Chromanol 293B, an inhibitor of KCNQ1 channels, enhances glucose-stimulated insulin secretion and increases glucagon-like peptide-1 level in mice

Abstract

Glucose-stimulated insulin secretion (GSIS) is a highly regulated process involving complex interaction of multiple factors. Potassium voltage-gated channel subfamily KQT member 1 (KCNQ1) is a susceptibility gene for type 2 diabetes (T2D) and the risk alleles of the KCNQ1 gene appear to be associated with impaired insulin secretion. The role of KCNQ1 channel in insulin secretion has been explored by previous work in clonal pancreatic β-cells but has yet to be investigated in the context of primary islets as well as intact animals. Genetic studies suggest that altered incretin glucagon-like peptide-1 (GLP-1) secretion might be a potential link between KCNQ1 variants and impaired insulin secretion, but this hypothesis has not been verified so far. In the current study, we examined KCNQ1 expression in pancreas and intestine from normal mice and then investigated the effects of chromanol 293B, a KCNQ1 channel inhibitor, on insulin secretion in vitro and in vivo. By double-immunofluorescence staining, KCNQ1 was detected in insulin-positive β-cells and GLP-1-positive L-cells. Administration of chromanol 293B enhanced GSIS in cultured islets and intact animals. Along with the potentiated insulin secretion during oral glucose tolerance tests (OGTT), plasma GLP-1 level after gastric glucose load was increased in 293B treated mice. These data not only provided new evidence for the participation of KCNQ1 in GSIS at the level of pancreatic islet and intact animal but also indicated the potential linking role of GLP-1 between KCNQ1 and insulin secretion.

Abbreviations

AUC

Area under the curve

DMSO

Dimethyl sulfoxide

GLP-1

Glucagon-like peptide-1

GSIS

Glucose-stimulated insulin secretion

GTT

Glucose tolerance test

GWAS

Genome wide association studies

ITT

Insulin tolerance test

IVGTT

Intravenous glucose tolerance tests

KCNQ1

Potassium voltage-gated channel subfamily KQT member 1

KRBH

Krebs-Ringer bicarbonate HEPES buffer

OCT

Optimal Cutting Temperature Compound

OGTT

Oral glucose tolerance tests

SNPs

Single nucleotide polymorphisms

T2D

Type 2 diabetes

Introduction

Type 2 diabetes (T2D)，which is increasing in epidemic proportions worldwide, is a complex metabolic disorder characterized by chronic, progressive hyperglycemia as the result of impaired insulin secretion and insulin resistance.^1,2 Although its exact etiology is still unknown, accumulating evidence recognizes T2D as a quintessential multifactorial disease which results from the complex interaction between environmental and genetic factors.^3,4 In the last few years, several single nucleotide polymorphisms (SNPs) located in intron 15 of the KCNQ1 gene have been identified by genome wide association studies (GWAS) to be strongly associated with T2D.^5-8

KCNQ1 is a member of the KCNQ gene family that encodes for the pore-forming subunits of voltage-dependent potassium (K⁺) channels.⁹ KCNQ1 channel is distributed widely in epithelial and non-epithelial tissues throughout the body and regulates many key physiological functions such as maintaining the membrane potential as well as controlling the water and salt homeostasis.^10,11 Mutations of KCNQ1 have been confirmed to be responsible in most cases for the long QT syndrome, a genetic disorder characterized by increases in the cardiac action potential duration, therefore KCNQ1 was well known as KvLQT1 (Kv7.1).^9,12

In rodent pancreas, KCNQ1 has been demonstrated to be expressed in cells of islets of Langerhans.¹³ In insulinoma cell lines, the blockade of the KCNQ1 channel promoted insulin secretion while the increased expression of KCNQ1 impaired insulin secretion.^14,15 All these data suggest a direct regulatory role of KCNQ1 in β-cell function. Yet insulin secretion is a multi-tiered process involving the integration of many factors, and KCNQ1 is widely expressed in a number of tissues throughout the body, it is worthful to investigate the effect of KCNQ1 inhibition on insulin secretion at the level of pancreatic islet and intact animal. Moreover, the variation in KCNQ1 was found to be associated with insulin secretion during oral glucose tolerance test (OGTT) but not intravenous glucose tolerance tests (IVGTT), implying the involvement of indirect mechanisms.^5,16 The glucose-stimulated incretin glucagon-like peptide 1 (GLP-1) was assumed to be a candidate linker between KCNQ1 and β-cells function,¹⁶ but this hypothesis has not been verified so far. A study in gene-targeted mice demonstrated that the KCNQ1 knockout results in enhanced insulin sensitivity and in turn inhibits insulin release,¹⁷ adding more attractiveness over the mechanism by which KCNQ1 exert its effects on β-cells in vivo.

In the current study, we first examined KCNQ1 expression in mice pancreas and intestine. As KCNQ1 channel is blocked by chromanol 293B,^14,15,18,19 we then investigated the effects of chromanol 293B on insulin secretion in vitro and in vivo. Our work showed that KCNQ1 was expressed in insulin-positive β-cells in pancreatic islets and GLP-1-positive L-cells in intestinal crypt. The KCNQ1 inhibitor chromanol 293B enhanced glucose induced but not basal insulin secretion. Blood concentration of bioactive GLP-1 during OGTT was higher in 293B treated mice. These results not only provide new evidence for the participation of KCNQ1 in glucose induced insulin secretion (GSIS) at the level of pancreatic islet and intact animal but also indicate the potential linking role of GLP-1 between KCNQ1 and insulin secretion.

Results

Expression of KCNQ1 in pancreas and intestine

As shown in Fig. 1A-F, in pancreas, KCNQ1 antibodies not only strongly labeled acinar cells in a similar pattern as described before²⁰ but also yielded good staining in insulin-positive cells, indicating the expression of KCNQ1 protein in pancreatic β-cells. In intestine (Fig. 1G-L), consistent with previous work,^21,22 KCNQ1 immunoreactivity was observed in basolateral membranes of crypt epithelial cells. Although we did not find colocalization of KCNQ1 and GLP-1 in intestinal villi, KCNQ1 was detected in the GLP-1 positive cell in intestinal crypt.

Figure 1.

Expression of KCNQ1 in pancreas and intestine. (A-F): Mouse pancreas sections were subjected to double-immunofluorescent staining for the insulin (green) and KCNQ1 (red). The merged image (C) from the same section (A and B) shows that KCNQ1 colocalizes with insulin in pancreatic islet. The absence of primary KCNQ1 antibody was used as negative controls (E). The scale bar in A applies also for B-F. (G-O): Mouse intestine sections were subjected to double-immunofluorescent staining for the GLP-1 (green) and KCNQ1 (red). The merged images from the same sections show that KCNQ1 colocalizes with GLP-1 in intestinal crypt (I, merged image of G and H) but does not colocalize with GLP-1 in intestinal villi (L, merged image of J and K). The absence of primary KCNQ1 antibody was used as negative controls (N). Arrowhead (G-I) indicates cell that is both GLP-1 positive and KCNQ1 positive in intestinal crypt. Arrow (J-L) indicates cell that is GLP-1 positive but KCNQ1 negative in intestinal villi. The merged images of these cells are depicted magnified in the corresponding inserts. The scale bar in G also applies for H-O.

Chromanol 293B promotes insulin secretion from primary islets

Chromanol 293B at 100 μM has been shown to enhance insulin secretion in insulinoma cell lines,^14,15 but the effect of this pharmacological tool on insulin secretion in primary islets has yet to be determined. As demonstrated in Fig. 2, glucose-stimulated insulin secretion (GSIS) elicited by 16.7 mM glucose was significantly enhanced in the 293B (100 μM) islets compared with vehicle (DMSO) islets, suggesting a direct role for KCNQ1 in the regulation of GSIS in primary islets. However, basal insulin secretion at 3.3 mM glucose was not altered by 293B application.

Figure 2.

The effect of chromanol 293B on insulin secretion in primary islets. Insulin secretion was assessed in 1-h static incubations. Batches of 10 islets each were incubated for 1 h in either 3.3 or 16.7 mM glucose in the presence of chromanol 293B (100 μM) or only vehicle (DMSO) . The amount of secreted insulin was expressed as a percentage of the total insulin content. Data are mean ± SE (n = 8 for each group). *P < 0.05 for 293B vs DMSO mice.

Chromanol 293B enhances glucose tolerance and glucose-stimulated insulin secretion in mice

Given its stimulatory effects on GSIS in cultured islets, we then assessed the in vivo effects of 293B on glucose metabolism and insulin secretion during intraperitoneal glucose tolerance test (IPGTT) and oral glucose tolerance test (OGTT).

As shown in Fig. 3 A-B, blood glucose levels before and 30 min after the 293B or vehicle (DMSO) administration are all comparable, indicating that, without glucose challenge, the glucose metabolism was not influenced by the administration of 293B. However, after glucose loading, mice treated with 293B exhibited apparently lower blood glucose levels both in OGTT and IPGTT. IPGTT showed significant difference in 60 min time points, while in OGTT, the difference was even greater at 30–120 min after glucose challenge. The area under the curve (AUC) _0–120 min of blood glucose during OGTT was also significantly lower in 293B-treated mice than that in controls.

Figure 3.

The effect of chromanol 293B on glucose tolerance and glucose-stimulated insulin secretion in normal mice. Intraperitoneal (A, C) and oral (B, D) glucose tolerance tests (IPGTT and OGTT, respectively) were performed in mice infused with DMSO vehicle or chromanol 293B. After a 30-minute equilibration period, a 20% D-glucose solution (10 μl/g BW) was administered at time 0 min. Each point represents the mean of 8 mice. The vertical bars are SEs. Insets depict quantification of the area under the curve for blood glucose (A, B) or insulin level (C, D). *P < 0.05, **P < 0.01 for 293B vs DMSO mice.

In parallel with enhanced glucose tolerance, insulin responses to glucose at the early phase were higher significantly in 293B-treated mice (Fig. 3C-D) yet the AUC_0–120 min of their plasma insulin was only slightly increased as compared to vehicle controls. It is noteworthy that, administration of 293B lead to an even stronger plasma insulin response in OGTT than that in IPGTT, although the blood glucose concentration in OGTT are not higher than that in IPGTT. This suggests that the blood glucose is not the only determinator of the insulin response to gastric glucose challenge in 293B mice.

Chromanol 293B does not alter insulin sensitivity in mice

Considering the complex interplay among factors that influence glucose homeostasis, whole-body insulin sensitivity was also assessed by insulin tolerance test (ITT) in mice. As illustrated in Fig. 4, the blood-glucose levels dropped to similar extents in DMSO and 293B mice after insulin injection (0.75 U insulin/kg), suggesting that 293B mice retained normal insulin sensitivity and that enhanced glucose tolerance and insulin response in 293B mice was not due to peripheral insulin resistance.

Figure 4.

The effect of chromanol 293B on insulin sensitivity in normal mice. Insulin tolerance tests (ITT) were performed in mice infused with DMSO vehicle or chromanol 293B. After a 30-minute equilibration period, regular human insulin (Humulin, 0.75 U/kg body weight) was administered at time 0 min. Each point represents the mean of 12 mice. The vertical bars are SEs.

Chromanol 293B augments plasma GLP-1 level after gastric glucose challenge in mice

GLP-1 is released by the gut during a meal and is the major hormonal mediator regulating postprandial insulin release. To explore the mechanisms mediating the even pronounced insulin response during OGTT in 293B mice, we examined plasma bioactive GLP-1 level after the gastric glucose administration. It was found that at 15 min post-challenge plasma active GLP-1 concentration was markedly higher in 293B-treated mice vs vehicle mice (Fig. 5), suggesting that the augmenting effect of 293B on insulin secretion during OGTT is at least partly mediated by GLP-1.

Figure 5.

The effect of chromanol 293B on plasma GLP-1 level after gastric glucose challenge in normal mice. Oral glucose tolerance tests (OGTT) were performed in mice infused with DMSO or 293B. Levels of active GLP-1 in plasma taken 15 min after oral glucose challenge were analyzed. Data are mean ± SE (n = 8 for each group). *P < 0.05 for 293B vs DMSO mice.

Discussion

Nutrient-induced insulin secretion relies on electrical spiking activity of the β-cells membrane which is regulated by a number of ionic and nonionic signaling pathways. In general, the β-cells adapt insulin secretion to prevailing blood glucose levels through glucose metabolism.²³ The closure of ATP-sensitive K⁺ channels (K_ATP channels) and the subsequent generation of action potentials via activation of voltage-dependent L-type Ca²⁺-channels are the key events linking elevated glucose metabolism to alterations of electrical activity and eventually the release of insulin granules. Insulin secretion terminates when the β-cell is repolarized by the opening of potassium channels including members of the voltage- and calcium-activated potassium channel families.^24,25 Thus, the amount of insulin secreted is directly coupled to the electrical spiking activity of the β-cells, and each potassium channel involved in the repolarizing phase of spike plays a regulatory role in glucose-induced insulin release.

The voltage-gated KCNQ1 potassium channel is expressed in the heart and pancreas as well as in the kidney, liver, lung, and intestine and contributes to a K⁺ current which is responsible for maintaining the resting potential, shortening the action potential as well as controlling the water and salt homeostasis in several tissues.^19,20,26,27 By several GWA studies, KCNQ1 has been identified as a transethnic susceptibility gene for T2D.⁶ Genetic variation in KCNQ1 appears to be associated with impaired insulin secretion.^5,6,28 A very recent clinical study showed that patients with loss of function mutations in KCNQ1 exhibit increased insulin release and lower levels of plasma glucose upon oral glucose stimulation.²⁹ Thus, it seems plausible that enhanced KCNQ1 function is responsible for the development of T2D.^6,14,28 In support of this assumption, KCNQ1 inhibition has been reported to result in broadened action potential and enhanced insulin secretion in clonal pancreatic β-cells.^14,15 Yet the regulation of insulin secretion is a multi-tiered process, occurring at the level of not only the single β-cell, but also the pancreatic islet, the whole pancreas, and the intact organism.³⁰ It is necessary to study the effect of KCNQ1 inhibition on insulin secretion at the level of primary islets as well as intact animals.

Our present work shows that KCNQ1 was expressed in islet β-cells. On cultured islet, the application of chromanol 293B did not cause significant change on basal insulin secretion but enhanced glucose induced insulin secretion significantly, indicating the contribution of 293B sensitive target to the process of GSIS in pancreatic islets. Accordingly, 293B infusion enhanced glucose tolerance and insulin response to glucose challenge without changing basal glucose and insulin level in intact mice. Collectively, these results indicate that the administration of 293B augments insulin secretion in a glucose-dependent manner at the level of islets and intact animal. As 293B has been shown to be an effective pharmacological blocker of KCNQ1 channel,^18,31 these data provide more evidence for the idea that the risk allele of KCNQ1 may increase the expression or up-regulate the function of KCNQ1 and thereby promote the development of T2D.⁶

To generate an adequate secretory response, β-cells must receive multiple regulatory signals relaying information about changes in the internal and external environments.³² As KCNQ1 is expressed in a number of tissues, presumably incorporating signals generated from many tissues contributes to the diabetes susceptibility conferred by the risk allele of KCNQ1 for T2D. For the same reason, chromanol 293B applied in present work might influence more tissues than just β-cells. Signals from other tissues contain KCNQ1 might also contribute to the effect of 293B on insulin secretion.

Our data from intact mice shows the insulinotropic effect of 293B was greater during OGTT than IPGTT. Plasma levels of active GLP-1 after gastric glucose administration also higher in 293B vs DMSO mice. This should not be seen as a mere coincidence. GLP-1 is an incretin hormone stored and secreted from intestinal L-cells. GLP-1 augments insulin secretion in a glucose-dependent manner, and, together with glucose-dependent insulinotropic polypeptide (GIP), is responsible for up to 70% of the insulin response to food intake in healthy individuals³³ and thus orchestrating the bodies' response to the availability of newly absorbable nutrients.³⁴ A study in German population suggested that altered incretin secretion after food intake might be a potential link between KCNQ1 gene variants and impaired β-cells function.¹⁶ In line with their assumption, the KCNQ1 was then be found by RT-PCR to be the most highly expressed K⁺ channel gene-family member in L-cells.²¹ Here by double-immunofluorescence staining, KCNQ1 was found to be expressed in GLP-1-positive L-cells in intestinal crypt but not in intestinal villi. This result is partly agree with the previous report that chromanol sensitive K⁺ current in crypts was larger than villi.^21,22 The role of KCNQ1 in L-cells is still unclear. It is possible that the KCNQ1 channels influence glucose-stimulated incretin secretion directly by the modulation of L-cells’ electric activity. As the expression of KCNQ1 is relatively higher in crypt, it seems also reasonable that the KCNQ1 might influence incretin level indirectly by altering crypt function or L-cells number.²¹ Nevertheless, it should be noted that the changes of GLP-1 during OGTT we observed in 293B-mice might just be a part of scenario. The immunofluorescence in pancreas indicates that KCNQ1 stains not only insulin positive but also some insulin negative cells of islet. This implies that the insulin secretion might also be regulated by cross-talk between the β-cells and neighboring KCNQ1 positive cells. Further studies based on better design are needed to identify and describe other potential signals orchestrating the glucose induced insulin response under KCNQ1 inhibition.

In our present work, insulin sensitivity of normal mice was nearly not altered by 293B infusion. This result is in apparent contradiction to the previous report that insulin sensitivity was enhanced in KCNQ1-deficient mice.¹⁷ The chromanol 293B dosage used here is within the range utilized in previous studies to inhibit KCNQ1 effectively in vivo,^19,35 but the insufficient pharmacokinetic exposure of the animal to the inhibitor cannot be completely excluded. On the other hand, it also cannot be ruled out that some compensatory or consequential alterations might contribute to the altered insulin sensitivity upon global KCNQ1 knockout. Further experiments are necessary to untangle the complex relationships among KCNQ1 function and glucose metabolism in vivo.

The present study has some limitations. Although Chromanol 293B is a well known KCNQ1 channel blocker and has been utilized to study the KCNQ1 current in insulin secretion cells,^14,15 several previous publications have reported separately that chromanol 293B can cause inhibition of channels besides KCNQ1 such as potassium voltage-gated channel subfamily A member 4 (Kv1.4) and the cystic fibrosis transmembrane conductance regulator (CFTR).^36,37 As many ion channels involved in the process of insulin secretion, it's cannot be excluded completely that other target of chromanol 293 might also contributed to the enhanced insulin secretion. Due to the limited experimental conditions, we did not detect the electrical activity of β-cells and L-cells. Further studies are needed to provide more precise information about the contribution of KCNQ1 current to the function of these cells. However, as mentioned before, the insulin secretion is a complex process involving the integration and interaction of multiple signals arising from many tissues. Our studies performed in islet and in vivo not only provided new evidence for the participation of KCNQ1 in insulin response in the context of interaction network but also indicated the potential linking role of GLP-1 between KCNQ1 and β-cell function. Given the importance of GLP-1 in orchestrating insulin secretion, the underlying mechanisms linking the KCNQ1 activity and GLP-1 secretion should be investigated in future studies.

Materials and Methods

Animals

Male ICR mice (25 ± 3g) were obtained from Qinglongshan Animal Center (Nanjing, China) and housed under standard conditions with a 12 h light/dark cycle and free access to food and water. All animal procedures were approved by the University Committee for Laboratory Animals of Southeast University, China.

Immunohistochemistry

The mice were perfused transcardially with PBS followed by 4% paraformaldehyde in PBS. Specimens of pancreas and intestine were dissected and fixed in 4% paraformaldehyde for 24 h, cryoprotected in 30% sucrose and embedded in Optimal Cutting Temperature compound (OCT, Leica). Tissue cryosections (12 μm) were then permeabilised with 0.05% Tween, blocked with 3% BSA for 1 h, and incubated overnight at 4 ℃ with anti-KCNQ1 (SC-10646; Santa Cruz) and anti-Insulin (SC-9168; Santa Cruz) or anti-GLP-1 (ab22625; Abcam), followed a one-hour incubation with Alexa 488-conjugated chicken anti-rabbit antibody (A-21441; Molecular Probes) and Cy3-conjugated donkey anti-goat antibody (ab6949; Abcam). Tissue samples stained with secondary antibody alone served as controls. Images of tissue sections were captured using an Olympus BX53 fluorescence microscope equipped with a DP72 digital camera and Image-Pro Express software.

Islet isolation and Glucose-stimulated insulin secretion (GSIS) assay

Pancreatic islets were isolated from overnight fasted mice by collagenase digestion and purified by manual selection picking under a dissecting microscope as described previously.³⁸ Isolated islets were cultured overnight in RPMI 1640 medium with 11.1 mM glucose supplemented with 10% FBS, 200 units/ml penicillin, and 200 mg/ml streptomycin at 37°C in an atmosphere of 5% CO₂. The next day, groups of 10 similarly-sized islets were washed and preincubated in Krebs Ringer bicarbonate HEPES (KRBH) buffer with 3.3 mM glucose for 30 min at 37°C. Islets were then incubated for another 1 hr in fresh KRBH buffer containing 3.3 or 16.7 mM glucose in the absence of chromanol 293B (100 μM; stock solution was prepared in dimethyl sulfoxide (DMSO)) or only DMSO vehicle. Thereafter, the supernatants were collected and islets were lysed overnight in a 2% acid: 80% ethanol solution for insulin content. Samples were stored at −20°C and assayed for insulin by using the Rat/Mouse Insulin ELISA Kit from Millipore (catalog #EZRMI-13K) according to the manufacturer's protocol. Results are expressed as insulin released as a percentage of total insulin content.³⁹

Administration of Chromanol 293B in vivo

After an overnight (for glucose tolerance test (GTT)) or a 4-hour (for ITT) fast, mice were weighed and then anaesthetized with a combination of pentobarbital sodium + ketamine + xylazine (40 mg/kg + 20 mg/kg + 10 mg/kg, respectively, i.p.). The body temperature maintained at 38°C with a thermostatic heating pad. As described previously,¹⁹ Chromanol 293B (Sigma; a DMSO stock added to 0.9% NaCl) were infused at 0.5 ml/h intravenously via a polyethylene cannula placed in the left jugular vein with an initial bolus of 8 mg/kg/h followed by a 4 mg/kg/h maintenance dose. An equivalent amount of DMSO vehicle was administered to control mice in the same way. Once the venous cannula was implanted, blood samples were collected from the tail vein and taken for the measurement of basal plasma glucose and insulin levels. Glucose tolerance tests or insulin tolerance tests were drawn after a 30-minute equilibration period.

Glucose tolerance tests and insulin release

For the GTT, mice were given 2 mg D-glucose/g body wt through a gavage tube which was inserted in the stomach (for OGTT) or via injection into the peritoneal cavity (for IPGTT). Controls were given saline. The volume load was 10 μl/g body weight. Blood samples were collected from the tail vein just prior to glucose administration (0 min) and at different time points (5, 10, 30, 60, and 120 min) after glucose load. Blood glucose levels were determined by OneTouch® UltraVue™ glucometer (Johnson & Johnson KK). Plasma insulin was measured as mentioned above.

For analysis of active GLP-1, as the volume of plasma required for insulin and GLP-1 measurements precluded sampling at multiple time points, a single blood sample was collected 15 min after an oral glucose challenge.⁴⁰ Collected blood was rapidly mixed with a dipeptidyl peptidase 4 inhibitor (Millipore; DPP4) and plasma GLP-1 concentration was determined by ELISA kit (EGLP-35K; Millipore) according to the manufacturer's instructions.

Insulin tolerance tests

Because glucose tolerance and plasma insulin levels might be altered due to peripheral insulin resistance, we then assessed insulin sensitivity in 293B or DMSO treated mice by ITT. Mice were given an intraperitoneal injection of regular human insulin (Humulin, 0.75 U/kg body weight) and blood glucose concentrations were monitored just prior to (0 min) and at 15, 30, 45, 60, 90, and 120 min after insulin injection.

Statistics

Data are expressed as mean ± standard error (SE). The level of statistical significance was determined using Student's 2-tailed t-test or 2-way ANOVA as appropriate, and significance assumed at the 5% level.

Disclosure of Potential Conflict of Interest

No potential conflicts of interest were disclosed.

Funding

This work was supported by the National Natural Science Foundation of China (81100548), the Technology Fund from Southeast University (9224000009), and the Analyze-Test Fund from Southeast University (2012-1160739030).