Cystic fibrosis (CF), a lethal genetic disease, leads to disruption of ion and fluid transport across the airway epithelium, with early, severe pathological changes occurring in the submucosal glands. Inhibition of both chloride and bicarbonate secretion prior to stimulation with the glandular secretagogue acetylcholine leads to mucus occlusion of the glands, resembling the changes seen in CF (Inglis et al. 1997). However, the exact mechanisms underlying ion transport in these glands are not known. Calu-3 human airway epithelial cells exhibit many features of the serous cells of the submucosal glands. When grown to confluence on Snapwells and mounted in Ussing chambers, they demonstrate an increase in short circuit current (I_sc) in response to acetylcholine (ACh) (M.C. Constable, unpublished data). This study investigates the ion transport processes involved in this response.

Calu-3 cells were grown to confluence on Snapwell filters (10 days approximately) before being mounted in Ussing chambers in a HCO₃^–/CO₂ buffer. The effect on short circuit current of 100 µM ACh (basolateral) was noted after pre-treatment with either an anion channel blocker (glibenclamide, diphenylamine-2-carbonsaeure (DPC), or 4,4â-diisothiocyanatostilbene-2,2â-disulfonic acid (DIDS)) or apical dimethylsulfoxide (DMSO). Glibenclamide and DPC are blockers of cystic fibrosis transmembrane conductance regulator (CFTR), while DIDS blocks other chloride channels but not CFTR. Glibenclamide pre-treatment resulted in a 37.2 ±11.7 % (mean ± S.E.M.) inhibition of ACh-stimulated I_sc compared to DMSO (n = 8, P = 0.03, Student’s paired t test, used in all statistical analysis). DPC caused an inhibition of 43.9 ± 15.6 % of the ACh response (n = 8, P = 0.02). However, DIDS caused no significant reduction in I_sc (n = 8, P = 0.68). These blocker sensitivities suggest that the response of Calu-3 cells to ACh is mediated through CFTR.

There was no significant difference between the responses to ACh in Hepes-Krebs solution and HCO₃^–/CO₂ buffers (n = 6, P = 0.38). When a Cl^–-free HCO₃^– solution was used instead of a Cl^–-containing HCO₃^–solution, the response to ACh was reduced by 53.6 ± 11.5 % (n = 9, P = 0.009), suggesting that the response to ACh in Cl^–-containing buffer is at least partly mediated by Cl^– secretion across the apical membrane. When a Cl^–-free Hepes solution was used, the response to acetylcholine was reduced by 82.8 ± 4.1 % relative to Cl^–-containing Hepes (n = 9, P = 0.00003). This supports the hypothesis that HCO₃^– is secreted in response to acetylcholine in HCO₃^–/CO₂ buffered solutions, and is responsible for approximately two-thirds of the residual current in Cl^–-free HCO₃^– buffer.

This work was supported by the Medical Research Council.