A conserved kinase cascade for MAP kinase activation in yeast.

Mitogen-activated protein kinases are regulators of proliferation and differentiation in many eukaryotes. Studies during the last year have revealed that functionally distinct signal pathways in yeast use related protein kinase cascades for mitogen-activated protein kinase activation. These cascades act as intracellular signaling modules that are likely to be conserved from yeast to mammals.

The proliferation of MAP kinase signaling pathways in yeast.

Mitogen-activated protein kinases function in at least five, physiologically distinct, signaling pathways in yeast. These include pathways that mediate response to mating pheromone, pseudohyphal development and invasive growth, cell integrity, sporulation, and response to high extracellular osmolarity. These kinases and their upstream activating kinases comprise signaling modules that, in at least some cases, exist as multiprotein complexes. Studies during the past year have revealed that the Ste5 protein of the mating pheromone response pathway serves as a scaffold to promote interactions among the protein kinases in this pathway, and to prevent their interaction with kinases of other modules.

Stimulatory effects of yeast and mammalian 14-3-3 proteins on the Raf protein kinase.

Intracellular signaling from receptor tyrosine kinases in mammalian cells results in activation of a signal cascade that includes the guanine nucleotide-binding protein Ras and the protein kinases Raf, MEK [mitogen-activated protein kinase (MAPK) or extracellular signal-regulated kinase (ERK) kinase], and MAPK. MAPK activation that is dependent on the coupling of Ras and Raf was reconstituted in yeast. Yeast genes were isolated that, when overexpressed, enhanced the function of Raf. One of them is identical to BMH1, which encodes a protein similar to members of the mammalian 14-3-3 family. Bacterially synthesized mammalian 14-3-3 protein stimulated the activity of Raf prepared from yeast cells expressing c-Raf-1. Thus, the 14-3-3 protein may participate in or be required for activation of Raf.

MCM1 binds to a transcriptional control element in Ty1.

Some Ty1 transposable-element insertion mutations of Saccharomyces cerevisiae activate adjacent-gene expression. These Ty1-activated genes are regulated similarly to certain mating genes. This report shows that the MCM1 protein, which binds to several mating genes, also binds to a transcriptional regulatory sequence in Ty1. The binding of MCM1 to Ty1 correlates with the ability of its binding site to function as a component of the Ty1 transcriptional activator. This correlation supports the idea that MCM1 is important for Ty1-activated gene expression. At mating-gene promoters, MCM1 binds with coactivators or repressors such as STE12, alpha 1, or alpha 2. In contrast, MCM1 binds without these associated DNA-binding proteins at its site in Ty1. This finding suggests that its role in Ty1-mediated transcription is different from that at mating genes.

Transcriptional analysis of Ty1 deletion and inversion derivatives at CYC7.

One class of Ty insertion mutation in Saccharomyces cerevisiae activates expression of adjacent structural genes. The CYC7-H2 mutation, in which a Ty1 element is inserted 5' to the iso-2-cytochrome c coding region of CYC7, causes a 20-fold increase in CYC7 expression. Deletion analysis of CYC7-H2 has shown that distal regions of the Ty1 element are not essential for the transcriptional activation at CYC7. In this report, we have analyzed Ty1 and CYC7 RNA from two CYC7-H2 deletion derivative genes to determine whether a direct correlation exists between transcription of Ty1 and transcription of the adjacent gene. Assuming that all Ty1 elements in the genome are transcribed equally, amounts of CYC7-H2 deletion derivative Ty1 RNA were found to be at least fivefold lower than the amount estimated for the average Ty1 element. These same Ty1 deletion derivatives caused a 20-fold increase in adjacent CYC7 expression. This finding suggests that the mechanism by which Ty1 activates adjacent gene expression does not require normal levels of Ty1 transcription. Two inversion derivatives of the CYC7-H2 Ty1 have also been analyzed. These derivatives did not produce any iso-2-cytochrome c or any normal CYC7 mRNA. Instead they were found to produce a Tyl-CYC7 fusion RNA. Consistent with our findings on CYC7-H2 Ty1 transcription, the amount of the fusion RNA was very low. In addition, the Ty1 inversion derivatives produced a new RNA that mapped to sequences upstream from the inverted Ty1 segment. Similar to Ty1 insertions that activate transcription, the new RNA was found to be transcribed away from Ty1.

Cell-type-dependent gene activation by yeast transposon Ty1 involves multiple regulatory determinants.

Ty transposable element insertion mutations of Saccharomyces cerevisiae can cause cell-type-dependent activation of adjacent gene expression. Several cis-acting regulatory regions within Ty1 that are responsible for these effects were identified. A 211-base-pair (bp) region functions as an activator. This region includes the so-called U5 domain of delta and 145 bp of adjacent epsilon sequences. Unlike activation by the intact Ty1, activation by the 211-bp Ty1 subfragment is cell-type independent. The presence of a 112-bp fragment from a more distal region of Ty1 confers cell-type specificity to the activator. The 112-bp fragment includes sequences with homology to mammalian enhancers and to a yeast a/alpha control site. In addition, Ty1 regions that exert negative effects on gene expression were identified. These results demonstrate that the Ty1 transcriptional control region consists of multiple components with distinct regulatory functions.

Ash1, a daughter cell-specific protein, is required for pseudohyphal growth of Saccharomyces cerevisiae.

Ash1 (for asymmetric synthesis of HO) was first uncovered in genetic screens that revealed its role in mating-type switching. Ash1 prevents HO expression in daughter cells. Because Ash1 has a zinc finger-like domain related to that of the GATA family of transcription factors, it presumably acts by repressing HO transcription. Nonswitching diploid cells also express Ash1, suggesting it could have functions in addition to regulation of HO expression. We show here that Ash1 has an essential function for pseudohyphal growth. Our epistasis analyses are consistent with the deduction that Ash1 acts separately from the mitogen-activated protein kinase cascade and Ste12. Similarly to the case in yeast form cells, Ash1 is asymmetrically localized to the nuclei of daughter cells during pseudohyphal growth. This asymmetric localization reveals that there is a previously unsuspected daughter cell-specific function necessary for pseudohyphal growth.

A Ty1 cell-type-specific regulatory sequence is a recognition element for a constitutive binding factor.

Ty transposable-element insertion mutations of Saccharomyces cerevisiae can cause cell-type-dependent activation of adjacent-gene expression. Several cis-acting regulatory regions within Ty1 are responsible for the effect of Ty1 on adjacent-gene expression. One of these is the block II sequence that was defined by its homology to mammalian enhancers and to the yeast a1-alpha 2 control site. Tandem copies of a 57-base-pair region encompassing block II caused an additive increase in expression of the CYC7 reporter gene in the absence of other Ty1 sequences. The activation of gene expression by the multiple repeats was abolished in a/alpha diploid cells. A specific complex between a constitutive factor in whole-cell extracts and the DNA regulatory element was observed. The protein-binding site for the constitutive factor coincided with the block II element. Base-pair substitutions within the binding site abolished the ability of the block II element to function as a component of the Ty1 activator and to form the factor-DNA complex. The correlation between complex formation and reporter gene expression indicates that factor binding to the cis-acting element is essential for this element to function as a component of the Ty1 activator.

MOT2 encodes a negative regulator of gene expression that affects basal expression of pheromone-responsive genes in Saccharomyces cerevisiae.

Pheromones induce haploid cells of Saccharomyces cerevisiae to differentiate into a mating-competent state. Ste11p is one of several protein kinases required to transmit the pheromone-induced signal and to maintain basal expression of certain mating-specific genes in the absence of pheromone stimulation. To identify potential regulators of Ste11p, we screened for suppressors that restored mating and basal transcriptional competence to a strain with a conditionally functional Ste11p. This screen uncovered a novel gene we call MOT2, for modulator of transcription. A mot2 deletion mutation leads to modest increases in the basal amounts of mRNA for several pheromone-responsive genes. Yet mot2 deletion does not affect the signal transmission activity of the pathway in either the presence or absence of pheromone stimulation. Therefore, we propose that Mot2p, directly or indirectly, represses basal transcription of certain mating-specific genes. Because mot2 deletion mutants also have a conditional cell lysis phenotype, we expect that Mot2p regulatory effects may be more global than for mating-specific gene expression.

An anomalous Ty1 structure attributed to an error in reverse transcription.

We have determined the nucleotide sequence of both delta elements of a Ty1 transposon inserted near the CYC7 gene in the Saccharomyces cerevisiae CYC7-H2 mutant. The upstream delta element in this Ty1 has an unusual inverted repeat structure that may have been formed by an error during reverse transcription.

Ty1 sequence with enhancer and mating-type-dependent regulatory activities.

Some insertion mutations in Saccharomyces cerevisiae activate the expression of adjacent structural genes. The CYC7-H2 mutation is a Ty1 insertion 5' to the iso-2-cytochrome c coding region of CYC7. The Ty1 insertion causes a 20-fold increase in CYC7 expression in a and alpha haploid cell types of S. cerevisiae. This activation is repressed in the a/alpha diploid cell type. Previous computer analysis of the CYC7-H2 Ty1 activator region identified two related sequences with homology both to mammalian enhancers and to a yeast a/alpha control site. A 112-base-pair (bp) DNA fragment encompassing one of these blocks of homology functioned as one component of the Ty1 activator. A 28-bp synthetic oligonucleotide with the wild-type homology block sequence was also functional. A single base pair mutation within the enhancer core of the synthetic 28-bp regulatory element reduced its activation ability to near background amounts. In addition, the 112-bp Ty1 fragment by itself functioned as a target for repression of adjacent gene expression in a/alpha diploid cells.

Identification of a Ty1 regulatory sequence responsive to STE7 and STE12.

Ty1 activation of gene expression observed in haploid cell types of Saccharomyces cerevisiae requires the STE7 and STE12 gene products. An activator sequence within Ty1 that is responsive to these two regulators has been defined. Complex formation between a factor in whole-cell extracts and the DNA regulatory element showed the same dependence on the STE7 and STE12 gene products as did reporter gene expression. Base pair substitutions within the binding site abolished the ability to form the factor-DNA complex and to activate gene expression. The correlation between complex formation and reporter gene expression indicates that factor binding to the cis-acting element is essential for gene activation. Because the predicted protein for the STE7 gene product is homologous to protein kinases, we suggest that protein phosphorylation may directly or indirectly regulate formation of this DNA-protein complex.

Cooperative binding interactions required for function of the Ty1 sterile responsive element.

The Ste12p transcription factor controls the expression of Ty1 transposable element insertion mutations and genes whose products are required for mating in Saccharomyces cerevisiae. The binding site for Ste12p is a consensus DNA sequence known as a pheromone response element (PRE). Upstream activating sequences (UASs) derived from known Ste12p-dependent genes have previously been characterized to require either multiple PREs or a single PRE coupled to a binding site for a second protein. The Ste12p-dependent UAS from Ty1, called a sterile response element (SRE), is of the second type and is comprised of a PRE and an adjacent TEA (TEF-1, Tec1, and AbaA motif) DNA consensus sequence (TCS). In this report, we show by UV cross-linking analysis that two proteins, Ste12p and a protein with an apparent size of 72 kDa, directly contact the Ty1 SRE. Other experiments show that Tec1p is required for formation of the Ty1 SRE protein-DNA complex and is physically present in the complex. These results establish a direct role for Tec1p in the Ty1 SRE and yet another set of combinatorial interactions that achieve a qualitatively distinct mode of transcriptional regulation with Ste12p.

Coordination of the mating and cell integrity mitogen-activated protein kinase pathways in Saccharomyces cerevisiae.

Mating pheromone stimulates a mitogen-activated protein (MAP) kinase activation pathway in Saccharomyces cerevisiae that induces cells to differentiate and form projections oriented toward the gradient of pheromone secreted by a mating partner. The polarized growth of mating projections involves new cell wall synthesis, a process that relies on activation of the cell integrity MAP kinase, Mpk1. In this report, we show that Mpk1 activation during pheromone induction requires the transcriptional output of the mating pathway and protein synthesis. Consequently, Mpk1 activation occurs subsequent to the activation of the mating pathway MAP kinase cascade. Additionally, Spa2 and Bni1, a formin family member, are two coil-coil-related proteins that are involved in the timing and other aspects of mating projection formation. Both proteins also affect the timing and extent of Mpk1 activation. This correlation suggests that projection formation comprises part of the pheromone-induced signal that coordinates Mpk1 activation with mating differentiation. Stimulation of Mpk1 activity occurs through the cell integrity phosphorylation cascade and depends on Pkc1 and the redundant MAP/Erk kinases (MEKs), Mkk1 and Mkk2. Surprisingly, Mpk1 activation by pheromone was only partially impaired in cells lacking the MEK kinase Bck1. This Bck1-independent mechanism reveals the existence of an alternative activator of Mkk1/Mkk2 in some strain backgrounds that at least functions under pheromone-induced conditions.

Novel members of the mitogen-activated protein kinase activator family in Xenopus laevis.

Mitogen-activated protein (MAP) kinases comprise an evolutionarily conserved family of proteins that includes at least three vertebrate protein kinases (p42, p44, and p55 MAPK) and five yeast protein kinases (SPK1, MPK1, HOG1, FUS3, and KSS1). Members of this family are activated by a variety of extracellular agents that influence cellular proliferation and differentiation. In Saccharomyces cerevisiae, there are multiple physiologically distinct MAP kinase activation pathways composed of structurally related kinases. The recently cloned vertebrate MAP kinase activators are structurally related to MAP kinase activators in these yeast pathways. These similarities suggest that homologous kinase cascades are utilized for signal transduction in many, if not all, eukaryotes. We have identified additional members of the MAP kinase activator family in Xenopus laevis by a polymerase chain reaction-based analysis of embryonic cDNAs. One of the clones identified (XMEK2) encodes a unique predicted protein kinase that is similar to the previously reported activator (MAPKK) in X. laevis. XMEK2, a highly expressed maternal mRNA, is developmentally regulated during embryogenesis and expressed in brain and muscle. Expression of XMEK2 in yeast cells suppressed the growth defect associated with loss of the yeast MAP kinase activator homologs, MKK1 and MKK2. Partial sequence of a second cDNA clone (XMEK3) identified yet another potential MAP kinase activator. The pattern of expression of XMEK3 is distinct from that of p42 MAPK and XMEK2. The high degree of amino acid sequence similarity of XMEK2, XMEK3, and MAPKK suggests that these three are related members of an amphibian family of protein kinases involved in the activation of MAP kinase. Discovery of this family suggests that multiple MAP kinase activation pathways similar to those in yeast cells exist in vertebrates.

Pheromone-induced signal transduction in Saccharomyces cerevisiae requires the sequential function of three protein kinases.

Protein phosphorylation plays an important role in pheromone-induced differentiation processes of haploid yeast cells. Among the components necessary for signal transduction are the STE7 and STE11 kinases and either one of the redundant FUS3 and KSS1 kinases. FUS3 and presumably KSS1 are phosphorylated and activated during pheromone induction by a STE7-dependent mechanism. Pheromone also induces the accumulation of STE7 in a hyperphosphorylated form. This modification of STE7 requires the STE11 kinase, which is proposed to act before STE7 during signal transmission. Surprisingly, STE7 hyperphosphorylation also requires a functional FUS3 (or KSS1) kinase. Using in vitro assays for FUS3 phosphorylation, we show that pheromone activates STE7 even in the absence of FUS3 and KSS1. Therefore, STE7 activation must precede modification of FUS3 (and KSS1). These findings suggest that STE7 hyperphosphorylation is a consequence of its activation but not the determining event.

Mid2 is a putative sensor for cell integrity signaling in Saccharomyces cerevisiae.

Hcs77 is a putative cell surface sensor for cell integrity signaling in Saccharomyces cerevisiae. Its loss of function results in cell lysis during growth at elevated temperatures (e.g., 39 degrees C) and impaired signaling to the Mpk1 mitogen-activated protein kinase in response to mild heat shock. We isolated the MID2 gene as a dosage suppressor of the cell lysis defect of an hcs77 null mutant. MID2 encodes a putative membrane protein whose function is required for survival of pheromone treatment. Mid2 possesses properties similar to those of Hcs77, including a single transmembrane domain and a long region that is rich in seryl and threonyl residues. We demonstrate that Mid2 is required for cell integrity signaling in response to pheromone. Additionally, we show that Mid2 and Hcs77 serve a redundant but essential function as cell surface sensors for cell integrity signaling during vegetative growth. Both proteins are uniformly distributed through the plasma membrane and are highly O-mannosylated on their extracellular domains. Finally, we identified a yeast homolog of MID2, designated MTL1, which provides a partially redundant function with MID2 for cell integrity signaling during vegetative growth at elevated temperature but not for survival of pheromone treatment. We conclude that Hcs77 is dedicated to signaling cell wall stress during vegetative growth and that Mid2 participates in this signaling, but its primary role is in signaling wall stress during pheromone-induced morphogenesis.

Identification of regulatory regions within the Ty1 transposable element that regulate iso-2-cytochrome c production in the CYC7-H2 yeast mutant.

The CYC7-H2 mutation in the yeast Saccharomyces cerevisiae was caused by insertion of a Ty1 transposable element in front of the iso-2-cytochrome c structural gene, CYC7. The Ty1 insertion places iso-2-cytochrome c production under control of regulatory signals that are normally required for mating functions in yeast cells. We have investigated the regions of the Ty1 insertion that are responsible for the aberrant production of iso-2-cytochrome c in the CYC7-H2 mutant. Five alterations of the CYC7-H2 gene were obtained by specific restriction endonuclease cleavage of the cloned DNA and ligation of appropriate fragments. The CYC7+, CYC7-H2, and modified CYC7-H2 genes were each inserted into the yeast vector YIp5 and used to transform a cytochrome c-deficient yeast strain. Expression and regulation of each allele integrated at the CYC7 locus have been compared in vivo by determination of the amount of iso-2-cytochrome c produced. These results show that distal regions of the Ty1 element are not essential for the CYC7-H2 overproducing phenotype. In contrast, alterations in the vicinity of the proximal Ty1 junction abolish the CYC7-H2 expression and give rise to different phenotypes.

Yeast MEK-dependent signal transduction: response thresholds and parameters affecting fidelity.

Ste7p and Mkk1p are MEK (MAPK/ERK kinase) family members that function in the mating and cell integrity signal transduction pathways in Saccharomyces cerevisiae. We selected STE7 and MKK1 mutations that stimulated their respective pathways in the absence of an inductive signal. Strikingly, serine-to-proline substitutions at analogous positions in Ste7p (position 368) and Mkk1p (position 386) were recovered by independent genetic screens. Such an outcome suggests that this substitution in other MEKs would exhibit similar properties. The Ste7p-P368 variant has higher basal enzymatic activity than Ste7p but still requires induction to reach full activation. The higher activity associated with Ste7p-P368 allows it to compensate for defects in the cell integrity pathway, but it does so only when it is overproduced or when Ste5p is missing. This behavior suggests that Ste5p, which has been proposed to be a tether for the kinases in the mating pathway, contributes to Ste7p specificity.

Structure of the Saccharomyces cerevisiae HO gene and analysis of its upstream regulatory region.

The HO gene product of Saccharomyces cerevisiae is a site-specific endonuclease that initiates mating type interconversion. We have determined the nucleotide sequence of a 3,129-base-pair (bp) segment containing HO. The segment contains a single long open reading frame encoding a polypeptide of 586 amino acids, which has unusual (unbiased) codon usage and is preceded by 762 bp of upstream region. The predicted HO protein is basic (16% lysine and arginine) and is calculated to have a secondary structure that is 30% helical. The corresponding transcript is initiated approximately 50 nucleotides prior to the presumed initiation codon. Insertion of an Escherichia coli lacZ gene fragment into the putative HO coding segment inactivated HO and formed a hybrid HO-lacZ gene whose beta-galactosidase activity was regulated by the mating type locus in the same manner as HO (repressed by a 1-alpha 2). Upstream regions of 1,360 and 762 bp conferred strong repression; 436 bp led to partial constitutivity and 301 bp to full constitutivity. Thus, DNA sequences that confer repression of HO by a1-alpha 2 are at least 250 nucleotides upstream of the transcription start point and are within 436 nucleotides of the HO initiation codon. The progressive loss of repression suggests that both the -762 to -436 and the -436 to -301 intervals contain sites for regulation by a1-alpha 2. The HO gene contains two distinct regions that promote autonomous replication of plasmids in S. cerevisiae. These regions contain sequences that are homologous to the two conserved sequences that are associated with ARS activity.