GPR55 deficiency is associated with increased adiposity and impaired insulin signaling in peripheral metabolic tissues

Abstract

Emerging evidence indicates that G-protein coupled receptor 55 (GPR55), a nonclassic receptor of the endocannabinoid system that is activated by L-α-lysophosphatidylinositol and various cannabinoid ligands, may regulate endocrine function and energy metabolism. We examined how GPR55 deficiency and modulation affects insulin signaling in skeletal muscle, adipose tissue, and liver alongside expression analysis of proteins implicated in insulin action and energy metabolism. We show that GPR55-null mice display decreased insulin sensitivity in these tissues, as evidenced by reduced phosphorylation of PKB/Akt and its downstream targets, concomitant with increased adiposity and reduced physical activity relative to wild-type counterparts. Impaired tissue insulin sensitivity coincided with reduced insulin receptor substrate-1 abundance in skeletal muscle, whereas in liver and epididymal fat it was associated with increased expression of the 3-phosphoinoistide lipid phosphatase, phosphatase and tensin homolog. In contrast, GPR55 activation enhanced insulin signaling in cultured skeletal muscle cells, adipocytes, and hepatocytes; this response was negated by receptor antagonists and GPR55 gene silencing in L6 myotubes. Sustained GPR55 antagonism in 3T3-L1 adipocytes enhanced expression of proteins implicated in lipogenesis and promoted triglyceride accumulation. Our findings identify GPR55 as a positive regulator of insulin action and adipogenesis and as a potential therapeutic target for countering obesity-induced metabolic dysfunction and insulin resistance.-Lipina, C., Walsh, S. K., Mitchell, S. E., Speakman, J. R., Wainwright, C. L., Hundal, H. S. GPR55 deficiency is associated with increased adiposity and impaired insulin signaling in peripheral metabolic tissues.

Keywords: adipogenesis; cannabinoid receptor; endocannabinoid; liver; skeletal muscle.

Figure 1

Comparison of food intake and energy expenditure in WT and GPR55-deficient (KO) mice. A) GPR55 mRNA expression quantified relative to that of 18S rRNA in epididymal fat tissue, liver, and gastrocnemius muscle of WT and GPR55 KO mice (n = 4/group). B–E) Body weight (B) and daily food intake (C) for WT and GPR55 KO mice were monitored over a 12-wk period. Body temperature (D), physical activity and energy expenditure (E) for WT and GPR55^−/− mice were determined along with average values in the light phase (LP) and dark phase (DP). All values presented are the means ± sem from 12 mice unless indicated otherwise. Asterisks denote statistically significant differences between WT and GPR55^−/− mice. *P < 0.05.

Figure 2

GPR55-deficient mice exhibit increased body fat mass and enhanced lipogenic drive. A, B) Mean fat (A) and lean masses (B) for WT and GPR55 KO mice at 22 wk of age were determined (n = 10 for WT, n = 8 for KO). C, D) Comparison of total protein content (per mg of muscle tissue) in gastrocnemius muscle (n = 6 for each group) (C) and mean fat depot mass (D) for brown adipose tissue (BAT), epididymal fat (Epi), retroperitoneal fat (Retro), mesenteric fat (Mesen), omental fat (Omen), and subcutaneous fat (Sub Cut) in WT and GPR55^−/− mice are presented (n = 12 for each group). E) Protein lysates prepared from epididymal fat tissue of WT and GPR55^−/− mice were immunoblotted using the antibodies indicated (4 individual animals/group). NS, not significant. Values presented are means ± sem. *P < 0.05.

Figure 3

GPR55-deficient mice display elevated circulating blood glucose and plasma triglyceride levels. A) Blood glucose (n = 12 for each group), fasting plasma insulin (n = 5 for each group), and plasma triglyceride (n = 4 for each group) concentrations were determined in WT and GPR55^−/− mice held without food for 6 h. B) Prandial blood glucose levels (n = 3 for each group) were measured in WT and GPR55-deficient mice treated with or without insulin (2 mU/g body weight for 10 min). *P < 0.05.

Figure 4

GPR55 deficiency is associated with impaired insulin signaling in peripheral tissues. Lysates prepared from epididymal fat tissue (A), liver (B), gastrocnemius muscle (C), and solei (D) of WT and GPR55^−/− mice stimulated with or without insulin (2 mU/g body weight for 10 min) were immunoblotted using the antibodies against phospho (Thr308) and native PKB, FOXO, GSK3, and α-tubulin as shown. Values presented are the mean ± sem from 5 individual animals. *P < 0.05.

Figure 5

Effects of GPR55 deficiency upon key modulators of insulin signaling. Lysates prepared from epididymal fat tissue (A), liver (B), and gastrocnemius muscle (C) in WT and GPR55^−/− mice were immunoblotted using the antibodies against IRβ (insulin receptor β subunit), IRS-1, PDK1, protein phosphatase 2A (PP2A), and PTEN. Values presented are the means ± sem from 4 individual animals (n = 5 for liver samples). *P < 0.05.

Figure 6

Confirmation of GPR55 expression in rat and human skeletal muscle and in cultured rat myotubes and that in myotubes modulation of GPR55 affects insulin signaling. A–C) Detection of GPR55 mRNA was performed using RT-PCR analysis in rat skeletal L6 myoblasts (Mb) (A) and differentiated L6 myotubes (Mt) (A, C), rat soleus (A), and human myotubes (B). The induction of GPR55 mRNA in L6 myotubes after treatment with LPI (3 µM for 48 h) was also determined by conventional RT-PCR analysis (representative of 3 independent experiments) (C). D, E) For insulin signaling studies, L6 myotubes were treated with 3 µM LPI, 3 µM O-1602, and/or 5 μM O-1918 for 48 h prior to stimulation with insulin (20 nM for 10 min) or vehicle control. Resulting cell lysates were immunoblotted with antibodies against phospho-ERK1/2 and phospho-CREB (D) or phospho-PKB^T308 (E). Values presented are the mean ± sem from 3 independent experiments. *P < 0.05 between the indicated bars.

Figure 7

Augmentation of insulin signaling in response to GPR55 activation can be blocked by ML-193 and by shRNA-mediated receptor silencing in L6 myotubes. A, B) L6 myotubes were treated with 3 µM LPI in the presence or absence of 20 µM ML-193 (A) or 20 μM CID16020046 (B) (GPR55 antagonists) as indicated for 48 h prior to stimulation with insulin (20 nM for 10 min) or vehicle control prior to cell lysis. Resulting cell lysates were immunoblotted with antibodies against phospho-PKB^T308 and PKB. C) GPR55 mRNA expression in L6 myotubes infected with lentiviral constructs containing a non-target control shRNA or GPR55 shRNA. D, E) Effects of 3 µM LPI (D) and 10 µM 1-(2,4-difluorophenyl)-5-[[2-[[(1,1-dimethylehyl)amino]thioxomethyl]hydrazinylidene]methyl]-1H-pyrazole-4-carboxylic acid methyl ester (ML-184) (E) (GPR55 agonist) on insulin signaling (phospho-PKB^S473) in control (nontarget shRNA) infected myotubes and myotubes in which GPR55 had been silenced (GPR55 shRNA). F) IRS-1 protein abundance in control and GPR55-silenced L6 myotubes was determined by immunoblotting as indicated. NS, not significant. The bar values presented are the means ± sem from at least 3 independent experiments. *P < 0.05 between the indicated bars.

Figure 8

Confirmation of GPR55 expression in differentiated murine 3T3-L1 adipocytes, rat H4IIE, and human HepG2 hepatoma cells and effects of GPR55 modulation in these cells on insulin signaling. Expression of GPR55 in 3T3-L1 adipocytes (A), rat H4IIE liver cells (B), and human HepG2 hepatoma cells (C) was determined by conventional RT-PCR analysis (representative of 3 independent experiments). Cells were treated for 24 h with 3 μM LPI, 10 μM ML-193, and/or 10 μM CID16020046 (CID) as indicated prior to stimulation with insulin (20 nM for 10 min) or vehicle control. Resulting cell lysates were immunoblotted using the antibodies against phospho-PKB^T308 or phospho-PKB^S473. Values presented are means ± sem from 3 independent experiments. *P < 0.05 between the indicated bar values.

Figure 9

Effects of sustained GPR55 antagonism using ML-193 on adipocyte differentiation/lipogenic markers and triglyceride accumulation. Fully confluent 3T3-L1 adipocytes were allowed to differentiate for up to 10 d in the presence or absence of ML-193 (10 μM). On the indicated days, relative mRNA abundance of PPARγ, FAS, and FABP4 (A) as well as triglyceride (C) content were determined. Immunoblot analysis was performed to determine protein levels of perilipin, FAS, acetyl CoA carboxylase (ACC), FABP4, and Actin (as an internal control) in 3T3-L1 adipocytes differentiated for 6 d in the presence or absence of ML-193 (10 μM) (B). All quantified values presented are the means ± sem from 3 independent experiments. *P < 0.05 between the indicated bars.

Figure 10

Effects of sustained GPR55 silencing on adipocyte differentiation/lipogenic markers and triglyceride accumulation. After transfection with control or GPR55-trageting siRNA, 3T3-L1 adipocytes were allowed to differentiate for up to 10 d. A) GPR55 mRNA abundance was determined in control and GPR55 siRNA-transfected 3T3-L1 adipocytes by qPCR analysis. B–D) On the indicated days of differentiation, relative mRNA abundance of FAS (B), PPARγ (C), and triglyceride (D) content were determined. All quantified values presented are mean + sem from 3 independent experiments. *P < 0.05 between the indicated bars.