Ca²⁺ signalling in interstitial cells of Cajal contributes to generation and maintenance of tone in mouse and monkey lower esophageal sphincters

Abstract

The lower esophageal sphincter (LES) generates tone and prevents reflux of gastric contents. LES smooth muscle cells (SMCs) are relatively depolarized, facilitating activation of Ca_v1.2 channels to sustain contractile tone. We hypothesised that intramuscular interstitial cells of Cajal (ICC-IM), through activation of Ca²⁺-activated-Cl⁻ channels (ANO1), set membrane potentials of SMCs favorable for activation of Ca_v1.2 channels. In some gastrointestinal muscles, ANO1 channels in ICC-IM are activated by Ca²⁺ transients, but no studies have examined Ca²⁺ dynamics in ICC-IM within the LES. Immunohistochemistry and qPCR were used to determine expression of key proteins and genes in ICC-IM and SMCs. These studies revealed that Ano1 and its gene product, ANO1 are expressed in c-Kit⁺ cells (ICC-IM) in mouse and monkey LES clasp muscles. Ca²⁺ signaling was imaged in situ, using mice expressing GCaMP6f specifically in ICC (Kit-KI-GCaMP6f). ICC-IM exhibited spontaneous Ca²⁺ transients from multiple firing sites. Ca²⁺ transients were abolished by CPA or caffeine but were unaffected by tetracaine or nifedipine. Maintenance of Ca²⁺ transients depended on Ca²⁺ influx and store reloading, as Ca²⁺ transient frequency was reduced in Ca²⁺ free solution or by Orai antagonist. Spontaneous tone of LES muscles from mouse and monkey was reduced ~80% either by Ani9, an ANO1 antagonist or by the Ca_v1.2 channel antagonist nifedipine. Membrane hyperpolarisation occurred in the presence of Ani9. These data suggest that intracellular Ca²⁺ activates ANO1 channels in ICC-IM in the LES. Coupling of ICC-IM to SMCs drives depolarization, activation of Ca_v1.2 channels, Ca²⁺ entry and contractile tone.

Graphical Abstract

Introduction

The lower esophageal sphincter (LES) is a thickened region of smooth muscle located at the gastroesophageal junction. The LES is composed of clasp and sling muscles. In the human LES, the clasp muscle generates twice as much tone as the sling and ten times more tone than the esophageal body (Lecea et al., 2011). The LES must develop contractile tone to restrict the reflux of gastric acid into the esophagus (Dodds et al., 1982). Failure to maintain LES tone often leads to uncontrolled reflux and exposure of the esophagus to gastric acid can cause serious damage to the esophageal mucosa, leading to erosion, inflammation and eventually remodeling and hyperplasia (Smid & Blackshaw, 2000). Transient reduction in LES tone occurs as part of the swallowing reflex to facilitate the transport of food and saliva into the stomach (Hornby & Abrahams, 2000; Goyal & Chaudhury, 2008). Inability to relax the LES, as in achalasia (Rieder et al., 2020), can lead to loading of the esophagus and subsequent aspiration of undigested food particles into the respiratory tract (Boeckxstaens, 2005). The mechanisms responsible for LES tone generation have been controversial, however conctractile tone in this tissue is generally believed to occur independently of neural inputs and is thus myogenic, as it is not abolished by tetrodotoxin (TTX) (Farré & Sifrim, 2008).

From studies of isolated, unloaded smooth muscle cells (SMCs), a biochemical mechanism for regulation of LES tone was proposed that involved spontaneous activity of phosphatidylinositol-specific phospholipase C (PLC) and phosphatidylcholine specific PLC that regulated diacylglycerol and protein kinase Cβ (Harnett et al., 2005). This mechanism was based on experimental data obtained from permeabilized SMCs, and the ideas were never tested comprehensively on intact LES muscles. Patch clamp experiments on isolated SMCs from human LES clasp muscles demonstrated an electrophysiological, Ca²⁺ dependent mechanism for LES tone (Kovac et al., 2005). Tone in intact LES muscles was inhibited at least 50% by nifedipine, an L-type Ca²⁺ channel antagonist. Reduction in extracellular Ca²⁺ (and 0.5 mM EGTA buffering) also inhibited tone over several minutes. Inward, voltage-dependent current was activated in LES SMCs by voltage clamping from a negative holding potential to −40 mV, and the amplitude of this current peaked at 0 mV. The current was doubled in magnitude by Bay K8644, an L-type Ca²⁺ channel agonist, and blocked by nifedipine. The properties and pharmacology of the conductance responsible for the inward current suggested that it was dependent upon L-Type Ca²⁺ channels, and splice variants of Ca_v1.2 were shown to be expressed in human LES and esophageal body muscles (Kovac et al., 2005). Others showed that systemic administration of nifedipine decreases LES pressure in healthy patient volunteers (Hongo et al., 1984; Castell, 1985), and dihydropyridines (inhibitors of L-type Ca²⁺ channels) reduce LES tone in achalasia patients (Traube et al., 1984; Castell, 1985).

Low-level activation (i.e. low, but resolvable open probability) of L-type Ca²⁺ channels is facilitated by the relatively depolarized membrane potentials of LES SMCs, which fall within the window current range for this conductance (Zhang & Paterson, 2003; Zhang et al., 2010). The reason why LES SMCs are so depolarized has not been determined, however there is evidence that a Cl⁻ conductance contributes to setting membrane potentials in the LES. Niflumic acid, 9-anthracenecarboxylic acid (9-AC) (Zhang et al., 2000; Zhang & Paterson, 2003) and 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS) (Saha et al., 1992) hyperpolarize membrane potentials of LES SMCs by several mV and reduce LES tone. However, these drugs are not specific Cl⁻ channel antagonists. Reduced extracellular Cl⁻, which may decrease the gradient for Cl⁻ efflux from cells (Zhu et al., 2016), also inhibited LES tone (Saha et al., 1992). Pharmacological agents that disrupt intracellular Ca²⁺ handling, such as caffeine and cyclopiazonic acid (CPA), hyperpolarize LES muscles, suggesting that the Cl⁻ conductance involved in generating tone could be Ca²⁺-dependent (Zhang & Paterson, 2003). However, evidence for Ca²⁺ activated Cl⁻ conductances in gastrointestinal (GI) SMCs is lacking. In most regions of the GI tract, a Ca²⁺-activated Cl⁻ conductance, encoded by Ano1, is prominently expressed in interstitial cells of Cajal (ICC), but not resolved in SMCs (Gomez-Pinilla et al., 2009; Hwang et al., 2009; Cobine et al., 2017). Membrane potential was hyperpolarized by about 6 mV in LES of W/W^V mice with markedly decreased numbers of ICC (Zhang et al., 2010; Iino et al., 2011). Thus, it is possible that ICC-IM, via activation of ANO1, facilitate the development and maintenance of tone of LES muscles.

In the present study we investigated the hypothesis that spontaneous Ca²⁺ transients in intramuscular ICC (ICC-IM) in the LES, the only class of ICC found in LES (Ward et al., 1998; Iino et al., 2011; Blair et al., 2012), cause inward currents mediated by ANO1 and depolarization. Depolarizing currents generated in ICC would be conducted to electrically coupled SMCs, driving membrane potentials into a range favorable for openings of L-type Ca²⁺ channels. Ca²⁺ entry in SMCs leads to excitation-contraction (E-C) coupling and development and maintenance of tone. Thus, we investigated Ca²⁺ handling mechanisms in ICC-IM of the LES using mice expressing the genetically encoded Ca²⁺ indicator, GCaMP6f, in a cell-specific manner. We also investigated whether the ANO1 conductance expressed in ICC-IM is linked to regulation of membrane potential in mouse LES and maintenance of tone in LES muscles of mice and non-human primates.

Methods

Ethical Approval

All animals used and the protocols performed throughout this study were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All procedures were approved by the Institutional Animal Use and Care Committee at the University of Nevada, Reno (IACUC protocol numbers 20-09-1073-1 and 21-02-1131). All experiments conform to the principles and regulations for reporting animal experiments as described by (Grundy, 2015).

Animals

C57/BL6 (wild-type) mice of both sexes were purchased from the Jackson Laboratory (Bar Harbor, MN, USA) and used for experiments at 6-12 weeks of age. Mice of both sexes expressing GCaMP6f selectively in ICC (Kit-KI-GCaMP6f; (Jin et al., 2021)) and expressing a fluorescent reporter (Kit^+/copGFP mice (Ro et al., 2010)) in ICC were generated and bred in house. Mice expressing a fluorescent reporter driven by the smooth muscle myosin heavy chain promoter (SmHC^{+ /eGFP} mice) were donated by Dr. Michael Kotlikoff of Cornell University (NY, USA). Mice were anaesthetized by isoflurane inhalation (5% in oxygen; Baxter, Deerfield, IL, USA) and killed by cervical dislocation.

Cynomolgus monkeys were housed and maintained at Charles River Laboratories, Preclinical Services, Reno, NV. The Institutional Animal Care and Use Committee (IACUC approval # I-000536, Charles River Laboratories) ensured compliance with the United States Department of Agriculture, Public Health Service Office of Laboratory Animal Welfare Policy and the Animal Welfare act. Cynomolgus monkeys of either sex (2-8 years of age) were sedated with ketamine HCL (10 mg/kg given intramuscularly; Zoetis). Then pentobarbital (Fatal-Plus; Vortech Pharmaceuticals, Ltd.) was given IV for euthanasia, and this was followed by exsanguination. The dosage for pentobarbital was 0–4.5 kg – 1.0 mL, 4.6–9.1 kg, 2.0 mL and >9.2 kg – 3.0 mL. Tissue samples from these animals were provided immediately after necropsy to our laboratory in the Department of Physiology and Cell Biology at the University of Nevada, Reno. No non-human primates were purchased, bred or killed specifically for the present study. All of the non-human primate tissues used in the present study came from animals that had been used previously for pharmaceutical testing.

Tissue preparation

Mice:

After euthanasia, the thoracic and abdominal cavities of mice were opened with a longitudinal incision and the upper gastrointestinal tract (distal esophagus to duodenum) was removed and placed in cold Krebs-Ringer-Bicarbonate (KRB) solution. The esophagus and stomach were opened by sharp dissection, cutting along the lesser curvature of the stomach and gastric contents were flushed away with ice-cold KRB solution. The gastroesophageal junction was visible at the distal esophagus where the esophagus transitioned from skeletal muscle to smooth muscle (see Fig. 1Aa and inset). Tissues were pinned flat in a Sylgard-bottomed glass dissecting dish filled with either ice-cold Krebs-Ringer-Bicarbonate (KRB) solution or ice-cold Ca²⁺ free Hanks’ solution (for cell isolation and FACS studies) and the mucosa and submucosa were removed. The thickened band of LES clasp muscle was then dissected free from the surrounding muscle of the distal esophagus and gastric fundus and antrum to generate a muscle strip (see Fig. 1Ab) .

Fig. 1: Mouse LES morphology and ICC within LES and esophagus.

A Anatomy of the mouse upper gastrointestinal tract. Sagittal section of the mouse upper gastrointestinal tract (from esophagus to duodenum) stained with a trichrome stain to identify SMCs (pink) and connective tissue (blue) (a). Additional structures staining pink include the mucosa. Inset (right) shows a higher magnification image of the LES clasp muscle. Muscular tissues in the mouse esophageal body are composed of skeletal muscle fibers whereas the muscular tissues of the LES, stomach, pyloric sphincter (PS) and duodenum are smooth muscles. The LES clasp is indicated by the dashed green line and is divided into bundles separated by connective tissue septa. An image of a mouse LES dissection indicating the portion used for functional experiments (b). The skeletal muscle of the esophagus was removed (indicated by dashed red line and scissors symbol) and the LES clasp muscles were used for contractile, microelectrode, calcium imaging and immunohistochemical experiments performed in this study. B Immunohistochemical labelling of ICC using a c-Kit antibody (Red: ICC biomarker) and an ANO1 antibody (Green) in mouse (a) and monkey LES clasp (b) and the body of the monkey esophagus (c). Spindle-shaped ICC-IM were distributed through the LES of both animals and these cells appeared to be shorter and of a lower density in the monkey esophagus. c-Kit and ANO1 co-localized in ICC in the LES and esophagus (Images of c-Kit and ANO1 merged), demonstrating that ICC-IM express ANO1 protein in esophageal tissues.

Cynomolgus monkeys:

After euthanasia, the necropsy team at Charles River Laboratories removed the upper gastrointestinal tract (proximal esophagus to duodenum) and tissues were placed in ice cold KRB solution and transported on ice to the University of Nevada, Reno. The esophagus and stomach were opened by sharp dissection, cutting along the lesser curvature of the stomach and gastric contents were flushed away with ice-cold KRB solution. The tissues were pinned flat in a dissecting dish containing ice-cold KRB solution and the mucosa and submucosa removed. The LES clasp muscle was identified as a distinct thickening of the circular muscle layer at the gastroesophageal junction and was dissected free from the surrounding muscle of the distal esophagus and gastric fundus and cut into muscle strips.

Histology

Histological examination of the mouse upper gastrointestinal tract (esophagus to duodenum) was performed to determine sphinter localization and structure. Tissues were removed, pinned in a Sylgard bottomed dish and fixed overnight at 4°C in 4% paraformaldehyde. After fixation, tissues were washed in 0.01 M PBS, then dehydrated and cleared with a Leica ASP6025 tissue processor using a 4-hour rapid cycle. After processing, specimens were embedded in paraffin wax before cutting into 3–4 μm thick sections and placing on glass slides. Paraffin wax was removed with xylene and rehydration of sections was achieved by immersion in graded alcohol solutions followed by water. Bouin’s fluid was added for 1 h at 56°C to fix the dye before washing with running water for 3 min. Slides were placed in modified Mayer’s haematoxylin for 7 min and then washed with running tap water for 3 min before placing in a one-step trichrome stain (blue & red) for 8 min, then washed with running tap water for 5 min. Preparations were then treated with 1% acetic acid for 5-10 sec. The sections were dehydrated by rinsing three times with absolute alcohol for 1 min, followed by immersion in xylene 3 x 1 min before covering with a coverslip using permanent mounting media. All solutions were received from American MasterTech (Lodi, CA, USA). Slides were imaged using Brightfield microscopy on an Echo Revolve microscope (San Diego, CA) at 4x magnification. Resulting images were overlain and stitched using Adobe Photoshop CC 2019 software. Final figures with labeling were created with CorelDraw 2020 software.

Immunohistochemistry

Immunohistochemical experiments were performed on whole mouse LES clasp tissues and on 100 μm thick tissue sections of monkey LES clasp muscles. Monkey LES tissue sections were cut parallel to the circular muscle layer with a Leica VT1200S vibratome (Wetzlar, Germany), to facilitate antibody penetration. Mouse LES tissues and monkey LES tissue sections were pinned in a Sylgard bottomed dish and fixed for 15 min with cold Zamboni’s fixative (2% paraformaldehyde). LES preparations were then washed in 0.01 M PBS for 4 h prior to blocking with 1% bovine serum albumin (BSA) for 1 h at 20°C to prevent non-specific binding. LES tissues were incubated in the first primary antibody (mSCFR, anti-Kit antibody, R&D Systems, Minneapolis, MN, USA; 1:1000 dilution) with a 0.5% Triton-X working solution for 48 h at 4°C. Then the tissues were washed in 0.01 M PBS for 4 h and incubated in secondary antibody (Alexa Fluor anti-goat 594, Invitrogen, Carlsbad, CA, USA; 1:1000 dilution) for 1 h at 20°C in the dark followed by washing in 0.01 M PBS overnight. Then the tissues were incubated in the second primary antibody (anti-ANO1 antibody, Abcam, Cambridge, MA, USA; 1:500 dilution), as described above. Following incubation and subsequent washing in 0.01 M PBS, the tissues were incubated in secondary antibody (Alexa Fluor anti-rabbit 488, Invitrogen; 1:1000 dilution) as described above. After washing overnight in 0.01 M PBS, the tissues were mounted on glass slides and covered with coverslips using Aquamount mounting medium (Lerner Laboratories, Pittsburgh, PA, USA). The mounted tissues were imaged with a Zeiss LSM510 confocal microscope (Carl Zeiss, Thornwood, NY, USA). The images generated are digital composites of Z-series of scans 0.5–1 μm optical sections. The construction of final images was completed using Zeiss LSM 5 Image Examiner software, Adobe Photoshop CC 2019 software and CorelDRAW 2020 software.

Cell isolation and Fluorescence Activated Cell Sorting (FACS)

LES muscles from Kit^+/copGFP mice or SmMHC^+/eGFP mice were dissected in ice cold Ca²⁺ free Hanks’ solution, as described above. LES strips were cut into three 1 mm x 1 mm segments, placed in a glass tube and incubated in Ca²⁺ free Hanks’ solution at 4°C for 30 mins. The muscle pieces were transferred to an enzyme cocktail containing the following per 1 mL of Ca²⁺ free Hanks’ solution: 4 mg collagenase type II (Worthington Biochemical Corporation, Lakewood, NJ, USA), 8 mg bovine serum albumin (BSA), 8 mg trypsin inhibitor, 100 nM dithiothreitol (DTT) (all Sigma-Aldrich, St. Louis, MO, USA) and incubated at 37°C for 20 mins. Tissues were titurated using a wide gauge glass pipette and then washed three times in Ca²⁺ free Hanks’ solution to remove all remaining enzyme solution. After washing, the tissues were placed into a 37°C water bath for 10 mins and again titurated using glass pipettes with decreasing gauges until cells were dispersed.

FACS was carried out using a Becton-Dickinson FACSAria II cell sorter with an excitation laser (488 nm) and emission filter (530/30 nm). Cells were passed through a 130 μm nozzle at a pressure of 12 psi and rate of 300 events/s. Events were first gated according to viability (exclusion of Hoechst 33258), viable cells were then gated by FITC fluorescence intensity and sorted into appropriate tubes to capture both GFP⁺ and GFP⁻ cells. For the purposes of comparisons, an aliquot of cells were removed from each sample prior to cell sorting.

RNA extraction and quantitative PCR

Total RNA was isolated from purified Kit^+/copGFP, SmMHC^{+ /eGFP} and corresponding GFP⁻ cells using a Direct-zol microprep RNA Kit (Zymo Research. Irvine, CA). First-strand cDNA was synthesized using qScript cDNA supermix (Quantabio, Beverley, MA) according to the manufacturer’s instructions. The PCR primers used and their GenBank accession numbers are listed in Table 1. Using GoTaq Green Master Mix (Promega, Madison, WI), PCR products were analyzed on 2% agarose gels and visualized by ethidium bromide. Quantitative PCR (qPCR) was performed with the same primers as PCR using SYBR green chemistry on the QuantsStudio 3 Real-Time PCR System (Applied Biosystems, Waltham, MA) and analyzed as previously described (Cobine et al., 2014; Cobine et al., 2017; Drumm et al., 2019b). qPCR analysis represents experiments performed in technical triplicates on substrate pooled from multiple animals (Kit^+/copGFP: N=10, N=6, N=6; SmMHC^{+ /eGFP}: N=2, N=8, N=10). Gene expression was normalized as the level of expression relative to the expression of the house keeping gene Gadph, mean ± standard deviation (SD).

Table 1:

List of gene primers and ascension numbers used in this study.

Gene Name	Sequence (Sense primer on top)		Accession #
Gapdh-F	GCCGATGCCCCCATGTTTGTGA		NM_008084
Gapdh-R	GGGTGGCAGTGATGGCATGGAC
Pdgfra-F	ATGACAGCAGGCAGGGCTTCAACG		NM_011058
Pdgfra-R	CGGCACAGGTCACCACGATCGTTT
Ano1-F	TAACCCTGCCACCGTCTTCT		NM_178642
Ano1-R	ATGATCCTTGACAGCTTCCTCC
Kit-F	CGCCTGCCGAAATGTATGACG		NM_021099
Kit-R	GGTTCTCTGGGTTGGGGTTGC
Myh11-F	CCCAAGCAGCTAAAGGACAA		NM_013607
Myh11-R	AGGCACTTGCATTGTAGTCC
Cacna1c-F	GTAAGGATGAGTGAAGAAGCCGAGTAC		NM_009781
Cacna1c-R	CAGAGCGAAGGAAACTCCTCTTTGG
Orai1-F	GTTCACTTCTACCGCTCCCT		NM_175423
Orai1-R	GTGCCCGGTGTTAGAGAATG
Orai2-F	CACAAGGGCATGGATTACCG		NM_178751
Orai2-R	CCCTGCTCAGGTAGAGCTTC
Orai3-F	GGCTGAAGTTGTTCTGGTGG		NM_198424
Orai3-R	TGGAAGGCTGTTGTGATGTG
Stim1-F	GAGTCTACCGAAGCAGAGTT		NM_009287
Stim1-R	TTCTTCCACATCCACATCACC
Stim2-F	TCCCTGTATGTCGCTGAGTC		NM_001081103
Stim2-R	ACTCTCGTCCACTTCGATCC
Itpr1-F	TTATCAGCACCTTAGGCTTGGTTGA		NM_010585
Itpr1-R	ATCTGTAGTGCTGTTGGCCCCG
Itpr2-F	CAACCCAGGCTGCAAAGAGGTGA		NM_019923
Itpr2-R	AGGTCGTCCGAAGGAAAATGTGCT
Itpr3-F	CTTTGGGGCTGGTGGATGACCGTTG		NM_080553
Itpr3-R	TGCAGCTTCTGCAGCAATACCACA
Ryr1-F	GTCAGTTCGAGCCCTGCAGGAG		NM_009109
Ryr1-R	GCAACTCAGGTACATACGACTGTGT
Ryr2-F	TCCCCCGGACCTGTCTATCTGC		NM_023868
Ryr2-R	GGCCTCCACCTTGAGCAGTCTTC
Ryr3-F	TCCTCGTCAGTGTGTCCTCTGAAA		NM_177652
Ryr3-R	CATGGCCACCGAGTAAGTATCCTTC

Calcium imaging

LES tissues were pinned to the bottom of a 60 mm dish coated with Sylgard elastomer (Dow Corning, Midland, MI) and perfused with oxygenated KRB solution warmed to 37°C for 1 h prior to the beginning of experiments. Following the equilibration period, Ca²⁺ imaging was performed on LES-ICC in situ with an Eclipse E600FN microscope (Nikon Inc., Melville, NY, USA) with a 60x 1.0 CFI Fluor lens (Nikon instruments INC, NY, USA). GCaMP6f was excited at 488 nm (T.I.L.L. Polychrome IV, Grafelfing, Germany). The pixel size using this acquisition configuration was 0.225 μm. Image sequences were collected at 33 fps with TILLvisION software (T.I.L.L. Photonics GmbH, Grafelfing, Germany). Movement artefacts were stabilized digitally with custom made software (Volumetry: Dr. Grant Hennig) prior to analysis of Ca²⁺ transients. Video sequences were collected for 20-30 sec. Pharmacological agents were added directly to the KRB solution and were perfused into the bath for 12-15 mins before images of responses were collected.

Analysis of Ca²⁺ transients

Ca²⁺ transients were imaged and analyzed in LES-ICC in situ as described previously (Drumm et al., 2019a). Briefly, movies of Ca²⁺ events were converted to a stack of TIFF (tagged image file format) images and imported into custom software (Volumetry G8c, GW Hennig) for initial processing. Whole-cell ROIs were created to generate spatio-temporal maps (STMs) of Ca²⁺ transients in individual LES-ICC within a field of view (FOV). STMs were imported as TIFF files into Image J (version1.52a, National Institutes of Health, MD, USA, http://rsbweb.nih.gov/ij) for post hoc analysis. Basal fluorescence was acquired from regions of cells that displayed the most uniform and least intense fluorescence (F₀). Then fluorescence values throughout the rest of the cell were divided by the F₀ value to calibrate the STM for the amplitudes of Ca²⁺ transients as F/F₀. Ca²⁺ event amplitude, duration and spread were then calculated from the STM. Ca²⁺ transient frequency was expressed as the number of events fired per cell per minute (min⁻¹). The amplitude of Ca²⁺ transients was expressed as ΔF/F₀, the duration of Ca²⁺ transients was expressed as full duration at half maximum amplitude (FDHM) and the spatial spread of Ca²⁺ transients was expressed as μm of cell propagated per Ca²⁺ transient.

Contractile recordings

LES clasp muscle strips (mouse: ~3-4 mm long and ~1 mm wide (whole LES clasp); monkey: 9 mm long and 3 mm wide) were attached to a stable mount and to a Gould strain gauge in the circular muscle direction and immersed in heated water-jacketed 7.0 ml tissue baths. The LES muscles were maintained in oxygenated (95% O₂/5% CO₂) KRB solution at 37°C. Muscles were stretched to an initial tension of 0.5 g (mouse) or 1 g (monkey) and equilibrated until peak tone was observed, i.e., 30-60 mins. To eliminate influences of neurotransmitter release, all experiments were carried out in the presence of atropine (1 μM), MRS2500 (1 μM), L-NNA (100 μM), tetrodotoxin (TTX; 1 μM) and guanethidine (1 μM) unless otherwise stated. Addition of ANO1 antagonist or L-type Ca²⁺ channel antagonists occurred ~10-15 minutes after the development of spontaneous tone reached a peak and stabilized. KCl contractions were repeated at least three times to ensure responses were reproducible before testing the effects of the ANO1 antagonist on these contractions. Contractile activity was recorded and analysed using AcqKnowledge software (3.9.1; Biopac Systems, Goleta, CA). The integral (area) of contraction after drugs were administered was measured and normalized to the integral during control activity. Drug-induced changes in contractile activity were then expressed as the percentage change from control activity. In all experiments, an absolute zero baseline was determined by the level of tone after adding a combination of nifedipine (1 μM) and sodium nitroprusside (SNP; 10 μM) or in the absence of extracellular Ca²⁺, as previously described (Cobine et al., 2017).

Electrophysiological recordings

LES muscles (see Fig. 1A) were isolated from C57BL/6 mice (2-3 months old). The final LES muscle strips used for electrophysiological experiments were 8 x 2mm. The muscle strips were pinned down in a recording chamber lined with Sylgard elastomer 184 (Dow, USA), and incubated at 37 ± 0.5°C with continuously flowing, oxygenated KRB solution for 1hr before beginning intracellular recordings. Cells in the LES were impaled with glass microelectrodes filled with 3 M KCl and having resistance 50-100 MΩ. Transmembrane potentials were measured using a high input impedance amplifier (Axopatch 2B, Molecular Devices Corp., Sunnyvale, CA, USA) and recorded with Axoscope 10.3 software. The data were analyzed by Clampfit 10.4 (Molecular Devices). All recordings were made in the presence of nicardipine (100 nM) to reduce movements and extend the durations of impalements.

LES muscles displayed continuous random oscillations of resting membrane potentials (RMP). Therefore, membrane potentials (referred to in this paper as membrane potential oscillations; MPOs) were analyzed by generating amplitude histograms from 1 min recordings using Clampfit 10.4 software (Molecular Devices). The median values and standard errors (SE) were calculated by a Gaussian Function for RMP and MPOs, respectively.

Drugs and solutions

Tissues used for imaging and electrophysiological experiments were perfused with KRB solution containing (mM): NaCl, 120.35; KCl, 5.9; NaHCO₃, 15.5; NaH₂PO₄, 1.2; MgCl₂, 1.2; CaCl₂, 2.5; and glucose, 11.5. KRB solution was oxygenated with a mixture of 97% O₂/3% CO₂ and warmed to 37°C. For experiments using 0 mM [Ca²⁺]_o, CaCl₂ was excluded from the KRB solution and 0.5 mM EGTA was added, resulting in a Ca²⁺ free KRB solution. Ca²⁺ free Hanks’ solution contained the following (mM): NaCl, 125; KCl, 5.36; Glucose, 10; Sucrose, 2.9; NaHCO₃, 15.5; KH₂PO₄, 0.44; Na₂HPO₄, 0.33; Hepes, 10. The pH was adjusted to 7.4 with NaOH.

Nifedipine, nicardipine, atropine, guanethidine, L-NNA, CPA, tetracaine, caffeine, EGTA, and GSK 7975A were purchased from Sigma-Aldrich (St. Louis, MO, USA). Ani9 and MRS2500 was purchased from Tocris Bioscience (Ellisville, Missouri, USA). TTX was purchased from Abcam (Cambridge, MA, USA). Atropine, guanethidine, L-NNA, MRS2500, TTX, CPA, tetracaine and caffeine were diluted in deionized water. Nifedipine and nicardipine were diluted in ethanol. Ani9 was diluted in DMSO. All stock solutions of drugs were diluted to the desired final concentrations with KRB solution.

Statistics

Unless otherwise stated in the text, data is represented as mean ± standard deviation (S.D.). Statistical analysis was performed using student’s t-tests, Welsh’s t-test or a one-way or two-way ANOVA with a Tukey post hoc test as appropriate. In all statistical analyses, P<0.05 was taken as significant. When describing data throughout the text, “N” refers to the number of animals used in a data set, “c” refers to the numbers of cells analysed in Ca²⁺ imaging experiments, and “n” refers to the number of tissues or for qPCR, the number of experiments.

Results

The mouse and monkey LES contain ICC-IM, as determined by c-Kit immunolabelling (Ward et al., 1998; Iino et al., 2011; Blair et al., 2012). ICC in other regions of the GI tract can be distinguished from SMC not only by their expression of c-Kit, but also by expression of ANO1 (Gomez-Pinilla et al., 2009; Cobine et al., 2017). However, co-expression of ANO1 in ICC-IM of the LES is less well defined. Double-immunolabeling was performed on whole mounts of LES clasp muscles from wild-type (C57Bl/6) mice and Cynomolgus monkeys. Cells immunopositive for c-Kit⁺ were spindle-shaped and distributed throughout mouse LES clasp muscles, running in parallel with SMCs (Fig. 1Ba). All c-Kit⁺ cells exhibited ANO1 immunoreactivity, and no ANO1⁺ cells were c-Kit⁻. Monkey LES clasp and esophageal body muscles also exhibited c-Kit⁺ spindle-shaped cells (Fig. 1Bb & Bc), though the cells in the esophageal body were shorter in length than cells in the LES. As in mice, c-Kit⁺ and ANO1⁺ co-localization was present in monkey muscles.

Further confirmation of the expression of ANO1 in ICC-IM in the LES was obtained by evaluating cell-specific gene expression. ICC and SMCs were dispersed from mice expressing fluorescent, cell-specific reporters (Kit^+/copGFP and SmHC^+/eGFP mice, respectively) and sorted by FACS. ICC expressed abundant gene transcripts for Ano1 and Kit, as compared to the copGFP⁻ cells sorted from these mice (e.g., Fig. 2A, Ano1; expression relative to Gapdh, ICC = 0.908 ± 0.069 vs copGFP⁻ cells = 0.014 ± 0.002 (two-way ANOVA with Tukey’s multiple comparison, P=0.0048), n=3; Kit; expression relative to Gapdh, ICC = 0.847 ± 0.039 vs copGFP⁻ cells = 0.02 ± 0.001 (two-way ANOVA with Tukey’s multiple comparison, P=0.0004), n=3). ICC exhibited low expression of other cellular markers, such as Myh11 (encodes myosin heavy chain and dominant in SMCs; expression relative to Gapdh, n=3; ICC = 0.301 ± 0.005 vs copGFP⁻ cells = 0.605 ± 0.018 (two-way ANOVA with Tukey’s multiple comparison, P=0.0003) and Pdgfra (encodes platelet-derived-growth factor alpha which is dominant in PDGFRα⁺ interstitial cells (Kurahashi et al., 2011; Kurahashi et al., 2012) (Fig. 2A, Pdgfra; expression relative to Gapdh, n=3; ICC = 0.013 ± 0.001 vs copGFP⁻ cells = 0.425 ± 0.05 (two-way ANOVA with Tukey’s multiple comparison, P=0.0131). In comparison to ICC, SMCs exhibited significantly lower expression of Ano1 and Kit transcripts but more abundant expression of Myh11 (Fig. 2A; Ano1; expression relative to Gapdh, SMCs= 0.014 ± 0.00006 vs ICC = 0.908 ± 0.069 (two-way ANOVA with Tukey’s multiple comparison, P=0.0048), n=3; Kit; expression relative to Gapdh, SMCs = 0.017 ± 0.00064 vs ICC = 0.847 ± 0.039 (two-way ANOVA with Tukey’s multiple comparison, P=0.0005), n=3; Myh11; expression relative to Gapdh, SMCs = 2.165 ± 0.18406 vs ICC = 0.301 ± 0.005 (two-way ANOVA with Tukey’s multiple comparison, P=0.0081), n=3).

Fig. 2: Cell identification and Ca²⁺ handling gene expression in LES ICC-IM.

A qPCR data of relative expression of gene transcripts for Ano1, Kit, Myh11, Pdgfra, Gja1 and Gja7 in FACS sorted GFP⁺ populations of LES-ICC (from Kit^+/copGFP mice) and LES-SMC (from SmHC^{+ /eGFP} mice) vs. sorted GFP⁻ cells from Kit^+/copGFP mice and GFP⁻ cells from SmHC^{+ /eGFP} mice (expression relative to Gapdh), n=3. B qPCR data of relative expression of gene transcripts for Itpr1, Itpr2, Itpr3, Ryr1, Ryr2, Ryr3, Orai1, Orai2, Orai3, Stim1 and Stim2 in FACS sorted GFP⁺ populations of LES-ICC (from Kit^+/copGFP mice) and LES-SMC (from SmHC^{+ /eGFP} mice) vs. sorted GFP⁻ cells from Kit^+/copGFP mice and GFP⁻ cells from SmHC^{+ /eGFP} mice (expression relative to Gapdh), n=3.

Activation of ANO1 channels in ICC-IM would result in Cl⁻ efflux (defined electrophysiological convention as inward current) and depolarization of ICC-IM. Conduction of inward current generated in ICC-IM to SMCs would increase muscle excitability by depolarizing cells into the activation range of voltage-dependent Ca²⁺ channels. For this to occur, ICC-IM must be electrically coupled to SMCs, as occurs in the canine esophagus (Daniel & Posey-Daniel, 1984) and in other regions of the GI tract (Mitsui & Komuro, 2003). Therefore, we also characterized expression of gene transcripts encoding gap junction proteins (i.e., connexins 43 & 45 encoded by Gja1 and Gja7 respectively, these are the two most common gap junction genes found in SMC and ICC (Chen et al., 2007; Lee et al., 2015; Lee et al., 2017)). ICC exhibited enriched expression for both Gja1 and Gja7 in comparison to SMCs (Fig. 2A; Gja1; expression relative to Gapdh, ICC = 0.315 ± 0.047 vs SMCs = 0.047 ± 0.0016 (two-way ANOVA with Tukey’s multiple comparison, P=0.027), n=3; Gja7; expression relative to Gapdh, ICC = 0.097 ± 0.012 vs SMCs = 0.05 ± 0.00399 (two-way ANOVA with Tukey’s multiple comparison, P=0.0376), n=3). The most abundant gap junction transcript in ICC was Gja1 (Fig. 2A; Gja1: ICC = 0.3151 ± 0.04726 vs Gja7: ICC = 0.09736 ± 0.01194; expression relative to Gapdh, Welsh’s t-test, P=0.0115, n=3).

Activation of ANO1 channels would necessitate intracellular Ca²⁺ signalling mechanisms in ICC-IM, as observed in ICC-IM in other GI muscles (Baker et al., 2016; Drumm et al., 2019b; Drumm et al., 2019c; Hannigan et al., 2020). In other regions of the GI tract, ICC Ca²⁺ signalling relies on Ca²⁺ release from endoplasmic reticulum (ER) stores (Baker et al., 2016; Drumm et al., 2017; Drumm et al., 2019b; Hannigan et al., 2020; Baker et al., 2021b). We therefore examined gene transcripts for proteins involved in ER Ca²⁺ handling such as genes for inositol-triphosphate receptors (IP₃Rs) and ryanodine receptors (RyRs), both of which mediate Ca²⁺ release from the ER. Analysis of gene transcripts by qPCR indicated low to no expression of Ryr1 or Ryr3 in ICC whereas significant Ryr2 expression was detected in these cells (Fig. 2B; expression relative to Gapdh, n=3; Ryr1; ICC = 0.00063 ± 0.0000264, Ryr2; ICC = 0.0163 ± 0.0010, Ryr3; ICC = 0.00 ± 0.00 (one-way ANOVA with Tukey’s multiple comparison, Ryr1 vs Ryr2 P=<0.0001; Ryr1 vs Ryr3 P= 0.408; Ryr2 vs Ryr3 P=<0.0001). In contrast, Ryr2 and Ryr3 were enriched in SMCs, as compared to either ICC or GFP⁻ cells, and were more highly expressed than Ryr1 (Fig. 2B; expression relative to Gapdh, n=3; Ryr1; ICC= 0.00063 ± 0.000026 vs SMCs = 0.005 ± 0.00013 (two-way ANOVA with Tukey’s multiple comparison, P=<0.0001); SMCs = 0.005 ± 0.00013 vs eGFP⁻ cells = 0.008 ± 0.0007 (two-way ANOVA with Tukey’s multiple comparison, P=0.0671); Ryr2; ICC = 0.0163 ± 0.001 vs SMCs = 0.0262 ± 0.0026 (two-way ANOVA with Tukey’s multiple comparison, P=0.0838); SMCs =0.0261 ± 0.0026 vs eGFP⁻ cells = 0.006 ± 0.0004 (two-way ANOVA with Tukey’s multiple comparison, P=0.0173), Ryr3; ICC = 0.00 ± 0.00 vs SMCs = 0.03 ± 0.002 (two-way ANOVA with Tukey’s multiple comparison, P=0.0019); SMCs = 0.03 ± 0.002 vs eGFP⁻ cells = 0.006 ± 0.0008 (two-way ANOVA with Tukey’s multiple comparison, P=0.0104). One-way ANOVA with Tukey’s multiple comparison, Ryr1 vs Ryr2 P=<0.0001; Ryr1 vs Ryr3 P=<0.0001; Ryr2 vs Ryr3 P=0.7411). Expression of Itpr1 (that encodes IP₃R1) was also high in ICC (Fig. 2Bb; expression relative to Gapdh, n=3, Itpr1; ICC = 0.761 ± 0.041; Itrp2; ICC = 0.076 ± 0.007; Itpr3; ICC = 0.043 ± 0.001, (one-way ANOVA with Tukey’s multiple comparison, Itpr1 vs Itpr2; P=<0.0001; Itpr1 vs Itpr3; P=<0.0001; Itpr2 vs Itpr3; P=0.2847). Transcripts for Itpr1 were ~10 times greater in sorted ICC than in copGFP⁻ cells (Itrp1; ICC= 0.761± 0.041 vs copGFP⁻ cells = 0.072 ± 0.012 (two-way ANOVA with Tukey’s multiple comparison, P=0.0048), expression relative to Gapdh, n=3).

We also examined expression of gene transcripts for the ER membrane bound stromal interacting molecule (STIM), which interacts with plasma membrane bound Orai Ca²⁺ influx channels to mediate store-operated-Ca²⁺-entry (SOCE) (Liou et al., 2005; Roos et al., 2005; Feske et al., 2006; Peinelt et al., 2006; Prakriya et al., 2006; Vig et al., 2006; Zhang et al., 2006). This mechanism has also been shown to be important in ICC (Zheng et al., 2018; Drumm et al., 2019b; Drumm et al., 2020a; Drumm et al., 2020b). Stim2 was enriched in ICC as compared to copGFP⁻ cells from these cell sorts or sorted SMCs (Fig. 2B; Stim2; ICC = 0.209 ± 0.016 vs copGFP⁻ cells = 0.025 ± 0.002 (two-way ANOVA with Tukey’s multiple comparison, P=0.0061); Stim2; ICC = 0.209 ± 0.016 vs SMCs = 0.026 ± 0.0027 (two-way ANOVA with Tukey’s multiple comparison, P=0.0087) expression relative to Gapdh, n=3). In contrast, Stim1 was enriched in SMCs as compared to eGFP⁻ cells from these cell sorts or sorted ICC (Fig. 2B; Stim1; SMCs = 0.156 ± 0.0016 vs eGFP⁻ cells = 0.09 ± 0.007 (two-way ANOVA with Tukey’s multiple comparison, P=0.0087); Stim1; SMCs = 0.156 ± 0.0016 vs ICC = 0.044 ± 0.0021 (two-way ANOVA with Tukey’s multiple comparison, P=<0.0001). Transcripts for Orai1 were enriched in SMC compared to ICC (Fig. 2Ba, Orai1; expression relative to Gapdh, n=3, SMC = 0.0545 ± 0.005 vs ICC = 0.0003 ± 0.00002 (two-way ANOVA with Tukey’s multiple comparison, P=0.0073)). Expression of transcripts for Orai2 and Orai3 were resolved in ICC, although these were not enriched compared to SMC (Fig. 2Ba; Orai2; ICC = 0.0082 ± 0.0001 vs SMCs = 0.008 ± 0.001 (two-way ANOVA with Tukey’s multiple comparison, P=0.9858); Orai3; ICC = 0.0075 ± 0.0002 vs SMCs = 0.071 ± 0.0062 (two-way ANOVA with Tukey’s multiple comparison, P=0.0088), expression relative to Gapdh, n=3).

Ca²⁺ events in ICC-IM were imaged in intact LES clasp muscles from Kit-KI-GCaMP6f mice (Baker et al., 2021a; Jin et al., 2021) to determine the characteristics of Ca²⁺ entry and release in these cells. Imaging was performed at 33 frames per second (FPS) with a 60x objective to visualize spontaneous Ca²⁺ transients ICC-IM in situ. Ca²⁺ transients occurred spontaneously and arose from discrete intracellular firing sites along the lengths of ICC-IM (Fig. 3A). Spatio-temporal maps (STM) allowed mapping of intracellular firing sites (Fig. 3B). The firing of Ca²⁺ transients in ICC-IM of LES muscles from Kit-KI-GCaMP6f mice was stochastic in nature, with no evidence of entrainment between firing sites, even when firing sites were spaced only a few microns apart in the same cell. This is illustrated in the STM in Fig. 3B, which shows that spontaneous Ca²⁺ transients in this ICC-IM emerged from five discrete firing sites. The activities of the five firing sites was plotted separately and then also as a merged trace in Fig. 3C. From these traces, it can be seen that some sites fired Ca²⁺ transients in a regular, rhythmic fashion (site 2, 3, 4). However, when these traces were merged, no temporal entrainment of sites was observed. The lack of entrainment was also apparent across multiple cells in a field of view (FOV). For example, STMs of Ca²⁺ transients in two ICC-IM are shown in Fig. 3D. The STMs were thresholded to be a uniform colour (red for cell 1 and green for cell 2). When the STMs were merged there was little overlap in the activities of the two cells. Thus, Ca²⁺ transients in ICC-IM were not coupled or phase-locked in multiple cells within FOVs or even amongst the firing sites in the same cell.

Fig. 3: LES ICC-IM exhibit dynamic intracellular Ca²⁺ signaling.

A Time-lapse images of an ICC-IM within an LES muscle strip from a Kit-Cre-GCaMP6f mouse (60x). Left most panel shows the raw GCaMP6f signal. All other panels show colour-coded Ca²⁺ transients occurring along the length of ICC-IM from multiple intracellular sites (arrows). B STM of Ca²⁺ transients in a single LES-ICC recorded in situ. Five different firing sites were observed (as indicated) and their activities are plotted as traces in Panel C. Bottom trace in C shows merged traces from the five firing sites. No correlation or entrainment between firing sites was observed. D Leftmost panel shows an image of two adjacent LES ICC-IM from a Kit-KI-GCaMP6f mouse. Merged STMs of the Ca²⁺ transient activity of the two cells highlighted are shown in the rightmost panel. Ca²⁺ activity transients were thresholded and shown in a combined STM in uniform red (cell 1) or green (cell 2) colours.

ICC-IM in the LES were extremely active and capable of firing hundreds of Ca²⁺ transients per minute. Analysis of Ca²⁺ transients in 79 cells from 25 animals, showed that ICC-IM fired Ca²⁺ transients at an average frequency of 246 ± 123.4 min⁻¹ (range was 3 to 576 min⁻¹; Fig. 4A). The amplitudes of Ca²⁺ transients ranged from 0.1 to 3.7 ΔF/F₀ and averaged 0.65 ± 0.5 ΔF/F₀ (range was 0.1 to 3.7 ΔF/F₀; Fig. 4B, 1247 events). Ca²⁺ transients had an average duration of 180 ± 55.2 ms (range was 80 to 650 ms; Fig. 4C, 1247 events), and the spatial spread of the Ca²⁺ transients averaged 15 ± 8.2 μm (range was 1.76 to 58.05 μm; Fig. 4D, 1247 events). The Ca²⁺ transients typically arose from multiple firing sites along the lengths of ICC-IM, and the number of firing sites averaged 7.6 ± 3.8 sites per cell (range was 1 to 20 firing sites per cell; Fig. 4E).

Fig. 4: Quantification of ICC-IM Ca²⁺ signaling in the LES.

A-E Histograms show values for Ca²⁺ transient frequency (A), amplitude (B), duration (C), spatial spread (D), number of firing sites per ICC-IM (E), 1247 events, c=79, N=25.

The significance of Ca²⁺ release was evaluated by testing the effects of a sarcoplasmic/endoplasmic reticulum Ca²⁺ ATPase (SERCA) pump inhibitor, cyclopiazonic acid (CPA, 10 μM). During control recordings from 12 ICC-IM, Ca²⁺ transient frequency was 263 ± 103.2 min^-1. CPA abolished the spontaneous Ca²⁺ transients within 10 min (CPA 10 μM = 0.00 ± 0.00 min⁻1; Fig. 5A, C, P<0.0001, paired t test, c=12, N=4). Ca²⁺ release from the ER was also disrupted by caffeine (10 mM) (Control = 137.1 ± 62.14 min⁻1; Caffeine (10 mM) = 0.00 ± 0.00 min⁻1; Fig. 5B, D, P<0.0001, paired t test, c=10, N=4). At this concentration, caffeine can act as an inhibitor of IP₃R1 channels (Saleem et al., 2014; Heathcote et al., 2019). Likely due to the relatively low expression of genes encoding ryanodine receptors (RyRs) (see Fig. 2B), the RyR antagonist, tetracaine (50 μM), had no effect on the Ca²⁺ transient frequency (Control = 87.00 ± 74.54 min⁻1; tetracaine = 91.71 ± 84.86 min⁻1; P=0.86, paired t test), amplitude (Control = 0.41 ± 0.08622 ΔF/F₀; tetracaine = 0.4584 ± 0.1753 ΔF/F₀; P=0.58, paired t test), duration (Control = 219.7 ± 50.98 ms; tetracaine = 230 ± 44.2 ms; P=0.5796, paired t test) or spatial spread (Control = 14.56 ± 4.743 μm; tetracaine = 14.63 ± 4.498 μm; P=0.97, paired t test) of ICC-IM Ca²⁺ transients (Fig. 6A-B, c=7, N=4). These data suggest that Ca²⁺ release from the ER in ICC-IM of the LES occurs primarily via IP₃R1 channels.

Fig. 5: Ca²⁺ release pathways mediate Ca²⁺ transients in ICC-IM of the LES.

A STMs showing the effects of the SERCA pump inhibitor, CPA (10 μM), on Ca²⁺ transients. B STMs showing the effects of caffeine (10 mM) on Ca²⁺ transients. C Summary data of the effects of CPA on Ca²⁺ transient frequency in ICC-IM (c=12, N=4). D Summary data of the effects of caffeine on Ca²⁺ transient frequency in ICC-IM (c=10, N=4). This concentration of CPA was chosen because it has shown efficacy in blocking ER mediated Ca²⁺ release in ICC other GI organs (Baker et al., 2016; Drumm et al., 2019b; Drumm et al., 2020b; Hannigan et al., 2020). This concentration of caffeine was chosen as it has shown efficacy in blocking IP₃R1 mediated Ca²⁺ release in model systems (Saleem et al., 2014) as well as endothelial cells that also rely on IP₃R mediated Ca²⁺ release (Heathcote et al., 2019).

Fig. 6: Ca²⁺ transients in LES ICC-IM do not rely on RyRs.

A STMs showing the effects of the RyR inhibitor, tetracaine (50 μM), on Ca²⁺ transients. B Summary data of the effects of tetracaine on ICC Ca²⁺ transient frequency, amplitude, duration and spatial spread (c=7, N=4). This concentration of tetracaine was chosen as it has shown efficacy in blocking RyR mediated Ca²⁺ release in several smooth muscle cells and tissues (Bai et al., 2009; MacQuaide et al., 2009; Hilliard et al., 2010; Laver & van Helden, 2011; Pritchard et al., 2018). Higher concentrations were not tested due to concerns with off-target effects of high tetracaine concentrations on IP₃Rs (MacMillan et al., 2005).

Ca²⁺ transients occurred at multiple sites, and a single ICC-IM could fire hundreds of Ca²⁺ transients per min^-1. Therefore, mechanisms must be present to replenish intracellular stores by Ca²⁺ influx to prevent rundown of activity. In a series of experiments to examine this, Ca²⁺ transients were examined and compared under control conditions (normal KRB solution) and in the absence of extracellular Ca²⁺. Ca²⁺ transient frequency averaged 251.1 ± 126.9 min⁻¹ in control conditions (Fig. 7A). When the KRB solution bathing the muscles was replaced with a Ca²⁺ free KRB solution (plus 0.5 mM EGTA) these events decreased to 196.6 ± 103.9 min⁻¹ after 4 mins (Fig. 7B, one-way ANOVA with Tukey’s multiple comparison, Control vs 4 mins = 0.033; c=13, N=5). Ca²⁺ transient frequency was further reduced to 71.9 ± 53.33 min⁻¹ after 10 mins (Fig. 7B, one-way ANOVA with Tukey’s multiple comparison, Control vs 10 mins = 0.0008, c=13, N=5). The amplitude of Ca²⁺ transients was reduced after 4 mins exposure to Ca²⁺ free solution decreasing from 0.7 ± 0.4 ΔF/F₀ in control to 0.6 ± 0.2 ΔF/F_0, after 10 mins in Ca²⁺ free solution, Ca²⁺ transient amplitude was further decreased to 0.3 ± 0.15 ΔF/F₀ (Fig. 7B, one-way ANOVA with Tukey’s multiple comparison, Control vs 4 mins = 0.3422; Control vs 10 mins = 0.0006, c=13, N=5). Removal of extracellular Ca²⁺ had no significant effect on the duration of Ca²⁺ transients (Control = 199.7 ± 59.16 ms; 4 mins = 214 ± 107.2 ms; 10 mins = 169.4 ± 61.23 ms; one-way ANOVA with Tukey’s multiple comparison, Control vs 4 mins = 0.8387; Control vs 10 mins = 0.7647, c=13, N=5), or the spatial spread of Ca²⁺ transients (Control = 12.76 ± 3.563 μm; 4 mins = 12.85 ± 4.237 μm; 10 mins = 11.48 ± 6.326 μm; one-way ANOVA with Tukey’s multiple comparison, Control vs 4 mins = 0.9952, Control vs 10 mins = 0.8863, c=13, N=5). The effects of Ca²⁺ free solution on Ca²⁺ transient frequency (Control = 251.2 ± 126.9 min⁻¹; Wash = 217.2 ± 118.9 min⁻¹; one-way ANOVA with Tukey’s multiple comparison, Control vs wash = 0.5958, c=13, N=5), amplitude (Control = 0.7 ± 0.4 ΔF/F₀; Wash = 0.5452 ± 0.2462 ΔF/F₀; one-way ANOVA with Tukey’s multiple comparison, Control vs wash = 0.2978, c=13, N=5), duration (Control = 199.7 ± 59.16 ms; Wash = 202.5 ± 58.86 ms; one-way ANOVA with Tukey’s multiple comparison, Control vs wash = 0.9816, c=13, N=5) and spatial spread (Control = 12.76 ± 3.563 μm; Wash = 11.41 ± 3.154 μm; one-way ANOVA with Tukey’s multiple comparison, Control vs wash = 0.4671, c=13, N=5) were reversible upon restoration of normal KRB solution for 10 mins (Fig. 7B).

Fig. 7: Ca²⁺ transients in ICC-IM of the LES are depend upon extracellular Ca²⁺.

A STMs showing the effects removing extracellular Ca²⁺ (Ca²⁺ free KRB solution, buffered with 0.5 mM EGTA). The frequency of Ca²⁺ transients in ICC is greatly reduced after 10 minutes exposure to Ca²⁺ free conditions. B Summary effects of Ca²⁺ free solution (0.5 mM EGTA) on ICC Ca²⁺ transient frequency, amplitude, duration and spatial spread after 2-10 min exposures (c=13, N=5).

Nifedipine (1 μM) had no significant effect on the frequency of stochastic Ca²⁺ transients (Fig. 8A-B, Control = 270 ± 175.2 min⁻¹; Nifedipine = 286.8 ± 143.1 min⁻¹, P=0.47, paired t test, c=10, N=4), amplitude (Fig. 8A-B, Control = 0.3405 ± 0.09441 ΔF/F₀; Nifedipine = 0.3883 ± 0.09929 ΔF/F₀; P=0.2, paired t test, c=10, N=4), duration (Fig. 8A-B, Control = 168.1 ± 25.79 ms; Nifedipine = 169.7 ± 16.72 ms; P=0.8, paired t test, c=10, N=4) or spatial spread (Fig. 8A-B, Control = 18.74 ± 7.411 μm; Nifedipine = 20.19 ± 6.279 μm; P=0.4, paired t test, c=10, N=4) in ICC-IM. However, the Orai antagonist, GSK 7975A (1 μM), reduced the frequency of Ca²⁺ transients from 251.2 ± 77.4 to 140.6 ± 76.7 min⁻¹ (Fig. 8C-D, =0.0012, paired t test, c=13, N=4). The amplitude of Ca²⁺ transients was also reduced by GSK 7975A from 0.7 ± 0.45 to 0.4 ± 0.1 ΔF/F₀ (Fig. 8D, =0.024, paired t test, c=13, N=4). Inhibition of Orai with GSK 7975A had no significant effect on the duration of Ca²⁺ transients (Control = 227.5 ± 43.04 ms; GSK 7975A = 217 ± 51.62 ms; paired t test, P=0.1644, c=13, N=4) or the spatial spread of Ca2+ transients (Control = 19.22 ± 5.695 μm; GSK 7975A = 15.97 ± 5.653 μm, paired t test, P=0.1208, c=13, N=4).

Fig. 8: Ca²⁺ transients in ICC-IM of the LES are sustained by Ca²⁺ influx via Orai channels.

A STMs showing the effects of nifedipine (1 μM) on Ca²⁺ transients. B Summary data showing that nifedipine has no effect on Ca²⁺ transient frequency, amplitude, duration and spatial spread (c=10, N=4) in ICC-IM of the LES. C STMs showing the effects of the Orai channel inhibitor GSK 7975A (1 μM) on Ca²⁺ transients. D Summary data of the effects of GSK 7975A on Ca²⁺ transient frequency, amplitude, duration and spatial spread (c=13, N=4). Ca²⁺ transient frequency and amplitude were reduced significantly by GSK 7975A. This concentration of nifedipine was chosen as it has shown efficacy in blocking Ca²⁺ influx through L-type Ca²⁺ channels in several smooth muscle cells and tissues (Cobine et al., 2017; Drumm et al., 2020b; Hannigan et al., 2020; Baker et al., 2021b). This concentration of GSK7975A was chosen as it has shown efficacy in blocking Ca²⁺ influx via Orai channels in several SMCs and tissues (Boie et al., 2017; Chen & Sanderson, 2017; Drumm et al., 2018a; Drumm et al., 2018b; Zheng et al., 2018; Drumm et al., 2020a).

Release of Ca²⁺ and activation of ANO1 channels in ICC-IM would tend to cause depolarization. Conduction of the inward currents to SMCs would also depolarize these cells, enhancing the open probability of Ca_v1.2 channels and excitation-contraction coupling. Previous studies reported that antagonists of Cl⁻ conductances reduced LES tone (Saha et al., 1992; Zhang et al., 2000; Zhang & Paterson, 2003), and it was assumed that the non-specific drugs tested were working on Cl⁻ channels expressed in SMCs. We tested the effects of a potent and selective ANO1 antagonist, Ani9 (Seo et al., 2016) on membrane potentials in mouse LES muscles. Cells in mouse LES clasp muscles were impaled with microelectrodes. Singular values for resting membrane potentials (RMPs) were difficult to tabulate in these muscles due to continuous and apparently random membrane potential oscillations (MPOs; Fig. 9A), previously described as unitary potentials (Zhang et al., 2010). Extended periods of intracellular recording in these tiny muscles were facilitated by treatment with an L-type Ca²⁺ channel antagonist, nicardipine (100 nM) to reduce movement. In the presence of nicardipine, Ani9 (3 μM) caused net hyperpolarization from −43.9 ± 3.2 mV to −48.7 ± 2.3 mV (Fig. 9A-D, =0.0007, paired t test, n=8,N=8). Due to the presence of MPOs, we analyzed the standard error of the variance in RMP, which decreased in the presence of Ani9 from 0.02 ± 0.01 to 0.008 ± 0.002 (Fig. 9E, =0.0082, paired t test, n=8; N=8). These data suggested that the ANO1 channels expressed in ICC-IM in the LES are involved in the regulation of RMP. It is also likely that the stochastic release of Ca²⁺ coupled to activation of ANO1 channels contributes to the generation of MPOs, as these events were reduced in amplitude by Ani9.

Fig. 9: The effect of Ani9 on membrane potentials in mouse LES clasp muscles.

A Representative trace showing the membrane potential oscillations (MPOs) obtained upon impalment of cells in mouse LES clasp muscle. Ani9 (3 μM) caused hyperpolarization and reduced the amplitudes of MPOs in the presence of nicardipine (100 nM). Activity returned to control after washout of the drug. B Expanded time scale of portions of the recording in A (a=control, b=Ani9, c=washout). C Amplitude histograms from 1 min recordings generated by Clampfit 10.3 (Methods) summarizing RMPs and showing the net hyperpolarizations induced by Ani9 (n=8, N=8). Median values for RMP and MPOs were calculated from Gaussian fits of the data. This concentration of Ani9 was chosen, as there were no effects on contractile responses to high K⁺ solutions that might suggest off-target effects of the drug (see Fig. 11). D Summary of the effects of Ani9 on RMP. E. Summary of the effects of Ani9 on MPOs (plotted as standard error of the variance in RMP (distribution).

The ANO1 antagonist Ani9 (3 μM) also reduced the spontaneous tone in mouse LES clasp muscles by 88.7% ± 5.476% (Fig. 10A, =0.0039, paired t test, n=6, N=6). Spontaneous tone also developed in monkey LES clasp muscles, and 3 μM Ani9 reduced tone in these muscles by 84.2% ± 11.28% (Fig. 10C, =0.0185, n=6, N=2). There was no significant difference in the degree of tone inhibition in mouse and monkey LES in response to Ani9 (Fig. 10C, Mouse = 88.7% ± 5.476%; Monkey = 84.2% ± 11.28%; P=0.41, Welch’s unpaired t test). The effects of Ani9 on LES tone was not attributed to non-specific effects on L-type Ca²⁺ channels, as Ani9 (3 μM) had no significant effect on contractions elicited by 60 mM KCl in mouse (% reduction of KCl-induced contraction by Ani9 (3 μM) = 99.51% ± 1.98%; Fig. 11A-C, =0.572, paired t test, n=6, N=6) or monkey (% change of KCl-induced contraction by Ani9 (3 μM) = 100.4% ± 4.498%; P=0.892, paired t test, monkey n=3, N=3) LES.

Fig. 10: ANO1 channel antagonist reduces LES tone.

A Representative contractile trace showing the effect of the ANO1 channel antagonist, Ani9 (3 μM), on tone in mouse LES (n=6, N=6). B Representative contractile trace showing the effect of Ani9 (3 μM) on tone in monkey LES (n=6, N=2). C Summary data for the effects of Ani9 on tone in mouse and monkey LES. This concentration of Ani9 was chosen because it had no effect on contractile responses to high K⁺ solutions (see Fig. 11). This concentration of sodium nitroprusside (SNP) was chosen as it has shown efficacy in producing maximum relaxation in other GI muscles (Cobine et al., 2017; Cobine et al., 2018).

Fig. 11: Tests to confirm concentrations of ANO1 channel antagonist did not have off-target effects on KCl-induced contractions of LES.

A Representative contractile trace showing the effects of the ANO1 channel antagonist, Ani9 (1 and 3 μM) and nifedipine (1 μM) on contraction of mouse LES induced by 60 mM KCl solution. B Representative contractile trace showing the effects of a range of Ani9 concentrations (300 nM – 10 μM) and nifedipine (1 μM) on contraction of monkey LES induced by 60 mM KCl solution. Ani9 at the concentrations used elsewhere in this study did not block high K⁺ contractions but the contractions were completely blocked by nifedipine. C Summary data for the effects of Ani9 (3μM) on KCl-induced contractions in mouse (n=6, N=6) and monkey LES (n=4, N=4).

We predicted that an antagonist of L-type Ca²⁺ channels should have similar inhibitory effects on LES tone as Ani9 if activation of ANO1 in ICC-IM leads to opening of Ca_v1.2 channels in SMCs. Nifedipine (1 μM) reduced mouse LES tone by 83.9% ± 13.69% (Fig. 12A, =0.0250, paired t test, n=7, N=7) and tone in the monkey LES by 82.6% ± 9.584% (Fig. 12B, =0.0067 paired t test, n=6, N=3). No significant difference in the degree of inhibition by nifedipine between mouse and monkey muscles was observed (Fig 12C, =0.851, unpaired t test).

Fig. 12: Ca_v1.2 channel antagonist reduces LES tone.

A Representative contractile trace showing the effect of the Ca_v1.2 channel antagonist, nifedipine (1 μM), on tone in mouse LES (n=7, N=7). B Representative contractile trace showing the effects of nifedipine (1 μM) on tone in the monkey LES (n=6, N=3). SNP (10 μM) was added at the end of the traces in A&B to determine if there was any remaining tone after nifedipine. C Summary data for the effects of nifedipine on mouse and monkey LES tone.

Discussion

This study demonstrates a novel role for ICC-IM of the LES in the generation of tone. We found that activation of a Cl⁻ conductance, encoded by Ano1 and expressed exclusively in ICC within the LES, is an important regulator of LES tone. The mechanism for activation of ANO1 was also investigated, and we found that spontaneous Ca²⁺ release events in ICC-IM are likely the principal mechanism of ANO1 activation. Ca²⁺ release events were highly dynamic, amounting to hundreds of Ca²⁺ transients per min. The frequency, amplitude, duration and spatial spread of the Ca²⁺ transients in hundreds of cells were tabulated, and we found that these events occurred independently and were stochastic in nature. Ca²⁺ release from single sites displayed evidence of periodicity (Fig. 3), but none of our observations suggested that the events were entrained, phase-coupled or performed as coupled oscillators. Merging of STMs demonstrated the independence of firing sites and a lack of coupling of Ca²⁺ transients in single cells or between adjacent cells within a field of view. ICC in the LES express Itpr1, and a high concentration of caffeine, shown previously to block IP₃ receptors (Saleem et al., 2014; Heathcote et al., 2019), inhibited spontaneous Ca²⁺ release. Tetracaine, an inhibitor of ryanodine receptors and used at a concentration demonstrated to affect RyRs but not IP₃Rs (Bai et al., 2009; MacQuaide et al., 2009; Hilliard et al., 2010; Laver & van Helden, 2011; Chernov-Rogan et al., 2018; Pritchard et al., 2018) had no effect on Ca²⁺ transients. This was consistent with the relatively lower expression of genes encoding RyRs in ICC. Ca²⁺ transients were abolished by addition of CPA or removal of extracellular Ca²⁺. Furthermore, addition of an Orai channel antagonist, GSK7975A, significantly reduced the frequency and amplitude of Ca²⁺ transients. ICC-IM in mouse and monkey LES clasp muscles expressed ANO1 channels, and localized Ca²⁺ release has been shown to activate ANO1 currents (spontaneous transient inward currents or STICs) in ICC (Zhu et al., 2015). Activation of STICs would tend to exert a depolarizing influence on the electrically-coupled cells of the SIP syncytium (SMC, ICC, PDGFR-alpha⁺ cells), and this was observed, as the ANO1 antagonist, Ani9, caused hyperpolarization.

Depolarization of the SMC component of the SIP syncytium would increase the open probability of Ca_v1.2 channels in SMCs and facilitate excitation- contraction coupling. Ani9 reduced tone in LES muscles from mouse and monkey by ~ 80% or about the same degree of inhibition caused by nifedipine. Our results are consistent with the hypothesis that Ca²⁺ transients in ICC-IM activate inward current and depolarization that conducts to SMCs and conditions the membrane potentials of these cells to enhance the open probability of L-type Ca²⁺ channels (Ca_v1.2). Firing of Ca²⁺ transients in thousands of ICC-IM in LES muscles could provide a state of tonic depolarization and continuous Ca²⁺ entry in SMCs. Thus, firing of Ca²⁺ transients in ICC-IM provides a new mechanistic basis for the development and maintenance of tone in the LES. This idea is consistent with previous studies on mice with mutations in Kit (W/W^V mice) that cause developmental defects and reduce the number of ICC in the LES (Ward et al., 1998). LES pressures were found to be hypotensive in W/W^V mice (Sivarao et al., 2001; Muller et al., 2014), which is logical if the mechanism underlying the generation of tone is intrinsic to ICC-IM.

Previous studies suggested the involvement of Cl⁻ channels, suspected of being Ca²⁺-activated, in LES tone and in responses to neural inputs (Crist et al., 1991; Saha et al., 1992; Zhang et al., 2000; Zhang & Paterson, 2003). Inhibitory junction potentials (IJPs) activated by stimulation of enteric inhibitory neurons in the opossum esophagus were not affected by apamin, tetraethylammonium chloride or 4-aminopyridine but were first enhanced and then inhibited by perfusion of muscles with low Cl⁻ solution (Crist et al., 1991). DIDS also inhibited IJPs. Tone in opossum LES was inhibited by niflumic acid with an IC₅₀ of 1.2 μM. LES tone was also reduced by exposure of muscles to solutions with reduced extracellular Cl⁻ (Saha et al., 1992). ANO1, or other putative Ca²⁺-activated Cl⁻ channels, have never been resolved in GI SMCs by immunohistochemistry (Gomez-Pinilla et al., 2009) or by analysis of gene transcripts, as shown in the current study. Due to the selective expression of ANO1 channels in ICC, the efficacy of an ANO1 channel antagonist (Ani 9 (Seo et al., 2016)) demonstrates that i) the Cl⁻ conductance affecting LES tone is present in ICC-IM, and ii) ICC-IM have a fundamental role in generation and maintenance of tone in LES muscles of mouse and monkey. Expression of several additional genes known to be important for Ca²⁺ release from intracellular stores or to serve as biomarkers for cellular components of the SIP syncytium were measured in sorted ICC and SMCs by qPCR (Fig. 2). The pharmacological and physiological tests performed in this study strongly suggested functional expression of proteins encoded by the genes expressed, however, with the exception of Ano1 and Kit (evaluated by immunohistochemistry; Fig. 1), analysis of protein expression was not performed.

A consequence of the stochastic Ca²⁺ transients in ICC-IM is likely to be activation of ANO1 channels, however direct proof of coupling between Ca²⁺ release and activation of ANO1 channels has not been reported for any type of ICC to date. Activation of clusters of ANO1 channels generates STICs and these currents cause spontaneous transient depolarizations (STDs; (Van Helden et al., 2000)). STDs have also been called unitary potentials by Hirst and colleagues, and these events appear to be a common phenomenon in many GI muscles (Edwards et al., 1999; Suzuki et al., 2003; Beckett et al., 2004). We do not favor the term ‘unitary potentials’ because STDs are variable in amplitude and clearly not a function of single channel openings or Ca²⁺ transients that are of constant amplitude or ‘unitary’ in nature. The term STDs seems more consistent with the stochastic nature of Ca²⁺ transients and the STICs that are generated and linked to transient changes in membrane potential. STDs are likely to depend upon the magnitude and spatial spread of Ca²⁺ transients that would activate variable numbers of ANO1 channels. Due to the stochastic nature of Ca²⁺ transients and the variable amplitudes of STICs and STDs, these events create noisy membrane potentials when intracellular recordings are made from SMCs in GI muscles (Edwards et al., 1999; Suzuki et al., 2003; Beckett et al., 2004).

Previous studies have shown that STDs in intact GI muscles are inhibited by perfusion of muscles with membrane-permeable Ca²⁺ buffers (e.g. BAPTA-AM or MAPTA-AM; (Van Helden et al., 2000)) or when Ca²⁺ release from stores is inhibited (Edwards et al., 1999; Van Helden et al., 2000; Kito et al., 2002), suggesting that the conductance responsible for STDs is Ca²⁺ dependent and linked to spontaneous Ca²⁺ release. In previous studies, noisy membrane potentials were recorded from LES muscles, and membrane noise was greatly attenuated by depleting Ca²⁺ stores (Zhang & Paterson, 2003), by Cl⁻ channel antagonists (Zhang & Paterson, 2003) or in LES muscles of W/W^V mice with reduced numbers of ICC-IM (Ward et al., 1998; Zhang et al., 2010). We also observed very noisy membrane potentials in mouse LES in the current study and termed this activity membrane potential oscillations (Fig 9). This activity was reduced in amplitude by Ani9 suggesting it was due, at least in part, to activation of ANO1 channels and therefore likely to be STDs. Blocking of ANO1 channels hyperpolarized membrane potentials of LES muscles, suggesting that inward currents generated by ANO1 channels in ICC-IM provides depolarizing drive on the SIP syncytium. Taken together, these observations suggest that stochastic release of Ca²⁺ is linked to generation of STDs in ICC-IM. The Ca²⁺ release events regulating the open probability of ANO1 channels are equivalent to the dynamic Ca²⁺ transients observed and quantified in the present study. Summation of STDs from thousands of ICC-IM exerts a tonic depolarizing influence on SMCs of the SIP syncytium. Future studies on mice with cell-specific knock down of Ano1 in ICC would provide additional verification of this hypothesis.

Membrane potentials of LES SMCs are relatively depolarized in comparison to membrane potentials of cells in phasic GI smooth muscles. Average mean resting potential in mouse LES averaged −43 mV in a previous study (Zhang et al., 2010) and −44 mV in the present study. Activation of resolvable inward Ca²⁺ currents attributable to L-type channels occur in GI SMCs when cells are depolarized positive to −50 mV (Langton et al., 1989). The increase in open probability of L-type Ca²⁺ channels is steep over the range of potentials from −50 to 0 mV, such that small changes in membrane potential can result in significant differences in the open probability of these channels. L-type Ca²⁺ channels also experience voltage-dependent inactivation, such that depolarization increases the open probability of channels but then channel inactivation also occurs. With L-type Ca²⁺ channels, however, inactivation is in incomplete, and residual enhanced open probability occurs at potentials between −50 and 0 mV. This phenomenon is known as window current (Cohen & Lederer, 1987), and it explains how constant influx of Ca²⁺ can occur through L-type Ca²⁺ channels at potentials experienced by LES muscles.

STDs, when first observed after dissection of pyloric muscle bundles, occurred in a non-rhythmic manner before development of slow wave activity (Van Helden et al., 2000). These authors concluded that STDs and slow waves both arose from the same mechanism, as they shared identical pharmacology and dependence on IP₃-dependent Ca²⁺ release. Synchronization was attributed to entrainment of STDs by a process described as Ca²⁺ phase waves (van Helden & Imtiaz, 2003). In spite of similar periods of incubation of LES muscles in the present study, synchronization of Ca²⁺ transients, and therefore presumably STDs, was never observed. In fact, Ca²⁺ transients, even within single cells, displayed independence and remained stochastic in nature. The difference between ICC-IM in the LES and pylorus are unclear at the present time but may be related to the lack of a voltage-dependent mechanism in LES that might ‘entrain’ Ca²⁺ release events, as in muscles where slow waves are generated (van Helden & Imtiaz, 2003; Edwards & Hirst, 2005; Drumm et al., 2017; Baker et al., 2021b).

In summary, this study shows how mechanisms intrinsic to ICC-IM in the LES generate and maintain basal tone in this sphincter. Ongoing, stochastic Ca²⁺ transients generated spontaneously from many sites in each ICC-IM were characterized. ANO1 is highly expressed in ICC-IM of both mouse and monkey LES, and a selective antagonist for ANO1, Ani9, hyperpolarized muscles in mouse LES suggesting that ANO1 channels contribute to the depolarized level of LES muscles. Ani9 also inhibited LES tone significantly in mouse and monkey. Our data also provide an explanation for why the LES is hypotensive in W/W^V mice (Sivarao et al., 2001; Muller et al., 2014), as these mice have significantly reduced numbers of ICC-IM in the LES. With reduced ICC-IM, the mechanism for generating inward currents through ANO1 channels and producing tonic depolarization of LES muscles is lacking. Taken together, our data suggest the following concept for tone generation and maintenance in the LES: stochastic release of Ca²⁺ in ICC-IM leads to activation of ANO1 channels, conduction of depolarizing currents to SMCs and activation of L-type Ca²⁺ channels (see Abstract Figure). This novel concept has important consequences for normal sphincteric functions of the LES and in developmental conditions in which ICC-IM are compromised and diseases or aging in which ICC-IM may be reduced or experience loss-of-function.

Key Points Summary.

• The lower esophageal sphincter (LES) generates contractile tone preventing reflux of gastric contents into the esophagus. LES smooth muscle cells (SMCs) display depolarized membrane potentials facilitating activation of L-type Ca²⁺ channels.

• Interstitial cells of Cajal (ICC) express Ca²⁺-activated Cl⁻ channels encoded by Ano1 in mouse and monkey LES. Ca²⁺ signaling in ICC activates ANO1 currents in ICC.

• ICC displayed spontaneous Ca²⁺ transients in mice from multiple firing sites in each cell and no entrainment of Ca²⁺ firing between sites or between cells.

• Inhibition of ANO1 channels with a specific antagonist caused hyperpolarization of mouse LES and inhibition of tone in monkey and mouse LES muscles.

• Our data suggest a novel mechanism for LES tone in which Ca²⁺ transient activation of ANO1 channels in ICC generates depolarizing inward currents that conduct to SMCs to activate L-type Ca²⁺ currents, Ca²⁺ entry and contractile tone.