The structure of the potassium channel: molecular basis of K+ conduction and selectivity

The potassium channel from Streptomyces lividans is an integral membrane protein with sequence similarity to all known K+ channels, particularly in the pore region. X-ray analysis with data to 3.2 angstroms reveals that four identical subunits create an inverted teepee, or cone, cradling the selectivity filter of the pore in its outer end. The narrow selectivity filter is only 12 angstroms long, whereas the remainder of the pore is wider and lined with hydrophobic amino acids. A large water-filled cavity and helix dipoles are positioned so as to overcome electrostatic destabilization of an ion in the pore at the center of the bilayer. Main chain carbonyl oxygen atoms from the K+ channel signature sequence line the selectivity filter, which is held open by structural constraints to coordinate K+ ions but not smaller Na+ ions. The selectivity filter contains two K+ ions about 7.5 angstroms apart. This configuration promotes ion conduction by exploiting electrostatic repulsive forces to overcome attractive forces between K+ ions and the selectivity filter. The architecture of the pore establishes the physical principles underlying selective K+ conduction.

Purification, characterization, and synthesis of an inward-rectifier K+ channel inhibitor from scorpion venom

We have purified a protein inhibitor of an inward-rectifier K+ channel, ROMK1, from the venom of the scorpion Leiurus quinquestriatus var. hebraeus. The inhibitor is Lq2, a previously discovered blocker of voltage- and Ca2+-activated K+ channels. Mutations were made on the channel and the inhibitor, and the resulting effects were examined using an electrophysiological assay. The data show that Lq2 blocks the pore of ROMK1, and that the interaction surface on Lq2 is the same for binding to inward-rectifier, voltage-activated, or Ca2+-activated K+ channels. These findings support the notion that different classes of K+ channels have different gates but a similar K+-selective pore structure.

Hanatoxin modifies the gating of a voltage-dependent K+ channel through multiple binding sites

We studied the mechanism by which Hanatoxin (HaTx) inhibits the drk1 voltage-gated K+ channel. HaTx inhibits the K+ channel by shifting channel opening to more depolarized voltages. Channels opened by strong depolarization in the presence of HaTx deactivate much faster upon repolarization, indicating that toxin bound channels can open. Thus, HaTx inhibits the drk1 K+ channel, not by physically occluding the ion conduction pore, but by modifying channel gating. Occupancy of the channel by HaTx was studied using various strength depolarizations. The concentration dependence for equilibrium occupancy as well as the kinetics of onset and recovery from inhibition indicate that multiple HaTx molecules can simultaneously bind to a single K+ channel. These results are consistent with a simple model in which HaTx binds to the surface of the drk1 K+ channel at four equivalent sites and alters the energetics of channel gating.

Mapping the receptor site for hanatoxin, a gating modifier of voltage-dependent K+ channels

Hanatoxin (HaTx) binds to multiple sites on the surface of the drk1 voltage-gated K+ channel and modifies channel gating. We set out to identify channel residues that contribute to form these HaTx binding sites. Chimeras constructed using the drk1 and shaker K+ channels suggest that the S3-S4 linker may contain influential residues. Alanine scanning mutagenesis of the region extending from the C terminal end of S3 through S4 identified a number of residues that likely contribute to form the HaTx binding sites. The pore blocker Agitoxin2 and the gating modifier HaTx can simultaneously bind to individual K+ channels. These results suggest that residues near the outer edges of S3 and S4 form the HaTx binding sites and are eccentrically located at least 15 A from the central pore axis on the surface of voltage-gated K+ channels.

Crystal structures of a complexed and peptide-free membrane protein-binding domain: molecular basis of peptide recognition by PDZ

Modular PDZ domains, found in many cell junction-associated proteins, mediate the clustering of membrane ion channels by binding to their C-terminus. The X-ray crystallographic structures of the third PDZ domain from the synaptic protein PSD-95 in complex with and in the absence of its peptide ligand have been determined at 1.8 angstroms and 2.3 angstroms resolution, respectively. The structures reveal that a four-residue C-terminal stretch (X-Thr/Ser-X-Val-COO(-)) engages the PDZ domain through antiparallel main chain interactions with a beta sheet of the domain. Recognition of the terminal carboxylate group of the peptide is conferred by a cradle of main chain amides provided by a Gly-Leu-Gly-Phe loop as well as by an arginine side chain. Specific side chain interactions and a prominent hydrophobic pocket explain the selective recognition of the C-terminal consensus sequence.

Contribution of the S4 segment to gating charge in the Shaker K+ channel

Voltage-activated ion channels respond to changes in membrane voltage by coupling the movement of charges to channel opening. A K+ channel-specific radioligand was designed and used to determine the origin of these gating charges in the Shaker K+ channel. Opening of a Shaker K+ channel is associated with a displacement of 13.6 electron charge units. Gating charge contributions were determined for six of the seven positive charges in the S4 segment, an unusual amino acid sequence in voltage-activated cation channels consisting of repeating basic residues at every third position. Charge-neutralizing mutations of the first four positive charges led to large decreases (approximately 4 electron charge units each) in the gating charge; however, the gating charge of Shaker delta 10, a Shaker K+ channel with 10 altered nonbasic residues in its S4 segment, was found to be identical to the wild-type channel. These findings show that movement of the NH2-terminal half but not the CO2H-terminal end of the S4 segment underlies gating charge, and that this portion of the S4 segment appears to move across the entire transmembrane voltage difference in association with channel activation.

Agitoxin footprinting the shaker potassium channel pore

In voltage-dependent K+ channels, each of the four identical subunits contributes one pore loop to the central ion selectivity unit at the interface between the subunits. The pore loop is also the target for scorpion venom peptide inhibitors. These inhibitors bind at the pore entryway between the four subunits and can assume any one of four orientations. The orientations become distinguishable only if the binding site symmetry is disrupted. We have used mutagenesis and site-directed chemical modification to alter pore loop amino acids in either one or four subunits. The effects of these alterations on inhibitor affinity define the eccentricity of amino acids in the pore entryway and imply a different secondary structure for the amino and carboxyl ends of the pore loop.

Spatial localization of the K+ channel selectivity filter by mutant cycle-based structure analysis

The structurally well-characterized scorpion toxin Agitoxin2 inhibits ion permeation through Shaker K+ channels by binding to the external pore entryway. Scanning mutagenesis identified a set of inhibitor residues critical for making energetic contacts with the channel. Using thermodynamic mutant cycle analysis, we have mapped channel residues relative to the known inhibitor structure. This study constrains the position of multiple channel residues within the pore-forming loops; in one stretch, we have been able to map five out of seven contiguous residues to the inhibitor interaction surface, including those involved in ion selectivity. One interaction in particular, that of K27M on the inhibitor with Y445F on the channel, is unique in that it depends on the K+ ion concentration. These results reveal a shallow vestibule formed by the pore loops at the K+ channel entryway. The selectivity filter is located at the center of the vestibule close to (approximately 5 A) the extracellular solution.

Divalent cation selectivity in a cyclic nucleotide-gated ion channel

Divalent metal cation selectivity was studied in guanosine 3',5'-cyclic monophosphate-gated ion channels. Channels from bovine retina were expressed in Xenopus laevis oocytes, and currents were measured using tight-seal patch recording methods. The ability of divalent cations to block Na+ currents was used to determine the occupancy of divalent cations in the ion conduction pore. At positive membrane voltages, where extracellular divalent cations are near equilibrium with their binding site, the occupancy reflects the affinity of the blocking ion. The selectivity sequence based on relative affinity was Ca2+ > Mg2+ = Sr2+ = Ba2+. In addition to its higher affinity, Ca2+ was more permeant and blocked with a weaker voltage dependence. Ca2+ was the only ion that blocked with a high Hill coefficient (n = 2.7), suggesting the presence of multiple binding sites. When Glu 363, located in the pore-forming region, was mutated to Asp, the affinity of all four ions increased and the selectivity sequence became Ca2+ > Sr2+ > Ba2+ > Mg2+. These results show that the channel is highly selective for Ca2+ and that Glu 363 mediates divalent cation selectivity of the channel.

Probing a potassium channel pore with an engineered protonatable site

Blockade by intracellular cations reduces outward conduction of K+ in inward rectifier K+ channels. Mutations of residue 171 in the second transmembrane (M2) segment of the ROMK1 channel have been found to affect the affinity for blockade by intracellular Mg2+ and polyamines. In the present study, we examined the mechanism by which this residue mediates blockade by placing a proton acceptor (histidine) at this position. The results allow us to draw two conclusions. First, the side chain of residue 171 is located in the ion conduction pore about halfway across the transmembrane voltage drop. Second, its side chain comes into close contact and interacts electrostatically with a blocking ion.

An inhibitor of the Kv2.1 potassium channel isolated from the venom of a Chilean tarantula

The Kv2.1 voltage-activated K+ channel, a Shab-related K+ channel isolated from rat brain, is insensitive to previously identified peptide inhibitors. We have isolated two peptides from the venom of a Chilean tarantula, G. spatulata, that inhibit the Kv2.1 K+ channel. The two peptides, hanatoxin1 (HaTx1) and hanatoxin2 (HaTx2) are unrelated in primary sequence to other K+ channel inhibitors. The activity of HaTx was verified by synthesizing it in a bacterial expression system. The concentration dependence for both the degree of inhibition at equilibrium (Kd = 42 nM) and the kinetics of inhibition (kon = 3.7 x 10(4) M-1s-1; koff = 1.3 x 10(-3) s-1), are consistent with a bimolecular reaction between HaTx and the Kv2.1 K+ channel. Shaker-related, Shaw-related, and eag K+ channels were relatively insensitive to HaTx, whereas a Shal-related K+ channel was sensitive. Regions outside the scorpion toxin binding site (S5-S6 linker) determine sensitivity to HaTx. HaTx introduces a new class of K+ channel inhibitors that will be useful probes for studying K+ channel structure and function.

Solution structure of the potassium channel inhibitor agitoxin 2: caliper for probing channel geometry

The structure of the potassium channel blocker agitoxin 2 was solved by solution NMR methods. The structure consists of a triple-stranded antiparallel beta-sheet and a single helix covering one face of the beta-sheet. The cysteine side chains connecting the beta-sheet and the helix form the core of the molecule. One edge of the beta-sheet and the adjacent face of the helix form the interface with the Shaker K+ channel. The fold of agitoxin is homologous to the previously determined folds of scorpion venom toxins. However, agitoxin 2 differs significantly from the other channel blockers in the specificity of its interactions. This study was thus focused on a precise characterization of the surface residues at the face of the protein interacting with the Shaker K+ channel. The rigid toxin molecule can be used to estimate dimensions of the potassium channel. Surface-exposed residues, Arg24, Lys27, and Arg31 of the beta-sheet, have been identified from mutagenesis studies as functionally important for blocking the Shaker K+ channel. The sequential and spatial locations of Arg24 and Arg31 are not conserved among the homologous toxins. Knowledge on the details of the channel-binding sites of agitoxin 2 formed a basis for site-directed mutagenesis studies of the toxin and the K+ channel sequences. Observed interactions between mutated toxin and channel are being used to elucidate the channel structure and mechanisms of channel-toxin interactions.

The cosmid CSSM25 assigns syntenic group U2 to bovine chromosome 9 and is localized to ovine chromosome 8

The cosmid-derived microsatellite CSSM 25 has previously been shown to map to bovine syntenic group U2 by linkage and hybrid somatic cell analysis. We have mapped the cosmid by fluorescent in situ hybridization to bovine Chromosome (Chr) 9q17-21 and ovine Chr 8q17-21 and hence assign U2 to Chr 9 in cattle. Bovine Chr 9 and ovine Chr 8 show strong banding pattern homology, and the localization of CSSM 25 to the same region confirms the strong conservation of gene locations on these chromosomes.

Revealing the architecture of a K+ channel pore through mutant cycles with a peptide inhibitor

Thermodynamic mutant cycles provide a formalism for studying energetic coupling between amino acids on the interaction surface in a protein-protein complex. This approach was applied to the Shaker potassium channel and to a high-affinity peptide inhibitor (scorpion toxin) that binds to its pore entryway. The assignment of pairwise interactions defined the spatial arrangement of channel amino acids with respect to the known inhibitor structure. A strong constraint was placed on the Shaker channel pore-forming region by requiring its amino-terminal border to be 12 to 15 angstroms from the central axis. This method is directly applicable to sodium, calcium, and other ion channels where inhibitor or modulatory proteins bind with high affinity.

Transfer of the scorpion toxin receptor to an insensitive potassium channel

Voltage-dependent potassium channels belong to a family of structurally related cation channels that underlie the electrical activity of excitable cells. Many potassium channels are blocked with high affinity by scorpion toxins, whereas others are completely insensitive. We transferred toxin sensitivity from the highly sensitive Kv1.3 (KV3) to the insensitive Kv2.1 (DRK1) potassium channel by transferring the stretch of amino acids between transmembrane domains 5/6. We provide evidence that this S5-S6 linker, which has been shown to comprise the pore-forming region, is probably the only part of the ion channel that directly interacts with bound toxin. Using site-directed mutagenesis, we identified specific residues in the S5-S6 linker that are responsible for the acquisition of toxin sensitivity by Kv2.1.

Two identical noninteracting sites in an ion channel revealed by proton transfer

The functional consequences of single proton transfers occurring in the pore of a cyclic nucleotide-gated channel were observed with patch recording techniques. These results led to three conclusions about the chemical nature of ion binding sites in the conduction pathway: The channel contains two identical titratable sites, even though there are more than two (probably four) identical subunits; the sites are formed by glutamate residues that have a pKa (where K(a) is the acid constant) of 7.6; and protonation of one site does not perturb the pKa of the other. These properties point to an unusual arrangement of carboxyl side-chain residues in the pore of a cation channel.

Electrostatic tuning of Mg2+ affinity in an inward-rectifier K+ channel

Inward-rectifier potassium channels conduct K+ across the cell membrane more efficiently in the inward than outward direction. This unusual conduction property is directly related to the biological action of these channels. One basis for inward rectification is voltage-dependent blockade by intracellular Mg2+ (refs 1, 7-9): strong inward-rectifier channels are so sensitive to intracellular Mg2+ that no outward K+ current is measurable under physiological conditions; weak inward rectifiers are less sensitive and allow some K+ to flow outwards. Background K1 channels and acetylcholine-regulated K+ channels from the heart are examples of strong inward rectifiers and ATP-sensitive K+ channels are weak rectifiers. Here we show that mutations at one position in the second transmembrane segment can alter the Mg2+ affinity and convert a weakly rectifying channel (ROMK1) into a strong rectifier. The amino acid at this position exposes its side chain to the aqueous pore and affects Mg2+ blockade as well as K+ conduction through an electrostatic mechanism.

A conductance maximum observed in an inward-rectifier potassium channel

One prediction of a multi-ion pore is that its conductance should reach a maximum and then begin to decrease as the concentration of permeant ion is raised equally on both sides of the membrane. A conductance maximum has been observed at the single-channel level in gramicidin and in a Ca(2+)-activated K+ channel at extremely high ion concentration (> 1,000 mM) (Hladky, S. B., and D. A. Haydon. 1972. Biochimica et Biophysica Acta. 274:294-312; Eisenmam, G., J. Sandblom, and E. Neher. 1977. In Metal Ligand Interaction in Organic Chemistry and Biochemistry. 1-36; Finkelstein, P., and O. S. Andersen. 1981. Journal of Membrane Biology. 59:155-171; Villarroel, A., O. Alvarez, and G. Eisenman. 1988. Biophysical Journal. 53:259a. [Abstr.]). In the present study we examine the conductance-concentration relationship in an inward-rectifier K+ channel, ROMK1. Single channels, expressed in Xenopus oocytes, were studied using inside-out patch recording in the absence of internal Mg2+ to eliminate blockade of outward current. Potassium, at equal concentrations on both sides of the membrane, was varied from 10 to 1,000 mM. As K+ was raised from 10 mM, the conductance increased steeply and reached a maximum value (39 pS) at 300 mM. The single-channel conductance then became progressively smaller as K+ was raised beyond 300 mM. At 1000 mM K+, the conductance was reduced to approximately 75% of its maximum value. The shape of the conductance-concentration curve observed in the ROMK1 channel implies that it has multiple K(+)-occupied binding sites in its conduction pathway.

Purification and characterization of three inhibitors of voltage-dependent K+ channels from Leiurus quinquestriatus var. hebraeus venom

Three new toxins from the venom of the scorpion Leiurus quinquestriatus var. hebraeus have been identified on the basis of their ability to block the Shaker K+ channel. These toxins have been purified using HPLC techniques and characterized as 38 amino acid peptides by mass spectroscopy, amino acid analysis, and sequence determination. Their chemical identity was confirmed by producing fully functional synthetic toxins using recombinant methods. These peptides are potent inhibitors of the Shaker K+ channel (Kd < 1 nM) as well as the mammalian homologues of Shaker. They are related to other previously described K+ channel toxins, but form a new subclass within the larger family of K+ channel inhibitors derived from scorpion venom. We have named these toxins agitoxin 1, 2, and 3, respectively.

Mutations in the K+ channel signature sequence

Potassium channels share a highly conserved stretch of eight amino acids, a K+ channel signature sequence. The conserved sequence falls within the previously defined P-region of voltage-activated K+ channels. In this study we investigate the effect of mutations in the signature sequence of the Shaker channel on K+ selectivity determined under bi-ionic conditions. Nonconservative substitutions of two threonine residues and the tyrosine residue leave selectivity intact. In contrast, mutations at some positions render the channel nonselective among monovalent cations. These findings are consistent with a proposal that the signature sequence contributes to a selectivity filter. Furthermore, the results illustrate that the hydroxyl groups at the third and fourth positions, and the aromatic group at position seven, are not essential in determining K+ selectivity.

Conduction properties of the cloned Shaker K+ channel

The conduction properties of the cloned Shaker K+ channel were studied using electrophysiological techniques. Single channel conductance increases in a sublinear manner with symmetric increases in K+ activity, reaching saturation by 0.6 M K+. The Shaker K+ channel is highly selective among monovalent cations; under bi-ionic conditions, its selectivity sequence is K+ > Rb+ > NH+4 > Cs+ > Na+, whereas, by relative conductance in symmetric solutions, it is K+ > NH+4 > Rb+ > Cs+. In Cs+ solutions, single channel currents were too small to be measured directly, so nonstationary fluctuation analysis was used to determine the unitary Cs+ conductance. The single channel conductance displays an anomalous molefraction effect in symmetric mixtures of K+ and NH+4, suggesting that the conducting pore is occupied by multiple ions simultaneously.

Functional stoichiometry of Shaker potassium channel inactivation

Shaker potassium channels from Drosophila are composed of four identical subunits. The contribution of a single subunit to the inactivation gating transition was investigated. Channels carrying a specific mutation in a single subunit can be labeled in a heterogeneous population and studied quantitatively with scorpion toxin sensitivity as a selection tag. Linkage within a single subunit of a mutation that removes the inactivation gate to a second mutation that affects scorpion toxin sensitivity demonstrates that only a single gate is necessary to produce inactivation. The inactivation rate constant for channels with a single gate was one-fourth that of channels with four gates. In contrast, the rate of recovery from inactivation was independent of the number of gates. It appears that each of the four open inactivation gates in a Shaker potassium channel is independent, but only one of the four gates closes in a mutually exclusive manner.