Endogenous α2A-Adrenoceptor–Operated Sympathoadrenergic Tones Attenuate Insulin Secretion via cAMP/TRPM2 Signaling

Adrenaline secreted from the chromaffin cells of the adrenal medulla and noradrenaline released from sympathetic nerve endings inhibit insulin release from the pancreatic islets and elevate blood glucose levels. In the islet β-cells, catecholamines are reported to inhibit insulin release via multiple mechanisms: activation of α2-adrenoceptors coupled to pertussis toxin–sensitive Gi/Go proteins resulting in inhibition of adenylyl cyclase (1), activation of hyperpolarizing K⁺ currents (2,3), inhibition of Ca²⁺ currents (4,5), and direct inhibition of exocytosis at a distal step (6–8). Nevertheless, the significance of these signaling pathways for physiological adrenergic functions in β-cells is largely unknown, because pharmacologically higher concentrations (µmol/L order) of catecholamines have been used for the analysis of these signaling mechanisms.

During the process of glucose-stimulated insulin secretion in β-cells, opening of the background inward current through nonselective cation channels (NSCCs) might facilitate depolarization after glucose metabolism–induced closure of the ATP-sensitive K⁺ (K_ATP) channels (9–13). We previously reported that the transient receptor potential melastatin 2 (TRPM2) channel, a type of NSCC, in β-cells plays an essential role in glucose-induced and incretin-potentiated insulin secretion (9). Glucose metabolism and glucagon-like peptide 1 (GLP-1) receptor stimulation both increase the activity of TRPM2 channels via cAMP signaling (9). Thus, Gi/Go-mediated inhibition of cAMP production is expected to attenuate the channel activity. However, whether the modulation of TRPM2 channels in β-cells is implicated in the α2-adrenoceptor–operated attenuation of insulin release remains unknown.

In this study, we examined whether insulinostatic adrenaline action is mediated via TRPM2 channel activity in islet β-cells. The results showed that adrenaline physiologically exerts its insulinostatic action via α2A-adrenoceptor–mediated attenuation of TRPM2 channels.

C57BL/6J mice (CLEA Japan, Inc.) were housed in accordance with the institutional guidelines for animal care in an air-conditioned room with a 12-h light/dark cycle. Food and water were available ad libitum. All animal experiments were approved by an institutional ethics committee. TRPM2-knockout (KO) mice were provided by Y. Mori (Kyoto University), and these mice were backcrossed with the C57BL/6J strain for at least nine generations at Jichi Medical University (9,14,15).

Islets of Langerhans were isolated from male C57BL/6J mice (8–12 weeks of age) by collagenase (Sigma-Aldrich, Tokyo, Japan) digestion, according to a previously reported method (16,17). Briefly, the animals were anesthetized intraperitoneally (i.p.) with pentobarbital (30–100 mg/kg), followed by an injection of collagenase (1 mg/mL) dissolved in HEPES-added Krebs-Ringer bicarbonate (HKRB) buffer solution containing (in mmol/L): 129 NaCl, 5.0 NaHCO₃, 4.7 KCl, 1.2 KH₂PO₄, 2.0 CaCl₂, 1.2 MgSO₄, and 10 HEPES at pH 7.4 with NaOH, supplemented with 5.6 mmol/L glucose and 0.1% BSA (purchased from Wako, Osaka, Japan) into the bile duct. The collagenase-injected pancreas was incubated at 37°C for 16 min in HKRB buffer. The islets collected were used for insulin release experiments and cAMP measurements. For electrophysiological experiments, the collected islets were dispersed into single cells and maintained for up to 2 days in Eagle’s minimal essential medium containing 5.6 mmol/L glucose supplemented with 10% FBS (purchased from Biowest, Nuaillé, France), 100 µg/mL streptomycin, and 100 units/mL penicillin in 95% air with 5% CO₂ at 37°C.

Measurement of insulin secretion was performed as follows: size-matched batches of 10 islets were incubated for 30 min at 37°C in HKRB with 2.8 mmol/L glucose for stabilization, followed by incubation for 60 min in HKRB with 2.8 mmol/L or 16.6 mmol/L glucose. Exendin (Ex)-4, a GLP-1 receptor agonist, and adrenaline (Wako) were present throughout the incubation. The concentration of secreted insulin was determined using an ELISA kit (Morinaga Institute of Biological Science, Yokohama, Japan).

Measurement of cAMP was performed as follows: size-matched batches of five islets were incubated for 30 min at 37°C in HKRB with 2.8 mmol/L glucose for stabilization, followed by incubation for 60 min in HKRB with 2.8 mmol/L, 5.6 mmol/L, or 16.6 mmol/L glucose containing 500 μmol/L 3-isobutyl-1-methylxanthine (IBMX). After incubation, islets were sonicated adequately, and cAMP production in islets was determined using an enzyme immunoassay kit (GE Healthcare, Buckinghamshire, U.K.).

Perforated whole-cell clamped currents were recorded with a pipette solution containing amphotericin B dissolved in 0.1% DMSO, using an amplifier (Axopatch 200B; Molecular Devices, Sunnyvale, CA) in a computer using pCLAMP 10.2 software, as previously reported (16,17). The pipette solution contained amphotericin B (200 µg/mL), 40 mmol/L K₂SO₄, 50 mmol/L KCl, 5 mmol/L MgCl₂, 0.5 mmol/L EGTA, and 10 mmol/L HEPES at pH 7.2 with KOH. Cell capacitances recorded from 204 single β-cells were 5.26 ± 0.08 picofarad (pF). The β-cells were voltage-clamped at a holding potential of −70 mV in the presence of 100 µmol/L tolbutamide, which is a sufficient concentration to specifically block the K_ATP channels; the residual current is a background current corresponding to NSCC conductance, as previously reported (9,10). Measurement of membrane potential was performed by current-clamp mode. These electrophysiological experiments were performed at 26–28°C. The following reagents were used: adrenaline, Ex-4, dibutyryl (db)-cAMP (Sigma-Aldrich), yohimbine (Sigma-Aldrich), BRL44408 (Abcam), JP1302 (Abcam), prazosin (Sigma-Aldrich), propranolol (Sigma-Aldrich), and 2-aminoethyl diphenylborinate (2-APB; Sigma-Aldrich).

Intraperitoneal glucose tolerance tests (IPGTTs) were performed with wild-type mice and TRPM2-KO mice fasted overnight, as previously reported (18,19). Glucose (2 g/kg) was injected intraperitoneally into the wild-type and TRPM2-KO mice, followed by blood sampling from the tail vein at 15 min, 30 min, 60 min, and 120 min, in addition to a baseline sampling. Saline (0.1 mL/10 g body weight, i.p.) or yohimbine (0.1 mg/kg, i.p.; Sigma-Aldrich) was administered 30 min before the glucose challenge. Blood glucose concentrations were measured using a GlucoCard DIA meter (ARKRAY, Kyoto, Japan).

Data are expressed as mean ± SEM. Statistical analyses were performed by the unpaired or paired Student t test, or one-way ANOVA, followed by Bonferroni multiple comparison tests with GraphPad Prism 6.0 software. P values <0.05 were considered statistically significant.

Effects of various concentrations of adrenaline on insulin release in mouse isolated islets were measured under stimulatory (16.6 mmol/L) glucose conditions. Insulin release from islets under batch-incubation conditions was greater with 16.6 mmol/L glucose than with 2.8 mmol/L glucose (P < 0.01), and treatment with Ex-4 (0.1 nmol/L) significantly enhanced the glucose (16.6 mmol/L)-induced insulin release (Fig. 1A). Adrenaline inhibited glucose-induced and Ex-4–potentiated insulin release in a concentration-dependent manner, with IC₅₀ values of 0.798 nmol/L and 0.797 nmol/L, respectively (Fig. 1A). At a low concentration of 1 nmol/L, the glucose (16.6 mmol/L)-induced and Ex-4 (0.1 nmol/L)–potentiated insulin release was reduced by 60% and 80%, respectively. Adrenaline (1 nmol/L) did not affect basal insulin secretion at 2.8 mmol/L glucose (Supplementary Fig. 1A). A higher concentration (5 µmol/L) of adrenaline displayed further inhibitory effects and completely attenuated glucose-induced and Ex-4–potentiated insulin release (Fig. 1A).

We examined whether the attenuation of the TRPM2 channel is involved in adrenaline-induced suppression of insulin release. In islets isolated from TRPM2-KO mice, the glucose (16.6 mmol/L)-induced insulin release was significantly lower than that from wild-type mice, whereas basal insulin release at low glucose remained unchanged (Fig. 1B), as previously reported (15). Similarly, Ex-4–potentiated insulin release was markedly reduced in the TRPM2-KO islets. Low concentrations of adrenaline did not attenuate glucose (16.6 mmol/L)-induced or Ex-4–potentiated insulin release in the TRPM2-KO islets (Fig. 1B). In contrast, a higher concentration of adrenaline (5 µmol/L) inhibited the insulin release in islets from TRPM2-KO and from wild-type mice (Fig. 1A and B). These results suggest that the physiological action of adrenaline to attenuate NSCC current through the TRPM2 channel is causally implicated in its insulinostatic action.

The NSCC current in mouse β-cells under amphotericin B–perforated whole-cell clamp was measured in the presence of 100 µmol/L tolbutamide, which inhibits the K_ATP channels; excluding the contamination of the K_ATP channel current. At a holding potential of −70 mV, an increase in external glucose concentration from 2.8 mmol/L to 16.6 mmol/L increased the NSCC current in β-cells (Fig. 2A), confirming previous reports (9). Adrenaline (1 nmol/L) markedly decreased the NSCC current elicited by 16.6 mmol/L glucose (Fig. 2B), and current densities were attenuated by adrenaline in a concentration-dependent manner (Fig. 2C). Adrenaline (1 nmol/L) did not affect basal NSCC current at 2.8 mmol/L glucose (Supplementary Fig. 1B). Furthermore, a GLP-1 receptor agonist, Ex-4 (0.1 nmol/L), increased the NSCC current in β-cells under 5.6 mmol/L glucose conditions (Fig. 2D), as previously reported (9). Adrenaline, at a concentration of 1 nmol/L or higher, strongly and dose-dependently inhibited the NSCC current (Fig. 2E and F). Ex-4 (0.1 nmol/L) also increased the NSCC current in β-cells under 16.6 mmol/L glucose conditions, as previously reported (9), and adrenaline (1 nmol/L) did attenuate it (Supplementary Fig. 1C).

We previously demonstrated that this glucose- and Ex-4–induced NSCC current is inhibited by 2-APB, a TRPM2 channel blocker, and is not elicited in TRPM2-KO β-cells, indicating that the current is passing through activated TRPM2 channels (9). As confirmed, neither glucose (16.6 mmol/L) nor Ex-4 (0.1 nmol/L) induced NSCC currents in β-cells in the presence of 2-APB (10 µmol/L) (Supplementary Fig. 2A and B), and current densities did not change compared with their controls (Supplementary Fig. 2C and D).

To identify the adrenergic receptor subtype involved in the inhibition of NSCC currents, β-cells were treated with subtype-selective adrenergic receptor antagonists. In the presence of yohimbine (1 μmol/L), an α2-adrenoceptor antagonist, adrenaline (1 nmol/L) failed to attenuate the NSCC currents elicited by Ex-4 (Fig. 2G). Furthermore, in β-cells treated with BRL44408 (10 µmol/L), which is a selective α2A-adrenoceptor antagonist, the Ex-4–induced currents were not altered by adrenaline (Fig. 2H). In contrast, JP1302 (1 µmol/L), a selective α2C-adrenoceptor antagonist, did not affect the inhibitory effects of adrenaline on the Ex-4–induced NSCC currents (Fig. 2I). Prazosin (10 µmol/L), an α1-adrenoceptor antagonist, and propranolol (10 µmol/L), a β-adrenoceptor antagonist, also did not affect the inhibitory effects of adrenaline (Fig. 2J and K). These pharmacological profiles of the receptor antagonists indicated that adrenaline affects the β-cell NSCC currents via the α2A-adrenoceptor but not via the α2C-, α1-, or β-adrenoceptors.

In the presence of the phosphodiesterase inhibitor, IBMX (500 µmol/L), 16.6 mmol/L glucose stimulated cAMP production in the islets under static incubation compared with 2.8 mmol/L glucose (P < 0.01) (Fig. 3A). The glucose (16.6 mmol/L)-induced cAMP increase was strongly suppressed by treatment with adrenaline (1 nmol/L) and could not be further inhibited by increasing the concentration (5 µmol/L). Furthermore, activation of the GLP-1 receptor with Ex-4 (0.1 nmol/L) at 5.6 mmol/L glucose increased cAMP production in the islets. This increase was strongly suppressed by 1 nmol/L adrenaline, and its cAMP inhibitory action was comparable to that at a higher concentration of 5 µmol/L (Fig. 3B). In isolated islets from TRPM2-KO mice, adrenaline at 1 nmol/L and 5 µmol/L significantly attenuated the Ex-4 (0.1 nmol/L)–induced cAMP production (Fig. 3C), indicating that α2-adrenoceptor–mediated inhibition of cAMP production is intact in the islets of TRPM2-KO mice.

We next examined whether the adrenaline-induced attenuation of cAMP production is involved in the adrenaline action to suppress NSCC currents. As shown in Fig. 4A, exposure of β-cells to 1 mmol/L db-cAMP, a membrane-permeable cAMP analog, with 5.6 mmol/L glucose induced an NSCC current increase, as previously reported (9). Adrenaline (1 nmol/L) failed to inhibit this db-cAMP–induced NSCC current increase (Fig. 4B and D), suggesting that physiological concentrations of adrenaline affect cAMP signaling and thereby attenuate the glucose-induced and GLP-1–potentiated TRPM2 current increase in β-cells. In contrast, a higher concentration of adrenaline (5 µmol/L) suppressed the db-cAMP–induced NSCC current increase (Fig. 4C and D), suggesting cAMP-independent action by pharmacological concentrations of adrenaline.

To determine whether adrenaline affects plasma membrane potential as a result of NSCC current attenuation, the membrane potential in β-cells was recorded. Under conditions of perforated whole-cell current-clamp mode, a rise in the perfusate glucose concentration from 2.8 mmol/L to 16.6 mmol/L depolarized the plasma membrane from −68.3 mV to −31.6 mV and elicited the firing of action potentials in β-cells (Fig. 5A). In contrast, in the presence of adrenaline (1 nmol/L), the membrane potential did not depolarize as with 16.6 mmol/L glucose alone (Fig. 5B); the membrane depolarization was slowed with subsequent repolarization after peak depolarization in adrenaline-treated β-cells. In addition, Ex-4–elicited depolarization was attenuated by low concentrations of adrenaline (Supplementary Fig. 3). The adrenaline-treated cells clearly displayed a prolonged latent period (lag time) to electrical burst of action potentials (259.3 ± 63.2 s) compared with the control cells (97.2 ± 13.4 s) (Fig. 5E), indicating that low concentrations of adrenaline also delayed the glucose-induced membrane depolarization. Furthermore, the adrenaline effects on membrane potential were counteracted by 10 µmol/L BRL44408 (Fig. 5C–E). These findings suggest that physiological concentrations of adrenaline could reduce the β-cell membrane excitability required for insulin secretion stimulated by glucose metabolism, via the α2A-adrenoceptor, which is prompt and adequate for depolarization. A higher concentration (5 µmol/L) of adrenaline reversed the β-cell membrane depolarization induced by high glucose (Fig. 5F and Supplementary Fig. 4A) and Ex-4 (Supplementary Fig. 4B), as previously reported (2,20–22).

Tolbutamide was used to further assess the effects of low and high concentrations of adrenaline on the membrane excitability in β-cells. Under the nonstimulatory glucose condition (2.8 mmol/L), tolbutamide (100 µmol/L)-induced depolarization in β-cells was not affected by a low concentration (1 nmol/L) of adrenaline but was significantly hyperpolarized by a high concentration (5 µmol/L) of adrenaline (Fig. 6A and B). In isolated islets, furthermore, tolbutamide-induced insulin secretion at 2.8 mmol/L glucose was not affected by a low concentration (1 nmol/L) of adrenaline but was suppressed by a high concentration (5 µmol/L) (Fig. 6C).

To assess the role of endogenous α2-adrenoceptor–operated sympathoadrenergic tones in vivo, the effects of α2-adrenoceptor antagonists on systemic glucose levels were studied in mice. In IPGTTs, when an α2-adrenoceptor antagonist, yohimbine (0.1 mg/kg), was administered i.p. 30 min before 2 g/kg glucose injection into the wild-type mice, increases in blood glucose levels at 30, 60, and 120 min were significantly attenuated (Fig. 7A), indicating the physiological functions of endogenous sympathoadrenergic tones to increase blood glucose in the IPGTT via the α2-adrenoceptor. To examine whether these functions were mediated by TRPM2 channels, IPGTT was performed in TRPM2-KO mice. The TRPM2-KO mice displayed a higher increase in blood glucose levels during IPGTT compared with wild-type mice (Fig. 7B). The area under the curve (AUC) of blood glucose increase for 120 min during IPGTT in TRPM2-KO mice was significantly higher than that in wild-type mice. Furthermore, administration of yohimbine (0.1 mg/kg, i.p.) failed to affect glucose tolerance in TRPM2-KO mice, in contrast to a decrease in the AUC of blood glucose rise in wild-type mice receiving the antagonist (Fig. 7C). The AUC of blood glucose in yohimbine-administered TRPM2-KO mice was significantly larger than that in the antagonist-administered wild-type mice (Fig. 7C). These results suggested that endogenous α2-adrenoceptor-operated sympathoadrenergic tones affect systemic blood glucose levels via TRPM2 channels to regulate glucose tolerance.

That catecholamines, such as adrenaline and noradrenaline, inhibit insulin secretion by activating α2-adrenoceptors in β-cells is well established (23–26). Catecholamines at pharmacological concentrations (µmol/L order) have been reported to exert their insulinostatic action via multiple mechanisms, including Gi/Go protein–mediated inhibition of cAMP production, activation of hyperpolarizing K⁺ currents, inhibition of Ca²⁺ currents, and inhibition of exocytotic machinery (1–8). However, signal transduction mechanisms mediating the physiological effects of adrenaline have not yet been identified in primary β-cells. Plasma concentrations of adrenaline were ∼0.2–0.4 nmol/L at rest (27,28) and reached to more than 1 nmol/L during exercise in healthy young men (27). The current study demonstrated that a low concentration of adrenaline (1 nmol/L) strongly attenuated glucose (16.6 mmol/L)- and Ex-4–elicited NSCCs current and membrane depolarization via the α2A-adrenoceptor in β-cells. In isolated islets, the low concentration of adrenaline inhibited the glucose-induced and Ex-4–potentiated insulin secretion in wild-type mice with IC₅₀ value of less than 1 nmol/L, but not in TRPM2-KO mice, indicating that physiological concentrations of adrenaline attenuate glucose-induced and GLP-1–potentiated insulin release by regulation of β-cell membrane excitability via α2A-adrenoceptor–operated TRPM2 channels. We further demonstrated that the blood glucose–increasing effects of endogenous catecholamines are mediated by the α2A-adrenoceptor–operated TRPM2 channels. The IPGTT study revealed that systemic administration of yohimbine, an α2-adrenoceptor antagonist, enhanced glucose tolerance in wild-type mice. In contrast, the α2-adrenoceptor antagonist failed to affect glucose tolerance in TRPM2-KO mice, indicating that endogenous α2-adrenoceptor–operated sympathoadrenergic tones affect systemic blood glucose via the TRPM2 channels.

Glucose metabolism–induced closure of K_ATP channels and membrane depolarization opened the voltage-dependent Ca²⁺ channels, triggering Ca²⁺ influx and insulin secretion (29,30). Incretin hormones, such as GLP-1 and glucose-dependent insulinotropic polypeptide, potentiate the glucose-induced insulin release by cytosolic cAMP productions via Gs-coupled receptors (23,31–33). We previously reported that the TRPM2 channel, a type of NSCC, plays a pivotal role in β-cells in glucose-induced and incretin-potentiated insulin secretion via cAMP signaling (9). The current study further uncovered a novel role of TRPM2 channels in insulinostatic adrenergic signaling in β-cells. In isolated islets, adrenaline at 1 nmol/L displayed strong inhibitory effects on cAMP production, which was similar in extent to that at 5 µmol/L (Fig. 3). In addition, low concentrations of adrenaline (1 nmol/L) markedly inhibited the glucose (16.6 mmol/L)- and GLP-1–induced activation of TRPM2 currents in β-cells (Fig. 2) but failed to affect the NSCCs current induced by a cAMP analog db-cAMP (1 mmol/L) (Fig. 4). These results suggested that physiological concentrations of adrenaline inhibit cAMP signaling to attenuate the glucose-induced and GLP-1-potentiated TRPM2 currents, which is consistent with our previous findings that ghrelin, a gastric hormone, attenuates TRPM2 activation via Gi-mediated inhibition of cAMP signaling (14), although it was reported that the major target of ghrelin in the islets is the δ-cell but not the β-cell (34,35). Whether other physiological insulinostatic hormones, including somatostatin and neuropeptide Y, also exert their inhibitory effects via suppression of TRPM2 currents remains to be clarified. In line with expected TRPM2 function to facilitate membrane depolarization, low concentrations of adrenaline significantly delayed the glucose-induced membrane depolarization with slight repolarization in β-cells (Fig. 5). Using a different approach with Ca²⁺ recordings in mouse islets, a similar prolongation by adrenaline of the lag time between glucose stimulation and the initial rise of cytosolic Ca²⁺ concentrations is reported (36). Furthermore, in the islets from TRPM2-KO mice, adrenaline (1 nmol/L) failed to attenuate glucose-induced and GLP-1–potentiated insulin release (Fig. 1B); in contrast, it enabled suppression of cAMP production in the KO islets as well as in wild-type islets (Fig. 3C). Adrenaline (1 nmol/L) did not affect NSCCs current and insulin secretion at low glucose in β-cells (Supplementary Fig. 1A and B). These results suggest that adrenaline physiologically suppresses insulin secretion at least partly by inhibiting the TRPM2 channels.

At pharmacological ranges of µmol/L order, adrenaline suppressed insulin secretion via distinct signaling pathways beyond the G-protein–coupled receptor–mediated cAMP/TRPM2 signaling. Adrenaline at 5 µmol/L but not 1 nmol/L attenuated insulin release in TRPM2-KO mice, and the potency of inhibition was identical to that in wild-type mice (Fig. 1). Adrenaline (5 μmol/L) suppressed the db-cAMP–induced NSCC-current increase in β-cells (Fig. 4C and D). Furthermore, 5 µmol/L adrenaline caused hyperpolarization of the plasma membrane depolarized by high glucose (Fig. 5F and Supplementary Fig. 4A), Ex-4 (Supplementary Fig. 4B), and tolbutamide (Fig. 6A and B) in wild-type β-cells, which is consistent with previous reports that adrenaline-induced hyperpolarization of mouse pancreatic islet cells is mediated by G protein–gated inwardly rectifying K⁺ channels (3). Debuyser et al. (20) reported concentration dependency of the effects of adrenaline on insulin secretion in isolated islets under various stimulatory conditions. Adrenaline inhibited glucose (15 mmol/L)-induced insulin release by ∼33% at 1 nmol/L and by more than 90% at 1 µmol/L. The inhibitory potency of adrenaline at 1 nmol/L was reduced by depolarizing agents, such as tolbutamide, arginine, and high K⁺, regardless of the mechanisms by which depolarization was produced. Under these strong depolarizing conditions exceeding the ability to control membrane excitability by the TRPM2 channel mediation, physiological concentrations of adrenaline might be difficult to attenuate insulin release via cAMP/TRPM2 signaling. High concentrations of adrenaline, however, inhibited cAMP-dependent and cAMP-independent insulin release nearly completely under any stimulatory conditions (6,20,22). It is necessary to clarify whether these distinct effects of adrenaline at different concentrations are mediated by a single α2-adrenoceptor coupled to multiple G-proteins with variable affinities for the agonists or by different receptor subtypes.

The effects of α-adrenoceptor blockade, especially α2-adrenoceptor antagonists, have been reported to decrease blood glucose levels in animals and in humans (37–39). Mice deficient in α2A-adrenoceptor exhibit increased plasma insulin levels, reduced blood glucose levels, and improved glucose tolerance (40,41). Conversely, an α2-adrenoceptor agonist increased the blood glucose levels and decreased insulin levels in wild-type mice but was without such effect in α2A-KO mice (40,41). These reports suggest that α2A-adrenoceptor–mediated tonic inhibition of insulin secretion contributes to blood glucose control in vivo. In the current study, we further demonstrated that administration of an α2-adrenoceptor antagonist, yohimbine, failed to affect glucose tolerance in TRPM2-KO mice, in contrast to a decrease in the blood glucose rise in wild-type mice receiving yohimbine (Fig. 7). These results suggest that endogenous α2-adrenoceptor–operated sympathoadrenergic tones worsen blood glucose control via TRPM2-mediated inhibition of insulin secretion. Further studies would be needed to clarify the physiological adrenergic effects via cAMP-dependent TRPM2 signaling in pancreatic β-cells.

In conclusion, the current study demonstrated that adrenaline physiologically attenuates glucose-induced and GLP-1–potentiated insulin release via α2A-adrenoceptor coupled to cAMP/TRPM2 signaling. From a clinical setting, an enhanced sympathetic activity is associated with chronic inflammation, including obesity (42,43), and increased α2A-adrenoceptor signaling may result in β-cell dysfunction and an increased risk of type 2 diabetes (44,45). Genetic variations in ADRA2A, a gene encoding the α2A-adrenoceptor, have been reported to be associated with type 2 diabetes (46,47), and insulin secretion defect in patients with a risk allele on ADRA2A expression was improved by administration of yohimbine (48). In transgenic mice with β-cell–specific overexpression of the α2A-adrenoceptor, basal blood glucose and insulin levels were normal, but glucose tolerance was deteriorated (49). Thus, elevated α2A-adorenoceptor activity could be causally implicated in type 2 diabetes, and blockade of the α2-adrenoceptor could improve glucose-stimulated insulin secretion. Hence, developing approaches to specifically intervene in α2A-adrenoceptor–cAMP/TRPM2 signaling in β-cells might provide a potential therapeutic tool to treat patients with type 2 diabetes.

Acknowledgments. The authors thank Harue Fukaya, Taeko Ohtani, and Chizuru Kobayashi at Jichi Medical University Saitama Medical Center for their technical assistance.

Funding. This work was supported by grants from the Japan Diabetes Foundation and Takeda Science Foundation to K.D. and by the Japan Society for the Promotion of Science Grant in Aid for Scientific Research (C) 15K09397 to K.D., 15K09396 to M.Kak., and Grant-in-Aid for Young Scientist (B) 26870532 and 16K19545 to M.Y.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. K.I. and K.D. designed the study, researched the data, and wrote, reviewed, and edited the manuscript. M.Y. designed the study, researched the data, contributed to the discussion, and reviewed the manuscript. H.Y. researched the data, contributed to the discussion, and reviewed the manuscript. R.M. and R.S.R. researched the data. S.O., K.T., M.Kaw., K.H., and Y.M. contributed to the discussion and reviewed the manuscript. T.Y. contributed to the discussion and reviewed and edited the manuscript. M.Kak. designed the study and wrote, reviewed, and edited the manuscript. K.I., K.D., and M.Kak. are the guarantors of this work, and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.