Expression of Four Protein Kinase C Isoforms in Rat Fibroblasts

THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1992 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 267, No. 18,Issue of June 25, pp. 12901-12910,1992 Printed in U.S.A.

Expression of FourProtein Kinase C Isoforms in Rat Fibroblasts DIFFERENTIALALTERATIONS

IN ras-, src-, AND fos-TRANSFORMED CELLS* (Received for publication, October 24, 1991)

Christoph Borner$$,Sarah Nichols GuadagnoSlI, WendyW.-L. HsiaoII , Doriano Fabbro**, Maureen Barr$#, and I. Bernard Weinstein$@$$ From the $Comprehensive Cancer Center and Institute forCancer Research, Columbia University, the TDepartmentof Pharmacology, Columbia University, New York, New York 10032, the **Department of Pharmaceutical Research, Oncology K125, Ciba-Geigy, Ltd., CH-4002 Basel, Switzerland, theIIDepartment of Molecular Biology and Biochemistry, University of California at Iruine, Imine, California 92171, and the §§Departmentof Genetics and Development, Columbia University, New York, New York 10032

In theaccompanying study (Borner, C. B., Guadagno, PKC in the samehost cell, and they suggest thatindiS. N., and Weinstein, I. B. (1992) J. Biol. Chem. 267, vidual isoforms may play distinct roles in mediating 12892-12899) we found that R6 embryo fibroblasts cellular transformation by specific oncogenes. express fourisoforms of PKC, cPKCa, nPKCt, nPKC6, and nPKCl whose subcellular distribution, activation, and down-regulation are differentially regulated. Furthermore, we demonstrated that overproduction of an The acquisition of a tumorigenic phenotype is a complex, exogenous cPKCj31 isoform in these cells (R6-PKC3) altered the TPA-induced down-regulation of nPKC6 multistep process which in several systems comprises three and nPKCc. In this paper we show that transformation distinct phases: initiation, promotion, and progression (1-3). of R6 or R6-PKC3 cells with a variety of different Initiation often involves the mutational activation of oncooncogenes results in differential alterations in expres-genes, whereas promotion results in the selective outgrowth sion of individual PKC isoforms. R6 orR6-PKC3 cells of initiated cells permitting the occurrence of secondary getransformed by an activated c-H-ras oncogene dis- netic changes which progressively convert the cell intoa played a marked increase in the expression of both tumorigenic phenotype. On mouse skin, the promotion phase cPKCa and nPKC6, decreased expression of nPKCc, can be induced by tumor-promoting phorbol esters, such as (TPA)‘ (1-3). It is now and no change in the expression of nPKCf. These al- 12-O-tetradecanoylphorbol-13-acetate terations occurred both at the mRNA and protein levels widely accepted that the cellular binding site for these combut did not significantly affect the subcellular distri- pounds is protein kinase C (PKC), a lipid-regulated serine/ bution of any of the four isoforms. Studies using acti- threonine proteinkinase (4), andmost, if not all, of the effects nomycin D and nuclear run-off assays indicated that of these compounds are mediated through this enzyme. the increased expressionof cPKCa in ras-transformed This scenario of multistep carcinogenesis suggests a close cells was due to increased de novo transcription rather cooperativity between oncogene encoded proteins and PKC than increased mRNA stability. Qualitatively similar, in regulating cellular growth and neoplastic transformation but less extensive changes in the expression of the four (5, 6). At the plasma membrane, oncogene products such as PKC isoforms wereseen inv-fos-transformedR6the epidermal growth factor receptor homolog neu/erb-B2 (7, PKC3 cells. Decreased expression of nPKCc was also 8), the G-protein p21 ras (9), andthe tyrosine protein kinase seen in the v-src-transformed R6- and R6-PKC3 lines; pp60 src (10) have been shown to increase the cellular levels however, the cellular level of cPKCBI appeared to bea of diacylglycerol (DAG) (11-18), an endogenous activator of limiting factor in mediating the effects of v-src on the increased expression of cPKCa and nPKC6. Interest- PKC (19-21). Additionally, numerous findings have been ingly, no major changes in the levels of expression of published which support a direct involvement of PKC in the any of the four PKC isoforms were found when R6 generation of cellular transformation by the ras or src oncocells were transformed by myc, neu/erb-B2, or mos genes (14,15, 22-31). However, the fact that certain ras-, src-, or neu/erb-B2-stimulated cellular events occur independently oncogenes. These results demonstrate that transformation of R6 cells by the oncogenes ras, src, and fos of PKC activation (10, 32, 33) indicates that the cooperation differentially alter theexpression of three isoforms of between these oncogenes and PKC cannot be explained by a simple linear pathway. In the nucleus, certain oncoproteins appear to function as * This work was supported in part by a National Cancer Institute ultimate effectors of PKC-dependent signalling or to coopergrant and an award from the Markey Charitable Trust (to I. B. W.) and by a Swiss National Science Foundation Grant (to D. F.). The ate with other PKC-regulated nuclear proteins in altering 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. § Recipient of a postdoctoral fellowship supported by the Swiss National Science Foundation. Current address: Inst. of Biochemistry, University of Lausanne, 155 Ch. des Boveresses, CH-1066 Epilanges, Switzerland. $3 To whom correspondence should be addressed. Comprehensive Cancer Ctr., and Inst. for Cancer Research, Columbia University, 701 W. 168th St., New York, NY 10032.

The abbreviations used are: TPA, 12-0-tetradecanoylphorbol-13acetate; PKC, proteinkinase C; cPKC, conventional Ca2+-dependent protein kinase C; nPKC, novel Ca2+-independentprotein kinase C; DAG,diacylglycerol;Ac-pep, synthetic peptide encompassing the pseudosubstrate region of nPKCc; EGTA, [ethylenebis(oxyethylenenitri1o)ltetraacetic acid; SDS, sodium dodecyl sulfate; kb, kilobase(s); bp, base pair(s); GAPDH, glyceraldehyde-phosphate dehydrogenase; Me2S0, dimethyl sulfoxide; DMEM, Dulbecco’s modified Eagle’s medium; HEPES, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid.

12900

Altered PKC Isoform Expression in TransformedFibroblasts gene expression. The transcription factors c-fos, c-jun, and Cmyc have been implicated in PKC action because they are rapidly induced by a variety of growth factors and other extracellular stimuli, including TPA (34, 35). Furthermore, treatment of cells with TPA activates the expression of several genes that carry the upstream regulatory element TPA response element which recognizesfos-jun heterodimers or junj u n homodimers as part of the AP-1 transcription factor complex (36,37).These various gene products may, therefore, provide a crucial link between signals at thecell surface, PKC activation, and the control of gene expression, growth, and differentiation. The complexity of the above-described synergistic interactions between PKC and specific oncogenes is compounded by the fact that PKC is a multigene family that encodes at least eight distinct isoforms. They can be grouped into Ca2+-dependent conventional PKCs (cPKC: a,D l , p2, y) (38-43) and Ca2+-independentnovel PKCs (nPKC: e, 6, 5; q/L) (41, 4447). The preservation of these multiple isoforms during evolution (48), their differential tissue distribution (48, 49), and subtle differences in enzymologic properties (50-53) and substrate specificities (54-56) suggest that each isoform might produce distinct biologic effects. However, this fact has not been directly demonstrated. Since single cells often express more than one isoform of PKC (48,49),to study the role of PKC incell transformation one must refine the analysis by analyzing specific isoforms. We have previously reported that theexpression of cPKCa is induced, while that of nPKCc is decreased, in R6 rat embryo fibroblaststransformed by an activated c-H-ras oncogene (57). Since these changes in the expression of two isoforms of PKC were a directconsequence of ras expression, we reasoned that they may represent an essential step in ras-mediated transformation. More recently we found that, in addition to cPKCa andnPKCc, R6 cells also express nPKC6 and nPKC< (76). We also found that these four isoforms differ in subcellular distribution and in TPA-induced activation and downregulation (76). As an extension of these findings, the present study examines alterations in the expression and subcellular distribution of the four endogenous PKC isoforms in R6 cells following the transformation of these cells by the oncogenes ras, src, neu, fos, myc, and mos. EXPERIMENTALPROCEDURES

Materials-PKC isoform-specific antibodies, cDNA probes, and the pseudosubstrate peptide analog Ac-pepwere from sources described elsewhere (76). Plasmids containing full length cDNAs for oncogenes were kindly provided as follows:pv-mycwas from Dr. S.Goff,' Columbia University, pMATV8 (pp60v-src) was from Dr. D. Shalloway, Cornel1 University (58), prc-mos (rearranged mouse cn o s ) , was from Dr. M. Horowitz, Weizmann Institute, Rehovot, Israel (59), psv2neuN and psvneuT were from Dr. M. C. Hung, M.D. Anderson Cancer Center, Houston, TX (60), pFBJ2 (v-jos) was from Dr. I. Verma, The Salk Institute, San Diego, CA (61), and pIBW (tkneo) was from Dr. R. Axel, Columbia University (30). Staurosporine derivatives CGP 41251 and CGP 42700 were a generous gift of Ciba Geigy Ltd. (Basel, Switzerland) (62). DEAE-Sephacel was obtained from Pharmacia LKB Biotechnology Inc., geneticin (G418) was purchased from GIBCO-BRL, and TPA was from LC Services Corporation (Woburn, MA). All other chemicals were reagent grade. Cell Lines-We previously reported the establishment of c-H-rastransformed derivatives of R6 (R6-C1/T24) (30) and R6-PKC3 (R6PKC3/T24) cells (31). In the present study, we employed two techniques to generate transformed derivatives of R6 cells that express other specific oncogenes. 1) Plasmids carrying neuN (normal protooncogene), neuT (transforming point-mutated oncogene), or v-myc Obtained from Dr. Steven Goff, unpublished data. v-mycwas inserted into aretroviral vector between two directly repeated Moloney murine leukemia virus long terminal repeats.

12901

were cotransfected into R6 cells with a plasmid containing the antibiotic resistance gene neo, and cells were selected for resistance to G418. 2) Plasmids carrying the v-src, rc-mos, v-fos, or v-myc genes were transfected into R6-PKC3 cells and into their vector control counterparts (R6-C1), without subsequent G418 selection since both target cell lines already expressed the antibiotic resistance gene neo (63), and the transfectants were then grown in agar (30) to select for transformed derivatives. The cell lines which displayed the highest level of expression of each oncogene, when examined by Northern blot analysis, and the appropriate normal control cells, were chosen for further studies. When this panel of cells was assayed for growth in agar we found that, whereas the control R6-Cl cells failed to grow, all of the derivatives that expressed either v-src (R6-Cllsrc, R6PKC3/src), rc-mos (RG-Cl/mos, R6-PKC3/mos), v-myc (R6-myc,R6PKC3/myc), neuT (R6-neuT),or c-H-raslT24 (R6-C1/T24, R6-PKC/ T24) or v-jos (R6-PKC3/jos) exhibited efficient colony formation in soft agar. The cell line that overexpressed the normal proto-oncogene neuN(R6-neuN)did not grow in agar. The cell line R6-PKC3lmyc.n~ was derived from a tumor established by injecting R6-PKC3/myctransformed cells into nude mice. The expression of the neogene alone does not alter the expression of PKC isoforms in R6 cells. Cell Culture Procedures and Treatment with Various Agents-All cells were grown in Dulbecco's modified Eagle's medium (DMEM, GIBCO) supplemented with 10% calf serum (HyClone, Logan, UT) and 50 pg/ml G418 in a 37 "C, 6% CO, humidified incubator (30, 31, 63). The medium was changed every 3 days; starving of the cultures and growth to postconfluence were strictly avoided. The cells were harvested for RNA, protein, enzyme extracts, or nuclei when they were subconfluent. Where indicated, R6-PKC3, RG-PKCS/jos, and RG-PKCBlsrccells were treated daily with the PKC-selective inhibitor CGP 41251 (500 nM) (Ciba-Geigy, Basel, Switzerland) (62) or its inactive analog CGP 42700 for 5 days. The compounds were added to the medium as a 10%dimethyl sulfoxide (Me2SO)solution to give a final concentration of 0.1% Me2S0. A Me&O solvent control revealed that this solvent had no effect. Where indicated, R6-C1 and R6-C1/T24 cells were treated with 10 Fg/ml actinomycin D (Sigma) for 2,4,6, and 8 h. After the indicated time total RNA was isolated, as indicated below. Actinomycin D was added to the medium in methanol to give a final concentration of 0.1% methanol. A methanol solvent control had no detectable effect. Soft Agar Growth-2 X lo4 cells were suspended in 2 ml of 0.3% Bacto-agar (Difco Laboratories) in DMEM containing 10% fetal calf serum (Cell Culture Laboratories, Cleveland, OH)(without G418) and overlaid above a layer of 5 ml of 0.5% agar in the same medium, on 35-mm Petri dishes. The cells were then overlaid with 2 mlof DMEM plus 10% fetal calf serum every 4 days. At the end of 30 days colonies were stained with a 40% solution of the vital stain 2-(piodophenyl)-3-(p-nitrophenyl)-5-phenyltetrazoliniumchloride hydrate (Sigma) for 48 h at 37 "C in an incubator with 6% CO,, and the numbers of colonies were counted under low power on an invertedphase microscope. Protein Extraction-For Western immunoblot analyses, cellular proteins were extracted by cell lysis in either preheated SDS (total extracts) or in a EGTA-containing ice-cold buffer with ensuing separation into cytosol and membrane fractions, as described elsewhere (57, 76). The membrane fraction was solubilized in 1%SDS, and the protein content was determined using a modification of the Lowry assay (64). For enzyme assays, protein was extracted as described above for total cell extracts with the exception that the ice-cold extraction buffer contained 1%Nonidet P-40 instead of hot SDS. This Nonidet P-40 extract was cleared by centrifugation at 100,000 X g for 60 min and the resulting pellet was discarded. The protein content was determined according to Bradford (65). Antibodies and Immunoblot Analysis-Polyclonal antisera against PKC-isoform specific peptides were raised in rabbits and characterized, as previously described (57, 76). Electroblotting of proteins onto Immobilon-NC nitrocellulose (Millipore) and subsequent PKC immunodetection with isoform-specific antibodies was also performed exactly as described (57,76).All antisera were used in a 1:500 dilution. RNA Isolation and Northern Blot Analysis-RNA isolation from the various cell lines was always performed in parallel with the extraction of proteins. Usually, cells were plated at a density of 1 X lo6 cells/l5-cm plate, refed the following day, and collected another 48 h later.Ras- and src-transformed cell lines were plated at a density of 1 X lo6 cells per 15-cm dish and collected 48 h after plating. Total RNA was then isolated by cesium chloride centrifugation as described (66). For the detection ofnPKC{, poly(A)+-RNA was isolated by

12902

Altered PKC Isoform Expression in TransformedFibroblasts

oligo(dT)-cellulose chromatography (67). RNA samples were electrophoresed, blotted onto Hybond-N (Amersham), and examined for the presence of PKC isoform-specific transcripts by hybridization to the respective 32P-labeledcDNA probes, as previously described (57, 76). The relative abundance of RNA per lane was judged to be similar by comparing the ethidium bromide staining of the ribosomal bands. For further confirmation, the blots were rehybridized with a probe for an endogenous housekeeping gene, glyceraldehyde-phosphatedehydrogenase (GAPDH). In all cases, the ethidium bromide staining reflected the results obtained by the GAPDH probe. Nuclear Run-on Assay-Nuclei were isolated from R6-Cl and R6Cl/T24 cells, stored at -70 "C in suspension buffer containing 40% glycerol, and assayed as previously described (68,69). After thawing, nuclei were pelleted at 1000 x g at 4 "C for 10 min. Approximately 10" nuclei were then resuspended and incubated at 37 "C for 30 min in 0.25 ml of run-on assay buffer (5 mM dithiothreitol, 5 mMMgC12, 90 mM KC1, 50% glycerol, 20 mM HEPES (pH 7.6), 1 mM GTP, 1 mM CTP, 3 mM ATP, 1 unit/ml RNasin (Promega), and 250 pCi [a'"PIUTP (800 Ci/mmol; Du Pont-New England Nuclear). The nuclei were then centrifuged as described above, and the pellet was washed three times in ice-cold lysis buffer (10 mM NaCl, 10 mM Tris-HC1 (pH 7.4), 3 mM MgC12, and 0.5% Nonidet P-40) followed by centrifugation. The RNA was then purified by ultracentrifugation, as described above, and resuspended in hybridization buffer at l X lo6 cpm/ml, and equal numbers of 10%tricholoroacetic acid-precipitable counts were added to each hybridization. For the preparation of DNA probes to detect run-on-labeled transcripts,supercoiled plasmid DNA containing the 3.0-kb cDNA fragment of murine cPKCa or control cDNAs were applied to aNytran filter in a slot blot apparatus (Schleicher and Schuell), as described (68).Enough plasmid DNA to provide 4.0 pg of each insert cDNA sequence was applied to each slot. The filters were baked and then prehybridized, hybridized, and washed, as described for Northern blot analyses (57, 76), except that, following the high stringency wash, they were incubated at 37 "C for 30 min in 10 pg/ml RNase A in 2 X SSC (1 X SSC: 0.15 M NaCl, 15 mM sodium citrate) prior to drying and exposure to X-ARB film. Partial Purification of PKC and PKC Assay-PKC was partially purified from Nonidet P-40-soluble total cellular extracts by DEAESephacel column chromatography as described (70). A 0.25 M NaCl column eluate was immediately used to measure PKC activity using as substrates histone III-S or the pseudosubstrate peptide analog Acpep, and theassay was performed exactly as described (76). Quantitation of Protein and RNA-The abundance of PKC proteins and mRNAs was quantitated by Betascope scanning of gels and also densitometry of autoradiographs, as previously described (57, 76). Both methods gave similar results. RESULTS

Expression of cPKCa Is Increased in Cells Transformed with c-H-ras, v-fos, and v-src, but Its Distribution between Cytosol and Membrane Fraction Is Not Chunged-We previously reported that theexpression of cPKCa mRNA and protein was increased by 4-5-fold in the c-H-ras-transformed cell lines R6-C1/T24 and R6-PKC3/T24 (57). Results obtained with the PKCinhibitorCGP 41251, astaurosporine derivative (62),provided evidence that theelevated expression of cPKCa mRNA in these cells was independent of intracellular PKC activity. Studies using R6 cells containing an inducible ras gene (57) indicated that the increased expression of cPKCa was a direct consequence of the action of the ras-oncogene and not due to clonal variations (57). In the present study, we examined changes in theexpression of several isoforms of PKC in a series of R6 and R6-PKC3 cells transformed by various oncogenes. The derivation and propertiesof these cell lines are described under "Experimental Procedures." We found a 2-%fold increase in the expression of cPKCa in RG-PKCSlfoscells, predominantly on the protein level (Figs. 1A and 2 A ) , when compared to the control R6-PKC3 cells. Treatment of these cells with a subtoxic dose of 500 nM CGP 41251 for 5 days did not reduce these levels of cPKCa (Fig. 3A). It would appear, therefore, thatthe increased expression of cPKCa in both c-H-ras- andv-fos-transformed cells is mediated by a PKC-independent pathway. RG-Cllsrc

cells displayed an increase in cPKCa expression of only 1.5fold higher than that seen in the control R6-C1 cells (Figs. lA and 2 A ) . On the other hand, in RG-PKCSlsrc cells the expression of both cPKCa mRNA and protein was 10-fold higher than in R6-PKC3 cells (Figs. lA, and 2 A ) . Incubation of RG-PKCS/src cells with 500 nM CGP 41251 for 5 days almost completely eliminated the increased expression of cPKCa in these cells, as evidenced by Northern RNA blot analysis (Fig. 3B). Thus,in contrast to thecase with activated c-H-ras and v-fos oncogenes, the ability of v-src to increase the expression of cPKCa appeared to depend on the presence of a high cellular level of the cPKCPI enzyme. Because the R6-Cllsrc cells displayed only a modest increase in cPKCa expression it is possible that in R6-C1 cells one or more of the endogenous isoforms of PKC is rate-limiting with respect to modulating the level of expression of cPKCa by the p ~ 6 0 " - ~ oncoprotein. R6-Cl or R6-PKC3 cells that expressed the rcmos, c-neuN, c-neuT, or v-myc oncogenes did not display increased expression of cPKCa (Figs. lA and 2 A ) . When Western blot analysis was performed on proteins obtained from the cytosol and membrane fractions of the above-described cells we found that the increased level of cPKCa in the cytosol was always accompanied by a similar increase in the membrane fraction (Fig. 4A). Thus, the distribution of total cellular cPKCa was about 70-80% in the cytosol and 20-30% in the membrane fraction in both the normal and oncogene-transformed cells (Fig. 2A). Expression of nPKCt Is Markedly Decreased in Both the Cytosol and Membrane Fractions of ras- or src-transformed Cells, and Slightly Decreased in Cells Expressing the v-myc, vfos, neu", or neuT Oncogenes-In contrast to theincrease seen with cPKCa, the level of expression of nPKCc was markedly decreased in the c-H-ras-transformed R6-C1/T24 and R6PKC3/T24 cells (57).This decrease was not reversed by treatment of the cells with the PKCinhibitor CGP41251 (57, and Figs. 1D and 2 0 in the present study). Although transformation by v-src produced a significant increase in the expression of cPKCa only in R6-PKC3 cells, it decreased the level of nPKCt mRNA and protein in both R6-C1 and R6PKC3 cells (Figs. 1D and 2 0 ) . Thus the effects of v-src on the expression of nPKCtappear to be less dependent on cellular levels of PKC than are its effects on the expression of cPKCa. The levels of expression of nPKCt were only slightly reduced in R6-myc, R6-muN, R6-neuT, R6-PKC3/ myc.nu, RG-PKCSlfos,and R6-PKC3/mos cells, and no effect was seen in R6-C1 cells transformed by rc-mos (Fig. 2 0 ) . Taken together theseresultsindicate thatthe decreased expression of nPKCt is more frequently seen following transformation of R6 cells by various oncogenes than is the increased expression of cPKCa. The fact that the R6-C1 rcmos-transformed cells did not show changes in the levels of either cPKCa or nPKCt indicates that thechanges seen with some of the other oncogenes are notsimply nonspecific effects of cell transformation. In previous studies we found that 6075% of the total cellular 89-kD nPKCt protein is associated with the membrane fraction (76). This was also the case in the various transformed cell lines examined in this paper (Figs. 2 0 and 4 0 ) . Therefore, the absolute levels of nPKCt protein are markedly decreased in both the membrane and the cytosol of ras- and src-transformed cells (Fig. 40). We have reported that Western blot analyses using nPKCtspecific antipeptide antibodies detected both a major 89-kDa protein and aminor 72-kDa protein (76). The 72-kDa protein might be a subspecies of nPKCc, possibly nPKCt' (45),which is an alternatively spliced form of nPKCt. Since the amino acid sequences of nPKCt and nPKCt', differ only in the

Altered PKC Isoform Expression Transformed in

A

Fibroblasts

12903

cPKCrr

- 2s s

GAPDH

B

nPKCC

D

nPKCE

kb

FIG.1. mRNA Levels of cPKCa, nPKCc, nPKCG, and nPKCf in R6- and R6-PKC3 cells transformed by various oncogenes. Northern blots showing the relative levels of the 8.1- and 3.5-kb transcripts of cPKCa ( A ) ,the 2.4- and 4.1-kb transcripts of nPKCC ( B ) ,the 3.2-kb transcript of nPKCG (C), and the 7.5-kb transcript of nPKCe (D), in R6 or R6-PKC3 cells transformed by the indicated oncogene. Poly(A)+ RNA was used to detect nPKC{, and total RNA was used to detect the other isoforms. The PKC-isoform specific cDNA probes were '"P-labeled and hybridized, as previously described (57). Numbers on the sides of the figures indicate the sizes of the transcripts in kilobases. The positions of 28 S and 18 S ribosomal RNAs are also indicated. The 2.2-kb transcript detected with a nPKCe-specific cDNA probe (RP16) is unrelated to nPKC (76). A probe for GAPDH served as a control for equal RNA loading. In B and C the autoradiographs were cut because samples from other cell lines were run on the same gel. For additional detailssee "Experimental Procedures." amino-terminal portion, the carboxyl-terminal-specific nPKCt antibodies used in our studies would also recognize nPKCt' (76). An alternative explanation is that the 72-kDa protein is derived from the 89-kDa nPKCc protein by in vivo proteolysis (76). In the present study, we found that the amount of this 72-kDa protein varied between the different transformed lines, and that itwas most abundant in ras- and mos-transformed cells (Figs. 2 0 and 4 0 ) , even though they displayed noticeably different levels of the 89-kDa nPKCe protein (Figs. 2 0 and 4 0 ) and nPKCt mRNA. Further studies, however, are required to determine the origin of the 72kDa protein and if the protein is encoded by a distinct gene whose expression is increased in ras- and mos-transformed cells. nPKCGmRNA and 74- and 76-kDa nPKCG Proteins Are Markedly Increased in Cells Transformed by ras and src, and the Proteins Are Predominantly Membrane-associatedNorthern blot analysis detected a single 3.2-kb mRNA transcript for nPKCG which was increased in abundance by about 3-4-fold in R6-C1/T24 and R6-PKC3/T24 cells (Fig. IC), when compared to theirrespective controls. A similar increase

in PKCG mRNA was found in src-transformed cells, but as with cPKCa (Fig. lA)this increase was seen only in cells whichhave high amounts of cPKCBI (R6-PKCIlsrc) (Fig. 1C). Furtherevidence that cPKCPI plays a criticalrole in the increased levels of cPKCG was obtained by incubation of R6PKC3/src cells with the PKC inhibitor CGP41251, since this resulted in a marked decrease in nPKCG mRNA (Fig. 3'2). In contrast, R6-Clderivatives that expressed rc-mos, muN,muT, or v-myc, and R6-PKC3 cells transformed by v-myc or v-fos, exhibited levels of nPKCG mRNA that were comparable to those of the corresponding control cells. We have previously shown that in R6-Cl and R6-PKC3 cells nPKCG is expressed as a protein doublet with apparent molecular masses of 74 and 76 kDa. Preliminary studies have indicated that the 76-kDa protein may be a phosphorylated and possibly more activated version of the 74-kDa protein (76). Using Western blot analysis, we found that the total levels of both the 74-kDa and the 76-kDa nPKCG proteins were markedly increased in R6-C1/T24, R6-PKC3/T24, and R6-PKC3/src cells and slightly increased in R6-muN, R6muT, R6-myc,RG-PKC3/mos, and R6-PKC/fos cells (Figs.

Altered PKC Isoform Expression in Transformed Fibroblasts

12904

B

. .-

- pa1

P72

cPKC a

nPKC