The cystic fibrosis transmembrane conductance regulator (CFTR) can conduct both Cl^– and HCO₃^– ions although there is a preference for Cl^– (Kunzelmann et al. 1991). In rat colon, direct manipulation of intracellular cAMP can stimulate HCO₃^– secretion through a pathway dependent on CFTR (Cuthbert et al. 1999). Here we have examined the effects of carbachol on HCO₃^– secretion through CFTR in rat distal colon.

The terminal segment of colon (5 cm) was removed from adult male Wistar rats that had been humanely killed by cervical dislocation. After washing in Krebs solution (mM: NaCl 119, KCl 4.7, NaHCO₃ 24.8, MgSO₄ 1.2, KH₂PO₄ 1.2, CaCl₂ 2.5, glucose 11.1) the smooth muscle was removed and the mucosal sheet mounted in an Ussing chamber (area 0.5 cm²) where apical and basolateral baths contained Krebs solution at 37 °C, gassed with 95 % O₂ and 5 % CO₂. Ion replacement experiments were performed using low-Cl^– Krebs solution prepared by iso-osmotically replacing NaCl with sodium gluconate. In addition, Cl^– and HCO₃^– were replaced using the following solution (mM: sodium gluconate 130, potassium gluconate 5, MgSO₄ 1.2, calcium gluconate 5.8, Hepes 10, glucose 10, pH 7.4 with NaOH). All values represent the change in I_SC (ΔI_SC) from baseline (mean ± S.E.M.), and Student’s unpaired t test was used for statistical analysis (P < 0.05 was considered significant).

Spontaneous short-circuit current (I_SC) was 17.0 ± 2.0 mA cm^-2 (n = 60) with a resistance of 105 ± 4 Ω cm^-2 and an open circuit transepithelial potential difference of -2.0 ± 0.2 mV. Basolateral carbachol (100 mM, n = 14) stimulated a triphasic response consisting of an initial increase in I_SC (phase 1, 12.4 ± 2.0 mA cm^-2) followed by a decrease (phase 2, 6.3 ± 2.4 mA cm^-2) then a large increase (phase 3, 55.9 ± 8.3 mA cm^-2) that slowly decayed to a stable plateau. Apical amiloride (100 mM, n = 19) did not alter the carbachol-stimulated response. Basolaterally applied bumetanide (100 mM, n = 10) significantly attenuated all three phases of the carbachol-stimulated ΔI_SC (phase 1, 4.6 ± 2.0 mA cm^-2, P < 0.02; phase 2, 0.9 ± 0.7 mA cm^-2, P < 0.05; phase 3, 7.4 ± 0.9 mA cm^-2, P < 0.001). The carbachol-evoked ΔI_SC was also significantly reduced compared with control, but not abolished, in low-Cl^– Krebs solution (P < 0.02 for all three phases, n = 10). In Cl^–– and HCO₃^–-free conditions, however, the carbachol-evoked ΔI_SC was abolished (n = 3). In low-Cl^– conditions where there is still a residual HCO₃^– secretion in response to carbachol, DPC (100 mM apically and basolaterally, n = 6) was found to significantly reduce phases 1 and 3 of the ΔI_SC (phase 1, 0.5 ± 0.1 mA cm^-2, P < 0.01; phase 3, 5.0 ± 3.2 mA cm^-2, P < 0.05).

These results suggest that carbachol-stimulated ΔI_SC is due to both Cl^– and HCO₃^– secretion, although Cl^– secretion is predominant, similar to results in porcine bronchial epithelia with acetylcholine (Trout et al. 1998). Blocking the Cl^– component of ΔI_SC, using bumetanide or low-Cl^–, reveals the HCO₃^– current, which can be significantly attenuated by the CFTR channel blocker DPC, suggesting that HCO₃^– is carried through CFTR in rat distal colon.

M.B. is supported by a Glasgow Caledonian University Research Studentship.

All procedures accord with current UK legislation.