Inflammatory Signals Enhance Piezo2-Mediated Mechanosensitive Currents Adrienne E. Dubin, Manuela Schmidt, Jayanti Mathur, Matthew J. Petrus, Bailong Xiao, Bertrand Coste, Ardem Patapoutian  Cell Reports  Volume 2, Issue 3, Pages 511-517 (September 2012) DOI: 10.1016/j.celrep.2012.07.014 Copyright © 2012 The Authors Terms and Conditions

Cell Reports 2012 2, 511-517DOI: (10.1016/j.celrep.2012.07.014) Copyright © 2012 The Authors Terms and Conditions

Figure 1 Piezo2-Induced MA Currents in HEK293T Cells Are Enhanced by BDKRB2 and G Protein Activation (A) MA currents recorded at −80 mV in transiently transfected piezo2+BDKRB2-HEK cells as a function of probe displacement intensity before (black) and during (red) exposure to 300 nM BK. Below: The ramp (1 μm/ms)-and-hold mechanical stimulation protocol. Inset: stimulus-response curves before (black) and during BK exposure (red). (B) AUC is 6-fold larger in piezo2+BDKRB2-HEK cells challenged with BK (red bar) compared with vehicle-treated cells (p < 0.0001). The fold change was determined for individual cells as the value in the presence of BK normalized to that obtained prior to BK exposure. Data are presented as box and whiskers plots in which the bottom and top of the box are the 25th and 75th percentiles (lower and upper quartiles, respectively), the band near the middle of the box is the median, and the ends of the whiskers represent the minimum and maximum of all the data. Initial values used to calculate fold changes are shown in Table S1. (C) The increase in piezo2 current amplitude and τ-inac (tau) by BK (plotted as fold change) is concentration dependent [EC50: 28 nM, 95% confidence interval (%CI) 6–140 nM; and 20 nM, %CI 4–100 nM, respectively]. (D) Fold changes are shown for piezo2-dependent current amplitude (left) and τ-inac (right) during application of vehicle only (clear bars), BK (red bars), BK together with H0E-140 after 5 min preincubation in HOE-140 (red hatched bars), and BK 10 min after intracellular perfusion with 667 μM GDPβS-3Li (red stippled bars). Similar changes in response to BK were observed for BK in the presence of the vehicles DMSO (≤0.3%), internal LiCl (2 mM), and ethanol (≤0.1%) for 5–10 min and values are combined (red bars). Application of BK at 100, 300, and 1,000 nM produced similar results and the data are combined. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001 compared with BK; one-way ANOVA with Newman-Keuls multiple-comparison post hoc test. (E) Intracellular GTPγS increases piezo2 amplitude, τ-inac, and AUC compared with vehicle (2 mM LiCl) in piezo2-HEK cells. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.005 using Student’s t test. The number of individual cells tested is shown above each bar. See also Table S1. Cell Reports 2012 2, 511-517DOI: (10.1016/j.celrep.2012.07.014) Copyright © 2012 The Authors Terms and Conditions

Figure 2 PKA and PKC Enhance Piezo2-Dependent Currents (A) Piezo2 MA current amplitude, time course of decay and AUC in Piezo2-HEK cells are enhanced by 8-BrcAMP (100–300 μM; pink bars) and PMA (1 μM; green bars), but not the inactive congener 4αPMA (1 μM; hatched bars). Effects during application of both agonists together were not significantly different from those observed with either agonist alone. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.01 compared with vehicle; one-way ANOVA with Newman-Keuls multiple-comparison post hoc test. (B) Recordings showing modulation of piezo2-dependent currents by 8-BrcAMP and PMA. Inset: Peak current-displacement curves for control (black), presence of both agonists (red), and washout (open circles). (C) Combined inhibition of PKA (H89, 10 μM) and PKC (BIM I, 1 μM; red hatched bars), but not H89 (orange bars) or BIM I (green bars) alone, abrogated BK effects on both MA amplitude (top) and τ-inac (bottom) in piezo2+BDKRB2-HEK cells. ∗p < 0.01; ∗∗p < 0.001 compared with BK; one-way ANOVA with Newman-Keuls multiple-comparison post hoc test. See also Table S1 and Figures S1 and S2. Cell Reports 2012 2, 511-517DOI: (10.1016/j.celrep.2012.07.014) Copyright © 2012 The Authors Terms and Conditions

Figure 3 Piezo2-Dependent Currents in a Class of DRG Neurons Are Enhanced by Endogenous BDKRB2 Activation (A) Inkblot neuron 1 day after plating in NT-4- and NGF-containing media. (B) siRNA pool against piezo2 reduces MA current density in Inkblot neurons 3 days after nucleofection (RNAi) compared with scrambled (Scr) controls (gray bars), and has no significant effect on MA current density of Rosette neurons (pink bars). Four separate experiments. (C) Families of whole-cell MA currents elicited by increasing probe displacement before (black) and with BK (1 μM) after 5 min of exposure (red); τ-inac values are shown. Inset: Peak current as a function of displacement. (D) BK increases amplitude and AUC, and prolongs the decay time course (tau) through activation of endogenous BDKRB2 (red bars) compared with vehicle (clear bars). Cells pretreated with 1–2 μM HOE-140 for 5 min prior to BK (100–300 nM) exposure in the continued presence of HOE-140 significantly suppressed BK effects. ∗∗p < 0.01; ∗∗∗p < 0.001; one-way ANOVA with Newman-Keuls multiple-comparison post hoc test. (E) PKA and PKC inhibition abolished BK effects on Inkblot neurons. ∗p < 0.05; ∗∗∗p < 0.001 compared with BK using Student’s t test. See also Table S1. Cell Reports 2012 2, 511-517DOI: (10.1016/j.celrep.2012.07.014) Copyright © 2012 The Authors Terms and Conditions

Figure S1 Piezo2 Currents in the Presence or Absence of BK Are Unmodified by Gq/PLC or Gi/o Pathway Modulators, Related to Figure 2 PLC activation by m-3M3FBS has no effect on piezo2 currents, and robust BK-induced enhancement of current amplitude (right) and slowing of inactivation (left) are still observed in the absence of PTX-sensitive Gi/o, TG-sensitive calcium stores, and U73122- and edelfosine-sensitive PLC. Piezo2+BK-HEK cells were challenged with vehicle or 300 nM BK after preincubation with the indicated pharmacological agents. Cells were incubated 6–24 hr in PTX (100–200 ng/ml), which irreversibly inhibits Gi/o by ribosylation, and tested for their modulation by BK in the absence of PTX (black bars). Cells were preexposed to the irreversible inhibitor TG (2 μM) for 25–60 min and tested in the absence of TG (blue bars). BK was applied in the presence of U73122 (2–3.5 μM; green bars) or edelfosine (ET-18-OCH3, 10–15 μM; green hatched bars) after 30–180 min or 9–34 min, respectively. There was no significant difference between BK effects in the presence or absence of inhibitors. ∗∗∗p < 0.001 compared with vehicle; one-way ANOVA with Newman-Keuls multiple-comparison post hoc test. External DMSO (≤0.3%) and ethanol (≤0.1%) had no effect and were combined with vehicle data. Piezo2-HEK cells were exposed to 25 μM m-3M3FBS, and the MA current families were monitored for the first 5 min. There was no difference compared with controls. Cell Reports 2012 2, 511-517DOI: (10.1016/j.celrep.2012.07.014) Copyright © 2012 The Authors Terms and Conditions

Figure S2 Locations of Consensus Sites along the Hydropathicity Plot of the Amino Acid Sequence of Piezo2, Related to Figure 2 Hydropathicity was determined using Protean software (DNAstar Lasergene 10) with a scanning window of nine amino acids. The putative consensus PKA and PKC consensus sites were predicted using the prosite database. Predicted sites for PKA phosphorylation (n = 8): S387, S1068, S1633, S1652, S2137, S2169, S2190, and S2357. Predicted sites for PKC phosphorylation (n = 31): T56, T102, S110, S195, T321, S459, S483, T642, T892, T971, T1021, T1327, T1375, T1407, S1795, T1933, S1967, S2031, T2119, S2140, S2176, S2183, S2186, S2192, S2197, S2205, S2210, T2230, T2352, S2454, and T2620. Of these 39 sites, four putative PKC sites (T971, T1327, T1375, and T1407) are likely to be in transmembrane domains according to the Kyte-Doolittle hydrophobicity prediction. Cell Reports 2012 2, 511-517DOI: (10.1016/j.celrep.2012.07.014) Copyright © 2012 The Authors Terms and Conditions