Modulators of MAPK pathway activity during filamentous growth in Saccharomyces cerevisiae

Abstract Mitogen-activated protein kinase (MAPK) pathways control the response to intrinsic and extrinsic stimuli. In the budding yeast Saccharomyces cerevisiae, cells undergo filamentous growth, which is regulated by the fMAPK pathway. To better understand the regulation of the fMAPK pathway, a genetic screen was performed to identify spontaneous mutants with elevated activity of an fMAPK pathway–dependent growth reporter (ste4 FUS1-HIS3). In total, 159 mutants were isolated and analyzed by secondary screens for invasive growth by the plate-washing assay and filament formation by microscopy. Thirty-two mutants were selected for whole-genome sequencing, which identified new alleles in genes encoding known regulators of the fMAPK pathway. These included gain-of-function alleles in STE11, which encodes the MAPKKK, as well as loss-of-function alleles in KSS1, which encodes the MAP kinase, and loss-of-function alleles in RGA1, which encodes a GTPase-activating protein (GAP) for CDC42. New alleles in previously identified pathway modulators were also uncovered in ALY1, AIM44, RCK2, IRA2, REG1, and in genes that regulate protein folding (KAR2), glycosylation (MNN4), and turnover (BLM10). Mutations leading to C-terminal truncations in the transcription factor Ste12p were also uncovered that resulted in elevated reporter activity, identifying an inhibitory domain of the protein from residues 491 to 688. We also find that a diversity of filamentous growth phenotypes can result from combinatorial effects of multiple mutations and by loss of different regulators of the response. The alleles identified here expand the connections surrounding MAPK pathway regulation and reveal new features of proteins that function in the signaling cascade.


Introduction
Organisms can execute a variety of responses when encountering external stimuli.The sensing and response to stimuli are facilitated by signal transduction pathways.One type of evolutionarily conserved signaling pathways are mitogen-activated protein kinase (MAPK) pathways (Saito 2010;Bahar et al. 2023).MAPK pathways are found ubiquitously in eukaryotes and control the response to stress, cell differentiation, cell proliferation, and diverse responses to many stimuli.When misregulated in humans, MAPK pathways can cause diseases (Dhillon et al. 2007).For example, mutations that result in activation of the RAS-RAF-MEK-ERK pathway are a main cause of many types of cancers (Li et al. 2016).Therefore, identifying proteins that regulate MAPK pathways and defining their impact on MAPK pathway activity and function remain at the forefront of studies in this area.
The budding yeast Saccharomyces cerevisiae contains welldefined MAPK pathways (filamentous growth, mating, highosmolarity glycerol (HOG) response, sporulation, and cell wall integrity) that govern the cellular response to external stimuli (Chen and Thorner 2007;Rispail et al. 2009;Saito 2010).In response to limiting nutrients, like availability of carbon or nitrogen sources, the filamentous growth MAPK (fMAPK) pathway regulates a cell differentiation response called filamentous/ invasive/pseudohyphal growth (Gimeno et al. 1992;Kumar 2021).
Filamentous growth is characterized by the formation of elongated cells that undergo distal unipolar budding, which attach and invade into surfaces in a response called invasive growth (Roberts and Fink 1994;Cullen and Sprague 2012).Filamentous growth commonly occurs in fungal pathogens (Rispail et al. 2009;Leach and Cowen 2014) and is required for virulence in some species (Lo et al. 1997).Therefore, studying how a MAPK pathway regulates filamentous growth in S. cerevisiae can provide general insights into this fungal growth response.
Like other MAPK pathways, the fMAPK pathway is regulated by evolutionarily conserved factors that operate in a signaling cascade (Fig. 1a, green).These include sensor proteins (mucin Msb2p and sensors Sho1p and Opy2p; Cullen et al. 2000Cullen et al. , 2004;;Wu et al. 2006;Karunanithi and Cullen 2012;Yamamoto et al. 2015), relay proteins (Rho GTPase, Cdc42;14-3-3 proteins, Bmh1/ 2p;PAK, Ste20p;MAPK cascade;Ste11p;Ste7p;and Kss1p;Stevenson et al. 1992;Leberer et al. 1996Leberer et al. , 1997;;Peter et al. 1996;Roberts et al. 1997), and transcription factors (including Ste12p and Tec1p; Roberts and Fink 1994;Gavrias et al. 1996;Madhani and Fink 1997;van der Felden et al. 2014).In addition to these positive factors, several proteins negatively regulate fMAPK pathway activity (Fig. 1a, red).One of these proteins is Rga1p, the main GTPase-activating protein (GAP) for Cdc42p in the fMAPK pathway  ste4), the FUS1-lacZ and FUS1-HIS3 reporters were used to measure fMAPK pathway activity in WT (PC538), ste12Δ (PC539), and dig1Δ (PC3039) strains.SD + AA media denotes dilution control.SD-HIS and SD-HIS + 10 mM 3-ATA media were used to measure fMAPK pathway activity.WT cells showing normal levels of fMAPK activity can grow on SD-HIS media, whereas ste12Δ mutant failed to grow on SD-HIS media due to downregulation of pathway activity.The dig1Δ mutant showing elevated activity of the pathway can grow on SD-HIS as well as on the SD-HIS + 10 mM 3-ATA media.Isolates that grow on SD-HIS + 10 mM 3-ATA presumably induce the fMAPK pathway.The β-galactosidase assays were performed as described.Experiments were performed in 3 independent replicates.Data were analyzed by 1-way ANOVA followed by a Tukey's pairwise comparison test to generate P-values.Asterisk denotes difference compared to WT and P < 0.05.Error bars denote standard error of the mean.c) Filamentous phenotype is depicted for WT, ste12Δ, and dig1Δ strains.To compare invasive growth, strains were spotted on YEPD media, grown for 3 days at 30°C, and colonies were washed under stream of water.Colonies were photographed before and after washing.To look at cell morphology, cells were scraped and visualized using DIC at 100× magnification (scale bar, 5 microns).Cells were examined by microscopy (100×) to view the changes in cell length and budding pattern in response to glucose limitation.(Smith et al. 2002;Patterson et al. 2021).Another set of proteins are the transcriptional repressors Dig1p and Dig2p (Cook et al. 1996;Madhani et al. 1997;Madhani and Fink 1997;Bardwell, Cook, Voora, et al. 1998;van der Felden et al. 2014).The MAPK Kss1p is itself a bifunctional protein that has both positive and negative regulatory inputs (Cook et al. 1996(Cook et al. , 1997)).The combination of positive and negative regulations of this pathway leads to the appropriate levels of transcriptional induction of target genes and other changes required to bring about the filamentous cell type.
Here, a genetic screen was conducted to isolate spontaneous mutants with elevated activity of an fMAPK pathway-dependent reporter, ste4 FUS1-HIS3, which gives a readout of fMAPK pathway activity in the absence of mating-specific genes (e.g.STE4).Whole-genome sequencing of select mutants identified mutations in expected genes known to impact the activity of the fMAPK pathway.The data validates previous findings that underscore key points of MAPK pathway regulation.The screen also uncovered potentially new points of regulation, including a region in the extreme C-terminus of Ste12p that has an inhibitory effect on the function of this homeobox transcription factor.The results furthermore reveal a diversity of phenotypes that presumably arise from combinations of mutations that impact the activity of the MAPK pathway.

Yeast strains, media, and growth conditions
Strains used in the study are haploids derived from the Σ1278b strain background (Liu et al. 1993).Strains are listed in Supplementary Table 1 and primers in Supplementary Table 2. Cells were grown at 30°C in YEPD, composed of yeast extract (1%), peptone (2%), and dextrose (2%), or synthetic complete (SD) media containing 0.67% yeast nitrogen base (Sigma-Aldrich, St. Louis, Missouri) and 2% glucose.Amino acids were added to the SD media as required.SD + AA denotes media containing all amino acids, SD-HIS denotes media containing all amino acids without histidine, and SD-URA denotes media containing all amino acids and lacking uracil.The His3p inhibitor 3-amino-1,2,4triazole (3-ATA; Klopotowski and Wiater 1965;Struhl and Davis 1977) was added to final concentrations of 2.5, 5, and 10 mM as indicated.To make S-GAL and YEP-GAL media, 2% galactose was used instead of 2% dextrose.SLAD media was made as described (Gimeno et al. 1992).Yeast strains were manipulated using standard techniques (Rose et al. 1990).Gene deletions were made by polymerase chain reaction (PCR)-based strategies as described (Longtine et al. 1998).To make the Ste12p G491* strain in the PC313 background, primers were designed with homology to the STE12 gene adding a NATMX6 drug resistance cassette to introduce a premature stop codon at the G491 position.Primers were used to amplify the cassette by PCR followed by transformation into a wild-type (WT) yeast strain and selection for NAT+ colonies (Goldstein and McCusker 1999).The integration was confirmed by PCR Southern analysis.The Ste12p G491*, Kss1p W277*, and Rga1p S679* mutations in PC538 background were made by a similar method using the KlURA3 cassette (Cullen et al. 2004).The pFUS1-lacZ (Basu et al. 2020) and pFRE-lacZ plasmids (Madhani and Fink 1997) have been described.
The ste4 mutation in the parent strain (PC538) was made using an integrating plasmid pSL1851 (CY2273), a pRS306-based vector (Schneider et al. 1995;Stevenson et al. 1995).The ste4 mutation was made by introducing URA3 gene at the STE4 gene locus and selection for URA3+ colonies to select for pRS306 followed by removal of the marker by counterselection on 5-FOA media.
Although it was assumed that the STE4 gene was deleted, a mutated form of the gene may be retained in the genome which underwent repair to form a functional gene in several mutants.The alleles of STE4 recovered in the mutant strains included the following insertions: TAGCTTGCATCTTGTATC, TAGCTTGCATCT TGTATCTTGTGTTTTGC, TAGCTTGCATC, and TAGCTTGCATC TTGTATCTTGTG.The WT allele in the parent strain is TAGCTTGC ATCTTGTATC, and the reference allele is T. The inserted nucleotides may cause frameshifts that result in a functional STE4 gene.

Genetic screen and characterization of mutants
An otherwise WT strain lacking the STE4 gene (PC538) was streaked onto SD + AA media for 2 days at 30°C.Individual colonies were used to inoculate 5 mL of YEPD media in separate tubes.Cells were grown for 16 h at 30°C with continuous orbital shaking and were harvested (400 μL) by centrifugation.After discarding the supernatant, cell pellets were resuspended in 100 μL of autoclave-sterilized water.Cell suspension was top spread on SD-HIS + 10 mM 3-ATA or SD-HIS + 5 mM 3-ATA media, and plates were incubated at 30°C for 3 days.Individual isolates presumably containing spontaneous mutations were collected and restreaked on SD-HIS + 10 mM 3-ATA media and grown at 30°C for 3 days.Strains were scraped and collected in 50% glycerol and were frozen at −80°C.Mutants were named based on the preculture number and degree of resistance to 3-ATA.Approximately 1 × 10 8 cells were screened for spontaneous mutations that hyperactivate the fMAPK pathway.Fifty-nine separate experiments were performed to identify independent isolates.
The activity of the fMAPK pathway was measured in mutants using the FUS1-lacZ reporter.Mutants were further characterized by screening for changes in invasive growth pattern by the platewashing assay (PWA), for changes in cell morphology by differential interference microscopy (DIC), and for changes in salt sensitivity by growth on media containing 1 M KCl.Mutants were ranked on a scale of 1 (no growth on salt) to 5 (robust growth on salt; Supplementary Table 3, column H).The PWA was performed as described (Roberts and Fink 1994).Mutants were scored on a scale from 1 to 5 (1, as invasive as ste12Δ; 3, as invasive as WT; 5, as invasive as dig1Δ; Supplementary Table 3, column J).For DIC images, invasive cells left behind after the PWA washes were observed at 100× using DIC filters on the Axioplan 2 fluorescence microscope (Zeiss, Jena, Germany) with a PLAN-APOCHROMAT 100×/1.4(oil) objective (N.A. 0.17; Zeiss, Jena, Germany).The single-cell invasive growth assay was performed as described (Cullen and Sprague 2000).WT cells (PC538) and the ste12Δ (PC539) and dig1Δ (PC3039) mutants were used as control strains.
β-Galactosidase assays β-Galactosidase assays were performed as described (Cullen et al. 2004).Cells were grown at 30°C overnight (16 h) in YEPD.Next day, 400 μL of overnight culture was centrifuged, and supernatant was discarded.The pellet was washed three times and resuspended in 100 μL of autoclave-sterilized water.The resuspension was then used to inoculate fresh 10 mL of YEPD media.Cells were grown to mid-log phase (A 600 ∼1.0), and 1 mL of cell culture was centrifuged; pellets were stored at −80°C.Next day, pellets were resuspended in 100 μL Z-buffer [44.32 mL H 2 O with 5 mL phosphate buffer (0.6 M Na 2 HPO 4 + 0.4 M Na 2 HPO 4 ), 0.5 mL 1 M KCl, 50 μL 1 M MgSO 4 , and 135 μL β-mercaptoethanol].Once resuspended, 2 μL of 5% sarkosyl (S) and 2 μL of toluene were added, and the pellets were incubated at 37°C for 30 min with open caps to allow for evaporation of toluene.Then, Z + O buffer containing ortho-nitrophenyl-β-galactosidase was added to the pellets.After color change was observed, reactions were stopped by adding 250 μL of 1 M Na 2 CO 3 , and reaction time was recorded.Reaction mixture was centrifuged (13,000 rpm for 3 min) to remove the cell extract, and 200 μL of supernatant was used to determine A 420 .(1000 × A 420 )/(A 600 × time) to calculate Miller units.Miller units demonstrated in graphs are averages of 3 or more independent experiments along with the standard error of the means.
For pheromone induction experiments using FUS1-lacZ, 5 mL of mid-log phase (A 600 ∼ 1.0) cultures were pelleted, washed twice with sterile water, and transferred to fresh 5 mL media containing 50 μL of α-factor (stock concentration: 1 mg/mL; induced condition) or 50 μL of water (uninduced condition).Cells were further grown for 2.5 h, and 1 mL of culture was pelleted and stored at −80°C.β-Galactosidase assays were performed on the pellets as described above.

Genomic DNA extraction, sequencing analysis, and identification of variants
For generating chromosomal DNA for sequencing analysis, the Gentra Puregene Yeast Bacteria DNA extraction kit was used (Qiagen, Hilden, Germany).Cells were grown for 16 h in 5 mL cultures and prepared according to the manufacturer's protocols.Sequencing and variant calling were performed at the Genomics & Bioinformatics shared resource of the Fred Hutchinson Cancer Center in Seattle, Washington.Genomic DNA was quantified using Life Technologies' Invitrogen Qubit 2.0 Fluorometer (Thermo Fisher, Waltham, MA) and fragmented using a Covaris LE220 ultrasonicator (Covaris, Woodburn, MA) targeting 400 bp.Sequencing libraries were prepared using 100 ng fragmented DNA with the KAPA HyperPrep Library Prep Kit (Roche, Indianapolis, IN) and NEXTFLEX UDI barcodes (PerkinElmer, Waltham, MA).Library quantification was performed using Life Technologies' Invitrogen Qubit 2.0 Fluorometer and size distribution validated using an Agilent 4200 TapeStation (Agilent Technologies, Santa Clara, CA).Individual libraries were pooled 30-plex at equimolar concentrations and sequenced on 1 lane of an Illumina HiSeq 2500 (Illumina, Inc., San Diego, CA) employing a paired-end, 50-base read length sequencing configuration.This yielded 3-5.5 M (average 4.5 M) read pairs covering the ∼12 Mb S. cerevisiae genome per sample.
Variant sites among all 34 samples (includes a pooled sample) with respect to the Σ1278b reference genome were identified (Supplementary Table 4a).To identify differences between WT and the mutant lines, results were filtered to 556 variants where at least 1 mutant line had a genotype different from both the reference genome and the WT.The resulting variant calls (in VCF format) were compiled into a single table providing an overview of detected variants (Supplementary Table 4b, one row per variant).A custom database was constructed for running snpEff, a tool to annotate variants, exported in html format.snpEff takes GATK-generated VCF file and generates a new VCF file with updated INFO field (snpEff.annotation/filtered.ann.vcf).The format is described in http://snpeff.sourceforge.net/SnpEff_manual.html#input.The relevant information was consolidated to snpEff.annotation/filtered.variants.annotation.xlsx.Each tab "legend" shows a description of each field.Intragenic variants identified in each mutant were compiled into a single sheet to show the genes, alternate alleles, amino acid change, and mutants that contain the alternate alleles (Supplementary Table 5).

Phospho-MAP kinase analysis
Cells were inoculated in YEPD or SD + AA media (5 mL) and grown for 16 h at 30°C with continuous shaking.Approximately 750 µL of cells from the saturated culture grown for 16 h were inoculated into 10 mL SD + AA media, and cells were grown until the culture reached mid-log phase (A 600 ∼1.0).Five milliliters of cells were harvested from mid-log phase cultures for immunoblot analysis.Cells were disrupted, and proteins were enriched by trichloroacetic acid precipitation as previously described (Basu et al. 2016).Protein samples were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis and were transferred from gels to a nitrocellulose membrane (Cat#10600003, Amersham Protran Premium 0.45 μm NC; GE Healthcare Life sciences).Membranes were probed with rabbit polyclonal p44/42 antibodies (Cell Signaling Technology, Danvers, MA; Cat #4370) diluted 1:10,000 in 5% bovine serum albumin (BSA) to detect P∼MAP kinases, P∼Fus3p and P∼Kss1p.Monoclonal mouse anti-Pgk1p antibodies (Life Technologies; Camarillo, CA; Cat #459250) were used at 1:10,000 dilution as a control for total protein levels.Secondary anti-mouse IgG-HRP (Cat#1706516; Bio-Rad, Inc.) and goat antirabbit IgG-HRP (Cat#115-035-003; Jackson ImmunoResearch Laboratories) were used to detect the primary antibodies.The nitrocellulose membrane was blocked with 5% nonfat dried milk for Pgk1p antibody or 5% BSA for the p44/42 antibody for 1 h prior to antibody incubation.Primary antibody incubations were performed at 4°C for 16 h, and secondary antibody incubations were performed at 22°C for 1 h.Immunoblots were visualized by Gel Doc XR Imaging System (Bio-Rad, Inc.), after addition of Chemiluminescent HRP substrate for chemiluminescent Westerns (Radiance Plus Substrate, Azure Biosystems).Raw images are included for the blots shown (Supplementary Fig. 4-a, membrane probed with p44/42 antibodies, and b, membrane probed with Pgk1p antibodies).
Densitometric analysis was performed with Image Lab Software (Bio-Rad).The same exposures in the linear range were used for measuring band intensities.Background subtraction was performed according to the guidelines provided by the manufacturer.Band intensities of phospho-proteins were normalized against total protein levels based on Pgk1p band intensity.

Halo assay
Halo assays were performed as described (Sprague et al. 1983).One milliliter of mid-log phase (A 600 ∼1.0) cells were serially diluted, and the 1:100 dilution was top spread on YEPD semisolid agar media.Plates were dried for 5 min, and 5 and 10 μL α-factor (stock concentration, 1 mg/mL) was spotted on the surface.Plates were incubated at 30°C for 2 days and photographed.

Statistical analysis
All statistical tests were performed with Minitab, LLC.2021, retrieved from https://www.minitab.com.Data were analyzed by 1-way ANOVA test followed by a Tukey's or Dunnett's (control group, WT) multiple comparison tests to generate P-values.

Strategy to identify modulators of the fMAPK pathway
The fMAPK pathway is composed of proteins that promote (Fig. 1a, green) or inhibit (Fig. 1a, red) MAPK pathway activity.In cells lacking an intact mating pathway (ste4), transcriptional reporters FUS1-HIS3 and FUS1-lacZ (Fig. 1b, both integrated into the genome) provide a readout of fMAPK pathway activity (Cullen et al. 2004).With respect to the FUS1-HIS3 reporter, WT cells (PC538) grow on SD-HIS media, whereas cells lacking an intact fMAPK pathway [e.g. the ste12Δ (PC539) mutant] do not.By comparison, cells lacking negative regulators of the pathway, such as the dig1Δ (PC3039) mutant, grow on media lacking histidine and containing 3-ATA, which is a competitive inhibitor of His3p enzyme (Fig. 1b).Therefore, mutants that grow on SD-HIS + 3-ATA media presumably show elevated fMAPK pathway activity.Likewise, WT cells show basal FUS1-lacZ reporter activity by β-galactosidase assays (Fig. 1b).The activity of this reporter is also dependent on the transcription factor Ste12p and is modulated by the transcriptional repressor Dig1p (Fig. 1b).
To observe aspects of filamentous growth (invasion, morphology, and budding pattern), WT cells along with the ste12Δ and dig1Δ mutants were spotted onto YEPD media and grown for 3 days at 30°C.Colonies were washed off the plate to reveal invasive scars (Fig. 1c, washed) by the PWA (Roberts and Fink 1994).The ste12Δ mutant showed less agar invasion than WT cells, whereas the dig1Δ mutant showed elevated invasive growth (Fig. 1c, washed).Microscopic examination showed that WT cells were elongated and remained attached to each other, whereas ste12Δ mutant cells were round and separated from each other.The dig1Δ mutant cells showed increased cell elongation (Fig. 1c, scraped).By the single-cell assay (Cullen and Sprague 2000), WT cells showed distal pole budding and increase in cell length, whereas the ste12Δ mutant showed a decrease in cell length.The dig1Δ mutant cells showed hyper cell polarization (Fig. 1c).Therefore, the filamentous growth phenotypes controlled by the fMAPK pathway parallel transcriptional reporter activity.

Genetic screen and characterization of mutants
A genetic screen was performed to identify genes that regulate the activity of the fMAPK pathway reporter (ste4 FUS1-HIS3).Expected mutations include loss-of-function mutations in genes encoding negative regulators and gain-of-function mutations in genes encoding positive regulators.Spontaneous mutants were isolated that were capable of growth on SD-HIS + 10 mM 3-ATA media.In total, 159 mutants were identified in 59 separate batches and were phenotypically characterized (Supplementary Table 3).Separate batches were used to eliminate duplicate mutations.Of the 159 mutants, 144 grew on SD-HIS + 10 mM 3-ATA media and 15 grew on SD-HIS + 5 mM 3-ATA media (Supplementary Table 3, column D).To eliminate potential contaminants, isolates were streaked on SD-URA media.No URA+ isolates were uncovered (Supplementary Table 3, column K), indicating that the mutants were derived presumably from the ura3-parent strain (Supplementary Fig. 1a).
Examining the mutants on the same media showed that some mutants grew better on SD-HIS + 10 mM 3-ATA than others (Fig. 2, inset).To quantitate MAPK pathway levels, the activity of the FUS1-lacZ reporter was determined by β-galactosidase assays.Most mutants showed elevated levels of reporter activity (Fig. 2, bar graph).Mutants that did not show elevated FUS1-lacZ activity may contain mutations in the ATR1 gene, which causes 3-ATA resistance or mutations in the FUS1-HIS3 promoter, and were not considered further.
Mutants were further screened by additional tests.To eliminate known mutations in the HOG pathway that are known to cause activation of the fMAPK pathway (specifically pbs2 and hog1; O'Rourke and Herskowitz 1998), mutants were screened for salt sensitivity by growth on media containing 1 M KCl.Mutants showed varying levels of salt sensitivity (Supplementary Fig. 1b and Table 3, column H).Mutants showing a growth defect on salt were not selected for further analysis.
Mutants were also examined by the PWA to identify changes in invasive growth (Fig. 3; the complete data set can be found in Supplementary Fig. 2a-d; Supplementary Table 3, column J).Invasive cells were also examined by microscopy (Fig. 3, right panels; the complete data set can be found in Supplementary File 1; a representative image is shown in Supplementary Table 3, column F).A subset of mutants showed growth defects on YEPD media (Supplementary Table 3, column G), which might arise due to strong hyperactivity of the fMAPK pathway.In lieu of dominance/recessive and complementation tests, which are powerful tools to identify unique gene categories, we elected to sequence alleles based on a rigorous phenotypic characterization.Thus, our approach may miss some unique classes of genes.To capture a broad collection of genotypes, a phenotypically diverse collection of mutants was selected for whole-genome sequencing analysis (Supplementary Table 3, column O).

Whole-genome sequencing and bioinformatics analysis to identify variants
For sequencing analysis, DNA was harvested from 32 mutants and the WT control strain by chromosomal preparation.DNA sequencing was performed with an Illumina HiSeq 2500 machine, which yielded an average of 4.5 M read pair coverage over the yeast genome.The total number of variants identified compared to the reference genome was 2,131 (Supplementary Table 4a).The variant rate was 1 variant every 4,943 base pairs and included single nucleotide polymorphisms, additions, and deletions.Overall, 196 missense, 21 nonsense, and 461 silent mutations were recovered.The missense to silent ratio was 0.4252.
To better define the relevant changes in the mutants, the variations in each mutant were compared to somatic variants that differed between the WT parental strain and the reference genome, Σ1278b.As a result, 557 mutations were identified across the mutants sequenced (Supplementary Table 4b).The mutations mapped to ORFs as well as intergenic regions upstream and downstream of the ORFs (Supplementary Table 4b).We focused on the mutations present in exons (Supplementary Table 5), which revealed a set of mutations expected to impact the activity of the fMAPK pathway.The mutations identified validate the utility of the screen by recovery of expected mutations.Many of the alleles (the term mutation and allele are used interchangeably) remain uncharacterized and could contribute to the phenotypic variation observed here.In addition, structural variation in genes (repeat regions) may not have been detected by short-read sequencing and might also contribute to phenotypic variation.These features could be examined in future analysis.

Alleles uncovered by analysis of DNA sequencing data
The screen identified mutations in genes previously known to regulate the fMAPK pathway (Table 1).These included mutations Modulators of MAPK pathway activity | 5 in four regulators of the pathway: Ste11p, Rga1p, Kss1p, and Ste12p (Fig. 1, daggers).Multiple alleles in each of the genes were uncovered, and each gene is discussed in detail below.

STE11
Six mutants contained mutations in the STE11 gene (Table 1).One of these mutants contained 2 missense mutations that led to changes at the P279H and P279A positions.A previous study uncovered an allele that resulted in the P279S change, called STE11-1 (Stevenson et al. 1992), which hyperactivated the MAPKKK.Another gain-of-function mutation was uncovered that resulted in the change T596I in the kinase domain of Ste11p (Fig. 4a), which was previously uncovered as the strong hyperactive STE11-4 allele (Stevenson et al. 1992).Three new mutations were also identified in the central region of the protein upstream of the kinase domain that would be predicted to change the following residues, L440Q, L300S, and F292I (Fig. 4a).Although not tested, these changes might also be expected to impact the activity of the Ste11p protein.

RGA1
Nine mutants contained mutations in the gene encoding Rga1p, the main GAP for Cdc42p that functions in the fMAPK pathway (Table 1).Eight of the mutations had either frameshifts (S679Fr) or premature stop codons (Fig. 4a, Y650*, E663*, Q713*, E758*, C797*, E803*, and L859*).These mutations would be expected to remove the C-terminal GAP domain.Loss of GAP activity is expected to result in elevated Cdc42p activity, resulting in higher levels of Cdc42p-dependent fMAPK pathway activity (Stevenson et al. 1995; see Supplementary Fig. 5a).One missense mutation was uncovered resulting in the amino acid substitution G826S, which changed a conserved glycine residue found in other Rho GAPs including Bem2p, Bem3p, and Lrg1p (Stevenson et al. 1995).Therefore, this change might also be expected to compromise GAP activity of the Rga1p protein and cause elevated reporter activity.

KSS1
Six mutants had mutations in the gene encoding the MAPK Kss1p (Table 1).Four were premature stop codons (Fig. 4a, Y312*, W277*, Y185*, and L178*).Two mutants contained the same missense mutation that led to the change Q126K.Loss-of-function alleles in Kss1p are expected because Kss1p has both positive and negative regulatory functions (Cook et al. 1997;Bardwell, Cook, Voora, et al. 1998).As shown below, loss of Kss1p function is known to result in elevated phosphorylation of a related MAP kinase, Fus3p, which would also be expected to induce fMAPK pathway reporter activity (Fig. 4; Supplementary Fig. 5b).

STE12
Eight mutants contained mutations in the gene encoding the transcription factor Ste12p.All of these mutations resulted in premature stop codons that would be expected to cause C-terminal truncations of the protein.Five mutants contained G491*, 2 mutants contained Q500*, and 1 mutant contained G587*.Loss of Ste12p would be expected to result in reduced reporter activity (Fig. 1b).Therefore, loss of the C-terminal domain of Ste12p is presumably inhibitory, leading to elevated fMAPK pathway activity (see below).

Other genes
Twelve mutants contained mutations in STE4, which might explain their hyperactive phenotypes.STE4+ revertant mutants were presumably recovered by repair of the STE4 gene.Overall, of the 32 mutants sequenced, 31 mutants had mutations in one of the abovementioned genes (STE11, RGA1, KSS1, STE12, and STE4), which could explain the increased reporter activity found in these strains.The remaining mutant contained a premature stop codon, S502*, in BLM10, which encodes a nonessential gene Fig. 2. Genetic screen identifies mutants that hyperactivate the fMAPK pathway.Inset: mutants that show hyperactivity of the FUS1-HIS3 reporter; a subset of 55 mutants that grow on SD-HIS + 10 mM 3-ATA media are shown.Validation of fMAPK pathway activity by the FUS1-lacZ reporter.WT (PC538), ste12Δ (PC539), and dig1Δ (PC3039) strains were used as controls.Experiments were performed in 3 independent replicates.Data were analyzed by 1-way ANOVA test followed by a Dunnett's multiple comparison test to generate P-values.Asterisk denotes significant difference compared to WT and P < 0.05.Error bars denote standard error of the mean.Differences in lacZ levels between experiments (e.g.Fig. 1b) may result from different growth conditions.The 32 indicated mutants were chosen for sequencing analysis.Colors represent mutations in the indicated genes: red, KSS1; green, RGA1; blue, STE11; and orange, STE12.
that functions as the activator of the 19S proteasome (Dange et al. 2011).The proteasome is required for turnover of Cdc42p (Gonzalez and Cullen 2022) and Ste20p (Gonzalez et al. 2023), which are positive regulators of the fMAPK pathway.Therefore, cells lacking the ability to degrade these proteins might show elevated activity of the fMAPK pathway.
Although multiple alleles of STE11, RGA1, KSS1, and STE12 were uncovered in the screen, each mutant contained a mutation in only one of these genes.Multiple mutations may cause growth defects that prevent detection by this approach.Comparing mutants that contained different alleles of abovementioned genes revealed unexpected variation in reporter activity (Fig. 2, compare orange Fig. 3. Secondary screens of mutants for invasive growth and cell morphology phenotypes.PWA was used to assess invasive growth phenotypes.Mutants were spotted in 5 μL volume on YEPD media and incubated at 30°C for 3 days.Plates were washed under a stream of water to reveal invasive scars.Photographs were taken before and after washing.The full data set is shown in Supplementary Fig. 2a-d.Invaded cells were scraped off from washed plates by a toothpick and observed at 100× magnification using DIC filter.Bar, 10 microns.A typical example of WT (PC538), ste12Δ (PC539), and dig1Δ (PC3039) from the experiment are shown as controls.Colors represent mutations in the indicated genes: red, KSS1; green, RGA1; blue, STE11; and orange, STE12.
to orange, etc.) and in filamentous growth phenotypes (Fig. 3, compare mutants of the same color).We speculate that this may be due to allele-specific differences for a given gene.Alternatively, phenotypic differences may arise due to mutations in multiple genes that generate the hyperactive phenotype, including mutations in intergenic regions that affect the expression of genes, although we have not tested this possibility.Variation in phenotype caused by alleles of the same gene may also be due to epigenetic factors that lead to stochastic differences within strains (Kaern et al. 2005;Zhang et al. 2013).
Mutations in several other negative regulators were identified.A mutation in ALY1 resulting in N224K amino acid change was uncovered.When overexpressed, the gene encoding the alpha arrestin, ALY1 (O'Donnell et al. 2013), dampens the fMAPK pathway activity due to mislocalization of the sensor protein Msb2p (Adhikari et al. 2015).A mutation in AIM44/GPS1 resulting in E150K amino acid was also uncovered.Loss of AIM44, a member of the negative polarity complex (Meitinger et al. 2014), is known to cause hyperactivity of the fMAPK pathway (Prabhakar et al. 2020).A mutation in RCK2 resulting in R216L amino acid change was also identified.RCK2 encodes a calmodulin-like kinase and a target of Hog1p (Bilsland-Marchesan et al. 2000;Teige et al. 2001), which dampens fMAPK pathway activity when overexpressed (Pitoniak et al. 2009).We also uncovered mutations in genes involved in protein folding (KAR2, G212R; Rose et al. 1989) and glycosylation (MNN4, V465L; Odani et al. 1996) in the endoplasmic reticulum.Defects in protein folding and glycosylation of Msb2p are known to hyperactivate the fMAPK pathway (Cullen et al. 2000;Pitoniak et al. 2009;Adhikari, Vadaie, et al. 2015;Jamalzadeh et al. 2020).A mutation in IRA2 resulting in T534N was uncovered.The IRA2 gene encodes a GAP for the GTPase Ras2p (Halme et al. 2004).Loss of IRA2 would be expected to increase Ras2p activity, which itself is a positive regulator of the fMAPK pathway (Mosch et al. 1996).A mutation in REG1 resulting in C475Y was uncovered.REG1 encodes a regulator of type I protein phosphatase Glc7p (Sanz et al. 2000).The REG1-GLC7 phosphatase complex is also known to negatively regulate glucose repressible genes.Glc7p is known to regulate filamentous growth (Cullen and Sprague 2002), and the mutations identified here may also impact fMAPK pathway activity.
Several other alleles were also identified and tested for a role on fMAPK pathway reporter activity.Gene deletions were constructed (nam7Δ, aly1Δ, bnr1Δ, esc1Δ, yke2Δ, pac10Δ, ssb1Δ, ena1Δ, tor1Δ, and snf5Δ) and tested for altered levels of FUS1-HIS3 reporter activity.None of the gene deletion mutants tested showed hyperactivity of the pathway based on this test (Supplementary Fig. 3).Apart from the abovementioned mutations that were recovered in nonessential genes, several mutations were identified in essential genes (Supplementary Table 5, column J).The roles that essential genes play in filamentous growth and fMAPK pathway activity are being explored in a separate study (Pujari et al., in preparation).Mutants with the same superscript contain mutations in more than 1 known regulator of the pathway. b The mutation causing substitution T596I was previously found using a similar approach as the dominant gain-of-function allele STE11-4. c The mutation causing substitution P279S was previously found using a similar approach as the dominant gain-of-function allele STE11-1.Two mutations at the same P279 position were found in the mutant resulting in unresolved amino acid change to H or A.

Examination of select mutants by P∼Kss1p analysis
To validate and extend these findings, the activity of the fMAPK pathway was measured in select mutants by phospho-immunoblot analysis (Fig. 4b).The level of phosphorylated Kss1p (P∼Kss1p) provides a readout of fMAPK pathway activity.The same antibodies (anti-p44/42) that recognize P∼Kss1p also recognize P∼Fus3p, which mainly regulates the mating pathway.The mating and the fMAPK pathways execute different responses yet share some components (Roberts and Fink 1994).Because several genes identified here are shared among MAPK pathways, the alleles identified in this study could have an effect on several MAPK pathways.The WT (PC538), ste11Δ (PC611), and STE11-4 (PC580) strains were used as controls to observe basal, defective, and hyper P∼Kss1p/P∼Fus3p levels, respectively.When comparing mutants to WT and the ste11Δ control, all the mutants showed hyperphosphorylation of either Kss1p or Fus3p or both MAP kinases, suggesting that mutations in these mutants lead to hyperactive MAPKs (Fig. 4b).The mutant F22c3 containing T596I, present in the kinase domain of Ste11p (Fig. 4a) and which is the same substitution in Ste11-4p (Stevenson et al. 1992), caused similar hyperactivation of P∼Kss1p and P∼Fus3p as seen in STE11-4 control (Fig. 4b, compare lanes 3 and 14).Like the Ste11-1p variant (P279S) identified previously (Stevenson et al. 1992), the mutant I36c3 (P279H/A) showed hyperphosphorylation of Kss1p and Fus3p (Fig. 4b, lane 7).Different alleles leading to specific amino acid substitutions in Ste11p and present in different mutants showed unique patterns of P∼Kss1p and P∼Fus3p levels (E20c1, L300S; I36c3, P279H/A; L53c1, L440Q; D14c2,F292I;and F22c3,T596I;Fig. 4b,compare lanes 5,7,10,12,and 14).Like the STE11 alleles, the RGA1 allele leading to C-terminal truncation of the protein (E758*) present in the K50c2 mutant showed hyperphosphorylation of Kss1p and Fus3p (Fig. 4b, lane 9).Interestingly, all 3 mutations in the KSS1 gene leading to premature stop codons, L178* and Y185*, present in B3c3 and E16c2, respectively, and Q126K, present in A2c1, show low levels of P∼Kss1p and high levels of P∼Fus3p (Fig. 4b, compare lanes 4, 11, and 13).The mutations leading to truncated Kss1p might cause reduced protein levels, which would explain loss of P∼Kss1p in those mutants.However, it is unclear how the missense mutation Q126K causes a defect in P∼Kss1p levels.Because Kss1p is known to have a negative regulatory function in the fMAPK pathway (Cook et al. 1997;Bardwell, Cook, Voora, et al. 1998), loss of Kss1p might lead to Fus3p-dependent activation of FUS1 reporter.In support of this idea, loss of Kss1p did not result in hyperinvasive growth in these mutants by the PWA (Fig. 3).Two mutants, H33c1 and K48c1, containing the same mutation that leads to premature stop codon at the G491 position in Ste12p, also caused similar hyperphosphorylation of Kss1p and Fus3p (Fig. 4b, compare lanes 6 and 8).This may indicate that the C-terminal domain of Ste12p negatively regulates fMAPK pathway activity.Underlined mutations were made in the parent strain and were checked for effect on fMAPK pathway activity using the FUS1-HIS3 and FRE-lacZ reporters (see Fig. 5 and Supplementary Fig. 5).b) Immunoblots of membranes probed with p44/42 antibodies to visualize P∼Kss1p and P∼Fus3p levels, and Pgk1p antibodies as a control for total protein levels.WT (PC538), ste11Δ (PC611), and STE11-4 (PC580) strains were used as standards.Band intensity was quantified using Image Lab.Phospho-band intensities were normalized to Pgk1p bands for each lane, and the WT phospho-protein levels were set to 1 to compare P∼MAP kinases in different mutants with respect to WT. Colors represent mutations in the indicated genes: red, KSS1; green, RGA1; blue, STE11; and orange, STE12.

The C-terminal domain of Ste12p negatively regulates aspects of mating and filamentous growth
The fact that the C-terminus of Ste12p may function in an inhibitory manner was unexpected.To confirm this result, a truncated version of Ste12p, based on the variant G491*, was generated in WT cells.Ste12p G491* showed hyperactivity of the FUS1-HIS3 reporter on SD-HIS media containing 3-ATA (Supplementary Fig. 5c).To examine the inhibitory role of the C-terminus, the Ste12p G491* strain was examined for mating and filamentous growth phenotypes.Ste12p G491* showed enhanced halo formation compared to WT cells in response to α-factor (Fig. 5a, α-factor), which provides a readout of Ste12p function in the mating pathway (Sprague et al. 1983;McCaffrey et al. 1987; Fields et al.G491* (PC7923) cells were spread on YEPD plates, and α-factor was applied at 5 μL (left) and 10 μL (right).Plates were incubated at 30°C for 2 days and photographed.At the right, FUS1-lacZ levels in indicated strains; all strains are STE4+.Minus sign, without pheromone; plus sign, with pheromone.Deviation was less than 10% between samples.Data were analyzed by 1-way ANOVA followed by a Tukey's pairwise comparison test to generate P-values.Asterisk denotes significant differences from WT (P < 0.05).b) Invasive growth.Ten microliters of cells in (a) were spotted onto YEPD media.The plate was incubated at 30°C for 3 days, and colonies were washed under a stream of water to reveal invasive scars.Photographs were taken before and after washing.At the right, typical examples of scraped cells at 100×.Bar, 5 microns.At the right, FRE-lacZ levels in indicated strains.Deviation was less than 10% between samples.Data were analyzed by 1-way ANOVA followed by a Tukey's pairwise comparison test to generate P-values.Asterisk denotes significant differences from WT (P < 0.05).c) Cell morphology in different media conditions.Strains from (a) were grown to mid-log phase (A 600 ∼1.0) and transferred to the indicated media and grown further for 5 h.Liquid cultures were looked at 100× using DIC filter.1988).Ste12p G491* cells also showed elevated basal FUS1-lacZ activity compared to WT (Fig. 5a, minus sign, no pheromone).For WT cells, the expected increase of reporter activity was seen in response to pheromone (plus sign, with pheromone), with no additional increase by Ste12p G491*.This may mean that removal of the C-terminal inhibitory region leads to constitutive activity of Ste12p.Thus, the region might be involved in inhibiting basal gene expression by the protein.Ste12p G491* also showed elevated invasive growth with a unique pattern (Fig. 5b, washed).Invasive cells showed hyperpolarized growth and unusual morphologies compared to WT cells by microscopy (Fig. 5b, scraped, arrow), which was similar to the mutants containing the G491* mutation (orange mutants, Scraped, Fig. 3).Ste12p G491* cells also showed elevated activity of the fMAPK reporter FRE-lacZ (Fig. 5b).Microscopic examination of cell morphology under different conditions showed that the Ste12p G491* mutant exhibited hyperpolarized cell morphology in SLAD and YEP-GAL media compared to SD and S-GAL media in reference to WT cells (Fig. 5c).The FRE-lacZ reporter was also 4.5-fold higher in SLAD media and 3.5-fold higher in YEP-GAL as compared to 2.2-fold in S-GAL and 1.7-fold in SD media (Fig. 5d).These results indicate that the C-terminus of Ste12p has an inhibitory role in regulating the mating and invasive growth pathways (Fig. 5e).These results also reveal conditionspecific inhibition of Ste12p function by its C-terminal domain.In line with these results, the extreme C-terminus of Ste12p has previously been shown to be dispensable for mating (Errede and Ammerer 1989;Kirkman-Correia et al. 1993).Moreover, a more extended deletion of Ste12p (Δ383-688; Kirkman-Correia et al. 1993) had a positive effect on mating.Therefore, amino acids from M383 to G491 may function as a positive regulatory region (PRR; Fig. 5e) while amino acids from G491 to the final residue in the C-terminus, T688, may function as a negative regulatory region (NRR, Fig. 5e).

Comparison of fMAPK pathway activity to invasive growth and cell morphology
Many of the mutants identified in the study showed higher FUS1-lacZ reporter activity than WT (Fig. 2, asterisk denotes significant difference).However, FUS1-lacZ reporter activity did not always correlate with invasive growth (Fig. 3) or P∼Kss1p levels (Fig. 4).Similarly, the polarized morphology of filamentous cells did not always correlate with invasive growth (Supplementary Fig. 2a-d and File 1).Therefore, we directly compared the effects of known hyperactive mutations on invasive growth, cell morphology, and FUS1-lacZ reporter activity (Fig. 6).All the hyperactive mutants tested showed upregulation of FUS1-lacZ reporter.The rga1Δ mutant showed a modest increase in reporter activity compared to the STE11-4, pmi40-101, and dig1Δ mutants (reporter activity in ascending order: ste12Δ < WT < rga1Δ < pmi40-101 < dig1Δ < STE11-4).By comparison, the dig1Δ mutant was more invasive than other hyperactive mutants, and the WT and ste12Δ controls (invasion in ascending order: ste12Δ < pmi40-101 < WT = STE11-4 < rga1Δ < dig1Δ).When comparing cell morphology, the dig1Δ mutant showed the most cell polarization compared to other mutants and controls (cell polarization in ascending order: pmi40-101 < ste12Δ < WT < STE11-4 = rga1Δ < dig1Δ).Thus, the output phenotypes of filamentous growth in the hyperactive mutants did not match each other or the activity of the fMAPK pathway.This may be because different regulators of the fMAPK pathway might control different aspects of filamentous growth, and genes that regulate fMAPK activity (such as PMI40 and STE11) have other functions, which may affect filamentous growth; PMI40 is mannose-6-phosphate isomerase and is required for protein mannosylation (Smith et al. 1992), and STE11 is a shared component with the HOG and the mating pathways (Zhou et al. 1993;Posas and Saito 1997).The fact that different hyperactive alleles lead to different output phenotypes makes it difficult to assign genotypic information based on phenotype.

Discussion
Here, a genetic screen was performed to identify mutants that show elevated activity of a MAPK pathway-dependent reporter.The FUS1 reporter provides a readout of several MAP kinase pathways.Expression of the mating-specific FUS1 gene is controlled by the transcription factor Ste12p, and the promoter contains several canonical Ste12p binding sites (Hagen et al. 1991).In cells lacking an intact HOG pathway (e.g.pbs2Δ and hog1Δ), FUS1 reporter activity is sensitive to salt and has been used as a "crosstalk" reporter (O'Rourke and Herskowitz 1998).In this genetic context, reporter activity is dependent on proteins Msb2p and Sho1p (O'Rourke and Herskowitz 2002).Msb2p and Sho1p also regulate filamentous growth.In the filamentous strain background lacking an intact Fig. 6.Comparison of phenotypes of mutants that hyperactivate the fMAPK pathway.Strains were spotted in 5 μL volumes on YEPD plates, grown for 3 days at 30°C, and washed under stream of water to reveal invasive scars.Pictures were taken before and after washing.Invasive cells were scraped and observed at 100× magnification using DIC filter.For FUS1-lacZ assays, strains were grown for 16 h in YEPD media, subcultured into YEPD, and grown till mid-log phase (A 600 ∼1.0).Mid-log phase cells were collected to prepare lysates for β-galactosidase assays.WT (PC538) values were compared with values for mutant defective for fMAPK activity (ste12Δ, PC539) or hyperactive for fMAPK activity (dig1Δ, PC3039; STE11-4, PC580; rga1Δ, PC3391; and pmi40-101, PC5014).Experiments were performed in 3 independent replicates.Data were analyzed by 1-way ANOVA test followed by a Tukey's pairwise comparison test to generate P-values.Asterisk denotes differences compared to WT (P < 0.05).Error bars denote standard error of the mean.Differences in lacZ levels between experiments (e.g.Fig. 1b) may result from different growth conditions.mating pathway (Σ1278b ste4), basal reporter activity detected is dependent on Msb2p and Sho1p (Cullen et al. 2004).The main benefit of using the reporter is its high sensitivity and complete dependency on Ste12p for activity; however, a potential limitation is that the reporter provides a readout for multiple MAPK pathways.
Whole-genome sequencing of a subset of mutants identified new alleles of well-established regulators of the MAPK pathway.These included gain-of-function alleles that induce constitutive activation of Ste11p, several of which had been identified previously (Stevenson et al. 1992).Likewise, loss-of-function mutations in RGA1, encoding the GAP for Cdc42p, are similar to those previously described (Stevenson et al. 1995).Although not previously identified, loss-of-function alleles predicted to truncate the MAP kinase Kss1p were also uncovered.These alleles might be expected to cause elevated activity of the fMAPK pathway, because Kss1p has an inhibitory role (Cook et al. 1997;Bardwell, Cook, Voora, et al. 1998).Unexpectedly, mutations predicted to truncate the C-terminal domain of Ste12p (from G491, Q500, and G587) were also uncovered.We infer that the C-terminus of Ste12p may function to inhibit the activity of this factor or alter its interaction with interacting partners.Multiple alleles of the abovementioned genes were identified.Alleles in other potential modulators (ALY1, AIM44, RCK2, BLM10, KAR2, MNN4, IRA2, and REG1) were uncovered at a low frequency and have been previously shown to impact the activity of the fMAPK pathway.At least some of these inputs presumably come from other pathways, such as the Ras2-cAMP-PKA pathway.Ras2p is known to regulate the fMAPK pathway (Mosch et al. 1996), and we identified mutations in the gene encoding for a Ras2p GAP, Ira2p, here.Furthermore, defects in protein glycosylation are also known to stimulate the activity of the fMAPK pathway (Cullen et al. 2000).Several other pieces of data support a causal relationship between the genes identified and reporter activity: (1) 2 strong gain-of-function alleles, STE11-1 and STE11-4, were identified in the screen and are known to activate the reporter (Stevenson et al. 1992); (2) Rga1p truncations lacking its GAP domain are known to activate the reporter (Stevenson et al. 1995; see Supplementary Fig. 5a); and (3) the phosphoblot analysis shows that no P∼Kss1p signal occurred in the Kss1p truncation mutants, even at low molecular weights (Supplementary Fig. 4), suggesting that the Kss1p protein is not produced at least in a form where the phosphorylated form of the protein can be detected.Loss of Kss1p protein is known to hyperactivate Fus3p MAPK (Cook et al. 1997;Bardwell, Cook, Voora, et al. 1998; see Supplementary Fig. 5b).
Raf-like MAPKKK and MEKK2/3-like MAPKKKs represent distinct classes of proteins that converge on and regulate MEKs (Lange-Carter et al. 1993).Ste11p is most similar to MEKK-like kinases with 41% identity to MEKKK3.Gain-of-function mutations in STE11 were also identified in this study.Mutations in STE11 were compared to mutations in RAF, which encodes the MAPKKK that regulates the ERK MAP kinase pathway in humans (Wellbrock et al. 2004).Mutations in RAF are associated with constitutive activation of the protein and cancer (Turski et al. 2016).Alignment of the proteins revealed a region in the kinase domain of Ste11p (VKITDFGISKK) that shows homology to the same region of B-RAF (VKIGDFGGLATV).In Ste11p, T596I identified here and previously (STE11-4; Stevenson et al. 1992) leads to constitutive activation of the kinase domain of the protein.Similarly, in B-RAF, changes to the adjacent amino acids D, F, and G have been linked to cancer.The valine at position 600 in B-RAF, which is not conserved in Ste11p, is the most changed residue found in cancer patients (Turski et al. 2016).A second site in Ste11p (PSEF) is conserved with (PSKS) in B-Raf, which corresponds to the previously identified STE11-1 allele (P279S; Stevenson et al. 1992) and the 2 alleles identified in this study (P279H/A).The other changes in Ste11p identified here (L440Q, L300S, and F292I) did not show homology to regions in the RAF protein.Therefore, at least 2 regions in MAPKKKs are conserved between yeast and humans, which when changed result in presumptive elevated kinase activity and phenotypic effects.
We show here that mutations resulting in loss of the extreme C-terminal domain of Ste12p are functional and show higher levels of MAPK pathway activity than the WT protein.Ste12p is a member of the homeodomain family of transcription factors (Rispail and Di Pietro 2010).Ste12p homologs exist in many fungal species and have been characterized in plant and animal pathogens (Rispail and Di Pietro 2010;Wong Sak Hoi and Dumas 2010;Fan et al. 2021;Purohit and Gajjar 2022).In plant pathogens, Colletotrichum lindemuthianum and Botrytis cinerea, the extreme C-terminus of Ste12p is modified by alternative splicing (Wong Sak Hoi et al. 2007;Schamber et al. 2010).Depending on the spliced products formed, different combinations of zinc-finger domains are produced at the end of the transcription factor, resulting in activating or inhibitory forms of the Ste12p proteins.The yeast Ste12p lacks Zn finger domains.Ste12p interacts with other proteins that lead to changes in gene expression.One group of proteins is the transcriptional repressors Dig1p and Dig2p (Tedford et al. 1997;Bardwell, Cook, Voora, et al. 1998;Bardwell, Cook, Zhu-Shimoni, et al. 1998;Zheng et al. 2010).Moreover, Ste12p can impact the organization of gene expression in the nucleus (Randise-Hinchliff et al. 2016), which occurs through a mechanism that involves the posttranslational modification of Dig2p (Randise-Hinchliff et al. 2016).In addition, Ste12p recognizes a specific consensus sequence on the DNA to regulate matingspecific genes and a different motif when in complex with Tec1p to control invasive growth (Madhani and Fink 1997;Heise et al. 2010;Dorrity et al. 2018;Zhou et al. 2020).Ste12p and Tec1p also regulate the cyclin-dependent kinase Cdk8p, controlling RNAPII-C-terminal domain to maintain genome integrity (Nelson et al. 2003;Aristizabal et al. 2015).
We also show that the NRR (Fig. 5c-e) may function in a regulated manner, dependent on the nutrient status of the cell.This might result from posttranslational modification of the C-terminal region.Ste12p contains multiple phosphorylated S/T residues in the inhibitory C-terminal region (Breitkreutz et al. 2010), including sites phosphorylated by the mitotic exit network (Zhou et al. 2021) and sites tied to the regulation of protein levels (Swaney et al. 2013).Ste12p is also phosphorylated by Cdk8p in response to nitrogen limitation (Nelson et al. 2003).An inhibitory protein may bind to this region of Ste12p to promote inhibition in a regulated manner, with candidates including Kss1p and Dig1p, as both of these proteins bind to Ste12p and have the capacity to function as repressors (Cook et al. 1996;Bardwell, Cook, Voora, et al. 1998).Some expected mutations were not identified by the screen.Internal deletions in the N-terminal inhibitory domain of Msb2p, which leads to elevated fMAPK pathway activity (Cullen et al. 2004), were not recovered.A version of transmembrane sensor Sho1p, Sho1p P120L , which hyperactivates fMAPK pathway (Vadaie et al. 2008), was also not uncovered by the screen.It is unclear why mutations in some genes were recovered multiple times (STE11, STE12, RGA1, and KSS1), while other mutations were underrepresented (e.g. in ALY1 and IRA2), and still other mutations were not recovered (e.g. in TEC1, STE20, and STE7).Perhaps some regions of the genome are more susceptible to spontaneous mutations than others.Alternatively, growth defects of some mutants may mask the pathway activity.These could include alleles in essential genes, like those controlling protein glycosylation.Other mutations (e.g. in DIG1, which was not found) could induce enhanced cell adhesion and polarized cell morphologies that may impact colony morphology and were missed.In addition, many mutations were uncovered in intergenic regions and upstream and downstream of the ORFs that were not explored here.Moreover, mutations that result in elevated MAPK pathway activity can lead to mutations in other parts of the genome, as retrotransposon mobility can be induced by the fMAPK pathway (Company et al. 1988;Laloux et al. 1994;Conte et al. 1998;Morillon et al. 2000).

Fig. 1 .
Fig. 1.Strategy to identify modulators of the fMAPK pathway.a) The fMAPK pathway.Positive components (in green) and negative regulatory inputs (in red) are shown.Daggers refer to genes with alleles identified in the study.b) In cells lacking the mating pathway (ste4), the FUS1-lacZ and FUS1-HIS3 reporters were used to measure fMAPK pathway activity in WT (PC538), ste12Δ (PC539), and dig1Δ (PC3039) strains.SD + AA media denotes dilution control.SD-HIS and SD-HIS + 10 mM 3-ATA media were used to measure fMAPK pathway activity.WT cells showing normal levels of fMAPK activity can grow on SD-HIS media, whereas ste12Δ mutant failed to grow on SD-HIS media due to downregulation of pathway activity.The dig1Δ mutant showing elevated activity of the pathway can grow on SD-HIS as well as on the SD-HIS + 10 mM 3-ATA media.Isolates that grow on SD-HIS + 10 mM 3-ATA presumably induce the fMAPK pathway.The β-galactosidase assays were performed as described.Experiments were performed in 3 independent replicates.Data were analyzed by 1-way ANOVA followed by a Tukey's pairwise comparison test to generate P-values.Asterisk denotes difference compared to WT and P < 0.05.Error bars denote standard error of the mean.c) Filamentous phenotype is depicted for WT, ste12Δ, and dig1Δ strains.To compare invasive growth, strains were spotted on YEPD media, grown for 3 days at 30°C, and colonies were washed under stream of water.Colonies were photographed before and after washing.To look at cell morphology, cells were scraped and visualized using DIC at 100× magnification (scale bar, 5 microns).Cells were examined by microscopy (100×) to view the changes in cell length and budding pattern in response to glucose limitation.

Fig. 4 .
Fig. 4. Analysis of the levels of phosphorylated Kss1p in selected mutants.a) Diagram of the Ste11p, Rga1p, Kss1p, and Ste12p proteins.Asterisks denote the position of stop mutations or changes resulting in the indicated amino acids.fs, frameshift.Underlined mutations were made in the parent strain and were checked for effect on fMAPK pathway activity using the FUS1-HIS3 and FRE-lacZ reporters (see Fig.5and Supplementary Fig.5).b) Immunoblots of membranes probed with p44/42 antibodies to visualize P∼Kss1p and P∼Fus3p levels, and Pgk1p antibodies as a control for total protein levels.WT (PC538), ste11Δ (PC611), and STE11-4 (PC580) strains were used as standards.Band intensity was quantified using Image Lab.Phospho-band intensities were normalized to Pgk1p bands for each lane, and the WT phospho-protein levels were set to 1 to compare P∼MAP kinases in different mutants with respect to WT. Colors represent mutations in the indicated genes: red, KSS1; green, RGA1; blue, STE11; and orange, STE12.

Fig. 5 .
Fig. 5. Role of the C-terminal region of Ste12p in regulating mating and filamentous growth.a) Halo assays.WT (PC7650), ste12Δ (PC5652), and Ste12p G491* (PC7923) cells were spread on YEPD plates, and α-factor was applied at 5 μL (left) and 10 μL (right).Plates were incubated at 30°C for 2 days and photographed.At the right, FUS1-lacZ levels in indicated strains; all strains are STE4+.Minus sign, without pheromone; plus sign, with pheromone.Deviation was less than 10% between samples.Data were analyzed by 1-way ANOVA followed by a Tukey's pairwise comparison test to generate P-values.Asterisk denotes significant differences from WT (P < 0.05).b) Invasive growth.Ten microliters of cells in (a) were spotted onto YEPD media.The plate was incubated at 30°C for 3 days, and colonies were washed under a stream of water to reveal invasive scars.Photographs were taken before and after washing.At the right, typical examples of scraped cells at 100×.Bar, 5 microns.At the right, FRE-lacZ levels in indicated strains.Deviation was less than 10% between samples.Data were analyzed by 1-way ANOVA followed by a Tukey's pairwise comparison test to generate P-values.Asterisk denotes significant differences from WT (P < 0.05).c) Cell morphology in different media conditions.Strains from (a) were grown to mid-log phase (A 600 ∼1.0) and transferred to the indicated media and grown further for 5 h.Liquid cultures were looked at 100× using DIC filter.d) Relative FRE-lacZ expression values are shown normalized to each respective WT in the media shown.Experiments were performed in duplicates; error bars denote standard error.Data were analyzed by 1-way ANOVA followed by a Tukey's pairwise comparison test to generate P-values.Asterisk denotes difference compared to WT, and P < 0.05.e) Model depicting regulation of mating and filamentous growth by the C-terminal region of Ste12p.PRR, positive regulatory region; NRR, negative regulatory region.The wavy arrow indicates that the NRR may have condition-specific effects on fMAPK pathway activity.
Fig. 5. Role of the C-terminal region of Ste12p in regulating mating and filamentous growth.a) Halo assays.WT (PC7650), ste12Δ (PC5652), and Ste12p G491* (PC7923) cells were spread on YEPD plates, and α-factor was applied at 5 μL (left) and 10 μL (right).Plates were incubated at 30°C for 2 days and photographed.At the right, FUS1-lacZ levels in indicated strains; all strains are STE4+.Minus sign, without pheromone; plus sign, with pheromone.Deviation was less than 10% between samples.Data were analyzed by 1-way ANOVA followed by a Tukey's pairwise comparison test to generate P-values.Asterisk denotes significant differences from WT (P < 0.05).b) Invasive growth.Ten microliters of cells in (a) were spotted onto YEPD media.The plate was incubated at 30°C for 3 days, and colonies were washed under a stream of water to reveal invasive scars.Photographs were taken before and after washing.At the right, typical examples of scraped cells at 100×.Bar, 5 microns.At the right, FRE-lacZ levels in indicated strains.Deviation was less than 10% between samples.Data were analyzed by 1-way ANOVA followed by a Tukey's pairwise comparison test to generate P-values.Asterisk denotes significant differences from WT (P < 0.05).c) Cell morphology in different media conditions.Strains from (a) were grown to mid-log phase (A 600 ∼1.0) and transferred to the indicated media and grown further for 5 h.Liquid cultures were looked at 100× using DIC filter.d) Relative FRE-lacZ expression values are shown normalized to each respective WT in the media shown.Experiments were performed in duplicates; error bars denote standard error.Data were analyzed by 1-way ANOVA followed by a Tukey's pairwise comparison test to generate P-values.Asterisk denotes difference compared to WT, and P < 0.05.e) Model depicting regulation of mating and filamentous growth by the C-terminal region of Ste12p.PRR, positive regulatory region; NRR, negative regulatory region.The wavy arrow indicates that the NRR may have condition-specific effects on fMAPK pathway activity.

Table 1 .
Mutations identified in genes regulating the fMAPK pathway. a