Airway mucus from asthmatics is often unusually solid. The death of a patient in status asthmaticus allowed the collection of 28 g of abnormal airway mucus at autopsy. Its chemical and physical properties were studied to reveal differences from more normal airway mucus. The gel plug taken from the airways could be dispersed in 6 M guanidinium chloride, but it took > 1 wk and 700 ml of extractant to disperse 3 g of exudate completely. In contrast, treatment with 10 mM dithiothreitol, which reduces disulfide bonds, dispersed the gel within seconds. Mucins accounted for 25% of the non-dialyzable material in the gel, while DNA constituted < 1% and proteoglycans could not be detected. The mucins were similar in architecture and general composition to other respiratory mucins and were present at a high concentration (approximately 40 mg/ml). The majority of mucins were of extreme size (mean M(r) 30-40 x 10(6)) and slow to dissolve, but sequential extraction experiments on the gel exudate demonstrated a proportion of mucins (15%), the most readily extracted, which had a higher density, 1.45-1.55 g/ml, a lower M(r) (11.5 x 10(6)) and were markedly more acidic than the bulk of the mucins. Both major and minor mucin populations were extremely heterogeneous in mass distribution. Electron microscopy of the major mucin species demonstrated extensive networks of molecules many microns in length. The major mucin species was distinctly less acidic than mucins previously described from either normal or diseased airways. Amino acid analysis of fractions across the charge distribution suggested the presence of at least two different mucin proteins occurring as distinct glycoforms.

This paper describes low-density mucus glycoconjugates released from feline trachea by dirhamnolipid (DRL), a toxin from Pseudomonas aeruginosa. Mucus glycoconjugates in feline tracheas were radiolabeled in vivo with 3H-proline and 14C-glucose. Control mucus and that released by 200 micrograms/ml DRL were dissolved in guanidine hydrochloride buffer (GuHCl) and chromatographed on Sepharose CL-2B. Molecules eluting in the void volume (V0) of the column were isolated by isopycnic density gradient centrifugation in CsCl/GuHCl. All samples gave peaks of radiolabeled and periodic acid/Schiff (PAS)-reactive material at rho = approximately 1.50 and approximately 1.60 g/ml, but DRL-stimulated samples contained low-density material (rho < 1.32 g/ml), also PAS-reactive and radiolabeled. Control secretions incubated with DRL in vitro did not form low-density material. In Triton X-100 (1% vol/vol), a nonionic detergent, low-density material behaved as smaller molecules, running in the partially included volume (Vi) of the column of Sepharose CL-2B, but still in the V0 of Sephacryl S-300. Incubation with chondroitinase ABC, heparinase II and III, and keratanase failed to change its elution profile on S-300, evidence against glycosaminoglycans; but proteolysis with trypsin or proteinase K gave two peaks, peptide fragments near the totally included volume of the column and glycopeptides in V0. The V0 glycopeptides banded between 1.50 and 1.55 g/ml in a CsCl gradient and eluted as a single peak in the Vi of Sephacryl S-400, suggesting a distinct homogeneous glycopeptide, smaller than those from normal mucins. The main 14C-labeled sugars in this glycopeptide were fucose, glucosamine, galactosamine, and galactose, consistent with a mucin. Thus, DRL releases stable but noncovalent complexes containing one or more distinct mucinlike glycoconjugates, probably combined with lipids and peptides. We discuss their possible relevance to airway diseases, including cystic fibrosis.

2. Vagal stimulation with regular pulses (10 V, 2 ms duration) at 1, 2X25, 4 5, 9 and 18 Hz produced frequency-dependent increases in the output of mucus glycoconjugates.3. The muscarinic agonist pilocarpine (0-1-10 ,uM), given intrasegmentally, produced dose-dependent increases in the output of mucus glycoconjugates.4. Pretreatment with atropine, phentolamine and propranolol reduced but did not, abolish the effects of vagal stimulation. Vagus nerve stimulation still caused frequency-dependent increases in the output of mucus glycoconjugates.5. High frequency stimulations at 22-5 and 47-5 Hz given intermittently (1 s burst then 4 s rest), whether in the absence or presence of cholinergic and adrenergic blockade, produced similar secretary responses as the same number of pulses delivered in regular trains at 4-5 and 9-5 Hz. This suggests that neither cholinergic nor non-adrenergic, non-cholinergic (NANC) nerve mechanisms in this system are potentiated by high frequency, intermittent burst stimulation.6. In the absence of atropine, regular vagal stimulation had a greater effect on heart rate than did the same number of pulses delivered in bursts.7. High molecular weight glycoconjugates from secretions were taken from the void volume of a Sepharose CL-2B gel filtration column and separated further by density-gradient centrifugation. Macromolecular components were observed at two densities, a typical mucin at 1'52 g ml-1, and a high density atypical component at 1-63 g ml-'. In secretions collected during vagal stimulation, either in the absence or presence of cholinergic and adrenergic blockade, the ratio of low density to high density macromolecules was higher than in unstimulated secretions. This can be explained if both cholinergic and NANC nervous vagal mechanisms stimulate the output of typical (density = 1-52 g ml-') mucins into the feline trachea.

No abstract