Intracellular signalling in arterial chemoreceptors during acute hypoxia and glucose deprivation: role of ATP

J Physiol 603.5 (2025) pp 1091–1107 1091 The Journal of Physiology Intracellular signalling in arterial chemoreceptors during acute hypoxia and glucose deprivation: role of ATP María Torres-López1,2 , Patricia González-Rodríguez1,2,3 , Olalla Colinas1,2,3 ,Hee-SoolRho 1, Hortensia Torres-Torrelo1,2 ,AntonioCastellano 1,2 ,LinGao 1,2,3, Patricia Ortega-Sáenz1,2,3 and José López-Barneo1,2,3 1Instituto de Biomedicina de Sevilla (IBiS), Hospital Universitario Virgen del Rocío/CSIC/Universidad de Sevilla, Sevilla, Spain 2Departamento de Fisiología Médica y Biofísica, Facultad de Medicina, Universidad de Sevilla, Sevilla, Spain 3Centro de Investigación Biomédica en Red sobre Enfermedades Neurodegenerativas (CIBERNED) Handling Editors: Harold Schultz & Andrew Holmes The peer review history is available in the Supporting Information section of this article (https://doi.org/10.1113/JP287130#support-information-section) Abstract figure legend Schematic representation of the distinct mechanisms involved in acute oxygen and glucose sensing by carotid body glomus cells. M. Torres-López and P. González-Rodríguez contributed equally to this work © 2025 The Author(s). The Journal of Physiology published by John Wiley & Sons Ltd on behalf of The Physiological Society. DOI: 10.1113/JP287130 This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. 1092 M. Torres-López and others J Physiol 603.5 Abstract The carotid body (CB) is the main oxygen (O2) sensing organ that mediates reflex hyperventilation and increased cardiac output in response to hypoxaemia. Acute O2sensing is an intrinsic property of CB glomus cells, which contain special mitochondria to generate signalling molecules (NADH and H2O2)thatmodulatemembraneK +channels in response to lowered O2tension (hypoxia). In parallel with these membrane-associated events, glomus cells are highly sensitive to mitochondrialelectrontransportchain (ETC)inhibitors.Itwassuggestedthatadecreaseinoxidative production of ATP is a critical event mediating hypoxia-induced cell depolarization. Here, we show that rotenone [an inhibitor of mitochondrial complex (MC) I] activates rat and mouse glomus cells but abolishes their responsiveness to hypoxia. Rotenone does not prevent further activation of the cells by cyanide (a blocker of MCIV) or glucose deprivation. Responsiveness to glucose deprivation is enhanced in O2-insenstive glomus cells with genetic disruption of MCI. These findings suggest that acute O2sensing requires a functional MCI but that a decrease in intracellular ATP, presumably produced by the simultaneous inhibition of MCI and MCIV, is not involved in hypoxia signalling. In support of this concept, ATP levels in single glomus cells were unaltered by hypoxia, but rapidly declinedfollowingexposureofthecellstolowglucoseortoinhibitorsofoxidativephosphorylation. These observations indicate that a reduction in intracellular ATP does not participate in physiological acute O2sensing. However, local decreases in ATP of glycolytic origin may contribute to low glucose signalling in glomus cells. (Received 18 June 2024; accepted after revision 27 January 2025; first published online 11 February 2025) Corresponding authors J. López-Barneo and P. Ortega-Sáenz: Instituto de Biomedicina de Sevilla (IBiS), Hospital Universitario Virgen del Rocío/CSIC/Universidad de Sevilla, Sevilla, Spain. Email: [email protected] and gor[email protected] Key points rThe carotid body contains oxygen-sensitive glomus cells with specialized mitochondria that generate signalling molecules (NADH and H2O2)toinhibitmembraneK +channels in response to hypoxia. rGlomus cells are highly sensitive to electron transport chain (ETC) blockers. It was suggested that a decrease in intracellular ATP is the main signal inducing K+channel inhibition and depolarization in response to hypoxia or ETC blockade. rRotenone,aninhibitorofmitochondrial complex(MC)I, activatesglomuscellsbutabolishestheir responsiveness to hypoxia. However, rotenone doesnotpreventfurtheractivationofglomuscells by cyanide (an MCIV blocker) or glucose deprivation. rSingle-cell ATP levels were unaltered by hypoxia, but decreased rapidly following exposure of glomus cells to 0 mM glucose or inhibitors of oxidative phosphorylation. rA reduction in intracellular ATP does not participate in signalling acute hypoxia. However, it may contribute to hypoglycaemia signalling in glomus cells. 0María Torres-López got her Degree in Biochemistry and Master’s in Physiology and Neuroscience at the University of Seville. She moved to the Institute of Biomedical Research of Barcelona to study axonal neurodegeneration in prodromal Parkinson’s disease. From 2020, she joined the Institute of Biomedicine of Seville to work on the molecular mechanism underlying O2sensing by carotid body glomus cells as part of her PhD project. Patricia Gonzalez-Rodriguez is a neuroscientist and physiologist whose research focuses on oxygen sensing, mitochondrial dysfunction and Parkinson’s disease. Dr González-Rodríguez obtained her degree in Biology and completed her PhD at the University of Seville, Spain. Her postdoctoral training has been conducted at four research institutions across three countries: EPFL in Switzerland, Northwestern University in Chicago (USA), the University of Minnesota Medical Center in Minneapolis (USA) and the University of Seville in Spain. © 2025 The Author(s). The Journal of Physiology published by John Wiley & Sons Ltd on behalf of The Physiological Society. 14697793, 2025, 5, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/JP287130 by Readcube (Labtiva Inc.), Wiley Online Library on [16/06/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License J Physiol 603.5 Intracellular signalling in arterial chemoreceptors 1093 Introduction The carotid body (CB) is the main arterial chemoreceptor and prototypical acute oxygen (O2)sensing organ in mammals that is responsible for the hypoxic ventilatory response (HVR). It is composed of clusters of glomus cells that are innervated by sensory nerve fibresandareinclosecontactwithcapillaries(for detailed reviews see Iturriaga et al., 2021; Ortega-Sáenz & López-Barneo, 2020). Glomus cells are electrically excitable and release neurotransmitters in response to lowered O2tension (hypoxia), thereby activating afferent fibres of the petrosal ganglion synapsing on brain stem respiratory centres (see López-Barneo, 2022). Hypoxia inhibits the activity of background and voltage-dependent K+channels in glomus cells (Buckler, 1997; Delpiano & Hescheler, 1989; Kim et al., 2009; Lopez-Barneo et al., 1988; López-López et al., 1989; Peers, 1990; Pérez-García et al., 2004; Stea & Nurse, 1991), leading to membrane depolarization, extracellular Ca2+influx and neurotransmitter secretion (Buckler & Vaughan-Jones, 1994; López-Barneo et al., 1993; Ureña et al., 1994). In addition to these membrane-associated events, it is known that CBs are strongly activated by mitochondrial inhibitors (for references see Mulligan & Lahiri, 1981; Wilson et al., 1994) and that several mitochondrial parameters (e.g. membrane potential or NADH levels) in dissociated glomus cells are modulated by changes in O2tension (Buckler & Turner, 2013; Duchen & Biscoe, 1992a, b). Similar to hypoxia, the activation of glomus cells (increased cytosolic Ca2+or secretory activity) by mitochondrial electron transport chain (ETC) blockers requires an influx of extracellular Ca2+, indicating that mitochondria can signal membrane ion channels (Ortega-Sáenz et al., 2003; Wyatt & Buckler, 2004). Over the past few years, studies on genetically modified mouse models have provided strong evidence to support a comprehensive mitochondrial-to-membrane signalling (MMS) pathway of acute O2sensing by glomus cells, which combines membrane and mitochondrial responses to hypoxia. CB glomus cells constitutively express high levels of HIF2α(Gao et al., 2017; Tian et al., 1998), which induces the expression of atypical mitochondrial complex (MC) IV subunit isoforms (Kadenbach & Huttemann, 2015) conferring upon mitochondria a high sensitivity to changes in O2tension (see Bishop & Ratcliffe, 2020; Colinas et al., 2023; Moreno-Domínguez et al., 2020). Hypoxia decreases MCIV activity and causes a backlog of electrons along the ETC, resulting in an increase in the reduced status of MCIII and in the ratio of reduced coenzyme Q (QH2) to coenzyme Q (Q). These dynamic changes slow down MCI NADH dehydrogenase activity andresultinanaccumulationofNADHandanincreased production of reactive oxygen species (ROS), generated in MCIII and MCI and converted to H2O2by superoxide dismutases. NADH and H2O2are mitochondrial signalsthatafterequilibrationwiththecytosolmodulate ion channel function (Arias-Mayenco et al., 2018; Fernández-Agüera et al., 2015; Jiménez-Gómez et al., 2023; Moreno-Domínguez et al., 2020). An unsolved mechanistic question regarding the MMS model of acute O2sensing is whether or not hypoxia causes a decrease in cytosolic ATP levels in glomus cells (for a detailed discussion, see López-Barneo & Ortega-Sáenz, 2022). A priori, the role of a drop in intracellular ATP in signalling physiological acute hypoxia is counterintuitive since ATP is needed to support essential functions in active cells, such as maintenance of ionic gradients, Ca2+ channel function and neurosecretion. However, it was reported that TASK-like channels, which mediate the background K+current inhibited by hypoxia in glomus cells, are activated by intracellular ATP (Varas et al., 2007). Therefore, it was suggested that a decrease in cytosolic ATP during hypoxia is the signal responsible for the hypoxia-induced decrease in K+conductance and cell depolarization (Wyatt & Buckler, 2004). Challenging this proposal, it was found that CB glomus cells from mice lacking TASK1 and TASK3 channel subunits that form ATP-activated background K+channels (Kim et al., 2009) have normal secretory responses to hypoxia (Ortega-Sáenz et al., 2010). In addition, several groups have shown that modulation of voltage-dependent K+channels by hypoxia is maintained in glomus cells dialysed with constant (3–5 mM) levels of intracellular ATP (López-López et al., 1989; Pérez-García et al., 2004). ToclarifythepotentialroleofATPinsignalling acute hypoxia, we have herein analysed the secretory responsesofglomuscells,inducedbyinhibitorsofthe mitochondrial ETC and oxidative metabolism (such as rotenone and cyanide), in conjunction with CB stimuli (hypoxia or glucose deprivation) which may compromise oxidative phosphorylation and ATP synthesis. We have also studied the bioenergetic characteristics of rat glomus cells and directly monitored single-cell changes in cytosolic ATP levels in response to hypoxia and other stimuli. In parallel studies we have tested whether recombinant TASK3 channel activity is modulated by mitochondria hypoxic signals (NADHandH2O2). Our data indicate that CB glomus cells have an active oxidative metabolism and that cytosolic ATP levels are decreased by either inhibition of cytochrome c oxidase with cyanide or exposure of thecellsto0mMglucose.Incontrast,acutehypoxia does not induce decreases in intracellular ATP. We also demonstrate that recombinant TASK3 channels are inhibited by intracellular NADH and H2O2.Thesedata indicate that cytosolic ATP does not participate in physiological acute O2sensing, although it may contribute to hypoglycaemia signalling in glomus cells. © 2025 The Author(s). The Journal of Physiology published by John Wiley & Sons Ltd on behalf of The Physiological Society. 14697793, 2025, 5, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/JP287130 by Readcube (Labtiva Inc.), Wiley Online Library on [16/06/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 1094 M. Torres-López and others J Physiol 603.5 Methods Ethical approval All procedures in this study were approved by the InstitutionalCommitteeforAnimalCareandUseatthe University of Seville (PN2019 07/04/2020/051). Handling of the animals was conducted in accordance with the European Community Council directive of 22 September 2010 (Directive 2010/63/EU) and the implementations of 5 June 2019 (Regulation 2019/2010) for the care and use of laboratory animals. All investigators understand the ethical principles under which the journal operates and that their work complies with the animal ethics checklist as outlined in the journal policy. For in vitro experiments,animalswerekilledbyi.p.injectionofalethal dose of anaesthesia (sodium thiopental, 120–200 mg/kg) to avoid suffering before the animals were killed and tissue dissection. Death was confirmed by the complete cessation of movement and the absence of breathing for over 2 min. Carotid bifurcations containing the CB were removed immediately after death. They were then transferred to a Petri dish with cold (∼4°C) PBS placed in the stageofastereomicroscope.Aftercarefulcleaningofthe surrounding tissues, each CB was dissected and extracted. MCI-deficient mouse model Mice (strain B6/129) carrying the floxed allele for the Ndufs2 gene were generated in our laboratory andbredwiththosecarryingCRErecombinaseunder control of the tyrosine hydroxylase (TH) promoter to generate embryonic conditional knockout mice with deletion of Ndufs2 in the catecholaminergic tissues (Fernández-Agüera et al., 2015; Jiménez-Gómez et al., 2023). Ndufs2 encodes a core subunit of mitochondrial respiratory chain NADH dehydrogenase (MCI). We used 2–4-month-old MCI-deficient mice in this work. Animal care Experimental animals used in this study were 4–8-week-oldWistarrats produced inouranimalfacilities and 2–4-month-old genetically modified MCI-deficient mice. Both female and male animals in approximately thesameproportionwereused.Animalswerehousedat 22 ±1°C on a 12 h light/12 h dark cycle with ad libitum access to food (Teklad global 14% protein, Envigo) and water. Preparation of CB slices CB slices were used to study the secretory activity of glomus cells by amperometry or to monitor the cytosolic ATP/ADP ratio and intracellular Ca2+by microfluorimetry. CB slices were prepared as previously described by our group (Gao et al., 2021; Pardal et al., 2000; Piruat et al., 2004). Briefly, dissected CBs from rats and mice were embedded in 1% (w/v) low-melting-point agarose at 42°C prepared in PBS. Sections of the carotid bodies, 150 μm thick, were cut using a vibratome (Leica, Wetzlar, Germany; VT1000S) and stored in cold PBS. Mice CB slices were digested for 5 min, at 37°C with shaking, in an enzymatic solution containing 50 μM CaCl2, 0.27 mg/ml trypsin (Sigma, St Louis, MO, USA; cat. no. #T8003), 0.6 mg/ml collagenase II (Sigma, cat. no. #C6885), and 1.25 U/ml porcine elastase (Millipore, Billerica, MA, USA; cat. no. #324682) prepared in PBS pH 7.4. Afterwards, freshly cut rat slices and digested mouse slices were washed twice withcoldPBSandincubatedat37°Cin5%CO 2for 24–48 h before use with the following culture medium: DMEM (0 glucose)/DMEM-F-12 (Gibco, Waltham, MA, USA; cat. no. #11966-025/21331-020) medium (3:1) supplemented with 100 U/ml penicillin and 10 mg/ml streptomycin (Gibco, cat. no. #15140-122), 2 mM l-glutamine (Gibco, cat. no. #25030-024), 10% fetal bovine serum (FBS; Gibco, cat. no.#10270-10), 84 U/l insulin (Actrapid, Novonordisk, Bagsværd, Denmark; cat. no. #EU/1/02/230/011) and 1.2 U/ml erythropoietin (Sandoz, Basel, Switzerland; cat. no. #EU 1/07/410/028). Dissociation of CB cells Dissociatedratglomuscellswereusedtoperform bioenergetic studies. Dissected CBs were digested in the same enzymatic solution used to digest mouse CB slices at37°Cin5%CO 2for 15 min without shaking. Then, the tissue was mechanically stretched with needles and incubated for another 5 min. Digestion was stopped by theadditionofcoldSeahorseassaymediumsupplemented with 10% FBS to the enzymatic solution. Seahorse assay medium contains Seahorse XF DMEM medium supplemented with 3.5 mM glucose, 0.1 mM pyruvate and 4.8 mM l-glutamine. When studying the effect of glucose deprivation on glomus cell ATP production, no glucose was added to the Seahorse assay medium. After centrifugation (5 min at 300 gat 4°C) the pellet was washedwith1mlofmediumandcentrifugedagain for 5 min at 300 g. Finally, cells were mechanically dispersed by pipetting and the final volume was adjusted to 70 μl. After counting, cells were seeded directly into the Seahorse PDL microplate and centrifuged at 200 gfor 2 min. Assay medium was added to each well to reach a final volume of 180 μl. Cells were incubated at 37°C for 45 min. © 2025 The Author(s). The Journal of Physiology published by John Wiley & Sons Ltd on behalf of The Physiological Society. 14697793, 2025, 5, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/JP287130 by Readcube (Labtiva Inc.), Wiley Online Library on [16/06/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License J Physiol 603.5 Intracellular signalling in arterial chemoreceptors 1095 Bioenergetic measurements in dispersed glomus cells Mitochondrial and glycolytic ATP production rate in rat CB glomus cells was determined based on oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) using Seahorse XFp Real-Time ATP rate assay kit (Agilent Technologies, Santa Clara, CA, USA) following the manufacturer’s instruction. Dissociated glomus cells were incubated in Seahorse assay medium (see‘DissociationofCBcells’).BasallevelsofOCRand ECAR and those in the presence of mitochondrial ETC inhibitors (1.5 μM oligomycin, 0.5 μM rotenone and antimycinA)werethenmeasuredinaSeahorseXFp analyser (Seahorse Bioscience) in order to calculate ATP production rate. ATP production rate was normalized by the number of cells used in each measurement. Amperometric recording of catecholamine secretion in single glomus cells Measurement of catecholamines secreted by glomus cells was performed as previously described in our laboratory (Ortega-Sáenzetal.,2003).Torecordthesecretoryactivity of glomus cells, CB slices were transferred to a recording chamber continuously perfused with extracellular solution (see ‘Recording solutions’). Secretory events were recorded with a 10 μm carbon fibre polarized at +750 mV to detect dopamine oxidation. Amperometric currents were recorded with an EPC-7 patch-clamp amplifier (HEKA Electronics, Lambrecht/Pfaltz, Germany), filtered at 100 Hz and digitized at 250 Hz before storage. Data acquisition was performed and analysed with an ITC-16 interface (Instrutech Corporation, Bellmore, NY, USA) and PULSE/PULSEFIT software (Heka Electronics). The secretion rate (pC/min) was calculated as the amount of charge transferred to the recording electrode during the last minute of exposure to the stimulus. Two-photon laser scanning microscopy imaging Two-photon laser scanning microscopy (2PLSM) was used to measure ATP/ADP ratio and cytosolic Ca2+in glomus cells in CB slices. Fluorescence was measured using a Scientifica multiphoton galvanometer scanning system (Scientifica Ltd, Ukfield, UK) with an Olympus Å∼60/1.0 NA water-dipping objective lens. A Chameleon Ultra II (680–1080 nm), 3.5 W Ti:sapphire laser system (Coherent laser group) provided the 2P excitation source. Laser power attenuation was achieved using two Pockels cell electro-optic modulators (S-MP-4700 and S-MP-6000-INT/UK) in series controlled by the Labview software (Scientifica). Non-descanned emission photons were detected with a GaAsP photomultiplier tube (PMT), S-MDU-PMT-50, green, 490–560 nm. In our experiments with PercevalHR, the genetically encoded ATP/ADP sensor, 2P excitation wavelengths of 820 nm (ADP-sensitive) and 950 nm (ATP-sensitive) were used. Each wavelength required separate adjustments before starting the experiment. Using the imaging software, power and image acquisition settings were preliminarily adjusted by starting with lower settings and gradually increasing laser power or gain as necessary. Imaging beganin‘live’modetoidentifytheregionofinterest (ROI), where parameters such as zoom, field of view, resolution, dwell time and frame rate were optimized. For experiments on CB slices, the imaging settings were set to a resolution of 256 ×256 pixels, a zoom factor of 4 and 10 μs dwell time, restricting the ROI so that the frame rate withthesesettingsis3–4f.p.s.(framespersecond). PercevalHR was expressed in CB cells using the adeno-associated viral vector (AAV9) containing the TH promoter fragment embedded. For infection, CB slices were incubated in complete culture medium supplemented with 2 ×1013 viral particles for 48 h before the experiment. For experiments, a CB slice was transferred to the recording chamber and continuously perfused with external solution. To estimate the ATP/ADP ratio using PercevalHR, the probe was excited with 950 and 820 nm light in rapid succession (González-Rodríguez et al., 2021; Tantama et al., 2013). Green channel (490–560 nm) fluorescent emission signals for both wavelengths were detected using a non-descanned Scientifica S-MDU-PMT-50-50 select GaAsP PMT. Two time series of five frames (rate of 3–4 f.p.s., 0.195 Å ∼0.195 mM pixels) were acquired for each wavelength. The ratio values from individual dual acquisitions were recorded as a time lapse series. A time-series acquisition protocol was developed to sequentially switch the 2P laser wavelengths. Manual adjustmentstopowerandgainsettingsweremadeforeach wavelength, with a 10 s stabilization period for the light path before data acquisition began. Laser power and PMT gain settings were adjusted and saved for each wavelength to ensure baseline fluorescence was bright but below the saturation threshold of 4095 fluorescence units. Baseline fluorescence was adjusted to ∼800 units, with background fluorescence typically ranging from 100 to 150 units. To ensure accurate measurement of fluorescence changes, the baseline ratio between fluorescence at 950 and 820 nm was maintained near 1. This facilitated monitoring during the experiment and simplified subsequent data analysis. Dual-wavelength acquisitions were collected every 3 min during time-series experiments, with five frames per wavelength acquired and averaged for analysis. The experiments aimed to examine the effects of various treatments or stimuli on the cellular ATP/ADP ratio. During metabolic manipulations, fluorescence levels at 950 and 820 nm typically exhibited opposite directional changes. Time series analysis was conducted © 2025 The Author(s). The Journal of Physiology published by John Wiley & Sons Ltd on behalf of The Physiological Society. 14697793, 2025, 5, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/JP287130 by Readcube (Labtiva Inc.), Wiley Online Library on [16/06/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 1096 M. Torres-López and others J Physiol 603.5 offline using FIJI. Multiple cytosolic ROIs and a background ROI were measured and subtracted. The 950/820 ratio was calculated for each ROI at each time point. Thecontributionofmitochondriatothebioenergetic status of each cell (the OXPHOS index) was estimated by comparing the decrease in the PercevalHR ATP/ADP ratioinducedbybathapplicationofoligomycin(10μM) with the one induced by addition of oligomycin (10 μM) with 2-deoxy-glucose (2-DG), a non-hydrolysable substitute for glucose. For cytosolic Ca2+measurement CB slices were loaded with 10 μM Fluo-3/AM (Invitrogen, Carlsbad, CA, USA; F1242) in Tyrode solution without calcium at 37°C for 15 min, followed by 10 min at room temperature. After loading, CB slices were transferred to the recording chamber and continuously perfused with external solution. The probe was excited at 950 nm to analyse the Ca2+signal using Fluo-3/AM. Green channel (490–560 nm) fluorescent emission signals were detected using a non-descanned Scientifica S-MDU-PMT-50-50 selectGaAsPPMT. Cytosolic[Ca2+]signalsweredigitized at a sampling interval of 500 ms. Immunohistochemistry CB slices expressing TH-PercevalHR after 48 h of infection were fixed with 4% paraformaldehyde in PBS for 30 min at room temperature, permeabilized with PBS-Triton X (PBS-T) 0.1%, blocked with blocking serum solutionfor1hatroomtemperatureandincubated with the TH (Novus Biological, Littleton, CO, USA; cat. no. #BD300-109, RRID: AB_10077691, 1:1000) and green fluorescent protein (GFP) (AvesLab, Davis, CA, USA; cat. no. #GFP-1202, RRID:AB_2734732, 1:400) primary antibodies overnight at 4°C. Finally, slices were incubated for 2 h at room temperature with the Alexa568 (Invitrogen, cat. no. #A10042, RRID:AB_2534017, 1:400) and Alexa488 (Invitrogen, cat. no. #A11039, RRID:AB_2534096, 1:400) secondary antibodies and nuclei were stained with 4,6-diamidino-2-phenylindole (DAPI). Immunofluorescent images were acquired using a Leica Stellaris 8 confocal microscope (Leica, RRID:SCR_02 4664). Patch clamp recordings from HEK cells expressing TASK3 channels HEK293 cells were maintained in high glucose and pyruvate DMEM medium (Gibco, cat. no. #21969-035) supplemented with 10% FBS (Gibco, cat. no. #10270-10), 1% penicillin/streptomycin (Gibco, cat. no. #15140-122) and 1% l-glutamine (Gibco, cat. no. #25030-024). Cells were grown as a monolayer in the bottom of 100 mm Petri dishes to 70% confluence on the day of transfection and washed with PBS. Cells were dissociated by adding 2 ml trypsin to the Petri dish and incubating for 2 min at 37°C. Afterwards, 9 ml of supplemented DMEM was used to resuspend cells previous to centrifugation at 200 gfor 5 min in a 15 ml tube. The pellet was resuspended in 1 ml DMEM and cells were counted using trypan blue solution. In total, 15 ×104cells for each transfection were pelleted again, resuspended in 15 μl of Neon Resuspension Buffer R (Neon transfection system, Invitrogen, cat. no. #MPK10096) and mixed with 2 μg plasmid DNA (pcDNA3.1) encoding TASK3 or GFP. Electroporation was carried out according to the manufacturer’s instructions. Resuspended cells were pulsedtwicewithavoltageof1.100Vandawidthof20ms. After electroporation cells were plated on glass coverslips treatedwithpoly-l-lysineandkeptinthesameculture medium at 37°C in a 5% CO2and 21% O2incubator for 24 h. Macroscopic ionic currents were recorded at room temperature from HEK cells transiently transfected with TASK3 and/or GFP using the whole cell configuration of the patch clamp technique. Patch clamp pipettes had resistances ranging from 2 to 4 Mwhen filled with the internal solution (see ‘Recording solutions’). Voltage-clamp recordings were obtained with an EPC-10 amplifier with an integrated AD interface (ITC-1600, HEKA). Data acquisition and analysis was performed using the PatchMaster/FitMaster software (HEKA). HEK cells were held at −70 mV and then subjected to different pulse protocols generated by the PatchMaster software. Recording solutions For amperometric and microfluorimetric recordings, slices were transferred to a recording chamber continuously perfused with control bicarbonate buffered solution, containing (in mM) 125 NaCl, 23 NaHCO3, 4.5 KCl, 1 MgCl2,2.5CaCl 2, 5 sucrose and 5 glucose. In the high K+solution, KCl substitutes NaCl equimolarly. For the low glucose experiments, cells were exposed to a glucose-free solution, where sucrose replaced glucose (in mM, 125 NaCl, 23 NaHCO3, 4.5 KCl, 1 MgCl2,2.5CaCl 2and 10 sucrose). The 2-DG solution was glucose/sucrose-free and contained 5 mM 2-DG. Pharmacological agents were added to the control solution at the desired concentration when required. The ‘normoxic’ solutions were continuously bubbled with a gas mixture containing 5% CO2, 20% O2and 75% N2(O2tension ∼145 mmHg). The ‘hypoxic’ solutions were bubbled with a gas mixture of 5% CO2and 95% N2, reaching an O2tension of ∼10–15 mmHg in the recording chamber. Osmolality of solutions was ∼300 mOsm/kg with pH 7.4. Experiments were performed at ∼35°C. © 2025 The Author(s). The Journal of Physiology published by John Wiley & Sons Ltd on behalf of The Physiological Society. 14697793, 2025, 5, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/JP287130 by Readcube (Labtiva Inc.), Wiley Online Library on [16/06/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License J Physiol 603.5 Intracellular signalling in arterial chemoreceptors 1097 Complete replacement of solutions in the chamber was achieved in ∼60 s. For patch-clamp recordings, pipettes were filled with internal solution containing (in mM) 150 KCl, 3 MgCl2, 10 HEPES and 5 EGTA (pH 7.2 adjusted with KOH). The composition of the bath solution was (in mM) 145 NaCl, 2.5 KCl, 3 MgCl2,1CaCl 2,10glucoseand10HEPES(pH was adjusted using HCl or NaOH to the desired values). TostudytheeffectofNADHandH 2O2on macroscopic currents, these agents were added to the internal solution at the desired concentration (NADH: 200 and 400 μM, H2O2: 200 and 500 μM). Statistical analysis Statistical analysis was performed using Prism Version 8.2.1. (279) for MacOS. Normality of the data obtained in the experiments was assessed with Shapiro-Wilk, D’Agostino & Pearson, Anderson-Darling and Kolmogorov-Smirnov tests. In some cases, a log transformation of the data was carried out to normalize the distribution prior to parametric analysis. For graphical representation of the data with a normal distribution, we used bar diagrams with a scatterplot of data points superimposed. These data are given as mean ±SD and the number (n) of experiments. For graphical representation of non-parametric data, we used box plots where median, quartiles and whiskers (highest and lowest values) are indicated. Comparisons between two groups were performed using paired or unpaired ttests for data with a normal distribution, and Wilcoxon or a Mann–Whitney test for non-parametric data. Comparisons of multiple groups with a normal distribution were done using a one-way ANOVA followed by Tukey’s multiple comparisons post hoc tests. Multiple non-parametric data were compared using Kruskal–Wallis followed by Dunn’s multiple comparison tests. Statistical tests used are also indicated in the figure legends. P<0.05 was considered statistically significant and the resulting value is represented in each figure panel. Results Blocking MCI abolishes responsiveness of glomus cells to hypoxia but not to cyanide Hypoxiainglomuscellsisknowntoproduce dose-dependent increases in the mitochondrial production of NADH and ROS and a neurosecretory response, which are mimicked by blocking MCI with rotenone. Rotenone and the genetic disruption of MCI are both able to abolish any effect of hypoxia on mitochondrial signalling or cell secretory activity, indicating that a functional MCI is essential for acute O2sensing by glomus cells (Fernández-Agüera et al., 2015; Ortega-Sáenz et al., 2003). Supporting this model (Fig. 1A), in amperometric experiments performed here, hypoxia and rotenone separately elicited clear increases in the secretory activity of glomus cells in CB slices. However, responsiveness to hypoxia was not only abolished by rotenone but rotenone-induced secretory activity also decreased during hypoxia (Fig. 1B,C,F). This finding, which is reminiscent of the hypoxic inhibition of rotenone-induced ROS production in mitochondria (Arias-Mayenco et al., 2018), contrasts with what would be expected if a decrease in intracellular ATP (indicated by the question mark in Fig. 1A)participatesinboth rotenone and hypoxia signalling. Itcouldbearguedthattheflowofelectronsfedtocytochrome c oxidase is drastically reduced in the presence of rotenone and that under these conditions hypoxia has no effect because it cannot produce further inhibition of MCIV. To test this hypothesis, we recorded the secretory activity induced by cyanide in cells already exposed to rotenone for several minutes. Cyanide is a potent MCIV blocker which has no effect on glomus cells when the ETC is interrupted (e.g. due to genetic disruption of MCIII) and electrons do not reach MCIV (Cabello-Rivera et al., 2022). In all cells tested (n=11), application of cyanide in the presence of rotenone induced a potent and reversible secretory response (Fig. 1D–F). These data indicate that MCIV is still functional and inhibitable by cyanide in cells treated with rotenone for several minutes. However, even withafunctionalMCIV,blockingMCImakesglomus cells unresponsive to hypoxia. Differential effect of MCI disruption on sensitivity to hypoxia and hypoglycaemia TheCBisoneoftheorganswiththehighestO 2 consumption for which glucose provides the fuel for oxidative phosphorylation. On the other hand, glomuscellsareactivatedbyhypoxiaandhypoglycaemia (García-Fernández et al., 2007; Pardal & López-Barneo, 2002; Zhang et al., 2007). We compared the effect of inhibition of the ETC with rotenone on the responsiveness of rat glomus cells to hypoxia and glucose deprivation. Whereas the secretory response of cells to hypoxia was occluded in the presence of rotenone, responsiveness to 0 mM glucose was maintained (Fig. 2A,B). Similar differences in responses to hypoxia and low glucose in the presence of rotenone were observed in glomus cells of mice (Fig. 2C,D). Moreover, glomus cells from genetically modified mice lacking a functional MCI in the catecholaminergic territory (see Fernández-Agüera et al., 2015; Jiménez-Gómez et al., 2023) were unresponsive to hypoxia but showed a robust secretory response to glucose deprivation, which was even higher than in © 2025 The Author(s). The Journal of Physiology published by John Wiley & Sons Ltd on behalf of The Physiological Society. 14697793, 2025, 5, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/JP287130 by Readcube (Labtiva Inc.), Wiley Online Library on [16/06/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 1098 M. Torres-López and others J Physiol 603.5 wild-type glomus cells (Fig. 2E–H). In MCI-deficient glomus cells obtained from genetically modified mice, sensitivity to glucose deprivation is totally independent of oxidative phosphorylation because in these cells, in which MCI has been disrupted for over 2 months, mitochondrial ETC and oxidative ATP synthesis are strongly inhibited as they are insensitive to cyanide, although maintain normal responses to CO2(Fig. 2I–L) (see also Jiménez-Gómez et al., 2023). The fact that sensitivity to hypoxia and glucose deprivation (both compromising oxidative phosphorylation) depends on separate pathways differentially affected by rotenone or genetic MCI disruption further suggests that changes in mitochondrial production of ATP are not involved in acute hypoxia signalling. However, modifications in non-oxidative cytosolic ATP levels could contribute to the responsiveness of glomus cells to low glucose. 0.0009 D A MCI MCII CoQH2MCIII MCIV O2 H2O Hypoxia NADH ROS ROS Rot CN ATP? Hypoxia Rotenone Hypoxia CN 5 pA B 5 pA 1 min Hypoxia Rotenone Hypoxia 1 min 1 min 2 pA Rotenone CN CN 0 5 10 15 Secretion rate (pC/min) C Rot Rot Hx E F 0.0106 0 5 10 15 Secretion rate (pC/min) 20 25 Rot Rot CN Figure 1. Blockade of MCI abolishes responsiveness of glomus cells to hypoxia but not to cyanide. A, scheme illustrating the electron transport chain (ETC) and its dynamic changes in response to hypoxia (red lines and symbols). Note the accumulation of reduced coenzyme Q (CoQH2), NADH and ROS during exposure to hypoxia. The inhibition of mitochondrial complex I (MCI) and IV (MCIV) by rotenone (Rot) and cyanide (CN) are illustrated in grey and blue, respectively. B, representative amperometric recording of catecholamine secretion of glomus cells in rat CB slices in response to hypoxia (O2tension ∼15 mmHg), rotenone (1 μM) and rotenone +hypoxia. C, average secretion rate induced by rotenone (Rot, 1–5 μM) and rotenone +hypoxia (Rot +Hx) in rat CB glomus cells. Values in pC/min are: Rot (5.4 ±3.1, n = 8 cells/5 rats), Rot +Hx (1.6 ±0.8, n = 8 cells/5 rats). D, representative amperometric traces recorded from rat CB glomus cells in slices in response to cyanide (CN, 300 μM), rotenone (1 μM) and rotenone +cyanide. E, average secretion rate induced by rotenone (Rot 1–5 μM) and rotenone +cyanide (Rot +CN) in rat CB glomus cells. Values in pC/min are: Rot (3.5 ±3.5, n = 7 cells/3 rats), Rot +CN (11.6 ±5.7, n = 7 cells/3 rats). F, representative amperometric recording (out of four similar experiments) of catecholamine secretion of glomus cells from a rat CB slice in response to hypoxia, rotenone (1 μM), rotenone +hypoxia and rotenone +cyanide (300 μM). Note the decrease in the secretory activity when the cell is exposed simultaneously to rotenone and hypoxia. By contrast, inhibition of MCIV by cyanide in the presence of rotenone strongly activates the cell. Data are expressed as mean ±SD with all data values superimposed. P-values (<0.05) calculated by a two-tailed paired t test are indicated in the panel. © 2025 The Author(s). The Journal of Physiology published by John Wiley & Sons Ltd on behalf of The Physiological Society. 14697793, 2025, 5, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/JP287130 by Readcube (Labtiva Inc.), Wiley Online Library on [16/06/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License J Physiol 603.5 Intracellular signalling in arterial chemoreceptors 1099 0.0282 Basal Secretion rate (pC/min) 20 15 10 5 0Rot < 0.0001 Rot 0 Glu <0.0001 Secretion rate (pC/min) CN Secretion rate (pC/min) Basal CN 10 5 0 Basal Control EHypoxia 0 Glu 5 pA 40K 1 min ARotenone 5 pA 1 min Hypoxia 0 Glu Hypoxia 0 Glu B Secretion rate (pC/min) 10 5 00 Glu F MCI-deficient Hypoxia 0 Glu 5 pA 40K 1 min G Secretion rate (pC/min) 20 10 0 0 Glu Basal <0.0001 H Rat Mouse Control MCI-deficient Hypoxia CN 5 pA 1 min IJK CNHypoxia 5 pA CO2CO2 1 min L C D Secretion rate (pC/min) 15 10 5 0Rot Rot 0 Glu 1 min Rotenone 5 pA Hypoxia 0 Glu Hypoxia 0 Glu hypoxia 0 Glu 0.0008 15 10 5 0 15 Hx Hx 0.0035 0.537 0.121 >0.999 >0.999 Figure 2. Differential effect of MCI disruption on sensitivity to hypoxia and glucose deprivation. A, representative amperometric traces of the secretory activity recorded from rat CB glomus cells in slices in response to hypoxia (O2tension ∼15 mmHg), 0 mM glucose (0 Glu), rotenone (1 μM), rotenone +hypoxia and rotenone +0 Glu. B, average secretion rate (pC/min) induced by rotenone (Rot, 1–5 μM) and rotenone +0glucose (Rot +0 Glu) in rat CB glomus cells. Data are expressed as mean ±SD with all data values superimposed. Values in pC/min are: Rot (4.7 ±3.2, n = 19 cells/11 rats) and Rot +0 Glu (12.1 ±3.7, n = 6 cells/3 rats). P-value (<0.05) calculated by unpaired two-tailed t test is indicated. C, representative amperometric traces of secretory activity recorded from mouse CB glomus cells in slices in response to hypoxia (O2,tension∼15 mmHg), 0 mM glucose (0 Glu), rotenone (5 μM), rotenone +hypoxia and rotenone +0 Glu. D, average secretion rate (pC/min) recorded from control mouse glomus cells in response to rotenone (Rot, 5 μM) and rotenone +0 glucose (Rot +0 Glu). Values in pC/min are: Rot (2.3 ±2.3, n = 7 cells/4 mice), Rot +0 Glu (10.1 ±3.5, n = 7 cells/4 mice). Comparison between both groups was performed using a paired two-tailed t test. P-value is indicated in the panel. E and G, representative recordings of the secretory activity induced by hypoxia, glucose deprivation (0 Glu) and high K+in glomus cells in CB slices from control mice (E) and MCI-deficient mice (G). F and H, quantification of the secretory activity (pC/min) recorded in basal conditions and during 0 Glu in glomus cells from control mice (F; n = 18 cells/15 mice) © 2025 The Author(s). The Journal of Physiology published by John Wiley & Sons Ltd on behalf of The Physiological Society. 14697793, 2025, 5, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/JP287130 by Readcube (Labtiva Inc.), Wiley Online Library on [16/06/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 1106 M. Torres-López and others J Physiol 603.5 Kilfoil, P. J., Tipparaju, S. M., Barski, O. A., & Bhatnagar, A. (2013). Regulation of ion channels by pyridine nucleotides. Circulation Research,112(4), 721–741. Kim, D., Cavanaugh, E. J., Kim, I., & Carroll, J. L. (2009). Heteromeric TASK-1/TASK-3 is the major oxygen-sensitive background K+channel in rat carotid body glomus cells. The Journal of Physiology,587(12), 2963–2975. Kim, Y., Bang, H., & Kim, D. (2000). TASK-3, a new member of the tandem pore K(+) channel family. Journal of Biological Chemistry,275(13), 9340–9347. Leach,R.M.,Sheehan,D.W.,Chacko,V.P.,&Sylvester,J. T. (2000). Energy state, pH, and vasomotor tone during hypoxia in precontracted pulmonary and femoral arteries. American Journal of Physiology-Lung Cellular and Molecular Physiology,278(2), L294–L304. Lopez-Barneo, J. (2022). Neurobiology of the carotid body. Handbook of Clinical Neurology,188, 73–102. López-Barneo, J., Benot, A. and Ureña, J. (1993). Oxygen sensing and the electrophysiology of arterial chemoreceptor cells. Neural Information Processing Systems,8, 191–195. Lopez-Barneo, J., Lopez-Lopez, J. R., Urena, J., & Gonzalez, C. (1988). Chemotransduction in the carotid body: K+current modulated by PO2 in type I chemoreceptor cells. Science, 241(4865), 580–582. Lopez-Barneo, J., & Ortega-Saenz, P. (2022). Mitochondrial acute oxygen sensing and signaling. Critical Reviews in Biochemistry and Molecular Biology,57(2), 205–225. Lopez-Lopez, J., Gonzalez, C., Urena, J., & Lopez-Barneo, J. (1989). Low pO2selectively inhibits K+channel activity in chemoreceptor cells of the mammalian carotid body. Journal of General Physiology,93(5), 1001–1015. Moreno-Dominguez, A., Ortega-Saenz, P., Gao, L., Colinas, O., Garcia-Flores, P., Bonilla-Henao, V., Aragones, J., Huttemann, M., Grossman, L. I., Weissmann, N., Sommer, N., & Lopez-Barneo, J. (2020). Acute O2 sensing through HIF2alpha-dependent expression of atypical cytochrome oxidasesubunitsinarterialchemoreceptors.Science Signaling,13(615), eaay9452. Moreno-Dominguez, A., Colinas, O., Arias-Mayenco, I., Cabeza, J. M., Lopez-Ogayar, J. L., Chandel, N. S., Weissmann, N., Sommer, N., Pascual, A., & Lopez-Barneo, J. (2024). Hif1α-dependent mitochondrial acute O2sensing and signaling to myocyte Ca2+channels mediate arterial hypoxic vasodilation. Nature Communications,15(1), 6649. Mulligan, E., & Lahiri, S. (1981). Dependence of carotid chemoreceptor stimulation by metabolic agents on PaO2 and PaCO2. Journal of Applied Physiology: Respiratory, Environmental and Exercise Physiology,50, 884–891. Nurse, C. A. (2014). Synaptic and paracrine mechanisms at carotid body arterial chemoreceptors. The Journal of Physiology,592(16), 3419–3426. Ortega-Saenz, P., Levitsky, K. L., MT, M.-A., Bonilla-Henao, V., Pascual, A., & Lopez-Barneo, J. (2010). Carotid body chemosensory responses in mice deficient of TASK channels. Journal of General Physiology,135(4), 379–392. Ortega-Saenz, P., & Lopez-Barneo, J. (2020). Physiology of the carotid body: From molecules to disease. Annual Review of Physiology,82(1), 127–149. Ortega-Saenz, P., Pardal, R., Garcia-Fernandez, M., & Lopez-Barneo, J. (2003). Rotenone selectively occludes sensitivity to hypoxia in rat carotid body glomus cells. The Journal of Physiology,548(3), 789–800. Papreck, J. R., Martin, E. A., Lazzarini, P., Kang, D., & Kim, D. (2012). Modulation of K2P3.1 (TASK-1), K2P9.1 (TASK-3), and TASK-1/3 heteromer by reactive oxygen species. Pflugers Archiv: European Journal of Physiology,464(5), 471–480. Pardal, R., & Lopez-Barneo, J. (2002). Low glucose-sensing cells in the carotid body. Nature Neuroscience,5(3), 197–198. Pardal, R., Ludewig, U., Garcia-Hirschfeld, J., & Lopez-Barneo, J. (2000). Secretory responses of intact glomus cells in thin slices of rat carotid body to hypoxia and tetraethylammonium. Proceedings National Academy of Science USA,97(5), 2361–2366. Peers, C. (1990). Hypoxic suppression of K+currents in type I carotid body cells: Selective effect on the Ca2(+)-activated K+current. Neuroscience Letters,119(2), 253–256. Perez-Garcia, M. T., Colinas, O., Miguel-Velado, E., Moreno-Dominguez, A., & Lopez-Lopez, J. R. (2004). Characterization of the Kv channels of mouse carotid body chemoreceptor cells and their role in oxygen sensing. The Journal of Physiology,557(2), 457–471. Piruat, J. I., Pintado, C. O., Ortega-Saenz, P., Roche, M., & Lopez-Barneo, J. (2004). The mitochondrial SDHD gene is required for early embryogenesis, and its partial deficiency results in persistent carotid body glomus cell activation with full responsiveness to hypoxia. Molecular and Cellular Biology,24(24), 10933–10940. Rajan, S., Wischmeyer, E., Xin Liu, G., Preisig-Muller, R., Daut, J., Karschin, A., & Derst, C. (2000). TASK-3, a novel tandem pore domain acid-sensitive K+channel. An extracellular histiding as pH sensor. Journal of Biological Chemistry,275(22), 16650–16657. Rakoczy, R. J., Schiebrel, C. M., & Wyatt, C. N. (2022). Acute oxygen-sensing via mitochondria-generated temperature transients in rat carotid body type I cells. Frontiers in Physiology,13, 874039. Sahoo, N., Hoshi, T., & Heinemann, S. H. (2014). Oxidative modulation of voltage-gated potassium channels. Antioxidants & Redox Signaling,21(6), 933–952. Stea, A., & Nurse, C. A. (1991). Whole-cell and perforated-patch recordings from O2-sensitive rat carotid body cells grown in shortand long-term culture. Pflugers Archiv: European Journal of Physiology,418(1–2), 93–101. Tantama, M., Martinez-Francois, J. R., Mongeon, R., & Yellen, G. (2013). Imaging energy status in live cells with a fluorescent biosensor of the intracellular ATP-to-ADP ratio. Nature Communications,4(1), 2550. Tian, H., Hammer, R. E., Matsumoto, A. M., Russell, D. W., & McKnight, S. L. (1998). The hypoxia-responsive transcription factor EPAS1 is essential for catecholamine homeostasis and protection against heart failure during embryonic development. Genes & Development,12, 3320–3324. © 2025 The Author(s). The Journal of Physiology published by John Wiley & Sons Ltd on behalf of The Physiological Society. 14697793, 2025, 5, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/JP287130 by Readcube (Labtiva Inc.), Wiley Online Library on [16/06/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License J Physiol 603.5 Intracellular signalling in arterial chemoreceptors 1107 Torres-Torrelo, H., Ortega-Saenz, P., Gao, L., & Lopez-Barneo, J. (2021). Lactate sensing mechanisms in arterial chemoreceptor cells. Nature Communication,12(1), 4166 Urena, J., Fernandez-Chacon, R., Benot, A. R., Alvarez de Toledo, G. A., & Lopez-Barneo, J. (1994). Hypoxia induces voltage-dependent Ca2+entry and quantal dopamine secretion in carotid body glomus cells. Proceedings National Academy of Science USA,91(21), 10208–10211. Varas, R., Wyatt, C. N., & Buckler, K. J. (2007). Modulation of TASK-like background potassium channels in rat arterial chemoreceptor cells by intracellular ATP and other nucleotides. The Journal of Physiology,583(2), 521–536. Wilson, D. F., Mokashi, A., Chugh, D., Vinogradov, S., Osanai, S., & Lahiri, S. (1994). The primary oxygen sensor of the cat carotid body is cytochrome a3 of the mitochondrial respiratory chain. Federation of European Biochemical Societies Letters,351(3), 370–374. Wyatt, C. N., & Buckler, K. J. (2004). The effect of mitochondrial inhibitors on membrane currents in isolatedneonatalratcarotidbodytypeIcells.The Journal of Physiology,556(1), 175–191. Zhang, M., Buttigieg, J., & Nurse, C. A. (2007). Neurotransmitter mechanisms mediating low-glucose signaling in cocultures and fresh tissue slices of rat carotid body. The Journal of Physiology,578(3), 735–750. Additional information Data availability statement All data supporting the results in the paper are in the paper itself. Raw data and recordings are available in the laboratory and will be provided upon request. Competing interests The authors have no conflicts of interest to declare. Author contributions M.T.-L., P.G.-R., O.C., H.-S. R., H.T.-T., A.C., L.G., P. O.-S. and J. L.-B. performed the experiments and participated in the interpretation of data. M-T.-L., P.G.-R., O.C., L.G., P. O.-S. and J.L.-B. designedthefiguresandwrotethefirstdraftofthepaper.P. O.-S. and J.L.-B wrote the final draft of the paper and supervised the project. All authors read and approved the final paper. All authors have approved the final version of the manuscript and agreetobeaccountableforallaspectsofthework.Allpersons designated as authors qualify for authorship, and all those who qualify for authorship are listed. Funding This research was supported by Spanish Ministries of Science and Innovation and Health (Grants PID2019-106410RBI00, PID2019-110817R, PID2022-138131OB-I00 and PID2023146862OB-100 funded by MCIN/AEI/10.13039/501100011033 to J.L.-B., L.G. and P.O.-S) and the European Research Council (ERC Advanced Grant PRJ201502629). Acknowledgements We thank Alfredo Caro-Maldonado (Agilent Technologies) for his technical help in setting up the metabolic analysis using the Seahorse analyser. We also thank technical help provided by the staff at the Core Facilities of Instituto de Biomedicina de Sevilla (IBiS). Keywords acute oxygen sensing, carotid body glomus cells, cytosolic ATP, electron transport chain inhibitors, glucose sensing, H2O2,hypoxia,mitochondria-to-membranesignalling,NADH, TASK3 channels Supporting information Additional supporting information can be found online in the Supporting Information section at the end of the HTML view of the article. Supporting information files available: Peer Review History © 2025 The Author(s). The Journal of Physiology published by John Wiley & Sons Ltd on behalf of The Physiological Society. 14697793, 2025, 5, Downloaded from https://physoc.onlinelibrary.wiley.com/doi/10.1113/JP287130 by Readcube (Labtiva Inc.), Wiley Online Library on [16/06/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License