The effects of mitochondrial inhibitors on Ca²⁺ signalling and electrical conductances required for pacemaking in interstitial cells of Cajal in the mouse small intestine

Abstract

Interstitial cells of Cajal (ICC-MY) are pacemakers that generate and propagate electrical slow waves in gastrointestinal (GI) muscles. Slow waves appear to be generated by the release of Ca²⁺ from intracellular stores and activation of Ca²⁺-activated Cl⁻ channels (Ano1). Conduction of slow waves to smooth muscle cells coordinates rhythmic contractions. Mitochondrial Ca²⁺ handling is currently thought to be critical for ICC pacemaking. Protonophores, inhibitors of the electron transport chain (FCCP, CCCP or antimycin) or mitochondrial Na⁺/Ca²⁺ exchange blockers inhibited slow waves in several GI muscles. Here we utilized Ca²⁺ imaging of ICC in small intestinal muscles in situ to determine the effects of mitochondrial drugs on Ca²⁺ transients in ICC. Muscles were obtained from mice expressing a genetically encoded Ca²⁺ indicator (GCaMP3) in ICC. FCCP, CCCP, antimycin, a uniporter blocker, Ru360, and a mitochondrial Na⁺/Ca²⁺ exchange inhibitor, CGP-37157 inhibited Ca²⁺ transients in ICC-MY. Effects were not due to depletion of ATP, as oligomycin did not affect Ca²⁺ transients. Patch-clamp experiments were performed to test the effects of the mitochondrial drugs on key pacemaker conductances, Ano1 and T-type Ca²⁺ (Ca_V3.2), in HEK293 cells. Antimycin blocked Ano1 and reduced Ca_V3.2 currents. CCCP blocked Ca_V3.2 current but did not affect Ano1 current. Ano1 and Cav3.2 currents were inhibited by CGP-37157. Inhibitory effects of mitochondrial drugs on slow waves and Ca²⁺ signalling in ICC can be explained by direct antagonism of key pacemaker conductances in ICC that generate and propagate slow waves. A direct obligatory role for mitochondria in pacemaker activity is therefore questionable.

Graphical abstract

1. Introduction

The gastrointestinal (GI) tract contains specialized populations of cells termed interstitial cells of Cajal (ICC), that regulate the contractile behaviors of GI smooth muscle cells (SMCs) [1]. In the small intestine, ICC at the plane of the myenteric plexus (ICC-MY) are electrically coupled in a network and are pacemakers that generate and propagate slow waves that activate voltage-gated Ca²⁺ channels in SMCs, and thus coordinate the rhythmic contractions of intestinal muscles [1–6]. GI motility can be augmented by intramuscular ICC (ICC-IM), or as in the small intestine, ICC located in the deep muscular plexus (ICC-DMP). ICC-IM/ICC-DMP have no voltage-dependent mechanism to allow formation of slow waves, but they receive inputs from enteric motor neurons and conduct responses to electrically coupled SMCs [7–10].

All classes of ICC in the GI tract express Ca²⁺-activated Cl⁻ channels (ANO1), and this conductance is fundamental for slow wave activity and responsiveness to motor neurons [11–14]. Both ICC-MY and ICC-DMP generate spontaneous transient inward currents (STICs) resulting from periodic activation of clusters of ANO1 channels [11,15]. With confocal imaging techniques and genetically encoded Ca²⁺ sensors expressed in ICC, we recently characterized the differences in Ca²⁺ signaling behaviours in ICC-MY and ICC-DMP in the small intestine. In ICC-MY, Ca²⁺ transients fire clusters of Ca²⁺ at the frequency of slow waves [16]. Ca²⁺ transients within the clusters occur asynchronously from multiple sites in the ICC-MY network. The duration of the Ca²⁺ clusters matches the duration of the plateau phase of slow waves [16]. Thus the asynchronous firing constitutes a means to sustain Ca²⁺ release and openings of ANO1 channels, creating the relatively long duration slow wave depolarizations. Slow waves propagate through the ICC-MY network by a voltage-dependent mechanism that relies on the activation of T-type Ca²⁺ channels [16]. In contrast, ICC-DMP fire Ca²⁺ transients in a stochastic manner with no voltage dependency, no coordination between firing sites in single cells, and no coordination between cells [10]. In both cell types, Ca²⁺ transients arise from release of Ca²⁺ from endoplasmic reticulum (ER) stores via ryanodine (RyR) and inositol-triphosphate (IP₃) receptors (IP₃Rs) [10,16,17]. Ca²⁺ transients appear to represent the fundamental signalling events regulating the open probability of ANO1 channels that provide the regulatory functions of ICC in GI motility. Thus, factors modulating Ca²⁺ signaling in ICC are important topics for investigation.

Previous studies have suggested that mitochondria play an important role in ICC function, as these organelles were found in abundance in ultrastructural studies of ICC [18–20]. Mitochondria act as a reversible Ca²⁺ storage unit with Ca²⁺ transport in and out of these organelles affecting cytosolic Ca²⁺ concentrations and whole cell Ca²⁺ signalling [21,22]. Electrical slow waves in the mouse small intestine, guinea-pig stomach and canine colon are blocked by pharmacological agents that inhibit mitochondrial Ca²⁺ uptake by reducing the mitochondrial transmembrane potential with the protonophores carbonyl cyanide-4 (trifluoromethoxy) phenylhydrazone (FCCP) or carbonyl cyanide m-chlorophenyl hydrazine (CCCP) or with inhibitors of the electron transport chain (ETC) such as antimycin [23]. CCCP not only blocked slow wave generation but also blocked active propagation of slow waves in the canine colon (a property intrinsic to ICC; [23,24]), mouse colon [25] and gastric antrum [26].

In the present study, imaging techniques were used to monitor Ca²⁺ signaling in intact small intestinal muscles from mice expressing a genetically encoded Ca²⁺ indicator (GCaMP3) in ICC. The effects of mitochondrial inhibitors on Ca²⁺ signals in ICC were investigated. We found that mitochondrial inhibitors potently blocked Ca²⁺ transients in ICC-MY and were less effective on ICC-DMP. Patch clamp experiments were performed to determine whether these drugs affect key conductances essential for pacemaking in ICC (i.e. ANO1 and T-type Ca²⁺ conductances expressed in HEK 293 cells). Mitochondrial drugs had significant inhibitory effects on these important conductances for slow wave generation and propagation and thus their inhibitory effects on Ca²⁺ signalling in ICC might be explained by inhibition of slow waves and not a direct effect on mitochondrial Ca²⁺ handling. Our findings suggest that models of ICC pacemaker activity, which have previously emphasized a role for mitochondrial Ca²⁺ handling based on pharmacological evidence, should be interpreted with caution.

2. Methods

2.1. Animals

All animals used and the protocols carried out in 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. B6.129S7-Kit^tm1Rosay/J (Kit^+/copGFP mice) were generated and bred in house [27]. GCaMP3-floxed mice (B6.129S-Gt(ROSA)26Sor^{tm38(CAG-GCaMP3)Hze}/J)) and their wild-type siblings (C57BL/6) were purchased from the Jackson Laboratories (Bar Harbor, MN, USA). c-Kit^+/Cre-ERT2 (Kit-Cre mice) were gifted from Dr. Dieter Saur (Technical University Munich, Germany). Once these mice were crossed, (Kit-Cre-GCaMP3), they were injected with tamoxifen at 6–8 weeks of age (2 mg for three consecutive days) as described previously [10,16] to induce the activation of cre recombinase and the resulting expression of GCaMP3 (confirmed by genotyping). Mice were anaesthetized by inhalation with isoflurane (Baxter, Deerfield, IL, USA) and killed by cervical dislocation 50 days post final injection date.

2.2. Tissue preparation

Following an abdominal incision, segments of jejunum (2 cm) were removed from mice and bathed in ice-cold Krebs-Ringer bicarbonate solution (KRB). The jejunal segments were opened along the mesenteric border and luminal contents were washed away with cold KRB. The mucosa and sub-mucosa layers were removed by sharp dissection. Tissues were pinned in a 5ml, 60mm Sylgard coated dish with the serosa facing downwards.

2.3. Calcium imaging

Intestinal muscles were mounted on the stage of an Eclipse E600FN microscope (Nikon Inc., Melville, NY, USA) and perfused continuously with KRB solution at 37°C for an equilibration period of 1 hour before beginning experiments. Following the equilibration period, in situ Ca²⁺ imaging data of small intestinal ICC was acquired with a 60× 1.0 CFI Fluor lens (Nikon instruments INC, NY, USA). GCaMP3 was excited at 488 nm (T.I.L.L. Polychrome IV, Grafelfing, Germany), as previously described [10,16]. Ca²⁺ imaging sequences were collected at 33 fps with TILLvisION software (T.I.L.L. Photonics GmbH, Grafelfing, Germany; [pixel size 0.225 μm). All Ca²⁺ imaging experiments were performed in the presence of the L-type Ca²⁺ channel blocker, nicardipine (1 μM), to block smooth muscle action potentials and contraction-induced artefacts. Nicardipine has been shown previously to have no effect on Ca²⁺ signalling in small intestinal ICC; [10,16]. Imaging of fields of view (FOVs) was performed for an initial 30 sec control period and then drugs were perfused into the muscle chamber for 12–15 mins before acquiring another 30 sec recording to asses drug effects.

2.4. Analysis of Ca²⁺ transients

ICC-MY and ICC-DMP exhibit different Ca²⁺ transient behaviours, and therefore image sequences were analyzed differently. ICC-MY form an interconnected network and displays clusters of Ca²⁺ transients (each cluster lasts for ~1 sec). Ca²⁺ transients from various sites within the cells in the FOV fire asynchronously within the clusters, and clusters occur at a frequency of ~30 per minute [16]. We have referred to these clusters as Ca²⁺ transient clusters (CTCs). The Ca²⁺ transients occurring within CTCs were quantified using particle analysis, as described previously [16]. Briefly, movies were imported into custom software Volumetry G8d. A differential (Δt = ± 66–70 ms) and Gaussian filter (1.5 × 1.5μm, StdDev 1.0) was applied before a particle analysis routine was applied by use of a flood-fill algorithm that marked the structure of all adjoining pixels that had intensities above a threshold. Ca²⁺ transient particles (PTCLs) were brighter and larger than noise particles and the point at which noise particles emerged and reduced the average particles size was marked as the threshold. The movie was then saved as a coordinate based PTCL movie. Any remaining noise in the PTCL file was filtered out by only including PTCLs >6 μm² in analysis. To identify Ca²⁺ firing sites, only particles that did not overlap with any particles in the previous frame but overlapped with particles in the next 70 ms were considered Ca²⁺ firing sites. The frequency of CTCs, the number of Ca²⁺ firing sites and the PTCL area (μm²) and total PTCL count for the entire recording was then calculated. PTCL area and PTCL count were expressed as PTCL area / frame and PTCL count / frame.

In contrast to ICC-MY, ICC-DMP do not form a connected network and exhibit spontaneous Ca²⁺ signals that are stochastic in nature and not coordinated or arranged into temporal clusters by a voltage-dependent mechanism [10]. Analysis of Ca²⁺ activity in ICC-DMP was performed using spatio-temporal (ST) mapping as described previously [10,28,29]. Movies of Ca²⁺ activity in ICC-DMP were converted to a stack of TIFF (tagged image file format) images and imported into (Volumetry G8c) for preliminary processing analysis. Whole cell ROIs were used to generate ST maps of Ca²⁺ activity in individual ICC-DMP within a given FOV. ST maps were then imported as TIFF files into Image J (version1.40, National Institutes of Health, MD, USA, http://rsbweb.nih.gov/ij) for post hoc analysis of Ca²⁺ events. The basal Ca²⁺ fluorescence was acquired from regions of the cell that displayed the most uniform and least intense fluorescence (F₀). Fluorescence values in the entire cell were then divided by the F₀ value to calibrate the ST map as F/F₀. Ca²⁺ transient frequency in ICC-DMP 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.

2.5. Drugs and solutions for calcium imaging

Tissues were maintained and perfused with KRB containing (mmol/L): NaCl, 120.35; KCl, 5.9; NaHCO₃, 15.5; NaH₂PO₄, 1.2; MgCl₂, 1.2; CaCl₂, 2.5; and glucose, 11.5. KRB was bubbled with a mixture of 97% O₂ – 3% CO₂ and warmed to a physiological temperature of 37 ± 0.2 °C. Antimycin A, CGP-37157, Oligomycin and Ru360 were purchased from Sigma-Aldrich (St Louis, MO, USA). Thapsigargin, FCCP and CCCP were purchased from Tocris Bioscience (Ellisville, Missouri, USA).

2.6. Patch clamp recordings

Whole-cell patch clamp recordings were performed using an Axopatch 200B amplifier and pClamp 9.0 software (Axon Instruments, Union city, CA). Pipettes with resistance between 3~6 MΩ were used for whole cell patch clamp recordings. Data was digitized and acquired using pClamp software (Clampex 10.0.0.61, Axon Instruments) and analyzed using Clampfit (v9.02, Axon Instruments, CA, USA) and GraphPad Prism (version 5.0, GraphPad Software Inc., San Diego, CA, USA). The Ano1 AC splice variant tagged with GFP was expressed in HEK 293 cells (ATCC, Manassas, VA), as described previously [30]. Ca_V3.2, T-type Ca²⁺ channels, stably expressed in HEK 293 cells (provided by Dr. E. Perez-Reyes; University of Virginia; [31]), were used as described previously [16]. Conventional dialyzed whole cell patch-clamp was used to record Ano1 and Ca_V3.2 currents from HEK 293 cells. Step pulses from −80 mV to +70 mV in 10 mV increment from a holding potential of −80 mV were applied to obtain Ano1 and Ca_V3.2 currents. SOCE current (I_CRAC) was recorded in HEK 293 cells stably expressing Orai1-V102C-CFP and STIM1-yellow fluorescent protein (YFP) (donated by Dr Donald Gill; The Pennsylvania State University College of Medicine). HEK 293 stably expressing each channel were derived and maintained in Dulbecco’s modified Eagle’s medium (Mediatech) supplemented with FBS (10%, vol/vol, Gibco), penicillin-streptomycin (1%, vol/vol, Gibco), and glutamax (1%, vol/vol; Gibco). Plasmid transfections were performed by the Bio-Rad Gene Pulser Xcell via electroporation in OPTI-MEM medium. Transfected cells were cultured on coverslips in OPRI-MEM with 5% FBS, penicillin and streptomycin. Transfected cells were moved to 12-well plates. The next day, cells were demounted and used for patch-clamp recordings.

2.7. Solutions used for patch clamp recordings

The external solution for the recording of Ano1, Ca_V3.2 and SOCE whole cell currents had the following composition (mM): 150 NMDG, 150 HCl, 2 CaCl₂, 1 MgCl₂ and 10 HEPES adjusted to pH 7.4 with NMDG for Ano1 currents and 5 KCl, 135 NaCl, 2 CaCl₂, 10 glucose, 1.2 MgCl₂, and 10 Hepes adjusted to pH 7.4 with Tris for Ca_V3.2 and SOCE currents, respectively. The pipette solutions contained the following (mM): 147.7 NMDG, 147.7 HCl, 10 EGTA, 4.1 CaCl₂, 10 HEPES adjusted to pH 7.2 with NMDG for Ano1 currents and 135 CsCl, 3 MgATP, 0.1 NaGTP, 10 BAPTA, 10 HEPES, 10 Glucose adjusted to pH 7.2 with Tris for Ca_V3.2 and SOCE currents, respectively.

2.8. Statistics

Unless otherwise stated in the text, data is represented as mean ± standard error (S.E.M.). Statistical analysis was performed using student’s t-tests or a ANOVA with a Dunnett post-hoc test where appropriate. In all statistical analyses, P<0.05 was taken as significant. Probabilities < 0.05 are represented by a single asterisk (*), probabilities < 0.01 are represented by two asterisks (**) and probabilities < 0.001 are represented by three asterisks (***).

3. Results

3.1. The effect of mitochondrial blockers on ICC-MY

We used a genetically encoded Ca²⁺ sensor (GCaMP3) to monitor Ca²⁺ signaling in ICC. As described previously, ICC-MY of the small intestine of these mice displayed spontaneous Ca²⁺ signals which manifested as Ca²⁺ release events originating from multiple sites in ICC-MY somata and processes [16]. The Ca²⁺ release events were clustered together temporally within a ~1 sec period and formed an ICC-MY network wide Ca²⁺ response which we termed Ca²⁺ transient clusters (CTCs; [16]). CTCs were rhythmic and occurred at a frequency of 23.6 ± 1.2 min⁻¹ with 65.3 ± 5.8 Ca²⁺ firing sites per FOV (60x objective), which confirms previous findings [16]. Fig.1Ai shows a representative heat map of summated CTCs activity over a 30 sec period in an ICC-MY network. Using PTCL analysis, we quantified the CTC frequency, the number of Ca²⁺ firing sites per FOV, PTCL areas and PTCL counts, as shown in the traces in Fig.1Aii.

Fig. 1: The effects of protonophores on Ca²⁺ signaling in ICC-MY.

Ai Representative heat map showing summation of Ca²⁺ transient particles in ICC-MY for an entire recording period under control conditions. Aii Traces of Ca²⁺ transient particle activity in the ICC-MY network in control conditions showing PTCL area (blue) and PTCL count (red). Bi Representative heat map showing summation of Ca²⁺ transient particles in ICC-MY for an entire recording period in the presence of FCCP (1 μM). Bii Traces of Ca²⁺ transient particle activity of the ICC-MY network in FCCP (1 μM). C Summary data showing the effects of FCCP (1 μM) on CTC frequency (i), number of Ca²⁺ sites / FOV (ii), PTCL area (iii) and PTCL count (iv), n=6. Di Representative heat map showing summation of Ca²⁺ transient particles in ICC-MY for an entire recording under control conditions. Dii Traces of Ca²⁺ transient particle activity in the ICC-MY network in control conditions showing PTCL area (blue) and PTCL count (red). Ei Representative heat map showing summation of Ca²⁺ transient particles in ICC-MY for an entire recording in the presence of CCCP (1 μM). Eii Traces of Ca²⁺ transient particle activity in the ICC-MY network in CCCP (1 μM). F Summary data showing the effects of CCCP (1 μM) on CTC frequency (i), number of Ca²⁺ sites / FOV (ii), PTCL area (iii) and PTCL count (iv), n=6.

We sought to evaluate the role of mitochondrial Ca²⁺ handling in modulating CTCs in ICC-MY by assessing the effects of the protonophores FCCP and CCCP. These substances collapse mitochondrial membrane potential by increasing the permeability to protons, reducing the ability of the uniporter to take up Ca²⁺ [32]. In 4 out of 6 muscles, FCCP (1 μM) completely abolished CTCs in ICC-MY (Fig.1Bi,ii). On average, FCCP reduced CTC frequency by ~ 92% from 21.7 ± 2.6 to 1.7 ± 1.1 min⁻¹ (Fig.1Ci, P=0.002, n=6). The number of Ca²⁺ firing sites per FOV in ICC-MY networks was reduced by FCCP from 68 ± 16.9 to 9 ± 5.6 (Fig. 1Cii, P=0.026, n=6). PTCL area of Ca²⁺ transients was also reduced from 76.1 ± 25.4 μm² to 3.3 ± 2.9 (Fig.1Ciii, P=0.039, n=6). PTCL count was also reduced by FCCP from 2.7 ± 0.6 to 0.1 ± 0.07 (Fig.1Civ, P=0.039, n=6). Similar results were obtained with CCCP (1 μM; Fig. 1D–E), which decreased CTC frequency from 23 ± 2.9 min⁻¹ to 6 ± 3.72 min⁻¹ (Fig. 1Fi, P=0.03, n=6). CCCP reduced the number of Ca²⁺ firing sites per FOV in ICC-MY from 75.5 ± 16.1 to 18.8 ± 11.7 (Fig.1Fii, P=0.01, n=6). PTCL area was decreased from 108.5 ± 40.7 μm² in control to 11.4 ± 10.3 μm² in the presence of CCCP (Fig.1Fiii, P=0.04, n=6). CCCP also significantly reduced PTCL count from 2.7 ± 0.8 to 0.25 ± 0.18 (Fig.1Fiv, P=0.025, n=6).

Mitochondrial Ca²⁺ handling can also be disrupted with electron transport chain (ETC) inhibitors. As shown in Fig.2A–B, application of antimycin (10 μM) greatly reduced CTCs. CTCs were completely abolished by antimycin in 4 of 5 muscles, and on average, antimycin reduced CTC frequency from 28.8 ± 2.2 to 0.4 ± 0.4 min⁻¹ (Fig.2Ci, P=0.0002, n=5) and reduced the number of Ca²⁺ firing sites (Fig.2Cii, P=0.0012, n=5), PTCL area (Fig.2Ciii, P=0.02, n=5), and PTCL count (Fig.2Civ, P=0.004, n=5).

Fig. 2: The effects of Antimycin and CGP-37157 on Ca²⁺ signaling in ICC-MY.

Ai Representative heat map showing summation of Ca²⁺ transient particles in ICC-MY for an entire recording under control conditions. Aii Traces of Ca²⁺ transient particle activity in the ICC-MY network in control conditions showing PTCL area (blue) and PTCL count (red). Bi Representative heat map showing summation of Ca²⁺ transient particles in ICC-MY for an entire recording in the presence of Antimycin (10 μM). Bii Traces of Ca²⁺ transient particle activity in the ICC-MY network in Antimycin (10 μM). C Summary data showing the effects of Antimycin (10 μM) on CTC frequency (i), number of Ca²⁺ sites / FOV (ii), PTCL area (iii) and PTCL count (iv), n=5. Di Representative heat map showing summation of Ca²⁺ transient particles in ICC-MY for an entire recording under control conditions. Dii Traces of Ca²⁺ transient particle activity in the ICC-MY network in control conditions showing PTCL area (blue) and PTCL count (red). Ei Representative heat map showing summation of Ca²⁺ transient particles in iCC-MY for an entire recording in the presence of CGP-37157 (30 μM). Eii Traces of Ca²⁺ transient particle activity in the ICC-MY network in CGP-37157 (30 μM). F Summary data showing the effects of CGP-37157 (30 μM) on CTC frequency (i), number of Ca²⁺ sites / FOV (ii), PTCL area (iii) and PTCL count (iv), n=6.

Mitochondrial Ca²⁺ handling can also be disrupted by blocking the mitochondrial Na⁺/Ca²⁺ exchanger (NCXmit), which can influence cytosolic Ca²⁺ signals [33]. The NCXmit blocker, CGP-37157 (30 μM), reduced CTC frequency in ICC-MY by 65% (from 25.4 ± 1.8 to 8.9 ± 1.52 min⁻¹; Fig.2D–Fi, P=0.0001, n=6) and reduced Ca²⁺ firing sites from 45.1 ± 5.3 to 23.4 ± 8.5 per FOV (Fig.2Fii, P=0.007, n=6). PTCL area and PTCL count were also reduced by CGP-37157, decreasing from 134.8 ± 8 to 26.05 ± 10.7 μm² (Fig.2Fiii, P=0.01, n=6) and 1.9 ± 0.3 to 0.3 ± 0.1 respectively (Fig.2Fiv, P=0.04, n=6)

3.2. The effect of mitochondrial blockers on ICC-DMP

Murine small intestinal muscles contain another population of ICC at the level of the deep muscular plexus (ICC-DMP). ICC-DMP do not generate or propagate slow waves, but they display Ca²⁺ transients that activate Ca²⁺-activated Cl⁻ channels and generate STICs [10,17]. ICC-DMP fire Ca²⁺ transients continuously that are stochastic in nature and do not cluster temporally to generate CTCs, as in ICC-MY [10]. An example of this continuous stochastic firing pattern is displayed in the ST map in Fig. 3Ai. We investigated whether mitochondrial inhibitors might affect Ca²⁺ signals in a similar manner as CTCs in ICC-MY.

Fig. 3: The effects of protonophores on Ca²⁺ signaling in ICC-DMP.

Ai-ii Representative ST maps of Ca²⁺ transient firing within a single ICC-DMP recorded in situ during control conditions (i) and after addition of FCCP (1 μM) (ii). Bi-iv Summary data showing the effects of FCCP (1 μM) on ICC-DMP Ca²⁺ transient frequency (i), amplitude (ii), duration (iii) and spatial spread (iv), (n=6). Ci-ii Representative ST maps of Ca²⁺ transient firing within a single ICC-DMP recorded in situ during control conditions (i) and after addition of CCCP (1 μM) (ii). Di-iv Summary data showing the effects of CCCP (1 μM) on ICC-DMP Ca²⁺ transient frequency (i), amplitude (ii), duration (iii) and spatial spread (iv), (n=6).

FCCP (1 μM) decreased the frequency of Ca²⁺ transients in ICC-DMP from 116.7 ± 12.8 min⁻¹ to 16.9 ± 7.6 min⁻¹ (Fig. 3A&Bi, P<0.0001, n=6). The duration of Ca²⁺ transients was reduced by FCCP from 268 ± 12.7 to 162 ± 39.6 ms (Fig. 3Biii, P=0.036, n=6) and the spatial spread was reduced from 6.9 ± 0.7 μm in control to 3.2 ± 0.9 μm in FCCP (Fig.3Biv, P=0.0009, n=6). The amplitude of Ca²⁺ transients was unaffected by FCCP (Fig.3Bii, P=0.11, n=6). CCCP reduced Ca²⁺ transient frequency significantly from 123 ± 21.3 to 48.5 ± 8 min⁻¹ (Fig.3Di, P=0.015, n=6). The amplitude of Ca²⁺ transients decreased from 0.4 ± 0.05 ΔF/F₀ in control to 0.3 ± 0.04 ΔF/F₀ in the presence of CCCP (Fig. 3Dii, P=0.003, n=6). Finally, the spatial spread of Ca²⁺ transients decreased in CCCP from 7.7 ± 0.55 to 5 ± 0.7 μm (Fig.3Div, P=0.006, n=6) and there was no significant effect on duration (Fig.3Diii, P=0.45, n=6).

In contrast to its effects on ICC-MY, antimycin did not abolish Ca²⁺ transients in ICC-DMP, but it reduced their frequency, amplitude and spatial spread (Fig. 4A). Antimycin (10 μM) reduced Ca²⁺ transient frequency in ICC-DMP by ~55%, from 76.6 ± 9.1 to 33.6 ± 8.5 min⁻¹ (Fig.4Bi, P=0.0004, n=6) and reduced the amplitude of Ca²⁺ transients from 0.4 ± 0.04 to 0.26 ± 0.04 ΔF/F₀ (Fig.4Bii, P=0.03, n=6). Antimycin decreased the spatial spread of Ca²⁺ transients from 6.5 ± 0.2 to 4.8 ± 0.5 μm (Fig.4Biv, P=0.04, n=6) and there were no significant effects on duration (Fig.4Biii, P=0.76, n=6).

Fig. 4: Effects of Antimycin and CGP-37157 on Ca²⁺ signaling in ICC-DMP.

Ai-ii Representative ST maps of Ca²⁺ transient firing within a single ICC-DMP recorded in situ during control conditions (i) and after addition of Antimycin (10 μM) (ii). Bi-iv Summary data showing the effects of Antimycin (10 μM) on ICC-DMP Ca²⁺ transient frequency (i), amplitude (ii), duration (iii) and spatial spread (iv), (n=5). Ci-ii Representative ST maps of Ca²⁺ transient firing within a single ICC-DMP recorded in situ during control conditions (i) and after addition of CGP-37157 (30 μM) (ii). Di-iv Summary data showing the effects of CGP-37157 (30 μM) on ICC-DMP Ca²⁺ transient frequency (i), amplitude (ii), duration (iii) and spatial spread (iv), (n=5).

The NCXmit blocker CGP-37157 (30 μM) caused a ~65% reduction in CTC frequency in ICC-MY as well as significant decreases in other CTC parameters such as PTCL area and PTCL count. However, the effects of CGP-37157 (30 μM) on Ca²⁺ transients in ICC-DMP were much less potent (Fig.4C). CGP-37157 (30 μM) had no significant effects on Ca²⁺ transient amplitude (Fig.4Dii, P=0.24, n=5), duration (Fig.4Diii, P=0.93, n=5), or spatial spread (Fig.4Div, P=0.98, n=5), in ICC-DMP. However, there was a ~38% decrease in Ca²⁺ transient firing frequency from 64.9 ± 13.4 in control to 40.3 ± 9.8 min⁻¹ in CGP-37157 and this was found to be significant (Fig.4Di, P=0.046, n=5).

We also tested the effects of a mitochondrial uniporter antagonist, Ru360 [34] on ICC. In ICC-MY, Ru360 (10 μM) caused a small but significant decrease in CTC frequency from 29.5 ± 1.1 min⁻¹ to 19.3 ± 4.3 min⁻¹ (Fig.5A–C, P=0.043, n=5), while having no significant effects on the number of Ca²⁺ firing sites (Fig. 5Cii, P=0.2, n=5), PTCL area (Fig. 5Ciii, P=0.22, n=5) or PTCL count (Fig. 5Civ, P=0.12, n=5). In ICC-DMP, Ru360 (10 μM) had no significant effects on Ca²⁺ transient frequency (Fig. 5D–Ei, P=0.07, n=5), amplitude (Fig. 5Eii, P=0.67, n=5), duration (Fig. 5Eiii, P=0.92, n=5) or spatial spread (Fig. 5Eiv, P=0.99, n=5).

Fig. 5: Effects of Ru360 on Ca²⁺ signaling in ICC-MY and ICC-DMP.

Ai Representative heat map showing summation of Ca²⁺ transient particles in ICC-MY for an entire recording under control conditions. Aii Traces of Ca²⁺ transient particle activity of the ICC-MY network in control conditions showing PTCL area (blue) and PTCL count (red). Bi Representative heat map showing the summation of Ca²⁺ transient particles in ICC-MY for an entire recording in the presence of Ru360 (10 μM). Bii Traces of Ca²⁺ transient particle activity of the ICC-MY network in Ru360 (10 μM). C Summary data showing the effects of Ru360 (10 μM) on CTC frequency (i), number of Ca²⁺ sites / FOV (ii), PTCL area (iii) and PTCL count (iv), n=5. Di-ii Representative ST maps of Ca²⁺ transient firing within a single ICC-DMP recorded in situ during control conditions (i) and after addition of Ru360 (10 μM) (ii). Ei-iv Summary data showing the effects of Ru360 (10 μM) on ICC-DMP Ca²⁺ transient frequency (i), amplitude (ii), duration (iii) and spatial spread (iv), (n=5).

The inhibitory effects of mitochondrial active substances on ICC Ca²⁺ signalling were unlikely due to ATP depletion as the ATP synthase inhibitor oligomycin (1 μM), failed to have significant effects on ICC-MY CTC frequency (Fig.6A–Ci, P=0.39, n=6), number of Ca²⁺ sites (Fig.6Cii, P=0.19, n=6), PTCL area (Fig.6Ciii, P=0.76, n=6) or PTCL count (Fig.6Civ, P=0.92, n=6). Similarly, oligomycin (1 μM) had no significant effects on Ca²⁺ transient frequency (Fig.6D–Ei, P=0.85, n=6), amplitude (Fig.6Eii, P=0.06, n=6), duration (Fig.6Eiii, P=0.08, n=6) or spatial spread in ICC-DMP (Fig.6Eiv, P=0.43, n=6).

Fig. 6: Responses to mitochondrial blockers are not due to ATP depletion.

Ai Representative heat map showing summation of Ca²⁺ transient particles in ICC-MY for an entire recording under control conditions. Aii Traces of Ca²⁺ transient particle activity of the ICC-MY network in control conditions showing PTCL area (blue) and PTCL count (red). Bi Representative heat map showing the summation of Ca²⁺ transient particles in ICC-MY for an entire recording in the presence of Oligomycin (1 μM). Bii Traces of Ca²⁺ transient particle activity of the ICC-MY network in Oligomycin (1 μM). C Summary data showing the effects of Oligomycin (1 μM). on CTC frequency (i), number of Ca²⁺ sites / FOV (ii), PTCL area (iii) and PTCL count (iv), n=6. Di-ii Representative ST maps of Ca²⁺ transient firing within a single ICC-DMP recorded in situ during control conditions (i) and after addition of Oligomycin (1 μM) (ii). Ei-iv Summary data showing the effects of Oligomycin (1 μM) on ICC-DMP Ca²⁺ transient frequency (i), amplitude (ii), duration (iii) and spatial spread (iv), (n=6).

3.3. Effects of antimycin and FCCP on store operated Ca²⁺ entry

It has been reported that mitochondria can influence store operated calcium entry (SOCE) by modulating the Ca²⁺ sensitization state of Orai channels [35,36]. We thus tested if FCCP or antimycin affected SOCE by examining their effects on Ca²⁺ release activated Ca²⁺ current (I_CRAC). Patch clamp experiments were performed on HEK 293 cells with stable expression of Orai1 and Stim1 to determine the effects of FCCP and antimycin on I_CRAC. Cells were treated with thapsigargin (1 μM) in external Ca²⁺-free PSS for 10 min to passively deplete ER Ca²⁺ stores. Cs⁺-rich pipette solution containing BAPTA was used in these experiments, and cells were held at −50 mV. Under these conditions, restoration of 2 mM [Ca²⁺]_o induced inward current (Fig.7A&C). Ramp potentials (from −80 to +80 mV) were applied before and after adding [Ca²⁺]_o. [Ca²⁺]_o (2 mM) induced I_CRAC, increasing inward current from −4.4 ± 0.4 to −88.9 ± 5.1 pA at −80 mV (n = 6, P<0.001). FCCP (1 μM) had no effect on I_CRAC (−89.0 ± 5.3 pA, n=6, P=0.94, Fig.7B&E). After activation of I_CRAC by restoration of 2 mM [Ca²⁺]_o (−4.8 ± 0.3 pA in 0 mM [Ca²⁺]_o, −86.2 ± 2.8 pA in 2 mM [Ca²⁺]_o; n=6, p<0.001), antimycin (10 μM) did not inhibit I_CRAC (−84.8 ± 2.9 pA; n=6, P=0.29; Fig.7D&E). These data suggested that FCCP and antimycin did not exert their inhibitory effects on Ca²⁺ transients in ICC by decreasing Ca²⁺ entry via SOCE and depletion of Ca²⁺ stores.

Fig. 7: Effect of FCCP and antimycin on I_CRAC.

I_CRAC was measured in HEK 293 cells expressing Orai1 and Stim1 using whole-cell voltage-clamp (holding potential = −50 mV). Cells were treated with Ca²⁺-free PSS containing thapsigargin (1 μM) for 10 min to deplete ER Ca²⁺. A shows the response elicited by reintroduction of 2 mM [Ca²⁺]_o: an inward current was activated and FCCP (1 μM) had no effect on this current. Ramp potentials from −80 to +60 mV were run during the experiment (examples noted by a, b and c in th trace in A) B shows responses to ramp potentials in Panel A (a is control, black trace, b is after addition of 2 mM [Ca²⁺]_o, red trace, and c is after addition of FCCP (1 μM), green trace). C shows Antimycin had no effect on the current activated by reintroduction of 2 mM [Ca²⁺]_o. D shows responses to ramp potentials in Panel C (a is control, black trace, b is after addition of 2 mM [Ca²⁺]_o, red trace, and c is after addition of Antimycin A (10 μM), green trace). E Summary data showing the effects of FCCP and Antimycin A on I_CRAC at −80 mV (n=6, ***p<0.001 vs 0 mM [Ca²⁺]_o; ns, not significant).

3.4. Effects of CCCP, Antimycin and CGP-37157 on ANO1 and Ca_v3.2 currents

Protonophores can have off target effects, including indirect effects on the Na/K pump [37], depolarization due to the activation of proton and Na⁺ pumps [38], impairment of lyosomal function [39] and changes in intracellular pH and generation of free radicals [40]. In ICC-MY, the propagation of CTCs relies on the activation of a voltage dependent Ca²⁺ conductance mediated by T-type Ca²⁺ channels [16,41], and CaV3.2 is the T-type conductance most highly expressed in ICC [41]. T-type Ca²⁺ channel blockers inhibit slow wave propagation and CTCs in ICC-MY [16,42–44], in a similar manner to the blockade seen in the present study using antimycin, protonophores and CGP-37157. The coordinated nature of CTCs is also dependent on ANO1 activation, as previous studies have shown that pharmacological blockade of ANO1 leads to a loss of slow wave propagation and a disruption in the generation of coordination of Ca²⁺ signals in ICC-MY [14,45]. We next tested whether the inhibition of Ca²⁺ transients in ICC by the ETC inhibitor, protonophores and NCXmit drugs might be explained by non-specific effects of these compounds on either Cav3.2 or ANO1 currents or a combination of both.

HEK 293 cells transfected with ANO1-AC were dialyzed with 100 nM Ca²⁺ and displayed outwardly-rectifying currents at positive potentials (Fig.8A&B; [30]). Antimycin (1&10 μM), which abolished Ca²⁺ signals in ICC-MY, reduced ANO1 currents from 49.8 ± 5.2 pA/pF to 21.8 ± 6.0 pA/pF and 4.5 ± 0.8 pA/pF at +70 mV, respectively (n=5, P<0.01 for each concentration of antimycin; Fig.8A&B). The protonophore CCCP (1–3 μM) slightly, but not significantly, increased ANO1 current (Fig.8C&D). The NCXmit blocker, CGP-37157 (10, 30 and 50 μM) inhibited ANO1 currents from 61.0 ± 5.1 pA/pF to 34.9 ± 3.8, 15.3 ± 2.5 pA/pF and 5.3 ± 1.8 pA/pF at +70 mV, respectively (n=5, P<0.01 for each concentration of CGP-37157; Fig.8E&F). Ru360 (10 μM), had no significant effect on ANO1 current (data not shown, n=6, p=0.07).

Fig. 8: The effect of antimycin, CCCP and CGP-37157 on Ano1 currents.

Ano1 (splice variant AC) expressed in HEK 293 cells were voltage-clamped and stepped from −80 mV to +70 mV in 10 mV increments (from holding potential = −80 mV). A Representative currents were evoked with 100 nM Ca²⁺ pipette solution. The currents were inhibited by 1 μM antimycin and blocked by 10 μM antimycin. B I-V relationships for the effects of antimycin on Ano1-AC currents. C Representative currents evoked in HEK 293 cells with 100 nM Ca²⁺ pipette solution were not affected significantly by 1 μM and 3 μM CCCP. D I-V relationships for the effects of CCCP on Ano1-AC currents. E Representative currents were evoked in HEK 293 cells expressing Ano1-AC with a 100 nM Ca²⁺ pipette solution. These currents were inhibited by CGP-37157 in a concentration-dependent manner and summary data are plotted in F.

Ca_V3.2 channels stably expressed in HEK 293 cells displayed T-type Ca²⁺ currents (Fig.8A). Maximum current was evoked at −40 mV. Ca_V3.2 currents were reduced by CCCP and FCCP (Fig.9A–D), from −54.6 ± 3.7 pA/pF to −26.9 ± 6.0 pA/pF by CCCP (3 μM; n=7, P<0.01, Fig.9C) and from −40.4 ± 7.8 pA/pF to −12.7±3.6 pA/pF by FCCP (3 μM; n=6, P<0.01, Fig.9D) at −40 mV, respectively. CGP-37157 decreased Ca_V3.2 currents from −45.4 ± 8.1 to −7.0 ± 2.0 pA/pF and to −1.3 ± 0.5 pA/pF by 10 μM and 30 μM at −40 mV, respectively (n=5, P<0.01 for each dose; Fig.9E&F). Ru360 (10 μM) decreased Ca_V3.2 currents from −57.6 ± 8.0 to −36.3 ± 7.5 pA/pF at −40 mV F (n=5, P<0.05 Fig.9G&H).

Fig. 9: The effects of CCCP, FCCP, CGP-37157 and Ru360 on Cav3.2 currents.

HEK 293 cells expressing Ca_v3.2 were stepped from −80 mV to +60 mV in 10 mV increments from a holding potential of −80 mV. A Representative currents were evoked using a Cs⁺-rich pipette solution. Ca_v3.2 currents were reduced by CCCP dose-dependently. B Ca_v3.2. currents were also reduced by FCCP dose-dependently. C I-V relationships showing the effects of CCCP on Ca_v3.2 currents. D I-V relationships showing the effects of FCCP on Ca_v3.2 currents. E Representative currents were evoked by Cs⁺-rich pipette solution in HEK 293 cells expressing Ca_v3.2. Currents were reduced by CGP-37157 dose-dependently. F I-V relationships for the effects of CGP-37157 on Ca_v3.2 currents. G Representative currents were evoked by Cs⁺-rich pipette solution in HEK 293 cells expressing Ca_v3.2. Currents were reduced by Ru360. H I-V relationships for the effects of Ru360 on Ca_v3.2 currents.

3.5. Use-dependent block of Ca_v3.2 currents by Antimycin

A repetitive step protocol (steps from −80 to −10 mV every 2 sec) was used to evoke Ca_V3.2 currents (Fig.9A). This protocol was designed to simulate the depolarizations experienced by ICC in intact muscles due to slow wave activity [43]. The Ca_V3.2 currents in HEK 293 cells did not run down significantly even when depolarizing test potentials were applied repeatedly every 2 sec for several minutes (Fig.10A). Antimycin (10 μM) caused marginal inhibition of Ca_V3.2 currents at room temperature in 3 of 6 cells (Fig.10B). However, when temperature was increased to 30^o C and the repetitive stepping protocol was repeated (Fig.10C&D), Ca_V3.2 currents were inhibited by antimycin within a few minutes (10 μM; n=4).

Fig. 10: The effects of antimycin on Ca_v3.2 currents.

Repetitive step depolarizations were applied to HEK 293 cells expressing Ca_v3.2. The cells were stepped from −80 mV to −10 mV for 500 ms every 2 sec. A Representative currents recorded at room temperature (RT) showing activation of T-type Ca²⁺ currents at 0, 1 and 3 min of repetitive step depolarizations. The first trace shows the superposition of all current responses. Note little or no run-down in the currents during the repetitive step depolarizations. B Representative currents recorded at room temperature (RT) after addition of antimycin (10 μM). Under these conditions antimycin had little effect on Ca_v3.2 currents. C Representative control currents recorded at 30^o C showing T-type Ca²⁺ currents over time using the repetitive stepping protocol. D Representative currents recorded at 30^o C in the presence of antimycin (10 μM). Under these conditions, which better simulate the conditions in which antimycin is applied to intact tissues (see Figure 2), antimycin caused a use-dependent block of Ca_v3.2 currents.

4. Discussion

In this study we examined the effects of pharmacological antagonists of mitochondrial function, such as protonophores, ETC inhibitors and a blocker of the mitochondrial Na⁺/Ca²⁺ exchanger (NCXmit) on Ca²⁺ transients in murine small intestinal ICC. We found that FCCP and CCCP, antimycin and CGP-37157 potently blocked CTC activity in ICC-MY networks. These compounds also inhibited Ca²⁺ transients in ICC-DMP, however the effects were noticeably less striking. Using patch clamp recordings, we found that these pharmacological agents did not block SOCE current (I_CRAC) but antagonized both Ano1 and Ca_V3.2 conductances, both of which are required for the pacemaker function of ICC in the small intestine [11,12,14,16,41,45].

Ca²⁺ uptake into mitochondria occurs through a low sensitivity mitochondrial uniporter, implying that mitochondria act more efficiently as a Ca²⁺ repository when positioned within or near microdomains of Ca²⁺ release [21,46–48]. Ca²⁺ uptake into mitochondria can potentiate or inhibit Ca²⁺ signals from the ER either by reducing Ca²⁺ levels at ER release channels (RyRs or IP₃Rs) sufficiently to prevent their activation or by preventing Ca²⁺ induced inhibition of these channels [40,49–54]. Mitochondrial Ca²⁺ uptake may also affect Ca²⁺ influx through store-operated Ca²⁺ entry (SOCE) by modulating Ca²⁺ inhibition of Orai Ca²⁺ channels [35,36].

Previous studies showed that FCCP, CCCP and antimycin inhibited electrical slow waves in the mouse small intestine, guinea-pig stomach and canine colon [23]. These findings, and the morphological observations that ICC contain an abundance of mitochondria [19,20], suggested that Ca²⁺ handling in mitochondria may, in some manner, regulate Ca²⁺ release events from ER stores in ICC. In addition to generation of pacemaker activity, other studies also suggested that mitochondrial Ca²⁺ handling is also important in regeneration (i.e. active propagation) of slow waves through the ICC-MY network and in the discharge of STICs and their voltage-responses, spontaneous transient depolarization (aka unitary potentials) in canine colon [23,24], mouse colon [25], gastric antrum [26,55] and stomach [56]. Outside of the GI tract, FCCP, CCCP, CGP-37157 and antimycin have also been shown to inhibit Ca²⁺ signals in ICC-like cells in the urethra [57,58], gallbladder [59] and vasculature [60].

We found that FCCP and CCCP potently blocked CTCs in ICC-MY. However, these protonophores are known to have non-specific effects on a variety of cellular functions other than mitochondrial Ca²⁺ handling, such as indirect effects on the Na/K pump [37], depolarization due to the activation of proton and Na⁺ pumps [38], impairment of lyosomal functions [39] and changes in intracellular pH and generation of free radicals [40]. In the small intestine, CTC and slow wave propagation depend upon activation of T-type Ca²⁺ channels [16,41–44]. Ca²⁺ influx into ICC is thought to initiate Ca²⁺ induced Ca²⁺ release from multiple ER Ca²⁺ release sites, and this sustains activation of Ano1 channels forming the slow wave plateau [16]. Regeneration of slow waves in one cell leads to further depolarization in coupled cells and active propagation through the ICC-MY network. Control experiments were performed to ascertain the effects of FCCP and CCCP on Ano1 and Cav3.2 currents expressed in HEK 293 cells. While neither of these compounds significantly affected Ano1 currents, they were effective in antagonizing Cav3.2 currents. Agents that block T-type Ca²⁺ channels inhibit slow waves in the small intestine [16,42–44]. The protonophores mimicked the effects of T-type Ca²⁺ channels blockers [16], and therefore the inhibitory effects of these compounds on CTCs and slow waves in ICC-MY could be due to blockade of T-type Ca²⁺ channels.

Other mitochondrial drugs had confounding effects on key ionic conductances responsible for slow waves in ICC-MY. Antimycin abolished CTCs in ICC-MY in the majority of tissues, however we also found that this compound has significant inhibitory effects on Ano1 currents. Previously studies have shown that Ca²⁺ signalling in ICC-MY is disrupted when Ano1 is pharmacologically blocked [45] or deleted genetically [14,45]. Thus, inhibitory effects of antimycin on CTCs could be due to its effects on Ano1 channels in ICC. Furthermore, we found that antimycin had minor effects on Ca_V3.2 currents at room temperature. However, T-type Ca²⁺ channels are quite sensitive to temperature [61], and when the temperature in our experiments was increased, the degree of block increased dramatically. In addition, ICC-MY generate about 30 slow waves per min. Thus, T-channels are activated and cells depolarize on average about every 2 sec.

When a voltage-clamp protocol with these characteristics was utilized (i.e. depolarization steps applied every 2 secs), antimycin caused block of Ca_V3.2 currents (i.e. use-dependent block) within the time required for this compound to block slow waves in intact muscles. The NCXmit blocker CGP-37157 has also been shown to inhibit the electrical activity of intestinal ICC [62], and other reports suggest it has effects on Ca²⁺ signalling in ICC-MY [63]. In the rabbit urethra, CGP-37157 was reported to inhibit Ca²⁺ shuttling between the mitochondria and the cytosolic face of ER Ca²⁺ release channels [58]. However, similar to FCCP and CCCP, we found that CGP-37157 had significant inhibitory effects on Ca_V3.2 currents. These data are also consistent with the idea that the inhibitory effects of CGP-37157 could be due to its channel blocking effects and not due to uncoupling of mitochondrial Ca²⁺ handling from Ca²⁺ release events in ICC-MY. Ru360, a compound often considered to be a selective antagonist of the mitochondrial uniporter, had minor effects on ICC-MY in comparison to the other mitochondrial inhibitors tested. Ru360 caused a 35% reduction in CTC frequency without affecting other properties of CTCs (Fig.5). This small effect might have been accounted for by the 37.5 % reduction in Cav3.2 currents caused by Ru360 (Fig.9). It should be noted that T-channel blockers reduce the frequency of electrical slow waves in gastrointestinal muscles, but a significant safety factor exists such that significant antagonism of T-channels is necessary for slow wave block [16,64].

We also found that the mitochondrial drugs tested had inhibitory effects on Ca²⁺ transients in ICC-DMP. The inhibitory effects of FCCP, CCCP, antimycin and CGP-37157 were significant in ICC-DMP, but did not approach the degree of blockade exhibited on CTCs in ICC-MY. For example, antimycin only reduced (and did not abolish) the frequency, amplitude and spatial spread of transients in ICC-DMP, and CGP-37157 reduced Ca²⁺ transient frequency in ICC-MY by 65% but only by 38% in ICC-DMP. One difference between Ca²⁺ transients in ICC-DMP vs. ICC-MY is that ICC-DMP fire Ca²⁺ transients in a continuous, stochastic manner, and voltage-dependent mechanisms appear to have little influence on the firing pattern [10]. ICC-DMP show minimal expression of voltage-gated Ca²⁺ channels [65], and thus the inhibitory effects of mitochondrial blockers are unlikely to be attributable to inhibition of Ca_V3.2 currents, as apparently the case in ICC-MY. Similarly, the lack of cell-to-cell coordination between firing of Ca²⁺ transients in adjacent ICC-DMP suggests that blocking Ano1 would also not have effects on Ca²⁺ transients in ICC-DMP. Finally, we found no effect of the mitochondrial drugs on Orai currents (I_CRAC), suggesting that the inhibitory effects of these drugs is not due to unloading of ER Ca²⁺ after application of the drugs. Thus, it is possible that the potent effects of the mitochondrial drugs on ICC-MY is largely due to block of key membrane conductances, but in the absence of a primary role for these conductances in regulating Ca²⁺ transients, as in ICC-DMP, more accurate effects of mitochondrial involvement are unmasked. It is also possible that other processes, fundamental to Ca²⁺ release in ICC (e.g. Ca²⁺ uptake into stores, Ca²⁺ release from ryanodine and IP₃ receptors) might also be affected non-selectively by these mitochondrial drugs.

A shortcoming of this study is that we were unable to identify a mitochondrial drug that has been certified to disrupt mitochondrial Ca²⁺ handling that did not have non-specific effects on other processes upon which ICC depend. Thus, the exact role of mitochondria in pacemaker activity, if any, remains uncertain. It appears that all previous work on mitochondrial function in ICC have relied upon some or all of the pharmacological tools tested in the current study [23–25,55,56,62,63] and therefore models of pacemaker activity that have incorporated mitochondrial Ca²⁺ uptake as a contributing variable [66–69] may require reconsideration. However, the fact remains that ICC contain an abundance of mitochondria [18–20], and thus mitochondria may be important in sustaining the rigorous electrical behaviours of pacemaker cells. For example, these organelles may be critical for long-term metabolic maintenance of slow waves, as others suggested in past evaluations of pacemaker activity [70–72]. Additional experiments, including more selective deactivation of mitochondrial Ca²⁺ handling (for example mice lacking the mitochondrial uniporter [73]) will be needed to further resolve the role of these organelles in pacemaking in GI muscles. The use of mitochondrial-targeted opsins may also be used to influence mitochondrial metabolism and Ca²⁺ signalling in ICC [74]. Furthermore, the development of genetically encoded mitochondrial Ca²⁺ indicators that fluoresce in the red spectrum [75] could be combined with simultaneous recordings of GCaMP expressed in ICC, allowing detailed study of the relationship between mitochondrial and cytosolic Ca²⁺ in pacemaker activity.

Highlights.

• Interstitial cells of Cajal (ICC-MY) are pacemakers that generate and propagate electrical slow waves in gastrointestinal (GI) muscles.

• Mitochondrial Ca²⁺ handling is currently thought to be critical for ICC pacemaking as mitochondrial inhibitors blocked slow waves in several GI muscles, but their effects on Ca²⁺ signalling in ICC are largely unknown.

• We utilized in situ Ca²⁺ imaging of mice expressing a genetically encoded Ca²⁺ indicator (GCaMP3) selectively in ICC to determine the effects of mitochondrial drugs on ICC Ca²⁺ transients in the small intestine.

• FCCP, CCCP, antimycin, a mitochondrial Na⁺/Ca²⁺ exchange inhibitor, CGP-37157 and a mitochondrial uniporter blocker, Ru360, inhibited Ca²⁺ transients in ICC-MY.

• Using patch-clamp experiments we found that FCCP, CCCP, antimycin and CGP-37157 inhibited key pacemaker conductances essential for slow wave generation and propagation in ICC (T-type Ca²⁺ (Ca_V3.2) and Ano1).

• The inhibitory effects of mitochondrial drugs on slow waves and Ca²⁺ signalling in ICC can be explained by direct antagonism of key pacemaker conductances in ICC that generate and propagate slow waves.

Abbreviations

CTC

Ca²⁺ transient cluster

ETC

Electron transport chain

FOV

Field of view

GI

Gastrointestinal

ICC

Interstitial Cells of Cajal

ICC-DMP

Interstitial cells of Cajal at the level of the deep muscular plexus

ICC-IM

Intramuscular interstitial cells of Cajal

SIP syncytium

Electrical syncytium formed by smooth muscle cells, ICC and PDGFRα⁺ cells in GI muscles

GCaMP

Genetically encoded Ca²⁺ indicator composed of a single GFP

IP₃R

Inositol triphosphate receptor

KRB

Krebs Ringer Bicarbonate

PDGFRα

Platelet derived growth factor receptor α

PTCL

Ca²⁺ transient particle

ROI

Region of interest

RyR

Ryanodine receptor

SIP syncytium

Smooth muscle, ICC, PDGFRα cells forming an electrical syncytium