Yoda molecules agonize PIEZO2

Article activity feed

1. 

   eLife

   Oct 28, 2026

   eLife Assessment

   This study shows that the compound Yoda potentiates the function of the PIEZO2 channel, challenging the assumption that they selectively activate PIEZO1. This is an important finding and the evidence supporting this claim is solid. The authors have addressed the methodological and conceptual issues raised during review. Overall, this work will be of broad interest to researchers working with PIEZO channels across various biological scales.

   Read the original source
2. 

   eLife

   Oct 28, 2026

   Reviewer #1 (Public review):

   [Editor's note: this version has been assessed by the Reviewing Editor without further input from the original reviewers. The authors have addressed the comments raised in the previous round of review.]

   Summary:

   In this work, T. Wijerathne et al. investigated and reported the agonistic effect of Yoda1 and Yoda2 over PIEZO2 function using patch clamp electrophysiology, Ca2+ imaging, and molecular dynamics. They find that Yoda1 sensitizes PIEZO2 to membrane tension, can induce Ca2+ influx, and decreases its inactivation to a lesser degree than it does to PIEZO1 channels. Additionally, their data shows that Yoda2 sensitizes PIEZO2 channels to membrane indentation to a greater extent, but it has a weaker effect on channel inactivation than Yoda1. Interestingly, they report that a mutation in a conserved arginine …

   Reviewer #1 (Public review):

   [Editor's note: this version has been assessed by the Reviewing Editor without further input from the original reviewers. The authors have addressed the comments raised in the previous round of review.]

   Summary:

   In this work, T. Wijerathne et al. investigated and reported the agonistic effect of Yoda1 and Yoda2 over PIEZO2 function using patch clamp electrophysiology, Ca2+ imaging, and molecular dynamics. They find that Yoda1 sensitizes PIEZO2 to membrane tension, can induce Ca2+ influx, and decreases its inactivation to a lesser degree than it does to PIEZO1 channels. Additionally, their data shows that Yoda2 sensitizes PIEZO2 channels to membrane indentation to a greater extent, but it has a weaker effect on channel inactivation than Yoda1. Interestingly, they report that a mutation in a conserved arginine between PIEZO channels can be used to abolish PIEZO1-mediated Ca2+ flux in response to Yoda molecules. As a whole, the results presented here should be put into perspective against previous and future works involving systems where both PIEZO1 and PIEZO2 might be expressed. This is especially true for works where Yoda1 has been used as a basis for determining the absence of PIEZO2.

   Strengths:

   The authors use multiple techniques to investigate how Yoda molecules affect the three most important biophysical aspects of PIEZO channels that, when changed, result in pathophysiological responses: a) sensitivity to mechanical stimuli, b) Ca2+ entry, and c) channel inactivation. Lastly, they find a specific amino acid/region that could be exploited for drug design and/or development.

   Read the original source
3. 

   eLife

   Oct 28, 2026

   Reviewer #2 (Public review):

   Summary:

   This manuscript challenges the long-standing assumption that Yoda1 and Yoda2 are PIEZO1-selective activators. Using patch-clamp electrophysiology and calcium imaging in HEK293TΔPZ1 cells overexpressing PIEZO2, the authors demonstrate that Yoda1 potentiates PIEZO2 stretch-activated currents to a similar extent as PIEZO1 and slows PIEZO2 poking-current inactivation (albeit with lower efficacy). They further show that the more potent analog Yoda2 affects PIEZO2 at nanomolar concentrations and use mutagenesis and molecular dynamics simulations to propose that Yoda2's benzoic acid group forms a transient salt bridge with R1724 in the putative Yoda binding pocket, explaining its enhanced potency.

   Strengths:

   The authors are established Piezo/biophysics experts; the study is highly important, technically …

   Reviewer #2 (Public review):

   Summary:

   This manuscript challenges the long-standing assumption that Yoda1 and Yoda2 are PIEZO1-selective activators. Using patch-clamp electrophysiology and calcium imaging in HEK293TΔPZ1 cells overexpressing PIEZO2, the authors demonstrate that Yoda1 potentiates PIEZO2 stretch-activated currents to a similar extent as PIEZO1 and slows PIEZO2 poking-current inactivation (albeit with lower efficacy). They further show that the more potent analog Yoda2 affects PIEZO2 at nanomolar concentrations and use mutagenesis and molecular dynamics simulations to propose that Yoda2's benzoic acid group forms a transient salt bridge with R1724 in the putative Yoda binding pocket, explaining its enhanced potency.

   Strengths:

   The authors are established Piezo/biophysics experts; the study is highly important, technically competent, and carries significant implications for the reinterpretation of prior work that used Yoda compounds as PIEZO1-selective probes.

   The core finding that Yoda1 modulates PIEZO2 stretch currents is convincing and important.

   Read the original source
4. 

   eLife

   Oct 28, 2026

   Reviewer #3 (Public review):

   Summary:

   The manuscript reports that Yoda1 and Yoda2 agonize PIEZO2 in a manner similar to PIEZO1, increasing open probability and stretch sensitivity, but the mechanism underlying this sensitivity is incomplete. Mutagenesis was shown exclusively in PIEZO1, with no corresponding mutagenesis in PIEZO2, so the proposed mechanism in PIEZO2 is inferred by homology rather than directly tested. All experiments use mouse PIEZO2, and the human ortholog should be used before generalizing the proposed reinterpretation of the field.

   Strengths:

   The pressure-clamp electrophysiology demonstrating a shift in half-activation pressure for PIEZO2 is compelling evidence in support of the central claim.

   Read the original source
5. 

   eLife

   Oct 28, 2026

   Author response:

   The following is the authors’ response to the original reviews.

   Public Reviews:

   Reviewer #1 (Public review):

   Summary:

   In this work, T. Wijerathne et al. investigated and reported the agonistic effect of Yoda1 and Yoda2 over PIEZO2 function using patch clamp electrophysiology, Ca2+ imaging, and molecular dynamics. They find that Yoda1 sensitizes PIEZO2 to membrane tension, can induce Ca2+ influx, and decreases its inactivation to a lesser degree than it does to PIEZO1 channels. Additionally, their data shows that Yoda2 sensitizes PIEZO2 channels to membrane indentation to a greater extent, but it has a weaker effect on channel inactivation than Yoda1. Interestingly, they report that a mutation in a conserved arginine between PIEZO channels can be used to abolish PIEZO1mediated Ca2+ flux in response to Yoda …

   Author response:

   The following is the authors’ response to the original reviews.

   Public Reviews:

   Reviewer #1 (Public review):

   Summary:

   In this work, T. Wijerathne et al. investigated and reported the agonistic effect of Yoda1 and Yoda2 over PIEZO2 function using patch clamp electrophysiology, Ca2+ imaging, and molecular dynamics. They find that Yoda1 sensitizes PIEZO2 to membrane tension, can induce Ca2+ influx, and decreases its inactivation to a lesser degree than it does to PIEZO1 channels. Additionally, their data shows that Yoda2 sensitizes PIEZO2 channels to membrane indentation to a greater extent, but it has a weaker effect on channel inactivation than Yoda1. Interestingly, they report that a mutation in a conserved arginine between PIEZO channels can be used to abolish PIEZO1mediated Ca2+ flux in response to Yoda molecules. As a whole, the results presented here should be put into perspective against previous and future works involving systems where both PIEZO1 and PIEZO2 might be expressed. This is especially true for works where Yoda1 has been used as a basis for determining the absence of PIEZO2.

   Strengths:

   The authors use multiple techniques to investigate how Yoda molecules affect the three most important biophysical aspects of PIEZO channels that, when changed, result in pathophysiological responses: a) sensitivity to mechanical stimuli, b) Ca2+ entry, and c) channel inactivation. Lastly, they find a specific amino acid/region that could be exploited for drug design and/or development.

   Weaknesses:

   The methods and discussion sections are lacking enough detail to fully evaluate the findings and put them into perspective, respectively.

   We have included new results, expanded our discussion, and added additional details in our Methods section.

   Reviewer #2 (Public review):

   Summary:

   This manuscript challenges the long-standing assumption that Yoda1 and Yoda2 are PIEZO1selective activators. Using patch-clamp electrophysiology and calcium imaging in HEK293TΔPZ1 cells overexpressing PIEZO2, the authors demonstrate that Yoda1 potentiates PIEZO2 stretch activated currents to a similar extent as PIEZO1 and slows PIEZO2 poking-current inactivation (albeit with lower efficacy). They further show that the more potent analog Yoda2 affects PIEZO2 at nanomolar concentrations and use mutagenesis and molecular dynamics simulations to propose that Yoda2's benzoic acid group forms a transient salt bridge with R1724 in the putative Yoda binding pocket, explaining its enhanced potency.

   Strengths:

   The authors are established Piezo/biophysics experts; the study is highly important, technically competent, and carries significant implications for the reinterpretation of prior work that used Yoda compounds as PIEZO1-selective probes.

   The core finding that Yoda1 modulates PIEZO2 stretch currents is convincing and important. However, several conceptual, methodological, and presentational issues need to be addressed before acceptance, as detailed below.

   Weaknesses:

   (1) The abstract states that Yoda1 potentiates PIEZO2 "as efficaciously as PIEZO1." This claim is accurate only for stretch currents and single-channel open probability, but the paper itself demonstrates important asymmetries: i) Yoda molecules slow PIEZO2 poking-current inactivation ~2-fold, versus ~5-10 fold for PIEZO1 (Figure 3b and ref #60). ii) Spontaneous Ca²⁺ entry via PIEZO2 requires non-physiological conditions (high extracellular Ca²⁺, hypertonic solutions) that are unlikely to occur in native cells. The abstract should be revised to clearly qualify where equivalence holds and where efficacy differences exist. IMO, the current wording risks overcorrecting the historical bias (PIEZO1-only) by going too far in the other direction.

   We have changed our language throughout the manuscript to convey more accurately differences of Yoda agonism between the two Piezo homologs.

   (2) Related concern: the PIEZO2 Ca²⁺ signal in Figure 2 is only detectable using a Ca²⁺-boosted solution (CBS ie 30 mM Ca²⁺). Physiological extracellular Ca²⁺ and cells normally do not experience sustained hypertonicity at these magnitudes. The authors should explicitly clarify that the practical implication of their findings is primarily for electrophysiological (patch-clamp) experiments and that the Ca²⁺ imaging caveat applies only under amplified conditions. Specifically, the authors should state that in standard Ca²⁺ imaging assays with physiological buffers, PIEZO2 is unlikely to confound Yoda1 results.

   Agreed. As mentioned above, we have modified our wording to convey this point more clearly.

   Related point: Can cytochalasin D (CytoD) restore a Yoda1-dependent Ca²⁺ signal in physiological saline? This would help determine whether the weak PIEZO2 response is primarily a membrane tension issue (cytoskeletal tethering) versus intrinsically lower channel expression or permeability. The authors already have tagged PIEZO1/2 constructs and could, in principle, normalize by surface expression.

   We have repeated our acute pharmacological treatment in cells pre-treated with CytoD (or 0.5% DMSO vehicle control). These experiments were done in normal saline solution. We found that Yoda1 produces a very small (but significant) calcium influx in cells treated with CytoD. This result strengthens our conclusion that Yoda1 is capable of producing an activatory effect on Piezo2 in absence of external mechanical stimulation - albeit this effect is very small and only detectable in non-physiological (calcium-rich or destabilized cytoskeleton) conditions. It remains difficult to definitely conclude about why Yoda molecules exert different agonistic effects. We can only discuss possibilities. Differences in Ca2+ permeability is unlikely to contribute to this difference, as pore properties and selectivity for monovalent ions are similar between the two channels (see PMID: 39543876). Slight differences in how Yoda bind the Yoda binding region could explain why Yoda1 robustly modulates Piezo2 stretch currents but fails to produce robust calcium influx, as recent work suggests that these two agonist effects may originate from two distinct Yoda binding poses in this region (PMID: 38060566; PMID: 41402296). Piezo2’s intrinsically high threshold for stretch-activation could also represent a contributing factor, as figures 1e-f show that Piezo2’s midpoint activation pressure in the presence of Yoda1 (about -30 mmHg vs. -60 mmHg) might still be too high to enable robust Yoda-induced activation at the resting membrane tension of the cell. Since Piezo1 has been consistently shown (by us and others) to be expressed at a higher level than Piezo2 in transfected HEK293 cells (e.g., PMID: 20813920; PMID: 32227462, PMID: 37102984; PMCID: PMC13423835), we have not replicated these experiments with Piezo1. We have modified our manuscript to include these new results and discussion.

   (3) The mean inactivation tau values for wild-type PIEZO2 poking currents in both DMSO and Yoda1 conditions (Figure 3b, approximately 15-40 ms range) appear substantially higher than values reported in published literature (typically 5-10 ms; eg, PMID: 20813920). This discrepancy needs to be addressed.

   This discrepancy likely arises because inactivation kinetics of Piezo1 and/or 2 poking currents can be modulated by many biological and experimental variables that are difficult to control. Biological variables include membrane potential (PMID: 20813920), extracellular pH (PMID: 25561736), temperature (PMID: 35858335; PMID: 31413193), the presence of modulatory proteins (PMID: 37590348) and extracellular matrix (PMID: 28164706). Experimental variables include the mechanics of probe displacement (velocity and impact angle) and the precise positioning of the time-cursor near the peak current for exponential fitting. In addition, the decaying portion of Piezo poking currents is likely contributed by both inactivation and deactivation, as a recent study demonstrates relaxation of the poke stimulus during mechanical indentation in N2a cell (PMID: 42649207). Cellular viscoelastic properties and cytoskeletal organization are thus anticipated to dictate how fast this relaxation occurs, influencing the proportion of inactivating and deactivating currents in poking recordings. Since this point deviates from our main goal - to test an effect of Yoda on channel kinetics measured side-by-side – we prefer not to discuss it.

   (4) The authors perform all MD simulations on a truncated PIEZO1 model and justify this choice by noting that the Yoda binding region is highly conserved between homologs. This is a reasonable and defensible starting point given the availability of well-validated PIEZO1 simulation set ups in their lab. A few points are nonetheless worth addressing: While PIEZO2 simulations are not strictly required, the authors are encouraged to briefly discuss whether any long-range structural differences between PIEZO1 and PIEZO2 (outside the binding site itself) could influence Yoda2 binding dynamics, particularly in light of the chimera data showing that PIEZO2 sequence in repeat A abolishes Yoda1 sensitivity. This reviewer still doesn't understand the reason behind this discrepancy despite it being acknowledged in the text.

   We have now conducted three replicas of all-atom MD simulations of the Yoda2-bound PIEZO2 system. The new results show that the equivalent Arg residue in PIEZO2 (R2029) also forms a salt bridge with Yoda2, consistent with our observations from the Yoda2-bound PIEZO1 simulations. Taken together, these simulation results seem to point to the fact that Yoda molecules bind to a similar region in both PIEZO1 and PIEZO2. The lack of Yoda1-sensitivity in the chimera is quite puzzling, and we can only speculate about why. The swapped chimeric region includes 103 residues, 21 of which are non-conserved between PIEZO1 and PIEZO2 and distributed around the binding region (PMID: 29795280). These mutations could collectively and allosterically alter the binding free energy landscape in a manner that either 1) strongly reduce Yoda binding affinity or 2) equalize Yoda binding affinity for both curved and flat conformations, effectively rendering Yoda molecules silent binders (similar to how we previously described Dooku1 binding to Piezo1 PMID: 38060566).

   Another MD-related comment is that three simulation replicas (which is impressive for such a big system) show markedly different salt bridge occupancy (82.6%, 49.7%, 99.8%; stated in the text). This wide variation suggests incomplete sampling in at least one replica. The authors should provide RMSD plots for ligand and protein backbone to assess convergence and possibly discuss whether the 49.7% replica represents a genuinely distinct binding mode or incomplete equilibration.

   We have extended PIEZO1-Yoda2 replica #2 from 600 to 1400 nanoseconds. This replica shows that the salt-bridge turned ON and OFF during the simulation because the Arg side chain tends to flip. This flipping behavior is evident in the 20 µs Replica #1 (Figure 6c). In Figure 6-figure supplement 1, we analyzed our previous five replicas of MD simulations using the PACE force field. We show that the angle distributions indeed vary between replicas and subunits at the timescales currently affordable by MD technique for this large system. Therefore, it is not too surprising that the salt bridge occupancy differs between three all-atom simulations. We now provide RMSD of ligand and protein backbone for each Replica as Figure 6 supplement 1. The time evolution of Yoda2 binding poses across all simulation replicas are now illustrated in Figure 6-figure supplement 2, which shows that Yoda2 is mobile, yet remain stable in this overall hydrophobic binding site, consistent with our previous report (PNAS2023 PMID: 38060566). We also added all corresponding analysis of Yoda2-PIEZO2 in the same figures.

   (5) The Discussion proposes that PIEZO2's weaker Ca²⁺ response to Yoda1 could partly reflect lower membrane expression. Since the authors already have fluorescently tagged PIEZO1 and 2 constructs, a simple fluorescence intensity comparison between the two (acknowledging it would reflect total rather than surface expression) could provide at least indirect support for this claim. Alternatively, if such a comparison is not feasible, the authors may consider removing membrane expression from the list of proposed explanations or explicitly acknowledging that this remains unsubstantiated speculation. The max poking currents may somewhat and roughly indicate the level expression difference too, if done exactly side by side.

   PIEZO2 expression is known to be lower than PIEZO1 in heterologous systems (PMID: 20813920; PMID: 32227462, PMID: 37102984; PMCID: PMC13423835), hence we have not attempted to test if this was the case with our tagged constructs.

   (6) The abstract or concluding remarks should highlight that Dooku1 is not PIEZO1-selective in its agonist-like action on PIEZO2, and that Cmpd15/Cmpd64 appear to be better PIEZO1-selective tools. This nuance is buried in the Results section.

   We now list these results in the abstract.

   (7) The authors should not cite PMID 31015490. Clearly, any work on MCC13 is confounded by the overwhelming expression of PIEZO1 (PMID: 42084270). Instead, the authors should also cite the literature from others who have clearly recorded stretch currents from PIEZO2 before the cited studies (eg, PMID: 37590348).

   Agreed!

   Reviewer #3 (Public review):

   Summary:

   The manuscript reports that Yoda1 and Yoda2 agonize PIEZO2 in a manner similar to PIEZO1, increasing open probability and stretch sensitivity, but the mechanism underlying this sensitivity is incomplete. Mutagenesis was shown exclusively in PIEZO1, with no corresponding mutagenesis in PIEZO2, so the proposed mechanism in PIEZO2 is inferred by homology rather than directly tested. All experiments use mouse PIEZO2, and the human ortholog should be used before generalizing the proposed reinterpretation of the field.

   Strengths:

   The pressure-clamp electrophysiology demonstrating a shift in half-activation pressure for PIEZO2 is compelling evidence in support of the central claim.

   Weaknesses:

   (1) In the single-channel recordings (Figure 1a), it's unclear how many channels were present in those patches. After applying -60 mmHg pressure, multiple channels would be activated (as seen in Figure 1e). The number of channels in the patch and their inactivation rate could significantly influence the open probability in such experiments. To overcome this, in the original Yoda1 article (Syeda, Ruhma, et al. eLife 2015), no additional pressure was used.

   To clarify the confusion, we record single-channel PIEZO2 currents less than 24 hrs after transfection and record macroscopic PIEZO2 currents 2-3 days after transfection. Hence, unlike the cited study, our single and macro currents are obtained from different patches. When patching cells for single recordings, most patches do not show any mechano-currents and those that do typically show only one current level of single-channel conductance through the recording, even when the patch is highly pressurized.

   Additionally, the reported open probability comparison (n=7 Yoda1 vs n=17 DMSO patches) has an SEM nearly as large as the effect itself (0.30 {plus minus} 0.11), consistent with a small number of outliers driving this. The underlying mean open and shut times are reported without any statistical test; only the derived open probability receives a p-value. Additionally, in Figure 1a, the Yoda1 condition noise is different from the control. This should be stated if noise filtering was applied and how, given that this could affect open probability analysis.

   This is a fair point. We have removed two outliers and have performed statistical analysis on mean dwell times. The effect of Yoda1 is still highly significant. We have replaced current traces with two of similar noise levels. We acquired all single-channel current traces at 100kHz, filtered them at 5kHz, and displayed them at 500 Hz.

   (2) The calcium imaging data in Figure 2 raise significant concerns regarding the chemical activation claim. The calcium-boosted solution (30 mM Ca2+) is not physiological and appears to be generally stressing cells rather than specifically activating PIEZO2: the control condition under CBS already shows an elevated signal, consistent with cells being unwell at this calcium concentration, and adding Yoda1 on top of this shifted baseline raises further questions about specificity rather than confirming it. Separately, it is unclear why DMSO alone produces measurable PIEZO2-associated calcium influx in HBSS, a result that is not addressed in the text. Figure 2 should clearly indicate when DMSO/Yoda1 perfusion was initiated, and y-axis labels are missing from panels A and B.

   We agree and have toned down our chemical activation claim, insisting that this activation is very weak in absence of external mechanical stimulus, being only detectable under extreme, non-physiological conditions. We also re-analysed our data (and added fluorescence scale bars, thank you) showing time-course and scatter plots from the same cells in Figure 2a. We think the “bump” seen upon DMSO perfusion in PIEZO2 cells could be amplified by the low expression of GC6 in these cells. We now discuss this point in our revisions.

   (3) In the poke experiments, an activation threshold should be calculated and reported, and amplitude data (e.g., peak current versus indentation depth) should be shown rather than only inactivation tau values. It is also unclear why mClover3- and N-GFP-tagged constructs were used in these experiments, since electrophysiological recording already confirms channel expression without requiring a fluorescent tag.

   Valid point, we are now comparing peak currents for different poking distances. We have removed our Tau analysis for our tagged-PIEZO2 clones.

   (4) For inactivation kinetics (Figure 3b), the authors use unpaired comparisons across separate cells, whereas the deactivation experiments (Figure 3c) use paired; it should be applied to the inactivation experiments as well. Deactivation kinetics for PIEZO2 itself should be shown.

   These experiments have different goals which do not require using the same protocol. Deactivation kinetics experiments were used for the chimera because this parameter was shown by us to be the most sensitive to Yoda1. Hence, if a Yoda modulation exists, but is hard to detect in the chimera, using this protocol with paired samples provide us with the highest detection sensitivity. This level of sensitivity was not needed to detect an effect of Yoda1 on PIEZO2.

   If the claim is that Yoda1 acts on PIEZO2 through the same mechanism proposed for PIEZO1, then a PIEZO1/2 chimera should be expected to show a corresponding effect on deactivation tau; instead, this chimera is reported as completely Yoda1-insensitive despite both parental channels being Yoda1-sensitive, as shown in this study.

   This discrepancy is acknowledged in the text. We do not know why the chimera is Yoda insensitive and can only speculate at this point. Of note, one must acknowledge that this chimeric swapping introduces 21 amino acid changes (vs. WT PIEZO1) which may have profound allosteric effect on binding affinity and/or efficacy. MD studies would be useful to explore this possibility in a future study.

   (5) Given that this reflects a different experimental paradigm for Yoda EC50, PIEZO1 should be included within Figure 4b. Additionally, EC50 bar plots should be present on this figure.

   Showing EC50 plots here is not possible because each cell was exposed to only one dose of Yoda1 (unpaired measurements). We have included a Yoda1 dose-response curve against PIEZO1 inactivation Tau from our prior study (PMID: 35858335) for comparison as a Figure Supplement.

   The inactivation time constant for PIEZO2 without Yoda1 is inconsistent across figures, below 20 ms in Figure 3b but above 20 ms in Figure 4c.

   As explained above in our answer to Reviewer #2, these inconsistencies likely occur because PIEZO inactivation kinetics are susceptible to modulation by a number of biological and experimental factors that are not easily traceable, making it difficult to define absolute channel kinetics.

   (6) Finally, the modeling is performed exclusively on PIEZO1, whereas the manuscript's central focus is PIEZO2. It is therefore unclear whether the proposed structural mechanism, including the basis for Yoda2's reduced efficacy on PIEZO2, can be directly extrapolated to PIEZO2.

   We have now conducted three replicate all-atom MD simulations of the Yoda2-bound PIEZO2 system. The new results show that the equivalent Arg residue in PIEZO2 (R2029) also forms a salt bridge with Yoda2, consistent with our observations from the Yoda2-bound PIEZO1 simulations. However, the current simulations do not allow us to distinguish differential efficacy of Yoda2 toward PIEZO1 versus PIEZO2. The reduced efficacy of Yoda2 on PIEZO2 may instead arise from differences in its binding affinity for the open versus closed conformations, or from contributions of cellular components that are not represented in our MD simulations, such as the cytoskeleton.

   Recommendations for the authors:

   Reviewer #1 (Recommendations for the authors):

   Main:

   (1) Adding more depth and detail to the methods section will help others replicate the findings.

   Agreed.

   (2) Methods: line 325. Describe the way each compound was prepared for each concentration without undissolved particles. Avoid using the phrase "as described before" in the methods section.

   Compounds were dissolved in DMSO and simply diluted in recording solution at the time of experiments with vortex mixing.

   (3) Figure 1B (comparison of open probability): only three points out of 7 (with Yoda1) seem outside the data range of the control (DMSO). It seems very unreliable for an agonist. Moreover, in Figure 1F, the data seem very noisy. I would suggest adding more n.

   We have eliminated two outliers and reanalyzed our data, showing statistics for mean dwell times and open probability.

   (4) The authors should add to the "Yoda1 activates PIEZO2 in the absence of external mechanical stimulus" section (or the methods) how perfusion of Yoda1 was applied.

   We used manual pipetting.

   (5) Lines 165 - 171. This part should be moved to the discussion and expanded on. Particularly since the reference they allude to comes from the same group.

   This discussion part is needed to introduce the chimera experiments and thus should remain in Results.

   (6) Figure 3B. The noise range of the time constants of inactivation is a lot. Specially for transfected cells. I would suggest a couple of sentences explaining why the authors think this happens.

   This discussion is now included.

   (7) Figure 4C. Tau inac of DMSO control seems very high when compared to others. I would suggest a couple of sentences explaining why the authors think this happens.

   One difference is that DMSO concentration here is higher (0.25% vs 0.075%) to match the higher concentration of Yoda1 tested. Inactivation kinetics are subject to modulation by many parameters; hence we focus on side-by-side comparisons instead of absolute values.

   (8) Line 213. Which standard binding equation? Include in the methods.

   Done.

   Rethink calculating the EC50 with dose responses that saturate. Figure5C. Are these absolute or relative EC50s?

   EC50 of mutants are normalized to WT.

   (9) Discussion: "Our discovery that Yoda1 modulates PIEZO2 stretch currents as effectively as PIEZO1's strongly suggests that some of the reported biological roles attributed to PIEZO1 using this agonist could be (at least partly) transduced by PIEZO2. For instance, our discovery seamlessly reconciles the puzzling fact that reducing PIEZO2 expression in cells that co-express PIEZO1 also reduces the effects of Yoda1 (49, 76, 77). Along." In the referenced works (49, 76, and 77) its not checked if PIEZO1 levels remain the same when PIEZO2 was knocked out. Here, the authors should discuss this possibility.

   Good point!

   Minor:

   (1) Indicate in the figure legends, result or methods section, what the error bars represent.

   Done.

   (2) Indicate the p-value for Figure 1B.

   Done.

   (3) Figure 2. Include the ionomycin value that each group is normalized against.

   We provide now vertical scale bars for all samples.

   (4) Why not use PIEZO2 for the MD simulations?

   Done.

   Reviewer #2 (Recommendations for the authors):

   Minor comments

   (1) In the Abstract: The sentence "Although a structural framework for PIEZO1 activation by Yoda1 exists" may be confusing for readers outside the field. The authors may mean "mechanistic" or "computational" framework. Please clarify or reword.

   We have reworded the abstract.

   (2) Residue number in abstract: The abstract states "a conserved arginine in the Yoda binding site" without identifying it. Please add the residue number (R1724 in mouse PIEZO1) for clarity.

   Done

   (3) Figure 1 legend or in the method: recording mode and filter settings: The figure legend should explicitly state that single-channel recordings were obtained in cell-attached mode, and should report the filter frequency (kHz) and sampling rate used for idealization. This is essential for readers to evaluate dwell-time analysis.

   Done, but details are in the method.

   (4) Several instances throughout the text use "native" where "naïve" is clearly intended.

   Corrected, thank you.

   (5) The statement that PIEZO2 requires more mechanical force than PIEZO1 (~−60 vs ~−40 mmHg P1/2) should be interpreted cautiously without this information, and the authors should clarify whether pipette geometry was controlled between PIEZO1 and PIEZO2 experiments.

   We have shown that PIEZO2 requires more force to be opened by membrane tension relative to PIEZO1 using MscL gating as a tension-calibration system (PMID: 41703062). Pipette geometry is consistent for each type of experiment (macroscopic or single-channel recordings). We have expanded our Methods.

   (6) The finding that Dooku1slows PIEZO2 inactivation similarly to Yoda1 is striking and is currently underplayed. This raises important questions: Does Dooku1 elicit Ca²⁺ entry in PIEZO2-expressing cells? It could at least be discussed further in the text.

   Likely not, since Yoda-induced calcium influx is very weak and requires non-physiological conditions to be detectable. But this is worth trying at some point.

   Read the original source
6. 

   Version published to 10.7554/elife.111912.2 on eLife

   Oct 28, 2026
7. 

   Version published to 10.7554/elife.111912 on eLife

   Oct 28, 2026
8. 

   Version published to 10.64898/2026.05.08.723777v2 on bioRxiv

   Oct 13, 2026
9. 

   eLife

   Jul 9, 2026

   eLife Assessment

   This study presents an important finding regarding the effect of Yoda molecules on PIEZO2 function, challenging the assumption that they selectively activate PIEZO1. The evidence supporting this claim is solid, but several methodological and conceptual issues need to be addressed. Overall, this work will be of broad interest to researchers working with PIEZO channels across various biological scales.

   Read the original source
10. 

    eLife

    Jul 9, 2026

    Reviewer #1 (Public review):

    Summary:

    In this work, T. Wijerathne et al. investigated and reported the agonistic effect of Yoda1 and Yoda2 over PIEZO2 function using patch clamp electrophysiology, Ca2+ imaging, and molecular dynamics. They find that Yoda1 sensitizes PIEZO2 to membrane tension, can induce Ca2+ influx, and decreases its inactivation to a lesser degree than it does to PIEZO1 channels. Additionally, their data shows that Yoda2 sensitizes PIEZO2 channels to membrane indentation to a greater extent, but it has a weaker effect on channel inactivation than Yoda1. Interestingly, they report that a mutation in a conserved arginine between PIEZO channels can be used to abolish PIEZO1-mediated Ca2+ flux in response to Yoda molecules. As a whole, the results presented here should be put into perspective against previous and future …

    Reviewer #1 (Public review):

    Summary:

    In this work, T. Wijerathne et al. investigated and reported the agonistic effect of Yoda1 and Yoda2 over PIEZO2 function using patch clamp electrophysiology, Ca2+ imaging, and molecular dynamics. They find that Yoda1 sensitizes PIEZO2 to membrane tension, can induce Ca2+ influx, and decreases its inactivation to a lesser degree than it does to PIEZO1 channels. Additionally, their data shows that Yoda2 sensitizes PIEZO2 channels to membrane indentation to a greater extent, but it has a weaker effect on channel inactivation than Yoda1. Interestingly, they report that a mutation in a conserved arginine between PIEZO channels can be used to abolish PIEZO1-mediated Ca2+ flux in response to Yoda molecules. As a whole, the results presented here should be put into perspective against previous and future works involving systems where both PIEZO1 and PIEZO2 might be expressed. This is especially true for works where Yoda1 has been used as a basis for determining the absence of PIEZO2.

    Strengths:

    The authors use multiple techniques to investigate how Yoda molecules affect the three most important biophysical aspects of PIEZO channels that, when changed, result in pathophysiological responses: a) sensitivity to mechanical stimuli, b) Ca2+ entry, and c) channel inactivation. Lastly, they find a specific amino acid/region that could be exploited for drug design and/or development.

    Weaknesses:

    The methods and discussion sections are lacking enough detail to fully evaluate the findings and put them into perspective, respectively.

    Read the original source
11. 

    eLife

    Jul 9, 2026

    Reviewer #2 (Public review):

    Summary:

    This manuscript challenges the long-standing assumption that Yoda1 and Yoda2 are PIEZO1-selective activators. Using patch-clamp electrophysiology and calcium imaging in HEK293TΔPZ1 cells overexpressing PIEZO2, the authors demonstrate that Yoda1 potentiates PIEZO2 stretch-activated currents to a similar extent as PIEZO1 and slows PIEZO2 poking-current inactivation (albeit with lower efficacy). They further show that the more potent analog Yoda2 affects PIEZO2 at nanomolar concentrations and use mutagenesis and molecular dynamics simulations to propose that Yoda2's benzoic acid group forms a transient salt bridge with R1724 in the putative Yoda binding pocket, explaining its enhanced potency.

    Strengths:

    The authors are established Piezo/biophysics experts; the study is highly important, technically …

    Reviewer #2 (Public review):

    Summary:

    This manuscript challenges the long-standing assumption that Yoda1 and Yoda2 are PIEZO1-selective activators. Using patch-clamp electrophysiology and calcium imaging in HEK293TΔPZ1 cells overexpressing PIEZO2, the authors demonstrate that Yoda1 potentiates PIEZO2 stretch-activated currents to a similar extent as PIEZO1 and slows PIEZO2 poking-current inactivation (albeit with lower efficacy). They further show that the more potent analog Yoda2 affects PIEZO2 at nanomolar concentrations and use mutagenesis and molecular dynamics simulations to propose that Yoda2's benzoic acid group forms a transient salt bridge with R1724 in the putative Yoda binding pocket, explaining its enhanced potency.

    Strengths:

    The authors are established Piezo/biophysics experts; the study is highly important, technically competent, and carries significant implications for the reinterpretation of prior work that used Yoda compounds as PIEZO1-selective probes.

    The core finding that Yoda1 modulates PIEZO2 stretch currents is convincing and important. However, several conceptual, methodological, and presentational issues need to be addressed before acceptance, as detailed below.

    Weaknesses:

    (1) The abstract states that Yoda1 potentiates PIEZO2 "as efficaciously as PIEZO1." This claim is accurate only for stretch currents and single-channel open probability, but the paper itself demonstrates important asymmetries: i) Yoda molecules slow PIEZO2 poking-current inactivation ~2-fold, versus ~5-10 fold for PIEZO1 (Figure 3b and ref #60). ii) Spontaneous Ca²⁺ entry via PIEZO2 requires non-physiological conditions (high extracellular Ca²⁺, hypertonic solutions) that are unlikely to occur in native cells.

    The abstract should be revised to clearly qualify where equivalence holds and where efficacy differences exist. IMO, the current wording risks overcorrecting the historical bias (PIEZO1-only) by going too far in the other direction.

    (2) Related concern: the PIEZO2 Ca²⁺ signal in Figure 2 is only detectable using a Ca²⁺-boosted solution (CBS ie 30 mM Ca²⁺). Physiological extracellular Ca²⁺ and cells normally do not experience sustained hypertonicity at these magnitudes. The authors should explicitly clarify that the practical implication of their findings is primarily for electrophysiological (patch-clamp) experiments and that the Ca²⁺ imaging caveat applies only under amplified conditions. Specifically, the authors should state that in standard Ca²⁺ imaging assays with physiological buffers, PIEZO2 is unlikely to confound Yoda1 results.

    Related point: Can cytochalasin D (CytoD) restore a Yoda1-dependent Ca²⁺ signal in physiological saline? This would help determine whether the weak PIEZO2 response is primarily a membrane tension issue (cytoskeletal tethering) versus intrinsically lower channel expression or permeability. The authors already have tagged PIEZO1/2 constructs and could, in principle, normalize by surface expression.

    (3) The mean inactivation tau values for wild-type PIEZO2 poking currents in both DMSO and Yoda1 conditions (Figure 3b, approximately 15-40 ms range) appear substantially higher than values reported in published literature (typically 5-10 ms; eg, PMID: 20813920). This discrepancy needs to be addressed.

    (4) The authors perform all MD simulations on a truncated PIEZO1 model and justify this choice by noting that the Yoda binding region is highly conserved between homologs. This is a reasonable and defensible starting point given the availability of well-validated PIEZO1 simulation set ups in their lab. A few points are nonetheless worth addressing: While PIEZO2 simulations are not strictly required, the authors are encouraged to briefly discuss whether any long-range structural differences between PIEZO1 and PIEZO2 (outside the binding site itself) could influence Yoda2 binding dynamics, particularly in light of the chimera data showing that PIEZO2 sequence in repeat A abolishes Yoda1 sensitivity. This reviewer still doesn't understand the reason behind this discrepancy despite it being acknowledged in the text.

    Another MD-related comment is that three simulation replicas (which is impressive for such a big system) show markedly different salt bridge occupancy (82.6%, 49.7%, 99.8%; stated in the text). This wide variation suggests incomplete sampling in at least one replica. The authors should provide RMSD plots for ligand and protein backbone to assess convergence and possibly discuss whether the 49.7% replica represents a genuinely distinct binding mode or incomplete equilibration.

    (5) The Discussion proposes that PIEZO2's weaker Ca²⁺ response to Yoda1 could partly reflect lower membrane expression. Since the authors already have fluorescently tagged PIEZO1 and 2 constructs, a simple fluorescence intensity comparison between the two (acknowledging it would reflect total rather than surface expression) could provide at least indirect support for this claim. Alternatively, if such a comparison is not feasible, the authors may consider removing membrane expression from the list of proposed explanations or explicitly acknowledging that this remains unsubstantiated speculation. The max poking currents may somewhat and roughly indicate the level expression difference too, if done exactly side by side.

    (6) The abstract or concluding remarks should highlight that Dooku1 is not PIEZO1-selective in its agonist-like action on PIEZO2, and that Cmpd15/Cmpd64 appear to be better PIEZO1-selective tools. This nuance is buried in the Results section.

    (7) The authors should not cite PMID 31015490. Clearly, any work on MCC13 is confounded by the overwhelming expression of PIEZO1 (PMID: 42084270). Instead, the authors should also cite the literature from others who have clearly recorded stretch currents from PIEZO2 before the cited studies (eg, PMID: 37590348).

    Read the original source
12. 

    eLife

    Jul 9, 2026

    Reviewer #3 (Public review):

    Summary:

    The manuscript reports that Yoda1 and Yoda2 agonize PIEZO2 in a manner similar to PIEZO1, increasing open probability and stretch sensitivity, but the mechanism underlying this sensitivity is incomplete. Mutagenesis was shown exclusively in PIEZO1, with no corresponding mutagenesis in PIEZO2, so the proposed mechanism in PIEZO2 is inferred by homology rather than directly tested. All experiments use mouse PIEZO2, and the human ortholog should be used before generalizing the proposed reinterpretation of the field.

    Strengths:

    The pressure-clamp electrophysiology demonstrating a shift in half-activation pressure for PIEZO2 is compelling evidence in support of the central claim.

    Weaknesses:

    (1) In the single-channel recordings (Figure 1a), it's unclear how many channels were present in those patches. …

    Reviewer #3 (Public review):

    Summary:

    The manuscript reports that Yoda1 and Yoda2 agonize PIEZO2 in a manner similar to PIEZO1, increasing open probability and stretch sensitivity, but the mechanism underlying this sensitivity is incomplete. Mutagenesis was shown exclusively in PIEZO1, with no corresponding mutagenesis in PIEZO2, so the proposed mechanism in PIEZO2 is inferred by homology rather than directly tested. All experiments use mouse PIEZO2, and the human ortholog should be used before generalizing the proposed reinterpretation of the field.

    Strengths:

    The pressure-clamp electrophysiology demonstrating a shift in half-activation pressure for PIEZO2 is compelling evidence in support of the central claim.

    Weaknesses:

    (1) In the single-channel recordings (Figure 1a), it's unclear how many channels were present in those patches. After applying -60 mmHg pressure, multiple channels would be activated (as seen in Figure 1e). The number of channels in the patch and their inactivation rate could significantly influence the open probability in such experiments. To overcome this, in the original Yoda1 article (Syeda, Ruhma, et al. eLife 2015), no additional pressure was used. Additionally, the reported open probability comparison (n=7 Yoda1 vs n=17 DMSO patches) has an SEM nearly as large as the effect itself (0.30 {plus minus} 0.11), consistent with a small number of outliers driving this. The underlying mean open and shut times are reported without any statistical test; only the derived open probability receives a p-value. Additionally, in Figure 1a, the Yoda1 condition noise is different from the control. This should be stated if noise filtering was applied and how, given that this could affect open probability analysis.

    (2) The calcium imaging data in Figure 2 raise significant concerns regarding the chemical activation claim. The calcium-boosted solution (30 mM Ca2+) is not physiological and appears to be generally stressing cells rather than specifically activating PIEZO2: the control condition under CBS already shows an elevated signal, consistent with cells being unwell at this calcium concentration, and adding Yoda1 on top of this shifted baseline raises further questions about specificity rather than confirming it. Separately, it is unclear why DMSO alone produces measurable PIEZO2-associated calcium influx in HBSS, a result that is not addressed in the text. Figure 2 should clearly indicate when DMSO/Yoda1 perfusion was initiated, and y-axis labels are missing from panels A and B.

    (3) In the poke experiments, an activation threshold should be calculated and reported, and amplitude data (e.g., peak current versus indentation depth) should be shown rather than only inactivation tau values. It is also unclear why mClover3- and N-GFP-tagged constructs were used in these experiments, since electrophysiological recording already confirms channel expression without requiring a fluorescent tag.

    (4) For inactivation kinetics (Figure 3b), the authors use unpaired comparisons across separate cells, whereas the deactivation experiments (Figure 3c) use paired; it should be applied to the inactivation experiments as well. Deactivation kinetics for PIEZO2 itself should be shown. If the claim is that Yoda1 acts on PIEZO2 through the same mechanism proposed for PIEZO1, then a PIEZO1/2 chimera should be expected to show a corresponding effect on deactivation tau; instead, this chimera is reported as completely Yoda1-insensitive despite both parental channels being Yoda1-sensitive, as shown in this study.

    (5) Given that this reflects a different experimental paradigm for Yoda EC50, PIEZO1 should be included within Figure 4b. Additionally, EC50 bar plots should be present on this figure. The inactivation time constant for PIEZO2 without Yoda1 is inconsistent across figures, below 20 ms in Figure 3b but above 20 ms in Figure 4c.

    (6) Finally, the modeling is performed exclusively on PIEZO1, whereas the manuscript's central focus is PIEZO2. It is therefore unclear whether the proposed structural mechanism, including the basis for Yoda2's reduced efficacy on PIEZO2, can be directly extrapolated to PIEZO2.

    Read the original source
13. 

    Version published to 10.7554/elife.111912.1 on eLife

    Jul 9, 2026
14. Mentioned by @biorxiv-biophys.bsky.social on Bluesky

    May 13, 2026

    Yoda molecules agonize PIEZO2 https://www.biorxiv.org/content/10.64898/2026.05.08.723777v1

15. Mentioned by @biorxivpreprint.bsky.social on Bluesky

    May 13, 2026

    Yoda molecules agonize PIEZO2 https://www.biorxiv.org/content/10.64898/2026.05.08.723777v1

16. 

    Version published to 10.64898/2026.05.08.723777v1 on bioRxiv

    May 12, 2026