JOURNAL OF CELLULAR PHYSIOLOGY 142:231-235 (1990)

Augmented Desensitization To Epidermal Growth Factor (ECF) Immediate Actions: A Novel Mechanism for Altered EGF Growth Response in Mutant A431 Cells AVRAHAM KARASIK, SETHU S.-K. REDDY, R. BLAKE PEPINSKY, T O M M Y BROCK, AND C. R O N A L D K A H N * Resedrch Division, lodin Diabetes Center (A.K., S..S.-K.R.,C.R.K.) and Department of Medicine and Pathology, Brigham and Wornen’b Hospital (T.B.), Harvard Medical School, Boston 022 7.5, and Biogen Research Corporation, (R.B.P.) Cambridge 02 742, Massachusetts Epidermal growth factor (EGF! may either stimulate or inhibit cell growth. To elucidate the mechanim of these varied effects, we compared EGF action in parental A431 cells in which cell growth is inhibited, and clone 15, a mutant of these cells resistant to ECF growth inhibition. In both lines, EGF receptor was present in similar concentrations and underwent tyrosine phosphorylation to the same extent. Likewise, in both lines, acute exposure to EGF stimulated an increase in free cytoplasmic iCa2+l, as well as a similar increase in phosphorylation of lipocortin 1, a major substrate for the ECF receptor kinase whose phosphorylation i s calcium-dependent. O n the other hand, pretreatment of clone 15 cells with EGF for 72 h abolished EGF-induced phosphorylation of lipocortin 1 and led to a loss of the increase in cytoplasmic free [Ca’+], whereas no such desensitization was seen in the parental A431 cells. These data indicate a link between EGF-induced increase in cytoplasmic calc i um, I i pocort in phosphorylation, and cell growth and suggest that differences in mechanisms of desensitization to these immediate actions of EGF may lead to altered growth response to this hormone.

The binding of epidermal growth factor to its cell surface receptor initiates multiple immediate changes in the target cells. The earliest change is activation of the protein tyrosine kinase that is intrinsic to the receptor (Downward et al., 1984; Ullrich et al., 1984). This activation results in receptor autophosphorylation, as well a s phosphorylation of tyrosine residues in intracellular substrates, the best characterized of which is lipocortin 1 (Sawyer and Cohen, 1985; Pepinski and Sinclair, 1986). Activation of the tyrosine kinase is essential for the other immediate effects of EGF (Chen et al., 1987: Moolenaar e t al., 1988), including a rise in free cytoplasmic Ca2+ (Moolenaar et al., 1986), activation of the Na+iH’ antiporter (Rothenberg et al., 1983), and activation of the phosphoinositide second-messenger system (Pike and Eakes, 1987). In most cells, this chain of events leads to enhanced cell growth (Carpenter and Cohen, 1979). In A431 cells, a human epidermal carcinoma cell line that possess a large number of EGF-receptors, this is associated with a n inhibition of growth (Gill and Lazar, 1981; Bravo et al., 1985). Mutants of A431 cells resistant to the growth inhibitory effects of EGF have been developed and used to study the mechanisms of EGF action (Bravo et al., 1985; Gill et al., 1982). In the present study, we have used one such cell line termed clone 15, initially developed and characterized by Bravo et al. (1985) to exam-

0 1990 WILEY-I,ISS, INC.

ine the role of early EGF effects on the final outcome on growth. We find t h a t although EGF binding, EGF-induced receptor autophosphorylation, phosphorylation of lipocortin 1, and increases in cytoplasmic Ca2+ in clone 15 are similar t o the parental A431 cells, there is a unique pattern of desensitization to EGF in clone 15, which may account for the differences in long-term effects of EGF on cell growth.

EXPERIMENTAL PROCEDURES Materials 1 3 2 P l o r t h ~ p h ~ ~ p h[3H]thymidine, ate, and Triton X100 were obtained from New England Nuclear. lZ5IEGF and EGF were from Collaborative Research. Clone 15 A431 cells were a generous gift from Dr. R. Bravo (Heidelberg, FRG). Fura-2 acetoxymethylester was from Molecular Probes. Reagents for SDSiPAGE Received June 12, 1989; accepted September 22, 1989.

*To whom reprint requestsicorrespondence should be addressed. Abbreviations used: BSA, bovine serium albumin; DMEM, Dulbecco’s modified Eagle’s medium; EGF, epidermal growth factor; PAGE, polyacrylamide gel electrophoresis; PBS, phosphate-buf‘fered saline; PMSF, phenylmethylsulfonyl fluoride; SDS, sodium dodecyl sulfate.

232

KARASIK ET AL

from Bio-Rad Laboratories, and all other chemicals were from Sigma. Antilipocortin 1 serum was generated in rabbits as described by Huang et al. (1986). Antiphosphotyrosine antibody was purified from the sera of rabbits immunized with phosphotyramine coupled to keyhole limpet hemocyanin by affinity chromatography as described by Pang et al. (1985).

Cell counts Cells were plated at a density of 2 x lo3 per well in a 24-well culture plates (Costar). After 24 h in DulbecCO'S modified Eagle's medium (DMEM) containing 10% fetal calf serum, the cells were washed and grown for 72 h in DMEM containing 1% HSA with and without EGF. At the end of this period, cells were suspended by trypsin treatment and counted in a Coulter counter. Thymidine incorporation Subconfluent cultures of cells in 24-well culture plates were incubated 24 h in DMEM with BSA and further incubated for 16 h in the BSA containing medium plus EGF and for 1 h in the same medium containing 0.5 FCi/ml [3H]thymidine. The labeling solution was removed and the monolayers washed three times with ice-cold phosphate buffered saline (PBS). The cells were dissolved in 1 ml of 0.1% SDS and the DNA precipitated with 2 ml of 20% trichloroacetic acid a t 4" C. The precipitate was collected by centrifugation and suspended in 1N NaOH. The base was neutralized, and the radioactivity was measured in a scintillation counter.

EGF binding assay l2"I-1abeledEGF binding was performed in confluent monolayers cultured in 24-well plates. Cultures were washed three times with PBS and then incubated for 3 h a t 4" C with Earle's balanced salt solution containing 0.1% BSA, 0.2 nM lZ5I-EGFand increasing concentrations of nonradioactive EGF. At the end of the incubation, cells were washed with PBS containing 0.1% BSA and solubilized with 0.1 M NaOH in 1%SDS, and the radioactivity was measured. Binding was expressed as percent Iz5I-EGFbound per 0.1 mg protein. Nonspecific binding was measured in the presence of 1 Fgiml of unlabeled EGF and subtracted from total binding to give specific binding. Triplicate assays were carried out a t each EGF concentration.

remove insoluble material, the supernatants were diluted to equalize protein concentration, and each supernatant was divided into two equal aliquots. These supernatants were subjected to immunoprecipitation with either antiphosphotyrosine or antilipocortin antibodies as described by Kasuga et al. (1984). Immunoprecipitates were analyzed by SDSiPAGE and subjected to autoradiography.

EGF-stimulated changes in [Caz'li Cells were grown in the presence and absence of EGF under conditions detailed in phosphorylation assay. Cells were loaded with 2 pM Fura-2 acetoxymethylester for 20 min a t 37°C and placed in a temperature controlled perfusion chamber mounted on the stage of a Nikon Diaphot microscope. Cells were perfused (2 mli min) with a balanced salt solution for 10 min and then exposed to 500 ngiml EGF. Fluorescence measurements were made using a SPEX Fluorolog I1 spectrofluorometer (Edison, NJ) equipped with a beam splitter and two excitation monochromators which allowed the alternating excitation of Fura-2 at 340 and 380 nm. A x 40 Flour Nikon objective connected in series with a narrow band pass filter (500 ~f10) and photomultiplier tube was used to collect the emitted fluorescence signals from five t o ten cells. Coverslips were Calibrated against 5 pM ionomycin in the presence of 1.5 mM CaC1, and 2 mM EGTA to obtain maximum and minimum signals, respectively. Autofluorescence was determined using a 2 mM MnCl solution and subtracted prior to calculation of [Ca2 ' 1,.

RESULTS AND DISCUSSION To establish the difference in EGF effects on cell growth, A431 and clone 15 cells were plated at the same density, and growth was monitored for 72 h in medium with and without EGF. In A431 cells, 100 ngi ml of EGF led to 54 i 7% decrease in cell number over 72 h as compared with cells grown without EGF, whereas with clone 15 cell, a 21 i 11%increase was observed in response to EGF (Fig. 1, right). A similar pattern of response was seen when the effect of EGF on thymidine incorporation was measured (Fig. 1, left). Incubation of A431 cells with 100 ngiml EGF abolished thymidine incorporation into DNA, and a milder effect was seen even after exposure to 1 ng/ml EGF. By contrast, exposure of clone 15 cells to 1 ngiml EGF increased thymidine incorporation more than threefold, Phosphorylation in intact cells and and 100 ng/ml produced almost a doubling in thymiimmunoprecipitation of dine incorporation. phosphotyrosine-containing proteins To determine if these differences in growth effects Subconfluent cells in 6-well plates were incubated could be explained by altered receptor number, EGF for 72 h in DMEM containing 1%BSA with or without binding assays were performed in confluent A431 and 100 ngiml EGF (indicated chronic to allow full ex- Clone 15 cells. Tracer binding was slightly higher in pression of EGF effects. After a n additional 8 h without clone 15 cells, but on Scatchard analysis receptor numhormone, cells were labeled for 2 h with 32P-orthophos- ber and affinity were similar in both cell lines (Fig. 2). phate (0.5 mCi/ml) in phosphate-free DMEM and then The next essential step in the EGF signal transmisstimulated with EGF (250 ng/ml) for 5 mim'4 h. The sion is EGF stimulation of the intrinsic protein tyreaction media was aspirated, and the phosphorylation rosine kinase activity, which leads to receptor autoreaction stopped by quick freezing the cells with liquid phosphorylation on tyrosine residues. Comparison of Nz. Cells then were allowed to thaw into a 4" C solution receptor autophosphorylation in intact cells was ascontaining 50 mM HEPES, 1% Triton, 5 mM EDTA, sessed by exposure of "P-labeled A431 and clone 15 100 mM sodium fluoride, 2 mM PMSF, 10 mM sodium cells to 250 ng/ml EGF precipitation of tyrosine phospyrophosphate, and 2 mM sodium vanadate. Following phorylated proteins using antiphosphotyrosine antiultracentrifugation at 200,OOOg for 60 min a t 4" C to body and analysis of SDS-PAGE and autoradiography.

233

EGF ACTION IN MUTANT A431 CELLS

A431

C l o n e 15

A431

Clone 15

T

T

JZ t-

I

I

)r)

'L

T

0

0

0

1 100

1100

0

Fig. 1. Effect of EGF on thymidine incorporation and cell number in parental A431 and in clone 15 cells. Effect of different concentrations of EGF on growth in both clones were assessed by monitoring thymidine incorporation (left panel) and cell number (right panel) after incubation with EGF as described in Experimental Procedures. Val-

10

50

100

EGF concentration (ng/ml) Fig. 2. EGF binding to A431 and clone 15 cells. '"1-labeled EGF binding was performed in confluent monolayers of both clones cultured in 24-well plates. Specific binding was expressed as percent bound per 0.1 mg protein. Triplicate assays were carried out a t each EGF concentration.

The most prominent tyrosine phosphorylated protein in both cell lines in response to EGF was the EGF receptor (M, = 170 kDa) (Fig. 3, upper panel, lanes b,f). The extent of EGF receptor autophosphorylation was similar in both cell lines and did not change substantially in cells chronically exposed (72 h) to EGF prior to the acute stimulation (Fig. 3, upper panel, lanes d,h), i.e., similar to the conditions under which the changes in growth are observed. Thus, in clone 15, the resistance to EGF growth inhibitory effects does not stem from a change in receptor number or phosphorylation (since phosphorylation is the product of these two factors), but rather is based on a subsequent event in the transduction of the signal. There are some differences in background 32P-labeled bands in the

i

0

1 100

0

1100

ucs given for both sets of experiments were standardized such that control cells, i.e., cells not created with EGF, were arbitrarily assigned the value 1 and others calculated relative to this value. Each value represents mean i SEM of thrce independent experiments.

clone 15 cells, which may reflect some nonspecific changes these cells have undergone during selection. Activation of the EGF receptor kinase also results in tyrosine phosphorylation of lipocortin 1; however, as we have previously shown, this protein is not precipitated by our antiphosphotyrosine antibody or detected in SDS gels of the precipitates (Karasik et al., 1988). Thus, to evaluate EGF-induced phosphorylation of lipocortin 1, the same cell extracts were subjected to immunoprecipitation with a specific antilipocortin 1 antibody (Fig. 3, lower panel). Acute stimulation with EGF led to phosphorylation of lipocortin 1in both A431 and clone 15 cells. This could be identified as a 36 kDa phosphorylated band on the autoradiogram of these immunoprecipitates (Fig. 3, lower panel, lanes b,f). In clone 15 cells, chronic exposure to EGF prevented lipocortin 1phosphorylation in response to acute hormone stimulation (lane h), and no phosphorylation of lipocortin 1 was detected up to 4 h after exposure to the hormone. By contrast, there was a small change in EGF's ability to stimulate lipocortin phosphorylation after chronic exposure of A431 cells to EGF (Fig. 3, lower panel, lane d). The dramatic decrease in lipocortin 1 phosphorylation in clone 15 cells suggests that in these cells a specific augmented desensitization occurs in response t o long-term exposure t o the hormone. As lipocortin 1phosphorylation is calcium dependent (Fava and Cohen, 1984), and EGF is known to cause a n immediate rise in cytoplasmic free Ca2+ ([Ca2+],)in A431 cells (Moolenaar et al. 1986), we measured EGFinduced changes in [Ca2'I, in both cell lines under the conditions associated with lipocortin phosphorylation. In both cell lines, acute exposure to EGF induced a rapid rise in [Ca2 ' 1, as measured by microfluorescence of Fura-2 loaded cells. The rapid rise was followed by a gradual decrease in [Ca'+:, returning to baseline after 5-7 min (Fig. 4A,B). Chronic exposure to EGF changed the patterns of [Ca2+l,in both cell lines, but in a different fashion. In pretreated A431 cells EGF induced a

234

KARASIK ET AL.

Anti-Phoephotyrosine Antibody

Clone 15

A43 1

___

Mr x 1 C ' ~~

- 200 - 116 - 92 - 66 - 45 - 31 A

B

C

D

E

F

G

H

Anti-Lpocortin 1 Antibody A431

Clone 15

- 200 -

115

- 92 - 66 - 45 Ilp 1 *

+lip1

A

B

C

D

E

F

G

31

H

Fig. 3. EGF-induced phosphorylation of tyrosine-containing proteins and of lipocortin 1 in A431 and in clone 15 cells. Subconfluent cells were incubated for 72 h in DMEM containing 1% BSA with and without 100 ng/ml EGF (indicated chronic * )to allow full expression of EGF effects. Cells then were labeled with 32P-orthophosphateand stimulated with EGF (indicated as acute .t).Phosphorylated proteins were immunoprecipitated with either antiphosphotyrosine (upper panels) or antilipocortin-1 (lower panel 1 antibodies. Immunoprecipi-

tates were analyzed by SDS-PAGE and subjected to autoradiography. The autoradiograms in the upper panel represent immunoprecipitates with antiphosphotyrosine antibody and in the lower panel with antilipocortin 1 antibody. The left part of each panel represents A431 cells and the right clone 15 cells. The left lanes in each group (A, B, E, F) are cells stimulated after 72 h precincubation with EGF, and the right lanes (C, D, G, H) represent cells without EGF.

rapid increase in [Ca2+l, of a similar magnitude as seen in controls. However, the [Ca2+],remained high for the entire length of the tracing and did not return to the basal level (Fig. 4C). In clone 15 cells exposed t o EGF for 72 h, acute addition of EGF resulted in no increase in [Ca2+],(Fig. 4D). Thus, prolonged exposure to EGF led to desensitization of clone 15 cells to the acute effects of EGF on both lipocortin 1 phosphorylation and changes in [Ca2+Ii. The altered patterns of hormone-induced modulation

of [Ca2 ' 1, following prolonged exposure to EGF mimic patterns that may be seen after manipulation of protein kinase C activity in cells (Pandiella et al. 1987). Protein kinase C has been suggested to mediate the gradual decline of intracellular calcium after the acute rapid increase induced by EGF (Moolenaar et al., 1986).Activation of protein kinase C by pretreatment with phorbol ester blocks the increase in [Ca2'1, in response to EGF, resulting in a response similar to that seen after chronic exposure of clone 15 cells t o EGF.

235

EGF ACTION IN M1JTANT A431 CELLS

7

f 1

%

D.

EGF

I.----

z

.+-

rnin

M

+

EGF

0

v

Fig. 4. EGF-stimulated incrcases in [Ca'+ J, in A431 and Clone 15 cells. A431 and Clone 15 cells were grown on coverslips with or without EGF for 72 h. Cells were loaded with 2 pM Fura-2, and fluorescence signals from five to ten cells reflecting changes in LCaL-l,were recorded as described in Experimental Procedures. EGF-induced changes in [Ca" I, are depicted in A431 cells before (A) and after (C) 72 h pretreatment with EGF and in clone 15 cells before (B) and after (D) such pretreatment.

Downregulation of protein kinase C by chronic ex o sure to phorbol esters prevents the decline in [Ca!P' Ii resulting in a pattern similar to that observed in A431 cells pretreated with EGF (Pandiella et al., 1987). Our data show that the clone 15 cells differ from parental A431 cells in their ability to undergo desensitization to the immediate effects of EGF. This suggests a novel mechanism by which cells escape from EGF growth inhibitory effects and that may involve activation of protein kinase C. From the multiple, interrelated early events induced by EGF upon binding to its receptor, lipocortin 1 phosphorylation and/or increase in cytoplasmic-free ICa2'1, may be important in mediating EGF effects on growth.

ACKNOWLEDGMENTS The authors are grateful t o Dr. Rodrigo Bravo for generously providing the cell lines used in these studies and to Terri-Lyn Bellman for secretarial assistance. This work was supported in part by National Institutes of Health grant DK 33201 (CRK), and by the Joslin Diabetes and Endocrinology Research Center grant DK 36836. Dr. Karasik is a recipient of a Capps Scholarship. LITERATURE CITED Bravo, R., Burckhardt, J., Curran, T., Muller, R. (1985) Stimulation and inhibition of growth by EGF in different A431 cell clones is

accompanied by the rapid induction of c-fos and c-myc proto-oncogenes. EMBO J., 4r1193-1197. Carpenter, G., and Cohen, S. (1979) Epidermal growth factor. Annu. Rev. Riochem., 48:193-216. Chen, W.S., Lazar, C.S., Poenie, M., Tsien, R.Y., Gill, G.N., and Rosenfeld, M. (1987) Requirments for intrinsic protein tyrosine kinase in the immediate and late actions of the EGF receptor. Nature, 328r820-823. Downward, J., Yarden, Y., Mayes, E., Scrace, G., Totty N., Stockwell, P., Ullrich, A., Schlessinger, J., and Waterfield, M.D.; (1984) Close similarity of epidermal growth factor receptor and v-erb-B oncogene protein sequences. Nature, 307:521-527. Fava, R.A., Cohen, S. (1984) Isolation of a calcium-dependent 35kilodalton substrate for the epidermal growth factor receptorikinase from A-431 cells. J. Biol. Chem., 2592636-2645. Gill, G.N., and Lazar, C.S. (1981) Increased phosphotyrosine content and inhibition of proliferation in EGF-treated A431 cells. Nature, 293r305-307. Gill, G.N., Buss, J.E., Lazar, C.S., Lifshitz, A,, Cooper, J.A. (19821 Role of epidermal growth factor-stimulated protein kinase in control of proliferation of A431 cells. J . Cell Biochem., 19:249 -257. Huang, K.S., Wallner, B.P., Mattalino, R.J., Tieard, R., Burne, C!., Frey, A,, Hession, C., McGray, P., Sinclair, L.K., Chow, E.P., Browning, J.L., Ramachandran, K.L., Tang, J., Smart, J.E., and Pepinsky, R.B. (1986) Two human 35 kd inhibitors ofphospholipase A, are related to substrates of pp60'~""and of the epidermal growth factor receptorikinase. Cell, 46r191-199. Karasik, A., Pepinsky, R.B., Shoelson, S.E., and Kahn, C.R. (1988) Lipocortins 1and 2 as substrates for the insulin receptor kinase in rat liver. J. Biol. Chem., 263r11862-11867. Kasuga, M., White, M.F., and Kahn, C.R. (1984) Phosphorylation of the insulin receptor in cultured hepatoma cells and a solibilized system Methods Enzymol., 109r609-621. Moolenaar, W.H., Aerts, R.J., Tertoolen, L.G.J., and deLaat, S.W. (1986)The epidermal growth factor-induced calcium signal in A431 cells. J . Biol. Chem., 261:279-284. Moolenaar. W.H., Bierman, A.J., Tilly, B.C., Verlaan, I., Defize, L.H.K., Ilonegger, A.M.. Ullrich, A., and Schlessinger, J. (1988) A point mutation at the ATP-binding site of the EGF-receptor abolishes signal transduction. EMBO J., 7:707-710. Pandiella, A,, Vicentini, L.M., Meldolesi, J . (1987) Protein kinase C-mediated feed back inhibition of the Ca' response a t the EGF receptor. Biochem. Biophys. Res. Commun., 149:145-152. Pang, D.T., Sharma, B.R., and Shafer, J.A. (1985) Purification of the catalytically active phosphorylated form of the insulin receptor kinase by affinity chromatography with 0-phosphotyrosyl-binding antibodies. Arch. Biochem. Biophys., 242:176-186. Pepinsky, R.B., and Sinclair, L.K. (1986) Epidermal growth factordependent phosphorylation of lipocortin. Nature, 322 r81-84. Pike, L.J., and Eakes, A.T. (1987) Epidermal groeth factorstimulates the Droduction of ahosnhatidvlinositolmonoahosahate and the breakdown of polypgosphoinositides in A431 &lls.'J. Biol. Chem., +

262:1644-1651 ~~.~

~

~~~~

Rothenberg, P., Glaser, L., Schlessinger, P., and Cawel, D. (1983) Activation of Na'lH exchange by epidermal growth factor elevates intracellular pH in A431 cells. J. Biol. Chem., 261.279-289. Sawyer, S.T., and Cohen, S. (19851 Epidermal groeth factor stimulates the phosphorylation of the calcium dependent 35,000-dalton substrate in intact A431 cells J. Biol. Chem., 260r8233-8237. Ullrich, A,, Coussens, J.S., Hayflick, T.J., Dull, A,, Gray, A.W.. Lee J. Yarden, Y., Liberman, T.A., Schlessinger, J., Downward, J., Mayes, E.L.V., Whittle, N., Waterfield, M.D., and Seeburg, P.H. (19841 Human epidermal growth factor receptor cDNA sequence and abberant expression of the amplified gene in A431 epidernioid carcinoma cells. Nature, 309:418-425.