Mucin 5B (MUC5B) has an essential role in mucociliary clearance that protects the pulmonary airways. Accordingly, knowledge of MUC5B structure and its interactions with itself and other proteins is critical to better understand airway mucus biology and improve the management of lung diseases such as asthma, cystic fibrosis, and chronic obstructive pulmonary disease (COPD). The role of an N-terminal multimerization domain in the supramolecular organization of MUC5B has been previously described, but less is known about its C-terminal dimerization domain. Here, using cryogenic electron microscopy (cryo-EM) and small-angle X-ray scattering (SAXS) analyses of recombinant disulfide-linked dimeric MUC5B dimerization domain we identified an asymmetric, elongated twisted structure, with a double globular base. We found that the dimerization domain is more resistant to disruption than the multimerization domain suggesting the twisted structure of the dimerization domain confers additional stability to MUC5B polymers. Size-exclusion chromatography-multiangle light scattering (SEC-MALS), SPR-based biophysical analyses and microscale thermophoresis of the dimerization domain disclosed no further assembly, but did reveal reversible, calcium-dependent interactions between the dimerization and multimerization domains that were most active at acidic pH, suggesting that these domains have a role in MUC5B intragranular organization. In summary, our results suggest a role for the C-terminal dimerization domain of MUC5B in compaction of mucin chains during granular packaging via interactions with the N-terminal multimerization domain. Our findings further suggest that the less stable multimerization domain provides a potential target for mucin depolymerization to remove mucus plugs in COPD and other lung pathologies.

The colonic epithelial surface is protected by an inner mucus layer that the commensal microflora cannot penetrate. We previously demonstrated that Entamoeba histolytica secretes a protease capable of dissolving this layer that is required for parasite penetration. Here, we asked whether there are bacteria that can secrete similar proteases. We screened bacterial culture supernatants for such activity using recombinant fragments of the MUC2 mucin, the major structural component, and the only gel-forming mucin in the colonic mucus. MUC2 has two central heavily O-glycosylated mucin domains that are protease-resistant and has cysteine-rich N and C termini responsible for polymerization. Culture supernatants of Porphyromonas gingivalis, a bacterium that secretes proteases responsible for periodontitis, cleaved the MUC2 C-terminal region, whereas the N-terminal region was unaffected. The active enzyme was isolated and identified as Arg-gingipain B (RgpB). Two cleavage sites were localized to IR↓TT and NR↓QA. IR↓TT cleavage will disrupt the MUC2 polymers. Because this site has two potential O-glycosylation sites, we tested whether recombinant GalNAc-transferases (GalNAc-Ts) could glycosylate a synthetic peptide covering the IRTT sequence. Only GalNAc-T3 was able to glycosylate the second Thr in IRTT, rendering the sequence resistant to cleavage by RgpB. Furthermore, when GalNAc-T3 was expressed in CHO cells expressing the MUC2 C terminus, the second threonine was glycosylated, and the protein became resistant to RgpB cleavage. These findings suggest that bacteria can produce proteases capable of dissolving the inner protective mucus layer by specific cleavages in the MUC2 mucin and that this cleavage can be modulated by site-specific O-glycosylation.