Sam68 interacts with IRS1.

Sam68 (Src associated in mitosis) is a RNA binding protein that links cellular signaling to RNA processing. In previous studies we found that insulin promotes Sam68 relocalization in the cytoplasm allowing Sam68 to associate with p85PI3K, Grb2, GAP and probably the insulin receptor (IR), modulating insulin action positively. In the present work, we wanted to define the role of Sam68 in the first stages of IR signaling. Both BRET and co-immunoprecipitation assays have been used for the study of Sam68 binding to IR, IRS1 and p85-PI3K. BRET saturation experiments indicated, for the first time, that Sam68 associates with IRS1 in basal condition. To map the region of Sam68 implicated in the interaction with IRS1, different Sam68 mutants deleted in the proline-rich domains were used. The deletion of P0, P1 and P2 proline rich domains in N-terminus as well as P4 and P5 in C-terminus of Sam68 increased BRET(50), thus indicating that the affinity of Sam68 for IRS1 is lower when these domains are missing. Moreover, in IR-transfected HEK-293 cells, BRET saturation experiment indicated that insulin increases the affinity between Sam68-Rluc and IRS1-YFP. In conclusion, our data indicate that Sam68 interacts with IRS-1 in basal conditions, and insulin increases the affinity between these two partners.

O-GlcNAc modification, insulin signaling and diabetic complications.

O-GlcNAc glycosylation (O-GlcNAcylation) corresponds to the addition of N-acetylglucosamine on serine and threonine residues of cytosolic and nuclear proteins. O-GlcNAcylation is a dynamic post-translational modification, analogous to phosphorylation, that regulates the stability, the activity or the subcellular localisation of target proteins. This reversible modification depends on the availability of glucose and therefore constitutes a powerful mechanism by which cellular activities are regulated according to the nutritional environment of the cell. O-GlcNAcylation has been implicated in important human pathologies including Alzheimer disease and type-2 diabetes. Only two enzymes, OGT and O-GlcNAcase, control the O-GlcNAc level on proteins. Therefore, O-GlcNAcylations cannot organize in signaling cascades as observed for phosphorylations. O-GlcNAcylations should rather be considered as a "rheostat" that controls the intensity of the signals traveling through different pathways according to the nutritional status of the cell. Thus, OGT attenuates insulin signal by O-GlcNAcylation of proteins involved in proximal and distal steps in the PI-3 kinase signaling pathway. This negative feedback may be exacerbated when cells are chronically exposed to elevated glucose concentrations and could thereby contribute to alterations in insulin signaling observed in diabetic patients. O-GlcNAcylation also appears to contribute to the deleterious effects of hyperglycaemia on excessive glucose production by the liver and deterioration of ß-cell pancreatic function, resulting in worsening of hyperglycaemia (glucotoxicity). Moreover, O-GlcNAcylations directly participate in several diabetic complications. O-GlcNAcylation of eNOS in endothelial cells have been involved in micro- and macrovascular complications. In addition, O-GlcNAcylations activate the expression of profibrotic and antifibrinolytic factors, contributing to vascular and renal dysfunctions.

Interaction of PTPB with the insulin receptor precursor during its biosynthesis in the endoplasmic reticulum.

PTP1B is a protein tyrosine-phosphatase predominantly located on the cystosolic surface of the endoplasmic reticulum. This tyrosine-phosphatase plays a major role in the regulation of the activity of the insulin receptor (IR). We have studied the interaction of the IR with PTP1B in living cells using bioluminescence resonance energy transfer (BRET). The IR was fused to Renilla luciferase and a substrate-trapping mutant of PTP1B was fused to the yellow variant of the green fluorescent protein (YFP). When the two partners interacted, an energy transfer occurred between the luciferase and the YFP, and a fluorescent signal, emitted by the YFP, could be detected. The interaction of the IR with PTP1B could be monitored in real time for more than 30 min. Insulin rapidly and dose-dependently stimulated this interaction. The basal (insulin-independent) interaction of IR with PTP1B was much lower with a soluble form than with the endoplasmic reticulum-targeted form of PTP1B, indicating that this basal interaction mainly occurred in the endoplasmic reticulum. In the basal state, PTP1B and the IR indeed co-localized in the endoplasmic reticulum, as demonstrated by confocal microscopy and cell fractionation experiments. Moreover, inhibition of IR processing with tunicamycin indicated that the basal interaction of PTP1B with IR occurred during biosynthesis of the IR precursor in the endoplasmic reticulum. These results strongly suggest that PTP1B not only dephosphorylates the insulin receptor that has been activated by insulin, but also regulates the insulin receptor precursor during its biosynthesis. Localisation of PTP1B to the endoplasmic reticulum may be important to prevent insulin-independent autonomous activity of the immature insulin receptor precursor.

The formation of an intrachain disulfide bond in the leptin protein is necessary for efficient leptin secretion.

Leptin is a cytokine secreted by the adipose tissue that is involved in the control of body weight. We previously showed that a point mutation (R105W) in leptin results in leptin deficiency, marked obesity and hypogonadism in humans adults. Expression in COS1 cells showed impaired secretion and intracellular accumulation of the mutated protein. However, impaired secretion of the mutant leptin had not been demonstrated in adipose cells. In this work, we demonstrate that secretion of R105W mutant is impaired in rat and human adipocytes. We also show that R105W mutant expressed in COS1 cells and in PAZ6 adipocytes forms large molecular aggregates that cannot cross a filtration membrane with a cut-off of 100 kDa. Moreover, we have engineered, by site directed mutagenesis, the cDNAs coding for leptin in which either Cys 117, Cys 167, or both, were replaced by a serine. When expressed in COS1 cells or PAZ6 adipocytes, cysteine mutants also show impaired secretion and formation of large molecular aggregates. Therefore, our work indicates that the formation of an intramolecular disulfide bridge is necessary for normal processing and secretion of leptin. Moreover, the similarity of the behavior of R105W mutant and cystein mutants suggests that the lack of secretion observed with the naturally occurring mutant could result from impaired disulfide bond formation.

Looking for an insulin pill? Use the BRET methodology!

Insulin exerts its biological effects through a plasma membrane receptor that possesses a tyrosine-kinase activity. This tyrosine-kinase activity depends on the autophosphorylation of the receptor on tyrosine residues and on its dephosphorylation by protein tyrosine-phosphatases. The discovery of pharmacological agents that specifically stimulate the autophosphorylation of the insulin receptor or inhibit its dephosphorylation will be of great importance for the treatment of insulin resistant or insulin deficient patients. Bioluminescence Resonance Energy Transfer (BRET) has developed in recent years as a new technique to study protein-protein interactions. In the BRET technique, one partner is fused to Renilla luciferase, whereas the other partner is fused to a fluorescent protein (e.g. YFP, Yellow Fluorescent Protein). The luciferase is excited by addition of its substrate, coelenterazine. If the two partners interact, resonance energy transfer occurs between the luciferase and the YFP, and a fluorescent signal, emitted by the YFP, can be detected. Our work indicates that this methodology could be an important tool for the search of molecules that activate insulin receptor autophosphorylation or that inhibit its dephosphorylation. Indeed, we first showed that the activation of the insulin receptor by different ligands can be monitored using a chimeric receptor with one B-subunit fused to Renilla luciferase and the other B-subunit fused to YFP. The conformational changes induced by different ligands could be detected as an energy transfer (BRET signal) between the luciferase and the YFP, that reflects the activation state of the receptor. This methodology allows for rapid analysis of the effects of agonists on insulin receptor activity and may therefore be used in high-throughput screening for the discovery of molecules with insulin-like properties. More recently, we demonstrated that the BRET methodology could also be used to monitor the interaction of the insulin receptor with protein tyrosine-phosphatase 1B, one of the main tyrosine-phosphatase that controls its activity. HEK cells were co-transfected with the insulin receptor fused to Renilla luciferase and a substrate-trapping mutant of PTP1B (PTP1B-D181A) fused to YFP. Insulin-induced BRET signal could be followed in real time for more than 30 min. Therefore, this methodology can also be used in high-throughput screening for the search of molecules that will specifically disrupt the interaction between the insulin receptor and PTP1B.

Hypoxia increases leptin expression in human PAZ6 adipose cells.

AIMS/HYPOTHESIS: Leptin, an adipose tissue-derived cytokine involved in the control of body weight, also participates in a variety of biological functions, including angiogenesis. Because reduced oxygen availability is a major inducer of angiogenesis, we hypothesized that low cellular oxygen tension could regulate leptin expression in adipose cells. METHODS: Differentiated PAZ6 adipocytes were cultured for 48 h in the presence of chemical inducers of cellular hypoxia (cobalt chloride or desferrioxamine) or in an atmosphere containing only 6% oxygen. The effect of hypoxia on the expression of leptin and several adipose genes was assessed by semi-quantitative RT-PCR. The effect of hypoxia on leptin promoter activity was tested in PAZ6 cells transiently transfected with a luciferase reporter construct, containing 1.87 kb of the human leptin promoter. Leptin secretion in the culture medium was determined by radioimmunoassay. RESULTS: Hypoxia increased leptin mRNA expression, leptin promoter activity and leptin secretion in the culture medium by two- to threefold ( p<0.05). The expression of the glucose transporter isoform 1 (GLUT-1) mRNA, a well known hypoxia inducible gene, was also increased. In contrast, glucose transporter isoform 4 (GLUT-4), hormone sensitive lipase (HSL), fatty acid binding protein (aP2) and uncoupling protein 2 (UCP2) mRNAs were markedly reduced by hypoxia. In addition, a similar hypoxia-induced increase in leptin mRNA and secretion was observed in primary rat adipose cells. CONCLUSION/INTERPRETATION: Hypoxia markedly and specifically increased leptin gene expression through activation of the leptin gene promoter, and this resulted in an increased leptin production by human PAZ6 adipocytes.

Monitoring the activation state of the insulin receptor using bioluminescence resonance energy transfer.

We have developed a procedure based on bioluminescence resonance energy transfer (BRET) to monitor the activation state of the insulin receptor in vitro. Human insulin receptor cDNA was fused to either Renilla luciferase (Rluc) or enhanced yellow fluorescent protein (EYFP) coding sequences. Fusion insulin receptors were partially purified by wheat-germ lectin chromatography from human embryonic kidney 293 cells cotransfected with these constructs. The conformational change induced by insulin on its receptor could be detected as an energy transfer (BRET signal) between Rluc and EYFP. BRET signal parallels insulin-induced autophosphorylation of the fusion receptor. Dose-dependent effects of insulin, insulin-like growth factor 1, and epidermal growth factor on BRET signal are in agreement with known pharmacological properties of these ligands. Moreover, antibodies that activate or inhibit the autophosphorylation of the receptor have similar effects on BRET signal. This method allows for rapid analysis of the effects of agonists on insulin receptor activity and could therefore be used in a high-throughput screening test for discovery of molecules with insulin-like properties.

Effect of dexamethasone on adipocyte differentiation markers and tumour necrosis factor-alpha expression in human PAZ6 cells.

AIMS/HYPOTHESIS: Adipose tissue-derived tumour necrosis factor-alpha (TNF-alpha) has been implicated in the insulin resistance observed in animal models of obesity. Moreover, TNF-alpha has inhibitory effects on adipocyte differentiation. Glucocorticoids play important roles in the regulation of insulin sensitivity and adipose tissue distribution. We therefore studied the effect of dexamethasone on TNF-alpha expression and adipocyte differentiation in human PAZ6 cells. METHODS: The expression of TNF-alpha and adipocyte differentiation markers was assessed by reverse-transcription polymerase chain reaction in PAZ6 cells. RESULTS: In cells cultured for 15 days in the presence of dexamethasone, adipocyte differentiation marker expression was higher and TNF-alpha expression was lower than in cells cultured in the absence of dexamethasone. The presence of dexamethasone was necessary during the whole period of differentiation because removal of dexamethasone during the second week resulted in poorly differentiated adipocytes that express higher levels of TNF-alpha. Dexamethasone also reduced TNF-alpha expression during early stages of differentiation. The use of a TNF-alpha-neutralising antibody showed, however, that endogenously-produced TNF-alpha did not play an important part in the control of PAZ6 cell differentiation. During early stages of adipocyte differentiation, dexamethasone induced the expression of the transcription factors PPAR gamma (peroxisome proliferator activated receptor gamma) and C/EBP alpha (CCAAT/enhancer binding protein alpha) while inhibiting the expression of the inhibitor of DNA binding Id2. CONCLUSION/INTERPRETATION: The effect of dexamethasone on human adipocyte differentiation is not mediated by reduction of TNF-alpha expression but more likely by regulation of the expression of nuclear factors such as PPAR gamma, CEBP alpha and Id2.

Effect of thiazolidinediones on expression of UCP2 and adipocyte markers in human PAZ6 adipocytes.

AIMS/HYPOTHESIS: Thiazolidinediones, a new class of insulin sensitizers, up-regulate the expression of uncoupling protein 2 in rodent adipocytes. It is not known, however, whether thiazolidinediones influence uncoupling protein 2 expression in human adipocytes. We therefore investigated the effect of these drugs on uncoupling protein 2 expression in the recently immortalized human PAZ6 adipocyte cell line. METHODS: Immortalized human PAZ6 preadipocytes were differentiated into adipocytes in the presence or absence of thiazolidinediones. The effect of the drugs on uncoupling protein 2 expression and adipocyte differentiation was measured by reverse transcription-polymerase chain reaction of mRNA of uncoupling protein 2 and of five adipocyte differentiation markers. RESULTS: When cells were differentiated 15 days in the presence of thiazolidinediones, uncoupling protein 2 expression was 2.1-fold higher than in the absence of the drugs. The expression of five adipocyte differentiation markers was, however, also increased by thiazolidinediones. Short-term incubation for 4 and 24 h with thiazolidinediones increased uncoupling protein 2 expression 1.35-fold and 2.3-fold, respectively. The expression of adipocyte markers studied in parallel was also augmented. CONCLUSION/INTERPRETATION: Thiazolidinediones rapidly increase the expression of uncoupling protein 2 in human PAZ6 adipocytes but the increase of uncoupling protein 2 expression is always associated with an augmentation of the expression of all adipocyte markers studied in parallel. This indicates that the effect of thiazolidinediones on uncoupling protein 2 mRNA reflects a general increase in adipocyte differentiation rather than a specific augmentation of uncoupling protein 2 gene expression.

Stimulation of the extracellular signal-regulated kinase 1/2 pathway by human beta-3 adrenergic receptor: new pharmacological profile and mechanism of activation.

We present evidence that stimulation of the human beta-3 adrenergic receptor (AR), expressed in Chinese hamster ovary/K1 cells, specifically activates the mitogen-activated protein kinases extracellular signal-regulated kinase (ERK)1 and 2, but not JNK or p38. The extent and kinetics of the ERK stimulation by the beta-3 AR are identical with those of the endogenic insulin receptor. However, insulin augments cellular proliferation, whereas beta-3 AR agonists inhibit proliferation due to the production of cyclic AMP. The pharmacological profile of the ERK activation by the beta-3 AR differs significantly from its activation of adenylyl cyclase. The order of potency and intrinsic activities of both natural ligands, norepinephrine and epinephrine, is inversed between both signaling pathways. In addition, BRL 37344 and propranolol, ligands that act as agonists in the stimulation of cyclase, act as antagonists for ERK activation. The activation of ERK1/2 is sensitive to pertussis toxin, suggesting that the beta-3 AR, in addition to its interaction with Gs, can couple to Gi/o. Furthermore, the activation of ERK by the beta-3 AR is sensitive to PD98059, wortmannin, and LY294002, indicating a crucial role for mitogen-activated protein kinase kinase and phosphatidylinositol-3 kinase (PI3K), respectively. A beta-3 AR-mediated stimulation of PI3K is confirmed by the observation that the selective agonist CGP 12177A specifically activates protein kinase B. As was observed for the activation of ERK, the activation of protein kinase B is inhibited by preincubation with pertussis toxin and PI3K inhibitors, suggesting that both are a consequence of a Gi/o-mediated activation of PI3K.

Fibroblast growth factor-2 has opposite effects on human breast cancer MCF-7 cell growth depending on the activation level of the mitogen-activated protein kinase pathway.

In human breast cancer MCF-7 and MCF-7ras cells, we demonstrated that whereas insulin had a mitogenic effect on both cell lines, fibroblast growth factor-2 (FGF-2) had opposite effects, stimulating MCF-7 and inhibiting MCF-7ras cell proliferation. The inhibitory signal induced by FGF-2 was related to sustained mitogen-activated protein kinase (MAPK) activation in MCF-7ras cells, while transient MAPK activation was associated with MCF-7 cell proliferation. FGF-2 was further used in combination with insulin or cAMP. In MCF-7 cells, insulin and cAMP reversed the mitogenic effect of FGF-2. In MCF-7ras cells, insulin did not modify the inhibitory effect of FGF-2, but cAMP markedly enhanced it. These effects were also associated with an increased level and duration of MAPK activation. PD98056 abolished the effect of FGF-2 on DNA synthesis in both cell lines, demonstrating that the dual effect of FGF-2 on cell proliferation is dependent on the activity of the extracellular-signal-regulated kinase 1 and 2 (ERK1/2) signalling pathway.

Rat uncoupling protein 2 (UCP2): expression in obese ventromedial hypothalamus (VMH)-lesioned animals.

BACKGROUND: The family of uncoupling proteins is thought to play an important role in the regulation of energy metabolism by uncoupling the respiratory chain reactions from ATP synthesis. The recently discovered uncoupling protein 2 (UCP2) is upregulated in genetically obese rodent models and during long term high fat feeding. AIM: We have examined the UCP2 mRNA levels in liver, heart and white adipose tissue (WAT) of obese ventromedial hypothalamus (VMH)-lesioned rats, during the dynamic and the early stage of the static phase of obesity, before the appearance of most of the metabolic perturbations associated with long term established obesity. RESULTS: The amount of UCP2 mRNA was not increased in any tissue of VMH-lesioned rats relative to control animals during the dynamic phase nor during the early static phase of obesity. CONCLUSION: These results indicate that in the rat, obesity does not necessarily lead to an increase in UCP2 expression and suggest that the up-regulation of UCP2 described in other models may be secondary to metabolic perturbations, rather than to a direct adaptative response to the increased adipose tissue content of the organism.

Desensitization of the beta-adrenergic response in human brown adipocytes.

Activation of adenylyl cyclase by beta-adrenergic receptors (betaARs) plays a major role in adipose tissue homeostasis. The increase in cAMP promotes lipolysis in white adipose tissue, activates both thermogenesis and lipolysis in brown adipose tissue (BAT), and induces BAT hypertrophy. Previous studies indicated that among the three betaAR subtypes present in adipose tissue, beta3AR could be a potential target for antiobesity treatments in humans. We studied immortalized human brown adipocytes (PAZ6 adipocytes) as a model of beta-adrenergic response in human BAT. PAZ6 adipocytes and freshly isolated mature human brown adipocytes display the same proportions of betaAR subtypes, with beta3AR being the most abundant (approximately 80% of the total). However, beta3AR was poorly coupled to the adenylyl cyclase pathway in PAZ6 cells, contributing to only 10% of the isoproterenol-induced accumulation of cAMP, whereas 20% and 70% of the signal depended on beta1- and beta2-subtypes, respectively. Upon isoproterenol stimulation, beta1- and beta2AR down-regulated with a half-life of about 3 h and the beta3AR with a half-life of 30-40 h. Long term stimulation with both saturating (micromolar) and nonsaturating (nanomolar) concentrations of beta-adrenergic agonists caused a complete desensitization of the beta-adrenergic response at the adenylyl cyclase level and loss of stimulated protein kinase A activity and CREB phosphorylation. These results suggest that cAMP-dependent processes will be desensitized upon permanent treatment with beta3AR agonists. Further studies should establish whether the beta3AR is coupled to other signaling pathways in human brown adipocytes and whether these may contribute to BAT hypertrophy and/or thermogenesis.

Molecular basis of insulin action.

Considerable progress has been made in our understanding of the molecular mechanisms of insulin action. The insulin receptor is a membrane receptor possessing tyrosine kinase activity. The binding of insulin to its receptor induces autophosphorylation of the receptor on tyrosine residues and thereby stimulates its tyrosine kinase activity towards intracellular substrates such as Shc or IRS1. This tyrosine kinase activity, which plays a crucial role in the transmission of the signal, is decreased in several insulin-resistance situations. This decrease was initially attributed to the phosphorylation of the receptor on serine or threonine residues, but this mechanism is now seriously questioned. Tyrosine phosphorylation of IRSs and Shc by the insulin receptor permits the activation of two major signalling pathways, the MAP kinase pathway and the Pl 3-kinase pathway. MAP kinases are involved in proliferation and differentiation processes, in particular by regulating the transcriptional activity of the nucleus. The MAP kinase pathway does not appear to play a significant role in the transmission of the metabolic effects of insulin. In contrast, the Pl 3-kinase pathway is involved in several of the metabolic effects of the hormone, such as glucose transport, glycolysis and glycogen synthesis. The Pl 3-kinase pathway also plays a crucial role in the regulation of protein synthesis by insulin. Moreover, this pathway is involved in cell growth and transmits a strong anti-apoptotic signal.

Increased mitogen-activated protein kinase expression and activity in white adipose tissue of ventromedial hypothalamus-lesioned rats.

Ventromedial hypothalamus lesions in rats induce hyperphagia and hyperinsulinaemia associated with a rapid growth of white adipose tissue resulting in massive obesity. It has been shown previously that at an early stage after the lesion, during the dynamic phase of obesity, the white adipose tissue is hyper-responsive to insulin. In the present work, we show that the effects of insulin on the autophosphorylation of the insulin receptor and on its tyrosine-kinase activity towards endogenous substrates are similar in intact adipocytes of control and ventromedial hypothalamus lesioned rats. One week after the lesion, the expression of phosphatidylinisitol 3-kinase and RAF-1 kinase, evaluated by Western-blotting, was similar in control and ventromedial hypothalamus lesioned rats. In contrast, an important increase in the expression of extracellular signal regulated kinase 1 protein was observed in white adipose tissue of ventromedial hypothalamus lesioned compared to control animals. No difference in the expression of extracellular signal regulated kinase 1 mRNA was observed in adipose tissue of control and ventromedial hypothalamus lesioned rats, suggesting that a posttranscriptional mechanism is involved in the over-expression of extracellular signal regulated kinase 1. The kinase activity of extracellular signal regulated kinase 1 and 2 is also markedly increased in adipocytes of ventromedial hypothalamus lesioned compared to control rats, both in the basal state and after insulin stimulation. Six weeks after the ventromedial hypothalamus lesion, this increase in mitogen-activated protein kinase expression and activity was still observed in adipocytes of ventromedial hypothalamus lesioned rats. These results suggest that an early and sustained increase in the expression and activity of mitogen-activated protein kinase may participate in the development of white adipose tissue in ventromedial hypothalamus lesioned rats.

Human immortalized brown adipocytes express functional beta3-adrenoceptor coupled to lipolysis.

Human brown pre-adipocytes were immortalized by microinjection of the genes encoding simian virus 40 T and t antigens under the control of the human vimentin promotor. The transfected pre-adipocytes were cultured for several months with no loss of their morphological characteristics. These cells accumulate lipids and differentiate into adipocytes when treated with insulin, triiodothyronine and dexamethazone. The mRNA of various adipocyte markers was detected by reverse transcriptase-polymerase chain reaction analysis, including hormone-sensitive lipase, lipoprotein lipase, adipsin, glucose transporters 1 and 4, the uncoupling protein (specific of brown adipocytes), and leptin, the product of the ob gene. Pharmacological analyses indicated that the beta3-adrenoceptor is the predominant beta-adrenoceptor subtype in PAZ6 cells and that this receptor subtype is functionally coupled to adenylate cyclase and lipolysis. The immortalization of human adipocytes will permit pharmacological analysis of the human beta3-adrenoceptor function in adipose cells and will allow detailed studies of human adipocyte differentiation.