Incomplete dissociation of glibenclamide from wild-type and mutant pancreatic K ATP channels limits their recovery from inhibition

Abstract

Background and purpose: The antidiabetic sulphonylurea, glibenclamide, acts by inhibiting the pancreatic ATP-sensitive K(+) (K(ATP)) channel, a tetradimeric complex of K(IR)6.2 and sulphonylurea receptor 1 (K(IR)6.2/SUR1)(4). At room temperature, recovery of channel activity following washout of glibenclamide is very slow and cannot be measured. This study investigates the relation between the recovery of channel activity from glibenclamide inhibition and the dissociation rate of [(3)H]-glibenclamide from the channel at 37 degrees C.

Experimental approach: K(IR)6.2, K(IR)6.2DeltaN5 or K(IR)6.2DeltaN10 (the latter lacking amino-terminal residues 2-5 or 2-10 respectively) were coexpressed with SUR1 in HEK cells. Dissociation of [(3)H]-glibenclamide from the channel and recovery of channel activity from glibenclamide inhibition were determined at 37 degrees C.

Key results: The dissociation kinetics of [(3)H]-glibenclamide from the wild-type channel followed an exponential decay with a dissociation half-time, t(1/2)(D) = 14 min; however, only limited and slow recovery of channel activity was observed. t(1/2)(D) for K(IR)6.2DeltaN5/SUR1 channels was 5.3 min and recovery of channel activity exhibited a sluggish sigmoidal time course with a half-time, t(1/2)(R) = 12 min. t(1/2)(D) for the DeltaN10 channel was 2.3 min; recovery kinetics were again sigmoidal with t(1/2)(R) approximately 4 min.

Conclusions and implications: The dissociation of glibenclamide from the truncated channels is the rate-limiting step of channel recovery. The sigmoidal recovery kinetics are in quantitative agreement with a model where glibenclamide must dissociate from all four (or at least three) sites before the channel reopens. It is argued that these conclusions hold also for the wild-type (pancreatic) K(ATP) channel.

Figure 1

Experiments with K_IR6.2/SUR1. a. Association and dissociation of kinetics of [³H]-glibenclamide ([³H]-GBC) in the presence of 1 mmol·L⁻¹ MgATP at 37°C. Association was measured in the presence of [³H]-glibenclamide = 0.73 nmol·L⁻¹. The kinetics were described by an exponential function (Eqn 1 in Methods) and gave an apparent rate constant of association (k_app) of 0.54 ± 0.03 min⁻¹. Dissociation was initiated by addition of 100 nmol·L⁻¹ glibenclamide and the fit of the data to Eqn 3a gave the dissociation rate constant, k_-(D), a value of 0.047 ± 0.003 min⁻¹ corresponding to a half-time of 15 min. b. Channel inhibition by glibenclamide and washout measured in the whole-cell configuration at −30 mV and 37°C. Dialysis of the cell with 1 mmol·L⁻¹ MgATP and 0.3 mmol·L⁻¹ MgGDP elicited a current that showed some run-down. Glibenclamide (3 and 100 nmol·L⁻¹) was applied to the bath as indicated by the grey (3 nmol·L⁻¹) and black bars (100 nmol·L⁻¹). Note the very small recovery of current upon washout of glibenclamide (see text).

Figure 2

Experiments with K_IR6.2ΔN5/SUR1. a. Association and dissociation kinetics of [³H]-glibenclamide ([³H]-GBC) in the presence of MgATP (1 mmol·L⁻¹) and MgGDP (0.3 mmol·L⁻¹). The association kinetics, measured in the presence of [³H]-glibenclamide = 0.89 nmol·L⁻¹, gave k_app = 0.38 ± 0.03 min⁻¹ and the dissociation kinetics initiated by addition of 30 nmol·L⁻¹ glibenclamide (GBC) gave the dissociation rate constant, k_-(D), a value of 0.125 ± 0.009 min⁻¹ corresponding to a half-time of 5.6 min. The inset shows the dependence of k_app on [³H]-glibenclamide concentration (Lo*). From the fit of Eqn 2 to the experimental data, the association rate constant k₊ was determined to be 0.30 ± 0.03 nmol·L⁻¹ min⁻¹ and the equilibrium dissociation constant for glibenclamide binding to the channel was calculated to be 0.43 ± 0.05 nmol·L⁻¹. b and c. Inhibition of the K_IR6.2ΔN5/SUR1 channel by glibenclamide and recovery from block (whole-cell recording at -60 mV). Glibenclamide (30 nmol·L⁻¹, applied for 5 min) abolished the current and washout for 15 min led to substantial recovery. The small bar at the beginning of the traces indicates the zero current level. b. The dotted line represents the exponential kinetics expected from the one-site model (Eqn 4c with k_-(D) = 0.12 min⁻¹) and the continuous curve, the fit to the four-site model (Eqn 4c giving k_-(R) = 0.16 min⁻¹). c. Fit to the four-site model gave k_-(R) = 0.14 min⁻¹.

Figure 3

Inhibition of the K_IR6.2ΔN10/SUR1 channel by glibenclamide (GBC) and recovery. Glibenclamide and Ba²⁺ were applied as indicated by the bars; conditions were as in Figure 2B,C; the small bar at the beginning of the trace indicates the zero current level. The close-ups show the recovery kinetics after washout of 10 and 100 nmol·L⁻¹ glibenclamide and the fit of Eqn 4c, giving k_- values of 0.49 and 0.34 min⁻¹ for 10 and 100 nmol·L⁻¹ respectively.