The Phosphorylation Site for Ste20p-like Protein Kinases Is Essential for the Function of Myosin-I in Yeast
1997; Elsevier BV; Volume: 272; Issue: 49 Linguagem: Inglês
10.1074/jbc.272.49.30623
ISSN1083-351X
AutoresCunle Wu, Viktoria Lytvyn, David Y. Thomas, Ekkehard Leberer,
Tópico(s)Protein Kinase Regulation and GTPase Signaling
ResumoThe budding yeast Saccharomyces cerevisiae has two functionally redundant myosin-I isoforms encoded by the MYO3 and MYO5 genes. The function shared by these myosin proteins is required for proper yeast budding. Serine residue 357 in the head domain of Myo3p, conserved among myosin-I proteins including yeast Myo5p, was identified as a unique phosphorylation site for the serine/threonine protein kinase Ste20p and its closely related isoform Cla4p. These protein kinases share a function that is also essential for budding. Replacement of serine 357 with alanine disrupted the in vivo function of Myo3p, whereas this function was maintained by changing the serine residue to aspartate. This mutant version failed to compensate the growth defect of cells which lack both Ste20p and Cla4p, suggesting that myosin-I is not the only essential target of these protein kinases. Our results suggest that phosphorylation of the head domain by Ste20p-like protein kinases plays an essential role in the function of myosin-I in yeast cells. The budding yeast Saccharomyces cerevisiae has two functionally redundant myosin-I isoforms encoded by the MYO3 and MYO5 genes. The function shared by these myosin proteins is required for proper yeast budding. Serine residue 357 in the head domain of Myo3p, conserved among myosin-I proteins including yeast Myo5p, was identified as a unique phosphorylation site for the serine/threonine protein kinase Ste20p and its closely related isoform Cla4p. These protein kinases share a function that is also essential for budding. Replacement of serine 357 with alanine disrupted the in vivo function of Myo3p, whereas this function was maintained by changing the serine residue to aspartate. This mutant version failed to compensate the growth defect of cells which lack both Ste20p and Cla4p, suggesting that myosin-I is not the only essential target of these protein kinases. Our results suggest that phosphorylation of the head domain by Ste20p-like protein kinases plays an essential role in the function of myosin-I in yeast cells. Myosin-I proteins are unconventional single-headed, nonfilamenting myosins consisting of a heavy chain and one or more light chains (for a review, see Refs. 1Brzeska H. Korn E.D. J. Biol. Chem. 1996; 271: 16983-16986Abstract Full Text Full Text PDF PubMed Scopus (79) Google Scholar and 2Hasson T. Mooseker M.S. J. Biol. Chem. 1996; 271: 16431-16434Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar). The amino-terminal portion of the heavy chain forms a head domain, which displays actin-stimulated Mg2+-ATPase activity and generates the mechanochemical force for actin-based motile processes. The carboxyl-terminal portion forms a globular tail domain with binding sites for phospholipids and, in some isoforms, an ATP-independent filamentous actin binding site and a Src homology 3 domain. While first described inAcanthamoeba, myosin-I has now been identified in many eukaryotic cells, including yeast, and is known to exist in many multiple isoforms (1Brzeska H. Korn E.D. J. Biol. Chem. 1996; 271: 16983-16986Abstract Full Text Full Text PDF PubMed Scopus (79) Google Scholar, 2Hasson T. Mooseker M.S. J. Biol. Chem. 1996; 271: 16431-16434Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar). However, the physiological roles of these isoforms and their in vivo regulation are not well understood. In the budding yeast Saccharomyces cerevisiae, two functionally redundant isoforms encoded by the MYO3 andMYO5 genes have been shown to be required for the organization of the actin cytoskeleton during budding (3Goodson H.V. Anderson B.L. Warrick H.M. Pon L.A. Spudich J.A. J. Cell Biol. 1996; 133: 1277-1291Crossref PubMed Scopus (186) Google Scholar), and Myo5p has also been shown to fulfill a unique role in actin-based endocytosis (4Geli M.I. Riezman H. Science. 1996; 272: 533-535Crossref PubMed Scopus (233) Google Scholar). Depending on the yeast strain, disruption of both genes causes either extremely poor growth (3Goodson H.V. Anderson B.L. Warrick H.M. Pon L.A. Spudich J.A. J. Cell Biol. 1996; 133: 1277-1291Crossref PubMed Scopus (186) Google Scholar) or lethality (4Geli M.I. Riezman H. Science. 1996; 272: 533-535Crossref PubMed Scopus (233) Google Scholar) indicating an important role of myosin-I in supporting cellular viability. The biochemical characterization of myosin-I isoforms fromAcanthamoeba and Dictyostelium has demonstrated that phosphorylation of a single site in the heavy chain head domain is necessary to attain maximal actin-stimulated Mg2+-ATPase activity and to contract and move actin filaments in in vitro assay systems (5Zot H.G. Doberstein S.K. Pollard T.D. J. Cell Biol. 1992; 116: 367-376Crossref PubMed Scopus (83) Google Scholar, 6Fujisaki H. Albanesi J.P. Korn E.D. J. Biol. Chem. 1985; 260: 11183-11189Abstract Full Text PDF PubMed Google Scholar, 7Lynch T.J. Brzeska H. Miyata H. Korn E.D. J. Biol. Chem. 1989; 264: 19333-19339Abstract Full Text PDF PubMed Google Scholar, 8Lee S.-F. Côté G.P. J. Biol. Chem. 1995; 270: 11776-11782Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar). However, the physiological significance of this type of regulation has not yet been demonstrated. The protein kinases capable of carrying out this phosphorylation reaction were recently identified as members of the Ste20p family of serine/threonine protein kinases (9Wu C. Lee S.-F. Furmaniak-Kazmierczak E. Côté G.P. Thomas D.Y. Leberer E. J. Biol. Chem. 1996; 271: 31787-31790Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar, 10Lee S.-F. Egelhoff T.T. Mahasneh A. Côté G.P. J. Biol. Chem. 1996; 271: 27044-27048Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar, 11Brzeska H. Knaus U.G. Wang Z.-Y. Bokoch G.M. Korn E.D. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 1092-1095Crossref PubMed Scopus (70) Google Scholar, 12Brzeska H. Szczepanowska J. Hoej J. Korn E.D. J. Biol. Chem. 1996; 271: 27056-27062Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar). These protein kinases are thought to be involved in triggering morphogenetic processes in response to external signals in organisms ranging from yeast to mammals (13Leberer E. Thomas D.Y. Whiteway M. Curr. Opin. Genet. Dev. 1997; 7: 59-66Crossref PubMed Scopus (191) Google Scholar, 14Lim L. Manser E. Leung T. Hall C. Eur. J. Biochem. 1996; 242: 171-185Crossref PubMed Scopus (273) Google Scholar, 15Sells M.A. Chernoff J. Trends Cell Biol. 1997; 7: 162-167Abstract Full Text PDF PubMed Scopus (265) Google Scholar). Two members of this protein kinase family, Ste20p and Cla4p, have been well characterized in S. cerevisiae (16Cvrckova F. De Virgilio C. Manser E. Pringle J.R. Nasmyth K. Genes Dev. 1995; 9: 1817-1830Crossref PubMed Scopus (312) Google Scholar, 17Leberer E. Dignard D. Harcus D. Thomas D.Y. Whiteway M. EMBO J. 1992; 11: 4815-4824Crossref PubMed Scopus (348) Google Scholar). Both kinases share an essential function for polarized growth during budding (16Cvrckova F. De Virgilio C. Manser E. Pringle J.R. Nasmyth K. Genes Dev. 1995; 9: 1817-1830Crossref PubMed Scopus (312) Google Scholar,18Leberer E. Wu C. Leeuw T. Fourest-Lieuvin A. Segall J.E. Thomas D.Y. EMBO J. 1997; 16: 83-97Crossref PubMed Scopus (167) Google Scholar). This function requires binding of the small Rho-like G-protein Cdc42p to a small domain (18Leberer E. Wu C. Leeuw T. Fourest-Lieuvin A. Segall J.E. Thomas D.Y. EMBO J. 1997; 16: 83-97Crossref PubMed Scopus (167) Google Scholar, 19Peter M. Neiman A. Park H.-O. Van Lohuizen M. Herskowitz I. EMBO J. 1996; 15: 7046-7059Crossref PubMed Scopus (193) Google Scholar) that is highly conserved in members of the Ste20p family, including the mammalian homologs of Ste20p, the p21-activated kinases (PAKs) 1The abbreviations used are: PAK, p21-activated protein kinase; GST, glutathione S-transferase; PAGE, polyacrylamide gel electrophoresis. (20Manser E. Leung T. Salihuddin H. Zhao Z. Lim L. Nature. 1994; 367: 40-46Crossref PubMed Scopus (1305) Google Scholar, 21Martin G.A. Bollag G. McCormick F. Abo A. EMBO J. 1995; 14: 1970-1978Crossref PubMed Scopus (304) Google Scholar, 22Bagrodia S. Taylor S.J. Creasy C.L. Chernoff J. Cerione R.A. J. Biol. Chem. 1995; 270: 22731-22737Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar). The PAKs are thought to be effectors of Rho-like small G-proteins (13Leberer E. Thomas D.Y. Whiteway M. Curr. Opin. Genet. Dev. 1997; 7: 59-66Crossref PubMed Scopus (191) Google Scholar, 14Lim L. Manser E. Leung T. Hall C. Eur. J. Biochem. 1996; 242: 171-185Crossref PubMed Scopus (273) Google Scholar, 15Sells M.A. Chernoff J. Trends Cell Biol. 1997; 7: 162-167Abstract Full Text PDF PubMed Scopus (265) Google Scholar), which in turn play a role in mediating signals that trigger membrane ruffling and the assembly of focal adhesions in fibroblasts (23Nobes C.D. Hall A. Cell. 1995; 81: 53-62Abstract Full Text PDF PubMed Scopus (3746) Google Scholar, 24Kozma R. Ahmed S. Best A. Lim L. Mol. Cell. Biol. 1995; 15: 1942-1952Crossref PubMed Scopus (883) Google Scholar). We have recently shown that Ste20p and Cla4p (9Wu C. Lee S.-F. Furmaniak-Kazmierczak E. Côté G.P. Thomas D.Y. Leberer E. J. Biol. Chem. 1996; 271: 31787-31790Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar), like theDictyostelium (10Lee S.-F. Egelhoff T.T. Mahasneh A. Côté G.P. J. Biol. Chem. 1996; 271: 27044-27048Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar) and Acanthamoeba (12Brzeska H. Szczepanowska J. Hoej J. Korn E.D. J. Biol. Chem. 1996; 271: 27056-27062Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar) homologs of Ste20p and the mammalian PAKs (9Wu C. Lee S.-F. Furmaniak-Kazmierczak E. Côté G.P. Thomas D.Y. Leberer E. J. Biol. Chem. 1996; 271: 31787-31790Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar, 11Brzeska H. Knaus U.G. Wang Z.-Y. Bokoch G.M. Korn E.D. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 1092-1095Crossref PubMed Scopus (70) Google Scholar), are capable of phosphorylating and activating the heavy chain of myosin-I from amoeboid cells. Here, we show that Ste20p and Cla4p are able to phosphorylate serine residue 357 in the head domain of the yeast myosin-I isoform Myo3p in vitro. We further demonstrate, by a mutational analysis in yeast, that this phosphorylation site is essential for the in vivo function of myosin-I and that the requirement for this phosphorylation site can be bypassed by introduction of an amino acid with a negatively charged side chain at this position, through replacement of serine by aspartate. The S. cerevisiae strains used in this study were YEL206 (MAT a ade2 leu2 trp1 ura3 his3 can1 ste20Δ-3::TRP1) (25Wu C. Whiteway M. Thomas D.Y. Leberer E. J. Biol. Chem. 1995; 270: 15984-15992Abstract Full Text Full Text PDF PubMed Scopus (150) Google Scholar), YEL252-1A (MAT a ade2 leu2 trp1 ura3 his3 can1 cla4Δ::TRP1) (18Leberer E. Wu C. Leeuw T. Fourest-Lieuvin A. Segall J.E. Thomas D.Y. EMBO J. 1997; 16: 83-97Crossref PubMed Scopus (167) Google Scholar), YEL257–1A-1 (MAT α ade2 leu2 trp1 ura3 his3 can1 ste20Δ-3::TRP1 cla4Δ::TRP1(pRL21::URA3)) (18Leberer E. Wu C. Leeuw T. Fourest-Lieuvin A. Segall J.E. Thomas D.Y. EMBO J. 1997; 16: 83-97Crossref PubMed Scopus (167) Google Scholar), and RH3384 (MAT a his3 leu2 lys2 trp1 ura3 sst1 myo3Δ::HIS3 myo5Δ::TRP1 (pMYO5::URA3)) (4Geli M.I. Riezman H. Science. 1996; 272: 533-535Crossref PubMed Scopus (233) Google Scholar). Yeast manipulations were carried out as described (26Rose M.D. Winston F. Hieter P. Methods in Yeast Genetics. A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1990Google Scholar). To construct plasmid pVL96 carryingMYO3, a SalI to SmaI fragment ofMYO3 from nucleotide positions −1,255 to 4,110 was amplified by the polymerase chain reaction (27Saiki R.J. Gelfand D.H. Stoffel S. Scharf S.J. Higuchi R. Horn G.T. Mullis K.B. Erlich H.A. Science. 1988; 239: 487-491Crossref PubMed Scopus (13515) Google Scholar) using the oligodeoxynucleotides OCW67 (5′-ACGCGTCGACAATGTCACTCAGAGCAGTCC-3′) and OCW68 (5′-TTCCCCGGGAAAATTAAGGAGGGTTTACTG-3′) 2The newly created SalI andSmaI sites, respectively, are underlined. as primers and plasmid pEKG090 (kindly provided by M. Ramezani-Rad and C. Hollenberg, Heinrich Heine University, Düsseldorf) as a template and subcloned into pRS315 (28Sikorski R.S. Hieter P. Genetics. 1989; 122: 19-27Crossref PubMed Google Scholar). Serine residue 357 of Myo3p was changed to an alanine residue using a two-step polymerase chain reaction procedure. First, two polymerase chain reactions were performed using the oligodeoxynucleotide primer pairs OCW67 and OCW69AR (5′-GGGTTCAAAGGTACATGATAACGGCGCCTCTCTTCATTCC-3′) and OCW69A (5′-GGAATGAAGAGAGGCGCCGTTTATCATGTACCTTTGAACCC-3′) 3The introduced nucleotide changes are underlined. and OCW68, respectively, and plasmid pVL96 as a template. The amplified fragments were then purified, mixed, and used as templates for an additional polymerase chain reaction using the oligodeoxynucleotides OCW67 and OCW68 as primers. The resulting fragment was then digested with SalI and SmaI and cloned into pRS315 to yield plasmid pVL97. The same procedure was used to create plasmid pVL98 carrying an alteration of serine residue 357 to an aspartate residue. The mutagenizing oligodeoxynucleotides in the first two polymerase chain reactions were OCW63DR (5′-GGGTCCAAAGGTACATGATAGACGTCGCCTCTCTTCATTCC-3′) and OCW63D (5′-GGAATGAAGAGAGGCGACGTCTATCATGTACCTTTGAACCC-3′),3respectively. To convert the alanine or aspartate residues at position 357 back to serine residues, plasmids pVL97 and pVL98, respectively, were mutagenized by the two-step polymerase chain reaction described above using the oligodeoxynucleotide primer pairs OCW67 and OCW72AR (5′-GGGTTCAAAGGTACATGATAAACTGAGCCTCTCTTCATTCC-3′) and OCW72A (5′-GGAATGAAGAGAGGCTCAGTTTATCATGTACCTTTGAACCC-3′) and OCW68, respectively. The fragments were subcloned into pRS315 to yield plasmids pVL104 and pVL105, respectively. All the predicted nucleotide changes were confirmed by DNA sequencing. Fusions of MYO3 to the gene encoding glutathione S-transferase (GST) were constructed by cloning the DraIII (blunt-ended) toBglII fragments from plasmids pVL96, pVL97, and pVL98, respectively, into the SmaI and BamHI sites of pGEX-4T-3 (Pharmacia Biotech Inc.) to yield plasmids pVL99 (GST-MYO3), pVL100 (GST-MYO3 S357A), and pVL101 (GST-MYO3 S357D), respectively. The GST fusion proteins were expressed in Escherichia colistrain UT5600 (New England Biolabs Inc.) and purified over glutathione-Sepharose (25Wu C. Whiteway M. Thomas D.Y. Leberer E. J. Biol. Chem. 1995; 270: 15984-15992Abstract Full Text Full Text PDF PubMed Scopus (150) Google Scholar). Antibodies to Ste20p and Cla4p were as described previously (9Wu C. Lee S.-F. Furmaniak-Kazmierczak E. Côté G.P. Thomas D.Y. Leberer E. J. Biol. Chem. 1996; 271: 31787-31790Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar, 25Wu C. Whiteway M. Thomas D.Y. Leberer E. J. Biol. Chem. 1995; 270: 15984-15992Abstract Full Text Full Text PDF PubMed Scopus (150) Google Scholar). Antibodies to bacterially produced GST were raised in rabbits and affinity-purified as described (25Wu C. Whiteway M. Thomas D.Y. Leberer E. J. Biol. Chem. 1995; 270: 15984-15992Abstract Full Text Full Text PDF PubMed Scopus (150) Google Scholar). Ste20p and Cla4p and their catalytically inactive mutant versions were immunopurified from yeast as described (9Wu C. Lee S.-F. Furmaniak-Kazmierczak E. Côté G.P. Thomas D.Y. Leberer E. J. Biol. Chem. 1996; 271: 31787-31790Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar, 25Wu C. Whiteway M. Thomas D.Y. Leberer E. J. Biol. Chem. 1995; 270: 15984-15992Abstract Full Text Full Text PDF PubMed Scopus (150) Google Scholar). Western blot analyses were performed as described previously (25Wu C. Whiteway M. Thomas D.Y. Leberer E. J. Biol. Chem. 1995; 270: 15984-15992Abstract Full Text Full Text PDF PubMed Scopus (150) Google Scholar). Prior to the phosphorylation reaction, Ste20p and Cla4p immune complexes were incubated in buffer A (20 mm Tris-HCl, pH 7.5, 10 mmMgCl2, 1 mm dithiothreitol, 5 μg/ml aprotinin, and 5 μg/ml leupeptin) with 0.2 mm ATP for 20 min at 30 °C to stimulate autophosphorylation of the protein kinases. The immune complexes were then washed three times in buffer A and added to the phosphorylation mixture consisting of 2 μg of GST-Myo3p fusion protein and 5 μm[γ-32P]ATP (1 × 104 Ci/mol) in 50 μl of buffer A. After incubation for 20 min at 30 °C, the reaction was stopped by addition of 50 μl of 2 × Laemmli buffer (29Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (207516) Google Scholar). The reaction mixture was then boiled for 5 min and analyzed by SDS-PAGE (29Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (207516) Google Scholar) and autoradiography. Our recent finding that the heavy chain from myosin-ID fromDictyostelium is a target for the Ste20p and Cla4p protein kinases from yeast (9Wu C. Lee S.-F. Furmaniak-Kazmierczak E. Côté G.P. Thomas D.Y. Leberer E. J. Biol. Chem. 1996; 271: 31787-31790Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar) prompted us to investigate whether these protein kinases are capable of phosphorylating the yeast myosin-I isoform Myo3pin vitro and whether the phosphorylation site recognized by these kinases is of physiological significance for the function of Myo3p in vivo. We found that a fragment of Myo3p from residues 19 to 619 fused to GST and purified from E. coli is an in vitro substrate for both Ste20p and Cla4p, which were immunopurified from yeast (Fig. 1). The catalytically inactive mutant versions of these kinases, in which the highly conserved lysine residues in the catalytic sites were replaced with arginine residues, were unable to phosphorylate the GST-Myo3p fusion protein (Fig. 1). The phosphorylated residues responsible for the in vitroactivation of Acanthamoeba myosin-I isoforms have been identified as either serine or threonine residues within the sequence motif (K/R)X(S/T)XY that is conserved in the heavy chains of all myosin-I isoforms from Acanthamoeba, Dictyostelium, and Aspergillus, and in theDrosophila and mammalian class VI myosins (30Brzeska H. Lynch T.J. Martin B. Korn E.D. J. Biol. Chem. 1989; 264: 19340-19348Abstract Full Text PDF PubMed Google Scholar, 31Bement W.M. Mooseker M.S. Cell Motil. Cytoskeleton. 1995; 31: 87-92Crossref PubMed Scopus (148) Google Scholar). This regulatory residue is conserved at position 357 in the sequence motif KRGSVY of yeast Myo3p (32Goodson H.V. Spudich J.A. Cell Motil. Cytoskeleton. 1995; 30: 73-84Crossref PubMed Scopus (70) Google Scholar). In the Myo5p isoform, the same sequence motif is conserved at the identical position (3Goodson H.V. Anderson B.L. Warrick H.M. Pon L.A. Spudich J.A. J. Cell Biol. 1996; 133: 1277-1291Crossref PubMed Scopus (186) Google Scholar, 4Geli M.I. Riezman H. Science. 1996; 272: 533-535Crossref PubMed Scopus (233) Google Scholar). We have changed serine residue 357 in Myo3p to either alanine or aspartate and analyzed the mutant fragments fused to GST as substrates for Ste20p and Cla4p. Consistent with the view that serine residue 357 of Myo3p represents the site phosphorylated by either Ste20p or Cla4p, we found that the mutant fusion proteins purified from E. coli were not phosphorylated by either of these protein kinases (Fig.2). These results indicate that serine residue 357 represents a unique phosphorylation site in the head domain of Myo3p for the Ste20p and Cla4p protein kinases. To investigate whether serine residue 357 plays a significant role for the function of Myo3p in yeast cells, we used a plasmid-shuffling procedure to replace the MYO3 wild type gene with the mutant versions encoding the mutant proteins in which serine residue 357 was replaced with alanine and aspartate residues, respectively. Plasmids carrying wild type MYO3 and the mutant versions were transformed into a myo3Δ myo5Δ strain whose viability was supported by a plasmid carrying MYO5 and URA3as a selectable marker. Loss of the MYO5 plasmid was then promoted by the presence of 5-fluoroorotic acid (33Boeke J.D. Trueheart J. Natsoulis G. Fink G.R. Methods Enzymol. 1987; 154: 164-171Crossref PubMed Scopus (1083) Google Scholar). As illustrated in Fig. 3, the Myo3pS357A mutant protein was unable to fulfill the essential function of Myo3p. However, the Myo3pS357D mutant was able to do so (Fig. 3). The wild type function of Myo3p was regained when the alanine residue in the mutant protein was reverted to a serine residue confirming that the mutant phenotype was caused by the serine to alanine mutation at position 357 and not by a secondary mutation (data not shown). These results suggest that phosphorylation of serine 357 may be essential for the function of Myo3p and that replacement of the serine residue with an aspartate residue may constitutively activate the myosin protein through introduction of a negative charge at this position. It seems reasonable to suppose that under wild type conditions this activation is brought about by phosphorylation through the Ste20p and Cla4p protein kinases, respectively. Cells deleted for the CLA4 gene alone are viable but have defects in cytokinesis (16Cvrckova F. De Virgilio C. Manser E. Pringle J.R. Nasmyth K. Genes Dev. 1995; 9: 1817-1830Crossref PubMed Scopus (312) Google Scholar). Cells deleted for STE20 alone are also viable but have defects in mating (17Leberer E. Dignard D. Harcus D. Thomas D.Y. Whiteway M. EMBO J. 1992; 11: 4815-4824Crossref PubMed Scopus (348) Google Scholar). Deletion of both genes causes lethality (16Cvrckova F. De Virgilio C. Manser E. Pringle J.R. Nasmyth K. Genes Dev. 1995; 9: 1817-1830Crossref PubMed Scopus (312) Google Scholar). Thus, Ste20p and Cla4p have unique functions in cytokinesis and mating and also share an essential function during budding. If Myo3p were the only essential target for Ste20p and Cla4p required for budding, the constitutively active Myo3pS357Dmutant would be expected to bypass the requirement of Ste20p and Cla4p for cellular viability. Therefore, in a plasmid-shuffling experiment we asked whether the Myo3pS357D mutant protein might compensate the growth defect of cells lacking both Ste20p and Cla4p. Plasmids carrying wild type MYO3 and the S357A and S357D mutant versions were transformed into a cla4Δ ste20Δstrain whose viability was supported by a plasmid carryingCLA4 and URA3 as a selectable marker. Loss of theCLA4 plasmid was then promoted with 5-fluoroorotic acid (33Boeke J.D. Trueheart J. Natsoulis G. Fink G.R. Methods Enzymol. 1987; 154: 164-171Crossref PubMed Scopus (1083) Google Scholar). We found that none of the MYO3 versions, including the S357D mutant, could support viability of the cells (Fig.4). A microscopic analysis of these cells showed a pleiotropic population of large and irregularly shaped cells and revealed many ghost cells indicative of cell lysis (data not shown). This phenotype is typical for cells which lack Ste20p and Cla4p (18Leberer E. Wu C. Leeuw T. Fourest-Lieuvin A. Segall J.E. Thomas D.Y. EMBO J. 1997; 16: 83-97Crossref PubMed Scopus (167) Google Scholar) and suggests that the constitutively active Myo3p mutant version is unable to bypass the essential functions of Ste20p and Cla4p. The most likely interpretation of this result is that myosin-I is not the only essential target of these protein kinases. This view is also supported by the observation that the lysis phenotype of cells lacking Ste20p and Cla4p could not be observed in cells lacking Myo3p and Myo5p. Thus, in addition to myosin-I, Ste20p and Cla4p may have several other targets that are required for viability of yeast cells. We also found that the cytokinesis defect caused by deletion ofCLA4 alone was not complemented by the Myo3pS357D mutant suggesting that myosin-I is not the primary target for Cla4p during cytokinesis. Moreover, defects in pheromone-induced morphogenesis during mating in cells deleted forSTE20 alone could not be cured by expression of the mutant protein. Thus, myosin-I is unlikely to be a target of Ste20p in morphogenesis during mating. This view is also supported by the observation that cells depleted for Myo3p and Myo5p have no discernible mating defects (3Goodson H.V. Anderson B.L. Warrick H.M. Pon L.A. Spudich J.A. J. Cell Biol. 1996; 133: 1277-1291Crossref PubMed Scopus (186) Google Scholar). Cells deleted for MYO3 and MYO5 whose viability was supported by expression of the MYO3 S357Dmutant version did not reveal any discernible mutant phenotype (data not shown). These cells showed normal growth rates and had a normal cellular morphology. The actin cytoskeleton, as revealed by phalloidin staining (18Leberer E. Wu C. Leeuw T. Fourest-Lieuvin A. Segall J.E. Thomas D.Y. EMBO J. 1997; 16: 83-97Crossref PubMed Scopus (167) Google Scholar), appeared normal. Growth arrest, the formation of mating-specific morphologies, and the induction of a pheromone-responsive reporter gene, FUS1::lacZ, in response to pheromone stimulation (18Leberer E. Wu C. Leeuw T. Fourest-Lieuvin A. Segall J.E. Thomas D.Y. EMBO J. 1997; 16: 83-97Crossref PubMed Scopus (167) Google Scholar) were also normal. Moreover, mating efficiencies were indistinguishable from those of wild type cells. Thus, although phosphorylation of serine residue 357 appears to be essential for the function of Myo3p, this type of regulation may not be the only mechanism that controls the in vivo function of myosin-I. Other targets of modification may include protein-protein interactions mediated by the Src homology 3 domain of the heavy chain of myosin-I and the myosin-I light chains (1Brzeska H. Korn E.D. J. Biol. Chem. 1996; 271: 16983-16986Abstract Full Text Full Text PDF PubMed Scopus (79) Google Scholar, 2Hasson T. Mooseker M.S. J. Biol. Chem. 1996; 271: 16431-16434Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar). Phosphorylation of the heavy chain by Ste20p-like kinases may be only one essential step in the control of myosin-I function through multiple mechanisms. Together, our results demonstrate that a conserved serine residue in the head domain of the myosin-I isoform Myo3p from yeast can serve as a phosphorylation site for members of the Ste20p protein kinase family. This residue is essential for the in vivo function of myosin-I, suggesting that its phosphorylation by Ste20p or Cla4p plays an essential role in the regulation of this myosin-I isoform. This study shows for the first time in a genetically tractable model organism that the conserved phosphorylation site of myosin-I is relevant for its in vivo regulation. In view of the high degree of evolutionary sequence conservation of both the myosin-I and the Ste20p isoforms and the finding that the PAKs from mammalian cells are capable of phosphorylating myosin-I heavy chains fromDictyostelium (9Wu C. Lee S.-F. Furmaniak-Kazmierczak E. Côté G.P. Thomas D.Y. Leberer E. J. Biol. Chem. 1996; 271: 31787-31790Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar) and Acanthamoeba (11Brzeska H. Knaus U.G. Wang Z.-Y. Bokoch G.M. Korn E.D. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 1092-1095Crossref PubMed Scopus (70) Google Scholar), myosin-I regulation by members of the Ste20p family protein kinases may be an essential mechanism for morphogenetic processes in organisms ranging from yeast to higher eukaryotes. We thank M. I. Geli, H. Riezman, M. Ramezani- Rad, and C. Hollenberg for providing plasmids and yeast strains, and A. Nantel and D. Harcus for helpful discussions and comments on the manuscript.
Referência(s)