Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 136 (6), 1085-97

The Ste5 Scaffold Directs Mating Signaling by Catalytically Unlocking the Fus3 MAP Kinase for Activation


The Ste5 Scaffold Directs Mating Signaling by Catalytically Unlocking the Fus3 MAP Kinase for Activation

Matthew Good et al. Cell.


The scaffold protein Ste5 is required to properly direct signaling through the yeast mating pathway to the mitogen-activated protein kinase (MAPK), Fus3. Scaffolds are thought to function by tethering kinase and substrate in proximity. We find, however, that the previously identified Fus3-binding site on Ste5 is not required for signaling, suggesting an alternative mechanism controls Fus3's activation by the MAPKK Ste7. Reconstituting MAPK signaling in vitro, we find that Fus3 is an intrinsically poor substrate for Ste7, although the related filamentation MAPK, Kss1, is an excellent substrate. We identify and structurally characterize a domain in Ste5 that catalytically unlocks Fus3 for phosphorylation by Ste7. This domain selectively increases the k(cat) of Ste7-->Fus3 phosphorylation but has no effect on Ste7-->Kss1 phosphorylation. The dual requirement for both Ste7 and this Ste5 domain in Fus3 activation explains why Fus3 is selectively activated by the mating pathway and not by other pathways that also utilize Ste7.


Figure 1
Figure 1. Ste5 scaffold protein is required for mating pathway signaling
(A) The MAPKKK Ste11 and MAPKK Ste7 function in both the mating and filamentation pathways in yeast. Ste7 must select the appropriate MAPK to phosphorylate in response to input (Fus3 for α-factor, and Kss1 in response starvation). (B) During mating, stimulation with α-factor leads primarily to phosphorylation of Fus3. This reaction requires the scaffold protein Ste5. Starvation input specifically induces the filamentation response through phosphorylation of Kss1. The Ste5 scaffold is not required for filamentation. (C) Expression of a constitutively active allele of MAPKKK Ste11 in a strain lacking Ste5 results only in Kss1 phosphorylation (both Kss1 and Fus3 phosphorylation are observed in strains with Ste5), further indicating that the Ste5 scaffold is required, in vivo, for Ste11➔Ste7➔Fus3 signaling (Flatauer et al., 2005).
Figure 2
Figure 2. Fus3 is intrinsically a poor substrate for Ste7, unless the Ste5 scaffold is present
(A) Fus3 and Kss1 both bind tightly to docking motifs (D-motifs) on Ste7 (KD ~100nM for each MAPK). (B) Coomassie stained gel showing purified components of the mating and filamentation MAPK pathways. (C & D) Activation of Kss1 and Fus3 by Ste7EE in vitro measured using the Trulight kinase assay - in which phosphorylation of a MAPK-specific labeled peptide substrate results in a decrease in fluorescence over time (the peptide quenches signal of a sensor bead coated with fluorescent polymers) (See Supp. Fig. 2A–C). 50nM of each protein was used in these assays. Ste7EE rapidly activates Kss1, and addition of the Ste5 scaffold has no impact on the reaction. (D) Fus3 cannot be activated by Ste7EE, unless ΔN-Ste5 is added. These results demonstrate that Fus3 is intrinsically a very poor substrate for Ste7, and that Ste5 is a required co-activator in Ste7➔Fus3 phosphorylation
Figure 3
Figure 3. Ste5 contains a novel domain required for Ste7➔Fus3 phosphorylation
(A) Ste5 is a large protein (917aa) that contains previously identified binding sites for the mating pathway kinases. Canonical tethering model proposes that Ste5 co-localizes three kinases in the mating pathway (Ste11, Ste7, Fus3) to promote signaling. (B) Deletion mapping identifies minimal region of Ste5 required for Ste7EE➔Fus3 phosphorylation in vitro. As in Figure 2, Trulight assay was used to measure Fus3 activation by Ste7EE. Amino acids 593–786 of Ste5 define the ‘minimal scaffold’ domain (Ste5-ms) sufficient to promote Ste7➔Fus3 phosphorylation. (C) Confirmation that the Fus3-binding region (KD = 1 µM) in Ste5 is not required for phosphorylation of Fus3 by Ste7EE. ΔN-Ste5-ND (green curve) is a variant of ΔN-Ste5 (black curve) bearing a mutation in the Fus3 binding region that disrupts interaction with Fus3. For panels C-E all reaction components are at 50 nM. (D) Ste5-ms domain is as active at the larger scaffold protein (ΔN-Ste5). (E) MAPK docking motifs on Ste7EE (KD ~ 100nM) are necessary for Fus3 activation. Mutation of these sites disrupt Ste7➔ Fus3 phosphorylation, even in the presence of Ste5 (purple curve). (F) Ste5-ms binds to Ste7 but not to Fus3. Interactions were measured with fluorescence polarization (anisotropy) using 5nM of fluoroscein-labeled Ste5-ms. (G) Minimal interactions necessary for formation of the Ste5-Ste7-Fus3 signaling complex.
Figure 4
Figure 4. Ste5-ms domain selectively improves kcat (not KM) for the substrate, Fus3
(A) Simple kinetic scheme for Ste7➔MAPK phosphorylation. Ste7EE enzyme converts substrate (MAPK) into doubly-phosphorylated product (MAPK-pp). Fus3 and Kss1 phosphorylation by Ste7EE was quantified using in vitro western blots with an anti-phospho p44/42 MAPK antibody (See Supp Fig. 1A–D and Supp Fig. 4). (B) Michaelis-Menten plots show Fus3 phosphorylation requires Ste5-ms, Kss1 phosphorylation does not. Ste7EE-ND2 (which contains only one MAPK docking motif, KD ~ 100nM), and Fus3-K42R (which is catalytically dead) were used to simplify the analyses. Kinase reactions contain 50nM Ste7EE-ND2, and a saturating concentration (1000nM, where appropriate) of Ste5-ms (see panel 4E). Fus3 activation by Ste7EE-ND2, in the absence Ste5-ms, is very slow but can be measured (inset graph). (C) Ste5-ms enhances the kcat of Ste7➔Fus3 phosphorylation by ~ 5000-fold, with negligible effect on KM. Ste5-ms has little or no effect on the kcat or KM of Kss1 phosphorylation. Overall specificity (kcat /KM) of Ste7 for Fus3 and Kss1 is comparable (~105 M−1 s−1). (D) Effect of Ste5 on Ste7➔MAPK phosphorylation reaction parameters, plotted as the fold-change in kcat, 1/KM, and kcatK/M for Fus3 and Kss1 activation by Ste7EE-ND2. Major effect of Ste5 is enhancement of the kcat for Fus3 phosphorylation. (E) Determination of the concentration of Ste5-ms required to drive Ste7➔Fus3 phosphorylation. 50nM Ste7EE was used along with a saturating amount of Fus3, (750nM, based on Fig. 4B). Rate of Fus3 activation reaches half-maximum at ~ 161nM Ste5-ms (+/− 60nM), which we infer is an apparent dissociation constant for the Ste7/Ste5-ms interaction. 1000 nM Ste5-ms, used in experiments described in panel 4B, represents a saturating concentration. (F) Reaction free energy diagram illustrating how Ste5-ms selectively lowers the energy of the transition state for Fus3 phosphorylation (dotted line).
Figure 5
Figure 5. Ste5-ms is a folded domain with distinct surfaces important for kinase-binding and catalysis
(A) Crystal structure of the Ste5 ms domain (1.6 Å resolution; data collection and refinement statistics can be found in Supp. Table 2). Structural figures were made using Pymol (DeLano, 2002). (B) Structural alignment using DALI illustrates the Ste5-ms domain is homologous to the von-Willebrand Type-A (VWA) domain. Cartoon of VWA domain fold and topology. (C) Ste5-ms has two distinct surfaces critical for Fus3 phosphorylation by Ste7 (identified by surface mutant scan of Ste5-ms for mutations with > 100-fold decrease in activity - see Supp Fig. 6 for full list of mutants used in the scanning experiment). One interface, the ‘coactivator loop’ (745–756) is critical for catalyzing Ste7➔Fus3 phosphorylation (phenotypes are represented by mutant ‘C’, N744A/D746A), and another interface is necessary for Ste7-binding (represented by mutant ‘B’, deletion of 778–786). (D) kcat of Ste7➔Fus3 phosphorylation reduced 100-fold for Ste5-ms mutant B and reduced nearly 1000-fold for mutant C. These mutants have no effect on the KM of Ste7-Fus3 phosphorylation (data not shown). Ste5-ms variants present at 1 µM, a concentration that saturates binding to Ste7 for Ste5-ms wild-type. (E) Pull-down assays show Ste5-ms mutant B is defective in binding to Ste7; mutant ‘C’ maintains Ste7 binding. Ste5-ms mutants were expressed as fusions to maltose binding protein (MBP) as a pull-down affinity tag. (F) Catalysis of Ste7➔Fus3 reaction by Ste5-ms mutant B, but not mutant C, can be restored by adding much higher concentrations of the mutant scaffold domain. Vmax for Ste7➔Fus3 reaction, measured using 50nM Ste7EE and 750nM Fus3. Point of half-max activation (Kact) gives apparent dissociation constants of Ste5-ms variants for Ste7. As expected, wild-type Ste5-ms has a Kactiv of 150nM, while Mutant ‘B’ had greatly diminished Kact = 15,500 nM, consistent with a defect in Ste7 binding. At high enough concentrations, mutant B can promote signaling to near wild-type levels. Ste5-ms mutant C shows a Kact close to wild-type (71nM) ( Supp Fig. 7B), but its Vmax at saturating concentrations is 1000-fold lower than wildtype (bar graph to right). This behavior is consistent with a defect in the catalytic step of Ste7➔Fus3 phosphorylation.
Figure 6
Figure 6. Ste5-ms catalytically unlocks Fus3 for phosphorylation by Ste7
(A) Two potential models for how Ste5-ms enhances Ste7➔Fus3 phosphorylation. One model proposes that Ste5-ms primarily acts on Ste7; Ste7 is a poor enzyme that requires Ste5-ms binding to increase its activity (top). An alterative model hypothesizes that Ste5-ms acts primarily on Fus3 - converting it from a poor substrate to a good one (bottom). (B) Ste5-ms has no effect on overall catalytic activity of Ste7EE as tested against the general kinase substrate, Myelin Basic Protein (MBP) using a 32P kinase assay. (C) To identify elements in Fus3 that make it a poor substrate compared to Kss1, we made mutations in Fus3 that make it more similar in sequence to Kss1 (Supp. Fig. 8A,B). These mutants were tested for their ability to be phosphorylated by Ste7EE in the absence of Ste5 (Supp. Fig 8C–D). A combined mutation of I161L with replacement of the 243–254 ‘MAPK insertion loop’ (with the same region from Kss1) created a Fus3 mutant with a 20-fold increase in kcat compared to wild-type (reaction contains 50nM Ste7EE-ND2, 750nM Fus3 variant, no scaffold). (D) Crystal structure of Fus3 (Remenyi et al., 2005), showing positions of critical mutations in red (I161L, and MAPK insert 243–254). The activation loop (shown as dotted line; not fully visible in the crystal structure) sits between these two regions. Residues that become phosphorylated (T180 and Y182) shown in green. (E) A model for Ste5-ms action: Fus3’s activation loop normally adopts a “locked” conformation, but Ste5-ms interaction with Fus3 transiently (and only in the presence of Ste7) stabilizes a transition state in which Fus3’s activation loop is accessible to Ste7.
Figure 7
Figure 7. A new model for how the Ste5 scaffold controls information flow in the mating MAPK pathway
(A) Ste5 has upregulatory (activating) and downregulatory interactions with Fus3. The strong, previously identified Fus3 binding site on Ste5 (Fus3-BD, KD = 1 µM) is not required for Ste7➔Fus3 phosphorylation, but rather is important for tuning down pathway output in vivo. Interactions that promote Fus3 phosphorylation involve the Ste5-ms domain (in cooperation with Ste7). (B) Cartoon summarizing various activities of Ste5. The Ste5-mediated complex has several critical tethering interactions (Ste5-Ste1 1, Ste5-Ste7, and Ste7-Fus3) essential for linear propagation of the mating pathway signal. In addition, Ste5-ms domain is an essential co-factor promoting catalysis of the Ste7➔Fus3 phosphorylation reaction. (C) Detailed model of minimal interactions in the mating scaffold complex required for Ste7➔Fus3 phosphorylation. Ste7 binds strongly to both Ste5-ms domain (via surface on Ste5-ms colored blue) and Fus3 (docking motifs on Ste7 bind to docking groove on Fus3 - colored gray), thereby tethering two proteins that normally interact only very weakly. Ste5-ms contains a coactivator loop (red surface) which promotes Fus3’s phosphorylation by Ste7. Fus3’s activation loop is colored red. Interaction affinities, where know, are indicated. Interactions that modulate kcat and KM of Fus3 phosphorylation by Ste7 are indicated by black boxes. Models for Fus3 (PDB code 2B9F) and Ste5-ms (this study) are derived from crystal structures. Ste7’s kinase domain was modeled from the structure of a homologous mammalian MAPKK (MKK7) using the threading program Phyre (Bennett-Lovsey et al., 2008).

Comment in

  • A MAPK Scaffold Lends a Helping Hand
    MA Seeliger et al. Cell 136 (6), 994-6. PMID 19303841.
    The scaffold proteins of signaling pathways are thought to act as passive tethering devices bringing together catalytic components of signaling cascades. Good et al. (200 …

Similar articles

See all similar articles

Cited by 79 PubMed Central articles

See all "Cited by" articles

Publication types

MeSH terms


Associated data