Muscarinic Receptor Antagonists: Effects on Pulmonary Function

Abstract

In healthy lungs, muscarinic receptors control smooth muscle tone, mucus secretion, vasodilation, and inflammation. In chronic obstructive pulmonary disease (COPD) and asthma, cholinergic mechanisms contribute to increased bronchoconstriction and mucus secretion that limit airflow. This chapter reviews neuronal and nonneuronal sources of acetylcholine in the lung and the expression and role of M₁, M₂, and M₃ muscarinic receptor subtypes in lung physiology. It also discusses the evidence for and against the role of parasympathetic nerves in asthma, and the current use and therapeutic potential of muscarinic receptor antagonists in COPD and asthma.

1 Acetylcholine in the Lung

The lung’s primary role is gas exchange, but it also serves as both a barrier and defense against pathogens and environmental contaminants. Acetylcholine produced and released by both neuronal and nonneuronal sources acts through muscarinic receptors to regulate these important physiological functions. Acetylcholine contracts airway smooth muscle to control tone and regulate patency of the conducting airways. In blood vessels, acetylcholine causes smooth muscle relaxation and vasodilation. At mucosal glands and epithelial cells, acetylcholine regulates mucus secretion and, via ciliary beat frequency, mucus clearance. Acetylcholine also modulates inflammation.

1.1 Neuronal Acetylcholine in the Lungs

Parasympathetic nerves synthesize and release acetylcholine and are the primary source of acetylcholine in the lung. They innervate all conducting airways, from the trachea to the bronchioles (Canning and Fischer 1997), and pulmonary blood vessels (Cavallotti et al. 2005; Haberberger et al. 1997).

Preganglionic parasympathetic nerves originate in the brain in the medulla oblongata, and their axons travel in the right and left vagus nerves to synapse with postganglionic nerves in the trachea and bronchi (Kalia 1981; McAllen and Spyer 1978). Postganglionic nerve cell bodies are clustered in ganglia of 2–38 cells (Baker et al. 1986; Canning and Fischer 1997), with axons projecting to airway smooth muscle (Canning and Fischer 1997; Daniel et al. 1986), mucous glands (Basbaum 1984), and blood vessels (both arteries and large veins) (Cavallotti et al. 2005; Haberberger et al. 1997; Knight et al. 1981). Acetylcholine released from cholinergic parasympathetic nerves induces smooth muscle contraction (Cabezas et al. 1971; Olsen et al. 1965) and mucus secretion (Baker et al. 1985; Gallagher et al. 1975; Ramnarine et al. 1996) in the conducting airways, and vasodilation in pulmonary arteries and large veins (Laitinen et al. 1987).

Cholinergic parasympathetic nerves fire tonically (Widdicombe 1966; Widdicombe et al. 1962) to contract airway smooth muscle during normal breathing in humans and animals (Jammes and Mei 1979; Kesler and Canning 1999; Roberts et al. 1988; Sheppard et al. 1982; Widdicombe 1966). Thus, vagotomy (severing the vagus nerve) relaxes smooth muscle and decreases airway tone while electrically stimulating the distal ends of cut vagus nerves contracts airway smooth muscle. Airway smooth muscle contraction narrows conducting airways resulting in bronchoconstriction, which decreases the volume of conducting airways and increases resistance to airflow (Cabezas et al. 1971; Kesler and Canning 1999; Olsen et al. 1965; Severinghaus and Stupfel 1955).

Acetylcholine, released by parasympathetic nerves upon stimulation, acts directly at muscarinic receptors on airway smooth muscle to cause bronchoconstriction. Therefore, as with vagotomy, muscarinic receptor antagonists decrease smooth muscle tone (Kesler and Canning 1999; Severinghaus and Stupfel 1955; Sheppard et al. 1982) and prevent bronchoconstriction induced by electrical stimulation of the vagus nerves. Furthermore, bronchoconstriction is increased by acetylcholinesterase inhibitors, which prevent metabolism of acetylcholine (Colebatch and Halmagyi 1963).

Cholinergic parasympathetic nerves are the efferent arm of vagal reflexes initiated by both physical (ex. dust and cold air) and chemical (ex. allergens, histamine, methacholine, and irritant gasses such as SO₂) stimuli (Gold et al. 1972; Holtzman et al. 1980; Sheppard et al. 1982; Wagner and Jacoby 1999; Widdicombe et al. 1962). These stimuli directly or indirectly activate afferent sensory nerves in the respiratory tract, which conduct signals to the brain via the vagus nerve (Widdicombe et al. 1962). Parasympathetic neurons carry the reflex response back to the lungs via acetylcholine release that leads to bronchoconstriction and mucus secretion in the airways.

Convincing evidence for reflex bronchoconstriction comes from experiments where a stimulus is applied to one location in the respiratory tract and a cholinergic response is measured in a separate location innervated by the efferent arm of the reflex arc (Nadel et al. 1965; Wagner and Jacoby 1999). For example, SO₂ gas applied to the larynx of tracheostomized animals causes bronchoconstriction in the trachea and bronchi. Severing either the afferent sensory nerves that leave the larynx or the efferent vagus nerves completely prevents this reflex bronchoconstriction (Boushey et al. 1972; Nadel et al. 1965). Mucus secretion is also initiated by reflexes but is challenging to measure in the lower airways. However, in the human upper respiratory tract, histamine applied to one nostril induces secretions in both nostrils. Secretions in the contralateral untreated nostril are completely prevented by muscarinic receptor antagonists, supporting a cholinergic reflex (Baroody et al. 1993). From a therapeutic perspective, vagal reflexes are important because muscarinic receptor antagonists are capable of blocking bronchoconstriction and mucus secretion initiated by a variety of chemical and physical stimuli in humans (Baroody et al. 1993; Holtzman et al. 1980; Sheppard et al. 1982).

It is also important to note that some chemicals act both directly at smooth muscle and indirectly through vagal reflexes to cause bronchoconstriction. This is true for both inhaled methacholine (muscarinic agonist) and inhaled histamine (Belmonte et al. 1998; Holtzman et al. 1980; Wagner and Jacoby 1999), two drugs commonly used to test airway responsiveness. For instance, most bronchoconstriction induced by inhaled histamine in humans is prevented by pharmacological blockade of vagal ganglionic neurotransmission with a neuronal nicotinic receptor antagonist (Holtzman et al. 1980). While no evidence exists for a similar effect of methacholine in people, there is substantial evidence for methacholine-induced reflex bronchoconstriction in diverse species such as sheep and rats, strongly suggesting that a similar reflex may also occur in humans (Belmonte et al. 1998; Wagner and Jacoby 1999).

1.2 Nonneuronal Acetylcholine in the Lungs

Acetylcholine synthesis and release is not limited to cholinergic neurons in the lung. Both epithelial cells (Proskocil et al. 2004; Reinheimer et al. 1998) and endothelial cells (Haberberger et al. 1997, 2000) in the lung contain the cellular machinery required to synthesize and release acetylcholine including choline acetyltransferase (ChAT), which catalyzes the synthesis of acetylcholine; hemicholinium-3 sensitive choline transporters that transport choline into cells (Ferguson et al. 2003); and vesicular acetylcholine transporters, which package acetylcholine into vesicles. In addition, ciliated epithelial cells express organic cation transporters OCT1 and OCT2 in their luminal membranes. These polyspecific organic cation transporters can transport acetylcholine and are believed to directly release acetylcholine into the airway lumen (Kummer et al. 2006; Lips et al. 2005).

Airway epithelial cells contain and release acetylcholine as measured by high-pressure liquid chromatography (Proskocil et al. 2004; Reinheimer et al. 1996, 1998). Epithelial cells from freshly isolated human bronchi contain 23±6 pmol acetylcholine per gram bronchus. This is just 1% of the 2,600±500 pmol acetylcholine per gram bronchus contained in the whole bronchial wall, but it is likely physiologically important for increasing ciliary beat frequency via muscarinic receptors on ciliated airway epithelial cells (Corssen and Allen 1959; Klein et al. 2009; Reinheimer et al. 1998; Wong et al. 1988).

Acetylcholine acts at muscarinic receptors in pulmonary arteries to induce vasodilation (Greenberg et al. 1987; McMahon and Kadowitz 1992). One source of acetylcholine is presumably the endothelial cells since they contain the machinery for synthesizing and releasing acetylcholine. In pulmonary arteries, endothelial ChAT expression is mosaic suggesting acetylcholine production may be related to differences in local mechanical forces due to blood flow (Haberberger et al. 2000).

2 Muscarinic Receptors in the Lungs

Sir Henry Dale first divided the actions of acetylcholine, and other choline derivatives, into nicotinic and muscarinic, based on their similarity to responses elicited by either nicotine or muscarine (Dale 1914). Nicotinic acetylcholine receptors are ligand-gated ion channels and muscarinic receptors are G protein-coupled. Although nicotinic receptors are also present throughout the lungs and are crucial for neurotransmission between pre- and postganglionic parasympathetic nerves, muscarinic receptors are a major physiological target for acetylcholine in the lungs and are the primary focus of this chapter.

Five muscarinic receptor subtypes, M₁, M₂, M₃, M₄, and M₅, are recognized by the International Union of Pharmacology (Caulfield and Birdsall 1998). M₁, M₃, and M₅ receptors typically couple to Gα_q/11 while M₂ and M₄ receptors typically couple to Gα_i/o. All five muscarinic receptor subtypes are expressed in the lungs. Currently, strong evidence for a functional role only exists for M₁, M₂, and M₃ receptors, and muscarinic receptor antagonists that target these receptors are used to treat several lung diseases, including asthma and chronic obstructive pulmonary disease (COPD).

The earliest studies of muscarinic receptor distribution in lung tissues employed autoradiographic labeling of muscarinic receptor density in lung sections (van Koppen et al. 1987, 1988). These studies demonstrated that muscarinic receptor density is actually highest in parasympathetic ganglia, followed by mucous glands, smooth muscle, and nerve fibers (van Koppen et al. 1988). Subtype-specific distribution and function has subsequently been determined using pharmacological analysis, in situ hybridization, RT-PCR, and knockout mice. The distribution of receptor subtypes in tissues based on these assays is in good agreement. However, it is important to note that most “selective” muscarinic antagonists have at most a 10-fold selectivity for one muscarinic receptor subtype over other subtypes, where selectivity is usually defined as 100-fold higher affinity (Caulfield and Birdsall 1998). In addition, many commercially available receptor antibodies may not be specific, based on the presence of similar antibody staining patterns in wild-type and muscarinic receptor gene-deficient mice (Jositsch et al. 2009; Pradidarcheep et al. 2008).

2.1 Muscarinic Receptors on Airway Nerves and Ganglia

Both pre- and postganglionic parasympathetic nerves innervating the lungs express muscarinic receptors that are densest at the ganglia (van Koppen et al. 1987, 1988). Muscarinic receptors on parasympathetic nerves modulate synaptic neurotransmission between the pre- and postganglionic nerves, and also limit release of acetylcholine by postganglionic nerves at target tissues (smooth muscle, glands).

Preganglionic autonomic nerves release acetylcholine onto nicotinic receptors at their synapses with postganglionic nerves. In the lungs, muscarinic receptors modulate neurotransmission across this synapse (Myers 2001). Preganglionic neurons contain inhibitory M₂ receptors at the synapse, which limit acetylcholine release in guinea pig bronchi. Thus, activating M₂ receptors during electrical stimulation of the preganglionic nerves reduces acetylcholine release into the synapse, which in turn decreases the amplitude of nicotinic fast excitatory postsynaptic potentials recorded in bronchial ganglia (Myers and Undem 1996).

M₁ muscarinic receptors are found in cell bodies of postganglionic nerves, although there are species differences regarding the importance of these receptors in modulating synaptic neurotransmission. In guinea pigs, M₁ receptors depolarize the resting membrane potential in approximately 50% of ganglion cells, which would be expected to facilitate neurotransmission at the synapse and increase bronchoconstriction (Myers and Undem 1996). However, blocking ganglion M₁ receptors does not reduce smooth muscle contraction induced by electrically stimulating vagal preganglionic nerves in guinea pigs (Undem et al. 1990). Similarly, no functional role for M₁ receptors has been identified in rat tracheal ganglia (Murai et al. 1998). This contrasts with rabbit, where blocking M₁ receptors in the bronchi reduces smooth muscle contraction following vagal stimulation (Bloom et al. 1988). Finally, in atopic humans, blocking M₁ receptors in the lung with pirenzepine decreases inhaled SO₂-induced reflex bronchoconstriction by approximately 50%. These data indirectly support a role for M₁ receptors facilitating parasympathetic neurotransmission in allergic humans. However, the importance of M₁ receptors in healthy humans is unclear, since pirenzepine inhibits only 17% of vagal tone in healthy women (Fujimura et al. 1992).

Of greater physiological importance are inhibitory M₂ receptors on postganglionic parasympathetic nerves that were first described in guinea pig (Fryer and Maclagan 1984). These neuronal M₂ receptors are activated by acetylcholine to inhibit further acetylcholine release in a feedback mechanism that limits vagally induced bronchoconstriction and mucus secretion in healthy animals and humans (Ayala and Ahmed 1989; Blaber et al. 1985; Fryer et al. 1996; Fryer and Maclagan 1984; Minette et al. 1989; Ramnarine et al. 1996). When neuronal M₂ receptors in trachea are blocked, the amount of acetylcholine released by nerve stimulation significantly increases (Baker et al. 1992). Pharmacologically blocking inhibitory M₂ receptors with gallamine in guinea pigs or deleting them genetically in mice significantly potentiates vagally induced bronchoconstriction in vivo, while selectively activating M₂ receptors with low doses of pilocarpine inhibits vagally induced bronchoconstriction 80% (Fisher et al. 2004; Fryer and Maclagan 1984). Similarly, in healthy humans, low doses of muscarinic agonists reduce bronchoconstriction induced by a vagal reflex, demonstrating a role for inhibitory neuronal M₂ receptors in limiting acetylcholine release (Ayala and Ahmed 1989; Minette et al. 1989).

2.2 Muscarinic Receptors on Airway Smooth Muscle

In the lungs, acetylcholine causes bronchoconstriction via smooth muscle contraction (Haddad et al. 1991; Roffel et al. 1988, 1990; Stengel et al. 2000; Struckmann et al. 2003). The presence of M₂ and M₃ receptors on airway smooth muscle is supported by radioligand binding data, autoradiography, in situ hybridization, and genetic deletion in humans, cows, guinea pigs, dogs, and mice (Fernandes et al. 1992; Haddad et al. 1991; Mak and Barnes 1990; Mak et al. 1992; Roffel et al. 1987; Struckmann et al. 2003). Physiological data support M₃ receptors as having the dominant role in smooth muscle contraction.

Functional experiments demonstrate that contraction induced by muscarinic ligands in isolated trachea and bronchi is mediated by M₃ receptors in all species including humans (Haddad et al. 1991; Roffel et al. 1988, 1990; Struckmann et al. 2003). In addition, in vivo experiments in muscarinic receptor gene-deficient mice demonstrate that only M₃ receptors contribute to bronchoconstriction induced by electrical stimulation of the vagus nerves or intravenous methacholine (Fisher et al. 2004). Recently, it has also been shown that smooth muscle M₃ receptors are activated in the absence of acetylcholine by membrane depolarization induced chemically with KCl (Liu et al. 2009). This ligand-independent activation of M₃ receptors has only thus far been demonstrated in mice. It is not known whether membrane depolarization activates M₃ receptors in vivo.

M₂ receptors often outnumber M₃ receptors but have an indirect role in airway smooth muscle contraction. M₂ receptors on airway smooth muscle inhibit relaxation induced both by β-adrenoreceptor agonists and adenylyl cyclase activation with forskolin (Fernandes et al. 1992). Thus, M₂ receptors contribute to smooth muscle contraction by functionally antagonizing Gα_s-induced relaxation. In isolated trachea from mice deficient for M₂ receptors, muscarinic agonist potency is reduced, however maximum contraction is still achieved (Stengel et al. 2000). This suggests M₂ receptors contribute to acetylcholine-induced smooth muscle contraction, but that M₃ receptors alone are sufficient for smooth muscle contraction. In vitro, airway narrowing induced by muscarine can only be completely prevented in airways from mice deficient in both M₂ and M₃ receptor genes. In vivo, however, only M₃ receptors contribute to bronchoconstriction induced by vagal stimulation and intravenous methacholine, since bronchoconstriction is absent in mice deficient for M₃ receptors.

2.3 Muscarinic Receptors on Airway Submucosal Glands

Parasympathetic nerves also stimulate mucus secretion from submucosal glands in the lungs. Mucus is an aqueous solution that includes electrolytes, mucins (large glycoproteins), enzymes, and antibacterial agents (Rogers 2001), and is beneficial in airway defense and for trapping particles. Particles are then removed along with the mucus by ciliary clearance into the mouth and esophagus.

Both constitutive and induced release of mucus occur in vivo (Gallagher et al. 1975) and in isolated glands in vitro (Baker et al. 1985; Dwyer et al. 1992; Gallagher et al. 1975). Constitutive mucus release does not depend on cholinergic nerves, since neither tetrodotoxin, which prevents action potentials in nerves by blocking sodium channels, nor vagotomy change baseline mucus secretion (Baker et al. 1985; Borson et al. 1984; Gallagher et al. 1975). However, vagal stimulation and exogenous acetylcholine both increase mucus secretion from submucosal glands (Borson et al. 1984; Gallagher et al. 1975). Acetylcholine-induced mucus release is rapid and transient, lasting only 2–6 min, followed by a relative refractory period where further mucus cannot be released by additional acetylcholine exposure. This could be due to receptor desensitization or acetylcholine may initiate a slower inhibitory response along with the secretory response (Dwyer et al. 1992).

Mucus secretion caused by vagal stimulation or exogenous acetylcholine is blocked by the nonselective muscarinic antagonist atropine (Borson et al. 1984; Gallagher et al. 1975). In submucosal glands, muscarinic receptors are found on both serous cells that secrete fluid and mucous cells that secrete mucins (Mak and Barnes 1990; Ramnarine et al. 1996; van Koppen et al. 1988). Both M₁ and M₃ receptors are present in human and animal submucosal glands (Mak and Barnes 1990; Mak et al. 1992). M₃ receptors are responsible for both vagal and exogenous acetylcholine-induced mucin secretion. This is supported by experiments that use selective antagonists (4-DAMP, methoctramine, and telenzepine) at concentrations that only block high-affinity binding sites (Ramnarine et al. 1996). Despite their presence, a direct role for M₁ receptors has not been demonstrated in airway submucosal glands, but it has been hypothesized that these receptors may be responsible for fluid or electrolyte release by serous cells (Yang et al. 1988).

2.4 Muscarinic Receptors on Pulmonary Arteries

While neuronal acetylcholine does not contribute to resting tone in pulmonary blood vessels, stimulation of the vagus nerves causes vasodilation (Laitinen et al. 1987). However, exogenous acetylcholine will only relax precontracted human pulmonary arteries if the endothelium is intact (Greenberg et al. 1987). Acetylcholine likely acts at muscarinic receptors on endothelial cells to stimulate production of nitric oxide, which relaxes smooth muscle (Furchgott and Zawadzki 1980; Greenberg et al. 1987; McMahon and Kadowitz 1992).

M₃ muscarinic receptors are important for vasodilation in vivo. This is supported by the inability of electrical stimulation of the vagus nerves in M₃ receptor-deficient mice to maximally decrease blood pressure (Fisher et al. 2004). Endothelial cells isolated from pulmonary trunk arteries in pigs express mRNA for both M₂ and M₃ receptors. Moreover, acetylcholine increases intracellular calcium concentrations in these cells in a manner consistent with M₃ receptor activation (Kummer and Haberberger 1999). Arterial smooth muscle cells may also contain M₂ and M₃ muscarinic receptors, however this needs to be confirmed because the evidence is based only on immunohistochemistry (Kummer and Haberberger 1999).

2.5 Muscarinic Receptors on Airway Epithelium

Activating muscarinic receptors in epithelial cells transiently increases intracellular calcium (Salathe et al. 1997) and increases ciliary beat frequency (Klein et al. 2009; Salathe et al. 1997; Seybold et al. 1990), which would increase transport of mucus and particulates out of the lung. Muscarinic signaling also increases the velocity of liquid (Seybold et al. 1990) and particle transport (Klein et al. 2009) upward in isolated tracheas.

M₃ receptor mRNA is found in human airway epithelium by in situ hybridization (Mak et al. 1992), while mRNA for both M₃ and M₁ receptors has been identified in mouse epithelia (Klein et al. 2009). Experiments using muscarinic receptor gene-deficient mice demonstrate that M₃ receptors are both required and sufficient for the full increase in ciliary beat frequency and particle transport speed induced by muscarine in wild-type mice. A role for M₃ receptors is further supported by pharmacological experiments in sheep showing that an antagonist with selectivity for M₃ receptors (4-DAMP) blocks acetylcholine-induced calcium signaling and ciliary beat frequency, while an antagonist with selectivity for M₁ receptors (pirenzepine) does not have this effect (Salathe et al. 1997).

While M₃ receptors provide the dominant control of ciliary beat frequency, M₁ and M₂ receptors can also contribute. M₁ receptors increase ciliary transport speed, but this function is only uncovered in mice that are deficient for both M₂ and M₃ receptors. M₂ receptor activation prevents increases in ciliary beat frequency initiated by M₁ receptors and also by nonmuscarinc stimuli such as ATP. Inhibition of ciliary beat frequency mediated by M₂ receptors is likely indirect, since M₂ mRNA and protein are not detectable in epithelial cells but are found in neighboring cells (Klein et al. 2009).

2.6 Muscarinic Receptors and Immune Responses

A functional role for muscarinic receptors in immune responses has been demonstrated in lung mast cells, alveolar macrophages, and airway epithelial cells.

Experimentally, anti-IgE antibodies and calcium ionophore both evoke histamine release from mast cells, an effect that is blocked by acetylcholine and other muscarinic agonists in isolated human bronchi (Reinheimer et al. 1997, 2000; Wessler et al. 2007). The role of inhibitory muscarinic receptors in human mast cells is confirmed since atropine (nonselective muscarinic antagonist) blocks the ability of acetylcholine to inhibit evoked histamine release (Reinheimer et al. 1997). Pharmacological data suggest this is mediated by M₁ receptors. In rats, however, acetylcholine enhances rather than inhibits evoked histamine release (Reinheimer et al. 2000).

Alveolar macrophages phagocytose foreign substances and initiate immune responses against invading pathogens. Acetylcholine induces release of leukotriene B₄ and other factors from alveolar macrophages that induce human peripheral blood monocyte, neutrophil, and eosinophil chemotaxis, and M₃ receptor antagonists prevent acetylcholine-induced release of chemotactic activity from macrophages (Reinheimer et al. 1998). Additionally, acetylcholine may also contribute to inflammation by inducing release of chemotactic factors from airway epithelial cells (Koyama et al. 1992, 1998).

2.7 Muscarinic Receptors and Airway Remodeling

Airway remodeling describes measureable changes in airway structure that occur as a pathological feature of lung diseases such as asthma and COPD. Acetylcholine may contribute to airway remodeling by acting at muscarinic receptors to increase proliferation of both fibroblasts and smooth muscle cells. In primary cultures of human fibroblasts and fibroblast cell lines, acetylcholine stimulates collagen production and proliferation through MAPK activation (Haag et al. 2008; Matthiesen et al. 2006, 2007; Pieper et al. 2007). While acetylcholine does not directly increase smooth muscle cell proliferation, it enhances proliferation induced by growth factors, including platelet-derived growth factor and epidermal growth factor (Gosens et al. 2003; Krymskaya et al. 2000). Muscarinic receptor antagonists block the proliferative effects of acetylcholine in both fibroblast and smooth muscle cells.

Human lung fibroblasts contain mRNA for M₁, M₂, and M₃ muscarinic receptors with trace levels of M₄ receptors (Haag et al. 2008; Matthiesen et al. 2006). It is likely that M₂ receptors are dominant since the proliferative response in fibroblasts is pertusis toxin sensitive and can be blocked with selective muscarinic antagonists that suggest M₂ receptors are responsible (Matthiesen et al. 2006). Acetylcholine-enhanced proliferation of human airway smooth muscle cells is M₃ receptor-dependent and is lost when M₃ receptor expression is decreased with cell passage in vitro (Gosens et al. 2003).

2.8 Muscarinic Receptors in Normal Lung Function

The contribution of M₁, M₂, and M₃ receptors to pulmonary physiology is summarized in Table 1. Most lung tissues express more than one muscarinic receptor subtype, but the function of one muscarinic subtype is often dominant. Where the functions of additional muscarinic receptor subtypes are known, they either inhibit or supplement the dominant receptor’s function. For example, M₂ receptors on postganglionic parasympathetic nerves inhibit acetylcholine release, and this function is inhibited by M₁ receptors in ganglia, which increase acetylcholine release by facilitating neurotransmission. In airway smooth muscle, M₂ receptors supplement contraction mediated via M₃ receptors. In healthy individuals, this muscarinic physiology is balanced and results in airway smooth muscle tone, vasodilation, mucus secretion, and mucociliary clearance.

Table 1.

Function of muscarinic receptor subtypes in lung

	M1	M2	M3
Parasympathetic nerves	Increase neurotransmission at ganglia	Limit acetylcholine release
Smooth muscle		Inhibit relaxationa	Contraction
Submucosal glands	Unknown		Mucus secretion
Endothelial cells		Unknown	Vasodilationa
Airway epithelium	Increase ciliary beat frequencya (if M2 and M3 blocked)	Reduce ciliary beat frequencya	Increase ciliary beat frequency
Immune function	Limit evoked histamine release from mast cells		Induce release of chemotactic factors from alveolar macrophagesa
Airway remodeling		Increase proliferation in fibroblasts	Enhance proliferation induced by growth factors in smooth muscle

Receptor subtype involvement is based on human data unless noted with ^a. Shading indicates physiological functions that are especially important in COPD and asthma, lung diseases characterized by airflow limitation

However, in obstructive lung diseases some muscarinic receptor functions contribute to disease symptoms. For example, excessive bronchoconstriction and increased mucus secretion limit airflow in asthma. Acetylcholine released by the vagus nerves onto M₃ receptors mediates both of these physiological functions. Blocking M₃ receptors is therefore therapeutically very beneficial for reducing symptoms and improving lung function. Conversely, blocking inhibitory M₂ receptors on parasympathetic nerves is counterproductive since this increases acetylcholine release, resulting in increased bronchoconstriction and mucus secretion.

3 Effects of Therapeutic Muscarinic Antagonists in Lung Disease

Muscarinic antagonists are used therapeutically as bronchodilators to treat both COPD and asthma, lung diseases characterized by airflow obstruction and underlying airway inflammation.

In COPD, airflow limitation is not fully reversible, and is caused by structural changes and narrowing of peripheral airways along with parenchymal destruction (GOLD 2009). Muscarinic antagonists increase airflow in COPD by blocking cholinergic tone at airway smooth muscle. However, asthma is different in that airflow limitation is generally fully reversible and caused by bronchoconstriction. In more severe asthma, edema due to mucus hypersecretion also contributes to airflow limitation. The airways are hyperresponsive and bronchoconstrictor responses are exaggerated (NHLBI 2007). Muscarinic antagonists increase airflow in asthma by blocking cholinergic tone and also by blocking reflex bronchoconstriction mediated by the vagus nerves. They may also inhibit secretion and clearance of mucus.

3.1 Therapeutic Muscarinic Receptor Antagonists

Atropine and other naturally occurring muscarinic receptor antagonists found in plants of the Datura genus have been effectively used as bronchodilators for centuries. In western medicine, the leaves and roots of D. stramonium were administered in cigarettes to treat respiratory diseases starting in the 1800s (Gross and Skorodin 1984). However, while atropine is an effective bronchodilator, its use is associated with side effects. Therefore, when beta adrenoreceptor agonists, which directly relax airway smooth muscle by stimulating β₂ receptors became available they largely replaced atropine. Since then, however, synthetic derivatives of atropine have been developed that contain a quaternary ammonium. This next generation of drugs, which include ipratropium and tiotropium, have limited bio-availability and are unable to cross the blood–brain barrier, and thus have fewer side effects. They are currently administered by inhalation to treat both COPD and asthma. Atropine, ipratropium, and tiotropium are all competitive antagonists (Casarosa et al. 2009), and thus contribute to bronchodilation primarily by blocking acetylcholine binding to M₃ receptors on airway smooth muscle. The pharmacological properties of atropine, ipratropium, and tiotropium are discussed below and summarized in Table 2.

Table 2.

Comparison of binding affinities and duration of binding for atropine, ipratropium, and tiotropium at human muscarinic receptors

		Atropine	Ipratropium	Tiotropium
Ki (nM)a	M1	0.170	0.398	0.016
	M2	0.339	0.295	0.020
	M3	0.209	0.263	0.010
	M4	0.107	0.224	0.010
	M5	0.316	0.851	0.110
Dissociation half-life (h)a, b	M1		0.10	10.5
	M2		0.03	2.6
	M3	0.04	0.22	27.0

^aAntagonist affinities determined in heterologous competition binding experiments against [³H] NMS. Dissociation kinetics using Motulski and Mahan method (Casarosa et al. 2009)

^bDowling and Charlton (2006)

3.1.1 Atropine

Atropine is a nonselective muscarinic antagonist with similar affinities for all five muscarinic receptor subtypes (Casarosa et al. 2009). Relative to the quaternary ammonium derivatives, atropine is also well absorbed across the gastrointestinal tract into systemic circulation. Total absorption of atropine across the intestine is approximately 25% in rat (Levine 1959), while bioavailability following intramuscular injection in humans is reported to be 50% (Goodman et al. 2006). As a result, atropine has many undesirable side effects including at low doses dry mouth, urinary retention, and accelerated heart rate. In addition, atropine is also able to cross the blood–brain barrier (Virtanen et al. 1982). Thus, at high doses side effects include coma, fever, and hallucinations.

3.1.2 Ipratropium Bromide

Ipratropium bromide is a quaternary ammonium derivative of atropine used clinically as a second-line bronchodilator behind β₂-agonists. It was also the first muscarinic antagonist widely used to treat COPD. Like atropine, ipratropium is nonselective and has similar affinities for all five muscarinic receptor subtypes (Casarosa et al. 2009). The major differences between ipratropium and atropine are the inability of ipratropium to cross the blood–brain barrier and its poor absorption in the gastrointestinal tract. Ipratropium is better absorbed when administered by inhalation (Ensing et al. 1989), which may be due to uptake by organic cation/carnitine transporters (OCTN) in airway epithelium. OCTN2, and to a lesser extent OCTN1, transport both ipratropium and tiotropium in a human bronchial epithelial cell line (Nakamura et al. 2010). Ipratropium produces peak bronchodilation within 60–90 min of inhalation and its duration of action is 4–6 h, requiring four times daily administration.

3.1.3 Tiotropium Bromide

Like ipratropium, tiotropium bromide also contains a quaternary ammonium. However, tiotropium has a much higher affinity for muscarinic receptors and a much longer duration of binding to muscarinic receptors than either atropine or ipratropium (see Table 2). However, tiotropium’s most interesting property is its significantly greater duration of binding to M₁ and M₃ receptors than M₂ receptors, which provides tiotropium with kinetic selectivity for these receptors (Casarosa et al. 2009; Disse et al. 1993). Functionally, tiotropium blocks M₂ receptors on parasympathetic nerves early after administration to increase acetylcholine release. However, following washout, neuronal acetylcholine release returns to baseline within 2 h, a time point when smooth muscle contraction via M₃ receptors is still completely blocked. M₃ receptor function only begins to return after 7 h (Takahashi et al. 1994). Tiotropium’s onset of bronchodilation in humans is very slow, reaching peak bronchodilation in 3–4 h, but tiotropium then has a very long duration of action (1–2 days) and can be administered daily (Maesen et al. 1995). The slow onset of action makes tiotropium inappropriate for a rescue medication, but the duration of action makes it useful as a once-daily bronchodilator.

3.2 Therapeutic Use of Muscarinic Receptor Antagonists in COPD

In COPD patients, airflow is limited by destructive and fibrotic changes in the lungs that narrow the airways. These changes are not reversible, but some bronchodilation can be achieved by blocking cholinergic tone. Because of the limited treatment options for COPD, bronchodilators are central to the management of symptoms. Cholinergic tone may be higher in patients with COPD than in healthy patients and is effectively reversed with muscarinic receptor antagonists (Gross et al. 1989). Ipratropium is currently recommended for use as a four times daily short-acting bronchodilator by the Global Initiative for Chronic Obstructive Lung Disease 2009 global strategy for diagnosis, management, and prevention of COPD (GOLD 2009). Tiotropium is recommended for use as a once-daily long-acting bronchodilator. In COPD patients, a single inhaled dose (10–80 μg) of tiotropium results in a dose-dependent 19–26% improvement in the volume of air that is exhaled during the first second of forced exhalation (FEV₁) (Maesen et al. 1995). However, once steady-state plasma concentrations are reached following multiple once-daily dosings (4.5–36 μg), higher doses add little additional improvement. Thus, low doses with limited adverse side effects can be effectively used. Based on these data repeated daily dosing with the recommended 18 μg leads to continued bronchodilation (Littner et al. 2000; van Noord et al. 2002).

In the 4-year UPLIFT randomized, double-blind, placebo-controlled trial, 5,993 COPD patients were treated with either tiotropium or a placebo control. Tiotropium improved airflow (as measured by FEV₁), improved health-related quality of life scores, significantly delayed onset of exacerbations and associated hospitalizations, and reduced respiratory failure (Tashkin et al. 2008). These results are consistent with results from previous smaller and shorter studies (Casaburi et al. 2000; Dusser et al. 2006; Niewoehner et al. 2005). However, COPD is a progressive disease, and while tiotropium remained efficacious over the study period it was not able to significantly slow the rate of decline in mean FEV₁ (Tashkin et al. 2008). This is not unexpected, since only cessation of smoking has been shown to reduce this decline in patients with COPD (Anthonisen et al. 1994).

3.3 Therapeutic Use of Muscarinic Antagonists in Asthma

Asthma is characterized by inflammation and airway hyperresponsiveness, which is defined as excessive bronchoconstriction to contractile stimuli. There is no correlation between contractile responses of bronchial smooth muscle isolated from asthma and nonasthma patients and methacholine responsiveness in these same patients in vivo (Roberts et al. 1984; Whicker et al. 1988). Thus airway hyperresponsiveness in asthma is not simply due to increased smooth muscle sensitivity to contractile agents. However, maximum contractile responses in tracheal smooth muscle are greater in tissues from humans who died of fatal asthma than controls (Bai 1990; Haddad et al. 1996). It is unclear whether these results reflect the use of tracheal tissue instead of bronchi or are unique to fatal asthma patients. Since airway hyperresponsiveness occurs in vivo where vagal reflexes are present but not in vitro where reflexes are absent, this supports the role of parasympathetic nerves in airway hyperresponsiveness in asthma patients.

The contribution of parasympathetic nerves to airway hyperresponsiveness is further supported by published reports of surgical treatment in humans with severe asthma, where autonomic and sensory nerves supplying the lung were severed. These uncontrolled studies show improvements in 50% or more of the patients (Balogh et al. 1957; Overholt 1959; Phillips and Scott 1929). Dimitrov-Szokodi et al. denervated the lungs in 19 patients. Prior to surgery, asthma attacks were actually induced in eight patients with histamine. In these patients, histamine-induced asthma attacks ceased when neurotransmission in autonomic and sensory nerves was blocked with novocaine administered in the neck. Surgeries to denervate the airways were then carried out in all patients, and the cut vagus nerves were carefully sutured to prevent reinnervation. Following surgical denervation of the airways, exogenous histamine no longer induced asthma attacks. Of the 19 patients treated, 10 no longer needed any pharmacological treatment for their asthma, 7 had improved symptoms that could easily be controlled with drugs, and only 2 patients were not improved. In addition, denervation of the airways altered airway inflammation by reducing or abolishing eosinophils in sputum and blood (Balogh et al. 1957).

However, pharmacological evidence supporting a role for parasympathetic nerves and vagal reflexes in asthma was controversial for many years. Numerous studies in humans showed limited or no benefit of muscarinic receptor antagonists in their ability to block bronchoconstriction induced by nonspecific stimuli such as histamine, sulfur dioxide, exercise, cold air, and antigens (Casterline et al. 1976; Chan-Yeung et al. 1971; Cockcroft et al. 1978; Fish et al. 1977; Fisher et al. 1970; Nadel et al. 1965; Rosenthal et al. 1977; Ruffin et al. 1978; Woenne et al. 1978). The majority of these studies used a single dose of inhaled muscarinic antagonist, which was chosen because it effectively blocked either cholinergic tone or bronchoconstriction induced by inhaled methacholine. Conversely, other studies showed that muscarinic antagonists are effective, and they were able to inhibit bronchoconstriction induced by these same stimuli (Chan-Yeung 1977; Chen et al. 1981; Holtzman et al. 1980; Nadel et al. 1965; Sheppard et al. 1982; Widdicombe et al. 1962; Yu et al. 1972). These discrepancies are due to the dose of muscarinic antagonist administered, the degree of bronchoconstriction induced, and the method by which the antagonist is administered, all of which contribute to the degree of blockade of bronchoconstriction induced by vagal reflexes.

In acute asthma, the dose of muscarinic antagonist administered is very important because of the competitive nature of therapeutic antagonists. Doses of muscarinic receptor antagonists that only block or reduce airway tone may be ineffective for inhibiting asthma attacks when acetylcholine concentrations are increased. For example, in asthma patients low doses of inhaled atropine inhibit baseline cholinergic tone and prevent bronchoconstriction induced by inhaled methacholine. However, much higher doses of atropine are required to inhibit cold air-induced bronchoconstriction that is mediated by the vagus nerves. In one study, cold air-induced bronchoconstriction could be abolished in five and reduced in two of seven patients with higher doses of atropine. In these two patients, atropine could abolish cold air-induced bronchoconstriction when the level of cold air-induced bronchoconstriction was decreased by reducing the exposure time (Sheppard et al. 1982). Thus, effective blockade can be achieved by increasing the muscarinic antagonist concentration or decreasing the agonist challenge (in this case cold air-induced acetylcholine release). Similarly, when increasing doses of ipratropium were given by nebulization to patients admitted to the hospital with acute asthma, 500 μg were required to achieve maximum bronchodilation, presumably via blockade of endogenous acetylcholine from vagus nerves. This is a ten times greater dose than the 40–80 μg dose (from a metered dose inhaler) that blocks vagal cholinergic tone and bronchoconstriction induced by inhaled exogenous methacholine (Baigelman and Chodosh 1977; Cockcroft et al. 1978). Thus, it is not surprising that the lower dose (80 μg) used in early studies had very little benefit in acute asthma (Cockcroft et al. 1978; Ruffin et al. 1978).

The method by which muscarinic receptor antagonists are administered is also important for achieving complete vagal blockade in animals and humans (Holtzman et al. 1983; Sheppard et al. 1983). In dogs, intravenous atropine blocks bronchoconstriction induced by inhaled acetylcholine or electrical stimulation of the vagal nerves equally well. This is in contrast to atropine administered by inhalation, which blocks bronchoconstriction induced by inhaled acetylcholine at significantly lower doses than are required to block bronchoconstriction induced by vagal stimulation (Holtzman et al. 1983). This study shows that intravenous administration of muscarinic receptor antagonists results in effective blockade of bronchoconstriction regardless of whether the agonist is inhaled or released by nerves, whereas inhaled antagonists, which are deposited in similar sites in the airway as inhaled agonists, are not able to block neuronal acetylcholine as effectively. Similar results are also found in humans (Sheppard et al. 1983).

One of the best-understood mechanisms for airway hyperresponsiveness in asthma is loss of inhibitory M₂ receptor function on the parasympathetic nerves (Ayala and Ahmed 1989; Minette et al. 1989). Loss of negative feedback through M₂ receptors in the efferent half of vagal reflexes leads to increased acetylcholine release and excessive bronchoconstriction to diverse stimuli.

In animals, airway hyperreactivity is due to M₂ receptor dysfunction following antigen challenge (Fryer and Wills-Karp 1991), ozone exposure (Schultheis et al. 1994), and viral infection (Fryer and Jacoby 1991), and it is closely associated with airway inflammation. Eosinophils, which are inflammatory cells associated with asthma, are clustered around the nerves in airways of sensitized guinea pigs and humans who have died of fatal asthma (Costello et al. 1997). Following antigen challenge, eosinophils are recruited to airway nerves (Fryer et al. 2006) and activated to release an endogenous M₂ receptor-selective antagonist, major basic protein (Jacoby et al. 1993). M₂ receptor dysfunction and hyperreactivity mediated by the vagus nerves are prevented by eosinophil depletion (Elbon et al. 1995), and by neutralizing eosinophil major basic protein or removing it from M₂ receptors (Evans et al. 1997; Fryer and Jacoby 1992). Eosinophils also mediate M₂ receptor dysfunction following ozone exposure (Yost et al. 1999) and virus infection in sensitized guinea pigs (Adamko et al. 1999). In addition, M₂ receptor dysfunction on parasympathetic nerves also occurs through eosinophil-independent mechanisms. In the absence of antigen sensitization, viral neuraminidases reduce agonist affinity for M₂ receptors by removing sialic acid. The muscarinic agonist, carbachol, has tenfold lower affinity for M₂ receptors following desialation with neuraminidase (Fryer et al. 1990). Additionally, interferon-γ and tumor necrosis factor-α, cytokines produced during the inflammatory response to virus or inhaled antigen, reduce M₂ receptor gene expression resulting in decreased function (Jacoby et al. 1998; Nie et al. 2009). The mechanisms by which neuronal M₂ receptor function is lost in humans with asthma are not known, but the increased association of eosinophils with nerves in the lungs of humans who died of fatal asthma (Costello et al. 1997) suggests a role for eosinophils.

In humans, muscarinic receptor antagonists have been shown to provide significant bronchodilation in virus-induced asthma (Aquilina et al. 1980; Empey et al. 1976), allergic asthma (Yu et al. 1972), exercise-induced bronchospasm (Borut et al. 1977; Godfrey and Konig 1975), nocturnal asthma (Catterall et al. 1988; Morrison et al. 1988), and psychogenic asthma (McFadden et al. 1969; Rebuck and Marcus 1979). Muscarinic receptor antagonists are effective at blocking vagally induced bronchoconstriction and decreasing tone but are less effective for blocking direct effects of noncholinergic agents on airway smooth muscle. Current guidelines for asthma management recommend β₂-agonists be used as first-line bronchodilators, and ipratropium be used in combination with short-acting β₂-agonists in moderate-to-severe asthma exacerbations. Ipratropium is also recommended as the treatment of choice for bronchospasm due to beta-blocker medications (NHLBI 2007).

A double-blind, randomized, prospective trial compared the use of short-acting β₂-agonists and β₂-agonists combined with high doses of ipratropium (504 μg per hour for 3 h) in 180 patients admitted to the emergency department with acute asthma. Patients who received both ipratropium and β₂-agonist had a 48.1% greater improvement in FEV₁ than those who received β₂-agonist alone. Additionally, they had a 49% reduction in the risk of hospital admission. Patients with severe airway obstruction and patients with a longer duration of symptoms were more likely to benefit from the addition of ipratropium (Rodrigo and Rodrigo 2000). Similarly, a meta-analysis of 32 randomized controlled trials in children and adults with acute asthma showed that use of muscarinic receptor antagonists along with β₂-agonists significantly improved airflow measurements and decreased the risk of hospital admissions by 30% when compared to treatment with β₂-agonists alone. Muscarinic receptor antagonists were particularly beneficial in patients with moderate to severe obstruction, and there was a greater bronchodilation benefit when patients were treated with more than one dose of muscarinic receptor antagonist (Rodrigo and Castro-Rodriguez 2005).

Muscarinic receptor antagonists are not recommended for long-term management of stable asthma because β₂-agonists inhibit bronchoconstriction and corticosteroids effectively inhibit inflammation (NHLBI 2007). However, studies in animals suggest that tiotropium is as effective as budesonide (corticosteroid) in inhibiting several aspects of allergen-induced airway remodeling and inflammation including smooth muscle thickening, mucous gland hypertrophy, and eosinophilia in airway tissue (Bos et al. 2007; Gosens et al. 2005). In addition, a small study in humans with severe asthma found that tiotropium bromide administered daily for 4 weeks improved FEV₁, and this was especially true for patients who had a large proportion of neutrophils in their sputum (Iwamoto et al. 2008). This may be due to muscarinic receptor antagonist inhibition of neutrophil recruitment into lungs, since tiotropium inhibits acetylcholine-induced release of chemotactic factors for neutrophils from human alveolar macrophages in vitro (Buhling et al. 2007). Together these data suggest that muscarinic receptor antagonists such as tiotropium may have additional anti-inflammatory benefits that need to be researched further and could be exploited in the future.

4 Conclusions

In the lungs, acetylcholine released from parasympathetic nerves provides the dominant control over airway smooth muscle tone. Muscarinic receptors found on glands, airway smooth muscle, and nerves control airway tone and mucus secretion. Additionally, nonneuoronal acetylcholine stimulates muscarinic receptors on epithelial cells and endothelial cells to increase ciliary beat frequency and cause vasodilation. Parasympathetic nerves act as the efferent arm in vagal reflexes to various chemical and physical stimuli, and in asthma changes in neuronal M₂ receptor function contribute to airway hyperresponsiveness. Muscarinic receptor antagonists are currently used as bronchodilators to treat airflow limitation in COPD and asthma and recent data suggest they may also be useful for treating airway remodeling.

Abbreviations

ChAT

Choline acetyltransferase

COPD

Chronic obstructive pulmonary disease

FEV₁

Forced expiratory volume in 1 s