ABSTRACT
The P2Y family of receptors comprises seven transmembrane G-protein coupled receptors, for which the naturally occurring ligands appear to be purine and pyrimidine nucleotides (1). The P2Y2 receptor is found on the apical surface of a variety of airway epithelia, including the ciliated epithelial cells and goblet cells of the trachea, bronchi, and bronchioles. Recent in vitro data indicate that extracellular adenosine triphosphate (ATP) and uridine triphosphate (UTP) nucleotides interact with P2Y2 receptors to initiate Cl“ secretion on the apical membrane of human airway epithelia (2,3), principally via calcium-regulated Cl-channels (4). Benali et al. (3) showed that this enhanced Cl~ secretion resulted in increased hydration of the airway surface with the application of equimolar concentrations of ATP or UTP. ATP has also been shown to stimulate ciliary beat frequency (CBF) via release of intracellular calcium stores both in vitro (5,6) and in vivo (7). Furthermore, Geary et al. (6) showed that the increase in CBF with ATP was twice that observed by equimolar treatment with isoproterenol. Finally, extra347
cellular triphosphate nucleotides (ATP and UTP) have been shown to stimulate goblet cell degranulation and mucin release onto apical surfaces of both normal and cystic fibrosis epithelial explants (4).Impaired mucociliary clearance may contribute to the pathophysiology of several airway diseases such as cystic fibrosis and chronic bronchitis. These patients have difficulty clearing secretions from their airways, resulting in chronic obstructive changes and repeated pulmonary infections. Similar problems occur in patients with primary ciliary dyskinesia (PCD), a genetic disease characterized by an ultrastructural defect in dynein arms of cilia, which manifests itself as abnormal or absent ciliary movement and thus, impaired mucociliary clearance. Patients with PCD rely on effective cough clearance to clear their airways of secretions (8). The integrated effects of ATP or UTP on the airway epithelium described above may facilitate mucociliary or cough clearance in patients with airways disease. To test the effects of UTP on mucociliary and cough clearance in humans, a series of studies was performed in normal and compromised subjects. Uridine 5'-triphosphate (Prolucin, Inspire Pharmaceuticals, Inc.) was chosen rather than ATP for the human studies because of bronchoconstriction observed with inhalation of adenosine and adenosine 5'-monophosphate (AMP) in patients with asthma (9,10) and chronic bronchitis (11). In addition, intravenous infusions of ATP cause a dose-dependent dilation of pulmonary and systemic vasculature in humans (12), which could potentially lead to systemic hypotension. Studies in anesthetized dogs suggested no adverse effects of aerosolized UTP on airway mechanics, gas exchange, or hemodynamics, whereas infusion of ATP caused hypotension not seen with comparable doses of UTP (13).Mucociliary clearance rates can be measured in humans by assuming that a nonpermeating, inhaled marker depositing on the airway surface moves out of the lung at the same rate as the airway secretions in which it is immersed. The most common technique is to use inhaled, radiolabeled particles, aqueous or dry, that upon deposition in the lung can be followed by gamma camera or scintillation detectors to determine their rate of egress from the lung. Technetium 99m (99mTc), generally the radiolabel of choice, may be bound to such impermeable markers as iron oxide, sulfur colloid, red blood cells, albumin, teflon, and polystyrene latex spheres. The aerosol may be generated by a number of means (e.g. nebulizers, spinning top, condensation) for inhalation by the subject. To assess the effects of UTP on mucociliary clearance, we used radiolabeled iron oxide particles (4-5 pm MMAD) generated by a spinning top (14,15). After inhalation of these markers, retention of activity in the lung (as a percent of initial deposition) is monitored as a function of time over a period of 1 -6 hours to determine clearance rates (Fig. 1). Whole lung retention is strongly dependent on the sites of particle deposition within the lung and may be confounded by spontaneous coughing episodes during monitoring (15). In the first case, it is important to control particle size characteristics and breathing pattern of the aerosol to repro-
duce deposition patterns as similarly as possible within and between patients. Second, while patients may be advised to avoid coughing during the measurements, spontaneous coughing will inevitably occur. To quantify episodes of spontaneous coughing, the number of coughs should be recorded during the clearance measurements. Then, in the study design the number of spontaneous coughs can be considered as a covariate in the analysis to insure that frequency of these coughs does not overly influence any differences due to therapy effects. Finally, while uncontrolled coughing can confound measures of mucociliary clearance, measuring cough clearance by incorporating controlled, voluntary coughs during the measures of particle clearance may provide a more sensitive indicator of rheological changes in airway secretions (16). EFFECT OF UTP ON MUCOCILIARY CLEARANCE IN THE NORMAL LUNG
Before testing the effects of aerosolized UTP in patients, we first tested its safety and effects on mucociliary clearance in healthy, nonsmoking adults (14). To achieve optimal clearance in cystic fibrosis via UTP’s action on Cl-secretion it may also be necessary to block the excess Na+ absorption in these patients (17). Therefore, we tested the effects of UTP alone and in combination with amiloride, an epithelial Na+ channel blocker, in the healthy subjects as well. The mucociliary clearance protocol design was a double-blind crossover study in 12 healthy male subjects with four aerosol treatment arms: aerosolized saline (0.12%) (vehicle), UTP (10-2 M), amiloride (1 .3 X 10~3 M), or UTP/amiloride combined. Following inhalation of the 99mTc-labeled iron oxide particles and an initial deposition scan (two 2 minutes scans) by gamma camera, subjects inhaled the aerosolized treatment (3.5 mL in a Devilbiss 646 jet nebulizer) over a period of 15-20 minutes. Lung scans (2 minutes each) were sequentially acquired over a period of 2 hours, including the treatment period (Fig IB). During the period of 80-120 minutes, four sets of 15 standardized coughs (16) were interspersed to assess the effect of cough clearance. Subjects returned 24 hours after inhalation of the radiolabeled particles for a half-hour scan to index the slow cleared (alveolar) retention of particles.The results showed that UTP alone and in combination with amiloride stimulated mucociliary clearance in the healthy subjects (14). Figure 1A shows gamma camera images from a subject before and 15 minutes after aerosolized vehicle or UTP/amiloride illustrating the latter’s dramatic effect on particle clearance. Clearance rates through 50 minutes (linear slope through the break point in the biphasic curve: Fig. IB) for the whole lung were enhanced from 0.47%/ min with vehicle to 1.14 and 1.01%/min with UTP/amiloride and UTP, respectively, in normal subjects. On the one hand, the similar enhancement with UTP/ amiloride and UTP alone might suggest that the UTP effects on ciliary beating
and goblet cell degranulation were more important than any hydration effects for enhancing mucociliary clearance in the normal lung. Interestingly, however, we did see a mild but significant enhancement of clearance with amiloride (to 0.63%/ min) in these healthy subjects. Our data mirror that of Benali et al. (Fig 2) (3), who showed that fluid absorption (apical to basal) in cultured nasal epithelial cells was inhibited by amiloride but had no effect in the presence of ATP or UTP. These authors suggested that ATP or UTP may stimulate the same amount of fluid secretion in the presence or absence of amiloride by either (1) ATP/UTP stimulation of calcium-dependent potassium channels that may be enhanced in the absence of amiloride (18,19), or (2) ATP/UTP inhibition of sodium absorption (20). The possibility that increasing periciliary fluid volume on the normal airway surface can enhance mucociliary clearance is supported by recent studies in patients with pseudohypoaldosteronism (PHA) (21), a rare inherited disease characterized by poor Na+ and volume absortion from the airway surface. These
patients’ airways are “ overhydrated,” appearing very wet by bronchoscopy. The baseline mucociliary clearance in three adult patients was found to be “ supernormal,” four times the average normal rate during the first 20 minutes of measurement. The mechanism by which an increased periciliary fluid layer results in enhanced mucociliary clearance is unclear. The efficiency of mucociliary clearance is thought to be dependent on contact of ciliary tips with the inner surface of a layer of mucus. However, the enhanced mucociliary clearance observed in PHA suggests that an increase in the volume of airway surface fluid does not “ float” mucus off the tips of cilia, slowing mucociliary clearance. On the contrary, it seems likely that the increased fluid facilitates the swelling of mucus, leading to the formation of a highly hydrated mucous blanket with optimal physical properties for an efficient mucociliary transport (3). Thus, it is possible that the enhanced clearance observed with UTP and UTP/amiloride in normal subjects (Fig. IB) is due to hydration of the normal airway surface in addition to increased mucus and ciliary beating effects.No enhancement of cough clearance was seen during the 80-to 120-minute period of observation with either UTP or UTP/amiloride. This was likely due to the loss of radiolabeled tracer from the airways by 80 minutes postdeposition (Fig. IB) following treatment with UTP or UTP/amiloride. It is unlikely, however, that in the normal lung UTP would significantly affect cough clearance, as cough plays a lesser role than mucociliary clearance in the transport of secretions from the normal lung (22). All treatments produced some degree of spontaneous coughing in this study, but the number of coughs was not different between the treatment groups and was not predictive of clearance during the 0 to 50-minute interval where clearance was greatly enhanced by UTP and UTP/amiloride (Fig. IB).Immediately following the single treatment of UTP alone (10"2 M), there was a small but statistically significant decrement in FEV) and maximal midexpir-atory flow (MMEF) (14). A small decrease in Pao2 was also noted in subjects immediately following treatment with UTP/amiloride. These changes were clinically insignificant but likely reflected the effect of liquid movement into the airways. Once most of the increased liquid cleared the bronchi (within 15 minutes following treatment; Fig. 1) these small decrements in lung function returned to normal.Aerosolized UTP was shown to cause a marked increase in mucociliary transport in our normal subjects. It is also possible that endogenous ATP acts in vivo to modulate or accentuate all aspects of mucociliary function, stimulation of CBF, mucus production, and Cl" and water secretion onto the airway surfaces. Basal levels of ATP on airway surfaces have been found to be in the submicromolar range (slightly less than the ED50 for activation of P2Y2 receptors) in both normal subjects and cystic fibrosis (CF) patients (23). The autocrine (or paracrine) secretion of ATP by airway cells may represent one component of mucociliary transport regulation.
