Insulin Action Rapidly Decreases Multifunctional Protein Kinase Activity in Rat Adipose Tissue*

Transcription

1 THE JOURNAL OF BIOLOGICAL CHEMISTRY 1988 by The American Society for Biochemistry and Molecular Biology, Inc. Vol. 263, No. 25, Issue of September 5, pp ,1988 Printed in U.S.A. Insulin Action Rapidly Decreases Multifunctional Protein Kinase Activity in Rat Adipose Tissue* (Received for publication, April 7, 1988) Seethala Ramakrishna and William B. Benjamin From the Diabetes Research Laboratory, Department of Physiology and Biophysics, School of Medicine, Health Sciences Center, State University of New York, Stony Brook, New York ATP-citrate lyase in vivo contains three phosphorylation sites on two tryptic peptides (peptides A and B). These phosphorylation sites are under hormonal control. Multifunctional protein kinase (MFPK) from rat liver phosphorylates peptide B on serine and threonine residues whereas CAMP-dependent protein kinase phosphorylates peptide A on a serine residue (Ramakrishna, S., and Benjamin, W. B. (1985) J. Biol. Chern. 26, ). We now report that rat adipose tissue MFPK also phosphorylates serine and threonine residues of peptide B of ATP-citrate lyase. When the activity of MFPK was assayed using partially purified (by chromatography on phosphocellulose) cytosol fractions from insulin-treated adipose tissue, it was found that MFPK activity was decreased by over 55%. This decrease in MFPK activity occurs at physiological con- centrations of insulin (EGO = 1 X 1 O M). Its onset is rapid and almost maximal at 5 min after the addition of insulin. Even when new protein synthesis is inhibited by cycloheximide, extracts from insulin-treated fatpads have less MFPK activity compared to the control. The insulin effect is maintained after further chromatography on a gel filtration column suggesting that the decrease in MFPK activity is not due to a low molecular weight inhibitor. The insulin-induced decrease in MFPK activity is due to a decrease in V, whereas the affinity of this enzyme toward ATP-citrate lyase or ATP is unchanged. Reversible phosphorylation of proteins is a common mechanism in the regulation of enzymes active in the control of cell metabolism (1). Hormones often regulate the protein kinases that phosphorylate multisite phosphorylated proteins at specific sites (2). Cyclic nucleotide-dependent protein kinases, calcium-calmodulin-dependent protein kinases, and protein kinase C are examples of kinases whose activities are controlled by hormones acting through their specific receptors by known general mechanisms. However, the mechanism for the coupling of receptors like the insulin receptor that possesses intrinsic kinase activity to their postulated controlled kinases is not known (1). Protein kinases that might be coupled to these receptors are now being sought (3-6). The identification of a serine and threonine kinase whose activity is controlled by insulin would contribute to our understanding of the mechanism of insulin action (7). Insulin stimulates incorporation of 32Pi into serine and threonine residues of ATP-citrate lyase (8,9), insulin receptor (1, ll), ribosomal protein S6 (4, 12, 13), and M, 23, protein (14). ATP-citrate lyase in uiuo contains several sites of phosphorylation in two tryptic peptides, peptides A and B, that are under hormonal control (8, 9, 15). Isoproterenol increases the phosphorylation of both peptides A and B. Insulin increases the phosphorylation of only peptide A but decreases the phosphorylation of peptide B of ATP-citrate lyase. These hormonal effects on ATP-citrate lyase phosphorylation remain unexplained (9, 16). We have isolated from rat liver a MFPK that has a limited substrate spectrum (17-19). It phosphorylates ATP-citrate lyase at peptide B, acetyl-coa carboxylase at a site different from that phosphorylated by CAMP-dependent protein ki- nase, and glycogen synthase at sites 2 and 3, a, b, and c (18, 2, 21). A similar protein kinase has been purified from rat skeletal muscle that also phosphorylates sites 3 and 2 of glycogen synthase (22). MFPK had been considered by us (18) to be a kinase whose activity could be regulated by insulin, because insulin decreases the phosphorylation of peptide B of ATP-citrate lyase (9,15) and because skeletal muscle glycogen synthase from alloxan diabetic rabbits has an increased phosphate content at sites 3 and 2 which decreases to control values by insulin treatment (23). Furthermore, in short-term experiments, insulin treatment of normal rabbits treated with propanolol decreases the phosphate content of site 3 (24). Interestingly, acute treatment of rat hemidiaphragms and adipocytes with insulin and glucose also decreases the incorporation of 32Pi at sites 3 and 2 (25, 26). In this report we describe experiments which demonstrate that insulin at physiological concentrations rapidly decreases MFPK activity in rat adipose tissue. EXPERIMENTAL PROCEDURES Materials-Phosphocellulose P-11 and K-2 TLC cellulose plates were purchased from Whatman. DEAE-Sephacel and Sephacryl S- 2 were from Pharmacia Biotechnology Inc. ATP, PMSF, leupeptin, soybean trypsin inhibitor, and casein (partially hydrolyzed and dephosphorylated) were from Sigma. [-p3 P]ATP was from ICN radiochemicals. ATP-citrate lyase and MFPK were purified from rat liver as described previously (18, 27). Catalytic subunit of CAMP-dependent protein kinase was purified from rabbit skeletal muscle (15, 28). Porcine insulin was a gift from Lilly. The sources of other chemicals were as described previously (18). Incubation of Fat Pods with Hormones-Epididymal fat pads from * This investigation was supported by a grant from the Suffolk Heart Association and by Grants AM 1895 and AM 3215 from the National Institute of Arthritis, Metabolism, and Digestive Diseases, National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked aduertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. In this report, the term in uiuo refers to intact tissue or cells, whereas the term in uitro refers to a cell-free system. The abbreviations used are: MFPK, multifunctional protein kinase; PMSF, phenylmethylsulfonyl fluoride; DTT, dithiothreitol; Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid; EGTA, [ethylenebis(oxyethylenenitrilo)]tetraacetic acid; HPLC, high pressure liquid chromatography

2 12678 Insulin Decreases MFPK Activity g male Sprague-Dawley rats (Taconic Farms, Inc., Germantown, NY) were incubated in modified Krebs-Ringer Hepes buffer, ph 7.4, containing 3% bovine serum albumin and 5 mm glucose at 37 "C as previously described (9, 16, 29). After preincubation, insulin (.1-1 nm) was added as indicated in the legends. After incubation, the adipose tissue was rinsed in Krebs-Ringer bicarbonate buffer and either rapidly frozen on an aluminum block cooled to -7 "C and processed or processed directly as described below. Purification of MFPK from Fat Pads-Adipose tissue was homogenized in 5 volumes of buffer A (2 mm Tris-HC1, ph 7.5, 5 mm EDTA, 1 mm NaF, 1 mm sodium pyrophosphate, 2 PM sodium orthovanadate, 1 mm PMSF, and 225 mm sucrose) and centrifuged at 1, X g for 1 min followed by centrifugation at 15, X g for 1 h. MgClz and Hepes, ph 7.4, were added to the supernatant to final concentrations of 12 and 7 mm, respectively, and loaded on a DEAE- Sephacel column (.8 X 6 cm) equilibrated in buffer B (25 mm Hepes, ph 7.4, 1 mm NaF, 2 PM sodium vanadate, 1 mm PMSF, and 2 mm DTT). The column was washed with 2 ml of buffer B. The wash was pooled with the material not retained on the column and concentrated in Centricon 1 concentrator (Amicon Corp.). Note that pyrophosphate and NaF inhibit MFPK activity (18). The concentrate was diluted with an equal volume of buffer C (4 mm p-glycerophosphate, ph 7.4, 1 mm EGTA, 7.5 mm MgC12,.2 mm PMSF, 2.5 mm DTT, 2 PM sodium vanadate, and leupeptin, pepstatin, and soybean trypsin inhibitor, 1 mg/liter) and loaded on a phosphocellulose P- 11 column (.8 X 6 cm) equilibrated in buffer C. CAMP-dependent protein kinase was not retained on the P-11 column. The column was washed with 1 ml of buffer C, and MFPK was eluted with a linear gradient of -.5 M KC1 in buffer C. 1-ml fractions were collected. MFPK activity was assayed in each fraction by using ATP-citrate lyase fully phosphorylated at peptide A. Fractions with the most MFPK activity were pooled and concentrated in a Centriprep 1 concentrator (Amicon Corp.). In some experiments, MFPK was eluted from the phosphocellulose column with.35 M NaCl in buffer C in 1-ml fractions by a batch procedure. MFPK activity was eluted in fractions 2 and 3, pooled and concentrated as described. Because a high salt concentration inhibits MFPK activity, the fractions were either concentrated and dialyzed to remove the salt or used at a 1- fold dilution. Note that all phosphorylation activity values were linear with time and with the amount of added kinase. In some experiments MFPK was further purified by gel filtration on a Sephacryl S-2 column (1 X 55 cm) equilibrated with 5 mm NaCl in buffer C. Protein Kinase Assays-MFPK activity was measured in the column fractions by using rat liver ATP-citrate lyase that had been phosphorylated on peptide A as described below. The kinase reaction was initiated by the addition of the reaction mixture (final concentrations: 5 mm Hepes, ph 7.,2 mm DTT, 8 mm magnesium acetate,.3 mm EGTA,.5 mm EDTA, and 5 PM [y-32p]atp). Samples were incubated at 37 "C for 5-2 min. The reaction was terminated by adding sodium dodecyl sulfate sample buffer. Samples were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and the phosphate incorporated into peptide B of ATP-citrate lyase was measured as described previously (18). Preparation of ATP-Citrate Lyase Phosphorylated at Peptide A- Rat liver ATP-citrate lyase (2 mg/ml) was dephosphorylated with.4 mg/ml Escherichia coli alkaline phosphatase as described previously (19). The dephospho-atp-citrate lyase was isolated by gel filtration on a Sephacryl S-2 column (1 X 55 cm). The dephospho-lyase was phosphorylated at peptide A with unlabeled ATP by incubation with the catalytic subunit of CAMP-dependent protein kinase at 37 "C for 6 h. The A site-phosphorylated lyase was isolated with a Sephacryl S-2 column. In parallel assays using radioactive ATP it was determined that about 98% of the A site was phosphorylated. HPLC and Phosphoamino Acid Analysis-ATP-citrate lyase was phosphorylated with [y-32p]atp by using either purified rat liver MFPK or MFPK (P-11 or Sephacryl S-2 fractions) from control or insulin-treated adipose tissue. The phosphorylated ATP-citrate lyase was further incubated with 5 mm ATP, 2 mm potassium citrate, 1 mm MgS4,.3 mm CoA, and 1 mm DTT to discharge histidine phosphate. The mixture was passed through a Sephacryl S-2 column (.8 X 21 cm) equilibrated in 1 mm NH4HC3. The fractions containing [32P]ATP-citrate lyase were pooled and lyophilized. The lyophilized sample was dissolved in 5 mm NHhHC3, digested with tosylphenylalanyl chloromethyl ketone-trypsin (trypsin:substrate, 1:5) at 37 'C for 18 h, and lyophilized. The peptides were dissolved in.1% trifluoroacetic acid and resolved by reverse-phase HPLC on an Ultrasphere C,, (Altex) column (9, 19). The lyophilized [32P]ATP-citrate lyase- or peptide B-containing fractions that were desalted and lyophilized were hydrolyzed in 6 N HCl at 11 'C for 2 h and lyophilized. The residue was dissolved in electrophoresis buffer, ph 2.8 (acetic acidformic acidpyridine:water, 12:1:187) and spotted along with phosphoserine, phosphothreonine, and phosphotyrosine standards onto a K2-cellulose plate (Whatman) and electrophoresed at 15 V for 1.5 h (15). Phosphoamino acids were detected by ninhydrin staining, and the radiolabeled phosphoamino acids were detected by autoradiography. RESULTS AND DISCUSSION Fat pads from normal fed rats were incubated with insulin and processed as described. Fractions not retained on DEAE- Sephacel were adsorbed to a phosphocellulose column. MFPK activity was eluted between.12 and.25 M KCl. Fig. 1 gives representative chromatographic elution profiles of MFPK from control and insulin-treated adipose tissue from phosphocellulose and Sephacryl columns. Phosphocellulose fractions from insulin-treated fat pads contained 58% less MFPK activity than fractions obtained from controls (Table I). When MFPK was further purified on a Sephacryl S-2 column, almost all of the activity eluted as a single peak. About 5% less MFPK activity was found in fractions from insulintreated adipose tissue (Fig. lb). Because this decrease in MFPK activity from insulin-treated adipose tissue persists after multiple column chromatographic steps (Fig. lb) it is unlikely that it is due to a low molecular weight inhibitor. As shown in Table 11, the specific activity of MFPK increases and the insulin inhibitions persist through the partial purification of the enzyme. We believe that the kinase assayed is MFPK because 1) its chromatographic behavior on the phos- a 7 16 E 12" 2 8;: /y I" 2 4: FRACTION NUMBER A A~~'A E,, ~ - A d? / - & i < ~ ~, - & - A. ~ /A-A. -A // \A -A I FRACTION NUMBER FIG. 1. Chromatography of MFPK from adipose tissue extracts. a, the high-speed supernatants were loaded on a DEAE- Sephacel column. The material not retained was loaded on a phosphocellulose (P-11) column equilibrated with buffer C. After washing the column with 1 ml of buffer C, MFPK activity was eluted with a linear gradient of-.5 M KC1 in buffer C. 1-ml fractions were collected. MFPK activity, protein, and salt concentrations were measured in the fractions. The specific activity of MFPK in the fractions from the control (A) and insulin-treated (A) fat pads was determined. Conductivity () in millimhos in the fractions was read in a.78-cm cell at 2 "C in a conductivity meter (Radiometer Copenhagen). Note that MFPK activity was eluted as a single peak at the same salt concentration (data not shown) when the peak fractions (8-14) and fractions (15-2) in the trailing edgewerepooled separately and rechromatographed on P-11. This finding suggests that the MFPK activity in the broad peak is due to a single species of MFPK. b, the phosphocellulose pool was concentrated and loaded on a Sephacryl S-2 column (1 X 55 cm) equilibrated in 5 mm NaCl in buffer C and eluted with the same buffer..9-ml fractions were collected. MFPK activity in the fractions from control (A) and insulin-treated (A) samples was assayed as described in the text. 4

3 Insulin Deweases MFPK Activity TABLE I Effect of insulin on MFPK activity in rat adipose tissue Fat pads were incubated with or without insulin (1nM) in modified Krebs-Ringer Hepes buffer (16)containing 3% bovine serum albumin for 3 min. After incubation, fat pads were quickly rinsed and homogenized in buffer containing phosphatase inhibitorsunless otherwise indicated. Fat pads that received cycloheximidewere preincubated with cycloheximide (1 pg/ml) for 15 min. Hormone was added and incubations continued as described. MFPK was purified from high speed supernatants through phosphocellulose chromatography as described. MFPK was assayed using substrates.1 mg/ml ATP-citrate lyase (peptide A-phosphorylated),.3 mg/ml native ATP-citrate lyase,.3 mg/mlglycogen synthase, and.1 mg/ml acetyl-coa carboxylase. The numbers in parenthesesare thenumber of independent experiments with 1-3 measurements in each experiment. The data arepresented as mean f S.E. of the tryptic peptides of these phosphorylated samples (Fig. 2) showed that only serine and threonine residues of peptide B are phosphorylated. Furthermore, as previouslydemonstrated (18, 21), CAMP-dependent protein kinase, glycogen synthase kinases 3, 4, and 5, casein kinase 1 and 2, phosphorylase kinase, and calmodulin-dependent kinase 1 do not phosphorylate the peptide B of ATP-citrate lyase. For all the above reasons we believe that these enzyme assays using ATPcitrate lyase fully phosphorylated at peptide A measure the activity of MFPK. Inhibition of MFPK by insulin was also noted (Table I) when native ATP-citrate lyase, glycogen synthase, and acetylcoa carboxylase were usedas substrates. Histone, phosvitin, phosphorylase b, and casein were poor substrates for phosa Substrate pmol/min/mg protein phosphorylated) (6) Lyase (native enzyme) (2) Glycogen synthase (2) Acetyl-coA carboxylase (2) 57.6 f f Experimental design and purification were as described in Table I and under Experimental Procedures. MFPK activity was assayed using ATP-citrate lyase that was phosphorylated a t peptide A. MFPK activity is difficult to assay in the supernatants anddeae-sephacel flow-through fractions partly dueto NaF and other inhibitors as had been noted (18).MFPK activity in the high-speed supernatant and DEAE-Sephacel fractions was lost after 3 days of storage a t C. Note that the t o t a l recoveries for control and insulin samples are similar to those reported for the ratliver MFPK (18).The values are the average of two independent experiments. Volume Activity/ml Specific activity Purification step ControlInsulinControlInsulinControlInsulin pmollmin C pmhwmg protein a 1 A b, 6- i 4-. TABLE I1 Purification ofmfpk from rat adipose tissue ml b % f2.35 f f13.9 f f.12 f NOphosphatase inhibitors Cyclo , I heximide, 1 pg/ml a Using [ C]leucine, cycloheximide was found to inhibit protein synthesis by 97%. 1. Extract 2. DEAE-Sephacel flow-through 3. Phosphocellulose 4. Sephacryl S phocellulose column is precisely that found for the rat liver enzyme, 2) it is strongly inhibited by the same agents that inhibit the liver enzyme,3) it has the same substrate specificity as theliver enzyme,and 4) it phosphorylates only peptide B of ATP-citrate lyase (18). To show that adipose tissue MFPK phosphorylates only ATP-citrate lyase peptide B, native ATP-citrate lyase was phosphorylated using MFPK (Sephacryl S-2 fractions) of control and insulin-treated adipose tissue. Analysis by HPLC C - 8 TIME, min FIG. 2. HPLC mapsof tryptic fragments of ATP-citrate lyase. ATP-citrate lyase was phosphorylated with MFPK from control ( a ) or insulin-treated adipose tissue ( b ) or liver MFPK (c). To the [32P]ATP-citratelyase, unlabeled lyase was added and digested with TPCK-trypsin (trypsin:enzyme, 1:5). The tryptic digest was lyophilized, dissolved in 1 pl of.1% trifluoroacetic acid, and 5 pl was chromatographed on a reversed-phase Ct8 column equilibrated with 5% acetonitrile in.2 M phosphate buffer, ph 3.5. Peptides were eluted a t a flow of 1 ml/min for 5 min with 5% acetonitrile and then by a linear gradient of 5-4% acetonitrile for 7 min followed by 5 min of elution with 4% acetonitrile in phosphate buffer. Fractions (1 ml) were collected and analyzed for radioactivity by Cerenkov radiation. Peptide B was eluted a t R,58 min. Note that thefractions eluted between 3 and 42 min were determined to be smaller fragments of peptide B containing bothphosphoserine and phosphothreonine (19).The inset shows the radiolabeled phosphoamino acids from ATP-citrate lyase that were phosphorylated by MFPK from control ( a ),insulin-treated ( b )adipose tissue, and liver (c). Following are the ratios of phosphoserine: phosphothreonine in ATP-citrate lyase phosphorylated by MFPKcontrol, 65:35; insulin, 7:3; and liver, 5743.Phosphoamino acid analyses of peptide B from the above HPLC samples showed both phosphoserine and phosphothreonine in similar ratios as for ATP-citrate lyase.

4 I.12] 1268 Insulin Decreases MFPK Activity phorylation by adipose tissue MFPK as has been found previously with the rat liver enzyme (18). If phosphatase inhibitors were omitted from the homogenization buffer (Table I) MFPK activity from insulin-treated fat pads was similar to control values. Note that these control values were lower than controls prepared using the buffer containing phosphatase inhibitors. We suggest that the inhibitory effect of insulin on MFPK activity is due to covalent modification of the kinase as the insulin effect is dependent on the presence of phosphatase inhibitors in the homogenization and subsequent buffers. To determine if the insulin-induced inhibition of MFPK activity requires new protein synthesis, fat pads were incubated with cycloheximide (1 pg/ml) to stop protein synthesis. The insulin effect on MFPK activity persists after cycloheximide inhibition of protein synthesis (Table I) suggesting that the effect of insulin on MFPK activity does not require new protein synthesis. The half-maximal effect of insulin on decreasing MFPK activity is found at an insulin concentration of.1 nm suggesting that the hormone-induced reduction is probably due to the stimulation of insulin receptors (Fig. 3). Note that there is a comparable insulin-induced decrease in MFPK activity when natural substrates of intermediary metabolism (native e P \.- C ATP, pm E - T b P / LYASE-, p ~ FIG. 4. Kinetic analyses of ATP-citrate lyase (peptide A phosphorylated) phosphorylation by the phosphocellulose fractions of control and insulin-treated fat pads. a, effect of ATP concentration at.56 PM ATP-citrate lyase (peptide A phosphorylated) and b, ATP-citrate lyase concentration at 5 PM ATP on the MFPK (Sephacryl S-2 fraction) activity in control () and insulin-treated () fat pads. These data are presented as a doublereciprocal plot of ATP concentration and MFPK (panel a) and ATPcitrate lyase concentration and MFPK activity (panel b). Lines were fitted by linear regression. The data points are the average of two determinations LOG INSULIN, M 2 4 TIME, rnin FIG. 3. Effect of insulin treatment on MFPK activity. a, dose response of insulin on MFPK activity using ATP-citrate lyase phosphorylated at peptide A (O), native ATP-citrate lyase (A), glycogen synthase (V), or acetyl-coa carboxylase () as substrate. Fat pads were incubated in the absence or presence of different concentrations of insulin (1 to 1 M) at 37 C for 3 min. Fat pads were homogenized in buffer A and purified through the phosphocellulose step as described under Experimental Procedures. MFPK activity in the phosphocellulose fractions at each insulin concentration was expressed as a percent of control. b, time course of the insulin effect on MFPK activity. Fat pads were incubated in the absence or presence of insulin (1 nm) at 37 C for the indicated times. MFPK activity was determined in the phosphocellulose fraction using ATP-citrate lyase phosphorylated at peptide A as substrate. Each data point represents the average of duplicate determinations. ATP-citrate lyase, glycogen synthase, and acetyl-coa carboxylase) were used for these assays. The time course of insulin action is shown in Fig. 3b. About 8% of the maximal reduction in MFPK activity is found by 5 min after the addition of insulin. This finding is consistent with previous data that demonstrate a marked insulin effect on ATP-citrate lyase phosphorylation by 5 min after the addition of insulin to either fat pads or fat cells (29). Fig. 4b shows a Lineweaver-Burk plot of the relationship between ATP-citrate lyase (phosphorylated at peptide A) concentration and MFPK activity in partially purified extracts of control and insulin-treated adipose tissue. The V,, was markedly decreased (66%) 26 to 7 pmol/min/mg protein by insulin. The apparent K,,, ( p ~ for ) lyase was unchanged by insulin. Similarly, the K,,, (1-11 p ~ for ) ATP was unchanged by insulin, but the V,, was decreased by 55% from 8 to 36 pmol/min/mg protein (Fig. 4a). Because the Vmax values were determined at a subsaturating concentration of ATP-citrate lyase, the values with varying concentrations of ATP were generally lower than those obtained at fixed (5 p ~ ATP. ) Using native ATP-citrate lyase, similar results were obtained (data not shown). The Vmax was decreased (48%) by insulin from 222 to 116 pmol/min/mg protein without a change in the K,,, ( FM) for native ATP-citrate lyase. These results suggest that the decreased V,, of MFPK in response to insulin was due to a decreased amount of active MFPK and not due to a change in its affinity toward ATPcitrate lyase and ATP. The decreased MFPK activity was not due to a decreased synthesis of MFPK because cycloheximide did not block the insulin effect. The decrease in MFPK by insulin could be due to a covalent modification (phosphoryl-

5 ~ ~~~ ~~~ ation state) of the enzyme because the insulin effect is lost when phosphatase inhibitors were omitted from the buffers. Although the spectrum of enzymes phosphorylated by MFPK is restricted (18), the consequences of phosphorylation by this kinase are most interesting. MFPK phosphorylation of glycogen synthase (18, 21) and acetyl-coa carboxylase inhibits enzyme3 activity whereas MFPK phosphorylation of ATP-citrate lyase has no effect on enzyme activity (18). Insulin increases the activity of some enzymes by causing dephosphorylation of sites that are inhibitory for enzyme activity when phosphorylated (3, 31). Activation of a broad spectrum phosphatase has been invoked to account for the observed enzyme dephosphorylation (32), though how specific dephosphorylations occurs has not been determined. We have suggested that the phosphorylation of sites 3a, b, and c of glycogen synthase could be decreased if insulin decreases the activity of a kinase with specificity for these key sites (18, 21). In this communication we have presented evidence that insulin action decreases the activity of a kinase with specificity for glycogen synthase sites 3a, b, c, and 2. The mechanism for the insulin-induced decrease in MFPK activity is unknown. Insulin could cause a change in the chromatographic behavior of the kinase leading to a loss of the enzyme in the eluates from the phosphocellulose column. However, we were unable to detect any MFPK activity in fractions eluted from the DEAE-Sephacel column even after dialysis and concentration to remove inhibitors and enhance the sensitivity of the assay. Insulin could cause a shift of the enzyme from an active to an inactive form which could also involve a change in its chromatographic behavior. It should be noted that (as shown in Table 11) the high-speed supernatant and DEAE-Sephacel fractions from insulin-treated tissue also show a decrease in the specific activity of MFPK when assayed with site A phosphorylated ATP-citrate lyase. The mechanism for the insulin-induced decrease in MFPK activity could involve a change in the phosphorylation state of MFPK as suggested by the loss of the insulin effect when phosphatase inhibitors were left out of the buffers. This change in phosphorylation could be due to the action of 1) the insulin receptor, 2) a specific phosphatase, or 3) a MFPK kinase. Whatever the precise signaling mechanism, the relation of the insulin-induced decrease in MFPK activity to the overall phosphorylation of ATP-citrate lyase and by analogy to the complex phosphorylations of glycogen synthase and acetyl-coa carboxylase is of importance in cell metabolism. Acknowledgment-We thank Ratnakumari Seethala for excellent technical assistance and typing the manuscript. S. Ramakrishna and W. B. Benjamin, manuscript in preparation. Insulin Decreast?s MFPK Activity REFERENCES 1. Cohen, P. (1982) Nature 296, Krebs, E. G. (1986) in The Enzymes (Boyer, P. D., and Krebs, E. G., eds) Vol. 17A, pp. 3-2, Academic Press, Orlando, FL 3. Tabarini, D., Heinrich, J., and Rosen,. M. (1985) Proc. Natl. Acad. Sci. U. S. A. 82, Cobb, M. H. (1986) J. Bwl. Chem. 261, Ray, L. B., and Sturgill, T. W. (1987) Proc. Natl. Acad. Sci. U. S. A. 84, Sommercorn, J., Mulligan, J. A., Lozeman, F. J., and Krebs, E. G. (1987) Proc. Natl. Acad. Sci. U. S. A. 84, Rosen,. M. (1987) Science 237, Ramakrishna, S., and Benjamin, W. B. (1979) J. Biol. Chem. 254, Pucci, D. L., Ramakrishna, S., and Benjamin, W. B. (1983) J. Bwl. Chem. 258, White, M.F., Takayama, S., and Kahn, C.R. (1985) J. Biol. Chem. 26, Pertruzelli, L., Herrera, R., and Rosen,. M. (1984) Proc. Natl. Acad. Sci. U. S. A. 81, Smith, C. J., Rubin, C. S., and Rosen,. M. (198) Proc. Natl. Acad. Sci. U. S. A. 77, Lastick, S. M., and McConkey, E. H. (1981) J. Biol. Chem. 266, Ramakrishna, S., and Benjamin, W. B. (1983) Biochem. Biophys. Res. Commun. 117, Ramakrishna, S., Pucci, D. L., and Benjamin, W. B. (1981) J. Bwl. Chem. 256, Ramakrishna, S., Pucci, D. L., and Benjamin, W. B. (1984) Bwchem. Biophys. Res. Commun. 122, Ramakrishna, S., and Benjamin, W. B. (1981) FEBS Lett. 124, Ramakrishna. S.. and Beniamin. W. B. (1985).. J. Biol. Chem. 26, Ramakrishna. S.. Pucci. D. L.. and Beniamin. W. B. (1983).. J. Biol. Chem. 258, Ramakrishna, S., and Benjamin, W. B. (1983) Biochem. Biophys. Res. Commun. 117, Sheorain, V. S., Ramakrishna, S., Benjamin, W. B., and Soderling, T. R. (1985) J. Bwl. Chem. 26, Hegazy, M. G., Thysseril, T. J., Schlender, K. K., and Reimann, E. M. (1987) Arch. Biochem. Biophys. 258, Sheorain, V. S., Juhl, H., Bass, M., and Soderling, T. R. (1984) J. Bwl. Chem. 269, Parker, P. J., Caudwell, F. B., and Cohen, P. (1983) Eur. J. Bwchem. 13, Lawrence, J. C., Jr., Hiken, J. F., DePaoli-Roach, A.A., and Roach, P. J. (1983) J. Bwl. Chem. 268, Lawrence, J. C., Jr., James, C., and Hiken, J. F. (1986) J. Bwl. Chem. 261, Ramakrishna, S., and Benjamin, W. B. (1981) Biochem. J. 195, Bechtel, P. J., Beavo, J. A., and Krebs, E. G. (1977) J. Bwl. Chem. 262, Benjamin, W. B., and Singer, I. (1975) Biochemistry 14, Lamer, J., and Villar-Palasi, C. (1971) Curr. Top. Cell. Regul. 3, Parker, P. J., Embi, N., Caudwell, B., and Cohen, P. (1982) Eur. J. Biochem. 124, Cohen, P. (1986) in The Enzymes (Boyer, P. D., and Krebs, E. G., eds) Vol. 17A, pp , Academic Press, Orlando, FL