Microscopic mechanism of PIEZO1 activation by pressure-induced membrane stretch

Abstract

Mechanosensitive PIEZO1 ion channels open in response to membrane stretch. Yet, the underlying microscopic mechanism of this activation remains unknown. To probe this mechanism, we used cell-attached pressure-clamp recordings to measure single channel currents at different steady-state negative pipette pressures, spanning the full range of the channel’s pressure sensitivity. Pressure-dependent activation occurs through a sharp reduction of the mean shut duration and through a moderate increase of the mean open duration. Across all tested pressures, the distribution of open and shut dwell times best follows sums of two and three exponential components, respectively. As the magnitude of the pressure stimulus increases, the time constants of most of these exponential components gradually change, in opposite directions for open and shut dwell times, and to a similar extent. In addition, while the relative amplitudes of fast and slow components remain unchanged for open intervals, they fully reverse for shut intervals, further reducing the mean shut duration. Using two-dimensional dwell time analysis, Markov-chain modeling, and simulations, we identified a minimal five-states model which recapitulates essential characteristics of single channel data, including microscopic reversibility, correlations between adjacent open and shut intervals, and asymmetric modulation of dwell times by pressure. This study identifies a microscopic mechanism for the activation of PIEZO1 channels by pressure-induced membrane stretch and deepens our fundamental understanding of mechanotransduction by a vertebrate member of the PIEZO channel family.

Analysis of pressure-activated Piezo1 open and subconductance states at a single channel level

“…Our data and conclusions closely mirror independent studies published by Wijerathne et al. ( 27 ), that is, Piezo1 shut state (and dwell times) is highly dependent on applied pressure; however, the previous studies did not report the occurrence of subconductance state. This could be due to the key differences in data acquisition, analysis, and presentation between our studies and Wijerathne et al.…”

“…4 , A and B ). As noted previously ( 27 ), we observed clear effect of pressure on shut dwell times distributions (mean shut time decreases with increasing pressure), but the SCLs and open-state dwell times remains the same in response to increasing negative pressure (no statistical significance p > 0.05, Table 1 ). To assess the frequency and occurrence of SCLs in response to pressure, we calculated occupancy of the channel in each state; shut, SCLs, and open ( Fig.…”

“…Our observation of the SCLs also warrants the development of a new single channel model for Piezo1 gating, as the previous models only focused on shut, open, and inactivated states ( 27 , 28 ). Furthermore, these models did not incorporate the dependence of the channel’s kinetics on voltage.…”

“…The all-point current-amplitude histograms (constructed from data acquired at −60 mV and −30 mm Hg) indicate four main states: (i) shut (current = 0 pA), (ii) first SCL (current = −0.5 ± 0.03 pA), (iii) a second SCL (current = −1.1 ± 0.02 pA), and (iv) fully open-state (current −1.5 ± 0.05 pA). There is also strong evidence for the existence of an inactivated state with current = 0 pA ( 27 , 36 ). The existence of two states with 0 pA current is also clear from the dwell time histogram of the shut-state ( Fig.…”

“…While single-channel models for Piezo1 have been developed before ( 27 , 28 , 45 ), the transition rate for those models are given at discrete pressure values. That is, data obtained at fixed pressure was used to obtain the rates.…”