Inhibition of voltage-dependent potassium channels mediates cAMP-potentiated insulin secretion in rat pancreatic β cells

ABSTRACT

Insulin secretion is essential for maintenance of glucose homeostasis. An important intracellular signal regulating insulin secretion is cAMP. In this report, we showed that an increase of cAMP induced by adenylyl cyclase (AC) activator forskolin or by cAMP analog db-cAMP not only potentiated insulin secretion but also inhibited Kv channels, and these effects were reversed by AC inhibitor SQ22536. The cAMP-mediated Kv channel inhibition resulted in prolongation of action potential duration, which partly accounts for the elevation of intracellular Ca²⁺ induced by activation of cAMP signaling. Taken together, the results suggest that Kv channels are involved in cAMP-potentiated insulin secretion in pancreatic β cells.

Introduction

Cyclic AMP (cAMP), an ubiquitous second messenger, plays essential roles in a variety of fundamental cell functions ranging from transcriptional regulation and apoptosis, to cell growth and differentiation.^1-3 In pancreatic islets, cAMP is pivotal in regulating insulin secretion, the related mechanisms that have been proposed include mobilization of Ca²⁺ from intracellular stores,^4,5 and modulation of ion channel activities from ATP-sensitive K⁺ channels (K_ATP channels),⁶ the L-type voltage-dependent Ca²⁺ channels^7,8 and the non-selective cation channels.⁹ These actions results in elevation of intracellular Ca²⁺ concentration ([Ca²⁺]_i) and subsequently insulin release in the presence of stimulatory glucose concentrations.¹⁰ In addition, cAMP regulated-insulin secretion is also attributed to exocytotic process.^11,12

As electrically excitable cells, pancreatic β cells secrete insulin in a stimulus-secretion coupling manner. During this process, action potentials stimulated by secretagogues are the primary electrical signal which affecting insulin secretion.¹³ The generation of action potentials is influenced by a number of ion channels. Among these channels, K_ATP channels and voltage dependent Ca²⁺ channels play important roles in membrane depolarization, voltage dependent potassium channels (Kv channels) are largely responsible to membrane repolarization.^13,14

It has been established that inhibition of Kv channels potentiate insulin secretion via prolongation of action potential duration.^13,15 In the present study, we therefore sought to investigate whether Kv channels are involved in cAMP-regulated insulin secretion. By using forskolin, an activator of adenylyl cyclase (AC), and dibutyryl cAMP (db-cAMP), a cAMP analog, we found that either endogenous or exogenous enhancement of cAMP both profoundly inhibited Kv channels, which in turn prolonged action potential duration and increased [Ca²⁺]_i. The results indicate that the Kv channel is the downstream target of cAMP signaling and participate the cAMP-regulated insulin secretion.

Material and methods

Animal

Male SD rats (about 180∼250 g) which obtained from the Animal Facility Center of Shanxi Medical University were housed in a temperature and humidity controlled room on a 12 h light-dark cycle with free access to pellet-type food and tap water. We followed the principles of laboratory animal care and the study was approved by the Animal Care and Use Committee of the Shanxi Medical University.

Preparation and culture of pancreatic islets and islet cells

After SD male rat was sacrificed, the pancreas was quickly removed when it was inflated by injecting 10 ml medium containing 1mg/ml of collagenase P (Roche, Indianapolis.USA), then digested in a 37°C water bath for 11 min and density gradient centrifuged by histopaque-1077. Subsequently, islets were handpicked under a dissecting microscope to ensure high purity. DispaseII was applied to disperse pancreatic islets into single cells. Pancreatic islets or dispersed islet cells were cultured in Hyclone RPMI 1640 medium (Hyclone Beijing, China) supplemented with 10% fetal bovine serum, 0.004% β-mercaptoethanol, 100 U/ml penicillin, 100 mg/ml streptomycin and 11.1 glucose, at 37°C in 5% CO₂ humidified atmosphere.

Insulin secretion assay

Islets from multiple rats were pooled and allowed to recover from the isolation procedure overnight in medium at 37°C in an atmosphere of humidified air (95%) and CO₂ (5%). Handpicked islets (five islets per tube) were equilibrated for 30 min at 37°C in Krebs-Ringer bicarbonate-HEPES (KRBH) buffer contained (mM): 128.8 NaCl, 4.8 KCl, 2.5 CaCl₂, 1.2 KH₂PO₄, 1.2 MgSO₄, 5 NaHCO₃, 10 HEPES and 2% BSA at pH 7.4. Glucose concentration was 2.8 mM during the equilibration period. Next, islets were incubated for 30 min in KRBH with various drugs as indicated. At the end of incubation, a fraction of supernatant liquid was taken for insulin measurement, then 70% acid-ethanol (Ethanol/water/HCl (vol./vol.) = 150:47:3) was added to the remaining volume to measure the total insulin contents. Insulin concentration was measured by Iodine [¹²⁵I] Insulin Radioimmunoassay Kit (North biological technology research institute of Beijing).

Measurements of cAMP levels

For cAMP measurement, islets (10 islets per tube) were exposed to 8.3 mM glucose for 1 h at 37°C in the presence of the phosphodiesterase inhibitor 3-isobutyl-1-methylxanthine(IBMX) and 4-(3-butoxy-4-methoxybenzyl) imidazolidone (RO 20–1724) to prevent the cAMP degradation and relevant drugs were applied during the experiment. At the end of incubation, islets were lysed by ultrasonic cell smash instrument and cAMP levels were determined by EIA kit (Institute of isotopes CO.Ltd, HU).

Electrophysiology

The whole-cell Kv currents in islet cells were measured using an EPC-10 amplifier and PULSE software from HEKA Electronik (Lambrecht, GER) at room temperature (22-25°C). Before experiment, islet cells were cultured on glass coverslips for 24 h in RPMI 1640 medium and then transferred to a recording chamber installed on an inverted microscope (Nikon Diaphot-TMD). Patch pipettes were pulled from borosilicate glass capillaries with the resistance of 4–7 MΩ by a two-stage Narishige MODEL PP-830 micropipette puller (Narishige Co., Tokyo, Japan) and MICRO-FORGEMF-200 (World Precision Instruments Inc., USA). The cell capacitance which was greater than 7 pF was perceived as β cell.¹⁶

To record Kv currents, the intracellular solution was used as follows (mM): 140 KCl, 1 MgCl₂, 0.05 EGTA, 10 NaCl and 10 HEPES (pH 7.4 with KOH). The extracellular solution contained (mM): 141.9 NaCl, 5.6 KCl, 1.2 MgCl₂, 11.1 glucose and 5 HEPES (pH 7.4 with NaOH). Kv currents were elicited using voltage steps from a holding potential (−70 mV) to +80 mV in 10 mV steps with pulses of 400 ms duration.

The current-clamp mode was applied to record action potentials, β cells were stimulated by a 4 ms, 150 pA current injection. The intracellular solution was used as follows (mM): 140 KCl, 1 MgCl₂, 0.05 EGTA, 10 NaCl and 10 HEPES (pH 7.4 with KOH). The extracellular solution contained (mM): 138 NaCl, 5.6 KCl, 2.6 CaCl₂, 1.2 MgCl₂, 11.1 glucose and 5 HEPES (pH 7.4 with NaOH). The time that initiation of action potential until the membrane potential returned to within 10 mv of the resting membrane potential was considered as action potential duration.

Calcium imaging

For cytosolic calcium measurements, islet cells were cultured on coverslips coated by cell adherent reagent. Calcium changes were monitored in pancreatic islet cells preincubated for 30 min at 37°C with 2 μM Fura2-AM (Dojindo Laboratories, Japan) in KRBH contained 2.8 mM glucose and then washed two times before the experiments. Ratiometric measurements of Fura2-AM were performed in OLYMPUS IX71 inverted microscope with filter set at 510 nm for emission and excitation at 340 and 380 nm. Islet cells were constantly perfused with Krebs-Ringer bicarbonate-HEPES buffer (KRBH: 128.8 mM NaCl, 4.8 mM KCl, 2.5 mM CaCl₂, 1.2 mM KH₂PO₄, 1.2 mM MgSO₄, 5 mM NaHCO₃, 10 mM HEPES, pH 7.4) and reagents as indicated. All operations were performed at 33°C. The change of intracellular calcium (ΔF340/F380 _(F-F0)) was determined by subtracting the initial ratio (F₀) from the maximum F340/F380 for each treatment (F).

Statistical analysis

Date is presented as mean ± SEM. Statistical significance was assessed with Student´s t-test or one-way analysis of variance (ANOVA). Difference were considered statistically significant when p < 0.05.

Results

Activation of AC/cAMP signaling amplifies insulin secretion

To examine the effect of cAMP signaling on insulin secretion, forskolin was applied in the static secretion experiments of rat pancreatic islets. As shown in Fig. 1A, 8.3 mM glucose significantly stimulated insulin secretion compared to 2.8 mM glucose, and forskolin (10 µM) further potentiated insulin secretion in the presence of 8.3 mM glucose. SQ22536, an AC inhibitor, attenuated the forskolin-induced elevation of insulin secretion, while SQ22536 alone did not influence basal secretion. To examine the involvement of intracellular cAMP in the forskolin-potentiated insulin secretion, we measured intracellular cAMP levels in rat pancreatic islets. As shown in Fig. 1B, the exposure of forskolin (10 µM) significantly increased intracellular cAMP levels, and this effect was reduced by SQ22536. We then employed Dibutyryl-cAMP (db-cAMP), an exogenous membrane-permeable cAMP analog, to further confirm the effect of cAMP signaling on insulin secretion. As expected, db-cAMP also potentiated glucose-stimulated insulin secretion (Fig. 1C).

Figure 1.

Activation of AC/cAMP signaling amplifies insulin secretion. (A) Effects of SQ22536 (SQ, 10 uM) and forskolin (FSK, 10 uM) on insulin secretion at 8.3 mM glucose (8.3G) condition. (B) Effects of SQ22536 (SQ, 10 uM) and forskolin (FSK, 10 uM) on intracellular cAMP at 8.3 mM glucose condition. (C) Effects of db-cAMP (1 mM) on insulin secretion at 8.3 mM glucose condition. Each experiment was repeated three times. *P < 0.05, **P <0.01, ***P <0.001.

We would have liked to test the effects of higher concentrations of SQ22536 in order to determine whether more complete adenylyl cyclase inhibition would cause greater inhibition of forskolin activation. However, this would have required exposing the cells to a larger volume of solvent (dimethyl sulfoxide). We and others have found that higher concentrations of dimethyl sulfoxide independently increase intracellular calcium concentration in pancreatic β cells (not shown) and in other cell types.¹⁷ This might influence the appropriate interpretation of the result directly caused by SQ22536.

Kv channels are inhibited by the AC/cAMP signaling pathway

To determine if cAMP signaling influence Kv channel activities in rat pancreatic β cells, we examined the Kv channel currents in response to forskolin. The result demonstrated that sustained Kv currents were potently suppressed by forskolin compared to control. SQ22536 did not cause changes of Kv currents compared to control, but reversed the effect of forskolin on Kv channels (Fig. 2A and B). To confirm the effect of cAMP signaling on Kv channels, db-cAMP was used to observe the Kv channel activity in response to exogenous cAMP. In addition, membrane-impermeable cAMP was also directly dialysed into β cell through the patch pipette to examine the direct effect of cAMP on Kv channels. As shown (Fig. 3A and B), addition of either db-cAMP or cAMP significantly inhibited Kv channels in β cells.

Figure 2.

Kv channels are inhibited by AC/cAMP signaling pathway in β cells. (A) Representative current traces recorded under the different treatments as indicated. (B) Current-voltage curves of Kv channels (left panel) and summary of the mean current density of Kv at +80 mV (right panel) under the different treatments as indicated. Control (Ctrl); SQ22536 (SQ, 10 uM); forskolin (FSK, 10 uM). Each experiment was repeated three times. *P < 0.05, **P < 0.01.

Figure 3.

cAMP inhibited Kv channels in β cells. Current-voltage curves of Kv channels (left panel) and summary of the mean current density of Kv at +80 mV (right panel) under the different treatments. (A) β cells were treated with membrane-permeable db-cAMP (1 mM) in the extracellular solution. (B) β cells were treated with membrane-impermeable cAMP (1 uM) in the intracellular solution. Each experiment was repeated three times. **P < 0.01, ***P < 0.001.

Activation of AC/cAMP signaling pathway prolongs action potential duration

We then investigated the effect of forskolin on action potentials to see how forskolin modulates β cell excitability. The action potentials were elicited by current injections under current-clamp recording mode (Fig. 4A–D). Compared with control, application of forskolin remarkably lengthened the action potential duration. SQ22536 did not influence action potentials compared to control, but attenuated the effect of forskolin on action potential duration (Fig. 4E).

Figure 4.

Activation of AC/cAMP signaling pathway prolongs action potential duration in β cells. (A-D) Representative action potential waveforms recorded under the different treatments as indicated. Action potential was elicited with a 4 ms, 150 pA current injection. (E) Summary of the mean action potential durations. Control (Ctrl); SQ22536 (SQ, 10 uM); forskolin (FSK, 10 uM); Each experiment was repeated three times. *P < 0.05, **P <0.01.

Kv channels are involved in the cAMP signaling-regulated [Ca²⁺]_i level

Using Ca²⁺-sensitive fluorescent dye, Fura-2 AM, the effect of forskolin on [Ca²⁺]_i level was investigated in rat β cells. As shown in Fig. 5A and B, addition of forskolin (10 uM) raised [Ca²⁺]_i level at 8.3 mM glucose condition. Similarly, addition of Kv channel inhibitor tetraethylammonium chloride (TEA) at the same condition also elevated [Ca²⁺]_i level. While in the presence of TEA, forskolin only slightly further elevated [Ca²⁺]_i level (Fig. 5B and D).

Figure 5.

Kv channels are involved in the cAMP signaling-regulated [Ca²⁺]_i level in β cells. The level of [Ca²⁺]_i were plotted by the ratio of F340/F380. (A) Cells were perfused with forskolin (FSK, 10 uM) at 8.3 mM glucose condition. (B) Cells were perfused with 20 mM TEA in the absence or presence of forskolin (FSK, 10 uM) at 8.3 mM glucose condition. (C, D) The mean value of ΔF340/F380 (F-F₀) in response to different treatments as indicated (F: The maximum F340/F380 ratio for each treatment; F₀: The initial F340/F380 ratio for 2.8G). Each experiment was repeated five times. *P < 0.05, **P <0.01, ***P <0.001.

Discussion

In pancreatic β cells, Ca²⁺ is believed to be the primary triggering signal for insulin secretion. In addition, cAMP is also a pivotal component in regulating insulin secretion. Several Gs-coupled receptor agonists, including glucagon-like peptide-1 (GLP-1), gastric inhibitory polypeptide and glucagon, are known to enhance glucose-stimulated insulin secretion through activating cAMP signaling pathway in β cells.¹⁸ On the contrary, Gi-coupled agonists like noradrenaline, somatostatin, ghrelin suppress insulin secretion by reducing cAMP.^19,20 Indeed, in the present study we showed that the increase of endogenous cAMP by forskolin or addition of exogenous cAMP and db-cAMP, all potentiated glucose-stimulated insulin secretion.

Studies have shown that cAMP-regulated insulin secretion exhibited glucose-dependent characteristic. At low glucose concentration (1.0-2.8 mM), cAMP has little or no effect on insulin secretion.^21,22 However, cAMP combined with high glucose induces the potentiation of insulin secretion.^23,24 Although the underlying mechanism is not fully understood, it has been attributed, at least in part, to the interaction between cAMP and Ca²⁺ signal, because low glucose concentration do not induce a Ca²⁺ rise in β cells.^21,22

Here, our study suggested that Kv channels are also involved in the cAMP-regulated insulin secretion. We found that forskolin inhibited Kv channels, which could be reversed by SQ22536, indicating that activation of AC/cAMP pathway has inhibitory effects on Kv channels. In support of this, the suppression on Kv channels has also been observed when β cells were either exposed to cell permeable db-cAMP or dialysed with cell impermeable cAMP. It is now well established that Kv channels participate in the repolarization phase of the action potential. Genetic ablation or pharmacological inhibition of K_V channels results in prolongation of action potential duration, and in turn elevates [Ca²⁺]_i and insulin secretion.^14,15,25 This notion is consistent with our data that forskolin prolonged action potential duration compared to control. The effect of forskolin on action potential is mediated by AC/cAMP pathway because SQ22536 suppressed this effect. Additionally, our calcium imagining study revealed that forskolin, similar to Kv channel inhibitor TEA, significantly elevated [Ca²⁺]_i level, respectively. While in the condition of pretreatment with TEA, forskolin only slightly further elevated [Ca²⁺]_i level, indicating that TEA and forskolin share the major pathway in elevating [Ca²⁺]_i level. Thus, these results collectively imply that Kv channels are involved in the process of forskolin-regulated [Ca²⁺]_i, which appears due to the increase of Ca²⁺ entry by inhibiting Kv channel-induced repolarization through cAMP signaling pathway.

Of note, inhibition of Kv channels has been reported to results in potentiation of insulin secretion in a glucose-dependent manner, because physiologically, the Kv channel is only open in response to membrane depolarization at high glucose condition.²⁶ Considering the similar regulatory effects of cAMP and Kv channels on insulin secretion, we speculate that the Kv channels play a pivotal role in shaping the glucose-dependent effect of cAMP on insulin secretion. This notion is also supported by the studies related to GLP-1 receptor agonists which potentiate insulin secretion in an glucose-dependent manner through cAMP pathway. Macdonald et. al. revealed that antagonism of Kv channels mediates the cAMP-dependent insulinotropic effect of exendin-4, which is a widely used GLP-1 receptor antagonist.²⁷ Our recent study show that Kv channel inhibition is also the cAMP downstream effect induced by geniposide, which has been identified as a GLP-1 receptor agonist and influences insulin secretion in a glucose-dependent manner.²⁸

Previous studies have shown that activation of Ca²⁺ channels and mobilization of Ca²⁺ from intracellular stores are responsible for cAMP-regulated cellular functions.^8,29,30 In the present study, we provide new evidence showing that the ability of cAMP signaling to enhance [Ca²⁺]_i and insulin secretion is also due to the inhibition of Kv channels, suggesting that targeting cAMP/Kv channel pathway may be a therapeutic strategy for diabetes treatments. Especially, considering the insulinotropic effect is glucose dependent, the potential agents developed based on this strategy might possess the promising antidiabetic effects with less risk of inducing hypoglycemia, which is a complication often encountered with the currently used antidiabetic sulfonylures.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Funding

This work was supported by NSFC (81373464, 81670710, 81273564), Advanced Programs of Shanxi for the Returned Overseas Chinese Scholars (2016–97), Natural Science Foundation of Shanxi Province (201401143, 2013011047-3) and NSFC (81270882).