THE JOURNAL OF 269, No. 24, Issue of Inc. Printed in U.S.A ...THE JOURNAL OF BIOL~GICAL CHEMISTRY 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc. Vol. 269,

THE JOURNAL OF BIOL~GICAL CHEMISTRY 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 269, No. 24, Issue of June 17, pp. 16821-16828, 1994 Printed in U.S.A.

Protein Kinase C in Yeast CHARACTERISTICS OF THE SACCHAROMYCES CEREVISZAE PKCl GENE PRODUCT*

(Received for publication, January 7, 1994, and in revised form, March 23, 1994)

Bruno Antonsson, Sylvie Montessuit, Laurence Friedli, Mark A. Payton, and Gerhard ParaviciniS From the Glaxo Institute for Molecular Biology, 14 Chemin des Aulx, 1228 Plan-les-Ouates, Geneva, Switzerland

The Saccharomyces cerevisiae PKCl gene encodes a homolog of mammalian protein kinase C (Levin, D. E., Fields, F. O., Kunisawa, R., Bishop, J. M., and Thorner, J. (1990) Cell 62, 213-224). A protein of 150 kDa is recog- nized by a polyclonal antiserum raised against a trpE- Pkcl fusion protein. In subcellular fractionations, Pkclp associates with the 100,000 x g particulate frac- tion. This association is resistant to extraction with high salt concentrations, alkali buffer, or nonionic deter- gents, suggesting that Pkclp may be associated with a large protein complex. Pkclp modified at its COOH ter- minus with two repeats of the Staphylococcus aureus protein A IgG-binding fragment (ZZ sequence tag) was able to fully restore the growth defects of apkclta strain at restrictive temperature. ZZ-tagged Pkclp was par- tially purified by chromatography on DEAE-Sepharose, followed by IgG-Sepharose. In vitro, Pkclp phospho- rylates the pseudosubstrate peptide and myelin basic protein, but not histones. Replacing an isoleucine with an arginine 2 amino acids COOH-terminal of the accep- tor serine in the substrate peptide resulted in a 10-fold decrease of K,. Pkclp activity was independent of cofac- tors such as phospholipids, diacylglycerol, and Ca2+, known to activate several mammalian protein kinase C isoenzymes, making it a rather distantly related mem- ber of the protein kinase C superfamily.

The protein kinase C (PKC)l family is a phospholipid-de- pendent group of serinekhreonine-specific protein kinases that play a central role in mediating a variety of cellular responses induced by receptor-mediated hydrolysis of phosphatidylinosi- to1 4,5-bisphosphate to diacylglycerol and inositol 1,4,5- trisphosphate (Hokin, 1985; Nishizuka, 1988). Diacylglycerol serves as a second messenger to activate PKC (Kishimoto et al., 1980; Nishizuka, 1986, 19881, and inositol 1,4,5-trisphosphate functions to mobilize Ca2+ from intracellular stores (Berridge, 1987; Berridge and Imine, 1984). Several different subtypes of mammalian PKC have been described (Kikkawa et al., 1989; Nishizuka, 1988, 1992), and they can be divided into two sub- sets. One subset, designated cPKCs, includes the isoforms a, PI, PII, and y; the other group, designated nPKCs, includes the isoforms 6, E , 6, 7, and 0 (Kikkawa et al . , 1989; Stabel and Parker, 1991; Weinstein et al., 1991). Optimal activation of the

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ To whom correspondence should be addressed. Tel.: 41-22-706-9601;

The abbreviations used are: PKC, protein kinase C; yPKC, yeast protein kinase C; PAGE, polyacrylamide gel electrophoresis; n, novel; c, conventional.

Fax: 41-22-794-6965.

cPKCs, but not of the nPKCs, requires Ca2+ (Kishimoto et al., 1980; Ohno et al . , 1988; Ono et al., 1989).

Mammalian PKC is thought to be involved in the regulation of a range of cellular functions through its activation by growth factors, hormones, and other agonists. These functions com- prise cell growth and division (Persons et al., 1988; Rozengurt et al., 1984), secretion and exocytosis (Ohmura and Friesen, 1985), and modulation of ion conductance channels (Farley and Auerbach, 1986; Madison et al., 1986). The evidence available to date suggests that activation of PKC ultimately induces transcription of a number of genes, including the proto-onco- genes c-myc, c-fos, and c-sis (Colamonici et al., 1986; Coughlin et al., 1985; Greenberg and Ziff, 1984; Kelly et al . , 1983; Kruijer et a l . , 19841, the human collagenase gene (Angel et al . , 19871, the human metallothionein 11, gene (Lee et al., 19871, and the SV40 early genes (Imbra and Karin, 1986). Several transcrip- tion factors such as the AP-1 complex, AP-2, and NF-KB (Bae- uerle and Baltimore, 1988; Chiu et d . , 1987; Imagawa et al., 1987; Karin, 1991) have been shown to be involved in this PKC-mediated response.

Genes predicted to encode PKCs have been isolated from simple eukaryotes such as the budding yeast Saccharomyces cerevisiae (Levin et al., 1990) and the fission yeast Schizosac- charomyces pombe (Mazzei et al . , 1993; Toda et al., 1993). This suggests that PKCs and PKC-mediated pathways may have a fundamental function common to all eukaryotes. The PKCl gene of S. cereuisiae was isolated as a homolog of mammalian PKCP and PKCy and is essential for cell growth and mitosis (Levin et al., 1990). Loss of PKCl function results in osmotic remedial deficiencies in cell wall construction (Levin and Bart- lett-Heubusch, 1992; Paravicini et al., 1992a). The 5’. pombe genes pck l+ and pckZ+(pkcl+) share an overlapping essential function, which is not suppressed by hypertonic medium (Toda et al., 1993). Finally, in Dictyostelium, cloning of the gene en- coding myosin heavy chain kinase revealed that the predicted myosin heavy chain kinase protein is a distantly related mem- ber of the PKC family, although it carries out a very specific cellular function (Ravid and Spudich, 1989, 1992).

In yeast, phospholipid-dependent protein kinase C activity has been demonstrated only for the S. pombepkcl+(pckZ+) gene product (Mazzei et al., 1993). In the budding yeast S. cerevisiae, PKC-like enzyme activities in cell extracts have been detected and partially purified (Ogita et al., 1990; Simon et a l . , 1991), but the corresponding genes have not been isolated. Here, we report the in vivo and in vitro characterization of the S. cereui- siae PKCl gene product. We have partially purified Pkclp and have analyzed its biochemical properties including substrate specificities and cofactor requirements.

EXPERIMENTAL PROCEDURES Strains and Media-The S. cereuisiae strains MCY15-3A (MATa

cly15-1 ade2 trpl ura3 leu21 and GPY1115 (MATapkclAl::HIS3 leu2- 3,112 ura3-52 his3-A200 trpl-A901 suc2-A9 ade2-101) used in this

16821

16822 Yeast Protein Kinase C study were described earlier (Paravicini et al., 1992a), and it was shown previously that the cly15 mutation is allelic to pkcl (Paravicini et al., 1992a). Strain SEY6210 (MATa leu2-3,112 ura3-52 his3-A200 trpl- A901 lys2-801 suc2-A9 GAL.) was used with the permission of Scott Emr (University of California, San Diego, CA). The Escherichia coli strains used were JMlOl (supE thi-1 A(Zac-proAE)) (F' traD36 proAE lacPZAM15) (Casadaban and Cohen, 1980) and CJ236 (F' cat (=pCJ105; M13sCmr)/dut ung 1 thi-1 reZAl spoTl mcrA) (Bio-Rad). LB and M9 media supplemented with antibiotics and amino acids as re- quired were used to propagate E. coli strains (Miller, 1972). YPD rich and SD minimal medium used to cultivate S. cereuisiae were as de- scribed by Sherman et al. (1979). When required, Wickerham's minimal proline medium (Wickerham, 1946) supplemented with the appropriate amino acids was also used to cultivate S. cereuisiae.

Reagents and Materials-New England Biolabs Inc. and Boehringer Mannheim were the suppliers for all the DNA restriction and modifying enzymes. Staurosporine and Triton X-100 were also purchased from Boehringer Mannheim. DEAE-Sepharose and IgG-Sepharose were from Pharmacia Biotech Inc. Zymolyase lOOT and T r a ~ ~ ~ ~ S - l a b e l were from ICN Radiochemicals (Irvine, CAI, [32Plorthophosphate was ob- tained from Amersham Corp., and [y3'P]ATP (>6000 Ci/mmol) was from DuPont NEN. Protein phosphatase 2Aand okadaic acid were from Upstate Biotechnology, Inc. (Lake Placid, NY). The antiserum to alka- line phosphatase was a gift from Scott Emr. Human IgG Fc fragment was from Organon Tecnica (Durham, NC), and horseradish peroxidase- conjugated swine anti-rabbit IgG was from DAKO (Glostrup, Den- mark). All other chemicals including the antibodies to glucose-6-phos- phate dehydrogenase were from Sigma.

Plasmid Constructions-For multicopy expression of PKCl, a 4.3- kilobase SphI fragment containing the PKCl gene under its own pro- moter was inserted into the SphI site of YEpl3 (pGY80). To modify the COOH terminus of Pkclp, a EgZII site was introduced by site-directed mutagenesis after the last codon of the PKCl open reading frame. A 0.9-kilobase BamHI fragment from plasmid XaZZ (constructed by Ber- nard Allet, Glaxo Institute) carrying two repeats of the 60-amino acid IgG-binding domain of S. aureus protein A (ZZ repeats) (Nilsson et al., 1987) was inserted into this EgZII site to generate a PKCl-ZZ fusion gene. A 5.2-kilobase SphI fragment containing this fusion gene was then inserted into the SphI site of YEpl3 to generate pGY82.

To construct the trpE-PKCl fusion gene, plasmid PATH1 containing the trpE coding sequence (Dieckmann and Tzagoloff, 1985) was cut with SrnaI, and a 810-base pair ScaIIStuI PKCl-containing fragment was inserted into the SmaI site in the correct orientation, generating plas- mid pGY85.

Preparation of Antiserum Directed against Pkclp-The trpE-PKCl fusion gene plasmid pGY85 was used to transform bacterial strain JM101. After induction with indoleacrylic acid, the fusion protein was purified from the insoluble protein fraction as described earlier (Diec- kmann and Tzagoloff, 1985). Pkclp-specific antiserum was raised in rabbits by the Glaxo Institute in-house antibody facility.

Cell Labeling and Immunoprecipitation-For whole cell labeling, cells were grown in YPD rich medium to an absorbance of 1 a t 550 nm (A,5o). To initiate labeling, Tran3,S-label was added (at final concentra- tion of 150 mCi/ml) to cells at 10 A,,, units/ml in Wickerham's minimal proline medium, and the cells were incubated for 5 min a t 30 "C (Para- vicini et al., 199213). When employed, the labeling reaction was chased with 10 mM unlabeled methionine for the appropriate time period. The reaction was terminated by the addition of trichloroacetic acid to a final concentration of 5%. For labeling of spheroplasts (Vida et al., 1990), cells were grown in YPD rich medium as described above. Zymolyase lOOT (10 pg/1.5 x lo7 cells) was added, and the culture was incubated for 30 min at 30 "C. Spheroplasts were washed in 1 M sorbitol and labeled with Tran35S-label in Wickerham's minimal proline medium containing 1 M

sorbitol. Immunoprecipitations were performed as described earlier (Klionsky et al., 1988; Paravicini et al., 1992b). Samples were separated on 10% SDS-polyacrylamide gels (Laemmli, 1970). After electrophore- sis, gels were fixed in 10% acetic acid, 10% isopropyl alcohol for 20 min; washed twice with water; and then treated with 1 M sodium salicylate, 1% glycerol for 20 min. Gels were dried and fluorographed at -80 "C with intensifier screens.

In Vivo Phosphorylation Analysis-Whole cells were labeled with [32Plorthophosphate as described by Herman et al. (1990), except that bovine serum albumin was omitted from the labeling reaction. Immu- noprecipitations of Pkclp were performed as described previously (Klionsky et al., 1988; Paravicini et al., 1992b). After electrophoresis, the SDS-polyacrylamide gels were fixed as described above, except that 1 M sodium salicylate was omitted from the 1% glycerol solution. The gels were dried and exposed to x-ray film (Kodak X-Omat) at -80 "C

with intensifier screens. Phosphoamino Acid Analysis-10 A,, units of MCY15-3A harboring

the PKCl gene on a 2-p plasmid (pGY80) were labeled as described above. Pkclp was immunoprecipitated and separated on an 8% SDS- polyacrylamide gel. The gel was dried onto Whatman No. 3" paper, and Pkclp was detected by autoradiography. The Pkclp-containing gel piece was excised, removed from the filter paper, and treated for phos- phoamino acid analysis as described by Meisenhelder and Hunter (1991). The extracted sample was dried under vacuum, resuspended in constant boiling HCI, and hydrolyzed at 110 "C for 1-4 h. The hydro- lyzed sample was dried; resuspended in water three times; and finally resuspended in a standard mixture of phosphoserine, phosphothreo- nine, and phosphotyrosine (5 mg/ml). 10 pl of the sample were applied to a cellulose plate for thin-layer chromatography and submitted to two-dimensional electrophoresis. The first dimension was run in acetic acidlformic aciawater (78:25:897), pH 1.9, at 400 V for 3 h; and the second dimension was in acetic acid/pyridine/water (50:5:945), pH 3.5, at 400 V for 2 h. The plate was dried and developed with 0.1% ninhydrin in ethanol and analyzed by autoradiography.

Subcellular Fractionation ofPkclp-Fractionation experiments were done with the MCY15-3A pkcl'" strain harboring plasmid pGY80. 40 A,,, units of cells were harvested, spheroplasted with zymolyase lOOT, and labeled with Trad5S-label for 10 min at 30 "C. The spheroplasts were sedimented a t 2000 x g and resuspended in 2 ml of lysis buffer (200 mM Tris-HC1, pH 8.0,6 mM MgCl,, 1 mM EGTA, 1 mM benzamidine, 1 mM phenylmethylsulfonyl fluoride, 200 pg/ml antipain, 10 pg/ml soy- bean trypsin inhibitor and 2 pg/ml each pepstatin A, aprotinin, al- antitrypsin, and leupeptin), l volume of acid-washed glass beads was added, and the mixture was vortexed four times for 30 s in a minibead beater (Biospec Products, Inc., Bartlesville, OK) with 30 s of cooling in between. The lysate was centrifuged at 750 x g for 10 min at 4 "C. The resulting supernatant was adjusted with lysis buffer to 4 ml. 0.8-ml aliquots of this supernatant fraction were adjusted to 1 ml with lysis buffer or with one of the following reagents to the indicated final con- centrations: 2 M urea, 0.5 M NaCI, 1% Triton X-100, or 0.1 M Na,CO,, pH 11. The samples were incubated on ice for 20 min and subsequently centrifuged a t 100,000 x g for 45 min a t 4 "C. The supernatants were precipitated by adding trichloroacetic acid to a final concentration of 6%, and the pellets were resuspended in lysis buffer and immediately precipitated in 6% trichloroacetic acid. These fractions were immuno- precipitated as described above and analyzed by SDS-PAGE and fluorography.

Purification of 22-tagged Pkclp-MCY15-3A cells containing plas- mid pGY82 were cultivated overnight in YPD rich medium at 37 "C, harvested, and stored a t -80 "C. The frozen yeast cells were suspended in 4 volumes (4 ml of buffer/g of cells) of buffer A (25 mM Hepes, 5 mM 2-mercaptoethanol, 0.5 mM EGTA, 0.5 mM benzamidine, 0.5 m~ phenyl- methylsulfonyl fluoride, 50 pg/ml antipain, 10 pg/ml soybean trypsin inhibitor, and 2 pg/ml each pepstatin A, aprotinin, a,-antitrypsin, and leupeptin, pH 8.0). The cells were broken by passage through a French pressure cell three times a t a cell pressure of 16,000 p.s.i. The resulting extract was centrifuged twice a t 20,000 x g for 20 min. The supernatant was applied to a DEAE-Sepharose column (2 ml of geVg of broken cells) equilibrated in buffer B (30 mM Tris-HC1, pH 7.5, 0.2 mM benzamidine, and 0.2 mM phenylmethylsulfonyl fluoride). The column was washed with 3 column volumes of buffer B and eluted with 0.45 M NaCl in buffer B. Fractions containing Pkclp were pooled. The pool was applied to an IgG-Sepharose column equilibrated in buffer C (50 mM Tris-HC1, pH 7.6, 150 mM NaCl). The column was washed with 5 volumes of buffer C followed by 2 volumes of buffer D (50 mM Tris-HC1, pH 8.0, 100 mM NaCl). Bound ZZ-tagged Pkclp was eluted from the column as follows. The column was drained to the top, and 3 volumes of buffer E (50 mM Tris-HC1, pH 8.0, 100 mM NaC1, 66% glycerol, 8 p~ dithiothreitol, 1 mM benzamidine, and 1 mM phenylmethylsulfonyl fluoride) were added. Human IgG Fc fragment was added to a final concentration of 5-10 mg/ml, and the gel suspension was incubated on rotation at 4 "C for 16 h. At the end of the incubation period, the column was drained and washed with 1 column volume of buffer E. The elution step with IgG Fc fragment was repeated, and the resulting eluates were pooled. The eluted fractions contained ZZ-tagged Pkclp. Pkclp was stored in 50% glycerol at -20 "C after the addition of dithiothreitol to a final concen- tration of 0.3 m ~ . This enzyme could be kept for at least 2 months without detectable loss of activity.

Protein Kinase C Assay-Pkclp activity was measured through in- corporation of [32Plphosphate into the pseudosubstrate peptide GGL- HRHGSIINRKEEI (House and Kemp, 19871, corresponding to positions 394-409 in the predicted Pkclp amino acid sequence (Levin et al., 1990), with a serine replacing the original alanine at position 401. In

Yeast Protein Kinase C 16823

Copy#: A L H H H

Chase: 0 0 0 30 60

- 205 kDa

- 106 kDa - 80 kDa

Pkclp - a 1 I -

FIG. 1. Identification of PKCl gene product. Whole yeast cells were labeled with Tran3"S-label for 10 min at 30 "C. Labeling was ter- minated by the addition of trichloroacetic acid to 5% (first through third lanes), or a 30- or 60-min chase period was included prior to trichloro- acetic acid precipitation (fourth and fifth lanes). The cells were lysed with glass beads, immunoprecipitations using Pkclp-specific antibodies were performed with equivalent amounts of the clarified lysates, and the antigen-antibody complexes were subjected to SDS-PAGE and flu- orography. Copy# refers to the gene dosage ofPKCl in the strains used: A, GPYlll5 (no copy); L, SEY6210 (single copy); H, MCY15-3A harhor- ing pGY80 (multicopy).

the inhibition studies, the short pseudosubstrate peptide GGLHRHG- SIINRK and the inhibitory peptide GGLHRHGAIINRK were used. To test various substrates, the pseudosubstrate peptide was replaced with the substrates indicated in the text. The assay mixture contained 40 mM Tris-HC1, pH 7.5, 10 mM MgCl,, 10 mM 2-mercaptoethanol, 50 PM [y-:'2PlATP (specific activity of 1000-3000 cpdpmol), 50 pg/ml pseudosubstrate peptide, 2 mM EGTA, and Pkclp in a total volume of 40 pl. The mixture was incubated at 30 "C for 15 min. At the end of the incubation period, 15 pl were spotted onto 15 x 15-mm P-81 filter paper squares in duplicate. The squares were air-dried for 5 min and washed three times for 5 min in 75 mM phosphoric acid, followed by one wash each in water, acetone, and diethyl ether. The filters were air-dried and counted for radioactivity in a Beckmann TA6000 liquid scintillation counter. 1 activity unit corresponds to the amount of Pkclp required for the incorporation of 1 pmol of 1"Plphosphate into the substrate peptide in 1 min under the assay conditions described above.

Effect of Mammalian PKC Activators on Pkclp Activity-To deter- mine the effect of cofactors known to activate mammalian PKC, the assay mixture described above was modified as follows. Phospholipid and diacylglycerol from chlorofoxdmethanol stock solutions were dried under a stream of nitrogen and subsequently incorporated into the assay buffer by sonication three times for 30 s. When assays were performed in the presence of calcium, EGTA was omitted from the mixture.

Autophosphorylation of Pkclp-Pkclp was incubated as described above with the following modifications. No pseudosubstrate peptide was added, and at the end of the incubation period, the reaction was stopped by the addition of 1 volume of SDS-PAGE sample buffer. The sample was submitted to SDS-PAGE, and phosphorylated proteins were de- tected by autoradiography.

Immunoblots-The samples were separated on 7.5% SDS-polyacryl- amide gels according to the method of Laemmli (1970), and the proteins were subsequently electrophoretically transferred to a polyvinylidene difluoride membrane (Immobilon). The membrane was preincubated three times for 20 min in blocking buffer (phosphate-buffered saline containing 0.5% gelatin, 0.2% bovine serum albumin, and 0.1% Tween 20) and subsequently incubated in blocking buffer containing horserad- ish peroxidase-conjugated anti-rabbit IgG diluted 1:3000. The mem- brane was rinsed three times for 15 min in blocking buffer and devel- oped using the enhanced chemiluminescence (ECL) procedure from Amersham Corp.

RESULTS Identification of Pkclp-A trpE-Pkcl fusion protein was used

to raise antiserum against Pkclp for the detection of this pro- tein in yeast cell extracts. A ScaIIStuI PKCl-containing frag- ment capable of encoding a 271-amino acid peptide (amino acids 787-1057) was fused in-frame to the COOH-terminal cod- ing region of the E. coli trpE gene. This gene fusion was ex- pressed a t high levels in E. coli. The resultant hybrid protein

was purified and used for the production of polyclonal anti- serum (see "Experimental Procedures"). This antiserum de- tected a single protein of -150 kDa in radiolabeled wild-type yeast cell extracts (Fig. 1, second lane). This value is slightly higher than the predicted molecular mass of 132 kDa (Levin et al., 1990) and may be due to the phosphorylation of Pkclp (see below). The 150-kDa protein was seen only in PKCl-containing cells. No protein was recognized by this antiserum in a Apkcl strain (Fig. 1,first lane). Furthermore, when PKCl was present on a multicopy plasmid vector, an -5-8-fold increase in the 150-kDa protein was seen (Fig. 1, third through fifth lanes). Quantitative comparison of the relative abundance of Pkclp with the well studied vacuolar protein carboxypeptidase Y (-0.1% of total cell protein) (Koehrer and Emr, 1993) expressed in the same cells indicated that Pkclp is a rare protein repre- senting <0.01% of total cell protein in a logarithmically growing culture of wild-type cells. Cells overproducing Pkclp did not exhibit any obvious phenotype such as temperature-sensitive growth or the need for osmotically supported growth medium. Pkclp also appears to be a relatively stable protein in vivo. Following a 60-min chase, no detectable loss of Pkclp was observed.

Mammalian PKCs are known to be phosphoproteins. We therefore tested whether Pkclp is a phosphoprotein in vivo. Whole yeast cells of the Apkcl strain GPY1115 and of thephcl'" strain MCY15-3A harboring the PKCl gene on a multicopy plasmid (pGY80) were labeled with [32Plorthophosphate. The cells were lysed with glass beads, and Pkclp was immunopre- cipitated from the clarified extracts. As shown in Fig. 2 A , phos- phorylated Pkclp was precipitated from the Pkclp-overex- pressing strain, whereas no 32P-labeled Pkclp was detected in the Apkcl strain. To determine if Pkclp was phosphorylated on serine, threonine, or tyrosine, we performed a phosphoamino acid analysis. I n vivo "P-labeled Pkclp was immunoprecipi- tated and subjected to SDS-PAGE, and Pkclp was extracted from the gel. After acid hydrolysis, the amino acid mixture was separated by two-dimensional thin-layer electrophoresis (Fig. 2B). The phosphoamino acids were visualized by ninhydrin staining, and incorporated radioactive phosphate was detected by autoradiography. Fig. 2B shows that Pkclp was predomi- nantly phosphorylated on serine, but also that phosphothreo- nine was detected. There was no indication of any phospho- rylated tyrosine residues. Therefore, like mammalian PKCs, Pkclp is phosphorylated exclusively on serine and threonine residues.

Subcellular Fractionation of Pkclp-To obtain information on the intracellular location of Pkclp, we performed differen- tial centrifugation experiments. MCY15-3A (pkcl'") sphero- plasts mildly overexpressing Pkclp were radiolabeled and ly- sed with glass beads in a minibead beater. The crude lysates were centrifuged at 500 x g for 10 min to remove unlysed spheroplasts, and the resulting supernatant was centrifuged at 100,000 x g for 45 min. The 100,000 x g pellet and supernatant fractions were assayed for the presence of Pkclp by immuno- precipitation. Under these conditions, >90% of the Pkclp was recovered in the insoluble fraction (Fig. 3A) together with the vacuolar membrane protein alkaline phosphatase, whereas the cytosolic marker glucose-6-phosphate dehydrogenase remained in the supernatant fraction (Fig. 3B). The nature of the asso- ciation of Pkclp with the particulate fraction was further in- vestigated by treating the cleared lysate with various reagents (1% Triton X-100, 2 M urea, 0.5 M NaCI, and 0.1 M Na,CO,) for 15 min at 4 "C before centrifugation at 100,000 x g and subse- quent immunoprecipitation. None of these reagents efficiently solubilized Pkclp from the particulate fraction. Moreover, sev- eral other nonionic or zwitterionic detergents were also inef- fective in solubilizing Pkclp into the supernatant fraction.

Pkcl p

Yeast Protein Kinase C A

pkc 1 ts I Apkc 7 PKC 7 -Zp

-.

P-Ser

p-Tyr P-Thr

Start I 2nd dimension pH3.5

ysis of Pkclp. A, whole yeast cells were labeled with [32Plorthophos- FIG. 2. In vivo phosphorylation and phosphoamino acid anal-

phate for 20 min a t 30 "C. The cells were lysed with glass beads, and immunoprecipitations were performed on the clarified lysates using the Pkclp-specific antiserum. Strains are GPY1115 (Appkcl) and MCY15-3A (pkcl") harboring pGY80. B, strain MCY15-3A harboring plasmid pGY80 was labeled with [32P]orthophosphate as described for A, and Pkclp was immunoprecipitated and eluted from an SDS-polyacrylam- ide gel. After acid hydrolysis, the amino acid mixture was separated by two-dimensional thin-layer electrophoresis together with the nonradio- active phosphoamino acid standards phosphoserine (P-Ser), phospho- threonine (P-Thr), and phosphotyrosine (P-Fyr). The phosphoamino acid standards were stained with ninhydrin, and their migration posi- tions are indicated by dashed circles. Start indicates the loading posi- tion.

Only complete denaturation of the proteins with SDS solubi- lized the enzyme (data not shown). These data indicate that Pkclp associates with the particulate fraction through strong protein-protein interactions with other cellular proteins or pro- tein complexes.

Purification of Pkclp-The sequence homology between Pkclp and mammalian PKCs suggests that the yeast protein is a member of the PKC family (Levin et al., 1990). In crude yeast extracts of MCY15-3A carrying pGY80 (PKCl multicopy), how- ever, no Pkclp-specific kinase activity could be detected. In addition, Pkclp activity was not found in crude extracts from cells expressing the PKCl gene under control of the GAL1 promoter. Moreover, as shown above, the 100,000 x g superna- tant fraction contained no more than 5% of total Pkclp ex- pressed in the cell, and Pkclp in any pellet fraction was com- pletely inactive and resistant to activation with Ca2+, phospholipids, and diacylglycerol or phorbol ester. Therefore, it was necessary to devise a means to facilitate the purification of Pkclp to permit assessment of its enzymatic activity. The car- boxyl terminus of Pkclp was modified with two repeats of the S. aureus protein A IgG-binding domain (ZZ tag) (Nilsson et al., 1987) to allow isolation on IgG-Sepharose and to facilitate im- munodetection due to its binding to the Fc portion of immuno-

1 Yo 2M 0.5M pH11 buffer Triton Urea NaCl Na$O3

Pkclp -I

B Pi00 s100

Pkcl p 95% 5%

ALP 295% <2%

GGPDH 5% 95%

FIG. 3. Association of Pkclp with PlOO particulate fraction. A, MCY15-3A cells harboring plasmid pGY80 were converted to sphero- plasts, labeled with Tran35S-label, and lysed with glass beads. The lysate was cleared of unbroken spheroplasts and subsequently treated with the following reagents prior to centrifugation a t 100.000 x g to generate pellet (P) and supernatant ( S ) fractions: lysis buffer alone (first and second lanes), 1% Triton X-100 (third and fourth lanes), 2 M urea (fifth and sixth lanes), 0.5 M NaCl (seventh and eighth lanes), and 0.1 M Na,CO,, pH 11 (ninth and tenth lanes). B, shown is a summary of the relative levels of Pkclp (multicopy), alkaline phosphatase (ALP), and glucose-6-phosphate dehydrogenase (GGPDH) in the PlOO and SlOO fractions. The centrifugation procedure was performed as de- scribed for A.

globulins (Nilsson et al., 1987). When reintroduced into the pkcl'" strain, the modified PKCl gene was able to fully comple- ment the growth defect of thepkcl'" mutant strain at restrictive temperature on medium lacking osmotic support, indicating that it had maintained its in vivo functions (Fig. 4).

S. cerevisiae MCY15-3A cells carrying the ZZ-tagged PKCl gene on a multicopy plasmid (pGY82) were grown in YPD rich medium at 37 "C. Cells were harvested and broken by three passages through a French pressure cell, and the extract was centrifuged and separated into soluble and particulate frac- tions at 20,000 x g. The 20,000 x g supernatant was applied to a DEM-Sepharose column to clear the lysate from the bulk of cellular lipids and proteins. The fractions containing Pkclp were further purified on IgG-Sepharose (Fig. 5A). The IgG- Sepharose column was washed extensively (see "Experimental Procedures"), and Pkclp kinase activity could be assayed di- rectly on the IgG-Sepharose resin. Alternatively, Pkclp was eluted from the IgG-Sepharose column by overnight incubation with an excess of human IgG Fc fragments and was subse- quently assayed in solution. No qualitative differences in the properties of Pkclp could be detected between these two en- zyme preparations, and the results described below were ob- tained with eluted ZZ-tagged Pkclp. When extracts from cells overexpressing untagged Pkclp were loaded on an IgG-Sepha- rose column, no kinase activity was detected on the IgG-Sepha- rose resin (data not shown). Furthermore, as shown in the accompanying paper by Watanabe et al. (1994), a Pkclp con- taining a mutated ATP-binding site showed strongly reduced levels of protein kinase activity, suggesting that the protein kinase activity associated with Pkclp is indeed an intrinsic property of this protein. During the purification, column frac- tions were assayed for kinase activity using the Pkclp pseudosubstrate peptide as substrate (see "Experimental Pro- cedures"). The presence of Pkclp in these fractions was verified

Yeast Protein Kinase C 16825

pkcl'" PKCl

FIG. 4. Complementation of growth defect by ZZ-tagged gene. Pkclp was modified with two re- peats of the S. aureus protein A IgG-bind- ing domain at the COOH terminus. Strains MCY15-3A (phcl") , MCY15-3A harboring plasmid pGY80 (PKCI multi- copy), and MCY15-3A harboring plasmid pGY82 (ZZ-tagged PKCI multicopy) were streaked on YPD rich medium and incu- bated for 48 h at 25 or 37 "C as indicated.

A 1 2 3

Pkclp -

pkc 7 ts / PKC 7 zZ-2~

m'' ' PKC7-2p

B 1

205 kDa

116kDa - 116kDa

- 80 kDa

Pkclp -,- I- 0-

80 kDa

2 3

- 205 kDa

25°C 37°C

FIG. 5. Partial purification of ZZ-tagged Pkclp. Cells of strain MCY15-3Aharboring plasmid pGY82 (ZZ-tagged PKCl multicopy) were grown in YPD rich medium at 37 "C, broken in a French pressure cell, and centrifuged a t 20,000 xg. The resulting supernatant was applied to a DEAE-Sepharose column, and the fractions containing ZZ-tagged Pkclp were pooled. Pkclp from this pool was then further purified on IgG-Sepharose. Samples from each purification step were analyzed by SDS-PAGE. The gel was either stained with Coomassie Blue (A) or blotted to a nitrocellulose membrane, and ZZ-tagged Pkclp was immu- nodetected with horseradish peroxidase-coupled anti-rabbit antiserum and visualized by chemiluminescence ( B ) . Lane I , S20 fraction; lane 2, DEAE-Sepharose pool containing ZZ-tagged Pkclp; lane 3, ZZ-tagged Pkclp after elution from IgG-Sepharose. Note that -400 times more protein was applied to lane 3 in A than to the corresponding lane in B.

with immunoblots using horseradish peroxidase-coupled anti- rabbit antiserum (Fig. 5B ).

Biochemical Properties of Pkclp-Pkclp has an absolute re- quirement for M e ; the K, was determined to be 0.6 mM. The K, for ATP was 22 PM. Pkclp assays were performed with the addition of cofactors known to influence the activity of some mammalian PKCs. Similar to mammalian PKCG, PKCE, and PKCC and S. pombe pkcl+, Pkclp activity was independent of the Ca2+ concentration (data not shown). Unlike mammalian PKCs, however (Kishimoto et al., 19801, Pkclp kinase activity was not activated by painvise combinations of several phospho- lipids and dioleoylglycerol (Fig. 6A). Also, several different dia- cylglycerols or phorbol 12-myristate 13-acetate together with phosphatidylserine were unable to promote Pkclp activation (Fig. 6B), as was arachidonic acid (data not shown).

Pkclp activity was markedly inhibited by staurosporine, a potent inhibitor of PKC, with an IC,, of 4 x M (Fig. 7). Also, Pkclp was competitively inhibited by the pseudosubstrate pep- tide containing the naturally occurring alanine (Fig. 8). The Ki was determined as 2.8 pg/ml (1.6 p~).

We also tested the ability of Pkclp to phosphorylate various substrates in vitro. Fig. 9 shows that the pseudosubstrate pep- tide served as the preferred phosphate acceptor (K, = 19 pg/ml (9.3 p ~ ) ) , closely followed by myelin basic protein (K, = 550 pg/ml). In contrast, several histones, protamine, and dephos-

phorylated casein were only poorly phosphorylated or they were not substrates at all. This profile of substrate preference is similar to that found for mammalian PKCG and PKCE (Koide et al., 1992; Ogita et al., 1992). To determine the requirements for Pkclp substrate binding in more detail, several modified pseudosubstrate peptides were synthesized, and relative affini- ties for Pkclp were determined (Table I). When both NH,- and COOH-terminal basic amino acids flanking the acceptor serine were removed, the peptide was no longer a substrate for Pkclp. Removing the basic amino acids either NH,- or COOH-terminal of the serine increased the K,,, 13- and 100-fold, respectively. Replacing the COOH-terminal arginine with lysine did not sig- nificantly alter the substrate affinity. Inserting additional basic amino acids on either side of the acceptor serine, however, decreased the K, value up to 10-fold. Removing negatively charged amino acid residues at the COOH terminus also re- sulted in a decreased K,,. These results indicate that additional positive charges close to the acceptor serine significantly im- prove substrate affinity for Pkclp.

To determine the capability of Pkclp to autophosphorylate in vitro, the protein was incubated in the presence of radiolabeled ATP without external substrate (Fig. 10). At the end of the incubation period, the catalytic fragment of protein phospha- tase 2A was added to dephosphorylate Pkclp (Fig. 10). Upon addition of the protein phosphatase 2A inhibitor okadaic acid and reincubation with radiolabeled ATP, Pkclp was able to reautophosphorylate. This result indicates that Pkclp remains active after dephosphorylation. Thus, contrary to what was observed for mammalian PKCa (Cazaubon and Parker, 19931, protein phosphatase 2A was unable to inactivate S. cerevisiae Pkclp. Phosphoamino acid analysis of in vitro autophosphory- lated Pkclp carried out as described above for in vivo phospho- rylated Pkclp indicated that in vitro phosphorylated Pkclp contains phosphorylated serine and threonine residues (data not shown).

DISCUSSION

The high homology between S. cerevisiae Pkclp and mam- malian PKCs, particularly in the regions that are unique to the PKC family, suggests that S. cerevisiae Pkclp is a member of the PKC family (Levin et al., 1990). This conclusion is sup- ported by some of the properties exhibited by s. cerevisiae Pkclp: like all PKCs, i t undergoes autophosphorylation on ser- ine and threonine residues; its preferred in vitro substrate is a peptide corresponding to the pseudosubstrate site defined in mammalian PKCs (House and Kemp, 1987; House et al., 1987); and like some mammalian PKCs (Kiley and Jaken, 1990; Mo-

16826 Yeast Protein Kinase C

)r

> 0

.- c, Y

CI

a

100

75 . 50

25

0 no lipid PS PI PC PE

100

75

50

25

0

B

DOG OAG DOcG PMA lipids. Pkclp was assayed under the conditions described under “Ex-

FIG. 6. Pkclp kinase activity in presence of various phospho-

perimental Procedures,” except that various phospholipids, diacylglyc- erols, and phorbol ester were added as indicated. A, Pkclp activity in the presence of 10 pg/ml sn-1,2-dioleoylglycerol and a 100 pg/ml con- centration of the phospholipid indicated: phosphatidylserine (PS) , phosphatidylinositol (PI) , phosphatidylcholine (PC) , and phosphati- dylethanolamine (PE). B, Pkclp activity in the presence of 80 pg/ml phosphatidylserine and a 8 pg/ml concentration of the diacylglycerol indicated: sn-1,2-dioleoylglycerol (DOG), sn-1-oleoyl-2-acetylglycerol (OAG), and sn-1,2-dioctanoylglycerol (DOcG). Phorbol 12-myristate 13- acetate (PMA) was added at 1 x M. 100% corresponds to the activity obtained without addition of any cofactor.

chly-Rosen et al., 1990; Zalewski et al., 1990), Pkclp associates with the Triton X-100-insoluble particulate fraction.

We followed the fractionation of Pkclp in a strain overex- pressing Pkclp -5-fold. Under these conditions, -95% of the Pkclp sedimented in the Triton X-100-insoluble particulate fraction at 100,000 x g . Several reports demonstrate that upon activation, mammalian PKC also becomes associated with the Triton X-100-insoluble fraction of cells. This includes associa- tion with both the cytoskeleton (Kiley and Jaken, 1990; Mo- chly-Rosen et al., 1990; Zalewski et al., 1990) and the nucleus

75 i u >

a

25

1

Staurosporine (M) FIG. 7. Inhibition of Pkclp by staurosporine. Pkclp was assayed

as described under “Experimental Procedures” with the pseudosub- strate peptide as substrate in the presence of increasing concentrations of staurosporine. The IC,, was determined as 4 x lo-’ M.

I / / 1

0.08 1 f /

3 0.04

0.00 /I 0.00 0.10 0.20 0.30 0.40 0.50

FIG. 8. Inhibition of Pkclp by inhibitory alanine pseudosub- strate peptide. Shown is the Lineweaver-Burk plot for Pkclp inhibi- tion by the short pseudosubstrate peptide containing the original ala- nine (GGLHRHGAIINRK). The same peptide with a serine replacing the alanine was used as substrate (Table I, peptide 9). Pkclp was assayed at increasing substrate concentrations in the presence of the inhibitory peptide. Double-reciprocal plots at each inhibitor concentra- tion are shown. 0, no inhibitor; m, 4 pg/ml; A, 20 pg/ml; A, 40 p g h l . The intersection of the lines shows competitive inhibition, and the K, was determined as 2.8 pg/ml (1.6 p~).

(Masmoudi et al., 1989). The observations described above sug- gest that the majority of Pkclp is invariably associated with a large multiprotein complex. In uitro, Pkclp present in the par- ticulate fraction was completely inactive under all the assay conditions tested. It appears unlikely, however, that this rep- resents the situation in uiuo, but so far, the regulatory mecha- nism of Pkclp activation remains elusive.

A yeast PKC (yPKC) with cofactor requirements similar to those of mammalian PKC has been described by Ogita et al. (1990). For three reasons, we believe that Pkclp is not the same

Yeast Protein Kinase C 16827

100

A

E 75

n ”

Pseudosub. MBP Protamine Casein Histone Histone Ill-s VIII-s

FIG. 9. Relative enzymatic activities of Pkclp with various phosphate acceptors. Pkclp was incubated with various substrates under standard assay conditions as described under “Experimental Pro- cedures.” The activity obtained at a saturating concentration of pseudosubstrate peptide was set to 100%. The K,,, was determined to be 19 pg/ml (9.3 PM) for the pseudosubstrate peptide and 550 pg/ml for myelin basic protein ( M B P ) . The other substrates were not significantly phosphorylated up to a concentration of 2 mg/ml.

TARLE I Relative affinities of Pkclp for modified pseudosubstrate peptides

Pkclp was assayed for protein kinase activity as described under “Experimental Procedures” using different derivatives of the Pkcl pseudosubstrate peptide (peptide 1) as substrates. Underlined amino acids indicate changes in the modified pseudosubstrate peptides.

Peptide Sequence K”,

I.lM

1 G G L H R H G S I I N R K E E I 9.3 2 G G L H A H G S I I N A A E E I NS“ 3 G G L H R H G S I I N A A E E I 127 4 G G L H q H G S I I N R K E E I 935 5 G G L H R H G S I I N K K E E I 12 6 G G L H R K G S I I N R K E E I 1.9 7 G G L H R K G S I E N R K E E I 0.88 8 G G L H R H G S I R N R K E E I 0.86 9 G G L H R H G S I I N R K 2.1

“ No substrate.

enzyme as yPKC. (i) The reported size of yPKC (90 kDa) (Ogita et al., 1990) is much smaller than that of Pkclp (150 kDa), although in principle, yPKC could represent a proteolytic frag- ment of Pkclp. (ii) The substrate specificity of yPKC is quite different from that of Pkclp (Iwai et al., 1992). (iii) The protein kinase activity of yPKC was dependent on phosphatidylserine and was increased by Ca2+ and diacylglycerol, whereas we could not establish any effect of these cofactors on Pkclp activity.

To facilitate the isolation of the PKCl gene product, i t was modified by the addition of two repeats of the S. aureus protein A IgG-binding domain (ZZ tag). The modified Pkclp was able to fully complement the phcl‘” growth defect a t restrictive tem- perature. Moreover, the fractionation data obtained with ZZ- tagged Pkclp were identical to the results with unmodified Pkclp (data not shown). These results suggest that the ZZ sequence tag does not interfere with the in vivo functions of Pkclp. ZZ-tagged Pkclp was overproduced from a multicopy vector and subsequently isolated by chromatography on DEAE- Sepharose followed by IgG-Sepharose. This purification proce- dure allowed a substantial enrichment of the small fraction of soluble Pkclp. The DEAE chromatography step was included in the purification procedure to facilitate removal of contami- nating cellular phospholipids. In most applications, however,

Pkcl p + PP2a +okadaic acid +[y- 32 PIATP + [ Y - ~ ~ P I A T P

I 1 I I 1

5‘ 10‘ 15

autophosph. dephosph. reautophosph.

5’ lo’ 15

Pkcl p

FIG. 10. Pkclp autophosphorylation. Pkclp was incubated with [Y-:’~P]ATP (see “Experimental Procedures”) without exogenous sub- strate. After a 5-min incubation period, protein phosphatase 2A (PP2a) was added for dephosphorylation. After another 5 min of incubation, a 1 mM concentration of the protein phosphatase 2Ainhibitor okadaic acid was added, and the sample was further incubated for 5 min in the presence of freshly added radiolabeled ATP. The reaction was stopped with 1% SDS, and samples were analyzed by SDS-PAGE and autora- diography.

this step may not be necessary. Thus, the ZZ sequence tag allows massive enrichment of the modified protein in essen- tially a single step.

Pkclp activity was shown to be independent of activation by phospholipids, diacylglycerol, and Ca2+. These properties con- trast with those of mammalian PKCs, and Pkclp can be com- pared best with members of the subset of mammalian nPKCs: PKCS, PKCE, and PKCC also do not require Ca2+ for activity. Contrary to Pkclp, however, they require diacylglycerol or phorbol ester in addition to phosphatidylserine for activation (Koide et al., 1992; Ogita et al., 1992; Ways et al., 1992). The regulatory region of Pkclp does contain two of the cysteine-rich repeats known as the C1 region, the hallmark sequences found in the PKC family of kinases previously shown to be involved in phorbol ester and phospholipid binding (Luo et al., 1993). The two repeats share 34% sequence identity (47% similarity if conservative amino acid substitutions are considered) with the mammalian counterparts: human and rat PKCsP. Ahmed et al. (1991) identified a 15-amino acid consensus motif in the C1 region of mammalian PKCs including the 6 cysteine and 2 histidine residues potentially involved in Zn2+ binding. In Pkclp, both C1 repeats share 73% sequence identity (11/15) with the consensus sequence. Apart from rat PKCS, all other mammalian PKC isoenzymes sequenced so far possess at least one C1 domain with 100% identity to the consensus sequence. Thus, small differences in the cysteine-rich repeats may ac- count for the lack of phorbol ester and phospholipid binding activity. In the region of Pkclp that corresponds to the putative Ca2+-binding region of mammalian PKCs, the modest degree of sequence identity (17%) supports the notion that Pkclp is not dependent on Ca2+ for activity.

Pkclp can be dephosphorylated upon addition of protein phosphatase 2A without losing its capability to reautophospho- rylate. This observation is similar to the results obtained with mammalian PKCa after treatment with potato acid phospha- tase, which does not inactivate the mammalian protein; treat- ment of mammalian PKCa with protein phosphatase 2A, how- ever, inactivates the enzyme (Pears et al., 1992). Bacterially expressed or in vitro translated mammalian PKC is completely

16828 Yeast Protein Kinase C inactive (Cazaubon and Parker, 1993; Pears et al., 1992), and an overexpressed catalytically inactive form of PKCa competes with newly synthesized endogenous PKCP for phosphorylation (Cazaubon and Parker, 1993). Thus, it has been suggested that a PKC kinase may carry out initial phosphorylation on PKC to activate the enzyme (Cazaubon and Parker, 1993; Pears et al., 1992). Similar to these observations, S. cerevisiae Pkclp syn- thesized in vitro remains completely inactive (data not shown). Purified Pkclp, however, is not inactivated upon treatment with protein phosphatase 2A. Thus, native Pkclp may carry phosphate residues not accessible to protein phosphatase 2A. Therefore, like the mammalian system, yeast may contain a “PKC activator” necessary for an initial phosphorylation step in order to prime Pkclp activity.

There is little information as yet about the biochemical events mediating the connection between mammalian PKC ac- tivation and cellular response, and only very few potential in vivo PKC substrates are known in mammalian cells (Stabel and Parker, 1991; Woodgett et al., 1987). In the yeast S. cer- euisiae, several protein kinases have been shown by genetic means to participate in a protein kinase cascade controlled by PKCl (Irie et al., 1993; Lee et al., 1993; Lee and Levin, 1992), but none of the enzymes has been shown to be a physiological substrate for Pkclp. However, combining the biochemical and genetic approaches possible in yeast should allow the study in detail of the role of PKCs in eukaryotic signal transduction.

Acknowledgments-We thank Christopher Hebert for preparing the photographs, Charles Bradshaw for peptide synthesis, David Levin for communicating results prior to publication, the members of the Glaxo Institute yeast group for helpful discussions, and Jonathan Knowles for critically reading the manuscript.

REFERENCES

Ahmed, S., Kozma, R., Lee, J., Monfries, C., Harden, N., and Lim, L. (1991) Biochem. J . 280,233-241

Angel, P., Imagawa, R., Chiu, R., Stein, B., Imbra, R. J., Rahmsdorf, H. J., Jonat, C., Herrlich, P., and Karin, M. (1987) Cell 49, 729-739

Baeuerle, P. A,, and Baltimore, D. (1988) Cell 53,211-217 Bemdge, M. J. (1987) Annu. Rev. Biochem. 56,159-193 Bemdge, M. J., and Irvine, R. F. (1984) Nature 312,315-321 Casadaban, M. J., and Cohen, S. N. (1980) J. Mol. Biol. 138, 179-207 Cazaubon, S. M., and Parker, P. J. (1993) J. B i d . Chem. 268, 17559-17563 Chiu, R., Imagawa, M., Imbra, J., Bockover, J. R., andKarin, M. (1987)Nature 329,

Colamonici, 0. R., Trepel, J. B., Vidal, C. A., and Neckers, L. M. (1986) Mol. Cell.

Coughlin, S. R., Lee, W. M. F., Williams, P. W., Giels, G. M., and Williams, L. T.

Dieckmann, C. L., and Tzagoloff, A. (1985) J. Biol. Chem. 260, 1513-1520 Farley, J., and Auerbach, S. (1986) Nature 319,220-223 Greenberg, M. E., and Ziff, E. B. (1984) Nature 311,433438 Herman, P. K., Stack, J. H., DeModena, J. A,, and Emr, S. D. (1990) Cell 64,

Hokin, L. E. (1985) Annu. Rev. Biochem. 54,205-235 House, C., and Kemp, B. E. (1987) Science 238,1726-1728 House, C., Wettenhall, R. E. H., and Kemp, B. E. (1987) J. B i d . Chem. 262,

Imagawa, M., Chiu, R., and Karin, M. (1987) Cell 51,251-260 Imbra, R. J., and Karin, M. (1986) Nature 323, 555-558 Irie, K., Takase, M., Lee, K. S., Levin, D. E., Araki, H., Matsumoto, K., and Oshima,

Y. (1993) Mol. Cell. Biol. 13, 3076-3083 Iwai, T., Fujisawa, N., Ogita, K., and Kikkawa, U. (1992) J. Biochem. (Tokyo) 112,

7-10 Karin, M. (1991) in Molecular Aspects of Cellular Regulation (Cohen, P., and

Amsterdam Foulkes, J. G., eds) Vol. 6, pp. 235-253, Elsevier Science Publishers B. V.,

648-651

Biol. 6, 1847-1850

(1985) Cell 43, 243-251

425-437

772-777

Kelly, K, Cochran, B. H., Stiles, C. D., and Leder, P. (1983) Cell 35, 603410 Kikkawa, U., Kishimoto, A,, and Nishizuka, Y. (1989) Annu. Reu. Biochem. 58,

Kiley, S., and Jaken, S. (1990) Mol. Endocrinol. 4, 5 9 4 8 Kishimoto, A., Takai, Y., Mori, T., Kikkawa, U., and Nishizuka, Y. (1980) J. Biol.

Klionsky, D. J., Banta, L. M., and Emr, S. D. (1988) Mol. Cell. Biol. 8, 2105-2116

Koide, H., Ogita, K., Kikkawa, U., and Nishizuka, Y. (1992) Proc. Natl. Acad. Sci. Koehrer, K., and Emr, S. (1993) J. Biol. Chem. 268, 559-569

Kruijer, W., Cooper, J. A,, Hunter, T., and Verma, I. M. (1984) Nature 312,711-716 Laemmli, U. K (1970) Nature 227, 680-685 Lee, K. S., and Levin, D. E. (1992) Mol. Cell. Biol. 12, 172-182 Lee, W., Haslinger, A., Karin, M., and Tjian, R. (1987) Nature 325, 368-372 Lee, K. S., Irie, K., Gotoh, Y., Watanabe, Y., Araki, H., Nishida, E., Matsumoto, K.,

Levin, D. E., and Bartlett-Heubusch, E. (1992) J. Cell B i d . 116, 1221-1229 Levin, D. E., Fields, F. O., Kunisawa, R., Bishop, J. M., and Thorner, J. (1990) Cell

Luo, J. H., Kahn, S., O’Driscoll, K, and Weinstein, I. B. (1993) J. B i d . Chem. 268,

Madison, D. V., Malenka, R. C., and Nicoll, R. A. (1986) Nature 321, 695-697 Masmoudi, A,, Labourdette, G., Mersel, M., Huang, F. L., Huang, K. P., Vincendon,

Mazzei, G. J., Schmid, E. M., Knowles, J . K. C., Payton, M. A,, and Maundrell, K.

Meisenhelder, J., and Hunter, T. (1991) Methods Enzymol. 197, 288-305 Miller, J. (1972) Experiments in Molecular Genetics, Cold Spring Harbor Labora-

3 1 4 4

Chem. 255,2273-2276

U. S. A. 89, 1149-1153

and Levin, D. E. (1993) Mol. Cell. Biol. 13, 30673075

62,213-224

37154719

G., and Malviya, A. N. (1989) J. Biol. Chem. 264,1172-1179

G. (1993) J. Biol. Chem. 268, 7401-7406

tory, Cold Spring Harbor, NY

(1990) Cell Reeul. 1.693-706

”

Mochly-Rosen, D., Henrich, C. J., Cheever, L., Khaner, H., and Simpson, P. C.

Nilsson,’B., Moks, T., Jansson, B., Abrahmsen, L., Elmhlad, A., Holmgren, E.,

Nishizuka, Y (1986) Science 233, 305312 Henrichson, C., Jones, T. A,, and Uhlen, M. (1987) Protein Eng. 1, 107-113

Nishizuka, Y (1988) Nature 334, 661465 Nishizuka, Y (1992) Science 268,607414 Ogita, K., Miyamito, S., Koide, H., Iwai, T., Oka, M., Ando, K., Kishimoto,A., Ikeda,

K., Fukami, Y., and Nishizuka, Y. (1990) Proc. Natl. Acad. Sci. U. S. A. 87,

Ogita, K., Miyamoto, S., Yamaguchi, K., Koide, H., Fujisawa, N., Kikkawa, U., 5011-5015

Sahara, S., Fukami, Y., and Nishizuka, Y. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 1592-1596

I ,

Ohmura, E., and Friesen, H. G. (1985) Endocrinology 116,728-733 Ohno, S., Akita, Y., Konno, S., Imajoh, S., and Suzuki, K. (1988) Cell 53, 731-741 Ono, Y., Fuji, K., Ogita, U., Kikkawa, K., Igarashi, K., and Nishizuka, Y. (1989)

Paravicini, G., Cooper, M., Friedli, L., Smith, D. J., Carpentier, J.-L., Klig, L. S.,

Paravicini, G., Horazdovsky, B. F., and Emr, S. D. (1992b) Mol. Biol. Cell 3,415-

Pears, C., Stahel, S., Cazaubon, S., and Parker, P. J. (1992) Biochem. J. 283,

Persons, D. A., Wilkison, W. O., Bell, R. M., and Finn, 0. J. (1988) Cell 52,447458 Ravid, S., and Spudich, J. A. (1989) J. Biol. Chem. 264, 1514415150 Ravid, S., and Spudich, J. A. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 5877-5881 Rozengurt, E., Rodriguez-Pena, A,, Coombs, M., and Sinnet-Smith, J. (1984) Proc.

Sherman, E , Fink, G. R., and Lawrence, L. W. (1979) Methods in Yeast Genetics,

Simon,A. J . , Milner,Y., Saville, S. P., Dvir, A,, Mochly-Rosen, D., and Om, E. (1991)

Stabel, S., and Parker, P. (1991) Pharmacol. & Ther 51, 71-95

Vida, T. A,, Graham, T. R., and Ernr, S. D. (1990) J. Cell Biol. 111,2871-2884 ’Ibda, T., Shimanuki, M., and Yanagida, M. (1993) EMBO J. 12,1987-1995

Watanabe, M., Chen, C. Y., and Levin, D. E. (1994) J. Biol. Chem. 269, 16829- 16836

Ways, D. K., Cook, P. P., Webster, C., and Parker, P. J . (1992) J. B i d . Chem. 267, 4799-4805

Weinstein, I. B., Borner, C. M., Krauss, R. S., ODriscoll, K., Choi, P. M., Moritomi, M., Hoshina, S., Hsieh, L. L., Tchou-Wong, K. M., Guadagno, S. N., Ueffing, M., and Guillem, J. (1991) in Origins of Human Cancer (Brugge, J., Curraw, T., Harlow, E., and McCormick, F., eds) pp. 113-123, Cold Spring Harbor Labora- tory, Cold Spring Harbor, NY

Proc. Natl. Acad. Sci. U. S. A. 86, 3099-3103

and Payton, M. A. (1992a) Mol. Cell. Biol. 12, 4896-4905

427

515-518

Natl. Acad. Sci. U. S. A. 81, 5748-5752

Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

Proc. R. SOC. h n d . B Biol. Sci. 243, 165-171

Wickerham, L. J. (1946) J. Bacteriol. 52, 293301 Woodgett, J. R., Hunter, T., and Gould, K. L. (1987) in Cell Membrane: Methods and

Review (Elson, E., Frazier, W., and Glaser, L., eds) pp. 215-340, Plenum Press,

Zalewski, P. D., Forbes, I. J., Giannakis, C., Cowled, A., and Betts, W. H. (1990) New York

FEBS Lett. 273. 131-134