Yoda1 analogue (Dooku1) which antagonizes Yoda1-evoked activation of Piezo1 and aortic relaxation

Article abstract of DOI:10.1111/bph.14188

Background and Purpose: The mechanosensitive Piezo1 channel has important roles in vascular physiology and disease. Yoda1 is a small-molecule agonist, but the pharmacology of these channels is otherwise limited. Experimental Approach: Yoda1 analogues were generated by synthetic chemistry. Intracellular Ca²⁺ and Tl⁺ measurements were made in HEK 293 or CHO cell lines overexpressing channel subunits and in HUVECs, which natively express Piezo1. Isometric tension recordings were made from rings of mouse thoracic aorta. Key Results: Modification of the pyrazine ring of Yoda1 yielded an analogue, which lacked agonist activity but reversibly antagonized Yoda1. The analogue is referred to as Dooku1. Dooku1 inhibited 2?μM Yoda1-induced Ca²⁺-entry with IC₅₀s of 1.3?μM (HEK 293 cells) and 1.5?μM (HUVECs) yet failed to inhibit constitutive Piezo1 channel activity. It had no effect on endogenous ATP-evoked Ca²⁺ elevation or store-operated Ca²⁺ entry in HEK 293 cells or Ca²⁺ entry through TRPV4 or TRPC4 channels overexpressed in CHO and HEK 293 cells. Yoda1 caused dose-dependent relaxation of aortic rings, which was mediated by an endothelium- and NO-dependent mechanism and which was antagonized by Dooku1 and analogues of Dooku1. Conclusion and Implications: Chemical antagonism of Yoda1-evoked Piezo1 channel activity is possible, and the existence of a specific chemical interaction site is suggested with distinct binding and efficacy domains.

Full text of DOI:10.1111/bph.14188

Yoda1 analogue (Dooku1) which antagonises Yoda1-evoked activation of
Piezo1 and aortic relaxation
Running title: Yoda1 antagonist
Elizabeth L Evans1†, Kevin Cuthbertson2†, Naima Endesh1†, Baptiste Rode1, Nicola M Blythe1,
Adam J Hyman1, Sally J Hall1, Hannah J Gaunt1, Melanie J Ludlow1, Richard Foster2 and David
J Beech1.
1Leeds Institute of Cardiovascular and Metabolic Medicine, School of Medicine, University of Leeds,
Leeds LS2 9JT, UK, 2School of Chemistry, University of Leeds, Leeds LS2 9JT, UK
To whom correspondence should be addressed: Professor David J Beech, Leeds Institute of
Cardiovascular and Metabolic Medicine, LIGHT Building, Clarendon Way, School of Medicine,
University of Leeds, Leeds LS2 9JT, UK. E-mail d.j.beech@leeds.ac.uk .Telephone +44 (0) 113
3434323. Dr Richard Foster, School of Chemistry, University of Leeds, Leeds LS2 9JT, UK. E-mail
r.foster@leeds.ac.uk. Telephone +44 (0) 113 3435759.
† Equal contributors
ACKNOWLEDGMENTS
This work was supported by research grants from the Wellcome Trust, Medical Research Council and
British Heart Foundation and studentship from the BBSRC for AJH and the Libyan Government for
NE.
Keywords: calcium channel, drug screening, endothelial cell, inhibitor, ion channel,
mechanotransduction, medicinal chemistry, mouse, pharmacology, vascular biology
This article has been accepted for publication and undergone full peer review but has not been
through the copyediting, typesetting, pagination and proofreading process which may lead to
differences between this version and the Version of Record. Please cite this article as doi:
10.1111/bph.14188
This article is protected by copyright. All rights reserved.

ABSTRACT
Background and Purpose The mechanosensitive Piezo1 channel has important roles in vascular
physiology and disease. Yoda1 is a small-molecule agonist but the pharmacology of these channels is
otherwise limited.
Experimental Approach Yoda1 analogues were generated by synthetic chemistry. Intracellular Ca2+
and Tl+ measurements were made in HEK 293 or CHO cell lines overexpressing channel subunits and
in human umbilical vein endothelial cells (HUVECs) which natively express Piezo1. Isometric tension
recordings were made from rings of mouse thoracic aorta.
Key Results Modification of the pyrazine ring of Yoda1 yielded an analogue which lacked agonist
activity but reversibly antagonised Yoda1. The analogue is referred to as Dooku1. Dooku1 inhibited 2
μM Yoda1-induced Ca2+-entry with IC50s of 1.3 μM (HEK 293 cells) and 1.5 µM (HUVECs) yet failed
to inhibit constitutive Piezo1 channel activity. It had no effect on endogenous ATP-evoked Ca2+
elevation or store-operated Ca2+ entry in HEK 293 cells or Ca2+ entry through TRPV4 or TRPC4
channels overexpressed in CHO and HEK 293 cells. Yoda1 caused dose-dependent relaxation of aortic
rings which was mediated by an endothelium- and nitric oxide-dependent mechanism and which was
antagonised by Dooku1 and analogues of Dooku1.
Conclusion and Implications Chemical antagonism of Yoda1-evoked Piezo1 channel activity is
possible and the existence of a specific chemical interaction site is suggested with distinct binding and
efficacy domains.
INTRODUCTION
Piezo1 protein is important for mechanical force sensing and its transduction in higher organisms (Coste
et al., 2010, Ranade et al., 2015, Wu et al., 2016). It assembles as a trimer with a propeller-like structure
around a central ion pore which is permeable to the cations Na+, K+ and Ca2+ (Ge et al., 2015, Coste et
al., 2015, Coste et al., 2012, Wu et al., 2017, Guo and MacKinnon, 2017, Saotome et al., 2017, Zhao et
al., 2018). Mechanical forces which include membrane tension and laminar flow are able to activate the
channel (Lewis and Grandl, 2015, Li et al., 2014, Syeda et al., 2016, Coste et al., 2010).
Roles of Piezo1 have been identified in embryonic vascular maturation, blood pressure regulation,
physical performance, hypertension-dependent arterial structural remodelling, urinary osmoregulation,
epithelial homeostasis and axonal growth (Li et al., 2014, Ranade et al., 2014, Cahalan et al., 2015,
Koser et al., 2016, Gudipaty et al., 2017, Martins et al., 2016, Rode et al., 2017, Retailleau et al., 2015).
In addition, pathological significance of Piezo1 has been suggested in humans. Gain of function
mutations have been linked to a form of haemolytic anaemia (hereditary stomatocytosis) and loss of
function mutations have been linked to autosomal recessive congenital lymphatic dysplasia (Albuisson
et al., 2013, Andolfo et al., 2013, Zarychanski et al., 2012, Fotiou et al., 2015, Lukacs et al., 2015, Bae
et al., 2013).
Piezo1 pharmacology is in its infancy. Inhibitors of the channel are limited to generic inhibitors of the
ion pore (Gd3+ and ruthenium red) and the spider toxin GsMTx4 which inhibits a range of
mechanosensitive ion channels and may act indirectly via the lipid bilayer (Drew et al., 2002, Bowman
et al., 2007, Bae et al., 2011, Suchyna et al., 2004). The first chemical activator of the channel, Yoda1,
was discovered in 2015 through high-throughput screening (Syeda et al., 2015). Yoda1 is a useful
research tool, not faithfully mimicking mechanical stimulation of the channels but facilitating study of
Piezo1 channels from a practical perspective without the need for mechanical stimulation and
importantly lacking effect on Piezo2 channels (Lukacs et al., 2015, Cahalan et al., 2015, Wang et al.,
2016, Rode et al., 2017).
This article is protected by copyright. All rights reserved.

As a step towards improved understanding of Piezo1 and the development of more and better Piezo1
modulators, increased knowledge on the structure-activity relationship for Yoda1 activation of Piezo1
would be helpful. Here we addressed this knowledge gap by synthesising Yoda1 analogues and testing
them against the Piezo1 channel, other channels, and vascular contractile function.
METHODS
Piezo1 tetracycline inducible HEK 293 cell line
HEK T-REx™ cells which overexpress Piezo1 upon induction with tetracycline were made as described
in Rode et al, 2017 (Rode et al., 2017). Expression was induced by treating the cells for 24 hours with
10 ng.mL-1 tetracycline (Sigma) and analysed by quantitative RT-PCR and western blot.
Cell culture
HEK 293 cells stably expressing tetracycline-regulated human Piezo1 were utilised as described above.
HEK 293 cells stably expressing tetracycline-regulated human TRPC4 have been described (Akbulut
et al., 2015). For the TRPC4 expressing cells, selection was achieved by including 400 μg.mL-1 zeocin
and 5 μg.mL-1 blasticidin in the cell medium. To induce expression, cells were incubated with 1 μg.mL-
1 tetracycline for 24 h prior to experiments. All HEK 293 cells were maintained in Dulbecco’s Modified
Eagle’s medium (Invitrogen, Paisley, UK) supplemented with 10% foetal calf serum (Sigma-Aldrich)
and 1% penicillin/streptomycin (Sigma-Aldrich).
Human Umbilical Vein Endothelial Cells (HUVECs) purchased from Lonza were maintained in
Endothelial Cell Basal Medium (EBM-2). This media was supplemented with a bullet kit (Cell Media
and Bullet Kit, Lonza) containing the growth factors: 10 ng.mL-1 vascular endothelial growth factor, 5
ng.ml-1 human basic fibroblast growth factor, 1 µg.mL-1 hydrocortisone, 50 ng.mL-1 gentamicin, 50
ng.mL-1 amphotericin B and 10 µg.mL-1 heparin, in addition to 2% foetal calf serum (Sigma). HUVECs
were passaged 2-6 times.
CHO K1 cells stably expressing human TRPV4 were maintained in Ham’s F12 (ThermoFisher
Scientific) in the presence of 1 mg.mL-1 G418 (Sigma-Aldrich).
All cells were grown at 37°C and in 5% CO2 in a humidified incubator.
RT-PCR
Total RNA was extracted using TRI reagent (Sigma-Aldrich). 500 ng of total RNA was used for reverse
transcription using the Reverse Transcription System (Promega). Real-time PCR was conducted using
an Applied Biosystems 7500 Real-Time PCR system with intron spanning primers and Taqman probe
for human Fam38A (Piezo1) (Hs00207230_m1) and GAPDH (Hs99999905_m1) (Applied
Biosystems).
Western blot for Piezo1 protein
Cells were harvested in lysis buffer (10 mM Tris (pH 7.4), 150 mM NaCl, 0.5 mM EDTA and 0.5%
Nonidet P40 substitute) containing protease and phosphatase inhibitor cocktails (Sigma). Equal
amounts of protein were loaded onto a 4-20% gradient gel (BioRad) and resolved by electrophoresis.
Samples were transferred to PVDF membranes and labelled overnight with BEEC-4 (1:1000,
Cambridge Biosciences). Horseradish peroxidase-donkey anti-rabbit secondary antibody (Jackson
ImmunoResearch Laboratories) and SuperSignal Femto detection reagents (Pierce) were used for
visualisation.
Intracellular Ca2+ measurements
HEK 293 and CHO cells were plated in poly-d-lysine coated 96-well plates (Corning, NY, USA) and
HUVECs in clear 96-well plates (Corning, NY, USA) at a confluence of 90%, 24 h before
experimentation. Cells were incubated with 2 μM fura-2-AM (Molecular Probes™) or 4 μM fluo-4-
This article is protected by copyright. All rights reserved.

AM (for TRPV4 expressing CHO cells), in the presence of 0.01% pluronic acid (Thermo Fisher
Scientific) in standard bath solution (SBS) for 1 h at 37°C. For recordings with fluo-4, 2.5 mM
probenecid (Sigma Aldrich) was included in the SBS throughout the experiment. Cells were washed
with SBS for 30 min at room temperature. If inhibitors were being tested, these were added at this time,
immediately following an SBS wash and maintained during the rest of the experiment. Measurements
were made at room temperature on a 96-well fluorescence plate reader (FlexStation, Molecular Devices,
Sunnyvale, CA, USA) controlled by Softmax Pro software v5.4.5. For recordings using fura-2, the
change (Δ) in intracellular calcium was indicated as the ratio of fura-2 emission (510 nm) intensities for
340 and 380 nm excitation. For recordings using fluo-4, the dye was excited at 485 nm and emitted
light collected at 525 nm and measurements are shown as absolute fluorescence in arbitrary units. The
SBS contained (mM): 130 NaCl, 5 KCl, 8 D-glucose, 10 HEPES, 1.2 MgCl2, 1.5 CaCl2 and the pH was
titrated to 7.4 with NaOH. For the Ca2+ add-back experiments, Ca2+ free SBS was used (without CaCl2)
and Ca2+ add-back was 0.3 mM. For the washout experiments, inhibitors were washed 3 times with SBS
immediately prior to recording.
FluxOR™ intracellular Tl+ (thallium ion) measurements
Induced (Tet+) and non-induced (Tet-) Piezo1 HEK 293 cells were plated in poly-d-lysine coated 96-
well plates (Corning, NY, USA) and HUVECs in clear 96-well plates (Corning, NY, USA) at a
confluence of 90%, 24 h before experimentation. Cells were loaded with FluxOR dye for 1 hour at room
temperature, before being transferred to assay buffer for 20 min. If inhibitors were being tested, these
were added at this time and maintained throughout the experiment. Cells were stimulated with a Tl+ -
containing K+ -free solution according to the manufacturer’s instructions (Molecular Probes).
Measurements were made at room temperature on a 96-well fluorescence plate reader (FlexStation,
Molecular Devices, Sunnyvale, CA, USA) controlled by Softmax Pro software v5.4.5. FluxOR was
excited at 485 nm, emitted light collected at 520 nm and measurements were expressed as a ratio
increase over baseline (F/F0).
Yoda1 analogue chemical synthesis
Analogues of Yoda1 were synthesised using three general synthetic approaches: 11 compounds [2a-
2k] were synthesised using a one-step procedure (Scheme S1), compounds 7a and 7b using a four-step
procedure (Scheme S2), and compound 11 using a separate four-step procedure (Scheme S3). All
synthesised chemicals were purified by column chromatography or trituration and determined as >97%
1
pure by H NMR (proton nuclear magnetic resonance) and 13C NMR (carbon-13 nuclear magnetic
resonance). Synthetic and analytical details are reported in the Supporting Information.
Animals
12-16 week-old, wild-type male C57BL/6 mice were used for experiments. All mice were housed in
GM500 individually ventilated cages (Animal Care Systems) at 21°C, 50-70% humidity and with a 12
h alternating light/dark cycle. They had ad libitum access to RM1 diet (SpecialDiet Services, Witham,
UK) with bedding from Pure’o Cell (Datesand, Manchester, UK). All animal experiments were
authorised by the University of Leeds Animal Ethics Committee and the UK Home Office. Animal
studies are reported in compliance with the ARRIVE guidelines (Kilkenny et al., 2010, McGrath and
Lilley, 2015).
Aorta contraction studies
The wire myograph technique using vessels from mice is regarded as a useful model for studying
vascular reactivity (Outzen et al., 2015). Animals were culled by CO2 inhalation, according to Schedule
This article is protected by copyright. All rights reserved.

1 procedure approved by the UK Home Office. Thoracic aorta was dissected out and immediately
placed into ice-cold Krebs solution (125 mM NaCl, 3.8 mM KCl, 1.2 mM CaCl2, 25 mM NaHCO3, 1.2
mM KH2PO4, 1.5 mM MgSO4, 0.02 mM EDTA and 8 mM D-glucose, pH 7.4). Connective tissue and
fat were carefully removed under a dissection microscope. 1 mm length segments were mounted in an
isometric wire myograph system (Multi Wire Myograph System, 620M, Danish Myo Technology) with
two 40 μm-diameter stainless steel wires, bathed in Krebs solution at 37°C and bubbled with 95% O2,
5% CO2. The segment was then stretched stepwise to its optimum resting tension to a 90% equivalent
transmural pressure of 100 mmHg and equilibrated for 1 h prior to experiments. The stretch was
approximately equal to that expected at diastolic blood pressure (Rode et al., 2017).
Data and statistical analysis
The data and statistical analysis comply with the recommendations on experimental design and analysis
in pharmacology (Curtis et al., 2015). OriginPro 2015 (OriginLab, Northampton, MA, USA) was used
for all data analysis. Averaged data are presented as mean ± SEM, where n represents the number of
independent experiments for a given result and N indicates the total number of replicates within the
independent experiments. Technical replicates were used to improve the confidence in data from
independent experiments. In order to compare the pharmacological activity of Yoda1 analogues, data
were normalised to the response of Yoda1 (agonist experiments) or the response of Yoda1 following
pre-treatment with vehicle only (inhibitor experiments). Data subjected to statistical analysis contained
at least 5 independent experiments (n). For comparisons between two sets of data, Student’s t-tests were
used. For multiple comparisons, one-way ANOVA was used with Tukey’s post-hoc test. P < 0.05 was
deemed significant. For IC50 determination, data were normalised to the vehicle controls (DMSO) and
curves were fitted using the Hill1 (Origin Pro 2015) equation. The analogues were novel and so their
initial testing occurred without knowledge of what effects might occur. Later in the study analogues
were blinded for aorta contraction experiments and used in random order. Randomisation and blinding
were not otherwise used.
Materials
Unless stated otherwise, all commercially available chemicals were purchased from Sigma-Aldrich.
Stocks of chemicals were reconstituted in DMSO and stored at -20 °C unless stated otherwise. Fura-2-
AM and fluo-4-AM (Molecular Probes) were dissolved at 1 mM. Pluronic acid F-127 was stored at
10% w/v in DMSO at room temperature. Probenecid was freshly prepared in 0.5 M NaOH and diluted
1:200 in SBS to give a working concentration of 2.5 mM. Yoda1 (Tocris) was stored at 10 mM. All
Yoda1 analogues were synthesised and purified (for more information, see Supporting Information) and
prepared as 10 mM stock solutions. Stock solutions were diluted 1:500 into the recording solution to
give a final working concentration of 0.02% DMSO. Thapsigargin and 4α-phorbol 12, 13-didecanoate
were stored as 5 mM and 10 mM stocks respectively. (-)-Englerin A was prepared as a 10 mM stock
solution and stored at -80°C. In experiments (-)-Englerin A was used in SBS containing 0.01% pluronic
acid as a dispersing agent to minimise aggregation of compound. Phenylephrine was stored at 100 mM
in an aqueous solution. ATP was stored at 10 mM in an aqueous stock solution. U46619 was stored as
a 10 mM stock in water. SIN-1 was stored as a 20 mM stock. BEEC-4 anti-Piezo1 antiserum, diluted
1:1000 for experiments, was generated by Cambridge Biosciences in rabbits by presentation of the
Piezo1 peptide DLAKGGTVEYANEKHMLALA.
Nomenclature of Targets and Ligands
Key protein targets and ligands in this article are hyperlinked to corresponding entries in
http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to
PHARMACOLOGY (Harding et al., 2018), and are permanently archived in the Concise Guide to
PHARMACOLOGY 2017/18 (Alexander et al., 2017).
This article is protected by copyright. All rights reserved.

RESULTS
2,6-dichlorophenyl ring of Yoda1 is important for Piezo1 activity
We synthesised a series of Yoda1 analogues, focusing on simple modifications to the 2,6-
dichlorophenyl ring (Figure 1A). To reliably study the effects of Yoda1 analogues on overexpressed
Piezo1 channels we stably incorporated tetracycline-inducible human Piezo1 expression in HEK 293
T-REx™ cells. These cells, hereby referred to as Piezo1 T-REx cells, showed Piezo1 expression after
tetracycline induction but not without induction (Figure 1B,C) and displayed dose-dependent Ca2+ entry
in response to Yoda1, in comparison to normal HEK 293 T-REx™ cells (without Piezo1 incorporation)
which showed no response (Figure 1D,E). The Yoda1 analogues were screened at 10 μM for their ability
to cause Ca2+ entry in these Piezo1 T-REx cells and compared to the Ca2+ entry caused by the same
concentration of Yoda1 (Figure 1F). All of the structural changes caused Piezo1 activation to be lost or
mostly lost, with all compounds showing less than 30% activation compared to Yoda1 (Figure 1F).
The analogues were also screened for their ability to inhibit the Yoda1 response (Figure 1G). Each
analogue was pre-incubated with the cells for 30 min at 10 μM, prior to the application of 2 μM Yoda1
in the continued presence of the analogue. Pre-incubation with these analogues did not affect the Ca2+
entry evoked by Yoda1, apart from 2g which caused inhibition. These data suggest that the 2,6-
dichlorophenyl moiety of Yoda1 is essential for interacting with the Piezo1 channel. Only analogue 2g
had any effect, showing a slight inhibitory effect but little agonist effect; it is chemically similar to
Yoda1 but with one fluorine replacing one chlorine.
Identification of a Yoda1 analogue which antagonises Yoda1
To further investigate the structure-activity relationship of Yoda1, we synthesised analogues of the
pyrazine group (Figure 2A). Similarly, these analogues were tested at 10 μM for their ability to cause
Ca2+ entry in Piezo1 T-REx cells, compared to Yoda1 (Figure 2B,C). Modification to the pyrazine ring
significantly reduced activity in comparison to Yoda1, but analogue 7a reached 50% of Yoda1 activity
(Figure 2B,C). We then synthesised analogues of the thiadiazole group (Figure 2D) and tested these in
the same manner (Figure 2E,F). Analogues containing an oxadiazole in place of a thiadiazole were also
less active, but analogue 11, the most similar in structure to Yoda1, showed ~70% activity (Figure
2E,F). These data suggest that the ability of Yoda1 to activate Piezo1 channels is dependent on very
specific structural requirements but that changes to the pyrazine and thiadiazole groups can be tolerated.
To investigate whether these analogues could inhibit Yoda1 activity we pre-incubated cells with
analogues and then tested Yoda1 (Figure 3A-G). The Yoda1 response was reduced by all analogues
(Figure 3G). Analogues 2i (Figure 3A), 2j (Figure 3B), 7a (Figure 3D), 7b (Figure 3E) and 11 (Figure
3F) also had agonist activity, as shown by the elevated baseline Ca2+ signal compared to vehicle
(DMSO) control. In contrast, analogue 2k (Figure 3C) inhibited the Yoda1 response without changing
the baseline, and so lacked agonist activity (Figure 3C). Analogue 2k was found to cause concentration-
dependent inhibition of Yoda1-induced Ca2+ entry with an IC50 value of 1.30 μM (Figure 3H). Inhibition
was incomplete at 10 μM but higher concentrations of 2k were not investigated because of solubility
limitations. Recovery from the inhibitory effect of 2k occurred after its washout (Figure 3I). The
inhibitory effect of 2k was not significantly different at 37 ˚C compared with room temperature (Figure
3J, K). The data suggest that 2k is an antagonist of Yoda1 which lacks agonist capability. We named
2k, Dooku1.
Dooku1 (analogue 2k) has selectivity for Piezo1
Pre-treatment with 10 μM Dooku1 had no effect on endogenous Ca2+ release in native HEK 293 cells
in response to 20 µM ATP (Figure 4A). Dooku1 (10 μM) had no effect on store-operated Ca2+ entry in
HEK 293 cells: the Ca2+ addback response after intracellular Ca2+ store depletion by 2 μM thapsigargin
This article is protected by copyright. All rights reserved.

(Figure 4B). Dooku1 (10 μM) had no effect on Ca2+ entry through TRPV4 channels overexpressed in
CHO cells and activated by 4αPDD (Figure 4C) or on Ca2+ entry through TRPC4 channels
overexpressed in T-REx™ HEK 293 cells and activated by 100 nM (-)-Englerin A (EA) (Figure 4D).
The data suggest selectivity of Dooku1 for Piezo1 channels.
Dooku1 does not inhibit constitutive Piezo1 activity
To investigate whether the effect of Dooku1 depends on Yoda1 we took advantage of constitutive
Piezo1 channel activity observed in the Piezo1 T-REx cells (Rode et al., 2017). The activity can be
detected using an intracellular thallium (Tl+) sensitive FluxOR™ indicator dye whereby Tl+ influx acts
as a surrogate for Na+ influx (Rode et al., 2017). Cells were maintained in a Tl+ free solution until 2 μM
Tl+ was added extracellularly 30 s into the recording and the resulting elevation of intracellular Tl+ was
detected. To ensure that constitutive Piezo1 channel activity was being represented in this assay, we
compared the rate of Tl+ entry in tetracycline-induced (Tet+) Piezo1 overexpressing cells to control
cells to which tetracycline was not added (Tet-) (Figure 5A,B). The initial rate of Tl+ entry in the Tet+
cells was nearly double that of control Tet- cells (Figure 5A,B).
Pre-treatment with Dooku1 did not reduce constitutive Piezo1 channel activity as shown by comparing
the DMSO and Dooku1 DMSO data (Figure 5C,D). Yoda1 increased the rate of Tl+ entry by ~2.5 fold
and this effect was inhibited by 10 μM Dooku1 as shown by comparing the Yoda1 and Dooku1 Yoda1
data (Figure 5C,D). These data suggest that Dooku1 has no effect on constitutive Piezo1 channel activity
and therefore that its effect depends on the presence of Yoda1.
Dooku1 inhibits endogenous Yoda1-activated channels
The above studies were on overexpressed Piezo1 channels. To investigate the relevance to endogenous
Piezo1 channels we studied human umbilical vein endothelial cells (HUVECs) which robustly express
endogenous Piezo1 channels (Li et al., 2014) and display a Piezo1-dependent Yoda1 response (Rode et
al., 2017). Similar to observations in Piezo1 T-REx cells (Figure 3C), Dooku1 did not evoke Ca2+ entry
(Figure 6A). Dooku1 was however able to inhibit the Yoda1 response in HUVECs (Figures 6B,C).
Dooku1 had a concentration-dependent inhibitory effect against Yoda1-induced Ca2+ entry in HUVECs,
acting with an IC50 of 1.49 μM (Figure 6D), which was comparable to the value in Piezo1 T-REx cells
even though its maximum effect was less (Figure 3H). These data suggest that Dooku1 is also an
antagonist of Yoda1-induced Piezo1 channels in endothelial cells. To investigate the reason for reduced
Dooku1 effect against the endogenous Yoda1-activated channel, we compared the concentration-effect
curves of Yoda1 in HUVECs (Figure 6E) and Piezo1 T-REx cells (Figure 6F). Yoda1 had increased
potency in HUVECs with an EC50 of 0.23 μM, compared to 2.51 μM in Piezo1 T-REx cells, suggesting
that greater Yoda1 potency in HUVECs may explain the smaller effect of Dooku1 in HUVECs.
Yoda1 causes endothelium- and nitric oxide-dependent relaxation of aorta
To investigate physiological responses we made isometric tension recordings from isolated murine
thoracic aorta rings. Yoda1 had no effect in the absence of phenylephrine (PE), which is an agonist of
α1-adrenoreceptors (Figure 7A). Rings contracted in response to PE (Figure 7B) and Yoda1 caused
concentration-dependent relaxation following this pre-contraction, with an estimated EC50 of 2.3 µM
(Figure 7B). Endothelium-denudation abolished the Yoda1 response but did not affect the PE response
(Figure 7C,D). Response to acetylcholine (ACh) was a positive control for functional endothelium and
this response was present in endothelium-intact rings but absent in endothelium-denuded rings (Figure
7C). The data suggest that Yoda1 causes endothelium-dependent relaxation in mouse thoracic aorta.
To determine whether the relaxation caused by Yoda1 was dependent on nitric oxide synthase (NOS),
we exposed rings to NOS inhibitor, Nω-nitro-L-arginine methyl ester (L-NAME). L-NAME prevented
This article is protected by copyright. All rights reserved.

Yoda1- and ACh-induced relaxation (Figure 7E,F). The data suggest that Yoda1 causes endothelium-
dependent relaxation in mouse thoracic aorta by stimulating nitric oxide production via the endothelium.
Dooku1 inhibits Yoda1-induced relaxation of aorta
To determine if Dooku1 inhibits relaxation caused by Yoda1, aortic rings were pre-incubated with 10
μM Dooku1 for 20 min. Dooku1 strongly suppressed the Yoda1-induced relaxation (Figure 8A-C). To
characterise this phenomenon in more detail we tested four further Yoda1 analogues in the aorta assay.
The selected analogues showed various abilities to inhibit Yoda1 responses in Piezo1 T-REx cells:
analogues 2e (no activation and no inhibition) (Figure 1), 2g (slight activation and slight inhibition)
(Figure 1), 7b (slight activation and partial inhibition) (Figures 2 and 3) and 11 (slight activation and
partial inhibition) (Figures 2 and 3). Analogue 2e had no effect (Figure 8D-F). 2g, 7b and 11 in contrast
suppressed the Yoda1-induced relaxation (Figure 8G-K). Moreover, the ability of these analogues to
inhibit Yoda1-induced relaxation correlated with inhibition of Yoda1-induced Ca2+ entry (Figure 8L).
The data suggest strong efficacy of Dooku1 as an inhibitor of Yoda1-induced aortic relaxation which is
mediated through disruption of Yoda1-induced Piezo1 channel activity.
Dooku1 is selective for Yoda1-induced relaxation but partially inhibits agonist contractile
responses
Analysis of the PE response in the presence of Dooku1 revealed significant inhibition without effect on
baseline tension (Figure 9A,B). To determine whether Dooku1’s inhibition of PE-induced contraction
was specific to this contractile agent, we also tested the effect of Dooku1 against contraction induced
by U46619, a thromboxane A2 mimetic. Aortic rings were pre-contracted with 0.1 μM U46619 (Figure
9C,D). Addition of Dooku1 caused partial relaxation (Figure 9D,E). In contrast, Dooku1 had no effect
on relaxation evoked by ACh (1 µM) or the NO donor SIN-1 (10 µM) (Figure 9F,G). Investigation of
the PE response in the presence of the other 4 Yoda1 analogues revealed no inhibitory effect (Figure
10). The data suggest Dooku1 selectively inhibits Yoda1-induced relaxation but also partially inhibits
receptor-mediated agonist responses via unknown mechanisms.
DISCUSSION AND CONCLUSIONS
This study has provided insight into the structure-activity relationships for Piezo1 channel activation by
Yoda1 with the goal of generating new tools for investigating Piezo1 channel function. Through this
research we have identified and named Dooku1, an inhibitor of Yoda1-induced Piezo1 channel activity
which strongly inhibits Yoda1-induced relaxation of aorta. The data suggest that Dooku1 may compete
with Yoda1 at a binding site or act allosterically at another site to reduce the binding or efficacy of
Yoda1.
During the discovery of Yoda1, the 2,6-dichlorophenyl group of the compound was highlighted as
important with particular reference to the chlorine atoms (Syeda et al., 2015). Our findings support this
conclusion and add new knowledge by demonstrating that small changes to this group result in complete
loss of Piezo1 channel activation. Removing one of the chlorine atoms [2b] or altering the position of
the chlorine atom around the ring [2c/d] abolished activity. Replacing one or both of the chlorine atoms
with fluorine [2a/g] also abolished activity implying that both chlorine atoms are important for activity
and may interact with Piezo1 in a chlorine specific manner, potentially via a sigma-hole interaction,
such as a halogen-pi bond. The 4-methoxyphenyl [2e] and 4-nitrophenyl [2f] analogues were also
inactive. Investigating the inhibitory potential of the compounds showed that all but 2g, which is the
most similar in structure to Yoda1, were ineffective at inhibiting Yoda1 activity.
This article is protected by copyright. All rights reserved.

Piezo1 channel activation by analogues with modification to the pyrazine group was less than that of
Yoda1, with the most successful analogue, compound 7a, in which the pyrazine was replaced with a 3-
pyridyl group, exhibiting 50% of the activity of Yoda1. This demonstrates the importance of the
nitrogen atom in the 2-position of the pyrazine ring, with loss of this nitrogen resulting in a 50% drop
of activity. The remaining two compounds from the series, the phenyl [2i] and 2-pyrrolyl [7b]
analogues, were less active than 7a which suggests that the presence of the nitrogen atom at the 3-
position of the pyridine ring in 7a is also contributing to Piezo1 activation, supporting our understanding
of the importance of the nitrogen atom at the equivalent position on the pyrazine ring of Yoda1 to
activity. We next investigated replacement of the central thiadiazole ring by an oxadiazole [11]. This
change was largely tolerated with the new compound demonstrating 70% of the activity of Yoda1. The
other two compounds from the series were less active, although the data for the 2-pyridyl analogue [2j]
were interesting in that the partial activity observed for the analogue suggests that the position of the
nitrogen atom on the pyridine contributes to activity, reinforcing the importance of the equivalent N on
the pyrazine ring of Yoda1 to activity.
Investigation into the inhibitory potential of this set of left-hand and middle ring-modified analogues
provided compounds with potential promise of being pharmacological tools. All of the compounds from
the series had the ability to reduce Ca2+ entry evoked by Yoda1 by at least 40%, and as much as 75%
in the case of 2j. However, most of these compounds exhibited partial agonist activity. The most
promising compound, 2k (Dooku1) effectively reduced Yoda1 activity by 60%, without causing any
activation and was a strong inhibitor of the Yoda1 response in the physiological setting of murine aortic
rings. This shows that the pyrazine ring can be replaced to identify compounds which do not activate
the channel but do inhibit the Yoda1 response. It appears that analogues lacking the 2,6-dichlorophenyl
group do not activate the channels or inhibit Yoda1 whereas pyrazine-modified analogues show reduced
activation and ability to inhibit Yoda1. Therefore the di-chloro group seems to be critical for binding
while the pyrazine group is less important for binding but key for channel activation.
Currently the only available inhibitors of Piezo1 activity are not selective for Piezo1 (Bae et al., 2011,
Drew et al., 2002). Dooku1 is also not perfect as it does not directly block the channels but it is a new
tool compound which is useful for Piezo1 characterisation studies. It antagonises the action of Yoda1
and could facilitate understanding of an important small-molecule binding site on or near to Piezo1
channels. Without agonist activity, Dooku1 effectively inhibits Yoda1-induced Piezo1 activity. It does
so without disturbing several Ca2+ handling events in the cell or affecting other aortic relaxing agents.
Although these data suggest specificity of Dooku1 for Piezo1 channels, further studies to address this
point are warranted, especially given the inhibitory effect of Dooku1 against PE and U46619-induced
contractions of aortic rings which might reflect a Piezo1 mechanism or some other unknown effect of
Dooku1. It is possible that Dooku1 may be acting on Piezo1 in smooth muscle cells of the vessel,
partially inhibiting contraction. This assumes that the channels become activated via a Yoda1-like
mechanism during contraction. Piezo1 was found not be required for normal myogenic tone (Retailleau
et al., 2015) and so a non-Piezo1 target of Dooku1 should be considered.
Dooku1 only has activity against Yoda1-induced and not constitutive Piezo1 channel activity. Such an
effect is consistent with Dooku1 acting at the same or a similar site to Yoda1 and thereby occluding
access of Yoda1 to its agonist binding site. The reversibility of Dooku1 is consistent with the
reversibility of Yoda1 (Rocio Servin-Vences et al., 2017). It would be good to investigate if the Dooku1
effect is consistent with competitive antagonism but solubility limitations of the compounds prevented
construction of appropriate concentration-response curves. The inability of Dooku1 to have any effect
on constitutive activity suggests that the mechanism of background channel activity is different to that
of chemical activation with Yoda1.
Dooku1 partially inhibited Yoda1 in HUVECs but strongly inhibited it in aorta (Figure 6D cf Figure
8C). We initially speculated that the difference was due to the higher temperature of the contraction
This article is protected by copyright. All rights reserved.

studies (37 ˚C cf room temperature) but the Dooku1 effect was not significantly temperature dependent
(Figure 3K). An alternative explanation might be that Ca2+ entry is not directly proportional to nitric
oxide production, so that partial inhibition of Yoda-1 induced Ca2+ entry is sufficient to inhibit most of
the relaxation induced by Yoda1. Another divergence was that Yoda1 was more potent in HUVECs
than Piezo1 T-REx cells, showing a difference between native and over-expressed Piezo1 channels
(Figure 6E,F). We speculate that this difference reflected a higher basal state of activity of the channels
in endothelial cells, as described previously (Rode et al., 2017), making the channels more sensitive to
Yoda1 because they are better primed for opening.
In summary, this study has provided important insight into the structure-activity relationships of Yoda1
and supported the concept of a specific chemical binding site on or in close proximity to Piezo1
channels. It has also revealed the discovery of a useful tool compound, Dooku1, which effectively
antagonises Yoda1-induced Piezo1 channel activity, distinguishing it from constitutive Piezo1 channel
activity. The complete role of Piezo1 in vascular biology is still being established but the protein may
have significant clinical interest with emerging roles in genetic disease, blood pressure control,
hypertension-induced arterial remodelling and exercise capacity (Wang et al., 2016, Retailleau et al.,
2015, Rode et al., 2017). As yet, it is not clear whether activating or inhibiting this channel may be
advantageous, but increasing our pharmacological knowledge, alongside our physiological knowledge
of Piezo1 will be essential if therapeutic potential of this protein is to be harnessed in the future.
Learning more about Piezo1 channel interactions with small-molecules promises to be an important
aspect of the overall effort to understand Piezo1 biology.
CONFLICT OF INTEREST
The authors state no conflict of interest
AUTHOR CONTRIBUTIONS
E.L.E., K.C., N.E., B.R., N.M.B., A.J.H., S.J.H., H.J.G., and M.J.L. performed experiments and data
analysis. E.L.E., K.C., D.J.B. and R.F. designed the research. D.J.B. and R.F. raised funds to support
the work. E.L.E., K.C., R.F., and D.J.B. co-wrote the paper.
This article is protected by copyright. All rights reserved.

REFERENCES
AKBULUT, Y., GAUNT, H. J., MURAKI, K., LUDLOW, M. J., AMER, M. S., BRUNS,
A., VASUDEV, N. S., RADTKE, L., WILLOT, M., HAHN, S., SEITZ, T.,
ZIEGLER, S., CHRISTMANN, M., BEECH, D. J. & WALDMANN, H. 2015. (−)-
Englerin A is a Potent and Selective Activator of TRPC4 and TRPC5 Calcium
Channels. Angewandte Chemie International Edition, 54, 3787-3791.
ALBUISSON, J., MURTHY, S. E., BANDELL, M., COSTE, B., LOUIS-DIT-PICARD, H.,
MATHUR, J., FENEANT-THIBAULT, M., TERTIAN, G., DE JAUREGUIBERRY,
J. P., SYFUSS, P. Y., CAHALAN, S., GARCON, L., TOUTAIN, F., SIMON
ROHRLICH, P., DELAUNAY, J., PICARD, V., JEUNEMAITRE, X. &
PATAPOUTIAN, A. 2013. Dehydrated hereditary stomatocytosis linked to gain-of-
function mutations in mechanically activated PIEZO1 ion channels. Nat Commun, 4,
1884.
ALEXANDER, S. P. H., KELLY, E., MARRION, N. V., PETERS, J. A., FACCENDA, E.,
HARDING, S. D., PAWSON, A. J., SHARMAN, J. L., SOUTHAN, C., DAVIES, J.
A. & COLLABORATORS, C. 2017. THE CONCISE GUIDE TO
PHARMACOLOGY 2017/18: Other ion channels. British Journal of Pharmacology,
174, S195-S207.
ANDOLFO, I., ALPER, S. L., DE FRANCESCHI, L., AURIEMMA, C., RUSSO, R., DE
FALCO, L., VALLEFUOCO, F., ESPOSITO, M. R., VANDORPE, D. H.,
SHMUKLER, B. E., NARAYAN, R., MONTANARO, D., D’ARMIENTO, M.,
VETRO, A., LIMONGELLI, I., ZUFFARDI, O., GLADER, B. E., SCHRIER, S. L.,
BRUGNARA, C., STEWART, G. W., DELAUNAY, J. & IOLASCON, A. 2013.
Multiple clinical forms of dehydrated hereditary stomatocytosis arise from mutations
in PIEZO1. Blood, 121, 3925.
BAE, C., GNANASAMBANDAM, R., NICOLAI, C., SACHS, F. & GOTTLIEB, P. A.
2013. Xerocytosis is caused by mutations that alter the kinetics of the
mechanosensitive channel PIEZO1. Proceedings of the National Academy of
Sciences, 110, E1162-E1168.
BAE, C., SACHS, F. & GOTTLIEB, P. A. 2011. The mechanosensitive ion channel Piezo1 is
inhibited by the peptide GsMTx4. Biochemistry, 50, 6295-6300.
BOWMAN, C. L., GOTTLIEB, P. A., SUCHYNA, T. M., MURPHY, Y. K. & SACHS, F.
2007. Mechanosensitive Ion Channels and the Peptide Inhibitor GsMTx-4: History,
Properties, Mechanisms and Pharmacology. Toxicon : official journal of the
International Society on Toxinology, 49, 249-270.
CAHALAN, S. M., LUKACS, V., RANADE, S. S., CHIEN, S., BANDELL, M. &
PATAPOUTIAN, A. 2015. Piezo1 links mechanical forces to red blood cell volume.
Elife, 4.
COSTE, B., MATHUR, J., SCHMIDT, M., EARLEY, T. J., RANADE, S., PETRUS, M. J.,
DUBIN, A. E. & PATAPOUTIAN, A. 2010. Piezo1 and Piezo2 are essential
components of distinct mechanically-activated cation channels. Science (New York,
N.Y.), 330, 55-60.
COSTE, B., MURTHY, S. E., MATHUR, J., SCHMIDT, M., MECHIOUKHI, Y.,
DELMAS, P. & PATAPOUTIAN, A. 2015. Piezo1 ion channel pore properties are
dictated by C-terminal region. 6, 7223.
COSTE, B., XIAO, B., SANTOS, J. S., SYEDA, R., GRANDL, J., SPENCER, K. S., KIM,
S. E., SCHMIDT, M., MATHUR, J., DUBIN, A. E., MONTAL, M. &
This article is protected by copyright. All rights reserved.

PATAPOUTIAN, A. 2012. Piezo proteins are pore-forming subunits of mechanically
activated channels. Nature, 483, 176-81.
CURTIS, M. J., BOND, R. A., SPINA, D., AHLUWALIA, A., ALEXANDER, S. P. A.,
GIEMBYCZ, M. A., GILCHRIST, A., HOYER, D., INSEL, P. A., IZZO, A. A.,
LAWRENCE, A. J., MACEWAN, D. J., MOON, L. D. F., WONNACOTT, S.,
WESTON, A. H. & MCGRATH, J. C. 2015. Experimental design and analysis and
their reporting: new guidance for publication in BJP. British Journal of
Pharmacology, 172, 3461-3471.
DREW, L. J., WOOD, J. N. & CESARE, P. 2002. Distinct Mechanosensitive Properties of
Capsaicin-Sensitive and -Insensitive Sensory Neurons. The Journal of Neuroscience,
22, RC228-RC228.
FOTIOU, E., MARTIN-ALMEDINA, S., SIMPSON, M. A., LIN, S., GORDON, K., BRICE,
G., ATTON, G., JEFFERY, I., REES, D. C., MIGNOT, C., VOGT, J., HOMFRAY,
T., SNYDER, M. P., ROCKSON, S. G., JEFFERY, S., MORTIMER, P. S.,
MANSOUR, S. & OSTERGAARD, P. 2015. Novel mutations in PIEZO1 cause an
autosomal recessive generalized lymphatic dysplasia with non-immune hydrops
fetalis. Nat Commun, 6, 8085.
GE, J., LI, W., ZHAO, Q., LI, N., CHEN, M., ZHI, P., LI, R., GAO, N., XIAO, B. & YANG,
M. 2015. Architecture of the mammalian mechanosensitive Piezo1 channel. Nature,
527, 64-9.
GUDIPATY, S. A., LINDBLOM, J., LOFTUS, P. D., REDD, M. J., EDES, K., DAVEY, C.
F., KRISHNEGOWDA, V. & ROSENBLATT, J. 2017. Mechanical stretch triggers
rapid epithelial cell division through Piezo1. Nature, 543, 118-121.
GUO, Y. R. & MACKINNON, R. 2017. Structure-based membrane dome mechanism for
Piezo mechanosensitivity. eLife, 6, e33660.
HARDING, S. D., SHARMAN, J., FACCENDA, E., SOUTHAN, C., PAWSON, A.,
IRELAND, S., GRAY, A. J. G., BRUCE, L., ALEXANDER, S. P. H., ANDERTON,
S., BRYANT, C., DAVENPORT, A. P., DOERIG, C., FABBRO, D., LEVI-
SCHAFFER, F., SPEDDING, M., DAVIES, J. & NC-IUPHAR, N.-I. 2018. The
IUPHAR/BPS Guide to PHARMACOLOGY in 2018: Updates and expansion to
encompass the new Guide to IMMUNOPHARMACOLOGY. Nucleic Acids
Research.
KILKENNY, C., BROWNE, W., CUTHILL, I. C., EMERSON, M. & ALTMAN, D. G.
2010. Animal research: Reporting in vivo experiments: The ARRIVE guidelines.
British Journal of Pharmacology, 160, 1577-1579.
KOSER, D. E., THOMPSON, A. J., FOSTER, S. K., DWIVEDY, A., PILLAI, E. K.,
SHERIDAN, G. K., SVOBODA, H., VIANA, M., COSTA, L. D. F., GUCK, J.,
HOLT, C. E. & FRANZE, K. 2016. Mechanosensing is critical for axon growth in the
developing brain. Nat Neurosci, 19, 1592-1598.
LEWIS, A. H. & GRANDL, J. 2015. Mechanical sensitivity of Piezo1 ion channels can be
tuned by cellular membrane tension. Elife, 4.
LI, J., HOU, B., TUMOVA, S., MURAKI, K., BRUNS, A., LUDLOW, M. J., SEDO, A.,
HYMAN, A. J., MCKEOWN, L., YOUNG, R. S., YULDASHEVA, N. Y., MAJEED,
Y., WILSON, L. A., RODE, B., BAILEY, M. A., KIM, H. R., FU, Z., CARTER, D.
A., BILTON, J., IMRIE, H., AJUH, P., DEAR, T. N., CUBBON, R. M., KEARNEY,
M. T., PRASAD, K. R., EVANS, P. C., AINSCOUGH, J. F. & BEECH, D. J. 2014.
Piezo1 integration of vascular architecture with physiological force. Nature, 515, 279-
82.
LUKACS, V., MATHUR, J., MAO, R., BAYRAK-TOYDEMIR, P., PROCTER, M.,
CAHALAN, S. M., KIM, H. J., BANDELL, M., LONGO, N., DAY, R. W.,
This article is protected by copyright. All rights reserved.

STEVENSON, D. A., PATAPOUTIAN, A. & KROCK, B. L. 2015. Impaired
PIEZO1 function in patients with a novel autosomal recessive congenital lymphatic
dysplasia. Nat Commun, 6, 8329.
MARTINS, J. R., PENTON, D., PEYRONNET, R., ARHATTE, M., MORO, C., PICARD,
N., KURT, B., PATEL, A., HONORÉ, E. & DEMOLOMBE, S. 2016. Piezo1-
dependent regulation of urinary osmolarity. Pflügers Archiv - European Journal of
Physiology, 468, 1197-1206.
MCGRATH, J. C. & LILLEY, E. 2015. Implementing guidelines on reporting research using
animals (ARRIVE etc.): new requirements for publication in BJP. British Journal of
Pharmacology, 172, 3189-3193.
OUTZEN, E. M., ZAKI, M., ABDOLALIZADEH, B., SAMS, A., BOONEN, H. C. M. &
SHEYKHZADE, M. 2015. Translational value of mechanical and vasomotor
properties of mouse isolated mesenteric resistance‐sized arteries. Pharmacology
Research & Perspectives, 3, e00200.
RANADE, S. S., QIU, Z., WOO, S.-H., HUR, S. S., MURTHY, S. E., CAHALAN, S. M.,
XU, J., MATHUR, J., BANDELL, M., COSTE, B., LI, Y.-S. J., CHIEN, S. &
PATAPOUTIAN, A. 2014. Piezo1, a mechanically activated ion channel, is required
for vascular development in mice. Proceedings of the National Academy of Sciences,
111, 10347-10352.
RANADE, S. S., SYEDA, R. & PATAPOUTIAN, A. 2015. Mechanically Activated Ion
Channels. Neuron, 87, 1162-1179.
RETAILLEAU, K., DUPRAT, F., ARHATTE, M., RANADE, S. S., PEYRONNET, R.,
MARTINS, J. R., JODAR, M., MORO, C., OFFERMANNS, S., FENG, Y.,
DEMOLOMBE, S., PATEL, A. & HONORÉ, E. 2015. Piezo1 in Smooth Muscle
Cells Is Involved in Hypertension-Dependent Arterial Remodeling. Cell Reports, 13,
1161-1171.
ROCIO SERVIN-VENCES, M., MORONI, M., LEWIN, G. R. & POOLE, K. 2017. Direct
measurement of TRPV4 and PIEZO1 activity reveals multiple mechanotransduction
pathways in chondrocytes. eLife, 6, e21074.
RODE, B., SHI, J., ENDESH, N., DRINKHILL, M. J., WEBSTER, P. J., LOTTEAU, S. J.,
BAILEY, M. A., YULDASHEVA, N. Y., LUDLOW, M. J., CUBBON, R. M., LI, J.,
FUTERS, T. S., MORLEY, L., GAUNT, H. J., MARSZALEK, K.,
VISWAMBHARAN, H., CUTHBERTSON, K., BAXTER, P. D., FOSTER, R.,
SUKUMAR, P., WEIGHTMAN, A., CALAGHAN, S. C., WHEATCROFT, S. B.,
KEARNEY, M. T. & BEECH, D. J. 2017. Piezo1 channels sense whole body physical
activity to reset cardiovascular homeostasis and enhance performance. Nature
Communications, 8, 350.
SAOTOME, K., MURTHY, S. E., KEFAUVER, J. M., WHITWAM, T., PATAPOUTIAN,
A. & WARD, A. B. 2017. Structure of the mechanically activated ion channel Piezo1.
Nature.
SUCHYNA, T. M., TAPE, S. E., KOEPPE II, R. E., ANDERSEN, O. S., SACHS, F. &
GOTTLIEB, P. A. 2004. Bilayer-dependent inhibition of mechanosensitive channels
by neuroactive peptide enantiomers. Nature, 430, 235.
SYEDA, R., FLORENDO, M. N., COX, C. D., KEFAUVER, J. M., SANTOS, J. S.,
MARTINAC, B. & PATAPOUTIAN, A. 2016. Piezo1 Channels Are Inherently
Mechanosensitive. Cell Rep, 17, 1739-1746.
SYEDA, R., XU, J., DUBIN, A. E., COSTE, B., MATHUR, J., HUYNH, T., MATZEN, J.,
LAO, J., TULLY, D. C., ENGELS, I. H., PETRASSI, H. M., SCHUMACHER, A.
M., MONTAL, M., BANDELL, M. & PATAPOUTIAN, A. 2015. Chemical
activation of the mechanotransduction channel Piezo1. Elife, 4.
This article is protected by copyright. All rights reserved.

WANG, S., CHENNUPATI, R., KAUR, H., IRING, A., WETTSCHURECK, N. &
OFFERMANNS, S. 2016. Endothelial cation channel PIEZO1 controls blood pressure
by mediating flow-induced ATP release. J Clin Invest, 126, 4527-4536.
WU, J., GOYAL, R. & GRANDL, J. 2016. Localized force application reveals mechanically
sensitive domains of Piezo1. 7, 12939.
WU, J., LEWIS, A. H. & GRANDL, J. 2017. Touch, Tension, and Transduction – The
Function and Regulation of Piezo Ion Channels. Trends in Biochemical Sciences, 42,
57-71.
ZARYCHANSKI, R., SCHULZ, V. P., HOUSTON, B. L., MAKSIMOVA, Y., HOUSTON,
D. S., SMITH, B., RINEHART, J. & GALLAGHER, P. G. 2012. Mutations in the
mechanotransduction protein PIEZO1 are associated with hereditary xerocytosis.
Blood, 120, 1908-15.
ZHAO, Q., ZHOU, H., CHI, S., WANG, Y., WANG, J., GENG, J., WU, K., LIU, W.,
ZHANG, T., DONG, M.-Q., WANG, J., LI, X. & XIAO, B. 2018. Structure and
mechanogating mechanism of the Piezo1 channel. Nature.
This article is protected by copyright. All rights reserved.

Figure 1- The 2,6-dichlorophenyl group of Yoda1 is required for activation of Piezo1. (A)
Structures of Yoda1 and analogues. Structural variation to Yoda1 is highlighted by the box outline. (B)
Western blot of control T-REx and Piezo1 T-REx cells with anti-Piezo1 antibody, confirming Piezo1
expression (predicted size, 286 kDa). (C) Real-time PCR of Piezo1 mRNA levels relative to GAPDH
mRNA in T-REx and Piezo1 T-REx cells. Error bars indicate S.E.M (n = 3). (D and E) FlexStation
intracellular Ca2+ measurement data for T-REx cells (D) and Piezo1 T-REx cells (E) exposed to Yoda1
at the specified concentrations or exposed to the vehicle only (DMSO). (F) (Left) FlexStation
intracellular Ca2+ measurement data for Piezo1 T-REx cells exposed to 10 μM 2e or exposed to vehicle
only (DMSO). Error bars indicate S.E.M (N = 3). (Right) Summary for experiments of the type shown
on the left measured between 40-60 s after Yoda1 analogue application, expressed as a % of the 10 μM
Yoda1 response. Each data point represents a value from an independent experiment with mean values
and error bars representing S.E.M indicated in black (n=5). (G) (Left) FlexStation intracellular Ca2+
measurement data for Piezo1 T-REx cells exposed to 2 μM Yoda1 after pre-treatment with 10 μM 2e
or vehicle only (DMSO). Error bars indicate S.E.M (N = 3). (Right) Summary for experiments of the
type shown on the left, as for (F, right) except data are expressed as a % of the Yoda1 response when
pre-treated with vehicle only (DMSO) (n=5; 2b, 2c, 2e, 2g, 2h, n=7; 2a, 2d, 2f).
This article is protected by copyright. All rights reserved.

Figure 2- Changes to the pyrazine ring or replacing the thiadiazole with an oxadiazole give rise
to less active analogues. (A) Structures of Yoda1 and analogues with changes to the pyrazine ring.
Structural variation to Yoda1 is highlighted by the box outline. (B) FlexStation intracellular Ca2+
measurement data for Piezo1 T-REx cells exposed to 10 μM 7a or exposed to vehicle only (DMSO).
Error bars indicate S.E.M (N = 3). (C) Summary for experiments of the type shown in (B) measured
between 40-60 s after Yoda1 analogue application, expressed as a % of the 10 μM Yoda1 response.
Each data point represents a value from an independent experiment with mean values and error bars
representing S.E.M indicated in black (n=5). (D) Structures of Yoda1 analogues with an oxadiazole.
Structural variation to Yoda1 is highlighted by the box outline. (E) FlexStation intracellular Ca2+
measurement data for Piezo1 T-REx cells exposed to 10 μM 2j or exposed to vehicle only (DMSO).
Error bars indicate S.E.M (N = 3). (F) Summary for experiments of the type shown in (E), as for (C).
This article is protected by copyright. All rights reserved.

Figure 3- Yoda1 analogues are able to inhibit Yoda1-induced Piezo1 activity. (A-F) FlexStation
intracellular Ca2+ measurement data for Piezo1 T-REx cells exposed to 2 μM Yoda1 after pre-treatment
with 10 μM 2i (A), 2j (B), 2k (C), 7a (D), 7b (E), 11 (F) or vehicle only (DMSO). Error bars indicate
S.E.M (N = 3). (G) Summary for experiments of the type shown in (A-F) measured between 40-60 s
after Yoda1 analogue application, expressed as a % of the Yoda1 response when pre-treated with
vehicle only (DMSO). Each data point represents a value from an independent experiment with mean
values and error bars representing S.E.M indicated in black (n=5). (H) Mean data for the type of
experiment shown in (C) with cells pre-treated with indicated concentrations of 2k. Expressed as a %
of the Yoda1 response when pre-treated with vehicle only (DMSO). The fitted curve is the Hill equation
with IC50 1.30 μM (n = 5). (I) Summary of intracellular Ca2+ measurement data (as for G) for Tet+
Piezo1 T-REx cells exposed to 2 µM Yoda1, following pre-treatment with 10 μM 2k or vehicle only
(DMSO). 2k was washed out before the recording (n=5). (J) As for (C) but conducted at or 37°C. (K)
Summary for experiments of the type shown in J (n = 5).
This article is protected by copyright. All rights reserved.

Figure 4- Selectivity of Dooku1. Ca2+ indicator dyes were fura-2 (A, B, D) or fluo-4 (C). Experiments
conducted in native HEK 293 cells (A, B), CHO cells overexpressing TRPV4 (C) or HEK 293 cells
overexpressing TRPC4 (D). Intracellular Ca2+ measurement data for cells exposed to 20 μM ATP (A),
0.3 mM Ca2+ addback (B), 5 μM 4α-phorbol 12,13-didecanoate (4α-PDD) (C) or 100 nM (-)-Englerin
A (EA) (D) following pre-treatment with DMSO or 10 μM Dooku1 (left). Error bars indicate S.E.M (N
= 3). Summary for experiments of the type shown on the left measured between 10-30s (A), 60-90s (B),
220-240s (C) or 20-60s (D) after treatment application and normalised to the peak amplitude values for
the vehicle only (DMSO) pre-treatment condition (right). Each data point represents a value from an
independent experiment with mean values and error bars representing S.E.M indicated in black (n=5).
This article is protected by copyright. All rights reserved.

Figure 5- Dooku1 does not affect Piezo1 constitutive activity (A) Intracellular Tl+ measurement data
using FluxOR for Tet+ Piezo1 T-REx cells or control Tet- cells exposed to extracellular Tl+. The
FluxOR measurements are displayed as the fluorescence intensity (F) divided by the initial fluorescence
intensity (F0). Error bars indicate S.E.M (N = 3). (B) Summary for experiments of the type shown in
(A) measured between 0-30 s after Tl+ application, normalised to rate of change of F in the Tet-
response. Each data point represents a value from an independent experiment with mean values and
error bars representing S.E.M indicated in black (n=5). (C) Intracellular Tl+ measurement data for Tet+
Piezo1 T-REx cells exposed to extracellular Tl+ and 5 μM Yoda1 or vehicle (DMSO), following pre-
treatment with 10 μM Dooku1 or vehicle only (DMSO). Error bars indicate S.E.M (N = 3). (D)
Summary for experiments of the type shown in (C), as for (B) except data are normalised to the rate of
change of the vehicle only (DMSO) control condition (n = 5).
This article is protected by copyright. All rights reserved.

Figure 6- Dooku1 is effective against endogenous Piezo1 channel. (A) Intracellular Ca2+ in HUVECs
after exposure to 10 μM Dooku1 or vehicle only (DMSO). Error bars indicate S.E.M (N = 3). (B)
Intracellular Ca2+ in HUVECs after exposure to 2 μM Yoda1 after pre-treatment with 10 μM Dooku1
or vehicle only (DMSO). Error bars indicate S.E.M (N = 3). (C) Summary for experiments of the type
shown in (B) measured 40-60s after Yoda1 application and normalised to peak amplitudes for the
vehicle only group. Each data point represents a value from an independent experiment with mean
This article is protected by copyright. All rights reserved.

values and error bars representing S.E.M indicated in black (n=7). (D) Mean data for the type of
experiment shown in (B) with cells pre-treated with indicated concentrations of Dooku1. Data are
expressed as a % of the Yoda1 response when pre-treated with vehicle only (DMSO). The fitted curve
is the Hill equation with IC50 1.30 μM (n = 5). (E,F) Mean intracellular Ca2+ for HUVECs (E) or Piezo1
T-REx cells (F) exposed to the indicated concentrations of Yoda1. The fitted curve is the Hill equation
with EC50 of 0.23 μM (E) and 2.51 μM (F) (n = 3).
This article is protected by copyright. All rights reserved.

Figure 7- Yoda1 induced relaxation in mouse thoracic aorta is endothelium and nitric oxide
dependent. (A) Isometric tension recording from aorta exposed to the indicated concentrations of
Yoda1. (B) (left) As for (A) but following pre-constriction with 0.3 μM phenylephrine (PE). (Right)
Mean data for experiments of the type shown on the left expressed as % relaxation. The fitted curve is
the Hill equation with EC50 of 2.3 μM (n = 5). (C) Isometric tension recording of aorta pre-constricted
with PE and exposed to 5 μM Yoda1 (left) or 5 μM acetylcholine control (middle and right) with the
endothelial layer removed (left and middle) or intact (right). (D) Summary data for experiments of the
type shown in (B and C, left) expressed as % relaxation evoked by Yoda1 (left) or the response to PE
(right) in the presence (EC+) or absence (EC-) of the endothelial cell layer. Each data point represents
a value from an independent experiment with mean values and error bars representing S.E.M indicated
by the black lines (n=5). (E) As for (C) but following pre-incubation with 100 µM Nω-nitro-L-arginine
methyl ester (L-NAME). (F) As for (D) but for experiments of the type shown in (E).
This article is protected by copyright. All rights reserved.

Figure 8- Dooku1 inhibits Yoda1-induced dilation in aorta. (A-K) Isometric tension data from
mouse thoracic aorta with intact endothelium. (A) Pre-constricted with PE and exposed to 5 μM Yoda1.
(B) As for A but following 30 min pre-incubation with 10 μM Dooku1. (C) Summary data for
experiments of the type shown in (A,B) expressed as % relaxation evoked by Yoda1. Each data point
represents a value from an independent experiment with mean values and error bars representing S.E.M
indicated by the black lines (n=7). (D-F) (G-I) As for (A-C) but following pre-incubation with 10 µM
2e (D-F) or 7b (G-I) (n = 5 on F, I). (J, K) As for (C) but following pre-incubation with 10 µM 2g (J)
or 11 (K) (n = 5). (L) Comparison of the mean % inhibition of Yoda1-induced relaxation in mouse
thoracic aorta and the mean % inhibition of Yoda1-induced Ca2+ entry by the 5 compounds: 2e, 2g,
Dooku1, 7b and 11. The points are fit to a straight line with Pearson’s correlation coefficient of 0.78.
This article is protected by copyright. All rights reserved.

Figure 9- Specificity of Dooku1 in aorta. All experiments were performed on mouse thoracic aorta
with intact endothelium. (A, B) Summary data for experiments of the type shown in Figure 8A-B,
expressed as the response to PE (A) or resting tension (B) before and after pre-incubation with 10 μM
Dooku1. Each data point represents a value from an independent experiment with mean values and error
bars representing S.E.M indicated by the black lines (n=7). (C) Aorta were pre-constricted with 0.1 µM
U46619 and treated consecutively with DMSO, 1 µM ACh and 10 µM SIN-1. (D) As for C but pre-
treated with Dooku1 instead of DMSO. (E-G) Summary data for experiments of the type shown in
(C,D) expressed as % of the effect of Dooku1 on the contraction by U46619 (E) or % relaxation evoked
by ACh (F) or SIN-1 (G) before and after pre-incubation with 10 μM Dooku1. Each data point
represents a value from an independent experiment with mean values and error bars representing S.E.M
indicated by the black lines (n=5).
This article is protected by copyright. All rights reserved.

Figure 10 – Lack of effect of other Yoda1 analogues on PE-induced contraction. Summary data for
experiments of the type shown in Figure 8 D-E, G-H expressed as resting tension (left) or the response
to PE (right) following pre-incubation with 10 µM 2e (A), 2g (B), 7b (C) and 11 (D). Each data point
represents a value from an independent experiment with mean values and error bars representing S.E.M
indicated by the black lines (n=5).
This article is protected by copyright. All rights reserved.