MUC5AC drives COPD exacerbation severity through amplification of virus-induced airway inflammation

Aran Singanayagam; Joseph Footitt; Benjamin T Kasdorf; Matthias Marczynski; Michael T Cross; Lydia J Finney; Maria-Belen Trujillo Torralbo; Maria Calderazzo; Jie Zhu; Julia Aniscenko; Thomas B Clarke; Philip L Molyneaux; Nathan W Bartlett; Miriam F Moffatt; William O Cookson; Jadwiga Wedzicha; Christopher M Evans; Oliver Lieleg; Patrick Mallia; Sebastian L Johnston

Ehre et al evaluated responses to influenza infection in a transgenic Muc5ac overexpressing mouse strain and showed reduced viral titres and attenuated neutrophilic inflammation in this model 9 . This contrasts our findings using exogenous MUC5AC administration where there was no effect on virus loads and MUC5AC enhanced virus-induced inflammation. The transgenic strain employed by Ehre et al was associated with ~18-fold increase in airway MUC5AC levels 9 . This level of augmentation greatly exceeds the increases observed in our human analyses where a ~2-fold increase was observed in stable state levels of MUC5AC in COPD versus healthy subjects and ~2-to 3-fold increase was observed from baseline to exacerbation. The concentrations of exogenous MUC5AC protein administered in our animal experiments (0.25-0.5mg/mL) were ~ 1.5-3.5 fold higher than the baseline MUC5AC protein concentrations we measured in the mouse airway (~150µg/mL) and are therefore more likely to be physiologically relevant to human exacerbation. Further, the effect of MUC5AC in reducing virus loads at the much higher MUC5AC levels generated in the MUC5AC transgenic strain reported by Ehre et al may be virus-specific 9 . The authors reported that MUC5AC binds to the influenza virus receptor, α2,3-linked sialic acids in transgenic animals, consistent with a mechanism of influenza protection in vivo through impairment of efficient binding of virus to receptors on the bronchial epithelium leading to reduction in virus loads and consequently lesser inflammation. It is important to note that we used a non-human purified mucin protein in solution for our experiments which does not fully recapitulate the complex biophysical properties of mucin glycoprotein contained within endogenous airway mucus. We confirmed that the mucin preparation used was free of endotoxin and DNA but we cannot exclude that other mucin-bound proteins are present in the solution that may be contributing to the observed effects. Future validation of our findings using administration of human or mouse MUC5AC in similar mouse models is warranted, although extraction and purification of mucins from these sources is technically challenging and was not feasible for the current studies.