Histidine augments the suppression of hepatic glucose production by central insulin action

Abstract

Glucose intolerance in type 2 diabetes is related to enhanced hepatic glucose production (HGP) due to the increased expression of hepatic gluconeogenic enzymes. Previously, we revealed that hepatic STAT3 decreases the expression of hepatic gluconeogenic enzymes and suppresses HGP. Here, we show that increased plasma histidine results in hepatic STAT3 activation. Intravenous and intracerebroventricular (ICV) administration of histidine-activated hepatic STAT3 reduced G6Pase protein and mRNA levels and augmented HGP suppression by insulin. This suppression of hepatic gluconeogenesis by histidine was abolished by hepatic STAT3 deficiency or hepatic Kupffer cell depletion. Inhibition of HGP by histidine was also blocked by ICV administration of a histamine H1 receptor antagonist. Therefore, histidine activates hepatic STAT3 and suppresses HGP via central histamine action. Hepatic STAT3 phosphorylation after histidine ICV administration was attenuated in histamine H1 receptor knockout (Hrh1KO) mice but not in neuron-specific insulin receptor knockout (NIRKO) mice. Conversely, hepatic STAT3 phosphorylation after insulin ICV administration was attenuated in NIRKO but not in Hrh1KO mice. These findings suggest that central histidine action is independent of central insulin action, while both have additive effects on HGP suppression. Our results indicate that central histidine/histamine-mediated suppression of HGP is a potential target for the treatment of type 2 diabetes.

FIG. 1.

Phosphorylation of hepatic STAT3 after protein or histidine administration. A: Western blotting was performed to analyze the phosphorylation of STAT3 and Akt in the liver at 120 min after the oral administration of protein extract (2 g/kg body wt) and injection of glucose (2 g/kg body wt i.p.) or saline. B: Somatostatin was injected continuously at 3 μg ⋅ kg⁻¹ ⋅ min⁻¹ via the jugular vein, and phosphorylation of hepatic STAT3 and Akt was analyzed at 120 min after the oral administration of protein. C: Phosphorylation of hepatic STAT3 was analyzed by Western blotting at 120 min after the intraperitoneal administration of amino acids (0.5 mmol/kg). The experiment was performed in quadruplicate (n = 4), and samples were combined for Western blot analysis. 1 mH, 1-methyl-histidine; Ans, anserine; Car, carnosine; and hLy, hydroxylysine. D: Phosphorylation of STAT3 was analyzed at 120 min after the oral administration of protein or Low-His. E and F: A glucose tolerance test was performed with the intraperitoneal administration of histidine (His) or saline (Veh). Changes in blood glucose levels (E) (left panel), blood glucose levels at 120 min after glucose loading (E) (right panel), and changes in plasma insulin levels (F) are shown. *P < 0.05 (n = 10). G and H: Western blotting analysis of hepatic STAT3 and Akt phosphorylation (G) and quantitative PCR analysis of Pck1 and G6pc gene expression (H) were performed at 120 min after the intraperitoneal injection of histidine or saline (-) and glucose (2 g/kg body wt). *P < 0.05 (n = 5). I and J: A fructose (I) and pyruvate (J) tolerance test was performed after the intraperitoneal administration of histidine or saline. *P < 0.05 (n = 10). Cont, control.

FIG. 2.

Augmentation of the insulin-dependent suppression of HGP by histidine. A: Schematic of hyperinsulinemic-euglycemic clamping with continuous intravenous administration of histidine (His) or saline (-). B and C: GIR (B, left panel), R_d (B, center panel), and the suppression (%) of HGP (B, right panel) in the study described in A. Quantitative PCR analysis of Pck1 and G6pc gene expression levels in the liver is shown in C. *P < 0.05 (n = 7–9). D and E: Western blotting was performed to analyze the levels of hepatic PEPCK, G6Pase, and β-actin (D) and phosphorylation of STAT3 and Akt (E) at 120 min after insulin (Ins) administration in the hyperinsulinemic-euglycemic clamp study (A). Quantitation of PEPCK and G6Pase levels is normalized to β-actin (D). Quantitation of phospho-Akt and phospho-STAT3 is normalized to Akt and STAT3, respectively (E). Data are represented as means ± SE (n = 7–9) values. F: Schematic of euglycemic-pancreatic clamp tests with continuous intravenous administration of histidine or saline in addition to somatostatin injection. G: The GIR (left panel) and the suppression (%) of HGP (right panel) in the study described in F (n = 5).

FIG. 3.

Activation of hepatic IL-6/STAT3 signaling pathways is required for the histidine-mediated regulation of HGP. A and B: A glucose tolerance test was conducted after the intraperitoneal injection of histidine (His) in liver-specific STAT3-deficient (LST3KO) mice and controls (Con). Changes in blood glucose levels (A, left panel), blood glucose levels at 120 min after glucose loading (A, right panel), and changes in plasma insulin levels (B) are shown. *P < 0.05 (n = 10). C: Expression of hepatic Pck1 and G6pc genes was analyzed by quantitative PCR at 120 min after the intraperitoneal injection of His into LST3KO and control (Con) mice. *P < 0.05 (n = 6–8). D: Hyperinsulinemic-euglycemic clamping was conducted as described in Fig. 2A. The GIR (left panel), R_d (center panel), and suppression (%) of HGP (right panel) were measured. *P < 0.05 (n = 6). E: Primary hepatocytes were treated with the different concentrations of histidine and insulin (Ins) shown in the figure, and the phosphorylation of STAT3, Akt, p70 S6 kinase, and S6 ribosomal protein was analyzed. F–I: The effect of intraperitoneal histidine injection on liver function was investigated using an IL-6 neutralizing antibody. F: Phosphorylation of hepatic STAT3 was analyzed by Western blotting at 120 min after histidine administration. G–I: Hepatic G6pc gene expression (G), blood IL-6 levels (H), and hepatic Il6 gene expression (I) at 120 min after histidine administration are shown. *P < 0.05 (n = 7–9).

FIG. 4.

Liver Kupffer cells are essential for the regulation of HGP by histidine (His). A: Immunostaining with an anti-Mac-2 antibody was performed in Kupffer cells after the administration of CLD. Scale bars, 50 nm. B–D: Histidine was injected intraperitoneally into mice pretreated with CLD, and Emr1 (B), Il6 (C), and Pck1 and G6pc (D) gene expression in the liver was quantitated using PCR at 120 min after histidine administration. *P < 0.05 (n = 5–7). E: The intraperitoneal injection of histidine and a glucose tolerance test were conducted in mice pretreated with CLD. Changes in blood glucose levels (E, left panel) and blood glucose levels at 120 min after glucose loading (E, right panel) are shown. *P < 0.05 (n = 9). F–H: With the continuous administration of histidine, hyperinsulinemic-euglycemic clamping (Fig. 2A) was performed in mice pretreated with CLD, and the GIR (left panel) and suppression (%) of HGP (right panel) were measured. In addition, the expression of hepatic Pck1 and G6pc genes (G) and the phosphorylation of hepatic STAT3 and Akt (H) at 120 min after insulin (Ins) administration were examined. *P < 0.05 (n = 6).

FIG. 5.

Central histamine action is required for the regulation of HGP by histidine. A: Hypothalamic histamine (HA) content was measured at 30 and 120 min after the intraperitoneal administration of histidine (His). *P < 0.05 (n = 5). B and C: HA was injected into the ventricle at the concentrations shown in the figure, and the phosphorylation of hepatic STAT3 and Akt (B) and expression of hepatic Il6 gene (C) were analyzed at 180 min postadministration. *P < 0.05 (n = 5). D: Histidine was injected intraperitoneally at 15 min after the lateral ventricular administration of the histamine H₁ receptor antagonist pyrilamine (50 μg/mouse), and phosphorylation of hepatic STAT3 was analyzed at 120 min after histidine administration. E: The intraperitoneal administration of histidine was performed at 30 min after the lateral ventricular administration of the histamine H₃ receptor antagonist thioperamide (50 nmol/mouse), and phosphorylation of hepatic STAT3 was analyzed at 120 min after histidine administration. F: Phosphorylation of hepatic STAT3 was analyzed (n = 4) at 180 min after the lateral ventricular administration of the histamine H₁ receptor agonist pyridylethylamine at the concentrations indicated in the figure. G–I: A hyperinsulinemic-euglycemic clamp (Ins injection) was performed with the continuous administration of histidine (His) or saline (-) in the jugular vein and pyrilamine (PL) or artificial cerebrospinal fluid (-) in the lateral ventricle, as shown in the schematic (G). The GIR (H, left panel) and the suppression (%) of HGP (H, right panel) are shown. ANOVA for GIR (H, left panel) revealed a significant effect of pyrilamine and histidine, as well as an interaction between pyrilamine and histidine (P < 0.05, n = 6). ANOVA for HGP (H, right panel) revealed a significant main effect of pyrilamine (P < 0.05) and an interaction between pyrilamine and histidine (P < 0.01, n = 6). In addition, the expression of hepatic Pck1 and G6pc genes was analyzed at 120 min after insulin administration (I). *P < 0.05 (n = 6). Cont, control.

FIG. 6.

ICV administration of histidine (His) augments the insulin-dependent suppression of HGP. A–C: With use of the doses indicated in the figure, histidine was injected rapidly into the lateral ventricle, and hypothalamic histamine (HA) content (A), phosphorylation of hepatic STAT3 (B), and expression of the hepatic G6pc gene (C) were analyzed at 180 min postadministration. *P < 0.05 (n = 5–8). D–G: As shown in the schematic (D), hyperinsulinemic-euglycemic clamping was performed concurrently with the continuous administration of histidine or artificial cerebrospinal fluid (-) into the lateral ventricle. The GIR (E, left panel), R_d (E, center panel), and suppression (%) of HGP (E, right panel) were measured. *P < 0.05 (n = 6). In addition, the expression of hepatic Pck1 and G6pc genes (F) and phosphorylation of hepatic STAT3 (G) were analyzed at 120 min after insulin administration. ANOVA for Pck1 (F, left panel) revealed a significant main effect of Ins (P < 0.05) but no significant interaction between Ins and His (P = 0.060, n = 6). ANOVA for G6pc (F, right panel) revealed a significant main effect of Ins and histidine (P < 0.05), but no significant interaction between insulin and histidine treatments (P = 0.120, n = 6).

FIG. 7.

Central histidine (His) action augments the suppression of HGP by insulin ICV administration. A–D: As shown in the schematic (A), continuous lateral ventricular injection of histidine, artificial cerebrospinal fluid (aCSF) (-), and insulin (Ins) in pancreatic clamping was performed. The suppression (%) of HGP, *P < 0.05 (n = 6) (B), expression of hepatic Pck1 and G6pc genes at 180 min after the start of lateral ventricular administration (C), and phosphorylation of hepatic STAT3 (D) were analyzed. ANOVA for Pck1 expression (C, left panel) revealed a significant main effect of Ins (P < 0.05) but no significant interaction between insulin and histidine (P = 0.964, n = 6). ANOVA for G6pc expression (C, right panel) revealed a significant main effect of insulin and histidine (P < 0.05) but no significant interaction between insulin and histidine treatments (P = 0.116, n = 6).

FIG. 8.

Central histidine (His) action on the liver is independent of central insulin (Ins) action. A and B: Phosphorylation of STAT3 was analyzed at 180 min after the intravenous administration of somatostatin (3 μg ⋅ kg⁻¹ ⋅ min⁻¹) and the lateral ventricular administration of histidine (A and B) or insulin (B), which was performed at 15 min after the ICV preadministration of the histamine H₁ antagonist pyrilamine (50 μg/mouse). Quantitation is represented as means ± SE (n = 6–8). C and D: Somatostatin administration (3 μg ⋅ kg⁻¹ ⋅ min⁻¹) and the lateral ventricular administration of histidine (C) or insulin (D) were performed in Hrh1KO mice, and the phosphorylation of hepatic STAT3 was analyzed at 180 min postadministration. Quantitation is represented as means ± SE (n = 6). E: The ICV administration of insulin (100 μU/mouse, bolus) was performed at 30 and 120 min after the administration of the PI3-K inhibitor LY294002 (10 nmol/mouse; LY). The phosphorylation of hypothalamic Akt at 30 min after insulin administration is shown. F: Somatostatin administration (3 μg ⋅ kg⁻¹ ⋅ min⁻¹) and the lateral ventricular administration of histidine or insulin were performed at 120 min after the lateral ventricular administration of LY294002 (10 nmol/mouse; LY), and the phosphorylation of hepatic STAT3 was analyzed at 180 min postadministration. Quantitation is represented as means ± SE (n = 5). G: Somatostatin (3 μg ⋅ kg⁻¹ ⋅ min⁻¹) and histidine were administered into NIRKO mice, and the levels of hypothalamic insulin receptor (IRβ), β-actin, and hepatic STAT3 were analyzed at 180 min postadministration. Quantitation is represented as means ± SE (n = 6). aCSF, artificial cerebrospinal fluid; WT, wild type.