Insulin Stimulation of Heart Glycogen Synthase D Phosphatase (Protein Phosphatase)

Insulin rapidly produced an increase in per cent of total heart glycogen synthase in the I form in fed rats. In fasted rats the response was diminished and delayed. In diabetic animals there was no response over the 15-min time period studied. Since synthase phosphatase activity is necessary for synthase D to I conversion, the phosphatase activity was determined in extracts from these groups of animals. In the fasted and diabetic rats phosphatase activity was less than one-half of that in fed animals. Administration of insulin to fasting animals increased synthase phosphatase activity to a level approaching that of fed animals by 15 min. In diabetic animals insulin also stimulated an increase in synthase phosphatase activity but 30 min were required for full activation. Insulin had no effect in normal fed animals. Insulin activation of synthase phosphatase activity in heart extracts from fasted animals was still present after Sephadex G-25 chromatography and ammonium sulfate precipitation. Thus insulin had induced a stable modification of the phosphatase itself or of its substrate synthase D rendering the latter a more favorable substrate for the reaction. A difference in sensitivity of the reaction to glycogen inhibition was present between fed and fasted animals. Increasing concentrations of glycogen had only a slight inhibitory effect in extracts from fed animals but considerably reduced activity in extracts from fasted animals. Insulin administration reduced the sensitivity of the phosphatase reaction to glycogen inhibition. This could explain, at least in part, the increased phosphatase activity noted in the insulin-treated, fasted rats since glycogen was routinely added to the homogenizing buffer.


Insulin
rapidly produced an increase in per cent of total heart glycogen synthase in the I form in fed rats. In fasted rats the response was diminished and delayed.  (l-3). It also may be catalyzed by a non-CAMP-dependent protein kinase (4). Synthase D phosphatase may have a rather broad substrate specificity (5-7), but has not been as well studied as the kinase.  (13).
Part of this data has been presented previously in abstract form (14). When fasted rats were used, food was removed 19 to 21 h before the animals were killed. Animals were always killed between 9:30 and 11:30 a.m. Diabetes was induced using 40 mg/kg of alloxan injected intravenously after a 24-h fast,. These animals were killed 3 days later.
Rats were anesthetized with 40 mg/kg of Seconal intraperitoneally. Ten to fifteen minutes later 6 units/kg of insulin (approximately 4.2 x 10m5 M) or an equivalent volume of diluent (1.0 ml/kg of 0.003 M HCl in 0.9% saline) was injected intraperitoneally.
In some experiments 2.8 mmol/kg of glucose or an equivalent volume of saline (controls) was injected intravenously.
At the appropriate time after injection the chest was quickly opened, the heart removed and placed in ice-cold saline (0.9% NaCl solution). Blood was collected in heparinized tubes and placed on ice. The atria and aorta were trimmed from the heart, and the ventricles homogenized in 50 mM imidazole, 0.5% glycogen, pH 7.0 (l/3, w/v), using Potter-Elvehjem homogenizing tubes and motor-driven Teflon pestles. The homogenate was then centrifuged at O-4", 7700 x g for 10 min. The supernatant was treated twice with Dowex l-X4 resin to remove anionic inhibitors of the phosphatase reaction (15). Total synthase and synthase I activities are not significantly affected by this treatment. In most studies 13.3 mM EDTA (final concentration 10 mM) also was present in the homogenizing solution in order to inhibit protein kinase activity.
Synthase phosphatase was assayed by incubating the supernatant at 30" using endogenous substrate.
The reaction was stoppe-d by adding an aliquot of the supernant to cold 200 mM KF, 7.5 mM EDTA, 10 mM potassium phosphate, pH 7.8, in a ratio of l/8, v/v. The diluted sample was immediately placed on ice and used for synthase and phosphorylase assays. No phosphatase activity could be demonstrated in the pellet. Eighty-five to ninety per cent of the synthase was present in the supernatant and there was no difference in the per cent of enzyme in the supernatant of fed, fasted, or insulin-treated animals. When synthase was assayed directly or tissue obtained for metabolite studies the heart was quickly removed and immediately frozen using liquid nitrogen-cooled aluminum clamps (~ 196"). The atria and aorta were removed and the tissue was stored at -20" until assay later the same day. The frozen tissue was crushed to a fine powder in a liquid nitrogen-cooled stainless steel percussion mortar. For synthase assay the pulverized tissue was homogenized in 60% glycerol, 10 mM EDTA, 50 mM KF. DH 7.0 at -10" (l/2. w/v) in a Potter-Elvehiem homoeenizer. Homogenization was completed at 0" using 10 mM EDTA, 50 GM KF. pH 7.0 (l/10, w/v). The homogenate was centrifuged at 12,000 x g for ib min at O-4" and the super&ant used for assay. For determination of glycogen, 50 pl of the l/4 supernatant was digested in 500 ~1 of 30% KOH. Extracts for glucose-6-P and ATP determinations were prepared by placing 100 ~1 of the l/4 supernatant on 300 ~1 of frozen 3 M perchloric acid and thawing slowly. Glycogen synthase was assayed using the method of Thomas et al. (16). Units are micromoles of UDP-["Clglucose incorporated into glycogen per min. Glycogen phosphorylase was assayed by the method of Gilboe et al. (15,17). Units are micromoles of ["Clglucose-1-P incorporated into glycogen per min. Synthase I kinase was assaved as previously described (is). Glycogen was determined by a phenol/sulfuric acid method (19), ATP and glucose-6-P by the fluorometric method of Lowry (20), plasma glucoseby the Nelson method (21). proteins by the method of Zak and Cohen (22), and CAMP by radioimmunoassay using a commercial kit (Schwarz/Mann).

20hr Fasted Rats
In the column chromatography studies, 500 pl of tissue supernatant were applied to a Sephadex G-25 column (1 x 15 cm) which previously had been equilibrated with homogenizing buffer at 4". The protein fraction was eluted with the same buffer and collected visually. This sample was used for phosphatase assay.

Minutes After Insulin Administration
In some studies proteins in the 7700 x g supernatant were FIG. 1. Increase in per cent of synthase in the I form in rat heart after precipitated using ammonium sulfate, 38% final concentration, at 0". insulin administration. The number of animals studied at each time After mixing for 10 min the precipitate was collected by centrifugation point appears in parentheses at the 15-min time point. Controls were and resuspended in 1 ml of homogenizing buffer. All of the phosphatase grouped and noted at zero time. * indicates a statistically significant activity was present in this fraction. This sample was column treated response to insulin. Total synthase activity was similar for each group as described above before phosphatase assay.
Statistical analyses were done using the Student t test for unpaired variates.
A p value of less than 0.05 is the criterion for significance.

RESULTS
Insulin administration to fed rats resulted in rapid increase in the per cent of heart glycogen synthase in the I form as we (11) and others (12) have shown previously.
In rats fasted 19 to 21 h the initial per cent of synthase I was lower and there was no increase after insulin administration. In diabetic animals the initial per cent of synthase I was similar to that in the normal fasting animals and changed little over the time course studied (Fig.  1). In none of these experiments was there a significant change in total synthase activity.
Since an active synthase D phosphatase is necessary for synthase D to synthase I conversion, the activity of this enzyme was studied in extracts of heart from fed, fasted, and diabetic animals.
In extracts from fed rats the phosphatase reaction was essentially linear for 10 min and the per cent of synthase I increased from 13 to 46% of the total synthase activity present (Fig. 2  Total synthase activity was -0.12 unit/ml and remained stable throughout the incubation. Initial mean per cent of synthase I was 10.2% (Fed), 5.5% (Fasted), and 4.8% (Insulin). n = 3 for each group.
in skeletal muscle (23). The initial glucose-6-P concentration in the phosphatase assay mixtures was very low but increased considerably during the incubation. However, the concentrations were similar in extracts from fed and fasted animals (  Hearts from fed, fasted, and insulin-treated, fasted rats were homogenized in 50 mM imidazole, pH 7.0, with varying concentrations of glycogen. Glycogen concentration in the homogenizing solution was (0) 0 mg/ml (0.5 for fed), (0) 2.5 mg/ml, (A) 5.0 mg/ml, (A) 7.5 mg/ml, and (x) 10 mg/ml. Total synthase activity, initial per cent of synthase I and blood sugars were similar to those reported in Fig. 4. n = 7 for each line.
its substrate, or both. If small molecular weight activators are important in the mechanism they must be tightly bound to the proteins or have induced a stable change in the protein. Although glycogen has been reported to inhibit synthase phosphatase activity (32), it was routinely added (5 mg/ml) in our studies in order to assure stability of synthase D; this also reduced any variability in phosphatase activity due to variation in endogenous glycogen concentration.
In order to investigate the mechanism of insulin activation of synthase phosphatase the assay was done using variable concentrations of glycogen in the homogenizing buffer.
In extracts from fed animals glycogen had little effect on the reaction velocity ( Fig.  7) but in extracts from fasted animals considerable inhibition was produced. Insulin administration to the fasted animals Per cent phosphorylase a activity in extracts from fed, fasted, and insulin-treated rats. Total phosphorylase activity was similar in all animals (-40 units/g wet weight) and remained stable throughout the incubation. n = 13 for each group.
resulted in a reduction in glycogen inhibitability. When the reaction velocity versus glycogen concentration was plotted and the data extrapolated to zero glycogen concentration, the effect of insulin essentially disappeared (Fig. 8). The assays could not be done in the absence of glycogen due to instability of the synthase activity. In liver, phosphorylase a inhibits synthase phosphatase activity (33) in a noncompetitive manner' and has been implicated in the physiological regulation of the synthase system in that organ. In heart, phosphorylase a has been reported to be a substrate for a partially purified protein phosphatase having synthase phosphatase activity (6). Therefore, the possibility that the phosphorylase a concentration could be regulating synthase phosphatase activity was considered.
In extracts from fed, fasted, and insulin-treated, fasted rats the initial phosphorylase a activity was similar (Fig. 9). The per cent of phosphorylase in the a form decreased rapidly and by 5 min of 'Unpublished observations. incubation had reached a low stable level which was similar in the three groups ( Fig. 9) (44) and it has been suggested to be an intermediate in the action of insulin.
In the present study direct addition of insulin to the homogenates had no effect on synthase phosphatase activity nor did addition of cGMP to a final concentration of 1.0 mM. Thus, the mediator through which insulin activates heart synthase phosphatase remains unknown. Since this enzyme appears to be a rather broad specificity protein phosphatase (5-7), activation of the enzyme or alterations in its substrates could produce other intracellular effects on insulin.
In addition, activation by insulin of similar but distinctly different phosphoprotein phosphatases may be important both intracellularly and in the mediation of insulinstimulated transmembrane transport.