|
Figure 1. Multiple alignment of the primary structure of potassium channel pores from 21 different species computed using Clustalw (http://www.ebi.ac.uk/clustalw/, European Bioinformatics Institute). The signature sequence TVGYGD on the selectivity was used as reference for the alignments. The segments S5, External loop, and P-helix were assigned using prediction of transmembrane helices and topology of protein software (HMMTOP, Hungarian Academy of Sciences, http://www.enzim.hu/hmmtop/). Residues from the beginning of segment S5 to the end of S6 were included in the alignment. RCK2 (Kv1.6), Rattus norvegicus (GI:116435); Kv1.5, Mus musculus (GI:2493594); SkCa-2, Homo sapiens (GI:37955868); KCNA10, Homo sapiens (GI:5031819); Kv1.2, Oryctolagus cuniculus (GI:9652317); Kv2.2, Homo sapiens (GI:27436974); KvEBN1, Homo sapiens (GI:2801452); Shaker, Drosophila melanogaster (GI:288442); SqKv1A, Schistosoma mansoni (GI:510098); Kv1.7, Homo sapiens (GI:14485555); Slack, Rattus norvegicus (GI:11177892); KcsA, Streptomyces lividans (1K4C); KvAP, Aeropyrum pernix (1ORQ); MthK, Methanothermobacter thermautotrophic (1LNQ); cSlo2, Caenorhabditis elegans (GI:71986737); RbSlo, Oryctolagus cuniculus (GI:46396500); mSlo3, Mus musculus (GI:86990444); mslo, Mus musculus (GI:347144); dSlo, Drosophila melanogaster (GI:62472831); rSlo, Rattus norvegicus (GI:58339363); mSlo1, Mus musculus (GI:111607492); hslo, Homo sapiens (GI:46396283).
|
|
Figure 2. Molecular representation of the atomic model of the open BK pore embedded in an explicit POPC phospholipid membrane bathed in a 110 mM KCl aqueous salt solution after 1 ns of MD. The extracellular aqueous phase is on top of the bilayer and the intracellular phase is below the bilayer. The structure embedded into the bilayer was obtained by homology using the 3D structures of KcsA and MthK as templates. The extracellular loop that is into the extracellular aqueous phase was modeled using Loop Search. Potassium ions in the selectivity filter (red spheres) are located in sites S1 and S3. Two opposing monomers of BK are shown using ribbon representation for the secondary structure of the protein. Phospholipid hydrocarbon chains are shown as green sticks. Phosphate atoms are shown as yellow spheres and nitrogen atoms of the phospholipds are shown as blue spheres. Water molecules are drawn as light blue spheres, chloride ions are the green spheres, and the red spheres represent the potassium ions.
|
|
Figure 3. Calculation of the electrostatic potential, φ(r) by solving the PB equation. In the PB equation ε(r) and κ2(r) are the position-dependent dielectric constant and ionic screening constant, respectively, and ρ(r) is the total atomic charge distribution in the system. κ(r) is 0 in zones II, III, and IV, and it is 1/9.8 A−1 for zone I. Internally located residues E386 and E389 and externally positioned residues E322, D326, and E329 are represented as CPK space-filling mode. In solving the PB equation the bilayer was considered a continuum with a dielectric constant of two, the protein as a continuum with a dielectric constant of two and the solutions a continuum with a dielectric constant of 80.
|
|
Figure 4. Negatively charged amino acids located in the external loop have a marginal effect on single channel conductance. (A) Samples of single channel current fluctuations induced by the WT BK channel and the neutralization mutants and recorded at −100 mV in symmetric 110 mM K+. Current–voltage relationship for WT (B) and neutralization BK channel mutants (C–E). Single channel current was measured from the peaks of all-points histograms of the current intensity records of membranes containing 1, 2, or 3 channels. (C) Current–voltage relationship for the E329Q mutant. (D) Current–voltage relationship for the D326N mutant. (E) Current–voltage relationship for the D326N-E329Q double mutant. Experiments were performed at room temperature, 20–22°C.
|
|
Figure 5. Comparison of single channel currents fluctuations of the D326-E329Q double mutant with WT BK channel recorded at different K+ concentrations. Samples of single channel current fluctuations at −100 mV induced by the WT BK channel and the neutralization mutant D326N-E329Q at 30, 110, and 1,000 mM symmetrical K+.
|
|
Figure 6. Current–voltage relationship for the D326N-E329Q mutant and WT BK channel at different K+ concentrations. (A) I-V relation obtained in symmetrical 30 mM K+ for the WT (solid black circles) and the D326N-E329Q mutant (solid red circles). Points are the mean of three to five different experiments ± SD. (B) I-V relation obtained in symmetrical 110 mM K+ for the WT (solid black circles) and the D326N-E329Q mutant (solid red circles). Points are the mean of two to five different experiments ± SD. (C) I-V relation obtained in symmetrical 1,000 mM K+ for the WT (solid black circles) and the D326N-E329Q mutant (solid red circles). Points are the mean of three to six different experiments ± SD. (D) Bar graph of single channel conductance at different K+ for the mutant and for WT. *, P < 0.05; **, P < 0.005.
|
|
Figure 7. (A and B) TEA+ blockade. The figures are plots of the macroscopic current measured at +70 mV on outside out patches at various outside TEA+ concentrations normalized by the current measured in the absence of TEA+. (A) Control wt BK channels. (B) D326N/E329Q BK mutant channels. Data points represent the average and standard deviation of the normalized current measured in three membranes. TEA+ dissociation constants are 1.9 ± 0.3 × 10−4 M for the control channels and 4.9 ± 0.3 × 10−4 M for the double mutant channels. From the effect of the mutation on the TEA+ dissociation a constant surface potential difference of 13 ± 2 mV was calculated using Eq. 1. (C and D) CTX blockade. The figures are plots of the relative macroscopic current measured on outside out patches at various outside CTX concentrations normalized by the current measured in the absence of the blocker. (C) Control channels. (D) D326N/E329Q BK mutant channels. Data points represent the average and standard deviation of the normalized current measured in three membranes. CTX dissociation constants are 1.7 ± 0.3 × 10−9 M for the control channels and 5.3 ± 0.3 × 10−9 M for the double mutant channels. From the effect of the mutation on the CTX dissociation constant, a surface potential difference can be calculated using Eq. 1. Δφ' is 28.8 mV for z = 1 and 4.8 mV for z = 6. Assuming that the actual surface potential difference is that measured using TEA+, the apparent z of CTX is 2.1.
|
|
Figure 8. Ba2+ blockade. On rate constant of Ba2+ blockade as a function of membrane potential. The product of barium block on rate constant times the local barium concentration, kon[Ba], was calculated from the relaxation experiments. Since barium block is voltage dependent the kinetics of barium entry to the blocking site can be examined using voltage jump experiments and observing the decay of the current as channels are blocked. Constants kon[Ba] and koff were calculated from the extent of the blockade and the time constant according to Eqs. 3 and 4. In this figure, the logarithm of the on rate constant standardized to 1 μM bulk barium concentration is plotted as a function of membrane potential for the WT channels (black symbols) and the E386N/E389N double mutant BK channels. The vertical distance of the two straight lines is Δφ times 2F/RT according to Eq. 7, where Δφ is 17 ± 2 mV.
|
|
Figure 9. Electrostatic potential difference along the axis of the pore. Ordinate represents the change in local electrostatic potential, in mV, observed when the charge of the different residues is turned off. Error bars are the standard deviation of the potential calculated over 100 different structures sampled during a 1-ns MD simulation. Abscissa is the z coordinate in Angstrom units. z = 5 is defined as the geometric center of the selectivity filter. The entrance of the extracellular mouth of the pore is at z = 17 Å and the intracellular mouth is at z = −15.5 Å. The magenta line describes the electrostatic potential differences calculated for the neutralization of the acidic amino acids located at the inner opening of the channel. The light green line is the calculation for the neutralization of residue Glu386. The cyan line is the result of the neutralization of Glu389. The red line is the electrostatic potential associated with turning off the charges of residues Asp326 and Glu329. Green and blue lines represent the individual contributions of Asp326 and Glu329, respectively.
|
|
Figure 10. Removing the ring four negative charges (Glu322) in the external mouth of the BK channel induces outward rectification. (A) Single channel current measured as a function of membrane potential for mutant E322I (black dots and solid line) compared with the WT channels (red line) measured in symmetrical 110 mM K-MES and 1 M K-MES. (B) Column plot of the average single channel conductance measured at −100 mV for the control and double mutant channels at 110 mM K-MES and 1 M K-MES. (C) Electrostatic potential calculated associated with turning off the charges of residue Glu322.
|
|
Figure 11. Electrostatic potential maps (EPMs). The figures show in pseudo colors the electrostatic potential obtained solving the Poisson-Boltzmann equation. The map corresponds to the electrostatic potential on the plane x = y = 0. The extracellular space is represented on the top of each figure and the intracellular space is below the bilayer. In blue are shown the positive potential (>20 mV) and in red are shown the negative potential (<−20 mV). The map was merged with a 3D representation of the α carbon trace of the BK protein. Amino acids Asp326, Glu329, Glu386, Glu389 are shown in CPK space-filling mode. Calculations were performed using the PBEQ module of CHARMM28, and figures were drawn using the program DINO: Visualizing Structural Biology (2002), http://www.dino3d.org. (A) The EPM for the WT hSlo structure. The plane cuts through the bilayer, the protein, and the water-filled vestibules. The electrostatic potential inside the protein and the lipid bilayer is negative (red) and neutral on the bulk of the aqueous phase (white). The water-filled vestibules of the channel have negative electrostatic potential (red to pink color). (B) The EPM for the neutralization mutant E386N/E389N, in this mutant the intracellular vestibule is no longer at negative potential (white). (C) The EPM for the mutant D326Q, E329N. Neutralization of these external negative charges produces only marginal changes in the electrostatic potential map of the BK pore.
|
|
Figure 12. Calculated electrostatic potentials are not a simple functions of distance. Electrostatic potentials are plotted in the vertical axis and the distances to the carboxyl group in the horizontal axis. Electrostatic potentials calculated at the internal end of the selectivity filter are represented with black symbols, the coordinates of this site are x = 0, y = 0, z = −2 Å. Electrostatic potentials calculated at the internal end of the selectivity filter (x = 0, y = 0, z = 17) are represented with red symbols. Error bars on the horizontal axis are the distance ranges read from the radial distribution function computed of 100 snapshots of an MD simulation where the protein backbone atoms were under a 5 kcal per angstrom square harmonic restraint. The three lines drawn represent the electrostatic potential as a function of distance calculated using 35, 75, or 200 as dielectric constants.
|
