REVIEW ARTICLE

Integrating signals between cAMP and the RAS/RAF/MEK/ERK signalling pathways ¨ r Biochemie Based on The Anniversary Prize of the Gesellschaft fu und Molekularbiologie Lecture delivered on 5 July 2003 at the Special FEBS Meeting in Brussels Nicolas Dumaz and Richard Marais Signal Transduction Team, Cancer Research UK Centre for Cell and Molecular Biology, The Institute of Cancer Research, London, UK

Keywords cAMP; C-RAF; B-RAF; ERK; PKA; crosstalk Correspondence R. Marais, Signal Transduction Laboratory, Cancer Research UK Centre for Cell and Molecular Biology, Institute of Cancer Research, 237 Fulham Road, London SW3 6JB, UK

One of the hallmarks of cAMP is its ability to inhibit proliferation in many cell types, but stimulate proliferation in others. Clearly cAMP has cell type specific effects and the outcome on proliferation is largely attributed to crosstalk from cAMP to the RAS ⁄ RAF ⁄ mitogen-activated protein kinase (MAPK) and extracellular signal-regulated kinase (ERK) kinase (MEK)/ ERK pathway. We review the crosstalk between these two ancient and conserved pathways, describing the molecular mechanisms underlying the interactions between these pathways and discussing their possible biological importance.

(Received 23 March 2005, revised 28 April 2005, accepted 12 May 2005) doi:10.1111/j.1742-4658.2005.04763.x

Introduction The mechanism by which cells monitor their environment has been the subject of intense scrutiny, particularly over the latter half of the last century. Cells are constantly exposed to hormones, growth factors, cytokines, physical interactions with neighbouring cells, and also to chemical and physical stresses. Many of these signals activate membrane bound receptors, which then activate signalling cascades that stimulate metabolic, genetic and physical changes, allowing cells to mount specific responses to changing environmental conditions. Signalling through these pathways can alter cell fate decisions, so it must be carefully controlled to maintain normal homeostasis as disrupted signalling can alter cell behaviour, and is the underlying cause of diseases from rheumatoid arthritis to cancer and diabetes.

Both the duration and intensity of signalling are important, so intense short-term signals can produce different responses to weaker longer-lived signals even though the total amount of signalling may appear to be the same. Furthermore, many pathways are subject to feedback mechanisms that can either amplify or suppress their own signalling and there is considerable signalling from one pathway to another, a phenomenon known as ‘crosstalk’. Consequently, the responses that cells mount to specific environmental conditions depends on the sum of the intensity and duration of signals from several pathways and how they interact with each other. Here we review the crosstalk between the cyclic adenosine monophosphate (cAMP) and RAS ⁄ RAF ⁄ mitogen-activated protein kinase (MAPK) and extracellular signal-regulated kinase (ERK) kinase (MEK)/ERK pathways, two

Abbreviations AKAP, A-kinase anchoring protein; CRD, cysteine-rich domain; CREB, cAMP-activated transcription factor; EGF, epidermal growth factor; Epac, exchange proteins directly activated by cAMP; a-MSH, a-melanocyte stimulating hormone; NGF, nerve growth factor; PDE4, phosphodiesterase-4; PKA, protein kinase A; RBD, RAS binding domain; TSH, thyroid stimulating hormone.

FEBS Journal 272 (2005) 3491–3504 ª 2005 FEBS

3491

Crosstalk from cAMP to RAS/RAF/MEK/ERK

ancient and conserved signalling pathways that each regulates a plethora of cellular functions. We describe the underlying molecular mechanisms and discuss the biological consequences of this crosstalk.

The cAMP signalling pathway The cyclic nucleotide cAMP is one of the oldest signalling molecules known. It is produced from ATP by adenylyl cyclases and is degraded to AMP by phosphodiesterases (Fig. 1). Most adenylyl cyclases are membrane bound and are activated by heterotrimeric G-protein coupled receptors [1]. Phosphodiesterases are often targeted to membranes and signalling scaffold proteins and are subject to complex regulation [2]. Protein kinase A (PKA), a serine ⁄ threonine specific protein kinase that consists of two catalytic subunits (C) and two regulatory subunits (R), is the most well known cAMP effector. Binding of cAMP to the R subunits releases the active C subunits, which are then free to phosphorylate substrates. In mammals, genes for three catalytic (Ca, Cb and Cc) and four regulatory subunits (RIa, RIb, RIIa and RIIb) have been identified. The ratio of type I (RIa and RIb) to type II (RIIa and RIIb) holoenzyme varies in different cell types and stages of development, suggesting nonredundant functions for each holoenzyme [3]. The R subunits exhibit different affinity for cAMP and are localized to distinct subcellular compartments. The RI holoenzymes are predominantly cytoplasmic whereas the RII holoenzymes associate with cell structures and organelles, largely due to anchoring by a family of specialized scaffold proteins

N. Dumaz and R. Marais

called the A-kinase anchoring proteins, or AKAPs [4]. AKAPs form signalling complexes that localize the type II isoforms within specific subcellular compartments, providing important spatial and temporal regulation of cAMP signalling. AKAPs also bind to other proteins involved in cAMP signalling, such as phosphodiesterases and PKA substrates, forming signalling complexes that finely tune cAMP responses. Thus for the type II holoenzyme, it is thought that localized production of cAMP releases the C-subunits, which then phosphorylate the substrates bound to the AKAP, while phosphodiesterases are on hand to terminate the signal. The exchange proteins directly activated by cAMP, Epacs, are another family of cAMP effectors. These cAMP activated exchange factors catalyse the activation of the Rap family of small G-proteins [5]. There are four Rap isoforms in mammals and they play a key role in cell attachment and migration. Another set of cAMP effectors are the cAMP-gated membrane ion channels, which regulate the influx of cations into the cytosol in response to cAMP [6], a process that regulates the activity of a large number of cellular proteins and functions. Finally, cAMP also regulates some of the phosphodiesterases, providing feedback mechanisms that regulate the duration and intensity of cAMP signalling. This feedback is achieved through phosphorylation by PKA of the commonly expressed phosphodiesterase-4 (PDE4) isoforms. This phosphorylation event at a single serine residue in the regulatory regions of PDE4 activates the enzyme [7]. In cells, cAMP controls metabolism, the cytoskeleton and gene expression, and thereby regulates cell fate

Fig. 1. The cAMP and the RAS ⁄ RAF ⁄ MEK ⁄ ERK signalling pathways. The cAMP pathway is activated when hormones bind to receptors (GPCR) coupled to heterotrimeric G proteins and lead to the activation of adenylyl cyclase, which converts ATP into cAMP. The second messenger cAMP acts through many effectors (see text for details) and has many cellular effects. The RAS ⁄ RAF ⁄ MEK ⁄ ERK pathway is activated by growth factors binding to receptor tyrosine kinase (RTK), which leads to the activation of the small G-protein RAS. Subsequently, RAF, MEK and ERK are activated in a cascade of phosphorylation events. Through the phosphorylation of many targets, ERK regulates cell fate. These two pathways interact through the regulation of the RAF family of kinases by cAMP.

3492

FEBS Journal 272 (2005) 3491–3504 ª 2005 FEBS

N. Dumaz and R. Marais

and also more specialized cellular functions, such as muscle contraction and memory. Many of these effects are attributed to phosphorylation of specific substrates by PKA, but this is a very complex pathway. There are at least nine membrane associated and one cytosolic adenylyl cyclases, eight phosphodiesterase families that hydrolyse cAMP (generating over 50 protein isoforms), three PKA catalytic and four regulatory subunits, over 30 highly diverse AKAPs, two Epacs and up to six cAMP-gated channels [1,2,4–6]. Furthermore, the consequences of biological signalling clearly depend on cell type and specific contexts. Suffice it to say that the cAMP pathway is extremely complex and our understanding of the biological principles underlying its regulation and biological output is far from complete.

The RAS/RAF/MEK/ERK signalling pathway The RAS ⁄ RAF ⁄ MEK ⁄ ERK pathway is another ancient and conserved signalling pathway. RAS proteins are small G-proteins that are embedded on the inner surface of the plasma membrane [8,9]. They are activated downstream of a variety of transmembrane receptors, a process that involves the exchange of GTP for GDP. RAS proteins represent an important signalling branchpoint because they activate several signalling pathways through a number of effectors. One family of effectors are the protein kinases of the RAF family. These cytosolic protein kinases are the first components of a three-tiered protein kinase cascade, which includes two other cytosolic protein kinases, MEK and ERK (Fig. 1). ERK phosphorylates many substrates, thereby regulating numerous cellular

Crosstalk from cAMP to RAS/RAF/MEK/ERK

functions, such as gene expression, metabolism and morphology. Consequently ERK signalling plays an important role in determining cell fate, choosing between diverse responses such as proliferation, differentiation, senescence or survival and it also regulates specialist functions such as those in neurones and immune cells [10]. Both the duration and intensity of ERK activity are important [11]. In neuronal precursors, transient ERK signalling stimulates proliferation whereas sustained signalling induces differentiation. In fibroblasts, ERK signalling is essential for proliferation, but high intensity signals induce cell cycle arrest or senescence [12– 14]. Thus, ERK is a key regulator of cell behaviour and ERK signalling has long been associated with cancer where its signalling is disrupted in approximately 30% of cases [10].

RAF is activated by membrane recruitment and phosphorylation In mammals, there are three RAF genes, A-RAF, B-RAF and C-RAF. All three RAF proteins activate MEK, but with different intensities and the phenotypic differences between A-RAF, B-RAF and C-RAF null mice suggest that the individual family members perform distinct functions in development, possibly in part due to differences in expression patterns [15]. The RAF proteins share a common architecture, with a largely regulatory N-terminal domain that encompasses two conserved regions (CR1 and CR2) and a kinase domain in the C-terminus, which is harboured within a third conserved region (CR3) (Fig. 2). RAF activation occurs at the plasma membrane and involves many steps, including lipid binding, binding to other

Fig. 2. Schematic representation of C-RAF. The three regions conserved in all RAF proteins (CR1, CR2 and CR3) are shown. CR1 encompasses the RAS binding domain (RBD) and the cysteine-rich domain (CRD). CR3 contains the kinase domain, which contains the negative-charge regulatory region (N-region) and activation segment. Phosphorylation of residues marked in blue is required for C-RAF activation whereas phosphorylation of the residues marked in red blocks its activation. Residues marked with * are phosphorylated by PKA and those marked with † are phosphorylated by ERK.

FEBS Journal 272 (2005) 3491–3504 ª 2005 FEBS

3493

Crosstalk from cAMP to RAS/RAF/MEK/ERK

proteins, conformational changes and phosphorylation [9]. Membrane recruitment is stimulated by direct binding to activated RAS which is mediated by the RAS binding domain (RBD) and the cysteine-rich domain (CRD) both of which are in CR1 (Fig. 2). Phosphorylation is essential for activation. The C-RAF isoform requires phosphorylation on four sites within the kinase domain: S338 and Y341, which are within the negative-charge regulatory region, or N-region, and T491 and S494, within the activation segment (Fig. 2). All four sites are conserved in ARAF (S298, Y301, T452 and T455) and T491 and S494 are also conserved in B-RAF (T599 and S602, respectively; note that previously these residues were assigned as T598 and S601 due to an error in the B-RAF sequence deposited on public databases [92]. The correct assignments are T599 and S602. Similarly, S446 and D449 were previously assigned S445 and D448, respectively.) However, although S338 is conserved in B-RAF (S446), it is constitutively phosphorylated and Y341 is replaced with an aspartic acid (D449) [16]. Thus, the N-region of B-RAF is constitutively charged and consequently it has high basal kinase activity and requires fewer events to become fully active than do C-RAF and A-RAF. This is a fundamental difference in the regulation of the RAF isoforms [9].

Inhibition of RAS–ERK signalling by cAMP In the early 1990s it was shown that the cell permeable cAMP analogues 8-chloro-cAMP (8-Cl-cAMP) or dibutyryl cAMP blocked growth factor stimulated ERK activation in Rat1 and NIH3T3 fibroblasts [17–19], establishing the existence of crosstalk between these pathways. Since then it has been shown that cAMP inhibits ERK in several cell types. For example, glucagon can inhibit the mitogenic effect of insulin on adipocytes [20], whereas in smooth muscle cells cAMP can inhibit the stimulation of the ERK pathway by thrombin [21]. As discussed below, cAMP also counteracts some of the biological effects of oncogenes that activate ERK. It is generally agreed that the target of cAMP is C-RAF, but there is some debate as to the precise mechanism by which this occurs. It has been suggested that the small GTPase Rap1 is activated in a PKA dependent manner, sequestering C-RAF and preventing its activation by RAS [22,23]. However, because those studies relied in large part on overexpression of active Rap1, dominant-negative Rap1 (N17Rap) or the Rap1 inactivator, Rap1GAP, they should be treated with some caution, particularly 3494

N. Dumaz and R. Marais

as activation of endogenous Rap1 by extracellular stimuli does not appear to cause inhibition of endogenous C-RAF [24] and it has been shown that N17 Rap is not a true dominant-negative [25]. Furthermore, in an elegant series of experiments, using an Epac selective activator, 8-CPT(MeO)cAMP, that does not activate PKA, it was shown that Rap1 activation did not block ERK activation [26], demonstrating that Rap1 cannot be responsible for blocking C-RAF when cAMP levels are elevated. An alternative model suggested that PKA inhibited C-RAF by direct phosphorylation on S621 [27]. However, this phosphorylation event allows 14-3-3 adaptor ⁄ scaffold proteins to bind to the C-RAF C-terminus, which is essential for activity [28–32]. Furthermore, cAMP does not inhibit the activity of the isolated C-RAF kinase domain, or a version of the kinase domain that is fused to the hormone-binding domain of the oestrogen receptor (RAF:ER) and both of these constructs retain the S621 site [33,34]. It therefore appears that the negative regulation of C-RAF by PKA mediated by S621 phosphorylation occurs in vitro only. The third model that has been suggested also involves direct phosphorylation of C-RAF by PKA. PKA phosphorylates three sites within the C-RAF N-terminal domain (S43, S233 and S259) both in vitro and in vivo [18,33,35,36]. S43 is proximal to the RBD, S233 is between CR1 and CR2 and S259 is within CR2 (Fig. 2). Importantly, these sites work independently to block C-RAF activation [33,36], providing redundant pathways that ensure that C-RAF is inactivated when cAMP levels are elevated. All three sites block C-RAF interaction with RAS. Phosphorylated S43 does so by directly interfering through steric hindrance [18], and S233 and S259 by recruiting 14-3-3 proteins to the N-terminus of C-RAF [32,37]. The mechanism by which 14-3-3 antagonises RAS binding is unclear, but 14-3-3 also binds to the CRD in a manner that is mutually exclusive with RAS [38,39]. Thus 14-3-3 may directly compete with RAS for CRD binding, or it may induce a conformation change that masks or disrupts the RAS binding site on the CRD or on the RBD itself. One of the interesting observations in this area is that ERK phosphorylates the phosphodiesterase PDE4 isoforms. In the case of the long PDE4 isoforms, HSPDE4D3, ERK phosphorylation inactivates the enzyme resulting in an increase in the basal levels of cAMP [40]. A potential consequence of this is that PKA would become activated, leading to inhibition of C-RAF and therefore inhibition of ERK signalling. This would provide an intriguing alternative negative feedback mechanism by which ERK can suppress its FEBS Journal 272 (2005) 3491–3504 ª 2005 FEBS

N. Dumaz and R. Marais

Crosstalk from cAMP to RAS/RAF/MEK/ERK

own signalling. However, it should be noted that whereas long PDE4 isoforms are inhibited by ERK phosphorylation, short PDE4 isoforms are activated by ERK phosphorylation. Therefore, positive or negative coupling between cAMP and ERK pathways at the level of cAMP degradation by PDE is regulated by the ratio between short and long isoforms in cells [41]. The importance of the regulation of PDE by ERK has been highlighted in neuronal processing and in learning and memory processes [42,43].

and the C subunit is activated when RAS induces recruitment of the B subunit to this complex [47]. Although C-RAF can still bind to RAS when S259 is phosphorylated [32], its dephosphorylation is presumably necessary to prevent 14-3-3 from simply rebinding to CR2 and disrupting RAS binding and trapping C-RAF in an inactive conformation. When mutations disrupt the binding of 14-3-3 to CR2, C-RAF basal kinase activity increases and the RAS-related small G-proteins TC21 and R-RAS can activate C-RAF [32,46,48]. Thus S259 phosphorylation and 14-3-3 binding to CR2 are essential to maintain signalling fidelity and to suppress C-RAF signalling in resting cells. Indeed, mutation of S259 to prevent its phosphorylation leads to a protein that has increased activity and which stimulates MEK ⁄ ERK signalling in cells [38,46,48,49]. It is therefore somewhat surprising that cAMP can stimulate S259 phosphorylation in some resting cells [35,36] as this suggests that it is not stoichiometrically phosphorylated, which should result in elevated activity and loss of signalling fidelity. However, in our view this may be a cell culture artefact. S259 appears to be more highly phosphorylated in primary cells than in cells that have been cultured for extended periods (N. Dumaz & R. Marais, unpublished observations), and we speculate that in culture,

The role of 14-3-3 in C-RAF regulation by cAMP The 14-3-3 proteins play a complex role in C-RAF regulation, because binding to S621 is essential for activity, whereas binding to S259 is inhibitory. As 143-3 is a dimer that can accommodate two phosphorylated peptides, it has been suggested that simultaneous binding of 14-3-3 to S259 and S621 traps C-RAF in an inactive conformation [14] (Fig. 3). For activation to occur, 14-3-3 binding to CR2 must be disrupted and this is accompanied by S259 dephosphorylation, which is mediated by the heterotrimeric protein phosphatase PP2A [44–47]. It appears that the A subunit and catalytic C subunit of PP2A are bound to C-RAF

Hormone

Growth Factor

GPCR

RTK

Heterotrimeric G-protein Adenylyl Cyclase

plasma membrane

14

RasGDP

-3-

3

RasGDP PP CR1

ATP

P

43

233 -314

(4)

(2)

CR3

C-RAF: open, active 621 3

cAMP

CR2

PP

CR2

P

P

CR1

P

CR3

MEK P

621 259C-RAF closed and

PP2A

ERK

non-activatable PKA

(1)

14

(3)

-3-

259 P

3

14

-3-

3

CR1

CR2

CR3

P

621

C-RAF closed but activatable

P

P CR1

P P P CR2

CR3

P P

C-RAF: open, inactive

Fig. 3. Model of C-RAF regulation by PKA. In resting cells 14-3-3 traps C-RAF in an inactive, closed conformation by binding to phosphorylated S259 and S621 (1). Active RAS displaces 14-3-3 from CR2 and S259 is dephosphorylated by PP2A. C-RAF can now be recruited to the plasma membrane for activation through a process that is phosphorylation dependent (2). Active C-RAF stimulates MEK and ERK activity and ERK then terminates signalling through the pathway by directly phosphorylating C-RAF, locking it into an open, but inactive conformation (3). However, when cAMP levels are elevated, PKA phosphorylates C-RAF on S43 and S233 creating a new 14-3-3 binding site, causing a rearrangement that switches 14-3-3 from the S621 site to the S233 site, but maintaining contact with the S259 site. C-RAF is then locked in a closed conformation that cannot be activated because RAS cannot displace 14-3-3 from the N-region. Together with the steric hindrance caused by S43 phosphorylation, this prevents membrane recruitment and so the C-RAF is locked into a conformation that cannot be activated (4).

FEBS Journal 272 (2005) 3491–3504 ª 2005 FEBS

3495

Crosstalk from cAMP to RAS/RAF/MEK/ERK

cells are under a selective pressure that results in a loss of S259 phosphorylation because it provides a proliferation advantage.

A model of C-RAF regulation by cAMP Based on the results described above, we propose a model of C-RAF regulation by PKA (Fig. 3). We propose that S259 and S621 are fully phosphorylated in resting cells and this may allow 14-3-3 to trap C-RAF in an inactive, closed conformation. In our model, S259 is the major, or ‘gatekeeper’ binding site and S621 is the minor site, a concept developed by M. Yaffee [50]. Under normal conditions, RAS activation would lead to recruitment to the plasma membrane, displacement of 14-3-3 from CR2 and dephosphorylation of S259. C-RAF could now be activated through a process that is phosphorylation dependent. The free binding site of 14-3-3, which is now attached to the C-terminus of C-RAF, is required to couple signals to MEK, possibly through binding to a new set of client proteins. This activation cycle is self-limiting, because there is a feedback loop whereby ERK phosphorylates C-RAF, preventing its binding to RAS and consequently suppressing further signalling through the pathway [51,52] (Fig. 3). When cAMP levels are elevated, PKA phosphorylates C-RAF on S43 and S233. S233 phosphorylation creates a third 14-3-3 binding site in C-RAF within its N-terminus, and this causes a rearrangement that switches one of the 14-3-3 binding sites from the S621 site to the S233 site, but maintains contact with the gatekeeper site. This switching may occur because 14-3-3 has higher affinity for the S233 site than for the S621 site or because its conformation is more favourable. The model is supported by the observation that S233 phosphorylation does not recruit additional 14-3-3 to C-RAF [37] and by the observation that the S233 site conforms to a weak 14-3-3 binding motif, RYpSTP(231–235) [53]. We propose that whereas RAS can displace 14-3-3 from the N-terminus when it is only bound to S259, once it is bound to both S233 and S259 RAS is unable to displace 14-3-3 from the N-terminus. Together with the steric hindrance provided by S43 phosphorylation, this provides a strong barrier to RAS binding, resulting in a complete inhibition of plasma membrane recruitment and C-RAF activation.

Inhibition of proliferation by cAMP In fibroblasts and a number of other cell lines, cAMP antagonizes the proliferative signals stimulated by 3496

N. Dumaz and R. Marais

growth factors and other activators of the ERK pathway. 8-Cl-cAMP inhibits proliferation of K-RAS transformed cells [54] and infusion of 8-Cl-cAMP induces regression of LX-1 lung carcinoma xenografts in athymic mice [55]. Similarly, PKA activation reverses the transformed phenotype of v-RAF and v-Abl transformed cells [56,57] and in adipocytes glucagon inhibits the mitogenic effect of insulin [20]. The inhibition of proliferation has largely been attributed to the ability of cAMP to inhibit ERK signalling, particularly because in many cells ERK signalling is essential for proliferation. However, several studies suggest that ERK is not the only target that mediates the antiproliferative effects of cAMP. In some cells, such as CCL39 fibroblasts, cAMP does not inhibit ERK but only delays its activation, and yet cAMP still blocks the proliferation of these cells [58]. Similarly, in NIH3T3 cells, when all three PKA phosphorylation sites in C-RAF are mutated, C-RAF and ERK signalling are resistant to cAMP but proliferation is still inhibited [36]. Finally, fibroblasts can be made to proliferate in a tamoxifen dependent manner by expression of RAF:ER [34]. Nevertheless, and despite the fact that ERK activation by RAF:ER is insensitive to cAMP, proliferation stimulated by this construct is still inhibited by cAMP. Taken together these data demonstrate that cAMP targets other proteins that control proliferation. Curiously, some of these targets, such as the epidermal growth factor (EGF) receptor and the nonreceptor tyrosine kinase Src [23,59] are upstream of the RAS ⁄ RAF ⁄ MEK ⁄ ERK signalling pathway, suggesting that this pathway can be inhibited at several steps. Many of the downstream transcriptional targets of the ERK pathway are key cell cycle regulatory proteins, such as cyclin A, cyclin D, cyclin dependent kinase 2 (CDK2) and Cdc25A, and the expression of many of these is inhibited when ERK signalling is blocked by cAMP [34,60]. However, expression of these targets in the presence of cAMP by the use of a hormone regulated version of the transcription factor E2F (E2F:ER) does not restore cell cycle progression. Other targets of cAMP are nontranscriptional, such as the ERKinduced degradation of the cell-cycle inhibitory protein p27KIP1, which is blocked by cAMP [34,60,61]. Finally, cAMP clearly targets other pathways that are important for proliferation and which are downstream of RAS, such as the PI3K ⁄ PKB pathway [62–64] allowing activation of FoxO transcription factors, which also contribute to the antiproliferative effect of cAMP [65]. Clearly cAMP blocks the proliferation of many cell types through multiple targets. In the case of C-RAF, multiple independent mechanisms have evolved to FEBS Journal 272 (2005) 3491–3504 ª 2005 FEBS

N. Dumaz and R. Marais

ensure that its activity is inhibited. C-RAF would only become insensitive to cAMP if it acquires three independent point mutations, suggesting that inhibition of C-RAF by cAMP is very important. We do not currently understand why it is so important to inhibit C-RAF when cAMP levels are elevated. However these observations suggest that high cAMP levels are incompatible with C-RAF signalling, possibly because RAF signalling stimulates proliferation, whereas cAMP prepares cells for differentiation, or other fates that are not compatible with proliferation. It is intriguing that growth factor pathways are targeted at several levels and in the different branches that radiate out from RAS. We are also intrigued by the fact that several cell cycle check-points have evolved to ensure that proliferation is blocked when cAMP levels increase. The redundancy may allow flexibility that enables cAMP to inhibit the cell cycle at several points, which presumably enables it to shut down the cell cycle immediately and at any phase, allowing for a rapid response. It is also clear that we have not identified all of the cAMP check-points, and furthermore, as we shall describe below, cAMP does not inhibit proliferation in all cell types; in some cells it actually stimulates proliferation.

cAMP-stimulated activation of ERK in proliferation and differentiation We have described above the mechanisms underlying the inhibition of ERK and proliferation by cAMP. We now discuss the opposite effect, the activation of ERK by cAMP, describing three different cell systems where ERK and cAMP act in cooperation to regulate proliferation or differentiation. PC12 cells The rat pheochromocytoma PC12 cell line is widely used as a model for neuronal cell differentiation. Transient ERK activation in these cells, mediated by EGF is associated with proliferation, whereas sustained ERK activation induced by nerve growth factor (NGF), fibroblast growth factor (FGF) or cAMP is associated with differentiation into sympathetic neurons [66,67]. This is characterized by neurite outgrowth and the induction of specific genes through a combination of ERK signalling and activation of the cAMPactivated transcription factor, CREB [68]. In PC12 and other neuronal cells, NGF and cAMP are thought to stimulate ERK in a B-RAF dependent manner [69], a model that is supported by several observations. B-RAF is the major RAF isoform and FEBS Journal 272 (2005) 3491–3504 ª 2005 FEBS

Crosstalk from cAMP to RAS/RAF/MEK/ERK

MEK activator in brain extracts [69–74]. B-RAF has been shown to be an essential survival gene in cultured primary sensory and motor neurons [75]. The PKA motifs equivalent to S43 and S233 of C-RAF are not conserved in B-RAF, and indeed B-RAF activity is resistant to inhibition by cAMP in PC12 cells in the presence of serum [76]. Finally, the introduction of wild type B-RAF into B-RAF-deficient Rat-1 cells renders them resistant to the inhibitory effect of cAMP on both growth factor induced ERK activation and mitogenesis [76]. The mechanism by which B-RAF is activated by cAMP has also been the matter of some controversy, with both PKA dependent and PKA independent mechanisms having been proposed. A number of studies have suggested that the small G-protein Rap1 mediates B-RAF activation [77–80]. In vitro Rap1 binds to both C-RAF and B-RAF in a GTP dependent manner and it has been reported to activate B-RAF both in vitro and in vivo but to be incapable of activating C-RAF in vivo [81]. Moreover, many hormones and agonists that elevate intracellular cAMP activate Rap1 [5,82]. These observations have led to a model in which the RAF isoforms are proposed to perform different functions in PC12 cells under different culture conditions [82]. In response to NGF, cAMP activates Rap1 via the Epacs in a PKA independent manner and Rap1 then stimulates sustained B-RAF and ERK signalling, leading to PC12 cell differentiation. By contrast, in EGF stimulated cells, RAS-mediated C-RAF activation stimulates transient ERK signalling and consequently proliferation. Despite the appealing simplicity of this model, several studies have challenged it. Importantly, Rap1 activation by cAMP or the Epac-specific analogue 8-CPT(MeO)cAMP does not activate B-RAF in any cell [24,26]. Furthermore, expression of B-RAF and Rap1 within the same cell does not inevitably lead to ERK activation by cAMP [83]. Finally, we have been unable to repeat the direct activation of B-RAF by Rap1 in vivo (R. Marais, unpublished results). In our view, Rap1 does not directly activate B-RAF, although we cannot discount the possibility that Rap1 is necessary in some cells where its effects may be indirect. For example, 14-3-3 binds more efficiently to B-RAF in cells where cAMP activates ERK than in cells where cAMP inhibits ERK [83]. Because 14-3-3 binding to the catalytic domain of B-RAF is essential for PC12 cell differentiation [84], if Rap1 affected 14-3-3 binding to B-RAF it would indirectly affect B-RAF activation. This uncertainty leaves open the question of how B-RAF is activated downstream of NGF and cAMP 3497

Crosstalk from cAMP to RAS/RAF/MEK/ERK

in PC12 cells. One study suggests a role for the Src tyrosine kinase family and the protein phosphatase PP1. PP1 inhibits Src family kinases and blunts forskolin induced ERK activation, suggesting that Src may be required for ERK activation [85]. However, in fibroblasts Src is required for C-RAF and not B-RAF activation [86,87], suggesting that if Src does play a role, it is mediated by C-RAF and not B-RAF. Importantly, PKA has been shown to phosphorylate B-RAF, but the consequences of this are unclear and it has not been shown to lead to direct B-RAF activation [88–90]. Similarly, PKA can phosphorylate Rap-1 [91], Rap1GAP [92] and Src [23], but again the physiological relevance of these events is unclear. Finally, although early studies suggested that RAS was not required for cAMP stimulated ERK activation, it is now clear that, particularly in some neuronal cells, RAS is essential [26,93]. On balance, the data suggest that RAS is important and that the simple model of B-RAF activation by Rap1 is incorrect. A convincing model of how cAMP activates B-RAF is not currently available, but the accumulated data suggest that the mechanism is complex and involves a number of cellular components (Fig. 4). The effects of cAMP may also depend on cell type and importantly on their context. In the absence of serum, both C-RAF and B-RAF are inhibited by PKA in PC12 cells [76,88,94] suggesting that how the individual isoforms respond may depend on the growth conditions of the cells.

Fig. 4. Crosstalk between the cAMP and the RAS ⁄ RAF ⁄ MEK ⁄ ERK pathway. In the widely accepted model, the consequences of cAMP activation depends on the responses of C-RAF and B-RAF. The inhibition of ERK by cAMP is linked to C-RAF inhibition, whereas ERK activation is linked to B-RAF. cAMP uncouples C-RAF from RAS through direct phosphorylation of C-RAF by PKA. Our understanding of how cAMP activates ERK is less clear, but B-RAF, RAS, Rap1, Src, PKA and 14-3-3 have all been implicated.

3498

N. Dumaz and R. Marais

Melanocytes/melanoma Melanocytes are specialized pigment cells that are found in the skin. Melanocytes are neuronal in origin and they synthesize the pigment melanin in response to physical stimuli such as ultraviolet (UV) radiation or to hormones such as a-melanocyte stimulating hormone (a-MSH). Stimulation of the melanocortin 1 receptor by a-MSH stimulates cAMP production (as do agents such as forskolin and cholera toxin) and this induces melanocyte differentiation in vitro and in vivo [95,96]. In the skin, melanocytes are thought to divide very rarely, but they can be induced to proliferate in culture if provided with a complex set of signals including those from receptor and nonreceptor tyrosine kinases, PKC and – unexpectedly – the cAMP pathway. Indeed, a-MSH can stimulate melanocyte proliferation in vitro, but only when combined with signals from receptor tyrosine kinases or PKC [97]. Presumably proliferation is carefully balanced by an opposing signal from cAMP instructing the cells to differentiate. The relative strength of these signals and how they interact determines whether cells stop cycling and differentiate or whether they continue to proliferate. By providing the correct balance, it is possible to culture proliferating melanocytes that retain many of the characteristics of the differentiated melanocytes found in the skin. Surprisingly, the ERK pathway may mediate the effects of cAMP on proliferation. a-MSH and cAMP elevating agents induce ERK activation in both B16 mouse melanoma cells and in normal human melanocytes [98,99]. Furthermore, as melanocytes and PC12 cells are both neuronal in origin, it has been suggested that these cells share similar signalling pathways. Accordingly cAMP activates RAS, Rap1 and B-RAF in B16 cells, but only RAS and B-RAF mediate the cAMP effects on ERK [99]. The importance of B-RAF to melanocyte proliferation was underscored by the discovery that this gene is mutated in 30–70% of human melanomas [100]. However, the mechanism by which cAMP activates RAS in B16 melanoma and normal human melanocytes is not known, but it does not involve PKA, Epac or the classical RAS activator, son-of-sevenless (SOS) [99]. Recently a RAS specific activator called CNrasGEF (also known as PDZGEF1) was identified that is activated by cAMP in a PKA independent manner [101], providing a potential PKA independent route by which cAMP could activate RAS and B-RAF. However, the responsiveness of this exchange factor to cyclic nucleotides has recently been challenged [102]. Clearly, and in common with neuronal cells, many factors seem to be involved in FEBS Journal 272 (2005) 3491–3504 ª 2005 FEBS

N. Dumaz and R. Marais

ERK activation by cAMP in melanocytes (Fig. 4) and our understanding of this pathway is far from complete. Thyroid cells Thyroid stimulating hormone (TSH) and IGF-I ⁄ insulin regulate thyroid cell proliferation. IGF-I ⁄ insulin alone have little mitogenic activity, but they cooperate with TSH to stimulate proliferation of cell lines and primary cells [103]. TSH induces cAMP production through G-protein coupled receptors (GPCR) and adenylyl cyclase, and its mitogenic activity, alone or in the presence of IGF-I ⁄ insulin, is PKA dependent and is mimicked by various cAMP enhancers [103]. However, PKA activation alone is not sufficient to stimulate thyroid cell proliferation [104]. The RAS ⁄ RAF ⁄ MEK ⁄ ERK pathway has also been implicated in stimulating proliferation. Activating mutations in RAS, B-RAF and the receptor tyrosine kinase RET are found in approximately 60–70% of follicular thyroid adenomas and carcinomas [105– 109] and they appear to occur early in the disease [110], demonstrating an important role for this pathway in thyroid cell proliferation. Furthermore, RAS is required for cAMP dependent mitogenesis in Fisher (FRTL-5) and Wistar (WRT) rat thyroid cells [111,112], and TSH and cAMP activate ERK in FRTL-5 cells [113]. Finally, the MEK inhibitor PD098059 blocks TSH plus insulin-stimulated DNA synthesis in dog and human thyroid epithelial cells [114]. In FRTL-5 cells, ERK activation by cAMP is independent of PKA and linked to Rap1 activation through Epac. However in other models, TSH or forskolin stimulated Rap1 activation does not activate ERK, but curiously it does potentiate the proliferative effects of growth factors [104,115]. Thus, the role of ERK and the mechanism by which cAMP triggers mitogenesis in thyrocyte proliferation is still unclear, but it is important to note that different results are obtained in different cell systems [103]. Clearly further studies are required to elucidate how cAMP and ERK signalling cooperate to stimulate proliferation of these cells.

Conclusion Both the RAS ⁄ RAF ⁄ MEK ⁄ ERK and cAMP signalling pathways are important regulators of cell fate. Both control the activity of metabolic enzymes, transcription factors and cytoskeletal proteins and in many cells their activities appear to counteract each other. The cAMP pathway clearly suppresses ERK FEBS Journal 272 (2005) 3491–3504 ª 2005 FEBS

Crosstalk from cAMP to RAS/RAF/MEK/ERK

signalling in many cells through its ability to target C-RAF. The underlying mechanism is reasonably well characterized and involves the uncoupling of RAS signalling to C-RAF (Fig. 4). There is clear evidence of a complex between C-RAF and PKA [37], presumably mediated by an AKAP that integrates signalling between these pathways [4]. The identity of the AKAP and the other components within this complex are unknown and it is unclear how it is regulated or how it integrates the signals between these pathways. cAMP appears to be produced in restricted microdomains near the plasma membrane [116], producing ‘mists’ of signalling molecules whose effects are localized rather than acting uniformly throughout the cell [117]. These observations pose the interesting possibility that the subcellular localization of the C-RAF ⁄ PKA complex is important to the correct spatial ⁄ temporal regulation of C-RAF. If cAMP was produced in a highly localized fashion (say through contacts with neighbouring cells) or only within specific cellular structures (perhaps dendrites in the case of neurones) and if this coincided with global stimulation by growth factors, then only the C-RAF in those regions where cAMP levels were high would be inhibited and the remainder would be able to signal. This could provide a versatile mechanism by which the activity of C-RAF within specific regions of the cell could be carefully regulated. In contrast to C-RAF, B-RAF is linked to the stimulation of proliferation by cAMP. Models to explain ERK activation by cAMP are incomplete and in addition to B-RAF suggest the involvement of RAS, Rap1, Src, PKA and 14-3-3 (Fig. 4). It is clear that B-RAF activation is modulated by cell type specific factors and it is therefore conceivable that the net outcome of the cellular effects of cAMP may be dependent upon the dynamic expression of these specific factors as well as their specific subcellular localization. It is also possible that a specific AKAP is involved in regulating this positive crosstalk and that spatial ⁄ temporal controls are important here too. Future studies will be required to determine if the ability of cAMP to activate ERK signalling in some cells or some regions of a particular cell, while inhibiting it in others, is limited by subcellular localization of different RAF isoforms, other specific factors or by compartmentalization of cAMP production. Finally, we have focussed on the interaction between cAMP and RAF isoforms, but it should be noted that there are other points at which these pathways interact. For example, PKA phosphorylation of some G-protein coupled receptors can switch their signalling from Gs stimulation of adenylyl cyclase to Gi coupled activation of ERK [118]. Clearly, the 3499

Crosstalk from cAMP to RAS/RAF/MEK/ERK

regulation of cAMP and RAS ⁄ RAF ⁄ MEK ⁄ ERK signalling is complex at many levels. Future studies will also be required to investigate the biological consequences of integrating RAF and cAMP signalling in different cell types. As we have described, even when C-RAF signalling is restored, cAMP still inhibits proliferation through other cellular targets, but the physiological importance of this crosstalk is not understood. The role played by B-RAF in cAMP stimulated proliferation will also require further investigation. The underlying mechanism will need to be elucidated and it is interesting to note that B-RAF mutations are common in melanoma and thyroid cancers, cells in which cAMP is reported to activate ERK signalling rather than inhibit it. It is possible that the important role that B-RAF plays in the normal biology of these cells and the fact that cAMP cannot suppress B-RAF signalling predisposes these cells to B-RAF mutations. It will be interesting to determine whether cAMP signalling also activates ERK in colorectal cells and ovarian cells, the two other types of cancer in which B-RAF mutations are frequently found.

Acknowledgements Work in our laboratory is funded by The Institute of Cancer Research and Cancer Research UK (ref C107 ⁄ A3096).

References 1 Sunahara RK & Taussig R (2002) Isoforms of mammalian adenylyl cyclase: multiplicities of signaling. Mol Interv 2, 168–184. 2 Houslay MD & Adams DR (2003) PDE4 cAMP phosphodiesterases: modular enzymes that orchestrate signalling cross-talk, desensitization and compartmentalization. Biochem J 370, 1–18. 3 Cho-Chung YS, Nesterova M, Becker KG, Srivastava R, Park YG, Lee YN, Cho YS, Kim MK, Neary C & Cheadle C (2002) Dissecting the circuitry of protein kinase A and cAMP signaling in cancer genesis: antisense, microarray, gene overexpression, and transcription factor decoy. Ann NY Acad Sci 968, 22–36. 4 Wong W & Scott JD (2004) AKAP signalling complexes: focal points in space and time. Nat Rev Mol Cell Biol 5, 959–970. 5 Bos JL, de Rooij J & Reedquist KA (2001) Rap1 signalling: adhering to new models. Nat Rev Mol Cell Biol 2, 369–377. 6 Kaupp UB & Seifert R (2002) Cyclic nucleotide-gated ion channels. Physiol Rev 82, 769–824.

3500

N. Dumaz and R. Marais

7 MacKenzie SJ, Baillie GS, McPhee I, MacKenzie C, Seamons R, McSorley T, Millen J, Beard MB, van Heeke G & Houslay MD (2002) Long PDE4 cAMP specific phosphodiesterases are activated by protein kinase A-mediated phosphorylation of a single serine residue in Upstream Conserved Region 1 (UCR1). Br J Pharmacol 136, 421–433. 8 Robinson MJ & Cobb MH (1997) Mitogen-activated protein kinase pathways. Curr Opin Cell Biol 9, 180– 186. 9 Wellbrock C, Karasarides M & Marais R (2004) The RAF proteins take centre stage. Nat Rev Mol Cell Biol 5, 875–885. 10 Sebolt-Leopold JS & Herrera R (2004) Targeting the mitogen-activated protein kinase cascade to treat cancer. Nat Rev Cancer 4, 937–947. 11 Pouyssegur J & Lenormand P (2003) Fidelity and spatio-temporal control in MAP kinase (ERKs) signalling. Eur J Biochem 270, 3291–3299. 12 Kolch W (2000) Meaningful relationships: the regulation of the Ras ⁄ Raf ⁄ MEK ⁄ ERK pathway by protein interactions. Biochem J 351 Part 2, 289–305. 13 Hagemann C & Rapp UR (1999) Isotype-specific functions of Raf kinases, Exp Cell Res. 253, 34–46. 14 Avruch J, Khokhlatchev A, Kyriakis JM, Luo Z, Tzivion G, Vavvas D & Zhang XF (2001) Ras activation of the Raf kinase: tyrosine kinase recruitment of the MAP kinase cascade. Recent Prog Horm Res 56, 127–155. 15 Mercer KE & Pritchard CA (2003) Raf proteins and cancer: B-Raf is identified as a mutational target. Biochim Biophys Acta 1653, 25–40. 16 Mason CS, Springer CJ, Cooper RG, Superti-Furga G, Marshall CJ & Marais R (1999) Serine and tyrosine phosphorylations cooperate in Raf-1, but not B-Raf activation. EMBO J 18, 2137–2148. 17 Cook SJ & McCormick F (1993) Inhibition by cAMP of Ras-dependent activation of Raf. Science 262, 1069– 1072. 18 Wu J, Dent P, Jelinek T, Wolfman A, Weber MJ & Sturgill TW (1993) Inhibition of the EGF-activated MAP kinase signaling pathway by adenosine 3¢,5¢monophosphate. Science 262, 1065–1069. 19 Burgering BM, Pronk GJ, van Weeren PC, Chardin P & Bos JL (1993) cAMP antagonizes p21ras-directed activation of extracellular signal-regulated kinase 2 and phosphorylation of mSos nucleotide exchange factor. EMBO J 12, 4211–4220. 20 Sevetson BR, Kong X & Lawrence JC Jr (1993) Increasing cAMP attenuates activation of mitogen-activated protein kinase. Proc Natl Acad Sci USA 90, 10305–10309. 21 Osinski MT & Schror K (2000) Inhibition of plateletderived growth factor-induced mitogenesis by

FEBS Journal 272 (2005) 3491–3504 ª 2005 FEBS

N. Dumaz and R. Marais

22

23

24

25

26

27

28

29

30

31

32

33

34

phosphodiesterase 3 inhibitors: role of protein kinase A in vascular smooth muscle cell mitogenesis. Biochem Pharmacol 60, 381–387. Schmitt JM & Stork PJ (2001) Cyclic AMP-mediated inhibition of cell growth requires the small G protein Rap1. Mol Cell Biol 21, 3671–3683. Schmitt JM & Stork PJ (2002) PKA phosphorylation of Src mediates cAMP’s inhibition of cell growth via Rap1. Mol Cell 9, 85–94. Zwartkruis FJ, Wolthuis RM, Nabben NM, Franke B & Bos JL (1998) Extracellular signal-regulated activation of Rap1 fails to interfere in Ras effector signalling. Embo J 17, 5905–5912. van den Berghe N, Cool RH, Horn G & Wittinghofer A (1997) Biochemical characterization of C3G: an exchange factor that discriminates between Rap1 and Rap2 and is not inhibited by Rap1A (S17N). Oncogene 15, 845–850. Enserink JM, Christensen AE, de Rooij J, van Triest M, Schwede F, Genieser HG, Doskeland SO, Blank JL & Bos JL (2002) A novel Epac-specific cAMP analogue demonstrates independent regulation of Rap1 and ERK. Nat Cell Biol 4, 901–906. Mischak H, Seitz T, Janosch P, Eulitz M, Steen H, Schellerer M, Philipp A & Kolch W (1996) Negative regulation of Raf-1 by phosphorylation of serine 621. Mol Cell Biol 16, 5409–5418. Roy S, McPherson RA, Apolloni A, Yan J, Lane A, Clyde-Smith J & Hancock JF (1998) 14–3)3 facilitates Ras-dependent Raf-1 activation in vitro and in vivo. Mol Cell Biol 18, 3947–3955. Thorson JA., YuLW, Hsu AL, Shih NY, Graves PR, Tanner JW, Allen PM, Piwnica-Worms H & Shaw AS (1998) 14–3)3 proteins are required for maintenance of Raf-1 phosphorylation and kinase activity. Mol Cell Biol 18, 5229–5238. Tzivion G, Luo Z & Avruch J (1998) A dimeric 14– 3)3 protein is an essential cofactor for Raf kinase activity. Nature 394, 88–92. Yip-Schneider MT, Miao W, Lin A, Barnard DS, Tzivion G & Marshall MS (2000) Regulation of the Raf-1 kinase domain by phosphorylation and 14–3)3 association. Biochem J 351, 151–159. Light Y, Paterson H & Marais R (2002) 14–3)3 antagonizes Ras-mediated Raf-1 recruitment to the plasma membrane to maintain signaling fidelity. Mol Cell Biol 22, 4984–4996. Sidovar MF, Kozlowski P, Lee JW, Collins MA, He Y & Graves LM (2000) Phosphorylation of serine 43 is not required for inhibition of c-Raf kinase by the cAMP-dependent protein kinase. J Biol Chem 275, 28688–28694. Balmanno K, Millar T, McMahon M & Cook SJ (2003) DeltaRaf-1: ER* bypasses the cyclic AMP block of extracellular signal-regulated kinase 1 and 2

FEBS Journal 272 (2005) 3491–3504 ª 2005 FEBS

Crosstalk from cAMP to RAS/RAF/MEK/ERK

35

36

37

38

39

40

41

42

43

44

45

46

47

activation but not CDK2 activation or cell cycle reentry. Mol Cell Biol 23, 9303–9317. Dhillon AS, Pollock C, Steen H, Shaw PE, Mischak H & Kolch W (2002) Cyclic AMP-dependent kinase regulates Raf-1 kinase mainly by phosphorylation of serine 259. Mol Cell Biol 22, 3237–3246. Dumaz N, Light Y & Marais R (2002) Cyclic AMP blocks cell growth through Raf-1-dependent and Raf-1independent mechanisms. Mol Cell Biol 22, 3717–3728. Dumaz N & Marais R (2003) Protein kinase A blocks Raf-1 activity by stimulating 14-3-3 binding and blocking Raf)1 interaction with Ras. J Biol Chem 278, 29819–29823. Michaud NR, Fabian JR, Mathes KD & Morrison DK (1995) 14–3)3 is not essential for Raf-1 function: identification of Raf-1 proteins that are biologically activated in a 14-3-3- and Ras-independent manner. Mol Cell Biol 15, 3390–3397. Clark GJ, Drugan JK, Rossman KL, Carpenter JW, Rogers-Graham K, Der Fu H, CJ & Campbell SL (1997) 14-3-3 zeta negatively regulates raf)1 activity by interactions with the Raf-1 cysteine-rich domain. J Biol Chem 272, 20990–20993. Hoffmann R, Baillie GS, MacKenzie SJ, Yarwood SJ & Houslay MD (1999) The MAP kinase ERK2 inhibits the cyclic AMP-specific phosphodiesterase HSPDE4D3 by phosphorylating it at Ser579. EMBO J 18, 893–903. Baillie GS, MacKenzie SJ, McPhee I & Houslay MD (2000) Sub-family selective actions in the ability of Erk2 MAP kinase to phosphorylate and regulate the activity of PDE4 cyclic AMP-specific phosphodiesterases. Br J Pharmacol 131, 811–819. Gao Y, Nikulina E, Mellado W & Filbin MT (2003) Neurotrophins elevate cAMP to reach a threshold required to overcome inhibition by MAG through extracellular signal-regulated kinase-dependent inhibition of phosphodiesterase. J Neurosci 23, 11770–11777. Zhang HT, Zhao Y, Huang Y, Dorairaj NR, Chandler LJ & O’Donnell JM (2004) Inhibition of the phosphodiesterase 4 (PDE4) enzyme reverses memory deficits produced by infusion of the MEK inhibitor U0126 into the CA1 subregion of the rat hippocampus. Neuropsychopharmacol 29, 1432–1439. Jaumot M & Hancock JF (2001) Protein phosphatases 1 and 2A promote Raf-1 activation by regulating 14-3-3 interactions. Oncogene 20, 3949–3958. Dhillon AS, Meikle S, Yazici Z, Eulitz M & Kolch W (2002) Regulation of Raf-1 activation and signalling by dephosphorylation. EMBO J 21, 64–71. Kubicek M, Pacher M, Abraham D, Podar K, Eulitz M & Baccarini M (2002) Dephosphorylation of Ser-259 regulates Raf-1 membrane association. J Biol Chem 277, 7913–7919. Ory S, Zhou M, Conrads TP, Veenstra TD & Morrison DK (2003) Protein phosphatase 2A positively

3501

Crosstalk from cAMP to RAS/RAF/MEK/ERK

48

49

50

51

52

53

54

55

56

57

58

regulates Ras signaling by dephosphorylating KSR1 and Raf-1 on critical 14-3-3 binding sites. Curr Biol 13, 1356–1364. Dhillon AS, Meikle S, Peyssonnaux C, Grindlay J, Kaiser C, Steen H, Shaw PE, Mischak H, Eychene A & Kolch W (2003) A Raf-1 mutant that dissociates MEK ⁄ extracellular signal-regulated kinase activation from malignant transformation and differentiation but not proliferation. Mol Cell Biol 23, 1983–1993. Abraham D, Podar K, Pacher M, Kubicek M, Welzel N, Hemmings BA, Dilworth SM, Mischak H, Kolch W & Baccarini M (2000) Raf-1-associated protein phosphatase 2A as a positive regulator of kinase activation. J Biol Chem. 275, 22300–22304. Yaffe MB (2002) How do 14-3-3 proteins work? – Gatekeeper phosphorylation and the molecular anvil hypothesis. FEBS Lett 513, 53–57. Alessi DR, Cuenda A, Cohen P, Dudley DT & Saltiel AR (1995) PD 098059 is a specific inhibitor of the activation of mitogen-activated protein kinase kinase in vitro and in vivo. J Biol Chem 270, 27489–27494. Dougherty MK, Muller J, Ritt DA, Zhou M, Zhou XZ, Copeland TD, Conrads TP, Veenstra TD, Lu KP & Morrison DK (2005) Regulation of Raf-1 by direct feedback phosphorylation. Mol Cell 17, 215– 224. Yaffe MB, Rittinger K, Volinia S, Caron PR, Aitken A, Leffers H, Gamblin SJ, Smerdon SJ & Cantley LC (1997) The structural basis for 14–3)3: phosphopeptide binding specificity. Cell 91, 961–971. Tortora G, Ciardiello F, Ally S, Clair T, Salomon DS & Cho-Chung YS (1989) Site-selective 8-chloroadenosine 3¢,5¢-cyclic monophosphate inhibits transformation and transforming growth factor alpha production in Ki-ras-transformed rat fibroblasts. FEBS Lett 242, 363–367. Ally S, Clair T, Katsaros D, Tortora G, Yokozaki H, Finch RA, Avery TL & Cho-Chung YS (1989) Inhibition of growth and modulation of gene expression in human lung carcinoma in athymic mice by site-selective 8-Cl-cyclic adenosine monophosphate. Cancer Res 49, 5650–5655. Hafner S, Adler HS, Mischak H, Janosch P, Heidecker G, Wolfman A, Pippig S, Lohse M, Ueffing M & Kolch W (1994) Mechanism of inhibition of Raf-1 by protein kinase A. Mol Cell Biol 14, 6696–6703. Weissinger EM, Eissner G, Grammer C, Fackler S, Haefner B, Yoon LS, Lu KS, Bazarov A, Sedivy JM, Mischak H & Kolch W (1997) Inhibition of the Raf-1 kinase by cyclic AMP agonists causes apoptosis of v-abl-transformed cells. Mol Cell Biol 17, 3229–3241. McKenzie FR & Pouyssegur J (1996) cAMP-mediated growth inhibition in fibroblasts is not mediated via mitogen-activated protein (MAP) kinase (ERK) inhibition. cAMP-dependent protein kinase induces a

3502

N. Dumaz and R. Marais

59

60

61

62

63

64

65

66

67

68

69

70

temporal shift in growth factor-stimulated MAP kinases. J Biol Chem 271, 13476–13483. Piiper A, Lutz MP, Cramer H, Elez R, Kronenberger B, Dikic I, Muller-Esterl W & Zeuzem S (2003) Protein kinase A mediates cAMP-induced tyrosine phosphorylation of the epidermal growth factor receptor. Biochem Biophys Res Commun 301, 848–854. L’Allemain G, Lavoie JN, Rivard N, Baldin V & Pouyssegur J (1997) Cyclin D1 expression is a major target of the cAMP-induced inhibition of cell cycle entry in fibroblasts. Oncogene 14, 1981–1990. Kato JY, Matsuoka M, Polyak K, Massague J & Sherr CJ (1994) Cyclic AMP-induced G1 phase arrest mediated by an inhibitor (p27Kip1) of cyclin-dependent kinase 4 activation. Cell 79, 487–496. Kim S, Jee K, Kim D, Koh H & Chung J (2001) Cyclic AMP inhibits Akt activity by blocking the membrane localization of PDK1. J Biol Chem 276, 12864–12870. Lou L, Urbani J, Ribeiro-Neto F & Altschuler DL (2002) cAMP inhibition of Akt is mediated by activated and phosphorylated Rap1b, J Biol Chem. 277, 32799–32806. Mei FC, Qiao J, Tsygankova OM, Meinkoth JL, Quilliam LA & Cheng X (2002) Differential signaling of cyclic AMP: opposing effects of exchange protein directly activated by cyclic AMP and cAMP-dependent protein kinase on protein kinase B activation. J Biol Chem 277, 11497–11504. Kuiperij HB, van der Horst A, Raaijmakers J, Weijzen S, Medema RH, Bos JL, Burgering BM & Zwartkruis FJ (2005) Activation of FoxO transcription factors contributes to the antiproliferative effect of cAMP. Oncogene 24, 2087–2095. Marshall CJ (1995) Specificity of receptor tyrosine kinase signaling: transient versus sustained extracellular signal-regulated kinase activation. Cell 80, 179–185. York RD, Yao H, Dillon T, Ellig CL, Eckert SP, McCleskey EW & Stork PJ (1998) Rap1 mediates sustained MAP kinase activation induced by nerve growth factor. Nature 392, 622–626. Grewal SS, Fass DM, Yao H, Ellig CL, Goodman RH & Stork PJ (2000) Calcium and cAMP signals differentially regulate cAMP-responsive element-binding protein function via a Rap1-extracellular signalregulated kinase pathway. J Biol Chem 275, 34433– 34441. Jaiswal RK, Moodie SA, Wolfman A & Landreth GE (1994) The mitogen-activated protein kinase cascade is activated by B-Raf in response to nerve growth factor through interaction with p21ras. Mol Cell Biol 14, 6944–6953. Troppmair J, Bruder JT, App H, Cai H, Liptak L, Szeberenyi J, Cooper GM & Rapp UR (1992) Ras controls coupling of growth factor receptors and protein

FEBS Journal 272 (2005) 3491–3504 ª 2005 FEBS

N. Dumaz and R. Marais

71

72

73

74

75

76

77

78

79

80

81

82 83

kinase C in the membrane to Raf-1 and B-Raf protein serine kinases in the cytosol. Oncogene 7, 1867–1873. Catling AD, Reuter CW, Cox ME, Parsons SJ & Weber MJ (1994) Partial purification of a mitogen-activated protein kinase kinase activator from bovine brain. Identification as B-Raf or a B-Raf-associated activity. J Biol Chem 269, 30014–30021. Lange-Carter CA & Johnson GL (1994) Ras-dependent growth factor regulation of MEK kinase in PC12 cells. Science 265, 1458–1461. Traverse S & Cohen P (1994) Identification of a latent MAP kinase kinase kinase in PC12 cells as B-raf. FEBS Lett 350, 13–18. Morice C, Nothias F, Konig S, Vernier P, Baccarini M, Vincent JD & Barnier JV (1999) Raf-1 and B-Raf proteins have similar regional distributions but differential subcellular localization in adult rat brain. Eur J Neurosci 11, 1995–2006. Wiese S, Pei G, Karch C, Troppmair J, Holtmann B, Rapp UR & Sendtner M (2001) Specific function of B-Raf in mediating survival of embryonic motoneurons and sensory neurons. Nat Neurosci 4, 137–142. Erhardt P, Troppmair J, Rapp UR & Cooper GM (1995) Differential regulation of Raf-1 and B-Raf and Ras-dependent activation of mitogen-activated protein kinase by cyclic AMP in PC12 cells. Mol Cell Biol 15, 5524–5530. Vossler MR, Yao H, York RD, Pan MG, Rim CS & Stork PJ (1997) cAMP activates MAP kinase and Elk1 through a B-Raf- and Rap1-dependent pathway. Cell 89, 73–82. Grewal SS, Horgan AM, York RD, Withers GS, Banker GA & Stork PJ (2000) Neuronal calcium activates a Rap1 and B-Raf signaling pathway via the cyclic adenosine monophosphate-dependent protein kinase. J Biol Chem 275, 3722–3728. Guo FF, Kumahara E & Saffen D (2001) A CalDAGGEFI ⁄ Rap1 ⁄ B-Raf cassette couples M (1) muscarinic acetylcholine receptors to the activation of ERK1 ⁄ 2. J Biol Chem 276, 25568–25581. Kao S, Jaiswal RK, Kolch W & Landreth GE (2001) Identification of the mechanisms regulating the differential activation of the mapk cascade by epidermal growth factor and nerve growth factor in PC12 cells. J Biol Chem 276, 18169–18177. Ohtsuka T, Shimizu K, Yamamori B, Kuroda S & Takai Y (1996) Activation of brain B-Raf protein kinase by Rap1B small GTP-binding protein. J Biol Chem 271, 1258–1261. Stork PJ (2003) Does Rap1 deserve a bad Rap? Trends Biochem Sci 28, 267–275. Qiu W, Zhuang S, von Lintig FC, Boss GR & Pilz RB (2000) Cell type-specific regulation of B-Raf kinase by cAMP and 14-3-3 proteins. J Biol Chem 275, 31921– 31929.

FEBS Journal 272 (2005) 3491–3504 ª 2005 FEBS

Crosstalk from cAMP to RAS/RAF/MEK/ERK

84 MacNicol MC, Muslin AJ & MacNicol AM (2000) Disruption of the 14-3-3 binding site within the B-Raf kinase domain uncouples catalytic activity from PC12 cell differentiation, J Biol Chem. 275, 3803–3809. 85 Klinger M, Kudlacek O, Seidel MG, Freissmuth M & Sexl V (2002) MAP kinase stimulation by cAMP does not require RAP1 but SRC family kinases, J Biol Chem. 277, 32490–32497. 86 Marais R, Light Y, Paterson HF & Marshall CJ (1995) Ras recruits Raf-1 to the plasma membrane for activation by tyrosine phosphorylation. Embo J 14, 3136– 3145. 87 Marais R, Light Y, Paterson HF, Mason CS & Marshall CJ (1997) Differential regulation of Raf-1, A-Raf, and B-Raf by oncogenic ras and tyrosine kinases. J Biol Chem 272, 4378–4383. 88 Peraldi P, Frodin M, Barnier JV, Calleja V, Scimeca JC, Filloux C, Calothy G & Van Obberghen E (1995) Regulation of the MAP kinase cascade in PC12 cells: B-Raf activates MEK-1 (MAP kinase or ERK kinase) and is inhibited by cAMP. FEBS Lett 357, 290–296. 89 MacNicol MC & MacNicol AM (1999) Nerve growth factor-stimulated B-Raf catalytic activity is refractory to inhibition by cAMP-dependent protein kinase. J Biol Chem 274, 13193–13197. 90 Konig S, Guibert B, Morice C, Vernier P & Barnier JV (2001) Phosphorylation by PKA of a site unique to B-Raf kinase. C R Acad Sci III (324), 673–681. 91 Lerosey I, Pizon V, Tavitian A & de Gunzburg J (1991) The cAMP-dependent protein kinase phosphorylates the rap1 protein in vitro as well as in intact fibroblasts, but not the closely related rap2 protein. Biochem Biophys Res Commun 175, 430–436. 92 Polakis P, Rubinfeld B & McCormick F (1992) Phosphorylation of rap1GAP in vivo and by cAMP-dependent kinase and the cell cycle p34cdc2 kinase in vitro. J Biol Chem 267, 10780–10785. 93 Ambrosini A, Tininini S, Barassi A, Racagni G, Sturani E & Zippel R (2000) cAMP cascade leads to Ras activation in cortical neurons. Brain Res Mol Brain Res 75, 54–60. 94 Vaillancourt RR, Gardner AM & Johnson GL (1994) B-Raf-dependent regulation of the MEK-1 ⁄ mitogenactivated protein kinase pathway in PC12 cells and regulation by cyclic AMP. Mol Cell Biol 14, 6522– 6530. 95 Levine N, Sheftel SN, Eytan T, Dorr RT, Hadley ME, Weinrach JC, Ertl GA, Toth K, McGee DL & Hruby VJ (1991) Induction of skin tanning by subcutaneous administration of a potent synthetic melanotropin. J Am Med Assoc 266, 2730–2736. 96 Hunt G, Donatien PD, Lunec J, Todd C, Kyne S & Thody AJ (1994) Cultured human melanocytes respond

3503

Crosstalk from cAMP to RAS/RAF/MEK/ERK

97

98

99

100 101

102

103

104

105

106

107

to MSH peptides and ACTH. Pigment Cell Res 7, 217– 221. De Luca M, Siegrist W, Bondanza S, Mathor M, Cancedda R & Eberle AN (1993) Alpha melanocyte stimulating hormone (alpha MSH) stimulates normal human melanocyte growth by binding to high-affinity receptors. J Cell Sci 105, 1079–1084. Englaro W, Rezzonico R, Durand-Clement M, Lallemand D, Ortonne JP & Ballotti R (1995) Mitogen-activated protein kinase pathway and AP-1 are activated during cAMP-induced melanogenesis in B-16 melanoma cells. J Biol Chem 270, 24315–24320. Busca R, Abbe P, Mantoux F, Aberdam E, Peyssonnaux C, Eychene A, Ortonne JP & Ballotti R (2000) Ras mediates the cAMP-dependent activation of extracellular signal-regulated kinases (ERKs) in melanocytes. EMBO J 19, 2900–2910. Garnett MJ & Marais R (2004) Guilty as charged; B-RAF is a human oncogene. Cancer Cell 6, 313–319. Pham N, Cheglakov I, Koch CA, de Hoog CL, Moran MF & Rotin D (2000) The guanine nucleotide exchange factor CNrasGEF activates ras in response to cAMP and cGMP. Curr Biol 10, 555–558. Quilliam LA, Rebhun JF & Castro AF (2002) A growing family of guanine nucleotide exchange factors is responsible for activation of Ras-family GTPases. Prog Nucleic Acid Res Mol Biol 71, 391–444. Kimura T, Van Keymeulen A, Golstein J, Fusco A, Dumont JE & Roger PP (2001) Regulation of thyroid cell proliferation by TSH and other factors: a critical evaluation of in vitro models. Endocr Rev 22, 631–656. Dremier S, Vandeput F, Zwartkruis FJ, Bos JL, Dumont JE & Maenhaut C (2000) Activation of the small G protein Rap1 in dog thyroid cells by both cAMP-dependent and -independent pathways. Biochem Biophys Res Commun 267, 7–11. Fukushima T, Suzuki S, Mashiko M, Ohtake T, Endo Y, Takebayashi Y, Sekikawa K, Hagiwara K & Takenoshita S (2003) BRAF mutations in papillary carcinomas of the thyroid. Oncogene 22, 6455–6457. Kimura ET, Nikiforova MN, Zhu Z, Knauf JA, Nikiforov YE & Fagin JA (2003) High prevalence of BRAF mutations in thyroid cancer: genetic evidence for constitutive activation of the RET ⁄ PTC-RASBRAF signaling pathway in papillary thyroid carcinoma. Cancer Res 63, 1454–1457. Soares P, Trovisco V, Rocha AS, Lima J, Castro P, Preto A, Maximo V, Botelho T, Seruca R & SobrinhoSimoes M (2003) BRAF mutations and RET ⁄ PTC

3504

N. Dumaz and R. Marais

108

109

110

111

112

113

114

115

116

117 118

rearrangements are alternative events in the etiopathogenesis of PTC. Oncogene 22, 4578–4580. Xu X, Quiros RM, Gattuso P, Ain KB & Prinz RA (2003) High prevalence of BRAF gene mutation in papillary thyroid carcinomas and thyroid tumor cell lines. Cancer Res 63, 4561–4567. Williams SF & Smallridge RC (2004) Targeting the ERK pathway: novel therapeutics for thyroid cancer. Curr Drug Targets Immune Endocr Metabol Disord 4, 199–220. Lemoine NR, Mayall ES, Wyllie FS, Williams ED, Goyns M, Stringer B & Wynford-Thomas D (1989) High frequency of ras oncogene activation in all stages of human thyroid tumorigenesis. Oncogene 4, 159–164. Miller MJ, Rioux L, Prendergast GV, Cannon S, White MA & Meinkoth JL (1998) Differential effects of protein kinase A on Ras effector pathways. Mol Cell Biol 18, 3718–3726. Ciullo I, Diez-Roux G, Di Domenico M, Migliaccio A & Avvedimento EV (2001) cAMP signaling selectively influences Ras effectors pathways. Oncogene 20, 1186– 1192. Iacovelli L, Capobianco L, Salvatore L, Sallese M, D’Ancona GM & De Blasi A (2001) Thyrotropin activates mitogen-activated protein kinase pathway in FRTL-5 by a cAMP-dependent protein kinase A-independent mechanism. Mol Pharmacol 60, 924–933. Vandeput F, Perpete S, Coulonval K, Lamy F & Dumont JE (2003) Role of the different mitogen-activated protein kinase subfamilies in the stimulation of dog and human thyroid epithelial cell proliferation by cyclic adenosine 5¢-monophosphate and growth factors. Endocrinology 144, 1341–1349. Saavedra AP, Tsygankova OM, Prendergast GV, Dworet JH, Cheng G & Meinkoth JL (2002) Role of cAMP, PKA and Rap1A in thyroid follicular cell survival. Oncogene 21, 778–788. Rich TC, Fagan KA, Tse TE, Schaack J, Cooper DM & Karpen JW (2001) A uniform extracellular stimulus triggers distinct cAMP signals in different compartments of a simple cell. Proc Natl Acad Sci USA 98, 13049–13054. Schwartz JH (2001) The many dimensions of cAMP signaling. Proc Natl Acad Sci USA 98, 13482–13484. Baillie GS, Sood A, McPhee I, Gall I, Perry SJ, Lefkowitz RJ & Houslay MD (2003) beta-Arrestin-mediated PDE4 cAMP phosphodiesterase recruitment regulates beta-adrenoceptor switching from Gs to Gi. Proc Natl Acad Sci USA 100, 940–945.

FEBS Journal 272 (2005) 3491–3504 ª 2005 FEBS