Genetic Analysis of the Interface Between Cdc42p and the CRIB Domain of Ste20p in Saccharomyces cerevisiae

Corresponding author: Malcolm Whiteway, National Research Council, Biotechnology Research Institute, 6100 Royalmount Ave., Montreal, Quebec H4P 2R2, Canada., malcolm.whiteway{at}nrc.ca (E-mail)

Communicating editor: B. J. ANDREWS

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Mutagenesis was used to probe the interface between the small GTPase Cdc42p and the CRIB domain motif of Ste20p. Members of a cluster of hydrophobic residues of Cdc42p were changed to alanine and/or arginine. The interaction of the wild-type and mutant proteins was measured using the two-hybrid assay; many, but not all, changes reduced interaction between Cdc42p and the target CRIB domain. Mutations in conserved residues in the CRIB domain were also tested for their importance in the association with Cdc42p. Two conserved CRIB domain histidines were changed to aspartic acid. These mutants reduced mating, as well as responsiveness to pheromone-induced gene expression and cell cycle arrest, but did not reduce in vitro the kinase activity of Ste20p. GFP-tagged mutant proteins were unable to localize to sites of polarized growth. In addition, these point mutants were synthetically lethal with disruption of CLA4 and blocked the Ste20p-Cdc42p two-hybrid interaction. Compensatory mutations in Cdc42p that reestablished the two-hybrid association with the mutant Ste20p CRIB domain baits were identified. These mutations improved the pheromone responsiveness of cells containing the CRIB mutations, but did not rescue the lethality associated with the CRIB mutant CLA4 deletion interaction. These results suggest that the Ste20p-Cdc42p interaction plays a direct role in Ste20p kinase function and that this interaction is required for efficient activity of the pheromone response pathway.

A member of the small GTPase superfamily, Cdc42p (BOURNE et al. 1990 ), has been implicated in control of polarized growth in a variety of cell types (CHANT 1996 ; BISHOP and HALL 2000 ). Cdc42p interacts with a variety of downstream effector molecules, such as PAR6 (LIN et al. 2000 ), WASP (SYMONS et al. 1996 ), and Bni1p (EVANGELISTA et al. 1997 ). Many of these effectors interact with Cdc42p through a conserved motif, known as the CRIB or GBD domain (BURBELO et al. 1995 ). Among the Cdc42p targets are members of the STE20/PAK kinase family, a class of kinases identified in mammalian cells because of their ability to bind GTPases of the Cdc42p and Rac families (MANSER et al. 1994 ). In mammalian cells the binding of the CRIB domain region of the PAK kinases activates kinase activity (MANSER et al. 1994 ). The crystal structure of PAK suggests that this activation requires a structural rearrangement of the kinase molecule (LEI et al. 2000 ).

Cdc42p function has been extensively studied in the yeast Saccharomyces cerevisiae, where there are three members of the STE20/PAK kinase family: Ste20p, the founding member (LEBERER et al. 1992 ); Cla4p (CVRCKOVA et al. 1995 ); and Skm1p (MARTIN et al. 1997 ). Ste20p functions in a variety of signaling pathways in the yeast cell. These pathways include the pheromone response pathway (LEBERER et al. 1992 ), the pseudohyphal growth pathway (LEBERER et al. 1997 ), and the osmotic response or HOG pathway (RAITT et al. 2000 ). In addition, Ste20p and Cla4p have been shown to phosphorylate and activate members of the myosin I subfamily of motor proteins in vitro (WU et al. 1996 ). Ste20p and Cla4p together provide an essential function, as double mutants are not viable (CVRCKOVA and NASMYTH 1993 ). Cells that are defective in both functions show loss of polarized growth and disruptions in cortical actin localization (EBY et al. 1998 ; HOLLY and BLUMER 1999 ). Although loss of myosin I function in yeast can be lethal, myosin I activation is not the only essential function of Ste20p and Cla4p, as replacement of the myosin I phosphorylation site with an aspartate residue does not rescue the lethal phenotype of the Cla4 Ste20 double mutant (WU et al. 1997 ).

The role of Cdc42p in activation of the pheromone response pathway has been controversial. Initial evidence suggested a direct role because Cdc42p was required for pheromone induction of gene expression (SIMON et al. 1995 ). However, deletion of the CRIB domain of Ste20p did not affect mating (PETER et al. 1996 ; LEBERER et al. 1997 ), and the apparent requirement for Cdc42p function in pheromone-induced gene expression could be accounted for by the fact that the cells were trapped at G₁ (OEHLEN and CROSS 1998 ). These results for the CRIB deletion showed that localization of Ste20p to the bud sites was dependent on Cdc42p binding, but that mating-induced localization was Cdc42p independent (PETER et al. 1996 ; LEBERER et al. 1997 ). However, more recent evidence suggests that Cdc42p does play a role in the pheromone response pathway (MOSKOW et al. 2000 ). In the current study we have analyzed specific point mutants that influence the interaction between Cdc42p and the CRIB domain of Ste20p. The analysis of these mutants shows that the association of Cdc42p with Ste20p plays a role in the mating process. It appears that cellular localization of Ste20p controlled by Cdc42p is required for efficient mating. This role of the Ste20p CRIB domain residues in the mating process has been independently shown in a recent publication (LAMSON et al. 2002 ).

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

DNA purification:
Total yeast DNA was isolated as described (PHILIPPSEN et al. 1991 ). Plasmid DNA was isolated from Escherichia coli using a Biorobot 9600 (QIAGEN, Valencia, CA) following the manufacturer's instructions.

DNA sequencing:
The concentration of the plasmid DNA to be sequenced was determined by fluorimetry using the DyNA Quant 220 (Hoefer, San Francisco) following the manufacturer's protocol. Sequencing was done with the dRhodamine kit (Perkin-Elmer, Norwalk, CT) and was analyzed on a 377 XL DNA sequencer (Applied Biosystems, Foster City, CA).

PCR:
PCR was done using the standard protocol from the Expand High Fidelity PCR kit (Roche, Indianapolis).

Oligonucleotides:
The following oligonucleotides were used:

• KOLI22: CCACTACAATGGATGATG

• ORL19: TTTGTATACACTTATTTTTTTTATAAC

• ORL72: CTACAAAATaGtACtTTTTTTACTTTTCTTG, ScaI

• ORL71: CAAGAAAAGTAAAAAaGTaCtATTTTGTAG, ScaI

• ORL59: CAATGCCAAGgATATCCACCATG, EcoRV

• ORL60: CATGGTGGATATcCTTGGCATTG, EcoRV

• ORL61: GCATATCCACgAcGTcGGCGTGGACTCC, AatII

• ORL62: GGAGTCCACGCCgACgTcGTGGATATGC, AatII

• KOLI42: GATCCTCGACTAAAGATCTTAATGAGCAATGATCC, BglII

• OAF3: CTTCCAATGCATTGGCTGCAGCTAAGAATTCATGACGGGTGATTT, PstI

• OJA33: CCATATGCAAACGCgtAAGTGTGTTGTTGTCG, MluI

• OJA34: CGACAACAACACACTTacGCGTTTGCATATGG, MluI

• OJA37: CTATGCGGTGACgcgtATGATTGGTGATGAACC, MluI

• OJA38: GGTTCATCACCAATCATAcgcGTCACCGCATAG, MluI

• OJA41: GGTGATGAACCAcgTACGTTAGGTTTG, BsiWI

• OJA42: CAAACCTAACGTAcgTGGTTCATCACC, BsiWI

• OJA43: GGTGATGAGgCCggcTACGTTAGGTTTG, NaeI

• OJA44: CAAACCTAACGTAgccGGcTCATCACC, NaeI

• OJA45: CGATGAAGCTcgaGTGGCCGCCTTG, XhoI

• OJA46: CAAGGCGGCCACtcgAGCTTCATCG, XhoI

• OJA47: CGATGAAGCTgcaGTGGCCGCCTTG, PstI

• OJA48: CAAGGCGGCCACtgcAGCTTCATCG, PstI.

The specific restriction sites noted are underlined, and the nucleotides changed for the mutagenesis are in lowercase letters.

Two-hybrid plasmids and the construction of site-directed mutants in CDC42:
Plasmid pVL7 consists of the entire coding sequence of KSS1 (COURCHESNE et al. 1989 ) inserted into pBTM116 as an EcoRI to SalI fragment. Plasmid pRL39 (LEBERER et al. 1997 ) consists of full-length CDC42 from S. cerevisiae cloned in frame in the two-hybrid activation vector pGAD424 (CHIEN et al. 1991 ). Cysteine 188 of pRL39 was changed to serine by PCR-mediated site-directed mutagenesis. The mutagenic strategy employed involved complementary oligonucleotides containing the desired nucleotide substitution and the use of outside oligonucleotides to direct production of the rest of the gene and flanking sequences (HIGUCHI et al. 1988 ). In all site-directed modifications of CDC42 the flanking oligonucleotides were KOLI22 and ORL19, which encode sequences from the ADH1 promoter and ADH1 terminator regions of pGAD424, respectively. Oligonucleotides ORL72 (coupled with KOLI22 for the first cycles) and ORL71 (coupled with ORL19 for the first cycles), together with the flanking oligonucleotides, were used to generate a DNA fragment with cysteine 188 mutated to a serine. This PCR product was transformed into yeast strain L40a (MARCUS et al. 1994 ) together with plasmid pGAD424 linearized with BamHI and EcoRI, and recombinant plasmids were generated by in vivo recombination. Recombinant plasmids were isolated as total yeast DNA, transformed into E. coli, purified, tested for the introduction of the ScaI site engineered along with the serine substitution, and then sequenced, and a plasmid with the correct substitution was designated pRL202. Subsequent mutations of specific hydrophobic residues of CDC42^C188S were performed similarly, except that the starting PCR template was pRL202 rather than pRL39. The oligonucleotides for generating L4R were OJA33 and OJA34; for V44R, OJA37 and OAJ38; for Y51R, OJA41 and OAJ42; for Y51A, OJA43 and OAJ44; for I173R, OJA45 and OAJ46; and for I173A, OJA47 and OAJ48.

Construction of CRIB domain mutants:
Plasmid pAF3 (LEBERER et al. 1997 ) contains the N terminus of STE20 encompassing the CRIB domain cloned as an EcoRI to SalI fragment in the LexA DNA-binding domain plasmid pBTM116 (CHIEN et al. 1991 ). Derivatives containing the H345D and H348D alleles were created by site-directed PCR-mediated mutagenesis. PCR products containing the mutations were created by a first-round synthesis using primers ORL60 plus KOLI42 and ORL59 plus OAF3 for H345D and, for H348D, ORL62 plus KOLI42 and ORL61 plus OAF3. This amplification was followed by second-round synthesis using the outside primers KOLI42 and OAF3 for both constructs. The template plasmid in each case was pDH37 (LEBERER et al. 1997 ). These PCR products were cut with BglII and SalI and cloned into BamHI- and SalI-cut pRL99, and candidate H345D and H348D substitutions were identified through the coordinate introduction of restriction sites (EcoRV for H345D and AatII for H348D). Selected plasmids were sequenced; a correct plasmid with the H345D mutation was named pRL155, and a correct plasmid with the H348D substitution was named pRL156.

Construction of CRIB domain mutant-GFP fusions:
Plasmid pRL116 (LEBERER et al. 1997 ) contains a functional Ste20-green fluorescent protein (GFP) fusion protein that localizes to new buds and to the tips of growing buds during polarized bud growth (LEBERER et al. 1997 ; WU et al. 1998 ). The H345D and H348D substitutions within the CRIB domain of Ste20p were introduced by replacing the BamHI to SalI fragment containing the CRIB domain of Ste20 with the equivalent fragment from pRL155 and pRL156 to create plasmids pJA82 and pJA83, respectively.

Construction of random mutants of CDC42:
Plasmid pRL202 was used as a template for Taq polymerase-mediated PCR using oligonucleotides KOLI22 and ORL19. The PCR products were gel purified and cloned into EcoRI- and BamHI-linearized pGAD424 by in vivo recombination to generate a library of CDC42 genes mutated at the frequency characteristic of Taq polymerase nucleotide misincorporation. The library was generated in strain L40 already transformed with the pBTM116 derivative plasmid pRL155, and LEU2⁺ TRP1⁺ transformants were selected. These transformants were screened by replica plating to identify colonies that generated a plasmid-based growth on medium lacking histidine and containing 5 mM 3-aminotriazole (3-AT).

Separation of double mutants:
Plasmid pMJ13 contained two mutations, N39S and P69T. These were separated by digesting pMJ13 with EagI and SnaBI, which generates a 1.65-kb fragment containing the N39S mutation and a 5.6-kb fragment with the P69T mutation. An equivalent digestion of pRL202 was performed, and the fragment pairs were ligated together to generate plasmids with the separated mutations. Plasmids containing the independent mutations were sequenced, and a plasmid with N39S was designated pJA49, and P69T was designated pJA50.

Transfer of N39S and P69T mutations:
The N39S and P69T mutations were transferred to the wild-type CDC42 gene. In plasmid pRS315 CDC42 (ADAMS et al. 1990 ), a kind gift of D. Johnson (University of Vermont), the NotI (EagI) site of the multiple cloning site was destroyed. The resulting plasmid was then digested with EagI to cleave the CDC42 gene between amino acids 39 and 69 (at position 58). Protruding extremities were digested with mung bean nuclease and this linear plasmid was cotransformed with a PCR fragment of the mutant (N39S, P69T) CDC42 gene. Following in vivo recombination, plasmids were extracted and sequenced. A double mutant (N39S, P69T) was designated pJA55, while the single amino acid changes were called pJA53 (N39S) and pJA54 (P69T).

Strain constructions:
Strains are listed in Table 1. Strains JAY25 and JAY26 were derived from strain W303-1A by replacing, respectively, the histidines 345 and 348 of STE20 with aspartic acid. SalI to BamHI fragments of 1.1 kb containing the mutated region of STE20 from pRL155 and pRL156 were subcloned into pRS306 cut with SalI and BamHI to create plasmids pJA45 and pJA46. Plasmids pJA45 (H345D) and pJA46 (H348D) were targeted to the STE20 locus by cleavage with ClaI, leading to duplication of the STE20 locus flanking the vector and URA3 marker. Loop-outs of the URA3 marker were identified by growth on medium with 5-fluoroorotic acid (5-FOA), and strains with the histidines replaced with asparagines were detected as showing reduced mating in 4-hr mating tests. These replacement alleles were confirmed by digesting PCR products generated from strains JAY25 and JAY26 using oligonucleotides ORL1 and ODH77 with the diagnostic restriction enzymes AatII for JAY25 and EcoRV for JAY26. Subsequently, strains JAY25 and JAY26, together with W303-1A, were transformed to sst1::URA3 using plasmid pJGsst1 (RENEKE et al. 1988 ) cut with SalI and EcoRI. These strains were then grown on 5-FOA medium to select loss of the URA3 marker and then transformed to fus1::LacZ using plasmid pDH17 cut with HindIII. The W303-1A derivative containing the sst1 disruption and the fus1::lacZ reporter was named JAY54, while the JAY25 and JAY26 derivatives were JAY39 and JAY40, respectively.

ß-Galactosidase assays:
For the assay with strains JAY39, JAY40, and JAY54, induction with 0.4 µg/ml of -mating factor was done for 2 hr prior to the ß-gal assays. Standard conditions were used for ß-gal assays (JARVIS et al. 1988 ); the values shown are Miller units based on the following formula: units = 1000 x OD₄₂₀/T x V x OD₆₀₀, where T is the reaction time (in minutes) and V is the volume of culture used for the test (in milliliters).

Quantitative mating:
Quantitative matings were done on filter discs as described (LEBERER et al. 1993 ).

Kinase assays:
In vitro kinase assays were performed essentially as described (WU et al. 1995 ), except that the washed immune complexes were incubated at 30° for 15 min with 30 µl kinase buffer containing 100 µM ATP and then washed three times with kinase buffer, 1 ml each, without ATP. The kinase reactions with labeled ATP were started by addition of 30 µl of kinase buffer containing 10 µM ATP and 1 µl of [-³²P]ATP (10 µCi/µl at 6000 Ci/mmol) and 2 µg of myelin basic protein (MBP), and the reactions were incubated at 30° for 15 min. One set of mock reactions processed without addition of [-³²P]ATP was used for Western blot analysis to reveal the amount of Ste20p in the immune complexes.

Preparation of total yeast cell extracts:
Overnight cultures were diluted 1:20 and grown in selective media to an A₆₀₀ of 1.0. A total of 1 ml of these cultures was harvested by centrifugation at 3000 x g and resuspended in 100 µl of loading buffer. Cells were then disrupted by boiling for 5 min. The extracts were clarified by centrifugation in a microcentrifuge at top speed for 10 min at room temperature. Twenty microliters of the supernatants was loaded on 12% SDS-PAGE gels.

Western blot analysis:
Protein samples were resolved on 12% SDS-PAGE and transferred on Immobilon-P membrane (Millipore, Bedford, MA). The membranes were probed first with affinity-purified anti-CDC42 antiserum, incubated with anti-rabbit HRP (Santa-Cruz), and developed using Lumilight Plus chemiluminescent substrate (Roche). The anti-Cdc42p antibody was raised in rabbits against purified bacterially expressed glutathione S-transferase (GST)-Cdc42p. The antibody was affinity purified against GST-Cdc42p immobilized on nitrocellulose membranes essentially as described (TALIAN et al. 1983 ).

Fluorescence microscopy:
Cells of strain YCW563 containing the plasmids pRL116C.4, pJA82, pJA83, pRS315:CDC42, and pJA55 were grown overnight in selective medium. GFP was visualized with a Leica DMIRE2 inverted microscope (Leica Microsystemes, Montreal) equipped with a Hamamatsu cooled CCD camera and a Ludl motorized stage at x630 magnification. Openlab software (Improvision, Lexington, MA) and Adobe Photoshop were used for image acquisition and manipulation. Pheromone treated cells were incubated in the presence of 74 µg/ml -factor for 3 hr to trigger shmoo formation.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Recent structural studies (GUO et al. 1998 ; STEVENS et al. 1999 ; LEI et al. 2000 ; MORREALE et al. 2000 ) have provided a framework for the interface between members of the Cdc42p class of proteins and the CRIB domain of members of the Ste20/Pak kinase family. These studies have implicated specific residues as likely playing important roles in directing the association of these proteins. We have used classical yeast genetics coupled with the powerful two-hybrid protein association assay to investigate aspects of the interaction between Cdc42p and Ste20p.

Hydrophobic residues of Cdc42p are involved in association with Ste20p:
NMR analyses have identified residues of Cdc42p whose resonances are modified through an interaction with CRIB domain peptides (GUO et al. 1998 ; STEVENS et al. 1999 ; MORREALE et al. 2000 ). These residues are likely involved, either directly or indirectly, with the association of the GTPase and its CRIB domain containing target protein. Several of these residues are nonpolar, and a set of these residues forms a hydrophobic cluster within the Cdc42 protein. We used the two-hybrid protein association assay to test whether the residues of Cdc42p forming this hydrophobic cluster are involved in the interaction with the CRIB domain region of Ste20p. Mutations L4R, V44R, Y51A, Y51R, I173A, and I173R were created by PCR-mediated site-directed mutagenesis as described in MATERIALS AND METHODS. These mutant constructs were introduced into plasmid pRL202, which contains a C188S mutant version of Cdc42p fused in frame to the LexA DNA-binding domain of plasmid pBTM116. As shown in Fig 1, the protein levels generated from the various Cdc42p two-hybrid plasmids varied considerably. In particular, the I173A protein was expressed dramatically less than the wild-type protein or the other mutants. The V44R substitution protein was expressed as well as wild type, while the L4R, Y51A, Y51R, and I173R proteins were reduced relative to wild type. These mutant constructs were compared with the starting fusion protein in pRL202 for the level of activation of the reporter constructs HIS3 and LacZ. Although the V44R protein was well expressed, the substitution dropped the two-hybrid association measurement to close to background levels, suggesting that V44 plays an important role in Cdc42p association with the CRIB domain. In addition, the L4R and Y51R substitution proteins were expressed as well as the I173R mutant protein, but had reduced two-hybrid interactions (Fig 2). The very low expression of the I173A protein precluded interpretation of the reduced two-hybrid interaction; however, the lack of an influence of the I173R modification may suggest that the I173 residue is not critical for the two-hybrid association.

View larger version (20K): [in this window] [in a new window] [Download PPT slide]   Figure 1. Western blot of total protein extract detected with Cdc42p antibodies. The lower band represents the endogenous Cdc42p and the higher band is Cdc42p fused to the Gal4 activation domain. The endogenous protein serves as an internal control for sample loading, which shows that the amounts of the fusion proteins containing the amino acid substitutions are quite variable. TheV44R substitution protein is as abundant as the wild-type protein, while the I173A fusion protein is almost undetectable. The lane labeled L40a is the two-hybrid analysis strain with no plasmids, while the lane labeled pAF3 represents L40 transformed with the STE20 CRIB domain plasmid fused to the LexA DNA-binding domain. The lane labeled CDC42 contains endogenous Cdc42p as well as Cdc42p fused to the Gal4 activation domain in plasmid pGAD424.

View larger version (18K): [in this window] [in a new window] [Download PPT slide]   Figure 2. Two-hybrid interactions between Ste20p and Cdc42p. Interactions were determined between the STE20 gene (binding domain, BD) and the CDC42 gene (activation domain, AD) as described in MATERIALS AND METHODS. The ß-galactosidase determinations are the average of three independent samples. (Right) ß-Galactosidase readings with standard deviation error bars.

Conserved residues in the Ste20p CRIB domain are necessary for protein function:
An alignment of CRIB domains from several proteins has defined a core consensus region of 12 amino acids (BURBELO et al. 1995 ). Within this conserved region are several invariant amino acids whose conservation in Ste20/Pak kinases from yeast to humans suggests that they play important functional roles. We used site-directed mutagenesis to test the importance of a pair of closely located invariant histidine residues found near the center of the CRIB domain. These histidine residues were modified to aspartates by oligonucleotide-directed PCR mutagenesis as described in MATERIALS AND METHODS. The two mutant alleles, H348D and H345D, were both introduced into the genome of strain W303-1A to replace the wild-type STE20 gene to create strains JAY25 and JAY26 (Table 1).

Previous work had shown that the CRIB domain of the Ste20p kinase was essential for the Ste20p kinase to support cellular growth in the absence of Cla4p (PETER et al. 1996 ; LEBERER et al. 1997 ). An identical result was obtained when either the H345D or the H348D allele was crossed to a CLA4 deletion strain. Strains JAY26 and JAY25 were crossed to W303-1B cla4::TRP1 strain EL252-1B to generate diploids DM98001 and DM98014 (Table 1). These diploids were sporulated and dissected and the pattern of viable spores was observed. In 18 tetrads from DM98001 and 9 tetrads from DM98014 one-half of the expected cla4::TRP1 spores were missing. This pattern of inviability was that expected for the independent segregation of a mutation that would render the cla4::TRP1 cells inviable. Evidence that this mutation was either ste20^H345D or ste20^H348D was obtained by introducing plasmid pVT-STE20 into the strains DM98001 and DM98014 prior to dissection. Several Trp⁺ spores containing the URA3 plasmid with STE20 were unable to grow on 5-FOA-containing medium, as expected for strains in which the STE20 plasmid covered a lethal mutation. This establishes that a single point mutation in the CRIB domain of STE20 can render the kinase incapable of complementing the loss of CLA4 function.

Strains containing the H345D and H348D alleles were assayed for the effects of the mutations on Ste20p function in the pheromone response pathway. In contrast to the complete deletion of the CRIB domain, which had essentially no effect on the response of cells to mating (PETER et al. 1996 ; LEBERER et al. 1997 ), the H345D and H348D alleles caused detectable defects in mating. Strains JAY39, JAY40, and JAY54 (Table 1) were tested for pheromone-induced fus1::LacZ induction and for mating. As shown in Fig 3 (rows 1, 6, and 11), the H345D and H348D alleles reduced mating to approximately one-tenth the level of the wild-type strain in the absence of any modification to Cdc42p. Fus1::LacZ induction was similarly compromised by the H345D and H348D mutations. After a 2-hr induction with -factor, strains JAY39 and JAY40 generated 13 and 11 Miller units of activity, respectively, while the wild-type control strain JAY54 generated 120 Miller units under the same conditions. Thus the CRIB domain mutations also reduced pheromone-responsive gene induction 10-fold (Fig 4, rows 1, 6, and 11).

View larger version (32K): [in this window] [in a new window] [Download PPT slide]   Figure 3. Matings involving STE20 wild-type and mutant strains were done as described in MATERIALS AND METHODS. The strains contained pRS315-based plasmids that were either the control vector or recombinant plasmids containing a version of CDC42. The matings were the average of three independent measurements. The bars on the right are a graphical representation of the relative mating levels; the absolute levels are low due to the effects of the sst1 and fus1 mutations. (*) Values represent the number of diploids formed for each 100 in-put haploid cells.

View larger version (35K): [in this window] [in a new window] [Download PPT slide]   Figure 4. Pheromone-induced expression of the fus1::LacZ constructs in STE20 wild-type and mutant strains. ß-Gal expression was measured as described in MATERIALS AND METHODS. The strains contained pRS315-based plasmids: either the control vector or recombinant plasmids containing a version of CDC42. The ß-gal values were the average of three independent measurements; the bars on the right represent the ß-gal values graphically.

We determined the catalytic activity of the Ste20p kinase containing the modified histidines. The Ste20p kinase was immunoprecipitated from strains W303-1A, JAY25, and JAY26 as described (WU et al. 1995 ). The immunoprecipitated protein was assayed for autophosphorylation and for activity on MBP. As shown in Fig 5, the Ste20p immune complexes had a strong in vitro kinase activity on MBP, whereas the immune complexes containing the catalytically inactive Ste20pK649R did not phosphorylate MBP. The Ste20p mutants from JAY25 and JAY26 had intracellular concentrations similar to the wild-type Ste20p as judged by the Western blot (Fig 5A) and also showed similar kinase activity on the phosphorylation of MBP as judged by autoradiography (Fig 5B, bottom). The residual amount of ³²P labeling of Ste20p by autophosphorylation after the immune complexes were incubated with high concentration of nonlabeled ATP was also very similar among the mutants and the wild-type Ste20p. These results indicated that the Ste20p mutants and the wild-type Ste20p immune complexes prepared from total cell extracts had very similar in vitro kinase activities.

View larger version (45K): [in this window] [in a new window] [Download PPT slide]   Figure 5. In vitro kinase assay of Ste20p. Ste20p was isolated as immune complexes from yeast strains W303-1A (Ste20WT), JA25 (Ste20H345D), JA26 (Ste20H348D), and W303-1A (ste20K649R) expressing catalytically inactive Ste20p. The immune complexes were assayed for kinase activity using MBP as a substrate as described (WU et al. 1995 ). (A) The relative amount of Ste20p in each assay detected by Western blotting. (B) MBP phosphorylation (bottom) and Ste20p autophosphorylation (top).

The loss of biological function in the absence of modification to the kinase activity suggested that some other aspect of Cdc42p function was compromised. We constructed GFP fusions to the CRIB domain point mutants of Ste20p to analyze cellular localization. Previous work had established that wild-type Ste20p localized to sites of polarized growth and was concentrated in small buds and at the tips of larger buds (PETER et al. 1996 ; LEBERER et al. 1997 ; WU et al. 1998 ). The single point mutants of the CRIB domain prevented this localization, and the GFP signal was observed throughout the cells with no evidence for localization to the buds or bud tips (Fig 6). This confirms the result of the previous localization studies undertaken with CRIB deletion mutants (PETER et al. 1996 ; LEBERER et al. 1997 ). We also examined the localization of the Ste20p point mutants during the mating process. In contrast to the deletion mutants of the Ste20p CRIB domain (PETER et al. 1996 ; LEBERER et al. 1997 ), the Ste20-GFP fusion protein containing the histidine mutation failed to localize to the shmoo tip (Fig 6).

View larger version (44K): [in this window] [in a new window] [Download PPT slide]   Figure 6. GFP:Ste20p fusion protein localization. Yeast strains expressing the wild type (a and d), the H345D (b and e), or the H348D (c and f) CRIB domain mutant GFP:Ste20p fusion proteins were analyzed by fluorescence microscopy to determine the cellular localization of the GFP protein. The wild-type protein localizes to bud (a) or shmoo (d) tips. The CRIB domain mutant proteins fail to localize to either bud (b and c) or shmoo (e and f) tips and show instead a generalized cytoplasmic staining.

Conserved residues in the Ste20p CRIB domain are necessary for association with Cdc42p:
Both histidine substitution mutants dramatically affected the Ste20p-Cdc42p interaction as measured by the two-hybrid assay. Two-hybrid DNA-binding domain chimeras of the STE20 N terminus containing the wild-type sequence (pAF3), the H345D allele (pRL155), or the H348 allele (pRL156) were tested with the Cdc42p activation domain chimera pRL202. As shown in Fig 7, both the H345D and the H348D mutations reduced the two-hybrid association signal to background levels, showing that the residues are involved in the interaction between the two proteins.

View larger version (34K): [in this window] [in a new window] [Download PPT slide]   Figure 7. Two-hybrid assay for the association of Cdc42 and the Ste20 CRIB domain. CAAX box mutant Cdc42p constructs fused to the Gal4 activation domain (pGAD424 derivatives) were combined with pBTM116 (DNA-binding domain) versions of either the N terminus of Ste20p or mutant derivatives or with the Kss1p kinase. Values represent the average of three to six independent assays.

The two-hybrid interaction assay was used to identify mutant versions of Cdc42p that could reestablish protein-protein interaction with a bait plasmid containing the H345D version of the Ste20p CRIB domain. Plasmid pRL202 was subjected to random PCR mutagenesis as described in MATERIALS AND METHODS. The mutagenized PCR product was cotransformed with BamHI- and SalI-linearized plasmid pGAD424 (CHIEN et al. 1991 ) in strain L40 containing the bait plasmid pRL155 to create a library of mutagenized Cdc42p constructs. Cells were selected on -his + 5 mM 3-AT to detect significant two-hybrid interactions between the mutant Cdc42p and the H345D version of the Ste20p CRIB domain, as this concentration of 3-AT was the minimum that was capable of eliminating all background growth of the pRL155/pRL202 combination. One mutant derivative of pRL202 was identified in 2000 transformants screened that reproducibly provided plasmid-dependent HIS3⁺ activity in strain L40 containing pRL155. Sequencing this plasmid (pMJ13) established that it contained two nucleotide substitutions (A234 to C234 and A567 to C567) that resulted in two amino acid substitutions in Cdc42p, N39S, and P69T. We tested whether both substitutions were necessary for the reestablishment of the two-hybrid interaction with pRL155 by constructing derivatives of pRL202 with the two single mutations. Neither plasmid pJA49, containing the N39S substitution, nor plasmid pJA50, containing the P69T substitution, gave a strong two-hybrid interaction with pRL155. We also tested the specificity of the double mutant. pMJ13 had a normal two-hybrid interaction with pAF3, showing that the amino acid substitutions did not block interaction between the mutant Cdc42 protein and the wild-type CRIB domain. In addition, the double mutant was capable of interacting with pRL156, which contains the H348D substitution. However, the protein was still specific for interaction with the Ste20 N terminus; the mutant Cdc42p construct did not show any interaction with pVL7, which contains the Kss1p kinase (COURCHESNE et al. 1989 ) as negative control (Fig 7).

We investigated whether other combinations of mutations could reestablish the two-hybrid interaction with pRL155 containing the H435D substitution. Plasmids pJA49 and pJA50 were used as templates for a round of PCR mutagenesis as described. Mutant plasmids that conferred growth of L40 transformed with pRL155 on plates with 5 mM 3-AT were identified and sequenced (Table 2). When pJA49 (N39S) was used as a template, we identified two plasmids in which P69 was mutated to T, regenerating the combination identified in pMJ13. In addition to P69T, we found that G60S, G12S, G12H, D65G, and Y64H mutations in combination with N39S could confer growth in the presence of 5 mM 3-AT. When pJA50 (P69T) was used as the template, we identified three plasmids with the N39S mutation. In addition, we identified the Y64H change that also was found together with the N39S substitution. Finally, we identified one triple-mutant combination, which had both the N39S and the G12S substitutions, together with the template P69T change.

We determined whether the N39S P69T allele of Cdc42p was able to improve the function of Ste20p containing the H345D and H348D substitutions. Strains JAY54, JAY39, and JAY40 were transformed with pRS315 or derivatives of pRS315 plasmids containing wild-type Cdc42p or mutant versions of Cdc42p with the N39S, the P69T, or the double N39S/P69T substitutions. These transformants were tested to determine the consequence of the presence of the mutant CDC42 constructs. As previously noted in Fig 3, in cells with the vector plasmid pRS315, the H345D and H348D substitutions reduced mating 10-fold relative to that of the wild-type strain. These reductions were sensitive to the level of Cdc42p; introduction of an extra copy of wild-type CDC42 in strains JAY39 and JAY40 improved mating 2-fold, although introduction of an extra copy of the CDC42^P69T allele generated no improvement in mating in these strains. Intriguingly, the presence of the CDC42^P69T allele had a detrimental effect on mating in an otherwise wild-type strain, while the other plasmids had essentially no effect on mating levels in JAY54. In both strains JAY39 and JAY40, the presence of the N39S/P69T double mutant raised the mating level to essentially that of the wild-type strain with the double-mutant plasmid.

Similar results were obtained for pheromone-induced expression of the fus1::LacZ. The presence of the various CDC42 alleles had little effect on fus1::LacZ expression in strain JAY54, but the N39S/P69T allele improved induction in both JAY39 and JAY40 (Fig 7). Finally, we tested whether the N39S/P69T allele could rescue the inviability of the cla4::TRP1 allele coupled with either the ste20H345D or the ste20H348D allele. Strains M98001-2A (pVT STE20) and M98014-1A (pVT STE20; Table 1) were transformed with plasmid pSTE20, pRS315CDC42, or pRS315cdc42^N39S/P69T, and the resulting transformants were challenged on 5-FOA medium to determine whether the wild-type STE20 allele was still necessary to support viability. The introduction of the STE20 gene on the HIS3 plasmid allowed the growth of the transformants on 5-FOA medium, whereas the introduction of either the wild-type or the N39S/P69T versions of Cdc42p did not permit growth on 5-FOA plates. This establishes that although the N39S/P69T allele is able to suppress the mating and gene expression defects of the Ste20^H3345D and Ste20^H348D alleles, the modified Cdc42 protein was not capable of rescuing the need for Ste20p CRIB function in the absence of Cla4p (data not shown).

We tested whether the double-mutant version of CDC42 was able to improve the localization of the H345D and H348D versions of Ste20p. The wild-type and double-mutant versions of CDC42 were introduced into strains containing the mutant versions of Ste20GFP, and cellular localization was analyzed. As shown in Fig 8, the presence of the mutant Cdc42p did not dramatically change the cellular mislocalization of Ste20GFP during either normal vegetative growth or shmoo formation. In some cells there is evidence of improved localization of the Ste20GFP signal to sites of polarized growth, but overall the H345D- and H348D-substituted Ste20GFP fusions are poorly localized even when coexpressed with the modified Cdc42p.

View larger version (47K): [in this window] [in a new window] [Download PPT slide]   Figure 8. GFP:Ste20p fusion protein localization controlled by Cdc42p. Yeast strains expressing either the wild type or the H345D CRIB domain mutant GFP:Ste20p fusion proteins were analyzed by fluorescence microscopy to determine the cellular localization of the GFP protein. The wild-type protein localizes to bud (a) or shmoo (d) tips (see also Fig 3). The CRIB domain mutant protein is not localized to sites of polarized growth, and this mislocalization is not corrected by increasing the amount of Cdc42p for either buds (b) or mating projections (e). The N39S, P69T allele of Cdc42p did not dramatically improve the localization to sites of polarized growth in budding cells (c), although some mating-factor-treated cells exhibited weak localization to the shmoo tip (arrows in f).

Localization of the mutant residues on structures (ABDUL-MANAN et al. 1999 ; MOTT et al. 1999 ; GIZACHEW et al. 2000 ; MORREALE et al. 2000 ) of the Cdc42p-CRIB domain (and related GTPase-CRIB interactions) show that the N39S substitution is close to the conserved histidine residues of the CRIB domain (Fig 9). The P69T substitution is not closely associated with the CRIB domain in most of the determined structures, but direct assessment of the actual S. cerevisiae Cdc42p-Ste20p structure will be necessary to confirm this interpretation.

View larger version (67K): [in this window] [in a new window] [Download PPT slide]   Figure 9. Models of CRIB domain interactions. Mutations that enhanced the two-hybrid interactions between Cdc42p and the H245D/H348D Ste20p CRIB domain are mapped onto the representative structures of human Cdc42-CRIB complexes. Cdc42 is represented as a molecular surface and the CRIB fragment as a yellow worm. The location of the Ste20p CRIB mutants H345D/H348D are shown in red. The template mutations of Cdc42p, N39S, P69T that most frequently showed binding are colored green; other residues leading to changes are colored magenta. All four structures show the close proximity of N39 in Cdc42 to the mutated His residues in Ste20p. However, P69 in Cdc42 was consistently distant from the His mutations in Ste20p. Some mutated residues are not visible on all the structures. (A) Cdc42-PAK (MORREALE et al. 2000 ). (B) Cdc42-PAK (GIZACHEW et al. 2000 ). (C) Cdc42-WASP (ABDUL-MANAN et al. 1999 ). (D) Cdc42-ACK (MOTT et al. 1999 ). The figures were produced using GRASP (NICHOLLS et al. 1991 ).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Ste20p is the founding member of the PAK family of protein kinases (LEBERER et al. 1992 ). PAK kinases are defined by the ability to associate with small GTPases of the Rac and Cdc42 families through a binding region termed a CRIB domain. Mammalian PAK kinases require this binding for their activity, and recent studies have defined the structure of the binding interaction and have suggested a mechanism for the role of this association in kinase activation (LEI et al. 2000 ). In this study we have identified point mutants in both the Cdc42p GTPase and the CRIB domain of the Ste20p kinase that influence the association of these proteins and modify their function.

Previous structural studies had implicated a number of residues of the human Cdc42 protein in interactions with peptides synthesized to correspond to the CRIB domain region of PAK (GUO et al. 1998 ; STEVENS et al. 1999 ; LEI et al. 2000 ; MORREALE et al. 2000 ). These results suggest that the binding surface of the GTPase with the CRIB domain involved a more extended surface than that observed for the Ras/Raf association (EMERSON et al. 1995 ) and, in particular, implicated a block of physically adjacent residues of Cdc42p in the interaction (STEVENS et al. 1999 ). We have used the two-hybrid assay to test the role of the residues (Ile4, Val44, Tyr51, and Ile173) forming this hydrophobic block in the association of the yeast Cdc42p and the CRIB domain of Ste20p. Most of the substitutions reduced protein stability as measured by the relative steady-state protein amounts determined by Western blotting, suggesting that these hydrophobic residues have a role in overall protein structure. The most significant destabilization of the Cdc42 fusion protein was caused by the I173A substitution, while the I173R protein showed moderate stability. The I173R protein also showed an essentially wild-type two-hybrid interaction with the Ste20p CRIB domain. This suggests that the position 173 residue is important in overall protein folding, but that the I173A mutation does not have a specific role in the interaction with the PAK family kinase CRIB domain.

However, some of the hydrophobic residues appear directly implicated in the association with the CRIB domain. The V44R substitution did not influence protein stability, but eliminated the two-hybrid interaction with the Ste20p CRIB domain. Previous analysis of a V44A substitution (RICHMAN et al. 1999 ) suggested that this mutation blocked the association of Cdc42p with effectors such as Cla4p, Gic1p, and Gic2p, but did not influence the association with Ste20p. The dramatic consequence of the V44R change in the Ste20p association confirms that position 44 plays an important role in the selectivity of Cdc42p for CRIB motifs. In a manner similar to that of position V44, the L4 position appeared to modulate the CRIB-binding efficacy of Cdc42p. The protein stability was somewhat compromised by this substitution, but the fusion protein generated a weaker two-hybrid signal than that of other mutants of equivalent stability. Thus within the hydrophobic block are residues whose major contribution appears to be to protein stability, while other residues are implicated primarily in the protein association function.

We also examined the importance of residues within the Ste20p CRIB domain for binding to the Cdc42 protein. Two conserved histidine residues at positions 345 and 348 were mutated to aspartic acids. Both mutations blocked the association of the kinase and the GTPase as measured by the two-hybrid assay. This loss of association was correlated with a loss of biological function; when the H345D and H348D substitutions were introduced into the endogenous STE20 gene, neither the ste20^H345D nor the ste20^H348D alleles were able to rescue the deletion of CLA4. This confirms previous results obtained for CRIB domain deletions (PETER et al. 1996 ; LEBERER et al. 1997 ). The defect in Ste20p function caused by mutation of either CRIB domain histidine does not appear to be the result of reduced kinase activity. In vitro assays show that the kinase activity toward myelin basic protein of Ste20p immunoprecipitated from the mutant cells was identical to that of Ste20p immunoprecipitated from wild-type cells. This is perhaps surprising as the structure of the mammalian PAK suggests that conformational changes associated with Cdc42p binding are required for kinase catalytic activity (LEI et al. 2000 ). Either the yeast protein has a less critical requirement for Cdc42p-induced conformational change or the protein isolation conditions are sufficient to activate catalytic activity in the absence of Cdc42p. Recent data (LAMSON et al. 2002 ) suggest that even an "activated" Ste20p allele fails to complement the CLA4 deletion when it lacks Cdc42p-binding capacity.

Even though the kinase activity was normal, the strains containing the ste20^H345D and the ste20^H348D alleles were compromised in STE20 function in the pheromone response pathway. Strains containing the alleles showed reduced mating and reduced ability to induce the fus1::lacZ reporter construct when treated with mating pheromone. These phenotypes appeared to result from the defect in the association of Cdc42p with the Ste20p kinase. Mutations that reestablished the two-hybrid association with the H345D allele of the STE20 CRIB domain with Cdc42p were selected in Cdc42p. This strategy identified a double-mutant CDC42^N39S,P69T that was capable of a more effective association with both the H345D and the H348D CRIB mutations. The double mutant and the two single mutations were transferred to a nonfusion version of Cdc42p to test the consequences of the changes on the function of Cdc42p. The double mutant provided a significant improvement in the mating response of cells containing either the H345D or the H348D mutation. Similarly, the double mutant was able to restore essentially normal pheromone responsiveness to the H345D and H348D mutant strains. Analysis of the single mutants showed that neither was able to reestablish the two-hybrid association on its own. In the biological assays the P69T mutation actually reduced mating and pheromone response relative to the wild-type CDC42, while the N39S substitution caused a moderate improvement.

Although the double mutant was necessary for the enhanced association with the mutant CRIB domain, other double-mutant combinations, including N39S G60S, N39S G12S, and G12H, N39S D65G, N39S Y64H, and Y64H P69T, were capable of enhancing the two-hybrid association between Cdc42p and the Ste20 CRIB domain containing the H345D substitution. Although the initial N39S P69T allele did not influence residues implicated in the GTPase activity of Cdc42p, residues equivalent to G12 and Y64 have been found to reduce the GTPase activity of Cdc42 homologs in other systems (NUR et al. 1992 ; BEST et al. 1996 ; MURRAY and JOHNSON 2001 ). Thus it is possible that these latter changes prolong the time the protein spends in the active, CRIB-binding conformation and that this increase is sufficient to allow detection of the interaction in the two-hybrid assay. Direct assessment of the roles that the mutant residues play in protein structure and function will be necessary to test these models.

The reduced mating and pheromone inducibility of the H345D and H348D substitutions contrasts with what was observed for the deletion of the CRIB domains of Ste20p (PETER et al. 1996 ; LEBERER et al. 1997 ). This is consistent with the removal of the CRIB domain region deleting not only the Cdc42p-binding sequence, but also an inhibitory function. The point mutant constructs appear to influence only the CRIB-binding function and to leave the autoinhibitory function intact. Intriguingly, the reduced mating seen in the point mutants and not in the deletion was correlated with the ability to localize to shmoo tips. GFP versions of the point mutants did not localize to either the bud or the shmoo tips, while the deletion mutant protein was able to associate with the shmoo tip but not with the bud tip (LEBERER et al. 1997 ). It has been speculated that the bud tip localization of Ste20p is controlled by Cdc42p binding, while the association with the shmoo tip may be regulated by the interaction with Ste4p (LEBERER et al. 1997 ; LEEUW et al. 1998 ). We show here that CRIB domain point mutants that prevent Cdc42p interaction also block the shmoo tip localization of Ste20p. This may imply that a conformational change in Ste20p caused by Cdc42p binding is necessary for Gß binding and thus for subsequent localization to sites of polarized growth in mating cells. Alternately, the ability of the CRIB-deleted Ste20p to localize to shmoo tips during mating (PETER et al. 1996 ; LEBERER et al. 1997 ) may be related to the abnormal activity of this protein (LAMSON et al. 2002 ).

The improved function exhibited by the H345D and H348D strains in the presence of the N39S P69T version of Cdc42p was limited to improving mating and pheromone responsiveness. The modified Cdc42p did not rescue the inviability of cells containing the histidine substitutions in STE20 when they were coupled to the cla4 mutation, and the modified Cdc42p did not dramatically improve the localization of the histidine substitution mutants of the Ste20-GFP fusion protein. The two-hybrid association of the H345D and H348D alleles of Ste20p with the N39S P69T allele of Cdc42p was considerably less than the association measured for the wild-type proteins. Thus the improvement in some Ste20p functions and not in others may be attributed to the strength of the associations; weak improvements in function or localization may improve mating without allowing the rescue of the missing cla4 function.

Previous studies provided conflicting data on the requirement for Cdc42p in Ste20p function in the mating pathway (SIMON et al. 1995 ; PETER et al. 1996 ; LEBERER et al. 1997 ; MOSKOW et al. 2000 ). The present work shows that the interaction between the kinase and the GTPase plays a significant role in mating, and this interpretation has been independently shown by work that analyzed related but distinct CRIB domain mutations (LAMSON et al. 2002 ). This association appears critical for proper localization of the kinase to sites of polarized growth in both vegetative and shmooing cells, but not for kinase catalytic activity measured in vitro. Further work will be required to determine the structural role of Cdc42p in Ste20p action.

	FOOTNOTES

¹ Present address: Biochemistry Department, McGill University, Montreal, Quebec H3A 1B1, Canada.

	ACKNOWLEDGMENTS

We thank Peter Pryciak for comments on the manuscript. This is NRC publication no. 44853.

Manuscript received April 4, 2002; Accepted for publication October 7, 2002.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ABDUL-MANAN, N., B. AGHAZADEH, G. A. LIU, A. MAJUMDAR, and O. OUERFELLI et al., 1999  Structure of Cdc42 in complex with the GTPase-binding domain of the ‘Wiskott-Aldrich syndrome’ protein. Nature 399:379-383.[Medline]

ADAMS, A. E. M., D. I. JOHNSON, R. M. LONGNECKER, B. F. SLOAT, and J. R. PRINGLE, 1990  CDC42 and CDC43, two additional genes involved in budding and the establishment of cell polarity in the yeast Saccharomyces cerevisiae.. J. Cell Biol. 111:131-142.[Abstract/Free Full Text]

BEST, A., S. AHMED, R. KOZMA, and L. LIM, 1996  The Ras-related GTPase Rac1 binds tubulin. J. Biol. Chem. 271:3756-3762.[Abstract/Free Full Text]

BISHOP, A. L. and A. HALL, 2000  Rho GTPases and their effector proteins. Biochem. J. 348(Pt 2):241-255.[Medline]

BOURNE, H. R., D. A. SANDERS, and F. MCCORMICK, 1990  The GTPase superfamily: a conserved switch for diverse cell functions. Nature 348:125-132.[Medline]

BURBELO, P. D., D. DRECHSEL, and A. HALL, 1995  A conserved binding motif defines numerous candidate target proteins for both Cdc42 and Rac GTPases. J. Biol. Chem. 270:29071-29074.[Abstract/Free Full Text]

CHANT, J., 1996  Generation of cell polarity in yeast. Curr. Opin. Cell Biol. 8:557-565.[Medline]

CHIEN, C. T., P. L. BARTEL, R. STERNGLANZ, and S. FIELDS, 1991  The two-hybrid system: a method to identify and clone genes for proteins that interact with a protein of interest. Proc. Natl. Acad. Sci. USA 88:9578-9582.[Abstract/Free Full Text]

COURCHESNE, W. E., R. KUNISAWA, and J. THORNER, 1989  A putative protein kinase overcomes pheromone induced arrest of cell cycling in S. cerevisiae.. Cell 58:1107-1119.[Medline]

CVRCKOVA, F. and K. NASMYTH, 1993  Yeast G1 cyclins CLN1 and CLN2 and a GAP-like protein have a role in bud formation. EMBO J. 12:5277-5286.[Medline]

CVRCKOVA, F., C. DE VIRGILIO, E. MANSER, J. R. PRINGLE, and K. NASMYTH, 1995  Ste20-like protein kinases are required for normal localization of cell growth and for cytokinesis in budding yeast. Genes Dev. 9:1817-1830.[Abstract/Free Full Text]

EBY, J. J., S. P. HOLLY, F. VAN DROGEN, A. V. GRISHIN, and M. PETER et al., 1998  Actin cytoskeleton organization regulated by the PAK family of protein kinases. Curr. Biol. 8:967-970.[Medline]

EMERSON, S. D., V. S. MADISON, R. E. PALERMO, D. S. WAUGH, and J. E. SCHEFFLER et al., 1995  Solution structure of the Ras-binding domain of c-Raf-1 and identification of its Ras interaction surface. Biochemistry 34:6911-6918.[Medline]

EVANGELISTA, M., K. BLUNDELL, M. S. LONGTINE, C. J. CHOW, and N. ADAMES et al., 1997  Bni1p, a yeast formin linking cdc42p and the actin cytoskeleton during polarized morphogenesis. Science 276:118-122.[Abstract/Free Full Text]

GIZACHEW, D., W. GUO, K. K. CHOHAN, M. J. SUTCLIFFE, and R. E. OSWALD, 2000  Structure of the complex of Cdc42Hs with a peptide derived from P-21 activated kinase. Biochemistry 39:3963-3971.[Medline]

GUO, W., M. J. SUTCLIFFE, R. A. CERIONE, and R. E. OSWALD, 1998  Identification of the binding surface on Cdc42Hs for p21-activated kinase. Biochemistry 37:14030-14037.[Medline]

HIGUCHI, R., B. KRUMMEL, and R. K. SAIKI, 1988  A general method of in vitro preparation and specific mutagenesis of DNA fragments: study of protein and DNA interactions. Nucleic Acids Res. 16:7351-7367.[Abstract/Free Full Text]

HOLLY, S. P. and K. J. BLUMER, 1999  PAK-family kinases regulate cell and actin polarization throughout the cell cycle of Saccharomyces cerevisiae. J. Cell Biol. 147:845-856.[Abstract/Free Full Text]

JARVIS, E. E., D. C. HAGEN, and G. F. SPRAGUE, JR., 1988  Identification of a DNA segment that is necessary and sufficient for alpha-specific gene control in Saccharomyces cerevisiae: implications for regulation of alpha-specific and a-specific genes. Mol. Cell. Biol. 8:309-320.[Abstract/Free Full Text]

LAMSON, R. E., M. J. WINTERS, and P. M. PRYCIAK, 2002  Cdc42 regulation of kinase activity and signaling by the yeast p21-activated kinase Ste20. Mol. Cell. Biol. 22:2939-2951.[Abstract/Free Full Text]

LEBERER, E., D. DIGNARD, D. HARCUS, D. Y. THOMAS, and M. WHITEWAY, 1992  The protein kinase homologue Ste20p is required to link the yeast pheromone response G-protein beta gamma subunits to downstream signalling components. EMBO J. 11:4815-4824.[Medline]

LEBERER, E., D. DIGNARD, D. HARCUS, L. HOUGAN, and M. WHITEWAY et al., 1993  Cloning of Saccharomyces cerevisiae STE5 as a suppressor of a Ste20 protein kinase mutant: structural and functional similarity of Ste5 to Far1. Mol. Gen. Genet. 241:241-254.[Medline]

LEBERER, E., C. WU, T. LEEUW, A. FOUREST-LIEUVIN, and J. E. SEGALL et al., 1997  Functional characterization of the Cdc42p binding domain of yeast Ste20p protein kinase. EMBO J. 16:83-97.[Medline]

LEEUW, T., C. WU, J. D. SCHRAG, M. WHITEWAY, and D. Y. THOMAS et al., 1998  Interaction of a G-protein beta-subunit with a conserved sequence in Ste20/PAK family protein kinases. Nature 391:191-195.[Medline]

LEI, M., W. LU, W. MENG, M. C. PARRINI, and M. J. ECK et al., 2000  Structure of PAK1 in an autoinhibited conformation reveals a multistage activation switch. Cell 102:387-397.[Medline]

LIN, D., A. S. EDWARDS, J. P. FAWCETT, G. MBAMALU, and J. D. SCOTT et al., 2000  A mammalian PAR-3-PAR-6 complex implicated in Cdc42/Rac1 and aPKC signalling and cell polarity. Nat. Cell Biol. 2:540-547.[Medline]

MANSER, E., T. LEUNG, H. SALIHUDDIN, Z. S. ZHAO, and L. LIM, 1994  A brain serine/threonine protein kinase activated by Cdc42 and Rac1. Nature 367:40-46.[Medline]

MARCUS, S., A. POLVERINO, M. BARR, and M. WIGLER, 1994  Complexes between STE5 and components of the pheromone mitogen-activated protein kinase module. Proc. Natl. Acad. Sci. USA 91:7762-7766.[Abstract/Free Full Text]

MARTIN, H., A. MENDOZA, J. M. RODRIGUEZ-PACHON, M. MOLINA, and C. NOMBELA, 1997  Characterization of SKM1, a Saccharomyces cerevisiae gene encoding a novel Ste20/PAK-like protein kinase. Mol. Microbiol. 23:431-444.[Medline]

MORREALE, A., M. VENKATESAN, H. R. MOTT, D. OWEN, and D. NIETLISPACH et al., 2000  Structure of Cdc42 bound to the GTPase binding domain of PAK. Nat. Struct. Biol. 7:384-388.[Medline]

MOSKOW, J. J., A. S. GLADFELTER, R. E. LAMSON, P. M. PRYCIAK, and D. J. LEW, 2000  Role of Cdc42p in pheromone-stimulated signal transduction in Saccharomyces cerevisiae. Mol. Cell. Biol. 20:7559-7571.[Abstract/Free Full Text]

MOTT, H. R., D. OWEN, D. NIETLISPACH, P. N. LOWE, and E. MANSER et al., 1999  Structure of the small G protein Cdc42 bound to the GTPase-binding domain of ACK. Nature 399:384-388.[Medline]

MURRAY, J. M. and D. I. JOHNSON, 2001  The Cdc42p GTPase and its regulators Nrf1p and Scd1p are involved in endocytic trafficking in the fission yeast Schizosaccharomyces pombe. J. Biol. Chem. 276:3004-3009.[Abstract/Free Full Text]

NICHOLLS, A., K. A. SHARP, and B. HONIG, 1991  Protein folding and association: insights from the interfacial and thermodynamic properties of hydrocarbons. Proteins 11:281-296.[Medline]

NUR, E. K. M. S., A. SIZELAND, G. D'ABACO, and H. MARUTA, 1992  Asparagine 26, glutamic acid 31, valine 45, and tyrosine 64 of Ras proteins are required for their oncogenicity. J. Biol. Chem. 267:1415-1418.[Abstract/Free Full Text]

OEHLEN, L. J. and F. R. CROSS, 1998  The role of Cdc42 in signal transduction and mating of the budding yeast Saccharomyces cerevisiae. J. Biol. Chem. 273:8556-8559.[Abstract/Free Full Text]

PETER, M., A. M. NEIMAN, H. O. PARK, M. VAN LOHUIZEN, and I. HERSKOWITZ, 1996  Functional analysis of the interaction between the small GTP binding protein Cdc42 and the Ste20 protein kinase in yeast. EMBO J. 15:7046-7059.[Medline]

PHILIPPSEN, P., A. STOTZ, and C. SCHERF, 1991  DNA of Saccharomyces cerevisiae. Methods Enzymol. 194:169-182.[Medline]

RAITT, D. C., F. POSAS, and H. SAITO, 2000  Yeast Cdc42 GTPase and Ste20 PAK-like kinase regulate Sho1-dependent activation of the Hog1 MAPK pathway. EMBO J. 19:4623-4631.[Medline]

RENEKE, J. E., K. J. BLUMER, W. E. COURCHESNE, and J. THORNER, 1988  The carboxy-terminal segment of the yeast alpha-factor receptor is a regulatory domain. Cell 55:221-234.[Medline]

RICHMAN, T. J., M. M. SAWYER, and D. I. JOHNSON, 1999  The Cdc42p GTPase is involved in a G2/M morphogenetic checkpoint regulating the apical-isotropic switch and nuclear division in yeast. J. Biol. Chem. 274:16861-16870.[Abstract/Free Full Text]<