Berberine Attenuates Hyperglycemia by Inhibiting the Hepatic Glucagon Pathway in Diabetic Mice

Abstract

Dysregulated glucagon drives hyperfunction in hepatic glucose output, which is the main cause of persistent hyperglycemia in type 2 diabetes. Berberine (Zhang et al., 2010) has been used as a hypoglycemic agent, yet the mechanism by which BBR inhibits hepatic gluconeogenesis remains incompletely understood. In this study, we treated diabetic mice with BBR, tested blood glucose levels, and then performed insulin, glucose lactate, and glucagon tolerance tests. Intracellular cAMP levels in hepatocytes were determined by ELISA, hepatic gluconeogenetic genes were assayed by RT-qPCR, and the phosphorylation of CREB, which is the transcriptional factor controlling the expression of gluconeogenetic genes, was detected by western blot. BBR reduced blood glucose levels, improved insulin and glucose tolerance, and suppressed lactate- and glucagon-induced hepatic gluconeogenesis in ob/ob and STZ-induced diabetic mice. Importantly, BBR blunted glucagon-induced glucose production and gluconeogenic gene expression in hepatocytes, presumably through reducing cAMP, which resulted in the phosphorylation of CREB. By utilizing a cAMP analogue, adenylate cyclase (AC), to activate cAMP synthetase, and an inhibitor of the cAMP degradative enzyme, phosphodiesterase (PDE), we revealed that BBR accelerates intracellular cAMP degradation. BBR reduces the intracellular cAMP level by activating PDE, thus blocking activation of downstream CREB and eventually downregulating gluconeogenic genes to restrain hepatic glucose production.

Figure 1

BBR improves glucose metabolism in diabetic mice. (a–d) ob/ob mice were treated with saline (control) or BBR intraperitoneally once a day for 3 weeks while consuming a normal chow diet. (a) Feeding and fasting blood glucose. ^∗P < 0.05, compared with the control group (N = 8). (b) Fasting plasma insulin levels in ob/ob mice for 18 weeks. ^∗P < 0.05, compared with the control group (N = 8). GTTs (c) and ITTs (d) were performed in ob/ob mice for 19 and 20 weeks, and areas under the curve (AUCs) were calculated. ^∗P < 0.05, compared with the control group (N = 8). (e, f) The age-matched mice were treated with saline (normal), and the STZ-induced mice were treated with saline (STZ) or BBR (STZ+BBR) once a day for 3 weeks while consuming a normal chow diet. (e) Fasting blood glucose and 12 hr AUCs of the STZ and BBR groups. ^∗P < 0.05, compared with the STZ group (N = 7‐9). (f) Glucose tolerance tests were performed, and areas under the curve (AUCs) were calculated. Each value represents the mean ± S.E.^∗P < 0.05, compared with the STZ group (N = 7‐9).

Figure 2

In vivo regulation of the gluconeogenic program by BBR. Lactate tolerance test in diabetic mice. Fasted mice were injected intraperitoneally with either NaCl or sodium lactate. Blood glucose was measured across time and the AUC data. (a) Lactate tolerance test in ob/ob mice. ^∗P < 0.05, compared with the control group (N = 8). (c) Lactate tolerance test in STZ-induced diabetic mice. ^∗P < 0.05, ^∗∗P < 0.01, compared with the STZ group (N = 7‐9). Quantitative PCR analysis of Pepck, G6pc, and PGC1α mRNA levels in livers from diabetic mice and normalized to 36B4 levels. (b) PCR analysis in ob/ob mice. ^∗P < 0.05, ^∗∗P < 0.01, compared with the control group (N = 8). (d) PCR analysis in STZ-induced diabetic mice. ^∗P < 0.05, compared with the STZ group; ^#P < 0.05, ^##P < 0.01, compared with the normal group (N = 7‐9).

Figure 3

BBR represses glucogenesis in glucagon-induced primary mouse hepatocytes and livers of ob/ob mice. (a) Glucose production from C57BL/6J primary hepatocytes incubated in the absence or presence of 10 μM BBR and/or 50 nM glucagon. ^∗P < 0.05, compared with the control group (N = 3). (b) Gene expression measured by quantitative PCR in primary mouse hepatocytes. ^∗P < 0.05, compared with the control group (N = 3). (c) Glucagon tolerance test. ^∗P < 0.05, compared with the control group (N = 8). (d) CREB protein expression levels in liver homogenates and hepatocytes of ob/ob mice treated with vehicle, glucagon, or BBR, as indicated. The data are expressed as the mean ± S.E. (N = 3).

Figure 4

Effects of BBR on cAMP levels and CREB phosphorylation induced by various agonists. (a) cAMP in hepatocytes after glucagon treatment with and without BBR pretreatment. ^##P < 0.01, compared with the none group; ^∗P < 0.05, compared with the glucagon group (N = 5). (b) cAMP in the liver with and without intravenous glucagon treatment. ^##P < 0.01, compared with the none group; ^∗P < 0.05, compared with the glucagon group (N = 5). (c–e) Primary hepatocytes were preincubated with BBR for 15 min, then stimulated with 1 mM 8-bromo-cAMP (c), 10 μM forskolin (d), or 1 mM IBMX (e) for another 30 min, and protein was analyzed by western blot with the total CREB antibodies and the phospho- (p-) CREB antibody. CREB phosphorylation was normalized to total CREB levels. The data are expressed as the mean ± S.E.^#P < 0.05, compared with the untreated group; ^∗P < 0.05, compared with the agonist group (N = 3).