Pertussis toxin catalyzed ADP-ribosylation of a 41 kDa G-protein impairs insulin-stimulated glucose metabolism in BC3H-1 myocytes

A role for heterotrimeric GTP-binding proteins and ERK1/2 in insulin-mediated, nitric-oxide-dependent, cyclic GMP production in human umbilical vein endothelial cells

AIMS/HYPOTHESIS

Insulin is known to stimulate endothelial nitric oxide synthesis, although much remains unknown about the intracellular mechanisms involved. This study aims to examine, in human endothelial cells, the specific contribution of heterotrimeric Gi proteins and extracellular signal-regulated protein kinases 1/2 (ERK1/2) in insulin signalling upstream of nitric-oxide-dependent cyclic GMP production.

METHODS

Human umbilical vein endothelial cells were treated with 1 nmol/l insulin in the presence or absence of inhibitors of tyrosine kinases (erbstatin), Gi proteins (pertussis toxin) or ERK1/2 (PD098059 or U0126), and nitric oxide production was examined by quantification of intracellular cyclic GMP. Activation/phosphorylation of ERK1/2 by insulin was examined by immunoblotting with specific antibodies, and direct association of the insulin receptor with Gi proteins was examined by immunoprecipitation.

RESULTS

Treatment of cells with a physiological concentration of insulin (1 nmol/l) for 5 min increased nitric-oxide-dependent cyclic GMP accumulation by 3.3-fold, and this was significantly inhibited by erbstatin. Insulin-stimulated cyclic GMP production was significantly reduced by pertussis toxin and by the inhibitors of ERK1/2, PD098059 and U0126. Immunoblotting indicated that insulin stimulated the phosphorylation of ERK1/2 after 5 min and 1 h, and that this was completely abolished by pertussis toxin, but insensitive to the nitric oxide synthase inhibitor L-NAME. No direct interaction of the insulin receptor beta with Gialpha2 could be demonstrated by immunoprecipitation.

CONCLUSIONS/INTERPRETATION

This study demonstrates, for the first time, that nitric oxide production induced by physiologically relevant concentrations of insulin, is mediated by the post-receptor activation of a pertussis-sensitive GTP-binding protein and subsequent downstream activation of the ERK1/2 cascade.

IRS-1 and Vascular Complications in Diabetes Mellitus

The expected explosive increase in the number of patients with diabetes mellitus will increase the stress on health care. Treatment is focused on preventing vascular complications associated with the disorder. In order to develop better treatment regimens, the field of research has made a great effort in understanding this disorder. This chapter summarizes the current views on the insulin signaling pathway with emphasis on intracellular signaling events associated with insulin resistance, which lead to the prothrombotic condition in the vasculature of patience with diabetes mellitus.

The role of G proteins in insulin signalling

Insulin modulates many intracellular processes including cellular metabolism, cell proliferation and cell differentiation. Some of these processes involve significant changes in the traffic of intracellular vesicles or in the structural organization of the cell. These phenomena have been linked to the activity of regulatory GTP-binding proteins. Most, if not all functions, of the insulin receptor are associated with its tyrosine kinase activity. Thus, over the past few years, a significant effort has been dedicated to elucidate the cross-talk between the tyrosine kinase activity of the receptor and the regulation of G protein-mediated pathways. Recent progress indicates that G proteins may mediate the control of several of insulin's intracellular functions. These include the regulation of the MAP kinase pathway, the activation of phospholipase D and the regulation of glucose uptake. This article discusses some recent advances in this area.

Insulin-induced activation of NADPH-dependent H2O2 generation in human adipocyte plasma membranes is mediated by Galphai2

Human fat cells possess a multireceptor-linked H2O2-generating system that is activated by insulin. Previous studies revealed that manganese was the sole cofactor required for a hormonal regulation of NADPH-dependent H2O2 generation in vitro. In this report it is shown that the synergistic activation of NADPH-dependent H2O2 generation by Mn2+ and insulin was blocked by GDPbetaS (guanosine 5'-O-(2-thiodiphosphate)), pertussis toxin and COOH-terminal anti-Galphai1-2 or the corresponding peptide. Consistently, manganese could be replaced by micromolar concentrations of GTPgammaS (guanosine 5'-O-(3-thiotriphosphate)), which increased NADPH-dependent H2O2 generation by 20-40%. Insulin shifted the dose response curve for GTPgammaS to the left (>10-fold) and increased the maximal response. In the presence of 10 microM GTPgammaS, the hormone was active at picomolar concentrations, indicating that insulin acted via its cognate receptor. The insulin receptor and Gi were co-adsorbed on anti-Galphai and anti-insulin receptor beta-subunit (anti-IRbeta) affinity columns. Partially purified insulin receptor preparations contained Galphas, Galphai2, and Gbetagamma (but no Galphai1 or Galphai3). The functional nature of the insulin receptor-Gi2 complex was made evident by insulin's ability to modulate labeling of Gi by bacterial toxins. Insulin action was mimicked by activated Galphai, but not by Galphao or Gbetagamma, indicating that insulin's signal was transduced via Galphai2. Thus, NADPH oxidase is the first example of an effector system that is coupled to the insulin receptor via a heterotrimeric G protein.

Lactogen enhances Nb2 cell GTPase activity after 4 hours incubation

The lactogen receptor has been suggested to associate with one or more G proteins despite the absence of a 7-transmembrane spanning sequence. These studies were designed to determine whether lactogens acutely increase GTP binding to or GTPase activity in Nb2 cell membrane. Incubation of Nb2 cell membrane with either ovine PRL (10 ng/ml) or diluent for 0-1 h resulted in a decrease in total(35)S-GTP binding to both with no difference in GTP binding between PRL- and diluent-treated membranes. There was also no change in(35)S-GTP binding to Nb2 cell membrane incubated with increasing oPRL concentrations (0.001-100 ng/ml) for 60 min. α-(32)P-GTP photoaffinity labelling was used to evaluate changes in GTP binding to specific G proteins. Photoaffinity labelling of α-(32)P-GTP to no G protein was changed after preincubation with oPRL (10 ng/ml) for 0-60 min or with oPRL (0.01-10 ng/ml) for 60 min. Finally, it was determined whether oPRL had any acute effect on GTPase activity, as determined by release of(32)Pi from γ-(32)P-GTP. When Nb2 cell membrane was preincubated for 0-60 min with oPRL (10 ng/ml) or a range of oPRL concentrations (0-10 ng/ml), no change in GTPase activity was observed. However, when Nb2 cells were incubated with lactogen for 0-7 h, GTPase activity in equal quantities of Nb2 cell membrane prepared from those cells increased over time. Increased GTPase activity (64.9-74.4%;P<0.03 compared to 0 h) was observed after 4-7 h incubation with lactogen.In summary, addition of lactogen to Nb2 cell membrane did not acutely increase either GTP binding or GTPase activity. Yet when Nb2 cells were incubated with lactogen for 4 h prior to preparation of membrane, GTPase activity was significantly increased. This evidence, in addition to our previous results showing that 4 h incubation with lactogen increased G protein β subunit concentration and pertussis toxin-stimulated ADP-ribosylation of Gi, support a role for delayed lactogen modulation of one or more G proteins in the Nb2 cell, requiring at least 4 h for maximal effect.

Insulin-stimulated phosphatidylcholine hydrolysis, diacylglycerol/protein kinase C signalling, and hexose transport in pertussis toxin-treated BC3H-1 myocytes

Pertussis toxin was used to block insulin-stimulated phosphatidylinositol (PI)-glycan hydrolysis, consequent de novo synthesis of phosphatidic acid (PA) and the diacylglycerol (DAG) production that results from these two related processes in BC3H-1 myocytes. In contrast, pertussis toxin pretreatment did not inhibit insulin-stimulated hydrolysis of phosphatidylcholine (PC) which was found to be at least partly due to activation of a phospholipase D. Moreover, pertussis toxin-insensitive PC hydrolysis was accompanied by rapid biphasic increases in DAG and translocative activation of protein kinase C (PKC). Insulin-stimulated glucose transport was also insensitive to pertussis toxin pretreatment. Our findings suggest that insulin-stimulated PC hydrolysis pays an important role in DAG/PKC signalling during insulin action.

Guanine nucleotide binding regulatory proteins in liver from obese humans with and without type II diabetes: evidence for altered "cross-talk" between the insulin receptor and Gi-proteins

A novel pathway for physiological "cross-talk" between the insulin receptor and the regulatory Gi-protein has been demonstrated. We tested the hypothesis that a coupling defect between Gi and the insulin receptor is present in the liver of obese patients with and without type II diabetes. Insulin 1 x 10(-9) M (approximately ED50) and 1 x 10(-7) M (Max) inhibited pertussis toxin-catalyzed ADP ribosylation of Gi in human liver plasma membranes from lean and obese nondiabetic patients. However, 1 x 10(-7) M insulin was without effect in membranes from patients with type II diabetes. This coupling defect was not intrinsic to Gi, since Mg2+ and GTP gamma S inhibited pertussis toxin-catalyzed ADP ribosylation in both diabetic and nondiabetic patients. Binding of insulin of the alpha-subunit and activation of the tyrosine kinase intrinsic to the beta-subunit of the insulin receptor are not responsible for the coupling defect. 125I insulin binding is the same in obese patients with or without diabetes. Tyrosine kinase of the insulin receptor is decreased in diabetes. However, a monoclonal antibody to the insulin receptor (MA-20) at equimolar concentrations with insulin equally inhibits pertussis toxin-catalyzed ADP ribosylation of Gi without activating tyrosine kinase or insulin receptor autophosphorylation. Immunodetection of G-proteins suggested that Gi3 alpha was normal in diabetes and Gi1-2 alpha was decreased by 40% in the diabetic group as compared to the obese nondiabetic group but was normal when compared to the lean non diabetic group. We conclude that the novel pathway of insulin signaling involving the regulatory Gi proteins via biochemical mechanisms not directly involving the tyrosine kinase of the insulin receptor is altered in obese type II diabetes and offers a new target for the search of the mechanism(s) of insulin resistance.

Insulin stimulates GDP release from G proteins in the rat and human liver plasma membranes

Plasma membranes (1-2 mg protein) prepared from the livers of adult male rats and human organ donors were incubated with 0.6 microM [alpha-32P] guanosine triphosphate (GTP) in an adenosine triphosphate (ATP)-regenerating buffer at 37 degrees C for 1 h; during this incubation, the [32P]GTP is hydrolyzed and the nucleotide that is predominantly bound to the membranes is [32P] guanosine diphosphate (GDP). [32P]GDP release from the liver membranes was proportional to the protein concentration and increased as a function of time. At 5 mM, Ca2+, Mg2+, Mn2+, and Zn2+ maximally inhibited GDP release by 80-90%, whereas, 5 mM Cu2+ maximally stimulated the reaction by 100%. Therefore, cations were not included in the buffer used in the GDP release step. One microM Gpp(NH)p (5'-guanylylimidodiphosphate), a nonhydrolyzable analog of GTP, maximally stimulated [32P]GDP release in the liver membranes by up to 30%. Although 10 nM Gpp(NH)p had no effect on GDP release, it appeared to stabilize the hormonal effect by blocking further GDP/GTP exchange. In the rat membranes, 1-100 nM glucagon (used as a positive control) stimulated [32P]GDP release by about 17% (P < .05); similarly, 0.1-100 nM insulin stimulated [32P]GDP release by 10-13% (P < .05). In the human membranes, 10 pM to 100 nM insulin stimulated [32P]GDP release by 7-10%. In the rat membranes, 10 nM insulin stimulated [32P]GDP release by 17 and 24% at 2 and 4 min, respectively (P < .05); in the human membranes, 10 nM insulin stimulated [32P]GDP release by about 9% at 2 and 4 min.(ABSTRACT TRUNCATED AT 250 WORDS)

Glucose regulates the expression of Gi-proteins in cultured BC3H-1 myocytes

To examine whether glucose has regulatory effects on the expression of Gi-proteins, BC3H-1 myocytes were incubated for 24 hr in the presence of various concentrations of glucose (0-25 mM) and the amount of Gi-proteins was detected by pertussis toxin ADP-ribosylation and immunoblot analysis. Both detection methods showed a progressive decrease in the amount of Gi proteins in cells treated with increasing concentrations of glucose. A maximal reduction of 40% was observed after a 24 hr exposure to 25 mM glucose. The reduction in Gi-proteins correlated with a decrease in insulin-stimulated glucose transport.