Cyclic AMP activates Ras - Nature

Oncogene (2000) 19, 3609 ± 3615 ã 2000 Macmillan Publishers Ltd All rights reserved 0950 ± 9232/00 $15.00 www.nature.com/onc

Cyclic AMP activates Ras Oxana M Tsygankova1, Erik Kupperman1,2, Wei Wen1,3 and Judy L Meinkoth*,1 1

Department of Pharmacology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania, PA 19104-6084, USA

In addition to protein kinase A (PKA), cAMP regulates the activity of cAMP-gated channels and Rap1-speci®c guanine nucleotide exchange factors. We tested the hypothesis that the targets of cAMP might also include regulators of the Ras protooncogene. In rat thyroid cells, thyrotropin (TSH) stimulates proliferation through a cAMP-mediated pathway that requires Ras activity. Interference with Ras impairs DNA synthesis stimulated by TSH as well as cAMP elevating agents and analogs, demonstrating that the requirement for Ras lies downstream of cAMP. Although cAMP stimulates proliferation, microinjection of the puri®ed PKA catalytic subunit failed to do so, suggesting that factors in addition to PKA are required for cAMP-stimulated cell cycle progression. When added to thyroid cells expressing human Ha-Ras, TSH rapidly and markedly increased the proportion of GTP-bound Ras. Ras activity was increased within 1 min of TSH addition, maximal at 5 ± 15 min, and declined to basal levels 30 ± 60 min after hormone treatment. Cyclic AMP elevating agents elicited similar eects on Ras, indicating that TSH activates Ras through a cAMP-mediated pathway. Although cAMP-mediated, Ras activation by TSH and cAMP was independent of PKA activity. Moreover, cAMP-stimulated Ras activation was not impaired by tyrosine kinase inhibitors. These results indicate that cAMP activates targets in addition to PKA in thyroid cells, and that these targets may include regulators of Ras. The ability of cAMP elevating agents to activate Ras in addition to PKA may explain the inability of the PKA catalytic subunit to stimulate DNA synthesis in thyroid cells. Oncogene (2000) 19, 3609 ± 3615. Keywords: cAMP; Ras; proliferation; thyroid cells Introduction Cyclic AMP elicits dierential eects on cell proliferation (reviewed in Iyengar, 1996). In many cells, cAMP has no eect on cell proliferation, while in others it inhibits growth. In endocrine cells, secondary rat Schwann cells and Swiss 3T3 ®broblasts, elevations in cAMP are mitogenic. While the growth inhibitory eects of cAMP include impaired signaling from Ras to Raf-1, the mechanism through which cAMP stimulates proliferation remains to be resolved. Mutations that give rise to constitutive activity of cAMP*Correspondence: JL Meinkoth Current addresses: 2Department of Biochemistry and Biophysics, University of California, San Francisco, San Francisco, CA 941430448, USA; 3Department of Surgery, Harvard Medical School, Boston, MA 02115, USA Received 23 November 1999; revised 2 May 2000; accepted 10 May 2000

mediated signaling pathways have been identi®ed in human tumors. Mutations in Gas (Lyons et al., 1990; Suarez et al., 1991) and in the TSH receptor (Parma et al., 1993) that result in constitutive activation of cAMP-mediated signaling are prevalent in human thyroid tumors. Despite this, very little is known regarding how cAMP elicits proliferation in some cells and impairs it in others. TSH, the physiologic regulator of thyroid epithelial cells, stimulates proliferation and thyroid-speci®c gene expression through cAMP-mediated signaling pathways. Gas plays an important and essential role in TSH signaling. Expression of an activated Gas mutant reproduced the mitogenic eects of TSH in FRTL-5 cells (Muca and Vallar, 1994). Furthermore, microinjection of a Gas-speci®c antibody abolished TSHstimulated DNA synthesis in Wistar rat thyroid (WRT) cells (Meinkoth et al., 1992). Consistent with these ®ndings, treatment of rat thyroid cells with cholera toxin, forskolin or cAMP analogs is sucient to stimulate cell cycle progression, which requires PKA activity (Kupperman et al., 1993). Paradoxically, although cAMP is mitogenic, overexpression of the PKA catalytic subunit was insucient to stimulate mitogenesis in canine thyroid cells (Dremier et al., 1997). These results suggest that although PKA is required for cAMP-mediated DNA synthesis, it is insucient to mediate cAMP-stimulated cell cycle progression. We reported that microinjection of a dominant negative Ras protein impaired the mitogenic activity of cAMP elevating agents (Kupperman et al., 1993), implicating a role for Ras downstream of cAMP in thyroid cells. We now extend these ®ndings to demonstrate that TSH activates Ras through a novel cAMP-mediated signaling pathway. Furthermore, the eects of cAMP on Ras activation are independent of PKA, providing an explanation for the disparate eects of cAMP and PKA on thyroid cell proliferation. These ®ndings are the ®rst to demonstrate that the intracellular targets of cAMP include regulators of Ras activity. Results Cyclic AMP is a mitogenic signal for Wistar rat thyroid (WRT) cells. Interference with PKA by microinjection of the heat stable inhibitor impaired the ability of TSH to stimulate DNA synthesis (Kupperman et al., 1993). To determine whether PKA was sucient to stimulate DNA synthesis, we microinjected puri®ed PKA catalytic subunit (PKA-C) into quiescent WRT cells. We ®rst examined the activity of PKA-C in injected cells by monitoring its eects on CRE-regulated gene expression. We used WRT cells stably expressing a CRE-regulated lacZ

Cyclic AMP activates Ras OM Tsygankova et al

3610

gene for these studies (Kupperman et al., 1993; Meinkoth et al., 1992). As we have previously shown, none of these cells express b-galactosidase in the absence of cAMP elevation (i.e., Cass and Meinkoth, 1998). Following treatment with TSH or cAMP elevating agents, b-galactosidase is expressed in most cells (70 ± 80% depending upon the dose and times analysed). Following injection at 2.0 mg/ml, PKA-C stimulated b-galactosidase expression in 74% of the injected cells (n=695) at 6 h post injection, verifying its activity in the injected cells. PKA-C was then injected into quiescent WRT cells and its ability to stimulate DNA synthesis was examined. Injection of PKA-C at several concentrations was insucient to stimulate DNA synthesis measured at 48 h post injection (Table 1), although it was capable of stimulating CREregulated gene expression at these times (data not shown). These ®ndings support those observed in canine thyroid cells (Dremier et al., 1997) and suggest that factors regulated by cAMP, in addition to PKA, are required for cell cycle progression. We previously reported that TSH-stimulated DNA synthesis is impaired following microinjection of a dominant negative mutant Ras protein (RasN17) (Kupperman et al., 1993). As shown in Table 1, microinjection of a Ras interfering antibody dosedependently reduced TSH-stimulated DNA synthesis. Injection of the interfering Ras antibody also reduced DNA synthesis stimulated by cholera toxin and forskolin, con®rming our ®ndings that Ras is required for the mitogenic activity of cAMP elevating agents. To determine whether the requirement for Ras was conserved in other thyroid cells, similar experiments were performed in FRTL-5 cells. Injection of the 259 antibody reduced DNA synthesis stimulated by TSH and fully supplemented growth medium containing TSH (3H). These experiments verify that Ras is required for the mitogenic activity of cAMP in multiple thyroid cell lines. Based upon these results, together with the recent discovery of cAMP-regulated GEFs for Rap1 (DeRooij et al., 1998; Kawasaki et al., 1998), we examined whether cAMP activated Ras in thyroid cells. Because Ras is expressed at very low levels in rat thyroid cells, we were obliged to monitor Ras activation in thyroid cells stably overexpressing human H-Ras (RasG12 cells) (Kupperman et al., 1996). RasG12 cells, incubated in growth factor-de®cient

basal medium for 48 h, were stimulated with TSH for various times and Ras activity was examined using the RalGDS Ras binding domain (RBD) to selectively isolate GTP-bound Ras (Figure 1a). Very low levels of GTP-Ras were retrieved from cells in basal medium (row 2). TSH increased the proportion of GTP-Ras within 1 min of treatment, and exerted maximal eects on Ras activity at 5 ± 15 min. Subsequently, the levels of GTP-Ras slowly declined to basal levels 30 ± 60 min after treatment. The alterations in the level of GTPRas were not a consequence of changes in Ras expression in response to cAMP elevating agents (row 1). Densitometric analysis revealed that TSH increased the proportion of GTP-Ras by 2.7+0.3-fold (measured at 15 min; n=4, P50.005). Although GTP-Rap1 also binds to the RalGDS RBD (reviewed in Bos, 1998), the migration of HAtagged human H-Ras (30 kDa) on SDS ± PAGE diers

Figure 1 TSH and cAMP activate Ras. RasG12 cells incubated in basal medium (Cass and Meinkoth, 1998) for 48 h were stimulated with 1 mU/ml TSH (a) or 1 mM 8BrcAMP (b) for various times. Cell lysates were incubated with glutathioneSepharose coupled peptides corresponding to RalGDS RBD (Hofer et al., 1994) or Raf-1 RBD (Kolch et al., 1996) as described in Materials and methods. Bound GTP-Ras was eluted from the beads and analysed by immunoblotting with anti-Ras (aRas) or anti-HA (aHA) antibodies. Whole cell extracts were analysed in parallel with anti-Ras antibodies. Four independent experiments were performed for each cAMP-elevating agent with similar results

Table 1 Role of Ras and PKA in cAMP-stimulated DNA synthesis Cell line (treatment)

Injected protein

WRT (starved) WRT (starved) WRT (starved) WRT (starved) WRT (TSH) WRT (TSH) WRT (CTd) WRT (fskc) WRT (3H) WRT (TSH) FRTL-5 (3H) FRTL-5 (TSH)

PKA-Ca PKA-C PKA-C PKA-C 259 259 259 259 259 IgG 259 259

a

Concentration 0.5 1.0 2.0 4.0 5.0 2.5 5.0 5.0 5.0 6.0 2.5 2.5

mg/ml mg/ml mg/ml mg/ml mg/ml mg/ml mg/ml mg/ml mg/ml mg/ml mg/ml mg/ml

n (injected)

% BrdU-positive injected cells

n (control)

% BrdU-positive control cells

595 170 474 93 943 231 470 218 196 266 542 634

25 18 17 2 11 27 7 3 19 67 43 7

1028b 296 985 223 1569 346 689 355 387 626 650 1633

19 16 13 13 44 45 58 47 78 61 78 26

Recombinant C subunit was co-injected with rabbit IgG (6.0 mg/ml) in order to identify injected cells. DNA synthesis was examined at 48 h post injection. bControl cells are uninjected cells adjacent to injected cells on the same coverslips. BrdU incorporation was similar in control cells and cells injected with IgG alone under all conditions. cForskolin. dCholera toxin Oncogene


substantially from that of Rap1 (24 kDa) (data not shown). Nonetheless, the identity of the protein retrieved by the RalGDS RBD was con®rmed to be Ras by immunoblotting with an antibody to the HA epitope tag (row 3). In addition, we con®rmed the eects of TSH on Ras activity using the N-terminus of Raf-1 as an activation-speci®c probe to isolate GTPRas. Using this peptide, TSH activated Ras with kinetics similar to those seen using the RalGDS RBD (row 4). To determine if the eects of TSH on Ras activity were cAMP-mediated, RasG12 cells were treated with forskolin and 8BrcAMP. In four independent experiments with each reagent, both cAMP elevating agents activated Ras to a similar level as did TSH (forskolin, 4.0+1.2-fold, P50.05; 8BrcAMP, 3.3+0.7-fold, P50.05). 8BrcAMP stimulated Ras activity with kinetics similar to those in response to TSH (Figure 1b, row 2). This is the ®rst demonstration that cAMPelevating agents can directly activate Ras. While many of the eects of cAMP are mediated through PKA, cAMP directly binds cyclic nucleotidegated channels (Zufall et al., 1997) and Rap1-speci®c GEFs (DeRooij et al., 1998; Kawasaki et al., 1998). In WRT cells, cAMP-mediated signaling pathways stimulated by TSH diverge into PKA-dependent pathways to p70 ribosomal S6 kinase (p70S6k) and PKA-independent pathways to Akt (Cass et al., 1999). These ®ndings prompted us to examine whether TSH eects on Ras activity require PKA activity. We used two highly speci®c and independently acting PKA inhibitors, H89 (Chijiwa et al., 1990) and RpcAMPS (Rothermel and Botelho, 1988) for these studies. Pretreatment with 25 mM H89, but not with lower concentrations, blocked CREB phosphorylation stimulated by TSH (Figure 2a). These results agree well with our previous ®ndings that 25 mM H89 is required to abolish cAMP response element (CRE)-regulated gene expression in these cells (Cass et al., 1999). These ®ndings also verify that TSH activates PKA under these conditions. In contrast to its inhibitory eects on CREB phosphorylation and CRE-regulated gene expression, H89 had no eects on TSH-stimulated Ras activity (Figure 2b). Similarly, pretreatment with 250 mM RpcAMPS (the lowest concentration sucient to block CRE-regulated gene expression; Cass et al., 1999), had no eect on forskolin-stimulated Ras activity (Figure 2c). These experiments demonstrate that cAMP eects on Ras activity do not require PKA activity. Activation of the b2 adrenergic receptor stimulates Ras activity through a bg subunit-mediated pathway that requires tyrosine kinase activity (reviewed in Luttrell et al., 1997). To determine whether tyrosine kinase activity was required for TSH-induced Ras activation, Ras G12 cells were pretreated with genistein or herbimycin A prior to stimulation with TSH (Figure 3a). Neither inhibitor impaired Ras activation by TSH, although they individually reduced IGF-I-stimulated MAPK activity (Figure 3b). Similarly, forskolinstimulated Ras activity was not impaired in cells pretreated with PP2, a speci®c inhibitor of Src family tyrosine kinases (Figure 3c). Together, these data indicate that cAMP-mediated Ras activation is not mediated by intervening tyrosine kinases.

3611

Figure 2 TSH-induced Ras activation is PKA-independent. (a) Quiescent WRT cells were pretreated with indicated concentrations of H89 (mM) for 60 min and then stimulated with TSH for 45 min. After stimulation, cells were lysed and analysed by Western blotting with anti-phospho-CREB antibody (CREB-p). (b, c) RasG12 cells were pretreated with 25 mM H89 for 60 min (b) or 250 mM RpcAMPS for 20 min (c) and then stimulated with 1 mU/ml TSH (b) or 10 mM forskolin (c) for 15 min. Ras activity was monitored using RalGDS RBD-beads as described in Figure 1. Inclusion of H89 had no eect on basal levels of Ras activity. In some experiments, inclusion of RpcAMPS resulted in a small increase in basal levels of GTP-Ras

Overexpression of cellular Ras can alter intracellular signaling pathways. Under the conditions employed to measure Ras activity, RasG12 cells were quiescent (Table 2). Stimulation with TSH or 8BrcAMP at the same concentrations that activated Ras stimulated DNA synthesis (Table 2), suggesting that cAMPincreased Ras activity results in cell cycle progression. These data agree well with the requirement for Ras in cAMP-stimulated cell cycle progression in parental WRT cells (Kupperman et al., 1993). Given that TSH elicits only very transient and weak eects on MAPK phosphorylation (Miller et al., 1998) and that cAMPstimulated DNA synthesis is not impaired by interference with Raf-1 or MEK1 (Al-Alawi et al., 1995), we speculate that Ras activation stimulates DNA synthesis through alternate eector pathways. In support of this idea, TSH stimulated Akt phosphorylation in RasG12 cells, an eect that like Ras activation, was independent of PKA activity (Figure 4). TSH stimulated higher levels of Akt phosphorylation in RasG12 cells than in parental cells, consistent with the overexpression of cellular Ras in RasG12 cells. These results agree well with the ability of PI3K inhibitors to impair cAMP-stimulated DNA synthesis in parental thyroid cells (Cass et al., 1999). Discussion Although G protein-coupled receptors stimulate MAPK activity through a and bg mediated signaling pathways (reviewed in Luttrell et al., 1997), there are only a few examples where Ga subunits have been reported to activate Ras (i.e., Collins et al., 1996). Gscoupled b-adrenergic receptors stimulate MAPK Oncogene


3612

Figure 3 Ras activation does not require tyrosine kinase activity. (a) RasG12 cells were pretreated with 100 mM genistein (Gen) for 1 h or 1 mM herbimycin A (HerA) for 18 h and stimulated with 1 mU/ml TSH for 15 min. Ras activation was monitored using RalGDS RBD-beads. Inclusion of tyrosine kinase inhibitors had no eect on basal levels of GTP-Ras. Two experiments were performed with similar results for each inhibitor. (b) Cells pretreated with genistein or herbimycin A were stimulated with 200 ng/ml IGF-1 for 1 min, whole cell extracts prepared and analysed by immunoblotting with anti-phosphoMAPK (MAPK-P) antibody. (c) RasG12 cells were pretreated with 1 mM PP2 (Sigma) for 1 h and stimulated with 10 mM forskolin for 5 min. Ras activity was assessed using RalGDS RBD-beads. Two experiments were performed with similar results

Table 2 Cyclic AMP stimulates DNA synthesis in RasG12 cells Treatment Basal TSH (1 mU/ml) 8BrcAMP (1 mM)

% BrdU-positive nuclei 11+1.0 44.3+1.5* 56.7+17.8**

RasG12 cells, incubated in basal medium for 48 h, were stimulated with BrdU-containing basal medium or medium supplemented with 1 mU/ml TSH or 1 mM 8BrcAMP for 48 h. DNA synthesis was monitored by immunostaining as described in (Cass and Meinkoth, 1998). Results shown are mean+s.e. from three independent experiments (*P50.005, **P50.05)

Figure 4 TSH stimulates Akt phosphorylation in RasG12 cells through a PKA-independent mechanism. Parental WRT and RasG12 cells incubated in basal medium for 48 h were stimulated with TSH (1 mU/ml) for 45 min in the presence or absence of pre-treatment with H89 (25 mM) for 60 min. Total cell lysates were immunoblotted with a phospho-speci®c (S473) Akt antibody. A representative experiment of three experiments performed with similar results is shown

activity through a Ras-dependent signaling pathway (Crespo et al., 1994). However, in this instance, MAPK activation was mediated through a Gbg- and Rasdependent mechanism that was subject to repression by expression of an activated Gas mutant or PKA Oncogene

activation. Similarly, in HEK293 cells, stimulation of b-adrenergic receptors resulted in Gbg-, Src- and Rasdependent activation of MAPK. In this case, PKAmediated phosphorylation altered receptor coupling from Gas to Gai (Daaka et al., 1997). Activation of the Gs-coupled A2A-adenosine receptor activated Ras and MAPK, however these eects were shown to be independent of Gas and cAMP (Seidel et al., 1999). Glycoprotein hormones such as TSH regulate proliferation and dierentiation in their endocrine target tissues through cAMP-mediated signaling pathways. We reported that TSH, as well as cAMP elevating agents and analogs, require Ras to elicit their mitogenic eects in thyroid cells. We now extend these ®ndings to demonstrate that cAMP directly activates Ras in rat thyroid cells. This is the ®rst report to demonstrate Gas- and cAMP-mediated Ras activation. Ras activation by cAMP appears to occur by a novel mechanism. Our preliminary mechanistic studies revealed that Ras activation is mediated through Ga rather than Gbg subunits. The eects of TSH receptor activation on Ras were reproduced by direct activation of Gas by cholera toxin, or by downstream eectors of Gas including adenylyl cyclase and cAMP. Unlike Ras activation downstream from many G protein-coupled receptors, cAMP-stimulated Ras activation was independent of tyrosine kinase activity, further evidence that Ras is activated by the Gas/cAMP pathway and not through intervening Gbg subunits. Additional insight into the mechanism of cAMP-regulated Ras activity was provided by our ®nding that PKA activity is not required for the stimulatory eects of TSH or cAMP elevating agents on Ras activity. While the precise mechanism through which cAMP regulates Ras will require extensive analysis, there are several


potential mechanisms through which cAMP could activate Ras. First, the recent discovery of cAMPregulated Rap1GEFs (DeRooij et al., 1998; Kawasaki et al., 1998) establishes the precedent for the existence of cAMP-regulated RasGEFs, although to our knowledge these have not yet been described. It is intriguing, however, that novel calcium-activated RasGEFs have been identi®ed (Ebinue et al., 1998; Farnsworth et al., 1995). Second, the eects of cAMP on Ras activity could be mediated through eects on GAP activity. Several examples of crosstalk between Ga subunits and GAP proteins have been described. In addition to bifunctional RGS proteins that function both as terminators of serpentine receptor signaling and RhoGEFs (reviewed in Hall, 1998), G protein a subunits have been reported to interact directly with a variety of GAPs for small G proteins. Ga12 binds directly to Gap1m, a RasGAP (Jiang et al., 1998). Although this interaction resulted in increased GAP activity, interactions between unactivated Gao and a novel Rap1GAP resulted in increased Rap1 activity (Jordan et al., 1991). Novel Rap1GAPs have also been identi®ed as binding partners for Gai (Mochizuki et al., 1999). We do not favor a similar mechanism for cAMP-stimulated Ras activation, as interaction between Gas and RasGAPs does not readily explain how cAMP analogs stimulate Ras activity, without postulating the existence of cAMP-regulated RasGAPs. Still other mechanisms for cAMP-stimulated Ras activation can be envisioned. Cyclic AMP activates Rap1 in many cells (reviewed in Bos, 1998). As a consequence of their identical eector domains, Ras and Rap1 compete for binding to many of the same proteins, including RasGAP (reviewed in Bos, 1997). Sequestration of RasGAP by GTP-bound Rap1 could lead to increased levels of Ras activity. In this case, the role of cAMP might be limited to Rap1 activation. Which, if any, of these mechanisms is responsible for cAMP eects on Ras activity is presently under investigation. The ®nding that PKA is not required for cAMP stimulated Ras activity is consistent with recent observations from our laboratory and from others that cAMP activates multiple signaling pathways in thyroid cells. We reported that cAMP-mediated signals diverge to include PKA-dependent signals to p70S6k and PKA-independent signals to PI3K (Cass et al., 1999). Both pathways appear to be required for proliferation since interference with mTOR/p70S6k via treatment with rapamycin (Cass and Meinkoth, 1998) or treatment with highly speci®c PI3K inhibitors (Cass et al., 1999) impairs cAMP-stimulated proliferation. It is tempting to speculate that Ras initiates the latter pathway given its ability to bind to and activate PI3K (Downward, 1997), an interaction which may be particularly favored in thyroid cells where PKA impairs Ras signaling to Raf-1 (Al-Alawi et al., 1995). In support of this idea, TSH eects on Akt phosphorylation were enhanced in RasG12 cells. In addition, the eects of TSH on Akt phosphorylation, as on Ras activation, were independent of PKA. The eects of TSH on dierentiated gene expression were also shown to involve both PKA-dependent and -independent pathways in thyroid cells (Ohno et al., 1999). The regulation of Ras by cAMP provides a potential mechanism to explain the disparate eects of cAMP

and PKA on thyroid cell proliferation. As reported here, microinjection of the PKA catalytic subunit is insucient to stimulate DNA synthesis in thyroid cells, even though cAMP elevating agents and analogs do so. The ability of cAMP to activate Ras in the absence of PKA activity may explain the paradox that while cAMP elevating agents stimulate DNA synthesis, overexpression of the PKA catalytic subunit fails to do so in thyroid cells derived from multiple species (this report and Dremier et al., 1997). The ability of cAMP to activate Ras is compatible with the important role of both cAMP and Ras in the regulation of thyroid cell proliferation. Ras mutations are particularly frequent in human thyroid tumors (reviewed in Fagin, 1994; Wynford-Thomas, 1997), and expression of activated Ras confers transient TSH-independent proliferation to primary human thyrocytes (Lemoine et al., 1990) and continuous rat thyroid cells (Fusco et al., 1987; Miller et al., 1998). Although there are numerous examples of crosstalk where cAMP and Ras elicit opposing eects, for example on MAPK activity, in many instances cAMP and Ras elicit similar biological eects. Ras and cAMP stimulate neurite extension and survival in PC12 cells (Schubert et al., 1978) and sympathetic neurons (Creedon et al., 1996; Rydel and Greene, 1988). These ®ndings, together with those reported here, reveal that there are likely to be additional sites of crosstalk between intracellular signals initiated by Ras and cAMP.

3613

Materials and methods Reagents TSH, 8BrcAMP, forskolin, IGF-1 and RpcAMPS were from Sigma (St. Louis, MO, USA). Coon's modi®ed Ham's F12 medium was from Sigma and calf serum was from GIBCO BRL (Gaithersburg, MD, USA). Genistein, herbimycin A and H89 were from Calbiochem (La Jolla, CA, USA). The rat monoclonal Ras antibody (SC-35) and horseradish peroxidase (HRP) conjugated secondary antibodies were from Santa Cruz (Santa Cruz, CA, USA), the HA antibody was kindly provided by Dr Jerey Field (Department of Pharmacology, University of Pennsylvania). Phospho-speci®c antibodies to MAPK, CREB and Akt were obtained from Promega (Madison, WI, USA), Upstate Biotechnology (Lake Placid, NY, USA) and Biolabs (Beverly, MA, USA) respectively. Cell culture WRT and FRTL-5 cells were propagated in 3H medium (Coon's modi®ed Ham's F12 medium containing bovine TSH (1 mU/ml), insulin (10 mg/ml), transferrin (5 mg/ml), 5% calf serum) as described previously (Meinkoth et al., 1992). WRT cells expressing hemagglutinin (HA) epitope-tagged human H-Ras (RasG12 cells) (Kupperman et al., 1996) and CREregulated lacZ gene (CRE-WRT cells) (Kupperman et al., 1993; Meinkoth et al., 1992) were propagated at 378C in 5% CO2 in 3H medium supplemented with geneticin (150 mg/ml). Expression and purification of GST fusion proteins GST-RBD fusion proteins were isolated as described in (Miller et al., 1997). Brie¯y, sonicated bacteria were solubilized in 1% Triton X-100 and lysates were incubated Oncogene


3614

with glutathione-Sepharose overnight at 48C. Beads were washed with PBS and stored in 50% glycerol at 7208C. Microinjection and DNA synthesis PKA catalytic subunit (PKA-C) was expressed and puri®ed as described previously (Fantozzi et al., 1994). DNA synthesis in injected cells was assessed by bromodeoxyuridine (BrdU) incorporation as we described previously (Al-Alawi et al., 1995; Kupperman et al., 1993). Ras activation Ras activation was assessed as described in (DeRooij and Bos, 1997). Cells were grown to 80% con¯uence, starved in basal medium (Coon's modi®ed Ham's F12 medium devoid of all growth factors) for 48 h, and stimulated with TSH (1 mU/ml), forskolin (10 mM), 8BrcAMP (1 mM) or vehicle for the indicated times. In some experiments, basal medium was further supplemented with insulin (0.5 mg/ml). Inhibitors were added 60 min prior to stimulation. Following lysis in RIPA buer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 0.5% DOC, 1% NP40, 0.1% SDS, 25 mM NaF, 2 mM Na3VO4, aprotinin, leupeptin, pefabloc each 10 mg/ml), protein concentrations were determined using DC Protein Assay (BioRad Laboratories, Hercules, CA, USA). Total cell protein (400 mg) was incubated with glutathione-Sepharose coupled peptides corresponding to the Ras binding domains (RBDs) (5 mg/ml beads) of RalGDS (Hofer et al., 1994) or Raf-1 (Kolch et al., 1996) for 1 h at 48C. The beads were washed

three times with RIPA buer, bound proteins eluted in sample buer, heated for 5 min at 958C, separated on 12% SDS ± PAGE, transferred to nitrocellulose and subjected to immunoblotting with anti-Ras or anti-HA antibodies. Densitometric analysis was performed using a ScanMaker4 (Microtek) and NIH Image software (version 1.6.). Statistical analysis was made by paired Student's t-test. Immunoblotting Protein immunoblotting was performed as described previously (Cass and Meinkoth, 1998). Where protein phosphorylation was examined, total cell lysates were resolved on 10% SDS ± PAGE. Primary antibodies (0.25 mg/ml) were incubated overnight at 48C; HRP-secondary antibodies (1 : 3000) were incubated for 1 h at room temperature and detected using ECL (Amersham, Piscataway, NJ, USA) following manufacturer's recommendations.

Acknowledgments Since submission of this manuscript, a cAMP regulated Ras GEF was described (N. Pham, 2000, Curr Biol: 10, 555). We are grateful for helpful suggestions from Dr Margaret Chou regarding the Ras activation assay and from Lisa Cass regarding preparation of this manuscript. This work was supported by PHS grants DK45696 and DK02494 awarded to JL Meinkoth.

References Al-Alawi N, Rose DW, Buckmaster C, Ahn N, Rapp U, Meinkoth J and Feramisco JR. (1995). Mol. Cell. Biol., 15, 1162 ± 1168. Bos JL. (1997). Biochim. Biophys. Acta., 1333, M19 ± M31. Bos JL. (1998). EMBO J., 17, 6776 ± 6782. Cass LA and Meinkoth JL. (1998). Endocrinol., 139, 1991 ± 1998. Cass LA, Summers SA, Prendergast GV, Backer JM, Birnbaum MJ and Meinkoth JL. (1999). Mol. Cell. Biol., 19, 5882 ± 5891. Chijiwa T, Mishima A, Hagiwara M, Sano M, Hayashi K, Inoue T, Naito K, Toshioko T and Hidaka H. (1990). J. Biol. Chem., 265, 5267 ± 5272. Collins LR, Minden A, Karin M and Brown JH. (1996). J. Biol. Chem., 271, 17349 ± 17353. Creedon DJ, Johnson EM, and Lawrence JC. (1996). J. Biol. Chem., 271, 20713 ± 20718. Crespo P, Xu N, Simonds WF and Gutkind JS. (1994). Nature, 369, 418 ± 420. Daaka Y, Luttrell LM and Lefkowitz RJ. (1997). Nature, 390, 88 ± 91. DeRooij J and Bos JL. (1997). Oncogene, 14, 623 ± 625. DeRooij J, Zwartkruis FJT, Verheijen MHG, Cool RH, Nijman SMB, Wittinghofer A and Bos JL. (1998). Nature, 396, 474 ± 477. Downward J. (1997). Adv. Sec. Messenger Phosphopro. Res., 31, 1 ± 10. Dremier S, Pohl V, Poteet-Smith C, Roger P, Corbin J, Doskeland SO, Dumont JE and Maenhaut C. (1997). Mol. Cell. Biol., 17, 6717 ± 6726. Ebinue JO, Bottor DA, Chan EYW, Stang SL, Dunn RJ and Stone JC. (1998). Science, 280, 1082 ± 1086. Fagin JA. (1994). Annu. Rev. Med., 45, 45 ± 52. Fantozzi DA, Harootunian AT, Wen W, Taylor SS, Feramisco JR, Tsien RY and Meinkoth JL. (1994). J. Biol. Chem., 269, 2676 ± 2686.

Oncogene

Farnsworth CL, Freshney NW, Rosen LB, Ghosh A, Greenberg ME and Feig LA. (1995). Nature, 376, 524 ± 527. Fusco A, Berlinghieri MT, Fiore PPD, Portella G, Grieco M and Vecchio G. (1987). Mol. Cel. Biol., 7, 3365 ± 3370. Hall A. (1998). Science, 279, 509 ± 514. Hofer F, Fields S, Schneider C and Martin GS. (1994). Proc. Natl. Acad. Sci. USA, 91, 11089 ± 11093. Iyengar R. (1996). Science, 271, 461 ± 463. Jiang Y, Ma W, Wan Y, Kozasa T, Hattori S and Huang XY. (1998). Nature, 395, 808 ± 813. Jordan JD, Carey KD, Stork PJS and Iyengar R. (1999). J. Biol. Chem., 274, 21507 ± 21510. Kawasaki H, Springett GM, Mochizuki N, Toki S, Nakaya M, Matsuda M, Housman DE and Graybiel AM. (1998). Science, 282, 2275 ± 2279. Kolch W, Philipp A, Mischak H, Dutil EM, Mullen TM, Feramisco JR, Meinkoth JL and Rose DW. (1996). Oncogene, 6, 1304 ± 1314. Kupperman E, Wen W and Meinkoth JL. (1993). Mol. Cell. Biol., 13, 4477 ± 4484. Kupperman E, Woord D, Wen W and Meinkoth JL. (1996). Endocrinol., 137, 96 ± 104. Lemoine NR, Staddon S, Bond J, Wyllie FS, Shaw JJ and Wynford-Thomas D. (1990). Oncogene, 5, 1833 ± 1837. Luttrell LM, Biesen Tv, Hawes BE, Koch WJ, Krueger KM, Touhara K and Lefkowitz RJ. (1997). Adv. Sec. Mess. Phosphopro. Res., 31, 263 ± 277. Lyons J, Landis CA, Harsh G, Vallar L, Grunewald K, Feightinger H, Duh Q-Y, Clark OH, Kawasaki E, Bourne HR and McCormick F. (1990). Science, 249, 655 ± 659. Meinkoth JL, Goldsmith PK, Spiegel AM, Feramisco JR and Burrow GN. (1992). J. Biol. Chem., 267, 13239 ± 13245.


Miller MJ, Prigent S, Kupperman E, Rioux L, Park S-H, Feramisco JR, White MA, Rutkowski JL and Meinkoth JL. (1997). J. Biol. Chem., 272, 5600 ± 5605. Miller MJ, Rioux L, Prendergast GV, Cannon S, White MA and Meinkoth JL. (1998). Mol. Cell. Biol., 18, 3718 ± 3726. Mochizuki N, Ohba Y, Kiyokawa E, Kurata T, Murakami T, Ozaki T, Kitabatake A, Nagashima K and Matsuda M. (1999). Nature, 400, 891 ± 894. Muca C and Vallar L. (1994). Oncogene, 9, 3647 ± 3653. Ohno M, Zannini M, Levy O, Carrasco N and DiLauro R. (1999). Mol. Cell. Biol., 19, 2051 ± 2060. Parma J, Duprez L, Sande JV, Cochaux P, Gervy C, Mockel J, Dumont J and Vassart G. (1993). Nature, 365, 649 ± 651. Rothermel JD and Botelho LHP. (1988). Biochem. J., 251, 757 ± 762.

Rydel RE and Greene LA. (1988). Proc. Natl. Acad. Sci. USA, 85, 1257 ± 1261. Schubert D, LaCorbiere M, Whitlock C and Stallcup W. (1978). Nature, 273, 718 ± 723. Seidel MG, Klinger M, Freissmuth M and Holler C. (1999). J. Biol. Chem., 274, 25833 ± 25841. Suarez HG, Villard JAd, Cailou B, Schlumberger M, Parmentier C and Monier R. (1991). Oncogene, 6, 677 ± 679. Wynford-Thomas D. (1997). Horm. Res., 47, 145 ± 157. Zufall F, Shepherd GM and Barnstable CJ. (1997). Curr. Opin. Neurobiol., 7, 404 ± 412.

3615

Oncogene