The non-hydrolytic pathway of cystic fibrosis transmembrane conductance regulator ion channel gating

Abstract

It has been suggested that the cystic fibrosis transmembrane conductance regulator (CFTR) chloride channel may utilize a novel gating mechanism in which open and closed states are not in thermodynamic equilibrium. This suggestion is based on the assumption that energy of ATP hydrolysis drives the gating cycle. We demonstrate that CFTR channel gating occurs in the absence of ATP hydrolysis and hence does not depend on an input of free energy from this source. The binding of ATP or structurally related analogues that are poorly or non-hydrolysable is sufficient to induce opening. Closing occurs on dissociation of these ligands or the hydrolysis products of those that can be cleaved. Not only can channel opening occur without ATP hydrolysis but the temperature dependence of the open probability (Po.) is reversed, i.e. Po. increases as temperature is lowered whereas under hydrolytic conditions, Po. increases as temperature is elevated. This indicates that there are different rate-limiting steps in the alternate gating pathways (hydrolytic and non-hydrolytic). These observations demonstrate that phosphorylated CFTR behaves as a conventional ligand-gated channel employing cytoplasmic ATP as a readily available cytoplasmic ligand; under physiological conditions ligand hydrolysis provides efficient reversibility of channel opening.

Figure 1. Single CFTR channels gated by ATP with excess Mg²⁺ ions

Wild-type CFTR ion channels were recorded in symmetrical solutions: 300 mm NaCl, 2 mm MgCl₂, 1 mm EGTA, 20 mm Hepes/KOH, pH 7.2. Each trace is of 2 min duration. P_o values were calculated from all points histograms which are shown at the left of each trace. Prephosphorylated membrane vesicles and 300 μm Na₂ATP were added to the ‘cis’ compartment only. Transmembrane potential difference V_m=−75 mV. A, +25 °C, P_o= 0.16± 0.04 (n= 4). B, +30 °C, P_o= 0.26± 0.03 (n= 7). C, +35 °C, P_o= 0.38± 0.03 (n= 5).

Figure 2. ATP-gated CFTR channels in Mg²⁺-free buffer

In A-C, wild-type CFTR ion channels were recorded in symmetrical solutions containing: 300 mm NaCl, 100 mm NaF, 20 mm Hepes/KOH, pH 7.2. In D, 100 mm NaF was replaced with 2 mm EDTA. Other experimental details are as in Fig. 1. A, +25 °C, P_o= 0.83± 0.07 (n= 4). B, +30 °C, P_o= 0.51± 0.04 (n= 6). C, +35 °C, P_o= 0.25± 0.04 (n= 3).

Figure 3. CFTR channels gated by ATPγS and AMP-PNP

Wild-type CFTR ion channels were recorded in symmetrical solutions: 300 mm NaCl, 100 mm NaF, 20 mm Hepes/KOH, pH 7.2. Transmembrane potential difference V_m=−75 mV. Traces shown are with 5 mm ATPγS at +25 °C, P_o= 0.27± 0.04 (n= 5) (A); 5 mm ATPγS at +30 °C, P_o= 0.08± 0.04 (n= 5) (B); and 5 mm AMP-PNP at +25 °C, P_o= 0.11± 0.05 (n= 3) (C).

Figure 4. Gating of 8SE CFTR channels by ATP and AMP-PNP

8SE mutant CFTR channels were recorded in symmetrical solutions: 300 mm NaCl, 5 mm MgCl₂, 1 mm EGTA, 20 mm Hepes/KOH, pH 7.2. Membrane vesicles were not phosphorylated. Transmembrane potential difference V_m=−75 mV. Traces shown are with 5 mm MgATP at +25 °C, P_o= 0.09± 0.03 (n= 5) (A); 5 mm MgATP at +30 °C, P_o= 0.28± 0.04 (n= 8) (B); 5 mm MgATP at +35 °C, P_o= 0.52± 0.04 (n= 4) (C); and 5 mm MgAMP-PNP at +25 °C in symmetrical solutions: 300 mm NaCl, 100 mm NaF, 10 mm Hepes/KOH, pH 7.2. P_o= 0.11± 0.04 (n= 3) (D).