Bombesin and Neuromedin B Stimulate the Activation of P42mapk and P74raf-1 Via a Protein Kinase C-Independent Pathway in Rat-1 Cells

Bombesin and neuromedin B stimulate the activation of p42mapk and p74raf-1 via a protein kinase C-independent pathway in Rat-1 cells

Amanda Charlesworth and Enrique Rozengurt

Imperial Cancer Research Fund, PO Box 123, 44 Lincoln's Inn Fields, London, WC2A 3PX, UK

The mechanisms by which seven transmembrane recep- (PKC)-dependent pathways have been described (Howe tors activate p42mapk/p44mapk are not well de®ned although and Marshall, 1993; Crespo et al., 1994; Seuerlein and p21ras- and protein kinase C (PKC)-dependent pathways Rozengurt, 1995; Seuerlein et al., 1995, 1996; van have been implicated, typically for Gi- and Gq-coupled Biesen et al., 1996) typically for pertussis toxin (PTx)- receptors, respectively. Here, we demonstrate that in sensitive Gi and PTx-insensitive Gq-coupled receptors

Rat-1 cells transfected with the Gq-coupled bombesin/ respectively (Hawes et al., 1995). gastrin releasing peptide receptor, bombesin stimulated The peptides of the bombesin family including gastrin activation of p42mapk that was not inhibited by the speci®c releasing peptide (GRP) are potent mitogens for a PKC inhibitor GF 109203X or by down regulation of variety of cells including Swiss 3T3 cells (Rozengurt phorbol ester-sensitive PKC isoforms. In addition, and Sinnett-Smith, 1983; Zachary and Rozengurt, 1985) bombesin rapidly stimulated p74raf-1 activity that was and Rat-1 cells (Charlesworth et al., 1996) and have been also independent of PKC activity and insensitive to implicated as autocrine growth factors for small cell lung inhibition by pertussis toxin. Furthermore, addition of cancer (Cuttitta et al., 1985; Carney et al., 1987; Sethi et neuromedin B to Rat-1 cells transfected with the al., 1991). Bombesin binds to a speci®c Gq-coupled seven neuromedin B preferring receptor also activated p42mapk transmembrane domain receptor (Zachary and Rozen- and p74raf-1 in a PKC-independent and pertussis toxin- gurt, 1985, 1987; Coer et al., 1990; Battey et al., 1991; insensitive manner. Finally we show that addition of Oermans et al., 1994) and elicits rapid phospholipase C bombesin to Rat-1 cells stimulated the GTP loading of mediated hydrolysis of polyphosphoinositides leading to p21ras. Our results reveal a novel PKC-independent Ca2+ mobilisation from intracellular stores and activa- pathway in the action of Gq-coupled receptors and stress tion of PKC (Rozengurt, 1992). Addition of bombesin to the importance of cell context in de®ning the signal Swiss 3T3 cells also induces rapid activation of p42mapk transduction pathway(s) that link speci®c receptors to and p44mapk (Withers et al., 1995) that is completely the activation of the mitogen-activated protein kinase dependent on PKC (Pang et al., 1993; Seuerlein et al., cascade. 1996) and does not involve p21ras (Satoh et al., 1990; Mitchell et al., 1995), or p74raf-1 (Mitchell et al., 1995; ras mapk Keywords: p21 ;Gq; neuropeptide; MAP kinases; Seuerlein et al., 1996). Thus, the mechanism for p42 / signal transduction p44mapk activation by bombesin in Swiss 3T3 cells is

consistent with the model for general Gq-coupled receptor activation. However, some studies have indicated that p21ras could be involved in p42mapk/ mapk Introduction p44 activation by Gq-coupled receptors (Crespo et al., 1994; Koch et al., 1994; Mattingley et al., 1994) and The mitogen-activated protein kinases (MAPKs) are a currently there is increasing recognition of the impor- family of evolutionary conserved serine/threonine tance of the cellular context in which the MAPK kinases that are activated by a range of extracellular pathway is activated (Dikic et al., 1996; van Biesen et signals (Davis, 1993; Marshall, 1995). The two best al., 1996; Wan et al., 1996). characterised isoforms p42mapk/ERK-2 and p44mapk/ In the present study we examined the mechanism of ERK-1 are directly activated by phosphorylation on p42mapk activation by bombesin in Rat-1 cells stably speci®c tyrosine and threonine residues by the dual transfected with the bombesin/GRP preferring receptor speci®city MAPK kinase (or MEK) of which at least (Charlesworth et al., 1996). Our results reveal a novel two isoforms have been identi®ed in mammalian cells PKC-independent pathway in the action of Gq-coupled (Payne et al., 1991; Crews et al., 1992; Wu et al., 1993). receptors and stress the importance of cell context in Several pathways leading to MEK activation have been de®ning the signal transduction pathways that link demonstrated. Tyrosine kinase receptors induce p42mapk speci®c receptors to the activation of the MAP kinase and p44mapk activation via SOS-mediated accumulation cascade. of p21ras-GTP, which then activates a kinase cascade comprising p74raf-1, MEK and p42mapk/p44mapk (Davis, 1993; Marshall, 1995). The mechanisms by which mitogenic guanine nucleotide binding protein (G Results and Discussion protein) coupled receptors activate p42mapk and p44mapk Bombesin stimulates p42mapk activity in Rat-1 cells are less well de®ned although p21ras- and protein kinase transfected with the bombesin/GRP receptor

mapk Correspondence: E Rozengurt To examine bombesin stimulation of p42 activity in Received 1 November 1996; revised 29 January 1997; accepted Rat-1 cells transfected with the bombesin/GRP 31 January 1997 receptor, quiescent BOR 15 cells were treated with PKC-independent Gq activation of MAPK pathway A Charlesworth and E Rozengurt 2324 increasing concentrations of this agonist for 5 min. down regulation of phorbol ester-sensitive isoforms of Lysates from stimulated and unstimulated cells were PKC by prolonged treatment with 800 nM PDB, and immunoprecipitated with antibodies against p42mapk by using the speci®c PKC inhibitor GF 109203X. This and then the activity of this enzyme was measured by compound is a bisindolylmaleimide which selectively an immune-complex kinase assay, using myelin basic inhibits diacylglycerol-regulated PKC isoforms by protein (MBP) peptide as substrate. As shown in competing with ATP for the ATP binding site Figure 1 (left), bombesin induced a concentration (Toullec et al., 1991; Yeo and Exton, 1995). The dependent increase in p42mapk activity, the half results presented in Figure 1 (right) show that phorbol maximum and maximum concentration being 0.3 and ester-mediated p42mapk activation was signi®cantly 10 nM respectively. The maximum stimulation of attenuated by GF 109203X (66% inhibition) and p42mapk activity induced by bombesin (15-fold over completely abolished by chronic pretreatment with baseline levels) was comparable to the increase in PDB. As expected, EGF-induced p42mapk activation in p42mapk activity induced by epidermal growth factor Rat-1 cells was not signi®cantly inhibited by PKC (EGF) (Figure 1, inset). In contrast, addition of inhibition and only slightly by PKC down regulation. phorbol 12,13-dibutyrate (PDB) to parallel cultures In contrast to previous results with Swiss 3T3 cells, activated p42mapk to a level that was only 20 ± 30% of bombesin-induced p42mapk activation was not signifi- that induced by bombesin or EGF (Figure 1, inset). cantly inhibited by either GF 109203X or pretreatment Thus, bombesin stimulated a robust p42mapk activation with PDB (Figure 1, right). Thus, the results in Figure in Rat-1 cells transfected with the bombesin/GRP 1 demonstrate that bombesin induces a strong receptor. stimulation of p42mapk activity in Rat-1 cells trans- In view of the role of phorbol ester-sensitive fected with the bombesin GRP receptor through a mapk mapk isoforms of PKC in Gq-mediated p42 and p44 PKC-independent pathway. activation (Pang et al., 1993; Hawes et al., 1995; Seuerlein et al., 1996), it was important to assess the Bombesin stimulates p74raf-1 activity independently of contribution of these PKCs to the p42mapk response PKC in Rat-1 cells induced by bombesin in the bombesin/GRP receptor transfected Rat-1 cells. This was examined both by G protein-coupled receptors activate p42mapk and p44mapk through p74raf-1-dependent and independent pathways (Russell et al., 1994; Hawes et al., 1995; Mitchell et al., 1995; Seuerlein et al., 1995). Consequently, we examined whether bombesin stimulates p74raf-1 activity in Rat-1 cells transfected with the bombesin/GRP receptor. This was achieved using a sensitive assay in which the activation of immunoprecipitated p74raf-1 kinase from bombesin-stimulated cells

Figure 1 (Left) Bombesin-induced increase in p42mapk activity. BOR 15 cells in 100 mm dishes were washed twice in DMEM and incubated in 5 ml DMEM with increasing concentrations of bombesin for 5 min at 378C. Cells were lysed and p42mapk was immunoprecipitated with polyclonal anti-p42mapk antibodies. p42mapk activity was measured by an in vitro kinase assay using MBP peptide as a substrate as described in Materials and methods. Results shown are the mean of two independent experiments that were performed in duplicate. (Inset) p42mapk Figure 2 (a) Time course of bombesin-induced p74raf-1 activity. activation induced by various factors. BOR 15 cells in 100 mm BOR 15 cells in 100 mm dishes were washed twice in DMEM and dishes were washed twice with DMEM and incubated in 5 ml incubated in 5 ml DMEM with 1 nM bombesin for increasing raf-1 DMEM without (7) or with 1 nM bombesin (B), 200 nM PDB times at 378C. Cells were lysed and p74 was immunoprecipi- (P) or 5 ng/ml EGF (E) for 5 min at 378C. p42mapk was tated with polyclonal C12 anti-p74raf-1 antibody. p74raf-1 activity immunoprecipitated followed by an in vitro kinase reaction as was measured by a two stage in vitro immunecomplex kinase above. Results shown are from one experiment performed in assay as described in Materials and methods. Results shown are duplicate. Similar results were obtained in two independent the mean of two independent experiments performed in duplicate. experiments (Right) Eect of PKC inhibition and down (b) Eect of PKC inhibition and down regulation on p74raf-1 regulation on bombesin-stimulated p42mapk activity. Cultures of activity induced by various mitogens. BOR 15 cells in 100 mm BOR 15 cells (5 days after passage) were pretreated with (closed dishes were pretreated for 40 h with (closed bars) or without bars) or without 800 nM PDB for 48 h. Then the cells were 800 nM PDB. The cells were then washed twice with DMEM and washed twice in DMEM and incubated in 5 ml DMEM for 1 h in incubated in 5 ml DMEM for 1 h in the absence or presence the absence or presence of 3.5 mM GF 109203X (striped bars). The (striped bars) of 3.5 mM GF 109203X. Cells were then stimulated cells were stimulated without (7) or with 100 nM PDB, 1 nM for 3 min at 378C without (7) or with 200 nM PDB or 1 nM bombesin (Bom) or 5 ng/ml EGF for 5 min at 378C. The cells bombesin (Bom). p74raf-1 was immunoprecipitated and kinase were lysed and p42mapk was immunoprecipitated and kinase activity measured as above. Results are presented as a percent of activity assayed as above. Results shown are from one the bombesin control and are the mean of two independent representative experiment that was performed in duplicate. experiments performed in duplicate with the error bar showing Similar results were obtained in two independent experiments the range PKC-independent Gq activation of MAPK pathway A Charlesworth and E Rozengurt 2325 was measured by its ability to activate, in vitro,GST- fusion proteins of downstream kinases (i.e. MEK-1 and p42mapk). As shown in Figure 2a addition of bombesin at 1 nM induced a rapid and transient increase in p74raf-1 kinase activity. The maximum activation occurred 3 min after bombesin stimulation and was nearly all over by 10 min. Diacylglycerol-dependent isoforms of PKC (e.g. PKCa) have been proposed to directly activate p74raf-1, at least in some cell types (Sozeri et al., 1992; Kolch et al., 1993). However, PDB at all concentrations that induce a marked increase in the phosphorylation of the PKC substrate 80K/MARCKS in Rat-1 cells (Charlesworth et al., 1996) did not stimulate p74raf-1 activity in these cells suggesting that direct PKC activation does not lead to p74raf-1 stimulation in Rat-1 cells (Figure 2b). To determine the role of phorbol ester-sensitive isoforms of PKC in bombesin-stimulated p74raf-1 activity, quiescent BOR 15 cells were pretreated with 800 nM PDB for 40 h to down regulate PKC, or with 3.5 mM GF 109203X for 1 h prior to stimulation by various factors for 3 min. The results are shown in Figure 2b. Bombesin-induced p74raf-1 activity was not aected by inhibition or down regulation of PKC Figure 3 (Upper) Eect of pertussis toxin and Cytochalasin D on p74raf-1 activity induced by bombesin. (Left) BOR 15 cells in raf-1 demonstrating that bombesin activates p74 via a 100 mm dishes were washed twice in DMEM and incubated in PKC-independent route. In fact, a small stimulation 5 ml DMEM with 50 ng/ml pertussis toxin for 2 h (closed bars) was noticed in every case suggesting that GF 109203X as indicated. Cells were then stimulated for 3 min at 378C without (7) or with 1 mM bombesin (Bom), 2 mM LPA or 5 ng/ml EGF. and down regulation of PKC eliminates a constitutive raf-1 feed-back loop in Rat-1 cells. Thus, bombesin induces p74 was immunoprecipitated and kinase activity measured as above. Results shown are from one experiment that was p42mapk and p74raf-1 activation through a PKC- performed in duplicate with bars showing the range. Similar independent pathway in Rat-1 cells transfected with results were obtained in two independent experiments. (Right the bombesin receptor. BOR 15 cells in 100 mm dishes were washed twice in DMEM and incubated in 5 ml DMEM with 2 mM cytochalasin D for 2 h (closed bars) as indicated. Cells were then stimulated for 3 min at raf-1 Bombesin stimulates p74raf-1 and p42mapk via a pertussis 378C without (7) or with 1 nM bombesin (Bom). p74 was toxin and cytochalasin D-insensitive pathway immunoprecipitated and kinase activity measured as above. Results are the mean of three independent experiments Many G protein-coupled receptors (e.g. M1 acetylcho- performed in duplicate with bars showing the standard error. (Lower) p42mapk and p74raf-1 activation induced by neuromedin B line and those for lysophosphatidic acid (LPA) and in Rat-1 cells. NMBR-1 cells in 100 mm dishes were pretreated ras raf-1 mapk thrombin) activate p21 , p74 and p42 via a PTx- for 40 h with 800 nM PDB (DR). These and parallel culture of sensitive Gi pathway (Alblas et al., 1993; Van Corven cells were washed twice with DMEM and incubated in 5 ml et al., 1993; Winitz et al., 1993). We considered the DMEM in the absence or presence 3.5 mM GF 109203X for 1 h possibility that the transfected bombesin/GRP receptor (GF) or 50 ng/ml pertussis toxin for 2 h (PTx), as indicated. Cells were then stimulated for 5 min (p42mapk) or 3 min (p74raf-1)at could be coupled to G in Rat-1 cells. To assess mapk i 378C without (7) or with 1 nM neuromedin B (NmB). p42 or mapk whether Gi mediated bombesin-induced activation of p74 was immunoprecipitated followed by in vitro kinase p74raf-1, BOR 15 cells were pretreated for 2 h with reactions as described above. Results are shown as a percent of 50 ng/ml PTx and then stimulated with bombesin, LPA the neuromedin B control and are the mean of two independent experiments performed in duplicate with the error bar showing or EGF for 3 min. As shown in Figure 3 (upper, left) the range. Where the error bar cannot be seen it lies within the bombesin-induced p74raf-1 activation was not prevented dimensions of the bar. As in other experiments shown above, the by pretreatment with PTx in these cells. Similarly, pretreatments did not signi®cantly aect the basal activity bombesin induced PTx-insensitive p42mapk activation in BOR 15 cells (results not shown). We veri®ed that in parallel cultures LPA and EGF induced p74raf-1 activation through PTX-sensitive and PTX-insensitive determined whether tyrosine phosphorylated p125FAK pathways, respectively (Figure 3, upper left). Thus, could act upstream of p74raf-1 activation in bombesin- these ®ndings con®rm that the bombesin/GRP receptor stimulated Rat-1 cells transfected with the bombesin/ transfected in Rat-1 cells induces p74raf-1 and p42mapk GRP receptor. As shown in Figure 3 (upper, right) raf-1 activation via a PTx-insensitive Gq pathway. p74 activation induced by bombesin was not Bombesin stimulates the tyrosine phosphorylation of inhibited in BOR 15 cells that were preincubated for p125 focal adhesion kinase (p125FAK) through a PKC- 2 h with 2 mM cytochalasin D, a pretreatment that independent pathway that is completely dependent on completely abolished p125FAK tyrosine phosphorylation the integrity of the actin cytoskeleton in Rat-1 cells in response to bombesin. Indeed there was a slight (Charlesworth et al., 1996). Tyrosine phosphorylated stimulation of basal p74raf-1 activity upon pretreatment p125FAK has been proposed to direct the formation of a with cytochalasin D. These results demonstrate that signalling complex that leads to the activation of the tyrosine phosphorylation of p125FAK can be dissociated MAP kinase cascade in integrin-mediated signal- from p74raf-1 activation in Rat-1 cells transfected with transduction (Schlaepfer et al., 1994). We therefore the bombesin/GRP receptor. PKC-independent Gq activation of MAPK pathway A Charlesworth and E Rozengurt 2326 As our results with bombesin in Rat-1 cells 2.5 min with bombesin or EGF. p21ras was immuno- transfected with the bombesin/GRP receptor imply precipitated and the bound guanine nucleotides were the existence of a novel PCK-independent pathway eluted and separated by thin layer chromatography. leading to p74raf-1 and p42mapk activation, we examined The results are shown in Figure 4. Bombesin ras whether this pathway is utilised by other Gq coupled increased the proportion of p21 in the GTP bound receptors in Rat-1 cells, such as the transfected form from 28 to 43%. In parallel cultures, EGF- neuomedin B receptor (Lach et al., 1995). The induced stimulation increased p21ras GTP from 28 to neuromedin B receptor, which is closely related to the 61%. Thus, activation of the transfected bombesin/ bombesin/GRP receptor (Corjay et al., 1991; Wada et GRP receptor stimulated the GTP loading of p21ras in al., 1991), has been previously shown to activate Rat-1 cells. p42mapk although the mechanism has not been determined (Lach et al., 1995). Figure 3 (lower) shows Conclusions and implications that addition of neuromedin B induced p42mapk and p74raf-1 activation in a PKC-independent and PTx- In contrast to receptor tyrosine kinases, the signal insensitive manner in Rat-1 cells transfected with the transduction steps leading to the activation of p42mapk neuromedin B receptor. and p44mapk by seven transamembrane domain recep-

tors are not well understood. Recently, Gi-coupled receptors have been shown to induce the activation of Bombesin stimulates p21ras GTP loading in transfected p42mapk and p44mapk via a p21ras-dependent pathway Rat-1 cells (Koch et al., 1994; Hawes et al., 1995). A second PTx-

Previous results demonstrated that addition of sensitive pathway involving Go has also been identi®ed

bombesin to Swiss 3T3 cells does not increase the (van Biesen et al., 1996). In contrast, Gq-coupled active, GTP-bound state, of p21ras (Satoh et al., 1990; receptors are thought to induce PTx-insensitive p42mapk Mitchell et al., 1995). However, the PKC- and PTx- and p44mapk activation through a pathway dependent independent stimulation of p74raf-1 activity induced by on phorbol ester-sensitive PKC isoforms (Hawes et al., bombesin in Rat-1 cells transfected with the 1995). In accord with this model, bombesin induces bombesin/GRP receptor prompted us to examine p42mapk and p44mapk activation in Swiss 3T3 cells via a whether this agonist could activate p21ras in these PKC-dependent and PTx-insensitive pathway without cells. Quiescent BOR 15 cells were labelled overnight promoting an increase in p21ras GTP. Bombesin does with [32P]orthophosphate and then stimulated for not increase p74raf-1 activation in these cells.

GDP

GTP

origin

mAb –+ –+ –+

– Bom EGF Figure 4 Bombesin stimulation of p21ras GTP loading. (Left) BOR 15 cells in 100 mm dishes were washed twice in DMEM without phosphate and incubated overnight in 5 ml DMEM without phosphate supplemented with 1 mg/ml BSA, 20 mM HEPES 32 and 1.6 mCi Pi. Cells were stimulated without (7) or with 1 nM bombesin (Bom) or 5 ng/ml EGF for 2.5 min at 378C. The cells were lysed and p21ras was immunoprecipitated with mAb 259 in duplicate (+) with non-speci®c controls (7) from the detergent phase as described in Materials and methods. The guanine nucleotides were eluted from Ras and resolved by thin layer chromatography on PEI cellulose plates as described in Materials and methods and visualised by autoradiography. Results shown are from one representative experiment that was performed with duplicate immunoprecipitations from duplicate cultures. Similar results were obtained in two independent experiments. (Right) Labelled nucleotides were quanti®ed with a Phosphorimager using ImageQuant from Molecular Dynamics. The results are presented as the mean percent of Ras in GTP bound form (GTP/ (GTP+GDP)6100)+s.e. of duplicate immunoprecipitations from duplicate cultures PKC-independent Gq activation of MAPK pathway A Charlesworth and E Rozengurt 2327 The results presented here demonstrate that NaCl, 30 mM sodium pyrophosphate, 50 mM NaF, 100 mM mapk raf-1 bombesin induces p42 and p74 activation in Na3VO4, 1% Triton X-100 and 1 mM phenylmethylsulpho- Rat-1 cells transfected with the bombesin/GRP nyl ¯uoride (PMSF), pH 7.6). Lysates were clari®ed by receptor via a PKC-independent pathway. Further- centrifugation at 15 000 g for 10 min at 48C. Immunopre- cipitation was performed using a polyclonal anti-p42mapk more, the stimulation of p74raf-1 activity is PTx- antibody raised against the C-terminal peptide of p42mapk insensitive, indicating that the bombesin/GRP recep- (EETARFQPGYRS) and incubating the samples on a tor does not couple to either Gi or Go in Rat-1 cells. rotating wheel for 2 h. Washed protein A-agarose beads Bombesin also stimulated the GTP loading of p21ras in (50 ml 1 : 1 slurry) were added for the second hour. Immune these cells. In addition, we found that ligand activation complexes were collected by centrifugation and washed of the Gq-coupled receptor for neuromedin B also leads twice in lysis buer without PMSF and twice in kinase raf-1 mapk to p74 and p42 activation via a PKC-indepen- buer (15 mM Tris-HCl, 15 mM MgCl2). The kinase dent pathway in Rat-1 cells. Thus, our results reveal a reaction was performed by resuspending the pellet in 25 ml of kinase assay cocktail containing kinase buer novel PKC-independent pathway in the action of Gq- coupled receptors and imply that these receptors have with 0.5 mM EGTA, 1 mg/ml MBP peptide (APRTPGGRR), 50 mM ATP, 50 mCi/mlof[g-32P]ATP. the potential to activate p42mapk and p44mapk through Incubations were performed under linear assay conditions multiple pathways. The ®ndings also emphasise the (10 min) at 308C and terminated by spotting 25 mlofthe importance of cell context in determining the signal supernatant onto Whatman P81 cation exchanger chroma- transduction pathways that link a speci®c receptor to tography paper. Free [g-32P]ATP was removed by washing the activation of the MAP kinase cascade. ®lters four times for 5 min in 0.5% orthophosphoric acid. Filters were immersed in acetone to aid drying before Cerenkov counting. The average radioactivity of two blank Materials and methods samples containing no immune complex was subtracted from the result of each sample. The speci®c activity of Cell culture [g-32P]ATP used was 900 ± 1200 c.p.m/pmol. Stock cultures of Rat-1 cells were maintained as described previously (Higgins et al., 1992) except that transfected cell p74raf-1 kinase assay lines were propagated in the presence of 0.5 mg/ml G418 (Lach et al., 1995; Charlesworth et al., 1996). For p74raf-1 activity was assayed using a modi®cation of the experimental purposes, 56104 or 36105 cells were method described by Alessi et al. (1994). Overnight subcultured in 33 or 100 mm Nunc Petri dishes respec- cultures of Escherichia coli strain BL21 DE3 transformed tively with Dulbecco's modi®ed Eagle's medium (DMEM) with GST-p42mapk and GST-MEK expression vectors supplemented with 5% foetal calf serum (FBS) and 0.5 mg/ (pGEX-2T) were diluted 1 in 10 and grown for 1 h. ml G418 as necessary and incubated in a humidi®ed GST-p42mapk was induced with 1 mM isopropylthio-b-D- atmosphere of 10% CO2, 90% air at 378C until the galactoside for 4 h at 378C and GST-MEK induced with became con¯uent and quiescent. 30 mM isopropylthio-b-D-galactoside at 278Covernight. The cells were then pelleted, freeze-thawed and lysed. Cell debris was removed by centrifugation and GST-fusion Transfection of Rat-1 cells proteins were puri®ed by adding 0.5 ml glutathione- A 1.4 kb insert containing the complete cDNA coding Sepharose beads and rotating for 30 min at 48C. The region of the murine bombesin/GRP preferring receptor GST-p42mapk was cleaved from the GST in thrombin buer cloned into the mammalian expression vector pCD2-neo while the GST-MEK was eluted from the glutathione- (BNR-pCD2-neo), and a 2 kb insert containing the Sepharose beads with 5 mM glutathione in 50 mM Tris pH complete cDNA coding region of the rat neuromedin B 8. Both preparations were then dialysed and concentrated. receptor cloned into pCD2 and under the transcriptional The purity of each preparation was checked by subjecting control of the SV40 early region promoter, were kindly the proteins to SDS ± PAGE and staining the gels with provided by Dr J Battey, and transfected into Rat-1 cells as Coomassie blue. For the p42mapk typical yields were 10 mg/ described previously (Lach et al., 1995; Charlesworth et al., ml of culture, with a purity of 495%. For the GST-MEK 1996). The two clones used were BOR 15 and NMBR-1. typical yields were 5 mg/ml of culture with a purity of The binding properties of the transfected receptors are 495%. consistent with the binding properties of endogenous Quiescent cells were treated as indicated in the Figure receptors. It should be pointed out that the Rat-1 cell legends and lysed in lysis buer as above. p74raf-1 lines transfected with the bombesin/GRP receptor used in immunoprecipitation was performed with an anity purified this study (BOR 15) expressed 50% fewer receptors per cell rabbit polyclonal antibody for 2 h with protein A-agarose than Swiss 3T3 cells (Charlesworth et al., 1996). Also the added for the second hour or a mouse monoclonal antibody NMBR-1 cell line expressed neuromedin B receptors with anti-mouse IgG-agarose. The results obtained with these (36104 per cell) at a lower level than BOR 15 expressed two antibodies were essentially the same. Immune complexes bombesin/GRP receptors (Lach et al., 1995). This is of were collected by centrifugation and then washed twice in importance in analysis of cellular responses because lysis buer with no PMSF and twice with buer A (50 mM overexpression of G protein linked receptors could over- Tris-HCl pH 7.5, 0.1 mM EGTA, 0.5 mM Na3VO4 and 0.1% come the speci®city of coupling to eector systems b-mercaptoethanol). Pellets were then resuspended in 30 ml (Ashkenazi et al., 1987). Thus, the clones selected in this buer B (30 mM Tris-HCl pH 7.5, 0.1 mM EGTA, 0.1% b- study, provide a useful system to examine the signalling mercaptoethanol, 6.5 mg/ml GST-MEK, 100 mg/ml p42mapk, pathways activated by bombesin and neuromedin B in 0.03% Brij-35, 10 mM Mg-ATP and 20 mM N-octyl-b-D- receptor transfected cells. glucopyranoside) and incubated at 308C for 30 min. The reaction was then terminated by diluting the supernatant in 40 ml of buer A with 1 mg/ml of bovine serum albumin Immune-complex assay for p42mapk activation (BSA) and, after mixing, 10 ml of the supernatant was Quiescent Rat-1 cells in 100 mm dishes were treated with removed to a fresh tube. p42mapk activation was then factors as described in each experiment and lysed at 48Cin measured using the MBP peptide phosphorylation assay 1 ml of lysis buer (10 mM Tris/HCl, 5 mM EDTA, 50 mM essentially as above. PKC-independent Gq activation of MAPK pathway A Charlesworth and E Rozengurt 2328 GTP loading of p21ras runin1.2Mammonium formate, 0.8 M HCl. Results were quantitated on a phosphorimager and analysed by Con¯uent and quiescent Rat-1 cells in 100 mm dishes were ImageQuant (Molecular Dynamics). washed twice in DMEM-phosphate and incubated in 5 ml DMEM without phosphate with 20 mM HEPES pH 7.5 32 and 1 mg/ml tissue culture grade BSA with 1.5 mCi of Pi Materials and labelled for 16 ± 18 h. Cells were then stimulated with GF 109203X was obtained from Calbiochem. Protein A- factors as described in the Figure legend for 2.5 min at agarose conjugate and Protein G-Sepharose was obtained 378C then rinsed with cold phosphate-buered saline and from Boehringer Mannheim, Germany. PDB, bombesin, lysed in Ras lysis buer (50 mM Tris pH 7.5, 100 mM BSA, anti-mouse IgG (whole molecule)-agarose were NaCl, 1 mM EGTA, 10 mM benzamidine, 10 mg/ml trypsin obtained from Sigma. FBS was obtained from Gibco BRL, inhibitor, 10 mg/ml aprotinin, 10 mg/ml leupeptin, 1 mM Life Technologies. The cDNA of the murine bombesin/GRP -dithiothreitol, 1 mg/ml BSA) with 5 mM MgCl and 1% DL 2 preferring receptor in pCD2-neo (BNR-pCD2-neo) and the Triton X-114. After removal of cell debris by centrifuga- neuromedin B receptor were kindly provided by Dr J Battey, tion, phases were partitioned in the presence of 0.5 M NaCl Laboratory of Biological Chemistry, National Cancer for 2 min at 378C(Basuet al., 1992). The detergent gel Institute, Bethesda, Maryland, U.S.A). Anti-p21ras mAb containing p21ras was recovered by spinning down for clone 259 was obtained from the hybridoma development 2 min at 14 000 g at 208C and redissolved in Ras lysis unit, Imperial Cancer Research Fund, London, United buer containing 1% Triton X-100, 0.5% sodium Kingdom. The anti-p74raf-1 anity puri®ed rabbit polyclo- deoxycholate, 0.05% SDS and 0.5 M NaCl. p21ras was nal antibody was obtained from Santa Cruz Biotechnology immunoprecipitated for 2 h with 25 ml [1 : 1 slurry] anti- Ltd., USA and the anti-p72raf-1 monoclonal antibody was Ras mAb 259 pre-coupled to Protein G-Sepharose (0.1 mg from Transduction Laboratories. p42mapk antisera was a 259 mAb per 1 ml [1 : 1 slurry] Protein G-Sepharose). All generous gift from Dr D J Withers of Department of immunoprecipitations were carried out in duplicate along Medicine, Hammersmith Hospital, London, UK. GST- with non-speci®c controls. Immunoprecipitates were MEK and GST-p42mapk expression vectors were kind gifts washed with 661mlof50mM HEPES, pH 7.4, 0.5 M from Professor C Marshall, Chester Beattie Laboratory, NaCl, 5 mM MgCl , 0.1% Triton X-100, 0.005% SDS and 2 Institute for Cancer Research, London, UK. 32P (10 mCi/ the nucleotides were eluted with 2 mM EDTA, 2 mMDL- i ml) and [g-32P]ATP (5000 Ci/mol) were obtained from dithiothreitol, 0.2% SDS, 0.5 mM GTP, 0.5 mM GDP, at Amersham International, UK. All other materials were of 688C for 20 min. Eluted nucleotides were separated by thin the purest grade commercially available. layer chromatography on PEI-cellulose plates (MERCK)

References

Alblas J, van Corven EJ, Hordijk PL, Milligan G and Howe LR and Marshall CJ. (1993). J. Biol. Chem., 268, Moolenaar WH. (1993). J. Biol. Chem. 268, 22235 ± 22238. 20717 ± 20720. Alessi DR, Saito Y, Campbell DG, Cohen P, Sithanandam Koch WJ, Hawes BE, Allen LF and Lefkowitz RJ. (1994). G, Rapp U, Ashworth A, Marshall CJ and Cowley S. Proc.Natl.Acad.Sci.USA,91, 12706 ± 12710. (1994). EMBO J., 13, 1610 ± 1619. KolchW,HeideckerG,KochsG,HummelR,VahidiH, Ashkenazi M, Winslow JW, Peralta EG, Peterson GL, Mischak H, Finkenzeller G, Marme D and Rapp UR. Schimerlik MI, Capon DJ and Ramachandran J. (1987). (1993). Nature, 364, 249 ± 252. Science, 238, 672 ± 675. Lach EB, Broad S and Rozengurt E. (1995). Cell Growth Basu TN, Gutmann DH, Fletcher JA, Glover TW, Collins Di., 6, 1427 ± 1435. FS and Downward J. (1992). Nature, 356, 713 ± 715. Marshall CJ. (1995). Cell, 80, 179 ± 185. Battey JF, Way JM, Corjay MH, Shapira H, Kusano K, Mattingley RR, Sorisky A, Brann MR and Macara IG. Harkins R, Wu JM, Slattery T, Mann E and Feldman RI. (1994). Mol. Cell. Biol., 14, 7943 ± 7952. (1991). Proc. Natl. Acad. Sci. USA, 88, 395 ± 399. Mitchell FM, Heasley LE, Qian N-X, Zamarripa J and Carney DN, Cuttitta F, Moody TW and Minna JD. (1987). Johnson GL. (1995). J. Biol. Chem., 270, 8623 ± 8628. Cancer Res., 47, 821 ± 825. Oermans S, Heiler E, Spicher K and Schultz G. (1994). Charlesworth A, Broad S and Rozengurt E. (1996). FEBS Lett., 349, 201 ± 204. Oncogene, 12, 1337 ± 1345. Pang L, Decker SJ and Saltiel AR. (1993). Biochem. J., 289, Coer A, Fabregat I, Sinnett-Smith J and Rozengurt E. 283 ± 287. (1990). FEBS Lett., 263, 80 ± 84. Payne DM, Rossomando AJ, Martino P, Erickson AK, Her Corjay MH, Dobrzamski DJ, Way JM, Vallet J, Shapira H, J-H, Shabanowitz J, Hunt DF, Weber MJ and Sturgill Worland P, Sausville EA and Battey JF. (1991). J. Biol. TW. (1991). EMBO J., 10, 885 ± 892. Chem., 266, 18771 ± 18779. Rozengurt E. (1992). Curr. Opin. Cell Biol., 4, 161 ± 165. Crespo P, Xu N, Simonds WF and Gutkind JS. (1994). Rozengurt E and Sinnett-Smith J. (1983). Proc. Natl. Acad. Nature, 369, 418 ± 420. Sci. USA, 80, 2936 ± 2940. Crews CM, Alessandrini A and Erikson RL. (1992). Science, Russell M, Winitz S and Johnson G. (1994). Mol. Cell. Biol., 258, 478 ± 480. 14, 2343 ± 2351. Cuttitta F, Carney DN, Mulshine J, Moody TW, Fedorko J, Satoh T, Endo M, Nakafuku M, Nakamura S and Kaziro Y. Fischler A and Minna JD. (1985). Nature, 316, 823 ± 826. (1990). Proc. Natl. Acad. Sci. USA, 87, 5993 ± 5997. Davis R. (1993). J. Biol. Chem., 268, 14553 ± 14556. Schlaepfer DD, Hanks S, Hunter T and van der Geer P. Dikic I, Tokiwa G, Lev S, Courtneidge SA and Schlessinger (1994). Nature, 372, 786 ± 791. J. (1996). Nature, 383, 547 ± 550. Sethi T, Langdon S, Smyth J and Rozengurt E. (1991). Hawes BE, van Biesen T, Koch WJ, Luttrell LM and Cancer Res., 52, 2737s ± 2742s. Lefkowitz RJ. (1995). J. Biol. Chem., 270, 17148 ± 17153. Seuerlein T and Rozengurt E. (1995). J. Biol. Chem., 270, Higgins TE, Murphy AC, Staddon JM, Lax AJ and 1±9. Rozengurt E. (1992). Proc. Natl. Acad. Sci. USA, 89, Seuerlein T, Withers D, Broad S, Herget T, Walsh J and 4240 ± 4244. Rozengurt E. (1995). Cell Growth Di., 6, 383 ± 393. PKC-independent Gq activation of MAPK pathway A Charlesworth and E Rozengurt 2329 Seuerlein T, Withers DJ and Rozengurt E. (1996). J. Biol. Wan Y, Kurosaki T and Huang X-Y. (1996). Nature, 380, Chem., 271, 21471 ± 21477. 541 ± 544. Sozeri O, Vollmer K, Liyanage M, Frith D, Kour G, Mark Winitz S, Russell M, Qian N-X, Gardner A, Dwyer L and GE and Stabel S. (1992). Oncogene, 7, 2259 ± 2262. Johnson GL. (1993). J. Biol. Chem., 268, 19196 ± 19199. Toullec D, Pianetti P, Coste H, Bellevergue P, Grand-Perret Withers DJ, Bloom SR and Rozengurt E. (1995). J. Biol. T,AjakaneM,BaudetV,BoissinP,BoursierE,Loriolle Chem., 270, 21411 ± 21419. F, Duhamel L, Charon D and Kirilovsky J. (1991). J. Biol. Wu J, Harrison JK, Vincent LA, Haysted C, Haysted TAJ, Chem., 266, 15771 ± 15781. Michel H, Hunt DF, Lynch KR and Sturgill TW. (1993). van Biessen T, Hawes BE, Raymond JR, Luttrell LM, Koch Proc.Natl.Acad.Sci.USA,90, 173 ± 177. WJ and Lefkowitz RJ. (1996). J. Biol. Chem., 271, 1266 ± Yeo E-J and Exton J. (1995). J. Biol. Chem., 270, 3980 ± 3988. 1269. Zachary I and Rozengurt E. (1985). Proc. Natl. Acad. Sci. Van Corven EJ, Hordijk PL, Medema RH, Bos JL and USA, 82, 7616 ± 7620. Moolenaar WH. (1993). Proc. Natl. Acad. Sci. USA. 90, Zachary I and Rozengurt E. (1987). J. Biol. Chem., 262, 1257 ± 1261. 3947 ± 3950. Wada E, Way J, Shapira H, Kusano K, Lebacq-Verheyden AM, Coy D, Jenson R and Battey J. (1991). Neuron, 6, 421 ± 430.