Transgenic Humanized AHR Mouse Reveals Differences between ...

PharmSightTM DOI: 10.4255/mcpharmacol.09.29

Molecular and Cellular Pharmacology www.mcpharmacol.com

Compound-Induced Block of Ion Channel Pore Function: Inward-Rectifier Potassium Channels as a Model Kazuharu Furutani, Hiroshi Hibino, Atsushi Inanobe and Yoshihisa Kurachi Department of Pharmacology, Graduate School of Medicine and the Center for Advanced Medical Engineering and Informatics, Osaka University, Suita, Osaka 565-0871, Japan PharmSight on Furutani K et al., Mutational and in silico analyses for antidepressant block of astroglial inward-rectifier Kir4.1 channel. Mol Pharmacol 2009;75:1287-95. __________________________________________________________________________________________________________________

Abstract Small chemical compounds modulate ion channel functions. This is the reflection of ligand interactions with ion channels at their various sites. Many biophysical and biochemical researches have been performed on this subject and have provided important basic concepts on the structure-functional relationships of ion channels. Especially, ion channel blockers have been excellent tools for biophysical studies of ion channels and some of them are actually used for treating various diseases. The mechanisms underlying the blocking action of various chemical compounds, however, remain largely unknown at the atomic level, partly because of the promiscuous nature of the reaction. As one of the attempts to overcome the problem, we have adopted a novel approach combining molecular pharmacology and in silico analyses in the study of block of astroglial Kir4.1 channel by various antidepressants, including nortriptyline and fluoxetine. In molecular pharmacology experiments, we have demonstrated that Thr128 and Glu158 of Kir4.1 facing the central cavity play an important role in determining the sensitivities of the Kir channel to the antidepressants. On the other hand, we abstracted common sets of features from Kir4.1 channel blockers by the computer-aided technique that quantitatively correlates their chemical structures with IC50 values for Kir4.1 channel current block. By combining these two lines of studies, we modeled the channel-drug interaction for Kir4.1block, showing that the compound is ______________________ Received 11/03/09; accepted 12/04/09 Correspondence: Yoshihisa Kurachi, M.D., Ph.D. Division of Molecular and Cellular Pharmacology, Department of Pharmacology, Graduate School of Medicine, Osaka University, 2-2 Yamada-Oka, Suita, Osaka 565-0871, Japan. Tel. 81668793512, Fax. 81668793519. e-mail: [email protected]

Mol Cell Pharmacol 2009;1(5):234-244.

accommodated between Thr128 and Glu158 within the central cavity of the channel. This combined approach may be useful to obtain some insights in the structure-function relationship of various ion channels and will shed light on the basic understandings of ion permeation and block. Keywords: Drug-channel interaction; Kir channels; Antidepressants; Pharmacophore; Central cavity

Introduction: Earlier studies of potassium channel block Ion channel proteins provide a pathway for the rapid passive movement of selected ions across the plasma membrane, and therefore are responsible for essential physiological processes such as cardiac excitation, neural transmission, and hormones secretion (1). Ion channels are inhibited when interacting with particular types of small compounds, the reaction that is referred to as block. Structural and functional studies on permeation and block of the channels have provided pivotal information to understand the basic properties in the biological pores; which amino acids form the wall of the pore (2-7), how ions permeate through the pore (8), how the pore opens and closes (9-13), and how small compounds cause block of ion permeation (14-17). Actually, the majority of our early knowledge about the functional architectures of ion channels have been derived from the pharmacological experiments of ion channel currents (1). In spite of numerous experimental data, it remains unclear how at the atomic level drugs the drugs block the channel current. Recently a number of studies have been performed to examine the mechanisms underlying the interaction between inwardly rectifier

234

235

Kir Channel-drug Interaction

potassium (Kir) channels and drugs or toxins (1821). Because of the simple structure of Kir channels, the results have provided new insights into the blocking reaction of the ion permeation by various substances. This review article is devoted to summarize the recent advances of the molecular pharmacology of Kir channels, with referring to our recent results.

Physiology, pharmacology, and structure of Kir channels Kir channels are a class of K+ channels that can conduct much larger inward currents at membrane voltages negative to the K+ equilibrium potential (EK) than outward currents at voltages positive to EK (22-24). This property is quite different from the delayed rectification observed in voltage-gated K+ (Kv) channels (1). Of all the ion channels, the Kir subunit has the most simple structure; it possesses two trans-membrane and one poreforming regions and does not have any residues responsible for the voltage-sensing mechanisms. Kir channels are homomeric or heteromeric assemblies of four Kir subunits. They are one of the most ubiquitous ion channels and play indispensable roles in the maintenance of the deep resting potential, the regulation of the action potential duration, and the control of cell excitability by membrane receptors and intracellular metabolism (22, 24, 25). In addition, certain types of Kir channels are involved in K+ transport across the epithelial or astrocyte membrane of several organs such as kidney, stomach, and brain, and control K+ homeostasis in the body. It is now established that the inward rectification results from the current flowdependent block by intracellular substances, i.e., polyamines and Mg2+ (22-24), and these native blockers efficiently prevent the outward flow of potassium currents (26-29). The effect resembles anomalous rectification that is elicited when Kv channels are blocked by intracellularly applied quaternary ammonium (QA) pore blockers such as tetraethylammonium (TEA) (11, 30). In turn, QAs block not only Kv but also other K+ channels including Kir channel (1, 25, 31-34). It suggests that many K+ channels share the similar QAsbinding site and the mechanism of ion permeation. It has been also proposed that the pore domain is the functional module for K+ permeation, while the voltage sensor domain of Kv channel and the cytoplasmic domain of Kir channel just confer the regulation by voltage changes and intracellular


molecules on the channels, respectively (35). On the other hand, the effects of polyamines or Mg2+ are restricted to some of Kir channels. These substances interact with a negatively charged residue, aspartate or glutamate, which is exposed to the central cavity of the Kir channels, and prevent the permeation of K+ ions through the pore, although several residues outside the cavity are also involved in their blocking action (2, 5, 7, 3641). At present, it still remains uncertain which domains of the K+ channels are the sites for the drug actions. In spite of aforementioned observations, it had been controversial even whether the central cavity of Kir channels is the interaction site for small chemical compounds and whether the differences in the amino acid residues constituting the central cavity do account for different susceptibility of the channels to the compounds (42-48). We may emphasize that Kir channels have favorable features to examine the complex structure and function relationship for drug action, because of their simple structure. Of particular importance, the structures of a bacterial Kir homologue, KirBac1.1, Kir3.1 and a chimera channel of mouse Kir3.1 and KirBac1.3 provide the template for the plausible overall shapes of this family of K+ channels (49-52). Such information has revealed the common structure of Kir channels; the tertrameric assembly with a pore domain formed by two transmembrane (TM) segments and a selectivity filter containing the signature sequence. In addition, both N- and C-terminal segments of four subunits form another pore conserved among Kir channels, which is called the cytoplasmic pore. It extends the ion permeation pathway toward intracellular side (50). The Kir channels may be structurally related to the last two TM segments of the 6TM Kv channels except the cytoplasmic pore region, and lack a voltagesensor domain corresponding to S1-S4 segments. Thus, the Kir channels seem to be much simpler in the TM architecture than Kv channels (53). A number of preceding structural studies highlighted the physiological and pharmacological significance of Kv channel pore domain (2, 35, 54, 55). In the case of Kv channels, drugs can affect not only the pore function but also the voltagedependent gating mechanism (15, 38, 56, 57). On the other hand, Kir channels lack the voltagedependent gating mechanism and some of them such as Kir1.1, Kir2.1, or Kir4.1 channel are constitutively active and do not need any

Kir Channel-drug Interaction additional factors such as G proteins and nucleotides. Accordingly, those constitutively active Kir channels are a suitable model to examine the mechanism underlying the interaction between the pore of K+ channels and various drugs.

Recent reports for Kir channel blockers A variety of drugs can block Kir channel currents. For example, haloperidol, thioridazine, pimozide and clozapine (categorized as antipsychotics) block Kir3.x channel (58); haloperidol also blocks Kir6.2 channel (47); carvedilol (a non-selective b blocker) blocks Kir6.2 and Kir3.x channel (59); fluoxetine, desipramine, etc. (antidepressants) block Kir3.x channel (42, 43); ifenprodil (a NMDA receptor antagonist) blocks Kir3.x channel (44); genistein (a protein tyrosine kinase inhibior) blocks Kir2.3 channel (48); chloroquine and related quinines (antimalarial drugs) block Kir2.1 channel (19); ethosuximide (an

236

antiepileptic drug) blocks Kir3.x channel (60). In some cases, because the blood concentrations of drugs achieved clinically are enough to block the channels, their inhibition of Kir channels may be involved in the therapeutic actions and/or the side effects. Examinations of drug-channel interactions in Kir channels are fairly straightforward. Efforts have been made to explore the mechanism of the interactions and to understand block of the Kir channels. Some studies have identified important regions in Kir channels for the susceptibility to drugs (19, 48). The cytoplasmic pore domain of Kir2.1 is a key target for chloroquine, whereas the pore region of Kir2.3 seems to be significantly affected by genistein. By combining mutational studies with in silico analyses, we recently proposed a general approach for understanding of the mechanisms of drug-channel interactions in the study of Kir4.1 channel by antidepressants (18).

Figure 1. In silico analysis for common pharmacophores of Kir4.1 blockers and interaction between Kir4.1 and its blockers. (A) Construction of a pharmacophore for Kir4.1 blockers. Quantitative structure-activity relationship-based pharmacophores were generated by the Catalyst program. The four-point pharmacophore hypothesis comprises one hydrogen bond acceptor (green), two hydrophobic (pale blue) and one positively-ionizable (red) features. The hydrogen bond acceptor could also be characterized as a hydrophobic feature. The structures of fluoxetine and nortriptyline are fitted to the hypothesis. (B) Docking conformations of compounds in the Kir4.1 pore region. Open conformation models of Kir4.1 pore region docked with either fluoxetine (left) or nortriptyline (right). The two fluorine atoms and an amine in fluoxetine are close enough to interact respectively with side chains of Thr128 and Glu158. The secondary amine in nortriptyline could contact the side chain of Glu158. This figure was reproduced with permission from (18).


237


Antidepressants preferentially astroglial Kir4.1 channels

block

Su and Ohno et al. (45, 46, 61-64) reported that a series of antidepressants block Kir4.1 channels that are involved in astroglial K+ buffering. For example, nortriptyline blocks Kir4.1 channel expressed in HEK293T cells with a half blocking concentration of ~30 µM (46). Other effective antidepressants including amitriptyline, desipramine, imipramine, fluoxetine or sertraline also blocks Kir4.1 channel, with the IC50 ranging from 5 to 100 µM (45, 46). It seems likely that block of astroglial Kir4.1 channels is common to tricyclic antidepressants (TCAs) and selective serotonin reuptake inhibitors (SSRIs). In contrast, all these drugs have a moderate or a negligible effect on the neuronal Kir2.1 channel or the epithelial Kir1.1 channel, respectively (45, 46). Thus, we observe clear differences among Kir channels in the susceptibility to the drugs. Such findings set the stages for further mutational experiments to examine the chemical features in the vicinity of the receptor domain.

Properties of drug receptor domain in Kir4.1 The blockade of Kir4.1 by nortiptyline depends on the voltage-difference from EK; this reagent evokes the strong block at depolarized potential and the unblock at hyperpolarized potential (46). It implies that the current flows affect the channeldrug interaction. On the other hand, the blockade by fluoxetine is voltage-independent (45). Analyzing antidepressants’ interacting sites may help to elucidate the mechanisms underlying distinct modes of the action between nortriptyline and fluoxetine. One of the classic explanations of voltage dependence of drug-action is that the charged blocker enters the narrow confine area of the pore and interacts with sites located within the electric field (65). When based on this idea, it could be speculated that the two antidepressants interact with distinct sites of the channel. With mutation analyses, however, we found that the affinities for both nortriptyline and fluoxetine to Kir4.1 channel similarly depend on a subset of residues in the TM segments as following (18). Alanine substitutions at Thr128 between the pore helix and selectivity filter, and Glu158 located middle to the TM2 helix markedly reduced the blocking effect of each drug. Thus, these two residues, which are thought to face to its central


cavity, are critical for the drug-induced inhibition of the current in the both cases. Our results suggested that both nortriptyline and fluoxetine interact with the Thr128 and Glu158 of Kir4.1 subunit. In support of this idea, the closed and open conformation models of the Kir4.1 pore demonstrated that these two residues face the central cavity and are positioned within a geometrical distance permitting their interaction with the drugs (18). We next performed the docking simulation of the drugs to the estimated structure of Kir4.1 which was modeled with using the structure of the open conformation of Shaker K+ channel as a template. The results suggest that each antidepressant could be positioned at the top of the Kir4.1 central cavity where Thr128 and Glu158 locate. The simulation indicates that drugs would take multiple conformations in the central cavity. We should not injudiciously select the energy minimized docking models for presentation, because it is not necessarily the case that the conformation of the drug causes channel current block. Thus, to estimate the configuration of small compounds responsible for block of the ion channel currents, we applied another in silico approach from the compounds.

Pharmacophore of Kir channel blockers We observed differences in the affinities of structurally related compounds to the same channel. Using the 3D-quantitative structureactivity relationship (QSAR) analysis technique, we can extract from different chemical compounds the common features that elicit particular biological effect. The ensemble of the common features is also called ‘pharmacophore’. We have applied this technique to the antidepressants affecting Kir4.1 channel current and obtained the pharmacophore for block of the channel. The results indicate that all the drugs share a hydrogen bond acceptor and a positively charged moiety (Figure 1A) (18). Of particular importance, both 3D-structures and physicochemical features of receptor and ligand coincide well together. Thus, both analyses, each of which is either from protein side or chemical side, suggested the same configuration, which may be a strong candidate for the drug-form to cause channel current block. The geometrical arrangement of nortriptyline and fluoxetin indicated that they might be located in the region of the central cavity between a hydroxyl


238

Figure 2. Effects of different antidepressant compounds on Kir4.1 channel currents. (A) Chemical structures of fluoxetine, atomoxetine, and nisoxetine. Representative traces of Kir4.1 channel current responses to 30 µM fluoxetine, atomoxetine, and nisoxetine. The HEK293T + cells expressing Kir4.1 channels were held at −40 mV (~EK) in 30 mM [K ]o. Voltage pulses from 90 mV to 0 mV (10 mV increments) were applied. (B) Concentration–response effects of different antidepressant compounds upon Kir4.1 channel currents. The relationship between the current ratio at −110 mV and the drug concentration was fitted with Hill's equation (lines). Symbols represent the mean ± SEM of 3 to 5 separate experiments.

group of Thr128 and a carboxyl group of Glu158 that are derived from two different channel subunits diagonally facing the pore. Together, we can estimate the mechanism underlying the interaction between antidepressants and Kir4.1 channel at the atomic level: Most likely, the antidepressants bind to the pore residues through hydrogen and ionic interactions (Figure 1B) (18). Such interactions are significant for the pharmacological effects, but weak and transient. In addition, drug and channel structures are both highly flexible. In this context, it is noteworthy that the drug-channel interaction is completely different from the lock-and-key type model for molecular recognition in pharmacology.

Structural changes of fluoxetine affect its blocking property The models could have been further confirmed by examining the substituent effect of structural

analogues derived from antidepressants. Fluoxetine and sertraline, which are categorized into SSRIs, block Kir4.1 channel in voltageindependent manner. These drugs have a halogen substitution at the third or fourth position of phenoyl or phenyl ring, and this characteristic seems to be responsible for the specific effect of SSRIs on serotonin-transporter (66). As for fluoxetine, the pharmacophore analysis suggested that these SSRI halogens are involved in its interaction with Thr128 of Kir4.1. To test this idea, we examined the effect of fluoxetine-related compounds, atomoxetine and nisoxetine, which are respectively synthesized by substitutions of -CH3 and -OCH3 for the second position in the phenoxy ring of fluoxetine. The two compounds are categorized in norepinephrine-reuptake inhibitors (NRIs). We found that these NRIs suppressed Kir4.1-current much more weakly than fluoxetine (IC50 values in µM, 17.2 ± 0.8 for fluoxetine, 37.9 ±


239


Figure 3. Effect of chloroquine on Kir4.1 channel. (A) Features mapping on chloroquine. (B and C) Representative responses of Kir4.1 + channel currents to chloroquine (30 and 300 µM). Transiently transfected HEK293T cells were voltage-clamped at -40 mV (~EK) in 30 mM [K ]o. Voltage pulses from 90 mV to 0 mV (10 mV increments)(B), or a pair of voltage steps (0.5 sec in duration) to EK ± 70 mV (-110 and +30 mV) with a 20-msec interval (C) were applied.

2.2 for (R)-atomoxetine, and 204.9 ± 28.4 for nisoxetine, n = 3-5) (Figure 2). It implies that the trifluoromethyl group of fluoxetine is important for the drug-channel interaction. In addition, we also found the approximately equal Hill coefficients between fluoxetine and atomoxetine (2.0 ± 0.3 and 1.9 ± 0.5, respectively), although the forms of the two drugs we used were racemic mixture and Renantiomer, respectively. This not only indicates the possibility that Kir4.1 channels do not strictly discriminate the racemic mixture of the drugs but also indicates the flexible nature of the drugchannel interaction.

Differential drug interaction sites in Kir channels The differences of amino acid residues at the drug interaction site may account for the differential sensitivity to antidepressants among Kir channels. The rank order of the inhibitory effects of antidepressants on Kir channels is following; Kir4.1, Kir3.1, Kir2.1, and Kir1.1 (45, 46). Alignment of Kir subunits shows that Thr128 at the putative drug interaction site in Kir4.1 is conserved in all Kir subunits at this position,


whereas Glu158 in Kir4.1, which may be another interaction site, is not conserved. For instance, the residue at the corresponding site in Kir1.1 is Asn171. We demonstrated that, when Asn171 was mutated to Glu or Asp in Kir1.1, the channel became highly sensitive to the antidepressants (18). This result indicates that the drugs require a negatively charged carboxyl group in the channel for high affinity interaction whereas the length of the side chain is secondary for it. At the same time, the result also accounts for the mechanism underlying their preferential inhibitory action on Kir4.1 current. Therefore, the differential affinity of Kir channels for these drugs may primarily be due to a single amino acid residue at this position. The locations of Thr128 and Glu158 of Kir4.1 may affect ion energetics inside the pore (67), and these residues are likely to provide a favorable environment for the docking of chemical compounds at the site. All evidences support the idea that these antidepressants enter the channel pore and finally reach the central cavity (18). Side chains of Thr128 and Glu158 stabilize the accommodation of such drug inside the central cavity.

Kir Channel-drug Interaction Recent studies indicate that, in Kir channels, the central cavity is not the sole domain for interactions with chemical compounds. For example, tertiapine interacts with the extracellular mouse of the selectivity filter of Kir1.1 channel and KG channel composed of Kir3.x and blocks them (20, 21). The position identified as the chloroquine-interaction site in Kir2.1 channel is the cytoplasmic end of the conduction pathway, which is far from the central cavity (19). This drug may block the ion permeation at the cytoplasmic pore site. Chloroquine could not be matched with the Kir4.1 pore blockers (antidepressants)-derived QSAR model. This is due to the positively charged nitrogen in the quinoline ring of chloroquine (Figure 3A). Rodriguez-Menchaca et al. speculated that two positively ionizable sites of chloroquine including the nitrogen are crucial for the interaction of the drug with the cytoplasmic pore residues, Glu224 and Asp229, of Kir2.1 channel (19). Consistent with the result of the modeling, we found that chloroquine at < 30 µM had a negligible effect on Kir4.1 channels expressed in HEK cells, although the drug at higher concentrations could suppress the channels, especially the outward-going currents (Figure 3B). Thus, Kir4.1 has much lower affinity for chloroquine than Kir2.1 (IC50 = 8.7 ± 0.9 µM (19)). This observation suggests that the central cavity of Kir4.1 and the cytoplasmic pore of Kir2.1 provide distinct environment for various small chemical compounds. This may indicate the possibility of designing pharmacological tools to distinguish between Kir2.1 and Kir4.1 channels. As have been described in the above, the properties of drug-channel interaction can be better understood with using the approaches from both channel protein and chemical compound sides. This may help us to further elucidate the mechanisms for the interaction between the drug and the channel. The pharmacophore analysis from the compound side should be one very useful method for the purpose, although the feasibility of specific models should be carefully examined.

Antidepressants also block other K+ channels K+ channels are a highly diversified group of ion channels, including Kir channel, Kv channel, Ca2+activating K+ channel, and two-pore domain K+ channels. Like QAs, some drugs are known to

240

affect the multiple types of K+ channels. Of note, antidepressants have been reported to block other K+ channels than Kir4.1, such as hTREK1 twopore domain K+ channels and hERG Kv channel (68-72). Thus, antidepressants can block various K+ channels, although they can distinguish Kir4.1 and Kir2.1 in the Kir channel family. This should reflect the properties of the drugs and some moiety of channel protein amino acids, and thus the interaction between them. It is therefore conceivable that there are some basic rules in the drug-channel interaction common to a variety of K+ channels. To clarify the rules at atomic level, it is needed to examine how the antidepressants block the K+ channels other than Kir4.1 with our combined method. Alignment of the pore region of different K+ channel subunits shows that threonine at the putative drug interaction site 128 in Kir4.1 is conserved also in mTREK-1 and hERG channels at the corresponding position, whereas a residue corresponding to Glu158 is not. It may be Phe170 in mTREK-1 and Tyr 652 in hERG channels. However, we do not have any experimental information on the interacting site of those K+ channels. We do not know whether these residues contribute to the interaction of the channels with antidepressants. The pose of compounds interacted with the channels are still unpredictable from these primary sequences of channels. These should be estimated with the bidirectional approach adopted in our study.

Considerations, insights, perspectives When K+ ions moved along a narrow channel pore through the hydrophobic plasma membrane, they must cross the electrostatic energy barrier. The presence of the water-filled central cavity, which is halfway across the membrane, can be understood intuitively as one of the channel’s mechanisms for overcoming the barrier (54, 67, 73). Nevertheless, the basic properties and structurefunction relationship in the cavity are still not well understood. At the end of this article, pharmacological aspects of the central cavity are further discussed (see model of Kir4.1 channel pore domain as shown in Figure 4A). Similar to the case of Kv channels, the central cavity of Kir channels may serve as the receptor site for some drugs. Therefore, drugs are very useful for proving the common and/or diverse features of the cavity in K+ channels. It is thought


241


+

Figure 4. The focus on the central cavity of K channel. (A) Side view of Kir4.1 pore domain’s homology model highlighting residues Thr128 and Glu158. (B) Amino acid sequence alignment of Kir4.1, Kir2.1, Kir1.1, hERG (Kv11.1), Shaker (Kv1.1), and Kv1.5 channels. Putative druginteraction residues are colored in red (for Kir4.1 channel) and blue (for others) (2, 5, 7, 18, 36-41).

that, in Kv and Kir channels, the residues affecting their sensitivities to drugs are located in the two juxtapositions in its primary sequences (2, 5, 7, 3641), as shown in red- and blue-letters in Figure 4B. It seems probable that these residues are located at the vicinity in the tertiary structures. For examples, as mentioned above, Thr128 and Glu158 of Kir4.1 correspond to Ser624 and Tyr652 of hERG channel, respectively, which are known to be the residues for hERG-interaction with various drugs. Also, Thr441 of Drosophila Shaker (Kv1.1) channel which locates on the base of pore helix is


critical for the blockade by internal TEA. Unfortunately, the contribution of the neighboring Thr442 to the TEA block has remained uncertain because the mutation of this residue causes different voltage- and time-dependence of the activation gating. Thr469 positioned on the S6 domain determines the TEA sensitivity (5, 7). Ile508 of hKv1.5 channel is one of the key residues that determine the pharmacological properties of the channels (37). These residues may correspond to Thr128 or Glu158 of Kir4.1 channels (Figure 4B). We propose the possibility that this topological

Kir Channel-drug Interaction similarity may indicate a common feature for druginduced block of ion current flowing through various K+ channel pores. Thus, the configurations of the drug inhibiting the ion channel permeation may not be free but restricted in order to interact with these sites, even if drugs could take multiple poses in the central cavity. Our data suggest that distinct (especially, diagonal) subunits sandwich the blocker of asymmetric structure with the key residues of two concentric rings (Figure 1B). Therefore, for chemical-scale studies of ion channels, the bi-directional approach may be useful to estimate the representative pose of compounds within the cavity, which would allow us to obtain further insights for the common features of mechanism underling the block. Also, the pharmacological studies point out the diverse nature of the cavity even in Kir channels. The potencies of a certain compound for several Kir channels clearly differ (18, 45, 46, 48, 59). At least, the different susceptibility of the Kir channels to the antidepressants could be accounted for by the differences in the amino acid residues constituting the central cavity (18). This indicates their diversity in the electrostatic nature of the pore between Kir channels. Such differences may influence traveling of K+ ions and water molecules as well. In Kir3.2 channel, Asn184, which is expected to face the cavity, could play a key role in ion selectivity (74). In our study, several mutations of Thr128 or Glu158 such as T128L, T128I, E158D, E158N, and E158Q in Kir4.1 impaired the ability to conduct K+ currents (18), although the mechanism remains largely unclear. One possibility is that this phenotype is linked with the relationship between the electrostatics free energy of K+ ions and polyamines inside the central cavity (67). Using chemical compounds as probes, we will be able to further examine the nature of ion permeation pathway as well as the status of the intracellular gate in the blocking state of Kir channels.

Acknowledgements This work was supported by the Grant-in-Aid for Scientific Research (A) 20249012 (to Y. K.), Grant-in-Aid for Young Scientists (B) 20790207 (to K. F.), Grant-inAid for Scientific Research on Priority Areas 17079005 (to Y. K.), and Grant-in-Aid for Scientific Research on Priority Areas 17081012 (to H. H.) from the Ministry of Education, Science, Sports and Culture of Japan.

Conflicts of Interest No potential conflicts of interest to disclose.

242

References 1. Hille B: Ion channels of excitable membranes, 3rd ed. Sunderland, MA, Sinauer Associates, Inc., 2001. 2.del Camino D, Holmgren M, Liu Y, et al. Blocker protection in the pore of a voltage-gated K+ channel and its structural implications. Nature 2000;403:321-25. 3. Hidalgo P, MacKinnon R. Revealing the architecture of a K+ channel pore through mutant cycles with a peptide inhibitor. Science 1995;268:307-10. 4. MacKinnon R, Heginbotham L, Abramson T. Mapping the receptor site for charybdotoxin, a pore-blocking potassium channel inhibitor. Neuron 1990;5:767-71. 5. Yellen G, Jurman ME, Abramson T, et al. Mutations affecting internal TEA blockade identify the probable pore-forming region of a K+ channel. Science 1991;251:939-42. 6. MacKinnon R, Yellen G. Mutations affecting TEA blockade and ion permeation in voltage-activated K+ channels. Science 1990;250:276-79. 7. Choi KL, Mossman C, Aube J, et al. The internal quaternary ammonium receptor site of shaker potassium channels. Neuron 1993;10:533-541. 8. Hille B, Schwarz W. Potassium channels as multi-ion single-file pores. J Gen Physiol 1978;72:409-42. 9. Armstrong CM. Time course of TEA+-induced anomalous rectification in squid giant axons. J Gen Physiol 1966;50:491-503. 10. Armstrong CM. Inactivation of the potassium conductance and related phenomena caused by quaternary ammonium ion injection in squid axons. J Gen Physiol 1969;54:553-75. 11. Armstrong CM. Interaction of tetraethylammonium ion derivatives with the potassium channels of giant axons. J Gen Physiol 1971;58:413-437. 12. Armstrong CM, Hille B. The inner quaternary ammonium ion receptor in potassium channels of the node of ranvier. J Gen Physiol 1972;59:388-400. 13. Strichartz GR. The inhibition of sodium currents in myelinated nerve by quaternary derivatives of lidocaine. J Gen Physiol 1973;62:37-57. 14. Lenaeus MJ, Vamvouka M, Focia PJ, et al. Structural basis of TEA blockade in a model potassium channel. Nat Struct Mol Biol 2005;12:454-59. 15. Choi KL, Aldrich RW, Yellen G. Tetraethylammonium blockade distinguishes two inactivation mechanisms in voltage-activated K+ channels. Proc Natl Acad Sci U S A 1991;88:5092-5. 16. Hoshi T, Zagotta WN, Aldrich RW. Two types of inactivation in shaker K+ channels: Effects of alterations in the carboxy-terminal region. Neuron 1991;7:547-56. 17. Crouzy S, Berneche S, Roux B. Extracellular blockade of K+ channels by TEA: Results from molecular dynamics simulations of the KcsA channel. J Gen Physiol 2001;118:207-18. 18. Furutani K, Ohno Y, Inanobe A, et al. Mutational and in silico analyses for antidepressant block of astroglial inward-rectifier Kir4.1 channel. Mol Pharmacol 2009;75:1287-95.


243


19. Rodriguez-Menchaca AA, Navarro-Polanco RA, Ferrer-Villada T, et al. The molecular basis of chloroquine block of the inward rectifier Kir2.1 channel. Proc Natl Acad Sci U S A 2008;105:1364-8. 20. Jin W, Klem AM, Lewis JH, et al. Mechanisms of inward-rectifier K+ channel inhibition by tertiapin-Q. Biochemistry 1999;38:14294-301. 21. Ramu Y, Xu Y, Lu Z. Engineered specific and highaffinity inhibitor for a subtype of inward-rectifier K+ channels. Proc Natl Acad Sci U S A 2008;105:10774-8. 22. Isomoto S, Kondo C, Kurachi Y. Inwardly rectifying potassium channels: Their molecular heterogeneity and function. Jpn J Physiol 1997;47:11-39. 23. Lu Z. Mechanism of rectification in inward-rectifier K+ channels. Annu Rev Physiol 2004;66:103-29. 24. Nichols C, Lopatin A. Inward rectifier potassium channels. Annu Rev Physiol 1997;59:171-91. 25. Kubo Y, Adelman J, Clapham D, et al. International union of pharmacology. Liv. Nomenclature and molecular relationships of inwardly rectifying potassium channels. Pharmacol Rev 2005;57:509-26. 26. Horie M, Irisawa H, Noma A. Voltage-dependent magnesium block of adenosine-triphosphate-sensitive potassium channel in guinea-pig ventricular cells. J Physiol 1987;387:251-72. 27. Lopatin AN, Makhina EN, Nichols CG. Potassium channel block by cytoplasmic polyamines as the mechanism of intrinsic rectification. Nature 1994;372:366-9. 28. Matsuda H. Open-state substructure of inwardly rectifying potassium channels revealed by magnesium block in guinea-pig heart cells. J Physiol 1988;397:23758. 29. Vandenberg CA. Inward rectification of a potassium channel in cardiac ventricular cells depends on internal magnesium ions. Proc Natl Acad Sci U S A 1987;84:2560-4. 30. Armstrong CM, Binstock L. Anomalous rectification in the squid giant axon injected with tetraethylammonium chloride. J Gen Physiol 1965;48:859-72. 31. Blatz AL, Magleby KL. Ion conductance and selectivity of single calcium-activated potassium channels in cultured rat muscle. J Gen Physiol 1984;84:1-23. 32. Guo D, Lu Z. Mechanism of IRK1 channel block by intracellular polyamines. J Gen Physiol 2000;115:799814. 33. Makhina EN, Kelly AJ, Lopatin AN, et al. Cloning and expression of a novel human brain inward rectifier potassium channel. J Biol Chem 1994;269:20468-74. 34. Yellen G. Ionic permeation and blockade in Ca2+activated K+ channels of bovine chromaffin cells. J Gen Physiol 1984;84:157-86. 35. Lu Z, Klem AM, Ramu Y. Ion conduction pore is conserved among potassium channels. Nature 2001;413:809-13. 36. Chen J, Seebohm G, Sanguinetti M. Position of aromatic residues in the S6 domain, not inactivation,


dictates cisapride sensitivity of HERG and eag potassium channels. Proc Natl Acad Sci U S A 2002;99:12461-6. 37. Decher N, Pirard B, Bundis F, et al. Molecular basis for Kv1.5 channel block: Conservation of drug binding sites among voltage-gated K+ channels. J Biol Chem 2004;279:394-400. 38. Hosaka Y, Iwata M, Kamiya N, et al. Mutational analysis of block and facilitation of HERG current by a class III anti-arrhythmic agent, nifekalant. Channels 2007;1:198-208. 39. Kamiya K, Niwa R, Mitcheson J, et al. Molecular determinants of HERG channel block. Mol Pharmacol 2006;69:1709-16. 40. Mitcheson J, Chen J, Lin M, et al. A structural basis for drug-induced long QT syndrome. Proc Natl Acad Sci U S A 2000;97:12329-33. 41. Sánchez-Chapula J, Navarro-Polanco R, Culberson C, et al. Molecular determinants of voltage-dependent human ether-a-go-go related gene (HERG) K+ channel block. J Biol Chem 2002;277:23587-95. 42. Kobayashi T, Washiyama K, Ikeda K. Inhibition of G protein-activated inwardly rectifying K+ channels by various antidepressant drugs. Neuropsychopharmacology 2004;29:1841-51. 43. Kobayashi T, Washiyama K, Ikeda K. Inhibition of G protein-activated inwardly rectifying K+ channels by the antidepressant paroxetine. J Pharmacol Sci 2006;102:278-87. 44. Kobayashi T, Washiyama K, Ikeda K. Inhibition of G protein-activated inwardly rectifying K+ channels by ifenprodil. Neuropsychopharmacology 2006;31:516-24. 45. Ohno Y, Hibino H, Lossin C, et al. Inhibition of astroglial Kir4.1 channels by selective serotonin reuptake inhibitors. Brain Res 2007;1178:44-51. 46. Su S, Ohno Y, Lossin C, et al. Inhibition of astroglial inwardly rectifying Kir4.1 channels by a tricyclic antidepressant, nortriptyline. J Pharmacol Exp Ther 2007;320:573-80. 47. Yang SB, Proks P, Ashcroft FM, et al. Inhibition of atp-sensitive potassium channels by haloperidol. Br J Pharmacol 2004;143:960-7. 48. Zhao Z, Liu B, Zhang G, et al. Molecular basis for genistein-induced inhibition of Kir2.3 currents. Pflugers Arch 2007 49. Kuo A, Domene C, Johnson LN, et al. Two different conformational states of the KirBac3.1 potassium channel revealed by electron crystallography. Structure 2005;13:1463-72. 50. Kuo A, Gulbis JM, Antcliff JF, et al. Crystal structure of the potassium channel KirBac1.1 in the closed state. Science 2003;300:1922-26. 51. Nishida M, Cadene M, Chait BT, et al. Crystal structure of a Kir3.1-prokaryotic Kir channel chimera. Embo J 2007;26:4005-15. 52. Bichet D, Haass FA, Jan LY. Merging functional studies with structures of inward-rectifier K+ channels. Nat Rev Neurosci 2003;4:957-67.

Kir Channel-drug Interaction 53. Long SB, Tao X, Campbell EB, et al.. Atomic structure of a voltage-dependent K+ channel in a lipid membrane-like environment. Nature 2007;450:376-82. 54. Doyle DA, Morais Cabral J, Pfuetzner RA, et al. The structure of the potassium channel: Molecular basis of K+ conduction and selectivity. Science 1998;280:69-77. 55. Sanguinetti M, Tristani-Firouzi M. hERG potassium channels and cardiac arrhythmia. Nature 2006;440:4639. 56. Cha A, Bezanilla F. Structural implications of fluorescence quenching in the shaker K+ channel. J Gen Physiol 1998;112:391-408. 57. Jiang M, Dun W, Fan JS, et al. Use-dependent 'agonist' effect of azimilide on the HERG channel. J Pharmacol Exp Ther 1999;291:1324-36. 58. Kobayashi T, Ikeda K, Kumanishi T. Inhibition by various antipsychotic drugs of the G-protein-activated inwardly rectifying K+ (GIRK) channels expressed in xenopus oocytes. Br J Pharmacol 2000;129:1716-22. 59. Kikuta J, Ishii M, Kishimoto K, et al. Carvedilol blocks cardiac KATP and KG but not IK1 channels by acting at the bundle-crossing regions. Eur J Pharmacol 2006;529:47-54. 60. Kobayashi T, Hirai H, Iino M, et al. Inhibitory effects of the antiepileptic drug ethosuximide on G proteinchannels. activated inwardly rectifying K+ Neuropharmacology 2009;56:499-506. 61. Kofuji P, Newman E. Potassium buffering in the central nervous system. Neuroscience 2004;129:1045-56. 62. Poopalasundaram S, Knott C, Shamotienko O, et al. Glial heterogeneity in expression of the inwardly rectifying K+ channel, Kir4.1, in adult rat CNS. Glia 2000;30:362-72. 63. Simard M, Nedergaard M. The neurobiology of glia in the context of water and ion homeostasis. Neuroscience 2004;129:877-96. 64. Walz W. Role of astrocytes in the clearance of excess extracellular potassium. Neurochem Int 2000;36:291-300. 65. Heginbotham L, MacKinnon R: The aromatic binding site for tetraethylammonium ion on potassium channels. Neuron 1992;8:483-91. 66. Wong DT, Bymaster FP, Engleman EA. Prozac (fluoxetine, lilly 110140), the first selective serotonin uptake inhibitor and an antidepressant drug: Twenty years since its first publication. Life Sci 1995;57:411-41. 67. Robertson JL, Palmer LG, Roux B. Long-pore electrostatics in inward-rectifier potassium channels. J Gen Physiol 2008;132:613-32. 68. Choi J, Hahn S, Rhie D, et al. Mechanism of fluoxetine block of cloned voltage-activated potassium channel Kv1.3. J Pharmacol Exp Ther 1999;291:1-6. 69. Teschemacher A, Seward E, Hancox J, et al. Inhibition of the current of heterologously expressed HERG potassium channels by imipramine and amitriptyline. Br J Pharmacol 1999;128:479-85. 70. Thomas D, Gut B, Wendt-Nordahl G, et al. The antidepressant drug fluoxetine is an inhibitor of human ether-a-go-go-related gene (HERG) potassium channels. J Pharmacol Exp Ther 2002;300:543-48.

244

71. Kennard LE, Chumbley JR, Ranatunga KM, et al. Inhibition of the human two-pore domain potassium channel, TREK-1, by fluoxetine and its metabolite norfluoxetine. Br J Pharmacol 2005;144:821-9. 72. Sung M, Ahn H, Hahn S, et al. Open channel block of Kv3.1 currents by fluoxetine. J Pharmacol Sci 2008;106:38-45. 73. Roux B, MacKinnon R. The cavity and pore helices in the KcsA K+ channel: Electrostatic stabilization of monovalent cations. Science 1999;285:100-2. 74. Bichet D, Grabe M, Jan YN, et al. Electrostatic interactions in the channel cavity as an important determinant of potassium channel selectivity. Proc Natl Acad Sci U S A 2006;103:14355-60.
