Arrestin domain-containing 3 (Arrdc3) modulates insulin action and glucose metabolism in liver

Abstract

Insulin action in the liver is critical for glucose homeostasis through regulation of glycogen synthesis and glucose output. Arrestin domain-containing 3 (Arrdc3) is a member of the α-arrestin family previously linked to human obesity. Here, we show that Arrdc3 is differentially regulated by insulin in vivo in mice undergoing euglycemic-hyperinsulinemic clamps, being highly up-regulated in liver and down-regulated in muscle and fat. Mice with liver-specific knockout (KO) of the insulin receptor (IR) have a 50% reduction in Arrdc3 messenger RNA, while, conversely, mice with liver-specific KO of Arrdc3 (L-Arrdc3 KO) have increased IR protein in plasma membrane. This leads to increased hepatic insulin sensitivity with increased phosphorylation of FOXO1, reduced expression of PEPCK, and increased glucokinase expression resulting in reduced hepatic glucose production and increased hepatic glycogen accumulation. These effects are due to interaction of ARRDC3 with IR resulting in phosphorylation of ARRDC3 on a conserved tyrosine (Y382) in the carboxyl-terminal domain. Thus, Arrdc3 is an insulin target gene, and ARRDC3 protein directly interacts with IR to serve as a feedback regulator of insulin action in control of liver metabolism.

Keywords: Arrdc3; alpha arrestins; glucose metabolism; insulin action; liver.

Fig. 1.

Arrdc3 mRNA is induced in liver by hyperinsulinemia. (A) Arrdc3 mRNA levels in livers of fasted WT mice during a euglycemic-hyperinsulinemic clamp (12 mU/kg/min) for 20 or 180 min (n = 6). (B) Relative Arrdc3 mRNA levels in livers of C57/BL/6J mice fed on a normal chow (20% calories from fat) versus a HFD (60% calories from fat) for 10 wk (n = 5 to 6). (C and D) Arrdc3 mRNA levels in 10-wk-old Ob/Ob mice compared to WT controls (n = 4 to 6) and in 3-mo-old LIRKO mice compared to IR^fl/fl controls (n = 8). Data are means ± SEM; *P < 0.05, **P < 0.01, ***P < 0.001, Student’s t test.

Fig. 2.

Liver-specific Arrdc3 deletion regulates glucose homeostasis. (A) Arrdc3 mRNA levels in liver, muscle, WAT, and kidney of Arrdc3^fl/fl and L-Arrdc3 KO mice (n = 4 to 6). (B) Blood glucose levels in random-fed conditions after an overnight fast or 6 h refeeding at ∼16 wk of age (n = 5 to 6). (C) Fasting insulin and glucagon levels at ∼16 wk of age (n = 4 to 6). (D) Glucose, (E) insulin, and (F) pyruvate tolerance tests at 15 to 17 wk of age (n = 4 to 6). Data are means ± SEM; *P < 0.05, **P < 0.01, Student’s t test.

Fig. 3.

Liver-specific Arrdc3 deletion increases insulin sensitivity. Euglycemic-hyperinsulinemic clamps were performed on 20-wk-old L-Arrdc3 KO and Arrdc3^fl/fl controls. (A) Blood glucose levels at baseline and during the last 30 min of the clamp (average from 90 to 120 min). (B) Average glucose infusion rate during the entire 120 min of the clamp. Data are means ± SEM; *P < 0.05, Student’s t test. (C) Insulin-stimulated 2-deoxy-d-[1-¹⁴C]glucose uptake in gastrocnemius muscle, epididymal WAT, and BAT. (D) Whole-body glucose turnover, glycolysis, glycogen synthesis, and (E) insulin-stimulated suppression of hepatic glucose production were assessed during the clamp as described in Methods (n = 6 to 7). Data are means ± SEM; *P < 0.05, **P < 0.01, ****P < 0.0001, two-way ANOVA.

Fig. 4.

Arrdc3 regulates insulin signaling and PEPCK expression in liver. (A) Western blot analysis of insulin-signaling proteins in livers from overnight fasted L-Arrdc3 KO mice and Arrdc3^fl/fl controls collected 5 min after intravenous insulin (2 IU/mouse) or phosphate-buffered saline (minus insulin) injection (n = 2 to 4). (B) Relative phosphorylation of FOXO1 calculated as the ratio of phosphoprotein/total protein from data in A (n = 2 to 4). Data are means ± SEM; *P < 0.05, two-way ANOVA. (C) Liver glycogen levels were assessed after an overnight fast followed by 6 h of refeeding (n = 4 to 5). (D) Relative Pck1 and (E) Gck mRNA expression normalized to TBP in livers from overnight-fasted or 6 h refed Arrdc3^fl/fl and L-Arrdc3 KO mice (n = 4 to 7). Data are means ± SEM; *P < 0.05, ***P < 0.001, ****P < 0.0001, two-way ANOVA. (F) PEPCK protein expression normalized to vinculin was assessed by Western blotting in livers of overnight-fasted L-Arrdc3 KO and Arrdc3^fl/fl mice (n = 4). Data are means ± SEM; *P < 0.05, Student’s t test.

Fig. 5.

ARRDC3 interacts with the IR and regulates tyrosine kinase activity. (A) IR protein levels in membrane fractions isolated from livers of overnight-fasted and 6 h refed Arrdc3^fl/fl and L-Arrdc3 KO mice (n = 3). (B) Quantification of IR levels normalized to total protein indicated by ultraviolet imaging of a stain-free gel. Data are means ± SEM; *P < 0.05, Student’s t test. (C) Coimmunoprecipitation assay of FLAG-tagged hIR and HA-tagged hARRDC3 transiently overexpressed in HEK 293 cells. (D) Insulin-induced tyrosine phosphorylation (pY) of HA-hARRDC3. (E and F) IR/IGF1R tyrosine phosphorylation in total lysates of HEK 293 cells transiently overexpressing HA-hARRDC3. Cells were serum starved for 3 h and stimulated with 100 nM insulin for 10 min. Data are means ± SEM; **P < 0.01, Student’s t test. All in vitro experiments were done at least twice.

Fig. 6.

Tyrosine 382 residue in ARRDC3 C-tail regulates its interaction with IR. (A) schematics depicting structure and truncation mutations of the ARRDC3 C-terminal tail. (B) Coimmunoprecipitation assay of FLAG-tagged hIR and HA-tagged hARRDC3 truncation constructs transiently expressed in HEK 293 cells. (C) Tyrosine phosphorylation in anti-HA precipitates of full-length or 1 to 381 truncation mutant of HA-tagged hARRDC3. (D) Coimmunoprecipitation assay of FLAG-tagged hIR and HA-tagged WT or Y382A hARRDC3 in HEK 293 cells.