Insulin signalling: metabolic pathways and mechanisms for specificity.

Biological actions of insulin are mediated by the insulin receptor, a member of a large family of receptor tyrosine kinases (RTK). Signal transduction by the insulin receptor follows a paradigm for RTK signalling. Many intracellular signalling molecules contain multiple modular domains that mediate protein-protein interactions and participate in the formation of signalling complexes. Phosphorylation cascades are also a prominent feature of RTK signalling. Distal pathways are difficult to dissect because branching paths emerge from downstream effectors and several upstream inputs converge upon single branch points. Thus, insulin action is determined by complicated signalling networks rather than simple linear pathways. Interestingly, many signalling molecules downstream from the insulin receptor are also activated by a plethora of RTKs. Therefore, mechanisms that generate specificity are required. In this review we discuss recent advances in the elucidation of specific metabolic insulin signalling pathways related to glucose transport, one of the most distinctive biological actions of insulin. We also present examples of potential mechanisms underlying specificity in insulin signalling including interactions between multiple branching pathways, subcellular compartmentalization, tissue-specific expression of key effectors and modulation of signal frequency and amplitude.     

IRS2 and PTP1B: Two opposite modulators of hepatic insulin signalling

Type 2 Diabetes mellitus (T2D) is the most common endocrine disorder associated to metabolic syndrome (MS) and occurs when insulin secretion can no compensate peripheral insulin resistance. Among peripheral tissues, the liver controls glucose homeostasis due to its ability to consume and produce glucose. The molecular mechanism underlying hepatic insulin resistance is not completely understood; however, it involves the impairment of the insulin signalling network. Among the critical nodes of hepatic insulin signalling, insulin receptor substrate 2 (IRS2) and protein tyrosine phosphatase 1B (PTP1B) modulate the phosphatidylinositol (PI) 3-kinase/Akt/Foxo1 pathway that controls the suppression of gluconeogenic genes. In this review, we will focus on recent findings regarding the molecular mechanism by which IRS2 and PTP1B elicit opposite effects on carbohydrate metabolism in the liver in response to insulin. Finally, we will discuss the involvement of the critical nodes of insulin signalling in non-alcoholic fatty liver disease (NAFLD) in humans.     

Palmitate modulates the early steps of insulin signalling pathway in pancreatic islets

Insulin stimulates its own secretion and synthesis by pancreatic beta-cells. Although the exact molecular mechanism involved is unknown, changes in beta-cell insulin signalling have been recognized as a potential link between insulin resistance and its impaired release, as observed in non-insulin-dependent diabetes. However, insulin resistance is also associated with elevated plasma levels of free fatty acids (FFA) that are well known modulators of insulin secretion by pancreatic islets. This information led us to investigate the effect of FFA on insulin receptor signalling in pancreatic islets. Exposure of pancreatic islets to palmitate caused up-regulation of several insulin-induced activities including tyrosine phosphorylation of insulin receptor and pp185. This is the first evidence that short exposure of these cells to 100 microM palmitate activates the early steps of insulin receptor signalling. 2-Bromopalmitate, a carnitine palmitoyl-CoA transferase-1 inhibitor, did not affect the effect of the fatty acid. Cerulenin, an acylation inhibitor, abolished the palmitate effect on protein levels and phosphorylation of insulin receptor. This result supports the proposition that protein acylation may be an important mechanism by which palmitate exerts its modulating effect on the intracellular insulin signalling pathway in rat pancreatic islets.     

The insulin receptor and type I IGF receptor: Comparison of structure and function

The insulin receptor and type I IGF receptor are closely related in structure and function. The receptors are heterotetrameric glycoproteins, of structure αββα, which are widely distributed in mammalian tissues. A third member of this receptor family has been described, the insulin receptor-related receptor, for which a ligand has still to be identified. It has also been demonstrated that the insulin receptor and IGF receptor form αββ′α′ hybrids in cells expressing both receptors.The key elements in the function of any receptor are recognition of ligand and transmission of an intracellular signal. In the insulin and IGF receptors, determinants of binding specificity are contained within amino-terminal and cysteine-rich domains of the extracellular α-subunit. Intracellular signalling is dependent on ligand activated tyrosine kinase activity in the transmembrane β-subunit, which phosphorylates both the receptor itself and the specific substrate insulin receptor substrate-1 (IRS-1). Phosphorylated IRS-1 binds the enzyme phosphatidylinositol 3-kinase and may act as a multivalent docking site for SH2 domains of other proteins involved in signalling. The possibility that some signalling molecules interact directly with the receptors has not been ruled out.The specificity of action of insulin and IGFs in vivo depends on differences between the respective receptors in tissue distribution, ligand binding specificity and intrinsic signalling capacity. However, the detailed aspects of gene and receptor structure which underly these functional differences are still poorly understood. Moreover, the issue of specificity is complicated by the existence of hybrid and atypical receptors, which in principle could bind and respond to both insulin and IGF-I, although the physiological significance of these receptor subtypes is at present unclear.     

Insulin receptor substrate 1 and 2 (IRS1 and IRS2): what a tangled web we weave

The insulin receptor is a transmembrane tyrosine kinase that is essential for mediating multiple intracellular signalling cascades that lead ultimately to the biological actions of insulin Tyrosine phosphorylation o f the cytosolic proteins insulin receptor substrate 1 and 2 (IRS1 and IRS2) produces protein 'scaffolding' for the assembly of effector proteins containing Src homology 2 (SH2) domains, thereby generating multisubunit signalling complexes. Although IRS1 was originally isolated as a specific insulin receptor substrate, both IRS1 and IRS2 appear to play a broader role, functioning also as proximal substrates in growth hormone and cytokine receptor signalling. Current data establish IRS1 and IRS2 as critical effectors integrating various cell-type-specific signals into distinct, but overlapping, biological responses.     

Identification of the insulin-responsive tyrosine phosphorylation sites on IRSp53

The 53-kDa insulin receptor substrate protein (IRSp53) is part of a regulatory network that organises the actin cytoskeleton in response to stimulation by small GTPases, promoting formation of actin-rich cell protrusions such as filopodia and lamellipodia. It had been established earlier that IRSp53 is tyrosine phosphorylated in response to stimulation of the insulin and insulin-related growth factor receptors, but the consequences of tyrosine phosphorylation for IRSp53 function are unknown. Here, we have used a variety of IRSp53 truncation and point mutants to identify insulin-responsive tyrosine phosphorylation sites on IRSp53. We have found that the C-terminal half of IRSp53 (residues 251-521) undergoes tyrosine phosphorylation in response to insulin stimulation of the insulin beta receptor or epidermal growth factor stimulation via the epidermal growth factor receptor, and that the key residue for insulin receptor-mediated phosphorylation is tyrosine 310, located in a region between the N-terminal IRSp53/MIM homology domain (IMD, residue 1-250) and the central SH3 domain (residues 374-438) that is predicted to be natively unstructured. Mutation of tyrosine 310 to phenylalanine or glutamic acid abrogates the phosphorylation in response to insulin stimulation, but not in response to stimulation of the epidermal growth factor receptor. The N-terminal IMD, which mediates dimerisation of IRSp53, is required for efficient tyrosine phosphorylation downstream of either the insulin or epidermal growth factor receptor stimulation, yet does not appear to include a tyrosine-phosphorylated site itself. Thus, we have identified tyrosine 310 as a primary site of tyrosine phosphorylation in response to insulin signalling and we have shown that although IRSp53 is tyrosine phosphorylated in response to epidermal growth factor receptor signalling, tyrosine 310 is not crucial. Furthermore, the tyrosine phosphorylation status does not appear to affect the cell morphology and production of filopod-like structures upon expression of IRSp53.     

A modelling-experimental approach reveals insulin receptor substrate (IRS)-dependent regulation of adenosine monosphosphate-dependent kinase (AMPK) by insulin

Mammalian target of rapamycin (mTOR) kinase responds to growth factors, nutrients and cellular energy status and is a central controller of cellular growth. mTOR exists in two multiprotein complexes that are embedded into a complex signalling network. Adenosine monophosphate-dependent kinase (AMPK) is activated by energy deprivation and shuts off adenosine 5'-triphosphate (ATP)-consuming anabolic processes, in part via the inactivation of mTORC1. Surprisingly, we observed that AMPK not only responds to energy deprivation but can also be activated by insulin, and is further induced in mTORC1-deficient cells. We have recently modelled the mTOR network, covering both mTOR complexes and their insulin and nutrient inputs. In the present study we extended the network by an AMPK module to generate the to date most comprehensive data-driven dynamic AMPK-mTOR network model. In order to define the intersection via which AMPK is activated by the insulin network, we compared simulations for six different hypothetical model structures to our observed AMPK dynamics. Hypotheses ranking suggested that the most probable intersection between insulin and AMPK was the insulin receptor substrate (IRS) and that the effects of canonical IRS downstream cues on AMPK would be mediated via an mTORC1-driven negative-feedback loop. We tested these predictions experimentally in multiple set-ups, where we inhibited or induced players along the insulin-mTORC1 signalling axis and observed AMPK induction or inhibition. We confirmed the identified model and therefore report a novel connection within the insulin-mTOR-AMPK network: we conclude that AMPK is positively regulated by IRS and can be inhibited via the negative-feedback loop.     

Insulin receptor activation inhibits insulin secretion from human islets of Langerhans

There is no consensus on the role of insulin secreted from pancreatic β-cells in regulating its own secretion, either in rodent islets or in human islets. We have now investigated whether there is an autocrine signalling role for insulin in human islets by determining insulin receptor expression and assessing the effects of insulin receptor activation using a non-peptidyl insulin mimetic termed L-783,281. Human insulin receptor mRNA was detected by PCR amplification of human islet cDNA, and translation of the message in human islets was confirmed by Western blotting. Perifusion experiments revealed that both glucose-stimulated and basal insulin secretion were significantly inhibited following human islet insulin receptor activation with L-783,281, and that signalling through phosphatidylinositol 3-kinase (PI 3-kinase) was responsible, at least in part, for this inhibitory effect. These studies indicate that human islets express insulin receptors and that they are functionally coupled to a PI 3-kinase-dependent inhibition of insulin secretion.     

Oleoylethanolamide,a natural ligand for PPAR-alpha,inhibits insulin receptor signalling in HTC rat hepatoma cells

Oleoylethanolamide (OEA) is a lipid mediator belonging to the fatty acid ethanolamides family. It is produced by intestine and adipose tissue. It inhibits food intake and body weight gain, and has hypolipemiant action in vivo, as well as a lipolytic effect in vitro. OEA is a PPAR-alpha agonist, and recently it has been found that OEA is an endogenous ligand of an orphan receptor. Previously, we have shown that OEA inhibits insulin-stimulated glucose uptake in isolated adipocytes, and produces glucose intolerance in rats. In the present work, we have studied another insulin target cell, the hepatocyte using a rat hepatoma cell line (HTC), and we have studied the cross-talk of OEA signalling with metabolic and mitotic signal transduction of insulin receptor. OEA dose-dependently activates JNK and p38 MAPK, and inhibits insulin receptor phosphorylation. OEA inhibits insulin receptor activation, blunting insulin signalling in the downstream PI3K pathway, decreasing phosphorylation of PKB and its target GSK-3. OEA also inhibits insulin-dependent MAPK pathway, as assessed by immunoblot of phosphorylated MEK and MAPK. These effects were reversed by blocking JNK or p38 MAPK using pharmacological inhibitors (SP 600125, and SB 203580). Since OEA is an endogenous PPAR-alpha agonist, we investigated whether a pharmacologic agonist (WY 14643) may mimic the OEA effect on insulin receptor signalling. Activation of PPAR-alpha by the pharmacological agonist WY14643 in HTC hepatoma cells is sufficient to inhibit insulin signalling and this effect is also dependent on p38 MAPK but not JNK kinase. In summary, OEA inhibits insulin metabolic and mitogenic signalling by activation of JNK and p38 MAPK via PPAR-alpha.     

Calreticulin regulates insulin receptor expression and its downstream PI3 Kinase/Akt signalling pathway

Defects in insulin signalling and glucose metabolism are associated with the development of diabetes. Insulin signalling is initiated by the binding of insulin to its receptor and triggering cascades of events including activation of PI3kinase/Akt signalling pathway. Calreticulin (CRT) is a calcium binding chaperone molecule located in the endoplasmic reticulum. Targeted deletion of CRT in mice is embryonic lethal as a result of developmental and metabolic abnormalities. Rescued CRT null mice develop severe hypoglycemia the reason for which is not known. In addition, ventricular cardiomyocytes isolated from CRT null (crt-/-) mice have increased glycogen deposits. Therefore, the aim of this study was to investigate the changes in the glucose uptake and insulin signalling pathway (mainly PI3 kinase/Akt) in the absence of CRT. Here we show a significant increase in the glucose uptake by the crt-/- cells. This increase was accompanied by a significant increase in both insulin receptor beta expression, Insulin receptor substrate-1 phosphorylation, GLUT-1 expression and in insulin stimulated Akt phosphorylation and kinase activity in the crt-/- cells. Intriguingly, the increased expression of insulin receptor beta in the crt-/- was due to decreased levels of p53 protein. The current study is the first evidence for the up-regulation of insulin receptor density and activity in the absence of CRT function.     

Role of binding proteins to IRS-1 in insulin signalling

Insulin elicits its divergent metabolic and mitogenic effects by binding to its specific receptor, which belongs to the family of receptor tyrosine kinases. The activated insulin receptor phosphorylates the intracellular substrate IRS-1, which then binds various signalling molecules that contain SRC homology 2 domains, thereby propagating the insulin signal. Among these IRS-1-binding proteins, the Grb2-Sos complex and the protein tyrosine phosphatase SHP-2 transmit mitogenic signals through the activation of Ras, and phosphoinositide 3-kinase is implicated in the major metabolic actions of insulin. Although substantial evidence indicates the importance of IRS-1 in insulin signal transduction, the generation of IRS-1-deficient mice has revealed the existence of redundant signalling pathways.     

Activation of the insulin receptor (IR) by insulin and a synthetic peptide has different effects on gene expression in IR-transfected L6 myoblasts

Single-chain peptides have been recently produced that display either mimetic or antagonistic properties against the insulin and IGF-1 (insulin-like growth factor 1) receptors. We have shown previously that the insulin mimetic peptide S597 leads to significant differences in receptor activation and initiation of downstream signalling cascades despite similar binding affinity and in vivo hypoglycaemic potency. It is still unclear how two ligands can initiate different signalling responses through the IR (insulin receptor). To investigate further how the activation of the IR by insulin and S597 differentially activates post-receptor signalling, we studied the gene expression profile in response to IR activation by either insulin or S597 using microarray technology. We found striking differences between the patterns induced by these two ligands. Most remarkable was that almost half of the genes differentially regulated by insulin and S597 were involved in cell proliferation and growth. Insulin either selectively regulated the expression of these genes or was a more potent regulator. Furthermore, we found that half of the differentially regulated genes interact with the genes involved with the MAPK (mitogen-activated protein kinase) pathway. These findings support our signalling results obtained previously and confirm that the main difference between S597 and insulin stimulation resides in the activation of the MAPK pathway. In conclusion, we show that insulin and S597 acting via the same receptor differentially affect gene expression in cells, resulting in a different mitogenicity of the two ligands, a finding which has critical therapeutic implications.     

Expression of insulin signalling components in the sensory epithelium of the human saccule

Several studies have demonstrated a link between diabetes and the dysfunction of the inner ear. Few studies, however, have reported the signalling mechanisms involved in metabolic control in human inner ear cells. Knowledge of the expression and role of the insulin receptor and downstream signalling components in the inner ear is sparce. Our immunohistochemistry approach has shown that the insulin receptor, insulin receptor substrate 1 (IRS1), protein kinase B (PKB) and insulin-sensitive glucose transporter (GLUT4) are expressed in the sensory epithelium of the human saccule, which also exhibits expression of a calcium-sensitive cAMP/cGMP phosphodiesterase 1C (PDE1C) and the vasopressin type 2 receptor. IRS1 and PDE1C are selectively expressed in sensory epithelial hair cells, whereas the other components are expressed in sensory epithelial supporting cells or in both cell types, as judged from co-expression or non-co-expression with glial fibrillary acidic protein, a marker for supporting cells. Furthermore, IRS1 appears to be localized in association with sensory nerves, whereas GLUT4 is expressed in the peri-nuclear area of stromal cells, as is the case for aquaporin 2. Thus, the insulin receptor, insulin signalling components and selected cAMP signalling components are expressed in the human saccule. In addition to well-known mechanisms of diabetes complications, such as neuropathy and vascular lesions, the expression of these proteins in the saccule could have a role in the observed link between diabetes and balance/hearing disorders.     

G Protein-coupled receptor kinase 2 (GRK2): A novel modulator of insulin resistance

G protein-coupled receptor kinase 2 (GRK2) is emerging as a key, integrative node in many signalling pathways. Besides its canonical role in the modulation of the signalling mediated by many G protein-coupled receptors (GPCR), this protein can display a very complex network of functional interactions with a variety of signal transduction partners, in a stimulus, cell type, or context-specific way. We review herein recent data showing that GRK2 can regulate insulin-triggered transduction cascades at different levels and that this protein plays a relevant role in insulin resistance and obesity in vivo, what uncovers GRK2 as a potential therapeutic target in the treatment of these disorders.     

A 180,000 molecular weight glycoprotein substrate of the insulin receptor tyrosine kinase is present in human placenta and in rat liver, muscle, heart and brain plasma membrane preparations

Cell signalling for insulin may include insulin receptor tyrosine kinase catalysing the phosphorylation of one or more cell proteins. Since temporally the insulin receptor will encounter plasma membrane protein first, we have studied the in vitro phosphorylation of purified plasma membrane preparations. Two proteins were immunoprecipitated with anti-phosphotyrosine antibody from rat liver, muscle, heart and brain membranes and from human placenta membranes: the insulin receptor (detected as a phosphorylated-β-subunit) and a 180,000 molecular weight protein (pp180). pp180 is a monomeric glycoprotein that in the absence of dithiothreitol migrated in denaturing gels like a 150,000 molecular weight protein. pp180 was a substrate for the insulin receptor: (i) receptor and pp180 phosphorylation followed a similar insulin dose-response, although fold-stimulation of autophosphorylation was greater; and (ii) removal of insulin receptors with monoclonal antibodies prevented subsequent pp180 phosphorylation. Insulin-activated receptors increased the extent, but not the rate, of pp180 phosphorylation; the increased phosphate was incorporated into tyrosine and appeared to do so in three or four of pp180's 12 tryptic phosphopeptides. Some data suggest that pp180 is the same protein in each of the tested tissues. The occurrence of pp180, an insulin receptor substrate, in plasma membranes of several insulin responsive tissues suggests that it has a role in insulin signalling.     

Effect of food restriction on the insulin signalling pathway in rat skeletal muscle and adipose tissue

We tried to elucidate the effects of a brief and severe model of food restriction on insulin sensitivity in female rats, focusing on key proteins involved in the insulin signalling pathway in skeletal muscle and adipose tissue after 5, 10 and 15 days of food restriction. Using euglycemic clamp, we detected that food-restricted rats are significantly less sensitive to insulin action than control rats. However, the time of restriction promotes a progressive increase on insulin sensitivity. The analysis of the insulin signalling pathway showed a tissue-specific regulation of several proteins involved in insulin signalling. In skeletal muscle, insulin receptor substrate 1 and Glut4 are up-regulated at the end of the food restriction period, just the opposite of what we found in adipose tissue. In conclusion, a 50% reduction of food intake modulates insulin sensitivity through a tissue-specific regulation of the insulin signalling pathway in the main target tissues for this hormone.     

Improved glucose homeostasis and enhanced insulin signalling in Grb14-deficient mice

Gene targeting was used to characterize the physiological role of growth factor receptor-bound (Grb)14, an adapter-type signalling protein that associates with the insulin receptor (IR). Adult male Grb14(-/-) mice displayed improved glucose tolerance, lower circulating insulin levels, and increased incorporation of glucose into glycogen in the liver and skeletal muscle. In ex vivo studies, insulin-induced 2-deoxyglucose uptake was enhanced in soleus muscle, but not in epididymal adipose tissue. These metabolic effects correlated with tissue-specific alterations in insulin signalling. In the liver, despite lower IR autophosphorylation, enhanced insulin-induced tyrosine phosphorylation of insulin receptor substrate (IRS)-1 and activation of protein kinase B (PKB) was observed. In skeletal muscle, IR tyrosine phosphorylation was normal, but signalling via IRS-1 and PKB was increased. Finally, no effect of Grb14 ablation was observed on insulin signalling in white adipose tissue. These findings demonstrate that Grb14 functions in vivo as a tissue-specific modulator of insulin action, most likely via repression of IR-mediated IRS-1 tyrosine phosphorylation, and highlight this protein as a potential target for therapeutic intervention.     

Harmonic oscillator model of the insulin and IGF1 receptors’ allosteric binding and activation

The insulin and insulin‐like growth factor 1 receptors activate overlapping signalling pathways that are critical for growth, metabolism, survival and longevity. Their mechanism of ligand binding and activation displays complex allosteric properties, which no mathematical model has been able to account for. Modelling these receptors’ binding and activation in terms of interactions between the molecular components is problematical due to many unknown biochemical and structural details. Moreover, substantial combinatorial complexity originating from multivalent ligand binding further complicates the problem. On the basis of the available structural and biochemical information, we develop a physically plausible model of the receptor binding and activation, which is based on the concept of a harmonic oscillator. Modelling a network of interactions among all possible receptor intermediaries arising in the context of the model (35, for the insulin receptor) accurately reproduces for the first time all the kinetic properties of the receptor, and provides unique and robust estimates of the kinetic parameters. The harmonic oscillator model may be adaptable for many other dimeric/dimerizing receptor tyrosine kinases, cytokine receptors and G‐protein‐coupled receptors where ligand crosslinking occurs.     

High-fat diet induced alterations in plasma membrane cholesterol content impairs insulin receptor binding and signalling in mouse liver but is ameliorated by atorvastatin

A high-fat diet (HFD) impairs insulin binding and signalling and may contribute to the development of insulin resistance. In addition, in vitro studies have shown that alterations in plasma membrane cholesterol influence ligand binding and downstream signalling for several receptor-tyrosine kinases (RTKs), including the insulin receptor. Using an ex vivo approach, we explored the effects of a HFD on insulin binding and signalling in mouse liver and relate these to observed changes in plasma membrane cholesterol. Mice fed a HFD demonstrated decreased insulin signalling compared to mice fed a normal chow diet (ND), indicated by a 3-fold decrease in insulin binding (P < 0.001) and a similar decrease in insulin receptor phosphorylation (~2.5-fold; P < 0.0001). Interestingly, we also observed a marked decrease in the cholesterol content of liver plasma membranes in the HFD fed mice (P < 0.0001). These effects of the HFD were found to be ameliorated by atorvastatin treatment (P < 0.0001). However, in ND mice, atorvastatin had no influence on membrane cholesterol content or insulin binding and signalling. The influence of membrane cholesterol on insulin binding and signalling was also corroborated in HepG2 cells. To the best of our knowledge, this is the first demonstration of the effects of a HFD and atorvastatin treatment on changes in plasma membrane cholesterol content and the consequent effects on insulin binding and signalling. Collectively, these findings suggest that changes in membrane cholesterol content could be an important underlying reason for the long-known effects of a HFD on the development of insulin resistance.