1128 Diabetes Volume 67, June 2018

Adrenaline Stimulates Glucagon Secretion by Tpc2- Dependent Ca2+ Mobilization From Acidic Stores in Pancreatic a-Cells

Alexander Hamilton,1 Quan Zhang,1 Albert Salehi,2 Mara Willems,1 Jakob G. Knudsen,1 Anna K. Ringgaard,3,4 Caroline E. Chapman,1 Alejandro Gonzalez-Alvarez,1 Nicoletta C. Surdo,5 Manuela Zaccolo,5 Davide Basco,6 Paul R.V. Johnson,1,7 Reshma Ramracheya,1 Guy A. Rutter,8 Antony Galione,9 Patrik Rorsman,1,2,7 and Andrei I. Tarasov1,7

Diabetes 2018;67:1128–1139 | https://doi.org/10.2337/db17-1102

Adrenaline is a powerful stimulus of glucagon secretion. It The ability of the “fight-or-flight” hormone adrenaline to acts by activation of b-adrenergic receptors, but the down- increase plasma glucose levels by stimulating liver gluconeo- stream mechanisms have only been partially elucidated. genesis is in part mediated by glucagon, the body’sprincipal Here, we have examined the effects of adrenaline in mouse hyperglycemic hormone (1). Glucagon is secreted by the and human a-cells by a combination of electrophysiology, a-cells of the pancreas (2). Reduced autonomic stimulation 2+ imaging of Ca and PKA activity, and hormone release of glucagon secretion may result in hypoglycemia, a serious measurements. We found that stimulation of glucagon and potentially fatal complication of diabetes (3). It has been secretion correlated with a PKA- and EPAC2-dependent estimated that up to 10% of insulin-treated patients die of (inhibited by PKI and ESI-05, respectively) elevation of hypoglycemia (4). Understanding the mechanism by which [Ca2+] in a-cells, which occurred without stimulation of i adrenaline stimulates glucagon secretion and how it becomes

ISLET STUDIES electrical activity and persisted in the absence of extracel- perturbed in patients with diabetes is therefore essential. lular Ca2+ but was sensitive to ryanodine, bafilomycin, and The mechanism by which adrenaline stimulates glucagon thapsigargin. Adrenaline also increased [Ca2+] in a-cells in i secretion has only partially been elucidated. It is clear, how- human islets. Genetic or pharmacological inhibition of the b Tpc2 channel (that mediates Ca2+ release from acidic intra- ever, that it involves activation of -adrenergic receptors – cellular stores) abolished the stimulatory effect of adren- (5 7), leading to elevated intracellular cAMP ([cAMP]i), and 2+ aline on glucagon secretion and reduced the elevation of culminates in elevation of the cytoplasmic free Ca concen- 2+ 2+ 2+ tration ([Ca ] )withresultantCa -dependent stimulation [Ca ]i. Furthermore, in Tpc2-deficient islets, ryanodine i exerted no additive inhibitory effect. These data suggest of glucagon secretion (6,8,9). However, our understanding that b-adrenergic stimulation of glucagon secretion is con- of the spatiotemporal interrelationship between different 2+ 2+ a trolled by a hierarchy of [Ca ]i signaling in the a-cell that is intracellular Ca stores involved in -cell adrenaline signal- initiated by cAMP-induced Tpc2-dependent Ca2+ release ing remains patchy. from the acidic stores and further amplified by Ca2+-induced Here, we have explored how adrenaline triggers glucagon Ca2+ release from the sarco/endoplasmic reticulum. release by a combination of hormone secretion measurements,

1Oxford Centre for Diabetes, Endocrinology and Metabolism, Churchill Hospital, 8Section of Cell Biology and Functional Genomics, Department of Medicine, University of Oxford, Headington, U.K. Imperial College London, London, U.K. 2Institute of Neuroscience of Physiology, Department of Physiology, Metabolic 9Department of Pharmacology, University of Oxford, Oxford, U.K. Research Unit, University of Göteborg, Göteborg, Sweden Corresponding author: Andrei I. Tarasov, [email protected], or 3 Diabetes Research, Department of Stem Cell Biology, Novo Nordisk A/S, Måløv, Patrik Rorsman, [email protected]. Denmark Received 15 September 2017 and accepted 14 March 2018. 4Department of Biomedical Sciences, University of Copenhagen, Copenhagen, Denmark This article contains Supplementary Data online at http://diabetes 5Department of Physiology, Anatomy and Genetics, University of Oxford, Oxford, U.K. .diabetesjournals.org/lookup/suppl/doi:10.2337/db17-1102/-/DC1. 6Center for Integrative Genomics, Université de Lausanne, Lausanne, Switzerland © 2018 by the American Diabetes Association. Readers may use this article as 7Oxford National Institute for Health Research, Biomedical Research Centre, Oxford, long as the work is properly cited, the use is educational and not for profit, and the U.K. work is not altered. More information is available at http://www.diabetesjournals .org/content/license. diabetes.diabetesjournals.org Hamilton and Associates 1129 electrophysiology, and imaging of cytoplasmic Ca2+,cAMP, Pancreatic islets were isolated from mice by injecting and PKA activity in a-cells within intact mouse pancreatic collagenase solution into the bile duct, with subsequent islets. We have extended the measurements to human islets digestion of the connective and exocrine pancreatic tissue. and also tested how the responsiveness to adrenaline is Human pancreatic islets were isolated at the Oxford Di- affected when islets are cultured under hyperglycemic con- abetes Research & Wellness Foundation Human Islet Isola- ditions to establish whether and how this regulation becomes tion Facility according to published protocols (12,13). impaired in diabetes. Our data indicate that Ca2+ release from Unless otherwise stated, experiments were performed on acidic (lysosomal) stores plays a critical and unexpected role acutely isolated islets. For experiments aiming to emulate in adrenaline-induced glucagon secretion. chronic hyperglycemia, islets were cultured for 48 h in RPMI medium containing 30 mmol/L or the standard 11 mmol/L RESEARCH DESIGN AND METHODS glucose, supplemented with 10% FBS, 100 IU/mL penicillin, and 100 mg/mL streptomycin (all reagents from Life Tech- Chemicals nologies, Paisley, U.K.). Recombinant sensors were deliv- The following substances were used (source given in ered via adenoviral (GCaMP6f, AKAR3) or BacMam (cADDis) parentheses): the L-type Ca2+ channel blocker isradipine vectors at 105 infectious units per islet, followed by 24–36 h and the P/Q Ca2+ channel blocker v-agatoxin (Alomone culturing at 11 mmol/L glucose (as described above) to Laboratories, Jerusalem, Israel); the a -antagonist prazosin 1 express the sensors. (Abcam, Cambridge, U.K.); the membrane-permeable PKA inhibitor myr-PKI, the inositol triphosphate receptor in- 2+ Imaging of [Ca ]i, cAMP, and PKA Activity in Islet Cells hibitor Xestospongin C, the nicotinic acid adenine dinucle- 2+ Time-lapse imaging of [Ca ]i in intact freshly isolated mouse otide phosphate (NAADP) antagonist Ned-19, the V-ATPase islets was performed on an inverted Zeiss AxioVert 200 mi- fi inhibitor ba lomycin, the sarco/endoplasmic reticulum croscope equipped with the Zeiss 510-META laser confocal (sER) ATPase inhibitor thapsigargin, and noradrenaline scanning system. Prior to Ca2+ imaging, mouse islets were (Tocris Bioscience, Bristol, U.K.); the EPAC2 inhibitor ESI- loaded with 6 mmol/L Fluo-4 at room temperature or 05 (BioLog, Bremen, Germany); and the insulin receptor 10 mmol/L Fluo4FF (Molecular Probes) at 37°C for 90 min. antagonist (S961; Novo Nordisk, Måløv, Denmark). All other Both dyes were excited at 488 nm and emission was collected compounds were obtained from Sigma-Aldrich (Dorset, at 530 nm, using the 512 3 512 frame scanning mode with U.K.). the pixel dwell time of 6 ms (image frequency: 0.25 Hz). Time- 2+ Solutions lapse imaging of [Ca ]i in groups of mouse islets was done on The extracellular solution (EC1) used for imaging and an Axiozoom.V16 microscope (0.1 Hz) using GCaMP6f (14). current-clamp electrophysiology (Fig. 2C)experiments PKA activity was reported in pancreatic islet cells using fl contained (in millimoles per liter ) 140 NaCl, 4.6 KCl, 2.6 recombinant uorescence resonance energy transfer (FRET) probe AKAR3 (15). AKAR3 CFP fluorescence was excited at CaCl2,1.2MgCl2,1NaH2PO4,5NaHCO3,and10HEPES (pH 7.4 with NaOH). The pipette solution (IC1) consisted of 458 nm using an Axiozoom.V16 microscope, which has been shown previously to excite CFP selectively (16); the emitted 76 K2SO4, 10 NaCl, 10 KCl, 1 MgCl2, and 5 HEPES (pH 7.35 with KOH). The extracellular solution (EC2) for recordings of light was collected at 485 nm (CFP) and 515 nm (YFP) the depolarization-induced exocytosis (Fig. 3C and D) con- (image frequency: 0.0 Hz). Time-lapse imaging of [cAMP]i was performed using a recombinant Green Downward cADDis tained 118 NaCl, 5.6 KCl, 2.6 CaCl2,1.2MgCl2,5HEPES, 20 TEA-Cl, and 6 glucose (pH 7.4 with NaOH). In the latter sensor (Montana Molecular, Bozeman, MT). The cADDis fl experiments, the pipette-filling medium (IC2) consisted of uorescence was excited at 485 nm and the emission recorded at 515 nm using an Axiozoom.V16 microscope. 76 Cs2SO4, 10 NaCl, 10 KCl, 1 MgCl2, and 5 HEPES (pH 7.35 with CsOH). PKA and cAMP were imaged at a frequency of 16 mHz. All imaging at 34°C was performed using an open cham- Animals, Islet Isolation, and Culture ber. The bath solution containing various stimuli was peri- NMRI mice (Charles River) were used throughout the fused continuously at a rate of 200 mL/min. In experiments study except for the experiments involving the genetic involving PKI, ESI-05, ryanodine, Xestospongin C, Ned-19 2/2 manipulation of TPC1 and -2 (Fig. 5). Tpcn1 and thapsigargin, and the Ca2+ channel blockers v-agatoxin and 2/2 Tpcn2 mice were developed on C57Bl/6N background isradipine, cells were preincubated in the solution of the 2/2 as previously described (10). Cd38 mice were developed respective agent for 90 min. For the bafilomycin experiments, on C57Bl/6J background (11). Age- and sex-matched the preincubation time was 20 min. By preincubating the islets C57Bl/6N and C57Bl/6J wild-type animals were used as in the agonist/antagonist solutions, we aimed to reach the controls. Mice were kept in a conventional vivarium with saturation of the effect by the start of the recording. a-Cells a 12-h-dark/12-h-light cycle with free access to food and were identified as those exhibiting a response to 1 mmol/L water and were killed by cervical dislocation. All experiments glutamate (Fig. 1C–E) (17) and/or adrenaline (known to were conducted in accordance with the United Kingdom inhibit secretion of both insulin and somatostatin) (18). Animals (Scientific Procedures) Act (1986) and the Univer- Images were acquired using ZenBlack or ZenBlue soft- sity of Oxford ethics guidelines. ware (Carl Zeiss). 1130 Adrenaline Controls Ca2+ Release in a-Cells Diabetes Volume 67, June 2018

2+ Figure 1—The stimulatory effect of adrenaline on glucagon secretion is mediated by selective elevation of [Ca ]i in pancreatic a-cells. A: Glucagon secreted from isolated NMRI mouse islets in response to 3 mmol/L glucose with/without 5 mmol/L adrenaline. #P , 0.05 vs. the effect of 3 mmol/L glucose alone. B: Variance of the Fluo4 intensity when the islet was perifused with 3 mmol/L glucose with or without 5 mmol/L 2+ adrenaline or 20 mmol/L glucose (as indicated). The brighter cells are those in which [Ca ]i oscillates. The arrow indicates a cell that started spiking after adrenaline had been applied. C and D: Typical single a-cell responses to application of 1 mmol/L glutamate and 5 mmol/L adrenaline recorded 2+ in mouse (C)(n = 29) and human (D)(n = 55) islets at 3 mmol/L glucose. E:Representative[Ca ]i time course in the populations of a-(n =21)and non-a-cells (mostly b-cells [n = 75]), differentiated by the response to glutamate. The difference in magnitude of the glutamate and the adrenaline 2+ effects was not a consistent finding. F:[Ca ]i changes in a-cells quantified as pAUC at 3 mmol/L glucose with or without glutamate or adrenaline in mouse (NMRI, C57Bl/6N) and human islets. P , 0.05 vs. the respective effect observed in NMRI mice (*) or the effect of basal (3 mmol/L glucose) in the same recording (#). diabetes.diabetesjournals.org Hamilton and Associates 1131

Hormone Release Measurements mouse pancreatic islets by 3.8 6 0.8-fold (Fig. 1A), in line Batches of 10–20 size-matched freshly isolated mouse islets with previously reported results (5). were preincubated in 1 mL Krebs-Ringer buffer containing Glucagon secretion is a Ca2+-dependent process and is 2+ 1 mmol/L glucose and 0.2% BSA for 30 min at 37°C, followed stimulated by an elevation of [Ca ]i (5). We quantified the 2+ by a 1-h test incubation in 1 mL of the same medium adrenaline effect on [Ca ]i in a-cells within intact islets supplemented as indicated. Glucagon was determined by using time-lapse laser scanning confocal microscopy. At radioimmunoassay (Euro Diagnostica, Malmö, Sweden) (19). 3 mmol/L glucose, ,20% of the cells in pancreatic islets Tpcn1 Tpcn2 Cd38 2+ The experiments on the , ,or knockout isolated from NMRI mice were active and generated [Ca ]i mice (Fig. 5) and control mice were assayed using the Meso oscillations (Fig. 1B). Of the spontaneously active cells, Scale Discovery glucagon assay (Rockville, MD). .70% responded to glutamate (Supplementary Movie 1) and were thus identified as a-cells (17,24). In a-cells thus Electrophysiology identified, adrenaline induced a rapid and reversible in- The electrophysiological measurements were performed on crease in [Ca2+] (Fig. 1C–E). Similar effects of adrenaline a i -cells within freshly isolated intact islets (from NMRI or were observed at 1 mmol/L glucose (Supplementary Fig. 1A fi C57Bl/6 mice), using an EPC-10 patch-clamp ampli er and E). (HEKA Electronics, Lambrecht/Pfalz, Germany) and Pulse The majority of the islet cells (.80%) were inactive at software. All electrophysiological experiments were per- 3 mmol/L glucose but were stimulated when glucose was a fi formed at 34°C. -Cells were identi ed as those active at elevated to 20 mmol/L, as expected for b-ord-cells (Fig. 1E). d 2+ low (3 mmol/L) glucose and were differentiated from -cells At 3 mmol/L glucose, adrenaline did not affect [Ca ] in any fi i (some of which re action potentials, albeit at low frequency of these cells (Supplementary Fig. 1B)andat20mmol/L at this glucose concentration) by the distinct appearance of actually reduced [Ca2+] (not shown). Assessed as pAUC (see A i action potentials (Supplementary Fig. 2 ). For the mem- RESEARCH DESIGN AND METHODS and Supplementary Fig. 1C), C brane potential recordings (Fig. 2 ), the perforated patch responsiveness to adrenaline strongly correlated with spon- configuration was used as previously described (20) using 2+ taneous [Ca ]i oscillations at 3 mmol/L glucose (Pearson r = solutions IC1 and EC1. Exocytosis was measured as increases 0.78) and responsiveness to glutamate (r = 0.81) (Supple- a in membrane capacitance in -cells in intact islets as de- mentary Fig. 1D). Similar responses to adrenaline and fi scribed previously using the standard whole-cell con gura- glutamate were observed in human islets (Fig. 1D and E)and tion and IC2 and EC2. islets of C57Bl/6N mice (Fig. 1F). These data make it unlikely 2+ that paracrine effects influence the [Ca ]i responses to Data Analysis adrenaline in a-cells. Indeed, neither addition of exogenous Image sequences were analyzed (registration, background insulin nor inhibition of insulin receptor with S961 signif- subtraction, region of interest intensity vs. time analysis, fi a fi icantly modi edtheadrenalinesignalingin -cells (Supple- F/F0 calculation) using open-source FIJI software (http:// ji mentary Fig. 5B). .sc/Fiji). The numerical data were analyzed using IgorPro Glucagon-secreting a-cells are equipped with several package (WaveMetrics). For calculation of partial areas under types of voltage-gated Ca2+ channels (25). The effect of the curve (pAUCs), the recording was split into 30-s inter- adrenaline on glucagon secretion was abolished by inhibition vals, and area under the curve was computed for each v C of L-type (with isradipine) but not P/Q-type (with -agatoxin) individual interval (Supplementary Fig. 1 ), using the trap- Ca2+ channels (Fig. 2A). The changes in adrenaline-induced ezoidal integration. Numbers of measurements/cells are 2+ v fi fi [Ca ]i dynamics produced by isradipine and -agatoxin speci ed in gure legends; the experiments on human islets (Fig. 2B) correlated with the effects of these blockers on were performed on islets isolated from three donors. Sta- glucagon secretion (Fig. 2A). When isradipine was applied just tistical analysis was performed using R (21). Data are 6 U before adrenaline, the response to adrenaline was attenu- presented as mean SEM. The Mann-Whitney test or ated but not abolished (Fig. 2B). The effect of adrenaline on Wilcoxon paired test was used to compute the signifi- 2+ [Ca ]i was mimicked by noradrenaline (released by intra- cance of difference between independent and dependent islet adrenergic nerve endings [26]) and by the b-adrenergic samples, respectively. Multiple comparisons within one agonist isoprenaline but abolished by b-antagonist propran- experiment were performed using the Kruskall-Wallis test olol. By contrast, the a1-adrenergic inhibitor prazosin, with Nemenyi post hoc analysis (independent samples) or 2+ reported to inhibit the adrenaline-induced [Ca ]i increases the Friedman test with Nemenyi post hoc analysis (depen- in single a- cells (6), had no detectable effect (Fig. 2B). Thus, dent samples). a 2+ the effect of adrenaline on -cell [Ca ]i dynamics principally reflects activation of b-receptors. RESULTS We tested whether the adrenaline-induced increase in 2+ We tested the effect of adrenaline on glucagon secretion at [Ca ]i can be attributed to stimulation of a-cell action a glucose concentration that roughly approximates hypogly- potential firing by whole-cell perforated-patch measure- 2+ cemia in vivo (3 mmol/L) (22) and minimizes the activity ments (20). In agreement with the [Ca ]i measurements, of b-andd-cells (see ref. 23 and Supplementary Fig. 1B). a-cells were electrically active and fired action potentials at Adrenaline stimulated glucagon secretion from isolated afrequencyof;1 Hz when exposed to 3 mmol/L glucose. 1132 Adrenaline Controls Ca2+ Release in a-Cells Diabetes Volume 67, June 2018

Figure 2—Adrenaline’s effect in pancreatic a-cells depends on Ca2+ influx through L-type Ca2+ channels and is mediated by b-adrenergic signaling. A: Glucagon secreted from isolated NMRI mouse islets in response to 3 mmol/L glucose with or without 5 mmol/L adrenaline with or without isradipine or v-agatoxin (as indicated). P , 0.05 vs. the effect of adrenaline + 3 mmol/L glucose (*) or vs. the basal (3 mmol/L glucose + 2+ respective antagonist) group (#). B:[Ca ]i upon adrenergic (isoprenaline [n = 27] or noradrenaline [n = 19]) stimulation alone (n = 29) or with (as indicated) isradipine (n = 17, preincubated, and n = 29, acute), v-agatoxin (n =13),propranolol(n = 11), prazosin (n = 49), diazoxide (n = 20), or EGTA (n =84).P , 0.05 vs. the effect of adrenaline alone (*) or vs. basal [3 mmol/L glucose + respective (ant)agonist] of the same recording (#). C:Effectof adrenaline on a-cell action potential firing (representative of 11 experiments). Examples of action potentials recorded in the absence and presence of adrenaline (taken from the recording above as indicated) are shown.

Adrenaline depolarized a-cells by 3 mV, produced a 10-mV account for the pronounced and sustained (several minutes) C 2+ E F reduction of spike height (Fig. 2 and Supplementary Fig. increases in [Ca ]i (Fig. 1 and ). Indeed, neither hyper- F 2 ), and increased the duration of the action potential from polarizing the plasma membrane with KATP channel opener 4 to 8 ms but did not (except for a moderate stimulation diazoxide nor chelating the extracellular Ca2+ with EGTA during the initial 4 s) increase spike frequency (Supplemen- prior to the addition of adrenaline diminished its capacity to 2+ tary Fig. 2F and G). These small effects of adrenaline on increase [Ca ]i in a-cells (Fig. 2B and Supplementary Fig. a-cell electrical activity, possibly reflecting transient activa- 2C–E). tion of a depolarizing store-operated membrane conduc- b-Adrenergic signaling results in the Gs-mediated acti- tance after depletion of the sER Ca2+ stores (27), cannot vation of adenylyl cyclase and, hence, increases the cytosolic diabetes.diabetesjournals.org Hamilton and Associates 1133

concentration of cAMP ([cAMP]i) (28) (Supplementary Fig. The effects of adrenaline itself on glucagon secretion have 3), which in turn activates the downstream targets PKA previously been examined (5). Here, we explored the roles of and exchange protein directly activated by cAMP (EPAC2). PKA and EPAC2 on the cAMP-dependent stimulation of the We determined the significance of PKA and EPAC in late-stage depolarization-evoked exocytosis (monitored as increases 2+ the regulation of [Ca ]i and glucagon secretion using in membrane capacitance). Figure 3C shows the increase in the inhibitors myr-PKI (for PKA) and ESI-05 (for EPAC2). plasma membrane electrical capacitance (DCm)evokedby Both myr-PKI and ESI-05 inhibited adrenaline-induced glu- a 20-ms depolarization from 270 to 0 mV in the absence or cagon secretion (Fig. 3A)andsignificantly reduced the presence of cAMP in the patch pipette and after inhibition of 2+ adrenaline-induced increase in [Ca ]i (Fig. 3B), with ESI- PKA or EPAC2; this pulse duration is comparable with that 05 exerting a much stronger effect than myr-PKI on glucagon of the a-cell action potential (Fig. 2C). Under control con- secretion. The finding that EPAC2 and/or PKA, despite having ditions, exocytotic responses were small (,5fF).When 2+ similar effects on [Ca ]i, differentially affect adrenaline- cAMP was included in the pipette, exocytosis was stimu- induced glucagon secretion (P , 0.05) suggests that cAMP- lated by ;10-fold (Fig. 3C). The stimulatory effect of cAMP dependent activation of EPAC2 may be more tightly linked on depolarization-evoked exocytosis was resistant to PKI to exocytosis. but reduced by ESI-05 (Fig. 3C and D). Longer depolarizations

Figure 3—Adrenaline mediates its effects via elevation of [cAMP]i. A: Glucagon secretion from isolated NMRI mouse islets in response to 3 mmol/L glucose. Adrenaline, myr-PKI, or ESI-05 was added as indicated. P , 0.05 vs. basal (3 mmol/L glucose + respective antagonist) (#), vs. the effect of 2+ 3 mmol/L glucose + adrenaline (*), or vs. the effect of 3 mmol/L glucose + adrenaline in the presence of ESI-05 (¶). B:[Ca ]i upon application of (as indicated) adrenaline with or without myr-PKI (n =32)orESI-05(n =12).P , 0.05 vs. the basal of the same recording (#) or vs. the effect of 3 mmol/L glucose + adrenaline (*). C: Representative recording of the depolarization-induced increases in plasma membrane electrical capacitance. D: Exocytosis in a-cells. The pipette solution contained (as indicated) either no cAMP (n = 7) or 0.1 mmol/L cAMP (n = 11) and PKA (1 mmol/L PKI [n = 32]) or EPAC2 (25 mmol/L ESI-05 [n = 11]). P , 0.05 vs. the effect of addition of cAMP into the pipette solution (*) and vs. the control (cAMP-free) condition (#). E: The effect of adrenaline on PKA activity in a-cells (n =22)andb-cells (n = 85) within pancreatic islet cells isolated from Glu-RFP mice imaged using AKAR3 sensor on an LSM510 confocal microscope. (See also Supplementary Fig. 3.) The PKA activity is expressed as a change of the FRET ratio of the AKAR3 sensor. F: Comparison of the effects of 5 mmol/L adrenaline and 1 nmol/L–10 mmol/L forskolin on PKA activity of pancreatic islet cells (n = 870). The excerpt (lower part of panel) (n = 62) represents the data from cells activated by adrenaline (a-cells [see Fig. 3E]). Note the higher sensitivity of a-cells to forskolin: half-maximal effective concentration (EC50)=1876 50 nmol/L (n = 24) and 383 6 2+ 7nmol/L(n = 807) for a-andb-cells, respectively. G: Representative concentration–PKA activation dependence of [Ca ]i on forskolin in a-cells 2+ measured using Fluo4. H: Forskolin concentration-activation curves for PKA (solid line) (n =98)and[Ca ]i (dashed line) (n = 20). The effect of 5 mmol/L adrenaline on PKA measured in the same cells is mapped onto the curves as a shaded area (4.1 6 0.8 mmol/L forskolin). 1134 Adrenaline Controls Ca2+ Release in a-Cells Diabetes Volume 67, June 2018 evoked larger exocytotic responses, but the effects of cAMP report that the stimulatory effect of adrenaline on glucagon and the inhibitors were the same (Supplementary Fig. 4). secretion is dramatically reduced in EPAC2 knockout islets (5). fi We explored the effect of adrenaline on [cAMP]i and its The ndings that adrenaline produces a marked increase 2+ 2+ relationship to [Ca ]i further by imaging the activity of in [Ca ]i despite having only a marginal effect on a-cell cAMP’s downstream target PKA using the FRET sensor electrical activity and that it retains the capacity to stimulate AKAR3 (15). In most islet cells, the application of adrenaline glucagon secretion in the absence of extracellular Ca2+ and reduced [cAMP]i (Supplementary Fig. 3), probably reflecting inthepresenceofahighconcentrationoftheKATP channel the effect of a2-adrenergic receptors in b-(29)andd-cells activator diazoxide (to suppress electrical activity [20]) (30). However, in a subset of islet cells (;10%, correspond- suggest that it, via activation of b-receptors and cAMP- a ing to -cells), adrenaline increased [cAMP]i (Supplementary dependent stimulation of PKA and EPAC2, acts by mobiliz- C 2+ Fig. 3 ). These changes in [cAMP]i correlated with increased ing Ca from intracellular stores. We tested for the capac- a-cell activity and reduced b/d-cell activity of PKA (Fig. 3E). ity of adrenaline to induce glucagon secretion and increase 2+ We compared the adrenaline-induced increases in PKA [Ca ]i in the presence of inhibitors of inositol triphosphate activity in a-cells with those produced by increasing con- receptor (Xestospongin C [32]), sER Ca2+ ATPase (thapsigargin centrations of the adenylyl cyclase activator forskolin (Fig. or cyclopiazonic acid [CPA]), and Ca2+-induced Ca2+ release 3F and G). Forskolin dose-dependently increased PKA (ryanodine). Ryanodine, Xestospongin C, and thapsigargin 2+ activity and [Ca ]i; half-maximal effects were observed at all inhibited the stimulatory effect of adrenaline on glucagon 0.19 6 0.05 and 5.3 6 1.5 mmol/L, respectively. The secretion (Fig. 4A)but—unexpectedly—did not appear to m 2+ increase in PKA activity induced by 5 mol/L adrenaline affect the adrenaline-induced increase in [Ca ]i (Supple- was equivalent to the effect of 4.2 6 1.1 mmol/L forskolin mentary Fig. 5A). We reasoned that there might be multiple (Fig. 3H), in agreement with the observation that stimula- Ca2+-release mechanisms operating in parallel in the a-cell tion of glucagon secretion is only seen at micromolar and that Ca2+ release via any one of them might suffice to 2+ concentrations of adrenaline (5). The observation that the saturate the high-affinity Ca dye Fluo-4 (which has a Kd for 2+ 2+ adrenaline-induced increase in [Ca ]i requires high [cAMP]i Ca of 300 nmol/L). Indeed, when the response to adren- (= high forskolin) is consistent with the involvement of aline was instead assayed using the low-affinity indicator  25 m EPAC2 (Kd for cAMP 10 mol/L (31)) and the previous Fluo-4FF (Kd =9.7 mol/L), ryanodine as well as xestospongin

Figure 4—Adrenaline-induced glucagon secretion involves intracellular Ca2+ release. A: Glucagon secreted from isolated NMRI mouse islets at 3 mmol/L with or without adrenaline and ryanodine, Xestospongin C, or thapsigargin as indicated. P , 0.05 vs. basal (#) or vs. the effect of 2+ glucose + adrenaline (*). B: Representative recording of the effect of ryanodine on adrenaline-induced [Ca ]i increases visualized using Fluo4FF. 2+ The control (dashed) is superimposed with the experimental trace. C:[Ca ]i upon adrenaline stimulation alone (n =23)orinthepresenceof 21 ryanodine (n = 22), Xestospongin C (n = 30), or CPA (n = 14), as measured using Fluo4FF. D:[Ca ]i upon adrenaline stimulation alone or in the presence of Ned-19 (n =28),bafilomycin (n =36),myr-PKI+ESI-05(n =31),ormyr-PKI+ESI-05+Ned-19(n =21).P , 0.05 vs. the basal (3 mmol/L glucose + antagonist) of the same recording (#) or vs. the effect of 3 mmol/L glucose + adrenaline (*). Nonsignificant (P . 0.1) vs. the effect of 3 mmol/L glucose + adrenaline in the presence of PKI and ESI-05 (§). diabetes.diabetesjournals.org Hamilton and Associates 1135

2/2 C and CPA produced marked and statistically significant dim- Tpcn2 islets (Fig. 5C). The effect of adrenaline was 2+ B C Tpcn12/2 inution of the [Ca ]i-increasing effect (Fig. 4 and ). seemingly smaller in than in wild-type islets Prompted by the observations indicating the presence of (but this difference did not attain statistical significance). a ryanodine-independent component in adrenaline-induced Ryanodine had no inhibitory effect on glucagon secretion in 2 2 Ca2+ mobilization, we explored the possible involvement of Tpcn2 / islets (Fig. 5C). Likewise, pharmacological inhibi- acidic (lysosomal) Ca2+ stores (10) in this phenomenon. tion of NAADP (the intracellular messenger mobilizing Bafilomycin, an inhibitor of the vacuolar type-H+ ATPase, lysosomal Ca2+ via TPC2 channels) (38,39), which inhibited 2+ abolished the adrenaline effect on [Ca ]i in a-cells (Fig. 4D). glucagon secretion in wild-type islets, had no effect in 2 2 Ca2+ release from lysosomal stores is mediated by acti- Tpcn2 / islets. Genetic ablation of ADP-ribosyl cyclase 1 vation of Ca2+-andNa+-permeable (33,34) two-pore chan- (CD38), the enzyme producing NAADP, significantly reduced 2+ D B nels (TPCs) (10). We analyzed the role of these channels in the increase in [Ca ]i produced by adrenaline (Figs. 4 and 5 ) 2+ the adrenaline-induced [Ca ]i increase and glucagon secre- and abolished adrenaline-induced glucagon secretion (Fig. 5C). tion by generation of Tpcn1 or Tpcn2 knockout mice. The Finally, we tested the effects of culturing islets at mRNA of both Tpcn1 and Tpcn2 is expressed in mouse and 30 mmol/L glucose for 72 h (as an experimental paradigm 2+ human a-cells (35–37). While adrenaline increased [Ca ]i in of poorly controlled diabetes) on adrenaline’s capacity to 2/2 2+ a-cells in islets isolated from Tpcn1 mice (Fig. 5A and D), elevate [Ca ]i and stimulate glucagon secretion. “Hypergly- 2/2 it had very little effect in islets from Tpcn2 mice (Fig. 5B cemia” abolished the stimulatory action of adrenaline on D 2+ 2+ A B E and ). The effects of ablating TPC1 and TPC2 on [Ca ]i both [Ca ]i and glucagon secretion (Fig. 6 , ,and ). By correlated with those on glucagon secretion: adrenaline retained contrast, islets cultured at 11 mmol/L glucose (close to the 2/2 a stimulatory effect on glucagon secretion in Tpcn1 but not fed plasma glucose levels in mice) responded robustly to

2+ Figure 5—CD38 and TPC2 but not TPC1 mediate the adrenaline response in a-cells. A: Effect of adrenaline on [Ca ]i in a-cells within islets 2/2 2/2 2+ 2/2 2/2 isolated from TPC1 or TPC2 mice as indicated. B:[Ca ]i upon adrenaline stimulation in wild-type, TPC1 (n =84),TPC2 (n =27),or CD382/2 (n = 153) mouse a-cells. P , 0.05 vs. basal (3 mmol/L glucose) of the same recording (#) or vs. the effect of 3 mmol/L glucose + adrenaline in wild-type animals (§). C: Glucagon secretion from isolated mouse islets in response to low glucose (3 mmol/L) or adrenaline in the absence/presence of Ned-19 or ryanodine, measured in wild-type C57Bl/6n and TPC12/2,TPC22/2, and CD382/2 mice. P , 0.05 vs. basal (3 mmol/L glucose) (#), vs. the effect of 3 mmol/L glucose + adrenaline within the same genotype (*), or vs. the effect of 3 mmol/L glucose + adrenaline in wild-type animals (§). 1136 Adrenaline Controls Ca2+ Release in a-Cells Diabetes Volume 67, June 2018

2+ adrenaline with both a [Ca ]i increase and stimulation of dependent and -independent mechanisms and that EPAC2 glucagon secretion. “Hyperglycemia” did not affect the re- may act downstream of PKA (Fig. 3C and D). sponsiveness to glutamate (Fig. 6B)oreffectsofadrenaline Given the weak effect of adrenaline on a-cell electrical on [cAMP]i and PKA activity (Fig. 6C and D). activity, the high sensitivity of both adrenaline-induced 2+ glucagon secretion and [Ca ]i increase after inhibition of DISCUSSION L-type Ca2+ channels may seem paradoxical. However, this 2+ Adrenaline-induced glucagon secretion involves both PKA- effect may be secondary to depletion of intracellular Ca 2+ and EPAC2-dependent mechanisms (5). Here, we have ex- stores when Ca entry is blocked. Indeed, adrenaline re- 2+ plored the downstream mechanisms with a focus on intra- tained most of its effect on [Ca ]i when isradipine was cellular Ca2+ dynamics. Our data are suggestive of a hierarchy added just before adrenaline (Fig. 2B). 2+ 2+ of intracellular Ca signaling events and highlight a decisive The mechanism by which adrenaline increases [Ca ]i in 2+ role of lysosomal stores and NAADP-activated Ca2+ release a-cells has variably been attributed to increased Ca channel via TPC2 channels. activity (8) or activation of a1-adrenoreceptors (6). Our Pharmacological inhibition of EPAC2 nearly abolished the study unveils a previously unrecognized role of acidic Ca2+ stimulatory effect of adrenaline on both glucagon secretion stores in regulation of glucagon secretion. Ca2+ liberated by 2+ and [Ca ]i. By contrast, inhibition of PKA, while lowering adrenaline-induced PKA and EPAC2 signaling from the 2+ 2+ [Ca ]i as strongly as inhibition of EPAC2, had only a mod- acidic stores then triggers further Ca increase by activating erate effect on glucagon secretion. This suggests that a full ryanodine-sensitive Ca2+-induced Ca2+ release (40). This 2+ response to adrenaline requires the activation of both PKA- hierarchy of intracellular [Ca ]i signaling is supported by

2+ Figure 6—The adrenaline effect is attenuated by chronic hyperglycemia. A: Typical [Ca ]i responses to 5 mmol/L adrenaline recorded in mouse islet a-cells precultured at 11 mmol/L (n =102)or30mmol/L(n = 146) glucose for 48 h. B: Quantification of the data presented in A. P , 0.05 vs. the basal (3 mmol/L glucose) of the same recording (#) or vs. the effect of 3 mmol/L glucose + adrenaline in islets precultured in 11 mmol/L glucose (¶). C: cAMP responses to adrenaline in pancreatic islet a-cells precultured at 11 (n = 164) or 30 (n = 212) mmol/L glucose. D: PKA responses to adrenaline in pancreatic islet a-cells precultured at 11 (n = 1,277) or 30 (n = 1,323) mmol/L glucose. E: Glucagon secretion from pancreatic islets precultured at 11 (n =10)or30(n = 10) mmol/L glucose. P , 0.05 vs. the basal (3 mmol/L glucose) condition (#) or vs. the respective effect in islets preculturedin11mmol/Lglucose(¶). diabetes.diabetesjournals.org Hamilton and Associates 1137 the fact that ryanodine exerts no additive inhibitory effect seem at odds with the observation that adrenaline is without on glucagon secretion in Tpc2 knockout islets. effect on glucagon secretion in human islets (23). It is likely Our data suggest a link between PKA/EPAC2 signaling and that adrenaline produces a transient stimulation of glucagon the Tpc2 channel activity, withtheNAADPantagonistNed19 secretion that escapes detection during a 1-h incubation in having no additive inhibitory effect on the adrenaline-induced human islets. We can exclude though that this discrepancy 2+ [Ca ]i increase when PKA and EPAC2 activity was pharma- stems from different glucose concentrations used for secre- 2+ cologically suppressed (Fig. 4D). The production of NAADP tion measurements (1 mmol/L) and [Ca ]i imaging (3 mmol/L). is influenced by high concentrations of cAMP—as expected In a series of initial experiments, adrenaline also increased 2+ for an effect mediated by low-affinity cAMP sensor EPAC2— [Ca ]i in human islets exposed to 1 mmol/L glucose (Sup- and PKA-dependent phosphorylation of S66 in Tpc2 in- plementary Fig. 1A). creases the channel’s open probability (41). The involvement Circulating levels of adrenaline are below 1 nmol/L, and of NAADP production is also illustrated by the ability of the they remain below 100 nmol/L even during exercise (47). genetic knockdown of the NAADP-generating enzyme CD38 Interestingly, these low concentrations of adrenaline inhibit 2+ to prevent the effects of adrenaline on both [Ca ]i and glucagon secretion, while the stimulatory effect becomes glucagon secretion. NAADP-dependent adrenaline-induced detectable at concentrations .1 mmol/L and depends on Ca2+ release via Tpc2 channels was recently implicated in the activation of EPAC2 (5). Such high levels of agonist are transduction of b-adrenergic signals in cardiomyocytes (42) unlikely to occur anywhere except close to nerve terminals but not pancreatic b-cells (43,44). The observation that (48). Thus, our data suggest that the sympathetic signal 2 2 Tpcn2 / mice have normal fasting blood glucose (43) is not mediated by locally released noradrenaline (Fig. 2B)plays in conflict with the idea that adrenaline stimulates glucagon a key role in linking the physiological “flight-or-fight” re- secretion by a TPC2-dependent mechanism, as adrenaline sponse to the a-cell glucagon release. The schematic in Fig. 7 itself will stimulate hepatic glucose production even in the presents a model for adrenaline signaling in pancreatic absence of glucagon, and low glucose also stimulates gluca- a-cells. Hypoglycemia elevates systemic adrenaline (released gon secretion by a direct (intrinsic) mechanism independent from the adrenal glands) and/or triggers a local release of of systemic signals (20,45). noradrenaline from sympathetic nerve endings within or The innervation of mouse and human islets is rather adjacent to pancreatic islets. The catecholamine then binds fi ’ 2+ b a different (46). The nding that adrenaline s effects on [Ca ]i to a -adrenoreceptor on the -cell, resulting in increased are essentially the same in mouse and human a-cells may cytosolic cAMP levels, which activate PKA (which interacts with the Tpc2 channel residing inthemembraneoftheacidic vesicles) and EPAC2 (which facilitates the liberation of NAADP byCD38).ThetwosignalsconvergeintoareleaseofCa2+ from the acidic stores into the cytosol, which in turn triggers further liberation of Ca2+ from ER. Altogether, this leads to a fourfold stimulation of the glucagon release and, via stimulation of hepatic gluconeogenesis, rapid restoration of blood glucose levels. This process depends on background electrical activity of the a-cell and plasmalemmal Ca2+ entry (explaining the effect of isradipine) to refill intracellular Ca2+ stores. The mechanism underlying the attenuation of the sym- pathoadrenal response in diabetes remains debated. Diabetes is associated with the loss of sympathetic islet innervation, and this may, via reduced glucagon secretion, account for the increased risk of hypoglycemia (49). It is therefore of interest that our in vitro model of chronic hyperglycemia lowers the a-cell’s intrinsic responsiveness to adrenaline (Fig. 6) and thus can itself result in the sympathetic islet neuropathy. The exact mechanism remains to be identified but is likely to be downstream of cAMP and activation of PKA, which were both unaffected. Novel therapeutic strategies that bypass the innervation may help to restore normal counterregulation in patients with diabetes.

Acknowledgments. The authors thank Dr. Jin Zhang (Department of Phar- macology and Molecular Sciences, The Johns Hopkins University School of Medicine, Figure 7—Model of adrenaline-induced glucagon secretion, as de- Baltimore, MD) for the gift of AKAR3. scribed in detail in the main text. AC, adenylyl cyclase; Cav1.x, L-type voltage-gated Ca2+ channel; Gs, G-protein, s a-subunit; RyR, rya- Funding. This work was supported by Medical Research Council program grant nodine receptor. G0901521. A.H. is a recipient of a Diabetes UK PhD Studentship. Q.Z. is a Diabetes UK 1138 Adrenaline Controls Ca2+ Release in a-Cells Diabetes Volume 67, June 2018

RD Lawrence Fellow. G.A.R., A.G., and P.R. hold Wellcome Trust Senior Investigator 19. Panagiotidis G, Salehi AA, Westermark P, Lundquist I. Homologous islet amyloid awards (102828, 098424, and 095531). During the initial stages of the project, polypeptide: effects on plasma levels of glucagon, insulin and glucose in the mouse. A.I.T. held an Oxford Biomedical Research Council postdoctoral fellowship. Diabetes Res Clin Pract 1992;18:167–171 Duality of Interest. A.K.R. is a current employee of Novo Nordisk. No other 20. Zhang Q, Ramracheya R, Lahmann C, et al. Role of KATP channels in glucose- potential conflicts of interest relevant to this article were reported. regulated glucagon secretion and impaired counterregulation in type 2 diabetes. Cell Author Contributions. A.S., M.W., J.G.K., A.K.R., C.E.C., and R.R. per- Metab 2013;18:871–882 formed the experiments. A.H., Q.Z., A.G.-A., and A.I.T. performed the experiments 21. R: a language and environment for statistical computing [Internet], 2016. and analyzed data. N.C.S., M.Z., D.B., and A.G. provided reagents. M.Z., G.A. Available from https://www.R-project.org/. Accessed 6 April 2018 R., A.G., P.R., and A.I.T. contributed to data interpretation. G.A.R., A.G., P.R., 22. Cryer PE. Hypoglycemia, functional brain failure, and brain death. J Clin Invest and A.I.T. wrote the manuscript. P.R.V.J. provided human islets. P.R. and A.I. 2007;117:868–870 T. are the guarantors of this work and, as such, had full access to all the data in 23. Walker JN, Ramracheya R, Zhang Q, Johnson PR, Braun M, Rorsman P. the study and take responsibility for the integrity of the data and the accuracy of Regulation of glucagon secretion by glucose: paracrine, intrinsic or both? Diabetes thedataanalysis. Obes Metab 2011;13(Suppl. 1):95–105 24. Otter S, Lammert E. Exciting times for pancreatic islets: glutamate signaling in References endocrine cells. Trends Endocrinol Metab 2016;27:177–188 1. Cryer PE. Hypoglycaemia: the limiting factor in the glycaemic management of 25. Rorsman P, Ramracheya R, Rorsman NJ, Zhang Q. ATP-regulated potassium – type I and type II diabetes. Diabetologia 2002;45:937 948 channels and voltage-gated calcium channels in pancreatic alpha and beta cells: 2. Gromada J, Franklin I, Wollheim CB. Alpha-cells of the endocrine pancreas: similar functions but reciprocal effects on secretion. Diabetologia 2014;57:1749–1761 – 35 years of research but the enigma remains. Endocr Rev 2007;28:84 116 26. Esterhuizen AC, Spriggs TL, Lever JD. Nature of islet-cell innervation in the cat 3. Taborsky GJ Jr., Mundinger TO. Minireview: the role of the autonomic nervous pancreas. Diabetes 1968;17:33–36 system in mediating the glucagon response to hypoglycemia. Endocrinology 2012; 27. Liu YJ, Vieira E, Gylfe E. A store-operated mechanism determines the activity of – 153:1055 1062 the electrically excitable glucagon-secreting pancreatic alpha-cell. Cell Calcium 2004; 4. Skrivarhaug T, Bangstad HJ, Stene LC, Sandvik L, Hanssen KF, Joner G. Long- 35:357–365 term mortality in a nationwide cohort of childhood-onset type 1 diabetic patients in 28. Wallukat G. The beta-adrenergic receptors. Herz 2002;27:683–690 – Norway. Diabetologia 2006;49:298 305 29. Ismail NA, El-Denshary ES, Idahl LA, Lindström P, Sehlin J, Täljedal IB. Effects of 5. De Marinis YZ, Salehi A, Ward CE, et al. GLP-1 inhibits and adrenaline stimulates alpha-adrenoceptor agonists and antagonists on insulin secretion, calcium uptake, and glucagon release by differential modulation of N- and L-type Ca2+ channel-dependent rubidium efflux in mouse pancreatic islets. Acta Physiol Scand 1983;118:167–174 – exocytosis. Cell Metab 2010;11:543 553 30. Hauge-Evans AC, King AJ, Fairhall K, Persaud SJ, Jones PM. A role for islet 6. Vieira E, Liu Y-J, Gylfe E. Involvement of alpha1 and beta-adrenoceptors in somatostatin in mediating sympathetic regulation of glucagon secretion. Islets 2010;2: adrenaline stimulation of the glucagon-secreting mouse alpha-cell. Naunyn 341–344 Schmiedebergs Arch Pharmacol 2004;369:179–183 31. Ozaki N, Shibasaki T, Kashima Y, et al. cAMP-GEFII is a direct target of cAMP in 7. Johansson H, Gylfe E, Hellman B. Cyclic AMP raises cytoplasmic calcium in regulated exocytosis. Nat Cell Biol 2000;2:805–811 pancreatic alpha 2-cells by mobilizing calcium incorporated in response to glucose. 32. Gafni J, Munsch JA, Lam TH, et al. Xestospongins: potent membrane permeable Cell Calcium 1989;10:205–211 blockers of the inositol 1,4,5-trisphosphate receptor. Neuron 1997;19:723–733 8. Gromada J, Bokvist K, Ding WG, et al. Adrenaline stimulates glucagon secretion 33. Ruas M, Davis LC, Chen CC, et al. Expression of Ca2+-permeable two-pore in pancreatic A-cells by increasing the Ca2+ current and the number of granules close channels rescues NAADP signalling in TPC-deficient cells. EMBO J 2015;34:1743–1758 to the L-type Ca2+ channels. J Gen Physiol 1997;110:217–228 34. Wang X, Zhang X, Dong XP, et al. TPC proteins are phosphoinositide- activated 9. Pipeleers DG, Schuit FC, Van Schravendijk CF, Van de Winkel M. Interplay of sodium-selective ion channels in endosomes and lysosomes. Cell 2012;151:372–383 nutrients and hormones in the regulation of glucagon release. Endocrinology 1985; 35. DiGruccio MR, Mawla AM, Donaldson CJ, et al. Comprehensive alpha, beta and 117:817–823 delta cell transcriptomes reveal that ghrelin selectively activates delta cells 10. Calcraft PJ, Ruas M, Pan Z, et al. NAADP mobilizes calcium from acidic or- and promotes somatostatin release from pancreatic islets. Mol Metab 2016;5: ganelles through two-pore channels. Nature 2009;459:596–600 449–458 2+ 11. Lin WK, Bolton EL, Cortopassi WA, et al. Synthesis of the Ca -mobilizing 36. Adriaenssens AE, Svendsen B, Lam BY, et al. Transcriptomic profiling of messengers NAADP and cADPR by intracellular CD38 enzyme in the mouse heart: role pancreatic alpha, beta and delta cell populations identifies delta cells as a principal in b-adrenoceptor signaling. J Biol Chem 2017;292:13243–13257 target for ghrelin in mouse islets. Diabetologia 2016;59:2156–2165 12. Lake SP, Bassett PD, Larkins A, et al. Large-scale purification of human islets 37. Blodgett DM, Nowosielska A, Afik S, et al. Novel observations from next- utilizing discontinuous albumin gradient on IBM 2991 cell separator. Diabetes 1989; generation RNA sequencing of highly purified human adult and fetal islet cell subsets. 38(Suppl. 1):143–145 Diabetes 2015;64:3172–3181 13. Ricordi C, Lacy PE, Finke EH, Olack BJ, Scharp DW. Automated method for 38. Naylor E, Arredouani A, Vasudevan SR, et al. Identification of a chemical probe for isolation of human pancreatic islets. Diabetes 1988;37:413–420 NAADP by virtual screening. Nat Chem Biol 2009;5:220–226 14. Chen TW, Wardill TJ, Sun Y, et al. Ultrasensitive fluorescent proteins for imaging 39. Galione A. A primer of NAADP-mediated Ca(2+) signalling: from sea urchin eggs neuronal activity. Nature 2013;499:295–300 to mammalian cells. Cell Calcium 2015;58:27–47 15. Allen MD, Zhang J. Subcellular dynamics of protein kinase A activity visualized by 40. Cancela JM, Churchill GC, Galione A. Coordination of agonist-induced FRET-based reporters. Biochem Biophys Res Commun 2006;348:716–721 Ca2+-signalling patterns by NAADP in pancreatic acinar cells. Nature 1999;398:74–76 16. Liu X, Gong H, Li X, Zhou W. Monitoring calcium concentration in neurons with 41. Lee CS, Tong BC, Cheng CW, Hung HC, Cheung KH. Characterization of two-pore cameleon. J Biosci Bioeng 2008;105:106–109 channel 2 by nuclear membrane electrophysiology. Sci Rep 2016;6:20282 17. Cabrera O, Jacques-Silva MC, Speier S, et al. Glutamate is a positive autocrine 42. Capel RA, Bolton EL, Lin WK, et al. Two-pore channels (TPC2s) and nicotinic acid signal for glucagon release. Cell Metab 2008;7:545–554 adenine dinucleotide phosphate (NAADP) at lysosomal-sarcoplasmic reticular junc- 18. Samols E, Weir GC. Adrenergic modulation of pancreatic A, B, and D cells alpha- tions contribute to acute and chronic b-adrenoceptor signaling in the heart. J Biol adrenergic suppression and beta-adrenergic stimulation of somatostatin secretion, Chem 2015;290:30087–30098 alpha-adrenergic stimulation of glucagon secretion in the perfused dog pancreas. J 43. Arredouani A, Ruas M, Collins SC, et al. Nicotinic acid adenine dinucleotide Clin Invest 1979;63:230–238 phosphate (NAADP) and endolysosomal two-pore channels modulate membrane diabetes.diabetesjournals.org Hamilton and Associates 1139

excitability and stimulus-secretion coupling in mouse pancreatic b cells. J Biol 47. Stich V, de Glisezinski I, Crampes F, et al. Activation of antilipolytic alpha(2)- Chem 2015;290:21376–21392 adrenergic receptors by epinephrine during exercise in human adipose tissue. Am J 44. Cane MC, Parrington J, Rorsman P, Galione A, Rutter GA. The two pore channel Physiol 1999;277:R1076–R1083 TPC2 is dispensable in pancreatic b-cells for normal Ca2+ dynamics and insulin 48. Ludwig J, Gerlich M, Halbrügge T, Graefe KH. The synaptic noradrenaline secretion. Cell Calcium 2016;59:32–40 concentration in humans as estimated from simultaneous measurements of 45. Vieira E, Salehi A, Gylfe E. Glucose inhibits glucagon secretion by a direct effect plasma noradrenaline and dihydroxyphenylglycol (DOPEG). J Neural Transm Suppl on mouse pancreatic alpha cells. Diabetologia 2007;50:370–379 1990;32:441–445 46. Rodriguez-Diaz R, Abdulreda MH, Formoso AL, et al. Innervation patterns 49. Mundinger TO, Taborsky GJ Jr. Early sympathetic islet neuropathy in auto- of autonomic axons in the human endocrine pancreas. Cell Metab 2011;14: immune diabetes: lessons learned and opportunities for investigation. Diabetologia 45–54 2016;59:2058–2067