The Nucleotide-Binding Sites of SUR1: A Mechanistic Model

Abstract

ATP-sensitive potassium (KATP) channels comprise four pore-forming Kir6.2 subunits and four modulatory sulfonylurea receptor (SUR) subunits. The latter belong to the ATP-binding cassette family of transporters. KATP channels are inhibited by ATP (or ADP) binding to Kir6.2 and activated by Mg-nucleotide interactions with SUR. This dual regulation enables the KATP channel to couple the metabolic state of a cell to its electrical excitability and is crucial for the KATP channel's role in regulating insulin secretion, cardiac and neuronal excitability, and vascular tone. Here, we review the regulation of the KATP channel by adenine nucleotides and present an equilibrium allosteric model for nucleotide activation and inhibition. The model can account for many experimental observations in the literature and provides testable predictions for future experiments.

Figure 1

Domain organization of the K_ATP complex. (A) Hetero-octameric complex of the K_ATP channel, showing the Kir6.x tetrameric pore (Kir6.1 or Kir6.2) surrounded by four SUR subunits. (B) Schematic representation of Kir6.x and SUR protein topologies, indicating the three hydrophobic TMDs (TMD0, TMD1, and TMD2) and the two NBDs (NBD1 and NBD2) of SUR.

Figure 2

Structure of the NBSs. (A) Homology model of the NBSs of SUR1 based on the heteromeric structure of TM287/288 (18). (B) Schematic representation illustrating the functionally important regions of the NBSs.

Figure 3

Equilibrium gating model of Kir6.2/SUR1. (A) Schematic representing the gating of K_ATP channels as three interacting domains: the pore, the inhibitory NBS on Kir6.2, and the NBSs of SUR1. (B) Gating scheme in (A) expanded into a cubic model. C and O designate the closed and open states of the pore domain, respectively. The subscripts designate the occupancy of the two NBSs as either unbound (0) or nucleotide bound (A). The first subscript refers to the inhibitory site on Kir6.2 and the second refers to the NBDs. When both inhibition and activation by nucleotides are present, the open probability (P_o) is described by $P_o=\frac{K_{1}LD+K_{2}LE+\frac{L}{\left[ANP\right]}+K_1{K}_{2}LDEF\left[ANP\right]}{K_1+K_2+K_{1}LD+K_{2}LE+\frac{1}{\left[ANP\right]}+\frac{L}{\left[ANP\right]}+K_1{K}_{2}F\left[ANP\right]+K_1{K}_{2}LDEF\left[ANP\right]},$ where [ANP] is the nucleotide concentration and all other symbols are as described in the main text. (C) Cubic model (solid lines) fit to the nucleotide activation/inhibition data (open circles) from Proks et al. (33). Because the experimental data were normalized, the model was also normalized. For activation curves, this was achieved by subtracting the P_o at 0 nucleotide (L/(L + 1)) and dividing the difference by the maximum P_o (EL/(EL + 1)) minus the unliganded P_o. For inhibition curves, the equation was divided by the P_o in the absence of nucleotide (L/(L + 1)). When both activation and inhibition were present, the model was normalized to the P_o in the absence of nucleotide. (D) Simplified scheme for K_ATP activation via Mg-nucleotide binding at SUR1. This schematic assumes that SUR1 can hydrolyze MgATP, but that hydrolysis is not necessary for channel activation by MgATP.