The "dashpot" mechanism of stretch-dependent gating in MscS

Abstract

The crystal structure of the small conductance mechanosensitive channel (MscS) has been an invaluable tool in the search for the gating mechanism, however many functional aspects of the channel remain unsettled. Here we characterized the gating of MscS in Escherichia coli spheroplasts in a triple mutant (mscL-, mscS-, mscK-) background. We used a pressure clamp apparatus along with software developed in-lab to generate dose-response curves directly from two-channel recordings of current and pressure. In contrast to previous publications, we found that MscS exhibits essentially voltage-independent activation by tension, but at the same time strong voltage-dependent inactivation under depolarizing conditions. The MscS activation curves obtained under saturating ramps of pressure, at different voltages, gave estimates for the energy, area, and gating charge for the closed-to-open transition as 24 kT, 18 nm2, and +0.8, respectively. The character of activation and inactivation was similar in both K+ and Na+ buffers. Perhaps the most salient and intriguing property of MscS gating was a strong dependence on the rate of pressure application. Patches subjected to various pressure ramps from 2.7 to 240 mmHg/s revealed a midpoint of activation almost independent of rate. However, the resultant channel activity was dramatically lower when pressure was applied slowly, especially at depolarizing pipette voltages. It appears that MscS prefers to respond in full to abrupt stimuli but manages to ignore those applied slowly, as if the gate were connected to the tension-transmitting element via a velocity-sensitive "dashpot." With slower ramps, channels inactivate during the passage through a narrow region of pressures below the activation midpoint. This property of "dumping" a slowly applied force may be important in environmental situations where rehydration of cells occurs gradually and release of osmolytes is not desirable. MscS often enters the inactivated state through subconducting states favored by depolarizing voltage. The inactivation rate increases exponentially with depolarization. Based on these results we propose a kinetic scheme and gating mechanism to account for the observed phenomenology in the framework of available structural information.

Figure 1.

Responses of MscS populations to a 1-s linear pressure ramp followed by a 2-s plateau. (A) Superimposed raw traces obtained in the same patch at +20-mV pipette voltage. Arrows represent variability in maximal current. (B) One trace from the dataset in A converted into a dose–response curve by plotting P_o = I/I_max vs. pressure. (C) The P_o/P_c–tension dependence plotted in a semilogarithmic scale. The pressure scale was converted to tensions by equating the tension at the pressure midpoint (mean p_1/2 for all traces in A) to 5.5 dyne/cm and the linear part of the dependence was fitted with Eq. 1 (see materials and methods). (D) Correlated distribution of ΔE vs. ΔA parameters extracted from the traces obtained in the same patch (A).

Figure 2.

Responses of MscS populations to linear pressure ramps at different voltages. (A) Raw traces recorded at pipette voltages ranging from −100 to +100 mV (patch ruptures near the end of the −100-mV trace). Maximal conductance for the MscS population increases linearly with voltage in the range of moderate voltages (−30 to +40 mV). At high positive pipette voltages (>+60 mV), MscS behavior becomes more irregular and a population of channels remains open even after pressure has been released, increasing in number with higher voltages (right side, +80 and +100 mV). At very negative pipette voltages (<−40 mV), there is clear break in the modality of the traces as inactivation becomes prominent. (B) Maximum relative conductance as a function of voltage for four different patches. The data presented for KCl (closed circles) and NaCl (open triangles) recording buffers. (C) The energy of the closed-to-open transition (ΔE, fitted with the fixed ΔA = 17 nm²) as a function of voltage, for three independent patches in KCl (circles) and NaCl (triangles).

Figure 3.

Single-channel conductance of MscS as a function of voltage and the occupancy of substates. (A) Frequently observed single-channel conductances plotted as a function of pipette voltage. The solid line is the fit of maximal values representing the fully open state of MscS. The “break” of the curve at −40 and +60 mV mark the points where substates become prominent. (B and C) Traces illustrating the changes in MscS gating upon switching pipette voltage from +40 to −40 mV (B) and from +60 to −60 mV (C). The magnified episodes illustrate full transitions at positive pipette voltages and switching to subconducting states and inactivation at negative.

Figure 4.

MscS responses to ramps applied at different rate. (A) Raw current traces (top) recorded at pipette voltages of +20 and −20 mV under pressure ramps varying from 2.7 to 240 mm Hg/s (bottom). (B) Activation curves for different ramps (P_o = I/I_max, where I_max is scored at fastest ramp) presented in one scale as a function of pressure. Inset, maximum current as a function of rate at positive and negative voltages.

Figure 5.

Dynamics of MscS response at intermediate pressures. (A) Raw traces obtained with 20-s pressure prepulses of different amplitude followed by shorter (2 s) test pulses of saturating pressure. The current kinetics during the prepulse reflects the rate of channel inactivation, whereas the response to the test pulse checks for the MscS availability after the prepulse. The process of inactivation was fitted by single exponents shown by dotted lines. (B) The fraction of MscS inactivated during the prepulse as a function of pressure (open circles); the activation (P_o(p)) curve is presented to show the position of the pressure midpoint for the same patch. Inset, the characteristic time of inactivation as a function of pressure. (C) A trace obtained at a pressure near the midpoint from a patch having only five channels. The magnified episodes show the presence of substates at low positive pipette voltages.

Figure 6.

MscS inactivation is faster at depolarizing voltages. (A) Traces recorded with the prepulse pressure protocol at pipette voltages from −80 to +80 mV. The falling phase was fitted with a single exponent. (B) The characteristic time of inactivation as a function of pipette voltage. The high slope of the left part of the dependence corresponds to the barrier reduction equivalent to the energy of two positive charges (q = +2) per complex transferred outwardly across the entire transmembrane field.

Figure 7.

The hypothetical gating mechanism of MscS. (A) Kinetic scheme of MscS transitions presented on an area-charge plane shows transitions frequently observed at low voltages (solid lines), transitions favored by negative pipette voltages beyond −40 mV (dashed lines), and transitions which are possible but have not been characterized (dotted lines). The distances between the main states (closed, C; open, O; substate, S; and inactivated, I) reflect the changes of in-plane area (ΔA) and transmembrane movement of charges (Δq) associated with each transition. (B) Cartoon representation of helical positions associated with each state. An effective dashpot between the peripheral TM1–TM2 bundle and the pore-forming TM3 allows pore helices to disengage and collapse into a closed conformation, causing inactivation. A small upward swing of the peripheral helices generates gating charge associated with inactivation.