Insulin activates the plasma-membrane and dense-vesicle cyclic AMP phosphodiesterase in hepatocytes by distinct routes

Orthovanadate stimulates cAMP phosphodiesterase 3 activity in isolated rat hepatocytes through mitogen-activated protein kinase activation dependent on cAMP-dependent protein kinase

Orthovanadate (vanadate) as well as insulin stimulated phosphodiesterase 3 (PDE3) in the particulate fraction of rat hepatocytes. The vanadate-induced activations of PDE3 and mitogen-activated protein kinase (MAPK) were inhibited by H-89 and PD98059, suggesting that the MAPK activation via cAMP-dependent protein kinase (PKA) and MAPK kinase is involved in the vanadate action. On the other hand, the insulin-induced activations of PDE3 and Akt were inhibited by wortmannin, suggesting involvement of the Akt activation via phosphatidylinositol 3-kinase (PI3K) in the insulin action. The vanadate-induced activations of PKA and PDE3 were inhibited in part by propranolol or genistein, suggesting that vanadate may exert its actions via dual signaling pathways of beta-adrenergic receptors and receptor tyrosine kinases of growth factors. Vanadate, in contrast to insulin, did not promote the phosphorylation of insulin receptor substrate-1. The vanadate-induced increase in the phosphorylation of a main isoform of MAPKs, p44 protein, was detected by immunoblotting migration patterns of SDS-PAGE. A partially purified PDE3 activity was increased by addition of MAPK or Akt to the reaction mixture, suggesting that MAPK as well as Akt acts upstream of PDE3. The activation of PDE3 by insulin was independent of a transient increase in the MAPK activity, probably due to the dephosphorylated inactivation mediated by the induced activation of MAPK phosphatases (MKPs). Vanadate did not affect the MKP activity. These results indicate that vanadate stimulates the particulate PDE3 activity by activating mainly p44 MAPK via a PKA-dependent process, and that it differs from insulin with regard to a phosphorylation cascade of PDE3 activation.

Subcellular localization and hormone sensitivity of adipocyte cyclic AMP phosphodiesterase

Treatment of intact adipocytes with either or both insulin and adrenaline stimulated membrane cyclic AMP phosphodiesterase activity only in the endoplasmic reticulum subfraction. The cyclic GMP-inhibited cyclic AMP phosphodiesterase activity was also found in this fraction. Quantitative Western blotting using a specific polyclonal antibody, raised against the homogeneous 'dense-vesicle' cyclic AMP phosphodiesterase from rat liver, identified a single 63 kDa species which was localized in the adipocyte endoplasmic reticulum fraction. The ability of adrenaline to stimulate adipocyte membrane cyclic AMP phosphodiesterase was shown to be mediated via beta-adrenoceptors and not alpha 1-adrenoceptors. Membrane cyclic AMP phosphodiesterase was stimulated by glucagon but not by vasopressin, A23187 or 12-O-tetradecanoylphorbol 13-acetate (TPA). Treatment of adipocytes with either chloroquine or dansyl cadaverine failed to affect the ability of insulin to stimulate cyclic AMP phosphodiesterase activity. Treatment of an isolated adipocyte endoplasmic reticulum membrane fraction with purified protein kinase A increased its cyclic AMP phosphodiesterase activity some 2-fold. When this fraction was treated with purified protein kinase A and [32P]ATP, label was incorporated into a 63 kDa protein which was specifically immunoprecipitated with the antiserum against the liver 'dense-vesicle' cyclic AMP phosphodiesterase.

Insulin stimulates the tyrosyl phosphorylation and activation of the 52 kDa peripheral plasma-membrane cyclic AMP phosphodiesterase in intact hepatocytes

The 52 kDa subunit of the peripheral-plasma-membrane insulin-stimulated high-affinity cyclic AMP phosphodiesterase can be specifically detected by the antibody PM1 by Western-blotting procedures and also can be immunoprecipitated from a hepatocyte extract. PM1-mediated immunoprecipitation from hepatocyte extracts showed that insulin treatment of intact 32P-labelled hepatocytes caused the rapid phosphorylation of the peripheral-plasma-membrane cyclic AMP phosphodiesterase. Phosphoamino acid analysis and the use of a phosphotyrosine-specific antibody indicated that phosphorylation occurred on tyrosyl residue(s) of this phosphodiesterase. Prior treatment of hepatocytes with glucagon (10 nM) completely blocked the insulin-mediated tyrosyl phosphorylation of this 52 kDa protein, as detected with both the PM1 and the anti-phosphotyrosine antibodies. Treatment of hepatocytes with glucagon alone did not increase the phosphorylation state of the peripheral-plasma-membrane cyclic AMP phosphodiesterase. The specific anti-phosphotyrosine antibody also detected the insulin-stimulated phosphorylation of proteins of 180 kDa, 95 kDa and 39 kDa. Prior treatment of hepatocytes with glucagon decreased the ability of insulin to phosphorylate the 180 kDa and 39 kDa species, but not the 95 kDa species.

Activation and phosphorylation of the 'dense-vesicle' high-affinity cyclic AMP phosphodiesterase by cyclic AMP-dependent protein kinase

Incubation of a hepatocyte particulate fraction with ATP and the isolated catalytic unit of cyclic AMP-dependent protein kinase (A-kinase) selectively activated the high-affinity 'dense-vesicle' cycle AMP phosphodiesterase. Such activation only occurred if the membranes had been pre-treated with Mg2+. Mg2+ pre-treatment appeared to function by stimulating endogenous phosphatases and did not affect phosphodiesterase activity. Using the antiserum DV4, which specifically immunoprecipitated the 51 and 57 kDa components of the 'dense-vesicle' phosphodiesterase from a detergent-solubilized membrane extract, we isolated a 32P-labelled phosphoprotein from 32P-labelled hepatocytes. MgCl2 treatment of such labelled membranes removed 32P from the immunoprecipitated protein. Incubation of the Mg2+-pre-treated membranes with [32P]ATP and A-kinase led to the time-dependent incorporation of label into the 'dense-vesicle' phosphodiesterase, as detected by specific immunoprecipitation with the antiserum DV4. The time-dependences of phosphodiesterase activation and incorporation of label were similar. It is suggested (i) that phosphorylation of the 'dense-vesicle' phosphodiesterase by A-kinase leads to its activation, and that such a process accounts for the ability of glucagon and other hormones, which increase intracellular cyclic AMP concentrations, to activate this enzyme, and (ii) that an as yet unidentified kinase can phosphorylate this enzyme without causing any significant change in enzyme activity but which prevents activation and phosphorylation of the phosphodiesterase by A-kinase.

Insulin control of cyclic AMP phosphodiesterase

Cyclic AMP phosphodiesterase (PDE) is an enzyme involved in cellular homeostasis of cyclic AMP. It exists as multiple isozymes in cells, but only the high affinity, membrane-bound isozyme is sensitive to hormonal modulation. Several isozymes or isoforms of the low Km PDE have been detected. Data suggest that several mechanisms exist for hormonal modulation of PDE. Activity of the low Km PDE species may be modulated by phosphorylation/dephosphorylation, phospholipid substrate concentration, insulin second messenger, cyclic GMP, guanine nucleotide binding proteins, calmodulin, or aggregation/disaggregation of monomeric forms. Modulation of PDE isoforms by different hormones may be through different regulatory components or mechanisms.

An insulin mediator preparation serves to stimulate the cyclic GMP activated cyclic AMP phosphodiesterase rather than other purified insulin-activated cyclic AMP phosphodiesterases

An insulin mediator preparation was obtained from rat hepatocytes which had been treated with insulin. This preparation inhibited adenylate cyclase activity. It stimulated the activity of homogeneous preparations of both the cytosolic and membrane-bound forms of rat liver cyclic GMP-activated cyclic AMP phosphodiesterase. It failed to activate homogeneous preparations of both the peripheral plasma membrane and 'dense-vesicle' cyclic AMP phosphodiesterases. The insulin mediator preparation stimulated cyclic GMP-activated cyclic AMP phosphodiesterase activity in a dose-dependent fashion with a hill coefficient of 0.46. Insulin caused the dose-dependent production of mediator activity in intact hepatocytes with a Ka of 9 pM, although concentrations of insulin greater than 10 nM progressively reduced stimulatory activity.

Specific antibodies and the selective inhibitor ICI 118233 demonstrate that the hormonally stimulated 'dense-vesicle' and peripheral-plasma-membrane cyclic AMP phosphodiesterases display distinct tissue distributions in the rat

Polyclonal-antibody preparations DV1 and PM1, raised against purified preparations of rat liver insulin-stimulated 'dense-vesicle' and peripheral-plasma-membrane cyclic AMP phosphodiesterases, were used to analyse rat liver homogenates by Western-blotting techniques. The antibody DV1 identified only the 63 kDa native subunit of the 'dense-vesicle' enzyme, and the antibody PM1 only the 52 kDa subunit of the plasma-membrane enzyme. These antibodies also detected the subunits of these two enzymes in homogenates of kidney, heart and white adipose tissue from rat. Quantitative immunoblotting demonstrated that the amount of these enzymes (by wt.) varied in these different tissues, as did the expression of these two enzymes, relative to each other, by a factor of as much as 7-fold. The ratio of the dense-vesicle enzyme to the peripheral-plasma-membrane enzyme was lowest in liver and kidney and highest in heart and white adipose tissue. ICI 118233 was shown to inhibit selectively the 'dense-vesicle' cyclic AMP phosphodiesterase in liver. It did this in a competitive fashion, with a Ki value of 3.5 microM. Inhibition of tissue-homogenate cyclic AMP phosphodiesterase activity by ICI 118233 was used as an index of the contribution to activity by the 'dense-vesicle' enzyme. By this method, a tissue distribution of the 'dense-vesicle' enzyme was obtained which was similar to that found by using the immunoblotting technique. The differential expression of isoenzymes of cyclic AMP phosphodiesterase activity in various tissues might reflect a functional adaptation, and may provide the basis for the different physiological actions of compounds which act as selective inhibitors.

The insulin- and glucagon-stimulated 'dense-vesicle' high-affinity cyclic AMP phosphodiesterase from rat liver. Purification, characterization and inhibitor sensitivity

The hormone-stimulated 'dense-vesicle' cyclic AMP phosphodiesterase was solubilized as a proteolytically 'clipped' species, and purified to apparent homogeneity from rat liver with a 2000-3000-fold purification and a 13-18% yield. It appeared to be a dimer (Mr 112,000), of two Mr-57,000 subunits. Solubilization of either a liver or a hepatocyte membrane fraction, with sodium cholate in the presence of the protein inhibitor benzamidine, identified three protein bands which could be immunoprecipitated by a polyclonal antibody raised against the pure enzyme. The major band at Mr 62,000 is suggested to be the native 'dense-vesicle' enzyme, having a Mr-5000 extension which serves to anchor this enzyme to the membrane and which is cleaved off during proteolytic solubilization; the Mr-200,000 band is an aggregate of the Mr-62,000 species, and the Mr-63,000 species is possibly a precursor. The purified 'clipped' enzyme hydrolysed cyclic AMP with kinetics indicative of apparent negative co-operativity, with a Hill coefficient (h) of 0.43 and limiting kinetic constants of Km1 = 0.3 +/- 0.05 microM, Km2 = 29 +/- 6 microM, Vmax.1 = 0.114 +/- 0.015 unit/mg of protein and Vmax.2 = 0.633 +/- 0.054 unit/mg of protein. It hydrolysed cyclic GMP with Michaelis kinetics, Km = 10 +/- 1 microM and Vmax. = 4.1 +/- 0.2 units/mg of protein. Cyclic GMP was a potent inhibitor of cyclic AMP hydrolysis, with an IC50 (concn. giving 50% inhibition) of 0.20 +/- 0.01 microM-cyclic GMP when assayed at 0.1 microM-cyclic AMP. This enzyme was inhibited potently by several drugs known to exert positive inotropic effects on the heart, was extremely thermolabile, with a half-life of 4.5 +/- 0.5 min at 40 degrees C, and was shown to be distinct from the rat liver insulin-stimulated peripheral-plasma-membrane cyclic AMP phosphodiesterase [Marchmont, Ayad & Houslay (1981) Biochem. J. 195, 645-652].

The action of islet activating protein (pertussis toxin) on insulin's ability to inhibit adenylate cyclase and activate cyclic AMP phosphodiesterases in hepatocytes

Treatment of hepatocytes with islet activating protein (pertussis toxin) from Bordetella pertussis blocked the ability of insulin to inhibit adenylate cyclase activity both in broken plasma membranes and in intact hepatocytes. Such treatment of intact hepatocytes with pertussis toxin did not prevent insulin from activating the peripheral plasma membrane cyclic AMP phosphodiesterase although it did inhibit the ability of insulin to activate the 'dense-vesicle' cyclic AMP phosphodiesterase. The ability of glucagon pretreatment of hepatocytes to block insulin's activation of the plasma membrane cyclic AMP phosphodiesterase was abolished in pertussis toxin-treated hepatocytes. It is suggested that the ability of insulin to manipulate cyclic AMP concentrations by inhibiting adenylate cyclase and activating the plasma membrane and 'dense-vesicle' cyclic AMP phosphodiesterases involves interactions with the guanine nucleotide regulatory protein system occurring in liver plasma membranes.

The phorbol ester TPA prevents the expression of both glucagon desensitisation and the glucagon-mediated block of insulin stimulation of the peripheral plasma membrane cyclic AMP phosphodiesterase in rat hepatocytes

The phorbol ester TPA (12-O-tetradecanoyl phorbol-13-acetate) causes a dose-dependent inhibition of the glucagon-stimulated adenylate cyclase activity expressed in plasma membranes isolated from TPA-treated hepatocytes. However, no observable inhibitory effect of TPA on adenylate cyclase activity was observed in cells which had been exposed to glucagon for 5 min, prior to isolation, to desensitise adenylate cyclase. The degree of inhibition of adenylate cyclase elicited by both glucagon desensitisation and TPA treatment of hepatocytes was identical. Pre-treatment of hepatocytes with TPA was also found to prevent glucagon from blocking insulin's activation of the peripheral plasma membrane cyclic AMP phosphodiesterase in intact hepatocytes. TPA treatment also inhibited the ability of cholera toxin to activate the peripheral cyclic AMP phosphodiesterase in intact hepatocytes. It is suggested that in these particular instances TPA and glucagon elicit mutually exclusive processes rather than TPA mimicking glucagon desensitisation per se.

An assessment of the ability of insulin-stimulated cyclic AMP phosphodiesterase to decrease hepatocyte intracellular cyclic AMP concentrations

Treatment of hepatocytes with either NH4Cl (10mM) or fructose (10mM) blocks insulin's activation of the 'dense-vesicle' cyclic AMP phosphodiesterase. The ability of insulin (10 nM) to decrease intracellular cyclic AMP concentrations raised by glucagon (10 nM) was unaffected by pre-treatment with either NH4Cl (10 mM) or fructose (10 mM). It is concluded that the 'dense-vesicle' enzyme does not play a significant role in this action of insulin and that as yet unidentified cyclic AMP phosphodiesterase(s) must be activated by insulin. Treatment of hepatocytes with either NH4Cl or fructose appeared to increase, reversibly, cyclic AMP phosphodiesterase activity. When N6-(phenylisopropyl)adenosine was used to prevent glucagon from blocking insulin's activation of the plasma-membrane cyclic AMP phosphodiesterase activity, insulin's ability to decrease intracellular cyclic AMP concentrations in glucagon-treated hepatocytes was increased markedly. Insulin's activation of the plasma-membrane cyclic AMP phosphodiesterase activity can exert a potent effect in decreasing intracellular cyclic AMP concentrations elevated by glucagon.