Structure-function relationships in the beta-cell K(ATP) channel

The ATP-sensitive potassium (K(ATP)) channel plays a key role in controlling beta-cell membrane potential and insulin secretion. The channels are composed of two subunits, Kir6.2, which forms the channel pore, and SUR1, which contains binding sites for nucleotides and sulphonylureas and acts as a channel regulator. Our current studies are aimed at delineating the molecular interactions involved in assembly and ligand binding by K(ATP) channel proteins. We have employed a complementation approach in which SUR1 half-molecules are co-expressed in insect cells using a baculovirus system. Together with data from truncated SUR1 molecules and a fusion protein in which SUR1 is linked to Kir6.2, we have interpreted our findings in terms of a model for the structure of the K(ATP) channel. The main features of the model are: (i) the C-terminal end of SUR1 is close to the N-terminus of Kir6.2; (ii) the two nucleotide binding domains (NBDs) of SUR1--NBD1 and NBD2--are in proximity; (iii) transmembrane helix 12 of SUR1 is orientated in such a way that it can make contact with Kir6.2; (iv) formation of the glibenclamide binding site requires that the two cytosolic loops (CLs) CL3 and CL8 are located close to each other; (v) there are homomeric interactions between the NBD1 domains of neighbouring subunits. We suggest that binding of glibenclamide leads to conformational changes in CL3 and CL8 leading to rearrangement of transmembrane helices. These effects are transmitted to Kir6.2 to result in channel closure.     

Molecular modelling of K(ATP) channel blockers-ADP/ ATP carrier interactions

The modelling of molecule-molecule interactions has been widely accepted as a tool for drug discovery and development studies. However, this powerful technique is unappreciated in physiological and biochemical studies, where it could be extremely useful for understanding the mechanisms of action of various compounds in cases when experimental data are controversial due to complexity of the investigated systems. In this study, based on the biochemical data suggesting involvement of mitochondrial ADP/ATP carrier in K+ and H+ transport to mitochondrial matrix molecular modelling is applied to elucidate the possible interactions between the ADP/ATP carrier and its putative ligands--K(ATP) channel blockers glybenclamide, tolbutamide and 5-hydroxydecanoate. Results revealed that K(ATP) channel blockers could bind to the specific location proximal to H1, H4, H5 and H6 transmembrane helices within the cavity of the ADP/ ATP carrier. Analysis of the predicted binding site suggests that K(ATP) channel blockers could interfere with both the ADP/ATP translocation and possible cation flux through the ADP/ATP carrier, and supports the hypothesis that the ADP/ATP carrier is a target of K(ATP) channel modulators.     

How ATP inhibits the open K(ATP) channel

ATP-sensitive potassium (K(ATP)) channels are composed of four pore-forming Kir6.2 subunits and four regulatory SUR1 subunits. Binding of ATP to Kir6.2 leads to inhibition of channel activity. Because there are four subunits and thus four ATP-binding sites, four binding events are possible. ATP binds to both the open and closed states of the channel and produces a decrease in the mean open time, a reduction in the mean burst duration, and an increase in the frequency and duration of the interburst closed states. Here, we investigate the mechanism of interaction of ATP with the open state of the channel by analyzing the single-channel kinetics of concatenated Kir6.2 tetramers containing from zero to four mutated Kir6.2 subunits that possess an impaired ATP-binding site. We show that the ATP-dependent decrease in the mean burst duration is well described by a Monod-Wyman-Changeux model in which channel closing is produced by all four subunits acting in a single concerted step. The data are inconsistent with a Hodgkin-Huxley model (four independent steps) or a dimer model (two independent dimers). When the channel is open, ATP binds to a single ATP-binding site with a dissociation constant of 300 microM.     

Investigation of the molecular assembly of beta-cell K(ATP) channels

We have investigated the protein interactions involved in the assembly of pancreatic beta-cell ATP-sensitive potassium channels. The channels are a heterooligomeric complex of pore-forming Kir6.2 subunits and sulfonylurea receptor (SUR1) subunits. SUR1 belongs to the ATP binding cassette (ABC) family of proteins and has two nucleotide binding domains (NBD1 and NBD2) and 17 putative transmembrane (TM) sequences. Previously we showed that co-expression in a baculovirus expression system of two parts of SUR1 divided at Pro1042 between TM12 and 13 leads to restoration of glibenclamide binding activity, whereas expression of either individual N- or C-terminal domain alone gave no glibenclamide binding activity [M.V. Mikhailov and S.J.H. Ashcroft (2000) J. Biol. Chem. 275, 3360-3364]. Here we show that the two half-molecules formed by division of SUR1 between NBD1 and TM12 or between TM13 and 14 also self-assemble to give glibenclamide binding activity. However, deletion of NBD1 from the N-part of SUR1 abolished SUR1 assembly, indicating a critical role for NBD1 in SUR1 assembly. We found that differences in glibenclamide binding activity obtained after co-expression of different half-molecules are attributable to different amounts of binding sites, but the binding affinities remained nearly the same. Simultaneous expression of Kir6.2 resulted in enhanced glibenclamide binding activity only when the N-half of SUR1 included TM12. We conclude that TM12 and 13 are not essential for SUR1 assembly whereas TM12 takes part in SUR1 Kir6.2 interaction. This interaction is specific for Kir 6.2 because no enhancement of glibenclamide binding was observed when half-molecules were expressed together with Kir4.1. We propose a model of K(ATP) channel organisation based on these data.     

Molecular determinants of cardiac K(ATP) channel activation by epoxyeicosatrienoic acids

We have previously reported that epoxyeicosatrienoic acids (EETs), the cytochrome P450 epoxygenase metabolites of arachidonic acid, are potent stereospecific activators of the cardiac K(ATP) channel. The epoxide group in EET is critical for reducing channel sensitivity to ATP, thereby activating the channel. This study is to identify the molecular sites on the K(ATP) channels for EET-mediated activation. We investigated the effects of EETs on Kir6.2delta C26 with or without the coexpression of SUR2A and on Kir6.2 mutants of positively charged residues known to affect channel activity coexpressed with SUR2A in HEK293 cells. The ATP IC50 values were significantly increased in Kir6.2 R27A, R50A, K185A, and R201A but not in R16A, K47A, R54A, K67A, R192A, R195A, K207A, K222A, and R314A mutants. Similar to native cardiac K(ATP) channel, 5 microM 11,12-EET increased the ATP IC50 by 9.6-fold in Kir6.2/SUR2A wild type and 8.4-fold in Kir6.2delta C26. 8,9- and 14,15-EET regioisomers activated the Kir6.2 channel as potently as 11,12-EET. 8,9- and 11,12-EET failed to change the ATP sensitivity of Kir6.2 K185A, R195A, and R201A, whereas their effects were intact in the other mutants. 14,15-EET had a similar effect with K185A and R201A mutants, but instead of R195A, it failed to activate Kir6.2R192A. These results indicate that activation of Kir6.2 by EETs does not require the SUR2A subunit, and the region in the Kir6.2 C terminus from Lys-185 to Arg-201 plays a critical role in EET-mediated Kir6.2 channel activation. Based on computer modeling of the Kir6.2 structure, we infer that the EET-Kir6.2 interaction may allosterically change the ATP binding site on Kir6.2, reducing the channel sensitivity to ATP.     

Characterization of low-affinity binding sites for glibenclamide on the Kir6.2 subunit of the beta-cell KATP channel

The ATP-sensitive K+ channel, an octameric complex of two structurally unrelated types of subunits, SUR1 and Kir6.2, plays a central role in the physiological regulation of insulin secretion. The sulfonylurea glibenclamide, which trigger insulin secretion by blocking the ATP-sensitive K+ channel, interacts with both high and low affinity binding sites present on beta-cells. The high affinity binding site has been localized on SUR1 but the molecular nature of the low affinity site is still uncertain. In this study, we analyzed the pharmacology of glibenclamide in a transformed COS-7 cell line expressing the rat Kir6.2 cDNA and compared with that of the MIN6 beta cell line expressing natively both the Kir6.2 and the SUR1 subunits. Binding studies and Scatchard analysis revealed the presence of a single class of low affinity binding sites for glibenclamide on the COS/Kir6.2 cells with characteristics similar to that observed for the low affinity site of the MIN6 beta cells.     

ATP interaction with the open state of the K(ATP) channel

The mechanism of ATP-sensitive potassium (K(ATP)) channel closure by ATP is unclear, and various kinetic models in which ATP binds to open or to closed states have previously been presented. Effects of phosphatidylinositol bisphosphate (PIP2) and multiple Kir6.2 mutations on ATP inhibition and open probability in the absence of ATP are explainable in kinetic models where ATP stabilizes a closed state and interaction with an open state is not required. Evidence that ATP can in fact interact with the open state of the channel is presented here. The mutant Kir6.2[L164C] is very sensitive to Cd2+ block, but very insensitive to ATP, with no significant inhibition in 1 mM ATP. However, 1 mM ATP fully protects the channel from Cd2+ block. Allosteric kinetic models in which the channel can be in either open or closed states with or without ATP bound are considered. Such models predict a pedestal in the ATP inhibition, i.e., a maximal amount of inhibition at saturating ATP concentrations. This pedestal is predicted to occur at >50 mM ATP in the L164C mutant, but at >1 mM in the double mutant L164C/R176A. As predicted, ATP inhibits Kir6.2[L164C/R176A] to a maximum of approximately 40%, with a clear plateau beyond 2 mM. These results indicate that ATP acts as an allosteric ligand, interacting with both open and closed states of the channel.     

Molecular basis and structural insight of vascular K(ATP) channel gating by S-glutathionylation

The vascular ATP-sensitive K(+) (K(ATP)) channel is targeted by a variety of vasoactive substances, playing an important role in vascular tone regulation. Our recent studies indicate that the vascular K(ATP) channel is inhibited in oxidative stress via S-glutathionylation. Here we show evidence for the molecular basis of the S-glutathionylation and its structural impact on channel gating. By comparing the oxidant responses of the Kir6.1/SUR2B channel with the Kir6.2/SUR2B channel, we found that the Kir6.1 subunit was responsible for oxidant sensitivity. Oxidant screening of Kir6.1-Kir6.2 chimeras demonstrated that the N terminus and transmembrane domains of Kir6.1 were crucial. Systematic mutational analysis revealed three cysteine residues in these domains: Cys(43), Cys(120), and Cys(176). Among them, Cys(176) was prominent, contributing to >80% of the oxidant sensitivity. The Kir6.1-C176A/SUR2B mutant channel, however, remained sensitive to both channel opener and inhibitor, which indicated that Cys(176) is not a general gating site in Kir6.1, in contrast to its counterpart (Cys(166)) in Kir6.2. A protein pull-down assay with biotinylated glutathione ethyl ester showed that mutation of Cys(176) impaired oxidant-induced incorporation of glutathione (GSH) into the Kir6.1 subunit. In contrast to Cys(176), Cys(43) had only a modest contribution to S-glutathionylation, and Cys(120) was modulated by extracellular oxidants but not intracellular GSSG. Simulation modeling of Kir6.1 S-glutathionylation suggested that after incorporation to residue 176, the GSH moiety occupied a space between the slide helix and two transmembrane helices. This prevented the inner transmembrane helix from undergoing conformational changes necessary for channel gating, retaining the channel in its closed state.     

Crystal structure of E339K mutated human glucokinase reveals changes in the ATP binding site

Human glucokinase (GK) plays an important role in glucose homeostasis. An E339K mutation in GK was recently found to be associated with hyperglycemia. It showed lower enzyme activity and impaired protein stability compared to the wild-type enzyme. Here, we present the crystal structure of E339K GK in complex with glucose. This mutation results in a conformational change of His416, spatially interfering with adenosine-triphosphate (ATP) binding. Furthermore, Ser411 at the ATP binding site is phosphorylated and then hydrogen bonded with Thr82, physically blocking the ATP binding. These findings provide structural basis for the reduced activity of this mutant.     

Interaction of vanadate with the cloned beta cell K(ATP) channel.

Vanadate is used as a tool to trap magnesium nucleotides in the catalytic site of ATPases. However, it has also been reported to activate ATP-sensitive potassium (K(ATP)) channels in the absence of nucleotides. K(ATP) channels comprise Kir6.2 and sulfonylurea receptor subunits (SUR1 in pancreatic beta cells, SUR2A in cardiac and skeletal muscle, and SUR2B in smooth muscle). We explored the effect of vanadate (2 mM), in the absence and presence of magnesium nucleotides, on different types of cloned K(ATP) channels expressed in Xenopus oocytes. Currents were recorded from inside-out patches. Vanadate inhibited Kir6.2/SUR1 currents by approximately 50% but rapidly activated Kir6.2/SUR2A ( approximately 4-fold) and Kir6. 2/SUR2B ( approximately 2-fold) currents. Mutations in SUR that abolish channel activation by magnesium nucleotides did not prevent the effects of vanadate. Studies with chimeric SUR indicate that the first six transmembrane domains account for the difference in both the kinetics and the vanadate response of Kir6.2/SUR1 and Kir6. 2/SUR2A. Boiling the vanadate solution, which removes the decavanadate polymers, largely abolished both stimulatory and inhibitory actions of vanadate. Our results demonstrate that decavanadate modulates K(ATP) channel activity via the SUR subunit, that this modulation varies with the type of SUR, that it differs from that produced by magnesium nucleotides, and that it involves transmembrane domains 1-6 of SUR.     

Modeling K(ATP) channel gating and its regulation

ATP-sensitive potassium (K_ATP) channels couple cell metabolism to plasmalemmal potassium fluxes in a variety of cell types. The activity of these channels is primarily determined by intracellular adenosine nucleotides, which have both inhibitory and stimulatory effects. The role of K_ATP channels has been studied most extensively in pancreatic beta-cells, where they link glucose metabolism to insulin secretion. Many mutations in K_ATP channel subunits (Kir6.2, SUR1) have been identified that cause either neonatal diabetes or congenital hyperinsulinism. Thus, a mechanistic understanding of K_ATP channel behavior is necessary for modeling beta-cell electrical activity and insulin release in both health and disease. Here, we review recent advances in the K_ATP channel structure and function. We focus on the molecular mechanisms of K_ATP channel gating by adenosine nucleotides, phospholipids and sulphonylureas and consider the advantages and limitations of various mathematical models of macroscopic and single-channel K_ATP currents. Finally, we outline future directions for the development of more realistic models of K_ATP channel gating.     

Redox regulation of the mitochondrial K(ATP) channel in cardioprotection

The mitochondrial ATP-sensitive potassium channel (mK(ATP)) is important in the protective mechanism of ischemic preconditioning (IPC). The channel is reportedly sensitive to reactive oxygen and nitrogen species, and the aim of this study was to compare such species in parallel, to build a more comprehensive picture of mK(ATP) regulation. mK(ATP) activity was measured by both osmotic swelling and Tl(+) flux assays, in isolated rat heart mitochondria. An isolated adult rat cardiomyocyte model of ischemia-reperfusion (IR) injury was also used to determine the role of mK(ATP) in cardioprotection by nitroxyl. Key findings were as follows: (i) mK(ATP) was activated by O(2)(-) and H(2)O(2) but not other peroxides. (ii) mK(ATP) was inhibited by NADPH. (iii) mK(ATP) was activated by S-nitrosothiols, nitroxyl, and nitrolinoleate. The latter two species also inhibited mitochondrial complex II. (iv) Nitroxyl protected cardiomyocytes against IR injury in an mK(ATP)-dependent manner. Overall, these results suggest that the mK(ATP) channel is activated by specific reactive oxygen and nitrogen species, and inhibited by NADPH. The redox modulation of mK(ATP) may be an underlying mechanism for its regulation in the context of IPC. This article is part of a Special Issue entitled: Mitochondria and Cardioprotection.     

The I182 region of k(ir)6.2 is closely associated with ligand binding in K(ATP) channel inhibition by ATP

The ATP-inhibited potassium (K(ATP)) channel is assembled from four inward rectifier potassium (K(ir)6.x) subunits and four sulfonylurea receptor (SURx) subunits. The inhibitory action of ATP is mediated by at least two distinct functional domains within the C-terminal cytoplasmic tail of K(ir)6.2. The G334D mutation of K(ir)6.2 virtually eliminates ATP-dependent gating with no effect on ligand-independent gating, suggesting a role in linkage of the site to the gate or in the ATP binding site, itself. The T171A mutation of K(ir)6.2 strongly disrupts both ATP-dependent and ligand-independent gating, suggesting a role for T171 in the gating step. A neighboring mutation, I182Q, virtually eliminates ATP inhibition, but its effect on ligand-independent gating remained unknown. We have now characterized both the K(i) values for inhibition by ATP and the ligand-independent gating kinetics of 15 substitutions at position 182. All substitutions decreased ATP-dependent inhibition gating as measured by the K(i), many profoundly so, yet had little or no effect on ligand-independent gating kinetics. Thus, substitutions at position 182 are unlikely to act by disrupting inhibition gate movement. Our results indicate an indispensable role for I182 in a step of the ATP binding mechanism, the linkage mechanism coupling the ATP binding site to the inhibition gate, or both.     

The tolbutamide site of SUR1 and a mechanism for its functional coupling to K(ATP) channel closure.

Micromolar concentrations of tolbutamide will inhibit (SUR1/K(IR)6. 2)(4) channels in pancreatic beta-cells, but not (SUR2A/K(IR)6.2)(4) channels in cardiomyocytes. Inhibition does not require Mg(2+) or nucleotides and is enhanced by intracellular nucleotides. Using chimeras between SUR1 and SUR2A, we show that transmembrane domains 12-17 (TMD12-17) are required for high-affinity tolbutamide inhibition of K(ATP) channels. Deletions demonstrate involvement of the cytoplasmic N-terminus of K(IR)6.2 in coupling sulfonylurea-binding with SUR1 to the stabilization of an interburst closed configuration of the channel. The increased efficacy of tolbutamide by nucleotides results from an impairment of their stimulatory action on SUR1 which unmasks their inhibitory effects. The mechanism of inhibition of beta-cell K(ATP) channels by sulfonylureas during treatment of non-insulin-dependent diabetes mellitus thus involves two components, drug-binding and conformational changes within SUR1 which are coupled to the pore subunit through its N-terminus and the disruption of nucleotide-dependent stimulatory effects of the regulatory subunit on the pore. These findings uncover a molecular basis for an inhibitory influence of SUR1, an ATP-binding cassette (ABC) protein, on K(IR)6.2, a ion channel subunit.     

Fetal and adult cerebral artery K(ATP) and K(Ca) channel responses to long-term hypoxia.

High-altitude long-term hypoxia (LTH) alters cerebral vascular contractile and relaxation responses in both fetus and adult. We tested the hypotheses that LTH-mediated vascular responses were secondary to altered K+ channel function and that in the fetus these responses differ from those of the adult. In middle cerebral arteries (MCA) from both nonpregnant adult and fetal (approximately 140 days gestation) sheep, which were either acclimatized to high altitude (3,820 m) or sea-level controls, we measured norepinephrine (NE)-induced contractions and intracellular Ca2+ concentration ([Ca2+]i) simultaneously, in the presence or absence of different K+ channel openers or blockers. In adult MCA, LTH was associated with approximately 20% decrease in NE-induced tension and [Ca2+]i, with a significant increase in Ca2+ sensitivity. In contrast, in fetal MCA, LTH failed to affect significantly NE-induced contraction or [Ca2+]i but significantly decreased the ATP-sensitive K+ (K(ATP)) channel and Ca2+-activated K+ (K(Ca)) channel-mediated relaxation. The significant effect of K(ATP) and K(Ca) channel activators on the relaxation responses and the fact that K+ channels play a key role in myogenic tone support the hypotheses that K+ channels play an important role in hypoxia-mediated responses. These results also support the hypothesis of significant developmental differences with maturation from fetus to adult.     

K(ATP) channel activation reduces the severity of postresuscitation myocardial dysfunction

Postresuscitation myocardial dysfunction has been recognized as a leading cause of the high postresuscitation mortality rate. We investigated the effects of ischemic preconditioning and activation of ATP-sensitive K(+) (K(ATP)) channels on postresuscitation myocardial function. Ventricular fibrillation (VF) was induced in 25 Sprague-Dawley rats. Cardiopulmonary resuscitation (CPR), including mechanical ventilation and precordial compression, was initiated after 4 min of untreated VF. Defibrillation was attempted after 6 min of CPR. The animals were randomized to five groups treated with 1) ischemic preconditioning, 2) K(ATP) channel opener, 3) ischemic preconditioning with K(ATP) channel blocker administered 1 min after VF, 4) K(ATP) channel blocker administered 45 min before induction of ischemic preconditioning, and 5) placebo. Postresuscitation myocardial function, as measured by the rate of left ventricular pressure increase at 40 mmHg, the rate of left ventricular decline, cardiac index, and duration of survival, was significantly improved in both preconditioned and K(ATP) channel opener-treated animals. K(ATP) channel blocker administered 45 min before induction of ischemic preconditioning completely abolished the myocardial protective effects of preconditioning. We conclude that ischemic preconditioning significantly improved post-CPR myocardial function and survival. These results also provide evidence that the myocardial protective effects of ischemic preconditioning are mediated by K(ATP) channel activation.     

A specific binding site for K+ channel openers in rat aorta.

The K+ channel openers, including cromakalim, pinacidil, minoxidil sulfate, diazoxide, and nicorandil, form a chemically heterogeneous group of compounds, which relax smooth muscle by opening plasmalemmal K+ channels. At present it is not known whether these drugs elicit their effects by binding to the same target, presumably the K+ channel. In order to address this question, a binding assay for K+ channel openers has been developed in vascular smooth muscle. The novel tritiated K+ channel opener, [3H]P1075, an analogue of pinacidil, binds with high affinity (KD = 6 +/- 1 nM) to endothelium-denuded rings of rat aorta. Inhibition studies indicate that the different families of K+ channel openers bind to a common target. Evidence is presented to suggest that the binding site for the sulfonylurea, glibenclamide, the major blocker of the K+ channel openers, is coupled in a negative allosteric manner to the binding site(s) for the openers. The binding assay described here may open the way to the biochemical characterization of the drug receptor for the K+ channel openers.     

Initiation of remote microvascular preconditioning requires K(ATP) channel activity

The purpose of this study was to investigate vascular preconditioning of individual microvascular networks. Prior work shows that exposure of downstream arterioles to specific agonists preconditions upstream arterioles so that they exhibit an altered local vasoactive response [remote microvascular preconditioning (RMP)]. We hypothesized that mitochondrial ATP-sensitive K+ (K(ATP)) channels were involved in stimulation of RMP. Arteriolar diameter (approximately 15 microm) was observed approximately 1,000 microm upstream of the remote exposure site in the cheek pouch of pentobarbital sodium-anesthetized (70 mg/kg) male hamsters (n = 104); all agonists were applied via micropipette. RMP was initiated by application of pinacidil (Pin), diazoxide (DZ), sodium nitroprusside (SNP), or bradykinin (BK) to the downstream vessel. After 15 min, RMP was apparent at the upstream observation site from testing of local vasoactive responses to L-arginine. Pin, DZ, SNP, and BK each stimulated RMP. To evaluate a specific role for mitochondrial K(ATP) channels in this response, 5-hydroxydecanoate was applied (via a 2nd pipette) during downstream stimulation with agonist. 5-Hydroxydecanoate blocked RMP initiated by Pin, DZ, or SNP, suggesting that mitochondrial K(ATP) channels are involved before SNP signal transduction. To verify this, we applied N(omega)-nitro-L-arginine during DZ or SNP stimulation. RMP was blocked during SNP, but not during DZ, stimulation. Thus stimulation of the RMP response requires mitochondrial K(ATP) channel activity after stimulation by nitric oxide donors.     

Interactions of the sulfonylurea receptor 1 subunit in the molecular assembly of beta-cell K(ATP) channels

We have investigated protein interactions involved in pancreatic beta-cell ATP-sensitive potassium channel assembly. These channels, which are of key importance for control of insulin release, are a hetero-oligomeric complex of pore-forming Kir6.2 subunits and sulfonylurea receptor (SUR1) subunits with two nucleotide-binding domains (NBD1 and NBD2). We divided SUR1 into two halves at Pro-1042. Expression of either the individual N- or C-terminal domain in a baculovirus expression system did not lead to glibenclamide binding activity, although studies with green fluorescent protein fusion proteins showed that both half-molecules were inserted into the plasma membrane. However, significant glibenclamide binding activity was observed when the half-molecules were co-expressed (even when NBD2 was deleted from the C-terminal half-molecule). Simultaneous expression of Kir6.2 resulted in enhanced glibenclamide binding activity. We conclude that the glibenclamide-binding site includes amino acid residues from both halves of the molecule, that there is strong interaction between different regions of SUR1, that NBD2 is not essential for glibenclamide binding, and that interactions between Kir6.2 and SUR1 participate in ATP-sensitive potassium channel assembly. Investigation of NBD1-green fluorescent protein fusion protein distribution inside insect cells expressing C-terminal halves of SUR1 demonstrated strong interaction between NBD1 and NBD2. We also expressed and purified NBD1 from Escherichia coli. Purified NBD1 was found to exist as a tetramer indicating strong homomeric attractions and a possible role for NBD1 in SUR1 assembly.