INTRODUCTION

Protein phosphatase-2A (PP-2A)¹ is one of four major cytoplasmic serine/threonine phosphatases that are known to play an important role in the regulation of diverse cellular proteins, including metabolic enzymes, ion channels, hormone receptors, kinase cascade, and cell growth (1, 2, 3). Two forms of PP-2A, PP-2A₁ and PP-2A₂, have been found in the cytosol of all tissues examined (4, 5). The holoenzyme is a heterotrimer consisting of a catalytic subunit "C" with an apparent mass = 36 kDa, a 60-kDa regulatory subunit "A," and one of several "B" subunits of 54, 55, 72, and 130 kDa (6). The ABC, AC, and C forms of PP-2A are known to have different substrate specificities based upon in vitro assays with artificial substrates (7, 8). The physiologic importance of these forms and their regulation by hormones and extracellular stimuli has not been established. In vitro studies indicate that the catalytic subunit of PP-2A is regulated by a variety of post-translational modifications, including phosphorylation on tyrosine (9, 10, 11) or threonine residues (12, 13), and methylation on the carboxyl-terminal leucine (14). Phosphorylation of PP-2A on tyrosine or threonine residues inactivates the phosphatase (9, 12, 13), while methylation has been reported to activate the enzyme (14).

Recent studies from this laboratory have shown that insulin acutely activates protein phosphatase-1 and concomitantly inhibits PP-2A activity in the rat skeletal muscle cell line, L6, in a differentiation-dependent manner (15). These results, in conjunction with our recent observation, that TNF--induced insulin resistance is accompanied by an elevation in PP-2A activity (16), suggested that rapid activation and inactivation of PP-1 and PP-2A, respectively, may be an integral part of the insulin-stimulated signal transduction pathway. Therefore, in this study, we have attempted to examine the mechanism of PP-2A inactivation by insulin and its counter-regulation by cAMP agonists. Furthermore, in order to gain insight into the upstream signaling components that may participate in insulin inactivation of PP-2A, the contributions of PI3-kinase and 70-kDa S6 kinase signaling pathways were evaluated with the use of specific inhibitors of these pathways.

The results of this study suggest that insulin rapidly inactivates PP-2A in rat skeletal muscle cells by increasing tyrosine phosphorylation of the catalytic subunit. cAMP agonists counter-regulate the insulin effect by decreasing phosphorylation of PP-2A, thereby resulting in enhanced PP-2A activity. Wortmannin and rapamycin prevent insulin's effect on PP-2A inactivation, suggesting that PI 3-kinase and/or 70-kDa S6 kinase generated signals may participate in insulin inactivation of PP-2A.

EXPERIMENTAL PROCEDURES

Cell culture reagents, fetal bovine serum, phosphorylase kinase, and phosphorylase b were purchased from Life Technologies, Inc. [-³²P]ATP (specific activity > 3000 Ci/mmol), [³²P]orthophosphoric acid, and ¹²⁵I-protein A, were purchased from DuPont NEN. (S_p)-cAMP, wortmannin, and rapamycin were purchased from Biomol Research (Plymouth Meeting, PA). Electrophoresis and protein assay reagents were from Bio-Rad. Okadaic acid was from Moana Bioproducts (Honolulu, Hawaii). Protease inhibitors, sodium orthovanadate, and all other reagents were from Sigma. Porcine insulin was a kind gift from Eli Lilly Co. Antibody to the catalytic subunit of PP-2A was purchased from Upstate Biotechnology, Inc. (Lake Placid, NY).

The spontaneously fusing rat skeletal muscle cell line, L6, was a kind gift from Dr. Amira Klip (The Hospital for Sick Children, Toronto, Canada). Cells were grown and maintained in -minimal essential medium containing 2% fetal bovine serum and 1% antibiotic/antimycotic mixture in an atmosphere of 5% CO₂ at 37 °C as described previously (15). Completely differentiated myotubes were used for all the experiments after 16-18 h of starvation in serum-free Dulbecco's modified Eagle's medium.

Serum-starved myotubes were fed with serum-free medium containing 5 mM glucose and incubated at 37 °C for 1 h before treatment with insulin or (S_p)-cAMP. Identical dishes in triplicate were treated with and without (S_p)-cAMP (10⁴ M) for 0-60 min, followed by 10 nM insulin treatment for 5-10 min. In some experiments, cells were pretreated with freshly prepared orthovanadate (1 mM) for 20 min prior to (S_p)-cAMP or insulin exposure. At the end of the incubation period, the medium was removed and the cells were rinsed three times with ice-cold phosphate-buffered saline followed by the addition of PP-2A extraction buffer.

Control-, insulin-, and (S_p)-cAMP-treated cells were scraped off the dishes with 0.3 ml of phosphatase extraction buffer containing 20 mM imidazole-HCl, 2 mM EDTA, 2 mM EGTA, pH 7.0, with 10 µg/ml each of aprotinin, leupeptin, antipain, soybean trypsin inhibitor, 1 mM benzamidine, and 1 mM PhMeSO₂F. The cells were sonicated for 10 s and centrifuged at 2000 × g for 5 min, and the supernatants were used for the assay of phosphatase activities. The assay was performed in the presence and absence of 2 nM okadaic acid to inhibit PP-2A activity. The conditions of the assay allow measurement of PP-1 and PP-2A activities and not PP-2B and PP-2C enzymes which require divalent ions. Previous studies from this laboratory have shown that okadaic acid at 2 nM concentration inhibits only PP-2A, and the PP-1 activity remaining in the assay was comparable with the activity inhibited by inhibitor 2 (15). PP-2A activity was calculated by subtracting the activity measured in the presence of okadaic acid from the activity measured in the absence of okadaic acid. Purified [³²P]phosphorylase a was used as a substrate (15, 17). ³²P-Labeled phosphorylase a was prepared by reacting [-³²P]ATP with purified phosphorylase kinase and phosphorylase b (18).

Control- and agonist-treated cells were lysed in lysis buffer deprived of phosphatase inhibitors. Cell extracts containing equivalent amounts of protein (50-100 µg) were then incubated 2-4 h at 4 °C in the presence of 2 µg of anti-PP-2A C subunit antibody adsorbed overnight at 4 °C to protein A-Sepharose (19). The pellets were washed four times in lysis buffer. After an additional wash in kinase buffer, pellets were resuspended in 40 µl of reaction buffer. The phosphatase assay was initiated by the addition of 20 µl of ³²P-labeled phosphorylase a. After 10 min at 30 °C, the reaction was stopped with 40 µl of 3 × Laemmli sample buffer. In parallel, spontaneous dephosphorylation of ³²P-labeled phosphorylase a was evaluated by incubating it in the absence of cell extract. Samples were separated by electrophoresis on 7.5% SDS-polyacrylamide gels and then subjected to autoradiography. The specificity of the immunoprecipitation was tested by the addition of 2 nM okadaic acid, a concentration that specifically inhibits PP-2A activity. In all cases, the okadaic acid-resistant phosphatase activity did not exceed 10% of the activity measured in the absence of okadaic acid. After autoradiography, [³²P]phosphorylase dephosphorylation was measured by quantitation of the substrate bands by image analysis. Phosphatase activity was determined as the difference between the initial [³²P]phosphorylase level and the level remaining after the reaction.

L6 cells were serum-starved overnight. Next day, the medium was removed and replaced by 1 ml of phosphate-free Dulbecco's modified Eagle's medium and incubation was continued for 1 h. [³²P]Orthophosphate was added (0.5 mCi/ml), and the cells were incubated for 4 h followed by sodium orthovanadate (1 mM) or (S_p)-cAMP (10⁴ M) for 20 min. Insulin (10 nM) was added, and the incubation was continued for an additional 10 min. The cells were rinsed four times with 1 ml of ice-cold phosphate-buffered saline containing phosphatase and protease inhibitors and harvested in 0.5 ml of lysis buffer containing 20 mM triethanolamine, pH 7.2, 0.5 mM EGTA, 1 mM EDTA, 2 mM sodium vanadate, 100 mM sodium pyrophosphate, 100 mM sodium fluoride, 40 mM -glycerophosphate, 1 mM benzamidine, 0.1 mM phenylmethylsulfonyl fluoride, 10 µg/ml each of leupeptin, aprotinin, antipain, soybean trypsin inhibitor, and pepstatin A, 100 mM NaCl, 1 µM microcystin, and 1% Triton X-100. The cell lysates were centrifuged at 16,000 × g in a microcentrifuge for 10 min to remove cell debris. 400 µg of cell lysate protein was diluted to 1 ml with lysis buffer and precleared by incubation with rat IgG (5 µg/ml, coupled to protein A-Sepharose) at 4 °C for 1 h. The supernatants were immunoprecipitated with PP-2A catalytic subunit antibody (4 µg/sample) for 8 h at 4 °C, followed by treatment with 50 µl of protein A-Sepharose CL6B (50% v/v) for 1 h. In some experiments, the antibody was preincubated with the competing peptide before adding to cell lysates. The pellets were washed four times with 1 ml of lysis buffer and resuspended in 40 µl of 2 × SDS sample buffer. The samples were boiled for 5 min, followed by centrifugation (10,000 × g for 30 s) to pellet down the Sepharose beads. Electrophoresis of the immunoprecipitates was performed in 10% SDS-polyacrylamide gels, followed by autoradiography (20). The protein and phosphotyrosine contents of PP-2A catalytic subunits were determined by immunoprecipitating unlabeled cell lysates with anti-PP-2A C subunit antibody, followed by separation of immunoprecipitated proteins on SDS-PAGE. After transferring proteins to PVDF membranes, the membranes were probed with PP-2A C subunit antibody or phosphotyrosine antibody followed by detection with ¹²⁵I-protein A (0.2 µCi/ml) and autoradiography. The intensity of the signal was quantitated by densitometric analysis of the autoradiograms as well as by the "cut and count" technique.

The protein contents of the cell extracts were determined by the Bradford assay (21) or by using bicinchoninic acid (22).

Student's t test and analysis of variance were used to evaluate the significance of the effects of vanadate, (S_p)-cAMP, and insulin.

RESULTS

As indicated previously (15), acute exposure of L6 myotubes to physiologic concentrations of insulin for 5-10 min resulted in a rapid activation of PP-1 and an inhibition of PP-2A. In this study, we examined the mechanism of inactivation of PP-2A by insulin and the effect of cAMP analogues on insulin-mediated inhibition of PP-2A activity.

The kinetics of PP-2A inhibition by insulin is shown in Fig. 1. Treatment of L6 cells with 10 nM insulin resulted in a 35% reduction in cytosolic PP-2A activity within 10 min when compared with untreated control cells. A maximal reduction of 43% was observed after 20 min of incubation, and the effect was sustained for the entire 60-min period studied. Insulin dose-response studies demonstrated an inhibition of PP-2A activity at all concentrations of insulin tested (Fig. 2). Maximal reduction in enzyme activity was observed at 20 nM insulin, and the effect was sustained up to 100 nM insulin tested.

To further investigate whether PP-2A activity could be differentially regulated by insulin and the other counter-regulatory hormones, L6 cells were exposed to (S_p)-cAMP, a cAMP agonist, for 20 min prior to or after treatment with insulin. The presence of (S_p)-cAMP completely abolished the effect of insulin on PP-2A inhibition (Fig. 3). (S_p)-cAMP alone had very little effect on the basal PP-2A activity. Similar results were obtained with the other cAMP analogues, 8-bromo-cAMP and dibutyryl-cAMP (data not shown).

4

To examine whether (S_p)-cAMP blocked insulin inactivation of PP-2A by activating a phosphatase, thereby decreasing phosphorylation of the catalytic subunit, cells were preincubated for 20 min with 1 mM sodium orthovanadate (an inhibitor of tyrosine and dual specific phosphatases) before exposure to (S_p)-cAMP and insulin. As shown in Fig. 4, pretreatment of cells with sodium orthovanadate prevented the effect of (S_p)-cAMP on PP-2A activity. Direct addition of vanadate to cell extracts did not inhibit PP-2A activity. In contrast, treatment of intact cells with sodium vanadate alone decreased PP-2A when compared with control cells.

4

To directly assess changes in PP-2A activity in response to insulin and (S_p)-cAMP, immunoprecipitation of PP-2A was performed with a peptide antibody directed against the C subunit of human PP-2A (19). The immunoprecipitated PP-2A activity was measured with [³²P]phosphorylase a as a substrate. As indicated in Fig. 5, insulin caused 60% decrease in PP-2A activity when compared with control. The presence of (S_p)-cAMP during insulin exposure blocked insulin effect on PP-2A activity. Pretreatment with vanadate prevented (S_p)-cAMP effect on PP-2A. As indicated in Fig. 4, cells incubated with vanadate alone also showed an inhibition of PP-2A activity when compared with control cells. Addition of vanadate to immunoprecipitated PP-2A did not affect the enzyme activity.

Insulin and (S_p)-cAMP treatment did not change the abundance of PP-2A catalytic subunit (Fig. 6).

The results of Figs. 3, 4, 5 suggested that insulin and (S_p)-cAMP effects on PP-2A inactivation/reactivation may be mediated by phosphorylation/dephosphorylation mechanism. To examine this, we immunoprecipitated cell extracts with a peptide antibody raised against the catalytic subunit of PP-2A. The immunoprecipitates were separated by SDS-PAGE, transferred to PVDF membrane, and probed with an anti-phosphotyrosine antibody (Fig. 7A). Insulin treatment resulted in a >3-fold increase in tyrosine phosphorylation of PP-2A catalytic subunit when compared with control cells (compare Fig. 7A, lane 2 versus lane 1). The phosphotyrosine content of PP-2A in cells treated with (S_p)-cAMP alone (lane 3) was comparable with control cells. In contrast, cells treated with (S_p)-cAMP followed by insulin (lane 4) exhibited a reduction in the phosphotyrosine content of PP-2A. The observed alterations in PP-2A tyrosine phosphorylation seen in insulin- and (S_p)-cAMP-treated cells were not due to variations in the amounts of immunoprecipitated proteins (Fig. 7B).

Next, we examined the phosphorylation status of PP-2A catalytic subunit in ³²P-labeled cells that were exposed to insulin and (S_p)-cAMP. As shown in Fig. 8A, insulin treatment for 2, 5, 10, and 20 min (lanes 2, 5, 6, and 7) resulted in a rapid increase in the phosphorylation of the 36-kDa subunit of PP-2A when compared with control cells (lane 1). Cells exposed to (S_p)-cAMP for 20 min prior to (lane 3) or after 10 min of insulin treatment (lane 4) showed less phosphorylation of the PP-2A catalytic subunit when compared with insulin alone (lanes 5-7). Pretreatment with vanadate before (S_p)-cAMP (lane 9) prevented the (S_p)-cAMP effect on the phosphorylation status of the PP-2A catalytic subunit (compare lane 9 with lanes 3 and 4). Vanadate by itself (lane 8) caused a small increase in phosphorylation of PP-2A when compared with untreated control cells (lane 1). The extent of phosphorylation by (S_p)-cAMP alone was comparable with control cells (data not shown). Western blot analysis of immunoprecipitates from an identical experiment in which cells were incubated with cold P_i revealed similar amounts of PP-2A catalytic subunit in each treatment (Fig. 8B).

To gain insight into the upstream signaling components that might mediate insulin inactivation of PP-2A, we incubated cells with wortmaninn, a potent and selective inhibitor of PI-3 kinase, and rapamycin, an immunosuppressant that selectively blocks the phosphorylation and activation of 70-kDa S6 kinase. Both inhibitors blocked insulin's effect on PP-2A inactivation and restored the enzyme activity to control levels (Table I). Wortmannin alone had very little effect on the basal activity of PP-2A, but completely prevented insulin-mediated inactivation of the enzyme. In contrast, rapamycin alone had a stimulatory effect on PP-2A, and it completely abolished insulin-induced inactivation of PP-2A.

Table I. Wortmaninn and rapamycin prevent insulin-mediated inactivation of PP-2A activity L6 cells were treated with wortmaninn (100 nM) or rapamycin (20 ng/ml) for 20 min followed by insulin (10 nM) for 10 min. PP-2A activity was assayed as described under "Experimental Procedures." Results are the mean ± S.E. of three independent experiments in triplicate. *, p < 0.05 versus control; **, p < 0.05 versus insulin. Treatment PP-2A activity nmol Pi released/mg protein/min None (control) 1.66  ± 0.162 Insulin 0.88  ± 0.100* Rapamycin 1.99  ± 0.230 Rapamycin + insulin 1.87  ± 0.205** Wortmaninn 1.48  ± 0.137 Wortmaninn + insulin 1.42  ± 0.130**

DISCUSSION

This study suggests a possible mechanism for the previously reported inhibition of PP-2A activity by insulin in rat skeletal muscle cells. The insulin-mediated inactivation is accompanied by increased tyrosine phosphorylation of the PP-2A catalytic subunit. (S_p)-cAMP, a cAMP agonist, abrogates insulin inactivation of PP-2A by decreasing phosphorylation of the catalytic subunit and restores PP-2A enzyme activity to control levels. These results provide new evidence that PP-2A activity is regulated in vivo by insulin and counter-regulatory hormones via a complex phosphorylation/dephosphorylation mechanism.

Little is known about the in vivo regulation of PP-2A activity by hormones and growth factors. Studies by Chen et al. (9, 10) using transfected fibroblasts overexpressing the epidermal growth factor receptor, as well as the other in vitro studies (11, 12, 13), have shown that the catalytic subunit of PP-2A is regulated by phosphorylation. Thus, the insulin receptor kinase, the epidermal growth factor receptor kinase, PP60^v-src, and the insulin-stimulated protamine kinase have all been reported to catalyze tyrosine, threonine phosphorylation, and deactivation of purified PP-2A (9, 12, 13). An inactivation of PP-2A, as seen in the present study with insulin, may enhance the transmission of cellular signals through the kinase cascade. In contrast, the prevention of PP-2A inactivation by counter-regulatory hormones may block insulin signaling through the kinase cascade, due to dephosphorylation and inactivation of MAP kinase kinase and MAP kinase activities. In this respect, it is interesting to note that several previous studies have demonstrated an inhibition of insulin and growth factor-activated MAP kinase signaling by cAMP in various cell types including the rat adipose tissue (23, 24, 25, 26, 27). In contrast, cAMP did not block MAP kinase signaling in Swiss 3T3 fibroblasts and PC12 cells (28, 29). When observed, the cAMP inhibition was reported to be downstream of p21^ras activation. However, these studies did not address the role of phosphatases in cAMP-mediated inhibition of MAP kinase signaling. We have recently demonstrated an inhibition of insulin-activated MAP kinase signaling by TNF- in L6 cells (16). The inhibition was downstream of p21^ras and Raf-1. This inhibitory effect of TNF- on MAP kinase cascade was accompanied by elevations in PP-2A activity (16). Pretreatment with low concentrations of okadaic acid (a PP-2A inhibitor) prevented TNF--induced inhibition of MAP kinase signaling and restored insulin's effect on MAP kinase activation (16). (S_p)-cAMP also blocked insulin-mediated PP-1 activation (15) and MAP kinase signaling in L6 cells.² Therefore, PP-2A inactivation/reactivation may play an important role in insulin- and cAMP-mediated signal transduction.

It should be noted, however, that PP-2A activity measurements were performed using the conventional substrate phosphorylase a. This may not be the ideal substrate for PP-2A in skeletal muscle cells. Future studies should be directed toward the identification of in vivo cellular protein substrates that undergo changes in phosphorylation level as a consequence of changes in PP-2A activity.

The prevention of the (S_p)-cAMP effect on PP-2A activity with sodium orthovanadate suggests that the cAMP effect on PP-2A may be mediated either by self-autodephosphorylation of PP-2A as suggested by Chen et al. (9) or via the activation of a tyrosine phosphatase. In support of the latter observation, studies by Brautigan and Pinault (30) indicate that activators of cAMP-dependent protein kinase increase the activity of phosphotyrosine phosphatase-1B (PTPase-1B) in intact cells.

Although we have not directly measured tyrosine phosphatase activities in insulin- and (S_p)-cAMP-treated L6 cells, results with sodium orthovanadate, an inhibitor of tyrosine and dual specificity phosphatases, suggest that cAMP agonists and insulin may regulate PP-2A activity by altering tyrosine phosphorylation of the enzyme. Thus the effect of insulin on PP-2A phosphorylation may be due to inhibition of PTPases or due to an activation of a cytoplasmic tyrosine kinase or both. cAMP agonist may abrogate insulin's effect on PP-2A inactivation by activating a PTPase. An elevation of PTPase activity has been observed in insulin-resistant cells and in animal models of experimental diabetes (31). Studies with transfected cell lines overexpressing different forms of PTPases (e.g. CD45, LAR) demonstrate decreased activation of the insulin receptor tyrosine kinase, Insulin receptor substrate-1, and PI 3-kinase, thereby confirming that PTPases play a key role in the modulation of insulin signaling (32, 33). Further studies are warranted to clarify the exact role and mechanism of action of PTPases in insulin and cAMP signaling. Nonetheless, the results of this study add a new dimension to the well documented antagonism between cAMP and insulin-mediated kinase cascade.

Treatment of intact cells with orthovanadate alone caused a 40% inhibition of PP-2A activity. In contrast, an in vitro addition of vanadate to the immunoprecipitated PP-2A did not alter PP-2A activity. Thus our data lend support to the hypothesis that orthovanadate may act on intact cells indirectly by inhibiting a protein phosphatase that catalyzes the dephosphorylation of an inactivating phosphotyrosine residue of PP-2A (19). Orthovanadate did not further inhibit PP-2A activity when present together with insulin, suggesting that both vanadate and insulin may share a common mechanism(s). However, the effect of insulin on PP-2A phosphorylation appears to be more potent than vanadate, suggesting that vanadate may inhibit PP-2A activity by some other mechanism other than maximizing phosphorylation of PP-2A. Further studies are needed to clarify the exact role of vanadate in the in vivo regulation of PP-2A.

In an attempt to identify the upstream signaling components that may participate in insulin-mediated inactivation of PP-2A, cells were treated with wortmaninn, a potent and selective inhibitor of PI 3-kinase, and rapamycin, an inhibitor of 70-kDa S6 kinase (34). Both agents prevented insulin-mediated PP-2A inactivation and restored PP-2A activity to control values. These results suggests that (i) the insulin receptor may not directly phosphorylate PP-2A, and (ii) wortmaninn-sensitive PI 3-kinase and/or S6 kinase pathway may be participating in insulin signaling leading to PP-2A inactivation. The present studies do not rule out the possibility that insulin inactivation of PP-2A may be mediated through the activation of the recently reported inhibitors of PP-2A.

In summary, the results of the present study indicate that insulin signaling involves a rapid inactivation of PP-2A by increased tyrosine phosphorylation on its catalytic subunit, and counter-regulatory hormones block the insulin effect on PP-2A by decreasing tyrosine phosphorylation.