Biochimica et Bio~hysica Acta, 1091 (1991) 8i-86 © 1991 Elsevier ~cience Publisher.~, B.V. (Biomedical Division) 0167-4889/91/$03.50 ADONIS 0!6748899100056Z

81

BBAMCR 12825

The 38 kDa C a 2 +/membrane-binding protein of pig granulocytes needs a high C a 2+ concentration to be phosphorylated by protein

kinase C

Gy/Sngyi Farkas., Lfiszl6 Buday, Ferenc Antoni and Anna Farag6 1st Institute of Biochemistry, Semmelweis University Medical School, Budapest (Hungary)

(Received 20 February 1990) (Revised manuscript received 9 June 1990)

Key words: Calcium ion/membrane binding protein, 38 kDa; Lipocortin phosphorylation; Protein kinase C; Isoenzyme; Phospholipid-dependent calcium-independent protein kinase; (Granulocytes); (Pig)

The 38 kDa Ca2+/membrane-binding protein reported to be the dominant substrate of protein kinase C in the extracts of pig neutrophil granulocytes was purified partially and its phosphorylation was investigated. In pig wanulocytes type II protein kinase C was the major isoform, while type Ill isoenzyme was present only as a minor activity. Phosphorylation of the 38 kDa protein was performed with rat brain protein kinase C. Each of the three isoenzymes purified from rat brain was able to phosphorylate this protein, though on the conditions used in our experiments it was phosphorylated most intensively by type II protein kinase C. A phospholipid-dependent, but CaZ+-independent, form of protein kinase C was demonstrated with the aid of a synthetic oligopeptide substrate. Phosphorylation of the 38 kDa protein by the CaZ+-independent enzyme proceeded exclusively in the presence of Ca 2+. The Ca z+ concentration necessary for the phosphorylation of the 38 kDa protein by either form of protein kinase C was by orders of magnitude higher than that required for the activation of protein kinase C.

Introduction

A f.'.trmly of C a 2 +-dependent membrane binding pro- teins has been recognized recently. Two of the members of thi.,; family with a molecular mass between 35 kDa and 39 kDa have been named lipocortins or calpactins (lipocortin I is identical to calpactip II and lipocortin II is identical to calpae.tin I). Lipocortins may be im- plicated in the control of secretion and cell motility (reviewed in Refs. 1-3), and since they are known to be the substrates of some protein tyrosine kinases they may also be involved in the regulation of cell growth [4].

Protein kinase C (PKC), that takes part in the regu- lation of a number of cellular events (re,dewed in Refs. 5-7), also phosphorylates these proteins in vitro and in vivo [8-13]; however, the effect of phosphorylation by protein kinase C on the biological activity of lipocortins is not known exactly. Clarification of the conditions

Abbreviation: PKC, protein kinase C.

Correspondence: A. Faragb, 1st Institute of Biochemistry, Sem- melweis Univershy Medical School, PO Box 260, 1A.A.'I. Budapest 8, Hungary.

where this phospherylation proceeds may help to dis- cover its function.

Previously. we demonstrated that in the extracts of pig granulocytes a 38 kDa Ca2+/membrane-binding protein was the dominant substrate of protein kinase C [14]. Here we demonstrate that this 38 kDa protein itself needs Ca 2+ for the phosphorylation by protein kinase C and the Ca 2+ concentration necessary for this phosphorylation is with orders of magnitude higher than that necessary for the activation of the enzyme.

Materials and Methods

Preparation of extract from granulocytes Neutrophil granulocytes were isolated from pig blood

and sonicated ((2-3). 10 7 cells per ml) in a lysing medium containing EGTA as described previously [15]. The suspension was centrifuged at 100000 × g fol 30 min and the 38 kDa protein and protein kinase C isoenzymes were purified from the supernatant.

Purification of the 38 kDa protein 30 ml of the cellular extract was loaded onto a

DEAE-cellulo~,e column equilibrated with 5 mM potas- sium phosphate (pl-i 7.5). Tile flow-through fraction

82

was applied immediately to a hydroxyapatite column (15 × 40 mm). The column was washed with 5 mM potassium phosphate (pH 7.5) and proteins were eluted by 200 mM potassium phosphate. About 4 mi fractions were collected and 0.1 ml samples were investigated for the presence of the 38 kDa protein kinase C substrate. This protein was demonstrated by sodium dodecyl- sulphate polyacrylamide gel ¢lectrophoresis (SDS- PAGE) and autoradiography after phosphorylation (32p-labelling) by protein kinase C.

Separation of protein kinase C isoenzymes from the ex- tract of granulocytes

10 ml of the freshly prepared extract was diluted two times with 0.5 mM potassium phosphate (pH 7.5), the solution was filtered (Sartorius membrane 11307) and applied immediately to a Bio-Gel HPHT column (7.8 x 100 ram) connected to a HPLC system (DuPont 850). The column was washed with 10 ml of 5 mM potassium phosphate (pH 7.5) containing 1 mM dithiothreitol, and the isoenzymes were eluted with a linear concentration gradient of potassium phosphate. The chromatographic positions of type I, II and III isoenzymes of protein kinase C from rat brain were determined earlier in this system [16]. 0.4 ml fractions were collected and assayed immediately for protein kinase C activity.

Preparation of protein kinase C from rat brain Three rat brains were homogenized in 30 ml of a

solution comprising 10 mM EGTA, 2 mM EDTA, 2 mM phenylmethylsulphonyl fluoride (PMSF), 0.01% leupeptin, 10 mM benzamidine, 1 mM dithiothreitol (DTI'), 40 mM KCI, 20 mM Tris-HCl (pH 7.5), 250 mM sucrose and 10 -s M cAMP. The suspension was centrifuged at 40000 x g for 30 min. The supernatant was diluted with 30 ml of 5. mM potassium phosphate (pH 7.5) (containing 2 mM PMSF, 10 mM benzamidine and 1 mM DTT) and applied to a DEAE-Sephacel column (20 x 50 ram) The column was washed and PKC was eluted with a "inear concentration gradient of potassium phosphate (pH 7.5) containing I mM EGTA, 1 mM EDTA, 1 mM benzamidine and 1 mM DTT. 4,5 ml fractions were collected.

Fractions of the PKC peak were combined, potas- sium phosphate (pH 7.5) was added to the solution to 0.75 M final concentration and it was applied im- mediately to a phenyl-Sepharose column (10 x 30 ram). The column was washed with 0.75 M potasssium phos- phate and elution of PKC was performed with a de- creasing concentration gradient of potassium phosphate from 0.75 to 0.005 M. 4.5 ml fractions were collected. In several experiments (as indicated in the text) enzyme preparations obtained at this step of purification were used.

For separation of PKC isoenzymes the peak fractions eluted from phenyl-Sepharose were combined, the

potassium phosphate concentration was decreased to about 20 mM, the solution was completed with glycerol (10% v/v), 0.5 mM EGTA, 0.5 mM EDTA, 1 mM benzamidine and 1 mM DTT, and applied to a hydroxyapatite column (Bio-Gel HT, Bio-Rad) (140 x 100 mm). The isoforms of PKC were eluted with a linear concentration gradient of potassium phosphate (pH 7.5) (150 ml, 5 mM-250 mM) containing glycerol, EGTA, EDTA, DTT and benzamidine as above. 1.5 ml fractions were collected.

Assay of protein kinase C activity Assays (at 37°C for 10 rain) were carried out in 200

/~1 of a mixture comprising 50 mM Tris-HC! (pH 7.5), 10 mM MgCI2, 0.01 mM [),- 32p]ATP (about 250000 cpm per reaction mixture), 0.7 mg/ml of the synthetic nonapeptide substrate Ala-Ala-Ala- Ser-Phe-Lys Ala- Lys-Lys-amide [17] or 0.5 mg/ml of H1 histone (Sigma type IIIS) and 0.05 ml enzyme solution. Radioactivity incorporated into the oligopeptide substrate was mea- sured as described in Ref. 18 and the same method was used for measuring the radioactivity incorporated into histone H1.

Phosphorylation of the 38 kDa protein Phosphorylation was performed in 0.4 ml of a mix-

ture comprising 50 mM Tris-HC! (pH 7.5), 10 mM MgCI 2, 0.01 mM [),-32p]ATP ((1-2). 106 cpm per reac- tion mixture), 20 #g /ml phosphatidylserine + 50 ng/ml of diacylglycerol, Ca -'+ as indicated, 0.1 ml sample of a protein kinase C preparation and 0.1 ml sample of the partially purified protein. The phosphorylation (5 rain at 37°C) was stopped by the addition of trichloroacetic acid (5% final concentration), ~nd bovine serum al- bumin (10 #g per sample). After centrifugation the precipitate was dissolved for SDS-PAGE. SDS-PAGE was performed on 15~ polyacrylamide gels calibrated with a Pharmacia electrophoresis calibration kit.

Results and Discussion

Phosphorylation of the 38 kDa protein by different isoen- zymes of protein kinase C

As repc<ted previously in the extracts of pig neu- trophil granulocytes, a 38 kDa Ca2+/membrane-bind. ing protein was the dominant substrate of protein kinase C [14]. A part of the 38 kDa protein was bound to DEAE-cellulose, while the other part was not adsorbed. This latter component was purified further on a hy- droxyapatite column and its phosphorylation was in- vestigated.

Protein kinase C is known to be a family of distinct, but closely related isoenzymes [6,7]. The isoforms, which have been separated on hydroxyapatite, are designated as type I, II and III isoenzymes in the order of elution from hydroxyapatite [19]. Type I enzyme has been

found exlusively in the central nervous system [6,7]. Human neutrophil granulocytes contain mainly type III protein kinase C [20]. In the crude cytosolic extract of pig granulocytes type II protein kinase C was the domi- nant isoform, type III isoenzyme was present only as a minor activity (Fig. 1.). (The cytosolic extract contained about 7070 of total protein kinase C activity present in this cell type [15].) The isoenzyme pattern of pig neu- trophils resembled to that detected in bovine granulo- cytes [21].

The dominant character of type II PKC in the ex- tract of pig granulocytes suggests that, in these cells, mainly type II PKC is responsible for the phosphoryla- tion of the 38 kDa protein. However, in the present experiments the 38 kDa protein 'gas phosphorylated with the aid of rat brain PKC preparations. Therefore, we investigated its phosphorylation by each isoform separated from rat brain. The activity of the isoenzymes (Fig. 2.) was determined with the oligopeptide Ala-Ala- Ala-Ser-Phe-Lys-Ala-Lys-Lys-amide that is a selective substrate for PKC [17]. The apparent KM values of type I, II and Ill isoenzymes for the oligopeptide were 0.15 :t: 0.03 mg/ml, 0.0 o _+ 0.01 mg/ml and 0.23 _+ 0.03 mg/ml, respectively. The peptide ldnase activities mea- sured at 0.70 mg/ml peptide concentration were com- pared to the corresponding histone kinase activities determined at 0.5 mg/ml HI histone (Sigma IIIS) con- centration. The phosphotransferase activity was higher with the peptide than with H1 histone in each case, but type II PKC showed a higher relative histone kinase activity than the ~ r forms. The ratios of peptide kinase activity to histone kinase activity were 5.2 (n = 2) 2.7 + 0.5 (n -- 5) and 4.7 + 0.7 (n = 5) for type I, II and Ill isoenzymes, respectively (n is the number of pre- parations investigated).

> , 100 - .-,= :E

O 0.2 0 O .C:

o } : : ) , - O ,,- x E.. r- 1:: o c~ 01 ~.~ t - ¢0 t~

O ~ C3 r 0

t - O .

10 20 30 40 50

frochon number

Fig. 1. The dominant character of type II protein kinase C activity in the cytosofic extract of pig granulocytes. The isoenzyme profile was obtained by chromatography of the cytosolic e;ttract on a Br'~-Oel HPHT column connected to a HPLC system. The activity of !?KC was determined with the oligopeptide Ala-Ala-Ala-Ser-Phe-Lys-Ala- Lys-Lys-amide as a substrate in the presence of 0.5 mM EOTA (o), 0.5 mM EGTA+25 /~g/ml phosphatidylserine+50 ng/ml dia~yl- glycerol (x) and 0.5 mM CaCI~ +25 #g/ml phosphatidylserine-+. 50

ng/ml diacylglycerol (®).

83

~'- I. II. !

~ 70

4o o °

g 20

20 30 4O 50

f r o c t i o n n u m b e r

Fig. 2. Protein kinase C isoenzymes of rat brain. The isoenzyme profile was obtained by chromatography on a Bio-Gel HT column. The activity of PKC was measured with the oligopeptide substrate in the presence of 0.5 mM EGTA (o ) or in the presence of Ca2++ phosphatidylserine+diacylglycerol (O). For determination of kinc:ic parameters of the isoenzymes and phosphorylation of the 38 KUa protein, top fractions of peaks I and 11 and fractions 51 of peak III were used, respectively. Fraction 55 of peak Ill ~as also analyzed, the

results were identical to those obtained with fraction 51.

The phosphorylation of the 38 kDa protein was carried cut with enzyme samples of identical peptide kinase activity. It was phosphorylated more intensively by type ~I than by type Ill PKC, though each isoform was able ~o catalyze the phosphorylation (Fig. 3.) When the reactEon was carried out with type II and type Ill enzyme samples of identical histone kinase activity and not ~; identical peptide kinase activity, the difference in the intensity of phosphorylation was not significant (data no: shown). Our results indicate that the substrate specificities of the isotypes of PKC differ slightly, but the 38 kDa protein can be phosphorylated by either form of PKC.

Though type II and III isoforms were well separated from each other, experiments were performed to prove that the phosphorylation by the type Ill p~eparation was not due to a contamination by type II PKC. Arachidonic acid is known to activate PKC and type Ill PKC is activated to a higher extent than type II [6]. [n several experiments arachidonic acid (100 #M) was used to activate the enzyme instead of phosphatidylserine + diacylglycerol. In these cases the 38 kDa protein was phosphorylated more intensively by type III than by type II PKC (data not shown).

Phosphorylation of the 38 kDa protein with phospholipid- dependent but Ca 2 +.independent forms of protein kinase C

Protein kinase C is known to be extremely sensitive to limited proteolysis, which results Ca2+/phospho - fipid-independent forms of the enzyme [5-7,22,23]. In

84

4 5 6

6 7 -

4 2 -

3 0 -

2 0 -

Fig. 3. Phosphorylation of the 38 kDa Ca: +/membrane-binding pro- tvin of pig granulocytcs by different isoforms of rat brain protein kinasv C. Phosphorylation was performed in the presence of 0.5 mM EGTA (lanes 1, 2 and 3) or in the presence of 0.5 mM CaCI2 + 25 /~g/'ml phosphatidylserine + 50 ng/ml diacylglycerol (lanes 4, 5 and 6) by type I (lanes 1, 4), type II (lanes 2, 5) and type Iil (lanes 3, 6) protein kinase C. Each enzyme sample used for the phosphorylation of the 38 kDa protein had identical phosphotransferase activity (10

pmoi/min) with the synthetic oligopeptide as a substrate.

the e;aracts of pig granulocytes some Caa+/phospho- lipid-independent protein kinase C activity was always present [15], but in spite of this the 38 kDa protein was phosphorylated exclusively in the presence of Ca and

phospholipid [16]. This observation tempted us to in- vestigate in details the Ca 2+ requirement of this phos- phorylation.

With the aid of the synthetic peptide substrate, we previously demonstrated phospholipid-dependent, but Ca2+-independent or even Ca2+-inhibited forms of pro- tein kinase C in the crude extracts of brain [16,24]. Though among the protein kinase C isoenz~mes de- scribed in brain Ca2+-independent isoforms have also been suggested to exist [25-27], the Ca2+-independent form observed in our experiments seemed to be the product of degradation or denaturation of the Ca2+/phospholipid-dependent native enzyme. The Ca2+-independent form was not present in each pre- paration. Fig. 4 shows the chromatographic profiles of three preparations obtained after phenyI-Sepharose chromatography and used in the present experiments. (For these experiments the isoenzymes were, not sep- arated from each other.) One of the preparations used was absolutely Ca2+/phospholipid-dependent, while in the others less or more activity was measured in the absence of Ca '+. A fraction of the absolutely Caa+-de - pendent preparation was converted to a Ca2+-indepen- dent form, with the concomitant decrease in the total activity and a netto increase in the Ca2+/phospholipid independent activity when it was stored at -20°C for 3 days (Fig. 4 D,E). It is important to note that Ca2+-in - dependent but phospholipid-dependent activity could be detected with the aid of the peptide substrate even when this type of activity was not observed with HI histone as a substrate.

,,o ?

20

'°t ' ° o c:

L ;, 66 lo 1001 D R

[]

OL ~--~ r-I l

peptide H 1

50

30

2~

B

2 4 6 8 lO 100[

eo f E so

4o

2 0 ~ 0

peptide H I

so C

40

3o

20

I !

o i 4 6 e ~ o

£

0.6 'i ~2 R

Fig, 4, Phospholipid-dependent but Ca 2 +-independent protein kinase C activity in preparations obtained from rat brain. Parts A, B and C show chromatographic patterns of three preparations obtained by phenyl-Sepharose chromatography. About 4 ml fractions were collected and 0.05 ml samples were assayed with the oligopeptide substrate. The symbols of Fig. I are used. Parts D and E show the activity of fraction 5 from part A, immediately after preparation and "/2 h later, respectively. (It was stored at - 20 o C). Phosphotransferaae activity was measured in the presence of Ca 2+ +phosphatidylserine+diacylglycerol (striped columns); or EGTA (open columns); or EGTA+phosphatidylserine+diacylglycerol (dark columns), Phosphotransferase activities obtained with the oligopeptide or with HI histone (Sigma IllS) as substrates are expressed in per cent of

the initial peptide kinase activity measured in the presence of each activator.

A

94-

6 7 -

2 ....

3 0 -

2 3 4 5

:i~

20-

B

9 4 -

1 2 3 4 5

6 7 -

3 0 - ~

2 0 -

C

9 4 -

6 7 -

4 2 - -

3 0 - -

1 2 3 45 , ~

. . . . . ~-~

20

A

9 4

6 7 ~

4 2 m

2 0 -

2 3 4 5 6

85

B

4 .....

6 7 ~

4 2 u

3 0 ~

2O

i

Fig. 6. The Ca 2+ requirement of phosphoryladon of the 38 kDa protein. The 38 kDa protein was incubated with protein kinase C in the presence of 0.5 saM EGTA (lane 1); 5.10 -6 M Ca 2+ (lane 2); 10 -5 M Ca 2+ (luae 3); 5.10 -5 M Ca 2+ (lane 4); 10 -4 M Ca 2+ (lane 5) and 5.10 -4 M Ca 2+ (lane 6). 20/~g/ml phosphatidylserine and 50 ng/ml diacylglycerol were present in each incubation mixture. Parts A and B show results of two experiments carded out with different batches of the 38 kDa protein. Protein kinase C preparations

obtained after phenyI-Sepharose chromatography were used.

Fig. 5. Phosphorylation of the 38 kDa protein with phospholipid- dependent but Ca 2 +-independent forms of protein kinase C. A par- tially C~2+-independent protein kinase C preparation (combined 4-6 fractions ef C) was used for the phosphorylation of the 38 kDa protein and ~-I1 histone (Sigma IIIS) in parts A and B. respectively. and a completely Ca2+-independent preparation (fraction 8 of C) was used for the phospholTlation of the 38 kDa protein in part C. Incubations were performed in the presence of 0.5 mM EGTA (lane 1); 0.5 mM EGTA + phosphatidylserine + diacylglycerol (lane 2); 10-6 M Ca 2 + phosphatidylserine + diacylglycerol (lane 3); 0.5 mM CaC! 2 + diacylglycerol + phosphatidylserine (lane 4). Lane 5 shows the phos- phorylation of the enzyme preparations without added substrate in the presence of 0.5 mM CaCI 2 + phosphatidylserine + diacylglycerol.

86

The 38 kDa protein was phosphorylated by partially or completely Ca2+-independent enzyme fractions, but its phosphorylation was absolutely Ca2+-dependent in each case (Fig. 5). This suggested that the 38 kDa Ca2+/membrane-binding protein itself required Ca 2+ for phosphorylation by protein kinase C.

We compared the Ca 2+ concentrations needed for the activation of the native protein kinase C and for the phosphorylation of the 38 kDa protein. Under our experimental conditions (in the presence of 20 /tg/ml phosphatidylserine + 50 ng/ml diacylglycerol) the dif- ference was about two orders of magnitude. A protein kinase C preparation containing each isoenzyme without separation was practically saturated at 10-6 mol/l Ca 2 + concentration, when its activity was measured with H1 histone or with the synthetic peptide substrate. Under the same conditions phosphorylation of the 38 kDa protein was observed only when the Ca 2 + concentration was higher than 10 -s mol/l (Fig. 6.), and it was inten- sive at a Ca 2+ concentration higher than 10 -4 tool/1. The presence of phosphatidylserine was not essential for the phosphorylation of the 38 kDa protein, since its phosphorylation was observed at a Ca 2+ concentration of 5.10 -4 mol/l when protein kinase C (type III) was activated by arachidonic acid instead of phosphatidyi- serine + diacylglycerol (data not shown).

The Ca 2~ concentration required for the in vitro phosphorylation of the 38 kDa Ca2+/membrane-bind- ing protein seems to be very high. Presumably, inside the cell the phosphorylation may occur at a lower Ca 2 + concentration, but only when the cytosolic Ca 2+ con- centration is increased by extracellular signals and this protein is bound by Ca 2+ to the membrane. It is probable, however, that the phosphorylation of this protein does not occur or only to a very limited extent when the activity of PKC is stimulated by the treatment of cells with phorbol ester without a significant increase in the cytosolic Ca 2+ concentration. Our finding may explain the observations of Gould and co-workers [10], who have demonstrated a very low stoichiometry for the phosphorylation of iipocortin by PKC in cells stimu- lated by phorbol ester.

Acknowledgement

This work was supported in part by research grant OTKA 1/614/86.

References

1 Geisow, M.J. (1986) FEBS Lett. 203, 99-103. 2 Glenney, J.R. Jr. (1987) Biochem. Soc. Trans. 15, 798-800. 3 Klee, B.C. (1988) Biochemistry 27, 6645-6653. 4 Cooper, J.A. and hunter, T. (1985) Annu. Rev. Biochem. 54,

897-930. 5 Nishizuka, Y. (1988) Nature 334, 661-665. 6 Nishiz:.~ka, Y. (1989) Cancer 63, 1892-1903. 7 Kikkawa, U., Kishimoto, A. and Nishizuka Y. (1989) Annu. Rev.

Biochem. 58, 31-44. 8 Summers, T.A. and Creutz, C.E. (1985) J. Biol. Chem. 260, 2437-

2443. 9 Michener, M.L., Dawson, W.B. and Creutz, C,E. (1986),

10 Gould, K.L., Woodgett, J.R., lsacke, C.M. and Hunter, T. (1986) Mol. Cell. Biol. 6, 2738-2744.

11 Khanna, N.C., Tokuda, M. and Waisman, D.M. (1986) Biochem. Biophys, Res. Commun. 141, 547-554.

12 Schlaepfer, D.D. and Haigler, H.T. (1988) Biochemistry 27, 4253- 4258.

13 Varticovski, L., Chahwala, S.B., Whitman, M., Cantley, L., Schin- dler, D., Chow, E.P., Sinclair, L.K., Pepinsky, R.B. (1988) Bio- chemistry, 27, 3682-3690.

14 Buday, L., Farkas, G. and Farag6, A. (1989) Acta Biochim. Biophy~ Acad. Sci. Hung. 24, 101-106.

15 Buday, L., Sepr6di, J., Farkas, G., M6sz~iros, G., Romhfinyi, T., B~nhegyi, G., Mandl, J., Antoni, F. and Farag6, A. (1987) FEBS Lett. 223, 15-19.

16 Buday, L., M6sz~ros, G., Farkas, G., Sepr6di, J. Antoni, F. and Farag6, A. (1989) FEBS Lett. 249, 324-328.

17 Romh~nyi, T., Sepr6di, J., Antoni, F. M6szfiros G., Buday, L. and Farag6, A. (1986) Bichim. Biophys. Acta 888, 325-331.

18 Romh~nyi, T., Sepr6di, J., Antoni, F., M~sz~ros, G. and Farag6, A. (1985) Biochim. Biophys. Acta 827, 144-149.

19 Huang, K.-P., Nakabayashi, H. and Huang, F.U (1986) Proc. Natl. Acad. Sci. USA 83, 8535-8539.

20 Sekiguchi, K., Tsukuda, M., Ogita, K., Kikkawa, U. and Nishizuka, Y. (1987) Biochem. Biophys. Res. Commun. 145, 797-802.

21 Dianoux, A.-C., Stasia, M.-J. and Vignais, P.V. ~i989) Biochem- istry 28, 424-431.

22 Kishimoto, A., Mikawa, K., Hashimoh, K., Yas~tda l., Tanaka, S., Tomaniga, M., Kuroda, T. and Nishizuka Y. (1989) J. Biol. Chem. 264, 4088-4092.

23 Huang, F.L., Yoshida, Y., Cunha-Melo, J.R., Beaven, M.A. and Huang, K.-P. (1989) J. Biol. Chem. 264, 4238-4243.

24 Farag6, A., Farkas, G., M6sz~tros, G, Buday, L., Antoni, F. and Sepr6di, J. (1989) FEBS Left. 242, 328-332.

25 Ohno, S., Akita, Y., Konno, Y., Imajok, S. and Suzuki, K. (1988) Cell 53, 731-741.

26 Schaap, D., Parker, P.J., Bristol, A., Kriz, R. and Knopf, J. (1989) FEBS Left. 243, 351-357.

27 Ono, Y., Fujii, T., Ogita, K., Kikkawa, U., Igarashi, K. and Nishizuka, Y. (1989) Proc. Nat. Acad. Sci. 86, 3099-3103.