Evidence for a second peptide cleavage in the C-terminal domain of rodent intestinal mucin Muc3.

Rat intestinal mucin Muc3 (rMuc3), like its human homologue (MUC3) and several other membrane mucins, contains a C-terminally located SEA (sea urchin sperm protein, enterokinase and agrin) module, with an intrinsic proteolytic site sequence G downward arrow SIVV (where G downward arrow S is the glycine serine cleavage site). As shown previously [Wang, Khatri and Forstner (2002) Biochem. J. 366, 623-631], expression of the C-terminal domain of rMuc3 in COS-1 cells yields a V5 epitope-tagged N-terminal glycopeptide of 30 kDa and a Myc- and His epitope-tagged C-terminal glycopeptide of 49 kDa. The present study shows that the 49 kDa membrane-anchored fragment undergoes a further cleavage reaction which decreases its size to 30 kDa. Western blotting, pulse-chase metabolic incubations, immunoprecipitation and deglycosylation with N-glycosidase F were used to detect and identify the proteolytic products. Both the first and second cleavages are presumed to facilitate solubilization of Muc3 at the apical surface of enterocytes and/or enhance the potential for Muc3 to participate in ligand-receptor and signal transduction events for enterocyte function in vivo.

N-linked oligosaccharides play a role in disulphide-dependent dimerization of intestinal mucin Muc2.

Within the C-terminal domain of many secretory mucins is a 'cystine knot' (CK), which is needed for dimer formation in the endoplasmic reticulum. Previous studies indicate that in addition to an unpaired cysteine, the three intramolecular cystine bonds of the knot are important for stability of the dimers formed by rat intestinal mucin Muc2. The present study was undertaken to determine whether the two N-glycans N9 and N10, located near the first and second cysteines of the knot, also play a role in dimer formation. The C-terminal domain of rat Muc2 (RMC), a truncated RMC mutant containing the CK, and mutants lacking N9 and N10 sites, were expressed in COS-1 cells and the products monitored by radioactive [(35)S]Met/Cys metabolic pulse-chase and immunoprecipitation. Mutation of N9, but not N10, caused increased synthesis of dimers over a 2-h chase period. The N9 mutant remained associated with calreticulin for a prolonged period. About 34-38% of the total labelled products of RMC and its mutants was secreted into the media by 2 h, but the proportion in dimer form was dramatically reduced for the N9 mutant, suggesting lower dimer stability relative to RMC or its N10 mutant. We conclude that under normal conditions the presence of the N9 glycan functions to maintain a folding rate for mucin monomers that is sufficiently slow to allow structural maturation and stability of Muc2 dimers. To our knowledge this report is the first demonstration that a specific N-glycan plays a definitive role in mucin dimer formation.

Effect of chondroitinase ABC on purulent sputum from cystic fibrosis and other patients.

Cystic fibrosis (CF) patients develop chronic lung infections associated with airway obstruction by viscous and insoluble mucus secretions. Although mucus glycoproteins (mucins) are thought to be responsible for mucus plugs, other glycoconjugate components of airway secretions have not been systematically evaluated. The aim of the present study was to determine whether chondroitin sulfate proteoglycans (CSPG) contribute to the insolubility of CF sputum. Sputa obtained from 18 CF patients were incubated with chondroitinase ABC (ChABC) or buffer (control) for 18 h at 37 degrees C, and after centrifugation at 12,000 g, the volume of the insoluble pellet and turbidity of the supernatant were determined as measures of solubility. ChABC caused a 70-90% reduction in supernatant turbidity and a 60-70% decrease in pellet volume of the 13 purulent CF sputa, but had much less effect on the five nonpurulent CF sputa tested. Similar results were obtained with two non-CF purulent and two non-CF, nonpurulent sputa. Gel electrophoresis, Western blot, and slot blot immunoassays with antichondroitin sulfate and antimucin antibodies revealed that purulent sputa (CF and non-CF) contained more CSPG and less mucin than nonpurulent sputa. In vitro mixing experiments showed that mucin in nonpurulent sputa was reduced upon incubation with purulent sputa, presumably because of degradation or a loss of immunoreactive mucin epitopes from leukocyte and/or bacterial enzymes present in purulent sputa. Our results suggest that CSPG contribute more significantly than mucins to the insolubility of purulent tracheobronchial secretions from CF patients. Because purulent sputa from non-CF patients showed a similar pattern, our observations with CF sputa may have wider applicability.

SEA (sea-urchin sperm protein, enterokinase and agrin)-module cleavage, association of fragments and membrane targeting of rat intestinal mucin Muc3.

In a previous study we showed, by transient expression studies in COS-1 cells, that the C-terminal domain of rat intestinal membrane mucin Muc3 was cleaved between glycine and serine within a GSIVV (one-letter) amino acid sequence during its residence in the endoplasmic reticulum. The extracellular domain fragment remained linked to the membrane-associated fragment by non-covalent interactions. The present study demonstrates that cleavage depends not only on the presence of the G/SIVV site (where G/S is the glycine downward arrow serine cleavage site), but also on more distant C-terminal sequences in the SEA (sea-urchin sperm protein, enterokinase and agrin) module. Inhibition of N-glycosylation by tunicamycin treatment of transfected cells did not prevent re-association of fragments, although cleavage was partially impaired, as some of the non-glycosylated, non-cleaved products were seen to accumulate in cells. Membrane targeting of the Muc3 domain and its cleavage products occurred in transfected cells and was not impaired in mutants in which the cleavage site was mutated. Targeting was also not impaired for products devoid of N-linked oligosaccharides. Our studies thus indicate that (a) cleavage within the SEA module of rat Muc3 requires participation of peptide sequences located C-terminal of and distant from the cleavage site, (b) re-association of the fragments requires the SEA module, but is independent of N-linked oligosaccharides, and (c) membrane targeting of the mucin is independent of the SEA-module-cleavage reaction.

C-terminal domain of rodent intestinal mucin Muc3 is proteolytically cleaved in the endoplasmic reticulum to generate extracellular and membrane components.

Although human MUC3 and rodent Muc3 are both membrane-associated intestinal mucins, the present study has explored the possibility that rodent Muc3 might exist in soluble as well as membrane forms. No evidence was obtained for the existence of soluble splice variants; however, experiments with heterologous cells transfected with cDNA encoding the 381-residue C-terminal domain of rodent Muc3 showed that a definitive proteolytic cleavage occurs during processing in the endoplasmic reticulum. The products consisted of a V5-tagged 30 kDa extracellular glycopeptide and a Myc-tagged 49 kDa membrane-associated glycopeptide. Throughout their cellular transport to the plasma membrane, the two fragments remained associated by non-covalent SDS-sensitive interactions. Site-specific mutagenesis pinpointed the need for glycine and serine residues in the cleavage sequence Leu-Ser-Lys-Gly-Ser-Ile-Val-Val, which is localized between the two epidermal-growth-factor-like motifs of the mucin. A similar cleavage sequence (Phe-Arg-Pro-Gly downward arrow Ser-Val-Val-Val, where downward arrow signifies the cleavage site) has been reported in human MUC1 and analogous sites are present in human MUC3, MUC12 and MUC17. Thus early proteolytic cleavage may be a conserved characteristic of many membrane-associated mucins, possibly as a prelude to later release of their large extracellular domains at cell surfaces.