Diacylglycerol Modulates Binding and Phosphorylation of the ...

15 downloads 0 Views 2MB Size Report
Printed in U.S.A.. Diacylglycerol Modulates Binding and Phosphorylation of the. Epidermal Growth Factor Receptor*. (Received for publication, April 23, 1984).
Vol. 259, No. 20, Issue of October 25, pp. 12502-12507,1984

THEJOURNAL OF BIOLOGICAL CHEMISTRY

0 1984 by The American Society of Biological Chemists, Inc

Printed in U.S.A.

Diacylglycerol Modulates Binding and Phosphorylation of the Epidermal Growth FactorReceptor* (Received for publication, April 23, 1984)

Patricia G. McCaffrey, BethAnnFriedman, and Marsha Rich RosnerS From the Laboratory of Toxicology, Department of Nutrition and Food Science, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139

Tumor promoters cause a variety of effects in cul- derived growth factor, and somatomedins (1, 2). It has been tured cells, at least some of which are thought to result shown in certain quiescent cells that all three of these growth fromactivation of the Ca2+-phospholipid-stimulated factors are required for mitogenesis. In these systems, tumor protein kinaseC. One actionof tumor promotersis the promoters can substitute for anytwo of the above hormones modulation of the binding and phosphorylationof the epidermal growth factor(EGF) receptor in A431 cells. to stimulate growth (2). The synergism with growth factors, To determine if these compoundsact on the EGF recep- as well as the similarity in cellular responses, suggest that tor by substituting for the endogenous activator of C tumor promoter actioninvolves modulation of normal growth kinase, diacylglycerol, we compared the effects of the regulatory pathways. In support of this idea, recent results from our laboratory potenttumorpromoter12-0-tetradecanoylphorbol 13-acetate (TPA) with those of the synthetic diacyl- (3, 4) and others ( 5 , 6) have indicated that tumor promoters glycerol analog1-oleyl2-acetyldiglycerol (OADG). do in fact alter the actionof EGF by modulating the binding When A431 cells were treated with TPA, the subcel- and phosphorylation of the EGF receptor. Tumor promoters lular distribution of C kinase activity shifted from a have been shownto block EGF binding toa class of apparent predominantly cytosolic location toa membrane-associated state; OADG also caused the disappearance of highaffinity EGF receptors in several cell types (3, 7-9). Upon binding, EGF stimulates a receptor-associated tyrosylcytosolic C kinase activity. The shift in the subcellular carcinoma (A431) distribution of C kinase, caused by TPA or OADG, protein kinaseactivity. In human epidermal correlated with changes in binding and phosphoryla- cells, a primary substrate for this tyrosyl kinase is the EGF tion of the EGF receptor.OADG, like TPA, caused loss receptor itself (10). Potent tumor promoters from three repof binding to an apparent high affinity class of recep- resentative classes, namely phorbol esters, indole alkaloids, tors, blocked EGF-inducedtyrosine phosphorylationof and polyacetates, block EGF-stimulated tyrosine phosphorylthe EGF receptor, and stimulated phosphorylation of ation of the receptor (3, 5 ) . The loss of tyrosine phosphorylthe EGF receptor at both serine and threonine residues. ation in A431 cells in response to tumor promoters correlates No differencebetweenthephosphopeptidemaps of high affinity receptors from cells treated withOADG or TPA was withinhibition of EGF binding to apparent receptors (3). In addition to inhibition of EGF binding and observed. Thus, it appears that tumor promoters can exert their effects on the EGF receptorsby substituting loss of receptor phosphorylation at tyrosine residues, a third, for diacylglycerol, presumably by activating protein possibly related, modification of EGF receptors in A431 cells i n viuo enhancement of EGF kinase C. Further, these results suggest that endoge- by tumorpromotersisthe a role in nously produceddiacylglycerolmayhave receptor phosphorylation a t serine and threonineresidues ( 5 , normal growth regulatory pathways. 6). One potential mediator of the effects of tumor promoters on the EGF receptor i n vivo is thecyclic nucleotide-independCa2+-phospholipid-stimulated protein kinase C first deent, The mechanismby which tumor promoters enhance tumor formation in animals is unknown. Tumor promoters have a scribed by Nishizuka and co-workers (11).This enzyme is a variety of effects in cultured cells, many of which are also serine- andthreonine-specific protein kinase. The fact thatC induced by endogenous growth factors such asEGF,’ platelet- kinase co-purifies with phorbol ester bindingactivity (12) _ _ _ ~ suggests that this kinase is a major receptor for tumor pro* This work was supported by National Institutes of Health Grants moters. Representative compounds from threedifferent CA35541-01 and CA32267-01 to M. R. R. The costs of publication of classes of tumor promoters (TPA, teleocidin, and debromoathis article were defrayed in part by the payment of page charges. This articlemusttherefore be hereby marked“aduertisement” in plysiatoxin) have been shown to directly activate C kinase i n vitro in the presence of Ca2+and phospholipid (13, 14). TPA, accordance with 18 U.S.C. Section 1734 solely to indicat.e this fact. and presumably other tumor promoters, activateC kinase i n $ To whom correspondence should be addressed at the Laboratory of Toxicology, Department of Nutrition and Food Science, Building vitro by substituting for diacylglycerol (13), which must be E18-506, Massachusetts Institute of Technology, Cambridge, MA present in addition to phosphatidylserine for full activation 02139. The abbreviations used are: EGF, epidermal growth factor; TPA, at physiological concentrations of calcium(15). TPA also 12-0-tetradecanoyl phorbol13-acetate; OADG, 1-oleyl 2-acetyl di- directly activates C kinase i n viuo, as demonstrated in intact glycerol; C kinase or protein kinase C, Ca2+-phospholipid-dependent platelets where TPA induces phosphorylation of a 40-kDa protein kinase; PDGF, platelet-derived growth factor; Me2S0, di- protein substrate for C kinase (13). A variety of effectors, methyl sulfoxide; DME, Dulbecco’s modified Eagle’s medium; PBS, including growth factors, induce phosphatidylinositol breakphosphate-buffered saline; BSA, bovine serum albumin; EGTA, ethylene glycol his(@-aminoethyl ether)-N,N,N’,N’-tetraacetic acid down upon binding to theirspecific receptors, resulting in the PMSF, phenylmethylsulfonyl fluoride; SDS, sodium dodecyl sulfate; transient production of diacylglycerol (16-18). Thus, the production of diacylglycerol and subsequent activation of C kiPAGE, polyacrylamide gel electrophoresis. ~~

12502

Diacylglycerol Modulates EGF Receptor Action

12503

stimulated kinase activity. The DEAE-cellulose column profile shows a broad peak of Ca2+-phospholipid-stimulated kinase activity which may not all be due to C kinase; however, further resolution of the kinases by high-performance liquid chromatography resulted in substantialloss of activity leading toinaccuratedeterminations.Therefore, we used DEAEcellulose to isolateC kinase and focused on the peaks of activity which appeared to be responsive to TPA andOADG treatment. The demonstration that an endogenous-type activator of C kinase can causereversible loss of cytosolic C kinase activity suggests that, in uiuo, membrane association is important for enzyme activity. I n uitro, protein kinase C is inactive unless phospholipids are present (11).Full activation requires diacylglycerol, which increasesthe affinity of C kinase for calcium and phospholipid (15). Endogenously produced diacylglycerol could enhance the association of protein kinase C with membrane phospholipids, resulting in an increase in enzymeactivity. The effect of OADG on the distributionof C kinase is not EXPERIMENTAL PROCEDURES AND RESULTS’ as persistent as that of TPA, so it is likely that the loss of cytosolic activity seen even with quick fractionation of cells DISCUSSION after OADG treatment underestimates the actual extent of The mechanismby which tumor promoters act to modulatemembrane association of the enzyme in uiuo. Nonetheless, a the phosphorylation and binding of the EGF receptor A431 in shift insubcellular distribution is detectablewith both OADG cells i n vivo wasinvestigated. The tumor promoter TPA, and TPA atdoses which also producebiological effects. which causes a variety of effects at the cell surface,was We have shown that, in uiuo, both OADG and TPA cause compared with OADG, a synthetic analog of the endogenous enhanced serine and threonine phosphorylation of the EGF diacylglycerol produced as a result of phosphatidylinositol receptor. It is possible that activated C kinase might directly turnover. Treatment of A431 cells with TPA resulted in an phosphorylate the EGF receptor, given the specificity of C increase in membrane-associated protein kinase C activity kinase and the common membrane location of the two proand concurrent loss of the cytosolic enzyme activity. OADG teins. Alternatively, TPA and OADG may activate another caused asimilar loss of cytosolic Ckinase activitywhen added kinase, such as the Ca2+-independent, protease-activated kito intactA431 cells. The shift insubcellular distribution of C nase I1 (29). However, a recent report shows that purified kinase induced by TPA and OADG correlated with changes proteinkinase C canphosphorylatetheEGF receptorin in binding and phosphorylation of the EGF receptor. These membrane preparations from A431 cells and that, after such effects included loss of binding to a class of apparent high treatment, EGF-stimulated phosphorylation of the EGF reaffinity EGF receptors, loss of EGF-induced tyrosine phosphorylation, and stimulation of serine and threonine phos- ceptor a t tyrosine residues is reduced (5). Further, the sites phorylation. These results indicate that OADG can modulate phosphorylated by purified protein kinase C in uitro appear upon TPA treatmentof intact the EGF receptor in a manner similar to TPA, presumably similar to those phosphorylated via activation of C kinase. Further, the effects of OADG on cells as judged by tryptic phosphopeptide mapping. The i n uiuo work presented here, together with thesein uitro studies, the EGF receptor suggest that diacylglycerol itself may be provide strong evidence for a role of C kinase in the actionof important in the pathwayof growth factor action incells. The shift in the subcellular distribution of C kinase induced tumor promoters and OADG on the EGFreceptor. The demonstration thatOADG can modulate the phosphoby TPA in A431 cells is similar to that seen in at least two other cell types (25, 26). However, a recent study of C kinase rylation and bindingof the EGFreceptor implies that endogdistribution in A431 cellsconcluded that TPA induced no enously produced diacylglycerol has a role in modulation of change in cytosolic activity relative to the control, whereas growth factor receptors i n uiuo and may be an important EGF caused a marked increase (27). In the latter study, the component of the normal response to growth stimuli. EGF cytosol was assayed directlyfor enzyme act,ivltywithout prior stimulates limited phosphatidylinositol turnover in A431 and fractionation on DEAE-cellulose. Under these conditions,we 3T3 cells (17, 181, which presumably leads to diacylglycerol production in these cell types. It has recently been reported were unable to detect significant Ca”-phospholipid-stimuthat the EGF-stimulated increase indiacylglycerol production lated kinase activity, presumably due to inhibition b> other cytosolic components. A cytosolic inhibitor of C kinase activ- in A431 cells peaks between 5 and 10 min and returns to ity has recently been identified as a phosphatase which can control levels by 40 min(30). Ourobservation that EGF be separated from the kinase by DEAE-cellulose chromatog- induced nochangein C kinasedistribution at 40 min is raphy (28). Thus, further fractionationof the cytosol appears consistent with thisreport. However, because the membraneto be required for accurate assessment of Ca’+-phospholipid- C kinase association due to the presence of diacylglycerol is reversible, we might not be able to detect small changes in * Portions of this paper (including “Experimental Procedures,” the subcellular distribution of C kinase with our preparative “Results,” Figs. 1-6, and Tables 1 and 2) are presented in miniprint procedures. Others have noted some similarities between the at the end of this paper. Miniprint is easily read with the aid of a EGF receptor tryptic phosphopeptides derived from EGFstandard magnifying glass. Full size photocopies are available from treated A431 cells and those derived from TPA-treated cells the Journal of Biological Chemistry, 9650 Rockville Pike, Rethesda, MD 20814. Request Document No. 84M-1208, cite the authors, and (5, 6). Therefore, it is possible that EGF, through limited include a check or money order for $9.20 per set of photocopies. Full diacylglycerol production at early times, stimulates C kinase size photocopies are also included in the microfilm edition of the and by SO doing autoregulates the EGF-stimulated tyrosine Journal that is available from Waverly Press. kinase activity associated with EGF binding.

nase may also be a n early event in the normal response to growth stimuli. In order to determineif the effects of tumor promoters on in vivo reflect the abilityof these compounds the EGF receptor to substitute for diacylglycerol, we compared the effects of TPA with those of a synthetic diacylglycerol, OADG. The results presented here demonstrate that OADG mimics the effects of TPA on the binding and phosphorylation of the EGF receptor in A431 cells. These changes correlate with membrane association of C kinase and presumably enzyme activation. Thus, these results are consistentwith the possibility that TPA can act in this system by substituting for diacylglycerol, probably at the level of C kinase activation. These findings indicate thatdiacylglycerol can play a role in modulating growth factor receptors andmay be an important second messenger in the normal response to growth factors that induce phosphatidylinositol turnover.

12504

Diacylglycerol Modulates EGF Receptor Action

The findings presented here support a model (see Ref. 3) where TPA or diacylglycerol cause phosphorylation of the EGF receptor in A431 cells, either directly or indirectly, by a mechanism involving C kinase. The functional importanceof thisphosphorylationis suggested by the observation that inhibition of binding and loss of tyrosine phosphorylation occur a t doses of TPA which also cause increased serine and threonine phosphorylation of the receptor. The shift in subcellular distribution of C kinase to a membrane-associated state uponTPA or OADG treatment of cells supports therole of this enzyme as a likely mediator of the effects of tumor promoters and diacylglycerol on theEGF receptor. In light of these results, it would be expected that other ligands which stimulate diacylglycerol production should have the same effects on the EGF receptor. In particular, recent studies show that PDGF is a strong inducer of phosphatidylinositol turnover (18) and further suggest that this growth factor causes activation of C kinase (31). PDGF does reduce EGF binding to its receptor in 3T3 cells (32, 33). PDGF is a growth factor that appears to act early in the process of growth stimulation, whereas EGF acts later. Themodulation of the EGF receptor by a pathway involving diacylglycerol and presumably C kinase providesamechanismwhereby PDGF and EGF canactsequentiallyinstimulating cell growth. Acknowledgments-we are grateful to S. Decker for anti-EGF receptor antiserum and to Y. Nishizuka for OADG. We thank G . Foulkes for helpful comments and K. Schaefer for assistance in preparation of the manuscript. REFERENCES 1. Dicker, P., and Rozengurt, E. (1979) J. Supramol. Struct. 11, 79-93 2. Frantz, C.N., Stiles, C . D., and Scher, C. D. (1979) J. Cell. Physiol. 100,413-424 3. Friedman, B., Frackelton, A. R., Ross, A. H., Connors, J. M., Fujiki, H., Sugimura, T., and Rosner, M. R. (1984) Proc. Natl. Acad. Sci. U. S. A. 81, 3034-3038 4. Rosner, M. R., Friedman, B., Frackelton, A. R., Ross, A., and Sugimura, T. (1983) Fed. Proc. 4 2 , 1905 (abstr.) 5. Cochet, C., Gill, G . N., Meisenhelder, J., Cooper, J. A., and Hunter, T. (1984) J. Biol. Chem. 269,2553-2558 6. Iwashita, S., and Fox, C. F. (1984) J. Biol. Chem. 2 5 9 , 25592567

7. Lee, L. S., and Weinstein, I. B. (1978) Science 2 0 2 , 313-315 8. Shoyab, M.,De Larco, J. E., and Todaro, G . J. (1976) Nature (Lord.)279,387-391 9. Horowitz, A. D., Fujiki, H., Weinstein, I. B., Jeffrey, A., Okin, E., Moore, R. E., and Sugimura, T. (1983) Cancer Res. 43, 15291535 10. Hunter, T., and Cooper, J. A. (1981) Cell 2 4 , 741-752 11. Takai, Y., Kishimoto, A., Iwasa, Y., Kawahara, Y., Mori, T., and Nishizuka, Y. (1979) J. Biol. Chem. 254,3692-3695 12. Niedel, J. E., Kuhn, L. J., and Vandenbark, G . R. (1983) Proc. Natl. Acad. Sci. U. S. A. 8 0 , 36-40 13. Castagna, M., Takai, Y., Kaibuchi, K., Sano, K., Kikkawa, U., and Nishizuka, Y. (1982) J. Biol. Chem. 2 5 7 , 7847-7851 14. Fujiki, H., Tanaka, Y., Miyake, R., Kikkawa, U., Nishizuka, Y., and Sugimura, T. (1984) Biochem. Biophys. Res. Commun. 120, 339-343 15. Kishimoto, A., Takai, Y., Mori, T., Kikkawa, U., and Nishizuka, Y. (1980) J. Bwl. Chem. 255,2273-2276 16. Michell, R. H. (1979) Trends Bwchem. Sci. 4,128-131 17. Sawyer, S. T., and Cohen, S. (1981) Biochemistry 20,6280-6286 18. Habenicht, A. J. R., Glomset, J. A., King, W. C., Nist, C., Mitchell, C. D., and Ross, R. (1981) J. Biol. Chem. 2 5 6 , 12329-12335 19. Frackelton, A. R., Jr., Ross, A. H., and Eisen, H. N. (1983) Mol. Cell. Bwl. 3 , 1343-1352 20. Bradford, M. (1976) Anal. Biochem. 7 2 , 248-254 21. Peterson, G . L. (1977) Anal. Biochem. 8 3 , 346-356 22. Reynolds, F. H., Jr., Todaro, G . J., Fryling, C., and Stephenson, J . R. (1981) Nature (Lord.)292, 259-262 23. Scatchard, G . (1949) Ann. N . Y. Acad. Sci. 51,660-672 24. Kaibuchi, K., Takai, Y., Sawamura, M., Hoshijima, M., Fujikura, T., andNishizuka, Y. (1983) J. Biol. Chem. 258,6701-6704 25. Kraft, A. S., Anderson, W. B., Cooper, H. L., and Sando, J. J. (1982) J. Biol. Chem. 257, 13193-13196 26. Kraft, A. S., and Anderson, W.B. (1983) Nature ( L a n d . ) 3 0 1 , 621-623 27. Sahai, A., Smith, K. B., Panneerselvam, M., and Salomon, D. S. (1982) Bwchem. Biophys. Res. Commun. 109, 1206-1214 28. Sahyoun, N., LeVine, H., 111, McConnel, R., Bronson, D., and Cuatrecasas, P. (1983) Proc. Natl. Acad. Sci.U. S. A. 80,67606764 29. Gonzatti-Haces, M. I., and Traugh, J. (1984) J. Cell. Biochem. BA, 289 (abstr.) 30. Smith, K. B., Losonczy, I., Sahai, A., Panneerselvam, M., Fehnel, P., and Salomon, D. S. (1983) J. CeU. Physwl. 117,91-100 31. Rozengurt, E., Rodriguez-Pena, M., and Smith, K. A. (1983) Proc. Natl. Acad. Sci. U. S. A. 8 0 , 7244-7248 32. Bowen-Pope, D. F., Dicorleto, P. E., and Ross, R. (1983) J. Cell Bid. 96,679-683 33. Wharton, W., Leof, E., Pledger, W. J., and O’Keefe, E. J. (1982) Proc. Natl. Acad. Sci. U. S. A. 79,5567-5571

Diacylglycerol Modulates EGF Receptor Action

12505

supp1uentary materia1 to DIACYVILYCEROL UODUUTBS BINDING AND PHGSPBORYLITIGN OF TSE CDIDBRNAL G R O W H FACTOR RBCBPTOR by Patricia G. IcCaffray. BothAnn Prrsdman, and Uaraha Rich Aoaner

Er~arimental Procedure8

-

Mterials TPA was from Si-. O A D j wae the gsneroua gift of Y. Nillhizuka. Anti-EOP recsptor antiserum was kindly provided by 9. Decker. mnoclonal antipbosphotyroeine antibodiee *ere purified fron ascites fluid from mice injected with a hybridoma cell line dsscribed previously 1191.

-

Cell culture and fractionation A431 h-n epidermal carcinoma c e l l s were g r a n in W E supplemented with 108 fetal calf serum IGibcol. For cell fractionation experiments, cells were washed with PBS and incubated in DIE containing 0 . 5 1 BSA. Dimethyl sulfoxide lIe2SOl. or TPA dissolved in Ie2SO vere added all indicated. OALG e s sonicated in He2So:PBS (1:ll just before sddition. After incubation for 10 mlnutea at 37.C. plates were washad and scraped in 20 DM Tris-iiCl pH 7.5, 2 M EDTA, 0 . 5 mn E T A , 10 ug/ml aprotinln, 1111 PNSF and 30 W B-rcaptoethanol or 5 MI dithiothreitol. Lysed cell eltracts vere spun at 100,000rg for 60 minutes, yielding the cytosolicsupernatant. For isolation of -brans fr*Etions, -11s pretreated illi above were washed with PBS-0.5 M EDTA and allowed to lift off the plate. Cells were Ulen disrupted in a Dounee homogenizer in the buffer described above. and the extract Was *pun briefly at IOOOxg to pellet nuclei. The resulting supernatant was spun at 10D,OOOrg for 30 minutes to yield the crude membrane fraction and a CytosOliC supernatant, Nembroine proteins *ere solubilized in 18 Triton X-100 for 60 minutes on ice. and then spun for 30 minutes st 1Oo.oooxg to remove particulate material prior to fractionation on D ~ ~ - c e I l ~ l oass e described below. Protein conceitration van determined by the method Of Bradford 120) or by a nodifled Lorry method 1211.

-

DEAE-cellulose Fractionation One ml DEAE-cellulose (DE-52. Whatman) colunns were equllibrated with 20 mN TIis-HC1 PH 1.5, 2 ml4 EDTA. 0 . 5 M EGTA, 1 M PnSP and 3 0 mN B-nercaptoethanol or 1 MI dithiothreitol. Samples of cytosolic or salubililed m m b r a n e protein yere applied and the columna washed with 10 n l n of buffer. The columns were eluted with a linear gradient Of 20 mls Of buffer and 20 m1s of buffer containing 0.15 n NaC1. One m 1 fractione vere wllected and 50 "1 aliquots of each fraction were assayed for c kinase activity a8 described below. C K ~ n a 8 e~ n s a y - Activity of the Ca2+-phospholipid stimulated proteln kxnase c wae aaaayed in a reaction nix containing 20 mn Tria-HC1 pH 7.4. 10 mw IgC12, 1 M CaC1 400 Ug/ml histone ltype I l l - S , Sigmal, 5 0 Y I yJ2P-ATP 45-10 x lo5 cpm sample). 160 q / m l 1-7-phosphatidyl serlne and 1.6 ug/ml 1.2-diolein. For Control samples, phosphatidylserine and d m l e l n were omitted. Sample8 yere incubated for 5 minutes at 30'C and d i q u a t s were *potted o n whatman P81 phosphocellulose paper. Assays were washed in 30 m phosphoric acid. dried and counted. Under these conditmns, enzyme activity was linear with both time and protein concentration.

&

-

EGP ReCePtDr IsOlation The EGF receptor was precipitated with II m o n o ~ l o n a l antiphosphotyrosine antibody as previously described 119). Briefly, A431 ~ 1 1 a1105 cells per 3 5 m dish) were preincubated with 0 . 5 m c i 32P in phosphate-free DIE plu8 dialyzed fetal calf serum for 3 hr at 37-c. Cells were treated with Ie2SO. TPA I100 ng/mll. or O A K I 8 0 ug/mll for 10 minutes at 37'C. followed by EGP 1150 ng/mll for an additional 40 minutes. The cultures were then placed on ice, the binding media removed and the cells scraped and extractedwith buffer contalning 10 M Tris, 1% Trlton X-100, 5 0 M NaC1, 5 M EDTA, 1 m M PHSF. 0.18 BSA, 1 ug/ml aprotinin, 50 MI NaF, and 30 nul pyrophosphate, pH 8. Cell extracts were dialysed against the extraction buffer, preincubated with BSA coupled to Sephairose 48. and then absorbed to monoclonal antiphosphotyrosine antibody coupled Sepharose 48. The 32P-labelled phDsphDtyrOsine containing protein8 were eluted wlth buffer containing 40 mN phenylphoephate, denatured in sDs-B-mercaptoethanal for 10 minutes at IOO'C, separated on 7.5% polyacrylamide gels and visualized by autoradiography. For precipitatlon with polyclonal anti-receptor antlbody, cells were pretreated exactly as above, and solubilized in modlfLed RIPR buffer 111 Triton X - 1 0 0 , 0 . 5 8 deoxycholate, 0 . l e SDS. 100 M NaCl, 10 MI Trls, 2 rrm PMSF. 1 ug/ml aprotinin, 5 OH EDTA, 30 M pyrophosphate, 5 0 mn Nap. 0.1% BSA, pH 8 ) . Cell lysates were incvbated vith polyclonal antireceptor antibody at 4 - C and then absorbed to a 10-fold excess of protein A (lgsorb. The Enzyme Center, Inc.). Receptor-antibody complexes were eluted b7 boiling for 5 minute9 in buffer containing 100 M Trla pH 6.8, 2 IM EDTA,

t, " 5 w

v)

2 z o W

E

101

cI

1

0

IO 20 30 FRACTION NUMBER

40

Figure 1. Effect Of TPA or OALG treatment an Cyto801iC Ca2+-phospholipid stimulated protein kinase activity C in A 4 3 1 cells. Cytosolic protein* ( 4 mgl from c e l l a treated With 100 ng/ml E-phorbol IA) 100 ng/mI TPA IB). Or 80 ug/ml O A Z IC) Were fractionated on DEAL? cellulo& l l S described. FILCtiDns Were assayed in the preaence I O I or aboence I 0 I of 160 us/ml phosphatidyleerine and 1.6 ug/ml 1.2-diolein; Ca2+ was included L O a11 L B B ~ Y B . Protein kinase activity is expreesed as cpm incorporated into histone/ 5 min/ 5 0 "1 aliquot.

S u b ~ e l l u l a rdistrlbvtion Of C krnase follovlng TPA treatment was assessed using this assay. The CytOSoliC C kinase activlty vas dramatically reduced when c e l l s were pretreated vith 100 ng/ml of TPA for 10 minutel st 31-c IPigure 18). TO determine whether the reduction in CytOsOlx c kinase activity following TPA treatment reflected a redletributmn Of the enzyme, we also assayed C klnaae activity in the DEAE column eluate o f solubilized membrane fractions. In untreated c e l l s , only a small amount of Ca2+-phoepholipid dependent kinase was Obaerved (Figure 2111. Treatment of c e l l s wlth TPA c a n e d an ~ n c r e c i s ein the amount of activity aSSociated vlth the membrane fraction (P1gur.e 281. Thus the lose of cytosolic treatment Of celle wlth TPA correCa2+-phoaphollpid dependent C kinase &n lates with the appearance of this eneyme in association lith the membrane fraction Of cel?s.

-

Gel S l l C e S Ye1 Phoaphocimino Acid and Tryptic PhosphopeDtide AnalySls swelled and washed with several changes of 108 methanol. Slices were r e minced and incubated with 50 ug/ml diphenyl Carbamyl chloride-treated then (S~qmalin 50 Iw N H ~ H C O J ,pH 8.0 at roam temperature for 24 hours. POIItrypsin t x phosphopeptide maps. samples were treated cis described by Reynolds tryp3 et a 1 . 1221. Peptides were separated on cellulose thin layer plates by electrophoresis at pH 1.9. followed by chromatography in but~na1rpy~idine:acetic acid:rater l32:25:5:20). For separation Of phosphomlno acids. eluates from trypninzed gel slices were subjected to acid hydrolysis and Beparated by two-dimensional thin layer electrophoreslrI, first at pH 3.5 and then at pH 1.9 119).

Binding Of 125I-g;P to c e l l s - A431 cultures (105 cella per 35 m dish1 in W E contalninq2g.58 BSApretreated with TPA O A K , 01 Ie2SO at 37-C for 10 minutea. I-EGP was added to the Culturea At the Indicated concentrations and Incubation was continued at 37'c for 4 0 minutes. B i n d ~ n g media, first wash, second wash and cell associated radioactivity were Counted. Binding data was analyaed by the method Of Scatchard 123).

Change in subcellular distribution of protein kinase C znduceb by en*% activators. When A431 c e l l s were lysed n the presence Of EDTA and EGTA fractionated by centifvgation and the s k b l e p r o t e l m separated by DEA; cellulose chromatography, a p k k of calcium and phosphollpid sensitive protein kinaae activlty was observed [Figure 1 A I . Demonstration of kinase activ ~ r yrequired the addition of calcium phosphatldylserine and diaicylglycerol ldioleinl; neither Calcium alone nor phospholipid added with EDTA stimulated the enzyme (data not sharni. Thus, the kinaseactivity thisCpeak corresponds to the Ca2+, phospholipid-dependent protein in kinase activity originally described by Nishizuka and coworkers 1111. Becauee Ca2+ lipid, and diacylqlyceral dependencewa* Used an a criterion for c ldentificstion, It was necessary to measure activity under fully stimulated conditions. Therefore. the kinase activity detected by thls a9"y is a reflection of the relative amount of c kinaee present in a particular fract i o n , and not necessarily the actval & Yiva activity.

&h%z?ho-

-

0

IO 20 30 FRACTION NUMBER

Diacylglycerol Modulates EGFAction Receptor

12506

To determine whether an endogenous-type activator C a w e d a similar change in the subcellular distribution of Ckinase, we treated c e l l s with the synthetic diaCylglyCeIO1. 1-oleyl 2-acetyl diglycerol I O A D G l . OADG has been shown to stimulate protein kinase C in intact platelets 1241. when added to A431 cello in culture, OADG at a concentration of 50 ug/ml c a w e d a reduction in Cytosolic protein kinaee C activity by approximately 25% (data not shornl. Increaaing the OALG Concentration to BO ug/ml yielded up t o a 5 0 6 reductLon in CytDBOliC enzyme LlCtiYIty IFiqure 1Cl. Thus, the dlacylglycerol analog OADG caused a similar loss of Cytosolic C kinaseactivity as that seen for TPA. when cells were subjected to the longer fractionation procedure required for membrane isolation. no changes I n cytosolic aCtlVity or membrane aBeoCiation were detected [data Dot shown]. presumably due to metabolism of OADG during M r k u p o r disruption of the enlyme-dic.cylglyce.01 complex. This result indicates that the decrease in cytosolic activity in response to OADG is not due to inactivation of C kznase and/or loss Of CLZt-phospholipid dependence, but is instead due to L reversible membrane association. Since we were interested in determming whether the membrane association of c kinase mrrelated with tumor promoter a c t m n on the EGP receptor, it was important to determine if EGF could influence the distribution Of c kinase under the conditions O f O U T assay. Cultures yere treated with 150 ng/ml EGF at 37'c for 40 minutes and the cytosolic c kinase actlvity determined after fractionation on DEAE cellulase. Within the limits of resolution of the assay, w e dld not detect any significant difference (greater than lo$) in total LLCtiVltY relative to untreated cells (Fig. 3A.81. Addition of EGF to cells pretreated Wlth TPA for 10 minutes had no effect on the loss Of cytosolic C kinase activity relative to that observed in cells treated with TPA alone IFigure 3CI. Thus, EGP does not appeai to dramatically shift the distribution of c kinase from the cytosol to the membrane nor influence the redistribution induced by TPA.

0.4

-

'

Y8 0.3m

P

% O2 i ;0.1

L L " 0.4

.'.

0

-C 9

Y)

203040

a : f-

gu ?c 0.33 g 0.2

0.3-

-

0.2i

01 0

wzom40 EGF lnMl

0

0.1 0.2 0.3 0.4 0.5 EGF BOUND (mwles/105cslls)

Figure 4. Scatchard anlrlysill showing the effectOf TPA or OADG on specific binding of 1251-EGF to A431 cells. Cella were pretreated with Melso I A ) , 100 ng/ml TPA I B ) or 80 ug/ml OADG IC) for 10 minutes at 37'C. followed by addition Of 1251-ffiF for 40 minutes at 37'c. Inset ahows specific binding ltotal bindlnq minu. nonepecific binding measured in the pxesence of 1 ug/ml of unlabelled EGPI at increasing BGP concentrations.

navino observed reduction in EGF bxndina following OIDG treatment. we

0

10 20 30 4 FRACTION NUMBER

Figure 3. Effect Of EGF or EGF plus TPA on Cytosolic Ca2+-phospholipid stimulated protein kinase C activity in ~ 4 3 1cells. Cytosolic proteins 17 mgl from c e l l s treated with Me2SO (AI,150 ng/ml EGP 181. o r 100 n d m l TPA plus 150 ng/rnl EGF (Cl were fractionated on W A E LB described. Fractions were assayed in the presence ( @ I Or absence I D 1 of phoaphatldylserlne and diolein; Ca2+ wae included in a11 assays. Kinase activity is expressed as cpm incorporated into hiatone/ 5 nin/ 50 "1 aliquot.

nodulation of EGP ReCeDtOr Sindingand Phoaphorvlat&o_n_-b OADG. In previous studies. t u w r promoters have been shown to modulate b o t b i n q and phosphorylation of the EGF receptor in A431 cells 1 3 - 9 ) . We have recently described the loss of EGP binding to a population Of apparent high affinity EGF receptors in A431 c e l l s in response to three classes of tumor promoters 131. To determine whether TPA could be mediating thls effect by substituting for diacylglycerol. w e measured lP5I-ffiF binding to A431 cells which had been pretreated wxth Me2SO. TPA, or OADG as in the previous experiments. Scatchard analysis zndicated that OADG, like TPA, blocked binding to the apparent high affinity receptors without nigniflcantly affecting total binding (Figure 41. Although the signiflcance of this apparent high affinity binding is unclear, these results Indicate that OADG and TPA have a similar effect on EGF binding.

Table 1. Effect of TPA, OADG. and EGF OD tyrosine phosphorylation of the EGP receptor in A431 cells.

32P-prelabelled cella were treated LIB mdicated and tyrosine ihoepharylated protelna were immunoprecipitated using a monoclonal antipbosphotyrosine antibody. The proteina were separated by SDS-PAGE and the 170 koa EGF receptor band 1.3s visualized by autoradiography, excised and counted. The mean and range for two or t h k determlnntions &re shown.

TREATMENT Ile2so TPA 1100 ng/mll OR% ( 8 0 ug/mll EGP 1150 ng/mll TPA + BGGP . OADG + EGF

RECEPTOR PHOSPHORYLATION (cpm/105 27 k 14 176 C 84 283 k 45 3620 k 860 4 5 3 i 160 979 k 210

Diacylglycerol Modulates EGFAction Receptor Table 2

12507

In order to corn are the ~lpscificaites of phosphorylation, we analyxad tryptic digests of 39P-labellad EGP receptors Ieolated from e l l s treated with TPA or OADG. In untreated cella, a number of phoephopeptides were evident lPigure 6A). when cells were expoesd to TPA or OADG. no new peptides were observed: however the overall l e v e l of phosphorylation vas enhanced IPigure 6 , 8 and C). EGP treatment caused the appearance Of a number of new peptides (Figure 6D). presumably due to the aictlon Of a distinct EGF-stlmulaLed kinaee. The 81mllarity I n the phoaphopeptide pap0 of ECP receptors f m m O A Kand TPA treated cells ia m n s l s t e n t with t h e possibilitv that these two agents act through a c o m n pathway to cause t h e observed receptor phosphorylation.

Effect of TPA. OADG. and B3F on total phomphorylstion of the EGP re~eptor i n A431 cells

RECEPTOR PHOSPHORYLATION TREATUENT 111 ne250 TPA 1100 ng/mll OADG I 8 0 uglmll ECF 1150 nqlmll TPA

+

EGP

OADG + EGP

(111 W 2 Y ) T P A 10.01 ng/mll T P A I1 ng/mlI

1 ~ ~ 1 0 5 1960 4550 1060 4750 5140 3800

t 2 2 2 2 2

d i s ~

110 220 350 1170 80

A

B

530

2270 2 620 2930 2 400 4430 2 410

0

e9 ' OI

TO c a p a r e the residua that were phosphorylated. CeCePtOrS I ~ u n o p r e c i pleated with anti-ffiP receptor antiserum were subjected to phosphoamlno acid analysis. In untreated cells, EGP receptors were phosphorylated primarily at serine and threonine residues (Figure 5111. O A K stimulated the incorporation of phosphate exclusively into serine and threonine residues lPigure 581. Addltion of &P resulted In tyrosine phosphorylation of the receptors IPigure 5 C 1 , as previously reported 1 1 0 1 . O A K blocked the EGP arirnulated tyrosine phosphorylation of &P receptors IPiqure 5Dl. in agreement with data obtained using the antiphoephotyrosine antibody (see Table 11. These resY1ts indicate that OADG. like TPA 15.6, our unpubliahed l e s u l t s l . stimulates phosphorylation of the EGP receptor at serine and threonine residues, concomitant with inhibition of EGP-stimulated tyrosine phosphorylation.

. Tyr(P) A

..

..

C Figure 5. Phosphoamino acid analysis of EGP receptor. Imunoprecipitated from A431 cells using a plyclonal a n t i - E G P receptor antiserum. 32P-prelablelled cells were treated with Ue2SO (AI, 80 u g h 1 OADG 181, 150 ng/ml &F IC) or OADG plus EGP (Dl, the EGP receptor i'olated. hydrolysed and the phoaphoamino aclds separatedby two-dimsnaional electrophoresis (aee Uethodsl. Authentic phoaphoamino acid atandarda yere visualized rlth ninhydrin: radioacrivlry was detected by autoradiography.

0

Figure 6. 'No dimensional tryptic phosphopcptide analysisof EGP receptoes immunoprecipitated from A431 cells. 32P-prelsbsled cells were treated with Ue2SO (AI. 100 ng/ml TPA ( 8 1 80 u g h 1 O A K IC), or 150 ng/ml EGGP I D ) , and &P receptors were isolated a: an Uethods. Samples were trypsinired and the Peptides separated by electrophoresis at pH 1.9 in the horizontal direct i o n l a n o d e right) followed by ascendlnq chromatography. 0, orlgin.