PerspectiveSignal Transduction

Signaling Specificity in Yeast

See allHide authors and affiliations

Science  04 Feb 2005:
Vol. 307, Issue 5710, pp. 687-688
DOI: 10.1126/science.1109500

A central problem that continues to puzzle biologists is how cells translate myriad stimuli into highly specific responses. Signaling specificity is complicated in eukaryotic cells because signal transduction pathways that respond to different stimuli and perform distinct functions often share the same pathway components. In haploid budding yeast, for example, the mitogen-activated protein kinase (MAPK) cascades that regulate both mating and filamentous growth share the same three activating kinases—Ste20 MAPKKKK or PAK-type kinase, Ste11 MAPKKK, and Ste7 MAPKK (which are activated serially)—as well as the Ste12 transcription factor (see the figure). Two separate MAPKs are activated by the same Ste7 MAPKK: Fus3, which is essential for the mating program, and Kss1, which is essential for the filamentous growth program (also called invasive growth in haploids and pseudohyphal development in diploids). Fus3 and Kss1 are both activated in response to mating pheromone by the MAPK scaffold protein Ste5; Kss1 also can be activated by cell surface proteins such as mucin (Msb2) and other conditions that promote invasive growth [(13); see (4) for a summary of targets in the mating and invasive growth pathway]. How is pathway specificity maintained with so much redundancy? Two papers in a recent issue of Cell reveal how the Fus3 and Kss1 MAPK signaling pathways are insulated from each other, ensuring smooth execution of the mating program without erroneous activation of the filamentous growth program (5, 6).

Signaling specificity is possible in part because MAPKs choose different substrates. For example, during mating, Fus3 phosphorylates the cell cycle regulator Far1, whereas both Fus3 and Kss1 can phosphorylate Ste12 (see the figure). During filamentous growth, Kss1 regulates the Tec1 transcription factor indirectly through Ste12, which is guided by Tec1 to the promoters of filamentous growth genes (4). In addition, MAPKs use different activation mechanisms—for example, Fus3 must interact with the Ste5 scaffold to be activated, whereas Kss1 activation does not have this requirement (13). However, these levels of control do not explain how the kinases are insulated from erroneous cross-activation. For example, why doesn't activation of mating pheromone by Kss1 induce a filamentous growth program?

One level of insulation comes from attenuation of nonessential MAPKs under any given condition. During filamentous growth, selective repression of Fus3 by phosphatases such as Msg5 enforces the filamentous growth program (1). During pheromone stimulation, Fus3 attenuates the activation of Kss1 and influences its strength and duration of signaling (7). Another key level of insulation comes from the inhibition of filamentous growth by Fus3 and the expression of Tec1-dependent genes during pheromone signaling (8). The mysterious mechanism of this potent insulation is now revealed by the two Cell papers (5, 6) and a third study (9). These reports show that the Fus3 and Kss1 signaling cascades are insulated from each other because Fus3 controls degradation of Tec1, the transcription factor essential for activation of the filamentous growth program (see the figure). The beauty of these studies is that they use established yeast mutants that are hyperfilamentous as a result of mutations in Tec1 (Thr273 switched to methionine, and Pro274 switched to serine). These yeast mutants point the way to discovering the critical residue in Tec1 that is phosphorylated by Fus3, signaling the degradation of Tec1. Thr273 and Pro274 together form part of a consensus phosphorylation site for proline-directed MAPKs (5, 6). The threonine residue at position 273 is one of multiple sites phosphorylated by Fus3 in vitro (5, 6, 9) and in pheromone-treated cells in vivo (5). The proline residue at position 274 is predicted to be necessary for recognition of the threonine by a MAPK. However, Tec1 mutants with a methionine at residue 273 or a serine at residue 274 are resistant to pheromone-induced degradation and display almost the same absence of insulation as a yeast mutant that completely lacks the FUS3 gene (5, 6). These findings argue compellingly that down-regulation of the Tec1 protein is the major way in which Fus3 insulates the mating MAPK cascade from the filamentous growth MAPK cascade. In contrast, Kss1 does not phosphorylate Tec1 (5, 6, 9), ensuring that Tec1 remains stable during filamentous growth. Interestingly, the amount of Tec1 degraded is proportional to the concentration of the pheromone stimulus (5). This is consistent with earlier findings by several groups showing that a low concentration of mating pheromone induces production of TEC1 mRNA and invasive growth. Thus, the system may have evolved to permit dividing yeast both to forage by filamentous growth and to search for optimal mating conditions while cells are primed for mating, and before Tec1 is degraded.

No fuss about insulation.

(A) The Fus3 and Kss1 MAPKs are activated by the same signaling cascades—that is, they share the common activating kinases Ste20 MAPKKKK, Ste11 MAPKKK, and Ste7 MAPKK. Although Fus3 and Kss1 are both activated by mating-specific inputs that involve the Ste5 scaffold protein (red; pheromone receptor, G protein-MAPK scaffold), only Kss1 can be activated independently of Ste5 by inputs that promote filamentous growth (blue; the mucin Msb2 and nutrient deprivation). These two MAPKs have both common and unique substrates. Fus3 phosphorylates the Far1 cyclin-dependent kinase inhibitor and the Tec1/Ste12 transcription factor, two events critical for mating. Kss1 regulates the activation of Tec1/Ste12, which is necessary for filamentous growth. During mating, Kss1 activation switches on expression of both mating and filamentous growth genes, including the TEC1 gene (depicted by dashed arrows as they are not essential for mating and interfere with cell cycle arrest, which is needed for terminal differentiation). (B) Fus3 phosphorylation of the Tec1 protein on Thr273 leads to ubiquitin-dependent degradation of Tec1 by a SCF ubiquitin-ligase complex, neatly blocking entry into the filamentation program. Thus, the status of the Fus3 MAPK determines whether cells mate or enter a filamentous growth program. This regulatory device is not necessary in diploid cells, which do not mate or express Fus3.


Additional evidence suggests that phosphorylated Tec1 may be recognized for destruction by ubiquitin-dependent proteolysis through the action of an SCF (Skp1-Cdc53/Cul1-F-box) ubiquitin ligase complex (5, 6). Tec1 is ubiquitinated in the presence of mating pheromone (5, 6) but not if residue 273 is mutated or Fus3 is absent (6). In addition, Tec1 is completely stabilized in a cdc34-2 yeast mutant (which lacks the E2 enzyme) (5, 6) and a cdc53-1 mutant (6) in the presence of pheromone (although it is possible that these pleiotropic mutations could have interfered with the activation of Fus3). The Chou et al. (5) and Bao et al. (6) groups reach different conclusions about the identity of the F-box protein that recognizes phosphorylated Tec1 and links it to the SCF ubiquitin-ligase complex. Chou et al. (5) propose SCFCdc4 (WD40 class) as their candidate for the following reasons. Tec1 is stable in a cdc4-3 mutant although it is phosphorylated at residue 273; Fus3-phosphorylated Tec1 interacts with Cdc4 in vitro; and Tec1 cannot be ubiquitinated in a cdc4-3 yeast mutant. In their turn, Bao et al. (6) propose an as yet undefined SCFDia2 protein (LRR class by homology) on the basis of the stabilization of Tec1 in a dia2 mutant that is known to be hyperfilamentous (10). Although it is possible that multiple SCF complexes regulate Tec1 [and the data in Chou et al. support short-term stabilization of Tec1 in a dia2 mutant as well as in a grr1 mutant that is also hyperfilamentous (10)], more work is needed to dissect the possible interplay between different SCF complexes and between the direct and indirect effects of mutations on Tec1 production.

The most important notion gleaned from these studies is that pathway specificity can be regulated by the MAPKs downstream of the MAPK signaling cascades. Previous work supports the possibility that MAPK-induced proteolysis of MAPK targets may be a general means of controlling pathway specificity (11, 12). Such a regulatory device permits flexibility by focusing control on specific MAPK outputs while maintaining others that may be beneficial for the desired program. Proteolysis is advantageous because it is rapid and irreversible, allowing for sharp transitions. Phosphorylation-induced proteolysis of substrates provides a way to get rid of unwanted by-products of activated protein kinases at sites distal from the place of activation. There are additional opportunities for regulating substrate: at the place and time of substrate degradation, and during various phosphorylation events that are required for substrate to be recognized by the proteolysis machinery. These different methods of regulation are also applicable to feedback and cross-regulation of signal transduction pathways. Perhaps Nature has found a simple strategy to take full advantage of redundancies in a system that has been designed to be exquisitely sensitive.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
View Abstract

Stay Connected to Science

Navigate This Article