Role of transcription factor Kar4 in regulating downstream events in the Saccharomyces cerevisiae pheromone response pathway

Abstract

Yeast Kar4 is a putative transcription factor required for karyogamy (the fusion of haploid nuclei during mating) and possibly other functions. Previously known to be required only for the transcriptional induction of KAR3 and CIK1, microarray experiments identified many genes regulated by Kar4 in both mating and mitosis. Several gene clusters are positively or negatively regulated by mating pheromone in a Kar4-dependent manner. Chromatin immunoprecipitation and gel shift assays confirmed that Kar4 binds to regulatory DNA sequences upstream of KAR3. Together with one-hybrid experiments, these data support a model in which both Kar4 and Ste12 bind jointly to the KAR3 promoter. Analysis of the upstream regions of Kar4-induced genes identified a DNA sequence motif that may be a binding site for Kar4. Mutation within the motif upstream of KAR3 eliminated pheromone induction. Genes regulated by Kar4, on average, are delayed in their temporal expression and exhibit a more stringent dose response to pheromone. Furthermore, the induction of Kar4 by pheromone is necessary for the delayed temporal induction of KAR3 and PRM2, genes required for efficient nuclear fusion during mating. Accordingly, we propose that Kar4 plays a critical role in the choreography of the mating response.

FIG. 1.

(A) Kar4 functions at the promoter region of KAR3. Chromatin immunoprecipitation of Kar4::HA-bound upstream sequence of KAR3. Strain MY5792 harboring a chromosomal version of KAR4::HA (tagged Kar4) and the untagged control strain MY3375 were either treated with pheromone for 90 min or mock treated with MeOH. Immunoprecipitated (ChIP) or whole-cell extract (WCE) DNA was amplified with primers specific to upstream sequences of KAR3 and TEL1 (nonspecific control). PCR fragments were fractionated and visualized by using conventional agarose gel electrophoresis. Lane 1, MY5792 no-cross-link control; lane 2, MY5792 whole-extract control; lane 3, cross-linked tagged strain MY5792, not induced by pheromone; lane 4, pheromone-induced cross-linked MY5792; lane 5, cross-linked, pheromone-induced untagged strain MY3375. (B) Kar4 and Ste12 bind to the KAR3 promoter. The results of a gel shift assay using radioactively labeled KAR3 minimal promoter region (KAR3-min) in the presence of various amounts of E. coli expressed Ste12 and Kar4 is shown. The relative amount of each protein is indicated. (C) Kar4 binding confers pheromone inducible expression. The wild-type yeast two-hybrid reporter strain PJ69-4A (WT) or an isogenic strain containing a ste12Δ were transformed with either the GBD vector plasmid or with GBD-Kar4. Strains were grown to exponential phase, and serial dilutions were spotted onto selective plates with or without 3 μM pheromone (α-factor) dissolved in MeOH. All cells can grow on −TRP plates (GROWTH). Only cells expressing the GAL-HIS3 reporter gene can grow on −TRP,−HIS plates (EXPRESSION). (D) Fusing an activation domain to Kar4 does not remove the requirement for Ste12. S288C-derived strains MY3375 (KAR4 STE12) and MY4166 (kar4Δ STE12), as well as W303-derived strains MY3265 (KAR4 STE12) and MY5100 (KAR4 ste12Δ), were transformed with either the GAD vector plasmid or with the GAD-Kar4 plasmid. Each strain also harbored the reporter construct with the KAR3 minimal promoter region fused upstream of lacZ (pMR5620). Expression from the promoter is presented in β-galactosidase (β-gal) units from cultures grown in the presence (+) or absence (−) of 6 μM α-factor (pheromone).

FIG. 2.

A highly conserved region among Kar4 protein family. (A) Pileup sequences of Kar4-related proteins. Family members shown are from Saccharomyces cerevisiae (Sc), Eremothecium gossypii (Eg), Anopheles gambiae (Ag), Drosophila melanogaster (Dm), Homo sapiens (Hs), Mus musculus (Mm), Xenopus laevis (Xl), and Danio rerio (Dr). Identical residues () and similar residues (: or .) are indicated below the pileup. Boxed regions highlight the regions similar to methytransferase domains VIII′ and IX-N (5). Putative zinc-finger ligand residues are indicated in boldface. (B) Mutations in conserved cysteines do not decrease protein levels of Kar4. Western blot showing levels of MS3216 kar4Δ::HIS strain expressing wild-type HA epitope-tagged Kar4 (KAR4::HA on pMR2654) and mutagenized HA epitope-tagged Kar4 (kar4^{C214G C217G}::HA on pMR5511). The cells were grown in the absence (−) or presence (+) of pheromone. The arrows indicate the two forms of Kar4::HAp. (C) Microscopic analysis of mating. Homozygous matings (wild-type × wild-type, kar4^C214GC217G × kar4^C214GC217G, and kar4Δ × kar4Δ) were mated for 75 min before being fixed in methanol-acetic acid, stained with DAPI to reveal the positions of the nuclei, and examined by fluorescence microscopy. Zygotes were scored as being wild type (nuclei have fused [wt]) or karyogamy defective (nuclei separate and unfused [Kar⁻]). The percentages of each class are listed.

FIG. 3.

Kar4-dependent clusters after pheromone treatment. Strains MS3213 (wild type) and MS3215 (kar4Δ) were exposed to pheromone (+) or mock treated with methanol (−) for 90 min. RNA was extracted and prepared for microarray analysis as described in Materials and Methods. Genes were clustered according to their normalized expression profiles. Specifically, we identified seven clusters that show Kar4-independent pheromone induction (cluster 1), Kar4-independent repression in pheromone (cluster 2), Kar4-dependent pheromone induction (cluster 3), Kar4-dependent repression in pheromone (cluster 4), constitutive expression in wild-type cells but repression by pheromone in the kar4Δ strain (cluster 5), constitutive expression in wild-type cells but induction by pheromone in the kar4Δ strain (cluster 6), and constitutive expression in wild-type cells but higher levels of mitotic expression in the kar4Δ strain (cluster 7).

FIG. 4.

Kar4-dependent induction and repression of genes. (A) PRM2 is required for efficient nuclear fusion. Three different mating mixtures (wild-type × wild-type, n = 55; kar4Δ × kar4Δ, n = 148; and prm2Δ × prm2Δ, n = 195) were examined for karyogamy defects. Mating mixtures were stained with DAPI, and the phenotypes were observed and scored as fused (wt) or unfused (Kar⁻) nuclei. The percentages of each class are shown. (B) Northern blot analysis of wild type (MS3213) and kar4Δ (MS3215) mRNA in the absence (−) or presence (+) of pheromone. The blot was hybridized to a probe for PRM2 (upper panel) or CIS3 (middle panel). A duplicate Northern blot was hybridized with an ACT1 probe (bottom panel).

FIG. 5.

Defining Kar4 activation parameters. (A) KAR4-dependent minimal regulatory region. The DNA region upstream of KAR3 was fused upstream of the lacZ gene, lacking upstream activator sequences. Expression is presented as units of β-galactosidase specific activity in the presence of mating pheromone. Assays were performed with MY3265 (KAR4) or MY4166 (kar4Δ) harboring plasmids with deletions in the KAR3 promoter regions. L11 (pMR2806), L16 (pMR2802) and L18 (pMR2892) delimit the KAR4-dependent region to a 30-bp region upstream of the KAR3 gene. Flanking the lines are numbers that correspond to the endpoints of the deletions within the original 560-bp region (23). K3 min (pMR5620) contains the minimal 30-bp KAR4-dependent region indicated by the shaded box. The nucleotide sequence is shown below K3 min. The vertical line at position 485 indicates the endpoints for deletions L14 (pMR2808) and L15 (pMR2896). The results for L11, L14, L15 L16, and L18 were published previously (23). (B) Sequence alignment of upstream DNA from Kar4-regulated genes as identified by AlignAce analysis. Upstream regulatory regions of Kar4-dependent pheromone-induced genes showing a fourfold or greater induction were analyzed for sequence enrichment. See the Materials and Methods for descriptions of the MAP score and the specificity score. (C) Mutational analysis of the Kar4-dependent region. The results of β-galactosidase assays represented as a percentage of the KAR3-minimal region promoter activity (% K3 min) in the presence of mating pheromone are shown. MY3265 harboring plasmids containing mutated (31-2, pMR5621; 41-2, pMR5622) or wild-type (pMR5620) minimal KAR3 upstream regulatory region, as well as the vector control (pLG669-Z, vector), were analyzed for promoter activity. Mutated nucleotides are underlined. The consensus pheromone response element (PRE) is boxed.

FIG. 6.

Temporal and pheromone concentration regulation of Kar4-dependent genes. (A) Genes showing greater than 2.5-fold induction by pheromone from cluster 1 (Kar4-independent [34]) and cluster 2 (Kar4 dependent [29]) were selected, and datum points were taken representing the log₂ (ratio) of normalized expression at 0, 15, 30, 45, and 60 min after induction by pheromone (39). A pheromone response curve was generated for each gene, and the t_1/2 was extracted from the interpolated data. Primary data were from Roberts et al. (39). (B) Box plot representing the concentration of pheromone corresponding to the half-maximal induction of each of the Kar4-dependent and Kar4-independent pheromone response genes. The data determined from the log₂ (ratio) of normalized expression at 30 min after exposure of a bar1Δ strain to various concentrations of pheromone (0.15, 0.5, 1.5, 5, 15.8, 50, and 158 nM α-factor) Primary data were from Roberts et al. (39).

FIG. 7.

Kar4-dependent temporal regulation of PRM2, KAR3, and CIS3. (A) Western blot comparison of pheromone- and galactose-induced Kar4::HA. The MS3216 kar4Δ::HIS3 strain harbored either the wild-type KAR4::HA (pMR2654), P_GAL-KAR4-short::HA (pMR3356), or P_GAL-KAR4-long::HA (pMR3459) plasmid constructs. The cells were grown in the absence (−) or presence (+) of galactose and in the absence (−) or presence (+) of pheromone (α-factor). The arrows indicate the two forms of Kar4::HA. (B to F) Time course of kar4Δ cells (MS3216) harboring plasmid pMR2654 (⧫, wild-type Kar4) or the galactose-induced plasmids pMR3356 (▪, galactose-induced short form of Kar4, P_GAL-Kar4-short) or pMR3459 (▴, galactose-induced long form of Kar4 predominant, P_GAL-Kar4-long). Cells were grown in medium containing galactose, and pheromone was added at t = 0. Cells were harvested at the indicated times, and RNA was prepared. Northern blots were hybridized with probes to either PRM2 (B), KAR3 (D), CIS3 (F), or ACT1 (F). The graphs in panels C and E show the levels of mRNA in arbitrary intensity units.

FIG. 8.

(A) Proposed model of Kar4 action in the pheromone response. During the pheromone response, cytoplasmic Fus3 (Fus3-C) enters the nucleus (Fus3-N), where it phosphorylates Far1 and Dig1 and Dig2, leading to cell cycle arrest, and the activation of Ste12, which in turn induces the early pheromone response genes. Ste12 also induces the expression of the pheromone-specific short form of Kar4, which begins to accumulate in the cell. After sufficient Kar4 has accumulated, it facilitates Ste12 binding to the promoter regions of genes involved in functions required late in the pheromone response. (B) Representation of the feed-forward loop. The diagram was modified from that of Mangan and Alon (29). Coherent type I feed forward loop (AND gate logic) in which transcription factor X is represented by Ste12 and transcription factor Y is represented by Kar4 and the target gene, Z, is represented by KAR3. The inducer of X and Y is pheromone, although other inducers of Kar4 are possible. This type of feed-forward loop is suggested to act as a persistence detector where only persistent stimulus of both X and Y will lead to activation of Z (29). (C) Incoherent type I feed-forward loop (AND gate logic) in which X is represented by Ste12, Y is represented by Kar4, and Z is represented by CIS3. This type of feed-forward loop is characterized by the combined induction of X and Y by a stimulus (pheromone) in order to cause Z to be repressed by Y.