Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Curr Biol. Author manuscript; available in PMC 2011 January 7.
Published in final edited form as:
PMCID: PMC3017499

Signal Transduction: Turning a Switch into a Rheostat


MAP kinase cascades are inherently switch-like, but, during yeast mating, MAPK signaling is graded. A new study suggests that the Ste5 scaffold protein is responsible for making this switch less switch-like.

The individual elements in an electronic circuit can be analyzed in terms of how they convert input into output. Like electronic circuits, intracellular signal transduction networks also appear to contain modular elements, such as protein kinase cascades. Indeed, when the first protein kinase cascade was discovered, it was hypothesized to function as an amplifier. More recently, eukaryotic mitogen-activated protein kinase (MAPK) cascades have been shown to be inherently switch-like. Do modules like kinase cascades always carry out similar functions regardless of which signaling pathway they are placed in? If not, what are the mechanisms that modulate kinase function to allow different input–output mappings? In a recent issue of Current Biology, Takahashi and Pryciak [1] show that a scaffold protein can counteract the switch-like tendencies of the MAPK cascade to allow graded signaling, essentially turning an on–off switch into a dimmer.

MAPK cascades are found embedded in signaling networks that transmit many different signals, including those initiated by growth and development factors, inflammatory stimuli, and cellular stresses. The module consists of a MAP3K, which phosphorylates and thereby activates a MAP2K, which in turn phosphorylates and activates a MAPK. In the early 1990s, when components of mammalian MAPK cascades were cloned, it was discovered that they were highly homologous to several yeast kinases (Ste11MAP3K, Ste7MAP2K, Kss1MAPK and Fus3MAPK) known to function in the signal transduction pathway by which yeast cells respond to peptide mating pheromone [2]. Subsequent studies showed that not only were the components of MAPK pathways conserved between yeast and humans, but also their arrangement into a cascade was conserved. Thus began the idea of a conserved signaling module that must carry out some evolutionarily conserved function.

But what function? Why arrange proteins in a cascade? The textbook wisdom at the time pointed to signal amplification, with each activated kinase molecule in a given tier activating multiple kinase molecules in the next tier, like a triangular cascade of dominoes. This concept originated during studies of the prototypical kinase cascade that, during ‘fight or flight’, responds to small amounts of adrenaline by quickly releasing a lot of sugar into the bloodstream [3]. However, it was not clear that MAPK-mediated growth factor and developmental signaling had similar needs for speed and amplification. These pathways are primarily involved in binary decision making, e.g. ‘to divide or not?’, ‘to differentiate or not?’ and even ‘to be or not?’. Using a cascade that could rapidly amplify some spurious noise into a full-blown response seemed a poor way to make such decisions.

Beginning in 1996, in a series of important papers, Jim Ferrell and colleagues proposed that the MAPK cascade was not an amplifier, but a switch [4,5]. They showed that the output of the cascade (activated MAPK) was relatively insensitive to low levels of input noise (a property known as thresholding), but increased much more suddenly once a certain threshold of input had been reached (a property known as switching, or ‘ultrasensitivity’). Switching occurred even if the level of input MAP3K was ramped up steadily and gradually. Furthermore, the ultrasensitivity intrinsic to the MAPK cascade could be magnified by positive feedback into a bistable, all-or-none response, so that a cell would never decide to differentiate halfway, or ‘sort of be’ [6].

The thresholding and ultrasensitivity of the MAPK cascade that Ferrell and colleagues observed was unexpected, because simple biochemical reactions have a strong tendency to be graded: the output/response is linearly proportional to input/dose at low doses, but flattens out at high doses as the response asymptotically approaches its maximum [5]. This is attributable to the combination of simple kinetics with the diminishing returns that occur in a reversible reaction as the substrate is used up and the product accumulates. Theory indicates that the MAPK cascade behaves differently because of two key properties. First, a modest amount of thresholding and ultrasensitivity occurs at each step in the cascade because each kinase requires two or three phosphorylations by its upstream kinase in order to become activated, and these phosphorylations are distributive (where only one phosphate is added for each kinase–substrate collision) rather than processive (where multiple phosphates are added following a single collision) [4,7]. Second, the effect of each of the steps multiplies when they are chained together in a cascade [8].

Although there are several biological systems where the MAPK cascade has been shown to mediate a switch-like response (e.g. [6,9]), there are also important cases where the response has been shown to be more graded (e.g. [10]). In particular, even though it might seem that a haploid yeast cell which finds itself near a member of the opposite sex has a simple, binary decision to make (i.e., to mate or not), the transcriptional response to mating pheromone is graded [11,12]. Why? The likely explanation is, as with human mating rituals, there are a lot of intermediate steps between flirtation and consummation: courtship is not an all-or-none process. Rather, each successively more committed step requires a higher level of interest. A haploid yeast cell responds to a wisp of pheromone by upregulating the transcription of certain key genes that allow it to increase both its sensitivity to others’ pheromone and the production of its own pheromone. In other words, it pricks up its ears and dabs on some more perfume. At yet a higher concentration of pheromone it will go into cell-cycle arrest, forgetting about making more of itself in order to focus solely on romance. At even higher concentrations of pheromone the yeast cell, unable to move, will nevertheless make its move by extending a projection towards a nearby partner. This last bit requires gradient sensing, which would also be difficult if the initial response were all-or-none. Finally, courtship culminates when the two cells fuse to form a diploid.

If the MAPK cascade has a tendency to function as a switch, are there special mechanisms that allow it to function as a rheostat in cases, like yeast mating, where the physiological imperatives so demand? Takahashi and Pryciak [1] addressed this question by hooking up the gene encoding green fluorescent protein to a pheromone-inducible promoter and monitoring the transcriptional response to pheromone in single cells. They first hypothesized that the multiple layers of negative feedback known to operate on this pathway might counteract the MAPK cascade switch to allow graded signaling. However, they found that the graded nature of the response was robust to the elimination of several of the most prominent negative-feedback loops.

So perhaps the yeast MAPK cascade is not intrinsically ultrasensitive after all? To address this possibility, the authors expressed constitutively active alleles of various pathway components from an estrogen-inducible promoter. This allowed them both to activate the mating pathway at various points downstream of the pheromone receptor and to vary the level of activation over a broad range by adjusting the dose of estrogen. When signaling was triggered near the top of the pathway by expression of Ste4, the β subunit of the receptor-coupled G protein, the response remained graded. In contrast, when signaling was triggered at the entry point into the MAPK cascade by expression of a constitutively active allele of Ste11MAP3K, the response was markedly more switch-like. Thus, the yeast MAPK cascade does appear to be inherently ultrasensitive.

Given that the MAPK pathway was ultrasensitive in isolation, what was counteracting this ultrasensitivity when the full pathway was operating? The authors focused on the MAPK scaffold protein Ste5, which binds to all of the kinases of the MAPK cascade. Scaffold proteins are thought to work by tethering their bound kinases near to each other, increasing the chance for a productive reaction to occur. However, Ste5 has other functions in addition to scaffolding and is clearly more than just a passive platform upon which kinases assemble [13,14]. Early on during pheromone signaling, Ste5 concentrates at the plasma membrane by virtue of binding to activated Ste4 and membrane phospholipids. One of its key roles here is to tether Ste11MAP3K near membrane-bound Ste20MAP4K, so Ste20 can activate Ste11. It is likely that membrane localization also enhances Ste5’s ability to act as a scaffold and promote MAP2K and MAPK activation.

In a key experiment, Takahashi and Pryciak [1] showed that targeting various amounts of Ste5 to the plasma membrane in the absence of pheromone activated the MAPK cascade and did so in a graded fashion. The implication is that when the kinases of the MAPK cascade activate each other in the cytoplasm, signaling is ultrasensitive, but when the kinases activate each other while bound to Ste5 at the membrane, signaling is graded. Membrane-bound Ste5 thus blunts the switch, allowing graded signaling.

How does membrane-associated Ste5 promote graded signaling? One possibility is that, by tethering the kinases together, Ste5 increases the processivity of the phosphorylation reactions, as originally proposed in 2000 by Levchenko et al. [15]. Membrane association of Ste5 may increase its affinity for its ligands, and may also promote between-scaffold collisions. These effects will increase the number of collisions that an unactivated kinase molecule sustains before it dissociates from the scaffold and thereby increase the chance that it will leave with both its target residues phosphorylated (Figure 1). In contrast, cytoplasmic and nuclear Ste5 may have a much lower binding affinity for its ligand kinases and be unable to participate in between-scaffold interactions. Indeed, Takahashi and Pryciak [1] present evidence that cytoplasmic Ste5 actually makes signaling triggered by constitutively active Ste11MAP3K more ultrasensitive. However, this effect is perhaps related to the fact that Ste5 scaffolding is required for Fus3MAPK activation, but not Kss1MAPK activation, and Fus3 activation is (for unclear reasons) more ultrasensitive than Kss1 activation [1,16]. It is uncertain whether cytoplasmic Ste5 increases the ultrasensitivity of Fus3 activation or merely permits an intrinsically ultrasensitive step to occur [16].

Figure 1
Model for promotion of graded signaling by membrane-localized Ste5, based on the results of Takahashi and Pryciak [1]. (A) When MAP3K is activated (by noise or undesired crosstalk) in the absence of membrane-bound scaffold, distributive phosphorylations ...

One of the many interesting implications of the new work is that graded signaling is inducible. In unstimulated cells, Ste5 is found in the nucleus and cytoplasm, and thus MAPK activation has a relatively high threshold, which would minimize sensitivity to noise and to spurious crosstalk [14]. Pheromone stimulation would trigger Ste5 membrane recruitment, promoting graded signaling and, in effect, sensitizing the MAPK cascade.

The new work is consistent with the idea that the MAPK cascade has been conserved not because it performs a particular circuit function such as acting as a switch, but because it can be adapted to perform a variety of different functions depending on the performance objectives of the pathway in which it is embedded and the physiological requirements of the organism [17,18]. Versatility may be the most important design feature of the MAPK cascade.


1. Takahashi S, Pryciak PM. Membrane localization of scaffold proteins promotes graded signaling in the yeast MAP kinase cascade. Curr Biol. 2008;18:1184–1191. [PMC free article] [PubMed]
2. Bardwell L, Cook JG, Inouye CJ, Thorner J. Signal propagation and regulation in the mating pheromone response pathway of the yeast Saccharomyces cerevisiae. Dev Biol. 1994;166:363–379. [PubMed]
3. Press Release for the 1992 Nobel Prize in Physiology or Medicine to Edmond H. Fischer and Edwin G. Krebs. (
4. Huang CY, Ferrell JE., Jr Ultrasensitivity in the mitogen-activated protein kinase cascade. Proc Natl Acad Sci USA. 1996;93:10078–10083. [PubMed]
5. Ferrell JE., Jr Tripping the switch fantastic: how a protein kinase cascade can convert graded inputs into switch-like outputs. Trends Biochem Sci. 1996;21:460–466. [PubMed]
6. Ferrell JE, Jr, Machleder EM. The biochemical basis of an all-or-none cell fate switch in Xenopus oocytes. Science. 1998;280:895–898. [PubMed]
7. Gunawardena J. Multisite protein phosphorylation makes a good threshold but can be a poor switch. Proc Natl Acad Sci USA. 2005;102:14617–14622. [PubMed]
8. Ferrell JE., Jr How responses get more switch-like as you move down a protein kinase cascade. Trends Biochem Sci. 1997;22:288–289. [PubMed]
9. Bagowski CP, Besser J, Frey CR, Ferrell JE. The JNK cascade as a biochemical switch in mammalian cells. Ultrasensitive and all-or-none responses. Curr Biol. 2003;13:315–320. [PubMed]
10. Mackeigan JP, Murphy LO, Dimitri CA, Blenis J. Graded mitogen-activated protein kinase activity precedes switch-like c-Fos induction in mammalian cells. Mol Cell Biol. 2005;25:4676–4682. [PMC free article] [PubMed]
11. Paliwal S, Iglesias PA, Campbell K, Hilioti Z, Groisman A, Levchenko A. MAPK-mediated bimodal gene expression and adaptive gradient sensing in yeast. Nature. 2007;446:46–51. [PubMed]
12. Poritz MA, Malmstrom S, Kim MK, Rossmeissl PJ, Kamb A. Graded mode of transcriptional induction in yeast pheromone signalling revealed by single-cell analysis. Yeast. 2001;18:1331–1338. [PubMed]
13. Elion EA. The Ste5p scaffold. J Cell Sci. 2001;114:3967–3978. [PubMed]
14. Flatauer LJ, Zadeh SF, Bardwell L. Mitogen-activated protein kinases with distinct requirements for Ste5 scaffolding influence signaling specificity in Saccharomyces cerevisiae. Mol Cell Biol. 2005;25:1793–1803. [PMC free article] [PubMed]
15. Levchenko A, Bruck J, Sternberg P. Scaffold proteins may biphasically affect the levels of mitogen-activated protein kinase signaling and reduce its threshold properties. Proc Natl Acad Sci USA. 2000;97:5818–5823. [PubMed]
16. Hao N, Nayak S, Behar M, Shanks RH, Nagiec MJ, Errede B, Hasty J, Elston TC, Dohlman HG. Regulation of cell signaling dynamics by the protein kinase-scaffold Ste5. Mol Cell. 2008;30:649–656. [PMC free article] [PubMed]
17. Bhalla US, Ram PT, Iyengar R. MAP kinase phosphatase as a locus of flexibility in a mitogen-activated protein kinase signaling network. Science. 2002;297:1018–1023. [PubMed]
18. Bashor CJ, Helman NC, Yan S, Lim WA. Using engineered scaffold interactions to reshape MAP kinase pathway signaling dynamics. Science. 2008;319:1539–1543. [PubMed]