Pulmonary Pharmacology (1992) 5, 167-174 PULMONARY PHARMACOLOGY
The Effects of Intraluminal and Extraluminal Drug Application on Secretion and Smooth Muscle Tone in the Ferret Liquid-filled Trachea In Vitro S . Kitano, U . M. Wells*, S . E. Webber, J. G. Widdicombe Department of Physiology, St George's Hospital Medical School, London SW17ORE, UK
SUMMARY: With the ferret liquid-filled trachea in vitro, intraluminal methacholine (MCh), phenylephrine (PE) and histamine (Hist) increased smooth muscle tone and salbutamol (Salb) decreased tone . Lysozyme output was increased by intraluminal MCh and PE . Albumin transport into the lumen was not altered by intraluminal Hist, Salb or PE . The concentration-response curves for smooth muscle contraction and for lysozyme output to extraluminal MCh lay to the left of those for intraluminal MCh . Indomethacin shifted the smooth-muscle response curves to MCh significantly to the left but did not significantly alter lysozyme output . Extraluminal MCh produced a concentrationdependent increase in albumin output whilst intraluminal MCh did so in one of three studies . Albumin output in response to MCh was not significantly altered by indomethacin . Thus, MCh has a less potent effect on smooth muscle and lysozyme secretion and, to a lesser extent, on epithelial albumin transport when given intraluminally . This may be because the epithelium restricts diffusion of the drug or due to the production of a non-prostanoid factor which inhibits smooth muscle responsiveness. Smooth muscle responsiveness is enhanced by blocking cyclooxygenase activity, suggesting MCh-induced release of a prostanoid with relaxant activity .
little evidence about the effects on secretion of drugs given intraluminally and no study has compared concentration-response curves for drugs given by different routes . We have therefore investigated whether intraluminal application of drugs, whose extraluminal effects have previously been characterized in the same model," alter secretion of lysozyme from submucosal glands, the epithelial transport of albumin and smooth muscle tone in the ferret whole trachea in vitro . We then went on to compare the effects of routes of administration of methacholine (MCh), a potent secretagogue, on these variables . In addition we have investigated the possible involvement of cyclo-oxygenase products in the responses to MCh, using indomethacin .
INTRODUCTION Many bronchodilator drugs used in the treatment of respiratory disease and bronchoconstrictor agents used in provocation tests are applied as aerosols to the airway lumen ; they have to diffuse across the mucosa of the trachea and lungs from the luminal side to reach their site(s) of action, usually the smooth muscle . The epithelium has long been regarded as a possible barrier to the diffusion of intraluminal drugs, particularly those such as /t-agonists which are relatively lipid-insoluble, and evidence has accumulated that the epithelium may be stimulated by drugs to produce factors which reduce the responsiveness of smooth muscle .' , ' Little was known about the comparative effectiveness of drugs applied to different sides of the airway until recent studies showed that smooth muscle is more sensitive to stimuli applied to the submucosa than to the lumen of the trachea .` However, although numerous studies in vitro have demonstrated that various parameters of secretion can be influenced by drugs given either extraluminally 5' 9 or with unrestricted access to both sides of the airway, 10,1 •" there is
MATERIALS AND METHODS Experimental preparation Ferrets of either sex (weight 0.8-2 .1 kg) were anaesthetized by an intraperitoneal injection of sodium pentobarbitone (50 mg/kg ; Sagatal, May & Baker, Dagenham, UK) . The trachea was exposed and its
* For correspondence. 0952-0600/92/030167+08 $08 .00/0
© 1992 Academic Press Limited
S . Kitano et al
laryngeal end was cannulated about 5 mm below the larynx with a perspex conical collecting well with a narrow central bore . The ferret was killed with an overdose of anaesthetic injected into the heart and the trachea was exposed to the carina, cleared of surrounding tissue, and removed . A plastic cannula was then inserted at the carinal end . The trachea was mounted in an organ bath, laryngeal end down so that secretions were carried towards the collecting well by ciliary transport and gravity . A polyethylene tube (0 .5 mm internal diameter) was inserted into the central bore of the collecting well and its other end was attached to a syringe with a needle . By this means liquid could be washed into and withdrawn from the tracheal lumen . The submucosal side of the trachea was surrounded by Krebs-Henseleit solution containing bovine serum albumin (BSA, 4 mg/ml) and fluorescein-labelled BSA (0 .04 mg/ml) . The submucosal bathing medium was maintained at 37°C and gassed with 95% 0 2/5% C0 2. Before the start of an experiment, the lumen of the trachea was washed out three times with KrebsHenseleit solution, then refilled . The submucosal buffer was also changed and the trachea left for 20 min . At the beginning of the experiment the lumen of the trachea was filled with Krebs-Henseleit solution to a fixed point on the upper cannula . This solution was immediately removed, in order to provide a control sample at time zero in each collection period, and replaced by an equal volume . After 30 min the luminal liquid was withdrawn and its volume measured . The lumen was then refilled immediately and the submucosal buffer replaced by fresh Krebs-Henseleit solution . This procedure was repeated for all subsequent sampling periods . The samples of luminal fluid were stored at 4 ° C following collection and then at - 20°C after the experiment . The fluorescent albumin content and lysozyme activity of each sample were measured later . Changes in tracheal smooth muscle tone were measured during each collection period . In the first series of experiments this was done by measuring the change in volume of the tracheal lumen by determining the volume of liquid (in µl) displaced from or into the trachea . The maximum change (in mm) in the liquid level in the upper cannula was measured and since the internal radius of the cannula was known, the volume change could be calculated . A negative volume change denotes a relaxation of the tracheal smooth muscle . There was no significant difference between the initial and final sample volumes during any collection period, suggesting that any changes in liquid level were not due to a change in the volume of secretions in the trachea or to liquid transport across the tracheal wall. In the second and third studies intratracheal pressure was recorded by connecting the
upper tracheal cannula to a manometer (Micromanometer, Hilger-I.R.D. Ltd. or M12, Mercury Electronics Ltd ., Glasgow, UK) . Results are expressed in kPa .
Experimental protocol Three series of experiments were performed . Drugs were applied to the trachea either intraluminally or in the submucosal buffer during every third sampling period so that each stimulus was preceded by two 30min controls . In the first study, the intraluminal Krebs-Henseleit solution was replaced with KrebsHenseleit solution containing either methacholine (MCh), histamine (Hist), salbutamol (Salb) or phenylephrine (PE), all at 100 µM. In the second study, a comparison was made of the effects of intraluminal and extraluminal administration of MCh. The trachea was exposed to intraluminal MCh at 0 .1, 1 and 10 µM or extraluminal MCh at 0 .01, 0.1 and 1 µM. The concentrations were given in a random order to the inside and outside of the trachea alternately . In the third series of experiments, the tracheas were set up in pairs in separate organ baths to examine the effect of indomethacin on intra- and extraluminal MCh-induced responses . The test tracheas received indomethacin (1 µM) plus vehicle in both submucosal and intraluminal buffers throughout the experiment . The control tracheas received the vehicle alone . The vehicle consisted of 27 µM NaHCO3 and 0.009% (v/v) ethanol in Krebs-Henseleit solution . In one set of experiments both control and test tracheas received MCh in the concentration-range 0 .01-1 pm . In another set tracheas received concentrations in the range 1-100 gm . In each trachea intraluminal administration of a concentration of MCh was followed by extraluminal administration of the same concentration .
Measurement of albumin transport The fluorescence of the intraluminal fluid samples was measured with a fluorimeter (Perkin-Elmer, Beaconsfield) using an excitation wavelength of 490 nm and an emission wavelength of 550 nm . Using fluorescent BSA standards, a curve relating fluorescent albumin concentration (0 .1-7 .5 .tg/ml) to arbitrary fluorescent units was constructed . The concentration of fluorescent BSA in an intraluminal fluid sample was derived from the standard curve . The total concentration of BSA in the sample was obtained by multiplying the fluorescent albumin concentration by the ratio of non-fluorescent to fluorescent albumin used in the submucosal buffer . Previous studies have confirmed that this technique accurately reflects the
Drug Application in the Ferret Trachea
albumin concentration determined by affinity chromatography, suggesting that the fluorescent label remains attached to the albumin during transport .' Albumin output was calculated by dividing the total amount of albumin in the sample by the time during which the sample was collected .
Measurement of lysozyme secretion The lysozyme concentration in the samples was estimated using an assay which measures the reduction in turbidity caused by breakdown of the cell wall of the bacteria Micrococcus lysodeikticus by lysozyme . 13 Twenty microliters of each luminal fluid sample (diluted in 1 :10) were incubated in potassium phosphate buffer (50 mm, pH 7 .4) containing BSA (1 mg/ ml), sodium azide (1 mg/ml) and M . lysodeikticus (0 .3 mg/ml) . A standard curve was obtained by incubating known concentrations of hen egg-white lysozyme (0-100 ng/ml) under the same conditions . BSA was included in the assay for its protein-stabilizing effects and sodium azide was added to prevent the growth of bacteria in the incubating solutions . The reaction mixtures were incubated for 18 h at 37°C . After incubation, the optical density (OD) of the standards was measured at 450 nm. A standard curve was constructed by plotting OD against the logarithmic concentration of lysozyme in the solution . The lysozyme concentration in each of the diluted luminal samples was determined from the standard curve and the rate of secretion (ng/min) was calculated .
Data analysis The effects of drugs on tracheal pressure are shown as mean changes in intraluminal volume ± SE for the first set of experiments, and as change in tracheal pressure (kPa) ± SE for the second set of experiments. The effects on lysozyme and albumin outputs are shown as mean output ± SE and were analysed statistically by comparison with the preceding control using Student's two-tailed paired t-test .
Drugs Methacholine chloride, phenylephrine hydrochloride, histamine dihydrochloride, salbutamol, indomethacin, bovine serum albumin (BSA), fluorescein isothiocyanate BSA, hen egg-white lysozyme (Muramidase) and Micrococcus lysodeikticus were all obtained from Sigma Chemical Company Ltd (St Louis, MO, USA) . The composition (mm) of the Krebs-Henseleit solution was: NaCl 120 .8, KCl 4 .7, KH2PO4 1 .2, MgS04.7H 20 1 .2, NaHCO3 24 .9, CaCl2 2.4 and glucose 5 .6 .
RESULTS Resting outputs The mean outputs of albumin and lysozyme in intraluminal liquid samples taken at the start of each collection period were not significantly different from zero . The mean resting outputs of albumin and lyso.3±0 and 86 ± 21 ng/min 2 zyme were .3µg/min (n = 18) during the first control period in the first and second series of experiments . In the presence of indomethacin, the outputs were 3 .6±0 .5 µg/min and 31 ± 10 ng/min (n = 9), respectively, during the first control period, compared with 3 .0 f 0 .5 pg/min and 93 ± 50 ng/min in the paired controls . Tracheal pressure during the first control period decreased by - 0.22 ± 0 .06 kPa (n = 9) in indomethacin-treated tracheas compared with - 0.12 ± 0 .06 kPa in the paired controls . There were no significant differences in any of the baseline measurements between indomethacintreated and control tracheas. Intraluminal drugs Intraluminal MCh and PE decreased tracheal intraluminal volume by 155 ± 29µl (n=9, P