ABSTRACT
Mitogen-Activated Protein Kinase (MAPK) pathways control cell differentiation and the response to stress. MAPK pathways can share components with other pathways yet induce specific responses through mechanisms that remain unclear. In Saccharomyces cerevisiae, the MAPK pathway that controls filamentous growth (fMAPK) shares components with the MAPK pathway that regulates the response to osmotic stress (HOG). By exploring temporal regulation of MAPK signaling, we show here that the two pathways exhibited different patterns of activity throughout the cell cycle. The different patterns resulted from different expression profiles of genes encoding the mucin sensors (MSB2 for fMAPK and HKR1 for HOG). We also show that positive feedback through the fMAPK pathway stimulated the HOG pathway, presumably to modulate fMAPK pathway activity. By exploring spatial regulation of MAPK signaling, we found that the shared tetraspan protein, Sho1p, which has a dynamic localization pattern, induced the fMAPK pathway at the mother-bud neck. A Sho1p-interacting protein, Hof1p, which also localizes to the mother-bud neck and regulates cytokinesis, also regulated the fMAPK pathway. Therefore, spatial and temporal regulation of pathway sensors, and cross-pathway feedback, regulate a MAPK pathway that controls a cell differentiation response in yeast.
INTRODUCTION
Mitogen activated protein (MAP) kinase pathways are evolutionary conserved signaling modules that control growth (Johnson and Lapadat 2002; Lavoie et al. 2020), cell differentiation (Chen and Thorner 2007; Raman et al. 2007; Dinsmore and Soriano 2018), and the response to stress (Roux and Blenis 2004; Yoon and Seger 2006; Shaul and Seger 2007; Papa et al. 2019). Like other signaling pathways, MAP kinase pathways operate in interconnected networks where inputs from multiple pathways converge (Jordan et al. 2000; Fischer et al. 2018). Moreover, MAPK pathways are subject to exquisite spatiotemporal control, in that signaling occurs at a precise place, typically on the cell surface, and for a defined duration to elicit an appropriate response. Mis-regulation of MAPK pathways is linked to various diseases such as cancers, polycystic kidney disease, obesity, diabetes and developmental disorders (Lee et al. 2000; Hirosumi et al. 2002; Maekawa et al. 2005; Omori et al. 2006; Rodriguez-Viciana et al. 2006; Roberts and Der 2007; Lawrence et al. 2008). Therefore, understanding how MAPK pathways induce a specific signal an integrated network remains an important open question.
In the budding yeast Saccharomyces cerevisiae, MAP kinase pathways control well-characterized cell differentiation responses that occur in response to extrinsic cues (e.g mating and filamentous growth). Yeast MAP kinase pathways also control the response to stresses like high osmolarity (HOG). In response to nutrient limitation, yeast and other fungal species can undergo filamentous (or invasive/pseudohyphal) growth, where cells differentiate into interconnected and elongated filaments (Gimeno et al. 1992; Roberts and Fink 1994; Mosch et al. 1996; Peter et al. 1996; Leberer et al. 1997; Pan et al. 2000; Gancedo 2001; Adhikari et al. 2015b). In some fungal pathogens, this microbial differentiation response is critical for virulence (Lo et al. 1997). In yeast, the MAP kinase pathway that regulates filamentous growth (Fig. 1A, fMAPK, green) is controlled by the mucin-type glycoprotein Msb2p (Cullen et al. 2004). The underglycosylation of Msb2p occurs under nutrient-limiting conditions, which results in proteolytic processing and release of its inhibitory extracellular domain by an aspartyl-type protease, Yps1p (Adhikari et al. 2015b). Msb2p functions with a four-pass (tetraspan) adaptor protein, Sho1p (O’Rourke and Herskowitz 1998; Cullen et al. 2004), and another adaptor, Opy2p (Yamamoto et al. 2010; Karunanithi and Cullen 2012). Msb2p and Sho1p converge on the Rho GTPase Cdc42p, which regulates the fMAPK pathway by binding to and activating the p21-activated (PAK) kinase, Ste20p (Peter et al. 1996; Leberer et al. 1997; Johnson 1999; Bi and Park 2012).
When activated, Ste20p regulates a MAP kinase cascade composed of the MAPKKK (Ste11p), which phosphorylates and activates the MAPKK (Ste7p), which in turn phosphorylates the MAP kinase, Kss1p (Roberts and Fink 1994; Madhani et al. 1997). The active (phosphorylated) form of Kss1p regulates transcription factors including Ste12p and Tec1p, and the transcriptional repressor Dig1p, to induce target gene expression (Ma et al. 1995; Madhani and Fink 1997; Bardwell et al. 1998; Rupp et al. 1999; Roberts et al. 2000; van der Felden et al. 2014; Pelet 2017). The transcriptional targets of the fMAPK pathway encode products that induce differentiation to the filamentous cell type (Liu et al. 1993; Madhani and Fink 1997). Ste12p and Tec1p also stimulate the fMAPK pathway through positive feedback, because a subset of transcriptional targets of the fMAPK pathway encode pathway components (MSB2, KSS1, STE12, and TEC1)(Madhani et al. 1999; Roberts et al. 2000; Cullen et al. 2004; Adhikari and Cullen 2014).
The fMAPK pathway utilizes a subset of components that also function in other MAP kinase pathways, such as the HOG (O’Rourke and Herskowitz 2002; Cullen et al. 2004; Tatebayashi et al. 2006) and mating pathways (Bardwell et al. 1998; Sabbagh et al. 2001; Breitkreutz and Tyers 2002; Bardwell 2004; Schwartz and Madhani 2004; Bardwell 2006). The HOG pathway responds to changes in osmolarity and is composed of partially redundant branches (Fig. 1A). Like the fMAPK pathway, the Ste11p branch of the HOG pathway requires Sho1p, Cdc42p, Ste20p, and Ste11p (Tatebayashi et al. 2006). The pathways diverge after Ste11p, which phosphorylates Ste7p to regulate the fMAPK pathway, and Pbs2p to regulate the HOG pathway (Brewster et al. 1993; Maeda et al. 1994). Another difference between the pathways occurs at the level of the mucins, Msb2p and Hkr1p. Hkr1p regulates the HOG pathway, while Msb2p has been implicated in regulating both the fMAPK and HOG pathways [Fig. 1A; (O’rourke and Herskowitz 2002; Cullen et al. 2004; Tatebayashi et al. 2007; Pitoniak et al. 2009; Tanaka et al. 2014; Yamamoto et al. 2016)]. Msb2p and Hkr1p regulate the HOG pathway by different mechanisms (Tanaka et al. 2014). Importantly, Hkr1p does not regulate the fMAPK pathway, and overexpression of the mucins induce different target genes (Pitoniak et al. 2009). Moreover, the two pathways function in an antagonistic manner (Davenport et al. 1999; Adhikari and Cullen 2014). Therefore, selectivity in propagating a downstream signal presumably occurs at the level of the mucin glycoproteins and adaptors, although how this occurs is not well understood.
In this study, we explored the spatiotemporal regulation of the fMAPK and HOG pathways. One way that MAPK pathways exert precise temporal control is by regulating progression through the cell cycle. However, relatively little is known about how MAPK pathways are themselves regulated throughout the cell cycle. We previously found that the activity of the fMAPK pathway fluctuates throughout the cell cycle (Prabhakar et al. 2020), which provided an opportunity to explore this question in depth. We show here that the fMAPK and the HOG pathways show different patterns of cell-cycle regulation. While the activity of the fMAPK pathway peaked in M phase, the HOG pathway could be activated at any point in the cell cycle. The cell-cycle regulation of the fMAPK pathway stemmed from cell-cycle regulation of the mucin sensor, Msb2p. We also show that positive feedback through the fMAPK pathway induced the HOG pathway, generating cross-pathway feedback between the pathways. We have also previously shown that the fMAPK pathway is activated spatially by proteins that mark the cell poles (Basu et al. 2016). By examining the spatial localization of MAPK pathway sensors, we show that the tetraspan protein Sho1p was localized to the mother-bud neck during elevated fMAPK pathway activity. Sho1p-interacting proteins that also localize to the mother-bud neck were required for fMAPK pathway activity. Taken together, these results define new aspects of spatiotemporal regulation that leads to precise regulation of a MAPK pathway. Our findings may extend to other systems, where precise regulation of signaling pathways is required to induce cell differentiation.
RESULTS
MAPK Pathways that Share Components Show Different Patterns of Activity Throughout the Cell Cycle
Given that cell-cycle regulation of MAP kinase pathways is an under-explored aspect of MAP kinase pathway regulation, we explored the cell-cycle regulation of the fMAPK pathway that controls filamentous growth in yeast. In one approach, cells were synchronized by α-factor, which arrests cells in the G1 phase of the cell cycle (Elion et al. 1993; Peter et al. 1993; Peter and Herskowitz 1994). Release of cells into the cell cycle by α-factor removal showed low levels of phosphorylated (active) Kss1p early in the cell cycle (Fig. 1B, P~Kss1, 0 min to 80 min, G1, S and G2). When the cyclin Clb2p-Myc levels dropped in M phase (Richardson et al. 1992; Irniger et al. 1995; Wasch and Cross 2002; Cross et al. 2005; Eluere et al. 2007; Kuczera et al. 2010; Cepeda-garcia 2017), P~Kss1p levels increased (Fig. 1B, 100 min). Although the cell synchronization deteriorated in the second cycle (based on the failure of Clb2p-Myc to fully disappear), P~Kss1p levels decreased (at 140 min) and then increased again (at 160 min), presumably as cells progressed through the second cell cycle. An increase in P~Kss1p levels in the second cell cycle was also observed (from 17 to 24), which may be due to nutrient depletion at this point in the culture-growth cycle. Activation of the mating pathway might cause a reduction in fMAPK pathway activity through a variety of mechanisms, such as Tec1 degradation (Bao et al. 2004; BrÜckner et al. 2004; Chou et al. 2004) or altered TEC1 levels by the cell-cycle transcriptional regulator Swi5 (Spellman et al. 1998). In an independent approach, cells were synchronized by hydroxyurea (HU), which arrests cells in S phase (Slater 1973; Slater 1974; KoÇ et al. 2004). Release of cells from HU treatment also showed low P~Kss1p levels that increased as cells progressed through the cell cycle (Fig. S1A) In this case, P~Kss1p levels increased prior to the disappearance of Clb2-Myc levels. These results demonstrate that the activity of the fMAPK pathway fluctuates throughout the cell-cycle.
The fMAPK pathway shares components with the Ste11p-branch of the HOG pathway (Fig. 1A). To determine whether the activity of the HOG pathway changes throughout the cell cycle, the phosphorylation of the MAP kinase Hog1p was measured in cells lacking the Sln1p branch (Fig. 1C, ssk1Δ). As previously reported (Posas and Saito 1997), asynchronous ssk1Δ cells exposed to 1M sorbitol showed elevated P~Hog1p levels (Fig. 1C, P~Hog1, Async + 1M Sorb). Synchronized ssk1Δ cells exposed to 1M sorbitol also showed elevated P~Hog1p levels (Fig. 1C, P~Hog1, YEPD + 1M Sorb, 0, 30, 60, and 90 min). In synchronized cells not exposed to osmotic stress, basal P~Hog1p levels showed some periodicity, but compared to P~Kss1p, became reduced throughout the cell cycle (Fig. S1B). These results indicate that the HOG pathway can be activated at any point in the cell cycle.
Cells lacking an intact HOG pathway (pbs2Δ or hog1Δ mutants) exposed to osmotic stress exhibit crosstalk to the mating (O’Rourke and Herskowitz 1998) and fMAPK pathways [Fig. 1A, orange arrow (Pitoniak et al. 2009)]. In cells lacking Pbs2p, the fMAPK pathway showed elevated P~Kss1p levels in response to 1M sorbitol (Fig. 1C, P~Kss1, pbs2Δ, YEPD + 1M Sorb). Because P~Kss1p levels were high at all time points tested (0, 30, 60, and 90 min), the cell-cycle regulation of the fMAPK pathway appears to have been lost in the pbs2Δ mutant. P~Kss1p levels were also elevated to some degree in PBS2+ cells exposed to osmotic stress (Fig. 1C, P~Kss1, ssk1Δ, YEPD + 1M Sorb at 30, 60, and 90 min but not 0 min), which might result from basal crosstalk between the pathways (Hao et al. 2008). As expected, P~Hog1p was not detected in the pbs2Δ mutant exposed to sorbitol, because Pbs2p is required to phosphorylate Hog1p. Therefore, during crosstalk, the fMAPK pathway can be activated at early stages in the cell cycle. One implication of this result is that shared components between HOG and fMAPK pathways are present and have the capacity to function at early stages in the cell cycle.
Cell-Cycle Regulation of the fMAPK Pathway Occurs at the Level of Msb2p and Sho1p: MSB2 Expression is Cell-Cycle Regulated
To define how the fMAPK pathway is regulated throughout the cell cycle, genetic suppression analysis was performed. Genetic suppression analysis can order components in a pathway using gain- and loss-of-function alleles. Hyperactive versions of fMAPK pathway components, Msb2p* [GFP-Msb2p, (Adhikari et al. 2015b)], Sho1pP120L(Vadaie et al. 2008), and Ste11p-4 (Stevenson et al. 1992) were examined for the ability to bypass the low levels of fMAPK pathway activity seen early in the cell cycle. Ste11p-4 bypassed the low levels of fMAPK pathway activity (Fig. 2A). By comparison, Msb2p* and Sho1pP120L showed a partial bypass (Fig. S2, A-C). These results indicate that the cell cycle regulation of the fMAPK pathway occurs above Ste11p and at the level of Msb2p and Sho1p in the fMAPK cascade.
We also noticed that although wild-type cells showed low Clb2p-HA levels after 2 h of α-factor treatment (Fig. 2A, Clb2-HA, wT αf 2h) indicative of complete arrest, cells containing Ste11p-4 showed elevated Clb2p-HA levels (Fig. 2A, Clb2-HA, STE11-4 αf 2h), which is suggestive of a delay in G1 arrest. The delay might result from elevated fMAPK activity, which delays the cell cycle in G1 (Loeb et al. 1999; Madhani et al. 1999) and G2/M (Ahn et al. 1999; Rua et al. 2001; Vandermeulen and Cullen 2020). In line with this possibility, cells containing Ste11p-4 also showed a delay in accumulation in Clb2p-HA (Fig. 2A, compare WT at 60 min with STE11-4 at 60 min; see also Fig. 1C, compare Clb2-HA levels in WT [ssk1Δ] and pbs2Δ at 60 and 90 min).
The activity of the fMAPK pathway might fluctuate if the levels of one or more components change throughout the cell cycle. We first examined Msb2p levels, which directly impact fMAPK pathway activity (Cullen et al. 2004) and drive induction of fMAPK pathway targets (Pitoniak et al. 2009). Examining the level of a functional Msb2p-HA protein, expressed under the control of the MSB2 promoter at its endogenous locus in the genome, showed low levels early in the cell cycle, which peaked later in the cell cycle (Fig. 2B, Msb2) Msb2p-HA levels showed a similar pattern in cells synchronized by HU (Fig. S1A).
By comparison, the levels of the HOG pathway mucin, Hkr1p, (Tatebayashi et al. 2007; Pitoniak et al. 2009; Yang et al. 2009) also expressed as an HA fusion from its endogenous promoter in the genome, did not show an increase throughout the cell cycle (Fig. 2B, Hkr1). Because both mucins contained the same epitope fusion (HA) internal to both proteins in their glycosylated extracellular domains, the levels were able to be compared. Msb2p-HA levels were 15-fold higher than Hkr1p-HA levels in asynchronous (Fig. 2C, Async) and synchronized cultures (Fig. 2C, 100 and 160 min). Comparative proteomic studies show a similar trend (Breker et al. 2013; Yofe et al. 2016). The relative abundance of the mucins might impact the activities of the fMAPK and the HOG pathways throughout the cell cycle.
One way that protein levels are regulated is by changes in gene expression. To test whether the levels of Msb2p and Hkr1p result from different patterns of gene expression, mRNA levels of MSB2 and HKR1 were examined throughout the cell cycle. MSB2 mRNA levels were low following release from α-factor treatment and higher at cells progressed through the cell cycle (Fig. 2D, MSB2). By comparison, HKR1 mRNA levels were high early in the cell cycle (Fig. 2D, HKR1) but did not otherwise fluctuate. Thus, MSB2 and HKR1 genes show different patterns of gene expression throughout the cell cycle. Moreover, the MSB2 expression was 5-fold higher than HKR1 based on analysis of a previously published dataset that examined the levels of MSB2-lacZ and HKR1-lacZ fusions in asynchronous cells (Pitoniak et al. 2009).
Comparing fMAPK Pathway Activity Under Basal and Pathway-Inducing Conditions Throughout the Cell Cycle
Since Msb2p regulates the fMAPK pathway through positive feedback, Msb2p levels might precede, and therefore induce P~Kss1p levels at M/G1. Alternatively, MSB2 expression might be induced after P~Kss1p induction, because MSB2 is a target of the pathway. To determine if the rise in Msb2p levels precede Kss1p activation, the levels of Msb2p-HA and P~Kss1p were compared following release from α factor at short time intervals. Low levels of Msb2p-HA at the beginning of the cell cycle gradually increased and peaked by 40-50 min, which was prior to the increase in P~Kss1p levels (Fig. 3A). Although Msb2p levels are known to directly regulate fMAPK activity (ref-Cullen 2004, Hema and Nadia), the fact that increase at 40-50 min did not significantly increase P~Kss1p levels. There may be another regulation at this stage (G1/S boundary) preventing Kss1p activation, such as spatial regulation of Sho1p (see below). After a drop at 80 min, Msb2p-HA levels increased again to peak at 100 min, when P~Kss1p levels rose at M/G1 (Fig. 3A, P~Kss1). These fluctuations correlated with MSB2 mRNA levels (see Fig. 2D). Thus, Msb2p protein levels rise prior to the rise in P~Kss1p levels.
To this point, fMAPK pathway activity was examined under basal or non-inducing conditions. Nutrient limitation, including nitrogen (Gimeno et al. 1992) and carbon (Cullen and Sprague 2000) can trigger filamentous growth, which is thought to be a type of nutrient foraging response seen in fungi. The fMAPK pathway is induced by growth in the non-preferred carbon source galactose [YEP-GAL (Karunanithi and Cullen 2012)]. Cells grown in YEPD and synchronized by α factor that were released into YEP-GAL medium (inducing conditions) showed a delay in P~Kss1p accumulation, which indicates that the fMAPK pathway is cell-cycle regulated under inducing conditions (Fig. 3B), consistent with our previous observations (Prabhakar et al. 2020). Msb2p-HA levels also accumulated in YEP-GAL prior to the increase in P~Kss1p levels (Fig. 3B). Thus, the increase in Msb2p levels precedes and might therefore contribute to the accumulation of P~Kss1p levels throughout the cell cycle.
Clb2p accumulation showed an extended delay in YEP-GAL (Fig. 3B, 180 min) compared to YEPD media (Fig. 3A, 60-80 min). The extended delay in cell-cycle progression in YEP-GAL might result from glucose repression. Glucose repression involves the transcriptional repression of genes (GAL genes and many other genes) that metabolize non-preferred carbon sources (Carlson and Botstein 1982; Nehlin et al. 1991; Wilson et al. 1996; De Vit et al. 1997). To examine fMAPK pathway activity in response to a sustained induction, cells pregrown in YEP-GAL were synchronized and monitored for fMAPK activity. These cells also showed low levels of P~Kss1p after release from α-factor arrest (Fig. S3), but not the extended delay seen under pathway-inducing conditions. Thus, cell-cycle regulation of the fMAPK pathway is seen in basal, induced, and sustained-inducing conditions.
In YEP-GAL media, a new pattern of cell-cycle regulation was observed. Compared to basal conditions, where P~Kss1p levels peaked after Clb2p-HA levels fell (Fig. 3A), under inducing conditions, P~Kss1p levels peaked at the same time as Clb2p-HA accumulation (Fig. 3B). A similar trend was seen under sustained-inducing conditions (Fig. S3). Graphing P~Kss1p induction under the two conditions showed the differences in timing between basal (Fig. 3C, pink) and inducing (blue) conditions, compared to Clb2p-HA levels (yellow). The ability of the fMAPK pathway to partially bypass the cell-cycle regulation may result from GAL-dependent induction of the fMAPK pathway, which could occur by several mechanism including elevated proteolytic processing of Msb2p, which liberates an inhibitory glycodomain (Vadaie et al. 2008). As expected from previous observations, P~Kss1p levels were 12-fold higher under inducing conditions than under basal conditions (Fig 3C, compare left and right axes).
Altering Cell Cycle Regulation Impacts fMAPK Pathway Activity and Filamentous Growth
Cell-cycle progression is regulated by transcription factors. One set of transcription factors, SBF (Swi4/6 cell cycle box Binding Factor), induces transcription of genes required for the progression from G1 to S phase (Andrews and Herskowitz 1989; Nasmyth and Dirick 1991; Sidorova and Breeden 1993). We tested whether altering the normal cell-cycle progression, by loss of Swi4p and Swi6p, impacts fMAPK pathway activity and filamentous growth. The swi4Δ mutant had elevated levels of P~Kss1p compared to wild-type cells (Fig. 3D, P~Kss1). The swi6Δ mutant had a severe growth defect and was not tested further. The swi4Δ mutant also had elevated levels of Msb2p-HA (Fig. 3D, Msb2-HA), which may account for the elevated levels of fMAPK activity. Evaluation of the swi4Δ mutant by the plate washing assay (PWA), which measures invasive growth as a readout of fMAPK pathway activity (Roberts and Fink 1994), showed hyper-invasive growth compared to wild-type cells (Fig. 3E, washed). The swi4Δ mutant was not required for growth on high-osmolarity media (Fig. 3E, YEPD+1M Sorb). The swi4Δ mutant cells had an elongated appearance (Fig. 3E, 100X), which may result from hyperpolarized growth due to an extension of the G1 phase of the cell cycle (White et al. 2009) and/or increased fMAPK activity (Fig. 3D). Although the swi4Δ mutant responded to α factor, the cells did not synchronize, which prevented evaluation of P~Kss1p levels throughout the cell cycle in this mutant. These results indicate that normal cell-cycle progression is required for proper fMAPK pathway activity and filamentous growth.
Positive Feedback of the fMAPK Pathway Induces the HOG Pathway
The cell-cycle regulation of the fMAPK pathway might extend to transcriptional targets of the pathway. The fMAPK pathway regulates multiple target genes that encode proteins that bring about filamentous growth (Madhani et al. 1999; Roberts et al. 2000; Vandermeulen and Cullen 2020). A major target of the fMAPK pathway is the gene encoding the cell adhesion molecule Flo11p (Rupp et al. 1999; Vinod et al. 2008), which promotes cell adhesion during filamentous/invasive growth (Rupp et al. 1999). By qPCR analysis, FLO11 expression showed a similar periodicity as MSB2, which indicates that its expression is regulated throughout the cell cycle (Fig. 4A). Several transcriptional targets of the fMAPK pathway encode components of the pathway and are induced by positive feedback. The gene encoding the MAP kinase Kss1p (Roberts et al. 2000) (Fig. 4B) and the TEA/ATS type transcription factor Tec1p (KÖhler et al. 2002) (Fig. 4C) showed a similar pattern of cell-cycle regulation as MSB2. By comparison, the transcription factor Ste12p, which functions in the mating and the fMAPK pathways (Roberts and Fink 1994) showed a different pattern of expression, perhaps because its expression is induced by pheromone (Fig. 4D). Therefore, many of the transcriptional targets of the fMAPK pathway that we tested showed cell-cycle regulated expression profiles.
To further explore the relationship between positive feedback and fMAPK pathway activity, we examined the consequences of hyperactivating the fMAPK pathway in cells lacking the transcription factors that control positive feedback. Cells containing another hyperactive version of Msb2p, Msb2pΔ100-818, Msb2p**, hyperactivated the fMAPK pathway through positive feedback, because the transcription factor Ste12p was required [Fig. 4E, P~Kss1, (Cullen et al. 2004; Vadaie et al. 2008; Prabhakar et al. 2020)]. Positive feedback occurred in basal (YEPD) and inducing (YEP-GAL) conditions (Fig. 4E). Msb2p** expressed from an fMAPK-independent promoter also required Ste12p to hyperactive the fMAPK pathway [Fig. 4E, P~Kss1, GAL-MSB2** ste12Δ,(Prabhakar et al. 2020)], which indicates that in addition to MSB2 other components of the pathway (like Kss1p and Tec1p) are required for positive feedback.
We also examined the activity of the HOG pathway under this condition. In addition to osmotic stress, the HOG pathway can also be induced by the non-preferred carbon source, galactose (Adhikari and Cullen 2014). Unexpectedly, we found that Msb2p** also stimulated the HOG pathway in a Ste12p-dependent manner (Fig. 4E, P~Hog1). The effect was seen in basal and pathway-inducing conditions (Fig. 4E). This result indicates that positive feedback from the fMAPK pathway leads to activation of the HOG activity. Given that the HOG pathway functions antagonistically to the fMAPK pathway (Davenport et al. 1999; Adhikari and Cullen 2014), these results might describe a novel type of cross-pathway feedback, where positive feedback through one pathway (fMAPK) induces another pathway to modulate its activity. It might be interesting to note that positive feedback through the fMAPK pathway was higher (20-fold) than positive feedback through the HOG pathway (9-fold in GLU and 3-fold in GAL, Fig. 4F). Other measurements also showed key similarities and differences between the pathways. Thus, Msb2p induces both MAP kinase pathways at different levels to produce an appropriate response. This feedback regulation can be described by a simple model, which shows the relationship between the two pathways (Fig. 4G). In this relationship, Msb2p induces one pathway at high levels (fMAPK, green), and its antagonistic pathway to lower levels (HOG, red), resulting in a modulated response.
Sho1p Levels and Localization Change Throughout the Cell Cycle
To regulate the fMAPK pathway, Msb2p interacts with the tetraspan protein, Sho1p (O’Rourke and Herskowitz 1998; Cullen et al. 2004). The levels and expression of Sho1p were also examined. Sho1p-GFP levels rose prior to accumulation in Clb2p-Myc levels in G2/M (Fig. 5A) and dropped when P~Kss1p levels increased. The drop in Sho1p-GFP levels corresponding to fMAPK pathway activation might be due to turnover of active Sho1p. Specifically, a hyperactive version of Sho1p, Sho1pP120L, shows elevated turnover compared to the wild-type protein (Adhikari et al. 2015a). SHO1 mRNA showed a similar pattern of cell-cycle regulation (Fig. 5B, orange line). The increase in SHO1 mRNA levels correlated with the increase in protein levels. Incidentally, Sho1p-GFP protein levels increased after 30 min of α-factor treatment (Fig. 5A, Sho1-GFP), which might occur because Sho1p has a function in mating (Nelson et al. 2004). Therefore, genes encoding two of the sensors for the fMAPK pathway, MSB2 and SHO1,show similar patterns of cell-cycle regulated gene expression.
Although the Msb2p, Sho1p, and Opy2p proteins form a complex in the plasma membrane (Tatebayashi et al. 2015; Yamamoto et al. 2016), they have different patterns of localization and turnover (Adhikari et al. 2015a). Processed Msb2p is turned over by the E3 ubiquitin ligase Rsp5p and is mainly localized to the lysosome/vacuole (Adhikari et al. 2015a; Adhikari et al. 2015b; Prabhakar et al. 2019), whereas Sho1p primarily localizes to plasma membrane (Raitt et al. 2000; Reiser et al. 2003; Pitoniak et al. 2009) at the growth tip in developing buds and the mother-bud neck in large buds. To better understand the contribution of Msb2p and Sho1p in the cell-cycle regulation of the fMAPK pathway, the localization of the Msb2p-GFP and Sho1p-GFP proteins were examined as cells progressed through the cell cycle. Time-lapse fluorescence microscopy was performed by co-localization of GFP fusion proteins to Msb2p and Sho1p, and the septin and the cell-cycle marker, Cdc3p-mCherry (Kim et al. 1991; Lippincott et al. 2001). Due to its high turnover rate (Adhikari et al. 2015a; Adhikari et al. 2015b), Msb2p-GFP showed a predominately vacuolar localization pattern. Although Msb2p-GFP was at the periphery in some cells, its localization was not otherwise informative (Movie 1). By comparison, Sho1p-GFP localized to different parts of the cell throughout the cell cycle, including presumptive bud sites, the tip of developing buds and the mother-bud neck (Movies 2 and 3). The same pattern was seen under inducing conditions (Gal), except that Sho1p-GFP was polarized at the distal pole for an extended period (Movie 4). In cells grown in Gal, which bud distally, it was clear that Sho1p-GFP was localized to the at the mother-bud neck during prior to cytokinesis, when the septin ring splits into a double ring [Movie 4 (Kim et al. 1991; Lippincott et al. 2001; Bi and Park 2012)].
Because Sho1p was localized to the mother-bud neck during septin ring split, it appeared that Sho1p-GFP was at the mother-bud neck during the period of the cell cycle when cells experienced elevated fMAPK pathway activity. To further define the location of Sho1p during the induction of fMAPK pathway activity, the same cells harvested for immunoblot analysis (in Fig. 5A) were examined by microscopy for Sho1p-GFP localization. In synchronized cells, the increase in P~Kss1p levels (Fig. 5A, P~Kss1, 100 min) corresponded to an increase in the percentage of cells that showed Sho1p-GFP localization at the mother-bud neck. Synchronized cells containing Sho1p-GFP and Cdc3p-mCherry showed co-localization at the mother-bud neck during the time of elevated fMAPK pathway activity (Fig. S4A, 100 min, 45%).
A challenge connecting Sho1p localization to fMAPK pathway activity is that protein localization is evaluated by microscopy, whereas MAPK activity is evaluated by phosphoimmunoblot analysis. In mammalian cells, the localization of P~ERK has been evaluated by immunofluorescence (IF), which has revealed insights into the spatial and temporal nature of MAPK pathway signaling (Shapiro et al. 1998; Ingram et al. 2000; Molgaard et al. 2016). Antibodies that detect phosphorylated mammalian ERK also detect the phosphorylated forms of three yeast ERK-type MAPK kinases: Slt2p, for cell wall integrity pathway (Lee et al. 1993)], Kss1p, for fMAPK (Cook et al. 1997)]; and Fus3p, for the mating pathway (Elion et al. 1993)]. Therefore, a problem with iF of P~Kss1p is interference by other P~ERK type MAPKs. To circumvent this problem, the SLT2 gene was disrupted. In the slt2Δ mutant, P~Kss1p was the main band detected using a phospho-MAPK specific antibody that preferentially detects P~Kss1p over P~Fus3p (Fig. 5C, slt2Δ) As expected, by immunoblot analysis P~Kss1p levels were higher in cells carrying the hyperactive MSB2Δ100-818 mutant and reduced in cells lacking the MAPKKK Ste11p (Fig. 5C, ste11Δ). An Alexa 647 fluorophore-conjugated secondary antibody (Thermo fisher, Waltham, MA) showed the same pattern by immunoblot (Fig. 5C) and detected P~Kss1p by immunofluorescence (Fig. S4B). Cells grown in basal conditions showed brighter P~Kss1p levels than the no antibody control (Fig. S4B). Cells grown under pathwayinducing conditions (YEP-GAL) showed brighter P~Kss1p than cells grown in basal conditions (Fig. S4B). Therefore, IF of P~Kss1p is a feasible method to evaluate P~Kss1p activity.
Like many MAP kinases, mammalian ERK enters the nucleus upon activation (Chen et al. 1992; Lenormand et al. 1993; PouyssÉgur et al. 2002; Zehorai et al. 2010). By comparison, Kss1p has a unique regulatory mechanism. Unphosphorylated Kss1p is present in the nucleus in an inhibitory complex with Ste12p, Tec1p, and Dig1p (Bardwell et al. 1996; Cook et al. 1997; Bardwell et al. 1998). Upon phosphorylation by Ste7p, active Kss1p phosphorylates Ste12p, Tec1p and Dig1p and exits the nucleus (Ma et al. 1995; Bardwell et al. 1998; Pelet 2017). Consistent with this mechanism, P~Kss1p showed a punctate pattern in the cytoplasm (Fig. S4B). P~Kss1p levels were next evaluated throughout the cell cycle. P~Kss1p level were higher in M/G1, based on the signal intensity of mitotic and post-mitotic cells where the nucleus was visible in the mother and the daughter cell (Fig. S4C, normalized fluorescence). In addition, cells where Sho1p-GFP was localized to the mother-bud neck showed ~2-fold higher P~Kss1p levels than cells where Sho1p-GFP was localized in buds (Fig. 5D). These results demonstrate that Sho1p is localized to the mother-bud neck when cells experience elevated fMAPK pathway activity during M/G1, which defines spatial and temporal aspects to the regulation of fMAPK pathway signaling.
To further test whether Sho1p’s localization is critical for its activity in the fMAPK pathway, a version of Sho1p was examined where its cytosolic signaling domain was anchored to the plasma membrane by a myristylation tag [pMyr-Sho1p (Raitt et al. 2000)]. pMyr-Sho1 was not able to induce the fMAPK pathway (Fig. 5E), although it has been reported to function in HOG (Raitt et al. 2000). Similarly, a version that contains a point mutation in the myristylation site (pMyr-ASSho1) was also defective for fMAPK pathway signaling. Thus, based on this preliminary experiment, it appears that Sho1p’s cytosolic domain is not merely a passive scaffold in regulating the fMAPK pathway.
Cytokinesis Regulatory Protein Hof1p Regulates the fMAPK Pathway
The fact that Sho1p localizes to the mother-bud neck when cells experience elevated fMAPK pathway activity suggests that the mother-bud neck is a relevant site for fMAPK pathway signaling. To further test this possibility, the role of septins was examined. Septins are heterooligomers that form a cytoskeletal ring at the mother-bud neck and control bud emergence, cytokinesis, and mother-daughter asymmetry (Hall 2008; McMurray and Thorner 2009; Bi and Park 2012). Although septins are essential for viability, and temperature-sensitive alleles of septin genes allow evaluation of septin function. The cdc12-6 mutant shows normal growth at 25°C and is inviable at 37°C. At 30°C, the cdc12-6 mutant exhibits cytokinesis defects. At 30°C, the cdc12-6 mutant had a defect in fMAPK activity, based on immunoblot analysis of P~Kss1p levels (Fig. 6A). The cdc12-6 mutant also showed a defect in the activity of the FUS1-lacZ reporter (Fig. 6B) and the FUS1-HIS3 reporter (Fig. S5A), which in strains lacking an intact mating pathway (ste4Δ) shows dependency on the fMAPK pathway (Cullen et al. 2004). Sho1p-GFP was also mis-localized in the cdc12-6 mutant at 30°C (Fig. 6C, arrows), which may account for the signaling defect seen in this mutant. The cdc12-6 mutant also showed a defect in fMAPK pathway activity at 25°C (Fig. 6B, Fig. S5A), which we have previously reported is due to a bud-site-selection defect (Basu et al. 2016). Therefore, proper septin function is required for fMAPK pathway activity and Sho1p localization.
At the mother-bud neck, Sho1p interacts with proteins that regulate cytokinesis, including Hof1p, Cyk3p, and Inn1p (Labedzka et al. 2012). Hof1p is localized to the mother-bud neck at the initial stages of cytokinesis and moves to the actomyosin ring during cytokinesis (Vallen et al. 2000; Meitinger et al. 2011; Oh et al. 2013). The HOF1 and CYK3 genes, and several other genes that regulate aspects of cytokinesis, including BNI5 and SHS1, were disrupted in wild-type strains of the filamentous (Σ1278b) background. The hof1Δ mutant, but not the cyk3Δ, bni5Δ, or shsl.\ mutants, showed a defect in fMAPK activity based on P~Kss1 levels and FUS1-HIS3 activity (Fig. 6Dand E, Fig. S5B). Cyk3p might not regulate the fMAPK pathway because that protein does not interact with septins and has a distinct function from Hof1p (Oh et al. 2013). The hof1Δ mutant was also defective for invasive growth by the plate-washing assay (Fig. 6E) and the formation of filamentous cells (Fig. 6F) by the single-cell invasive growth assay (Cullen and Sprague 2000). These results establish Hof1p as a regulator of the fMAPK pathway.
Hof1p might regulate the fMAPK pathway by influencing Sho1p localization. Sho1p-YFP and Hof1p-CFP both localized at the mother-bud neck (Fig. 6G, kymograph, Movie 5), and Sho1p was mis-localized in the hof1Δ mutant (Fig. 6H, kymographs; Movie 6, normal cell; Movie 7, abnormal cell), although this phenotype was seen in only ~ 10% of cells, which also showed a cytokinesis defect. Sho1p showed genetic interactions with HOF1 in that overexpression of SHO1, which induces hyperpolarized growth (Vadaie et al. 2008; Pitoniak et al. 2015), exacerbated the growth defect of the hof1Δ mutant (Fig. S5C). Hof1p might alternatively impact bud-site-selection. Bud-site-selection proteins that control axial budding localize to the mother-bud neck (Chant et al. 1995; Sanders and Herskowitz 1996). Budsite-selection proteins also regulate the fMAPK pathway (Basu et al. 2016). The hof1Δ mutant had a defect in bud-site selection (Table 2, Fig. S5D), although the defect was less severe than seen in mutants lacking bud-site-selection proteins (Table 2, bud3Δ). The budding pattern defect of the hof1Δ mutant may not account for its signaling defect, because other cytokinesis regulators that had similar or more severe bud-site-selection defects did not impact fMAPK pathway signaling (Table 2, Fig. S5D). Thus, Hof1p may regulate the fMAPK pathway, in part, by regulating the localization of Sho1p.
DISCUSSION
Signaling pathways are regulated by extrinsic cues, by other pathways in integrated networks, and by spatiotemporal mechanisms to produce a precise signal that generates a physiologically-relevant response. Here, we provide evidence that the MAPK pathway that regulates filamentous growth in yeast is subject to temporal regulation throughout the cell cycle, and spatial regulation by proteins that function at the mother bud neck. We also show that amplification of the fMAPK pathway by positive feedback generates crosstalk to a pathway that shares components, which functions to modulate fMAPK pathway activity. Collectively, these regulatory processes may function to precisely match MAPK pathway activity during a cell differentiation response, as well as provide a mechanism for specific activation of a MAPK pathway that shares components with other MAPK pathways in the same cell type.
Cell-Cycle Regulation of the fMAPK Pathway
A common function for MAPK pathways is to alter cell-cycle progression. In yeast, MAPK pathways alter the progression of the cell cycle during mating (Strickfaden et al. 2007), during filamentous growth (Madhani et al. 1999), and in response to osmotic stress (Waltermann et al. 2010; Radmaneshfar et al. 2013). Mammalian ERKs induce entry into the cell cycle from a G0 state (Seger et al. 1994) and regulate G1/S (Aktas et al. 1997; Leone et al. 1997) and G2/M (Wright et al. 1999; Dangi et al. 2006) transitions including to regulate mitosis (Shapiro et al. 1998). By comparison, it is relatively less well understood how the activity of MAPK pathways might themselves be cell-cycle regulated. One reason for this is that cell synchronization experiments, followed by careful measurement of MAPK pathway activity by phosphoimmunoblot analysis, or by examining protein localization over the cell cycle by time-lapse fluorescence microscopy are technically challenging especially in complex metazoan systems.
Here we show that the activity of the fMAPK pathway is cell-cycle regulated (Fig. S6). The activity of the fMAPK pathway is low in G1, S, and G2, and up in M/G1. By comparison, the HOG pathway can be activated by osmotic stress at any point in the cell cycle. The fact that the HOG pathway is not cell-cycle regulated may not be surprising as cells might be expected to encounter an osmotic stress at any point in their life cycle. One point of regulation occurs at the level of expression of the gene encoding the mucin Msb2p. We also show that the G1/S transcription factor Swi4p regulates the fMAPK pathway activity and filamentous growth. Although Swi4p may regulate the fMAPK pathway in a number of ways, one possibility is by regulating MSB2 expression. The MSB2 promoter contains the regulatory element CACGAAA (Breeden and Nasmyth 1987) 466 bp upstream of the start site, which binds to Swi4p (Iyer et al. 2001; MacIsaac et al. 2006). This regulatory element is near two Ste12p-binding sites ([A]TGAAACA) at 474-481 and 522-530 bp (Cullen et al. 2004). Ste12p regulates the MSB2 promoter to induce its expression through positive feedback. Although Swi4/6p are positive regulators of G1/S transcription, the SBF complex is associated with the transcriptional repressor, Whi5p in early G1 (Costanzo et al. 2004; de Bruin et al. 2004; Palumbo et al. 2016). Therefore, the SBF complex might regulate MSB2 expression in a number of ways. Interestingly, growth of cells in the non-preferred carbon source, GAL partially overrides the cell-cycle regulation of the fMAPK pathway. This might result from induction of MSB2expression by a starvation-dependent transcription factor, and/or by elevated processing of the Msb2p protein, which occurs at elevated levels in galactose (Vadaie et al. 2008).
We also show that the gene encoding the adaptor Sho1p is also subject to cell-cycle control. Several transcription factors that regulate cell-cycle progression bind the SHO1 promoter, including Fhk1p (Ostrow et al. 2014) and Mbp1p (MacIsaac et al. 2006). Whether these proteins impart cell-cycle regulation of Sho1p levels remains to be determined. It has previously been shown that TEC1 expression is induced at M/G1 boundary by Swi5p transcription factor (Cho et al. 1998; Spellman et al. 1998; Wittenberg and Reed 2005). Therefore, cell-cycle regulation of the fMAPK pathway may occur through multiple mechanisms.
Coupling the activity of the fMAPK pathway to the cell cycle may occur for the pathway to regulate intrinsic polarity, which occurs in G1, under some conditions (Prabhakar et al. 2020). Coupling the activity of the fMAPK pathway to the cell cycle may also impact its ability to regulate filamentous growth. We show here that the major cell adhesion molecule and flocculin Flo11p is regulated throughout the cell cycle. Flo11p is required to control adhesion functions under a variety of conditions, including nutrient-replete conditions (Pitoniak et al. 2009; Basu et al. 2016) and may have biological effects on mating (Guo et al. 2000). Intriguingly, another target of the MAPK pathway, BUD8(Adhikari and Cullen 2014), which marks the distal pole and is required for distal budding during filamentous growth (Taheri et al. 2000; Harkins et al. 2001; Cullen and Sprague 2002), is also cell-cycle regulated (Schenkman et al. 2002). The fMAPK pathway also regulates cell-cycle progression by controlling CLN1 expression. Therefore the cell-cycle regulation of a differentiation-type MAPK pathway that itself alters the cell cycle might be critical for its morphogenetic responses to be coordinated. Interestingly, human MEK and ERK are also activated during mitosis in somatic cells to regulate the spindle assembly checkpoint (Shapiro et al. 1998; Horne and Guadagno 2003; Rosner 2007; Cao et al. 2010), proper entry into anaphase (Shapiro et al. 1998; Roberts et al. 2002), and fragmentation of Golgi cisternae (Acharya et al. 1998; Cha and Shapiro 2001; Shaul and Seger 2006). Therefore, MAP kinases may have a general role in regulating events that occur throughout the cell cycle (Pages et al. 1993; Mansour et al. 1994; Wright et al. 1999; Katz et al. 2007).
The Mother-Bud Neck: A Hub for fMAPK Pathway Signaling?
Cumulatively, we have now amassed evidence that proteins that primarily function at the mother-bud neck regulate the fMAPK pathway. These include axial markers that control budsite-selection (Basu et al. 2016), cytokinesis remnant proteins (Prabhakar et al. 2020), the septins themselves (this study), and the cytokinesis regulator Hof1 (this study). A subset of these proteins may function to regulate the localization and/or activity of the adaptor protein Sho1p, which also localized to the mother-bud neck. Sho1p interacts with Hof1p and has functions in cytokinesis (Labedzka et al. 2012). Interestingly, the fMAP kinase Kss1p and cell wall integrity kinase Slt2p also act in septum assembly during cytokinesis (PÉrez et al. 2016). Thus, a fMAPK pathway complex may function at the neck to coordinate cytokinesis and next round of bud emergence. The spatial localization of these proteins can be viewed as a type of compartmentalization, which is a classic mode for maintaining signal specificity (Ebisuya et al. 2005; Doncic et al. 2015). Compartmentalization occurs on many levels, by restricting signaling to different cell types, organelles, parts of the plasma membrane, and even at different points in the cell cycle (Henis et al. 2009).
A Positive Feedback Loop by One Pathway Sets up a Negative Feedback Loop by A Pathway with Shared Components
Most signaling pathways share components with other pathways. Pathway interactions can allow for modulation of pathway outputs. Here we provide evidence for cross-pathway feedback between the fMAPK pathway and the HOG pathway. Positive feedback through the fMAPK pathway induces HOG pathway activity, presumably to modulate fMAPK pathway activity. Interestingly, bleed through to the HOG pathway creates a loss of signal from the positivefeedback loop and a negative signal by stimulation of an antagonistic pathway. Such cross feedback might also impact target gene expression, although overexpression of Msb2p induces a non-overlapping set of targets as overexpression of Hkr1p (Pitoniak et al. 2009). The crosspathway feedback also fits with the fact that the HOG pathway is induced by non-preferred carbon sources (Adhikari and Cullen 2014). Such modulation may fine-tune fMAPK pathway activity, which needs to be at the right level to promote filamentous growth and bud emergence, and at elevated levels, can lead to morphogenetic problems.
General Ramifications to Pathway Specificity
Pathway specificity is one of the central questions in the signaling field. Signal duration, magnitude and subcellular compartmentalization of pathway regulators can have a profound impact on signal specificity and cellular outputs. It may be possible that the temporal regulation, by cell-cycle regulation of MSB2 and SHO1 gene expression, might be directly mechanistically connected to their spatial regulation, such as Sho1p’s localization at the mother-bud neck in M/G1. However, this need not necessarily be the case. Because both proteins are required to activate the fMAPK pathway, temporal and spatial control might acts in a coincidence manner to insure precise spatiotemporal activation of the fMAPK pathway. Although signaling pathways sense and respond to unique stimuli, often times multiple MAPK pathways collaborate to generate an appropriate response (Errede et al. 1995; Zarzov et al. 1996; Buehrer and Errede 1997; BaltanÁs et al. 2013; Adhikari and Cullen 2014; Prabhakar et al. 2020). Differences in activity between pathways that share components throughout the cell cycle may impact specificity and cell differentiation.
MATERIALS AND METHODS
Strains and Plasmids
Yeast strains are described in Table 1. Gene disruptions were made with antibiotic resistance markers KanMX6(Longtine et al. 1998), HYG and NAT(Goldstein and McCusker 1999) using PCR-based methods. Pop-in pop-out strategy was used to make internal epitope fusions (Schneider et al. 1995). Some strains were made ura3- by selection on 5-fluoroorotic acid (5-FOA). Gene disruptions were confirmed by PCR-based Southern analysis and also by phenotype when applicable. The swi6Δ mutant had a severe growth defect, which prevented evaluation of fMAPK and HOG pathway activities.
Most of the plasmids used in this study belong to the pRS series of plasmids (pRS315 and pRS316)(Sikorski and Hieter 1989). pGFP-MSB2(Adhikari et al. 2015b), pRS316-SHO1-GFP(Marles et al. 2004), pSHO1P120L(Vadaie et al. 2008), pSHO1-GFP::NAT(Prabhakar et al. 2020), pMyr-SHO1 and pMyrAS-SHO1(Raitt et al. 2000), YCp50-STE11-4(Stevenson et al. 1992) and pSTE4(Stevenson et al. 1992) have been described.
Microbial Techniques
Standard methods were followed during yeast and bacterial strain manipulations (Sambrook 1989; Rose 1990). Budding pattern was determined as described (Cullen and Sprague 2002). The activity of the FUS1-HIS3(McCaffrey et al. 1987) growth reporter in cells lacking an intact mating pathway (ste4Δ) is dependent on components of the fMAPK pathway (Cullen et al. 2004) and was determined by growth of cells on media lacking histidine and supplemented with ATA (3-amino-1,2,4-triazole) for 3 d. Beta-galactosidase assays to assess the activity of the FUS1-lacZ reporter were performed as described (Cullen et al. 2000). The single-cell invasive growth assay (Cullen and Sprague 2000) and the plate-washing assay (Roberts and Fink1994) have been previously described.
Immunoblot Analysis
Immunoblot analysis to detect phosphorylated MAP kinases has been described (Sabbagh et al. 2001; Lee and Dohlman 2008; Basu et al. 2016; Prabhakar et al. 2020). Proteins were precipitated from cell pellets stored at −80°C by trichloroacetic acid (TCA) and analyzed on 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Proteins were transferred to nitrocellulose membranes (Amersham™ Protran™ Premium 0.45 μm NC, GE Healthcare Life sciences, 10600003). For Msb2p-HA and Hkr1p-HA blots, 6% acrylamide gel was used.
ERK-type MAP kinases (P~Kss1p, P~Fus3p and P~Slt2p) were detected using α-p44/42 antibodies (Cell Signaling Technology, Danvers, MA, 4370 and 9101) at a 1:5,000 dilution. 9101 gave a stronger signal for P~Kss1p over P~Fus3p, while 4370 detected both proteins with similar strength. α-p38-type antibody at 1:5,000 dilution (Cell Signaling Technology, Danvers, MA 9211) was used to detect P~Hog1p. α-HA antibody at a 1:5,000 dilution (Roche Diagnostics, 12CA5) was used to detect Clb2p-HA, Msb2p-HA and Hkr1p-HA. Clb2p-Myc was detected using α-c-Myc antibody at 1:5,000 dilution (Santa Cruz Biotechnolog, Dallas, TX, 9E10) and Sho1p-GFP was detected using α-GFP antibody at 1:5,000 dilution (Roche Diagnostics, clones 7.1 and 13.1, 11814460001). α-Pgk1 antibody was used at a 1:5,000 dilution for total protein levels (Novex, 459250). For secondary antibodies, goat α-rabbit secondary IgG-HRP antibody was used at a 1:10,000 dilution (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, 111-035-144). Goat α-mouse secondary IgG-HRP antibody was used at a 1:5,000 dilution (BioRad Laboratories, Hercules, CA, 170-6516). Phospho-MAPK antibodies were incubated in 1X TBST (10 mM TRIS-HCl pH 8, 150 mM NaCl, 0.05% Tween 20) with 5% BSA. For all other antibodies, 1X TBST with 5% non-fat dried milk was used. Primary incubations were carried out for 16 h at 4°C. Secondary incubations were carried out for 1 h at 25°C.
Cell Synchronization and Cell-Cycle Experiments
Cell synchronization by elutriation (Rosebrock 2017) was not feasible for cells of the ∑1278b background because cells fail to separate, even those lacking the adhesion molecule Flo11p (Vandermeulen and Cullen 2020). Cell synchronization experiments were performed as previously described (Breeden 1997; Prabhakar et al. 2020). Overnight cultures were resuspended in fresh media and grown to an optical density (O.D.) A600 of 0.2 at 30°C. Strains that required synthetic media (SD-URA) to maintain plasmid selection were harvested and resuspended in equal volume of YEPD and incubated for 1 h at 30°C prior to α-factor treatment. 10 ml aliquot was harvested as asynchronous culture. To arrest cells in G1,α factor was added to a final concentration of 5 μg/ml and the culture was incubated for 2 h at 30°C. 10 ml aliquots were harvested at 5 min, 30 min and 2 h during α-factor treatment. To arrest cells in S phase, hydroxyurea (HU) (MilliporeSigma, Burlington, MA, H8627) was added to a final concentration of 400 mM and incubated for 4 h. Arrested cells were washed twice with water (pre-warmed at 30°C) and resuspended in fresh YEPD or YEP-GAL media (pre-warmed at 30°C) to release cells into the cell cycle. 10 ml aliquots were harvested every 10 or 20 min and stored at −80°C.
DIC and Fluorescence Microscopy
Differential-interference-contrast (DIC) and fluorescence microscopy using FiTC and TRiTC filter sets were performed using an Axioplan 2 fluorescent microscope (Zeiss, Oberkochen, Germany) with a Plan-Apochromat 100X/1.4 (oil) objective (N.A. 1.4) (cover slip 0.17) (Zeiss, Oberkochen, Germany). Digital images were obtained at multiple focal planes with the Axiocam MRm camera (Zeiss, Oberkochen, Germany) and Axiovision 4.4 software (Zeiss, Oberkochen, Germany). Adjustments to brightness and contrast were made in Adobe Photoshop (Adobe, San Jose, CA).
Time-lapse microscopy was performed on a Zeiss 710 confocal microscope (Zeiss, Oberkochen, Germany) equipped with a Plan-Apochromat 40x/1.4 Oil DIC M27 objective. For GFP, 488nm laser (496nm-548nm filter); for mCherry, 580 nm laser (589nm-708nm filter); for YFP, 517 nm laser (532-620 filter) and for CFP, 458 nm laser (462-532 filter) were used. For Sho1p-GFP time lapse, 9 z-stacks 1μm thick; for GAL-GFP-Msb2p time lapse 6 z-stacks 0.6 μm thick; and for Hof1p-CFP and Sho1p-YFP co-localization, 8 z-stacks 1.2 μm thick were captured at 10 min intervals.
Cells for time-lapse and co-localization studies were prepared as described in (Prabhakar et al. 2020). Cells were grown at 30°C for 16 h in SD-URA and diluted to < 0.1 O.D. 10 μL of diluted cells were placed under agarose pad (1%) prepared inside a 12 mm Nunc glass base dish (150680, Thermo Scientific, Waltham, MA). 100 μl of water was placed in the dish to prevent the agarose pad from drying and the petri dish was incubated at 30°C for 4 h prior to imaging.
Indirect Immunofluorescence
Indirect immunofluorescence was performed as previously described with following modifications (Amberg et al. 2006; Schnell et al. 2012). Cells were grown to mid-log stage and 8% fresh paraformaldehyde (PFA) prepared in PBS (pH 7.4) was added directly to the culture (final concentration, 4%) for 10 min with shaking. Cells were harvested for 3 min at 350g and resuspended in KM solution (40 mM KPO4 pH 6.5, 500 μM MgCl2) containing 4% PFA for 1 h at 30°C with gentle shaking. Cells were washed twice with KM solution and once with KM solution containing 1.2 M sorbitol. After the last wash, cells were resuspended in 500 μl KM solution containing 1.2 M sorbitol and 60 μl Zymolyase (50 mg/ml 20T) for 20 min at 37°C. During zymolyase treatment, samples were periodically examined by DIC microscopy for cells with dull gray appearance and intact morphology (Niu et al. 2011). After Zymolyase treatment, cells were washed at 300g with KM solution containing 1.2 M sorbitol and resuspended in the same solution. Wells of Teflon-faced slides (MP Biomedicals, Santa Ana, CA, 096041205) were coated with 20 μl poly-L-lysine (Cultrex Poly-L-Lysine, Bio-Techne, Minneapolis, MN, 3438100-01) and incubated for 10 min at 24°C in a humid chamber. All solutions were centrifuged at 16,000g for 20 min at 4°C prior to adding to the wells. Wells were washed 5 times with 20 μl water and air dried. 20 μl of cells were spotted onto poly-L-lysine coated wells for 10 min at 24°C in the humid chamber. Excess solution was aspirated, and the slides were plunged into a coplin jar containing cold methanol for 6 minutes followed by cold acetone for 30 sec. Fixed and permeabilized cells were blocked using Image-iT™ FX Signal Enhancer (Thermo Fisher, Waltham, MA, i36933) for 30 min at 24°C in the humid chamber with gentle shaking. Excess solution was aspirated, and wells were washed 5 times with blocking buffer [PBS (pH 7.4), 5% normal goat serum (50062Z, Thermo Fisher, Waltham, MA), 1% BSA (80055-674, MilliporeSigma, Burlington, MA), 2% TritonX-100]. Wells were blocked again with 20 μl blocking buffer for 30 min at 24°C in the humid chamber with gentle shaking. After aspiration, cells were incubated with 20 μl of rabbit anti-p44/42 primary antibody (Cell Signaling Technology, Danvers, MA, 9101), prepared in blocking buffer at 1:20 dilution for 12 h at 24°C in a humid chamber with gentle shaking. Wells were washed 5 times for 5 min each with blocking buffer and co-stained with Goat anti-Rabbit IgG (H+L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor 647 (Thermo Fisher, Waltham, MA, A-21245) and DAPI (4′,6-diamidino-2-phenylindole) at 1:1000 dilution each for 4 h at 24°C in a dark humid chamber with gentle shaking. Wells were washed 5 times for 10 min each with blocking buffer. After the last wash, wells were sealed with ProLong™ Diamond Antifade Mountant (Thermo Fisher, Waltham, MA, P36970) and covered with a cover slip. Slides were incubated in dark at 24°C for 24 h prior to imaging.
Image Analysis
Quantitation of P~Kss1p by immunofluorescence is discussed elsewhere (Prabhakar and Cullen, In Prep) Briefly, total fluorescent intensity for each cell was measured in ImageJ by subtracting background intensity from the mean fluorescent intensity, which was recorded using Analyze>Measure option. Normalized fluorescent intensity for each cell was quantified as previously described (Okada et al. 2017; Prabhakar et al. 2020) using a custom MATLAB (MATLAB R2016b, The MathWorks, Inc., Natick, MA) code (Prabhakar et al. 2020). For each cell, pixel intensities greater than the mean+2 STD were background subtracted, normalized to the peak value (which was set to 1), and summed.
For time-lapse microscopy, raw images were imported into ImageJ. Cells were registered using HyperStackReg plugin (ThÉvenaz et al. 1998; Sharma 2018) to remove drift in the position of cells that occurred during imaging. Grayscale fluorescence images were converted to maximum intensity projection and inverted. Kymographs were performed as described (Prabhakar et al. 2020).
Quantitative PCR Analysis
Quantitative RT-qPCR was performed as previously described (Adhikari and Cullen 2014; Chow et al. 2019a; Prabhakar et al. 2020). Samples harvested during cell-cycle experiments were used for total RNA extraction, which was done by hot acid phenol-chloroform treatment and further purified using RNeasy Mini Kit (Qiagen, Hilden, Germany, 74104). RNA stability was determined by agarose gel electrophoresis in 0.8% agarose Tris-Borato-EDTA (TBE, 89 mM Tris base, 89 mM Boric acid, 2mM EDTA). Concentration and purity were determined by absorbance using NanoDrop (NanoDrop, 2000C, Thermo Fisher Scientific, Waltham, MA). Concentration of total RNA was adjusted to 60 ng/μl and cDNA was synthesized using iScript Reverse Transcriptase Supermix (BioRad, Hercules, CA, 1708840). qPCR was performed using iTaq Universal SYBR Green Supermix (BioRad, Hercules, CA, 1725120) on BioRad thermocycler (CFX384 Real-Time System). Reactions contained 10 μl samples (2.5 μl 60 ng/μl cDNA, 0.2 μM each primer, 5 μl SYBRGreen master mix). Relative gene expression was calculated using the 2−ΔCt formula, where Ct is defined as the cycle at which fluorescence was determined to be statistically significant above background; ΔCt is the difference in Ct of the gene of interest and the housekeeping gene (ACT1). The primers used were: MSB2 forward (5′-CACTGCAAGCAGGTGGCTCT-3′), MSB2 reverse (5′-GAGGAGCCCGACAGTGTTGC-3′); HKR1 forward (5′-AAACCATGGGCGAAAATGGC-3′), HKR1 Reverse (5′-AAGGCAGGGGCTGTGAATAC-3′); KSS1 forward (5′-CCCAAGTGATGAGCCGGAAT-3′), KSS1 reverse (5′-TGGGCACTTCTTCCTCCTCT-3′); SHO1 forward (5′-AACTACGATGGGAGACACTTTG-3′), SHO1 reverse (5′-TCGTAAGCATCATCGTCATCAG-3′) (Adhikari and Cullen 2014); TEC1 forward (5′-ATGTTTCCAGAAGCCGTAGTT-3′), TEC1 reverse (5′-TTTAGCACCCAGTCCAGTATTT-3′) (Adhikari and Cullen 2014); STE12 forward (5′-GCAATCTTACCCAAACGGAATG-3′), STE12 reverse (5′-AATCGTCCGCGCCATAAA-3′) (Adhikari and Cullen 2014); FLO11 forward (5′-CACTTTTGAAGTTTATGCCACACAAG-3′), FLO11 reverse (5′-CTTGCATATTGAGCGGCACTAC-3′) (Chen and Fink 2006) and ACT1 forward (5′-TGGATTCCGGTGATGGTGTT-3′), ACT1 reverse (5′-CGGCCAAATCGATTCTCAA-3′) (Chow et al. 2019b). Experiments were performed with two independent biological replicates and two technical replicates for each biological replicate.
Statistical Analysis
Statistical tests and sample size (n) have been described in figure legends wherever applicable. Statistical analyses were performed in Microsoft Excel and Minitab (www.minitab.com). Oneway ANOVA with Tukey’s test and/or Dunnett’s test was used for statistical analysis.
ACKNOWLEDGEMENTS
The work was supported from a grant from the NIH (GM#098629). Thanks to Haruo Saito (University of Toyko) for reagents. Andrew Pitoniak helped with experiments.
Footnotes
The authors have no competing interests in the study.
ABBREVIATIONS
- ATA
- (3-amino-1,2,4-triazole)
- 5-FOA
- 5-fluoroorotic acid
- CFP
- cyan fluorescent protein
- D
- dextrose
- DAPI
- 4′,6-diamidino-2-phenylindole
- DIC
- differential interference contrast
- GAL
- galactose
- GAP
- GTPase activating protein
- GEF
- guanine nucleotide exchange factor
- GTPase
- guanine nucleotide triphosphatase
- GFP
- green fluorescent protein
- GLU
- glucose
- GAL
- galactose
- HA
- hemaglutinin
- HOG
- high osmolarity glycerol response
- HU
- hydroxyurea
- MAPK
- mitogen activated protein kinase
- O.D.
- optical density
- PAK
- p21 activated kinase
- PFA
- parafolmaldehyde
- RT-qPCR
- Reverse transcriptase quantitative polymerase chain reaction
- PM
- plasma membrane
- Rho
- Ras homology
- SDS-PAGE
- sodium dodecyl sulfatepolyacrylamide gel electrophoresis
- S.E.M
- standard error of mean
- TBE
- Tris-Borate-EDTA
- TCA
- trichloroacetic acid
- WT
- wild type
- YFP
- yellow fluorescent protein
- YNB
- Yeast Nitrogen base.