Variable Effects of Soman on Macromolecular by Ferret Trachea’** ROBERT
K. MCBRIDE, DANIEL J. ZWIERZYNSKI, KRISTA DAVID J. CULP, AND MATTHEW G. MARIN
Departments of Medicine and Dental Research, and Environmental Health Sciences Center, University of Rochester School of Medicine and Dentistry, Rochester, New York 14642
Received December 14, 1989; accepted August 15, 1990 Variable Effectsof Soman on Macromolecular Secretion by Ferret Trachea- MCBRIDE, ROBERT K., ZWIERZYNSKI, DANIEL J., STONE, KRISTA K., CULP, DAVID J., AND MARIN, MATTHEW G. (1991). Fundam. Appl. Toxicol. 16, 24-30. The purpose of this study was to examine the effect of the anticholinesterase agent, soman, on macromolecular secretion by ferret trachea, in vitro. We mounted pieces of ferret trachea in Us&g-type chambers. Secreted sulfated macromolecules were radiolabeled by adding 500 &i of ‘%04 to the submucosal medium and incubating for 17 hr. Soman added to the submucosal side produced a concentration-dependent increase in radiolabeled macromolecular release with a maximal secretory response (mean + SD) of 202 f 125% (n = 8) relative to the basal secretion rate at a concentration of lo-’ M. The addition of either low6 M pralidoxime (acetylcholinesterase reactivator) or 10e6M atropine blocked the response to lo-’ M soman. At soman concentrations greater than IO-’ M, secretion rate decreased and was not significantly different from basal secretion. Additional experiments utilizing acetylcholine and the acetylcholinesterase inhibitor, physostigmine, suggest that inhibition of secretion by high concentrations of soman may be due to a secondary antagonistic effect of soman on muscarinic rWqJtOl3.
0 199 I Society of Toxicology.
Soman and other similar but less toxic organophosphate agents are used as chemical warfare agents and as agricultural insecticides, respectively. As such, both intentional and inadvertent human exposures have produced fatalities (Sofer et al., 1989; Halle and Sloas, 1987; Bledsoe and Seymour, 1972; Eitzman and Wolfson, 1967). In most cases of organophosphate intoxication, death is attributed to
hypoxia due to interference with respiration by one or more mechanisms (Hayes, 1965): excessive respiratory tract secretions, bronchial constriction, paralysis of respiratory muscles, and/or failure of the respiratory center. Little information is available concerning the effects of organophosphate compounds on airway secretion. Because of the extreme toxicity of the anticholinesterase agent, soman (Grob, 1963), few physiological studies have been carried out. No prior studies have investigated its direct effect on airway secretion. We have shown previously that soman significantly increased mucociliary transport in ferret trachea (Mar-in et al., 1989). This effect may have been mediated by changes in the volume and/or the physical properties of the airway secretions. The major source of those secretions in man is the tracheobronchial submucosal glands
’ This work was supported in part by the U.S. Army Medical Research and Development Command, Contract DAMD17-85-C-5103. * In conducting the research described in this manuscript, the investigators adhered to the “Guide for the Care and Use of Laboratory Animals,” as promulgated by the Committee on Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, National Research Council (DHEW Publication NIH78-23, revised 1978). 0272-0590/91 $3.00 Copyright 0 I99 I by the society of Toxicology. All rights of reproduction in any form reserved.
EFFECTS OF SOMAN ON TRACHEAL
(Reid, 1959), which are cholinergically innervated (see review by Phipps, 198 1). Therefore, we investigated the direct effect of soman on airway secretion. For our studies we utilized the ferret as our experimental model. Ferret airways, like human airways, have large numbers of submucosal glands (Robinson et al., 1986) which are under autonomic control (Basbaum et al., 1983, 1981; Borson et al., 1980). The effects of soman on secretion were determined by quantitating basal and drugmediated release of acid-precipitable sulfated macromolecules. We found that soman increased secretion of labeled macromolecules up to a concentration of lo-’ M. This enhanced secretion was blocked by the soman antidotes, atropine and pralidoxime. At soman concentrations greater than lo-’ M, secretion rate decreased and was no different from basal levels. MATERIALS Preparation
AND METHODS of Ferret
Adult castrated male ferrets (Marshall Farms USA, North Rose, NY) were anesthetized with an intmperitoneal injection of sodium pentobarbital(lO0 mg/kg). Following euthanasia by thoracotomy and exsanguination, the trachea was removed, trimmed of adhering fat and connective tissue, and placed into 50 ml of L15 Medium Leibovitz containing 50 pg/ml gentamicin. We made a longitudinal incision through the posterior membranous portion and transversely cut the trachea into four pieces. Each tracheal piece was then mounted between the two-halves of an Ussing-type chamber in which 0.64 cm’ of trachea was exposed (Model CHMZ, World Precision Instrument, New Haven, CT). Each trachea supplied enough tissue for four chambers. The chambers were placed in a fume hood which had a face velocity of 100 ft/min and was equipped with an alarm system which was activated when the face velocity dropped to 70 ft/min. We added 15 ml of Medium 199 supplemented with gentamicin (50 &ml) to the luminal and submucosal sides of each chamber. Five hundred microcuries of Naz3?S04 (300-600 mCi/mmol; New England Nuclear, Boston, MA) were added to the medium on the submucosal side. The media in the chambers were maintained at 38°C. Circulation of the media through the luminal and submucosal chambers was accomplished by gas lift circulation reservoirs (Model U930 1, MRA, Clearwater, FL) using a gas mixture of 95% O2 and 5% CO2 . The reservoirs were equipped with water vapor
condensers. To achieve near-saturation labeling of secreted macromolecules, we incubated the mounted tracheal pieces for 17 hr (Boat, 1982). After this incubation period we removed the medium on the luminal side, washed the chamber three times, and refilled it with 15 ml of fresh Medium 199. Following a 60-min incubation (at Hour 18), a 2-ml sample was removed from the luminal side and replaced with 2-ml of fresh medium. At Hour 19, we removed a second 2-ml sample which was also replaced as before. Soman, physostigmine, antidotes, acetylcholine, or saline (controls) was added to the submucosal side. After a final 60-min incubation (at Hour 20), a 2-ml sample was removed from the luminal side.
of Bound Sul$ate Secretion
We quantitated 35SO&beled macromolecules using an acid precipitation procedure described earlier (Lundgren et al., 1988; Coles et al., 1984; Jennings et al., 1977) with the following modifications. Five hundred-microliter ahquots of media bathing the luminal surface were added to siliconized tubes containing 100 ~1bovine serum albumin (5 mg/ml). We then added 200 ~1 of a 20% trichloroacetic acid (TCA) and 4% phosphotungstic acid (PTA) solution to each tube. The tube was incubated at 4°C for 60 min to precipitate the labeled macromolecules. Ten milliliters of a TCA (5%):PTA (1%) solution was then added and each tube was vigorously vortexed and centrifuged at 700g for 10 min. We decanted the supematant and washed the precipitate with 5 ml of the 5%/l% acid solution. Following centrifugation, the pellet was washed with 4 ml of 95% ethanol. After centrifugation, the supematant was removed by aspiration, the pellet was dissolved by the addition of 300 ~1 of tissue solubilizer (T&l; Research Products International, Mount Prospect, IL), and the contents were placed in scintillation vials with 50 ~1 of glacial acetic acid and 6 ml of scintillation cocktail (3a7ob, Research Products International, Mount Prospect, IL). Radiolabeled sulfate in each vial (3sS) was determined by liquid scintillation spectrometry (Model LS3 133P scintillation counter, Beckman Instruments, Fullerton, CA). We calculated the amount of sulfate incorporated into acid-precipitated macromolecules which were secreted into the luminal bathing media from the following equation derived from Estep et al. (1981). A = ka P/S, where A = amount of sulfate incorporated into the acid-precipitated molecules expressed as pmol SO&m’ exposed tracheal tissue; k = constant = (chamber volume) t (precipitated sample volume) + (area of tracheal tissue); P = dpm of acid-precipitated ‘?G04 from the luminal sample; and S = specific activity of 35S04 in the submucosal medium = (dpm 35S03 + (pmol SO,). From the amounts of incorporated sulfate in the samples at Hours 18, 19, and 20, and with appropriate correction for media replacement, we determined basal secretion and drug-induced secretion rates.
Because the distribution and size of the submucosal glands in the mounted tracheal pieces were likely variable, we normalized the secretory response as a percentage change from basal secretion rate for each piece of tissue, where percentage change in basal secretion rate was equal to (drug-induced secretion rate) ~ (basal secretion rate) (basal secretion rate) x 100%. Basal secretion rate was taken as secretion during Hour 18 of each experiment. Secretion in response to drug exposure was measured during Hour 19. Tissue Viability We used two methods to assessthe viability of our in vitro preparation: lactate dehydrogenase release and the spontaneous electrical potential difference across the tracheal epithelium. Lactate dehydrogenase (LDH) release was determined according to the method of Wroblewski and IaDue (1955). Samples for LDH determination were obtained from the luminal side after 17 hr of incubation, prior to media replacement, and at the conclusion of the experiment at Hour 20. Tissue LDH content was quantitated by homogenizing tracheal pieces cut to an area of 0.64 cm2 (Polytron, Brinkmann Instruments, Westbury, NY) in 2 ml of Medium 199 at 4°C on setting 5 for 60 sec. Aliquots were taken for LDH determination as described above. The spontaneous potential difference was measured as described previously (Marin et al., 1976) with the following modifications. Agar bridges and calomel halfcells were replaced with Ag/AgCl half-cells prepared with 4% agar in 3 M KC1 and placed in ICC tuberculin syringe barrels. The half-cells were then snugly fit into the luer openings of the chambers. Data Analysis Each piece of trachea was randomly assigned with respect to the chamber utilized and the treatment. Each n value represents one piece of tracheal tissue from one animal unless noted otherwise. All results are presented as means + one standard deviation. The effects of drugs on the release of sulfated macromolecules was compared to basal secretion for each piece of trachea using the paired t test. The Newman-Keuls multiple range test was performed on multiple mean comparisons following a significant analysis of variance test. The (Ylevel for rejection of the null hypothesis was 0.05. Drugs and Reagents We obtained soman (pinacolyl methylphosphonofluoridate) from the U.S. Army Medical Research and Devel-
ET AL. opment Command (Aberdeen Proving Grounds, MD). This material was stored in small aliquots at a temperature of ~-70°C until immediately before use at which time it was thawed. Because of soman’s toxicity, all experiments were conducted with at least two people present in the laboratory. Emergency M 17A2 gas masks and Mark I autoinjector kits of atropine and pralidoxime were provided by the U.S. Army Medical Research and Development Command. At the conclusion of each experiment, the chambers, media, and tissue that came into contact with soman were decontaminated with NaOH at a final concentration of 1 N. The use, decontamination, and destruction of soman was monitored by the Environmental Health and Safety Officer of the University of Rochester. Pralidoxime was obtained from Ayerst Laboratories, Inc. (New York, NY). Sodium pentobarbital was obtained from Steris Laboratories, Inc. (Phoenix, AZ). Acetylcholine, atropine, physostigmine, tissue culture reagents, and all other chemicals were obtained from Sigma Chemical Co. (St. Louis, MO).
RESULTS Comparison of the mean basal secretory rate with the secretion rate during the final 60-min collection period for each control piece of tissue (n = 42) revealed no significant differences, 448 f 235 vs 459 + 213 pmoles S04/cm2/hr, respectively. The insignificant difference in secretion between the two collection periods suggested that steady-state labeling of the secretory pool had been achieved. The effect of soman concentration on sulfate bound macromolecular secretion is shown in Fig. 1. We found that soman produced a biphasic and concentration-dependent release of sulfated macromolecules. A maximum secretory response occurred at a soman concentration of lo-’ M. This represented a significant increase of 202% in secretion compared to basal secretion. The secretory response to soman decreased at concentrations greater than lo-’ M and were not significantly different from basal secretory rates. By comparison, pieces of ferret trachea secreted sulfated macromolecules in a dose-dependent manner when stimulated with acetylcholine (Fig. 2). The maximal secretory response was with an acetylcholine concentration of lO-5 M. The secretory responses to concentrations of 10m4 M or greater were not
EFFECTS OF SOMAN
-9 -6 -7 Soman Concentration
-6 -5 (log M)
FIG. 1. The effect of soman concentration on sulfated macromolecular secretion by ferret trachea. Values shown are the means + SD for (n) animals. *Indicates that the value is significantly different (p < 0.05) compared to basal secretion. TIndicates that the value is significantly different (p Q 0.05) compared to lo-’ M soman.
significantly different than the response to 1O-5 M. This indicated that the secretory response to acetylcholine was not biphasic. The viability of the tracheal epithelium during exposure to soman was assessed by the measurement of LDH release and the spontaneous potential difference across the tissue. In 28 control tissues measured, LDH release during the 20 hr of the experiment averaged 2.0 f 4.4% of the total tissue LDH content. In tissues taken from 24 tracheas, treatment with soman at concentrations from 10m9 to 10e5 M, resulted in LDH release that was independent of the soman concentration used, as determined by analysis of variance (p = 0.90). The mean release of LDH by tissues exposed to 10e5 M soman was 2.9 f 2.3% (n = 7). This value was not significantly different compared to paired control tissues (4.7 + 8.6%) as determined by the paired t test (p = 0.53). Potential difference was also independent of concentration of soman used. The mean potential difference of tissues exposed to 10e5 M soman was not significantly different compared to paired control tissues; 11.2 + 4.5 vs 11.4 k 5.8 mV (n = 8), respectively (paired ttest, p = 0.90). Macromolecular secretion induced by 1O-’ M soman was inhibited by agents that are known antidotes of soman. As shown in Table 1, both atropine and pralidoxime could each
FIG. 2. The effect of acetylcholine concentration on sulfated macromolecular secretion by ferret trachea. Values shown are the means + SD for(n) animals. *Indicates that the value is significantly different (p < 0.05) compared to basal secretion.
inhibit the secretory response to 10m7 M soman. When tested alone, atropine resulted in a slight decrease and pralidoxime a slight increase in secretion compared to basal values, in both cases the response was not significant. Additional experiments were conducted to investigate the lack of effect of high doses of soman ( lop5 M) on secretion. We found that pralidoxime could partially reverse the inhibTABLE 1 THE EFFEC~OF lo-'M SOMAN AND ITS ANTIDOTES ON SECRETION OF SULFATED MACROMOLECULES BY FERRETTRACHEA
Group low6 M atropine 10m6M pralidoxime 1O-’ M soman 1O-’ + 1O-’ +
M soman 10d6 M atropine M soman 10M6M pralidoxime
% Change relative to basal secretion -24+
21 (5) 142 23 (5) 202 + 125*
(5) 55+ 81 (3
Note. Values shown are the means k SD for (n) animals. * Indicates that the value is significantly different (p I 0.05) compared to basal secretion and to all other treatment groups.
itory effects on secretion of a high soman concentration (Table 2). We also determined the secretory effects of a high acetylcholine concentration in the presence of either soman or another acetylcholinesterase inhibitor, physostigmine. As shown in Table 2, acetylcholine at lop5 M produced a secretory response significantly greater than basal secretion. Addition of acetylcholine in the presence of soman (lo-’ M) failed to generate a significant response, whereas the combination of acetylcholine and physostigmine (lo-’ M) elicited a marked stimulation of secretion. DISCUSSION The anticholinesterase agent, soman, significantly altered the release of sulfated macromolecules by ferret trachea, in vitro. The response to soman was dose-dependent and biphasic with a maximal secretory response at a soman concentration of 10m7M. We believe that this secretory response was not related to cell death and release of the cytoplasmic contents. Epithelial cell viability, as assessed by release of LDH and spontaneous potential difference, was similar in control tissue and at all concentrations of soman tested. Soman-induced secretion was significantly inhibited by either atropine or pralidoxime. The site of action for atropine was probably the tracheal submucosal gland cells and not the surface epithelial cells. Tracheal glandular cells in several species have been shown to be cholinergically innervated and possess muscarinic receptors (Basbaum et al., 198 1, 1983; Culp and Marin, 1986; Murlas et al., 1980; Yang et al., 1988a). In addition, surface epithelial cells of the ferret trachea do not secrete macromolecules in response to cholinergic agonists (Borson et al., 1984). The trend for atropine alone to reduce secretion (Table 1) suggests that acetylcholine is available for activation of secretion in these preparations. The source of acetylcholine is probably cholinergic varicosities which have been shown to be within 10 pm of submucosal glands (Murlas
ET AL. TABLE THE EFFECX
ANTIDOTES, AND ACETYLCHOLINE OF SULFATED MACROMOLECULES
Group 10m5 M soman
ON THE SECRETION BY FERRET TRACHEA % Change relative basal secretion
10m5 M soman + 10m6 M pralidoxime lo-’ M acetylcholine
3k 10 (9) 83 -c 89 (5) 206 + 85*
lo-5 + lo-’ +
(7) 12+27 (5) 234 f 99* (7)
M soman 10e5 M acetylcholine M physostigmine lo-’ M acetylcholine
Note. Values shown am the means f SD for(n) animals. * Indicates that the value is significantly different (p 5 0.05) compared to basal secretion and to soman, soman/ acetylcholine, and soman/pralidoxime groups.
et al., 1980). Inhibition of soman-induced secretion by pralidoxime was likely mediated by its ability to reactivate acetylcholinesterase by forming an oxime-phosphonate complex with soman (Taylor, 1985). Inhibition of the stimulatory effect of lo-’ M soman by both atropine and pralidoxime suggests that inactivation of acetylcholinesterase by soman with the resultant activation of secretion was at least due, in part, to elevated levels of acetylcholine. A prior study demonstrated that soman, at a concentration of 10e7 M, reduced bronchial acetylcholinesterase activity 88% (Aas et al., 1986). An alternative hypothesis is that soman acted as a partial agonist at lower concentrations and thus contributed to activation of the stimulus-secretion coupling mechanism. Partial agonist activity of soman has been reported for nicotinic acetylcholine receptors found in Torpedo electric organ membranes (Bakry et al., 1988). However, this hypothesis is unlikely, because partial agonists normally elicit submaximal responses, and we found that the magnitude of the secretory response to soman at a concentration of 10m7 M was similar to the maximal secretory response found for tra-
cheal pieces stimulated (lo-’
Secretion was at or near basal levels at concentrations of soman greater than lop7 M. Addition of 1Od5M acetylcholine to 1O-5 M soman resulted in no significant increase in secretion (Table 2). These results were in contrast to those with another acetylcholinesterase inhibitor, physostigmine, in which 10e5 M acetylcholine in the presence of a high concentration of physostigmine (1 Oe5 M) elicited a maximal secretory response (Table 2). On the other hand, a lOOO-fold higher concentration ( 10d2 M) of acetylcholine by itself elicited a 160% increase in secretion (Fig. 2). Thus it is unlikely that high levels of acetylcholine were responsible for the attenuated secretory response observed with high soman concentrations. Furthermore, as discussed above, cytotoxic effects of soman are not a likely explanation for this decrease in secretory response. Howevc competitive and reversible binding of semi to a subtype of muscarinic receptors has bet reported (Bakry et al., 1988). It was found th, soman could antagonize the binding of ci [3H]methyldioxolane, a ligand reported I preferentially bind to the M2 subtype of mu! carinic receptors (Closse et al., 1987). The M subtype has been shown to represent abou 50% of the muscarinic receptors associate1 with isolated tracheal submucosal gland cell: from swine and to be the subtype responsiblt for mediating macromolecular secretion (Yang et al., 1988a,b). In the present study, il is possible that soman at higher concentrations interacted with muscarinic receptors and inhibited the binding of acetylcholine. Receptor antagonism by soman is also supported by experiments described above in which we substituted another acetylcholinesterase inhibitor, physostigmine, for soman. We found that when 10m5 M acetylcholine and IOH M physostigmine were added together, a maximal secretory response was evoked (Table 2). In addition, pralidoxime was able to partially reverse the effects of a high dose of soman and mediate a modest increase in secretion although this effect was not significant. Because
pralidoxime is known to react directly with soman by forming an oxime-phosphonate complex, it is possible that pralidoxime reduced the effective soman concentration enough to permit activation of a small number of tracheal gland cell muscarinic receptors. Further experiments, such as muscarinic receptor competition studies with soman, are required to verify this potential action of soman at high concentrations. Collectively, our results indicate that soman can significantly alter secretion of sulfated macromolecules by ferret trachea. Low concentrations of soman enhance secretion of macromolecules probably due to an increase in acetylcholine concentration at glandular receptors. Unexpectedly, higher concentrations failed to enhance secretion. Although this mechanism is not fully explained, our data indir*tlxz n-e----* *’ -
secretion from tracheal glands of ferrets.J. A&. Physiol. 51,451-466.
BORSON, D. B., CHINN, R. A., DAVIS, B., AND NADEL, J. A. (1980). Adrenergic and cholinergic nerves mediate fluid secretion from tracheal glands of ferrets. J. Appl. Physiol.
COLES,S. J., NEILL, K. H., AND REID, L. M. (1984). Potent stimulation of glycoprotein secretion in canine trachea by substance P. J. Appl. Physiol. 57, 1323-1327. CLOSSE, A., BITTIGER, H., LANGENEGGER, D., AND WANNER, A. (1987). Binding studies with [3H]cismethyldioxolane in different tissues. Naunyn-Schmiedebergs Arch. Pharmacol.
335, 372-3 7 1.
CULP, D. J., AND MARIN, M. G. (1986). Characterization of muscarinic cholinergic receptors in cat tracheal gland cells. J. Appl. Physiol. 61, 1375-1382. EITZMAN, D. V., AND WOLFSON, S. L. (1967). Acute parathion poisoning in children. Amer. J. Dis. Child. 114, 391-400.
ESTEP, J. A., ZORN, J. P., AND MARIN, M. G. (1981). Kinetics of sulfated mucous glycoprotein secretion in dog trachea in vitro. J. Appl. Physiol. 50, 383-39 1. GROB, D. (1963). Anticholinesterase Intoxication in Man and Its Treatment. In Handbuch der Experimentellen Pharmakologie. Cholinesterase and Anticholinesterase Agents (0. Eichler and A. Farad, Eds.), Vol. 15,
pp. 990- 1005. Springer-Verlag, Berlin. HALLE, H., AND SLOAS, D. D. (1987). Percutaneous organophosphate poisoning. South. Med. J. 80, 11791181. HAYES, JR., W. J. (1965). Parathion poisoning and its treatment. JAMA 192, 135-l 36. JENNINGS,M., CROSS,C. E.. AND LAST, J. A. (1977). Glycoprotein synthesis by tracheal explants from various mammalian species. Camp. Biochem. Physiol. A 51, 317-320.
LUNDCREN, J. D., HIRATA, F., MAROM, Z., LOCUN, C., STEEL, L., KALINER, M., AND SHELHAMER, J. (1988). Dexamethasone inhibits respiratory glycoconjugate se-
ET AL. cretion from feline airways in vitro by induction of lipocortin (lipomodulin) synthesis.Am. Rev. Respir. Dis. 137,353-357.
MARIN, M. G., DAVIS, B.. AND NADEL, J. A. (1976). Effect of acetylcholine on Cl- and Na+ fluxes across dog tracheal epithelium in vitro. .4mer. J. Physiol. 231, 15461549. MARIN, M. G., ZWIERZYNSKI,D.. MCBRIDE, R. K., AND CULP, D. J. (1989). Effects of soman and its antidotes on tracheal mucociliary transport of ferrets. J. Appl. Physiol. 66. 1390- 1396. MURLAS, C., NADEL, J. A., AND BASBAUM, C. B. (1980). A morphometric analysis of the autonomic innervation of cat tracheal glands. J. Auton. Nerv. Syst. 2, 23-31. PHIPPS, R. J. (1981). The airway mucociliary system. In International
iul0g.v II( (J. G. Widdicombe. Ed.). Vol. 23, pp. 213260. Univ. Park Press, Baltimore. REID, L. (1959). Chronic bronchitis and hypersecretion of mucus. Lect. Sci. Basis Med. 8, 235-255. ROBINSON, N. P., VENNING, L., KYLE, H.. AND WIDDICOMBE, J. G. (1986). Quantitation of the secretory cells of the ferret tracheobronchial tree. J. Anat. 145, 173188.
SOFER,S.. TAL, A.. AND SHAHAK, E. (1989). Carbamate and organophosphate poisoning in early childhood. Pediatr.
Care 5, 222-224.
TAYLOR, P. (1985). Anticholinesterase agents. In The Pharmacological Basis q/‘Therapeutics (A. G. Gilman. L. S. Goodman, T. W. Rail, and F. Murad. Eds.), 7th ed., pp. 110-l 29. Macmillan, New York. WROBLEWSKI, F., AND LADUE, J. S. (1955). Lactate dehydrogenase activity in blood. Proc. Sot. Exp. Biol. Med. 90,210-213.
YANG, C. M.. FARLEY. J. M., AND DWYER. T. M. (1988a). Muscarinic stimulation of submucosal glands in swine trachea. J. Appl. Physiol. 64, 200-209. YANG, C. M.. FARLEY, J. M., AND DWYER,T. M. (1988b). Acetylcholine-stimulated chloride flux in tracheal submucosal gland cells. J. Appl. Physiol. 65, I89 1- 1894.