Role of GLP-1 in the Hypoglycemic Effects of Wild Bitter Gourd

Abstract

This study aimed to examine the role of GLP-1 in the hypoglycemic activity of wild bitter gourd (Momordica charantia L., BG). In vitro, the GLP-1 secretion in STC-1, a murine enteroendocrine cell line, was dose dependently stimulated by water extract (WE), its fractions (WEL, >3 kD and WES, <3 kD), and a bitter compounds-rich fraction of BG. These stimulations were partially inhibited by probenecid, a bitter taste receptor inhibitor, and by U-73122, a phospholipase C β 2 inhibitor. These results suggested that the stimulation might involve, at least in part, certain bitter taste receptors and/or PLC β 2-signaling pathway. Two cucurbitane triterpenoids isolated from BG, 19-nor-cucurbita-5(10),6,8,22-(E),24-pentaen-3 β -ol, and 5 β ,19-epoxycucurbita-6,24-diene-3 β ,23 ξ -diol (karavilagenine E,) showed relative high efficacy in the stimulation. In vivo, mice fed BG diet showed higher insulinogenic index in an oral glucose tolerance test. A single oral dose of WE or WES pretreatment significantly improved intraperitoneal glucose tolerance. A single oral dose of WES significantly decreased glucose and increased insulin and GLP-1 in serum after 30 min. This acute hypoglycemic effect of WES was abolished by pretreatment with exendin-9, a GLP-1 receptor antagonist. Our data provide evidence that BG stimulates GLP-1 secretion which contributes, at least in part, to the antidiabetic activity of BG through an incretin effect.

Figure 1

Oral glucose tolerance and insulinogenic index of mice fed the BGP diet for 5 weeks. (a) Serum glucose and area under curve (AUC), (b) serum insulin and area under curve (AUC), and (c) insulinogenic index in an oral glucose tolerance test (OGTT). C57BL/6J male mice were fed a basal or a 5% BGP diet for 5 weeks. For OGTT, mice were feed deprived overnight and orally fed a glucose solution at the dose of 2 g/kg body weight. Insulinogenic index was calculated as the ratio of Δ serum insulin (15 min–0 min)/Δ serum glucose (15 min–0 min), and that of the basal group was taken as 1. Data are mean ± SD. *P < 0.05 and **P < 0.01 denote significant difference compared to the basal group analyzed by Student's t test.

Figure 2

Effects of various BG extracts on GLP-1 secretion in STC-1. After a 3 hr starvation in 10% FBS DMEM (low glucose), cells were treated with 0.2 % BSA DMEM (low glucose) containing (a) WE: water extract, EE: ethanol extract, or EAE: ethyl acetate extract; (b) WEL: large molecular weight fraction, >3 kD, of WE, WES: small molecular weight fraction, <3 kD, of WE, or Pf: P fraction, insulin-like peptide-rich fraction; (c) BGP-bi: crude bitter taste compounds extract for 1 hr. All data were calculated as GLP-1 secretion per cell and the secretion of vehicle-treated cells was taken as 1. The basal GLP-1 secretion of the vehicle-treated cells was 40~120 pM/well. Data are mean ± SD of 1~3 batches of experiments, with n = 1~3 for each treatment. *P < 0.05, **P < 0.01, and ***P < 0.001 denote significant difference compared to vehicle-treated cells analyzed by Student's t test with RCBD to adjust for the differences between separate batches of experiments.

Figure 3

Effects of probenecid or U73122 on the GLP-1 secretion stimulated by WE or BGP-Pi in STC-1. After a 3 hr starvation in 10% FBS DMEM (low glucose), cells were treated with 0.2 % BSA DMEM (low glucose) containing (a) WE (water extract) or (b) BGP-bi in the absence or presence of probenecid or U73122 for 1 hr. All data were calculated as GLP-1 secretion per cell, and the secretion of vehicle-treated cells was taken as 1. Data are mean ± SD of 1~3 batches of experiments, with n = 1~3 for each treatment. *P < 0.05, **P < 0.01, and ***P < 0.001 denote significant difference compared to vehicle-treated cells analyzed by Student's t test with RCBD to adjust for the differences between separate experiments.

Figure 4

Effects of various BG compounds on GLP-1 secretion in STC-1. After a 3 hr starvation in 10% FBS DMEM (low glucose), cells were treated with 0.2% BSA DMEM (low glucose) containing tested compounds for 1 hr. Tested compounds include (a) compound 1, cucurbita-6,22(E),24-trien-3β-ol-19,5β-olide; (b) compound 2, 5β,19-epoxycucurbita-6,22(E),24-triene-3β,19-diol; (c) compound 3, 3β-hydroxycucurbita-5(10),6,22(E),24-tetraen-19-al; (d) compound 4, 19-dimethoxycucurbita-5(10),6,22(E),24-tetraen-3β-ol; (e) compound 5, 19-nor-cucurbita-5(10),6,8,22-(E),24-pentaen-3β-ol; (f) Compound 6, 5β,19-epoxycucurbita-6,24-diene-3β,23ξ-diol (karavilagenine E,); (g) oleanolic acid; (h) conjugated linolenic acid (CLN) and other fatty acids (PA, palmitic acid; OA, oleic acid; LA, linoleic acid; LN, Linoleic acid; CLA, conjugated linoleic acids). All data were calculated as GLP-1 secretion per cell, and the secretion of vehicle-treated cells was taken as 1. Data shown are mean ± SD of 1~3 batches of experiments, with n = 1~3 for each treatment. *P < 0.05, **P < 0.01, and ***P < 0.001 denote significant difference compared to vehicle-treated cells analyzed by Student's t test with RCBD to adjust for the differences between separate experiments.

Figure 5

Acute effects of WE, WEL, and WES on serum glucose concentrations in an intraperitoneal glucose tolerance test (ipGTT). High-fat-fed mice were fasted for 6 hr and orally administered with a single dose of WE (2100 mg/kgBW), WES (1800 mg/kgBW) or WEL (300 mg/kgBW) or vehicle containing equivalent amount of glucose as in WE. Thirty minutes later, they were i.p. injected with 1 g/kg BW glucose for ipGTT. Changes in serum glucose were shown in (a) WE, WES versus Vehicle and (b) WE, WEL versus vehicle. Data are mean ± SD. *P < 0.05 and **P < 0.01 denote significant difference compared to vehicle-treated mice analyzed by Student's t test.

Figure 6

Effects of exendin-9 on the acute hypoglycemic effect of WES. Changes in plasma glucose concentration after a single oral dose of WES to high-fat-diet fed mice without (a) or with (b) a prior administration of Exendin-9, a GLP-1 receptor antagonist. Mice were fasted for 6 hr and blood samples collected at time 0. PBS (a) or 0.12 mg/kg BW Exendin-9 (b) were i.p. injected. Five minutes later, WES (3000 mg/kg BW) (S, SE) or vehicle (V, VE) containing equal amount of glucose as in WES (258 mg/kg BW) were administered orally and blood samples collected at the time points indicated. (c) area under curve of (a) and (b). Data are mean ± SD with n = 4 for each group. *P < 0.05 and **P < 0.01 denote significant difference compared to vehicle-treated mice analyzed by Student's t test.