Glycosylation differences between a cystic fibrosis and rescued airway cell line are not CFTR dependent

Altered glycosylation of mucus and membrane glycoconjugates could explain reported differences in binding of bacterial pathogens to cystic fibrosis (CF) versus normal tissue. However, because bacteria can alter cell surface glycoconjugates, it is not possible to assess the role of cystic fibrosis transmembrane conductance regulators (CFTR) in glycosylation in these studies. To address this issue, we have developed quantitative lectin binding assays to compare cell surface glycosylation in well-matched immortalized CF cells and rescued cell lines. The CF airway bronchial epithelial cell line IB3-1 consistently bound more peanut agglutinin (PNA) than its clonal derivative S9, which stably expresses functional wild-type CFTR. Pretreatment with neuraminidase increased PNA binding and abolished the difference between the two cell lines. However, infection of the IB3-1 cells with a replication-deficient recombinant adenovirus encoding CFTR restored CFTR function but did not alter PNA binding to cells. In contrast, treatment with the weak base ammonium chloride increased PNA binding to both cell lines as expected. Our data show that even clonally related CF and rescued cells can exhibit significant differences in carbohydrate processing. Although the differences that we found are consistent with the proposed role for CFTR in modulating intraorganellar pH, our data strongly suggest that they are CFTR independent. These studies add a cautionary note to the interpretation of differences in glycosylation between CF and normal primary tissues and immortalized cells.