Epidermal Growth Factor Stimulates Tyrosine ...

THEJOURNALOF BIOLOGICAL CHEMISTRY

Val. 258, No. 21, Issue of November 10, pp. 13089-13093,1983 Printed in U.S.A.

Epidermal Growth Factor Stimulates Tyrosine Phosphorylation of the Myosin Regulatory Light Chain from Smooth Muscle* (Receivedfor publication, June 10, 1983)

Byron Gallis, Arthur M. Edelman, John E. Casnellie, and Edwin G. Krebs From the Howard Hughes Medical Institute Laboratories, Department of Pharmacology, University of Washington, Seattle, Washington 98195

stimulates tyrosine rather than serine phosphorylation, we investigated whether the regulatory light chain of myosin could be phosphorylated in tyrosine residues. We now report that EGF stimulates the receptor-associated tyrosine protein kinase of A431 membranes to catalyze phosphorylation of two tyrosine residues in the M , = 20,000 myosin light chain. EXPERIMENTALPROCEDURES

Materials-EGF was purchased from Collaborative Research; phosphothreonine, phosphoserine, casein (C-78911, histone fl, histone a b , phosphoglycerate mutase, L-lactic dehydrogenase, and enolase from Sigma Chemical Co.; and [y3’P]ATP from New England Nuclear. Phosphotyrosine was a gift from Dr. Linda J. Pike of this laboratory. Phosphorylase was a gift from Dr. Edmond Fischer, Department of Biochemistry, University of Washington. Sodium orthovanadate wasfrom Fisher Scientific Co. TPCK-trypsin was from Worthington. Cells and Membrane-Human epidermoid carcinoma (A431) cells were grown as previously described (13). Membranes from A431 cells were prepared according to Brautigan et al. (13). Purification of Smooth Muscle Myosin Light Chin-The M, = 20,000 smooth muscle myosinlight chain was purified by a procedure to be described in detail elsewhere., Briefly, it involved the purification of myosin from fresh chicken gizzard by modifications of the Epidermal growth factor stimulates phosphorylation of ty- method of Sobieszek (14). Light chains were then dissociated and the rosine residues of several proteins of known molecular weight heavy chains precipitated with 5 M guanidine HC1,63% (v/v) ethanol, in human epidermoid carcinoma (A431) cells (1). It also according to Holt and Lowey (15). Purification ofLC20 from the stimulates the phosphorylation of synthetic peptides (2, 3) mixed light chain fraction was accomplished by chromatography on Sephacryl S200 a t 23 ‘C in the presence of 5 M guanidine HCl. LCzo and exogenous protein substrates (4, 5) in reactions contain- was dialyzed against a series of buffers of gradually decreasing NaCl ing A431 membranes. These phosphorylations are mediated concentrations to maintain solubility, concentrated by vacuum diby the EGF’ receptor, which has an associated tyrosine pro- alysis, and stored at -20 “C. Its concentration was determined by the tein kinase activity (6,7). In addition to inducing in vitro and method of Lowry et al. (16) using modifications suggested by Peterson in vivo phosphorylation, EGF induces a variety of other (17). The LC,, preparation was 91% pure as estimated by SDS-gel responses in cells that have EGF receptors (8). In A431 cells, electrophoresis and densitometry. Light C h i n Phosphorylation-Phosphorylation reactions which possess about 1-2 X lo6 EGF receptors/cell, EGF (90Myosin pl) containing 6 pgofA431 membranes, 100 ng of EGF, 0.1% induces changes in cell surface morphology within 30 min (9) Nonidet P-40; 2.5 mM MnCl,, 30 mM HEPES, pH 7.4,lOO p~ sodium and slower responses, including stimulation of DNA synthesis orthovanadate, 600 p~ [y3’P]ATP (specific activity 150-1000 cpm/ (10) and cell proliferation (11). pmol) and the indicated concentrations of LCzo were carried out at Ithas been hypothesized thatthe rapid morphological 30‘C for the indicated times. Phosphorylation ofLC20 was deterchanges produced by EGF are due to cytoskeletal alterations mined by spotting an aliquot of the reaction mixture on P81 phosphocellulose squares (Whatman), whichwere then acidified and controlled by the phosphorylation of myosin (9). Myosin is washed once with cold 10% trichloroacetic acid and three times with known to be phosphorylated in a serine residue in one of its 5% trichloroacetic acid. Nonspecific phosphorylation (A431memlight chains (Mr = 20,000 in smooth muscle and nonmuscle branes alone) was corrected for by subtraction of the cpm bound to cells) in a Ca2+-dependentmanner (12); this serine phospho- phosphocellulose in these reactions from reactions under identical rylation is thought to control smooth muscle contraction and conditions that contained LC2,,. The background cpm were always cytoskeleton related processes in nonmuscle cells. Since EGF 4 0 % of the cpm incorporated in the presence of LCzo. SDS-Gel Electrophoresis, Hydrolysis of Protein, and Phosphoamino * The costs of publication of this article were defrayed in part by Acid Analysis-Reaction mixtures containing phosphorylated LCzo the payment of page charges. This article must therefore be hereby were boiled in Laemmli sample buffer containing dithiothreitol and marked “aduertisement” in accordance with 18 U.S.C. Section 1734 the proteins were electrophoresed on 15% polyacrylamide slab gels. The gels were stained, destained, and dried as previously described solely to indicate this fact. (13). The LCZo wasexcised from the gel and eluted by incubation of The abbreviations used are: EGF, epidermal growth factor; LC,,,, the M, = 20,000 myosin regulatory light chain; TPCK-trypsin, L-1- the gel slice in 2 ml of 100 mM NH4HC03containing 200 pg of TPCKtosylamido-2-phenylethylchloromethyl ketone-treated trypsin; SDS, sodium dodecyl sulfate; HEPES, 4-(2-hydroxyethyl)-l-piperazineA. M. Edelman, D. Hunter, A. E. Hendrickson, and E. G. Krebs, manuscript in preparation. ethanesulfonic acid. ~

13089

Downloaded from www.jbc.org by guest, on July 13, 2011

Epidermal growth factor (EGF) stimulates thephosphorylation by A431 membranes of tyrosine residues inthe myosin regulatorylight chain (LC~O)from chicken gizzardsmooth muscle. In the presence of EGF, the K , of the EGF receptor kinase for LCz0 is 73 pM and the V,, is 17 nmol/min/mg. Two moles of phosphate are incorporated into tyrosine per mol of LCzo. Trypsin digestionof the phosphorylated LC20 produces two phosphopeptides which are phosphorylated to approximately equal extents. Sequential Edman degradation of the separated phosphopeptides shows that tyrosines 142 and 155 of the protein are phosphorylated. Tyrosine142 islocated within a sequence similar to that of autophosphorylated tyrosine kinases in that five of the sevenamino acids on the NHz-terminal side are acidic. Tyrosine155 has no acidic amino acids near its NHz-terminal side. A comparison of the initial rates of phosphorylation of the two tyrosines shows that tyrosine 142 isphosphorylated at three times the rate of tyrosine 155.

Tyrosine Phosphorylation

13090

of Myosin Light Chain

trypsin for 24 h at 30 “C.The elution was repeated with fresh protease Its amino acid sequence (20) suggests that cleavage of the and buffer for 24 h, and the pooled eluates were lyophilized. The molecule with trypsin should generate three distincttyrosinepeptides were subjected to partial acid hydrolysis and high voltage containing peptides. After phosphorylation of LC20, the propaper electrophoresis as described (2). polyacylamide gels, eluted with Analysis of Tryptic Peptides-After lyophilization, the tryptic pep- tein was fractionatedon tides were taken up in water and spotted onto thin layer cellulose TPCK-trypsin, and the tryptic fragments were fractionated acetate plates (10 X 10 cm) and electrophoresed in 1%(NH4)&03, by thin layer electrophoresis on cellulose acetate at pH 8.9. pH 8.9, for 20 min at 500 V. The plates were dried and autoradi- An autoradiogram of the separated peptides showed two major ographed. spots (Fig. 4), suggesting that two of the three tyrosines in Sequencing of Tryptic Peptides-LCzo (480 pg) wasphosphorylated to a level of 2 molof 32P/mol of protein and isolated by SDS- LC2, were labeled by A431 membranes. This was in agreement polyacrylamide gel electrophoresis. Phosphopeptides were generated by incubation with 25 pg of TPCK-trypsin in 100 mM NH4HC03for 18 h a t room temperature. After lyophilization, the peptides were A A B C D separated by HPLCona reverse phase octadecyl-silane column (Varian) with 0.1% trifluoroacetic acid as the aqueous phase and 0.08% trifluoroacetic acid in acetonitrile as the organic phase. The phosphopeptide containing tyrosine 142 was clearly separated from the othertrypticfragments while the phosphopeptide containing tyrosine 155 was further purified using an isocratic gradient held a t 23% of the organic phase. The sequence of each phosphopeptide was determined as described (18). u P-Tyr RESULTS

3B. Gallis, A. M. Edelman, J. E. Casnellie, and E. G. Krebs, unpublished observations.

-

P-Thr

P-Ser

u

FIG. 1. Phosphorylation of LC20 by A431 membranes and phosphoamino acid content of LC20.A, Autoradiogram of an SDSpolyacrylamide gel after fractionation of A431 kinase reactions. The reactions were for 20 min. The reactions were carried out in the absence (Lanes A and C)or presence (Lanes B and D)of EGF. Lanes C and D contained 30 p~ LCZO.B, Phosphoamino acid analysis of phosphorylated light chain. Reactions in which LC20 was phosphorylated to a stoichiometry of 1.9 mol/mol were fractionated by polyacrylamide gel electrophoresis. After localization by staining and autoradiography, the light chain was excised and the phosphoamino acid content was determined as described under “Experimental Procedures.” An autoradiogram after high voltage paper electrophoresis of the phosphoamino acids of LC, is shown.

t

-.02-.01 .01.02.03.04

1/S (JIM”)

0

no EQF

I

0

50

I

I

I

I

100 150 200 250

pM Myosin light chain

FIG.2. Rate of incorporation of 32Pinto LC20as a function of LC20concentration in the absence and presence of EGF. Reactions contained 6 pg of A431 membrane protein and were for 10 min, during which time phosphorylation rates were linear (not illustrated). ATP was 380 p~ (506 cpm/pmol). The inset shows a Lineweaver-Burke plot of the incorporation in the presence of EGF.


Addition of EGF to phosphorylation reactions containing A431 membranes and 30 ~ L LCzo M caused marked stimulation of phosphorylation of trichloroacetic acid-precipitable proteins. Fractionation of the proteins by polyacrylamide gel electrophoresis, followed byautoradiography, showed that the major phosphorylated protein was LCzo (Fig. l A , Lane D). Phosphorylation of LC2, in the absence of EGF (Fig. lA,Lane C ) was 8% of the level of phosphorylation of LCzo in the presence of hormone, as determined by trypsin elution of cpm from gel slices containing LC2,. Control reactions without LCzo (Fig. l A , Lanes A and B ) showed phosphorylation of small peptides from A431 membranes which ran with the bromphenol blue front, although the level of this phosphorylation was unaffected by EGF. A431 membranes phosphorylate the M , = 17,000 myosin light chain,but at a rate considerably less than that ofLC,,. The inclusion of 100 ~ L Msodium orthovanadate, reagent a thatinhibitsboth membrane ATPases and phosphatases (19), was necessary for maximal incorporation of phosphate into LC20.3 In order to determine which phosphoamino acids were present in LC20 after phosphorylation by A431 membranes, the phosphorylated protein was excised from the gel, eluted by incubation with trypsin, and subjected topartial acid hydrolysis. Phosphoamino acids were separated by high voltage paper electrophoresis. All of the radioactivity was found to co-migrate with phosphotyrosine (Fig. 1B).The strong stimulation of phosphorylation by EGF suggests that EGF receptor kinase was responsible for catalyzing this phosphorylation. Addition of increasing amounts ofLC2, to A431 kinase reactions containing EGF produced a hyperbolic increase in phosphorylation rates (Fig. 2). From a Lineweaver-Burk plot of these data (see inset), an apparent K, (73 p ~ and ) a V,,, (17 nmol/min/mg) were calculated. It was not possible to determine accurately an apparentK, and V,,, for the reaction in the absence of EGF, because the rate of incorporation was very low (1.5-2.0-fold above background). In order to determine the extent of phosphorylation of LC2,, it was incubated with A431 membranes inthe presence of EGF for an extended period of time to allow the reaction to go to completion (Fig. 3). The maximum amount of phosphate incorporated was 2 (* 0.02) mol/mol of LCzo. Chicken gizzard LC2, contains threetyrosine residues (20).


1

I

1

I

ofChain Myosin Light

13091

I

I A

F T D E E V D E M Y R

6 12 IO

B

Time (mid

c)

CYCLE NUMBER G N F N Y V E F T R

6

FIG. 5 . Sequence analysis of 3zP-labeled tryptic peptides from LCzo. The phosphopeptides from a tryptic digest of 480 pg of LC20 were isolated by HPLC. A , The phosphopeptide which eluted first from HPLC, 4 nmol, 900,000 cpm. B, The phosphopeptide which eluted last from HPLC, 0.65 nmol, 188,000 cpm. The fraction from each cycle of Edman degradation was counted by Cerenkov radiation and then subjected to HPLC for determination of the amino acid. The sequence obtained in given by the one-letter code for amino acid residues.

FIG. 4. Autoradiogram of tryptic phosphopeptides of LCzo. Reactions in which LC20 was phosphorylated to a stoichiometry of 1.7 mol/mol werefractionated by SDS-polyacrylamide gel electrophoresis and theLC20 was eluted and trypsinized. Phosphopeptides were fractionated by high voltage electrophoresis on thin layer cellulose acetate as described. The arrow indicates the origin. Migration is toward the anode.

with the observed stoichiometry of the phosphorylation reaction. In order to determine which two tyrosines in LCzo were phosphorylated, larger amounts of the two major tryptic pep-


FIG. 3. Stoichiometry of incorporation of 3zP into myosin light chain. ATP was 640pM (258 cpm/pmol) and theconcentration of LCmwas 9.4 p ~ .

tides were separated by HPLC and analyzed by sequential Edman degradation. One of the phosphopeptides gave the sequence Phe-Thr-Asp-Glu-Glu-Val-Asp-Glu-Met-XArg with a peakof radioactivity at cycle number 10 (Fig. 5A). This sequence corresponds to the published sequence (20), from which we deduce that thelabeled amino acid is tyrosine 142. The sequence of the otherphosphopeptide was Gly-AsnPhe-Asn-X-Val-Glu-Phe-Thr-Arg with a peak of radioactivity at cycle number five (Fig. 5B).This sequence corresponds to the published sequence from which we deduce that the labeled amino acid at cycle number five is tyrosine 155. While there are five acidic amino acid residues in the first seven amino acids NHz-terminal to tyrosine 142, there areno acidic residues to theNHz terminus of tyrosine 155. In prolonged reactions, tyrosines 142 and 155 are both phosphophorylated stoichiometrically (Figs. 2 and 3), but it was of interest todetermine whether these two tyrosines could be phosphorylated a t different rates. Separate kinase reactions containing light chainwere incubated for 10,30, and90 min and theLC20 tryptic phosphopeptides from each reaction were separated by thin layer electrophoresis and isolated by scraping and extracting the radioactive spots after determination of their location by autoradiography. The spots were counted and the ratesof incorporation of 32Pinto each were determined for the indicated times (TableI). During the first 30 min of the reaction, the rateof phosphorylation of tyrosine 142 is three times that of tyrosine 155.

13092


of Chain Light Myosin

TABLEI Extent of incorporation of 32Pinto tyrosine 142 and tyrosine 155 as a function of reaction time Total incorporation into LCzowas as determined as under “Experimental Procedures.” LCzo was eluted from gel slices with trypsin, and phosphopeptides were separated by thin layer electrophoresis. The phosphopeptides were located by autoradiography, scraped from the plates, extracted with 50% acetic acid, and counted. The recovery of cpm from the plateswas 52%. The number of moles of 32P/moI of LCzo was calculated from the specific activity of ATP (1660 cpm/ pmol) and the totalincorporation. The number of mols of 32P ineach peptide was calculated from the percentage of the total cpm in each peptide.


membranes in the presence of EGF for the following substrates: 0.28 mM and 7.5 nmol/min/mg for a synthetic peptide containing the tyrosine phosphorylation site ofpp60”” (2), 0.11 mM and 1.7 nmol/min/mgfora synthetic peptide of gastrin (31,240 p~ and 1.7 nmol/min/mg for a synthetic tyrosine-containing peptide of p21 Harvey and Kirsten viruses (221, and 26 p~ and 0.2 nmole/min/mg for gastrin (4). It isof interest to compare the kinetic parametersLC, for of A431 EGF receptor kinase with thoseof myosin light chain kinase. The K,,, and VmaXfor phosphorylation of isolated LCzo by myosin light chain kinase are 5 p~ and 30 pmol/min/mg, K, for LCzoof the tyrosine respectively (23,24). The apparent Time Tyr 142 Tyr 155 protein kinase is73 pM. Assuming recoveries of A431 receptor nin mol 3 Z P l m ~LCm l kinase inourmembranepreparationssimilartothose of 10 0.26 0.08 Carpenter et al. (25), we can calculate from the determination 30 0.61 0.20 of the number of receptors in A431 membranes that theV,,, 90 0.90 0.50 of the EGF receptorkinasefor LCzo is 5.3 pmol/min/mg receptor. The K,,, and V,,, values of these enzymes for LCzo TABLEI1 are not the same, perhapsbecause different sites onLCzoare Comparison of rates of incorporation of 32Pinto proteins inthe being phosphorylated by myosin light chain kinase and EGF presence of A431 membranes and EGF receptor kinase. All reactions were for 10 min, 30 “C. The concentration of each The M , = 20,000 lightchain is phosphorylated in two protein was 10 PM. The specific activity of the ATP was 430 cpm/ tyrosine residues, tyrosine 142 and tyrosine 155. There are pmol. Determination of 32Pincorporation was as described for myosin three aspartic acids and two glutamic acids in a sequence of light chain under “Experimental Procedures.” seven amino acids NHz-terminal to tyrosine 142. This site Protein pmol/min resembles other tyrosine phosphorylation sites found in tyLC20 67.0 rosine kinases (26-29) and other proteins (29), which have Histone f2b 26.0 one to three acidic amino acids immediately adjacent to the Casein 18.9 tyrosine on its NHz-terminalside. The other phosphorylated Histone f l 14.9 tyrosine, residue 155, has no acidic amino acids on its NHzterminal side; instead, thesequence is Gly-Asn-Phe-Asn-Tyr. The initial rate of phosphorylation of LCzowas compared Still, the initial rate of phosphorylation of this site is only with those of several other proteins (Table 11). Proteins were three times less than that of tyrosine 142. It is clear that incubated a t a concentration of 10 PM for 10 min with A431 factors other than the mere presence of acidic residues also membranes in the presence of EGF, and 32Pincorporation contribute to therecognition of sites of tyrosine phosphorylwas determined as describedfor LC2o. The initial rate of ation, since HLA antigens are phosphorylated at the sequence phosphorylation of LCzowas 2.6-5-fold greater than thatfor Lys-Gly-Gly-X-Tyr (30), and tyrosine 68 of LCzo,which has other proteins. Phosphotyrosine was the only phosphoamino two acidic residues preceding it, is notphosphorylated. acid found in any of these proteins (data not shown). The Tyrosine phosphorylationof LCzohas not been reported to following proteinsat 1 mg/ml were incubatedwith A431 occur in intact cells, but there are several reasons why this membranes and EGF and were not phosphorylated phospho- modification might have been missed. Hunter and his colrylase, a and (3 tropomyosin, enolase, L-lactic dehydrogenase, leagues (1, 21, 31) have carried out comprehensive searches and phosphoglycerate mutase. The last three proteins have for proteins phosphorylated in tyrosines in A431 and other been shown to be phosphorylated a t tyrosine residues in Rous transformed cells. This technique involves fractionation of sarcoma virus transformed cells (21). These experiments show labeled cellular proteins by isoelectric focusing from pH 6.0that, among the protein substrates tested under the conditions 8.0 followed by SDS-gel electrophoresis. LCzo has a PI = 5.0 (12) and itwould not havebeen detected in these experiments. of assay, LCzo is the best substrate for the EGF receptorassociated tyrosineproteinkinaseactivity of A431 mem- While the serine phosphorylation ofLC20 has been studied branes. extensively in tissues (12), these experimentsmay have been carriedoutin cells with low tyrosinekinase activity. In DISCUSSION addition,quantitation of phosphorylation of LC20 isfrequently carried out after extraction of the protein in buffers We have shown that a tyrosine protein kinase present in A431 membranes phosphorylatestwo tyrosine residues in the containing F- and EDTA, (32) inhibitors of phosphoserine M , = 20,000 smooth muscle myosin light chain. The extensive phosphatase activity. These two reagents activate phosphostimulation of this reactionby EGF indicates that the respon-tyrosine protein phosphatase (13). Chinkers et al. (9) have shown that within 20 min after sible kinase is the EGF receptor-associated tyrosine kinase. To our knowledge, LCzois the first substrate demonstrated toaddition of EGF, A431 cells show surface ruffles, rounding be phosphorylated stoichiometricallybya tyrosine kinase. up, and protrusionof nuclei from the planeof the cell monolayer. They speculated that these changes could be due to Myosin light chain was phosphorylated a t a greater initial rate than a series of substrates examined (Table 11). That cytoskeletal alterations regulated by myosinphosphorylation. LC2ois a favored substrate for the tyrosine kinase in A431 These morphological alterations correlatewell with the findmembranes is illustratedby a comparison of literature values ing that additionof EGF to A431 cells increases the cellular for the kinetic constants K , + VmaX.The apparent K,,, and level of phosphotyrosine 3-fold after 15 min and 4-fold after V,,, of the kinase in A431 membranes in thepresence of EGF 60 min (1).It is therefore possible that the phosphorylation for LCzOare 73 p~ and 17 nmol/min/mg, respectively. This of myosin ontyrosineratherthanserinemediatesthese can be compared toK , and V,,, values of the kinase in A431 cytoskeletal alterations.

Tyrosine Phosphorylation of Myosin Light Chain To date, experiments to demonstrate the phosphorylation of intact myosin by the EGF-stimulated kinaseof A431 membranes have not yielded conclusive results, possibly because there are contaminating phosphatases and myosin light chain kinase in the myosin preparation. Further work on this important question, which bears on the physiological significance of the tyrosine phosphorylation of the LCzo, is being undertaken. We are also determining whether LCzo is phosphorylated at tyrosine residues in cells in response to mitogenic hormones or after infection by Rous sarcoma virus. Acknowledgments-We thank Drs. Don Blumenthaland Linda Pike for many helpful discussions. We thank Brad McMullen and Edwin Beckman for excellent technical assistance. REFERENCES

12. Stull, J . T. (1980) Adu. Cyclic Nucleotide Res. 13,39-93 13. Brautigan, D. L., Bornstein, P., and Gallis, B. (1981) J. Biol. Chem. 256,6519-6522 14. Sobieszek, A. (1977) The Biochemistry of Smooth Muscle (Stephens, N. L., ed.) pp. 11413-11443, University Park Press, Baltimore 15. Holt, J. C., and Lowey, S. (1975) Biochemistry 1 5 , 4600-4609 16. Lowry, 0. H., Rosebrough, N. J., Farr, A.L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265-275 17. Peterson, G. L. (1977) Anal. Biochem. 83, 346-356 18. Hashimoto, E., Takio, K., and Krebs, E. G. (1981) J. Bwl. Chem. 256,5604-5607 19. Swarup, G., Cohen, S., and Garbers, D.L. (1982) Biochem. Biophys. Res. Commun. 107, 1104-1109 20. Maita, T., Chen, J.-I., and Matsuda, G. (1981) Eur. J. Biochem. 117,417-424 21. Cooper, J. A., Reiss, N. A., Schwartz, R. J., and Hunter, T.(1983) Nature (Lond.) 3 0 2 , 218-223 22. Baldwin, G. S., Stanley, I. J., and Nice, E. C. (1983) FEBS Lett. 153,257-261 23. Adelstein, R. S., andKlee, C . B. (1981)J . Biol. Chem. 256,75017509. 24. Stull, J. T., Blumenthal, D. K., Miller, J. R., and DiSalvo, J. (1982) J. Mol. Cell. Cardiol. 1 4 , Suppl 3, 105-110 25. Carpenter, G., King, L., Jr., and Cohen, S. (1979) J . Biol. Chem. 254,4884-4891 26. Casnellie, J. E., Harrison, M. L., Hellstrom, K. E., and Krebs, E. G. (1982) J. Biol. Chem. 257, 13877-13879 27. Smart, J. E., Oppermann, H., Czernilofsky, A. P., Purchio, A. F., Erikson, R. L., and Bishop, J. M. (1981) Proc. Natl. Acad. Sci. U. S. A. 78,6013-6017 28. Neil, J. C., Ghysdael, J., Vogt, P. K., and Smart, J. E. (1981) Nature (Lond.)2 9 1 , 675-677 29. Patchinsky, T., Hunter, T., Esch, F. S., Cooper, J. A,, and Sefton, B. M. (1982) Proc. Natl. Acad. Sci. U. S. A. 79,973-977 30. Guild, B. C., Erikson, R. L., and Strominger, J. L. (1983) Proc. Natl. Acad. Sci. U. S. A. 80, 2894-2898 31. Cooper, J. A., and Hunter, T. (1981) J. Cell Biol. 91, 878-883 32. Silver, P. J., and Stull, J. T. (1982) J. Biol. Chem. 257, 61376144


1. Hunter, T., and Cooper, J. A. (1981) Cell 2 4 , 741-752 2. Pike, L. J., Gallis, B., Casnellie, J. E., Bornstein, P., and Krebs, E. G. (1982) Proc. Natl. Acad. Sci. U. S. A. 79, 1443-1447 3. Baldwin, G. S., Burgess, A. W., and Kemp, B. E. (1982) Biochem. Biophys. Res. Commun. 109,656-663 4. Baldwin, G. S., Knesel, J., and Monckton, J. M. (1983) Nature (Lond.) 301,435-437 5. Swarup, G., Cohen, S., and Garbers, D. L. (1981) J. Biol. Chem. 256,8197-8201 6. Cohen, S., Ushiro, H., Stoscheck, C., and Chinkers, M. (1982) J . Biol. Chem. 257, 1523-1531 7. Buhrow, S. A,, Cohen, S., and Staros, J. V. (1982) J . Biol. Chem. 257,4019-4022 8. Carpenter, G., and Cohen, S. (1979) Annu. Rev. Biochem. 4 8 , 193-216 9. Chinkers, M., McKanna, J . A., and Cohen, S. (1979) J. Cell Biol. 83,260-265 10. Carpenter, G., and Cohen, S. (1976) J . Cell. Physiol. 88, 227-238 11. Kawamoto, T., Sato, J. D., Le, A., Polikoff, J., Sato, G. H., and Mendelsohn, J. (1983) Proc. Natl. Acad. Sci. U. S. A. 80, 13371341

13093