Src Tyrosine Kinase Mediates Stimulation of Raf-1 and Mitogen-Activated Protein Kinase by the Tumor Promoter Thapsigargin'

ICANCERRESEARCH57.3168-3173, August 1. 19971 Src Tyrosine Kinase Mediates Stimulation of Raf-1 and Mitogen-activated Protein Kinase by the Tumor Promoter Thapsigargin'

Tsung-Shu Oliver Chao,2 Mark Abe, Marc B. Hershenson, Ignatius Gomes, and Marsha Rich Rosner@

The Ben May Institute for Cancer Research and Departments of Pharmacological and Physiological SciencesiT-S. 0. C., 1.G., M. R. R.j and Pediatrics fM. A., M. B. H.). University of Chicago. Chicago. Illinois 60637

ABSTRACT signal-regulated kinase, or ERK). This activation is independent of @ protein kinase C or Ca2 influx (2) but requires the presence of Raf-l Thapsigargin is a non-phorbol ester-type tumor promoter that elevates (7). The activation of MAP kinase can lead to the specific phospho the intracellular Ca2@ (Ca@2') levels by blocking the microsomal Ca2@ rylation of EGF receptor on threonine 669 (8, 9) as well as a subse ATPase. At present, the consequence of this Ca@2@increaseand the nature quent stimulation of a membrane-associated phosphatase (10). These of the tumorigenicity of thapsigargin still remain to be elucidated. Previ ously, we demonstrated that thapsigargin activates the mitogen-activated events appear to underlie the negative regulation of EGF receptor by protein (MAP) kinase via Ca@2' but independently of protein kinase C or both phorbol ester (11) and thapsigargin action (10, 12, 13). More Ca2â€•Influx.Here, we show that thapsIgar@n also rapidly stimulates the over, they suggest that the MAP kinase signaling pathway can play a Src tyrosine kinase. Transfection of a v-Src gene into a hippocampal cell role in mediating the cellular influence of thapsigargin. line (H19â€”7)rendersa constitutive activation of MAP kinase, whereas Activation of MAP kinase is a rapid response shared by diverse transfection of a kinase-deficient Src mutant blocks the activation by groups of growth modulators and tumor promoters. Numerous reports thapsigargin, suggesting that Src is required for the thapsigargin-induced have documented a well-conserved signaling pathway for MAP ki MAP kinase activation. Cotransfection of a doednant-inifibitory Raf-1 nase activation that requires a sequential activation of Grb2, Sos, Ras, and the v-Src genes into H19â€”7 cells results in an Inhibition of the Raf, and MEKS (reviewed in Ref. 14). It is also known that alternative otherwise constitutively elevated MAP kinase activity, suggesting that pathways for activation of MAP kinase exist in parallel (15, 16). Raf-1 is required for the Src-dependent activation of MAP kinase. Sinai lady, in the LA-90 cells, expression of a temperature-sensitive allele of Although MAP kinase appears to integrate a network of intracellular v-Src constitutively activates Raf.1 and MAP kinase, whereas expression signals, it is of importance to identify and distinguish how different of a dominant-Inhibitory Raf-1 mutant abolishes the MAP kinase activa extracellular stimuli transfer their signals to MAP kinase. tion Induced by either v-Src or thapsigargin treatment Together, these In this report, we focus our investigation on the action of thapsi results suggest that thapsigargin stimulates MAP kinase signaling via Src gargin that is related to MAP kinase signaling. We show here that and Raf-1. The activation of this Src-MAP kinase pathway suggests a thapsigargin rapidly stimulates the Src tyrosine kinase; the Src kinase biochemical mechanism for the tumorigenic nature of thapsigargin. activity is required for the thapsigargin-induced activation of Raf-1 and MAP kinase. Because various cellular components, such as cy INTRODUCTION toskeletal proteins, focal adhesion kinase, and GTPase activating protein, are substrates of Src tyrosine kinase (17), it is conceivable Thapsigargin, originally isolated from the plant Thapsia garganica that the stimulation of Src, as well as the subsequent activation of L. (Apiaceae), is a non-phorbol ester-type tumor promoter (1). It Raf-l and MAP kinase, may contribute to the molecular basis of the inhibits the microsomal Ca2@ ATPase, causing the Ca2@to leak out of tumorigenicity of thapsigargin. the internal stores and elevating the Ca@2@4levels(1, 2). The depletion @ of internal Ca2 stores is then followed by a subsequent â€œcapacitative entryâ€•of extracellular Ca2@ through surface channels (3, 4). The MATERIALS AND METHODS effect of thapsigargin on Ca2@ mobilization has been studied exten EGF (receptor grade) was purchased from Biomedical Technologies sively (3, 5); however, the molecular mechanism for its tumorigenic (Stoughton, MA). Thapsigargin and MBP were from Sigma Chemical Co. ity remains to be elucidated. Several lines of evidence have suggested Anti-p2lras monoclonal antibody (Yl3â€”259) and anti-c-Raf-l antibody were that phosphorylation events are essential in relaying thapsigargin from Santa Cruz Biotechnology, Inc. Anti-Src antibody (Ab327) was a gift action. For example, the transcriptional activation of the glucose from Dr. J. Brugge (ARID Pharmaceuticals, Cambridge, MA). Anti-MAP regulated protein GRP78 promoter by thapsigargin required both kinase antiserum (Ab283) was developed as described previously (2). The tyrosine and serine/threonine kinases (6). On the other hand, alkaline plasmid encoding the kinase-inactive MEK was a gift from Dr. G. Johnson phosphatase abolished the activity of thapsigargin-activated calcium (National Jewish Center for Immunology and Respiratory Medicine, Denver, influx factors from Jurkat cells to induce rapid chloride current in the CO), andplasmidsencodingv-Srcandkinase-inactiveSrcwerea gift fromD. oocytes (4). Foster (Hunter College, New York, NY). Previously, our laboratory demonstrated that thapsigargin-induced Cell Cultures. The H19â€”7cellswere generated from rat hippocampal neurons (18). The cells were conditionally immortalized by stable transfection release of Ca@2@activated MAP kinase (also known as extracellular with a temperature-sensitive SV4O large T antigen. H19â€”7cells were grown at 33Â°Cin DMEMJ1O% fetal bovine serum and kept under G4l8 selection Received 1/13/97: accepted 5/28/97. throughout the experiments (18, 19). Under these conditions, the cells are The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with undifferentiated and proliferate in response to serum. 18 U.S.C. Section 1734 solely to indicate this fact. The dominant-negative Raf-1 cell line Rl was generated by stably trans I This work was supported by NIH Grant CA 35541 (to M. R. R.) and a gift from the fecting the LA-90 cells with the mutant Raf vector p301â€”1(20). The LA-MNC Cornelius Crane Foundation. control cell line was generated by transfecting with the parental vector to 2 Current address: SynthÃ©labo Research, Bagneux 92225, France. p301â€”1(pMNC).The LA-90 cells express a temperature-sensitive derivative 3 To whom requests for reprints should be addressed, at the Ben May Institute for Cancer Research, MC 6027, University of Chicago, 5841 South Maryland Avenue, of Src (20). Experiments with activated Src were performed by shifting the Chicago, IL 60637. Phone: (773) 702-1857; Fax: (773) 702-4634 or 6260; E-mail: cells to the permissive temperature (33Â°C)for10 mm; the control experiments [email protected]. were performed at the nonpermissive temperature (39Â°C;Refs.7 and 20). @@ 4 The abbreviations used are: Ca,2@, intracellular Ca2 MAP, mitogen-activated Transient Transfection of H19â€”7Hippocampal Cells. The H19â€”7cells protein; ERK, extracellular signal-regulated kinase; EGF, epidermal growth factor; MEK, MAP kinase kinase; MBP, myelin basic protein; LPA, lysophosphatidic acid; BK, were transfected with plasmids encoding the hemagglutinin-tagged ERK2, bradykinin; PMSF, phenylmethylsulfonyl fluoride. kinase-deficient Src Arg295 (21), v-Src, and dominant-inhibitory Rafp3Ol and 3168

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1997 American Association for Cancer Research. THAPSIGARGIN ACTIVATES Src, Raf-l, AND MAP KINASE A C4(7,20)usingthecalciumphosphateprecipitationmethod.TheSrcand @ 4 ______Raf-l constructs were about 4-fold in excess of the epitope-tagged ERK2 construct. A plasmid encoding @3-galactosidase driven by a cytomegalovirus â€”0- Thaps promoter was cotransfected to normalize for transfection efficiency. After transfection, the cells were allowed to grow in fresh medium with 10% FBS for U 16â€”24 h. The plates containing the cells from the same transfection were

@@ >1 pooled and replated to ensure homogeneous transfection efficiency (approxi :@ mately 15%). The replated cells were grown in fresh medium with 10% FBS @ :@ for 20â€”24hand subsequently starved for another 20â€”24h.The transfected 4 : cells were harvested for experiments within 60â€”72 h. @@@ 2 Sit KinaseAssay.Srctyrosmnekinaseactivitywasmeasuredbyphospho @ â€” rylation of rabbit muscle enolase (22, 23). Stimulated and control cells were @ :@ washedtwicewithPBSandthenlysedin50nmiTris,pH8.0;0.5%NP-40; @ 0 120 mt@i NaCl; 1 mM EDTA; 20 @tg/ml aprotinin; 200 mM Na3VO4; and 0.2 nmi c@5 PMSF. After removal of debris, the supematant was incubated with anti-Sec antibody (Ab327) for 30 mm. Protein A-Sepharose was added for another 30 mm. The mixtures were washed in radioimmunoprecipitation assay buffer and once in Tris-buffered saline (20 mrsi Tris, pH 7.5; 150 mM NaCl). The pellet was resuspended in 10 @.tlofkinase buffer (20 mai HEPES, pH 7.2; 5 mM @@@ 0 . . 20 30 MnCl2; and 200 p@M Na3VO4) containing 2â€”5 mg of acid- and heat-treated muscle enolase. The phosphorylation was initiated by adding 5 @tCi Time (mm) [y-32P]ATP to the mixtures and then incubated at 30Â°Cfor 10 mm. The reaction was terminated by adding 2X sample buffer and boiling for 2â€”5mm. B Thesampleswereresolvedona10%SDS-polyacrylamidegel.Thelevelof enolase phosphorylation was measured by scintillation counting of the excised bands. Epitope-tagged ERK2 Immune Complex Kinase Assay. The experiment was performed according to a modified procedure as described previously (24). The transfected cells were stimulated and lysed with 0.5 ml of ice-cold lysis buffer containing 50 mMTris-HCI, pH 7.5; 1% (w/v) Triton X-100; 40 mM

.@â€˜I @3-glycerophosphate; 100 mM NaCl; 50 m@i NaF; 2 mt@i EDTA; 200 @LM sodium @ :@ orthovanadate;and0.2mMPMSF.Insolublematerialswereremovedby @ @; centrifugationat14,000rpmfor 10mmat 4Â°C.Thecell lysateswerethen .! incubatedwith25 ml of proteinA precoupledwiththe antibody12CA5 @ â€˜a specifictothehemagglutininepitope(Babco,Emeryville,CA)for 16hat4Â°C. @ I Theimmunecomplexeswerewashedfourtimeswiththelysisbufferandonce @, with the kinase buffer containing 20 mM HEPES, pH 7.4; 10 mM MgCI2; 1 mM Dli'; 200 @LMNa3VO4and 10 mMp-nitrophenyl phosphate. The immune-complex kinase activity was measured by phosphorylation of the MBP. Briefly, the immune complexes were resuspended in a final volume of 40 ml of kinase buffer with 0.25 mg/ml MBP and 50 mM ATP (5 pCi [â€˜y-32P]ATP).The phosphorylation reaction was initiated by adding the AlP, control Thaps and samples were incubated at 30Â°Cfor 20 mm. The reactions were terminated by adding 20 ml of 2X Laemmli sample buffer. Samples were resolved on a 12% SDS gel. An autoradiogram was developed from the dried gel. The MBP C bandswereexcisedfromthegel,andradioactivitywascountedbyliquid scintillation. Raf-1 Immune Complex Kinase Assay. Raf- 1 kinase activity was ana 10 . lyzed by an immune complex kinase assay (25). Briefly, cells were washed U ERK2+Src twice with ice-cold PBS and lysed in 10 mM Tris-HCI, pH 7.4: 1% Triton X-lOO; 1 mM EDTA, 150 m@tNaCl; 50 ms,i NaF; 0.1% BSA; 20 @g/ml 8 aprotinmn; 200 mM Na3VO4; and 0.2 mM PMSF (lysis buffer). The cell debris @ and nuclei were removed by centrifugation, and the supernatants were incu

.@l : bated with anti.c-Raf antibody (Santa Cruz Biotechnology, Inc.) for 1 .5 h at @ :E 6 4Â°C.ProteinA-Sepharosewasaddedforanother30mm.Then,themixtures @@ weredilutedwith an equalvolumeof lysis bufferplus 10%sucroseand

â€” centrifuged for 20 mm at 2500 rpm. The pellet was washed twice with lysis @i 4 Iii IL ______phosphorylation of rabbit muscle enolase by the Src-immune complex precipitated with a specific antibody, Ab327, as described in â€œMaterialsand Methods.â€• Radioactivity of 2 phosphorylated enolase was quantitated by scintillation counting of the excised enolase bands from the dried gel. B and C. a kinase-deficient Src mutant inhibits epitope-tagged ERK2 activation by thapsigargin. H19â€”7cells were transfected with 4 @gofa hemag @ 0 â€” glutinin-tagged ERK2 plasmid with or without 16 @gofa kinase-deficient Src. After control EGF transfection. the cells were grown, pooled. and replated to obtain a homogeneous trans fected cell population as described in â€œMaterialsand Methods.â€•The serum-starved cells Fig. 1. A. thapsigargin activates Sec tyrosine kinase in Hl9â€”7cells. Confluent Hl9â€”7 were then stimulated with 1 sg/ml thapsigargin (Thaps: B) or 10 flMEGF (C) for 5 mm, cells were serum-starved for 48 h in DMEM before treatment with I @sg/mlthapsigargin lysed. and immunoprecipitated. The kinase activity was measured as MBP phosphoryla or 10 nst EGF for various times as indicated. The kinase activity of Src was measured as tion by the ERK2 immune complex. 3169

A buffer, twice with PAN (10 mMPIPES, pH 7.0; 100 mMNaC1;and 20 pg/mi aprotinin) plus 0.5% NP4O, and finally twice with PAN. The pellet was resuspended in 20 @lofkinase buffer (20 mrsiPIPES, pH 7.0; 10 mMMnCl2; 20 pg/mi aprotinin; and 200 @MNa3VO4) containing 25â€”50 ng of a recom @â€˜ binant polyhistidine-tagged, kinase-inactive MEK. The immune-complex ki :@__ nasereactionwasinitiatedbyadding10mCiof[â€˜y-32P]AlPtothemixture. @@ Sampleswereincubatedat30Â°Cfor15â€”20ruin.Thereactionwasterminated @ by adding 2X SDS-gel sample buffer and boiling for 2â€”5mm. The samples @ .E were resolved on a 10% SDS-gel, and the level of MEK phosphorylation was @ .E measuredbyscintillationcountingoftheexcisedbands. .@ @, Analysis ofMAP Kinase Activity by in-Gel Kinase Renaturation Assay. @ MAP kinase activity was analyzed by an in-gel assay (2). Briefly, cell extracts @ (5â€”10@goftotal protein) were resolved in a 10% SDS-polyacrylamide gel containing 0.1 mg/mI MBP. After electrophoresis, the gel was washed twice with buffer A (50 mMHEPES, pH 7.4; and 5 mMmercaptoethanol) plus 20% 2-propanol to remove SDS, then re-equilibrated with buffer A at room tern perature for 1 h followed by two changes of 6 Mguanidine-HCI in buffer A. B Subsequently,theenzymesonthegelwererenaturedwithtwochangesof500 ml of buffer A containing 0.04% Tween 20 at 4Â°Cfor16 h. The renatured gel

, was incubated with buffer B (25 ruM HEPES, pH 7.4; 10 mM MgCl2; 100 @M MEK-@ Na3VO4;and 5 mM @3-mercaptocthanol)at30Â°Cfor30 mm. The phosphoryl ation assay was conducted by layering 10 ml of buffer B plus 50 mM [â€˜y-32P]-ATP(250pCi) onto the renatured gel and incubating it at 30Â°Cfor 1 h. C TO FSK EGFFSK The reaction was terminated by immersing the gel in 5% trichloroacetic acid and 10 mMNa-PP1.and the gel was washed extensively with this solution to + + TG EGF remove unreacted radioactivity. A control experiment was performed with a gel in the absence of MBP to measure autophosphorylation by the renatured C kinases.AutoradiogramsweredevelopedbyexposingtheKOdakX-OMAT film to the dried gel with an intensifying screen. Quantitation of the MBP phosphorylation by the MAP kinases was measured by either digital optical scanning of the autoradiograms on an image analyzer (AMBIS, San Diego, CA) or scintillation counting of the gel bands that corresponded to the MAP kinases.

@. Western Blotting of MAP Kinase. Cell extracts were subjected to dcc :E.@. trophoresisona 10%SDS-polyacrylamidegelandtransferredtonitrocellulose @ @. sheets on a semi-dry transfer unit. The Ab283 antiserum wasused to detect the @ MAP kinase (2, 7). Western blots were visualized by using the enhanced @ cherniluminescence reagents (Arnersham Corp., Arlington Heights, IL).

RESULTS

Thapsigargin Activates Src Tyrosine Kinase in H19-7 Hip C pocampal CeUs. To test the possibility that the tyrosine kinase pp@Src functions as a mediator for thapsigargin/Ca@2@ signaling, we D directly analyzed Src tyrosine kinase activity in response to thapsi gargin in H19â€”7cells. The Src tyrosine kinase activity was measured 100000 by immune complex assay using acid-treated rabbit muscle enolase as substrate (22, 23). Fig. 1A shows that thapsigargin stimulates Src U ERK 80000 U ERK+Raf approximately 2-fold within 5 mm, and the kinase activity is main tamed for at least 30 mm without significant attenuation. The EGF also elicits a quick induction of Src activity to a extent similar to i 60000 thapsigargin.ThisresultsuggeststhatthapsigarginmayuseSrcfor @@ downstreamsignalingevents. @ Â£ Kinase-deficient Src Blocks Thapsigargin-induced Activation @ of MAP Kinase. We have shown thatthapsigargincanactivateMAP kinase in human foreskin (HSWP) fibroblasts, A431, LA-90, and H19â€”7cells (2, 7, 19). To test whether Src is required for mediating the thapsigargin-induced activation of MAP kinase, we transiently 0 . transfected the Hl9â€”7 cells with a kinase-deficient Src mutant (21)

Unstlmulated

Fig. 2. A, thapsigargin (Thaps) activates Raf-l in H19â€”7cells. Cells were grown and was analyzed as in A. C and D, dominant-inhibitory Raf-1 blocks thapsigargin but not stimulated with thapsigargin or EGF as in Fig. 1. The kinase activity of Raf-l was EGF-induced MAP kinase activity. H19â€”7cellswere transfected with 4 @.tgofa hemag measured by Raf-I immune complex phosphorylation assay using a recombinant poly- glutinin-tagged ERK2 plasmid with or without 16 @sgof a dominant-inhibitory Raf-1 lustidine-tagged, kinase-inactive MEK as substrate as described m â€œMatenalsandMeth ods.â€•Inset, a representative autoradiogram. B, forskolin (FSK) pretreatment inhibits (P301)allele. After transfection. the cells were crown. Pooled, replated, and serum-starved thapsigargin- and EGF-induced activation ofRaf-l. Hl9â€”7cellswere treated as indicated: as in Fig. 1. The cells were stimulated with 1 @.tg/mithapsigargin(C) or 10 nsi EGF (D) 1 ,.tg/ml thapsigargin (TG) for 5 mm; 10 nat EGF for 5 mm; FSK + TG or FSK + EGF for 5 minj lysed. and immunoprecipitated. The kinase activity was measured as MBP (10 mM) pretreated for 15 mm followed by thapsigargin or EGF for 5 rain. Raf-1 activity phosphorylation by the ERK2 immune complex as in Fig. 1. 3170

Dominant-Inhibitory Raf-1 Blocks Src-dependent Activation of MAP Kinase. To further determine whether Raf-l is required for mediating Src-dependent activation of MAP kinase, we performed various combinations of transient transfection in Hl9â€”7 cells with an oncogenic v-Src, a dominant-inhibitory Raf-l (C4), and the epitope tagged ERK2 constructs. As shown in Fig. 3, transfection with the v-Src allele induces an activation of epitope-tagged ERK 2 activity (approximately 3-fold). In contrast, cotransfection of the v-Src with LUO Li. the dominant-inhibitory Raf-l substantially blocks the otherwise con stitutively activated ERK2 activity. Another line of evidence indicating Src requires Raf-l for MAP kinase activation comes from studies with the LA-90 cells. The BALB/c 3T3-derived LA-90 cells contain a temperature-sensitive Src. control vSrc vSrc+Raf. The cells are cultured at the nonpermissive temperature for Src Fig. 3. Constitutive activation of MAP kinase by v-Sec is inhibited by dominant inhibitory Raf-l. H19â€”7cellswere transfected with 4 @sgofepitope-taggedERK2 plasmid activity (39Â°C) and can be shifted to the permissive temperature (control), a combination of 4 @.tgofepitope-tagged ERK2 and 12 @sgof v-Src (vSrc), or (33Â°C)to activate the endogenous Src kinase. Two derivatives of a combination of 4 @sgepitope-tagged ERK2, 12 @gofv-Src, and 12 @gofa dominant LA-90 cells were then established (7, 18): Rl, which contains an inhibitory Raf-l mutant C4 (vSrc+Rafâ€”).Calf thymus DNA was used to equalize the total DNA for transfection. A plasmid encoding @-galactosidasewas also cotransfected for expression vector for dominant-inhibitory Raf-l (p301â€”1),and MNC, normalization. Cells were harvested after 40 h of serum starvation. The epitope-tagged which contains the expression vector alone as a control. As shown in ERK2 was immunoprecipitated and analyzed for kinase activity as in Fig. 1. Fig. 4A, incubating the LA-90 cells at 33Â°Cfor 10 mm to activate the endogenous Src induced a significant Raf-l activation in the MNC together with an epitope-tagged ERK2 as described in â€œMaterialsand control cells (Fig. 4A), indicating that Src activation can lead to Raf-l Methods.â€•Using epitope-tagged ERK2-immune complex kinase as says, we found that in the presence of the kinase-deficient Src, thapsigargin-induced activation of ERK2 was substantially blocked (Fig. IB). This result indicates that Src is necessary for thapsigargin A dependent activation of MAP kinase. Interestingly, the EGF-induced ERK2 was only slightly inhibited by the kinase-deficient Src (Fig. 10, suggesting that Src is not crucial for EGF-activated MAP kinase signaling. S Thapsigargin Activates Raf-1 in H19â€”7 Hippocampal Cells. S Previously, we reported that Raf-l is involved in thapsigargin-induced MAP kinase activation in the LA-90 cell line (7). Therefore, to further characterize the thapsigargin-induced signaling pathway, we also cx amined the Raf-l activity in Hl9â€”7 cells. As shown in Fig. 2A, a 2-fold activation by thapsigargin is observed in 5 mm, whereas EGF induces a nearly 5-fold activation of Raf-l activity toward phospho rylation of a recombinant MEK (25). The level of thapsigargin induced Raf-l activation is consistent with that of MAP kinase acti 39 33 vation by thapsigargin in all cell types we have tested previously. In Temp ( Â°C) addition, the Raf-l activity induced by either thapsigargin or EGF is abolished by pretreatment with forskolin (Fig. 2B), indicating that the B Raf-l activity is sensitive to cAMP inhibition. Pretreatment of the H19â€”7hippocampal cells with dibutyryl cAMP and forskolin also completely blocks the activation of MAP kinase induced by thapsi @ gargin (data not shown), suggesting that in the H19â€”7hippocampal ERKI--* @ cells, the MAP kinase signaling pathway is antagonized by cAMP as ERK2â€”* in several other cell types (26). Dominant-Inhibitory Ref-i Blocks Thapsigargin-induced Acti vation of MAP Kinase To further determine the requirement of 39 33 TG EGF 39 33 TG EGF Raf-1 for thapsigargin-induced MAP kinase signalings, we transiently transfected the Hl9â€”7 cells with a a kinase-deficient, dominant MNC Ri inhibitory mutant of Raf-l p301â€”i(7, 20) together with a hemagglu Fig. 4. Effects of temperature-sensitive Src on Raf-l and MAP kinase activities in tinin-tagged ERK2 plasmid. As analyzed by an epitope-tagged ERK2 LA-90 cells. LA-90 (MNC, control; Ri, dominant-inhibitory Raf-l mutant) cells were immune complex assay (Fig. 2C), the dominant-inhibitory Raf-l grown and maintained at 39Â°C(nonpermissive temperature) as described in â€œMaterials blocks the epitope-tagged ERK2 activity induced by thapsigargin but and Methods.â€•To activate the temperature-sensitive Sit, cell plates were shifted from 39Â°Cto 33Â°Cfor 10 mm. A, activation of Src leads to activation of Raf-l. MEK fails to inhibit the ERK2 activation induced by EGF. The data indicate phosphorylation by the Raf-l immune complex precipitated from MNC (U) and Rl (0) that Raf-l is required for thapsigargin-induced MAP kinase activation cells at 39Â°C(nonpermissive) or 33Â°C(permissive). The level of phosphorylation was in H19â€”7cells, whereas EGF may use an alternative pathway by measured by scintillation counting of the MEK band excised from the gel. The fold increase was calculated after subtracting the background cpm. B, Raf-1 is required for passing Raf-l to activate MAP kinase. These results are also in thapsigargin- and Src-dependent activation of MAP kinase. MNC and RI cells were agreement with our previous observations regarding Raf- 1-dependent incubated at 39Â°Cor 33Â°Cfor 10 mm without treatment, or treated with 1 @tg/ml thapsigargin (TG) or 10 flM EGF at 39Â°Cfor 5 mm. Cell extracts were analyzed by and -independent mechanisms for MAP kinase activation in the Western blotting. Activation of ERKI and ERK2 is identified by the characteristic LA-90 cells (7). bandshifts. 3171

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1997 American Association for Cancer Research. ThAPSIGARGIN ACFIVATES Src, Raf-l, AND MAP KINASE activation, whereas in Ri cells, no Raf-l activity was observed (Fig. interact with a calcium/calmodulin-dependent protein kinase IV re 4A). Conceivably, it is due to overexpression of the dominant-inhib mains to be elucidated. itory Raf-l mutant in Ri cells (20). A recent report also suggested that Src may directly interact with Shifting LA-90 cells to the permissive temperature (33Â°C)also Raf- 1, and the association may increase the autophosphorylation of activated MAP kinase in MNC but not Ri cells (Fig. 4B). As analyzed Raf-1 (33). Because we have previously observed a Raf-l-dependent by the characteristic bandshift on Western blots, thapsigargin can only activation of MAP kinase by thapsigargin, it was of interest to stimulate MAP kinase in MNC but not Rl cells (Fig. 4B), further determine the involvement of Raf-1 in Src-induced MAP kinase suggesting that the thapsigargin-induced MAP kinase activation, like activation. The studies with the dominant-inhibitory Raf-l mutant that due to Src, requires Raf-1. Also, the fact that EGF remains a (Figs. 3 and 4) strongly suggest that Src can be an upstream activator potent activator in Rl cell indicates again that EGF can bypass Raf-l for Ra.f-l, and Raf-1 may be required for the Src-dependent activation to activate MAP kinase (Fig. 4B). of MAP kinase. In addition, the thapsigargin-induced activation of Raf-l is sen sitive to cAMP-inhibition (Fig. 2). This phenomenon is similar to DISCUSSION other cell types that are also sensitive to cAMP-inhibition (26) but represents a different signaling mechanism from that observed in The actions of thapsigargin on the microsomal Ca2@ ATPase and PC-12 cells in which cAMP activates MAP kinase (34). This intracellular CA12@have been well documented. However, the bio cAMP inhibition of thapsigargin-induced Raf-l and MAP kinase chemical consequences of these actions, particularly the tumorigenic activity, together with the result that a kinase-deficient, dominant effects, are still unclear. Through various approaches, we have shown inhibitory Raf-l mutant blocks thapsigargin-induced MAP kinase that exposure to thapsigargin leads to modulation of the EGF receptor activation, strongly indicates that Raf is essential for the thapsi (9, 12). The molecular basis for this phorbol ester-insensitive regula gargin-dependent action. In the case of EGF-induced MAP kinase tion appears to be the result of thapsigargin-induced activation of activation, neither Src or Raf-l appear to be absolutely required MAP kinase (8, 13, 27). Although thapsigargin, as well as other (Figs. 1 and 2). It is likely that different upstream MAP kinase calcium mobilizers, activates MAP kinase approximately 2â€”3fold,the activators also mediate EGF action, as observed previously for biological impact of the MAP kinase cascade could be substantial. In other growth factors (26, 35), in contrast to the more stringent this report, we further demonstrate that thapsigargin can stimulate the requirements for the thapsigargin-induced pathway. @@@Srctyrosine kinase and that Src is required for the subsequent Thapsigargin-elicited calcium transients can exert multiple effects MAP kinase activation induced by thapsigargin. and may use various routes for signaling (36, 37). There are a number As shown in Fig. 1, thapsigargin stimulates Src tyrosine kinase of different signaling intermediates, including Ca2@/calmodulin de activity in 5 mm. This rapid stimulation implicates Src as an efficient pendent protein kinases, the classic protein kinase C family, phos upstream activator for thapsigargin-induced immediate or early sig phatases (e.g., calcineurin), proteases (e.g., calpain), ion channels, and naling responses. Because pp(@()Sfltyrosine kinase is widely expressed various Ca2tbinding proteins that require Ca2@ for their physiolog and is implicated in various cell functions (17), such as cell contact, ical function. To assess the involvement of these intermediates, we cell proliferation, and differentiation, this suggests that thapsigargin is have tested various inhibitors for the Ca2@-dependent kinases, phos able to exert its effects on specific cellular processes through Src phatases, or proteases, but we found no inhibitory effect toward tyrosine kinase, in addition to signaling through Ca@[email protected], the thapsigargin-induced MAP kinase activation (data not shown). elevated Src kinase activity persisted (Fig. 1) even after the calcium Our results presented here, showing that Src mediates thapsigargin transient subsided (2). Therefore, Src could be a cellular component induced activation of Raf-l and MAP kinase, provide a possible that potentiates the effect of thapsigargin/Ca@2@. As in the case of biochemical basis for the tumorigenicity of thapsigargin. Because thapsigargin-induced MAP kinase activation, the requirement for Src many other Ca2@-induced cellular responses may also use a similar is strongly supported by results showing that kinase-deficient Src mechanism, continued elucidation of the mechanism by which thap blocks the thapsigargin-induced activation of epitope-tagged ERK2 sigargin/Ca2@ activates Src should provide more insights into onco activity (Fig. lB). It has recently been shown that Ca2F may use Src genic signaling pathways. during differentiation of keratinocytes (28) and neurite outgrowth in PC12 cells (29). Our results now extend these studies by showing that Src is also implicated in the action of a Ca24@-dependent tumor ACKNOWLEDGMENTS promoter. We thank L. Hill for assistance with the manuscript; 0. Bowie for photo Although Src can activate MAP kinase, the mechanism by which graphic assistance with the figures; and J. Brugge, G. Johnson, U. Rapp, and this occurs has not been elucidated. Although a number of studies D. Fosterfor generouslyprovidingreagents. have suggested that Src cooperates with Ras in some cellular signaling events (see Ref. 29), there is also evidence indicating that both can activate independent pathways (30). For example, LPA and BK have REFERENCES been shown to induce MAP kinase activation possibly via direct 1. Thastrup. 0., Dawson, A. P., Scharff, 0., Foder, B., Cullen, P. J., Drobak, B. K., interaction of Pyk2 and Src (31). Although both LPA and BK can Bjerrum, P. J., Christensen, S. B., and Hanley, M. R. Thapsigargin, a novel molecular probe for studying intracellular calcium release and storage. Agents Actions, 43: elicit Ca2@ transients, the activation of MAP kinase may be largely 187â€”193,1994. the consequence of receptor-mediated stimulation. Thus, LPA or BK 2. Chao, T-S. 0., Byron, K. L., Lee, K-M., Villereal, M., and Rosner, M. R. Activation of MAP kinase by calcium-dependent and calcium-independent pathways. J. Biol. binding to 0.- or Gq-coupled receptors could trigger interaction of Chem.,267: 19876â€”19883,1992. Pyk2 with Src, subsequently linking Grb2 and Sos. The 6â€”9-fold 3. Putney, J. W., Jr., and Bird, G. S. The signal for capacitative calcium entry. Cell, 75: activation of MAP kinase by LPA or BK is similar to that observed for 199â€”201,1993. 4. Thomas, D., and Hanley, M. R. Evaluation of calcium influx factors from stimulated other receptor-mediated MAP kinase activation. In addition, a Jurkat T-lymphocytes by microinjection into Xenopus oocytes. J. Biol. Chem., 270: calcium/calmodulin-dependent protein kinase IV has recently been 6429â€”6432,1995. 5. Putney, J. W., Jr. Excitment about calcium signaling in inexcitable cells. Science shown to activate three epitope-tagged MAP kinases, ERK2, JNK-l, (Washington DC), 262: 676â€”678,1993. and p38, to various extents in PC-l2 cells (32). Whether Src can 6. Cao, X., Thou, Y., and Lee, A. S. Requirement of tyrosine- and serine/threonine 3172

kinases in the transcriptional activation of the mammalian grp78BiP promoter by V. P. Hypoxic induction of human vascular endothelial growth factor expression thapsigargin. J. Biol. Chem.. 270: 494â€”502, 1995. through c-Sec activation. Nature (Land.), 375: 577â€”581,1995. 7. Chao, 1-S. 0., Qureshi, S. A., Hendrickson, M., Foster, D. A., Rapp, U., and Rosner, 22. Cooper, J. A., Each, F. S., Taylor, S. S., and Hunter, T. Phosphorylation sites in M. R. Differential Raf-1 requirement for activation of MAP kinase by growth factors enolase and lactate dehydrogenase utilized by tyrosine protein kinases in vivo and in and tumor promoters. J. Biol. Chem., 269: 7337â€”7341, 1994. vitro. J. Biol. Chem., 259: 7835â€”7841,1984. 8. Takishima,K.,Griswold-Prenner,I.,Ingebritsen,1.,andRosner,M. R. Epidermal 23. Courtneidge, S. A. Activation ofthe pp60c-src kinase by middle T antigen binding or growth factor (EGF) receptor T669 peptide kinase from 3T3â€”Ll cells is an EGF by dephosphorylation. EMBO J., 4: 1471â€”1477,1985. stimulated â€œMAPâ€•kinase.Proc. Nail. Acad. Sd. USA, 88: 2520-2524, 1991. 24. Meloche, S., Pages, G., and Pouyssegur, J. Functional expression and growth factor 9. Takishima, K., Friedman, B.. Fujiki. H., and Rosner, M. R. Thapsigargin, a novel activation of an epitope-tagged p44 mitogen-activated protein kinase, p44mapk. Mol. promoter, phosphotylates the epidermal growth factor receptor at threonine 669. Biol. Cell, 3: 63â€”71,1992. Biochem. Biophys. Rca. Commun., 157: 740â€”746, 1988. 25. Russell, M., Winitz, S., and Johnson, G. L. Acetylcholine muscarinic ml receptor 10. Griswold-Prenner, I., Carlin, C. R., and Rosner, M. R. MAP kinase regulates the EGF regulation of cyclic AMP synthesis controls growth factor stimulation of Raf activity. receptor through activation of a tyrosine phosphatase. J. Biol. Chem., 268: 13050â€” Mol.Cell.Biol.,14:2343â€”2351,1994. 13054, 1993. 26. Hershenson, M. B., Chao, T-S., 0., Abe, M. K., Solway, J.. and Rosner, M. R. 11. Morrison, P., Saltiel, A. R., and Rosner, M. R. Role of MAP kinase kinase in Histamine antagonizes serotonin and growth factor-induced MAP kinase activa regulation of the EGF receptor by protein kinase C. J. Biol. Chem., 271: 12891- lion in bovinetrachealsmoothmusclecells.J. Biol.Chem.,270: 19908â€”19913, 12896, 1996. 1995. 12. Friedman,B.,VanAmstenlam,J.,Fujiki,H.,andRosner,M.R. Phosphorylationat 27. Friedman, B., Fujiki, H., and Rosner, M. R. Regulation ofthe epidermal growth factor threonine-654 is not required for negative regulation of the epidermal growth factor receptor by growth modulating agents. Cancer Res.. 50: 533â€”538,1990. receptor by non-phosboltumor [email protected]. Aced. Sd. USA, 86: 812-816, 28. Thao, Y., Sudol, M., Hanafusa, H., and Krueger, J. Increased tyrosine kinase activity 1989. of c-Sec during calcium-induced keratinocyte differentiation. Proc. NatI. Acad. Sci. 13. Momson, P., TakIShiIna, K., and Rosner, M. R. Role of threonine residues in USA, 89: 8298â€”8302, 1992. regulation of the epidermal growth factor receptor by protein kinase C and mitogen 29. Rusanescu, G., Qi, H., Thomas, S. M., Brugge, J. S., and Halegoua, S. Calcium influx activated protein kinase. J. Biol. Chem., 268: 15536-15543, 1993. induces neurite growth through a Src-Ras signaling cassette. Neuron, 15: 1415â€”1425, 14. Marshall, C. J. Specificity of receptor tymsine kinase signaling: transient versus 1995. sustained extracellular signal-regulated kinase activation. Cell, 80: 179â€”185,1995. 30. Barone,M.V.,andCourtneidge,S.A. Mycbutnot Fosrescueof PDGFsignalling 15. Lange-Carter. C. A., Pleiman, C. M., Gardner, A. M., Blumer, K. J., and Johnson, block caused by kinase-inactive Sec. Nature (Land.), 378: 509â€”512, 1995. G. L A divergence in the MAP kinase regulatory network defined by MEK kinase 31. Dikic, I., Tokiwas, 0., Lev, S., Courtneidge, S. A., and Schlessinger, J. A role for and raf. Science (Washington DC), 260: 315â€”319,1993. Pyk2 and Sec in linking G-protein-coupled receptors with MAP kinase activation. 16. Johnson,0. L, Gardner,A. M., Lange-Carter,C.,Qian,N. X., Russell,M., and Nature (Lond.), 383: 547â€”550,1996. Winitz, S. How does the G protein, 0i2, transduce mitogenic signals. J. Cell 32. Enslen, H., Tolumitsu, H., Stork, P. J., Davis, R. J., and Soderlmg, T. R. Regulation Biochem., 54: 415â€”422,1994. of mitogen-activated protein kinases by a calcium/calmodulin-dependent protein 17. Taylor,S.J., andShalloway,D.Secandthecontrolofcelldivision.BioEssays,18: kinase cascade. Proc. Ned. Acad. Sci. USA, 93: 10803-10808, 1997. 9â€”11,1996. 33. Cleghon, V., and Monison, D. K. Raf-l interacts with Fyn and Sec in a non 18. Eves, E. M., Tucker, M. S., Roback, J. D., Downen, M., Rosner, M. R., and phosphotyrosine-dependent manner. 3. Biol. Chem., 269: 17749â€”17755,1994. Warner, B. H. Immortal rat hippocampal cell lines exhibit neuronal and glial 34. Erhardt, P., Troppmair, J., Rapp, U. R., and Cooper, 0. M. Differential regulation of lineages and neurotrophingene expression.Proc. Natl. Acad. Sd. USA, 89: Raf-l and B-Raf and Ras-dependent activation of mitogen-activated protein kinase by 4373â€”4377, 1992. cyclic AMP in PC12 cells. Mol. Cell. Biol., 15: 5524â€”5530, 1995. 19. Tucker,M. S., Eves, E. M., Wainer,B. H., and Rosner,M. R. Activationof 35. Faure, R., Baquiran, G., Bergeron, J. J. M.. and Posner, B. I. The dephosphorylation mitogen-activated protein kinase by epidermal growth factor in hippocampal neurons of insulin and epidermal growth factor receptors. I. Biol. Chem., 267: 11215â€”11221, and neuronal cell lines. J. Neurochem., 61: 1376â€”1387, 1993. 1992. 20. Qureshi.S. A., Joseph,C. K., Hendrickson,M.,Song,J., Gupta,R., Bruder,J., 36. Bading,H., Ginty,D. D., and Greenberg,M.E. Regulationofgeneexpressionin Rapp, U., and Foster, D. A. A dominant negative Raf-l mutant prevents hippocampal neurons by distinct calcium signaling pathways. Science (Washington V-Sec-induced transformation. Biochem. Biophys. Res. Commun., 192: 969â€”975, DC), 260: 181â€”186,1993. 1993. 37. Ghosh, A., and Greenberg, M. E. Calcium signaling in neurons: molecular mecha 21. Multhopadhyay, D., Tsiokas, L, Thou, X-M., Foster, D., Brugge, J. S., and Sukhatnse, nisms and cellular consequences. Science (Washington DC), 268: 239â€”247, 1995.

3173

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1997 American Association for Cancer Research. Src Tyrosine Kinase Mediates Stimulation of Raf-1 and Mitogen-activated Protein Kinase by the Tumor Promoter Thapsigargin

Tsung-Shu Oliver Chao, Mark Abe, Marc B. Hershenson, et al.

Cancer Res 1997;57:3168-3173.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/57/15/3168

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/57/15/3168. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.