Glucose-induced phospholipid-dependent protein phosphorylation in neonatal rat islets

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol248, No. 2, August 1, pp. 562-569, 1986

Glucose-Induced

Phospholipid-Dependent Protein Phosphorylation in Neonatal Rat Islets

MARJORIE University

of Melhnwne

E. DUNLOP’

Department

Royal

of Medicine,

Received

RICHARD

AND

February

Melbourne

G. LARKINS Hosp’tal,

Victoria

3050, Australia

13, 1986

The participation of calcium-activated, phospholipid-dependent protein kinase in the phosphorylation of endogenous islet proteins following the exposure of cultured, neonatal pancreatic islets to stimulatory glucose concentrations was investigated by two techniques. In the first technique, islets were prelabeled with 32Pi. The major endogenous substrates for glucose-induced phosphorylation had apparent molecular massesof 130,100 f 1010, 100,000 + ‘700,80,400 + 890,58,100 f 1200,39,800 + 700, and 29,400 f 700 Da. In the presence of 12-0-tetradecanoylphorbol 13-acetate (2 PM), an activator of calciumactivated phospholipid-dependent kinase, there was enhanced phosphorylation of proteins of 80,000,40,000, and 29,000 Da. In the second technique, exogenous phosphorylation by [y-32P]ATP of proteins in a postnuclear particulate fraction was studied in the presence and absence of cofactors for Ca2+-activated, phospholipid-dependent protein kinase (Ca2+, phosphatidylserine, and unsaturated diolein). These studies were performed in islets preexposed to low (1.7 mM) or high (16.7 mM) glucose concentration prior to preparation of the postnuclear particulate fraction. Following exposure of islets to low glucose concentration, three substrates (apparent molecular masses40,500 -t 600,57,100 + 700, and 79,400 + 600 Da) in the postnuclear particulate fraction exhibited enhanced phosphorylation in the presence of calcium ions, phosphatidylserine, and unsaturated diolein. In preparations of islets preexposed to 16.7 mM glucose, the phosphorylation of the protein of molecular mass about 40,000 Da was significantly reduced, indicating prior phosphorylation of the acceptor sites on this substrate in response to glucose exposure. It is concluded that stimulation of neonatal cultured islets by glucose induces the acute changes in calcium ion, phospholipid, and diacylglycerol concentration required to activate the calcium-activated phospholipid-dependent protein kinase and that the islet postnuclear particulate fraction contains at least one specific substrate for this kinase. 0 1986 Academic Press, Inc.

Many studies have indicated that glucose stimulation is associated with a change in the phosphorylation state of a number of islet proteins. These phosphorylation reactions have been shown to be dependent on either cyclic nucleotides of Ca’+/calmodulin (l-5). In addition, a cyclic nucleotide-independent protein kinase which is activated in association with membrane phospholipids in the presence of Ca2’ has 1 To whom

correspondence

should

0003-9861186 $3.00 Copyright 0 1986 by Academic Press. Inc. All rights of reproduction in any form resewed.

be addressed. 562

been described (6). In other tissues alteration in diacylglycerol appears to be the major cellular regulator of the activity of this kinase (7, 8). Although demonstrated to phosphorylate endogenous substrates in rat and mouse islets (9) and in an insulinsecreting tumor (lo), the extent to which Ca2+- and phospholipid-dependent protein kinase participates in glucose-induced insulin release is not known. Recent investigations from our laboratory have shown acute elevation of islet 1,2-diacylglycerol content derived from arachidonic acid-la-

GLUCOSE-INDUCED

PHOSPHOLIPID-DEPENDENT

beled phosphoinositide in response to exposure of the islets to high glucose concentration (ll), providing a potential mechanism for activation of the kinase. The purposes of this investigation were to determine whether the glucose-induced rise in diacylglycerol was accompanied by activation of the Ca2+-activated phospholipid-dependent kinase, to identify possible substrates for the kinase, and to examine the cellular location of the protein(s) phosphorylated by the kinase. EXPERIMENTAL

PROCEDURES

Materials. ‘aPi and [y-32P]ATP were obtained from the Radiochemical Centre (Amersham, U.K.). RPM1 1640 medium and 4-(Z-hydroxyethyl)-l-piperazineethanesulfonic acid (Hepes)’ were from Flow Laboratories. 4-Morpholineethanesulfonic acid (Mes) was from Sigma Chemical Company. Maintenance and labeling of nemzatal islets with “Pi in culture. Cultured neonatal islets were prepared from dispersion of l-day-old rat pancreas as described by Hellerstrom et aL (12). After culture for 40 h in RPM1 1640 medium containing 11.1 mM glucose and 10% bovine fetal serum =Pi (30 pCi/400 islets) was added and culture continued for a further 24 h. Islets were transferred to RPM1 1640 medium supplemented to the salt concentrations of Hepes-buffered Krebs Ringer Bicarbonate buffer (pH 7.4) containing bovine serum albumin (350 mg/lOO ml) and glucose (1.7 mM) for 20 min. After this time glucose (final concentration, 16.7 mM) was added to this medium for a stimulatory period of 15 min. Islets were pelleted by eentrifugation (1008, 5 min). Insulin concentration of supernatant medium was determined by radioimmunoassay using rat insulin standards. Measurement of SPP,-labekd phosphoproteins in glucose-stimulated islets. Islets free of supernatant medium were homogenized in Tris/HCl (10 mM, pH 7.4) with sucrose (320 mM) and EGTA (1 mM) in a handheld Tenhroeck homogenizer at 4°C. Nuclei and cell debris were removed by centrifugation (6OOg, 5 min) and a membrane particulate pellet was obtained after centrifugation (ll,OOOg, 4 min, Beckman microfuge). Membrane pellets were solubilized by heating for 2 min at 100°C in Tris/HCl(l25 mM, pH 6.8) containing sodium dodecyl sulfate (SDS) (I%),glycerol(20%),2-

’ Abbreviations used: Hepes, 4-(2-hydroxyethyl)-lpiperazineethanesulfonic acid; Mes, 4-morpholineethanesulfonic acid; EGTA, ethylene glycol bis(paminoethyl ether) N,N’-tetraacetic acid; TPA, 12-0tetradecanoylphorbol la-acetate; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis.

PROTEIN

PHOSPHORYLATION

563

mercaptoethanol (lO%),andbromphenol blue (0.01%). Protein content of the preparations was determined by the method of Bradford (13). Measurement of ssP,-labeled phosphoproteins in islets incubated with 1%&?etradecanoylphorbol B-acetate (TPA). In additional experiments islets were incubated in the supplemented incubation medium described above with the addition of glucose (1.7 mM) and TPA (2 pM) for 20 min and a membrane particulate pellet was prepared. Phosphorylation of endogenwus islet proteins with [yS8PJ4TP. Endogenous kinase activity was assayed by measuring the incorporation of 32Pi from [y32PjATP into islet cell particulate homogenates free of nuclei and cell debris, prepared from islets stimulated with 1.7 or 16.7 mM glucose. The net effect of activation of the kinase(s) in the present study was determined in the absence of protease and phosphatase inhibitors as calcium-dependent neutral protease activation of calcium-activated, phospholipid-dependent protein kinase has been described (14) and specific phosphoprotein phosphatases described for substrates phosphorylated by this protein kinase (15, 16). The reaction tube contained Mes, pH 7.5 (5 rmol), Mp (1.25 rmol), [y-=PJATP (2.5 nmol, lo6 cpm/nmol), and either cyclic AMP (0.28 pmol), calmodulin (1.5 pg), or a phosphatidylserine-rich fraction of brain phospholipid (4 pg) and unsaturated diolein (0.08 pg) in a final volume of 125 ~1. The lipid additions were sonieated in Mes buffer from dried chloroform solutions. A free calcium concentration of 40 pM, calculated from published stability constants as described previously (17), was maintained in all incubations. The reaction was initiated by the addition of 50 pg islet protein and continued from 10 s up to 3 min at 30°C. The incubation was terminated by the addition of boiling SDS, X-mercaptoethanol, glycerol buffer as described above and the reaction mixture boiled for a further 2 min. Sodium dodecgl sulfate/polyacrylamide gel eledrophcvesis (SDS/PAGE), detection, and quantitation of 99P-kbeled phosphoproteins. Proteins were fractionated by SDS/PAGE employing 4% stacking gels and 10% resolving gels using the method of Laemmli (18). Standard proteins, phosphorylase b (molecular weight, 97,000), bovine serum albumin (67,000), ovalbumin (43,000), carbonic anhydrase (30,000), and soybean trypsin inhibitor (20,100) were resolved concurrently. Gels were stained in Coomassie blue R-250 (0.02% in 50% trichloroacetic acid), destained, and dried by heating in vacua. Autoradiographs were obtained using Kodak XRP-1 film and Cronex Hi-Plus intensifying screens (DuPont). The radioactivity of specific protein bands was estimated by excising the bands and mincing with Soluene (4 h, 2O’C). Aquasol scintillation fluid was added and radioactivity determined by liquid scintillation counting.

564

DUNLOP

AND

LARKINS

stained for protein abundance indicated that these phosphorylations did not correspond to major proteins in the preparation. Table I indicates the extent of phosphorylation and further shows that in the presence of TPA, that was enhanced RESULTS phosphorylation of endogenous proteins M, 80,40, and 29 in the particulate fraction in The effect of exposure of intact neonatal islets prelabeled with 32Pi to 16.7 mM glua manner similar to that seen on glucose No effect of TPA on phoscose for 15 min on the phosphorylation of stimulation. phorylation of M, 130 and M, 58 proteins proteins in a postnuclear particulate preparations of intact islets is shown in Fig. 1. was seen, and there was a decrease rather Phosphorylation of several endogenous than an increase in phosphorylation of the M, 100 protein in response to TPA. proteins was consistently enhanced following exposure to high glucose concenThis acute exposure to elevated glucose and to TPA stimulated insulin release (0.87 tration. The proteins in which there was f 0.40 rig/IO islets/l5 min, I.7 mM &COW significantly greater phosphorylation following exposure of islets to 16.7 mM comversus 3.42 + 0.47 and 3.76 + 0.42 for 16.7 pared with 1.7 mM glucose were of apparent mM glucose and 2 PM TPA, respectively, P molecular masses 130,000 + 1010, 100,000 < 0.001). Endogenous ATP levels reached + 700, 80,400 f 890, 58,100 + 1200, 39,800 steady state by 6 h of 32P incubation and * 700, and 29,400 f 700 Da (Mr 130,100,80, were maintained over the 24-h incubation 58, 40, and 29). The corresponding gels period, 20,000 + 600 dpm/400 islets. At the end of the acute stimulatory period 17,200 + 900 (86% ) of endogenous 32P-labeled ATP remained. CB j2P M, x10-' Direct study of the effect of various cofactors for protein phosphorylation was 130undertaken by examining the transfer in 10 s of 32Pi from [Y-~‘P]ATP to proteins in 100-97 a postnuclear particulate preparation of 80pancreatic islets. In Fig. 2 (top), it can be -67 58seen that in the presence of 40 PM free calcium but in the absence of other cofactors there was slight phosphorylation of proteins of calculated molecular masses 79,400 f 600, 57,100 + 700, 43,200 + 400, 39,600 +- 400, and 29,100 + 400 Da. A consistent finding was an apparent shift in the mo167 17 167 Glucose (mMI 17 lecular mass of exogenously labeled substrates compared to those labeled with 32Pi FIG. 1. Endogenous phosphorylation of neonatal rat islet proteins in response to glucose. Islets prelabeled in culture. It is assumed that the exogewith “Pi were stimulated with glucose (1.7 or 16.7 nously labeled proteins of M, 79,57, and 29 mM, 15 min), a crude particulate fraction was prerepresent the endogenously labeled subpared, and proteins were separated on SDS-polystrates of M, 80,58, and 29, respectively. A acrylamide gels, as described under Experimental shift in electrophoretic mobility is known Procedures. Proteins were stained with Coomassie to accompany phosphorylation abundance blue (CB, right panels) and phosphorylated proteins (19), and this probably explains the appardetected by autoradiography (left panels). Apparent ent slight discrepancy in M, values between molecular weights (Afr X 10m3) of specific phosphoproendogenously and exogenously phosphorteins in a representative experiment are indicated on ylated proteins. The mean value for M, has the left axis as calculated from the molecular weight of standard proteins, electrophoresing at the positions therefore been employed for further deindicated on the right axis. scriptions of the phosphoproteins. In the CeUuJar [=PjAZ’P content. Following addition of unlabeled ATP (1 mg/ml) to islet cell homogenates =P content of ATP was determined by thin-layer chromatography on Whatman No. 1 paper in isobutyric acid:water:ammonia:EDTA (0.2 M) 100:56:4.2:1.

GLUCOSE-INDUCED

PHOSPHOLIPID-DEPENDENT

PROTEIN

TABLE

565

PHOSPHORYLATION

I

ENDOGENOUSPHOSPHOR~ATIONOFNEONATALRATISLETPROTEINSINAMEMBRANE PARTICULATE PELLET INRESPONSETO GLUCOSEANDTPA STIMLJLATIONOF~NTAKE ISLETS “*Pi

content

(cpm

X 10m3)

Glucose Phosphoprotein (M, x 10-3) 130 100 80 58 40 29

1.7 1.40 0.87 0.24 0.91 0.65 2.81

mM f f + + + f

0.22 0.06 0.06 0.20 0.11 0.20

16.7 mM 2.84 2.41 0.80 2.12 1.32 4.20

+I f + f +k

0.24*** 0.57* 0.23* 0.24 ** 0.18* 0.28***

TPA 1.34 0.36 0.96 0.84 1.42 3.81

+ f f f f IL

(2

/iM)

0.17 O.OS* 0.09*** 0.21 0.20* 0.28***

Islets prelabeled with 32Pi were stimulated with glucose (1.7 or 16.7 mM, 15 min) or TPA (2 pM) and the proteins of a postnuclear membrane particulate fraction (ll,OOOg, 4 min) separated by SDS/PAGE. Following autoradiography the phosphorus content of the excised proteins was determined. The values shown are means f SE (n = 6 observations). Statistically significant differences when compared to 32Pi content in the presence of 1.7 mM glucose are indicated by *P < 0.005, **P < 0.01, and ***P < 0.005, respectively.

Note.

presence of cyclic AMP, under the conditions studied, there was no significant increase in phosphorylation of these proteins, although there was an insignificant rise in the =Pi incorporation into the iW, 58 protein (Table II). In the presence of calmodulin (in addition to 40 pM Cazf) there was enhancement of phosphorylation of the M,. 80 and lW, 58 proteins. In the presence of phosphatidylserine and diolein (in addition to 40 PM Ca’+) there was enhancement of phosphorylation of the M, 80 and llf, 58 protein, and, in addition, there was enhanced phosphorylation of a protein of apparent molecular mass 40,500 f 600 Da (mean + SE, six observations). This protein appeared to be distinct from that of M, 39,600 but was not easily separated from the latter protein (Fig. 2). The proteins of the 39,600 to 40,500 molecular weight region have therefore been combined in further calculations, and referred to as the M, 40 protein. It is possible that they correspond to the M,. 40 (39,800 + 700) protein(s) phosphorylated endogenously. To clarify whether the calcium-activated phospholipid-dependent protein kinase is involved in the endogenous phosphorylation of proteins following exposure of pancreatic islets to high glucose concentra-

tions, the effect of preincubation of intact islets in 16.7 mM glucose for 15 min prior to study of transfer of 32Pi from [32P]ATP to endogenous substrates in the postnuclear particulate preparation in the absence and presence of the various protein kinase cofactors was examined. It can be seen that prior exposure of islets to high glucose concentration prevented the subsequent enhancement of phosphorylation of the M, 40 protein by the addition of the cofactors of the Ca2+-activated phospholipid-dependent protein kinase, but did not prevent the enhanced phosphorylation by these cofactors of the M, 80 and M, 58 proteins (Fig. 2). In contrast, prior exposure to high glucose concentration prevented the enhancement of phosphorylation of the M, 80 and M,. 58 proteins by calmodulin (Fig. 2). The decrease in the extent of cofactor-stimulated exogenous phosphorylation would be consistent with prior phosphorylation of the protein substrates for the Ca2+-activated phospholipid-dependent kinase (M,. 40) and Ca2+/calmodulin-dependent kinase (Mr 80 and 58). Figure 3 indicates that phosphorylation abundance in the endogenously labeled M, 40 protein is maximal 2-3 min following glucose stimulation of intact islets and de-

566

DUNLOP

M, 80

Protein

M, 58

Protem T

AND

LARKINS

creases after this time. This subsequent decrease in protein phosphorylation was seen for a number of islet proteins and may indicate endogenous glucose-induced stimulation of Ca’+-dependent phosphoprotein phosphatases or protease activity. The degree of phosphorylation of these proteins was not affected by the inclusion of 50 PM phenylmethylsulfonyl fluoride together with 1 mM EDTA during the membrane preparation (results are not shown), indicating an intracellular effect. At 15 min following exposure to stimulatory glucose concentration, enhanced endogenous phosphorylation of M, 40 protein was still evident compared to that seen at nonstimulatory glucose concentration (1.7 mM). This suggests that the apparent prevention of transfer from [T-~~P]ATP in particulate preparations prepared from glucose-stimulated islets illustrated in Fig. 2 is not due to phosphatase activation, and instead supports prior phosphorylation of the M, 40 protein as an explanation. DISCUSSION

Mr 40

Two experimental approaches were followed to determine whether the previously described alterations in islet phospholipids following exposure to glucose were associated with stimulation of the phosphorylation of endogenous protein substrates. In the first approach, it was shown that exposure of intact islets, prelabeled with “Pi, to a stimulatory glucose concentration led to enhanced phosphorylation of a number of substrates, M, 130,100,80,58,40, and 29 (endogenous phosphorylation). In the same preparation, the phorbol ester TPA, which can substitute for diacylglycerol in the

Protem

4

FIG. 2. Calcium ion, cyclic AMP, Caa+/calmodulin, and calcium/phospholipid-dependent protein phosphorylation in particulate preparations of cultured neonatal rat islets. Proteins of a crude particulate fraction from islet homogenate were phosphorylated in an exogenous protein phosphorylation assay (10 s in the presence of [Y-~]ATP, 50 pmol/pg protein) as described under Experimental Procedures. Phosphorylated proteins were separated by SDS/PAGE.

The resulting autoradiograph is shown in the top panel. Indicated are apparent molecular weights (Af, X 10-a of the proteins phosphorylated in the presence of calcium, lane 1; calcium and cyclic AMP, lane 2; calcium and calmodulin, lane 3; and calcium phosphatidylserine and unsaturated diolein, lane 4. The effect of glucose exposure (1.7 and 16.7 mM, in open and hatched columns, respectively) prior to exogenous phosphorylation on the phosphorylation of proteins, excised from the gel for solubilizing and arP determination, is shown in the lower panel. Values are means f SE for four determinations.

GLUCOSE-INDUCED

PHOSPHOLIPID-DEPENDENT

PROTEIN

TABLE ACTIVATION

II

OF PROTEIN KINASES IN A MEMBRANE OF NON-GLUCOSE-STIMULATED “Pi

transferred

from

PARTICULATE ISLETS

[T-~~P]ATP

80

Calcium (free Ca2+, 40 PM) Calcium + CAMP Calcium + calmodulin Calcium + phosphatidylserine/ 1,2-diolein

FRACTION

(pmol/min/mg

Phosphoprotein Addition

567

PHOSPHORYLATION

islet

protein)

(A4,)

58

43

40

29

12.4 + 1.4 12.0 r 1.9 21.0 * 2.1**

11.1 + 2.1 13.2 f 2.5 17.7 + 2.4*

4.4 + 1.2 4.7 f 1.1 5.4 f 1.0

6.4 f 1.7 7.1 + 1.9 7.2 f 2.1

4.1 f 1.7 4.2 + 1.4 5.4 +- 1.9

20.1 + 2.3*

17.0 + 2.2*

3.6 + 1.0

12.5 + 1.2**

5.7 f 2.1

Note. Proteins of a postnuclear membrane particulate fraction (ll,OOOg, 4 min) were labeled by exposure to [y-32P]ATP in the presence of the cofactors indicated. Phosphorus content of excised proteins was determined following separation by SDS/PAGE and autoradiography. The values shown are means + SE (n = 7-10 observations). Statistically significant differences were compared to phosphorylation in the presence of free 40 pM calcium are indicated by *P < 0.05 and **P i 0.01, respectively.

stimulation of Caz+-activated, phospholipid-dependent protein kinase, enhanced the phosphorylation of the M, 80,40, and 29 proteins. This suggests these proteins may be substrates for Caz+-activated phospholipid-dependent protein kinase activity in the intact islet.

0

5 Time

10

15

(men 1

FIG. 3. Time course of endogenous phosphorylation of phosphoprotein M, 40 in glucose-stimulated neonatal islets prelabeled with “Pi. Islets prelaheled with =Pi were stimulated with glucose (16.7 mM) at the indicated times over a period of 15 min. The proteins of crude particulate preparations were separated by SDS/PAGE. Following autoradiography the area of gel corresponding to the phosphoprotein ilf, 40 was excised and solubilized and the 9 content determined. Values shown are means f. SE for six observations.

In the second approach, the phosphorylation studies were performed using [32P]ATP in a postnuclear particulate preparation (exogenous phosphorylation) in the absence and presence of cofactors for the various protein kinases. M, 80 and 58 proteins were shown to undergo enhanced phosphorylation when Ca2+ and calmodulin were present, whereas M, 80, 58, and 40 proteins underwent enhanced phosphorylation when Ca”, phosphatidylserine, and diolein were present. This suggests that M, 80 and 58 proteins could serve as substrates for both Ca2+/calmodulin and Ca2+-activated, phospholipid-dependent kinase activity, and that the M, 40 protein was specific for the latter kinase. The experiments where these studies were performed in particulate fractions from islets previously exposed to glucose suggested strongly that, for the M, 80 and 58 proteins, only the phosphorylation sites for Ca2+/ calmodulin-dependent kinase were occupied in the intact islet following glucose exposure; the phosphorylation sites for the Ca2+-activated, phospholipid-dependent kinase were still available for exogenous phosphorylation in the presence of the appropriate cofactors for the enzyme. In contrast, the M,. 40 protein could no longer undergo enhanced phosphorylation following the addition of the cofactors for the Ca2+-

568

DUNLOP

AND

activated, phospholipid-dependent kinase, suggesting that glucose had been effective in stimulating the changes in Ca’+, diacylglycerol, and phospholipid necessary to activate this enzyme in the intact cell, and that the phosphorylation sites on this protein were occupied in the intact cell prior to the exogenous phosphorylation experiments. The time-course studies suggest the presence of glucose-activated phosphatase activity, but exclude this as a cause for the blocking of subsequent exogenous phosphorylation of the M,. 40 protein by prior glucose exposure of the islets. It is apparent that the two approaches were consistent in indicating that exposure of islets to glucose enhanced phosphorylation of a iV, 40 protein, apparently a specific substrate in this system for the Ca’+activated phospholipid-dependent protein kinase. The experiments in the endogenous phosphorylation system showed that TPA, an activator of this kinase, enhanced in addition the phosphorylation of M, 80 and 29 proteins. The experiments using the exogenous phosphorylation technique did not provide evidence that these latter substrates were phosphorylated by this enzyme in response to glucose. This apparent discrepancy may be due to other actions of TPA in the intact islet such as mobilization of Ca2+ (20) with secondary effects on protein phosphorylation or to compartmentalization of the effect of the Ca2+-activated phospholipid-dependent kinase following glucose stimulation. Alternatively, reliance on exogenous diolein for activation may be lessened if the kinase and substrate have a close membrane association. It is of interest that Brocklehurst and Hutton (10) have shown that combination of insulin granules with the soluble fraction of a rat islet cell tumor homogenate resulted in Ca’+-dependent phosphorylation of 100,000, 29,000, and 10,000 Da proteins, the phosphorylation of the 29,000 Da granule membrane protein being consistent with the action of Ca-activated, phospholipid-dependent kinase. With respect to the M, 80 protein, not described previously as a specific islet phosphoprotein, it is possible that it may reflect the autophosphorylated Ca2+activated, phospholipid-dependent kinase as described by Kikkawa et al. (21) and Cachet et al. (22). The phosphoprotein M,

LARKINS

58 may represent the protein(s) of molecular weight 55,000-60,000, shown to respond to cyclic AMP and Ca2+ and calmodulin in many studies (2-4, 9, 23). Whatever the significance of the apparent stimulation of phosphorylation the M,. 80 and 29 proteins by TPA, it is apparent from all the approaches used that a protein of M, 40 serves as an endogenous substrate for Ca2+-activated, phospholipid-dependent kinase activity stimulated by exposure to high glucose concentrations in the neonatal pancreatic islet. The identity of this M, 40 is not known. Conditions optimal for the demonstration of Ca’+-activated, phospholipid-dependent kinase have shown phosphorylation of proteins of 41,000 Da in cerebral cortex (24), 38,000 Da in pancreatic acinar and islet tissue (9,25), 41,000 Da in solubilized fractions of pig spleen (26), 38,000 Da in cardiac tissue (27), and 40,000 Da in platelet (28) and fetal membranes (29), which may be consistent with a common substrate for this kinase in these tissues. The rapid shape changes mediated by the cytoskeleton and induced by platelet activating factor in human platelets are paralleled by phosphorylation of a 40,000 Da protein and are dependent on phospholipase C-induced phospholipid turnover (28). With respect to islet cell function it is of interest to speculate on the identity of two possible protein substrates. The phosphorylation of lipomodulin, a protein of 40,000 Da, may follow stimulation by TPA and glucose. In the unphosphorylated form this protein, described by Hirata (30), inhibits the activity of phospholipase A2 but is inactivated by phosphorylation. We have demonstrated the rapid activation of phospholipase A2 in the islet following glucose stimulation (11). Indirect support comes from the finding that corticosterone at lo-? mol/liter, believed to induce lipomodulin in other tissues (30), is inhibitory to insulin release in islet cells. No increase in the phosphorylated form of the M,. 40 protein could be shown in islets after treatment with this steroid (results not shown). In addition, Ca-activated phospholipid-dependent protein kinase has been shown to phosphorylate the inhibitory guanine-nucleotide-binding regulatory component of adenylate cyclase (Mr 41,000) and inhibit its function in human platelet

GLUCOSE-INDUCED

PHOSPHOLIPID-DEPENDENT

membranes (31). In previous studies we have shown that inactivation of this component of adenylate cyclase by pertussis toxin-induced ADP ribosylation will enhance glucose-induced insulin release (32). The membrane preparation used in the present study included the extensive internal membrane systems of the nucleated cell, including endoplasmic reticulum. The M, 40 protein is associated with this fraction after glucose exposure and activation of the kinase. This study confirms that an increase in cytosolic Ca2+ in the p cell may promote enhanced protein phosphorylation. One substrate has been demonstrated for Ca2+activated, phospholipid-dependent protein kinase which is phosphorylated following glucose stimulation and present in a particulate fraction. Previously it has been shown that glucose can cause a rapid appearance of diacylglycerol in neonatal cultured islets (11). This study shows that the substrate protein can be rapidly phosphorylated on exposure to an unsaturated diacylglycerol in the presence of Ca2+ and phosphatidylserine and suggests that endogenous phosphorylation mediated by Ca2+-activated, phospholipid-dependent protein kinase may be an early event in glucose-induced insulin release. ACKNOWLEDGMENT This project was supported by the National and Medical Research Council of Australia.

Health

REFERENCES 1. SCHUBART, IJ. K., FLEISHER, N., AND ERLICHMAN, J. (1980) J. BioL Chem. 255,11063-11066. 2. LANDT, M., MCDANIEL, M. L., BRY, C. G., KOTOGAL, N., COLCA, J. R., LACY, P. E., AND MACDONALD, J. M. (1982) Arch. B&hem Biophys. 213, 148154. 3. HARRISON, D. E., AND ASHCROFT, S. J. H. (1982) B&him. Biophys. Actu 714,313-319. 4. GAGLIARDINO, J. J., HARRISON, D. E., CHRISTIE, M. R., GAGLIARDINO, E. E., AND ASHCROFT, S. J. H. (1980) Biochem. J. 192,919-927. 5. MACDONALD, M. J., AND KO~LURU, A. (1982) Diabetes 31, 566-570. 6. TANIGAWA, K.,KUZUYA, H., IMURA, H.,TANAGUCHI, H., BABA, S., TAKAI, Y., AND NISHIZUKA, Y. (1982) FEBS I.&t. 138,183-186. 7. TAKAI, Y., KISHIMOTO, A., KIKKAWA, U., MORI, T., AND NISHIZUKA, Y. (1979) B&hem. Biophys. Res. Commun. 91.1218-1224.

PROTEIN

PHOSPHORYLATION

569

8. MORI, T., TAKAI, Y., Yu, B., TAKAHASHI, J., NISHIZUKA, Y., AND FUJIKURA, T. (1982) J. Biochem. 91,427431. 9. HARRISON, D. E., ASHCROFT, S. J. H., CHRISTIE, M. R., AND LORD, J. M. (1984) Eccperientia 40, 1075-1084. 10. BROCKLEHURST, K. W., AND HUTTON, J. C. (1983) Biochem. J. 210,533-539. 11. DUNLOP, M. E., AND LARKINS, R. G. (1984) Biochem Biophys. Res. Commun 120,820-827. 12. HELLERSTROM, C., LEWIS, N. J., BORG, N., JOHNSTON, R., AND FREINKEL, N. (1979) Diabetes 28, 769-776. 13. BRADFORD, M. (1976) Anal. Biochem. 72,248-254. 14. KISHIMOTO, A., KAJIKAWA, N., SHIOTA, M., AND NISHIZUKA, Y. (1983) J. BioL Chem. 258,11561164. 15. SAHYOUN, N., LEVINE, H., MCCONNELL, R., BRONSON, D., AND CUATRECASAS, P. (1983) J. BioL Chem. 80,6760-6764. 16. LEVINE, H., SAHYOUN, N., MCCONNELL, R., BRONSON, D., AND CUATRECASAS, P. (1984) B&hem. Biophys. Res. Cmnmun 118.278-283. 17. DUNLOP, M. E., AND LARKINS, R. G. (1984) J. BioL Chem. 259,8407-8411. 18. LAEMMLI, U. C. (1970) Nature (London) 227, 680685. 19. WEIGEL, N. L., TASH, J. S., MEANS, A. R., SCHRADER, W. T., AND O’MALLEY, B. W. (1981) B&hem Biophys. Res. Commun 102,513-519. 20. MALAISSE, W. J., LEBRUN, P., HERCHUELZ, A., SENER, A., AND MALAISSE-LAGAE, F. (1983) Endocrinology 113,1870-1877. 21. KIKKAWA, U., TAKAI, Y., TANAKA, Y., MIYAKE, R., AND NISHIZUKA, Y. (1983) J. BioL Chem. 258, 11442-11445. 22. COCHET, C., GILL, G. N., MEISENHELDEZ, J., COOPER, J. A., AND HUNTER, T. (1984) J. BioL Chem. 259, 2553-2558. 23. KOWLURU, A., AND MCDONALD, M. J. (1984) B&hem. Biophys. Res. Commun. 118,797-804. 24. WRENN, R. W., KATOH, N., WISE, B. C., AND Kuo, J. F. (1980) J. BioL Chem 255,12042-12046. 25. WRENN, R. W. (1983) Life Sci. 32,2385-2392. 26. SCHATZMAN, R. C., RAYNOR, R. L., FRITZ, R. B., AND Kuo, J. F. (1983) Biochem. J. 209.435-443. 27. KATOH, N., WRENN, R. W., WISE, B. C., SHOJI, M., AND Kuo, J. F. (1981) Proc. NatL Ad Sci USA 784813-4817. 28. LAPETINA, E-G., AND SIEGEL, F-L. (1983) J. BioL Ch,em. 258.7241-7244. 29. OKAZAKI, T., BAN, C., AND JOHNSTON, J. M. (1984) Arch Biochem Biophys. 229.27-32. 30. HIRATA, F. (1981) J. BioL Chem. 256,7730-7733. 31. KATADA, T., GILMAN, A. G., WATANABE, Y., BAUER, S., AND JAKOBS, K. H. (1985) Eur. J. Biochem. 151,431-437. 32. DUNLOP, M. E., AND LARKINS, R. G. (1983) Biochem. Biophys. Res. Commun. 112,684-692.