- Email: [email protected]
c3T3 cells
Experimental Cell Research 183 (1989) 3644
Altered
Kinase C Function
in Transformed
BALB/c3T3
Cells’
HENRY C. YANG Division of Medical Oncology and Department of Medicine, Harbor-UCLA Medical Center, 1000 West Carson Street, Torrance, California 90509
Comparative studies of kinase C function were performed in an untransformed (A31) and the benzo[a]pyrene (BPA31), dimethylbenz[a]anthracene (DA31), and Kirsten sarcoma virus (KA31) transformed BALB/c 3T3 mouse tibroblast cell lines. The IO-kDa kinase C dependent phosphoprotein (pp80), an in viw marker of kinase C activity, was markedly decreased in the transformed cells although the amount of the SO-kDa substrate protein in the BPA31 cells was similar to that in the untransformed A31 cells. Total cell lysate kinase C levels were lower in the transformed cells but this difference could not account for the reduced pp80 phosphorylation. Increased affinity of kinase C for the membrane fraction in the BPA31 cells may account for decreased phosphorylation of pp80. @ 1989 Academic press, Inc.
Protein kinase C appears to serve as a signal transducer for a variety of biologically active substances (for review, see [l]). Kinase C also appears to be the receptor for the tumor promoter phorbol esters [2, 31. In Swiss 3T3 cells, a several-fold increase in the phosphorylation of an 80-kDa protein (~~80) is a marker of kinase C activity [4]. A similar protein of 87 kDa, which was first described in rat cerebral cortex synaptosomes after depolarization-induced calcium influx [5], is ubiquitously present in a number of species and a variety of tissues [6]. Previously, we have observed that a benzo[a]pyrene transformed (BPA31) BALB/c 3T3 cell line has a markedly decreased amount of pp80 at the end of low serum arrest and after serum or phorbol ester stimulation [7]. In this study, additional transformed BALB/c 3T3 cell lines were examined to determine whether decreased pp80 is a common characteristic of transformation. The mechanism for decreased pp8d in BPA31 cells was studied in detail by comparative assays of kinase C activity and localization and by immunoassays for the 80-kDa protein. EXPERIMENTAL
PROCEDURES
Phorbol 12-myristate 13-acetate (PMA) was from LC Services Corp. (Wobum, MA). [‘HlThymidine (84 Wmmol) and [‘*P]orthophosphoric acid (‘*P; 10 Wmmol) were from New England Nuclear Corp. (Boston, MA). Ampholines were from LKB Instruments, Inc. (Gaithersburg, MD). Protein ASepharose was from Pharmacia, Inc. (Piscataway, NJ). Cell culture. Untransformed (A31), benzo[a]pyrene transformed (BPA31), dimethylbenz[a]anthracene transformed (DA31), and Kirsten sarcoma virus transformed (KA31) BALB/c 3T3 mouse fibroblasts were obtained [8] and grown at 10% CO1 in Dulbecco’s modified Eagle’s medium (DME) Reagents.
’ This work was supported in part by Grant BRSG SO7RRO551from the National Institutes of Health, the Harbor Collegium, and the California Institute for Cancer Research. Copyright @ 1989 by Academic Press, Inc All rights of reproductmn I” any form reserved cull4-4827189 $03 00
36
Kinase C function
in transformed
cells
37
containing 10% calf serum and supplemented with 4 mM glutamine (complete medium). Cultures were used for experiments up to the 16th passage. New cultures were determined to be free of mycoplasma contamination by the uridine/uracil method [9]. Cell synchronization protocol. A31 cells were grown for 2-3 days in complete medium in 35 and lOO-mm plastic dishes or 24-multiwell plates to reach 25% confluence or approximately 2.5X104 cells/cm*. The medium was removed and the dishes were washed once with DME and then incubated with DME plus 0.5 % calf serum for 48 h. Transformed cells were plated in 35- and lOO-mmdishes or multiwell plates at 1.25~ lo4 cells/cm in DME plus 0.2 % calf serum and incubated for 72 h. Nuclear labeling. The medium from the low serum arrested A31 and transformed cells was aspirated from the multiwells and replaced with complete medium or DME containing 200 n&f PMA, plus 1 uCi [‘H]thymidine/ml. The multiwell plates were processed for autoradiography as previously described [lo]. The nuclear labeling index was determined by counting at least 200-400 cells in representative fields. Prepararion of ‘*P-1abeledproteins. The medium from the low serum arrested A31 and BPA31 cells was aspirated from the 35-mm dishes and replaced with 0.8 ml DME with 5 % of the usual phosphate content, 250 uCi “P/ml, and either 10% calf serum, 200 n&f PMA, 0.5% calf serum for A31 cells, or 0.05 % calf serum for transformed cells. Whole cell extracts were prepared at 4°C in lysis buffer containing 1 mM phenylmethylsulfonyl fluoride and 1% aprotinin for two-dimensional (2D) gel electrophoresis as previously described [ 11, 121.Also, 1% aprotinin, 20 mJ4 EDTA, 50 mM NaF, 7 mM Na&Or, and 100 pJ4 Na3V04 were added immediately after the initial cell lysis. Two-dimensional gel electrophoresis. Two-dimensional gel electrophoresis was performed as previously described [7, 11, 121 using 1.6%, pH 5-7, and 0.4%, 3.5-10, ampholines for isoelectric focusing in the first dimension and a 10% polyacrylamide separating gel with a 4% polyacrylamide stacking gel in the second dimension. Samples were normalized as previously described [7]. Protein kinase C studies. Kinase C was assayed as previously described [13]. The soluble fraction was the supematant obtained by centrifugation of cell Iysates at 13,OOOgfor 15 min. The detergent extractable particulate fraction was obtained after 0.2% Triton X-100 treatment of the pellet and similar centrifugation. Protein determination was by the method of Bradford [14] using bovine yglobulin as the standard. lmmunoprecipitation and quantiration of pp80. The polyclonal antibody to 87-kDa phosphoprotein was the generous gift of K. Albert and P. Greengard. Immunoprecipitation was performed as previously described [6]. pp80 was quantitated by a competitive binding assay and autoradiography. Quantitation of pp80 was relative to the amount present in A31 cells and was performed 3 h after stimulation of A31 and transformed cells with 10% serum. 32P-Labeled A31 cells from a 35-mm dish were lysed with 120 u1 sample buffer (0.5% SDS, 5 % Nonidet-P40, 180 m&f NaCl, 10 mM NaHZP04, 50 mM NaF, 2 mM EDTA, 2 mil4 EGTA, pH 7.4). Unlabeled A31, BPA31, DA31, and KA31 cells from 35mm dishes were scraped into 60 ul of phosphate-buffered saline (Ca”, Mg+ free), sonicated for protein determinations, and then diluted 1: 1 with 2x sample buffer (minus NaCl, NaI-IrPO,). Antibody (1.25 pl) and “P-labeled A31 cell lysate (10 ul) were incubated with different amounts of competing unlabeled cell lysates for 2 h at 4°C after bringing the final volume to 450 pl with solution II (150 m&f NaCl, 15 mM Hepes, 1 m&f Nar EDTA, 0.5% Nonidet-P40, pH 7.4). Fifty microliters of a protein A-Sepharose suspension was added to the mixture, which was then agitated at 4°C for 2 h. After six washes with solution II, the mixture was boiled for 2 min and the supematant in Laemmli sample buffer (0.05 Tris, 1% SDS, 0.14 M 2-mercaptoethanol, 20% glycerol, 0.002% bromphenol blue, pH 6.8) was subjected to one-dimensional gel electrophoresis. Radiolabeled phosphoproteins were quantitated as previously described [7] using the Bio-Rad Model 620 video densitometer. The percentage of (32Pbound/maximum 32Pbound) was plotted against serial amounts of A31 cell lysate to obtain a standard curve.
RESULTS Decreased Kinase C Dependent Phosphorylation 80-kDa Protein in Transformed Cells
of the
Kinase C dependent phosphorylation of the 80-kDa protein was examined in additional transformed cell lines to determine if decreased pp80 is a general characteristic of transformed cells. Figure 1 shows that like the BPA31 cells [7],
38
Henry C. Yang
Fig. 1. Decreased pp80 in transformed cells. DA31 cells at the end of low serum arrest (A) and 3 h after 10% serum!(B) and PMA (C) stimulation. KA31 cells (0, E, and F) and A31 cells (G, H, and I) at the same respective times. Arrows show the position of pp80.
the DA3 1 and KA3 1 cells had markedly decreased levels of pp80 at the end of low serum arrest and 3 h after restimulation with either 10% calf serum or PMA. In DA31 cells, the pp80 level was 17, 11, and 20% of that found in untransformed A31 cells at the end of low serum arrest and 3 h after 10% serum and PMA stimulation, respectively. In KA31 cells, the pp80 level was correspondingly 30, 9, and 6 % of that in A3 1 cells. The similar intensity of a 95kDa (pl=5.1) phosphoprotein seen on the gels in Fig. 1 in all three cell lines makes it unlikely that the difference in pp80 is due to different rates of inorganic phosphate uptake in these cells. More importantly, since 32Pincorporation may reflect turnover of phosphate in pp80 rather than the phosphorylation state of the 80-kDa protein, the A31, DA31, and KA31 cells were preequilibrated with 250 $i ‘*P/ml for 15 h prior to treatment with low serum, 10% calf serum, or PMA for 1 h. Reduced levels of pp80 were again seen in the DA31 and KA31 cells (data not shown), suggesting that reduced pp80 in these cells is not due to decreased 32Pturnover in pp80. Immunologic Identity of pp80 and the 87-kDa Phosphoprotein in Brain Synaptosomes
Although pp80 and the 87-kDa brain phosphoprotein have similar molecular weights and p1 [6, 71 it has not been established that these two proteins are identical. Figure 2A demonstrates that the antibody to the 87-kDa phosphopro-
Kinase C function in transformed cells
39
0
Fig. 2. Immunologic identity of the 87-kDa brain phosphoprotein and pp80. (A) The antibody to the 87-kDa protein immunoprecipitates a phosphoprotein of about 80-kDa in low serum-arrested A31 cells which have been stimulated with 10% serum for 3 h (lane I); no corresponding phosphoprotein is immunoprecipitated in low serum-arrested BPA31 cells (lane 2). (B) The arrow in the top panel shows the position of pp80 on a two-dimensional gel in A31 cells 3 h after serum stimulation. The bottom panel demonstrates the disappearance of pp80 after immunoprecipitation with the 87-kDa antibody.
tein immunoprecipitated the 80-kDa phosphoprotein in A31 cells. Figure 2 B shows that pp80 had disappeared when the residual supernatant was subjected to 2D gel electrophoresis, confirming that the 87 kDa antibody had immunoprecipitated pp80. Thus pp80 and the 87-kDa phosphoprotein share immunologic crossreactivity. Quantitation of 80-kDa Protein in Transformed Cell Lines To examine the possibility that decreased pp80 in the transformed cells was due to decreased 80-kDa substrate protein for kinase C, immunoassay of the 80kDa protein was performed in the untransformed A31 and the transformed BPA31 cells, using the antibody to the 87-kDa phosphoprotein. Since the 87-kDa phosphoprotein antibody is not satisfactory for immunoblotting of mouse material (K. Albert, personal communication), a competitive binding assay was developed. Figure 3A demonstrates that increasing amounts of unlabeled A31 and BPA31 cell lysates were able to displace the 32P-labeled A31 pp80. Figure 3 B shows the standard curve for A31 lysate. BPA31 cells had a level of 80-kDa protein (~80) equivalent to 62+23 % (SD, n=2) of that found in A31 cells. Precise determination of the p80 level in DA31 and KA3 1 cells could not be obtained because of displacement curves which were not parallel to that for A31 cells.
Henry C. Yang
A
6
12345678910
.
.
\ I 1 5
I I 10 20 A31 PROTEIN
I 40 OJql
I 60
Fig. 3. Quantitation of the 80-kDa protein in A31 and BPA31 cells 3 h after serum stimulation. (A) Immunoprecipitation of “P-labeled A31 pp80 without competing unlabeled material (lane I). In the presence of increasing (8, 16, 32, 64 ug) amounts of unlabeled A31 cellular protein (lanes 2-5) and increasing (7, 14, 28, 56, 112 ug) amounts of unlabeled BPA31 cellular protein (lanes 6-N). (B) Standard curve using cpm of lanes 2-5 divided by the maximum bound cpm of lane 1 as determined by densitometry vs amount of competing unlabeled A3 1 cellular protein in the corresponding lane.
Protein Kinase C Activity To test the hypothesis that differences in total kinase C activity levels could account for the difference in pp80 levels between the untransformed A31 and the transformed BPA3 1, DA3 1, and KA3 1 cells, kinase C activity in cell lysates was measured. Figure 4 shows that the A31 kinase C level was generally higher than those of the transformed cells at the end of low serum arrest and after 10% serum stimulation. The A31 cells had a kinase C level approximately twice that of the BPA31 cells at the three time points studied. However, when kinase C activity was measured in A3 1 and BPA3 1 cells 1 h after PMA stimulation, the levels were similar, 2.69f0.95 and 2.99kO.31 (SD, n=2) pmol/min/mg protein for A31 and BPA31 cells, respectively, although previous studies revealed a 24-fold difference in pp80 level between the two cell lines after PMA stimulation [7]. These data suggest that the difference between A3 1 and BPA3 1 cells in pp80 phosphorylation cannot simply be accounted for by the difference in total kinase C levels as measured in cell lysates. Previous studies have shown that both transformation and phorbol ester expo-
Kinase C function
k--r-T-
in transformed
cells
41
3 HOUR
Fig. 4. Total cellular kinase C activity (SD, n=2) at the end of low serum arrest and 1 and 3 h after 10% serum stimulation. A31 (0), BPA31 (0). DA31 (m) and KA31 (A) cells.
sure cause an increase in kinase C in the membrane-associated fraction and a decrease in the cytosolic fraction [U-17]. To test the hypothesis that the difference in pp80 phosphorylation is related to a difference in the localization of kinase C which might affect its function in uiuo, the compartmentalization of kinase C was compared in A31 and BPA31 cells. Figure 5 shows the membrane localization of kinase C in A31 and BPA31 cells at the end of low serum arrest and 1 h after serum and PMA stimulation. There appeared to be no difference between A31 and BPA31 cells at the end of low serum arrest and after serum stimulation. However, after PMA exposure, the membrane-associated fraction increased only modestly to 31+3% (SD, n=2) in A31 but markedly to 54-+3% (SD, n=2) in BPA31 cells. These data suggest that kinase C or a cofactor necessary for membrane localization in the presence of PMA is altered in the transformed BPA3 1 cells. Relationship
between pp80 Levels and PMA-Induced
Mitogenesis
To examine the functional significance of pp80 phosphorylation in the mitogenic pathway, the nuclear labeling index was measured after PMA treatment of low
A
6
C
D
E
F
Fig. 5. Membrane localization (SD, n=2) kinase C activity at the end of low serum arrest (A and 0) and 1 h after 10% serum (B and E) and PMA (C and F) stimulation in A31 (A-C) and BPA31 (D-F) cells.
42
Henry C. Yang TABLE 1 Effect of PMA on mitogenesis Cells
A31 BPA31 DA31 KA31
DME
1 4 22 31
10% serum
PMA
% labeled nuclei 86 2 94 22 93 62 88 47
DMSO
2 5 23 39
serum arrested cells for 24 h. Table 1 shows the labeling index at the end of low serum arrest and after 10% serum or PMA stimulation. There appeared to be no simple correlation between pp80 levels and PMA-induced mitogenesis. DISCUSSION This study suggests that reduced phosphorylation of the 80-kDa protein after low serum arrest and serum or PMA restimulation is a frequent feature of transformed BALB/c 3T3 mouse fibroblasts and may serve as a marker of transformation. Previously, we had shown that pp80 was minimally present in exponentially growing untransformed A31 and transformed BPA31 cells [7] and that in A31 cells but not BPA31 cells it appeared during low serum arrest and increased further after serum or PMA restimulation. Elevated diacylglycerol levels and decreased phosphorylation of the 80-kDa protein after PMA exposure have been observed in ras-transformed NIH 3T3 fibroblasts, although those cells also had increased basal phosphorylation of the 80-kDa protein as compared to the untransformed cells [18]. The current study suggests that the mechanism of reduced pp80 in transformation is not due to decreased ~80 substrate or total in vitro kinase C activity. Alternative possibilities are that kinase C has an altered activity for ~80 in uiuo or that there is increased dephosphorylation of pp80. The former possibility is suggested by an increased affinity of kinase C for the membrane fraction when studied in the presence of phorbol ester in the BPA31 cells. It has been shown that activation of kinase may require a Ca*+-dependent proteinase which releases kinase C from the membrane to the cytosol [19]. Possibly decreased available cytosolic kinase C results in decreased pp80 phosphorylation. Although the location of pp80 is not known, the related synaptosomal87-kDa phosphoprotein is a cytosolic protein [S}. The role of pp80 phosphorylation in the mitogenic pathway remains to be defined. It is a very early substrate for kinase C appearing within minutes after phorbol ester or serum exposure [4]. The transformed BPA31 and DA31 cells were previously found to have constitutively high levels of c-myc, thus suggesting that pp80 phosphorylation and c-myc transcription may be reciprocally related
Kinase C function in transformed cells
43
[20]. The relationship between pp80 phosphorylation and some of the known early actions of kinase C such as tyrosine phosphorylation [21, 221, protooncogene induction [23, 241, phosphorylation of growth factor receptors [22, 251 and physiologic substrates [26-281, and S6 kinase activation [29] is not known. By itself, PMA is not mitogenic in A31 cells although it markedly increases the pp80 level. In contrast, PMA is highly mitogenic for DA31 cells although it increases the pp80 level only modestly. Kinase C undoubtedly activates other independent pathways which lead to mitogenesis, as demonstrated by the DA31 cells. It is not possible to state that pp80 phosphorylation is unnecessary for mitogenesis since transformation may have bypassed an otherwise necessary pp80 phosphorylation step. The author gratefully acknowledges the assistance of Diana Chaco in the preparation of the manuscript.
REFERENCES 1. Nishizuka, Y. (1980) Nature (London) 388, 693-698. 2. Castagna, M. J., Takai, Y., Kaibuchi, K., Sano, K., Kikkawa, U., and Nishizuka, Y. (1982) J. Biol. Chem. 257, 7847-7851. 3. Yamanishi, J., Takai, Y., Kaibuchi, K., Sano, K., Castagna, M., and Nishizuka, Y. (1983) Biochem. Biophys. Res. Commun. 112, 778-786. 4. Rozengurt, E., Rodriguez-Pena, M., and Smith, K. A. (1983) Proc. N&l. Acad. Sci. USA 80, 7244-7248. 5. Wu, W. C.-S., Walaas, S. I., Naim, A. C., and Greengard, P. (1982) Proc. N&l. Acad. Sci. USA 79, 5248-5253. 6. Albert, K. A., Walaas, S. I., Wang, J. K.-T., and Greengard, P. (1986) Proc. Nutl. Acad. Sci. USA 83, 2822-2826. 7. Yang, H. C., and Pardee, A. B. (1987) J. Cell. Physiol. 133, 377-382. 8. Dubrow, R., Riddle, V. G. H., and Pardee, A. B. (1979) Cancer Res. 39, 2718-2726. 9. Schneider, E. L., Stanbridge, E. J., and Epstein, C. J. (1974) Exp. Cell Res. 84, 311-318. 10. Campisi, J., Medrano, E. E., Morreo, G., and Pardee, A. B. (1982) Proc. Nutl. Acad. Sci. USA 79, 43w. 11. O’Farrell, P. H. (1975) J. Biol. Chem. 250, 4007421. 12. Croy, R. G., and Pardee, A. B. (1983) Proc. Natl. Acud. Sci. USA 80, 4699-4703. 13. Kikkawa, U., Takai, Y., Minakuchi, R., Inohara, S., and Nishizuka, Y. (1982) J. Biol. Chem. 257, 13,341-13,348. 14. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254. 15. Kraft, A. S., and Anderson, W. B. (1983) Nuture (London) 301, 621-623. 16. Anderson, W. B., Estival, A., Tapiovaara, H., and Gopalakrishna, R. (1985) in Advances in Cyclic Nucleotide and Protein Phosphorylation Research (Cooper D. M. F., and Seaman, K. B., Eds.), Vol. 19, pp. 287-306, Raven Press, New York. 17. Halsey, D. L., Girard, P. R., Kuo, J. F., and Blackshear, P. J. (1987) J. Biol. Chem. 262, 2234-2243. 18. Wolfman, A., and Macara, I. G. (1987) Nuture (London) 325, 359-361. 19. Melloni, E., Pontremoli, S., Michette, M., Sacco, O., Sparatore, B., Salamino, F., and Horecker, B. L. (1985) Proc. Natl. Acad. Sci. USA 82, 6435-6439. 20. Campisi, J., Gray, H. E., Pardee, A. B., Dean, M., and Sonnenshein, G. E. (1984) Cell 36, 241-247. 21. Gilmore, T., and Martin, G. S. (1983) Nature (London) 306, 487-490. 22. Moon, S. O., Palfrey, H. C., and King, A. C. (1984) Proc. Nutl. Acad. Sci. USA 81,22%2302. 23. Coughlin, S. R., Lee, W. M. F., Williams, P. W., Giels, G. M., and Williams, L. T. (1985)Cell 43, 243-251. 24. Blackshear, P. J., Stumpo, D. J., Huang, J. K., Nemenoff, R. A., and Spach, D. H. (1987)J. Biol. Chem. 262. 7774-7781.
44
Henry C. Yang
25. Jacobs, S., Sahyoun, N. E., Saltiel, A. R., and Cuatrecasas, P. (1983) Proc. Natl. Acad. Sci. USA 80, 6211-6213. 26. Werth, D. K., Niedel, J. E., and Pastan, I. (1983) J. Eiol. Chem. 258, 11,423-11,426. 27. Feurstein, N., and Cooper, H. L. (1984) .I. Biol. Gem. 259, 2782-2788. 28. Gould, K. L., Woodgett, J. R., Isacke, C. M., and Hunter, T. (1986) Mol. Cell. Biol. 6, 2738-2744. 29. Tabarini, D., Heir&h, J., and Rosen, 0. M. (1985) Proc. Natl. Acad. Sci. USA 82,4369-4373. Received June 10, 1988 Revised version received December 23, 1988
Printed
in Sweden
Altered
Kinase C Function
in Transformed
BALB/c3T3
Cells’
HENRY C. YANG Division of Medical Oncology and Department of Medicine, Harbor-UCLA Medical Center, 1000 West Carson Street, Torrance, California 90509
Comparative studies of kinase C function were performed in an untransformed (A31) and the benzo[a]pyrene (BPA31), dimethylbenz[a]anthracene (DA31), and Kirsten sarcoma virus (KA31) transformed BALB/c 3T3 mouse tibroblast cell lines. The IO-kDa kinase C dependent phosphoprotein (pp80), an in viw marker of kinase C activity, was markedly decreased in the transformed cells although the amount of the SO-kDa substrate protein in the BPA31 cells was similar to that in the untransformed A31 cells. Total cell lysate kinase C levels were lower in the transformed cells but this difference could not account for the reduced pp80 phosphorylation. Increased affinity of kinase C for the membrane fraction in the BPA31 cells may account for decreased phosphorylation of pp80. @ 1989 Academic press, Inc.
Protein kinase C appears to serve as a signal transducer for a variety of biologically active substances (for review, see [l]). Kinase C also appears to be the receptor for the tumor promoter phorbol esters [2, 31. In Swiss 3T3 cells, a several-fold increase in the phosphorylation of an 80-kDa protein (~~80) is a marker of kinase C activity [4]. A similar protein of 87 kDa, which was first described in rat cerebral cortex synaptosomes after depolarization-induced calcium influx [5], is ubiquitously present in a number of species and a variety of tissues [6]. Previously, we have observed that a benzo[a]pyrene transformed (BPA31) BALB/c 3T3 cell line has a markedly decreased amount of pp80 at the end of low serum arrest and after serum or phorbol ester stimulation [7]. In this study, additional transformed BALB/c 3T3 cell lines were examined to determine whether decreased pp80 is a common characteristic of transformation. The mechanism for decreased pp8d in BPA31 cells was studied in detail by comparative assays of kinase C activity and localization and by immunoassays for the 80-kDa protein. EXPERIMENTAL
PROCEDURES
Phorbol 12-myristate 13-acetate (PMA) was from LC Services Corp. (Wobum, MA). [‘HlThymidine (84 Wmmol) and [‘*P]orthophosphoric acid (‘*P; 10 Wmmol) were from New England Nuclear Corp. (Boston, MA). Ampholines were from LKB Instruments, Inc. (Gaithersburg, MD). Protein ASepharose was from Pharmacia, Inc. (Piscataway, NJ). Cell culture. Untransformed (A31), benzo[a]pyrene transformed (BPA31), dimethylbenz[a]anthracene transformed (DA31), and Kirsten sarcoma virus transformed (KA31) BALB/c 3T3 mouse fibroblasts were obtained [8] and grown at 10% CO1 in Dulbecco’s modified Eagle’s medium (DME) Reagents.
’ This work was supported in part by Grant BRSG SO7RRO551from the National Institutes of Health, the Harbor Collegium, and the California Institute for Cancer Research. Copyright @ 1989 by Academic Press, Inc All rights of reproductmn I” any form reserved cull4-4827189 $03 00
36
Kinase C function
in transformed
cells
37
containing 10% calf serum and supplemented with 4 mM glutamine (complete medium). Cultures were used for experiments up to the 16th passage. New cultures were determined to be free of mycoplasma contamination by the uridine/uracil method [9]. Cell synchronization protocol. A31 cells were grown for 2-3 days in complete medium in 35 and lOO-mm plastic dishes or 24-multiwell plates to reach 25% confluence or approximately 2.5X104 cells/cm*. The medium was removed and the dishes were washed once with DME and then incubated with DME plus 0.5 % calf serum for 48 h. Transformed cells were plated in 35- and lOO-mmdishes or multiwell plates at 1.25~ lo4 cells/cm in DME plus 0.2 % calf serum and incubated for 72 h. Nuclear labeling. The medium from the low serum arrested A31 and transformed cells was aspirated from the multiwells and replaced with complete medium or DME containing 200 n&f PMA, plus 1 uCi [‘H]thymidine/ml. The multiwell plates were processed for autoradiography as previously described [lo]. The nuclear labeling index was determined by counting at least 200-400 cells in representative fields. Prepararion of ‘*P-1abeledproteins. The medium from the low serum arrested A31 and BPA31 cells was aspirated from the 35-mm dishes and replaced with 0.8 ml DME with 5 % of the usual phosphate content, 250 uCi “P/ml, and either 10% calf serum, 200 n&f PMA, 0.5% calf serum for A31 cells, or 0.05 % calf serum for transformed cells. Whole cell extracts were prepared at 4°C in lysis buffer containing 1 mM phenylmethylsulfonyl fluoride and 1% aprotinin for two-dimensional (2D) gel electrophoresis as previously described [ 11, 121.Also, 1% aprotinin, 20 mJ4 EDTA, 50 mM NaF, 7 mM Na&Or, and 100 pJ4 Na3V04 were added immediately after the initial cell lysis. Two-dimensional gel electrophoresis. Two-dimensional gel electrophoresis was performed as previously described [7, 11, 121 using 1.6%, pH 5-7, and 0.4%, 3.5-10, ampholines for isoelectric focusing in the first dimension and a 10% polyacrylamide separating gel with a 4% polyacrylamide stacking gel in the second dimension. Samples were normalized as previously described [7]. Protein kinase C studies. Kinase C was assayed as previously described [13]. The soluble fraction was the supematant obtained by centrifugation of cell Iysates at 13,OOOgfor 15 min. The detergent extractable particulate fraction was obtained after 0.2% Triton X-100 treatment of the pellet and similar centrifugation. Protein determination was by the method of Bradford [14] using bovine yglobulin as the standard. lmmunoprecipitation and quantiration of pp80. The polyclonal antibody to 87-kDa phosphoprotein was the generous gift of K. Albert and P. Greengard. Immunoprecipitation was performed as previously described [6]. pp80 was quantitated by a competitive binding assay and autoradiography. Quantitation of pp80 was relative to the amount present in A31 cells and was performed 3 h after stimulation of A31 and transformed cells with 10% serum. 32P-Labeled A31 cells from a 35-mm dish were lysed with 120 u1 sample buffer (0.5% SDS, 5 % Nonidet-P40, 180 m&f NaCl, 10 mM NaHZP04, 50 mM NaF, 2 mM EDTA, 2 mil4 EGTA, pH 7.4). Unlabeled A31, BPA31, DA31, and KA31 cells from 35mm dishes were scraped into 60 ul of phosphate-buffered saline (Ca”, Mg+ free), sonicated for protein determinations, and then diluted 1: 1 with 2x sample buffer (minus NaCl, NaI-IrPO,). Antibody (1.25 pl) and “P-labeled A31 cell lysate (10 ul) were incubated with different amounts of competing unlabeled cell lysates for 2 h at 4°C after bringing the final volume to 450 pl with solution II (150 m&f NaCl, 15 mM Hepes, 1 m&f Nar EDTA, 0.5% Nonidet-P40, pH 7.4). Fifty microliters of a protein A-Sepharose suspension was added to the mixture, which was then agitated at 4°C for 2 h. After six washes with solution II, the mixture was boiled for 2 min and the supematant in Laemmli sample buffer (0.05 Tris, 1% SDS, 0.14 M 2-mercaptoethanol, 20% glycerol, 0.002% bromphenol blue, pH 6.8) was subjected to one-dimensional gel electrophoresis. Radiolabeled phosphoproteins were quantitated as previously described [7] using the Bio-Rad Model 620 video densitometer. The percentage of (32Pbound/maximum 32Pbound) was plotted against serial amounts of A31 cell lysate to obtain a standard curve.
RESULTS Decreased Kinase C Dependent Phosphorylation 80-kDa Protein in Transformed Cells
of the
Kinase C dependent phosphorylation of the 80-kDa protein was examined in additional transformed cell lines to determine if decreased pp80 is a general characteristic of transformed cells. Figure 1 shows that like the BPA31 cells [7],
38
Henry C. Yang
Fig. 1. Decreased pp80 in transformed cells. DA31 cells at the end of low serum arrest (A) and 3 h after 10% serum!(B) and PMA (C) stimulation. KA31 cells (0, E, and F) and A31 cells (G, H, and I) at the same respective times. Arrows show the position of pp80.
the DA3 1 and KA3 1 cells had markedly decreased levels of pp80 at the end of low serum arrest and 3 h after restimulation with either 10% calf serum or PMA. In DA31 cells, the pp80 level was 17, 11, and 20% of that found in untransformed A31 cells at the end of low serum arrest and 3 h after 10% serum and PMA stimulation, respectively. In KA31 cells, the pp80 level was correspondingly 30, 9, and 6 % of that in A3 1 cells. The similar intensity of a 95kDa (pl=5.1) phosphoprotein seen on the gels in Fig. 1 in all three cell lines makes it unlikely that the difference in pp80 is due to different rates of inorganic phosphate uptake in these cells. More importantly, since 32Pincorporation may reflect turnover of phosphate in pp80 rather than the phosphorylation state of the 80-kDa protein, the A31, DA31, and KA31 cells were preequilibrated with 250 $i ‘*P/ml for 15 h prior to treatment with low serum, 10% calf serum, or PMA for 1 h. Reduced levels of pp80 were again seen in the DA31 and KA31 cells (data not shown), suggesting that reduced pp80 in these cells is not due to decreased 32Pturnover in pp80. Immunologic Identity of pp80 and the 87-kDa Phosphoprotein in Brain Synaptosomes
Although pp80 and the 87-kDa brain phosphoprotein have similar molecular weights and p1 [6, 71 it has not been established that these two proteins are identical. Figure 2A demonstrates that the antibody to the 87-kDa phosphopro-
Kinase C function in transformed cells
39
0
Fig. 2. Immunologic identity of the 87-kDa brain phosphoprotein and pp80. (A) The antibody to the 87-kDa protein immunoprecipitates a phosphoprotein of about 80-kDa in low serum-arrested A31 cells which have been stimulated with 10% serum for 3 h (lane I); no corresponding phosphoprotein is immunoprecipitated in low serum-arrested BPA31 cells (lane 2). (B) The arrow in the top panel shows the position of pp80 on a two-dimensional gel in A31 cells 3 h after serum stimulation. The bottom panel demonstrates the disappearance of pp80 after immunoprecipitation with the 87-kDa antibody.
tein immunoprecipitated the 80-kDa phosphoprotein in A31 cells. Figure 2 B shows that pp80 had disappeared when the residual supernatant was subjected to 2D gel electrophoresis, confirming that the 87 kDa antibody had immunoprecipitated pp80. Thus pp80 and the 87-kDa phosphoprotein share immunologic crossreactivity. Quantitation of 80-kDa Protein in Transformed Cell Lines To examine the possibility that decreased pp80 in the transformed cells was due to decreased 80-kDa substrate protein for kinase C, immunoassay of the 80kDa protein was performed in the untransformed A31 and the transformed BPA31 cells, using the antibody to the 87-kDa phosphoprotein. Since the 87-kDa phosphoprotein antibody is not satisfactory for immunoblotting of mouse material (K. Albert, personal communication), a competitive binding assay was developed. Figure 3A demonstrates that increasing amounts of unlabeled A31 and BPA31 cell lysates were able to displace the 32P-labeled A31 pp80. Figure 3 B shows the standard curve for A31 lysate. BPA31 cells had a level of 80-kDa protein (~80) equivalent to 62+23 % (SD, n=2) of that found in A31 cells. Precise determination of the p80 level in DA31 and KA3 1 cells could not be obtained because of displacement curves which were not parallel to that for A31 cells.
Henry C. Yang
A
6
12345678910
.
.
\ I 1 5
I I 10 20 A31 PROTEIN
I 40 OJql
I 60
Fig. 3. Quantitation of the 80-kDa protein in A31 and BPA31 cells 3 h after serum stimulation. (A) Immunoprecipitation of “P-labeled A31 pp80 without competing unlabeled material (lane I). In the presence of increasing (8, 16, 32, 64 ug) amounts of unlabeled A31 cellular protein (lanes 2-5) and increasing (7, 14, 28, 56, 112 ug) amounts of unlabeled BPA31 cellular protein (lanes 6-N). (B) Standard curve using cpm of lanes 2-5 divided by the maximum bound cpm of lane 1 as determined by densitometry vs amount of competing unlabeled A3 1 cellular protein in the corresponding lane.
Protein Kinase C Activity To test the hypothesis that differences in total kinase C activity levels could account for the difference in pp80 levels between the untransformed A31 and the transformed BPA3 1, DA3 1, and KA3 1 cells, kinase C activity in cell lysates was measured. Figure 4 shows that the A31 kinase C level was generally higher than those of the transformed cells at the end of low serum arrest and after 10% serum stimulation. The A31 cells had a kinase C level approximately twice that of the BPA31 cells at the three time points studied. However, when kinase C activity was measured in A3 1 and BPA3 1 cells 1 h after PMA stimulation, the levels were similar, 2.69f0.95 and 2.99kO.31 (SD, n=2) pmol/min/mg protein for A31 and BPA31 cells, respectively, although previous studies revealed a 24-fold difference in pp80 level between the two cell lines after PMA stimulation [7]. These data suggest that the difference between A3 1 and BPA3 1 cells in pp80 phosphorylation cannot simply be accounted for by the difference in total kinase C levels as measured in cell lysates. Previous studies have shown that both transformation and phorbol ester expo-
Kinase C function
k--r-T-
in transformed
cells
41
3 HOUR
Fig. 4. Total cellular kinase C activity (SD, n=2) at the end of low serum arrest and 1 and 3 h after 10% serum stimulation. A31 (0), BPA31 (0). DA31 (m) and KA31 (A) cells.
sure cause an increase in kinase C in the membrane-associated fraction and a decrease in the cytosolic fraction [U-17]. To test the hypothesis that the difference in pp80 phosphorylation is related to a difference in the localization of kinase C which might affect its function in uiuo, the compartmentalization of kinase C was compared in A31 and BPA31 cells. Figure 5 shows the membrane localization of kinase C in A31 and BPA31 cells at the end of low serum arrest and 1 h after serum and PMA stimulation. There appeared to be no difference between A31 and BPA31 cells at the end of low serum arrest and after serum stimulation. However, after PMA exposure, the membrane-associated fraction increased only modestly to 31+3% (SD, n=2) in A31 but markedly to 54-+3% (SD, n=2) in BPA31 cells. These data suggest that kinase C or a cofactor necessary for membrane localization in the presence of PMA is altered in the transformed BPA3 1 cells. Relationship
between pp80 Levels and PMA-Induced
Mitogenesis
To examine the functional significance of pp80 phosphorylation in the mitogenic pathway, the nuclear labeling index was measured after PMA treatment of low
A
6
C
D
E
F
Fig. 5. Membrane localization (SD, n=2) kinase C activity at the end of low serum arrest (A and 0) and 1 h after 10% serum (B and E) and PMA (C and F) stimulation in A31 (A-C) and BPA31 (D-F) cells.
42
Henry C. Yang TABLE 1 Effect of PMA on mitogenesis Cells
A31 BPA31 DA31 KA31
DME
1 4 22 31
10% serum
PMA
% labeled nuclei 86 2 94 22 93 62 88 47
DMSO
2 5 23 39
serum arrested cells for 24 h. Table 1 shows the labeling index at the end of low serum arrest and after 10% serum or PMA stimulation. There appeared to be no simple correlation between pp80 levels and PMA-induced mitogenesis. DISCUSSION This study suggests that reduced phosphorylation of the 80-kDa protein after low serum arrest and serum or PMA restimulation is a frequent feature of transformed BALB/c 3T3 mouse fibroblasts and may serve as a marker of transformation. Previously, we had shown that pp80 was minimally present in exponentially growing untransformed A31 and transformed BPA31 cells [7] and that in A31 cells but not BPA31 cells it appeared during low serum arrest and increased further after serum or PMA restimulation. Elevated diacylglycerol levels and decreased phosphorylation of the 80-kDa protein after PMA exposure have been observed in ras-transformed NIH 3T3 fibroblasts, although those cells also had increased basal phosphorylation of the 80-kDa protein as compared to the untransformed cells [18]. The current study suggests that the mechanism of reduced pp80 in transformation is not due to decreased ~80 substrate or total in vitro kinase C activity. Alternative possibilities are that kinase C has an altered activity for ~80 in uiuo or that there is increased dephosphorylation of pp80. The former possibility is suggested by an increased affinity of kinase C for the membrane fraction when studied in the presence of phorbol ester in the BPA31 cells. It has been shown that activation of kinase may require a Ca*+-dependent proteinase which releases kinase C from the membrane to the cytosol [19]. Possibly decreased available cytosolic kinase C results in decreased pp80 phosphorylation. Although the location of pp80 is not known, the related synaptosomal87-kDa phosphoprotein is a cytosolic protein [S}. The role of pp80 phosphorylation in the mitogenic pathway remains to be defined. It is a very early substrate for kinase C appearing within minutes after phorbol ester or serum exposure [4]. The transformed BPA31 and DA31 cells were previously found to have constitutively high levels of c-myc, thus suggesting that pp80 phosphorylation and c-myc transcription may be reciprocally related
Kinase C function in transformed cells
43
[20]. The relationship between pp80 phosphorylation and some of the known early actions of kinase C such as tyrosine phosphorylation [21, 221, protooncogene induction [23, 241, phosphorylation of growth factor receptors [22, 251 and physiologic substrates [26-281, and S6 kinase activation [29] is not known. By itself, PMA is not mitogenic in A31 cells although it markedly increases the pp80 level. In contrast, PMA is highly mitogenic for DA31 cells although it increases the pp80 level only modestly. Kinase C undoubtedly activates other independent pathways which lead to mitogenesis, as demonstrated by the DA31 cells. It is not possible to state that pp80 phosphorylation is unnecessary for mitogenesis since transformation may have bypassed an otherwise necessary pp80 phosphorylation step. The author gratefully acknowledges the assistance of Diana Chaco in the preparation of the manuscript.
REFERENCES 1. Nishizuka, Y. (1980) Nature (London) 388, 693-698. 2. Castagna, M. J., Takai, Y., Kaibuchi, K., Sano, K., Kikkawa, U., and Nishizuka, Y. (1982) J. Biol. Chem. 257, 7847-7851. 3. Yamanishi, J., Takai, Y., Kaibuchi, K., Sano, K., Castagna, M., and Nishizuka, Y. (1983) Biochem. Biophys. Res. Commun. 112, 778-786. 4. Rozengurt, E., Rodriguez-Pena, M., and Smith, K. A. (1983) Proc. N&l. Acad. Sci. USA 80, 7244-7248. 5. Wu, W. C.-S., Walaas, S. I., Naim, A. C., and Greengard, P. (1982) Proc. N&l. Acad. Sci. USA 79, 5248-5253. 6. Albert, K. A., Walaas, S. I., Wang, J. K.-T., and Greengard, P. (1986) Proc. Nutl. Acad. Sci. USA 83, 2822-2826. 7. Yang, H. C., and Pardee, A. B. (1987) J. Cell. Physiol. 133, 377-382. 8. Dubrow, R., Riddle, V. G. H., and Pardee, A. B. (1979) Cancer Res. 39, 2718-2726. 9. Schneider, E. L., Stanbridge, E. J., and Epstein, C. J. (1974) Exp. Cell Res. 84, 311-318. 10. Campisi, J., Medrano, E. E., Morreo, G., and Pardee, A. B. (1982) Proc. Nutl. Acad. Sci. USA 79, 43w. 11. O’Farrell, P. H. (1975) J. Biol. Chem. 250, 4007421. 12. Croy, R. G., and Pardee, A. B. (1983) Proc. Natl. Acud. Sci. USA 80, 4699-4703. 13. Kikkawa, U., Takai, Y., Minakuchi, R., Inohara, S., and Nishizuka, Y. (1982) J. Biol. Chem. 257, 13,341-13,348. 14. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254. 15. Kraft, A. S., and Anderson, W. B. (1983) Nuture (London) 301, 621-623. 16. Anderson, W. B., Estival, A., Tapiovaara, H., and Gopalakrishna, R. (1985) in Advances in Cyclic Nucleotide and Protein Phosphorylation Research (Cooper D. M. F., and Seaman, K. B., Eds.), Vol. 19, pp. 287-306, Raven Press, New York. 17. Halsey, D. L., Girard, P. R., Kuo, J. F., and Blackshear, P. J. (1987) J. Biol. Chem. 262, 2234-2243. 18. Wolfman, A., and Macara, I. G. (1987) Nuture (London) 325, 359-361. 19. Melloni, E., Pontremoli, S., Michette, M., Sacco, O., Sparatore, B., Salamino, F., and Horecker, B. L. (1985) Proc. Natl. Acad. Sci. USA 82, 6435-6439. 20. Campisi, J., Gray, H. E., Pardee, A. B., Dean, M., and Sonnenshein, G. E. (1984) Cell 36, 241-247. 21. Gilmore, T., and Martin, G. S. (1983) Nature (London) 306, 487-490. 22. Moon, S. O., Palfrey, H. C., and King, A. C. (1984) Proc. Nutl. Acad. Sci. USA 81,22%2302. 23. Coughlin, S. R., Lee, W. M. F., Williams, P. W., Giels, G. M., and Williams, L. T. (1985)Cell 43, 243-251. 24. Blackshear, P. J., Stumpo, D. J., Huang, J. K., Nemenoff, R. A., and Spach, D. H. (1987)J. Biol. Chem. 262. 7774-7781.
44
Henry C. Yang
25. Jacobs, S., Sahyoun, N. E., Saltiel, A. R., and Cuatrecasas, P. (1983) Proc. Natl. Acad. Sci. USA 80, 6211-6213. 26. Werth, D. K., Niedel, J. E., and Pastan, I. (1983) J. Eiol. Chem. 258, 11,423-11,426. 27. Feurstein, N., and Cooper, H. L. (1984) .I. Biol. Gem. 259, 2782-2788. 28. Gould, K. L., Woodgett, J. R., Isacke, C. M., and Hunter, T. (1986) Mol. Cell. Biol. 6, 2738-2744. 29. Tabarini, D., Heir&h, J., and Rosen, 0. M. (1985) Proc. Natl. Acad. Sci. USA 82,4369-4373. Received June 10, 1988 Revised version received December 23, 1988
Printed
in Sweden