Cdc42 regulation of kinase activity and signaling by the yeast p21-activated kinase Ste20

Abstract

The Saccharomyces cerevisiae kinase Ste20 is a member of the p21-activated kinase (PAK) family with several functions, including pheromone-responsive signal transduction. While PAKs are usually activated by small G proteins and Ste20 binds Cdc42, the role of Cdc42-Ste20 binding has been controversial, largely because Ste20 lacking its entire Cdc42-binding (CRIB) domain retains kinase activity and pheromone response. Here we show that, unlike CRIB deletion, point mutations in the Ste20 CRIB domain that disrupt Cdc42 binding also disrupt pheromone signaling. We also found that Ste20 kinase activity is stimulated by GTP-bound Cdc42 in vivo and this effect is blocked by the CRIB point mutations. Moreover, the Ste20 CRIB and kinase domains bind each other, and mutations that disrupt this interaction cause hyperactive kinase activity and bypass the requirement for Cdc42 binding. These observations demonstrate that the Ste20 CRIB domain is autoinhibitory and that this negative effect is antagonized by Cdc42 to promote Ste20 kinase activity and signaling. Parallel results were observed for filamentation pathway signaling, suggesting that the requirement for Cdc42-Ste20 interaction is not qualitatively different between the mating and filamentation pathways. While necessary for pheromone signaling, the role of the Cdc42-Ste20 interaction does not require regulation by pheromone or the pheromone-activated G beta gamma complex, because the CRIB point mutations also disrupt signaling by activated forms of the kinase cascade scaffold protein Ste5. In total, our observations indicate that Cdc42 converts Ste20 to an active form, while pathway stimuli regulate the ability of this active Ste20 to trigger signaling through a particular pathway.

FIG. 1.

Mutations to test whether the Ste20 CRIB domain behaves as a Cdc42-regulated autoinhibitory domain. (A) Model for activation of Ste20, supported by this study, involving transition between low-activity (closed) and high-activity (open) conformations. The CRIB domain (grey) is indicated as having regions that bind Cdc42 (notch) or the kinase domain (bump). Asterisks denote residues mutated in this study. The Gβγ-binding domain was defined previously (29). (B) Alignment of CRIB domains in PAKs from S. cerevisiae (S.c.), S. pombe (S.p.) and Homo sapiens (H.s.). Residues identical in three or more proteins are boxed in black, and mutated residues are indicated (∗). Also indicated are human PAK1 residues (30, 37, 66, 68) involved in binding Cdc42 (caret) or the PAK1 kinase domain (+). A consensus CRIB motif is shown at the bottom; sequences C-terminal to this motif are well conserved among PAKs (68) but not between PAKs and other Cdc42 targets, such as mammalian WASP and ACK or yeast Gic1 and Gic2 (8, 37). (C) Effects of CRIB domain mutations on binding to Cdc42 and Bem1, measured by two-hybrid assay. Wild-type (pB20N2) and mutant (pPP1059, pPP1060, pPP1061, and pPP1062) derivatives of a DNA-binding-domain fusion to Ste20^1-499 were coexpressed in PPY760 with activation domain fusions to Bem1^157-551 (pRL51.1) or Cdc42^G12V/C188S (pPP1027), or with vector (pGAD424). β-Galactosidase measurements (mean ± standard deviation [SD]; n = 4) were normalized to the wild-type (WT) Ste20 allele (= 100%) for each binding partner; values in parentheses are the mean units for the wild-type interactions. The negative control (control) gives the basal signal of pB20N2 with pGAD424 (mean units = 0.04); the mutant derivatives of pB20N2 gave similar basal signals (not shown). (D) Effects on subcellular localization. Representative fluorescence images of cells (PPY913) expressing the indicated GFP-Ste20 mutants or carrying only a vector (pRS316).

FIG. 2.

Interaction between the Ste20 CRIB domain and Cdc42 is required for mating pathway function, in a manner that is bypassed by CRIB domain deletion. (A) CRIB domain point mutations reduce pheromone response in proportion to their Cdc42-binding defect. Strain PPY913 harbored a vector (pRS316) or the indicated STE20 plasmids in either native (pPP1001-based) or GFP fusion (pRL116-based) contexts. Tests of pheromone-induced growth arrest (left) and FUS1-lacZ expression (middle) and of mating efficiency (right) were done as described in Materials and Methods. (B) Signaling role of the Cdc42-Ste20 interaction is independent of pheromone and Gβγ. FUS1-lacZ induction was stimulated without addition of pheromone and in the absence of Gβ (Ste4) by galactose-induced synthesis of Ste5-CTM (pH-GS5-CTM) in an ste4Δ ste5Δ ste20Δ strain (PPY866). The strain also harbored a vector (pRS316) or the indicated GFP-STE20 mutant plasmid. (C) Mating and FUS1-lacZ results using alleles integrated at the genomic STE20 locus. Strains were PPY496, PPY640, PPY1205, and PPY1203. αF, α-factor. In A to C, bars indicate mean ± SD for three or more repeats.

FIG. 3.

The Ste20 CRIB domain behaves as an autoinhibitory domain. (A) Mutation of Leu369 is sufficient to allow Ste20 to signal without binding Cdc42. FUS1-lacZ and mating assays used strain PPY913 harboring the vector (pRS316) or plasmids encoding the indicated native Ste20 or GFP-Ste20 derivatives, similar to Fig. 2. Bars, mean ± SD (n = 4). Cdc42 binding two-hybrid results show mean β-galactosidase units (n = 3) from PPY760 cells cotransformed with pPP1027 and either pBTM116, pB20N2, pPP1119, pPP1061, or pPP1115. (B) Schematic diagram of interaction between CRIB and kinase domains, as detected by two-hybrid analysis (in panel C) using activation domain (AD) and DNA-binding domain (DBD) fusions to the indicated Ste20 fragments. Positions of point mutations are denoted by asterisks as in Fig. 1A. (C) CRIB-kinase interaction results. Strain PPY760 harbored plasmids encoding the indicated DNA-binding domain fusions (pPP1309, pB20N2, and pBTM116) in combination with the vector (pGADXP) or activation domain fusions to wild-type (WT) and mutant derivatives of Ste20 residues 1 to 439 (pPP1037, pPP1321, pPP1322, and pPP1323). The kinase domain fragment (580 to 939) bears a kinase-inactivating mutation, K649M, in order to prevent toxicity. Values are means ± SD (n = 5).

FIG. 4.

Ste20 kinase activity and stimulation by Cdc42-GTP. (A) Hyperactive kinase activity displayed by Cdc42-independent Ste20 mutants. Strain PPY913 was transformed with the indicated GFP-STE20 plasmids or pRS316 (vector). Ste20 mutant proteins were immunoprecipitated with anti-GFP antibodies and tested for kinase activity by using myelin basic protein (MBP) as the substrate. Portions of the kinase reactions were also analyzed by anti-GFP immunoblot to determine protein levels. Bottom, quantification of two independent experiments (mean ± range), normalized in each experiment to the level of [³²P]MBP in the wild-type (WT) sample (= 1). (B) Cdc42-GTP stimulates kinase activity of wild-type but not mutant Ste20 derivatives. Ste20 kinase activity was assayed using strains PPY1234, PPY1238, and PPY1236, expressing Myc₁₂-tagged wild-type, S338A/H345G, and Δ334-369 derivatives of Ste20, respectively. The strains harbored a vector (pRS413) or a galactose-inducible CDC42^Q61L construct (pH-G42-L61), as denoted by − and + GAL-CDC42^Q61L, respectively. Transformants were induced with 2% galactose for 90 min in the absence (−) or presence (+) of 10 μM α-factor as indicated. The Myc₁₂-tagged proteins were immunoprecipitated, assayed for kinase activity, and tested for Ste20-Myc protein levels, as for panel A. Bottom, quantification of two independent experiments as in panel A.

FIG. 5.

Additional contribution of Cdc42 to Ste20 function, separable from kinase activity. (A) Hyperactive kinase mutants are toxic when overexpressed. Strain PPY398 was transformed with pPP1219, pDH166, pPP1248, pPP1272, pPP1274, or pPP1273, and then fivefold serial dilutions were spotted onto −His/glucose or −His/raffinose/galactose plates, as indicated, and incubated for 3 days at 30°C. (B) Loss of Cdc42-binding, not deregulated Ste20 activity, disrupts the Cla4-redundant essential function of Ste20. Strain KBY211 (ste20Δ cla4Δ YCp-TRP1-cla4-75^ts) was transformed with the indicated GFP-STE20 plasmids, and then fivefold serial dilutions were spotted onto −Ura plates and incubated for 3 days at 25 or 37°C, as indicated. (C) Loss of Cdc42 binding decreases Ste20 signaling activity even when kinase activity is already made Cdc42 independent by the L369G mutation. Strain PPY866 harbored the indicated GFP-STE20 plasmids plus either pH-GS5-CTM (Ste5-CTM) or pH-SL2 (Ste5^P44L-GST). Bars show FUS1-lacZ induction (mean ± SD, n = 3) after galactose-induced synthesis of the indicated Ste5 derivative.

FIG. 6.

Effects of Ste20 CRIB mutants on the filamentation pathway. (A) Agar invasion in haploid strains PPY966, PPY1209, PPY1200, and PPY1202. Duplicate patches of each strain were grown on a rich medium (YPD) plate, which is shown before and after being rinsed under a stream of water. Note that the Δ334-369 allele retains agar invasion activity, which differs from previous reports of agar invasion or pseudohyphal growth assays when this allele was tested in plasmid-borne form (27, 45). In contrast, the S338A/H345G allele is strongly defective. These results are representative of multiple assays and were confirmed by using independently derived genomic replacements with the Δ334-369 and S338A/H345G alleles. (B) Transcriptional expression of filamentation reporters. The strains tested in panel A harbored the indicated lacZ reporter constructs (plasmids pPP827, pPP1253, p2988, and p2987). Bars show β-galactosidase units (mean ± SD; n = 6 to 8). Note that the Δ334-369 allele activates the TyFRE and FLO11 reporters to a degree indistinguishable from the wild-type (WT) level but is detectably reduced for the other two reporters. (C) Effects of plasmid-borne alleles on filamentation reporter expression. β-Galactosidase activity was measured in strain PPY1209 harboring the indicated lacZ reporter (pPP827, pPP1252, or p2988) plus plasmids expressing native Ste20 derivatives as follows, from top to bottom: pPP1001, pPP1011, pPP1005, pPP1112, and pRS316. Bars show β-galactosidase units (mean + SD, n = 9). The flo11(10/9)-lacZ reporter contains a 440-bp segment of the 3-kb promoter present in the full-length FLO11pr-lacZ construct (53); it is included here to illustrate the range of transcriptional effects caused by removal of the Ste20 CRIB domain, from no effect (TyFRE and FLO11), to mild effect (flo11[10/9]), to moderate effect (YLR042C and KSS1).

FIG. 7.

General model for the production and utilization of activated Ste20. Based on results in this study, the model proposes that GTP-bound Cdc42 (presumably activated by its exchange factor, Cdc24) converts Ste20 to an active form (asterisk) by antagonizing the autoinhibitory influence of CRIB domain sequences in the Ste20 N terminus. As discussed in the text, this conversion is required for pheromone-dependent signaling, but there is no evidence that mating pheromones stimulate this conversion, and instead Cdc42-activated Ste20 is competent to signal even without prior exposure to pheromone. Therefore, pheromone-mediated activation of the Gβγ complex may promote signaling through the mating pathway by an existing pool of activated Ste20 kinase. The lifetime of activated Ste20 is unknown, as is the degree to which Ste20 activity requires continual interaction with Cdc42; evidence from other systems suggests that PAK autophosphorylation stabilizes the open conformation, allowing kinase activity to persist after dissociation of the GTPase (10, 66). The pathway leading to filamentous growth also requires activation of Ste20 by Cdc42. The question mark implies that while nutritional cues can trigger filamentous growth, they do not necessarily stimulate the Ste20-dependent filamentation MAP kinase cascade, as discussed previously (35); instead, nutritional cues may primarily modulate a separate (protein kinase A) pathway that is required in parallel for filamentous growth (12, 32). Dashed-line arrows denote that Ste20 also functions in other cellular processes (e.g., regulation of actin, cell polarity, and osmoregulatory signaling); these likely also depend on activation by Cdc42, but little is known about Ste20 regulation during these events.