NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Lodish H, Berk A, Zipursky SL, et al. Molecular Cell Biology. 4th edition. New York: W. H. Freeman; 2000.

  • By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.
Cover of Molecular Cell Biology

Molecular Cell Biology. 4th edition.

Show details

Section 20.5MAP Kinase Pathways

All Ras-linked RTKs in mammalian cells appear to utilize a highly conserved signal-transduction pathway in which the signal induced by ligand binding is carried via GRB2 and Sos to Ras, leading to its activation (see Figure 20-23). Activated Ras then induces a kinase cascade that culminates in activation of MAP kinase. This serine/threonine kinase, which can translocate into the nucleus, phosphorylates many different proteins including transcription factors that regulate expression of important cell-cycle and differentiation-specific proteins. In this section, we first examine the components of the kinase cascade downstream from Ras in RTK-Ras signaling pathways in mammalian cells. Then we discuss the linkage of other signaling pathways to similar kinase cascades and recent studies indicating that both yeasts and cells of higher eukaryotes contain multiple MAP kinases.

Activation of MAP kinase in two different cells can lead to similar or different cellular responses, as can activation in the same cell by stimulation of different RTKs. The mechanisms controlling the response specificity of MAP kinases are poorly understood and are not considered in this chapter.

Signals Pass from Activated Ras to a Cascade of Protein Kinases

A remarkable convergence of biochemical and genetic studies in yeast, C. elegans, Drosophila, and mammals has revealed a highly conserved cascade of protein kinases that operate in sequential fashion downstream from activated Ras as follows (Figure 20-28):


Activated Ras binds to the N-terminal domain of Raf, a serine/threonine kinase.


Raf binds to and phosphorylates MEK, a dual-specificity protein kinase that phosphorylates both tyrosine and serine residues.


MEK phosphorylates and activates MAP kinase, another serine/threonine kinase.


MAP kinase phosphorylates many different proteins, including nuclear transcription factors, that mediate cellular responses.

Figure 20-28. Kinase cascade that transmits signals downstream from activated Ras protein.

Figure 20-28

Kinase cascade that transmits signals downstream from activated Ras protein. In unstimulated cells, most Ras is in the inactive form with bound GDP (top); binding of a growth factor to its RTK leads to formation of the active Ras · GTP (see Figure (more...)

Several types of experiments have demonstrated that Raf, MEK, and MAP kinase lie downstream of Ras and their sequential order in the pathway. For example, constitutively active mutant Raf proteins induce quiescent cultured cells to proliferate in the absence of hormone stimulation. These mutant Raf proteins, which initially were identified in tumor cells, are encoded by oncogenes and stimulate uncontrolled cell proliferation. Conversely, cultured mammalian cells that express a mutant, defective Raf protein cannot be stimulated to proliferate uncontrollably by a mutant, constitutively active RasD protein. This finding establishes a link between the Raf and Ras proteins. In vitro binding studies have shown that purified Ras · GTP protein binds directly to Raf. An interaction between the mammalian Ras and Raf proteins also has been demonstrated in the yeast two-hybrid system, a genetic system in yeast used to select cDNAs encoding proteins that bind to target, or “bait” proteins (Figure 20-29). The binding of Ras and Raf to each other does not induce the Raf kinase activity.

Figure 20-29. Yeast two-hybrid system for detecting proteins that interact.

Figure 20-29

Yeast two-hybrid system for detecting proteins that interact. (a) Recombinant DNA techniques can be used to prepare genes that encode hybrid (chimeric) proteins consisting of the DNA-binding domain (purple) or activation domain (orange) of a transcription (more...)

The location of MAP kinase downstream of Ras was evidenced by the finding that in quiescent cultured cells expressing a constitutively active RasD, activated MAP kinase is generated in the absence of hormone stimulation. More importantly, in Drosophila mutants that lack a functional Ras or Raf but express a constitutively active MAP kinase specifically in the developing eye, R7 photoreceptors were found to develop normally. This finding indicates that activation of MAP kinase is sufficient to transmit a proliferation or differentiation signal normally initiated by ligand binding to an RTK. Biochemical studies showed that Raf does not activate MAP kinase directly. The signaling pathway thus appears to be a linear one: activated RTK → Ras → Raf → (?) → MAP kinase.

Finally, fractionation of cultured cells that had been stimulated with growth factors led to identification of MEK, a kinase that specifically phosphorylates threonine and tyrosine residues on MAP kinase, thereby activating its catalytic activity. (The acronym MEK comes from MAP and ERK kinase, where ERK is another acronym for MAP.) Later studies showed that MEK binds to the C-terminal catalytic domain of Raf and is phosphorylated by the Raf serine/ threonine kinase activity; this phosphorylation activates the catalytic activity of MEK. Hence, activation of Ras induces a kinase cascade that includes Raf, MEK, and MAP kinase.

Ksr May Function as a Scaffold for the MAP Kinase Cascade Linked to Ras

Two additional proteins not depicted in Figure 20-28 participate in the MAP kinase cascade downstream from Ras. Although the precise functions of these proteins, called 14-3-3 and Ksr, are not yet known, they both appear to play important roles in forming protein complexes necessary for signaling from Raf to MAP kinase.

In a resting cell prior to stimulation, Raf is present in the cytosol in an inactive conformation stabilized by a dimer of 14-3-3. Each 14-3-3 monomer binds to a phosphoserine residue in Raf, one to Ser-259 and the other to Ser-621. Ras · GTP, which is anchored to the membrane, recruits inactive Raf to the membrane and induces a conformational change in Raf that disrupts its association with 14-3-3. Ser-259 then is dephosphorylated, activating Raf’s kinase activity. After Ras returns to the GDP form, it dissociates from Raf and presumably can then be reactivated again, thereby recruiting additional Raf molecules to the membrane. In addition to its role in regulating Raf structure, the 14-3-3 dimer appears to have a more general function in linking together signaling components through phosphoserine residues.

Activation of Raf also requires Ksr, which contains binding sites for Raf, 14-3-3, MEK, and MAP kinase. Ksr may function as an adapter protein, providing a scaffold for formation of a large signaling complex that continues to operate after Raf dissociates from Ras · GDP. Although Ksr has a kinase domain, genetic studies in Drosophila have shown that this domain has a negative regulatory function, perhaps acting as part of a switch for regulating the turnover of components within the complex.

Phosphorylation of a Tyrosine and a Threonine Activates MAP Kinase

Biochemical and x-ray crystallographic studies have provided a detailed picture of how phosphorylation activates MAP kinase (Figure 20-30). In MAP kinase and other protein kinases, including the cytosolic domain of RTKs, the catalytic site in the inactive, unphosphorylated form is blocked by a segment of amino acids, the phosphorylation lip. Binding of MEK to MAP kinase destabilizes the lip structure, resulting in exposure of a critical tyrosine that is buried in the inactive conformation. Following phosphorylation of this tyrosine, MEK phosphorylates a neighboring threonine. Both the phosphorylated tyrosine and threonine residues in MAP kinase interact with additional amino acids, thereby conferring an altered conformation to the lip region, which in turn permits binding of ATP to the catalytic site. The phosphotyrosine residue also plays a key role in forming the binding site for specific substrate proteins on the surface of MAP kinase. Phosphorylation promotes not only the catalytic activity of MAP kinase but also its dimerization. The dimeric form of MAP kinase (but not the monomeric form) can be translocated to the nucleus where it regulates the activity of a number of nuclear localized transcription factors (see below).

Figure 20-30. Structures of MAP kinase in its inactive, unphosphorylated form (a) and active, phosphorylated form (b).

Figure 20-30

Structures of MAP kinase in its inactive, unphosphorylated form (a) and active, phosphorylated form (b). Phosphorylation of MAP kinase by MEK at tyrosine 185 (pY185) and threonine 183 (pT183) leads to a marked conformational change in the phosphorylation (more...)

Various Types of Receptors Transmit Signals to MAP Kinase

Although yeasts and other single-celled eukaryotes lack RTKs, they have been found to possess MAP kinase pathways. And in different cell types of higher eukaryotes, stimulation of receptors other than RTKs also can initiate signaling pathways leading to activation of MAP kinase. The mating pathway in S. cerevisiae is a well-studied example of a MAP kinase cascade that is linked to G protein – coupled receptors, in this case for two secreted peptide pheromones, the a and α factors. These pheromones control mating between haploid yeast cells of the opposite mating type, a or α. An a haploid cell secretes the a mating factor and has cell-surface receptors for the α factor; and an α cell secretes the α factor and has cell-surface receptors for the a factor (see Figure 14-4). Thus each type of cell recognizes the mating factor produced by the other opposite type.

As with the GPCRs discussed earlier, ligand binding to the yeast pheromone receptors triggers the exchange of GTP for GDP on the α subunit and dissociation of Gα · GTP from the Gβγ complex. In contrast to most GPCR systems, however, the dissociated Gβγ complex (not Gα · GTP) mediates all the physiological responses induced by activation of the yeast pheromone receptors. It does so by triggering a kinase pathway that is analogous to the one downstream from Ras. The components of this yeast kinase cascade were uncovered mainly through analyses of mutants that possess functional a and α receptors and G proteins but are defective in mating responses. The physical interactions between the components were assessed through immunoprecipitation experiments with extracts of yeast cells and other types of studies.

Based on these studies, scientists have proposed the kinase cascade depicted in Figure 20-31. Gβγ, which is tethered to the membrane via the γ subunit, binds to and activates Ste20, which in turn activates Ste11, a serine/threonine kinase analogous to Raf. Activated Ste11 then phosphorylates Ste7, a dual-specificity MEK, which in turn activates Fus3, a serine/threonine kinase equivalent to MAP kinase. After translocation to the nucleus, Fus3 promotes expression of target genes by activating nuclear transcription factors such as Ste12 by mechanisms that are not completely understood. The other component of the yeast mating cascade, Ste5, interacts with Gβγ as well as Ste11, Ste7, and Fus3. Ste5 has no obvious catalytic function and acts as a scaffold for assembling other components in the cascade.

Figure 20-31. Kinase cascade that transmits signals downstream from Gβγ in the mating pathway in S. cerevisiae.

Figure 20-31

Kinase cascade that transmits signals downstream from Gβγ in the mating pathway in S. cerevisiae. The receptors for yeast a or α mating factors are both coupled to the same trimeric G protein. Ligand binding leads to activation (more...)

Multiple MAP Kinase Pathways Are Found in Eukaryotic Cells

In addition to the MAP kinases discussed above, both yeasts and higher eukaryotic cells contain other functionally equivalent proteins, including the Jun N-terminal kinases (JNKs) and p38 kinases in mammalian cells and six yeast proteins described below. Collectively referred to as MAP kinases, all these proteins are serine/threonine kinases that are activated in the cytosol in response to specific extracellular signals and can be translocated to the nucleus. Activation of all known MAP kinases requires dual phosphorylation of analogous residues in the phosphorylation lip of the protein (see Figure 20-30). Thus in all eukaryotic cells, binding of a wide variety of extracellular signaling molecules triggers highly conserved kinase cascades culminating in activation of a particular MAP kinase. The different MAP kinases mediate specific cellular responses, including morphogenesis, cell death, and stress responses.

Current genetic and biochemical studies in the mouse and Drosophila are aimed at determining which MAP kinases are required for mediating the response to which signals in higher eukaryotes. This has already been accomplished in large part for the simpler organism S. cerevisiae. Of the six MAP kinases encoded in the S. cerevisiae genome, five have been assigned by genetic analyses to specific signaling pathways triggered by various extracellular signals, such as pheromones, starvation, high osmolarity, hypotonic shock, and carbon/nitrogen deprivation. Each of these MAP kinases mediates very specific cellular responses (Figure 20-32).

Figure 20-32. Overview of five MAP kinase pathways in S. cerevisiae.

Figure 20-32

Overview of five MAP kinase pathways in S. cerevisiae. Each pathway is triggered by a specific extracellular signal and leads to activation of a single MAP kinase, which mediates characteristic cellular responses. [Adapted from H.D. Madhani and G.R. Fink, (more...)

In both yeasts and higher eukaryotic cells, different MAP kinase cascades share some common components. For instance, in yeast, Ste11 functions in the signaling pathways that regulate mating, filamentous growth, and osmoregulation. Nevertheless, each pathway activates only one MAP kinase: Fus3 in the mating pathway, Kss1 in the filamentation pathway, and Hog1 in the osmoregulation pathway (see Figure 20-32). Similarly, in mammalian cells, common upstream signal-transducing molecules participate in activating multiple JNK kinases. Given this sharing of components among different MAP kinase pathways, how is the specificity of the responses to particular signals achieved? Recent studies of the MAP kinase pathways in yeast suggest possible answers to this question.

Specificity of MAP Kinase Pathways Depends on Several Mechanisms

As we’ve just explained, Ste11 is a component of the yeast mating, filamentation, and osmoregulatory MAP kinase pathways. A complex set of regulatory interactions appears to limit intracellular signaling in response to a particular extracellular signal to the appropriate pathway. Here we describe recent findings about two mechanisms that can determine the specificity of MAP kinase pathways.

Pathway-Specific Signaling Complexes

To illustrate one way multiple MAP kinases are segregated, we consider how a change in osmolarity activates Ste11 but does not lead to activation of downstream components in the mating pathway. There are two osmoregulatory MAP kinase pathways in S. cerevisiae; both lead to activation of the MAP kinase Hog1, but only one pathway requires Ste11. The dual-specificity MEK in this osmoregulatory pathway, called Pbs2, also functions as a scaffold for assembly of a large signaling complex. Pbs2 binds to Hog1, Ste11, and Sho1 (the osmolarity-sensitive receptor). Transmission of the signal from Sho1 to Hog1 occurs within the complex assembled by Pbs2. Recall that in the mating pathway, the scaffold protein Ste5 likewise stabilizes a large complex including Ste11. In both cases, the common component Ste11 is constrained within a large complex that forms in response to a specific extracellular signal, and signaling downstream from Ste11 is restricted to the complex containing it (Figure 20-33). As a result, exposure of yeast cells to mating factors induces activation of a single MAP kinase, Fus3, while exposure to a high osmolarity induces activation only of Hog1.

Figure 20-33. MAP kinase signaling complexes in the mating and osmoregulatory pathways in yeast.

Figure 20-33

MAP kinase signaling complexes in the mating and osmoregulatory pathways in yeast. Formation of such pathway-specific complexes prevents “cross-talk” between pathways that contain a common component, such as Ste11 in these two pathways. (more...)

Kinase-Independent Functions of MAP Kinases

Detailed genetic analysis of two different yeast MAP kinases, Fus3 (mating pathway) and Kss1 (filamentation pathway), have revealed another mechanism for restricting signaling to a single MAP kinase pathway. Mutant yeasts that express no Fus3 can still mate. In these “kinase-lacking” mutants, mating factors induce mating-specific genes, normally regulated by Fus3, by activating the catalytic activity of Kss1. Moreover, stimulation of these mutants by mating factors also induces filamentation-specific genes regulated by Kss1, which normally is activated in response to starvation. Kss1-mediated induction of mating-specific and filamentation-specific genes in response to pheromones requires Ste5, the scaffold protein in the mating pathway. Other yeast mutants, referred to as “kinase-dead” mutants, express a defective Fus3 without catalytic activity. In kinase-dead mutants, mating factors induce neither mating-specific nor filamentation-specific genes. Apparently, Kss1 cannot be activated in the presence of an altered, catalytically inactive Fus3.

The results with these two types of yeast mutants suggest that Kss1 can bind to Ste5, but it does so less efficiently than wild-type or kinase-dead Fus3. Under normal conditions, Fus3 is recruited into a Ste5 signaling complex in response to mating-factor stimulation of cells (see Figure 20-31). If Fus3 is missing altogether, as in the kinase-lacking mutants, then Kss1 is recruited into the complex. In contrast, the inactive Fus3 in kinase-dead mutants is effectively recruited into the complex, but cannot be activated. Hence, kinase-dead Fus3 acts as a plug that inhibits signal flow by physically preventing recruitment of Kss1 into the complex. Thus, Fus3 has two separable functions, a kinase-dependent function necessary to induce its appropriate target genes and a kinase-independent function that prevents activation of the closely related MAP kinase, Kss1.

Genetic experiments reveal that Kss1 also has kinasedependent and kinase-independent functions. In the absence of a filamentation-inducing signal, inactive Kss1 is bound in the cytosol to a transcription factor complex required for induction of filamentation-specific genes; in this form Kss1 inhibits transcription. Induction of filamentation-specific genes in wild-type yeast specifically requires activation of Kss1 kinase activity. In yeast mutants that lack Kss1, filamentation-specific genes are inappropriately induced by Fus3-mediated activation of the transcription factor complex in response to mating factors. However, kinase-dead Kss1, which can bind to the transcription factor complex, prevents this cross-talk presumably by physically preventing access of Fus3. It is not yet known, however, why activated Kss1 fails to activate Ste12 dimers promoting mating-specific gene expression in wild-type yeast.

Mammalian MAP kinases have been found to bind to specific proteins in a kinase-independent fashion, raising the possibility that MAP kinases in animals, like those in yeast, may restrict signal specificity through kinase-independent functions.


  •  Activated Ras promotes formation of signaling complexes at the membrane containing three sequentially acting protein kinases and a scaffold protein Ksr. Raf is recruited to the membrane by binding to Ras · GTP and then activated. It then phosphorylates MEK, a dual specificity kinase that phosphorylates MAP kinase. Phosphorylated MAP kinase dimerizes and translocates to the nucleus where it regulates gene expression (see Figure 20-28).
  •  RTKs, GPCRs, and other receptor classes can activate MAP kinase pathways. Single-cell eukaryotes, such as yeast, and multicellular organisms contain multiple MAP kinase pathways that regulate diverse cellular processes (see Figure 20-32).
  •  Although different MAP kinase pathways share some upstream components, activation of one pathway by extracellular signals does not lead to activation of others containing shared components.
  •  In MAP kinase pathways containing common components, the activity of shared components is restricted to only a subset of MAP kinases by their assembly into large pathway-specific signaling complexes (see Figure 20-33).
  •  Some MAP kinases have kinase-independent functions that can restrict signals to only a subset of MAP kinases.
Image ch20f23
Image ch14f4
Image ch20f16

By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.

Copyright © 2000, W. H. Freeman and Company.
Bookshelf ID: NBK21529