Hepatic protein phosphotyrosine phosphatase. Dephosphorylation of insulin and epidermal growth factor receptors in normal and alloxan diabetic rats

Effects of adenovirus-mediated liver-selective overexpression of protein tyrosine phosphatase-1b on insulin sensitivity in vivo

AIM

Protein tyrosine phosphatase-1B (PTP-1B) is an intracellular PTP known to dephosphorylate and inactivate upstream tyrosine phosphoproteins in the insulin signalling cascade. We and others reported increased abundance of catalytically impaired PTP-1B in tissue lysates from obese human subjects with and without type 2 diabetes, while genetic knockout of PTP-1B improves insulin sensitivity and prevents nutritionally mediated insulin resistance and obesity. The aim of the present work was to further elucidate the role of PTP-1B in glucose metabolism in vivo.

METHODS

We used adenoviral constructs incorporating cDNAs for either wild-type (W/T) or a catalytically inactive C(215)S (C/S) mutant PTP-1B to achieve liver-selective PTP-1B overexpression in young Sprague-Dawley rats using tail vein injection, based on the high degree of hepatotropism of adenovirus 5 (Ad5). An Ad5-lacZ construct encoding beta-galactosidase was used as a control for viral effects alone. A hyperinsulinaemic euglycaemic clamp was used to study whole body glucose disposal and endogenous glucose production rates.

RESULTS

Control studies in HIRcB cells confirmed catalytic activity and inactivity of W/T and C/S respectively. Mean PTP-1B abundance was 2.24 +/- 0.02- and 2.33 +/- 0.04-fold of saline-treated control in liver lysates of W/T and C/S rats respectively. Liver selective overexpression was confirmed by analysis of tissue lysates from liver, fat and muscle tissues. Ad5 treatment did not result in a statistically or clinically significant liver injury, as determined by serum alanine aminotransferase and histological examination. Seven days post injection, no significant difference in rate of weight gain, fasting blood glucose or insulin levels were seen in any group. Similarly, under steady-state glucose clamp conditions, glucose disposal rate (R(d)), endogenous glucose production rate (EGP) and serum insulin levels were similar in all groups.

CONCLUSION

We conclude that moderate medium-term overabundance, to a degree resembling that seen in insulin-resistant states, of PTP-1B in liver tissue does not alter insulin action on glucose metabolism and that the major site of action of PTP-1B is presumably at insulin-responsive target tissue or tissues other than the liver.

Alterations in skeletal muscle protein-tyrosine phosphatase activity and expression in insulin-resistant human obesity and diabetes

Obese human subjects have increased protein-tyrosine phosphatase (PTPase) activity in adipose tissue that can dephosphorylate and inactivate the insulin receptor kinase. To extend these findings to skeletal muscle, we measured PTPase activity in the skeletal muscle particulate fraction and cytosol from a series of lean controls, insulin-resistant obese (body mass index > 30) nondiabetic subjects, and obese individuals with non-insulin-dependent diabetes. PTPase activities in subcellular fractions from the nondiabetic obese subjects were increased to 140-170% of the level in lean controls (P < 0.05). In contrast, PTPase activity in both fractions from the obese subjects with non-insulin-dependent diabetes was significantly decreased to 39% of the level in controls (P < 0.05). By immunoblot analysis, leukocyte antigen related (LAR) and protein-tyrosine phosphatase 1B had the greatest increase (threefold) in the particulate fraction from obese, nondiabetic subjects, and immunodepletion of this fraction using an affinity-purified antibody directed at the cytoplasmic domain of leukocyte antigen related normalized the PTPase activity when compared to the activity from control subjects. These findings provide further support for negative regulation of insulin action by specific PTPases in the pathogenesis of insulin resistance in human obesity, while other regulatory mechanisms may be operative in the diabetic state.

Decreased skeletal muscle phosphotyrosine phosphatase (PTPase) activity towards insulin receptors in insulin-resistant Zucker rats measured by delayed Europium fluorescence

In order to measure the phosphotyrosine phosphatase (PTPase) activity in small muscle biopsies, a sandwich-immunofluorescence assay was developed using the phosphorylated human insulin receptor as a substrate, a C-terminal insulin receptor antibody as catching antibody and Europium-labelled anti-phosphotyrosine as detecting antibody. Soluble and particulate muscle fractions were prepared from soleus muscle of obese, diabetic (fa/fa) Zucker rats and their lean littermates (Fa/-). In the soluble muscle fractions of the obese (fa/fa) rats PTPase activity was significantly reduced compared to control (Fa/-) rats (45.2 +/- 2.6% vs 61.3 +/- 4.7%, p < 0.02). This reduction was completely prevented by 24 days of metformin treatment which decreased plasma glucose and plasma insulin levels. In particulate muscle fractions, however, no difference in PTPase activity was found among any groups of rats examined. These results show that the alterations in soluble PTPase activity in the insulin-resistant, diabetic Zucker rat vary with the abnormality in glucose homeostasis.

Alterations in specific protein-tyrosine phosphatases accompany insulin resistance of streptozotocin diabetes

To test whether protein tyrosine phosphatases (PTPases) may play a role in the insulin resistance of insulinopenic diabetes, we assessed PTPase activity as well as the protein and mRNA abundance of three major candidate PTPases in subcellular fractions of liver and skeletal muscle of streptozotocin-diabetic rats before and after insulin treatment. PTPase activity against the insulin receptor in liver and muscle cytosol increased to 120-125% of control in the diabetic animals and by an additional 5-10% after insulin treatment. In the particulate fraction, PTPase activity decreased to 65-70% of control in diabetic liver and muscle and increased to 115-120% of control after insulin treatment. Protein for the leukocyte common antigen-related PTPase paralleled the changes in the PTPase activity in the particulate fraction. SH-PTP2/syp and PTPase 1B were both significantly increased in diabetes. SH-PTP2/syp also exhibited an increased ratio of particulate to cytosol distribution in diabetic tissues (1.8-1.9) that was reversed after insulin treatment (0.79-0.95). Northern analysis suggested that the PTPases were regulated at a pretranslational level. These changes in the abundance and distribution of specific PTPases may be involved in the pathogenesis of insulin resistance in insulinopenic diabetes.

Elevated protein tyrosine phosphatase activity and increased membrane viscosity are associated with impaired activation of the insulin receptor kinase in old rats

Insulin resistance is very common in the elderly, and may be associated with glucose intolerance or frank diabetes. In previous studies we demonstrated that insulin resistance in old Wistar rats is associated with decreased autophosphorylation and activation of the hepatic insulin receptor kinase (IRK) in vivo. We now show that this defect can be reproduced in vitro, where the extent of insulin-induced activation of IRK in liver membranes of old rats was decreased by approximately 50% compared with young controls. The defect could be largely abolished after solubilization of the membranes with Triton X-100. We also show that: (a) the viscosity of membranes from the old rats was significantly (P < 0.001, n = 4) higher (by 15%) compared with young controls; (b) incubation of plasma membranes from old animals with lecithin liposomes, which lowered their cholesterol levels, partially abolished the defect in IRK activation; and (c) Triton extracts of liver membranes prepared from old rats did not interfere with the activation of IRK derived from young controls. Additionally, non-membrane components did contribute to the development of this defect. We observed a significant (approximately 30%) (P < 0.001, n = 18) elevation of cytosolic protein tyrosine phosphatase (PTP) activity directed against the beta subunit of the insulin receptor in livers of old rats. No such elevation of PTP activity could be demonstrated with synthetic substrates. Our findings are consistent with a model in which increased membrane viscosity as well as enhancement of a cytosolic PTP activity both markedly inhibit the activation in vivo of the hepatic IRK in old animals.

Insulin receptor dephosphorylation by phosphotyrosine phosphatases obtained from insulin-resistant obese mice

To study the possible involvement of phosphotyrosine phosphatases in insulin resistance, the ability of cytosolic and membrane preparations to dephosphorylate insulin receptors was examined in lean and goldthioglucose-treated insulin-resistant and obese mice. Preparations were obtained from liver, heart, diaphragm and hindleg muscle and their phosphotyrosine phosphatase activities were measured using an immunoenzymatic assay with phosphorylated insulin receptors as substrate. Liver cytosolic and particulate phosphotyrosine phosphatases were more potent than preparations from other tissues and were able to almost completely dephosphorylate the insulin receptor in a dose- and time-dependent manner. No change was observed in cytosolic and membrane-associated phosphotyrosine phosphatases in liver, diaphragm, and heart of obese mice compared with lean mice. In contrast, cytosolic, but not membrane-associated, phosphotyrosine phosphatase activity was decreased in hindleg muscles of obese mice. These results suggest that the regulation of phosphotyrosine phosphatases is tissue-specific. In addition, alterations in total phosphotyrosine phosphatase activity do not appear to play an important role in insulin resistance in all tissues of obese mice, although specific changes cannot be excluded.

Insulin receptor tyrosine kinase domain auto-dephosphorylation

We have observed dephosphorylation of the soluble, 48 kDa insulin receptor tyrosine kinase domain following its tyrosine autophosphorylation. Dephosphorylation was associated with generation of inorganic phosphate, thereby making catalysis by reversal of the kinase reaction unlikely. The kinase domain preparations could not be shown to contain detectable, contaminating protein tyrosine phosphatase activity. In addition, dephosphorylation was insensitive to protein phosphatase inhibitors. However, it was blocked by the kinase inhibitor staurosporine. These results are consistent with insulin receptor kinase domain auto-dephosphorylation via catalysis involving the kinase itself. These findings raise the possibility of a novel mechanism for termination of the insulin receptor signal.

Differential regulation of mRNAs encoding three protein-tyrosine phosphatases by insulin and activation of protein kinase C

Protein-tyrosine phosphatases (PTPases) play an essential role in the control of signalling through phosphotyrosine pathways. Since little is known about the regulation of these enzymes, we examined the effect of insulin and phorbol 12-myristate 13-acetate (PMA) treatment of well-differentiated rat hepatoma (Fao) cells on the expression of mRNAs encoding three major PTPase homologs in liver: PTPase1B, an intracellular enzyme with a single conserved PTPase domain, and two tandem-domain, transmembrane PTPases, known as LAR and LRP. Treatment of serum-deprived cells with 100 nM insulin increased the abundance of the 4.3 kb and 1.6 kb mRNAs encoding PTPase1B on Northern analysis by 1.6 and 3.1-fold, respectively (p < or = 0.02). Similarly, exposure to 100 ng/ml PMA increased the 4.3 and 1.6 kb PTPase1B mRNAs by 4.5 and 5.7-fold, respectively (p < or = 0.035). In contrast, treatment with insulin or PMA had no significant effect of the abundance of mRNA encoding either LAR or LRP. PMA appeared to have a transcriptional effect on the PTPase1B gene by a protein kinase C-mediated mechanism. The increase in PTPase1B mRNA expression by insulin and PMA suggests that this PTPase may provide feed-back regulation of signalling through the insulin action pathway as well as a potential link between the action of protein kinase C and the regulation of specific phosphotyrosine residues in cells.

Differential regulation of multiple hepatic protein tyrosine phosphatases in alloxan diabetic rats

The involvement of tyrosine phosphorylation in insulin action led us to hypothesize that increased activity of protein tyrosine phosphatases (PTPases) might contribute to insulin resistance in alloxan diabetes in the rat. Hepatic PTPase activity was measured using two artificial substrates phosphorylated on tyrosine: reduced, carboxyamidomethylated, and maleylated lysozyme (P-Tyr-RCML) and myelin basic protein (P-Tyr-MBP), as well as an autophosphorylated 48-kD insulin receptor tyrosine kinase domain (P-Tyr-IRKD). Rats that were made alloxan diabetic exhibited a significant increase in hepatic membrane (detergent-soluble) PTPase activity measured with P-Tyr-MBP, without a change in activity measured with P-Tyr-RCML or the P-Tyr-IRKD. The PTPase active with P-Tyr-MBP behaved as a high molecular weight peak during gel filtration chromatography. Characterization of this enzyme indicated it shared properties with CD45, the prototype for a class of transmembrane, receptor-like PTPases. Our results indicate that alloxan diabetes in the rat is associated with an increase in the activity of a large, membrane-associated PTPase which accounts for only a small proportion of insulin receptor tyrosine dephosphorylation. Nonetheless, increased activity of this PTPase may oppose tyrosine kinase-mediated insulin signal transmission, thus contributing to insulin resistance.

Dephosphorylation of human insulin-like growth factor I (IGF-I) receptors by membrane-associated tyrosine phosphatases

The insulin-like growth factor-I (IGF-I) receptor exhibits structural and functional similarities to the insulin receptor. Although the regulation of the insulin-receptor tyrosine kinase has been extensively investigated, the mechanisms involved in phosphorylation/dephosphorylation of the IGF-I receptor have received only little attention. To obtain a better understanding of the mode of IGF-I action, we have investigated the effects of protein phosphotyrosine phosphatases (PTPases) on the phosphorylation status of the IGF-I receptor. The dephosphorylation of the human IGF-I receptor by membrane-associated tyrosine phosphatases was studied by an immuno-enzymic assay based on the recognition of phosphotyrosine residues by anti-phosphotyrosine antibodies. Using intact IGF-I receptors as substrates, we show that they could be completely dephosphorylated by different cellular PTPases. Three pieces of evidence indicate that receptor dephosphorylation takes place on phosphotyrosine, i.e. the inhibition profile of phosphatase activity by zinc and vanadate, its absolute requirement for thiol compounds and the diminution of [32P]phosphotyrosine labelling of the beta subunit assessed by SDS/PAGE and phosphoamino acid analysis. Tyrosine kinase activity and autophosphorylation of the IGF-I receptor were decreased in a dose-dependent manner by PTPases, indicating that partial dephosphorylation of the receptor was associated with a decrease in its intrinsic activity. The sensitivity of the activated human IGF-I receptor to dephosphorylation on tyrosine leads to the speculation that IGF-I receptor activity might be regulated by mechanisms such as those described for the insulin receptor. Further investigation of the pathways of IGF-I receptor dephosphorylation will contribute to define the role(s) of PTPases in the overall mechanism of IGF-I signalling.

Role of cAMP in mediating effects of fasting on dephosphorylation of insulin receptor

We studied the effect of fasting on phosphotyrosine phosphatase (PTPase) activities in particulate (PF) and cytosolic (CF) fractions of rat adipocytes and liver. PTPase activity was assessed using [32P]tyrosine insulin receptor (IR). In adipocytes, 48 h fasting significantly inhibited PTPase activity. Dephosphorylation of IR by PF and CF PTPases was reduced by 80 and 65%, respectively. Similar reductions of lesser magnitude were observed in fasted rat livers. The effect of fasting was completely reversed by either refeeding or by incubating "fasted" adipocytes for 2 h in tissue culture medium containing 5 mM glucose. Neither 20 mM glucose nor the presence of insulin influenced phosphatase activity. Because fasting is accompanied by elevated protein kinase C (PKC) and adenosine 3',5'-cyclic monophosphate (cAMP) levels, we examined their influence on adipocyte PTPases. Neither activation (1 microM 12-O-tetradecanoylphorbol-13-acetate) nor inhibition (20 microM sphingosine) of PKC affected PTPase activity. In contrast, cAMP (2 mM) significantly inhibited PTPase activity (80% inhibition at 2 h), and its effect was prevented by a cAMP antagonist RpcAMP. Fasting- and cAMP-induced inhibition of PTPase activity was restored by incubating PF with trypsin (4 micrograms/ml for 5 min), which separated the putative inhibitors from the phosphatases. We conclude that fasting-induced inhibition of PTPases is mediated by elevated cAMP levels, most likely by activating phosphatase inhibitors.

Hepatic insulin and EGF receptor phosphorylation and dephosphorylation in fetal rats

Hepatic insulin receptor and epidermal growth factor (EGF) receptor phosphorylation and dephosphorylation were studied in normal and growth-retarded fetal rats. Insulin receptor autophosphorylation at a subsaturating ATP concentration (0.5 microM) increased by 10-fold from day 17 to 21 of gestation and decreased by 50% in term growth-retarded fetuses of fasted mothers. In vitro kinase activation at 0.5 mM ATP did not change with gestation or maternal fasting. EGF receptor autophosphorylation increased in parallel with receptor number with advancing gestation and did not change with maternal fasting. Protein tyrosine phosphatases (PTPases), which might attenuate receptor signaling in livers from growth-retarded fetuses, were measured using polybasic and polyacidic artificial substrates as well as the insulin receptor kinase domain. Fetal membrane PTPase activities were twofold higher than in the adult and declined with advancing gestation. However, activities were similar in normal and growth-retarded fetuses. We conclude that decreased hepatic growth in growth-retarded fetuses may involve decreased insulin receptor tyrosine kinase activation in vivo, as indicated by diminished receptor autophosphorylation at subsaturating ATP concentrations. Changes in EGF receptor kinase activity and PTPases could not be implicated based on our in vitro findings.

Protein-tyrosine phosphatases and the regulation of insulin action

Protein-tyrosine phosphatases (PTPases) play an important role in the regulation of insulin action by dephosphorylating the active (autophosphorylated) form of the insulin receptor and attenuating its tyrosine kinase activity. PTPases can also modulate post-receptor signalling by catalyzing the dephosphorylation of cellular substrates of the insulin receptor kinase. Dramatic advances have recently been made in our understanding of PTPases as an extensive family of transmembrane and intracellular proteins that are involved in a number of pathways of cellular signal transduction. Identification of the PTPase(s) which act on various components of the insulin action cascade will not only enhance our understanding of insulin signalling but will also clarify the potential involvement of PTPases in the pathophysiology of insulin-resistant disease states. This brief review provides a summary of reversible tyrosine phosphorylation events in insulin action and available data on candidate PTPases in liver and skeletal muscle that may be involved in the regulation of insulin action.

Two nonradioactive assays for phosphotyrosine phosphatases with activity toward the insulin receptor

Two highly sensitive, nonradiolabeled assays for protein phosphotyrosine phosphatase (PTPase) have been developed. The first assay is based on the use of chemically synthesised phosphotyrosine-containing peptides that can be separated from the dephosphorylated peptide products by HPLC. In this assay, partially purified placental PTPase 1B dephosphorylated three dodecaphosphopeptides (corresponding to insulin receptor autophosphorylation sites at positions PY1146, PY1150, and PY1151) with approximately equal affinity (Km 1.3-2.5 microM), indicating that PTPase 1B shows no distinct preference for the site of dephosphorylation in these peptides. The second assay employs either a phosphopeptide or an autophosphorylated tyrosine kinase domain immobolized on microtiter plate wells. After reaction with PTPase, the remaining unconverted phosphosubstrate is detected in an ELISA using anti-phosphotyrosine antibodies. The latter assay was used to monitor PTPase activity during purification procedures and for characterizing PTPases. Modulation of PTPase activity by orthovanadate, heparin, Zn2+, and EDTA gave similar results in both assays. The immobilized autophosphorylated IR tyrosine kinase domain was a poor substrate for bovine liver alkaline phosphatase and seminal fluid acid phosphatase. The second assay also offers the potential for comparing PTPase activity toward several autophosphorylated tyrosine kinase domains, including those of the insulin, epidermal growth factor, and platelet-derived growth factor receptors.

The 180000 molecular weight plasma membrane insulin receptor substrate is a protein tyrosine phosphatase and is elevated in diabetic plasma membranes

Wheat germ agglutinin-purified non-diabetic and diabetic human placenta membranes were or were not depleted of EGF receptor with monoclonal anti-EGF receptor antibody B1D8, and subsequently phosphorylated. Phosphorylated insulin receptor beta-subunit was lower and pp180 was higher in diabetic placenta membranes than in non-diabetic membranes. Phosphorylated-beta-subunit was also lower in diabetic (streptozotocin-induced) rat liver whereas the amount of pp180 was dependent on membrane protein concentration. When rat liver tyrosine-phosphorylated proteins were incubated 30 min, 4 degrees C with EDTA-terminated 32P-phosphorylation reaction mixtures of wheat germ agglutinin-purified rat liver proteins, less phosphorylated proteins were immunoprecipitated with antiphosphotyrosine. The decrease in tyrosine-phosphorylated products suggested that pp180 was a protein tyrosine phosphatase. Taken together, the results suggest that diabetic plasma membranes contain more tyrosine phosphatase than non-diabetic membranes.

Hepatic protein tyrosine phosphatases in the rat

Regulation of cell growth and metabolism by protein tyrosine phosphorylation involves dephosphorylation via the action of protein tyrosine phosphatases (PTPases). We have characterized the membrane PTPases in rat liver, monitoring their activity by measuring the dephosphorylation of P-Tyr-reduced, carboxyamidomethylated and maleylated lysozyme (P-Tyr-RCML) and P-Tyr-myelin basic protein (P-Tyr-MBP). Separation of membrane PTPases by poly (L-lysine) chromatography yielded three peaks of PTPase, termed I, II and III. PTPases I and II were most active with P-Tyr-RCML, whereas PTPase III showed greater activity with P-Tyr-MBP than with P-Tyr-RCML (ratio of activities 4:1). Separation of membrane proteins by gel-filtration chromatography yielded two peaks of activity. Based on substrate specificity, sensitivity to inhibitors and requirement for thiol-containing compounds, the activity peak with an Mr of approximately 400,000 corresponded to PTPase III, whereas that with an Mr of approx. 40,000 contained PTPases I and II. All three PTPases dephosphorylated epidermal growth factor receptors and insulin receptors, but only PTPases I and II were active with P-Tyr-asialoglycoprotein receptors. Although none of the above characteristics distinguished between PTPases I and II, only PTPase I reacted in a Western immunoblotting procedure with anti-peptide antibodies directed towards human placental PTPase. We conclude that the membrane fraction from rat liver contains at least three distinct PTPases.