Potassium channel families arranged according to their subunit structure. Potassium channel families can be grouped in those having two transmembrane segments (2TM; Kir), 4TM (2‐pore domain), 6TM (voltage gated and SK), and 7TM (Slo). Note that for the sake of simplicity the large‐conductance Slo channel family includes the Slo2.x channels, which have only six transmembrane domains. The 6TM domain class can be divided into four families: Voltage‐gated Kv, voltage‐gated KCNQ‐type (KCNQ); ether‐a‐go‐go (Eag), and Ca²⁺‐activated channels (SK). Subdivisions of the voltage‐gated Kv channels into four subfamilies and Eag into three subfamilies are also named according to the Drosophila melanogaster genes. In the SK family IKCa1 stands for intermediate conductance Ca²⁺‐activated K⁺ channel.

Phylogenetic tree of Kir channels and their current‐voltage curves. (A) Amino acid sequence alignments and phylogenetic analysis for the 15 known members of the human Kir family. International Union of Pharmacology and Hugo Gene Nomenclature Committee names of the genes are shown. The subunits were classified into four functional groups following Hibino et al. (203). (B) Inward rectification and conductance are strongly external K⁺ concentration‐dependent. I‐V relationships are of the starfish egg cell membrane at four different K_ext concentrations in Na⁺‐free media. Continuous and broken line indicates instantaneous and steady‐state current, respectively (adapted. with permission, from reference 185). Notice that K⁺ conductance develops at voltages negative to the equilibrium potential for K⁺ (E_K). (C) I‐V relationship characteristic of a “weak” inward rectifier. (D) In “strong” inward rectifiers K⁺ conductance tends to zero as the membrane potential is depolarized and contrary to expectations the crossover phenomena produces an increase in K⁺ conductance at voltages larger than the crossover voltage despite the decrease in the K⁺ driving force.

Kir2.1 induces a smooth muscle cell hyperpolarization when K_ext increases. (A) The average current densities at three different [K_ext] were obtained in response to a voltage ramp from −130 to 0 mV lasting for 140 ms. (B) Ba²⁺‐sensitive currents densities recorded in the same condition as in A. (C) Elevation of K_ext from 3 mmol/L to 15 mmol/L caused a membrane potential hyperpolarization of smooth muscle cells [adapted, with permission, from Filosa et al. (139)]. (D) Chord conductance‐voltage curve at the same experimental conditions as in B. Notice that there is an appreciable increase in smooth muscle cells Kir conductance as the [K_ext] is raised at physiological membrane potentials (−50 to −40 mV).

Dual modulation of K_G channels by G protein‐coupled receptor (GPCR) and the topology and structure of Kir6.x. (A) Agonist activation of GPCR coupled to pertussis (PTX)‐sensitive α_i/o‐type of G protein promotes activation of K_G channels. Activation of K_G channels is induced by binding to the channel‐forming protein of the βγ complex of the G protein. Agonist binding to α_q‐type of G protein results in channel inhibition that is a consequence of the activation of phospholipase C (PLC), which in turn hydrolyses phosphatidyl‐inositol‐4,5‐bisphosphate (PIP₂). Other modulators include tyrosine kinase (TK), Ca²⁺‐calmodulin‐dependent kinase 2 (CAMK2), and protein phosphatase (PP1). Modified, with permission, from reference 332. For more details, see text. (B) SUR subunits contain 17 transmembrane segments assembled in three domains, TMD0‐2, and containing two nucleotide‐binding domains (NBD) contained between TMD1 and TMD2 (NBD1) and in the C‐terminus (NBD2). The structures show top and side views of the entire K_ATP channel complex analyzed at 18 Å resolution. Blue represents Kir6.x. Red represents the rest of SUR and yellow represents TMD0 of SUR [adapted, with permission, from Mikhailov et al. (367)].

Kir channel crystal structure and cation‐binding sites. (A) Voltage dependence in Kir channels arises as a consequence of the movement of K⁺ ions contained in the cytoplasmic pore. (B) Crystal structure of a Kir3.1‐prokaryotic Kir channel chimera determined at 2.2 Å. Seven Rb⁺ ions were located in the conduction pore. Two constriction sites, F181 side chain and residues 302‐309 Cα atoms in the G‐loop are colored in blue. For the sake of clarity only two subunits are shown [adapted, with permission, from Nishida et al. (394)]. (C) Crystal structure of the cytoplasmic pore of S225E mutant of Kir3.1 (yellow) and the Kir chimera (308) (cyan). Na⁺ ions are represented by orange spheres and Rb⁺ ions by pink spheres. (D) Crystal structure model of the cytoplasmic pore of S225E mutant Kir3.1 corresponding to the boxed region in A. The residues, Q225, G227, G261, D260, F255, and S256, interact directly or through water molecules with the Na⁺ ions located at S8‐S11. The positions of the phenylalanines coordinating the Na⁺ through π‐cation interactions at site S10 are shown [adapted, with permission, from Xu et al. (594)].

Molecular determinants of inward rectification and location of modulators binding sites in the cytoplasmic domain of K_G channels. (A, B) Amino acid residues in the cytoplasmic pore determining inward rectification in Kir2.1 channels. (C) Current‐voltage relationships for different Kir2.1 point mutants [adapted, with permission, from Pegan et al. (425)]. (D) The structure shown contains the cytoplasmic domains of Kir3.1 a G protein‐gated channel and the transmembrane domains and pore region of the chimeric Kir channel. The regions implicated in Na⁺, PIP2, G protein, and alcohol binding are shown [adapted, with permission from Luscher and Slensinger (332)].

Diversity of 2‐pore (2P)‐domain K+ channel (K2P) subunits and membrane topology. (A) The alignment was made using the web tool: Phylogeny.fr (109), with different sequences of human two pore K+ channels obtained from gene bank accession numbers from KCNK1 to KCNK18: NP_002236, NP_055032, NP_002237.1, NP_201567.1, NP_003731.1., NP_004814.1., NP_005705.1., NP_057685.1., NP_066984.1., NP_071338.1., NP_071337.2., NP_071753.1., NP_115491.1., NP_113648.2., and NP_862823.1. (B) Putative membrane topology of the two‐pore domain K⁺ channels. Green spheres indicate pH sensing residues and their predicted location in the first turret loop and M4 transmembrane domain. (C) Multiple sequence alignment of the outer and inner helix region of KcsA, hERG, and several K_2P K⁺ channels. Amino acid residues colored in red show the K⁺ channel signature sequence, corresponding to the selectivity filter.

Polymodal nature of K_2P channels receptors. (A) TREK‐1 channels are modulated by stretch, heat, intracellular acidosis, depolarization, lipids, general anaesthetics, and tonically inhibited by the actin cytoskeleton [adapted, with permission, from Patel and Honoré (417)]. (B) Polymodal regulation of TREK‐1 and TREK‐2. Activation of the Gs/cAMP/protein kinase A (PKA) and the Gq/phospholipase C (PLC)/Diacyl Glycerol (DAG)/protein kinase C (PKC) signaling pathway inhibit TREK channels by phosphorylating serine residues present on the C‐terminal. TREK‐1 is activated via the NO/cGMP/Protei kinase G (PKG) pathway, but the PKG phosphorylation consensus site is missing in TREK‐2. (Arrows indicate stimulation; lines with T ending represent inhibition.) [Modified, with permission, from Enyedi and Czirják (128).] (C) Regulation of TASK‐1 and TASK‐3. The channels are inhibited by extracellular acidification (EC) acidification as a result of protonation of histidine98 in the second extracellular loop. Anandamide inhibits both TASK‐1 and TASK‐3. Hypoxia inhibits TASK current indirectly. TASK channels are activated by halothane and isoflurane but they are not influenced by chloroform or ether. The polycation ruthenium red and Zn²⁺ allow pharmacological distinction between the two closely related channel subunits. Dashed lines represent effects on targets; arrows indicate stimulation; lines with T ending represent inhibition. [Modified, with permission, from Enyedi and Czirják (128).]

K2P channel activation by different stimuli. (A) Top: TREK‐1 activation was graded with membrane stretch in a cell‐attached patch from oocytes expressing TREK‐1. The inset shows channel openings with an enlarged time scale. In this patch, a small conductance endogenous channel was also present. Bottom‐graded reversible negative pressure activation of hTRAAK in physiological K⁺ conditions. The patch was held at 0 mV and the zero current is indicated by a dashed line [from Patél et. al (418) and Lesage and Lazdunski (305)]. (B) TREK‐1 channels show outward rectification. Single‐channel currents recorded in absence of Mg²⁺ at −100 mV (left trace), 0 mV (middle trace), and 100 mV (right trace) at atmospheric pressure (top traces) and at a steady pressure of −30 mm Hg (bottom traces). Po denotes open probability [adapted, with permission, from Maingret et al. (342)]. (C) Thermosensitivity. Cell‐attached patches from COS‐7 cells incubated at different bath temperatures are shown for TREK‐1 and TRAAK [adapted, with permission, from Kang et al. (253)]. (D) The C‐terminus of TREK‐2 is required for sensitivity to fatty acids and pH. (Top) Wild‐type TREK‐2 expressed in COS‐7 cells is robustly activated by a decrease in intracellular pH. Middle. The pH sensitivity is abolished in a chimeric mutant that consists of the core transmembrane segments of TREK‐2 and the C‐terminus of TASK‐3 (red) (chimera TREK‐2–TASK‐3C) indicating that C‐terminus of TREK‐2 is require to pH sensitivity. (Bottom) The sensitivity of a TRAAK–TASK‐3C chimera to pH is similar to wild‐type TRAAK, which indicates that the C‐terminus of TRAAK is unlikely to mediate activation by pH [adapted, with permission, from Kim (260)]. (E) Left. TREK‐1 is reversibly opened by chloroform (0.8 mmol/L). Voltage was linearly depolarized with a voltage ramp from −120 to 100 mV. Current becomes zero at a membrane potential equal to the equilibrium potential for K⁺ (−80 mV). Inset: stimulation of the K⁺ current by chloroform (CHCl₃) is concentration dependent and observed at pharmacologically relevant concentrations. Right top. Chloroform (0.8 mmol/L) induces reproducible membrane hyperpolarizations. Right bottom. Halothane (1 mmol/L; 0 mV) induces TREK‐1 single‐channel activity characterized by rapid flickering between closed and open states [modified, with permission, from Franks and Honore (147)]. (F) Fatty acid activation of TRAAK and TREK in COS cells. (Left side) Current‐voltage relationships obtained in an inside‐out patch with voltage ramps ranging from –150 to +50 mV, 500 ms in duration, before (control), after 3 min perfusion with 10 μmol/L AA and after the wash. Inset: effects of 10 μmol/L AA on TRAAK currents recorded in an inside‐out patch clamped at +20 mV. The zero current level is indicated by an arrow. (Right side) Inside out patch currents were recorded at 0mV from transfected COS cells. The zero current levels are indicated by a dotted line. The histograms represent the ratio of the mean currents recorded before (I_control) or after 10 μmol/L of AA application (I), gray and black color denotes absence or coexpression of A‐kinase anchoring protein (AKAP150), respectively [adapted, with permission, from Sandoz et al. (468) and Fink et al. (141)].

K_2P channel structure. (A) Homology model for ΔK2PØ (K2PØ channel variant lacking AA from 299 to 1000) channel shows bilateral symmetry with a 4‐fold symmetric selectivity filter. Color red indicates monomer A (from residue 1 to 152) and color blue monomer B (from residue 174 to 276). (B) Extracellular (top) and cytoplasmic (bottom) sides reveals overall symmetry like a parallelogram. The model includes residues 1 to 276 without the TM1‐P1 loop (residues 30‐91), TM2‐TM3 linker (residues 153‐173), and TM2‐P2 loop (residues 225‐238). (Bottom) Side view of domain I of both subunits. The glutaminase interacting protein (GIP) motif (G129‐I130‐P131) in TM2 is indicated. Side view of domain II of both subunits. Proline residue 183 and 192 in TM3 are indicated. (C) Structure of a mutant TASK‐3 channel modeled in an open state, using the structure of KvAP [adapted, with permission, from Jiang et al. (238)] as template. It is hypothesized that channels open through flexion of M2 and M4 around hinge glycines G117 and G231. The positions of these hinge glycines are indicated as spheres in the helices M2 and M4. Gain of function mutants stabilizes the open state through altered side chain‐side chain interactions between residues. A possible H‐bond between Thr in position 237 of M4 (in mutant A237T) and N133, which may contribute to stabilizing the open state, is indicated. The model gives a bond length of 3.2 Å. (D) K2P3.1 model, illustrating the interactions of a water molecule with the backbone of Tyr‐96 and Gly‐97 and the side chains of Thr‐89 and His‐98 in the unprotonated form of His‐98, according to molecular dynamic simulations, based on Yuill et al. (608). (E) pH‐sensing mechanism of human K2P2.1. Ribbon representation of one subunit of the bacterial KcsA potassium channel, based on the published structure [Doyle et al. (115)]. Predicted hydrogen bonds between KcsA residues are presented as orange lines. The side chain of Glu‐51 is predicted to form hydrogen bonds with the backbone amide groups of Val‐84 and Thr‐85 and the side chain hydroxyl group of Thr‐85. The homologous K2P2.1 residues are Glu‐84 (red), Arg‐166, and Thr‐167 (blue), respectively. KcsA Ala‐54 and Leu‐59 were replaced in this presentation by histidines, as present at the homologous positions in K2P2.1 [i.e., His‐87 and His‐141 (green), respectively] based on Cohen et al. (82). (F) Homology model of the TASK‐3 K2P channel. Illustrating the proximity of the two E30 (yellow) and two T103 (blue) residues (view looking from the top down). The model was created using Modeller 9v7 (465) based on the KcsA structure as template [originally solved by Doyle et al. (115)].

Organization of the voltage‐dependent K⁺ channel superfamily. Phylogenetic tree for the Kv1‐12 families. Amino acid sequence alignments of the human channel Kv proteins were created using CLUSTALW. Only the hydrophobic cores (S1‐S6) were used for analysis. The IUPHAR and HGNC names are shown together with the genes’ chromosomal localization and other commonly used name. The alignment was made using the web tool: Phylogeny.fr (109), with different sequences of human two pore K⁺ channels obtained from gene bank accession numbers: KCNH1: NM_002238.3, KCNH2: NP_000229.1., KCNH3:NP_036416.1., KCNH5: NP_647479.2., KCNH6: NP_110406.1., KCNH7: NP_150375.2., KCNH8: NP_653234.2., KCNQ1: NP_000209.2., KCNQ2: NP_004509.2., KCNQ3: NP_004510.1., KCNQ4: NP_004691.2., KCNQ5: NP_062816.2., KCNS1: NP_002242.2., KCNS2: NP_065748.1., KCNS3: NP_002243.3., KCNV1: NP_055194.1., KNCG1: NP_002228.2., KCNG4: NP_758857.1., KCNF1: NP_002227.2., KCNV2: NP_598004.1., KCNG3: NP_579875.1., KCND1: NP_004970.3., KCND2: NP_036413.1., KCND3: NP_004971.2., KCNB1: NP_004966.1., KCNB2: NP_004761.2., KCNC1: NP_004967.1., KCNC2: NP_631874.1., KCNC3: NP_004968.2., KCNC4: NP_004969.2., KCNA1: NP_000208.2., KCNA2: NP_004965.1., KCNA3: NP_002223.3., KCNA4: NP_002224.1., KCNA5: NP_002225.2., KCNA6: NP_002226.1., KCNA7: NP_114092.2., KCNA10: NP_005540.1.

Organization and structure of the Kv1.2/Kv2.1 chimeric channel (PDB_ID: 2RAR). Lateral (left) and top (right) views of the protein embedded in the membrane. Arginine residues important for voltage dependence are shown in sticks. For clarity, two monomers are shown in light gray. The secondary structure of the amino acid sequence (below) is color coded to match the respective transmembrane and functional segments of the protein. Potassium ions are represented in green and the oxygen of water molecules in red. The cytosolic structure hanging from the main protein body is the tratramerization domain, T.

Kvβ1 inactivate currents of a Kv1 channel. (A) Delayed rectifier currents elicited by voltage steps in the absence of Kvβ‐subunit. (B) Coexpression with Kvβ (α+β). (C) A single‐voltage pulse shown in a large time scale. More details in reference 446.

K⁺ currents diversity in Kv channels family. The indicated rat Kv channels were transiently expressed in HEK 293 cells. For each channel, whole‐cell K⁺ currents at +40 mV were measured in similar physiological conditions. Modified, with permission, from reference 62.

K currents from Kv7 and EAG families. Modulation of heteromeric KCNQ2/3 current by extracellular H⁺ ions. (A) Whole‐cell KCNQ2/3 currents from a HEK‐293 cell in bathing solutions of differing pH were elicited by depolarizing voltage steps (1.5 s duration) from a holding potential of −70 mV. (B) Whole‐cell KCNQ2/3 current activation curves in bathing solutions of different pHs (437). (C) Isochronal activation of human ether‐a‐go‐go‐related gene (HERG) channels. Membrane potential was stepped from −80 mV to a test potential between −70 and 100 mV, in intervals of 10 mV, for 2 s, followed by step to −50 mV. The HERG characteristic rapid rise in the tails of current account for a very fast recovery from inactivation and a slower inactivation (378).

Gating currents elicited by the squid potassium channels. (A) Superimposed 10 ms traces of gating and ionic currents recorded at three different voltages taken a 20°C degrees. Na⁺‐gating currents are missed because at this temperature they are too fast for the recording system (modified, with permission, from reference 44). (B) Voltage dependency of the gating charge (open symbols) and the ionic conductance (filled symbols). (C) Kinetics of the gating and ionic currents (B and C modified, with permission, from reference 572).

Structural determinants for the voltage sensitivity in voltage‐gated K⁺ channels. (A) Structure of a single monomer depicting the voltage‐sensor domain (VSD) and the pore domain. Arginines R1, R2, R3, and R4 (corresponding to Shaker R362, R365, R368, and R371) are represented in stick form. (B) Possible trajectories for the gating charges (for more details see text).

Structural design of the K⁺ conduction system. (A) Ions in the pore of the KcsA bacterial channel (PDB_ID: 1K4C). All possible K⁺‐binding sites are shown. Hydration water molecules are shown in red with a Van der Waal radius of 0.5 Å. (B) Ion conduction is due to two alternating and energetically equivalent configurations in ion occupancy (for more details see text).

Mechanical movements of the voltage‐sensitive pore opening. (A) Side and enlarged bottom views of the residues that change in accessibility during the opening of the Shaker activation gate (residues 470‐474; in blue), that do no change in accessibility during gating (residues 481‐486, in red), and residues that may form the gate (residues 475‐479; in green). After reference 313. (B) Allosteric surface proposed for the interaction between the S4 and S5 linker (in blue and gray) with the S6 C‐terminal half of two adjacent subunits (in yellow and orange).

Phylogenetic tree of Slo channels family in mammals and membrane topology of the α‐ and β‐subunits of Slo1 channels. (A) The four genes present in Slo channels families: Slo, Slo2.1, Slo 2.2, and Slo3. (B) The α‐subunit of Slo1 contains seven transmembrane segments divided in two domains [voltage‐sensor domain, (VSD) and pore region] that is normally associated to β‐subunits consisting of two transmembrane segments. β2 and β3 have an inactivating particle on their N‐terminus able to interact with the channel internal vestibule and block the passage of K⁺ through the channel. The α‐subunit contains a long C‐terminus domain in which two regulators of K⁺ conductance domains (RCK1 and RCK2) are present. Spread throughout in the BK C‐terminus are located the binding sites for Ca²⁺ and Mg²⁺ (for more details on the divalent cation‐binding sites see Section “Carboxy terminus”). (C) (Top) The α‐subunit has a voltage‐sensing domain formed by the S0 to S4 segments. Four charged residues contribute to the channel voltage membrane sensitivity, D153, R167 in S2, D186 in S3, and R213 in S4. (Bottom) The pore region formed by S5, the pore helix, the pore loop, and the S6 transmembrane. Three amino acid residues have been identified in the BK pore as partially responsible for the channel high conductance, D292, E321, and E324.

Polymodal activation of Slo channels. (A) Slo1 channel single‐channel activity increases its open probability in response to an increase in the membrane voltage. Upper trace was taken at −60 mV. Openings are downward deflections in the current. Lower trace was taken at 80 mV. Upward deflections are opening events. The opening of two independent channels can be appreciated in this current record. Open probability also increases with increasing intracellular calcium (1, 10, and 100 μmol/L) at a fixed voltage (+60 mV). (B) Slo2.2 channels single‐channel activity increases with high intracellular sodium. Perfusion using 80 mmol/L intracellular sodium elicited four conductance levels, which are reduced to just one with nominal 0 sodium concentration (top). (Bottom) The same type of experiment performed at a compressed time scale. (C) Single‐channel activity of cloned Slo3 increases with the alkalinization and depolarizing voltages. Upper, single‐channel activity increases at positive potentials. Bottom, single‐channel recordings at +80 mV at several intracellular pHs.

Functional differences between β‐subunits. (A) Macroscopic currents were elicited by voltage pulses between −200 and +200 mV at 5 nmol/L (left) and 2.8 μmol/L (right) intracellular calcium. All currents were recorded in the inside‐out configuration. Notice the change in the activation and the deactivation kinetics when β1 and β2IR (β2 inactivation removed) are coexpressed with the α‐subunit. Current records in the third line were obtained by coexpressing the α with the β2‐subunit. Notice that currents inactivate. (B) Voltage activation curves obtained from tail currents (the currents measured at the beginning of the repolarizing pulse; −60 mV) of recordings showed in A at 5 nmol/L (open symbols) and 2.8 μmol/L intracellular calcium (filled symbols) (modified, with permission, from reference 406). (C) Macroscopic currents of α + β4 channels (upper), and the activation curves at different calcium concentrations (lower). Notice the slower activation and deactivation kinetic produced by the β4‐subunit. (D) Comparison of the voltage activation curves at different Ca²⁺ concentrations between channels formed by expressing the α‐subunit alone (left) or by expressing α + β4 (adapted, with permission, from reference 564).

Physiological roles of Slo1 channels. (A) Proposed physiological roles of Slo1 channels. α‐ and β1‐subunits are shown as cartoons. (Adapted, with permission, from reference 407.) (B) Thanks to the close proximity of Slo1 (BK_Ca) and voltage‐dependent Ca²⁺ channels (VDCC), the increase of Ca²⁺ concentration induced by the opening of VDCC (up to 10 μmol/L in the neighborhood of Slo1 channels) promotes the opening of Slo1 channels (top). (Bottom) Current‐voltage relationship obtained in an oocyte expressing only Slo1 (open circles) and coexpressing Slo1 and VDCC. The colocalization of these two channels allows an increase in the K⁺ current that decreases when the potential approaches the reversal potential for Ca²⁺ indicating that K⁺ currents were elicited by the increase in internal Ca²⁺ concentration induced by the VDCC opening. (Adapted, with permission, from reference 131.) (C) In vascular smooth muscle cells, β1‐subunits confer the required Ca²⁺ sensitivity for effective coupling between Ca²⁺ sparks and spontaneous outward currents. [Adapted, with permission, from reference 61.] (D) In chromaffin cells, slowed Slo1 deactivation kinetics allows β2‐subunit‐expressing cells to fire a train of action potentials. (Adapted, with permission, from reference 508.)

Allosteric models for Slo1 activation by voltage and Ca²⁺. (A) Allosteric scheme for channel activation by voltage. J is the equilibrium constant governing the equilibrium between resting and active configuration of the voltages sensor. D is the allosteric factor and L is the intrinsic equilibrium for channel opening. Notice that the channel can open when all voltage sensors are in their resting configuration. (Adapted, with permission, from reference 217.) (B) Allosteric kinetic scheme for activation by Ca2⁺. K is the equilibrium constant for calcium sensor activation and C is an allosteric factor. (C) The combination of A and B produces a two‐tiered 50‐state kinetic model. [Adapted, with permission, from reference 218.) (D) The complete allosteric model taking into account that Slo1 channels are tetramers and including some interaction between the voltage sensor and Ca²⁺ binding (allosteric factor E). In this type of mechanism neither voltage, nor Ca²⁺ binding is strictly coupled to channel opening, these three processes are independent equilibria that interact allosterically with each other. (Adapted, with permission, from reference 406.)

Structural organization of the Slo 1 channel and the crystal structure of the gating ring. (A) Transmembrane segments location using the cysteine cross‐linking technique. Kv1.2/Kv2.1 chimera S1 to S6 with superimposed, labeled circles, uniquely colored for each subunit. White numbered circles correspond to TM1 and TM2 of the β1‐subunit. (Adapted, with permission, from reference 565.) (B) Slo1 20 Å structure resolved with electron cryomicroscopy. The large protrusion at the periphery of the voltage sensor has been suggested to correspond to S0 and the external N‐terminus. (C) Superimposed to the Slo1 structure shown in C is the structure of the transmembrane (TM) domains of Kv1.2 and the gating ring of the MthK channel (adapted, with permission, from reference 565). (D) Slo1 channel RCK1 and RCK2 domains of one subunit showing the position of the Ca²⁺‐binding site (calcium bowl) in the RCK2 domain. Calcium (yellow ball) is coordinated by D892/D895/D897/Q889 (modified, with permission, from reference 583). (E) Slo1 gating ring at 6 Å resolution. The ring is viewed down the 4‐fold symmetry axis with RCK1 in blue and RCK2 in red. Calcium ions are shown as yellow spheres. (F) The open gating ring structure from the MthK channel viewed down the 4‐fold axis of symmetry. Notice that a Ca²⁺ binds to the assembly interface in the Slo1 gating ring whereas two Ca²⁺ ions bind to the flexible interface in the MthK gating ring. (Modified, with permission, from reference 607.)

KCa2 channels activation: single‐channel currents. Single‐channel current from arterial chemoreceptor cells. The inside‐out patch containing one observable open channel during 200‐ms depolarizations from −80 mV to the indicated membrane potentials. Solutions: 130 mmol/L K, 0.01 mmol/L Ca²⁺//130 mmol/L K, 10 mmol/L EGTA. (Adapted, with permission, from reference 149.)

Topology of KCa2 channels and family dendogram. (A) Dendogram of the human SK channels genes constructed using t coffee and ClustalW. Genebank accession numbers: NC_000019 (KCNN1), NC_000005 (KCNN2), NC_000001 (KCNN3), and NC_000019 (KCNN4). (B) Proposed topology for KCa2 channels, showing a canonical six‐transmembrane segments organization (S1‐S6) whence S5 and S6 form the ion‐conduction pathway (shown in cyan) and the S4 segment. The intracellular Ca²⁺ regulation is given by the calmodulin‐binding domain (CaMBD) located in the C‐terminus (black segment). (C) Sequence alignment of the human SK channels (hSK1, hSK2, and hSK3). The transmembrane segments, S1 to S6, are boxed in gray. The pore region (P‐Region) is boxed in cyan. The CaMBD is indicated by black bars. Orange boxed amino acids and red residues show different phosphorylations sites conserved along the family. (Adapted, with permission, from reference 424.)

KCa2 channel C‐terminal calmodulin‐binding domain. Calmodulin protein and KCa2 C‐terminal calmodulin‐binding domain complex was crystallized at a 1.6 Å resolution (PDB: 1G4Y). Calmodulin protein is shown in cyan with two of the four calcium bowls occupied by Ca²⁺ (yellow balls). The center of the calmodulin molecule is in contact with the KCa2 C‐terminal domain (pale brown) (486).

Physiological functions of KCa2 channels. Schematic representation of SK channel function in central nervous system (A) afterhyperpolarization (AHP): CA1 pyramidal neuron whole cell current clamp recording. Twenty action potentials were elicited at 50 Hz in control (black) or apamin (red, 100 nmol/L) bath solutions. The control trace shows the development of an interspike AHP and a posttetanus AHP that is blocked by apamin. Plateau potentials: apamin prolonged the duration of the plateau potential but did not affect the amplitude. (B) Substantia nigra. Pacemaker: perforated‐patch current‐clamp recording of a dopamine neuron in control or apamin (300 nmol/L) bath solutions. On the left is a 4 s trace representative of a 5‐min recording. On the right, the interspike interval (ISI) frequency distribution is plotted for each recording. Apamin significantly decreased the pacemaker precision as shown by the increase in the coefficient of variation (CV). (C) Cerebellum. Trimodal firing: extracellular field recordings of individual cerebellar Purkinje neurons the tonic activity of the cells changed to random bursting when 100 nmol/L apamin was bath applied. (D) Auditory hair cells. Continuous firing: whole cell patch current‐clamp recording from inner ear hair cells in the acutely dissected organ of Corti of a P5 rat. Voltage responses induced by a continuous 30 pA depolarizing current from the resting potential of –59 mV are shown. Bath application of 300 nmol/L apamin gradually abolished the evoked action potentials, indicating that KCa2 channel activity is necessary for continued firing. (Modified, with permission, from reference 54.)