
Figure 1. dSlo1 activation exhibits higher Ca2+ sensitivity than mSlo1. (A) Current traces of mSlo1 and dSlo1. Test potentials were −80 to +200 mV and −150 to +200 mV for 5.7 μM and 89 μM [Ca2+]i, respectively, holding and repolarization potentials were −50 mV. (B) Steadystate GV relations of mSlo1 and dSlo1. Smooth curves are fits of the Boltzmann equation (see MATERIALS AND METHODS) with parameters for mSlo1, in 0 [Ca2+]i: V1/2 = 179 mV, z = 1.2, in 5.7 μM [Ca2+]i, V1/2 = 64.6 mV, z = 1.4, in 89 μM [Ca2+]i, V1/2 = −2.8 mV, z = 1.2; for dSlo1, in 5.7 μM [Ca2+]i, V1/2 = 157 mV, z = 1.3, in 89 μM [Ca2+]i, V1/2 = −2.9 mV, z = 1.1. (C) Free energy provided by Ca2+ binding for channel activation when [Ca2+]i changes from 0 or 5.7 μM to 89 μM, as indicated in parentheses under the abscissa. m, mSlo1; d, dSlo1.


Figure 2. Chimeric channels of mSlo1 and dSlo1 retain features of their native WT channels. (A, top) Current traces of Chim2 and mSlo1 at [Ca2+]i of 2, 10, and 100 μM. Test potentials were −80 to +200 mV, holding and repolarization potentials were −50 mV. (Bottom) Steadystate GV relations of Chim2 and mSlo1. Smooth curves are fits of the Boltzmann equation with parameters for Chim2, in 2 μM [Ca2+]i: V1/2 = 240.1 mV, z = 0.95, in 10 μM [Ca2+]i: V1/2 = 76.1 mV, z = 1.5, in 100 μM [Ca2+]i: V1/2 = 43.5 mV, z = 1.5; for mSlo1, in 2 μM [Ca2+]i: V1/2 = 84.8 mV, z = 1.5, in 10 μM [Ca2+]i: V1/2 = 31.4 mV, z = 1.5, in 100 μM [Ca2+]i: V1/2 = 14.4 mV, z = 1.4. (B, top) Current traces of Chim6 and mSlo1 at [Ca2+]i of 0, 10, and 500 μM. Test potentials were −80 to +200 mV, holding and repolarization potentials were −50 mV. (Bottom) Steadystate GV relations of Chim6 and mSlo1. Smooth curves are fits of the Boltzmann equation with parameters for Chim6, in 0 [Ca2+]i: V1/2 = 206.7 mV, z = 1.3, in 10 μM [Ca2+]i: V1/2 = 101.1 mV, z = 1.2, in 500 μM [Ca2+]i: V1/2 = 52.5 mV, z = 1.2; for mSlo1, in 0 [Ca2+]i: V1/2 = 189.7 mV, z = 1.1, in 10 μM [Ca2+]i: V1/2 = 33.8 mV, z = 1.7, in 500 μM [Ca2+]i: V1/2 = −13.2 mV, z = 1.8.


Figure 3. Activation of dSlo1 saturates around 89 μM [Ca2+]i. (A and B) Current traces of dSlo1 for indicated [Ca2+]i elicited by voltages from −150 to 200 mV. (C) Steadystate GV curves of dSlo1. Solid curves are fits of the Boltzmann equation. V1/2 obtained from the fits of the GV relations at each [Ca2+]i is indicated. Dotted lines indicate MWC model simulations of the GV relation at [Ca2+]i 89, 100, 200, and 300 μM using the same parameters as indicated in Fig. 10 C.


Figure 4. The AC region in the RCK1 domain is important for Ca2+ sensitivity. (A) Ca2+ sensitivity of activation in chimeric channels of mSlo1 and dSlo1. The vertical axis shows schematic representation of chimera constructs with dSlo1 portions shaded gray and mSlo1 black. Rectangles are transmembrane segments or RCK1 domains (Jiang et al., 2002a), ovals are the Ca2+ bowl (Schreiber et al., 1999). Free energy increase in response to increase in [Ca2+]i (shown at the right) for each chimera and WT channel was normalized against that for mSlo1. (B) Plot of V1/2 versus [Ca2+]i for mSlo1 (thin black line), dSlo1 (thick black line), and the chimeric channels as defined in A. V1/2 values are obtained by fitting the GV relations of the channels at various [Ca2+]i with the Boltzmann equation.


Figure 5. The AC region switches Ca2+ sensitivity between mSlo1 and dSlo1. Plot of (A) V1/2 and (B) z versus [Ca2+]i for mSlo1, dSlo1, m[dAC], and d[mAC]. V1/2 and z are obtained by fitting the GV relations of the channels at various [Ca2+]i with the Boltzmann equation.


Figure 6. Ca2+ sensitivity change is not related to the Ca2+ binding site in the AC region. (A) Sequence alignment of the AC region of the RCK1 domain (Jiang et al., 2002a) from mSlo1 (Butler et al., 1993), dSlo1 (Adelman et al., 1992), and the archeon MthK (Jiang et al., 2002a). Numbers indicate the position of the rightmost residues in the primary sequence of respective proteins. Secondary structures βAC and αAC are indicated by underlines. Boxed amino acids labeled as motifs 1, 2, and 3 are regions showing significant sequence differences between mSlo1 and dSlo1. Motif 1 is important for Ca2+dependent activation (Shi et al., 2002; Xia et al., 2002). Effects of switching motif 1 between mSlo1 and dSlo1 are shown in B and C. (B, left) Steadystate GV relations of the mutant channel m[d1] (motif 1 from dSlo1 in the mSlo1 background). Solid lines are fits of the Boltzmann equation with the following parameters: at 0 [Ca2+]i, V1/2 = 185 mV, z = 1.2; at 89 μM [Ca2+]i, V1/2 = 15.9 mV, and z = 1.2. Dotted lines are GV relations of mSlo1. (B, right) Steadystate GV relations of the mutant channel d[m1] (motif 1 from mSlo1 in the dSlo1 background). Solid lines are fits of the Boltzmann equation with the following parameters: at 89 μM [Ca2+]i, V1/2 = −52.5 mV, z = 0.83; at 5.7 μM [Ca2+]i, V1/2 = 95.1 mV and z = 0.89; at 0 [Ca2+]i, the GV relation was too right shifted for z and V1/2 values to be determined. Dotted lines are GV relations of dSlo1. (C) Free energy of activation provided by Ca2+ binding in mSlo1, d[m1], m[d1], and dSlo1 when [Ca2+]i increased from 0 to 89 μM (for m[d1]) and from 5.7 to 89 μM (for d[m1] and dSlo1), normalized against corresponding free energy values for mSlo1.


Figure 7. Switching sequence differences in AC region between mSlo1 and dSlo1 either singly or in combination does not switch Ca2+ sensitivity of BKCa gating. (A) Steadystate GV relations of m[dMotif2] and m[dMotif3]. Dotted lines are GV relations of mSlo1. Parameters of the Boltzmann fits (solid lines): for m[dMotif2], at 0 [Ca2+]i, V1/2 = 195.7 mV, z = 1.1; at 100 μM [Ca2+]i, V1/2 = 8.2 mV, z = 1.2; for m[dMotif3], at 0 [Ca2+]i, V1/2 = 213.2 mV, z = 1.1; at 100 μM [Ca2+]i, V1/2 = 9.2 mV, z = 1.2. (B) Steadystate GV relations of m[dMotif12] and m[dMotif13]. Dotted lines are GV relations of mSlo1. Parameters of the Boltzmann fits (solid lines): for m[dMotif12], at 0 [Ca2+]i, V1/2 = 186.5 mV, z = 1.2; at 100 μM [Ca2+]i, V1/2 = 22.2 mV, z = 1.1; for m[dMotif13], at 0 [Ca2+]i, V1/2 = 211.9 mV, z = 1.4; at 100 μM [Ca2+]i, V1/2 = 38.4 mV, z = 1.3. (C) Free energy of activation provided by Ca2+ binding for the chimeric channels. Free energy change for each channel was in response to [Ca2+]i change shown in parentheses for mSlo1 plotted alongside each group.


Figure 8. High Ca2+ sensitivity in dSlo1 is not related to the low affinity metal binding site. (A) Steadystate GV relations of mSlo1 and dSlo1 at 89 μM [Ca2+]i and indicated [Mg2+]i. Boltzmann fits (smooth curves) gave following parameters: mSlo1, for 0 [Mg2+]i, V1/2 = −2.8 mV, z = 1.2, for 10 mM [Mg2+]i, V1/2 = −73.8 mV, z = 1.1; dSlo1, for 0 [Mg2+]i, V1/2 = −2.9 mV, z = 1.1, for 10 mM [Mg2+]i, V1/2 = −69.2 mV, z = 0.94. (B) Steadystate GV relations of the E413R mutant dSlo1 channel. Boltzmann fits (solid curves) gave the following: E413R, for 5.7 μM [Ca2+]i, V1/2 = 163 mV, z = 1.1, for 89 μM [Ca2+]i, 0 [Mg2+]i, V1/2 = −24.6 mV, z = 0.95, for 89 μM [Ca2+]i, 10 mM [Mg2+]i, V1/2 = −46.7 mV, z = 0.87. Dotted lines are GV relations for dSlo1 at 5.7 μM [Ca2+]i and 89 μM [Ca2+]i, 0 [Mg2+]i. (C) Free energy provided by Mg2+ binding for channel activation in mSlo1, dSlo1, and E413R dSlo1 when [Mg2+]i changes from 0 to 10 mM, measured at [Ca2+]i of 89 μM. (D) Free energy provided by Ca2+ binding for channel activation in WT and E413R dSlo1 when [Ca2+]i changes from 5.7 to 89 μM.


Figure 9. Molecular dynamics simulations of the AC region of mSlo1 and dSlo1. Top panels show the extrema (green and orange) of the motion along the principal eigenvector for the AC region from mSlo1 (left) and dSlo1 (right) (see MATERIALS AND METHODS). Part of helix B highlighted in purple shows the region of largest dynamic difference between the structures of mSlo1 and dSlo1. Bottom panel shows the RMS fluctuations for the Cα's of the above structure. The purple box denotes the same region (αB) highlighted above. Amino acid in the sequence at the bottom corresponds to the Cα whose dynamics are plotted above. Molecular graphics were produced using visual molecular dynamics (VMD) (Humphrey et al., 1996).


Figure 10. Modulation of BKCa activation by the AC region depends on Ca2+ occupancy. (A) Current traces of mSlo1, dSlo1, d[mAC], and m[dAC] at 0 [Ca2+]i. Test voltages were from −80 to 200 mV with a holding and repolarization potential of −50 mV. (B) Steadystate GV relations of above channels at 0 (open symbols) and 89 μM (filled symbols) [Ca2+]i. Smooth curves are fits to the Boltzmann equation. At 89 μM [Ca2+]i, for mSlo1: V1/2 = −2.8 mV, z = 1.2; for d[mAC]: V1/2 = 7.7 mV, z = 1.4; for dSlo1: V1/2 = −2.9 mV, z = 1.1; for m[dAC]: V1/2 = 41.6 mV, z = 1.2. At 0 [Ca2+]i, for mSlo1: V1/2 = 179 mV, z = 1.2; for d[mAC]: V1/2 = 181 mV, z = 1.3. The voltage range of dSlo1 and m[dAC] activation at 0 [Ca2+]i is too positive to record any current. (C) Box plot of ΔGV = zV1/2 for the above channels at 0 and 89 μM [Ca2+]i. The percentile values shown are 10, 25, 50, 75, and 90 for each channel. ΔGV for m[dAC] and dSlo1 at 0 [Ca2+]i are too large to be determined.


Figure 11. The AC region affects Ca2+ binding to closed channels. (A and B) GV relations of mSlo1, dSlo1, d[mAC], and m[dAC] channels at [Ca2+]i of 0, 1.7, 2.3, 5.7, 11.2, 28.5, and 89 μM. Each dataset was fit (smooth curves) by the MWC model (Eq. 4), n = 4 (A) or 8 (B). (C) Parameters of fits obtained from A and B.


Figure 12. Spring effect of the AC region in channel gating. In the absence of bound Ca2+ and in the closed conformation, the AC region adopts a conformation that inhibits channel opening. Ca2+ binding (+Ca2+) or channel opening by depolarization (+V) removes this inhibition, rendering the channel gate more favorable to the open conformation. In the open conformation or when the Ca2+ binding sites are occupied, the AC region has little effect on channel gating.
