increases in airway smooth muscle cholinergic tone are poorly understood. In vitro studies of asthmatic airways reveal little or no evidence for an increase in responsiveness to acetylcholine or increased efficacy of airway cholinergic nerves. Morphological studies have also failed to consistently document alterations in airway innervation (Canning, 1999). This suggests that manifestation of airways obstruction is necessarily dependent upon the milieu of the airways and/or intact vagus nerves. The hypothesis that prejunctional M2 receptors are dysfunctional has had widespread appeal (Fryer and Maclagan, 1984; Barnes, 1999). Dysfunction of M2 receptors is, however, just one of many potential mechanisms of increased cholinergic nerve activity, and complete blockade of M2 receptors with selective antagonists only marginally alters airways reactivity. It would be surprising, therefore, if M2 receptor dysfunction accounted

entirely for the alterations in airway neural control seen in asthma and COPD. Conversely, the coincident increase in coughing associated with obstructive airways diseases makes it likely that alterations in airway afferent excitability or activity contributes to the increase in airway autonomic tone. A potential mechanism of elevated parasympathetic nerve activity in COPD may be an altered ability of slowly adapting stretch receptors to respond to lung inflation. Pathological abnormalities in the peripheral airways that precede development of the emphysematous lung could alter the sensory nerve terminals of stretch receptors, or alter the mechanical forces put upon airway afferent nerves during breathing. This might reduce stretch receptor activation, thereby reducing inhibitory input to the airway parasympathetic nerves and permitting excessive effects of excitatory afferent nerve input (e.g. rapidly adapting receptors, C-fibers, central drive). Such a pathological abnormality might also account for the sensation of breathlessness and dyspnea in COPD despite normal blood gases. Although difficult, further assessment of potential abnormalities in airway afferent excitability in COPD and asthma are warranted. 6.2. Airways hyperresponsi6eness There is compelling evidence to support the hypothesis that autonomic nerve dysfunction and/ or dysregulation contributes to the pathogenesis of airways hyperresponsiveness (Barnes, 1986; Canning and Undem, 1994; Gross, 1997; Canning, 1999). Anticholinergics such as atropine and ipratropium bromide markedly reduce or abolish airways responsiveness to a variety of challenges in asthmatics (Table 2). These effects on airways reactivity by anticholinergics vary between complete abolishment of responses (beta blocker-induced bronchospasm, post-viral infection hyperreactivity to histamine in nonasthmatic subjects, psychogenic stimuli) to partial reductions in airways reactivity in asthmatics (histamine, prostaglandin D2). The differential effects of anticholinergics on airways reactivity likely relates to the nature of the challenging agents and the ability to directly contract airway smooth muscle.

B.J. Canning, A. Fischer / Respiration Physiology 125 (2001) 113–127

Noncholinergic parasympathetic relaxant nerves might also be dysfunctional in airways diseases (Barnes, 1986). Studies in animals suggest that dysregulation or dysfunction of airway relaxant nerves can precipitate airways hyperreactivity (Canning, 1999; Kesler and Canning, 2000). Consistent with these observations, an inability to dilate with deep inspiration, and not excessive constriction of the bronchial musculature, may be an important mechanism of airways hyperresponsiveness in asthma (Skloot et al., 1995). Unfortunately, how these nerves function in the healthy lung, let alone how they might dysfunction in disease, remains poorly characterized. Functional studies of airway noncholinergic nerves in the healthy and diseased lungs are warranted.

125

vation of airway smooth muscle nerves can evoke near maximal constriction, or conversely, can completely dilate a constricted airway. The primary contractile innervation of airway smooth muscle is cholinergic and parasympathetic in nature. Both adrenergic and nonadrenergic relaxant nerves have been described. Nonadrenergic, noncholinergic nerves are the primary dilating nerves in many species, and the only functional dilating nerves innervating human airway smooth muscle. These noncholinergic parasympathetic nerves are physiologically and morphologically distinct from parasympathetic cholinergic nerves and remain poorly characterized. Both parasympathetic pathways are regulated by vagal afferent nerve activity. Dysfunction or dysregulation of these nerves plays a primary role in the pathogenesis of airways disease.

7. Conclusion Nerves innervating airway smooth muscle play an essential role in regulating airway caliber. ActiTable 2 Effect of anticholinergics on airways hyper-responsiveness in asthmaa Provocation

Effect of anticholinergic on responsiveness

Prostaglandin D2 Thromboxane A2 Bradykinin Histamine Capsaicin Distilled water Beta blockers Hyperpnea Psychogenic stimulation Exercise Reflux/Esophageal acidification

12–22-fold increase in PC20 23-fold increase in PC20 Fivefold increase in PD35 Tenfold increase in PC100 SRaw 60% reduction in response 50–100% reduction in response Abolished response Abolished response in children Abolished response

a

30% reduction in response Abolished response

Anticholinergics used were either ipratropium bromide or atropine delivered via aerosol. Results reviewed in detail elsewhere (Barnes, 1986; Coleridge et al., 1989; Canning and Undem, 1994; Gross, 1997; Canning, 1999). Abbreviations: PC20 and PD35: provocative concentration (or dose) of agonist producing a 20% or 35% fall in forced expiratory volume in 1 second (FEV1). PC100 SRaw: provocative concentration of agonist producing a 100% increase in specific airways resistance.

Acknowledgements The research presented in this manuscript was made possible through grants provided by the NIH (USA) and the BMBF (Germany). The authors thank Sandra Reynolds for technical assistance.

References Barnes, P.J., 1986. Neural control of human airways in health and disease. Am. Rev. Respir. Dis. 134, 1289– 1314. Barnes, P.J., Belvisi, M.G., 1993. Nitric oxide and lung disease. Thorax 48, 1034– 1043. Barnes, P.J., 1999. Airway muscarinic receptors. In: Spector, S.L. (Ed.), Anticholinergic Agents in the Upper and Lower Airways. Marcel Dekker, New York, Basel, pp. 31 – 58. Cabezas, G.A., Graf, P.D., Nadel, J.A., 1971. Sympathetic versus parasympathetic nervous regulation of airways in dogs. J. Appl. Physiol. 31, 651– 655. Canning, B.J., Undem, B.J., 1993. Evidence that distinct neural pathways mediate parasympathetic contractions and relaxations of guinea-pig trachealis. J. Physiol. London 471, 25 – 40. Canning, B.J., Undem, B.J., 1994. In: Raeburn, D., Gymbiecz, M. (Eds.), Airways Smooth Muscle: Structure, Innervation and Neurotransmission. Birkhauser Verlag, Basel, pp. 43 – 77. Canning, B.J., 1997. Potential role of tachykinins in inflammatory diseases. J. Allergy Clin. Immunol. 99, 579– 582.

126

B.J. Canning, A. Fischer / Respiration Physiology 125 (2001) 113–127

Canning, B.J., Fischer, A., 1997. Localization of cholinergic nerves in the airways of the guinea pig using antisera to choline acetyltransferase. Am. J. Physiol. 272, L731–L738. Canning, B.J., 1999. Inflammation in asthma: role of nerves and potential influence of gastroesophageal reflux disease. In: Stein, M.R. (Ed.), GERD and Airway Diseases. Marcel Dekker, New York, Basel, pp. 19–54. Clerici, C., Macquin-Mavier, I., Harf, A., 1989. Nonadrenergic bronchodilation in adult and young guinea pigs. J. Appl. Physiol. 67, 1764–1769. Coleridge, H.M., Coleridge, J.C.G., Schultz, H.D., 1989. Afferent pathways involved in reflex regulation of airway smooth muscle. Pharmacol. Ther. 42, 1–63. Dey, R.D., Altemus, J.B., Rodd, A., Mayer, B., Said, S.I., Coburn, R.F., 1996. Neurochemical characterization of intrinsic neurons in ferret tracheal plexus. Am. J. Respir. Cell Mol. Biol. 14, 207–216. Diamond, L., O’Donnell, M., 1980. A nonadrenergic vagal inhibitory pathway to feline airways. Science 208 (11), 185– 188. Diamond, L., Altiere, R.J., 1988. Airway nonadrenergic noncholinergic inhibitory nervous system. In: Lenfant, C. (Ed.), Lung Biology in Health and Disease, vol. 33. Marcel Dekker, Basel, New York, pp. 343–394. Dixon, W.E., Brodie, T.G., 1903. Contributions to the physiology of the lungs. Part I. The bronchial muscles, their innervation, and the action of drugs upon them. J. Physiol., London 29, 97 – 173. Dixon, M., Jackson, D.M., Richards, I.M., 1979. The effects of histamine, acetylcholine and 5-hydroxytryptamine on lung mechanics and irritant receptors in the dog. J. Physiol., London 287, 393–403. Dixon, W.E., Ransom, F., 1912. Broncho-dilator nerves. J. Physiol., London 45, 413–428. Ellis, J.L., Sham, J.S.K., Undem, B.J., 1997. Tachykinin-independent effects of capsaicin on smooth muscle tone in human isolated bronchi. Am. Rev. Respir. Crit. Care Med. 155, 751– 755. Fischer, A., Canning, B.J., Kummer, W., 1996. Correlation of vasoactive intestinal peptide and NO synthase with choline acetyltransferase in the airway innervation. Ann. N.Y. Acad. Sci. 805, 717– 722. Fischer, A., Hoffmann, B., 1996. Nitric oxide synthase in neurons and nerve fibers of lower airways and in vagal sensory ganglia of man: correlation with neuropeptides. Am. J. Respir. Crit. Care Med. 154, 209–216. Fryer, A.D., Maclagan, J., 1984. Muscarinic inhibitory receptors in pulmonary parasympathetic nerves in the guineapig. Br. J. Pharmacol. 83, 973–978. Goldie, R.G., Paterson, J.W., Lulich, K.M., 1990. Adrenoceptors in airway smooth muscle. Pharmacol. Ther. 48, 295– 322. Grider, J.R., Jin, J.G., 1993. Vasoactive intestinal peptide release and L-citrulline production from isolated ganglia of the myenteric plexus: evidence for regulation of vasoactive intestinal peptide release by nitric oxide. Neuroscience 54 (2), 521– 526.

Gross, N.J., 1997. Anticholinergic drugs. In: Barnes, P.J., Grunstein, M.M., Leff, A.R., Woolcock, A.J. (Eds.), Asthma, vol. 2. Lippincott-Raven, Philadelphia, New York, pp. 1555– 1568. Hobbs, A.J., Tucker, J.F., Gibson, A., 1991. Differentiation by hydroquinone of relaxations induced by exogenous and endogenous nitrates in non-vascular smooth muscle: Role of superoxide anions. Br. J. Pharmacol. 104, 645– 650. Ichinose, M., Inoue, H., Miura, M., Yafuso, N., Nogami, H., Takishima, T., 1987. Possible sensory receptor of nonadrenergic inhibitory nervous system. J. Appl. Physiol. 63, 923– 929. Ind, P.W., 1994. Role of the sympathetic nervous system and endogenous catecholamines in the regulation of the airways smooth muscle tone. In: Raeburn, D., Gymbiecz, M. (Eds.), Airways Smooth Muscle: Structure, Innervation and Neurotransmission. Birkhauser Verlag, Basel, pp. 29 – 41. Inoue, H., Ichinose, M., Miura, M., Katsumata, U., Takishima, T., 1989. Sensory receptors and reflex pathways of nonadrenergic inhibitory nervous system in feline airways. Am. Rev. Respir. Dis. 139, 1175– 1178. Jammes, Y., Mei, N., 1979. Assessment of the pulmonary origin of bronchoconstrictor vagal tone. J. Physiol., London 291, 305– 316. Joos, G.F., Lefebvre, R.A., Bullock, G.R., Pauwels, R.A., 1997. Role of 5-hydroxytryptamine and mast cells in the tachykinin-induced contraction of rat trachea in vitro. Eur. J. Pharmacol. 338, 259– 268. Kesler, B.S., Canning, B.J., 1999. Regulation of baseline cholinergic tone in guinea-pig airway smooth muscle. J. Physiol., London 518, 843– 856. Kesler, B.S., Canning, B.J., 2000. Nitric oxide dependent modulation of smooth muscle tone by airway parasympathetic nerves. Am. J. Respir. Crit. Care Med., submitted. Kummer, W., Fischer, A., Kurkowski, R., Heym, C., 1992. The sensory and sympathetic innervation of guinea pig lung and trachea as studied by retrograde neuronal tracing and double-labeling immunohistochemistry. Neuroscience 49, 715– 737. Lama, A., Delpierre, S., Jammes, Y., 1988. The effects of electrical stimulation of myelinated and non-myelinated vagal motor fibres on airway tone in the rabbit and cat. Respir. Physiol. 74, 265– 274. Li, C.G., Rand, M.J., 1991. Evidence that part of the NANC relaxant response of guinea-pig trachea to electrical field stimulation is mediated by nitric oxide. Br. J. Pharmacol. 102, 91 – 94. Lundberg, J.M., Brodin, E., Saria, A., 1983. Effects and distribution of vagal capsaicin-sensitive substance P neurons with special reference to the trachea and lungs. Acta Physiol. Scand. 119, 243– 252. Macklin, C.C., 1929. The musculature of the bronchi and lungs. Physiol. Rev. 9, 1 – 60. Manzini, S., 1992. Bronchodilatation by tachykinins and capsaicin in the mouse main bronchus. Br. J. Pharmacol. 105, 968– 972.

B.J. Canning, A. Fischer / Respiration Physiology 125 (2001) 113–127 Martin, C.A.E., Naline, E., Emonds-Alt, X., Advenier, C., 1992. Influence of ( 9)-CP-96345 and SR 48968 on electrical field stimulation of the isolated guinea-pig main bronchus. Eur. J. Pharmacol. 224, 137–143. Mitchell, R.A., Herbert, D.A., Baker, D.G., 1985. Inspiratory rhythm in airway smooth muscle tone. J. Appl. Physiol. 58, 911– 920. Mitchell, R.A., Herbert, D.A., Baker, D.G., Basbaum, C.B., 1987. In vivo activity of tracheal parasympathetic ganglion cells innervating tracheal smooth muscle. Brain Res. 437, 157– 160. Modin, A., Weitzberg, E., Lundberg, J.M., 1994. Nitric oxide regulates peptide release from parasympathetic nerves and vascular reactivity to vasoactive intestinal polypeptide in vivo. Eur. J. Pharmacol. 261, 185–197. Molfino, N.A., Slutsky, A.S., Julia-Serda, G., Hoffstein, V., Szalai, J.P., Chapman, K.R., Rebuck, A.S., Zamel, N., 1993. Assessment of airway tone in asthma. Am. Rev. Respir. Dis. 148, 1238–1243. Nadel, J.A., Widdicombe, J.G., 1962. Effect of changes in blood gas tensions and carotid sinus pressure on tracheal volume and total lung resistance to air flow. J. Physiol., London 163, 13 – 33. Pack, A.I., Delaney, R.G., 1983. Response of pulmonary rapidly adapting receptors during lung inflation. J. Appl. Physiol. 55, 955– 963. Partanen, M., Laitinen, A., Hervonen, A., Toivanen, M., Laitinen, L.A., 1982. Catecholamine- and acetylcholinesterase-containing nerves in human lower respiratory tract. Histochemistry 76, 175–188. Richardson, C.A., Herbert, D.A., Mitchell, R.A., 1984. Modulation of pulmonary stretch receptors and airway resistance by parasympathetic efferents. J. Appl. Physiol.

127

57, 1842– 1849. Sheppard, D., Epstein, J., Holtzman, M.J., Nadel, J.A., Boushey, H.A., 1983. Effect of route of atropine delivery on bronchospasm from cold air and methacholine. J. Appl. Physiol. 54 (1), 130– 133. Skloot, G., Permutt, S., Togias, A., 1995. Airway hyperresponsiveness in asthma: A problem of limited smooth muscle relaxation with inspiration. J. Clin. Invest. 96, 2393– 2403. Takahashi, N., Tanaka, H., Abdullah, N., Jing, L., Inoue, R., Ito, Y., 1995. Regional difference in the distribution of L-NAME-sensitive and -insensitive NANC relaxations in cat airway. J. Physiol., London 488, 709– 720. Wagner, E.M., Jacoby, D.B., 1999. Methacholine causes reflex bronchoconstriction. J. Appl. Physiol. 86, 294– 297. Ward, J.K., Barnes, P.J., Springall, D.R., Abelli, L., Tadjkarimi, S., Yacoub, M.H., Polak, J.M., Belvisi, M.G., 1995. Distribution of human i-NANC bronchodilator and nitric oxide-immunoreactive nerves. Am. J. Respir. Cell Mol. Biol. 13, 175– 184. Widdicombe, J.G., 1963. Regulation of tracheobronchial smooth muscle. Physiol. Rev. 43, 1 – 37. Widdicombe, J.G., 1966. Action potentials in parasympathetic and sympathetic efferent fibres to the trachea and lungs of dogs and cats. J. Physiol., London 186, 56 – 88. Widdicombe, J.G., Wells, U.M., 1994. Vagal reflexes. In: Raeburn, D., Gymbiecz, M.A. (Eds.), Airway Smooth Muscle: Structure, Innervation and Neurotransmission. Birkha¨user Verlag, Basel, pp. 279– 307. Yip, P., Palombini, B., Coburn, R.F., 1981. Inhibitory innervation to the guinea pig trachealis muscle. J. Appl. Physiol. 50, 374– 382.

.