INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Chronic obstructive pulmonary disease (COPD) is a clinical condition characterized by a progressive airflow obstruction. Although this process is irreversible, therapy with bronchodilator agents improves lung function in many patients. Unlike the situation in asthma, patients with COPD achieve an equal, if not better, bronchodilator response when treated with anticholinergic drugs compared with -agonists (1, 2).

Anticholinergic drugs act by inhibiting the neurotransmitter acetylcholine which is released from the parasympathetic nerves in the vagus. These nerves provide the dominant innervation of the airway smooth muscle and cause its contraction by releasing acetylcholine onto M₃ muscarinic receptors (3). Even under resting conditions the vagus nerves maintain a baseline tonic contraction of the airway smooth muscle. Acetylcholine also binds to M₂ muscarinic receptors on postganglionic cholinergic nerves, stimulation of which limits the magnitude of vagally mediated bronchoconstriction (4, 5).

Previously, we and others have shown in animal models that there is dysfunction of M₂ muscarinic receptors and that this leads to vagally mediated hyperreactivity (6). The function of pulmonary neuronal M₂ receptors has not been investigated in subjects with COPD, although there are several observations that suggest that they may be dysfunctional in this condition. For example, recently it has been shown that there is an age-related loss of function of M₂ receptors in the heart (9). Experimental data in animals indicate that factors that contribute to symptoms in subjects with COPD, such as eosinophils and viruses, inhibit the function of these receptors (7, 8). Thus, there is a considerable amount of supporting data to suggest that there is loss of function of M₂ muscarinic receptors in subjects with COPD. Because M₂ receptors limit the magnitude of vagally induced bronchoconstriction, dysfunction of these receptors may be a mechanism whereby anticholinergic agents could act as effective therapy in patients with COPD.

In this study we have tested the hypothesis that there is loss of function of pulmonary neuronal M₂ muscarinic receptors and an associated increase in vagally induced bronchoconstriction in patients with clinically stable COPD. To do this we have used a nasal cold dry air challenge to induce a vagally mediated bronchoconstriction. Unlike other methods of stimulating the vagus nerves, we and others have shown that this challenge produces a vagally mediated bronchoconstriction that is easily detected in healthy control subjects (10, 11). This means that M₂ receptor function can be tested in vivo in humans and that the magnitude of vagally induced bronchoconstriction can be directly compared between patients with airways diseases and healthy control subjects.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subject Recruitment

Patients under hospital follow-up with COPD were recruited for the study. COPD was defined physiologically by spirometry criteria as FEV₁ < 80% predicted and FEV₁/FVC ratio < 70% without significant day-to-day variation (12). All subjects were breathless on exertion and reported having an intermittent or persistent cough and wheeze. All were current or former smokers with a greater than 20 pack-years smoking history, and all used ₂ receptor agonists for symptomatic relief. Subjects with a personal or family history of asthma and those with atopy or hay fever were excluded from the study, as were those with other chest diseases, including ₁ antitrypsin deficiency. No subject had a clinical history of a COPD exacerbation, and none had taken antibiotics or oral corticosteroids for at least 6 wk before enrolment into the study. Two control groups were also studied; one was a group of healthy nonsmokers with normal lung function and the second a group of current smokers with normal lung function. These subjects were enrolled by advertisement in the hospital and all were blinded to the purpose of the study. All participants gave written informed consent before any tests were performed; the study protocol was approved by the local research ethics committee.

Measurement of Airway Resistance (Raw) and Spirometry

Raw was measured using the technique of impulse oscillometry (Master Screen Impulse Oscillometry; Erich Jaeger, Hoechberg, Germany). This involves the application of small pressure oscillations at the mouth by an external generator and the recording of reflected oscillation pressure and flow. Raw at 5 Hz (R5) measured using this technique provides a measure of total Raw and is performed noninvasively without having the subject perform specific respiratory maneuvers, in particular forced maneuvers that require deep inhalation, because these may cause bronchodilation. Measurements were made during tidal breathing with the subjects seated. The mean value of two measurements was recorded. Spirometry was performed using a pneumotachograph attached to the impulse oscillometry system (Master Screen Impulse Oscillometry; Erich Jaeger). Subjects were instructed to breathe in rapidly and breathe out as forcibly as possible without an end-inspiratory pause, the better of two reproducible measures being recorded.

Nasal Cold Dry Air Challenge

A domestic freezer was adapted for the nasal cold dry air challenges, as described previously (13) (designed and manufactured by Mr. J. Haycock, Ormskirk, Lanchashire, UK). In brief, a suction pump was attached to the side of the freezer, room air was drawn over a column of anhydrous calcium chloride and then circulated through a metal circuit designed to increase the surface area of the freezer. The cold dry air was vented through a second port in the freezer through an insulated tube (length 50 cm) to a silicone nasal mask (ResMed Ltd., Oxfordshire, UK) at a flow rate of 30 L/min. The temperature of air emitted from the freezer was recorded before and after each challenge and was maintained at 10° C.

The nasal cold dry air challenge was performed in the following manner: the subject inhaled cold dry air for 5 min through the nasal mask and then exhaled through the mouth. Breathing frequency was maintained at a rate of 15 breaths/min by the use of a metronome. Subjects were shown the ventilation maneuvers before the challenge was commenced. The conditions chosen for the challenge were based on the results of prior studies (10, 11).

Study Design

At an initial visit the subjects completed a questionnaire detailing respiratory health, medication use, as well as occupational and smoking history. A blood eosinophil count and skin-prick tests against five common aeroallergens (Biodiagnostics, Worcestershire, UK) were performed.

The subjects attended the laboratory on three separate occasions within 1 wk and were instructed to withhold all inhaled medications and to abstain from caffenated beverages for at least 12 h before any of the airway challenges. The rationale for each intervention used in each of these visits is shown in Figure 1. At each visit Raw and spirometry values were recorded. The subjects were then pretreated in random order with either placebo (saline 0.9%), or with the nonselective muscarinic receptor antagonist ipratropium bromide (500 µg), or with the selective M₂ muscarinic receptor agonist pilocarpine (5 mg/ml). A control group of age-matched smokers with normal lung function attended on one occasion for a nasal cold dry air challenge to assess the magnitude of vagal reflex-induced bronchoconstriction.

View larger version (23K): [in this window] [in a new window]   Figure 1.   The experimental design and rationale for the drug treatments administered to the subjects before the nasal cold dry air challenge.

Nasal cold air challenge to assess the magnitude of cold air-induced bronchoconstriction. After measurement of Raw the subjects were pretreated with saline 0.9% delivered through a dosimeter (Spira Elektro 2 Inhalation Dosimeter, Haemnlinna, Finland). They inhaled from FRC to total lung capacity at a flow rate of 0.6 L/s and after inhalation of 250 ml the nebulization was triggered for 2.5 s. When the subjects reached total lung capacity they were instructed to breath-hold for 10 s. These maneuvers were repeated five times. The Raw was measured again. Then the subjects inhaled cold dry air for 5 min and Raw was repeated immediately afterwards.

Nasal cold air challenge to assess the cholinergic dependence of nasal cold air-induced bronchoconstriction. After measurement of Raw the subjects were pretreated with the nonspecific muscarinic receptor antagonist ipratropium bromide (500 µg), delivered through a nebulizer. Forty minutes later Raw and spirometry were rerecorded. The subjects then inhaled cold dry air for 5 min and the measurement of Raw was repeated immediately afterward.

Nasal cold air challenge to assess M₂ receptor function. After measurement of Raw the subjects were pretreated with pilocarpine (5 mg/ml) delivered using a dosimeter, as described previously. In some cases bronchoconstriction occurred owing to the effect of pilocarpine on M₃ muscarinic receptors. Raw was recorded at 5-min intervals until the bronchoconstriction induced by pilocarpine stabilized (values returned to within at least 10% of baseline and were similar on two readings 5 min apart). If more than 30 min elapsed after the inhalation of pilocarpine and the Raw had not stabilized, the cold dry air challenge was not performed. In those subjects who stabilized a cold dry air challenge was performed and Raw was recorded immediately afterward.

Using a histamine challenge to induce a vagal reflex and study M₂ muscarinic receptor function. The magnitude of vagal reflex response to histamine and the function of M₂ receptors were also tested using a histamine challenge in 10 of the subjects with COPD (subject characteristics are shown in Table 1). The rationale for these tests is that histamine causes bronchoconstriction in humans in part through an indirect, vagally mediated effect on airway smooth muscle (14, 15). In these studies the subjects were pretreated with ipratropium bromide (500 µg/ml) dissolved in sterile saline to demonstrate that there was a reflex response to histamine in subjects with COPD, and on other study days they were pretreated with pilocarpine (5 mg/ml) or a placebo (0.9% saline). The subjects were then administered saline 0.9%, to establish baseline spirometry values, and then administered doubling doses of histamine through a dosimeter until the FEV₁ decreased by 20% of the postsaline value. The geometric mean concentration of histamine causing the FEV₁ to fall by 20% (PC₂₀) was calculated by linear interpolation.

Statistics

Baseline values of subject characteristics, spirometry, and Raw values at 5 Hz (R5) were compared using analysis of variance (ANOVA) with Fisher's correction for multiple comparisons. Correlation between the magnitude of vagally induced bronchoconstriction and spirometric, gas transfer data, M₂ receptor function, and after ipratropium bromide were made by simple regression analysis, using the computer software program StatView 5.0 (SAS Institutes Inc., Chicago, IL) (16). Changes in Raw after treatments were compared using paired t tests. Values are mean ± SEM.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The baseline characteristics of the subjects reported in this study are shown in Table 2. The subjects with COPD had a moderate degree of airflow obstruction (FEV₁ percentage of predicted ranged from 30 to 68%) and none was atopic, as judged by history, allergen skin tests, or blood eosinophil levels. Four subjects with COPD and five subjects with normal lung function did not return for at least one of the three tests owing to altered symptoms after one of the tests (n = 3) or due to an unwillingness to participate in further tests (n = 6).

Effect of a Nasal Cold Air Challenge on Raw

In control nonsmokers the baseline R5 was 0.35 ± 0.03 kPa/L/ s; after saline pretreatment, R5 was 0.34 ± 0.03 kPa/L/s. After the nasal cold dry air challenge R5 increased significantly to 0.39 ± 0.03 kPa/L/s (see Figure 2, open circles; comparing R5 before the cold air challenge with R5 after the challenge p = 0.002, n = 13). In smokers with normal lung function the baseline R5 was 0.41 ± 0.05 kPa/L/s. After saline pretreatment R5 was 0.44 ± 0.06 kPa/L/s and this increased significantly after the nasal cold air challenge to 0.51 ± 0.06 kPa/L/s (Figure 2, open diamonds; comparing R5 before the cold air challenge with R5 after the challenge, p < 0.01, n = 11). In subjects with COPD R5 was 0.68 ± 0.06 kPa/L/s after saline pretreatment and this increased significantly to 0.74 ± 0.07 kPa/L/s after the cold air challenge (Figure 2, open boxes; comparing R5 before the cold air challenge with R5 after the challenge, p < 0.01, n = 22). There was no significant correlation between the percentage change in Raw after the cold air challenge and either the baseline R5 or the percentage of predicted FEV₁. There was no difference in the percentage change in Raw after the cold air challenge between subjects with COPD who were current and ex-smokers. Among current smokers (n = 9) with COPD R5 increased 12.9 ± 4% after the cold air challenge and increased 9.8 ± 3.5% in the former smokers (p = not significant [NS]). In those subjects treated with inhaled corticosteroids Raw increased 8.1 ± 7% and in those not taking this therapy Raw increased to 11.3 ± 4% after the cold air challenge (p = NS).

View larger version (15K): [in this window] [in a new window]   Figure 2.   A nasal cold dry air challenge caused a significant increase in Raw in control nonsmokers (open circles, p < 0.01), in control smokers (open diamonds, p < 0.01), and in subjects with COPD (open boxes, p < 0.01). The group mean (closed symbols) and the individual subject values for each of the groups before (prechallenge) and after (postchallenge) the cold air challenge are shown. The rise in Raw was not different among either of the control subjects compared with the COPD group, which suggested that there may be an absence of vagally mediated hyperreactivity in the subjects with COPD. Data are shown as the Raw measured at 5 Hz.

Effect of Ipratropium Bromide on Nasal Cold Air-Induced Bronchoconstriction

After treatment with ipratropium bromide, R5 decreased from 0.35 ± 0.03 to 0.33 ± 0.04 kPa/L/s in subjects with normal lung function. In these subjects R5 was 0.33 ± 0.04 kPa/L/s after the cold air challenge (Figure 3A, comparing R5 before the challenge with R5 after the challenge, p = NS, n = 9).

View larger version (13K): [in this window] [in a new window]   Figure 3.   Nasal inhalation of cold dry air induced a vagally mediated bronchoconstriction in control (A) and COPD subjects (B). Raw at baseline, after pretreatment with either saline 0.9% (open symbols) or ipratropium bromide 500 µg (filled symbols) and after a nasal cold dry air challenge is shown. Pretreatment with ipratropium bromide prevented the cold air-induced bronchoconstriction, which indicated that the cold air challenge induced a vagally mediated bronchoconstriction. Data are shown as mean Raw measured at 5 Hz.

In subjects with COPD R5 decreased from 0.65 ± 0.05 to 0.54 ± 0.05 kPa/L/s (Figure 3B, compared with baseline R5 p < 0.01, n = 19). After the nasal cold dry air challenge R5 increased only to 0.58 ± 0.06 (comparing R5 before the challenge with R5 after the challenge p = NS). There was no correlation between the baseline R5 and fall in R5 following ipratropium bromide, r = 0.19; p = 0.5. Similarly, there was no correlation between the decline in R5 after ipratropium bromide and the baseline percent predicted FEV₁, r = 0.18, p = 0.4.

Effect of the Muscarinic Receptor Pilocarpine on the Cold Air-Induced Bronchoconstriction

In healthy control subjects R5 increased from 0.37 ± 0.03 to 0.45 ± 0.05 kPa/L/s immediately after pilocarpine treatment (p = 0.03, n = 10). Within 10 min R5 had returned to baseline levels and was 0.38 ± 0.03 kPa/L/s (Figure 4A). In these subjects R5 was 0.39 ± 0.02 kPa/L/s after the cold dry air challenge (comparing R5 before the challenge with R5 after the challenge, p = NS).

View larger version (13K): [in this window] [in a new window]   Figure 4.   Pretreatment with the muscarinic receptor agonist pilocarpine (5 mg/ml) prevented the nasal cold dry air-induced bronchoconstriction in control (A) and COPD subjects (B). Raw at baseline, after pretreatment with either saline 0.9% (open symbols) or pilocarpine 5 mg/ml (filled symbols) and after a nasal cold dry air challenge is shown. Pilocarpine pretreatment prevented the cold air challenge-induced rise in Raw, which indicated normal function of M2 muscarinic receptors. Data are shown as mean Raw measured at 5 Hz.

Pretreatment with pilocarpine also caused a significant increase in Raw in the subjects with COPD; R5 increased from 0.6 ± 0.06 to 0.74 ± 0.04 kPa/L/s (p = 0.01, n = 19). After an average of 10 min (range 5 to 25 min) Raw returned to a stable baseline value of 0.66 ± 0.04 kPa/L/s in 18 subjects. In these subjects R5 was 0.67 ± 0.05 kPa/L/s after the nasal cold dry air challenge (Figure 4B, p = NS). When the differences in the response to cold air were compared between the two study days, there was a significantly greater rise in Raw on the day when the subjects were saline pretreated compared with when they were pilocarpine pretreated (p = 0.02).

Among all subjects there was a weak correlation between the magnitude of cold dry air-induced bronchoconstriction after saline pretreatment (a measure of vagally induced bronchoconstriction) with the magnitude of cold dry air-induced bronchoconstriction when these subjects had been pretreated with pilocarpine (a measure of the effectiveness of M₂ receptors) (Figure 5, r = 0.36; p = 0.07).

View larger version (17K): [in this window] [in a new window]   Figure 5.   The correlation between the magnitude of vagally induced bronchoconstriction and the degree of M2 muscarinic receptor function in both control (open circles) and COPD subjects (filled boxes) is shown. The increase in Raw after the cold air challenge after saline pretreatment represents the degree of vagally induced bronchoconstriction and the increase after pilocarpine represents the function of the M2 receptor. An impaired response to pilocarpine represents a less functional M2 muscarinic receptor. Combining the data from both groups of subjects, r = 0.36, p = 0.07.

Histamine-induced Vagal Reflex Bronchoconstriction

Ten subjects with COPD attended the laboratory for a histamine challenge. The mean FEV₁ after the administration of saline was 1.48 ± 0.01 L, the histamine PC₂₀ for this group was 1.5 ± 0.4 mg/ml. On a separate study day these subjects returned to the laboratory and after pretreatment with ipratropium bromide, FEV₁ increased from 1.54 ± 0.01 to 1.65 ± 0.01 L, p = 0.01. On this occasion the FEV₁ postsaline was 1.57 ± 0.02 L; despite this, bronchodilation histamine PC₂₀ was 1.89 ± 0.26 mg/ml (comparing the PC₂₀ after placebo pretreatment with that after ipratropium bromide pretreatment, p = NS). Thus, a significant vagal response to histamine could not be detected in these subjects and so the results of the studies with pilocarpine are not reported.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study was designed to investigate the function of M₂ muscarinic receptors in subjects with COPD. In order to do this it was necessary to develop a method of inducing a vagally mediated bronchoconstriction, and for these studies a nasal challenge with cold dry air was used. This challenge induced bronchoconstriction in the subjects with COPD, as well as in smoking and nonsmoking subjects with normal lung function. The bronchoconstriction was inhibited by pretreatment with a muscarinic receptor antagonist, indicating that it was vagally mediated. These data confirm our own observations in younger individuals with normal lung function as well as those of Fontaneri and colleagues (10, 11), that nasal inhalation of cold dry air causes a vagally mediated bronchoconstriction.

Having established that the nasal cold air challenge induced a vagally mediated bronchoconstriction, we tested the function M₂ receptors by pretreating the subjects with the selective M₂ muscarinic receptor agonist pilocarpine. In previous studies in animal models, it has been shown that pilocarpine inhibits vagally induced bronchoconstriction when these receptors are functional (4, 5, 17). In this study we demonstrated that pilocarpine also prevented the cold air-induced bronchoconstriction in control subjects as well as those with COPD, indicating that in both groups there was normal function of M₂ receptors. The subjects with COPD had a high baseline Raw, almost twice the value of control subjects. Thus, it is possible that the attenuation of the cold air-induced bronchoconstriction was due to failure of either the airways to contract further or for this to be detected. This is unlikely, because impulse oscillometry can detect changes of resistance of greater than 50% of baseline values (18) and because Raw increased significantly immediately after the administration of pilocarpine. Thus, it is unlikely that Raw did not increase further because the airways were already maximally constricted. Instead, it is likely that pilocarpine attenuated the cold air-induced bronchoconstriction.

To confirm our observation of normal neuronal M₂ receptor function in subjects with COPD, we used a second technique to induce a vagally mediated bronchoconstriction, through a histamine challenge. Inhaled histamine has been reported to induce bronchoconstriction both by a direct effect on airway smooth muscle and by an indirect vagally mediated mechanism (14, 15). We were unable to demonstrate a vagal reflex response to histamine in subjects with COPD, because pretreatment with ipratropium bromide before the histamine challenge did not increase the histamine PC₂₀. Thus, it was not possible to use data from these experiments to investigate the function of M₂ muscarinic receptors in patients with COPD.

Under normal circumstances M₂ muscarinic receptors limit the degree of vagally induced bronchoconstriction. This means that when these receptors are dysfunctional there is an increased magnitude of vagally induced bronchoconstriction. The baseline differences in airway caliber between the groups of subjects make direct comparisons of the response to cold air between the two groups difficult. In preliminary studies we attempted to control for the differences in airway caliber by inducing a bronchoconstriction with histamine in healthy control individuals. However, the bronchoconstriction induced by histamine was not sustained for a sufficient period to allow the cold air challenge to be performed. Thus, it was not possible to directly compare the magnitude of increase of Raw between the control and COPD groups. It is likely that the more constricted airway in COPD would contract to a greater extent for a given stimulus than a healthy one (19). However, there was no difference in the degree of vagally induced bronchoconstriction between the groups. Indirectly, this suggests an absence of increased vagal reactivity in COPD and is in keeping with our conclusion that there is normal function of M₂ muscarinic receptors in this condition.

Although on average there was normal function of M₂ muscarinic receptors in both groups of subjects, there was a considerable heterogeneity in the magnitude of vagally induced bronchoconstriction in all subject groups. The degree of vagally induced bronchoconstriction was correlated with the degree of M₂ muscarinic receptor function, in both healthy and COPD subjects. The cause underlying the variability in M₂ receptor function in control and COPD subjects is not clear. This variability may reflect either the effects of inflammatory cell products or other factors that control the expression of the receptor or genetic differences in the function of these receptors; further studies will be required to investigate this observation.

This is the first study to have investigated the function of M₂ receptors and to have attempted to directly compare the magnitude of vagally induced bronchoconstriction between control and subjects with COPD. Brodde and colleagues recently demonstrated that there is a decline in the function of cardiac M₂ muscarinic receptors in older subjects (9). Although our subjects were of a comparable age to those patients reported in Brodde's study there was no evidence that there was a similar decline in M₂ muscarinic receptor function in the lungs, suggesting that the changes noted in the heart were organ-specific. We speculated that there may be dysfunction of neuronal M₂ muscarinic receptors in subjects with COPD because of the inhibitory effects of macrophage- or parainfluenza-derived neuraminidase or eosinophil major basic protein (MBP) on these receptors. Because increased concentrations of these compounds appear to be released into the airways, in particular during acute exacerbations of COPD (20), and because we specifically studied subjects with stable COPD, we cannot exclude the possibility that M₂ muscarinic receptor dysfunction occurs during exacerbations of COPD.

The preservation of M₂ receptor function in stable COPD is clinically significant. In the lungs, in addition to M₂ and M₃ muscarinic receptors M₁ receptors are found on cholinergic ganglia where they function to facilitate neurotransmission (21). The currently available anticholinergic drugs cause bronchodilation by inhibiting the action of acetylcholine on M₃ and M₁ muscarinic receptors. However, these drugs also inhibit M₂ muscarinic receptors on the postganglionic nerves and this leads to a facilitation of acetylcholine release (22, 23), which limits their effectiveness. The development of new functionally selective M₁ and M₃ receptor antagonists may be more effective than the currently available agents in improving airflow in patients with COPD (23, 24).

The results of this study suggest that there is normal function of neuronal M₂ muscarinic receptors in subjects with stable COPD. The presence of functional M₂ muscarinic receptors suggests that anticholinergic agents that selectively inhibit M_3/1 but not M₂ muscarinic receptors may be of particular benefit in this condition.