| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Received 5 March 1998; accepted after revision 18 May 1998.
 |
ABSTRACT |
|---|
 TopAbstract IntroductionMethodsResultsDiscussionReferences |
|---|
interactions and by backbone carbonyl groups (Thr142 and Gly144).
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Potassium channels fall into two main structural groups. The voltage-gated and Ca2+-activated K+ channels (KCa, Kv) form one class, having six membrane-spanning segments (S1 to S6) in each of four subunits (MacKinnon, 1991; Liman et al. 1992). A short stretch of amino acids between S5 and S6, known as the H5 or P region, projects into the centre of the pore and forms the K+ selectivity filter (Heginbotham et al. 1994). The inwardly rectifying K+ channel family (Kir) make up the second class, having only two membrane-spanning segments (M1 and M2) per subunit in the tetramer (Ho et al. 1993; Kubo et al. 1993; Yang et al. 1995), but retaining the pore-forming H5 loop between M1 and M2 (Fig. 1).
Sequence alignment of Kv and Kir channels reveals that the two classes of K+ channel share little sequence similarity apart from within the H5 region, where they are highly homologous over a stretch of eight amino acids: T(S)xxTxGY(F)G. This homologous region is often referred to as the 'K+ channel signature sequence' and probably represents a common catalytic domain that determines the ionic selectivity and permeation characteristics of K+ channels (Heginbotham et al. 1994). Consistent with this hypothesis of a catalytic domain, mutation of many of the residues within the signature sequence render Kv channels non-selective for monovalent cations (Yool & Schwarz, 1991; Taglialatela et al. 1993; Heginbotham et al. 1994). Similarly, mutation of a glycine in the conserved GYG motif in G-protein-regulated Kir channels results in the loss of K+ selectivity (Slesinger et al. 1996; Kofuji et al. 1996; Navarro et al. 1996).
The structure of the H5 region of Kv channels has been studied extensively. Pore-lining residues whose side chains may interact with permeating K+ have been identified using the substituted cysteine accessibility method (Kürz et al. 1995; Lü & Miller, 1995; Pascual et al. 1995), and the shape of the Kv channel pore has been mapped with remarkable resolution using pore-blocking peptide neurotoxins as structural templates (Aiyar et al. 1995, 1996; Hidalgo & MacKinnon, 1995; Ranganathan et al. 1996). Much less is known about the pore structure of the Kir channel family. However, Lu & MacKinnon (1997) have shown recently that the Leiurus quinquestriatus scorpion toxin Lq2, a potent inhibitor of both Kv and KCa channels, can also block the pore of the weak inward rectifier, Kir1.1 (ROMK1). The interaction surface or 'footprint' of the toxin is the same for binding to Kir, Kv and KCa channels, suggesting that a broadly similar K+-selective pore structure exists across the different K+ channel families, possibly reflecting the pore design of an ancestral channel.
We have used an existing molecular model of the H5 region of the voltage-gated K+ channel Kv1.3 (mapped using the scorpion venom, kaliotoxin; Aiyar et al. 1996) to construct a model of the H5 region of the strong inward rectifier, Kir2.1. Residues whose side chains are predicted by the model to line the pore were tested by the substituted cysteine accessibility method (Akabas et al. 1992). Here individual residues are mutated in turn to cysteine and the exposure of the highly reactive sulfhydryl group of the substituted Cys may be tested in the resultant channels for susceptibility to block by Ag+. Where necessary, the positions of amino acid side chains in the model were refined so as to be consistent with the experimental findings. Point mutations within the H5 region often result in non-functional channels, since the mutation is carried on each subunit and is repeated 4 times in the tetramer. To rescue mutants that did not express functional channels as monomers, dimeric and tetrameric Kir2.1 cDNAs were constructed (Leyland et al. 1997; Dart et al. 1998). Expression of dimers containing a single mutant subunit would be expected to yield tetrameric channels carrying a point mutation on only two of the four subunits. Our results support the idea that the different classes of K+ channel share a similar, but not identical, K+-selective pore structure. Preliminary accounts of this work have been given (Dart et al. 1997a,b).
|
METHODS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The details of methods of cell culture, molecular biology and electrophysiology are given in the previous paper (Dart et al. 1998). All experiments were carried out on channels in which the exposed Cys residue (149) had been replaced with Ser, since we have shown this mutation prevents blockage of wild-type channels by Ag+ (Dart et al. 1998).
Channel modelling
Our initial, 'unrefined' model of Kir2.1 was produced by replacing the amino acid residues in the model of Kv1.3 (Aiyar et al. 1996; co-ordinates generously supplied by K. G. Chandy and G. A. Gutman) with the corresponding residues of Kir2.1, using the program MODELLER (Sali & Blundell, 1993). The positions of the amino acid side chains in the resulting model were then compared with the results of the scanning cysteine mutagenesis/Ag+ blockage experiments. The residues that required repositioning were refined automatically, and the salt bridge between Glu138 and Arg148 in adjacent subunits (Yang et al. 1997) added, using distance restraints within the program XPLOR 3.843 (Brunger, 1996). The model was refined using simulated annealing followed by energy minimization, with restraints defined as follows. The 
-carbon atoms of residues Glu125-Ile137, Thr139 and Val150-Pro155 were fixed to their original positions; this procedure allowed the side chains to reposition themselves where steric clashes existed yet retained the starting topology for these residue positions. Those residue positions blocked by Ag+ were restrained using an upper distance from the centre of the channel to the 
-atom of 0·45 nm. (This atom is equivalent to the sulphur in the Cys sulfhydryl group.) The intersubunit salt bridge was restrained using distance ranges of 0·25-0·33 nm for Arg148 NH1-Glu138 OE1 and 0·18-0·23 nm for Arg148 HH1 (a hydrogen atom on NH1)-Glu138 OE1. The nomenclature used for the side chain atoms follows the IUPAC-IUB (1985) convention. The structure was also restrained to have 4-fold symmetry.
|
RESULTS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Cysteine-scanning mutagenesis of the H5 region of Kir2.1
We mutated the H5 amino acid residues shown in Fig. 1B in turn to cysteine (Akabas et al. 1992) and tested the resultant mutant channels for their sensitivity to blockage by Ag+. Channel blockage will occur only if a cysteine residue projects its side chain into the pore lumen so that a bound Ag+ prevents K+ transfer. Ag+ was chosen as a probe primarily because of its small size (radius 0·126 nm compared with 0·133 nm for K+), enabling it to reach into the narrow confines of the pore.
 |
View larger version [in this window] [in a new window] |
|
|
A, Kv channel subunits (left) contain 6 membrane-spanning segments (S1-S6) with a short loop of amino acids known as the H5 or P region between S5 and S6. Inward rectifier subunits (right) contain only 2 transmembrane segments but retain H5. B, sequence alignment of the H5 regions of two Kv channels and the strong inward rectifier Kir2.1. The K+ channel signature sequence is shown in bold. | ||
Point mutations made within the highly conserved H5 region often result in the production of non-functional channels. We investigated ten residues in the H5 region of Kir2.1 from Thr139 to Arg148 (Fig. 1B). After expressing monomers, we found currents with only two mutants, I143C and F147C. We therefore used covalently linked tandem dimers of Kir2.1 in which the residues of interest were changed to Cys in only one of the two coding regions. The channels formed in this way have residues substituted by Cys in two of the four subunits.
Using this technique, we found currents in a further six cases: T139C; T141C; T142C: G144C; Y145C; and G146C. We did not find currents after replacing the Gln residue at position 140 (Q140) in one coding region in the dimeric construct, but a tetrameric construct with only one Cys substitution did produce functional channels.
The only position at which mutation to cysteine did not lead to significant currents in dimers or tetramers was Arg148. This residue forms a salt bridge with a glutamate (E138) on the adjacent subunit, an interaction which helps stabilize the central pore structure (Yang et al. 1997). Any mutation that disrupts this intersubunit ion pairing results in non-functional channels. In cells transfected with R148C dimers (subunit A carrying the single mutation C149S and subunit B the double mutation C149S-R148C), we were able to record only very small whole-cell currents (amplitude < 5 pA pF-1), approximately 100-fold less than our normal levels of expression. It has been shown that dimeric proteins (A-B) preferentially associate in an A-B-A-B fashion (Liman et al. 1992; Heginbotham & MacKinnon, 1992; Yang et al. 1995). However, a very small fraction of the dimers contribute only one subunit (A or B) to the functional channel, leaving the other half of the dimer outside the complex (Heginbotham & MacKinnon, 1992). We assume that the small currents we recorded from the R148C dimers (i.e. [C149S]-[R148C, C149S]) resulted from the C149S half of four dimers coming together to form a C149S channel.
For each mutant, the normalized chord conductance was plotted against voltage and fitted with a single Boltzmann equation, giving values for V0·5 (the voltage at which the normalized conductance is 0·5) and k (a steepness factor). None of the whole-cell currents expressed by the mutant channels differed significantly from wild-type channels, as measured in terms of V0·5 and k values (data not shown).
Determining which residues line the channel pore
Figure 2A shows whole-cell currents recorded from Chinese hamster ovary (CHO) cells expressing Kir2.1 mutants. The current records shown are in response to voltage steps from a holding potential of -17 mV (equilibrium potential for K+; EK) to test potentials of +33 and -102 mV, before and after the addition of external 200 nM Ag+ (exposure time, 90 s).
 |
View larger version [in this window] [in a new window] |
|
|
A, whole-cell currents recorded from single CHO cells transfected with the dimeric Kir2.1 constructs encoding for cysteine-substituted channels. 'C149' indicates a C149S-wild-type dimer. Currents recorded in response to voltage steps from a holding potential of -17 mV (EK; [K+]o = 70 mM; [K+]i = 140 mM) to test potentials of +33 mV and -102 mV. Control records are indicated by | ||
Residues which, when mutated to cysteine, were inhibited by the application of 200 nM external Ag+ were: Thr141; Thr142; Ile143; Tyr145; and Phe147. We predict that the side chains of these residues line the channel pore. Residues which, when mutated to cysteine, remained insensitive to the application of external Ag+ were: Gly146; Gly144; Gln140; and Thr139, predicting that the side chains of these residues are not exposed to the aqueous pore. The results of cysteine-scanning mutagenesis of the H5 region of Kir2.1 are summarized in the histogram in Fig. 2B and in Table 1.
The right-hand column of Table 1 indicates whether the experimental results agree with the initial version of the model which is based on the H5 region of Kv1.3. Whilst the overall agreement with the model is good, the experimental results require the repositioning of three residues. Thr141 was initially too far (0·9 nm) from the centre of the channel for a bound Ag+ to block the pore; Cys149, known to be exposed in wild-type channels, was partially hidden in the initial model behind the aromatic ring of Phe147; and Ile143 was again slightly too far away (0·56 nm) from the centre of the pore.
Table 1. Results of cysteine-scanning mutagenesis and agreement with the initial model
| Mutant | Percentage inhibition by 200 nM Ag+ (90 s exposure) | Time constant,  (200 nM Ag+) (s) |
Side chain orientation | Agreement with initial model |
| T139C | -0·9 ± 2·5 (n = 3) | > 3000 | Buried | Yes |
| Q140C | 1·3 ± 1·7 (n = 4) | > 3000 | Buried | Yes |
| T141C | 16·7 ± 4·4 (n = 4) | 444 ± 60 | Exposed | No |
| T142C | 13·4 ± 2·5 (n = 5) | 636 ± 71 | Exposed | Yes |
| I143C | 21·1 ± 5·7 (n = 6) | 311 ± 18 | Exposed | No |
| G144C | 0·3 ± 1·7 (n = 3) | > 3000 | Buried | Yes |
| Y145C | 19·8 ± 2·6 (n = 5) | 318 ± 23 | Exposed | Yes |
| G146C | -1·6 ± 2·6 (n = 3) | > 3000 | Buried | Yes |
| F147C | 25·7 ± 5·7 (n = 4) | 246 ± 7 | Exposed | Yes |
| R148C No expression | No expression | No expression | - | |
| C149 | 43·8 ± 4·1 (n = 7) | 149 ± 5 | Exposed | Yes? * |
Rate of block by Ag+
The rate at which the different mutant channels are blocked gives an indication of how accessible the cysteine-substituted residue is to externally applied Ag+. Residues located deep in the pore should produce channel blockage more slowly than residues located at the external mouth and hence more easily accessible to the bulk extracellular solution. Figure 3A shows the fractional remaining current for four of the cysteine-substituted dimers plotted against exposure time to 200 nM external Ag+. The decay of the currents has been fitted with a single exponential (see Dart et al. 1998). Considering just two of the residues, when cysteine replaced Thr141 on two of the four subunits, Ag+ ions blocked the resultant channels with a time constant of 444 ± 60 s. When two cysteine residues were present at position 149, however, Ag+ blocked much faster, with a time constant of 149 ± 5 s (Table 1). If we compare the rate of blockage with that found at Cys149, the rates are in the ratio I143C : T142C : T141C : Y145C : F147C : C149 (wild-type) = 0·11 : 0·23 : 0·34 : 0·47 : 0·61 : 1 (Fig. 3B). (The rate of blockage with the I143C mutant monomers is compared with that in channels formed from wild-type monomers; there blockage occurs with = 33·2 ± 2·9, n = 4.) The relative rate of block of a given residue correlated well with the depth of the residue in the pore as predicted by the molecular model (Fig. 3B; see also Fig. 4), with I143 the deepest and C149 the most superficial residue.
 |
View larger version [in this window] [in a new window] |
|
|
A, fractional remaining current (log scale) for 4 cysteine-substituted mutants all expressed from dimeric constructs plotted against exposure time to 200 nM external Ag+. Results are fitted with a single exponential to give a time constant for the rate of decay. Time constants are given in Table 1. B, the relative rate of blockage by Ag+ and the distance along the pore axis at which blockage occurs. The distance is measured in the model shown in Fig. 4 from the ring formed by the 4 sulphur atoms of Cys149 lining the pore (i.e. 1 per subunit) to the ring formed by the equivalent 4 atoms in the respective residue. | ||
In Kv channels, the residue equivalent to the deeply positioned Thr141 is accessible from the cytoplasmic side of the membrane, where it forms a binding site for internal tetraethylammonium (TEA) ions (Yellen et al. 1991). The residue in Kv2.1 (Val) and Kv3.1 (Leu) equivalent to the deepest accessible residue found here (I143) affects both the ability to select between K+ and Rb+ and blockage by internal TEA+ (Taglialatela et al. 1993). Also in Kv, the residue equivalent to the superficially positioned Cys149 is located at the point where the wide outer vestibule first begins to narrow into the pore. Mutation of this residue in Kv to an aromatic converts the position into a high-affinity binding site for externally applied TEA (Heginbotham & MacKinnon, 1992).
Molecular model of the pore region of Kir2.1
The final version of the model of the H5 region of Kir2.1 is shown in Fig. 4. The residues accessible to externally applied Ag+ are Cys149, Phe147 and Tyr145, Thr141, Thr142 and Ile143. The central region of the pore is shown in greater detail in the space-filling models of Fig. 4B and C. Cys149, the residue responsible for Ag+ block in wild-type channels (Dart et al. 1998) lies just above the aromatic residue Phe147. The other aromatic, Tyr145, also lines the pore and lies on the other side of Phe147 from Cys149. Thr141, Thr142 and Ile143 are positioned below Tyr145 (Fig. 4C).
 |
View larger version [in this window] [in a new window] |
|
|
A, schematic 'ball and stick' representation of the pore region of Kir2.1, as viewed from outside the cell. Residues found to be blocked by Ag+ are shown in red (Cys149, Phe147 and Tyr145); residues Thr141 and Ile143 are not visible from this view. The salt bridge between Glu138 and Arg148 in adjacent subunits is indicated in cyan. Residues not blocked by Ag+ are shown in blue. The other parts of each subunit are coloured differently from each other. B, space-filling representation of the central pore region viewed from outside the cell. C, as B but viewed from inside the channel looking out. Note that although a methyl group on Ile143 appears to block the pore, this can reorientate so that it moves away from the centre of the pore. | ||
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Sequence similarity and toxin binding studies raise the possibility that all K+ channels, regardless of their overall structure, share a common K+-selective pore design. Based on this idea, we have modelled the pore-forming H5 region of Kir2.1 using as our starting point an existing model of the pore region of the voltage-gated channel Kv1.3. Residues whose side chains were predicted by the model to line the pore were tested using cysteine-scanning mutagenesis and subsequent blockage by sulfhydryl-reactive Ag+. The results of these experiments (summarized in Table 1) show good agreement with the initial model, suggesting that the pore regions of Kv and Kir channels indeed have similar topologies. Our results are also closely similar to those obtained by Lü & Miller (1995) from a cysteine scan of the H5 region of the Kv channel Shaker.
Toxin mapping of the H5 region
In voltage-gated K+ channels, the interaction between the channel protein and scorpion toxins has been used as an indirect means of assessing the spatial locations of the pore residues. These neurotoxins block Kv channels by binding tightly to the external mouth of the pore. This binding allows the dimensions of the channel pore to be mapped relative to the known structure of the toxin, provided that pore residues that interact with the blocker can be identified. Toxin mapping has provided remarkable resolution of the pore region of several members of the Kv family and reveals that the H5 pore loops form a wide, shallow outer vestibule (approximately 3 nm wide and 0·8 nm deep) with the selectivity filter sitting in a 0·5 nm depression at the centre (Aiyar et al. 1995, 1996; Ranganathan et al. 1996).
Lu & MacKinnon (1997) have recently shown that Lq2, a scorpion toxin that inhibits both Kv and KCa channels, can also block the weak inward rectifier ROMK1 (Kir1.1). The interaction surface between the toxin and the channel is structurally similar for Kv and Kir channels, suggesting that they share the same basic pore shape. However, a key interaction pairing between a lysine at the tip of the toxin (K27) and a tyrosine in the (GYG) K+ channel signature sequence was found to be much less strong in ROMK1 than in Kv channels. This lack of interaction with the toxin implies subtle differences between Kv and Kir pores. Lq2 has no effect on Kir2.1, reflecting differences in amino acid composition between ROMK1 and Kir2.1 (Lu & MacKinnon, 1997).
Similarity between Kir2.1 and Kv channels
Our result that Thr141, Thr142, Ile143, Tyr145, Phe147 and Cys149 are exposed to the narrow part of the ion conduction pathway is in broad agreement with the study of equivalent residues in the voltage-gated K+ channel Shaker (Lü & Miller, 1995). Two differences occur, at Thr141 and at Arg148. Thr141 is exposed in Kir2.1, but the accessibility of the equivalent residue in Shaker (Thr441) could not be clearly interpreted, although Thr441 has been shown to be a binding site for internally applied TEA (Yellen et al. 1991). This difference suggests that Thr141 may be closer to the centre of the ion conduction pathway in Kir2.1 than is Thr441 in Shaker. Possible reasons for this difference are the existence of the salt bridge between Glu138 and Arg148 in Kir2.1, which may alter the position of Thr141, and the presence of a conserved hydrophilic residue in Kir, Q140, in place of a conserved hydrophobic residue in Kv. We were unable to obtain any information regarding the level of exposure of Arg148, since substituting this residue did not produce functional channels. In Shaker the equivalent residue (M448) is exposed and susceptible to blockage by Ag+.
Molecular mechanism of selectivity
K+ channels are over 100-fold more permeable to K+ than to Na+ (e.g. Hille, 1973). One explanation for the molecular mechanism underlying selectivity is that K+ in the pore is co-ordinated in a cage of electrons generated at the face of aromatic rings (Kumpf & Dougherty, 1993). In our model, the aromatic side chains of residues Phe147 and Tyr145 point in towards the pore lumen (Fig. 4) and would be able to interact with K+ in the pore in such a cation- interaction. This is in contrast to the modelling studies of Ranatunga et al. (1998) who suggest that the hydroxyl group on the equivalent Tyr residue rather than cation- interactions helps confer K+ selectivity.
Based on mutagenesis studies of the signature sequence of Kv channels, however, Heginbotham et al (1994) concluded that K+ selectivity is not conferred by cation- interactions, but instead by interactions with oxygen atoms as first suggested by Bezanilla & Armstrong (1972) and Hille (1973). In this alternative model, the signature sequence provides oxygen atoms through backbone carbonyl oxygens.
Our molecular model suggests that, in addition to the aromatic rings of Phe147 and Tyr145, the carbonyl groups of Thr142 and Gly144 could also contribute to selectivity. Interestingly, Gly144 corresponds to the residue in G-protein-regulated Kir channels which, when mutated, results in loss of K+ selectivity (Slesinger et al. 1996; Kofuji et al. 1996; Navarro et al. 1996). In our current model, these carbonyl groups are significantly further apart (
0·9 nm) than the diameter of K+ (0·266 nm) suggesting, if they are involved in selectivity, the need for further refinement to move these groups closer to the centre of the pore. Thus, the model suggests that selectivity is conferred by aromatic residues (Tyr145, Phe147) perhaps with a contribution from carbonyl groups (Thr142, Gly144).
Crystal structure
Since the original submission of this manuscript, Doyle et al (1998) have reported the crystal structure of the K+ channel from the bacterium Streptomyces lividans (KcsA). Their elegant results indicate that residues within the signature sequence of KcsA (Thr-Val-Gly-Tyr-Gly) orient their side chains away from the pore so that main chain carbonyl oxygens line the selectivity filter. These carbonyl oxygen atoms are geometrically constrained so that a dehydrated K+ ion fits precisely (the carbonyl oxygens acting as 'surrogate water'), but a Na+ ion is too small to co-ordinate properly.
Side chains from the signature sequence that orient away from the channel pore make important interactions with residues from the -helical amino-terminal portion of the P region (the pore helix). Specifically, each Tyr residue from the GYG signature sequence interacts through hydrogen bonding with a Trp residue from the pore helix. This forms a network of aromatic residues which acts to structurally constrain the selectivity filter, holding the carbonyl oxygens at precisely the distance needed to optimally co-ordinate a K+ ion.
Based on our cysteine-scanning study, the selectivity filter of the inward rectifying K+ channel, Kir2.1, appears to be different from KcsA. For example, when the Tyr residue in the GYG sequence is mutated to cysteine (Y145C), the channel can be blocked by the application of extracellular Ag+. This suggests that the Tyr side chain is exposed to the channel pore. An alternative explanation is that the Tyr orients as in the KcsA crystal structure, but mutation to the shorter side chain Cys allows the thiol to reorient so as to be accessible to Ag+ in the pore. This reorientation, however, would be expected to disrupt selectivity and is not supported by our data. Also, our experiments were conducted in the presence of 70 mM external Na+ (with 70 mM [K+]o) and under these conditions, wild-type and Y145C whole-cell currents reversed at the same potential, indicating no change in the Na+/K+ permeability ratio. Additionally, Kir2.1 lacks the pore helix Trp residues which hydrogen bond with the Tyr of the GYG sequence in KcsA. It is possible that the Tyr side chain could hydrogen bond to a different residue, thereby conferring rigidity on the selectivity filter via an alternative set of hydrogen bonds. Even if this alternative exists in K+ channels where Trp is absent, site-directed mutagenesis to Cys (which removes hydrogen-bonding capability at this position) should reduce K+ selectivity, particularly towards smaller cations such as Na+, since the selectivity filter would be able to 'collapse' to co-ordinate such cations more effectively.
We conclude that the general mechanism proposed by Doyle et al. (1998) for K+ selectivity may not apply to Kir2.1, and therefore may not be universal among K+ channels.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
| Aiyar, J., Rizzi, J. P., Gutman, G. A. & Chandy, K. G. (1996). The signature sequence of voltage-gated potassium channels projects into the external vestibule. Journal of Biological Chemistry 271, 31013-31016 | [Abstract/Full Text] |
| Aiyar, J., Withka, J. M., Rizzi, J. P., Singleton, D. H., Andrewes, G. C., Lin, W., Boyd, J., Hanson, D. C., Simon, M., Dethlefs, B., Lee, C.-L., Hall, J. E., Gutman, G. A. & Chandy, K. G. (1995). Topology of the pore-region of a K+ channel revealed by the NMR-derived structures of scorpion toxins. Neuron 15, 1169-1181 | [Medline] |
| Akabas, M. H., Stauffer, D. A., Xu, M. & Karlin, A. (1992). Acetylcholine receptor channel structure probed in cysteine-substitution mutants.Science 258, 307-310 | [Medline] |
| Bezanilla, F. & Armstrong, C. M. (1972). Negative conductance caused by entry of sodium and cesium ions into the potassium channels of squid axons. Journal of General Physiology 60, 588-608 | [Medline] |
| Brunger, A. T. (1996). XPLOR Manual Version 3.843. Yale University, CT, USA. | |
| Dart, C., Leyland, M. L., Barrett-Jolley, R., Shelton, P. A., Spencer, P. J., Conley, E. C., Sutcliffe, M. J. & Stanfield, P. R. (1998). The dependence of Ag+ block of a potassium channel, murine Kir2.1, on a cysteine residue in the selectivity filter. The Journal of Physiology 511, 15-24. | [Abstract/Full Text] |
| Dart, C., Leyland, M. L., Barrett-Jolley, R. A. & Stanfield, P. R. (1997a). Block of inward rectifier IRK1 by sulphydryl-specific reagents. Biophysical Journal 72, A255. | |
| Dart, C., Leyland, M. L., Spencer, P. J., Conley, E. C., Stanfield, P. R. & Sutcliffe, M. J. (1997b). Possible topology of the H5 region of an inward rectifier K+ channel, murine IRK1. The Journal of Physiology 504.P, 155P. | |
| Doyle, D. A., Cabral, J. M., Pfuetzner, R. A., Kuo, A., Gulbis, J. M., Cohen, S. L., Chait, B. T. & MacKinnon, R. (1998). The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science 280, 69-77 | [Abstract/Full Text] |
| Heginbotham, L., Lu, Z., Abramson, T. & MacKinnon, R. (1994). Mutations in the K+ channel signature sequence. Biophysical Journal 66, 1061-1067 | [Abstract] |
| Heginbotham, L. & MacKinnon, R. (1992). The aromatic binding site for tetraethylammonium ion on potassium channels. Neuron 8, 483-491 | [Medline] |
| Hidalgo, P. & MacKinnon, R. (1995). Revealing the architecture of a K+ channel pore through mutant cycles with a peptide inhibitor. Science 268, 307-310 | [Medline] |
| Hille, B. (1973). Potassium channels in myelinated nerve. Selective permeability to small cations. Journal of General Physiology 61, 669-686 | [Medline] |
| Ho, K., Nichols, C. G., Lederer, W. J., Lytton, J., Vassilev, P., Kanazirska, M. V. & Herbert, S. C. (1993). Cloning and expression of an inwardly rectifying ATP-regulated potassium channel. Nature 362, 31-38 | [Medline] |
| IUAPAC-IUB Joint Commission on Biochemical Nomenclature (1985). Nomenclature and symbolism for amino acids and peptides. Journal of Biological Chemistry 260, 14-42. | |
| Kofuji, P., Hofer, M., Millen, K. J., Millonig, J. H., Davidson, N., Lester, H. A. & Hatten, M. (1996). Functional analysis of the weaver mutant GIRK2 K+ channel and rescue of weaver granule cells. Neuron 16, 941-952 | [Medline] |
| Kubo, Y., Baldwin, T. J., Jan, Y. N. & Jan, L. Y. (1993). Primary structure and functional expression of a mouse inward rectifier potassium channel. Nature 362, 127-133 | [Medline] |
| Kumpf, R. A. & Dougherty, D. A. (1993). A mechanism for ion selectivity in potassium channels: computational studies of cation- interactions. Science 261, 1708-1710 |
[Medline] |
| Kürz, L. L., Zühlke, R. D., Zhang, H.-J. & Joho, R. H. (1995). Side-chain accessibilities in the pore of a K+ channel probed by sulfhydryl-specific reagents after cysteine-scanning mutagenesis. Biophysical Journal 68, 900-905 | [Abstract] |
| Leyland, M. L., Shelton, P. A., Dart, C., Conley, E. C. & Stanfield, P. R. (1997). Single subunit mutations of the murine K+ channel IRK1. The Journal of Physiology 499.P, 125-126P. | |
| Liman, E. R., Tytgat, J. & Hess, P. (1992). Subunit stoichiometry of a mammalian K+ channel determined by construction of multimeric cDNAs. Neuron 9, 861-871 | [Medline] |
| Lü, Q. & Miller, C. (1995). Silver as a probe of pore-forming residues in a potassium channel. Science 268, 304-307 | [Medline] |
| Lu, Z. & MacKinnon, R. (1997). Purification, characterization and synthesis of an inward-rectifier K+ channel inhibitor from scorpion venom. Biochemistry 36, 6936-6940 | [Medline] |
| MacKinnon, R. (1991). Determination of the subunit stoichiometry of a voltage-activated potassium channel. Nature 350, 232-235 | [Medline] |
| MacKinnon, R. & Yellen, G. (1990). Mutations affecting TEA blockade and ion permeation in voltage-activated K+ channels. Science 250, 276-279 | [Medline] |
Navarro, B., Kennedy, M. E., Velimirovic, B., Bhat, D., Peterson, A. S. & Clapham, D. E. (1996). Nonselective and G -insensitive weaver K+ channels. Science 272, 1950-1953 |
[Abstract] |
| Pascual, J. M., Sheih, C.-C., Kirsch, G. E. & Brown, A. M. (1995). K+ pore structure revealed by reporter cysteines at inner and outer surfaces. Neuron 14, 1055-1063 | [Medline] |
| Ranatunga, K. M., Kerr, I. D., Adcock, C., Smith, G. R. & Sansom, M. S. P. (1998). Protein-water-ion interactions in a model of the pore domain of a potassium channel - a simulation study. Biochimica et Biophysica Acta 1370, 1-7 | [Medline] |
| Ranganathan, R., Lewis, J. H. & MacKinnon, R. (1996). Spatial localization of the K+ channel selectivity filter by mutant cycle-based structure analysis. Neuron 16, 131-139 | [Medline] |
| Sali, A. & Blundell, T. L. (1993). Comparative protein modelling by satisfaction of spatial restraints. Journal of Molecular Biology 234, 779-815 | [Medline] |
| Slesinger, P., Patil, N., Liao, Y. J., Jan, Y. N., Jan, L. Y. & Cox, D. R. (1996). Functional effects of the mouse weaver mutation on G-protein-gated inwardly rectifying K+ channels. Neuron 16, 321-331 | [Medline] |
| Taglialatela, M., Drewe, J. A., Kirsch, G. E., Debiasi, M., Hartmann, H. A. & Brown, A. M. (1993). Regulation of K+/Rb+ permeability and internal TEA blockade by mutations at a single site in K+ pores. Pflügers Archiv 423, 104-112. | |
| Yang, J., Jan, Y. N. & Jan, L. Y. (1995). Determination of the subunit stoichiometry of an inwardly rectifying potassium channel. Neuron 15, 1441-1447 | [Medline] |
| Yang, J., Yu, M., Jan, Y. N. & Jan, L. Y. (1997). Stabilization of ion selectivity filter by pore loop ion pairs in an inwardly rectifying potassium channel. Proceedings of the National Academy of Sciences of the USA 94, 1568-1572 | [Abstract/Full Text] |
| Yellen, G., Jurman, M. E., Abramson, T. & MacKinnon, R. (1991). Mutations affecting internal TEA blockade identify the probable pore-forming region of K+ channel.Science 251, 939-942 | [Medline] |
| Yool, A. J. & Schwarz, T. L. (1991). Alteration of ionic selectivity of a K+ channel by mutation of the H5 region. Nature 349, 700-704 | [Medline] |
Acknowledgements
We thank K. George Chandy and George A. Gutman for generously supplying the co-ordinates for their model of Kv1.3. We thank The Wellcome Trust and The Royal Society for their support. M. J. S. is a Royal Society University Research Fellow.
Corresponding author
P. R. Stanfield: Ion Channel Group, Department of Cell Physiology and Pharmacology, University of Leicester, PO Box 138, Leicester LE1 9HN, UK.
Email: prs{at}le.ac.uk
This article has been cited by other articles:
|
|
 |
|
  Z. T. Schug, P. C. A. da Fonseca, C. D. Bhanumathy, L. Wagner II, X. Zhang, B. Bailey, E. P. Morris, D. I. Yule, and S. K. Joseph Molecular Characterization of the Inositol 1,4,5-Trisphosphate Receptor Pore-forming Segment J. Biol. Chem., February 1, 2008; 283(5): 2939 - 2948. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Bichet, Y.-F. Lin, C. A. Ibarra, C. S. Huang, B. A. Yi, Y. N. Jan, and L. Y. Jan Evolving potassium channels by means of yeast selection reveals structural elements important for selectivity PNAS, March 30, 2004; 101(13): 4441 - 4446. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Alagem, S. Yesylevskyy, and E. Reuveny The Pore Helix Is Involved in Stabilizing the Open State of Inwardly Rectifying K+ Channels Biophys. J., July 1, 2003; 85(1): 300 - 312. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Fujiwara and Y. Kubo Ser165 in the Second Transmembrane Region of the Kir2.1 Channel Determines its Susceptibility to Blockade by Intracellular Mg2+ J. Gen. Physiol., October 29, 2002; 120(5): 677 - 693. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. A. Schwalbe, A. Rudin, S.-L. Xia, and C. S. Wingo Site-directed Glycosylation Tagging of Functional Kir2.1 Reveals That the Putative Pore-forming Segment Is Extracellular J. Biol. Chem., June 28, 2002; 277(27): 24382 - 24389. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Sheng, J. Li, K. A. McNulty, D. Avery, and T. R. Kleyman Characterization of the Selectivity Filter of the Epithelial Sodium Channel J. Biol. Chem., March 17, 2000; 275(12): 8572 - 8581. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Sheng, J. Li, K. A. McNulty, T. Kieber-Emmons, and T. R. Kleyman Epithelial Sodium Channel Pore Region. STRUCTURE AND ROLE IN GATING J. Biol. Chem., January 5, 2001; 276(2): 1326 - 1334. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, currents recorded from the same cell after 90 s exposure to 200 nM external Ag+ are indicated by 
. B, histogram showing the percentage inhibition of the whole-cell current for the H5 mutants measured at -102 mV following 90 s exposure to 200 nM external Ag+. 
Q140C as a tetramer. ² I143C was expressed as a monomer. R indicates the Arg148 mutant, which did not give currents (see text). * P < 0·05.