|
Figure 1. Cartoons summarizing the idea of antechamber occupancy and lateral access of β2 N termini to the BK channel pore. (A) The pathway for access of the β2 N-terminal inactivation domain to the BK channel central cavity is schematized. N termini must enter the central cavity by passing through the side portals separating the BK channel pore domain from the cytosolic domains involved in Ca2+ binding. The lateral distance from the center of the pore to the position where the N terminal attaches to the β2 subunit TM1 domain is estimated to be ∼45–60 Å (Zhang et al., 2006). Each ball in the schematized N terminus represents an amino acid, with red indicating basic residues and blue indicating the FIW hydrophobic triplet essential for inactivation. (B) Cartoons schematically summarize proposed configurations of β2 N termini during gating and inactivation. Each channel contains four β2 subunits (containing a triplet of hydrophobic residues [blue] at the N terminus and two digestible basic residues, R8 and R19 [red]), each of which can potentially enter the channel antechamber (equilibrium, Ba) through side portals. The central pore is indicated by the shaded, inner circle (smaller, closed channel; larger, open channel). In this scheme, only one N terminus can occupy the antechamber at a time. Channels open in accordance with equilibrium constant L. Open channels with a β2 N terminus in the antechamber may also inactivate (equilibrium constant, Bi).
|
|
Figure 2. Stochastic model for protection of β2 N termini from digestion by trypsin. (A) A kinetic scheme describing the proposed digestion of β2 N termini by trypsin is shown, including channel gating (L(V, Ca)), protection of N termini within the antechamber (P), and inactivation. States highlighted in the box are those in which inactivation has been completely removed by trypsin. Flattened red ovoids correspond to N termini in positions outside the antechamber, while, when black, are within the antechamber. Blue circles on the axes of the pore correspond to an N terminus in an inactivation position. (B–G) Simulated digestion time courses based on this scheme. (B) The simulated digestion time courses are shown for parameters given in Table II for closed-channel conditions (Po = 0.01) or inactivating conditions (Po = 0.99; Bi = 0.93). The fit of Eq. 1 to the time course under closed-channel conditions yielded τd = 28.5 s (n = 2.09). Under inactivating conditions, τd = 327.3 s (n = 1.17). (C) The affinity of the N terminus for the site involved in closed-channel protection (affinity for antechamber, Ba) was reduced by increasing the dissociation rate from the site 50-fold. Under these conditions (red circles), τd = 7.2 s (n = 3.98). (D) The effective digestion rate was varied (approximating changes in [trypsin] [black symbols]) for closed-channel conditions as defined in A and for the case that antechamber binding affinity is reduced (red symbols). With reduced antechamber binding affinity, the slope of the digestion time course is steeper, but the slope is unaffected by the effective rate of the digestion process. (E) The effective digestion rates are plotted as a function of effective trypsin concentration for the cases of modest closed-channel protection (open circles) or without such closed-channel protection (red circles). The slope of the lines shows the approximately fourfold difference in effective digestion rate that arises because of the difference of occupancy in the antechamber, despite the fact that the same underlying molecular rate for the cleavage step was used in the two cases. (F) The effect of weakening the affinity of the inactivation domain for the channel pore (ΔBi) is illustrated. The rate of dissociation of the inactivation domain from its blocking position was increased 10-fold, changing steady-state inactivation from 0.93 to 0.56. In this case, under inactivating conditions, the digestion time course (red circles) approaches that observed for closed-channel conditions (solid black circles) with τd = 55.8 s (n = 1.63), whereas under there is no difference between the two cases for closed-channel conditions (not depicted). (G) The effect of a fivefold weakening of both binding in the antechamber (Ba) and at the inactivation site (Bi) is shown. This alters digestion both under closed-channel conditions (solid symbols; τd = 11.1 s; n = 3.43) and under inactivated conditions (open symbols; τd = 21.7 s; n = 2.45). Values for simulations and resulting measurements of τd and n are given in Table II.
|
|
Figure 3. The W4G mutation in the β2 FIW motif speeds up trypsin-mediated removal of inactivation under closed-channel conditions. (A) Currents resulting from α + β2(W4G) channels were activated by depolarizing steps to +80 mV with 10 µM Ca2+. Between voltage steps, the patch was held at 0 mV with 0 Ca2+ and 0.01 mg/ml trypsin was applied for defined periods of time. Times given on the panel are the cumulative time of trypsin application. (B) The time course of digestion of β2-W4G under closed-channel conditions with 0.01 mg/ml trypsin (black circles and line) is compared with the digestion of β2 wild type with 0.02, 0.1, and 0.5 mg/ml trypsin (red lines). (C) Similar test traces are shown for a patch in which trypsin was applied in the presence of 10 µM Ca2+ at +80 mV, a condition in which β2-(W4G) channels are >95% inactivated. (D) The time course of removal of inactivation is plotted for sets of patches for the conditions shown in A and C. For comparison, the solid red line and dotted red line correspond, respectively, to the digestion of wild-type β2 N termini under closed-channel conditions (0 µM Ca2+; 0 mV) with 0.1 mg/ml trypsin and under inactivating conditions (10 µM Ca2+; 0 mV) with 0.1 mg/ml trypsin. (E) The time course of digestion of β2 and β2-W4G N termini is plotted on a log-log scale to compare the slope of the digestion process. For these fits, the power term for β2 digestion was 2.23 ± 0.39, and for β2-W4G it was 3.45 ± 0.13.
|
|
Figure 4. Some manipulations of the β2 N terminus can alter inactivation-dependent protection against digestion, with minimal effects on rates of digestion during closed-channel conditions. (A) Traces show digestion at different times by 0.l mg/ml trypsin applied in the presence of 0 Ca2+ at 0 mV (closed-channel conditions) for the MGGGFIW construct. Displayed currents were obtained between trypsin applications with 10 µM Ca2+ as in Fig. 3. (B) Traces are shown for the MGGGFIW construct during digestion by 0.1 mg/ml trypsin in 10 µM Ca2+ at +80 mV (inactivated conditions). For this construct, at +80 mV, the steady-state non-inactivating current is ∼30–40% of the maximal current. (C) The digestion time courses for MGGGFIW are plotted for the two conditions (black symbols, closed channel; open symbols, inactivated), along with the fitted results for the wild-type β2 subunit (dashed red line, inactivated conditions [10 µM Ca2+; 0 mV]; red line, closed-channel conditions [0 Ca2+; 0 mV]). (D and E) Traces are shown for the MFIWGGG construct for digestion by 0.1 mg/ml trypsin either at 0 Ca2+, 0 mV (D) or at 10 µM Ca2+, +80 mV (E). For MFIWGGG, the steady-state non-inactivating current at +80 mV is ∼5% of the maximal current. (F) Digestion time courses are plotted for the MFIWGGG construct along with the time course for β2. (G and H) Digestion time courses are plotted for MGFIW (G) and MWWWTSG (H), both under closed-channel and inactivating conditions.
|
|
Figure 5. Inactivation-associated protection against digestion by trypsin is differentially sensitive to cytosolic channel blockers. (A) Traces show test sweeps at different time points during digestion of β2 by 0.1 mg/ml trypsin under conditions of low activation (0 Ca2+; 0 mV) but in the presence of 20 mM TBA. (B) Traces show test sweeps during digestion of β2 by 0.1 mg/ml trypsin with 20 mM TBA under inactivating conditions (10 µM Ca2+; 0 mV). (C) The time course of removal of inactivation is plotted for the indicated conditions. Neither 20 nor 50 mM TBA abolishes the inactivation-associated slowing of digestion. The red lines show the normal removal of inactivation of β2 subunits with 0.1 mg/ml trypsin in the absence of TBA for closed-channel (solid line) and inactivating (dotted line) conditions. (D) Traces show removal of β2-mediated inactivation by 0.1 mg/ml trypsin applied under closed-channel conditions, but with 1 mM bbTBA. (E) The digestion by 0.1 mg/ml trypsin occurred at 0 mV, 10 µM Ca2+, with 1 mM bbTBA. (F) The time course of removal of β2-mediated inactivation is plotted for closed-channel and inactivating conditions, both with 1 mM bbTBA, showing the ability of bbTBA to prevent the inactivation-associated protection from digestion while not changing the digestion rate under closed-channel conditions.
|
|
Figure 6. Digestion of mβ3a N termini is much faster than for β2. (A) Traces show removal of hβ2-mediated inactivation under closed-channel conditions (0 Ca2+; 0 mV) with 0.1 mg/ml trypsin. (B) Traces show removal of β2-mediated inactivation by 0.1 mg/ml trypsin under inactivating conditions (10 µM Ca2+; 0 mV). (C) The digestion time course is plotted for both closed (solid circles) and inactivating (open circles) hβ2 channels. (D) Sweeps monitor removal of mβ3a-mediated inactivation under closed-channel conditions, but with 0.01 mg/ml trypsin. Trypsin abolishes both the slow β3a tail currents and removes inactivation. (E) The time course of removal of mβ3a-mediated inactivation by 0.01 mg/ml trypsin is shown under inactivating conditions. Note that even at 900 s, appreciable inactivation still remains and that the slow β3a tail current persists. (F) The digestion time course for mβ3a channels is plotted for both closed-channel (solid circles) and inactivating (open circles) conditions.
|
|
Figure 7. Comparison of removal of inactivation mediated by hβ2 and mβ3a. (A) The removal of mβ3a-mediated inactivation is plotted for four different trypsin concentrations. Fitted time constants of digestion were 5.83 ± 1.2 s (0.05 mg/ml), 10.5 ± 1.42 s (0.025 mg/ml), 23.72 ± 1.41 s (0.01 mg/ml), and 44.22 ± 3.74 s (0.005 mg/ml). (B) The effective digestion rate (min−1) is plotted as a function of trypsin concentration for both mβ3a and hβ2 (from Zhang et al., 2006). The line through the mβ3a points corresponds to a linear fit with a slope of 260 min−1/mg/ml, which assumes a molecular weight of 24 kD for trypsin corresponds to 9.97 × 104 M−1 s−1. For β2, the line corresponds to the slope through the two lowest trypsin concentrations, yielding a maximal effective rate of 6.82 × 103 M−1 s−1. (C) A log-log plot of the digestion time course for hβ2 (black circles) and mβ3a (red circles) compares the slope of the digestion process. For β2, n = 2.21 ± 0.23; for β3a, n = 3.62 ± 0.39. Note that in A, except for the time course observed with 0.05 mg/ml, the steeper slope of β3a digestion is independent of trypsin concentration. Similarly, the slope of β2 digestion is independent of trypsin concentration (not depicted).
|
|
Figure 8. bbTBA and TBA differentially influence inactivation-associated protection against trypsin digestion for β3a subunits. (A) Traces show progressive removal by 0.01 mg/ml trypsin of inactivation mediated by β3a subunits with 20 mM TBA applied together with trypsin under inactivating conditions (10 µM Ca2+; 0 mV). (B) The time course of the digestion process under inactivating conditions with TBA is compared with the normal time course of digestion of β3a under closed-channel (red line) or inactivated (dotted red line) conditions. (C) Traces show removal of inactivation when 0.01 mg/ml trypsin was applied along with 1 mM bbTBA under inactivating conditions. (D) Time course plots of removal of inactivation show that 1 mM bbTBA abolishes the protection normally produced by inactivation (dotted red line), but has little effect on the digestion time course (red line) under closed-channel conditions.
|
|
Figure 9. Exchange of hydrophobic and charged segments between β2 and hβ3a N termini. (A) The time courses of digestion under closed-channel conditions (0 Ca2+; 0 mV) of the MFIWTSGR-β3a and MFIWTSGQ-β3a constructs are compared with wild-type mβ3a (red line) for 0.01 mg/ml trypsin. (B) The time course of digestion of β3a1-12β213-end (MQPFSIPVQITL from β3a appended to β2 at position 13) with 0.1 mg/ml trypsin is compared with the β2 digestion time course. (C) A β3a segment (residues 12–21 containing R16-18 and R21) replaced residues 14–22 in a β2 construct in which all N-terminal basic residues were replaced with Q. The plot shows the digestion time course for both 0.1 and 0.01 mg/ml trypsin compared with that of β2 (blue line). (D) The time course of digestion with 0.1 mg/ml trypsin of construct mβ3a13-21β214-22 is compared with that of β2 (blue line). Dotted line represents a two-component exponential fit (analogous to Eq. 1) to the digestion time course, and the solid represents a fit of Eq. 1. (E) Possible configurations of N termini in a closed channel are schematized. Case 1, an N terminus is bound in the antechamber, and basic residues are shielded from digestion; Case 2, the N terminus is bound in the antechamber, but, although basic residues are outside the antechamber, they are structurally constrained, thereby hindering digestion; Case 3, the N terminus is bound in the antechamber, but basic residues are freely mobile and accessible to attack by trypsin; Case 4, an N terminus is freely mobile outside the antechamber allowing easy digestion by trypsin.
|
|
Figure 10. Impact of charge and the β3a12-21 segment inserted in artificial N-terminal linkers on trypsin digestion rates under closed-channel conditions. (A) The time course of trypsin-mediated removal of inactivation (0.1 mg/ml trypsin) is compared for constructs FIW-8Q-2K-20Q-β2 and FIW-20Q-2K-8Q-β2, in which the N terminus length is identical, but the basic residues are positioned at either 8 or 20 residues from the FIW motif. The digestion time course of wild-type β2 is given by the red line. (B) The time courses of digestion for FIW-8Q-3K-20Q-β2 and FIW-20Q-3K-20Q-β2 are shown. (C) The time course of digestion for FIW-8Q-4K-20Q-β2 is compared with FIW-20Q-4K-8Q-β2. (D) The digestion time course with 0.01 mg/ml trypsin for a construct (FIW-20Q-β3a12-21-8Q-β2) in which segment β3a12-21 was inserted in an artificial N terminus is displayed (τd =28.9 ± 4.0 s; n = 2.3 ± 0.5) and compared with that of FIW-20Q-KKK-8Q with 0.1 mg/ml trypsin (τd = 52.5 ± 4.8 s; n = 2.0 ± 0.2 for a different set of patches than in B). (E) The time course of digestion by 0.01 mg/ml trypsin is shown for two constructs in which the β3a12-21 segment (LQGGRRRQGR) contained additional substitutions. In one, R16Q and R17Q resulted in an N terminus in which only R18 and R21 remained (LQGGQQRQGR; τd = 45.4 ± 3.7 s; n = 2.1 ± 0.2). In the other, all glycines in the β3a segment were mutated to Q (τd = 40.3 ± 5.2 s; n = 1.8 ± 0.3).
|
|
Figure 11. Summary of digestion rates and power terms for various constructs. (A) Apparent digestion rates were calculated for β3a-derived constructs. For the calculation of effect rates, the trypsin molecular weight was assumed to be 24 kD. Red line indicates rate measured for wild-type β3a, and the black line indicates the mean value for wild-type β2. Blue symbols indicate those constructs with the most strongly slowed digestion for β3a. (B) The fitted values for n are for β3a-derived constructs shown to cluster around n = 4. Although there seems to be a trend for FIW-tagged constructs to have a lower value of n, ANOVA comparisons of n from sets of patches revealed no statistically significant difference between β3a and any FIW-tagged construct. (C) Apparent digestion rates for β2-related constructs are plotted, with the three with the largest increase in digestion rate highlighted in red. (D) Power terms for β2 constructs cluster around n = 2.5. Values for FIG (W4G; P = 0.00542) and KvβTSGR (P = 7.01E-05) were found to be statistically different from wild-type β2 by ANOVA analysis, whereas comparison of n values for other constructs showed no difference with β2. Power terms for β2 and β3a were also significantly different (P = 0.00542).
|
|
Figure 12. A model in which two N termini can dwell within the antechamber at a time may better account for the data. (A) The two left-hand columns of states correspond to Scheme 1 under conditions of low Po and no inactivation and, therefore, approximate closed-channel conditions for a model in which one N terminus can occupy the antechamber at a time. The additional states highlighted in the dotted rectangle propose that an additional N terminus can also bind, presumably in the antechamber, thereby being protected from digestion by trypsin. Red and black ovoids correspond to unprotected and protected N termini, respectively. (B) Both one- and two-site models were used to simulate the trypsin-mediated removal of inactivation over a range of values of P, the binding constant for the site producing closed-channel–associated protection. Fractional occupancy was calculated as: Pa = pf/(pf+pr), where pf and pr are defined as the rates of binding and unbinding of an N terminus in the antechamber (Table II). Both models predict a similar prolongation of τd (filled circles) with occupancy within the antechamber, whereas the two differ in regards to the limiting power term at high fractional occupancies (open circles). Note that, for the one-site model, the relationship between fractional occupancy, prolongation of τd, and n were plotted incorrectly in our previous report (Zhang et al., 2006), as described in Materials and methods. (C) The prolongation of τd is plotted as a function of the power term for the one-site (red) and two-site (blue) models. The horizontal lines mark the range of prolongations, relative to the case for n = 4, expected for power terms in the range of 2 to 3. The filled black circles correspond to measurements of n and calculations of prolongation of τd for wild-type β2 and mutated β2 constructs, assuming that the digestion rate of β3a most closely reflects digestion of an N terminus without protection in the antechamber. If the rate of digestion of the W4G construct is used as the τd(min), the points are shifted leftward, but still favor the idea that two sites are involved in closed-channel–associated protection.
|