Telmisartan Potentiates Insulin Secretion via Ion Channels, Independent of the AT1 Receptor and PPARγ

Abstract

Angiotensin II type 1 (AT1) receptor blockers (ARBs), as antihypertensive drugs, have drawn attention for their benefits to individuals with diabetes and prediabetes. However, the direct effects of ARBs on insulin secretion remain unclear. In this study, we aimed to investigate the insulinotropic effect of ARBs and the underlying electrophysiological mechanism. We found that only telmisartan among the three ARBs (telmisartan, valsartan, and irbesartan) exhibited an insulin secretagogue role in rat islets. Independent of AT1 receptor and peroxisome proliferator-activated receptor γ (PPARγ), telmisartan exerted effects on ion channels including voltage-dependent potassium (Kv) channels and L-type voltage-gated calcium channels (VGCCs) to promote extracellular Ca²⁺ influx, thereby potentiating insulin secretion in a glucose-dependent manner. Furthermore, we identified that telmisartan directly inhibited Kv2.1 channel on a Chinese hamster ovary cell line with Kv2.1 channel overexpression. Acute exposure of db/db mice to a telmisartan dose equivalent to therapeutic doses in humans resulted in lower blood glucose and increased plasma insulin concentration in OGTT. We further observed the telmisartan-induced insulinotropic and electrophysiological effects on pathological pancreatic islets or β-cells isolated from db/db mice. Collectively, our results establish an important insulinotropic function of telmisartan distinct from other ARBs in the treatment of diabetes.

Keywords: AT1 receptor; Kv channel; L-type VGCC; insulin secretion; telmisartan.

FIGURE 1

Only telmisartan among the three ARBs potentiated insulin secretion in rat islets (n = 7 tubes of islet). (A) Rat islets were treated with various doses (1, 10, and 50 μM) of telmisartan under 2.8 and 8.3 mM glucose (denoted as 2.8 G and 8.3 G) conditions. (B) Islets were treated with 10 μM telmisartan under different glucose concentrations (2.8, 11.1, and 16.7 mM). (C,D) Rat islets were treated with various doses (1, 10, and 50 μM) of valsartan or irbesartan under 2.8 and 16.7 mM glucose conditions. The islets compared between the groups in each graph are collected from the same animal. All results are normalized to basal secretion at 2.8 G, and reported as the means ± SEM. Statistical differences among three or more groups were compared using one-way analysis of variance (ANOVA) and Student–Newman–Keuls method post hoc analysis. Statistical differences between two groups (with or without telmisartan) under the same glucose condition in (B) were determined using an unpaired two-tailed Student’s t test. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 2

Only telmisartan among the three ARBs increased intracellular [Ca²⁺] ([Ca²⁺]_i) concentration in rat pancreatic β-cells. (A) The trace shows the changes of [Ca²⁺]_i concentration in β-cells treated with 1, 10, and 50 μM telmisartan (Tel) under 2.8 mM glucose conditions; 500 μM tolbutamide (Tolb) was used as a positive control. (B) The average value during 30 s F340/F380 spikes for each test in response to different doses of telmisartan under 2.8 mM glucose conditions as indicated (n = 9 cells). (C,D) The trace shows the changes of [Ca²⁺]_i concentration in β-cells treated with different doses of telmisartan under 11.1 mM glucose conditions, and the average value during 30 s F340/F380 spikes for each test as indicated (n = 9 cells). (E,F) The trace shows the changes of [Ca²⁺]_i concentration in β-cells treated with different doses of telmisartan under 16.7 mM glucose conditions, and the average value during 30 s F340/F380 spikes for each test as indicated (n = 9 cells). (G–J) The trace shows the changes of [Ca²⁺]i concentration in β-cells treated with 1, 10, and 50 μM of valsartan (Val) or irbesartan (Irb) under 16.7 mM glucose conditions respectively, and the average value during 30 s F340/F380 spikes for each test as indicated. KCl (60 mM) was used as a positive control (n = 9 cells). The cells compared between the groups in each graph are isolated from the same animal. All results are reported as the means ± SEM. Statistical differences among three or more groups were compared using one-way ANOVA, and followed by Student–Newman–Keuls Method post hoc analysis in (D,F,J), or Tukey post hoc analysis in (B,H). *p < 0.05, ***p < 0.001.

FIGURE 3

PPARγ does not participate in the pathway of telmisartan-induced insulin secretion and elevation of [Ca²⁺]_i levels (A) Telmisartan (10 μM) potentiated glucose-stimulated insulin secretion in the presence or absence of the PPARγ inhibitor GW9662 (10 μM) (n = 7 tubes of islet). All insulin secretion results are normalized to basal secretion at 2.8 Mm glucose concentration. (B) The trace shows the changes of [Ca²⁺]_i concentration in β-cells treated with GW9662 (10 μM) alone or in combination with telmisartan (Tel 10 μM) under 16.7 mM glucose conditions. (C) The average value during 30 s F340/F380 spikes for each test in response to GW9662 alone or in combination with telmisartan under 16.7 mM glucose conditions as indicated (n = 9 cells). The islets or cells compared between the groups in each graph are collected from the same animal. All results are reported as the means ± SEM. In (A), statistical differences between two groups (with or without telmisartan) were compared using an unpaired two-tailed Student’s t test, and difference among three groups without telmisartan were compared using one-way ANOVA and Student-Newman-Keuls method post hoc analysis. In (C), difference among three groups was determined by one-way ANOVA and Tukey Test post hoc analysis. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 4

Telmisartan enhances [Ca²⁺]_i levels through extracellular Ca²⁺ influx, rather than intracellular Ca²⁺ stores release. (A) The trace shows the changes of [Ca²⁺]_i concentration in β-cells treated with telmisartan (Tel, 10 μM) under 16.7 mM glucose conditions in Ca²⁺-free KRBH medium. (B) The average value of F340/F380 during each test in response to telmisartan (Tel, 10 μM) in Ca²⁺-free KRB medium (n = 23 cells). (C) The trace shows the changes of [Ca²⁺]_i concentration in β-cells treated with telmisartan (Tel 10 μM) under 16.7 mM glucose conditions with addition of the L-type VGCC blocker azelnidipine (Az, 0.1 μM). (D) The mean value of F340/F380 during each test in response to telmisartan (10 μM) with added azelnidipine (0.1 μM). Thapsigargin (Thap, 0.1 μM) was used as a positive control (n = 12 cells). The cells compared between the groups in each graph are isolated from the same animal. All results are reported as the means ± SEM. Statistical differences among three or more groups were determined by one-way ANOVA, followed by Tukey Test post hoc analysis. *p < 0.05.

FIGURE 5

Pancreatic β-cells treated with telmisartan exhibit reduced Kv currents and extended APD. (A) Kv currents were recorded in voltage-clamp mode with holding potential from −70 to +80 mV in 10 mV increments. Representative current traces recorded in control and telmisartan-treated (10 μM) β-cells. (B) Current-voltage relationship curves of Kv channels and summary of the mean current density of Kv channels recorded at 80 mV depolarization (control n = 9 cells, telmisartan n = 7 cells). (C) Action potentials were elicited by 4 ms, 150 pA current. Representative action potential waveforms for β-cells treated without or with telmisartan (10 μM) and summary of the mean APDs (n = 7 cells). (D) Representative Kv current traces recorded under treatment of 100 nM Guangxitoxin-1E (GxTX-1E) alone or in combination with 10 μM telmisartan (Tel). (E) Current-voltage relationship curves of Kv channels and summary of the mean current density of Kv channels recorded at 80 mV depolarization (n = 7 cells). The cells compared between the groups in each graph are collected from the same animal. Statistical differences between two groups were determined using an unpaired two-tailed Student’s t test. For comparing the effects of GxTX-1E groups, Tukey Test post hoc analysis was applied. ***p < 0.001, *p < 0.05.

FIGURE 6

The AT-1 receptor and PPARγ are not involved in the telmisartan-induced inhibition of Kv channels, whereas telmisartan exerts a direct effect on Kv2.1 channels. (A) Representative current traces recorded upon treatment with valsartan (10 μM) and irbesartan (10 μM) in β-cells. (B) Current-voltage relationship curves and the summary of the mean current density of Kv channels recorded at 80 mV depolarization (control n = 7 cells, valsartan n = 8 cells, irbesartan n = 6 cells). (C) Representative current traces recorded under treatment of telmisartan (10 μM) alone or in combination with GW9662 (10 μM) in β-cells. (D) Current-voltage relationship curves and the summary of the mean current density of Kv channels recorded at 80 mV depolarization (control n = 8 cells, GW9662 n = 12 cells, telmisartan n = 7 cells, telmisartan+GW9662 n = 10 cells). (E) The CHO-Kv2.1 cell line was constructed using a lentivirus vector overexpressing Kv2.1 channels. Representative current traces recorded without or with telmisartan (10 μM) in CHO-Kv2.1 cells. (F) Current-voltage relationship curves and the summary of the mean current density of Kv channels recorded at 80 mV depolarization (control n = 10 cells, telmisartan n = 8 cells). The cells compared between the groups in each graph are isolated from the same animal except CHO cells. All results are reported as the means ± SEM. Statistical differences between two groups were determined using an unpaired two-tailed Student’s t test. Statistical differences among three or more groups were compared using one-way ANOVA. For comparing the effects of GW9662 groups, Tukey Test post hoc analysis was applied. *p < 0.05, **p < 0.01.

FIGURE 7

L-type VGCCs also mediate telmisartan-induced insulin secretion and increase of [Ca²⁺]_i levels, independent of the AT-1 receptor and PPARγ. (A) Rat islets were treated with telmisartan (10 μM) in the presence or absence of TEA (20 mM) under 2.8 and 11.1 mM glucose conditions and insulin secretion was measured (n = 7 tubes of islet). All insulin secretion results are normalized to basal secretion at 2.8 Mm glucose condition. (B) The changes of ([Ca²⁺]_i) concentration in β-cells treated with 20 mM TEA and in combination with 10 μM telmisartan (Tel) under 11.1 mM glucose conditions, and the average value during 30 s F340/F380 spikes for each test (n = 9 cells). (C) VGCCs were recorded in voltage-clamp mode with test potentials from −50 to 30 mV in 10 mV increments. Representative current traces recorded in control and telmisartan-treated (10 μM) β-cells. (D) Current-voltage relationship curves of VGCCs and summary of the mean Ca²⁺current density recorded at 0 mV(Continued)

FIGURE 7

depolarization (control, n = 7 cells; telmisartan, n = 8 cells). (E) Representative VGCC current traces recorded under treatment of 0.1 μM azelnidipine (Az) alone or in combination with 10 μM telmisartan (Tel). (F) Current-voltage relationship curves of VGCC and summary of the mean current density of Kv channels recorded at 0 mV depolarization (control, n = 6 cells; Az, n = 7 cells; Az+Tel, n = 6 cells). (G) Representative current traces recorded with treatment of valsartan (10 μM) and irbesartan (10 μM) in β-cells. (H) Current-voltage relationship curves and the summary of the mean Ca²⁺current density recorded at 0 mV depolarization (n = 7 cells). (I) Representative current trances recorded under treatment of telmisartan (10 μM) alone or in combination with GW9662 (10 μM) in β-cells. (J) Current-voltage relationship curves and the summary of the mean Ca²⁺ current density recorded at 0 mV depolarization (control, n = 10 cells; GW9662, n = 6; telmisartan, n = 8 cells; telmisartan+GW9662, n = 6 cells). The cells compared between the groups in each graph are isolated from the same animal. All results are reported as the means ± SEM. Statistical differences between two groups were determined using an unpaired two-tailed Student’s t test. Statistical differences among three or more groups were compared using one-way ANOVA and Student–Newman–Keuls method post hoc analysis. Effects on VGCCs between telmisartan and control in (D) were compared using the Mann–Whitney Rank Sum Test. For comparing the effects of GW9662 groups in (J), Dunn’s method post hoc analysis was applied. *p < 0.05, ***p < 0.001.

FIGURE 8

Telmisartan improves glucose tolerance in db/db mice, elevates GSIS levels in isolated islets, and exerts similar electrophysiological effects on β-cells of db/db mice. (A) OGTT was performed and AUCs for OGTT were calculated from the data in 8-wk-old mice. (B) OGTT and AUC for OGTT in 11-wk-old mice. (C) Serum insulin levels at corresponding times and AUC in 11-wk-old mice. (D) Db/db mice islets were treated with or without telmisartan (10 μM) under different glucose concentrations (2.8, 16.7, and 30 mM) (n = 6 tubes of islet). (E) Representative Kv channels current trances recorded with treatment of telmisartan (10 μM) in β-cells. (F) Current-voltage relationship curves and the summary of the mean current density of Kv channels recorded at 80 mV depolarization (n = 6 cells). (G) Representative Ca²⁺ current trances recorded with treatment of telmisartan (10 μM) in β-cells. (H) Current-voltage relationship curves and the summary of the mean Ca²⁺(Continued)

FIGURE 8

current density recorded at 0 mV depolarization (control, n = 6 cells; telmisartan, n = 7 cells). In (D), all the islets were isolated from the same mouse, and the insulin secretion results are normalized to basal secretion at 2.8 Mm glucose concentration. In (E,G), the cells compared between the groups in each graph are collected from the same animal. All results are reported as the means ± SEM. Statistical differences between two groups were determined using the unpaired two-tailed Student’s t test unless otherwise stated. Glucose levels in 11-wk-old mice at indicated time points were compared using the Mann–Whitney Rank Sum Test except at 0 min. AUCs calculated from the data of glucose levels and plasma insulin levels in 11-wk-old mice were compared using the Mann–Whitney Rank Sum Test. As for the insulin assay in (D), statistical differences among three groups (without telmisartan) were compared using one-way ANOVA and followed by Student–Newman–Keuls method post hoc analysis, and differences between two groups under the same glucose conditions were compared using the paired t test. *p < 0.05.