|
Figure 1. (A) Radial distribution of the α carbons for the residues V275 to N292 along the S6 transmembrane segment computed for the closed KCa3.1 structure generated using the KcsA channel as template. The Z axis refers to the pore central axis of diffusion with z = 0 centered at the level of the V275 residue. (B) Amino acid sequence alignment of the S6 transmembrane segment in KCa3.1, KcsA, and Shaker channels. The alignment was based on the highly conserved glycine hinge at position 274 in KCa3.1. Conserved residues are shaded in black (100%) and gray (67%), respectively. The S6 segment in KCa3.1 is presented as extending from V266 to A286.
|
|
(SCHEME 1).
|
|
(SCHEME 2).
|
|
Figure 2. Inside-out current recordings obtained in 25 μM internal Ca2+ conditions illustrating the various effects of the positively charged MTSEA (1 mM) and negatively charged MTSES (10 mM) reagents on KCa3.1 channels where residues predicted to be facing the channel pore along the S6 transmembrane segment were substituted to cysteines. Perfusion with a zero Ca2+ solution is marked by a filled square. The symbol c refers to the current level for the closed KCa3.1 channel. There was no detectable effect of MTSEA applied for 4 min on KCa3.1 wild type. However, internal application of MTSEA caused a time-dependent inhibition of the V275C and T278C mutants while resulting in a robust current increase for cysteines engineered at positions 283 and 286. Also illustrated in insets are the theoretical curves computed by curve fitting to Eq. 2, the time-dependent current variation initiated following MTSEA application. Only 1 point out of 100 is presented for clarity. The time course of the current inhibition for the V275C and T278C mutants was best reproduced assuming the binding of two MTSEA molecules. In contrast, the results from the A283C and A286C are in accordance with the formation of four MTSEA–cysteine complexes, with the binding of each MTSEA leading to an increase of the channel open probability coupled to a 10–15% decrease of the channel unitary conductance per MTSEA. Current recordings performed in symmetrical K2SO4 conditions at −60 mV membrane potential.
|
|
Figure 3. Example of single channel recording illustrating the stimulatory action of MTSEA on the 286C mutant. The amplitude of the current jumps recorded before MTSEA application was estimated at 2.4 pA at −60 mV membrane potential with a Po of 0.01 based on the entire recording. The unitary current amplitude following MTSEA binding to A286C decreased to 1 pA and 2.6 pA at −60 and −120 mV membrane potential, respectively, with a Po of 0.35 based on the entire recording. Recording obtained in symmetrical 200 K2SO4 conditions at saturating Ca2+ concentration (25 μM).
|
|
Figure 4. (A and B) Column graph representations of the modification rates measured in 25 μM Ca2+ conditions for residues along the S6 segment predicted to be facing the channel pore using MTSEA (A) and MTSET (B) as reagents. The rates of modification per cysteine KC estimated using Model I are illustrated as dense pattern columns, whereas the modification rates ki computed from Model II are shown as sparse pattern columns. In both cases the interaction parameters f1 and \documentclass[10pt]{article}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{pmc}
\usepackage[Euler]{upgreek}
\pagestyle{empty}
\oddsidemargin -1.0in
\begin{document}
\begin{equation*}f_{1}^{{\mathrm{i}}}\end{equation*}\end{document} were set to 0.98. The modification rates were computed for each mutant channel by curve fitting to Eq. 2 the time-dependent current variation triggered following the internal addition of MTSEA or MTSET in 25 μM internal Ca2+. The lowest rates of modification were obtained with the T278C and V282C mutants. (C) Column graph showing the modification rates measured for the V275C and A286C mutant channels using either a positively charged (MTSEA and MTSET), negatively charged (MTSES), or neutral (MTSACE) reagent. The modification rates were obtained by curve fitting to Eq. 2 the current variations induced by each MTS reagent. The modification rate KC given by Model I were found to strongly depend on the charge of the MTS with faster rates of modification with the positively charged MTSEA and MTSET as compared with the neutral MTSACE or negatively charged MTSES.
|
|
Figure 5. (A) Protocol used to measure the modification by MTSEA of a cysteine engineered in the channel central cavity (V275C) in zero (EGTA) internal Ca2+ conditions. MTSEA was applied for 3 s during 5.5-s pulses in zero internal Ca2+ conditions at a frequency of 0.1 Hz. The modification rates by MTSEA of the closed V275C mutant were estimated by fitting to Eq. 2 the time course of the test currents measured in 25 μM internal Ca2+ for each pulse following the washout of MTSEA (see inset). As illustrated, MTSEA could access the cysteine at position 275 in the channel central cavity with KCa3.1 in a closed configuration. (B) Channel activation induced by MTSET (1 mM)reacting with the A286C mutant in zero (EGTA) internal Ca2+ conditions. MTSET was applied for 1.0 s during a 2.0-s perfusion period with a Ca2+-free solution. The time course of the current variation induced by MTSET was measured as described in A. This recording demonstrates that the cysteine substituting for the alanine at 286 reacted with MTSET in conditions where KCa3.1 was maintained in a closed state configuration. All recordings were obtained at −Vp = −60 mV in symmetrical 200 K2SO4 conditions. The symbol c refers to the current level for the closed KCa3.1 channel.
|
|
Figure 6. Column graph representations of the modification rates measured in zero Ca2+ conditions for residues along the S6 segment predicted to be facing the channel pore using MTSEA (A) and MTSET (B) as reagents. Dense pattern and sparse pattern columns are defined as described in Fig. 4. These results indicate that the modification rates measured with MTSEA varied by <10-fold (V275C relative to A286C) for residues located along the S6 segment, whereas a 104-fold decrease in modification rates was measured with MTSET for cysteines at 275C in the channel central cavity as compared with cysteines distal to V282 (positions 283–286). Stars refer to modification rates <10−3 M−1s−1.
|
|
Figure 7. (A) Bar graph illustrating the state dependence of the modification rates measured for cysteine residues substituting for amino acids predicted to be facing the channel pore except for the V284 residue. The rates of modification per cysteine KC obtained with MTSET (circles) in 25 μM (open symbols) and zero (filled symbols) internal Ca2+ conditions differed by 103-fold at the level of the cavity lining residue 275, in contrast to MTSEA (triangles) where modification rates differed by less than sevenfold. (B) Bar graph illustrating the state dependence of the modification rates by Et-Hg+ (squares) and Ag+ (circles) of the cysteine residues at position 275. The open/closed modification rates estimated using a simple two-state model with a transition rate given by K0 were nearly identical. A modification rate of 3.1 ± 0.6 × 107 s−1M−1 was obtained with Ag+ for the closed V275C mutant. This value is close to the diffusion-limited modification rate expected for this ion, arguing for the C-terminal end of S6 being nonobstructive to Ag+ diffusing inside the channel cavity. Taken together, these results suggest that the C-terminal end of S6 does not constitute the Ca2+-dependent active gate of KCa3.1.
|
|
Figure 8. (A) Inside-out recording illustrating the inhibition of the V275C channel by Et-Hg+ (200 μM) applied internally. The time course of the current inhibition was fitted to a single exponential function. (B) Protocol used to measure the modification by Et-Hg+ of the closed V275C channel. Et-Hg+ (120 μM) was applied for 0.5 s during 5-s pulses in zero internal Ca2+ conditions at a frequency of 0.12 Hz. The modification rates by Et-Hg+ of the closed V275C mutant were estimated by fitting to a single exponential function the time course of the test currents measured in 25 μM internal Ca2+ for each pulse following the washout of Et-Hg+ (see inset). The time course of the current inhibition was identical for the open and closed state. (C) Perfusion protocol as described in B with Ag+ (25 nM) as modifying agent. These experiments required substantial buffering of the Ag+ concentration (60 mM EDTA) for reproducibility and stability, thus limiting the use of Ag+ to the closed configuration only as a μM concentration of free Ca2+ would have required to reduce the buffering of Ag+. These experiments confirmed that Ag+ could access the channel central cavity with the channel closed. Current recordings performed in symmetrical K2SO4 conditions at −Vp = −60 mV membrane potential.
|
|
Figure 9. (A) Dose–response curves of the V275C inhibition rate K0 as a function of the MTSEA concentration measured either in 25 μM (filled squares) or zero (empty squares) internal Ca2+ conditions. The modification rates were computed by fitting to a single exponential function the time course of the current inhibition induced by MTSEA using Eq. 2 with g =1 and K0 given by Eq. 3. The rates of modification obtained in zero internal Ca2+ were proportional to [MTSEA] for concentrations ranging from 0.1 to 30 mM, indicating a [MTSEA]1/2 ≫ 30 mM. The rates of modification measured in 25 μM Ca2+ conditions could be fitted to Eq. 1, leading to km = 7 s−1 and [MTSEA]1/2 = 35 mM. These results support an analysis of the modification rates Kn based on Eq. 1. In addition, the requirement in Model I and Model II of Kn being directly proportional to [MTS] is demonstrated to be valid for [MTSEA] < 5 mM. (B) Dose–response curves of the V275C modification rate measured either in 25 μM (filled circles) or zero (empty circles) internal Ca2+ conditions using Et-Hg+ as thiol modifying agent. The data points obtained with Et-Hg+ could be fitted to a linear equation for concentrations ranging from 20 to 200 μM. Slightly higher modification rates were measured in zero than 25 μM internal Ca2+ over the entire concentration range considered.
|
|
Figure 10. pH dependence of the modification rates measured for the V275C and A286C channels in zero internal Ca2+ conditions. Curve A (empty circles) shows the pH dependence obtained for the A286C mutant channel using MTSET (1 mM) as modifying agent. The data points (KC from Model I) could be fitted to a titration curve with a pKa of 8.4, close to the pKa predicted for the protonation of a cysteine residue (8.5). This result confirms that the observed pH dependence of the modification rates Kn (Eq. 5) can be interpreted in terms of an effect on the reaction rate km (Eq. 4), with little or no modulation of the entry and exit rates ki and ko. Curve B (filled squares) shows the pH dependence of the modification rate measured for MTSEA acting on the V275C mutant in zero Ca2+. The data points were obtained by curve fitting to a single exponential the time course of the current inhibition induced by MTSEA (K0). The resulting curve was fitted to a titration curve with a pKa of 7.8 for a km/ko ≈ 0.07 at pH 7.4 . With K0 = 0.112 s−1 at 1 mM MTSEA Eq. 5 leads to an entry rate ki = 480 M−1s−1. Curve C (empty squares) presents the pH dependence of the modification rate by MTSEA determined for the closed V275C mutant with TBA in the channel central cavity. The resulting titration curve led to a pKa of 6.9. As the modification rate K0 = 0.040 s−1 for 1 mM [MTSEA], Eq. 5 leads to ki = 54 M−1s−1. Curve D (empty triangles) illustrates the pH dependence of the modification rate by MTSEA measured for the A286C mutant in zero Ca2+. The titration curve was computed assuming a pKa ≈ 6.3 for a ratio km0/ko ≫ 1. ki can then be directly estimated from Kn (Model II in Materials and methods) with ki = 961 ± 63 (n = 3) M−1s−1.
|
|
Figure 11. (A) Example of the perfusion protocol implemented to study the effect of TBA on the modification rate by MTSEA of the closed V275C mutant. TBA (1 mM) was first applied during a 0.5-s period prior replacing the Ca2+-containing solution (25 μM) with a zero Ca2+ solution containing TBA at the same concentration. MTSEA (5 mM) was applied for 1 s during the zero Ca2+ perfusion period. The time course of the current variation induced by MTSEA was assayed according to the procedure described in Fig. 5. Experiments performed at −Vp = −60 mV. (B) Modification of the closed (EGTA) A286C mutant by the large MTS reagent MTS-PtrEA. A typical current increase was observed following the internal application of MTS-PtrEA (5 mM) in zero Ca2+, indicating that MTS-PtrEA had access to cysteines at position 286 with KCa3.1 in the closed configuration. A subsequent application of MTS-PtrEA in 25 μM internal Ca2+ failed to induce an additional current activation, suggesting that the binding sites were occupied. With dimensions equivalent to a cylinder 12 Å long by 9.5 Å in diameter, the MTS-PtrEA results suggest a pore diameter >9.5 Å, which can hardly be accounted for by a pore structure for the closed KCa3.1 based on the KcsA channel structure. (C) Experimental protocol used to measure the modification rate by MTSEA of cysteines at position 275 in the channel central cavity for the V275C-A286C double mutant channel. An identical protocol was employed to study the V275C-A283C double mutant. MTSET–cysteine complexes were first formed in zero Ca2+ conditions (MTSET cannot access the channel cavity under these conditions), and the rate of modification by MTSEA (5 mM) of the cysteines at 275 measured according to the protocol described in Fig. 5. (D) Column graph representation of the modification rates by MTSEA of cysteines at 275 with and without MTSET complexes formed at position 283 and 286, respectively. The modification rates of cysteines at position 275 measured with MTSET–cysteine complexes at position 286 and 283 were estimated at 168 ± 8 M−1s−1 (N = 3) and 104 ± 12 M−1s−1 (N = 3), respectively, compared with 112 ± 1 M−1s−1 (N = 7) for the V275C simple mutant. The presence of charged residues within the alleged channel inner vestibule for the closed KCa3.1 failed to modify the accessibility of MTSEA to cysteine residues located inside the channel central cavity. Taken together these results support a pore structure for the closed KCa3.1 with a large inner vestibule starting at V282.
|
|
Figure 12. Topology of the closed KCa3.1 channel inner vestibule. (A and B) Side view (A) and view from the cytosolic side (B) of the predicted structure for the closed KCa3.1 derived from KcsA illustrating the presence of a triple histidine ring system surrounding the channel inner vestibule. A double ring of eight histidines at positions 297 and 299 is predicted to be located directly at the pore entrance, whereas four additional histidines on the S5 transmembrane segment should be located at the level of the A286 residues. Because histidine residues are expected to be 90% protonated at pH 5.5, this structure predicts the equivalent of ≈11 positive charges in the channel inner vestibule. Also illustrated are two K+ ions along the channel pore. (C) Modification rates estimated for the V275C simple mutant and for the triple V275C-H297A-H299A and quadruple V275C-H203A-H297A-H299A mutant channels measured at pH 5.5. The modification rates were estimated by fitting to a single exponential the MTSEA-induced current inhibition for each mutant (K0). The substitution of the histidine residues by alanine did not alter the MTSEA cysteine accessibility at 275, suggesting that the packing of the TM2 transmembrane segments in KcsA can hardly account for the structure of the closed Kca3.1 channel. Molecular representations were produced with InsightII (Accelry) using the PDB file included as online supplemental material.
|
|
(SCHEME 3).
|
