Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Growth factor activation of MAP kinase requires cell adhesion

Growth factor activation of MAP kinase requires cell adhesion The EMBO Journal Vol.16 No.18 pp.5592–5599, 1997 Growth factor activation of MAP kinase requires cell adhesion Ras, Raf (Leevers et al., 1994), or MEK1 (Crowley et al., Mark W.Renshaw, Xiang-Dong Ren and 1994; Mansour et al., 1994) is transforming. Furthermore, Martin Alexander Schwartz inhibition of Ras, Raf, MEK1 or ERKs inhibits normal Department of Vascular Biology, The Scripps Research Institute, and transformed cell growth (Feramisco et al., 1985; 10550 N. Torrey Pines Road, La Jolla, CA 92037, USA Mulcahy et al., 1985; Feig and Cooper, 1988; Kolch et al., Corresponding author 1991; Crowley et al., 1994; Khosravi-Far et al., 1995). e-mail: [email protected] Ras, however, has a number of other effectors such as Rho family GTPases, PI 3-kinase and Ral GDS, which The MAP kinase pathway is a major regulator of also make important contributions to cell cycle progression both normal and oncogenic growth. We report that and transformation (Rodrigez-Viciana et al., 1994; activation of the MAP kinase ERK2 by serum or Khosravi-Far et al., 1995; Qiu et al., 1995; White et al., purified growth factors is strongly dependent on cell 1996). adhesion to extracellular matrix proteins. This effect It is well established that normal cell growth requires is specific to soluble growth factors, since suspended cell adhesion to extracellular matrix proteins as well as cells still activate ERK2 in response to plating on stimulation by growth factors (reviewed in Clark and fibronectin, and is reversible. Analysis of endogenous Brugge, 1995; Schwartz et al., 1995). The ability of Ras and Raf show that these proteins are still activated integrins to transduce a variety of intracellular signals by serum in suspended cells, whereas MEK activity is most likely mediates this effect. One such signaling inhibited. Conversely, activation of ERK2 by activated pathway involves the MAP kinases ERK1 and ERK2. mutants of Ras and Raf is still adhesion-dependent but Plating of suspended cells on extracellular matrix proteins activation by MEK is not. Consistent with these results, such as fibronectin (FN) triggers a rapid, transient activ- activated MEK enhances growth of ras-transformed ation of MAP kinases (Chen et al., 1994; Morino et al., cells in suspension but not when adherent. These results 1995; Zhu and Assoian, 1995). However, it is not at all identify a novel synergism between cell adhesion- and clear how this transient phenomenon contributes to growth growth factor-regulated pathways, and explain how of stably adherent cells. oncogenic activation of MAP kinases induces both Conversely, how constitutive activation of ERKs leads serum- and anchorage-independent growth. not only to serum-independent but also to anchorage- Keywords: cell adhesion/growth factor regulation/ independent growth is poorly understood. Anchorage- integrin/signal transduction independence is the in vitro characteristic that correlates best with tumorigenicity in vivo (Freedman and Shin, 1974). Two groups have analyzed the effect of cell adhesion on growth factor activation of MAP kinase and Introduction found that placing 3T3 cells in suspension had either a modest effect (Zhu and Assoian, 1995) or no effect The MAP kinase pathway is a major cellular signaling (Hotchin and Hall, 1995) on activation of ERKs by growth pathway that mediates effects of growth factors on cell factors. These results are paradoxical, since growth factors cycle progression (reviewed in Crews and Erikson, 1993; do not induce DNA synthesis in suspended cells, yet Cobb et al., 1994; Marshall, 1995). Growth factors induce activated variants of Ras, Raf or MEK1 do. conversion of c-Ras to its GTP-bound form, which then To resolve this paradox, we investigated in greater binds the protein kinase c-Raf. Localization of Raf to the detail the involvement of cell adhesion in growth factor plasma membrane leads to its activation via mechanisms activation of the MAP kinase pathway. We found that that are incompletely understood but that appear to involve activation of ERK2 was in fact strongly dependent on its phosphorylation (Morrison et al., 1988; Jelinek et al., adhesion to extracellular matrix. Furthermore, we identi- 1996). Once activated, Raf phosphorylates and activates fied activation of MEK by Raf as the adhesion-dependent the protein kinase MEK1, which phosphorylates and step. Thus, constitutive activation of the MAP kinase activates the MAP kinases ERK1 and ERK2. These kinases pathway by oncogenes can bypass the requirement for phosphorylate a number of substrates that participate in both growth factors and cell adhesion. cell cycle regulation, including the transcription factor Elk-1 (Janknecht et al., 1993), phospholipase A (Lin RSK et al., 1993) and p90 (Sturgill et al., 1988; Chung Results et al., 1991). This pathway also plays a key role in oncogenic Inhibition of ERK2 activity in suspended cells transformation, as activated forms of Ras are among the To test the possible dependence of the MAP kinase ERK2 most common mutations found in human tumors (Cobb on adhesion, cells were detached and either held in et al., 1994; Marshall, 1995) and expression of activated suspension in medium with 0.4% serum for 24 h or 5592 © Oxford University Press MAP kinase activation requires cell adhesion Fig. 1. ERK2 activity in adherent and suspended cells. (A) Stimulation by serum. Cells that had been incubated for 24 h in 0.4% serum while attached or suspended were stimulated by addition of 10% serum for the indicated times. Endogenous ERK was immunoprecipitated and analyzed for kinase activity using the in-gel assay and for ERK2 protein by Western blotting. Quantitation of the kinase assay is shown graphically. Data are representative of eight experiments. (B) Stimulation by PDGF. Attached or suspended cells that were incubated in 0.4% serum were stimulated by addition of 20 ng/ml PDGF for the indicated times. Endogenous ERK2 activity and amounts of immunoprecipitated protein were assayed as in (A). Quantitation of the kinase assay by densitometry is shown graphically. Data are representative of three experiments. (C) Stimulation by LPA. Attached or suspended cells were incubated in 0.4% serum, then stimulated with 2 μg/ml LPA for the indicated times. Endogenous ERK2 was immunoprecipitated and its activity determined in the immunoprecipitates. Kinase activity is shown graphically. Data are representative of four experiments. (D) Stimulation by TPA. Adherent cells kept in low serum for 24 h, serum-starved cells that were trypsinized and kept in suspension for 3 h, or cells kept in suspension in low serum for 24 h were stimulated with 100 nM TPA for 10 min. ERK2 kinase activity and immunoprecipitated protein amounts were assayed as before. The lower panel shows the quantitation of ERK2 kinase activity normalized for protein levels; values are means  standard deviations from three experiments. replated on tissue culture plastic in low serum for the sion for short periods of time still activate MAP kinase same period. Subsequent stimulation with 10% serum in response to growth factors (Hotchin and Hall, 1995; induced strong activation of ERK2 in adherent cells but Zhu and Assoian, 1995). To resolve the conflict between only a slight increase in suspended cells (Figure 1A). In our results and these reports, we analyzed the effect of eight experiments, inhibition varied between 97% and leaving cells in suspension for various lengths of time. 77%, averaging 90.2  9.8%. Platelet-derived growth We found that cells which were detached and held in factor (PDGF) and lysophosphatidic acid (LPA) are two suspension for short periods of time retained the ability of the major mitogens in serum. We therefore examined to activate ERK2 in response to serum (Figure 2). With the stimulation of ERK2 activity by purified PDGF and longer times in suspension, ERK2 activation diminished, LPA. Both mitogens strongly activated ERK2 in adherent with a half-time of ~6 h. These results indicate that loss cells but showed a substantially reduced response in non- of structures such as focal adhesions or dephosphorylation adherent cells (Figure 1B and C) of focal adhesion proteins such as focal adhesion kinase, cas Previous studies have shown that cells placed in suspen- paxillin, p130 or tensin, which occur rapidly upon loss 5593 M.W.Renshaw, X.-D.Ren and M.A.Schwartz Fig. 2. Time in suspension. Cells were kept attached in 0.4% serum or were trypsinized and maintained in suspension for varying periods of time. Total time in 0.4% serum was kept constant by varying the length of starvation in low serum prior to detachment. Cells were then stimulated with 10% serum for 60 min and MAP kinase assayed. Values are means  standard deviations from four experiments. Similar results were obtained with cells stimulated with serum for 15 min (not shown). of cell adhesion, are not likely to account for the failure of suspended cells to activate the MAP kinase pathway. Consistent with this idea, treatment of cells with cytochala- sin D for up to 90 min, which disrupts actin filaments and causes loss of focal adhesions, did not inhibit MAP kinase activation by serum (not shown). A prolonged period of ‘adhesion deprivation’ is required, analogous to the prolonged serum starvation that is required to de-activate growth factor-regulated pathways. Fig. 3. Recovery of MAP kinase activation. (A) Replating in the It is noteworthy that both LPA and PDGF, which bind absence of serum. Cells were incubated in suspension for3hor24h; a G protein-linked receptor and a protein tyrosine kinase total time in low serum was approximately constant. Cells were then receptor, respectively, are affected in the same manner by replated on plastic coated with 25 μg/ml FN. ERK2 activity was loss of cell adhesion. These observations argue against assayed as before. Values are means  standard deviations from three the likelihood that changes in receptor expression or experiments. (B) Response to serum. Cells were maintained in suspension in low serum for 24 h, then replated on plastic coated with function can account for the failure of non-adherent cells 20 μg/ml FN (upper panel). At the indicated times, cells were to activate ERK2. Consistent with this idea, previous work stimulated with serum for 15 min and MAP kinase assayed. has shown that the PDGF receptor is still functional Alternatively, as shown in the lower panel, cells were replated on even after prolonged incubation of cells in suspension plastic with different coating. These were: 20 μg/ml FN; 100 μg/ml polylysine (PL); 20 μg/ml anti-mouse β integrin IgG (β1); 25 μg/ml (McNamee et al., 1992). To test this idea further, we goat anti-mouse IgG followed by 20 μg/ml anti-mouse β IgG (2nd examined the activation of ERK2 by the phorbol ester 1 β1). In these experiments, the plastic was subsequently blocked with TPA, which activates Raf via protein kinase C, bypassing 10 mg/ml heat-denatured BSA to prevent adhesion via secreted FN. Ras and the membrane receptors. The ability of TPA to After 3 h, cells were stimulated with 10% serum for 15 min and MAP activate ERK2 was also greatly attenuated in suspended kinase assayed. Four experiments yielded similar results. Similar results were obtained with cells stimulated with serum for 60 min (not cells in a time-dependent manner (Figure 1D). shown). Reversibility To examine whether MAP kinase activation could be stimulus (adhesion). Thus, the failure to respond to serum restored by replating cells on the extracellular matrix is specific to soluble growth factors and cannot be due to protein FN, two experiments were performed. First, cells general toxicity or apoptosis. that were kept in suspension for either 3 or 24 h were To test whether replating restores the ability to respond replated on FN. The cells that had been in suspension for to serum, 24 h suspended cells were replated on FN in 24 h adhered and spread at a significantly slower rate, low serum for various times and then stimulated with though in both cases cells were well spread by 60 min serum. ERK2 activity was then assayed. We observed that (data not shown). Assay of ERK2 activity in these cells upon replating, cells regained the ability to activate ERK2 showed that the activation of ERK2 in cells kept in in response to serum (Figure 3B). The activation by suspension for 24 h was delayed relative to cells kept in replating alone in the absence of serum was, as in Figure suspension for 3 h (Figure 3A). Presumably this delay is 3A, readily detectable at 1 h, followed by a return to due to the delay in attachment and spreading. However, baseline by 2 h. However, the increase in activity due to the extent of ERK2 activation was equivalent in the cells serum was readily detected at the 1 h time point. Interest- suspended for 24 h. This result shows that ERK2 can still ingly, the serum response was diminished at the 2 h time be activated in 24 h suspended cells by an appropriate point. This effect is most likely due to the fact that 5594 MAP kinase activation requires cell adhesion Fig. 5. Raf kinase activity. Attached or suspended cells incubated in low serum for 24 h were stimulated with 10% serum, then lysed and c-Raf immunoprecipitated. Raf kinase activity, using kinase-defective Fig. 4. Ras GTP loading. Attached or suspended cells in low serum MEK-1 as a substrate, and Raf protein levels were assayed (A)as were labeled with P-orthophosphate and stimulated with serum for described in Materials and methods. Kinase activity was normalized 8 min. Endogenous Ras was immunoprecipitated and the bound for immunoprecipitated protein and data presented graphically (B). nucleotide was analyzed by chromatography (A). Values shown Values are means  standard deviations from three experiments. graphically (B) are means  standard deviations from four experiments. Next, we assayed Raf function by immunoprecipitating stimulation of MAP kinases by replating alone may render cellular Raf and carrying out an in vitro kinase assay using cells refractory to subsequent stimulation by serum, due kinase-defective MEK1 as a substrate. These experiments to induction of the ERK phosphatase MKP-1 (Sun et al., showed that Raf kinase was still activated in suspended 1993). Recovery of MAP kinase activation was also cells, with only a slight (~10%) decrease relative to observed when cells were replated on an antibody to the adherent cells (Figure 5). Thus, activation of Raf is not integrin β1 subunit, but not on polylysine. Treatment of significantly adhesion-dependent. cells with cycloheximide did not block the recovery of We then immunoprecipitated cellular MEK1, and MAP kinase activation (data not shown), but did reduce examined its kinase activity using kinase-defective ERK2 cell spreading and slightly reduced MAP kinase activation. as a substrate. These experiments revealed that activation These results show that the inhibition of ERK2 in sus- of endogenous MEK1 in response to serum was greatly pended cells is reversible and is integrin-dependent. reduced in non-adherent cells (Figure 6), to an extent that was similar to the inhibition of ERK2. Thus, the ability Activation of intermediates of Raf to activate MEK1 appears to be the step that is It is well established that growth factors activate ERK2 sensitive to adhesion. via a pathway that involves Ras, Raf and MEK1. To determine which step(s) in the pathway require cell adhe- Activated components sion, we assayed the activation of these components in To confirm the results of experiments analyzing the adherent and suspended cells. Ras activation was assayed activities of cellular Ras, Raf and MEK1, we carried out by immunoprecipitating endogenous Ras protein and ana- a second series of experiments to study the activation of lyzing the bound nucleotide. We observed that the baseline ERK2 by constitutively activated mutants. Cells were GTP content was somewhat lower in suspended cells; transiently transfected with ras G12V, which is constitu- however, treatment with serum stimulated GTP loading tively activated by a mutation that reduces its GTPase; to the same level in adherent and suspended cells with raf BxB, in which the kinase is constitutively activated (Figure 4). Thus, the adhesion-dependent step must lie by truncation of the regulatory domain; or by ΔN3, S222 downstream of Ras. MEK1, which is activated by a deletion and a point 5595 M.W.Renshaw, X.-D.Ren and M.A.Schwartz Fig. 6. MEK kinase activity. Attached or suspended cells incubated in Fig. 7. Effect of activated mutants. NIH 3T3 cells were co-transfected low serum for 24 h were stimulated with 10% serum, then lysed and with HA-tagged ERK2 together with activated mutants of Ras, Raf or MEK-1 immunoprecipitated. MEK kinase activity was assayed using MEK. At 24 h after transfection, cells were replated on tissue culture kinase-defective ERK2 as a substrate (top panel), and plastic in 0.4% serum or placed in suspension in 0.4% serum. After an immunoprecipitated MEK-1 protein was analyzed by Western blotting additional 24 h, cells were harvested and the transfected ERK2 (lower panel). For graphical data, kinase activity was normalized for immunoprecipitated using anti-HA antibody. Both the kinase activity immunoprecipitated MEK-1 protein. Values are means  standard of the precipitated ERK2 and the ERK2 protein levels were assayed as deviations from three experiments. described in Materials and methods. Negligible ERK2 activity was observed in the absence of co-transfected activated Ras, Raf or MEK. Values shown in the lower panels are kinase activity normalized for mutation. In each case, the activated protein was co- ERK2 protein, averaged from three experiments  standard deviations. transfected with an ERK2 that had been tagged with the hemagglutinin (HA) epitope. Control cells were co- transfected with HA–ERK2 plus empty vector. At 24 h with MEK alone induced foci rather weakly, as reported after transfection, cells were placed in suspension or previously (Khosravi-Far et al., 1995; White et al., 1995). replated on tissue culture plastic in low serum, and This result is consistent with the concept that ras induces incubated a further 24 h. The HA–ERK2 was then immuno- transformation via combined effects on a number of precipitated and its activity assayed. These results show effectors. Co-transfection of MEK with ras did not increase that ERK2 activation by both Ras and Raf were decreased the number of foci or colonies, but this may be due to in suspended cells to ~15% of the levels in adherent cells the already high efficiency of transformation by ras alone. (Figure 7). By contrast, activation of ERK2 by activated However, MEK substantially increased the size of the MEK1 remained high in non-adherent cells. These results colonies in soft agar (Table II). Measurement of their confirm that the adhesion-dependent step lies between Raf volume showed that MEK1 enhanced the rate of growth and MEK. It should be noted, however, that the activation of cells in suspension by ~7-fold. By contrast, MEK1 did of ERK2 by these oncogenes in adherent cells is ~2.5- not increase the size of adherent ras-transformed colonies fold higher than that triggered by serum in adherent in low serum. These results show that preventing the cells (not shown). Hence, the activation by oncogenes in decline in MAP kinase in non-adherent cells significantly suspended cells, while strongly attenuated, is ~40% of augments their growth in suspension but does not enhance that found in adherent cells treated with serum. the ability of ras to promote serum-independent growth in adherent cells. Cell growth The results obtained above make a surprising prediction. Discussion Present views of the MAP kinase pathway would argue against the idea that activated MEK1 should augment Our results show that in NIH 3T3 cells, activation of the transformation by Ras. However, our data suggest that MAP kinase pathway by serum or growth factors is expression of MEK1 should enhance MAP kinase activity strongly dependent on integrin-mediated cell adhesion to and growth in suspended cells. To test this hypothesis, extracellular matrix protein. Similar results were obtained cells were transfected with activated ras alone, activated with CHO cells (P.E.Hughes and M.H.Ginsberg, personal MEK1 alone, or both. Growth of transfected cells in communication), but other cell types have yet to be monolayer and in suspension was then examined. examined. Several results demonstrate that this effect is As shown in Table I, ras induced a significant number due to a specific blockade rather than to general toxicity: of foci in monolayer culture and a similar number of suspended cells retain the ability to activate MAP kinase colonies in soft agar. The number of foci and colonies in response to adhesion, they retain the ability to activate was nearly equal to the total number of hygromycin- Ras and Raf in response to growth factors, and the effect resistant colonies, suggesting that ras induced transform- is reversed when cells are replated on FN. ation of NIH 3T3 cells with high efficiency. Transfection Two independent lines of experimentation show that 5596 MAP kinase activation requires cell adhesion Table I. Stable transformation assay Vectors HygR colonies Foci in 10% CS Foci in 0.5% CS Soft agar colony Control 1340  60 0 00 00  0 MEK 1120  45 45  11 11  11 0  0 Ras 1100  45 1584  88 1408  220 1452  44 Ras  Mek 1260  98 1537  227 1588  151 1411  101 NIH 3T3 cells were co-transfected with RSVHyg and combinations of the plasmids expressing the activated MEK or Ras mutant proteins or the empty expression vector (Control). Relative DNA amounts of the mutants were kept constant (0.8 μg) in the transfections, so in transfections containing only one of the mutants the remainder of the DNA was made up by the empty expression vector. Transfected cells were trypsinized and portions replated in media containing 10% CS, 10% CS with 200 μg/ml hygromycin, 0.5% CS, or soft agar to measure foci and colony formation. Values shown represent the total number of colonies or foci calculated for the entire transfection and represent values obtained from two separate experiments. this effect occurs at the level of activation of MEK by Table II. Soft agar and minimal media colony size Raf. First, as mentioned above, endogenous Ras and Raf are still maximally activated by serum in non-adherent Vector Soft agar Foci in 0.5% CS cells, but MEK1 is not. Second, expression of activated Colony Relative Foci size Relative mutants of these proteins showed that signals from onco- volume genic Ras and Raf were still attenuated in suspended cells, whereas the signal from activated MEK1 was Ras 0.59  0.17 1.00 87  7 1.00 undiminished. Thus, the ability of Raf to activate MEK1 Ras  MEK 4.12  1.07 6.99 81  6 0.91 maximally must require cell adhesion. Finally, the bio- Transfected cells from Table I were further analyzed by visually logical relevance of these effects was indicated by the measuring soft agar colony size and calculating the volume (4/3πr ). result that activated MEK1 significantly enhanced the Values represent the mean  standard error from 15 colonies each. growth of ras-transformed cells in suspension, but had no Minimal media focus size was determined by trypsinizing the foci and replating in soft agar to determine the number of transformed cells effect on their growth when adherent. per focus. Values represent the mean  standard deviation from two Previous studies have been puzzling in that either no separate experiments in which at least 35 foci from each transfection effect or a modest effect of cell adhesion on the activation were trypsinized. of the MAP kinase pathway by growth factors was observed. Yet transforming activators of this pathway, et al., 1994), or as yet unidentified protein kinases or such as mutated forms of Ras, Raf and MEK1, induce phosphatases that may regulate MEK1 activity. Further growth of cells in suspension. If MAP kinase were fully work will be required to resolve these issues. activated by growth factors in normal, suspended cells, The majority of key cellular decisions about growth, then the activation of MAP kinase by oncoproteins would differentiation and survival are made on the basis of input be redundant. Our results resolve these questions and from both soluble mediators (growth factors, cytokines, suggest a model relating MAP kinase to anchorage- hormones) and from extracellular matrix receptors. Thus, dependence and -independence of growth. Normal cells understanding how cellular pathways are controlled by are anchorage-dependent in part because growth factor both soluble factors and cell adhesion to extracellular activation of MAP kinase depends on cell adhesion. matrix is a major goal. Our results identify a novel Conversely, cells transformed by ras and other oncogenes synergism between cell adhesion and growth factors in that work through the MAP kinase pathway are anchorage- the activation of the MAP kinase pathway. These results independent because activation of MAP kinase bypasses are likely to be important in a variety of systems where a requirement for both adhesion and growth factors. gene expressions, cell growth and cell survival depend This model, however, raises the question why activated upon cell adhesion. Ras or Raf are able to induce growth in suspension at all. It must be noted that the reduction of the ERK2 signal in non-adherent cells is not complete (Figure 7). Indeed, the Materials and methods activation of the MAP kinase pathway by these oncogenes Reagents is so strong that the levels of ERK activity in suspended Anti-ERK2 polyclonal antibody (C-14) was purchased from Santa Cruz cells are still significant, ~40% of the level in adherent, Biotechnology (Santa Cruz, CA). LipofectAMINE, Dulbecco’s modified serum-treated control cells. Thus, very strong activation Eagle’s medium (DMEM), serum and other reagents for cell culture of the pathway can at least partially overcome the require- were from Gibco-BRL (Gaithersburg, MD). Anti-mouse β1 integrin IgG ment for adhesion. (HMβ1-1) was purchased from Pharmingen (San Diego, CA). Fibronectin was prepared from human plasma as described (Miekka et al., 1982). The mechanism by which cell adhesion potentiates the Myelin basic protein was prepared from bovine spinal cord as described activation of MEK by Raf is currently unknown. The (Deibler et al., 1972). Platelet-derived growth factor (PDGF), lysophos- existence of focal adhesions or the phosphorylation of phatidic acid (LPA), agarose, methylcellulose and other reagents were focal adhesion proteins are not likely to be the immediate purchased from Sigma Chemicals (St. Louis, MO) unless otherwise noted. cause, since they are rapidly lost upon detachment whereas Cell culture the decline in ERK2 activation occurs more slowly NIH 3T3 cells were cultured in DMEM supplemented with 10% bovine (Figure 1D). Possible explanations include the regulation calf serum. For MAP kinase experiments, cells in DMEM with 0.4% of scaffolding proteins that control the proximity of Raf serum were plated in 60 mm tissue culture plastic dishes at ~80% of and MEK, analogous to the Ste5 protein in yeast (Choi confluence for adherent cells. Alternatively, equal numbers of cells in 5597 M.W.Renshaw, X.-D.Ren and M.A.Schwartz DMEM with 0.4% serum and 0.5% methylcellulose were plated in scanning densitometry using the deskscan software with a Scanjett IIP 10 cm dishes coated with 1 ml of 1% agarose which was equilibrated scanner (Hewlett Packard). with DMEM. Methyl cellulose reduces cell–cell aggregation but does not otherwise affect growth or MAP kinase activity (data not shown). Measurement of MEK 1 activity Cells were incubated as described in the text, stimulated with 10% serum Endogenous MEK 1 was immunoprecipitated from 100 μg of cell lysate or other factors, harvested and analyzed as described below. using 1 μg of anti MEK 1 (Santa Cruz Biotechnology). Immunoprecipit- ates were washed three times in lysis buffer, and then once in kinase Plasmids and transfection buffer (Chen et al., 1996) 10 mM Tris, pH 7.5, 10 mM MgCl ,1 mM For transfections, cells were plated at a density of 410 cells per 6 cm DTT. IPs were then split, using one-fifth to measure the amount of MEK dish 24 h before transfection. Cells were transfected with LipofectAMINE 1 protein and four-fifths to measure kinase activity. MEK kinase activity as described (Renshaw et al., 1996) using 0.2 μg of pCMV5 ERK2, 32 was measured in kinase buffer containing 25 μMATP,5 μCi [γ- P]ATP 0.2 μg of pCMV5 βgal, and 1.6 mg of either pDCR ras G12V (White and 2 μg of kinase-dead GST–ERK2 (Hipskind et al., 1994) for 30 min et al., 1995), pZIP Neo raf BXB (Bruder et al., 1992) or pMCL MEK at room temperature. Samples were electrophoresed on 10% SDS– ΔN3, S222D (Mansour et al., 1994) per plate. At 24 h after transfection, polyacrylamide gels, which were then dried and exposed to film. cells were transferred to medium containing 0.5% serum for an additional Autoradiographs were quantitated using a model I.S. 1000 digital imaging 24 h for adherent cells and for cells suspended for only 3 h. After 24 h system from Alpha-Innotech Corp. in 0.5% CS, cells from indicated plates were then trypsinized and suspended for 3 h in serum-free DMEM containing 0.1% BSA Ras-GTP loading (Calbiochem, nuclease-, protease-free), and 0.25 mg/ml soybean trypsin NIH 3T3 cells were maintained adherent in 0.4% serum or in suspension inhibitor (Sigma) over dishes which had been coated with 1% heat- as before for 12 h. Cells were then transferred to phosphate-free DMEM denatured BSA (Sigma fraction V). Alternatively, for cells kept in containing 0.4% serum and 50 μCi/ml P-labeled orthophosphate for suspension for 24 h, cells were trypsinized 24 h after transfection and an additional 12 h. Cells were then rinsed with cold Tris-buffered saline placed in suspension in DMEM media containing 0.5% methyl cellulose or phosphate-free DMEM and stimulated with 10% serum for 6– and 0.4% calf serum in dishes coated with 1% agarose as above. 10 min at room temperature. (Preliminary experiments demonstrated that maximal GTP loading of Ras occurred at this time interval.) Cells Cell growth and transformation assays were then rinsed three times with cold PBS, and lysed in 380 μlof For stable transformation assays, cells were transfected with RSVHyg 50 mM HEPES, pH 7.4, containing 1% NP40, 0.1% BSA, 100 mM and 1.6 μg of either the empty control vector, pDCR Ras G12V, pMCL NaCl, 15 mM MgCl , 0.1 mM GTP, 0.1 mM GDP, 1 mM ATP, 10 μg/ml MEK ΔN3, S222D, or combinations thereof. After 24 h, the cells were aprotinin and leupeptin, 1 mM PMSF, and 2 μg of either a rat monoclonal changed to fresh media and allowed to grow for two more days. Cells th anti-Ras antibody (Santa Cruz Biotechnology) or non-immune rat IgG were then trypsinized and a portion (1/20 ) replated to measure focus as a control. Lysates were centrifuged at 13 000 r.p.m. for 12 min, the formation in normal media (with 10% serum), minimal media, or in soft supernatants transferred to fresh tubes and adjusted to 0.5M NaCl, 0.5% agar (with 10% serum) as previously described (Renshaw et al., 1995). deoxycholate, 0.05% SDS. They were incubated on ice for 30 min, then Cells were also replated in media containing 200 μg/ml Hygromycin 10 μl protein G PLUS-Agarose (Santa Cruz Biotechnology) added and (Boehringer-Mannheim) to determine the total number of stably trans- the samples rotated for1hat 4°C for 60 min. The beads were washed fected clones. Colonies were scored visually 2 weeks after transfections. eight times, and the bound nucleotides extracted and chromatographed Soft agar colony volume was determined by visually measuring colony on polyethyleneimine paper as described (Downward et al., 1990). diameter against a scale, from which colony volume was calculated. Chromatograms were analyzed by autoradiography on Kodak BioMax Size of foci in low serum (‘minimal medium’) was determined by MS film and quantitated by densitometry. trypsinizing the foci and replating them in soft agar. The number of soft agar colonies was then counted after 2 weeks to determine the number Raf kinase assay of transformed cells per original minimal medium focus. Adherent or suspended cells were stimulated with serum for 8 min, rinsed with cold PBS and lysed in buffer containing 50 mM HEPES, Measurement of ERK2 activity pH 7.4, 1% NP40, 100 mM NaCl, 2 mM EDTA, 10 μg/ml aprotinin Cells were stimulated as described in the Results section, rinsed with and leupeptin, 1 mM PMSF, 5 mM sodium vanadate, 20 mM NaF, cold phosphate-buffered saline (PBS) and extracted in ice-cold buffer 10 mM sodium pyrophosphate and 3 mM β-glycerophosphate. Cell containing 0.5% NP-40, 20 mM Tris, pH 7.6, 250 mM NaCl, 5 mM lysates were centrifuged at 13 000 r.p.m. in a microfuge for 15 min, and EDTA, 3 mM EGTA, 20 mM sodium phosphate, 20 mM sodium precleared with 10 μl of protein A–agarose beads (Pierce). Protein in pyrophosphate, 3 mM β-glycerophosphate, 1 mM sodium orthovanadate, each sample was assayed using the Pierce BCA Kit. Normalized samples 1 mM PMSF, 10 mM NaF, and 10 μg/ml each of leupeptin and aprotinin. Lysates were centrifuged 15 min at maximum speed in a microfuge and were adjusted to 0.5 M NaCl, 0.1% SDS and 0.5% deoxycholate, and the supernatants removed. For assays of transfected HA-ERK2 activity, immunoprecipitated for 2 h at 4°C using 3 μg of mouse monoclonal 0.5 μg of anti-HA (12CA5) antibody purified over an HA affinity column anti-Raf1 IgG (Transduction Laboratories) and 10 μl of protein A– was used for each immunoprecipitation. For endogenous ERK2, 0.5 μg agarose beads prebound with 10 μg of rabbit anti-mouse IgG (Cappel). of anti-ERK2 was used for each immunoprecipitation. Approximately The beads were rinsed three times with lysis buffer containing 0.5 M 100 μg of cell protein was immunoprecipitated in each sample. For NaCl and once with lysis buffer. The in vitro Raf kinase assay was transfected cells, the amount of cell lysate used in the IP was normalized carried out exactly as described (Morrison, 1995) using His-tagged to βgal activity levels to account for transfection efficiencies. βgal kinase-defective MEK1 as a substrate (Gardner et al., 1994). A portion activity levels were measured as previously described (Herbomel et al., of each sample was analyzed by Western blotting with the monoclonal 1984) using 20 μg of the cell lysate. For assays of endogenous ERK2 anti-Raf1 antibody to determine the amount of Raf in the immuno- activity, samples were normalized to protein using the BCA kit from precipitates. Pierce Chemicals (Rockford IL). For all immunoprecipitation, one-fifth of the samples were saved and run on a 10% SDS–polyacrylamide gels, transferred to Hybond C (Amersham), and immunoblotted using the Acknowlegements anti-ERK2 antibody, to measure the amount of ERK2 protein immuno- precipitated. In some cases, activity of endogenous ERK2 was measured This work was supported by National Institutes of Health grants #RO1 by carrying out kinase reactions in the immune precipitates using myelin GM47214 to M.A.S. and F32 GM18298 to M.W.R. basic protein (MBP) as a substrate (Minden et al., 1994). Phosphorylated MBP was separated by SDS–PAGE, bands cut from the gel and counted for radioactivity. Alternatively, in some experiments endogenous ERK2 References activity was assayed using an in-gel kinase assay as previously described (Kamashita and Fujisawa, 1989). For transiently transfected ERK2, the Bruder,J.T., Heidecker,G. and Rapp,U.R. (1992) Serum-,TPA-,and Ras- in-gel kinase assay was used exclusively. Briefly, samples were run on induced expression from Ap-1/Ets-driven promoters requires Raf-1 12.5% SDS–polyacrylamide gels containing 0.25 mg/ml myelin basic kinase. Genes Dev., 6, 545–556. protein and kinase reactions were performed by soaking gels in buffer Chen,Q., Kinch,M.S., Lin,T.H., Burridge,K. and Juliano,R.L. (1994) containing 25 mCi/ ml [γ- P]ATP with 10 mM non-radiolabeled ATP. Integrin-mediated cell adhesion activates mitogen-activated protein Gels were washed exhaustively and analyzed by autoradiography and kinases. J. Biol. Chem., 269, 26602–26605. 5598 MAP kinase activation requires cell adhesion Chen,Q., Lin,T.S., Der,C.J. and Juliano,R.L. (1996) Integrin-mediated McNamee,H.M., Ingber,D.E. and Schwartz,M.A. (1992) Adhesion to activation of mitogen-activated protein (MAP) or extracellular signal fibronectin stimulates inositol lipid synthesis and enhances PDGF- related kinase kinase (MEK) and kinase is independent of Ras. J. Biol. induced inositol lipid breakdown. J. Cell Biol., 121, 673–678. Miekka,S.I., Ingham,K.C. and Menache,D. (1982) Rapid methods for Chem., 271, 18122–18127. isolation of human plasma fibronectin. Thromb. Res., 27, 1–14. Choi,K.Y., Satterberg,B., Lyons,D.M. and Elion,E.A. (1994) Ste5 tethers Minden,A., Lin,A., McMahon,M., Lange-Carter,C., Derijard,B., multiple protein kinases in the MAP kinase cascade required for Davis,R.J., Johnson,G.L. and Karin,M. (1994) Differential activation mating in S. cerevisiae. Cell, 78, 499–512. of ERK and JNK mitogen-activated protein kinases by Raf-1 and Chung,J., Pellech,S.L. and Blenis,J. (1991) Mitogen-activated Swiss MEKK. Science, 266, 1719–1723. mouse 3T3 RSK kinases I and II are related to pp44mpk from sea Morino,N. et al. (1995) Matrix/integrin interaction activates the mitogen star oocytes and participate in the regulation of pp90rsk activity. Proc. activated protein kinase p44erk-1 and p42erk-2. J. Biol. Chem., 270, Natl Acad. Sci. USA, 88, 4981–4985. 269–273. Clark,E.A. and Brugge,J.S. (1995) Integrins and signal transduction Morrison,D. (1995) Activation of Raf1 by Ras in intact cells. Methods pathways: the road taken. Science, 268, 233–239. Enzymol., 255, 301–310. Cobb,M.H., Xu,S., Hepler,J.E., Hutchinson,M., Frost,J. and Robbins,D.J. Morrison,D.K., Kaplan,D.R., Rapp,U. and Roberts,T.M. (1988) Signal (1994) Regulation of the MAP kinase cascade. Cell. Mol. Biol. Res., transduction from membrane to cytoplasm: growth factors and 40, 253–256. membrane bound oncogene products increase Raf-1 phosphorylation Crews,C.M. and Erikson,R.L. (1993) Extracellular signals and reversible and associated protein kinase activity. Proc. Natl Acad. Sci USA, 85, protein phosphorylation: what to MEK of it all. Cell, 74, 215–217. 8855–8859. Crowley,S., Paterson,H. and Marshall,C.J. (1994) Activation of MAP Mulcahy,L.S., Smith,M.R. and Stacey,D.W. (1985) RAS protein in kinase kinase is necessary and sufficient for PC12 differentiation and growth control of 3T3s. Nature, 313, 241–243. for transformation of NIH 3T3 cells. Cell, 77, 841–852. Qiu,R.G., Chen,J., Kirn,D., McCormick,F. and Symons,M. (1995) An Deibler,G.E., Martenson,R.E. and Kies,M.W. (1972) Large scale essential role for Rac in Ras transformation. Nature, 374, 457–459. preparation of myelin basic protein from central nervous tissue of Renshaw,M.W., McWhirter,J.R. and Wang,J.Y.J. (1995) The human several mammalian species. Prep. Biochem., 2, 139–165. leukemia oncogene bcr-abl abrogates the anchorage requirement but Downward,J., Graves,J.D., Warne,P.H., Rayter,S. and Cantrell,D.A. not the growth factor requirement for proliferation. Mol. Cell. Biol., ras (1990) Stimulation of p21 upon T cell activation. Nature, 346, 15, 1286–1293. 719–723. Renshaw,M.W., Toksoz,D. and Schwartz,M.A. (1996) Involvement of Feig,L.A. and Cooper,G.M. (1988) Inhibition of NIH 3T3 cell the small GTPase Rho in integrin-mediated activation of MAP kinase. proliferation by a mutant ras protein with preferential affinity for J. Biol. Chem., 271, 21691–21694. GDP. Mol. Cell. Biol., 8, 3235–3243. Rodrigez-Viciana,P., Warne,P.H., Dhand,R., Vanhaesebroeck,B., Gout,I., Feramisco,J.R., Clark,R., Wong,F., Arnheim,N., Milley,R. and Fry,M.J., Waterfield,M.D. and Downward,J. (1994) Phosphatidyl-3- McCormick,F. (1985) Microinjecting anti-ras Ab causes reversion. OH kinase as a direct target of Ras. Nature, 370, 527–532. Nature, 314, 639–641. Schwartz,M.A., Schaller,M.D. and Ginsberg,M.H. (1995) Integrins: Freedman,V.H. and Shin,S. (1974) Cellular tumorigenicity in nude mice: emerging paradigms of signal transduction. Annu. Rev. Cell Biol., 11, correlation with cell growth in semisolid medium. Cell, 3, 355–359. 549–599. Gardner,A.M., Lange-Carter,C.A., Vaillancourt,R.R. and Johnson,G.L. Sturgill,T.W., Ray,L.B., Erikson,E. and Maller,J.L. (1988) Insulin- (1994) Measuring activation of kinases in mitogen-activated protein stimulated MAP-2 kinase phosphorylates and activates ribosomal kinase regulatory network. Methods Enzymol., 238, 258–276. protein S6 kinase II. Nature, 334, 715–718. Herbomel,P., Bourachot,B. and Yaniv,M. (1984) Two distinct enhancers Sun,H., Charles,C.H., Lau,L.F. and Tonks,N.K. (1993) MKP-1 (3CH134), with different cell specificities coexist in the regulatory region of an immediate early gene product, is a dual specificity phosphatase polyoma. Cell, 39, 653–662. that dephosphorylates MAP kinase in vivo. Cell, 75, 487–493. White,M.A., Nicolette,C., Minden,A., Polverini,A., Aelst,L.V., Karin,M. Hipskind,R.A., Baccarini,M. and Nordheim,A. (1994) Transient and Wigler,M.H. (1995) Multiple Ras functions can contribute to activation of Raf-1, MEK, and Erk2 coincides kinetically with ternary mammalian cell transformation. Cell, 80, 533–541. complex factor phosphorylation and immediate early gene promoter White,M.A., Vale,T., Camonis,J.H., Schaefer,E. and Wigler,M.H. (1996) activity in vivo. Mol. Cell. Biol., 14, 6219–6231. A role for Ral guanine nucleotide dissociation stimulator in mediating Hotchin,N.A. and Hall,A. (1995) The assembly of integrin adhesion Ras-induced transformation. J. Biol. Chem., 271, 16439–16442. complexes requires both extracellular matrix and intracellular rho/rac Zhu,X. and Assoian,R.K. (1995) Integrin-dependent activation of MAP GTPases. J. Cell Biol., 131, 1857–1865. kinase: a link to shape-dependent cell proliferation. Mol. Biol. Cell, Janknecht,R., Ernst,W.H., Pingoud,V. and Nordheim,A. (1993) Activation 6, 273–282. of ternary complex factor Elk-1 by MAP kinases. EMBO J., 12, 5097–5104. Received on May 20, 1997; revised on June 27, 1997 Jelinek,T., Dent,P., Sturgill,T.W. and Weber,M.J. (1996) Ras-induced activation of RAf-1 is dependent on tyrosine phosphorylation. Mol. Cell. Biol., 16, 1027–1034. Kamashita,I. and Fujisawa,H. (1989) A sensitive method for detection of calmodulin dependent protein kinase II activity in sodium dodecyl sulfate polyacrylamide gels. Anal. Biochem., 183, 139–143. Khosravi-Far,R., Solski,P.A., Clark,G.J., Kinch,M.S. and Der,C.J. (1995) Activation of Rac1, RhoA and mitogen-activated protein kinases is required for ras transformation. Mol. Cell. Biol., 15, 6443–6453. Kolch,W.G., Heidecker,G., Lloyd,P. and Rapp,U.R. (1991) Raf-1 kinase is required for growth of induced NIH 3T3 cells. Nature, 349, 426–428. Leevers,S.J., Paterson,H.F. and Marshall,C.J. (1994) Requirement for Ras in Raf activation is overcome by targeting Raf to the plasma membrane. Nature, 369, 411–414. Lin,L.-L., Wartmann,M., Lin,A.Y., Knopf,J.L., Seth,A. and Davis,R.J. (1993) cPLA2 is phosphorylated and activated by MAP kinase. Cell, 72, 269–278. Mansour,S.J., Matten,W.T., Hermann,A.S., Candia,J.S., Rong,S., Fukasawa,K., Woude,G.F.V. and Ahn,N.G. (1994) Transformation of mammalian cells by constitutively active MAP kinase kinase. Science, 265, 966–970. Marshall,C.J. (1995) Specificity of receptor tyrosine kinase signalling: transient versus sustained extracellular signal related kinase activation. Cell, 80, 179–185. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The EMBO Journal Springer Journals

Growth factor activation of MAP kinase requires cell adhesion

Download PDF
8 pages

Loading next page...
 
/lp/springer-journals/growth-factor-activation-of-map-kinase-requires-cell-adhesion-bTajAbD5ng

References (57)

Publisher
Springer Journals
Copyright
Copyright © European Molecular Biology Organization 1997
ISSN
0261-4189
eISSN
1460-2075
DOI
10.1093/emboj/16.18.5592
Publisher site
See Article on Publisher Site

Abstract

The EMBO Journal Vol.16 No.18 pp.5592–5599, 1997 Growth factor activation of MAP kinase requires cell adhesion Ras, Raf (Leevers et al., 1994), or MEK1 (Crowley et al., Mark W.Renshaw, Xiang-Dong Ren and 1994; Mansour et al., 1994) is transforming. Furthermore, Martin Alexander Schwartz inhibition of Ras, Raf, MEK1 or ERKs inhibits normal Department of Vascular Biology, The Scripps Research Institute, and transformed cell growth (Feramisco et al., 1985; 10550 N. Torrey Pines Road, La Jolla, CA 92037, USA Mulcahy et al., 1985; Feig and Cooper, 1988; Kolch et al., Corresponding author 1991; Crowley et al., 1994; Khosravi-Far et al., 1995). e-mail: [email protected] Ras, however, has a number of other effectors such as Rho family GTPases, PI 3-kinase and Ral GDS, which The MAP kinase pathway is a major regulator of also make important contributions to cell cycle progression both normal and oncogenic growth. We report that and transformation (Rodrigez-Viciana et al., 1994; activation of the MAP kinase ERK2 by serum or Khosravi-Far et al., 1995; Qiu et al., 1995; White et al., purified growth factors is strongly dependent on cell 1996). adhesion to extracellular matrix proteins. This effect It is well established that normal cell growth requires is specific to soluble growth factors, since suspended cell adhesion to extracellular matrix proteins as well as cells still activate ERK2 in response to plating on stimulation by growth factors (reviewed in Clark and fibronectin, and is reversible. Analysis of endogenous Brugge, 1995; Schwartz et al., 1995). The ability of Ras and Raf show that these proteins are still activated integrins to transduce a variety of intracellular signals by serum in suspended cells, whereas MEK activity is most likely mediates this effect. One such signaling inhibited. Conversely, activation of ERK2 by activated pathway involves the MAP kinases ERK1 and ERK2. mutants of Ras and Raf is still adhesion-dependent but Plating of suspended cells on extracellular matrix proteins activation by MEK is not. Consistent with these results, such as fibronectin (FN) triggers a rapid, transient activ- activated MEK enhances growth of ras-transformed ation of MAP kinases (Chen et al., 1994; Morino et al., cells in suspension but not when adherent. These results 1995; Zhu and Assoian, 1995). However, it is not at all identify a novel synergism between cell adhesion- and clear how this transient phenomenon contributes to growth growth factor-regulated pathways, and explain how of stably adherent cells. oncogenic activation of MAP kinases induces both Conversely, how constitutive activation of ERKs leads serum- and anchorage-independent growth. not only to serum-independent but also to anchorage- Keywords: cell adhesion/growth factor regulation/ independent growth is poorly understood. Anchorage- integrin/signal transduction independence is the in vitro characteristic that correlates best with tumorigenicity in vivo (Freedman and Shin, 1974). Two groups have analyzed the effect of cell adhesion on growth factor activation of MAP kinase and Introduction found that placing 3T3 cells in suspension had either a modest effect (Zhu and Assoian, 1995) or no effect The MAP kinase pathway is a major cellular signaling (Hotchin and Hall, 1995) on activation of ERKs by growth pathway that mediates effects of growth factors on cell factors. These results are paradoxical, since growth factors cycle progression (reviewed in Crews and Erikson, 1993; do not induce DNA synthesis in suspended cells, yet Cobb et al., 1994; Marshall, 1995). Growth factors induce activated variants of Ras, Raf or MEK1 do. conversion of c-Ras to its GTP-bound form, which then To resolve this paradox, we investigated in greater binds the protein kinase c-Raf. Localization of Raf to the detail the involvement of cell adhesion in growth factor plasma membrane leads to its activation via mechanisms activation of the MAP kinase pathway. We found that that are incompletely understood but that appear to involve activation of ERK2 was in fact strongly dependent on its phosphorylation (Morrison et al., 1988; Jelinek et al., adhesion to extracellular matrix. Furthermore, we identi- 1996). Once activated, Raf phosphorylates and activates fied activation of MEK by Raf as the adhesion-dependent the protein kinase MEK1, which phosphorylates and step. Thus, constitutive activation of the MAP kinase activates the MAP kinases ERK1 and ERK2. These kinases pathway by oncogenes can bypass the requirement for phosphorylate a number of substrates that participate in both growth factors and cell adhesion. cell cycle regulation, including the transcription factor Elk-1 (Janknecht et al., 1993), phospholipase A (Lin RSK et al., 1993) and p90 (Sturgill et al., 1988; Chung Results et al., 1991). This pathway also plays a key role in oncogenic Inhibition of ERK2 activity in suspended cells transformation, as activated forms of Ras are among the To test the possible dependence of the MAP kinase ERK2 most common mutations found in human tumors (Cobb on adhesion, cells were detached and either held in et al., 1994; Marshall, 1995) and expression of activated suspension in medium with 0.4% serum for 24 h or 5592 © Oxford University Press MAP kinase activation requires cell adhesion Fig. 1. ERK2 activity in adherent and suspended cells. (A) Stimulation by serum. Cells that had been incubated for 24 h in 0.4% serum while attached or suspended were stimulated by addition of 10% serum for the indicated times. Endogenous ERK was immunoprecipitated and analyzed for kinase activity using the in-gel assay and for ERK2 protein by Western blotting. Quantitation of the kinase assay is shown graphically. Data are representative of eight experiments. (B) Stimulation by PDGF. Attached or suspended cells that were incubated in 0.4% serum were stimulated by addition of 20 ng/ml PDGF for the indicated times. Endogenous ERK2 activity and amounts of immunoprecipitated protein were assayed as in (A). Quantitation of the kinase assay by densitometry is shown graphically. Data are representative of three experiments. (C) Stimulation by LPA. Attached or suspended cells were incubated in 0.4% serum, then stimulated with 2 μg/ml LPA for the indicated times. Endogenous ERK2 was immunoprecipitated and its activity determined in the immunoprecipitates. Kinase activity is shown graphically. Data are representative of four experiments. (D) Stimulation by TPA. Adherent cells kept in low serum for 24 h, serum-starved cells that were trypsinized and kept in suspension for 3 h, or cells kept in suspension in low serum for 24 h were stimulated with 100 nM TPA for 10 min. ERK2 kinase activity and immunoprecipitated protein amounts were assayed as before. The lower panel shows the quantitation of ERK2 kinase activity normalized for protein levels; values are means  standard deviations from three experiments. replated on tissue culture plastic in low serum for the sion for short periods of time still activate MAP kinase same period. Subsequent stimulation with 10% serum in response to growth factors (Hotchin and Hall, 1995; induced strong activation of ERK2 in adherent cells but Zhu and Assoian, 1995). To resolve the conflict between only a slight increase in suspended cells (Figure 1A). In our results and these reports, we analyzed the effect of eight experiments, inhibition varied between 97% and leaving cells in suspension for various lengths of time. 77%, averaging 90.2  9.8%. Platelet-derived growth We found that cells which were detached and held in factor (PDGF) and lysophosphatidic acid (LPA) are two suspension for short periods of time retained the ability of the major mitogens in serum. We therefore examined to activate ERK2 in response to serum (Figure 2). With the stimulation of ERK2 activity by purified PDGF and longer times in suspension, ERK2 activation diminished, LPA. Both mitogens strongly activated ERK2 in adherent with a half-time of ~6 h. These results indicate that loss cells but showed a substantially reduced response in non- of structures such as focal adhesions or dephosphorylation adherent cells (Figure 1B and C) of focal adhesion proteins such as focal adhesion kinase, cas Previous studies have shown that cells placed in suspen- paxillin, p130 or tensin, which occur rapidly upon loss 5593 M.W.Renshaw, X.-D.Ren and M.A.Schwartz Fig. 2. Time in suspension. Cells were kept attached in 0.4% serum or were trypsinized and maintained in suspension for varying periods of time. Total time in 0.4% serum was kept constant by varying the length of starvation in low serum prior to detachment. Cells were then stimulated with 10% serum for 60 min and MAP kinase assayed. Values are means  standard deviations from four experiments. Similar results were obtained with cells stimulated with serum for 15 min (not shown). of cell adhesion, are not likely to account for the failure of suspended cells to activate the MAP kinase pathway. Consistent with this idea, treatment of cells with cytochala- sin D for up to 90 min, which disrupts actin filaments and causes loss of focal adhesions, did not inhibit MAP kinase activation by serum (not shown). A prolonged period of ‘adhesion deprivation’ is required, analogous to the prolonged serum starvation that is required to de-activate growth factor-regulated pathways. Fig. 3. Recovery of MAP kinase activation. (A) Replating in the It is noteworthy that both LPA and PDGF, which bind absence of serum. Cells were incubated in suspension for3hor24h; a G protein-linked receptor and a protein tyrosine kinase total time in low serum was approximately constant. Cells were then receptor, respectively, are affected in the same manner by replated on plastic coated with 25 μg/ml FN. ERK2 activity was loss of cell adhesion. These observations argue against assayed as before. Values are means  standard deviations from three the likelihood that changes in receptor expression or experiments. (B) Response to serum. Cells were maintained in suspension in low serum for 24 h, then replated on plastic coated with function can account for the failure of non-adherent cells 20 μg/ml FN (upper panel). At the indicated times, cells were to activate ERK2. Consistent with this idea, previous work stimulated with serum for 15 min and MAP kinase assayed. has shown that the PDGF receptor is still functional Alternatively, as shown in the lower panel, cells were replated on even after prolonged incubation of cells in suspension plastic with different coating. These were: 20 μg/ml FN; 100 μg/ml polylysine (PL); 20 μg/ml anti-mouse β integrin IgG (β1); 25 μg/ml (McNamee et al., 1992). To test this idea further, we goat anti-mouse IgG followed by 20 μg/ml anti-mouse β IgG (2nd examined the activation of ERK2 by the phorbol ester 1 β1). In these experiments, the plastic was subsequently blocked with TPA, which activates Raf via protein kinase C, bypassing 10 mg/ml heat-denatured BSA to prevent adhesion via secreted FN. Ras and the membrane receptors. The ability of TPA to After 3 h, cells were stimulated with 10% serum for 15 min and MAP activate ERK2 was also greatly attenuated in suspended kinase assayed. Four experiments yielded similar results. Similar results were obtained with cells stimulated with serum for 60 min (not cells in a time-dependent manner (Figure 1D). shown). Reversibility To examine whether MAP kinase activation could be stimulus (adhesion). Thus, the failure to respond to serum restored by replating cells on the extracellular matrix is specific to soluble growth factors and cannot be due to protein FN, two experiments were performed. First, cells general toxicity or apoptosis. that were kept in suspension for either 3 or 24 h were To test whether replating restores the ability to respond replated on FN. The cells that had been in suspension for to serum, 24 h suspended cells were replated on FN in 24 h adhered and spread at a significantly slower rate, low serum for various times and then stimulated with though in both cases cells were well spread by 60 min serum. ERK2 activity was then assayed. We observed that (data not shown). Assay of ERK2 activity in these cells upon replating, cells regained the ability to activate ERK2 showed that the activation of ERK2 in cells kept in in response to serum (Figure 3B). The activation by suspension for 24 h was delayed relative to cells kept in replating alone in the absence of serum was, as in Figure suspension for 3 h (Figure 3A). Presumably this delay is 3A, readily detectable at 1 h, followed by a return to due to the delay in attachment and spreading. However, baseline by 2 h. However, the increase in activity due to the extent of ERK2 activation was equivalent in the cells serum was readily detected at the 1 h time point. Interest- suspended for 24 h. This result shows that ERK2 can still ingly, the serum response was diminished at the 2 h time be activated in 24 h suspended cells by an appropriate point. This effect is most likely due to the fact that 5594 MAP kinase activation requires cell adhesion Fig. 5. Raf kinase activity. Attached or suspended cells incubated in low serum for 24 h were stimulated with 10% serum, then lysed and c-Raf immunoprecipitated. Raf kinase activity, using kinase-defective Fig. 4. Ras GTP loading. Attached or suspended cells in low serum MEK-1 as a substrate, and Raf protein levels were assayed (A)as were labeled with P-orthophosphate and stimulated with serum for described in Materials and methods. Kinase activity was normalized 8 min. Endogenous Ras was immunoprecipitated and the bound for immunoprecipitated protein and data presented graphically (B). nucleotide was analyzed by chromatography (A). Values shown Values are means  standard deviations from three experiments. graphically (B) are means  standard deviations from four experiments. Next, we assayed Raf function by immunoprecipitating stimulation of MAP kinases by replating alone may render cellular Raf and carrying out an in vitro kinase assay using cells refractory to subsequent stimulation by serum, due kinase-defective MEK1 as a substrate. These experiments to induction of the ERK phosphatase MKP-1 (Sun et al., showed that Raf kinase was still activated in suspended 1993). Recovery of MAP kinase activation was also cells, with only a slight (~10%) decrease relative to observed when cells were replated on an antibody to the adherent cells (Figure 5). Thus, activation of Raf is not integrin β1 subunit, but not on polylysine. Treatment of significantly adhesion-dependent. cells with cycloheximide did not block the recovery of We then immunoprecipitated cellular MEK1, and MAP kinase activation (data not shown), but did reduce examined its kinase activity using kinase-defective ERK2 cell spreading and slightly reduced MAP kinase activation. as a substrate. These experiments revealed that activation These results show that the inhibition of ERK2 in sus- of endogenous MEK1 in response to serum was greatly pended cells is reversible and is integrin-dependent. reduced in non-adherent cells (Figure 6), to an extent that was similar to the inhibition of ERK2. Thus, the ability Activation of intermediates of Raf to activate MEK1 appears to be the step that is It is well established that growth factors activate ERK2 sensitive to adhesion. via a pathway that involves Ras, Raf and MEK1. To determine which step(s) in the pathway require cell adhe- Activated components sion, we assayed the activation of these components in To confirm the results of experiments analyzing the adherent and suspended cells. Ras activation was assayed activities of cellular Ras, Raf and MEK1, we carried out by immunoprecipitating endogenous Ras protein and ana- a second series of experiments to study the activation of lyzing the bound nucleotide. We observed that the baseline ERK2 by constitutively activated mutants. Cells were GTP content was somewhat lower in suspended cells; transiently transfected with ras G12V, which is constitu- however, treatment with serum stimulated GTP loading tively activated by a mutation that reduces its GTPase; to the same level in adherent and suspended cells with raf BxB, in which the kinase is constitutively activated (Figure 4). Thus, the adhesion-dependent step must lie by truncation of the regulatory domain; or by ΔN3, S222 downstream of Ras. MEK1, which is activated by a deletion and a point 5595 M.W.Renshaw, X.-D.Ren and M.A.Schwartz Fig. 6. MEK kinase activity. Attached or suspended cells incubated in Fig. 7. Effect of activated mutants. NIH 3T3 cells were co-transfected low serum for 24 h were stimulated with 10% serum, then lysed and with HA-tagged ERK2 together with activated mutants of Ras, Raf or MEK-1 immunoprecipitated. MEK kinase activity was assayed using MEK. At 24 h after transfection, cells were replated on tissue culture kinase-defective ERK2 as a substrate (top panel), and plastic in 0.4% serum or placed in suspension in 0.4% serum. After an immunoprecipitated MEK-1 protein was analyzed by Western blotting additional 24 h, cells were harvested and the transfected ERK2 (lower panel). For graphical data, kinase activity was normalized for immunoprecipitated using anti-HA antibody. Both the kinase activity immunoprecipitated MEK-1 protein. Values are means  standard of the precipitated ERK2 and the ERK2 protein levels were assayed as deviations from three experiments. described in Materials and methods. Negligible ERK2 activity was observed in the absence of co-transfected activated Ras, Raf or MEK. Values shown in the lower panels are kinase activity normalized for mutation. In each case, the activated protein was co- ERK2 protein, averaged from three experiments  standard deviations. transfected with an ERK2 that had been tagged with the hemagglutinin (HA) epitope. Control cells were co- transfected with HA–ERK2 plus empty vector. At 24 h with MEK alone induced foci rather weakly, as reported after transfection, cells were placed in suspension or previously (Khosravi-Far et al., 1995; White et al., 1995). replated on tissue culture plastic in low serum, and This result is consistent with the concept that ras induces incubated a further 24 h. The HA–ERK2 was then immuno- transformation via combined effects on a number of precipitated and its activity assayed. These results show effectors. Co-transfection of MEK with ras did not increase that ERK2 activation by both Ras and Raf were decreased the number of foci or colonies, but this may be due to in suspended cells to ~15% of the levels in adherent cells the already high efficiency of transformation by ras alone. (Figure 7). By contrast, activation of ERK2 by activated However, MEK substantially increased the size of the MEK1 remained high in non-adherent cells. These results colonies in soft agar (Table II). Measurement of their confirm that the adhesion-dependent step lies between Raf volume showed that MEK1 enhanced the rate of growth and MEK. It should be noted, however, that the activation of cells in suspension by ~7-fold. By contrast, MEK1 did of ERK2 by these oncogenes in adherent cells is ~2.5- not increase the size of adherent ras-transformed colonies fold higher than that triggered by serum in adherent in low serum. These results show that preventing the cells (not shown). Hence, the activation by oncogenes in decline in MAP kinase in non-adherent cells significantly suspended cells, while strongly attenuated, is ~40% of augments their growth in suspension but does not enhance that found in adherent cells treated with serum. the ability of ras to promote serum-independent growth in adherent cells. Cell growth The results obtained above make a surprising prediction. Discussion Present views of the MAP kinase pathway would argue against the idea that activated MEK1 should augment Our results show that in NIH 3T3 cells, activation of the transformation by Ras. However, our data suggest that MAP kinase pathway by serum or growth factors is expression of MEK1 should enhance MAP kinase activity strongly dependent on integrin-mediated cell adhesion to and growth in suspended cells. To test this hypothesis, extracellular matrix protein. Similar results were obtained cells were transfected with activated ras alone, activated with CHO cells (P.E.Hughes and M.H.Ginsberg, personal MEK1 alone, or both. Growth of transfected cells in communication), but other cell types have yet to be monolayer and in suspension was then examined. examined. Several results demonstrate that this effect is As shown in Table I, ras induced a significant number due to a specific blockade rather than to general toxicity: of foci in monolayer culture and a similar number of suspended cells retain the ability to activate MAP kinase colonies in soft agar. The number of foci and colonies in response to adhesion, they retain the ability to activate was nearly equal to the total number of hygromycin- Ras and Raf in response to growth factors, and the effect resistant colonies, suggesting that ras induced transform- is reversed when cells are replated on FN. ation of NIH 3T3 cells with high efficiency. Transfection Two independent lines of experimentation show that 5596 MAP kinase activation requires cell adhesion Table I. Stable transformation assay Vectors HygR colonies Foci in 10% CS Foci in 0.5% CS Soft agar colony Control 1340  60 0 00 00  0 MEK 1120  45 45  11 11  11 0  0 Ras 1100  45 1584  88 1408  220 1452  44 Ras  Mek 1260  98 1537  227 1588  151 1411  101 NIH 3T3 cells were co-transfected with RSVHyg and combinations of the plasmids expressing the activated MEK or Ras mutant proteins or the empty expression vector (Control). Relative DNA amounts of the mutants were kept constant (0.8 μg) in the transfections, so in transfections containing only one of the mutants the remainder of the DNA was made up by the empty expression vector. Transfected cells were trypsinized and portions replated in media containing 10% CS, 10% CS with 200 μg/ml hygromycin, 0.5% CS, or soft agar to measure foci and colony formation. Values shown represent the total number of colonies or foci calculated for the entire transfection and represent values obtained from two separate experiments. this effect occurs at the level of activation of MEK by Table II. Soft agar and minimal media colony size Raf. First, as mentioned above, endogenous Ras and Raf are still maximally activated by serum in non-adherent Vector Soft agar Foci in 0.5% CS cells, but MEK1 is not. Second, expression of activated Colony Relative Foci size Relative mutants of these proteins showed that signals from onco- volume genic Ras and Raf were still attenuated in suspended cells, whereas the signal from activated MEK1 was Ras 0.59  0.17 1.00 87  7 1.00 undiminished. Thus, the ability of Raf to activate MEK1 Ras  MEK 4.12  1.07 6.99 81  6 0.91 maximally must require cell adhesion. Finally, the bio- Transfected cells from Table I were further analyzed by visually logical relevance of these effects was indicated by the measuring soft agar colony size and calculating the volume (4/3πr ). result that activated MEK1 significantly enhanced the Values represent the mean  standard error from 15 colonies each. growth of ras-transformed cells in suspension, but had no Minimal media focus size was determined by trypsinizing the foci and replating in soft agar to determine the number of transformed cells effect on their growth when adherent. per focus. Values represent the mean  standard deviation from two Previous studies have been puzzling in that either no separate experiments in which at least 35 foci from each transfection effect or a modest effect of cell adhesion on the activation were trypsinized. of the MAP kinase pathway by growth factors was observed. Yet transforming activators of this pathway, et al., 1994), or as yet unidentified protein kinases or such as mutated forms of Ras, Raf and MEK1, induce phosphatases that may regulate MEK1 activity. Further growth of cells in suspension. If MAP kinase were fully work will be required to resolve these issues. activated by growth factors in normal, suspended cells, The majority of key cellular decisions about growth, then the activation of MAP kinase by oncoproteins would differentiation and survival are made on the basis of input be redundant. Our results resolve these questions and from both soluble mediators (growth factors, cytokines, suggest a model relating MAP kinase to anchorage- hormones) and from extracellular matrix receptors. Thus, dependence and -independence of growth. Normal cells understanding how cellular pathways are controlled by are anchorage-dependent in part because growth factor both soluble factors and cell adhesion to extracellular activation of MAP kinase depends on cell adhesion. matrix is a major goal. Our results identify a novel Conversely, cells transformed by ras and other oncogenes synergism between cell adhesion and growth factors in that work through the MAP kinase pathway are anchorage- the activation of the MAP kinase pathway. These results independent because activation of MAP kinase bypasses are likely to be important in a variety of systems where a requirement for both adhesion and growth factors. gene expressions, cell growth and cell survival depend This model, however, raises the question why activated upon cell adhesion. Ras or Raf are able to induce growth in suspension at all. It must be noted that the reduction of the ERK2 signal in non-adherent cells is not complete (Figure 7). Indeed, the Materials and methods activation of the MAP kinase pathway by these oncogenes Reagents is so strong that the levels of ERK activity in suspended Anti-ERK2 polyclonal antibody (C-14) was purchased from Santa Cruz cells are still significant, ~40% of the level in adherent, Biotechnology (Santa Cruz, CA). LipofectAMINE, Dulbecco’s modified serum-treated control cells. Thus, very strong activation Eagle’s medium (DMEM), serum and other reagents for cell culture of the pathway can at least partially overcome the require- were from Gibco-BRL (Gaithersburg, MD). Anti-mouse β1 integrin IgG ment for adhesion. (HMβ1-1) was purchased from Pharmingen (San Diego, CA). Fibronectin was prepared from human plasma as described (Miekka et al., 1982). The mechanism by which cell adhesion potentiates the Myelin basic protein was prepared from bovine spinal cord as described activation of MEK by Raf is currently unknown. The (Deibler et al., 1972). Platelet-derived growth factor (PDGF), lysophos- existence of focal adhesions or the phosphorylation of phatidic acid (LPA), agarose, methylcellulose and other reagents were focal adhesion proteins are not likely to be the immediate purchased from Sigma Chemicals (St. Louis, MO) unless otherwise noted. cause, since they are rapidly lost upon detachment whereas Cell culture the decline in ERK2 activation occurs more slowly NIH 3T3 cells were cultured in DMEM supplemented with 10% bovine (Figure 1D). Possible explanations include the regulation calf serum. For MAP kinase experiments, cells in DMEM with 0.4% of scaffolding proteins that control the proximity of Raf serum were plated in 60 mm tissue culture plastic dishes at ~80% of and MEK, analogous to the Ste5 protein in yeast (Choi confluence for adherent cells. Alternatively, equal numbers of cells in 5597 M.W.Renshaw, X.-D.Ren and M.A.Schwartz DMEM with 0.4% serum and 0.5% methylcellulose were plated in scanning densitometry using the deskscan software with a Scanjett IIP 10 cm dishes coated with 1 ml of 1% agarose which was equilibrated scanner (Hewlett Packard). with DMEM. Methyl cellulose reduces cell–cell aggregation but does not otherwise affect growth or MAP kinase activity (data not shown). Measurement of MEK 1 activity Cells were incubated as described in the text, stimulated with 10% serum Endogenous MEK 1 was immunoprecipitated from 100 μg of cell lysate or other factors, harvested and analyzed as described below. using 1 μg of anti MEK 1 (Santa Cruz Biotechnology). Immunoprecipit- ates were washed three times in lysis buffer, and then once in kinase Plasmids and transfection buffer (Chen et al., 1996) 10 mM Tris, pH 7.5, 10 mM MgCl ,1 mM For transfections, cells were plated at a density of 410 cells per 6 cm DTT. IPs were then split, using one-fifth to measure the amount of MEK dish 24 h before transfection. Cells were transfected with LipofectAMINE 1 protein and four-fifths to measure kinase activity. MEK kinase activity as described (Renshaw et al., 1996) using 0.2 μg of pCMV5 ERK2, 32 was measured in kinase buffer containing 25 μMATP,5 μCi [γ- P]ATP 0.2 μg of pCMV5 βgal, and 1.6 mg of either pDCR ras G12V (White and 2 μg of kinase-dead GST–ERK2 (Hipskind et al., 1994) for 30 min et al., 1995), pZIP Neo raf BXB (Bruder et al., 1992) or pMCL MEK at room temperature. Samples were electrophoresed on 10% SDS– ΔN3, S222D (Mansour et al., 1994) per plate. At 24 h after transfection, polyacrylamide gels, which were then dried and exposed to film. cells were transferred to medium containing 0.5% serum for an additional Autoradiographs were quantitated using a model I.S. 1000 digital imaging 24 h for adherent cells and for cells suspended for only 3 h. After 24 h system from Alpha-Innotech Corp. in 0.5% CS, cells from indicated plates were then trypsinized and suspended for 3 h in serum-free DMEM containing 0.1% BSA Ras-GTP loading (Calbiochem, nuclease-, protease-free), and 0.25 mg/ml soybean trypsin NIH 3T3 cells were maintained adherent in 0.4% serum or in suspension inhibitor (Sigma) over dishes which had been coated with 1% heat- as before for 12 h. Cells were then transferred to phosphate-free DMEM denatured BSA (Sigma fraction V). Alternatively, for cells kept in containing 0.4% serum and 50 μCi/ml P-labeled orthophosphate for suspension for 24 h, cells were trypsinized 24 h after transfection and an additional 12 h. Cells were then rinsed with cold Tris-buffered saline placed in suspension in DMEM media containing 0.5% methyl cellulose or phosphate-free DMEM and stimulated with 10% serum for 6– and 0.4% calf serum in dishes coated with 1% agarose as above. 10 min at room temperature. (Preliminary experiments demonstrated that maximal GTP loading of Ras occurred at this time interval.) Cells Cell growth and transformation assays were then rinsed three times with cold PBS, and lysed in 380 μlof For stable transformation assays, cells were transfected with RSVHyg 50 mM HEPES, pH 7.4, containing 1% NP40, 0.1% BSA, 100 mM and 1.6 μg of either the empty control vector, pDCR Ras G12V, pMCL NaCl, 15 mM MgCl , 0.1 mM GTP, 0.1 mM GDP, 1 mM ATP, 10 μg/ml MEK ΔN3, S222D, or combinations thereof. After 24 h, the cells were aprotinin and leupeptin, 1 mM PMSF, and 2 μg of either a rat monoclonal changed to fresh media and allowed to grow for two more days. Cells th anti-Ras antibody (Santa Cruz Biotechnology) or non-immune rat IgG were then trypsinized and a portion (1/20 ) replated to measure focus as a control. Lysates were centrifuged at 13 000 r.p.m. for 12 min, the formation in normal media (with 10% serum), minimal media, or in soft supernatants transferred to fresh tubes and adjusted to 0.5M NaCl, 0.5% agar (with 10% serum) as previously described (Renshaw et al., 1995). deoxycholate, 0.05% SDS. They were incubated on ice for 30 min, then Cells were also replated in media containing 200 μg/ml Hygromycin 10 μl protein G PLUS-Agarose (Santa Cruz Biotechnology) added and (Boehringer-Mannheim) to determine the total number of stably trans- the samples rotated for1hat 4°C for 60 min. The beads were washed fected clones. Colonies were scored visually 2 weeks after transfections. eight times, and the bound nucleotides extracted and chromatographed Soft agar colony volume was determined by visually measuring colony on polyethyleneimine paper as described (Downward et al., 1990). diameter against a scale, from which colony volume was calculated. Chromatograms were analyzed by autoradiography on Kodak BioMax Size of foci in low serum (‘minimal medium’) was determined by MS film and quantitated by densitometry. trypsinizing the foci and replating them in soft agar. The number of soft agar colonies was then counted after 2 weeks to determine the number Raf kinase assay of transformed cells per original minimal medium focus. Adherent or suspended cells were stimulated with serum for 8 min, rinsed with cold PBS and lysed in buffer containing 50 mM HEPES, Measurement of ERK2 activity pH 7.4, 1% NP40, 100 mM NaCl, 2 mM EDTA, 10 μg/ml aprotinin Cells were stimulated as described in the Results section, rinsed with and leupeptin, 1 mM PMSF, 5 mM sodium vanadate, 20 mM NaF, cold phosphate-buffered saline (PBS) and extracted in ice-cold buffer 10 mM sodium pyrophosphate and 3 mM β-glycerophosphate. Cell containing 0.5% NP-40, 20 mM Tris, pH 7.6, 250 mM NaCl, 5 mM lysates were centrifuged at 13 000 r.p.m. in a microfuge for 15 min, and EDTA, 3 mM EGTA, 20 mM sodium phosphate, 20 mM sodium precleared with 10 μl of protein A–agarose beads (Pierce). Protein in pyrophosphate, 3 mM β-glycerophosphate, 1 mM sodium orthovanadate, each sample was assayed using the Pierce BCA Kit. Normalized samples 1 mM PMSF, 10 mM NaF, and 10 μg/ml each of leupeptin and aprotinin. Lysates were centrifuged 15 min at maximum speed in a microfuge and were adjusted to 0.5 M NaCl, 0.1% SDS and 0.5% deoxycholate, and the supernatants removed. For assays of transfected HA-ERK2 activity, immunoprecipitated for 2 h at 4°C using 3 μg of mouse monoclonal 0.5 μg of anti-HA (12CA5) antibody purified over an HA affinity column anti-Raf1 IgG (Transduction Laboratories) and 10 μl of protein A– was used for each immunoprecipitation. For endogenous ERK2, 0.5 μg agarose beads prebound with 10 μg of rabbit anti-mouse IgG (Cappel). of anti-ERK2 was used for each immunoprecipitation. Approximately The beads were rinsed three times with lysis buffer containing 0.5 M 100 μg of cell protein was immunoprecipitated in each sample. For NaCl and once with lysis buffer. The in vitro Raf kinase assay was transfected cells, the amount of cell lysate used in the IP was normalized carried out exactly as described (Morrison, 1995) using His-tagged to βgal activity levels to account for transfection efficiencies. βgal kinase-defective MEK1 as a substrate (Gardner et al., 1994). A portion activity levels were measured as previously described (Herbomel et al., of each sample was analyzed by Western blotting with the monoclonal 1984) using 20 μg of the cell lysate. For assays of endogenous ERK2 anti-Raf1 antibody to determine the amount of Raf in the immuno- activity, samples were normalized to protein using the BCA kit from precipitates. Pierce Chemicals (Rockford IL). For all immunoprecipitation, one-fifth of the samples were saved and run on a 10% SDS–polyacrylamide gels, transferred to Hybond C (Amersham), and immunoblotted using the Acknowlegements anti-ERK2 antibody, to measure the amount of ERK2 protein immuno- precipitated. In some cases, activity of endogenous ERK2 was measured This work was supported by National Institutes of Health grants #RO1 by carrying out kinase reactions in the immune precipitates using myelin GM47214 to M.A.S. and F32 GM18298 to M.W.R. basic protein (MBP) as a substrate (Minden et al., 1994). Phosphorylated MBP was separated by SDS–PAGE, bands cut from the gel and counted for radioactivity. Alternatively, in some experiments endogenous ERK2 References activity was assayed using an in-gel kinase assay as previously described (Kamashita and Fujisawa, 1989). For transiently transfected ERK2, the Bruder,J.T., Heidecker,G. and Rapp,U.R. (1992) Serum-,TPA-,and Ras- in-gel kinase assay was used exclusively. Briefly, samples were run on induced expression from Ap-1/Ets-driven promoters requires Raf-1 12.5% SDS–polyacrylamide gels containing 0.25 mg/ml myelin basic kinase. Genes Dev., 6, 545–556. protein and kinase reactions were performed by soaking gels in buffer Chen,Q., Kinch,M.S., Lin,T.H., Burridge,K. and Juliano,R.L. (1994) containing 25 mCi/ ml [γ- P]ATP with 10 mM non-radiolabeled ATP. Integrin-mediated cell adhesion activates mitogen-activated protein Gels were washed exhaustively and analyzed by autoradiography and kinases. J. Biol. Chem., 269, 26602–26605. 5598 MAP kinase activation requires cell adhesion Chen,Q., Lin,T.S., Der,C.J. and Juliano,R.L. (1996) Integrin-mediated McNamee,H.M., Ingber,D.E. and Schwartz,M.A. (1992) Adhesion to activation of mitogen-activated protein (MAP) or extracellular signal fibronectin stimulates inositol lipid synthesis and enhances PDGF- related kinase kinase (MEK) and kinase is independent of Ras. J. Biol. induced inositol lipid breakdown. J. Cell Biol., 121, 673–678. Miekka,S.I., Ingham,K.C. and Menache,D. (1982) Rapid methods for Chem., 271, 18122–18127. isolation of human plasma fibronectin. Thromb. Res., 27, 1–14. Choi,K.Y., Satterberg,B., Lyons,D.M. and Elion,E.A. (1994) Ste5 tethers Minden,A., Lin,A., McMahon,M., Lange-Carter,C., Derijard,B., multiple protein kinases in the MAP kinase cascade required for Davis,R.J., Johnson,G.L. and Karin,M. (1994) Differential activation mating in S. cerevisiae. Cell, 78, 499–512. of ERK and JNK mitogen-activated protein kinases by Raf-1 and Chung,J., Pellech,S.L. and Blenis,J. (1991) Mitogen-activated Swiss MEKK. Science, 266, 1719–1723. mouse 3T3 RSK kinases I and II are related to pp44mpk from sea Morino,N. et al. (1995) Matrix/integrin interaction activates the mitogen star oocytes and participate in the regulation of pp90rsk activity. Proc. activated protein kinase p44erk-1 and p42erk-2. J. Biol. Chem., 270, Natl Acad. Sci. USA, 88, 4981–4985. 269–273. Clark,E.A. and Brugge,J.S. (1995) Integrins and signal transduction Morrison,D. (1995) Activation of Raf1 by Ras in intact cells. Methods pathways: the road taken. Science, 268, 233–239. Enzymol., 255, 301–310. Cobb,M.H., Xu,S., Hepler,J.E., Hutchinson,M., Frost,J. and Robbins,D.J. Morrison,D.K., Kaplan,D.R., Rapp,U. and Roberts,T.M. (1988) Signal (1994) Regulation of the MAP kinase cascade. Cell. Mol. Biol. Res., transduction from membrane to cytoplasm: growth factors and 40, 253–256. membrane bound oncogene products increase Raf-1 phosphorylation Crews,C.M. and Erikson,R.L. (1993) Extracellular signals and reversible and associated protein kinase activity. Proc. Natl Acad. Sci USA, 85, protein phosphorylation: what to MEK of it all. Cell, 74, 215–217. 8855–8859. Crowley,S., Paterson,H. and Marshall,C.J. (1994) Activation of MAP Mulcahy,L.S., Smith,M.R. and Stacey,D.W. (1985) RAS protein in kinase kinase is necessary and sufficient for PC12 differentiation and growth control of 3T3s. Nature, 313, 241–243. for transformation of NIH 3T3 cells. Cell, 77, 841–852. Qiu,R.G., Chen,J., Kirn,D., McCormick,F. and Symons,M. (1995) An Deibler,G.E., Martenson,R.E. and Kies,M.W. (1972) Large scale essential role for Rac in Ras transformation. Nature, 374, 457–459. preparation of myelin basic protein from central nervous tissue of Renshaw,M.W., McWhirter,J.R. and Wang,J.Y.J. (1995) The human several mammalian species. Prep. Biochem., 2, 139–165. leukemia oncogene bcr-abl abrogates the anchorage requirement but Downward,J., Graves,J.D., Warne,P.H., Rayter,S. and Cantrell,D.A. not the growth factor requirement for proliferation. Mol. Cell. Biol., ras (1990) Stimulation of p21 upon T cell activation. Nature, 346, 15, 1286–1293. 719–723. Renshaw,M.W., Toksoz,D. and Schwartz,M.A. (1996) Involvement of Feig,L.A. and Cooper,G.M. (1988) Inhibition of NIH 3T3 cell the small GTPase Rho in integrin-mediated activation of MAP kinase. proliferation by a mutant ras protein with preferential affinity for J. Biol. Chem., 271, 21691–21694. GDP. Mol. Cell. Biol., 8, 3235–3243. Rodrigez-Viciana,P., Warne,P.H., Dhand,R., Vanhaesebroeck,B., Gout,I., Feramisco,J.R., Clark,R., Wong,F., Arnheim,N., Milley,R. and Fry,M.J., Waterfield,M.D. and Downward,J. (1994) Phosphatidyl-3- McCormick,F. (1985) Microinjecting anti-ras Ab causes reversion. OH kinase as a direct target of Ras. Nature, 370, 527–532. Nature, 314, 639–641. Schwartz,M.A., Schaller,M.D. and Ginsberg,M.H. (1995) Integrins: Freedman,V.H. and Shin,S. (1974) Cellular tumorigenicity in nude mice: emerging paradigms of signal transduction. Annu. Rev. Cell Biol., 11, correlation with cell growth in semisolid medium. Cell, 3, 355–359. 549–599. Gardner,A.M., Lange-Carter,C.A., Vaillancourt,R.R. and Johnson,G.L. Sturgill,T.W., Ray,L.B., Erikson,E. and Maller,J.L. (1988) Insulin- (1994) Measuring activation of kinases in mitogen-activated protein stimulated MAP-2 kinase phosphorylates and activates ribosomal kinase regulatory network. Methods Enzymol., 238, 258–276. protein S6 kinase II. Nature, 334, 715–718. Herbomel,P., Bourachot,B. and Yaniv,M. (1984) Two distinct enhancers Sun,H., Charles,C.H., Lau,L.F. and Tonks,N.K. (1993) MKP-1 (3CH134), with different cell specificities coexist in the regulatory region of an immediate early gene product, is a dual specificity phosphatase polyoma. Cell, 39, 653–662. that dephosphorylates MAP kinase in vivo. Cell, 75, 487–493. White,M.A., Nicolette,C., Minden,A., Polverini,A., Aelst,L.V., Karin,M. Hipskind,R.A., Baccarini,M. and Nordheim,A. (1994) Transient and Wigler,M.H. (1995) Multiple Ras functions can contribute to activation of Raf-1, MEK, and Erk2 coincides kinetically with ternary mammalian cell transformation. Cell, 80, 533–541. complex factor phosphorylation and immediate early gene promoter White,M.A., Vale,T., Camonis,J.H., Schaefer,E. and Wigler,M.H. (1996) activity in vivo. Mol. Cell. Biol., 14, 6219–6231. A role for Ral guanine nucleotide dissociation stimulator in mediating Hotchin,N.A. and Hall,A. (1995) The assembly of integrin adhesion Ras-induced transformation. J. Biol. Chem., 271, 16439–16442. complexes requires both extracellular matrix and intracellular rho/rac Zhu,X. and Assoian,R.K. (1995) Integrin-dependent activation of MAP GTPases. J. Cell Biol., 131, 1857–1865. kinase: a link to shape-dependent cell proliferation. Mol. Biol. Cell, Janknecht,R., Ernst,W.H., Pingoud,V. and Nordheim,A. (1993) Activation 6, 273–282. of ternary complex factor Elk-1 by MAP kinases. EMBO J., 12, 5097–5104. Received on May 20, 1997; revised on June 27, 1997 Jelinek,T., Dent,P., Sturgill,T.W. and Weber,M.J. (1996) Ras-induced activation of RAf-1 is dependent on tyrosine phosphorylation. Mol. Cell. Biol., 16, 1027–1034. Kamashita,I. and Fujisawa,H. (1989) A sensitive method for detection of calmodulin dependent protein kinase II activity in sodium dodecyl sulfate polyacrylamide gels. Anal. Biochem., 183, 139–143. Khosravi-Far,R., Solski,P.A., Clark,G.J., Kinch,M.S. and Der,C.J. (1995) Activation of Rac1, RhoA and mitogen-activated protein kinases is required for ras transformation. Mol. Cell. Biol., 15, 6443–6453. Kolch,W.G., Heidecker,G., Lloyd,P. and Rapp,U.R. (1991) Raf-1 kinase is required for growth of induced NIH 3T3 cells. Nature, 349, 426–428. Leevers,S.J., Paterson,H.F. and Marshall,C.J. (1994) Requirement for Ras in Raf activation is overcome by targeting Raf to the plasma membrane. Nature, 369, 411–414. Lin,L.-L., Wartmann,M., Lin,A.Y., Knopf,J.L., Seth,A. and Davis,R.J. (1993) cPLA2 is phosphorylated and activated by MAP kinase. Cell, 72, 269–278. Mansour,S.J., Matten,W.T., Hermann,A.S., Candia,J.S., Rong,S., Fukasawa,K., Woude,G.F.V. and Ahn,N.G. (1994) Transformation of mammalian cells by constitutively active MAP kinase kinase. Science, 265, 966–970. Marshall,C.J. (1995) Specificity of receptor tyrosine kinase signalling: transient versus sustained extracellular signal related kinase activation. Cell, 80, 179–185.

Journal

The EMBO Journal – Springer Journals

Published: Sep 15, 1997

Keywords: cell adhesion; growth factor regulation; integrin; signal transduction

There are no references for this article.