transduced by IL-1 receptors remains poorly understood and controversial (for ... of the IL-1-activated kinase(s) with the newly described family of ERKs (40) is.
T H E JOURNAL Of BIOLOGICAL CHEMISTRY
Vol. 266, No. 33, Issue of November 25, pp. 22661-22670,1991 Printed in U.S.A.
$’ 1991 by The American Society for Biochemistry and Molecular Biology, Inc.
Interleukin-1 Represents a New Modality for the Activation of Extracellular Signal-regulated Kinases/Microtubule-associated Protein-2 Kinases* (Received for publication, March 28, 1991)
Timothy A. Bird$§, Paul R. Sleath8, Paul C. deRoos8, Steven K. Dower$, and G . Duke Vircall From the Departments of $Biochemistry and VProtein Chemistry, Immunex Corporation, Seattle, Washington98101
In this study we describe the activation of a protein Interleukin-1(IL-1)’is a potentimmunoregulatoryand kinase which phosphorylates a peptide, T669, com- proinflammatory cytokine secreted by a variety of cells in prising amino acids663-681 of the epidermal growth response to infection, activated lymphocyte products, microfactor receptor and containing the phosphate acceptorbial toxins, inflammatory and other stimuli (1).At present, site Pro-Leu-Thraee-Pro. In the human epidermoid the carIL-1 system is known to comprise two distantly related cinoma cell lineKB, T669 kinase activity incytosolic polypeptide agonistligands N and /3 (2, 3)anantagonist extracts peaked (up to 15-fold compared with basal polypeptide, IL-Ira (4,5) an 80-kilodalton single-chain receplevels) 15-30 min after addition of interleukin-1 (IL- tor present on T cells and connective tissue-derived cells (6, 1) and closely paralleled receptor occupancy with a 7), and an incompletely characterized 67-kilodalton receptor half-maximally effective concentrationof -100 p M IL- expressed by B cells and monocytes (8-11). Despite vigorous l a . IL-1 treatment elevated T669 kinase activity to a efforts over the last few years, the nature of the signal(s) variable extent in selected fibroblast lines, the hepatransduced by IL-1 receptors remainspoorly understood and toma cell line HepG2, and the murine thymoma EL4 controversial (forcurrent reviews seeRefs. 12-15). Some 6.1. An IL-1 receptor-negative EL4 variant and the B laboratories have proposed that cAMP is elevated in some cell lines70213, CB23, and RPMI 1788 did not respond (14-16) andthatconsequent in this way. All of the cell lines except 70213 showed cells afterexposuretoIL-1 increased levelsof T669 kinase when treat.ed with theactivation of protein kinaseA is a key step in the transcription of IL-1-responsive genes (17) whereas others have not deprotein kinase C activator phorbol myristate acetate induced tected such changes(18-21) or have reported that the and/or with epidermal growth factor. This finding is in agreement witha previous study (Countaway,J. L., level of cAMP is not sufficient to trigger a cellular response Northwood, I. C., and Davis, R. J. (1989) J. Biol. (22). Activation of protein kinase C has been suggested as an Chem. 264, 10828-10835). Activators of protein ki- important even in IL-1 signaling (23-25), but this view has nase A did not mimic the ability of IL-1 to stimulate also been challenged (26-31). Other reports suggest that both T669 kinase activity, nor did the protein kinase C pathways may be operative (32, 33). Despite these contradicinhibitor staurosporine abrogate the effect of IL-1. tory reports it is clear that IL-1 does stimulate protein serine/ T669 kinase activity from IL-1-stimulated KB cells threonine kinase activity in a wide variety of cell types, and was partially purified by ion exchange, hydrophobic some of the substrates for IL-1-dependent phosphorylation interaction, and size exclusion chromatography. The have recently been identified. IL-1 caused the phosphorylapartially purified enzyme phosphorylated myelin basic tion of a cytosolic 65-kDa protein in peripheral blood monoprotein, a characteristic substrate of microtubule-as- nuclear cells (21) now identified as L-plastin (34), the27-kDa sociated protein-2 kinase (MAP-2 kinase) and pepthe tide Arg-Arg-Arg-(Tyr-Ser-Pro-Thr-Ser-Pro-Ser),heat-shock protein (HSP27) in fibroblasts and HepG2 cells from RNA polymerase 11. Western blotting of chro- (35), EGF receptor in fibroblasts and KB cells (30, 31), 1920-kDa cytosolic proteins (possibly isoformsof stathmin (36)) matographic fractions revealed that T669 kinase acin AtT-20 pituitary cells (37), the 80-kDa protein kinase C tivity corresponded with two proteins of 43 and 45 kilodaltons which cross-reacted with antibodies raisedsubstrate in theD1O.A T helper cell line (32), the 80-kDa ILagainst peptide sequences of rat extracellular signal- 1 receptor (38) transfected into Chinese hamster ovary cells, regulated kinase- l/microtubule-associated protein-2 and talin in fibroblasts (39). develop cell-free systems in which kinase. T669 kinase activity was critically dependent At present, our aim is to be on the presence of phosphatase inhibitors. Since both thecharacteristics of theIL-1-activatedkinase(s)can studied and compared with t.hose of known protein kinases. the43-and45-kDaproteins,immunoprecipitated from [”2P]phosphate-labeled cells, demonstrated a dra- One such system, making use of a proline-rich peptide submatic increase in their levels of serine, threonine, and strate, is described in the present paper.Using this approach tyrosine phosphorylation after brief treatment with we have demonstrated the rapid IL-1-mediated activationof IL-1, we conclude that IL-1 modulates the activity of theseextracellularsignal-regulatedkinase/microtu’ The abbreviations used are: IL, interleukin; ERK, extracellular bule-associated protein-2 kinases by altering the level signal-regulated kinase; CTD4, Arg-Arg-Arg-(Tyr-Ser-Pro-Thr-Serof their phosphorylation. Pro-Ser),; EGF, epidermal growth factor;HEPES,4-(2-hydroxy* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. § To whom correspondence should be sent: Dept. of Biochemistry, Immunex Corporation, 51 University St., Seattle, WA 98101,
ethyl)-1-piperazineethanesulfonic acid; IL-lRa, interleukin-1 receptor antagonist; MAP-2, microtubule-associated protein-2; PMA, 4Dphorbol12-myristate13-acetate;ST-1, Arg-Arg-Arg-Glu-Leu-ValGlu-Pro-Leu-Thr-Pro-Ser-Gly-Glu; T669, Glu-Leu-Val-Glu-ProLeu-Thr-Pro-Ser-Gly-Glu-Ala-Pro-Asn-Cln-Ala-Leu-Leu-Arg; PBS, phosphate-buffered saline; SDS,sodium dodecyl sulfate; PAGE,polyacrylamide gel electrophoresis.
22661
22662
IL-1 SerinelThreonine Kinase -activated
Seger (University of Washington, Seattle). All other immunochemicals used for Western blotting were obtained from Sigma. Stimulation of Cells and Measurementof Kinase Activity in VitroConfluentadherent cells in six-well clustersor10-cmdishes or aliquots of approximately lo7 nonadherent cells in 1 mlof growth medium were stimulated with EGF, IL-1, or PMA (50, 20, and 100 ng/ml, respectively, unless otherwise stated) by the direct additionof these agents to the spent culture medium. Additions were made in PBS (EGF, IL-1) or dimethyl sulfoxide (PMA) from concentrated stock solutions such that the final concentration of vehicle did not exceed 0.1%. Vehicle wasadded to control cultures when appropriate. EXPERIMENTAL PROCEDURES Cells were returned to a 37 "C incubator for varying times (typically Materials-Human recombinant IL-la (referred to throughout this 15 min) asdescribed in the text. Nonadherent cells were sedimented paperas"IL-1") was expressedin Escherichia coli, purified, and in a benchtop microcentrifuge (Hoeffer Nanofuge,Hoeffer Instruradiolabeled ( 3 X 10'" cpm/mmol) as described previously (38).Pep- ments, San Francisco,CA) for 30 s, washed twice with ice-cold Ca'+, tides T669 (Glu-Leu-Val-Glu-Pro-Leu-Thr-Pro-Ser-Gly-Glu-AlaM F - f r e e P B S (0.05 M sodium phosphate, pH 7.2, 0.15 M sodium Pro-Asn-Gln-Ala-Leu-Leu-Arg)ST-1 (Arg-Arg-Arg-Glu-Leu-Val- chloride), then suspended in0.5 ml of cold extraction buffer (20 mM Glu-Pro-Leu-Thr-Pro-Ser-Gly-Glu), and CTD4 (Arg-Arg-Arg-(Tyr- HEPES, pH 7.4, 30 mM p-nitrophenyl phosphate, 10 mM NaF, 10 Ser-Pro-Thr-Ser-Pro-Ser),) were synthesized on an Applied BiosysmM MgC12, 2 mM EDTA, 5 mM dithiothreitol, 0.1 mM NaaV04,0.1 tems 430A peptide synthesizer using phenylacetamidomethyl resins mM Na2Mo04, 10 mM sodium 8-glycerophosphate,1 mM phenylmethand t-butoxycarbonyl chemistry. TheN-terminalbutoxycarbonyl ylsulfonyl fluoride, 10 p~ leupeptin, 10 p~ pepstatin A). After stimgroup was removedwith trifluoroacetic aciddichloromethane (1:l) ulation of adherent cells, the plates were aspirated and rinsed twice prior to hydrofluoric acid cleavage. Peptides were cleaved from the with cold PBS. The cell layers were then scraped into extraction resin with HF (1h, 0 "C) using anisole as scavenger and were purified buffer (2ml/dishor 0.5ml/well).After thisstep,adherentand on a Vydac Clx column using a 0-60% acetonitrile gradient in 0.1% nonadherent lines were treated in the same manner; the cells were trifluoroaceticacid. The identity and purity of each peptide were disrupted by forcing the suspension 20 timesthrough a 22-gauge confirmed by high performance liquid chromatography, amino acid hypodermic needle. Pilot experiments showed that T669 kinase activanalysis, and W f mass spectrometry using a Bio-Ion 20 mass spec- ity was recovered in a 100,000 x g supernatant, hut identical activity trometer. The casein kinase I1 substrate peptide Arg-Arg-Arg-Glu- was recovered in supernatants obtained after microcentrifugation of Glu-Glu-Thr-Glu-Glu-Glu was purchased from GibcoBRL Life Tech- the extracts at 13,000 rpm for 30 min a t 4 "C. The microcentrifuge nologies. P-81 phosphocellulose paper was obtained from Whatman. method was used routinely to process small samples. In some experiments, certain phosphatase inhibitorswere omitted from the extracStaurosporine, H7, and H8 were obtained from Calbiochem; [y-'"PI ATP, [y-'"PIGTP, and'"1-protein A were purchased from Amersham tion buffer; in this case those inhibitors (or diluents) were added to the microcentrifuged samples prior to assay of T669 kinase activity Corp.EGF(receptorgrade),48-phorbol12-myristate13-acetate of each. Samples (PMA), heat-stable protein kinase A inhibitor (Walsh inhibitor),W7, such that all the extracts contained the same amount could be stored in liquid N, for a t least 3 months without suffering (Arg),, (GIU),~,(Lys),, 8-glycerophosphate,para-nitrophenylphosphate, ATP, GTP, phosphoamino acids, 100-pm cellulose TLC plates, any apparent loss of enzyme activity. In uitro measurement of kinase activity using T669 as substrate proteinaseinhibitors,rabbit myelinbasic protein,histones,and HEPES were from Sigma. Sodium orthovanadate and electrophoresis was based on the method of Countaway et al. (42) with extensive modifications: 5 pl of cell extract was mixed with 15 pl of a reaction solvents were obtained from Aldrich. mM MgC12,T669 Cell Lines and Antibodies-KB epidermoid carcinoma cells (ATCC mixture which contained 20 mM HEPES, pH 7.4,lO CCL17), the human lymphoblastoid line RPMI 1788 (ATCC peptide (125 p ~ - 2 . 5mM, 730 p M in the standard assay), and 33 p M [y-'"PIATP or [y-:"P]GTP (2 pCi/nmol); other additions were made CCL156), 3T3-Ll murine preadipocyte (ATCC CCL92.1), and the fibroblast lines WI-38 (ATCC CCL75) and MRC-5 (ATCC CCL171) as indicated in the text. Blank incubations contained water inplace were originally obtained from the American Type Culture Collection of T669 peptide. The mixture was incubated for 20 min a t 30 "C (or (Rockville, MD) and were passaged extensively in our facility prior for 10 minfor determination of kinetic constants), duringwhich time to the commencement of the present studies. HepG2 human hepato- the rate of product formation was linear. Reactions were stopped by addition of 4 p1 of 88% formic acid, and the tubes were microcentricellular carcinoma cells were obtained from Dr. J. Bauer (Medicinfuged for 30 min at 4 "C. Aliquots (4-6 p1) of the supernatants were ische Universitaetsklinik, Freiburg, Germany), 70Z/3 murine pre-B spotted onto cellulose TLC plates along a central origin line (up to cells (10) were a gift from Dr. K. Bomsztyk (Dept. of Nephrology, 12 samples/20- X 20-cm plate). The plates were electrophoresed (in University of Washington, Seattle). Murine thymoma EL4 6.1 and an HTLE-7000 apparatus, C.B.S. Scientific, Del Mar, CA) in a pH the IL-1-R negative EL4:'+ lines were kindly provided by Dr. H. R. MacDonald (Ludwig Institute for Cancer Research, Lausanne, Swit- 3.5 buffer system (pyridine:acetic acidwater, 10:100:1890) at 1.4 kV y-"'P-nucleoside zerland). CB23, an Epstein-Barr virus transformed lymphocyte line for 1 h. Undertheseconditions,unincorporated (41) was provided by Dr. D. Benjamin (Division of Hematology and triphosphates (and alllabeled byproducts/impurities) migrate toward Oncology, Ohio State University Hospitals, Columbus), and primary theanode,andthephosphorylatedpeptidemigratestowardthe human gingival fibroblasts were the kindgift of Dr. E. E. Qwarnstrom cathode. The plates were cut along the origin line and the cathode halves dried and briefly exposed to x-ray film to locate the phospho(Dept. of Pathology, University of Washington, Seattle). RMPI 1788, CB20, and 70Z/3 were maintained in RPMI 1640 peptide spotswhich were excised and quantitatedby Cerenkov countmedium supplemented with 10% fetal bovine serum (Intergen Co., ing. Incorporation of "PO, into endogenouscellular protein COPurchase, NY), 50 units/ml penicillin, 50 pg/ml streptomycin sulfate, migrating with the phosphopeptide was negligible, but these counts were routinely subtracted.Netcounts were convertedinto molar and 300 pg/ml L-glutamine in a humidified atmosphere of air:CO, quantities by reference to the count rate given by a known amount (95:s). The other cells were maintained in Dulbecco's modified medium, with supplements as described above, in a humidified atmos- of labeled ATP spotted onto an appropriatelysized piece of cellulose phere of air enriched with 10% CO,. Five days prior touse, adherent TLC plate. As an alternative assay, we employed the peptide ST-1 cells were subculturedinto 10-cm Petridishesor six-well tissue which is T669 modified with a triplet of N-terminal arginineresidues. describedabove, withthe culture clusters (Costar, Cambridge, MA) and were not refed during Reactionconditions were thesameas this period. Nonadherent cells were used 3 days after feeding and substitution of ST-1 for T669. After centrifugation, 20-pl aliquots of the reaction mixtureswere spotted onto 2.5-cm-diametercircles of Pwereharvested by centrifugation (400 X g, 10 min) a t 15 "C and resuspended in an appropriate volume of the original spent medium. 81 phosphocellulose paper which were washed in four changes of 75 The neutralizing monoclonal antibodies M1 (mouse IgG1) and M4 mM H:,PO, (5 min/wash, 10-12 ml/paper), air-dried, and Cerenkov (rat IgGZb), raised against recombinant forms of the human 80-kDa counted. Phosphorylation of peptide CTD4, histones, myelin basic JL-1 receptor, have beendescribedpreviously (11).Mouse monoclonal protein, and casein kinase I1 peptide was also measured using the phosphocellulose paper binding method. anti-MAP-2 kinase (raised against a peptide conjugate containing Partial Purification of 7'669 Kinase Actiuity, Western Blotting, and amino acids 325-345 of the MAP-2 kinase/ERK-1 sequence) was cell extract was thawed, and 40 ml purchased from Zymed Laboratories Inc. (South San Francisco,CA). Immunoprecipitation-Prepared A rabbit antiserum raised against a synthetic peptide encompassing from 2.1 x 10" (IL-2-treated or untreated) KB cells was immediately loaded onto a DEAE-Sephacel (Pharmacia LKB Biotechnology Inc.) residues 305-327 of rat ERK-1 (40) was the kind gift of Dr. Rony
protein serine/threonine kinase activity with substrate specificity similar to thatof extracellular signal-regulated kinases (ERKs)/MAP-2 kinases. The identity of the IL-1-activated kinase(s) with the newly described family of ERKs (40) is supported by immunological and chromatographic parameters. As has been shown for otherextracellularsignaling molecules, IL-1 appears to activate these kinases by causing their phosphorylation.
SerineelThreonine IL-1Kinase -activated column (2.2 X 5 cm) equilibrated with buffer A (25 mM HEPES, pH 7.4, 2 mM EDTA, 1 mM dithiothreitol, 25 mM NaCI, 40 mM pglycerophosphate, 1 mM phenylmethylsulfonyl fluoride, 10 g~ pepstatin, 10 g~ leupeptin, and 10 mM para-nitrophenyl phosphate) at a flow rate of 1.5 ml/min. This, and all chromatographic steps were performed at 4 "C. After loading, the column was washed with 5 column volumes of buffer A and then eluted with 200 ml of buffer A containing 350 mM NaC1. The bulk eluate was immediately loaded onto a phenyl-Sepharose (Pharmacia) column (2.2 X 5 cm) by connecting the columns in series. The phenyl-Sepharose column had been equilibrated previously with buffer A containing 250 mM NaCl and 20% ethylene glycol. After loading, the column was washed with 6 column volumes of equilibration buffer. T669 activity was eluted with a 200-ml linear gradient in buffer A from 250 mM NaCl containing 20% ethylene glycol to 25 mM NaCl containing 80% ethylene glycol. This was followed by an additional 50 mlof the end buffer (buffer A containing 25 mM NaCl and 80% ethylene glycol). Fourmilliliter fractions were collected. Fractions containing T669 kinase activity were pooled (total volume, 72 ml) and concentrated to a final volume of -2 ml using Centriprep 10 and Centricon 10 concentrators (Amicon). Five hundred microliters of this concentrated phenylSepharose poolwas chromatographed on a Superose 12, HR10/30 column (Pharmacia) equilibrated with buffer A containing 20% ethylene glycol at a flow rate of0.2 ml/min; 0.5-1111 fractions were collected. Aliquots were removed from every fraction for assay, Western blotting, and immunoprecipitation, and the remainder of each fraction was immediately frozen with dry ice/methanol and stored at -80 "C. Aliquots of unfractionated KB or HepG2 cell extracts (approximately 45 gg of protein) or Superose 12 fractions were mixed with a concentratedSDS-PAGE sample buffer (43) and electrophoresed under reducing conditions in either 12.5% gels or 8-16% polyacrylamide gradient gels. The electrophoresed proteins were transferred t o nitrocellulose, blocked overnight with bovine serum albumin in PBS, and then incubated with anti-MAP-2 kinase monoclonal antibody(1:2,000) or rabbit polyclonal antiserum (1:1,000).Immune complexes were visualized using a combination of biotinylated goat anti-mouse immunoglobulins and streptavidin-conjugated horseradish peroxidase, horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin, or with "'I-protein A. For immunoprecipitations, dishes of KB cells were rinsed and preincubated for 30 min with phosphate and serum-free minimal essential medium. The medium was replaced with fresh phosphate-free medium supplemented with 0.5-1 mCi/ml [:lSP]orthophosphate,and incubation was continued for 2 h; during the last 15 min, IL-1 or vehicle was added to the dishes. The cell layers were processed and immunoprecipitated exactly as described by Boulton et al. (40). Immune complexes, containing radiolabeled ERKs, were visualized by SDS-PAGE and autoradiography. Phosphoproteins were excised and eluted from the dried gels, partially hydrolyzed in 6 N HC1, and subjected to two-dimensional phosphoamino acid analysis (44). '"P-Phospboarnino acids were visualized and quantitated using Kodak storage phosphor screens and a Phosphorimager (Molecular Dynamics Inc., Sunnyvale, CA). Radioligand Binding Assay-Equilibrium binding of "'I-IL-1 to KB cells after detachment from culture dishes with 5 mM EDTA in PBS was carried out and analyzed as described previously for EL4 and COS cells (6). To determine the amount of IL-1 bound by the cells in shorter periods under nonequilibrium conditions, KB cells were seeded in 12-well plates (Costar) and grown to a density of 9 X 10"cells/well. The plates were rinsed with two changes of RPMI 1640 buffered with 20 mM HEPES, pH 7.4, containing 10% bovine serum albumin (binding medium) and were then incubated for 15 min at 3 7 "C with 0.4 ml/well of binding medium supplemented with 0.0820 ng/ml 12'I-IL-le.For each concentration, replicate wells received an additional 50-fold molar excess of unlabeled IL-1 for the determination of nonspecific binding. After incubation with labeled IL-1 the wellswere rinsed three times with ice-cold binding medium supplemented with 0.1% sodium azide. Bound radioactivity was solubilized by the addition of 1 ml of 0.1 N NaOH, 1%Triton X-100 to each well. Aliquots of theresultant lysate were subjected to ycounting. RESULTS AND DISCUSSION
Demonstration of Increased T669 Kinase Activity in IL-Itreated Cell Extracts-To obtain a convenient substrate with which to assay IL-1-activated protein serine/threonine
22663 kinases, we searched the predicted sequences of known substrate proteins for shared potential protein kinase acceptor sites. We reasonedthat such motifsshould be repeated within HSP27, stathmin) some of the proteins (EGF receptor, talin, or peptide mapping since, by two-dimensional electrophoresis these proteins appear to be coordinately phosphorylated at multiple sites in response to IL-1 (30, 31, 35, 36, 39). One such sequence is X-Ser/Thr-Pro-Y where X is frequently a hydrophobic residue, and Y is often aspartic acidor glutamic acid; Leu-Ser-Pro-Glu occurs in L-plastin (34) and H S P 2 7 (79), stathmin has two Leu-Ser-Pro-Y motifs (36), rat talin contains 10 X-Ser/Thr-Pro-Y motifs (four with hydrophobic residues in t h e X position, and four with Glu in the Y position (45)). The cytoplasmic domainof t h e EGF receptor contains the sequence Pro-Le~-Thr'~~-Pro in which the threonine residue is known to be a major site of phosphorylation in vivo (46). We therefore synthesized the T669 peptide containing this sequence and used it to probe for protein kinase activity in cytoplasmic extracts made from IL-1-stimulated KB cells, which phosphorylate theEGF receptor in response toIL-1 i n vivo (31). An increase in T669 kinase activity was detectable 5 min after treatment of these cells with 20 ng/ml IL-la (Fig. l a ) . Maximal activity, a 12.9-fold increase over basal levels in the experiment shown, occurred between 15 and 30 min; thereafter, there was a rapid decline in activity so that by 2 h i t had returned almost to the original level. This time course was strikingly similar to that described for the phosphorylation of epidermalgrowthfactorreceptorsinintact IL-1treated cells (31). As determined by phosphoamino acid analysis of t h e eluted and hydrolyzed phosphopeptide (not shown), all of the incorporated radioactivity occurred as phosphothreonine, the neighboringSer'" residue was not phosphorylated. The degree of activation of T669 kinase, measured after 15min exposure of KB cells t o IL-1, was dependent on the dose of IL-1 (Fig. lb). In many previous studies,a significant spare receptor effect has been reported for IL-1-dependent systems (see Ref. 47 for a review and discussion). Although showing some spare receptors, the activation of T669 kinase occursat higherfractionalreceptoroccupancy than many IL-1 responses. However, activation of this enzyme(s) is an early event in the IL-1 signal cascade. It is therefore possible that phosphorylation of substrates by this enzyme, a process in which it obviously acts catalytically, could be partly or wholly responsible for the signal amplification implied by the greater spare receptor effect observed when later events (prostaglandin production, IL-2 gene transcription) are measured. To determine the generalityof this response we examined the ability of a range of cell types to increase T669 kinase activity after similar treatment withIL-1 and compared this with the effects of PMA and EGF, agents recently reported to stimulate a similar activity (42). All of the cells listed in Table I, except for EL43+, expressIL-1 receptors although the level of expression varies considerably from cell to cell. Some of these cells increased their levels of cytosolic T669 kinase IL-1 (Table I). T h i s activityafter a 15-minexposureto response occurred in the connective tissue-derived cells and a T-cell line, EL4, all of which express the 80-kDa IL-1 receptor b u t n o t i n a receptor-negative variant of EL4. There was no clear correlation between the level of IL-1 receptor expression and the extent of T669 activation. Interestingly, the cells that did not respond (CB23, RPMI 1788, a n d 702/3) are those which bear the 67-kDa "type11" IL-1R (8-11,41). In the case of HepG2 cells, which responded with large increases in T669
IL-l -activated SerinelThreonine Kinase
22664 a
lines by normalization to the protein contentof the extracts because some of the phosphatase inhibitorsnecessary to preserve kinase activity (see below) interfered in protein assays. All samples were derived from approximately the same number of cells of each type, however, and were analyzed in an identical manner. On thisbasis, KB, WI-38, and MRC-5cells contained the highest unstimulated (0.2-0.4 pmol of phosphate incorporated/min/5-p1 aliquot of extract) and maximally stimulated (4-7 pmol/min) levels of T669kinase; 0 30 60 90 120 time (rnin) HepG2 and EL4 cells contained lower levels (both about0.08 and 0.4 pmol/min for unstimulated andmaximally stimulated levels, respectively). IL-1 signaling ina number of cells has been reported tobe via CAMP-dependent pathways. We therefore tested the effects of the cell-permeable cAMP analogs 8-bromo cAMP and dibutyryl CAMP, and of forskolin, a direct activator of adenylate cyclase. None of these agentswas able to elevate T669 kinase activity (Fig. 2), indicating the lack of involvement of protein kinase A in the activation process. To confirm that the cytokine, and nota contaminant in our IL-1 preparation, was responsible forthe observedeffects, we preincubated IL1 (log" Moles/liter) HepG2 cells with a mixture of two monoclonal antibodies FK. 1. Time- and dose-dependent activation of peptide T669-phosphorylating activity inKB cells. a, confluent mono- which block binding of IL-1 to the human 80-kDa receptor (Fig. 3a). This treatment completely abrogated the effects of layers of KB cells were stimulated by the direct additionof 20 ng/ml (1.14 nM) IL-lct to the culture medium. After further incubation at IL-1 (Fig. 3b). It is noteworthy that at a level of antibody 3 7 "C for theindicatedtimes,the monolayers were washed, and which completely blocked the activation of T669 kinase,only cytoplasmic extracts were prepared asdescribed under "Experimental 80% of the specific binding was inhibited. Since the antibodies Procedures." Five microliters of the extracts were incubated in the do not cross-react with the 67-kDa type I1 receptors expressed presence of [r-:'ZP]ATP and the peptide substrate T669 (see "Experby HepG2 cells,3 the observed residual binding is presumably imentalProcedures"). Radioactivity incorporatedintothepeptide was separated from free labeled ATP by thin layer electrophoresis. to these receptors, and so our results would tend to support cells at least, T669 kinase activation Phosphopeptide spotswere excised from the electrophoretograms and the notion that in HepG2 quantitated by Cerenkov counting. Each point represents the mean is mediated solely through type I receptors. Without blocking activity of three extracts. Bars give the standard error of the mean antibodies to the type I1 receptor, however, we cannot rule where this quantity exceeded the symbol height. b, KB cells were out the possibility that the functional receptor on thesecells stimulated with increasing concentrations of IL-1 for 15 mina t 37 "C. contains both polypeptide chains. T669 kinase activitywas measured as described above (open squares, Effects of Kinase and Phosphatase Inhibitors in Vitro and mean k S.E. of three separate determinations). Ina separate experii n Viuo-In theexperiments described above, cytoplasmic ment, triplicate tissue culture wells of cells (9 X lo5cells/well) were of a mixture of potent rinsed with binding medium and then incubated at 37 "C for 15 min extracts were prepared in the presence with '"I-IL-ln at the same concentrations used to determine kinase phosphatase inhibitors. When these inhibitors were omitted activity. Afterwashingaway unbound radioligand, cell-associated from the extraction medium,a brief exposure of the resulting radioactivity was recovered by lysis of the cells with NaOH/Triton extract to near ambient (25 "C) temperature resulted ina X-100. Nonspecific binding(measuredinthepresence of excess substantial loss of activity (Fig. 4). Both basal and stimulated unlabeled IL-1) was subtracted for each radioligand concentration. The maximum mean bindingwas 1,578cpm/well, and the correspond- activities were lost, with the ratio remaining approximately of a phosphatase ing nonspecific binding was 382 cpm. The total number of receptors constant. With each subsequent addition occupied under equilibrium conditions (4 h at 4 "C) was determined inhibitor, greater incrementsof kinase activitywere retained. by Scatchard analysis. Results (filled circles) are expressed as the This phenomenon has been reported for a number of other calculated percentage of the total receptors that, on average, would cytosolic protein kinases including ribosomal protein S6 kibe occupied at each radioligand concentration. nase (50), insulin-dependent serine kinase (51), and microtubule-associated protein-2 (MAP-2)- kinase (52);a reasonakinaseactivity,the two forms of IL-1 receptor are co-ex- ble explanation would be that phosphorylationof the kinases pressed (48).2 themselvesis required for theiractivation.MAP-2kinase In agreement with a previous study (42), T669-phosphoryl- activity is dependent upon phosphorylationof threonine and ating activity was elevated after treatment with PMA (and tyrosine residues (53), and this enzyme in turn phosphorylates EGF in those lines bearing EGF receptors) in all the cells on serine/threonine residues and thereby activates a 92-kDa tested except702/3. Thelatterare known to be ableto ribosomal protein S6 kinaseI1 from Xenopus (54, but not the activate protein kinase C in response to PMA (10) or IL-1 fibroblast 70-kDa S6 kinase (55)) i n uitro. The serine/threo(27), so their failure to respond in the present study could nine kinase proto-oncogene product Raf-1 can also be actireflect a lack of the T669 kinase(s). Nevertheless, thesecells vated by phosphorylation on serine and/or tyrosine residues will transcribe the K-light chain gene and hence express sur- (56, 57). face IgM in response to IL-1 (49). Thedegree of stimulation Since protein kinase C might be the intermediate enzyme obtained with the three agents in any particular cell line responsible for the putative phosphorylation of T669 kinase showed no correlation; EGF was as potent as IL-1 in some we compared the effect of the protein kinase inhibitor staucell types but less potent in others. It was not possible to rosporine on the activation of T669 in KB cells by PMA, compare the actuallevels of kinase activity in the various cell EGF, and IL-1 (Table 11). Ata concentration of100 nM, 1
' T. A. Bird, P. R. Sleath, P. C. DeRoos, S. K. Dower, and G. D. Virca, unpublished observations.
K. Schooley, J. E. Sims, J. Slack, and S. K. Dower, unpublished observations.
22665
IL-1 -activated SerinelThreonine Kinase TABLEI Activation of T669 kinase activity in variouscell lines Cell lines were maintained as described under “Experimental Procedures” and stimulated for 15 min by the direct addition of the indicated agents to the culture medium. Control cultures received vehicle alone. Cytoplasmic extracts were prepared as described (approximately 2 X 10’ cells/ml of extraction buffer) and assayed for T669 phosphorylating activity in the presence of 730 PM T669 and 25 PM [-y-:”P]ATP. Three separate cultures were assayed under each condition. The results are expressed foldasstimulation, relative to untreated cultures. Numbers in parentheses refer to the numberof independent experiments carried out. Entries without parentheses represent single experiments. ND, not determined. Stimulus type Cell
line
activity
T669 kinase
IL-1 (20 ng/ml)
Cell
KB HepG2 WI-38 fibroblast MRC-5 EL4 6.1 EL4”’ mutant HGF lymphocyte CB23 3T3-Ll lymphocyte RPMI 1788 70213 Additions
Epidermoid carcinoma Hepatoma Lung fibroblast Lung thymoma Murine IL-1R-ve Gingival fibroblast B Murine Preadipocyte B pre-B Murine cell
KB cells
DMSO
7.2 k 1.3 (14) 5.4 k 0.7 (6) 2.4 4.3 8.2 2.5 -C 0.3 (2) 0.9 2.6 ND 0.81.3 1.6 0.9 activitv detectable No
I
PBS 285 pM ILl
0
1 2 T669 kinase activity (pmols/min)
3
FIG. 2. Activators of cyclic AMP-dependent protein kinase do not mimic the effect of IL-1 to cause elevation of T669 peptide kinase activity. Confluent monolayers of KB cells were incubated with the indicated concentrations of forskolin, dibutyryl CAMP, 8-bromo-CAMP, or IL-1 for 15 min. Control cells received dimethyl sulfoxide ( D M S O ) or PBS vehicle, as appropriate. Cell extracts were preparedandassayed for T669kinaseactivityas described under “Experimental Procedures.” The bars represent the mean activity measured in three separate cultures k S.E.
staurosporine almostcompletely abrogated theeffect of PMA. This resultwas expected since the drug has anICso of 2.7 nM for protein kinase C. Staurosporine also inhibited the effect of EGF by 60%; the reported ICso of staurosporine for the EGF receptor tyrosine kinase is 630 nM (58), more than six times the concentration used here,so it is possible that some of the activation of T669 kinase by EGF is mediated via protein kinase C: a similar conlcusion was reached by Northwood and Davis in a recent study using A431 cells (59). In contrast, IL-1-induced T669 kinase activity was insensitive t o inhibition by staurosporine in vivo. This result lends support to the idea thatIL-1-associatedkinaseactivationis protein kinase C-independent (26, 28-31). The effect of tyrosine kinase inhibitors in vivo was also examined. The isoflavone, genistein, is a specific inhibitor of tyrosine kinases which, at 100 yg/ml, hasbeenreportedtoinhibitEGFmediated tyrosine phosphorylation in A431 cells totally (60). Preincubation with genistein a t 20 pg/ml resulted in detectable inhibition of T669 kinase in cells subsequently treated with PMA or EGF (Table 11). Surprisingly,IL-1-mediated kinase activity was potentiated by this compound. Another tyrosine kinase inhibitor, erbstatin (20 pg/ml), had no effect (not shown). These experimentssuggest that tyrosine kinase activity is not an absolute requirement for activation of the T669 kinase by IL-1.
PMA (100 ng/ml)
6.4 f 0.8 (5) 4.7 3.9 12.6 18.1 9.2 ND 8.2 17.4 1.7
EGF (50 ng/ml)
7.6 f 1 (8) 1.9 14.6 ND ND 4.0 ND
A range of protein kinase inhibitors was tested for their ability to inhibit the in uitro phosphorylation of T669 peptide (Table 111). Most interesting, and in contrast to the in vivo experiments, was the observation that even in extracts from PMA-stimulated cells, staurosporine only inhibited 42% and H7 (ICsofor protein kinase C = 15 p M ) 22% of the T669 kinase activity. This, taken together with the resultsdescribed above, provides strong evidence for the involvementof intermediatekinaseactivities including, butnotrestrictedto, protein kinase C in the activation of T669 kinase. None of the agents tested, with the exception of amiloride at rather high concentrations, were particularly effective inhibitors of IL-1-activated T669 kinase,which is therefore unlikely to be either calcium/calmodulin-dependent kinase (which would be inhibitable by W7), cGMP- or CAMP-dependent kinase (inhibited by H7, H8, Walsh peptide), or an EGF-stimulatable protein kinase from A431 cells whichphosphorylated a variety of protein and peptide substrates and was completely inhibited by micromolar concentrations of polyarginine but not polylysine (61). Sodium fluoride, which has been reported to inhibit MAP-2 kinase (62, 63), is included in our extraction buffer and is present 2.5 at mM in the standard assay mixture; but, as shown in Table 111, at higher concentrations, fluoride markedly inhibits IL-1-stimulated T669 kinase activity. Heparin was found tohave a stimulatory effect upon T669 kinase activity (Table III), aswas reported for MAP-2 kinase/ERK1 (64, 65). Substrate Requirements of T669 Kinase-Further experiments were carried out in a preliminary attempt to identify the kinase activity (or mixture of activities) present in the IL-1-stimulated cell extracts. Like the majority of cytosolic serine/threonine kinases, T669 kinase shows a marked prefonly minimal erence for Mg’+ over Mn’+, the latter supporting activity at low concentrations (Fig. 5). No activity was detected in assays employingcalcium as the onlyexogenous cation (Fig. 5 ) . Using ST-1 as a substrate in the presence of 10 mM MgC12, we have found Ca2+ have to a weakly inhibitory effect on T669 kinase. In the presence of 625 ~ L MCa2+,T669 kinase activitywas still 90%of the controllevel but decreased to 44.5% a t 5 mM Ca’+ (all activities were measured in the presence of 500 phi EDTA). MAP-2 kinasefrom nerve growth factor-treated cells was inhibited by millimolar concentrations of calcium (65); in contrast, another MAP-2 kinasewas reported to be completely inhibited by 2 FM Ca2+ (66). Sur-
IL-1 SerinelThreonine -activated Kinase
22666 a
1
2
3
4
5
FIG. 4. The effect of phosphatase inhibitors on T669 kinase activ-
0
0
1 2 3 4 anti-Ill R Mab's (Fg/rnl)
5
b
1.25 2.5 5 anti-Ill R Mabs (pg/rnl) FIG. 3. Interleukin-1-mediated activation of T669 peptide kinase activity inHepG2 cells is blocked by neutralizing antibodies directed against the 80-kilodalton IL-1 receptor. 0
HepG2 cells were grown to near confluence in six-well tissue culture plates. The growth medium was removed and replaced with DulbecCO'S modified medium supplemented with 10 mM HEPES, pH 7.4, 0.5% fetal bovine serum, and the anti-type IIL-1 receptor monoclonal antibodies (Mab's)M1 andM4, each at theindicated concentrations. Control dishes contained a phosphate-saline vehicle. Incubation of the cells was continued for 2 hat 37 "C. Then, ina, each well received "'I-IL-la (5 ng/ml), added directly in a small volume of binding medium. Incubation was continued for a further 15 min. Some wells received a 50-fold molar excess of unlabeled IL-1 for the determination of nonspecificbinding. The amount of specific binding was determinedas described under"Experimental Procedures." Each point is the mean of duplicate determinations (which did not differ by more than 5% of the mean), expressed as a percentage of the specific counts hound in the absence of antibodies (6,941 cpm/well; nonspecific = 402 cpm/well). In b, triplicate wells of cells were stimulated with nonradioactive IL-1 (5ng/ml, shaded bars) or vehicle (open bars).T669 kinase activity againstST-1 peptide substrate was measured in cytosolic extracts made from the cells as described under "Experimental Procedures."
ity. KB cells were stimulated for 15 min with I L - l a (shaded bars) or vehicle (open bars); cytoplasmic extracts were made in the presence of: 1, no phosphatase inhibitors; 2, 30 mM p-nitrophenyl phosphate; 3, 30 mM p-nitrophenyl phosphate and 10mM 0-glycerophosphate; 4, 30 mM p-nitrophenyl phosphate, 10 mM 0-glycerophosphate, and 10 mM sodium fluoride; 5, 30 mM p-nitrophenyl phosphate, 10 mM 0glycerophosphate, 10 mM sodium fluoride and sodium orthovanadate and sodiummolybdate (each at 0.1 mM). After centrifugation the extracts were all incubated a t 25 "C for30 min, and then phosphatase inhibitors were added to the extracts thatlacked them such that all contained the concentrations listed above in an identical volume. T669 kinase activity, with ST-1 as substrate, is shown as the mean k S.E. of triplicate determinations.
TABLEI1 T h e effect of a serinelthreonine kinase anda tyrosine kinase inhibitor on T669 activation in vivo Confluent KB cells in six-well tissue culture clusterswere treated with the indicated kinase inhibitors or vehicle (dimethyl sulfoxide), added directly to the basalmedium, for 30 min a t 37 "C. The cultures were then treated for an additional 15 min with EGF, PMA,IL-1, or vehicle, and incubation at 37 "C continued.Cytoplasmic extracts were prepared and assayed for T669 kinase activity as described under "Experimental Procedures." Three individual wells were assyed for each condition per experiment.The amountof stimulation of kinase activity is shown,for each inhibitor treatment, asa percentage of the stimulation occurring in the absence of inhibitor. T669 kinase activity Pretreatment Control
IL-1
PMA
EGF
%
100 100 Vehicle 100 147+ 9 81 21 84 k 23 Genistein (20 pg/ml)" 7+5 117+11 58+8 Staurosporine (0.1I ~ M ) ~ Mean range of two independent experiments. Mean S.E. of four independent experiments.
+
* *
loaded onto a DEAE-Sephacel column,washed, and then eluted with a linear gradient of NaC1. IL-1-stimulated T669 kinase activity eluted in a broad series of poorly resolved peaks between100 and 350 mM NaCl (not shown). Batch prisingly, when unfractionated cell extracts were assayed, a t elution with 350 mM NaCl was used for subsequent larger least half as much labeled phosphate was transferred to pep- scale preparations. The elution behavior of DEAE-enriched shown in Fig. 6. The tide T669when [y-"'P]GTP (25 PM) was substituted for ATP T669 kinase on phenyl-Sepharose is singlepeak which majority of the kinase activity appeareda as as the phosphatedonor. During subsequent purification steps was eluted at low ionic strength in the presence of >50% (see below), the ability to use GTP as a substrate was lost. ethylene glycol, suggesting that the kinase isvery hydrophoWe assume that a nucleoside diphosphate kinase which can bic. The chromatographic behavior of T669 kinase on phenyltransfer y-phosphate fromlabeled GTP to ADP is present in Sepharose is similar to that reported previously for MAP-2/ the crude extracts. myelin basic protein kinases (63, 64, 68). The activity in the Identification of T669 Kinase as a MAP-2 Kinase/ERKphenyl-Sepharose peak was concentrated approximately 30T h e inhibition profile of T669 kinase was reminiscent of the fold and applied to a Superose 12column(Fig. 7 a ) . T669 growing family of ERK/MAP-2/myelin basic protein kinases activityelutedas a sharp symmetrical peak at a position that have been reported from various sources. These kinases expected for a protein of 42-45 kDa. At this stage, fractions have characteristic chromatographic properties including ad- were assayed for their ability to phosphorylate myelin basic sorption to anion exchangers, tight binding to phenyl-Seph- protein (Fig. 7a), a substrate for ERK-l/MAP-2 kinase (63, arose, and apparent molecular masses of 41-45 kDa by size 68).Myelinbasic protein was phosphorylated by a kinase exclusion chromatography (63-65, 67, 68). Accordingly, we activitypresentinthesamefractionsastheT669kinase attempted to determine if our T669 kinase exhibited similar activity. Increased myelin basic protein kinase activity was or IL-1- also found in crude extracts madefrom IL-1-stimulated cells chromatographic behavior. Extractsfromcontrol stimulated KB cells (from seven 175-cm2 flasks each) were and correlated well with the increased T669 activity (not
IL-1 -activated Seril aelThreonine Kinase
22667
TABLEI11 Effects of some protein kinase inhibitors on T669 kinase activity in vitro Samples of T669 kinase prepared from KB cells maximally stimulated with IL-1, PMA, EGF,or left unstimulatedwere assayed under standardized conditions using T669 as substrate in the presence or absence of the indicated inhibitors. Resultsare expressed as percentages of the activity obtained inthe absence of inhibitor and represent the means of triplicate determinations. (D (D
k-
%
Staurosporine H7 w7
H8 Amiloride Walsh peptide Erbstatin Herbimicin A Polyarginine Polylysine
0.1 p M
57 79
20 p M 20 p M 20 p M 1.0 mM 20 50 p M
Polyglutamate
1 mg/ml 0.2 mg/ml 1 mg/ml
Sodium fluoride Heparin
32.5 p M 50 d m 1
.._ .-. c
10586 69
71 95 96
80
78 126
83 49
83 50
61 38
80 53
58
106 74
50 p M 40 ccg/ml 0.2 mg/ml 1 mg/ml 0.2 mg/ml
68
111 90 75 71
81 80
90 82 29 151
2.51
controlICa controWMn controllMg
ILlICa ILlIMn ILlIMg
r
10
rnM cation
20
0.1 m
IL-’extract treated treated treatedEGF-
Concentration Inhibitor
0.2 -
30
FIG. 5. The effect of various divalent cations on T669 kinase activity. Cytosolic extracts were prepared from dishes of IL1-treated and unstimulated KB cells. Aliquots of the extracts were incubated with T669 (730 p ~ and ) 25 p M [y-””P]ATPtogether with the indicated concentrations of Ca2+,Mn’+, or M e (added as chlorides). The assay mixtures contained EDTA (present in the original extracts) at a final concentration of 0.5 mM. T669 kinase activity is expressed as the mean of duplicate incubations (+range, where this exceeds the symbol height). shown). The IL-1-stimulated phosphorylationof myelin basic protein occurred predominantly on threonine (not shown). T h e observation that T669 kinase activity from EGF-stimulated 3T3-Llcells co-purifiesthrough several steps with myelin basic protein kinase activity was also recently made by Takishima et al. (69). Myelinbasic protein contains a sequence,Pro-Arg-Thr-Pro(residues 94-97), similar to the T669 phosphate acceptor site; this is a major site of phosphorylation in vivo(70). Another protein that isa promiscuous targetfor protein kinases is the large subunit of eukaryotic RNA polymerase 11, the carboxyl terminus of which contains 52 tandem repeats of a similar consensus sequence Tyr-SerPro-Thr-Ser-Pro-Ser (71). Phosphorylation in this domain appears to be functionally correlated with the transition of the polymerase from initiation to elongation (72), and several groups have demonstrated that this protein or peptides dein rived from the consensus sequence can be phosphorylated vitro by murine cdc2 kinase (73), casein kinase I1 (74), or an uncharacterized template-bound kinase (75). We prepared a 28mer (CTD4) based upon this sequence, as described previously by Cisek and Corden (73) and assayed fractions from
0.0 7 0
10
20
30
40
50
60 70
Fraction Number
FIG. 6. Chromatography of IL- 1-stimulated T669 kinase on phenyl-Sepharose. Cytosolic extracts prepared from seven 175-cm’ flasks (-2.1 X 10* cells) of either IL-1-treated (closed diamonds) or unstimulated (open squares) KBcellswereappliedsequentially to DEAE-Sephacel and phenyl-Sepharose columns as described under “ExperimentalProcedures.”Tenmicroliters of selectedfractions eluted from phenyl-Sepharosewas assayed for T669phosphorylating activityusing ST-1 as substrate (see“Experimental Procedures”). Data are plotted as the mean (duplicate determinations) of the net amount of B2P incorporated into phosphocellulose-bound (P-81paper) ST-1. size exclusion chromatography for the ability to phosphorylateit. As was thecase formyelinbasic protein,CTD4phosphorylatingactivity(notshown) wassuperimposable upon T699 kinase activity (Fig. 7 a ) .The K , for phosphorylation of CTD4 was 490 PM, 690 PM for myelin basic protein, and 750 PM for T669. Material from the Superose 12 peak H1, Leu-Arg-Arghad no ability to phosphorylate histone Ala-Ser-Leu-Gly (Kemptide),or the casein kinase I1 substrate Arg-Arg-Arg-Glu-Glu-Glu-Thr-Glu-Glu-Glu. The partially purified kinase had a K,,, for ATP of 38 PM with T669 as substrate. To confirm that T669 kinase isimmunologically related to MAP-2 kinase/ERK, fractions from the Superose 12 column were Western blotted using two different antibodies raised againstsyntheticpeptidesfromthe carboxyl terminus of ERK-l/MAP-2 kinase (40). Amousemonoclonal antibody reacted strongly with a sharp double band of 43 kDa which was present in the same fractions that contained T669 kinase activity (Fig. 76); asingle band of thesameapproximate mobility was detected in crude extracts of untreated KB or HepG2 cells(Fig.7c). No bands were seen upon Western blotting with an irrelevant monoclonal antibody (not shown). Usinga rabbitantiserumdirectedagainstan overlapping ERK-2 sequence, we detected the same band seen with the monoclonal antibody and an additional doublet at about 45 kDa (Fig. 7d). The appearance of the bands was similar to that reported for putative myelin basic protein kinases from Swiss 3T3 cells that are phosphorylated by extracts made from EGF-stimulated cells (76). These bands, and no others, were readily detectable in unfractionated extracts of KB or HepG2 cells (Fig. 7e).The mobility of the bands in the KB extracts was significantly greater than that of the bands in from variations the HepG2 cells (Fig.7 e ) .Whether this results in post-translational modification of the same enzyme in the two human cell lines or whether the bands detected in the Western blots are different human ERK isozymes remains to be established. Close inspection of some Western blots made with crude extracts from IL-1-treated or control KB cells showed that IL-1 treatment caused the splitting of both bands from singlets to doublets, consistent with an IL-1-dependent covalent modification, presumably phosphorylation. To confirm that
I L - 1-activated SerinelThreonine Kinase
22668
1
2
3
Yr(kDa)
4
i2ca 97 69
*
30
20
10
30
50
40
m
21.5
Fraction Number h
MrfkDa) 25 26
27
28
29
. . . . .
30 31 I
32
.
,
33 34 35 ,
,
KB
HeoGP
' -69 "
.
-30
+
-
+ 11.1
-21 5
I
Mr(kDa) 27 26 *
30 31
29 28 .
I
.
I
I
32
33
34
,
.
,
KB
HepG2
69
-46
-
-30
. +
- _" -
+ IL-l
-
-
IL-la
-
anti-ERK
+
-
-
+
+
+
-
FIG. 8. Immunoprecipitation of E R K s from ["'Pjphosphatelabeled KR cells. (:onfluent dishes o f l i t 3 cells were Ial)~lcd with [ "'l']orthophosphate and extracted as descrihed under "Experimental I'rocedures." Some cells (Innrs .'I and 4 ) received 20 ng/ml 1 L - l ~ during the final 15 min of laheling. Equal aliquots of the cell extracts were denatured, precleared, and immunoprecipitated with polvclonal anti-ERK serum (Ionrs 1 and 3 ) or normd rahtit serum (/rrnr,.s 2 and 4 ) . Immunoprecipitates were collected with Pansortin, washed extensively. eluted hv hoiling in SDS-PAGE sample buffer, and electrophoresed ona 1 2 . 5 5 polvacrvlamide gel. Autoradiographv oft he dried gel was for 16 h. Molecular weight markers are indicatedon the rrplrf. a 43 kDa ERk &
1
b 45 kDa ERk
m b
I
pn 3 5 FIG. 7. Size exclusion chromatography of IL- 1-stimulated kinase activity andWestern blotting with anti-MAP-2 kinase/ ISHK antibodies. a. 'l'(XS1 kinasecontainingfractions eluted off pooled, concentrated. and chromatographed I)hc~nvl-Sel,har~)se were o n Superose 12. The column was calibrated with the following molecular mass markers: thvroglot)ulin monomer, 330 k h ; y-glohulin. 158 kDn; ovall)umin, 44 k I h ; mvoglohin, 17 k h ; and cvanocohalamin, ~. -~ ", 45kDa I 43kDa 1 3 5 k I h . Ten microliters o f selected fractions was assayed for its - . , I1 ahilitv t o phosphorylate either ST-1( o p m syunrrs) or myelin hasic Experlmenl (1) (21 I O (21 I phosphoseme 8-28 10.20 3.15 4.70 protein ( c / o s c d diornonds). Final suhstrate concentrations were 1 mM phospholhreonme 4.64 5.40 1.88 2-07 a n d 120 P M , respectively; maximum net incorporation of'"'1' into STphospholyrosme 5.73 5.26 1.RR 2-10 . " . . 1 was 44,746 cpm (fraction 3 0 ) . and into myelin basic protein was l6:1,58:1 cpm (fraction 3 0 ) . All assays were performed in duplicate and FIG. 9. Phosphoamino acid analysis of E R K s phosphoryla r e plotted as the mean. h. fractions from the Superose 12 column KKK ated in response to 1 1 ~ 1 .'~1'-1,~1t~e1ed immuno~,recipit~lt(.fl shown in o were electrophoresed on an S D S - P A G E gel, (I.'Rh) hands o f 4 : 1 kDa ( n ) and 45 kl)a Ih) from Fig. 8 , Irrnr. :{. wwc. transferred t o nitrocellulose, and stained with anti-MAP-2 kinase excised from polyacrylamide gels. eluted, and hydrolyzed as descritrd monoclonal antihodv followed bv an anti-mouse Ig-streptavidin/l>io- under "Experimental l'rocetlures." I'hosphoamino acids in I he hytin-peroxidasedetectionsystem.Fractionnumbersareindicated drolysates were separated bv two-dimensional high voltage elrctrnalong t h e top o f t he hlot. c, W e s t e r n t h t t i n gof unfractionated K H o r phoresis on thin laver cellulose. The elertropherograms we're exposed Hep(;'L cell extracts a s in h. d, a tluplicate N'estern blot to that shown for 24 h to storage-phosphor plates which were then visnalized and in h was stained with a rathit polvclonal anti-EKK serum. Immune quantitated in a phosphorimagerusingtheImageQuantprogram A. c, Western blotting of complexes were tletectetl with ""I-protein o f the screen (Molecular Dvnamics Inc.). (;revscale reprnduct ions unfractionntetl Kt3 o r Hep(;2 extractswiththeanti-EHKserum. imagesareshownhere. In theexperimmtshown. 200 ('erenkov Iktection was with peroxitlase-coupled goat anti-rathitIg. counts were applied for the 45-kIh band and 500 rounts fnr the 4 3 k I h hand. Quantitative analvsis o f two independent experiments is on the total pixelvolr~mt.within this was the case,KR cells were labeled with [:':!P]orthophos- also shown; units are arhitrarv. hased a n ellipse enclosing each phosphoamino arid spot. phate, briefly stimulated wit.h IL-1 or vehicle, and then ex-
4
"~ ~
~~~
~
I
~
~
~
~~
I
tracted. The ext.ract,s were immunoprecipitated with either anti-ERK serum or control serum and visualized hy SDS- IL-1. Taken together with the experiments in which phosphaPAGE and autoradiography. Extracts made from control cellstase inhibitors were omitted (Fig. 4). these data support the contained no radiolabeled protein bands that were specificallyhypothesis (5.3) thattheenzvmesareonlvactiveinthe precipitated by anti-EIZK serum (Fig. 8, lane I); IL-1-treated phosphorylated state. T h e n a t u r e of the phosphorvlation induced bv II.-l was cells contained two phosphoproteinsof 43 and 4.5 kDa precipitated by the antiserum(lanc 3 ) which presumably correspond examined by two-dimensionalphosphoaminoacidanalysis (Fig. 9). Both the 4.3- and 45-kt)a proteins contained phosto phosphorylated forms of t.he ERKs detected hy Western phoserine, phosphothreonine, and phosphotyrosinein the r a blotting. This experiment confirms that a change in phosphorylation stateof ERKs occurs upon treatmentof cells with tio 2:1:1, respectively. If we accept that 11,-1 activates tht. 4 3 -
IL-1 SerinelThreonine Kinase -activated and 45-kDa ERKs via an intermediate kinase (or kinases) thenthepresence of phosphotyrosineinthesesubstratesis particularly Of the I IL-' receptor (6) revealed thatitdidnot possess intrinsictyrosinekinase activity, and the same is true for the type I1 r e ~ e p t o r No .~ intrinsic kinase activitycould be demonstrated for IL-1receptors purifiedfrom EL4 cells (77), and onlyasingle direct reportexists of anIL-1-stimulatedtyrosinephosphorylation: a 41-kDa protein in plasma membranes of K562 leukemia
22669
10. Bomsztyk,K.,Sims, J. E., Stanton, T. H., Slack, J., McMahan, C. J., Valentine, M. A,, and Dower, S. K. (1989)Proc. Natl. Acad. Sci. U. S. A. 86, 8034-8038 11, Spriggs, M. K,, Lioubin, p. Slack, J., J., Dower, S. K,, Jonas, U,, Cosman, D., Sims, J. E., and Bauer, J. (1990) J . Biol. Chem. 265,22499-22505 12. Rosoff, P. M. (1990) Semin. Immunol. 2, 129-137 13. Guy, G. R.,Bee, N. s.,andPeng, c. s. (1990)Prog. Growth. Factor Res. 2, 45-70 14. Shirakawa, F., Yamashita, U., Chedid, M., andMizel, S. B. (1988) Proc. Natl. Acad. Sci. U. S. A. 85, 8201-8205 cells (78).An alternative hypothesis, not requiring the direct 15. Zhang, y., Lin, J.-X,,yip, y. K., and Vilcek, J , (1988) Pro,-. Natl, stimulation of tyrosine kinase activityby IL-1, would be that Acad. Sci. U. S. A. 85, 6802-6805 IL-1 might stimulatephosphorylation of ERKs solely on 16. Renkonen, R., Mattila, P., Hayry, P., and Ustinov,J. (1990) Eur. J . Immunol. 20, 1563-1567 serine/threonine residues and that this causes subsequent tyrosine phosphorylation by a separate kinase or stimulates 17. Chedid, M.9 and Mizel, s. B. (1990) Mol. Cell. Bid. 10, 38243827 autophosphorylation Of the ERKs On tyrosine residues' Of 18. Didier,M.,Aussel, C., Pelassy, C., and Fehlmann, M. (1988) J . course, if one accepts that ERKs might be activated entirely Immunol. 141,3078-3080 via autophosphorylation then the activatormolecule need not 19, carroll, G. J. (1986) B ~J,, Rheumatol, 25, 359-365 be a kinase at all; such a possibility has been considered by 20. Nakano, T., Ohara, O., Teraoka, H., and Arita, H. (1990) FEBS Ahn et al. (76). TheERKs compriseafamily of a t least four Lett. 261, 171-174 and possibly more kinases having distinct tissue and &vel- 21. Shiroo, M.3 and Matsushima, K. (1990) Cytokine 2, 13-20 22. Bomsztyk, K., T o i v o h B.9 Emery, D. w.9 RooneY, J. w., Dower, opmental distributions (40). An attractive possibility would S. K., Rachie, N. A., and Sibley, C. H. (1990) J. Biol. Chem. be the existence of afamily of receptor-specific activators 265,9413-9417 which would Provide for a further level of fine tuning of 23. Raz, A., Wyche, A., and Needleman, P. (1989)Proc. Natl. Acad. specificcellularresponses. The two IL-1-activatablekinases Sci. U. S. A . 86, 1657-1661 identified in the present paper must purified be and sequenced 24. Magnuson, D. K., Maier, R., and Pohlman, T.H. (1989) Surgery t o determinetheirrelationshipto (or identitywith)members 106,218-223 of the ERK family. ~ ~ ~ with ~ fthese ~ mo~ecu~es l l ~ as , sub- 25. Zucali, J. R., Morse, c., and Dinarello, C. A. (1990) Exp. Hematol. 18,888-892 in either (40) Or dephospho~latedform, 26, Abraham, R, T., Ho, S, N., Barns, T, J,, and McKean, D. J. i t will be possible tocharacterize biochemically the IL-1(1987) J. Bid. Chem. 262, 2719-2728 27. Ostrowski, J., Meier, K. E., Stanton, T. H., Smith, L. L., and associated activator and thereby move closer to an underBomsztyk, K. (1988) J . Bid. Chem. 263, 13786-13790 standing of IL-1 receptor function. 28. Mills, G. B., May, C., Hill, M., and Gelfand, E. W. (1989) J . Cell. Acknowledgments-We greatly appreciate the excellent technical Physiol. 141,310-317 assistance of Heather Schule, JenniferSlack, and Kumi Weerawarna. 29. Macchia, G.9 Baldari, C. T., Massone, A,, and Telford,J. L. (1990) Mol. Cell. Biol. 10, 2731-2737 We thank Dr. Rony Seger for his generous gift of anti-ERK serum, 30. Bird, T. A.7 and Saklatvala, J. (1989) J . Immunol. 142, 126-133 ~ ,sims and David and we are grateful to our colleagues D ~ John 31. Bird, T. A., and Saklatvala, J. (1990) J. Biol. Chem. 265, 235Gearing for critical reading of the manuscript. 240 32. Munoz, E., Beutner, U., Zubiaga, A., and Huber, B. T. (1990) J . REFERENCES Immunol. 144,964-969 1. Oppenheim, J. J., Kovacs, E. J., Matsushima, K., and Durum, S. 33. Case, J . P., Lafyatis, R., Kumkumian, G. K., Remmers, E. F., and K. (1986) Immunol. Today 7,44-45 Wilder, R. L. (1990) J. Immunol. 145, 3755-3761 2. Auron, P. E., Webb, A. C., Rosenwasser, L., Mucci, S. F., Rich, 34. Zu, Y., Kohno, M., Kubota, I., Nishida, E., Hanaoka, M., and A,, Wolff, S. M., and Dinarello, C. A. (1984) Proc. Natl. Acad. Namba, Y. (1990) Biochemistry 29,1055-1062 Sci. U. S. A. 81, 7904-7911 35. Kaur, P., Welch, W. J., and Saklatvala, J. (1989) FEBS Lett. 3. March, C. J., Mosley, B., Larsen, A,, Cerretti, D. P., Braedt, G., 258,269-273 Price, V., Gillis, S., Henney, C. S., Kronheim, S., Grabstein, 36. Doye, V., Soubrier, F., Bauw, G., Boutterin, M.-C., Beretta, L., K., Conlon, P. J., Hopp, T. P., and Cosman, D. (1985) Nature Koppel, J., Vandekerckhove, J., and Sobel, A. (1989) J . Biol. 315,641-647 Chem. 264, 12134-12137 4. Eisenberg, S. P.,Evans,R.J.,Arend,W. P., Verderber, E., 37. Fagarasan, M. O., Bishop, J. F., Rinaudo, M. S., and Axelrod, J. Brewer, M. T., Hannum, C. H., and Thompson, R. C. (1990) (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 2555-2559 Nature 343, 341-345 38. Gallis, B., Prickett, K. S., Jackson, J., Slack, J., Schooley, K., 5. Hannum, C. H., Wilcox, C. J., Arend, W. P., Joslin, F. G., Dripps, Sims, J. E., and Dower, S. K. (1989) J . Immunol. 143, 3235D., Heimdal, P. L., Armes, L. G., Sommer, A., Eisenberg, S. P., 3240 and Thompson, R.C. (1990) Nature 343, 336-340 39. Qwarnstrom, E. E., MacFarlane, S. A,, Page, R. C., and Dower, 6. Sims, J. E., March, C. J., Cosman, D., Widmer, M. B., MacS. K. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 1232-1236 Donald, H. R., McMahan, C. J., Grubin, C. E., Wignall, J. M., 40. Boulton,T. G., Nye, S. H., Robbins,D. J., Ip, N. Y., Radziejewska, Call, S. M., Friend, D., Alpert, A. R., Gillis, S., Urdal, D. L., E., Morgenbesser, S. D., DePinho, R.A., Panayotatos, N., Cobb, and Dower, S. K. (1988) Science 241, 585-589 M. H., and Yancopoulos, N. (1991) Cell 65,663-675 7. Sims, J. E., Acres, B., Grubin, C. E., McMahan, C. J., Wignall, 41. Benjamin, D., and Dower, S. K. (1990) Blood 75, 2017-2023 J. M., March, C. J., and Dower, S. K. (1989) Proc. Natl. Acad. 42. Countaway, J. L., Northwood, I. C., and Davis, R. J. (1989) J . Sci. U. S. A. 86,8946-8950 Bid. Chem. 264,10828-10835 8. Horuk, R., Huang, J. J.,Covington, M., and Newton, R. C. (1987) 43. Laemmli, U. K. (1970) Nature 227,680-685 J . Bid. Chem. 262, 16275-16278 44. Cooper, J. A., Sefton,B. M., and Hunter, T. (1983)Methods 9. Chizzonite, R., Truitt, T., Kilian, P. L., Stern, A. S., Nunes, P., Enzymol. 99, 387-402 Parker, K. P., Kaffka, K. L., Chua, A. O., Lugg, D. K., and 45. Rees, D. J. G., Ades, S. E., Singer, S. J., and Hynes, R. 0. (1990) Gubler, U. (1989) Proc. Natl. Acad. Sci. U. S. A. 86,8029-8033 Nature 347, 685-689 46. Heisermann, G. J., and Gill, G. N. (1988) J . Bid. Chem. 263, C. J . McMahan, J. Slack, B. Mosley, D. Cosman, S. D. Lupton, 13152-13158 L. L. Brunton, C. E. Grubin, J. M. Wignall, N. A. Jenkins, D. J. 47. Dower, S. K., Gillis, S., and Sims, J. E. (1989) in Interleukin-1, Gilbert, N. G. Copeland, K. Heubner, C. M. Croce, L. A. Cannizzarro, Inflammation and Disease (Bomford, S., and Henderson, B., D. Benjamin, S. K. Dower, M. K. Spriggs, and J. E.Sims, submitted eds) pp. 63-78, Elsevier, Amsterdam for publication. 48. Solari, R. (1990) Cytokine 2, 21-28
22670
IL-1 -activated SerinelThreonine Kinase
49. Stanton, T. H., Maynard, M., and Bomsztyk, K. (1986) J . Bid. Chem. 261,5699-5701 50. Novak-Hofer, I., and Thomas, G. (1985) J. Biol. Chem. 260, 10314-10319 51. Yu, K. T., Khalaf, N., and Czech, M. P. (1987) J . Biol. Chem. 262, 16677-16685 52. Ray, L. B., and Sturgill, T. W. (1987) J . Biol. Chem. 262, 15021506 53. Anderson, N. G., Maller, J. L., Tonks, N. K., and Sturgill, T. W. (1990) Nature 343, 651-654 54. Sturgill, T. W., Ray, L. B., Erikson, E., and Maller, J. L. (1988) Nature 334, 715-718 55. Ballou, L. M., Luther, H., and Thomas, G. (1991) Nature 349, 348-350 56. Morrison, D. K., Kaplan, D. R., Escobedo, J. A., Rapp, U. R., Roberts, T. M., and Williams, L. T. (1989) Cell 58, 649-657 57. Kovacina, K. S., Yonezawa, K., Brautigan, D. L., Tonks, N. K., Rapp, U. R., and Roth, R.A. (1990) J. Biol. Chem. 265,1211512118 58. Fujita-Yamaguchi, Y., and Kathuria, S. (1988) Biochem. Biophys. Res. Commun. 157,955-962 59. Northwood, I. C., and Davis, R. J. (1990) Proc. Natl. Acad. Sci. U. S. A. 87,6107-6111 60. Akiyama, T., Ishida, J., Nakagawa, S., Ogawara, H., Watanabe, S., Itoh, N., Shibuya, M., and Fukami, Y. (1987) J. Biol. Chem. 262,5592-5595 61. Giugni, T. D., Chen, K., and Cohen, S. (1988) J . Biol. Chem. 263, 18988-18995 62. Hathaway, G. M., Lubben, T . H., and Traugh, J. A. (1981) J . Biol. Chem. 255,8038-8041
63. Boulton, T. G., Gregory, J. S., and Cobb, M. H. (1991) Biochemistry 30, 278-2860 64. Ray, L. B., and Sturgill,T. W. (1988) J. Biol. Chem. 263, 1272112727 65. Miyasaka, T., Chao, M.V., Sherline, P., and Saltiel,A. R. (1990) J. Biol. Chem. 265,4730-4735 66. Ackerman, P., and Osheroff, N. (1989) J. Biol. Chem. 264,1195811965 67. Sanghera, J. S., Paddon, H. B., Bader, S. A,, and Pelech, S. L. (1990) J. Biol. Chem. 265, 52-57 68. Ahn, N. G., Weiel, J. E., Chan, C. P., and Krebs, E. G. (1990) J . Biol. Chem. 265, 11487-11494 69. Takishima, K., Griswold-Prenner,I., Ingebritsen, T., and Rosner, M. R. (1991) Proc. Natl. Acad. Sci. U. S. A . 88, 2520-2524 70. Martenson, R. E., Law, M. J., and Deibler, G. E. (1983) J . Biol. Chem. 258,930-937 71. Corden, J. L., Cadena, D. L., Ahearn, J. M., and Dahmus, M. E. (1985) Proc. Natl. Acad. Sci. U. S. A. 82, 7934-7938 72. Payne, J. M., Laybourn, P. J., and Dahmus, M. E. (1989) J . Bid. Chem. 264,19621-19629 73. Cisek, L. J., and Corden, J. L. (1989) Nature 339,679-684 74. Zandomeni, R., Zandomeni, M. C., Shugar, D., and Weinmann, R. (1986) J. Biol. Chem. 261, 3414-3419 75. Arias. J. A.. Peterson. S. R.. and Dvnan. W. S. (1991) . , J . Biol. Chem. 266,8055-8061 ’ 76. Ahn, N. G.. Seaer, R.. Bratlien, R. L., Diltz. C. D.. Tonks, N. K., and Krebs, E G. (1991) J . Blol. Chem. 266,4220-4227 77. Bird, T. A., Gearing, A. J. H., and Saklatvala, J. (1988) J . Biol. Chem. 263,12063-12069 78. Resch, K., Martin, M., Lovett, D. H., Kyas, U., and Gemsa, D. (1986) Zmmunobiology 172, 336-345 79. Hickey, E., Brandon, S. E., Potter, R., Stein, G., Stein, J., and Weber, L. A. (1986) Nucleic Acids Res. 14,4127-4145 “
I
