Introduction

Evolutionary considerations predict that fermentation, an energy generating pathway in which ATP yield is low but operates at a high rate, has a selective advantage when organisms compete for shared energy sources. On the other hand, respiration, a pathway that offers high ATP yield at a lower rate, lends itself for co-operative utilization of energy resources and may allow the cells to spatially organize into defined structures (Pfeiffer et al. 2001). A classic example that seems to follow this dictum is the fermentative mode of growth exhibited by the yeast Saccharomyces cerevisiae (Pfeiffer and Schuster 2005) in presence of abundant glucose. It switches over to respiration as glucose concentration decreases (van Dijken et al. 1993). In addition to this metabolic switch, it has been demonstrated that diploid cells of Saccharomyces cerevisiae exhibit a morphologic switch when exposed to abundant glucose but low ammonium, forming a spatially defined structure called pseudohypha at the end of 6 days of incubation (Gimeno et al. 1992; Lorenz and Heitman 1997). Recent evidence, however, has indicated that pseudohyphal differentiation occurs only when the glucose concentration is reduced to a level that results in a switch to the respiratory mode of growth (Iyer et al. 2008). This is consistent with the evolutionary considerations discussed above. That the respiratory pathway of energy production favors pseudohyphal differentiation is supported by other independent observations. For example, petite cells that are incapable of respiration are defective in pseudohyphal differentiation (Kang and Jiang 2005; Jin et al. 2008). SNF1, required to de-repress mitochondrial function as the glucose concentration decreases is necessary for pseudohyphal transition as well (Kuchin et al. 2002). Thus, the fundamental question is how does S. cerevisiae signal the shift from fermentative to respiratory mode of growth to the downstream elements that induce the pseudohyphal differentiation?

PKA mediated signaling is pivotal in regulating growth, filamentation and stress response depending upon nutrient availability (reviewed in Rubio-Texeira et al. 2009; Leadsham and Gourlay 2010; Papp et al. 2016). For example, rapid cell division in response to abundant glucose is mediated through the cAMP-PKA pathway (Smets et al. 2010) which also activates the expression of FLO11, an essential and the primeval target for filamentation response in S. cerevisiae (Rupp et al. 1999). However, FLO11 expression is repressed under high glucose through the well known glucose repression signaling mechanism (Gagiano et al. 1999; Kuchin et al. 2002). Thus, FLO11 is under positive regulation mediated by PKA pathway and negative control exerted by excess glucose. The paradox, however, is how does the alleviation of glucose repression of FLO11 which occurs as the glucose concentration decreases, co-ordinate with the signaling machinery that sustains the PKA mediated positive signal for filamentation. In other words, upon glucose depletion, what are the signaling mechanisms that trigger pseudohyphal differentiation?

MEP2, an ammonium transceptor, was shown to be essential for nitrogen mediated signaling of pseudohyphal differentiation (Lorenz and Heitman 1998). The same authors showed that a constitutive allele of GPA2 overcomes the filamentation defect of a mep2 mutant indicating that MEP2 triggers pseudophyphal differentiation by activating the cAMP-PKA pathway (Lorenz and Heitman 1998). A large body of data is available on MEP2 mediated signaling (Boeckstaens et al. 2007, 2014; Rutherford et al. 2008) including a recent study where the crystal structure of Mep2p has been reported (van den Berg et al. 2016). Nonetheless, how MEP2 signals the PKA pathway in a GPA2 dependent fashion in response to low glucose and low ammonium remains unclear.

GPR1, a glucose sensing GPCR protein, which was identified in a two-hybrid screen using the Gα protein GPA2 as bait (Yun et al. 1997; Xue et al. 1998; Kraakman et al. 1999) was also shown to trigger filamentation via the cAMP-PKA pathway (Lorenz et al. 2000). Subsequently, it was proposed that the redundant kelch repeat proteins KRH1 (GPB1) and KRH2 (GPB2) prevent pseudohyphal differentiation through negative regulation of PKA signaling by inhibiting GPA2 function (Harashima and Heitman 2002; Battle et al. 2003). These proteins, thought to be Gβ mimics (Harashima and Heitman 2002), were shown to inhibit RAS signaling (Harashima et al. 2006). Krh/Gpb proteins seem to interconnect signaling from GPR1 and RAS to regulate the cAMP-PKA pathway by way of their interaction with GPA2 as well as IRA1/2 (reviewed in Cullen and Spargue 2012). However, there are opposing reports on the interaction between KRH1/2 and IRA1/2 where on one hand evidence supports the possibility of IRA being stabilized by KRH (Harashima et al. 2006) while on the other hand it appears that KRH facilitates degradation of IRA (Phan et al. 2010). Deletion of either KRH1/GPB2 or KRH2/GPB1 resulted in enhanced filamentation response. It was demonstrated that KRH1/2 negatively regulated filamentation response by inhibiting GPA2-GPR1 coupling (Harashima and Heitman 2005). In contrast, another study (Peeters et al. 2006) showed that KRH1/2 increased the interaction between the catalytic and regulatory subunit of PKA which was later shown to be achieved through the inhibition of PKA mediated phosphorylation of BCY1 by KRH1/2 (Budhwar et al. 2010). A subsequent study showed that KRH1/2 mediated effect on BCY1 is an indirect consequence of their effect on PKA (Budhwar et al. 2011). Peeters et al. (2006) also showed that the deletion of PKA catalytic subunits abolished the down regulation of glycogen and trehalose accumulation observed upon KRH1/2 deletion, suggesting that KRH1/2 inhibit PKA. Based on these observations, and contrary to the previous report (Harashima and Heitman 2005), it was proposed (Peeters et al. 2006) that GPA2 activates PKA by inhibiting KRH1/2. Although there is evidence to indicate that KRH1/2 function is independent of cAMP (Lu and Hirsch 2005), an earlier report showed that KRH1/2 deletion resulted in elevation of intracellular cAMP concentrations (Harashima and Heitman 2002). Counter to the above studies which considered KRH1/2 to be functionally redundant in pseudohyphal differentiation, the only study that demonstrated distinct functions for KRH1 and KRH2 was with regard to their role in IRA2 mediated RAS signaling (Phan et al. 2010).

It has been suggested that the promiscuous interaction of KRH1/2 may facilitate the coupling of nutrient sensing to cAMP production to execute diverse responses (Harashima and Heitman 2005). It was also proposed that in addition to GPA2, the activity of KRH1/2 may be controlled by other factors, and thus KRH1/2 may play a role in integrating the diverse signals that impinge on PKA (Peeters et al. 2006). Based on our previous observation that cells committed to pseudohyphal differentiation become refractory to glucose mediated repression of pseudohyphae formation (Iyer et al. 2008) and on the ability of KRH1/2 to interact with GPR1, GPA2 as well as PKA, we hypothesized that KRH1 and KRH2 could play distinct roles. We surmised that such a differential role could be uncovered only on the induction of pseudohyphae in synthetic low ammonium and low dextrose (SLALD) medium (Iyer et al. 2008) but not in synthetic low ammonium dextrose (SLAD) medium (Gimeno et al. 1992). Our results clearly demonstrate that KRH1 and KRH2 are non-redundant with respect to their function in filamentation response. Further, our observations indicate that MEP2 probably activates the cAMP-PKA pathway through suppression of KRH2 but not KRH1. We suggest that these distinct regulatory mechanisms of KRH1 and KRH2 play an important role in processing and conveying the signal to the downstream elements in response to glucose as well as ammonium depletion to elicit the filamentation response.

Materials and methods

Yeast strains and plasmids

Strains used in this study are isogenic derivatives of Σ1278 strain and listed in Table 1. Gene deletions were carried out using marker based polymerase chain reaction (PCR) methods as described (Wach et al. 1994). Plasmid pRI3 (for KRH1 over-expression) was constructed by cloning 6his-tagged KRH1 as a SmaI-PstI fragment under GPD promoter in p426GPD vector. Plasmid pRI4 (for KRH2 over-expression) was constructed using recombination mediated in vivo ligation in yeast as described earlier (Oldenburg et al. 1997) to clone 6his-tagged KRH2 under GPD promoter in p426GPD vector.

Table 1 List of strains used in this study
Full size table

Assay for filamentation response

Standard methods were used to score pseudohyphal growth (Gimeno et al. 1992; Lorenz and Heitman 1997). Briefly, diploid yeast cells pre-grown in YPD or synthetic uracil drop-out media were washed and then spread on Synthetic low ammonium dextrose (SLAD) or synthetic low ammonium low dextrose (SLALD) media as indicated in the figures. Representative colonies were photographed at the end of 6 days of incubation at 30 °C. Glucose replenishment assay was carried out as described earlier (Iyer et al. 2008). In the current study, however, the wild-type or mutant strains transformed with the vector and exposed to glucose replenishment served as the corresponding experimental controls. Images of the colonies were captured at the end of 4 days of incubation at 30 °C after glucose addition. All experiments were repeated at least three times before representative colonies were photographed. Colonies were photographed at 10× magnification using a Nikon Coolpix camera attached to a Nikon microscope. For the invasive growth assay, haploid strains were patched onto Yeast extract Peptone Dextrose (YPD) agar and the plates were photographed before as well as after washing the agar surface as described earlier (Kuchin et al. 2002).

β galactosidase assay

FLO11 activity was measured using a P FLO11 -lacZ construct with the full-length promoter sequence as reported earlier (Rupp et al. 1999). β galactosidase assay using chlorophenyl β-D-galactopyranoside as a substrate was employed as described (Som et al. 1988). Assays were done in triplicates and repeated at least five times. Statistical significance of the result was determined using the student’s t test.

Western blot analysis

Cells from confluent 20 h old cultures in synthetic uracil drop-out medium were washed twice, transferred to SLAD or SLALD medium as required and grown for 12–14 h. The 6his-tagged Krh1/2p present in the cell free extract of the above cultures was allowed to bind to Ni–NTA agarose beads from QIAGEN. The bound fraction was eluted and subjected to western blot analysis using anti-His antibody from Santa Cruz Biotechnology Inc., as per the manufacturer’s recommendations. The enrichment of Krh1/2p was required as we were unable to detect either Krh1p or Krh2p in whole cell extracts obtained from cells grown on synthetic uracil drop-out medium. The western blot analysis was performed multiple times and a representative result is presented.

Results

MEP2 signals for pseudohyphal differentiation by inhibiting KRH2

Iyer et al. demonstrated that pseudohyphae formation is a response to ammonium depletion only when glucose is limiting (Iyer et al. 2008). The authors further showed that GPR1 suppressed pseudohyphae and that the filamentation defect of a mep2 mutant is overcome upon disruption of GPR1 in low glucose. Although experimental evidence indicates that MEP2 functions upstream of GPA2 to trigger pseudohyphal response through the cAMP-PKA pathway (Lorenz and Heitman 1998), no signaling partner of MEP2 has been identified till date. Based on the above and the observation that KRH1/2 inhibit pseudohyphae through GPR1-GPA2-cAMP signaling (Battle et al. 2003; Harashima and Heitman 2002, 2005; Peeters et al. 2006), we surmised that MEP2 may regulate filamentation response through KRH1/2.

To test this hypothesis, KRH1 and KRH2 were independently disrupted in a mep2Δ mutant strain to monitor pseudohyphae formation. Both the mep2Δkrh1Δ as well as the mep2Δkrh2Δ mutants were unable to form pseudohyphae in SLAD (Synthetic low ammonium dextrose) medium (Fig. 1). In SLALD (Synthetic low ammonium low dextrose), however, the inability of the mep2Δ mutant to put forth pseudohyphae was overcome only by KRH2 disruption, although the pseudohyphae formed were fewer and shorter. On the contrary, the mep2Δkrh1Δ double mutant was unable to filament indicating that KRH2 but not KRH1 suppresses filamentation in the absence of MEP2 in low glucose (Fig. 1). This result was further supported by the invasive growth assay where the mep2Δkrh2Δ strain exhibited invasive growth while the mep2Δkrh1Δ strain did not (Fig. 2). For biochemical corroboration of the above observation, FLO11 expression was determined as a function of β-galactosidase activity in SLAD as well as SLALD medium using a P FLO11 -lacZ reporter plasmid (Table 2). As expected, in SLALD medium, the mep2Δkrh2Δ mutant exhibited higher β-galactosidase activity as compared to that in SLAD medium. Moreover, FLO11 expression was significantly higher in mep2Δkrh2Δ strain as compared to the mep2Δkrh1Δ strain, consistent with the filamentation response of the two strains. The diminished filamentation exhibited by the mep2Δkrh2Δ strain in SLALD (see Fig. 1) is probably due to lower FLO11 expression when compared to that of the wild-type strain. The above observations indicate that MEP2 exerts an inhibitory effect on KRH2 thereby activating filamentation response. Thus, the results presented here provide the first genetic evidence of a possible interacting partner for MEP2. The genetic interaction of MEP2 with KRH2 but not KRH1 indicates that KRH1 and KRH2 may be functionally non-redundant.

Table 2 Relative expression of FLO11 in different strains
Full size table
Fig. 1
figure 1

Pseudohyphal growth phenotype of mutant strains Images of colonies on SLAD as well as SLALD media as indicated in the figure. Three representative colonies are shown for each mutant

Full size image
Fig. 2
figure 2

Haploid invasive growth phenotype of mutant strains Images of colony patches on YPD medium; BW Before wash, AW After wash. a, b and c represent three independent experiments

Full size image

KRH1 and KRH2 are functionally non-redundant

It is clear from the above results that the disparate behaviour of KRH1 and KRH2 could be ascertained only in SLALD but not SLAD medium. To further test whether extracellular low glucose concentration modulates the function of KRH1 and KRH2 differentially, pseudohyphae formation in response to over-expression of KRH1/2 under GPD promoter was monitored on SLAD as well as SLALD media. We deliberately used a heterologous promoter to ensure that expression is independent of the growth conditions which can otherwise confound our interpretation. As reported earlier (Battle et al. 2003; Harashima and Heitman 2002), over-expression of either KRH1 or KRH2 suppressed pseudohyphae on SLAD medium (Fig. 3). However, filamentation was suppressed upon KRH2 but not KRH1 over-expression in SLALD medium (Fig. 4, compare middle and bottom – panels a, b and c). Only in the krh2∆ strain, we observed a mild retardation in pseudohyphae formation upon KRH1 overexpression when compared with the corresponding vector control. It is possible that in the absence of KRH2, overexpression of KRH1 has a mild, non-specific negative effect by virtue of being a paralogue of KRH2. Nonetheless, the pseudohyphae formed were more pronounced as compared to the wild-type strain under the same conditions. It is clear from the above data that KRH1 and KRH2 may have different roles only when both glucose as well as ammonium are limiting. Surprisingly, western blot analysis revealed that although KRH1 as well as KRH2 were expressed in synthetic uracil drop-out medium (with abundant glucose and ammonium), only KRH2 was expressed in SLAD or SLALD medium (Fig. 5) indicating that KRH1 may be suppressed when ammonium levels are low. In other words, when ammonium limitation occurs simultaneously with glucose limitation, MEP2 is probably required to suppress only KRH2 but not KRH1 to trigger filamentation response. To examine further whether glucose too has a role in the regulation of KRH1/2 we employed our earlier observation that in a wild-type strain, glucose replenishment was unable to inhibit filamentation once the cells are committed to differentiation. That is, glucose replenishment suppressed pseudohyphae formation only in SLAD but not in SLALD medium (Iyer et al. 2008). To determine whether KRH1/2 have any role in this glucose dependent commitment to pseudohyphae formation, the effect of glucose replenishment was monitored under SLAD as well as SLALD conditions in strains wherein KRH1/2 is either deleted or over-expressed. Upon glucose replenishment in SLAD medium with abundant glucose, pseudohyphae were suppressed in a krh1Δ but not krh2Δ mutant (Fig. 6, compare b and c of left panel) demonstrating that KRH1 and KRH2 are functionally non-redundant. In addition, under the same condition of growth, filamentation in the krh2Δ mutant was suppressed upon KRH2 over-expression (Fig. 6, compare left and right panels of c), indicating that KRH2 function is not suppressed by glucose in SLAD medium. However, in SLALD medium, upon glucose replenishment, pseudohyphae were formed even when KRH2 was over-expressed (Fig. 6, compare c and g of right panel) indicating that glucose suppressed the function of KRH2. Thus, it is evident from the difference in the phenotypes observed in SLAD as compared to SLALD medium that glucose mediated suppression of KRH2 function is dependent on initial glucose concentration.

Fig. 3
figure 3

Effect of over-expression of KRH1/2 in SLAD medium Over-expression of KRH1 and KRH2 independently in the wild-type (WT), krh1Δ as well as krh2Δ strains. When over-expressed both KRH1 and KRH2 inhibit pseudohyphae formation in SLAD medium. Images of three representative colonies are shown for each experimental condition

Full size image
Fig. 4
figure 4

Effect of over-expression of KRH1/2 in SLALD medium KRH1 and KRH2 are separately over-expressed in the WT, krh1Δ as well as krh2Δ strains. In SLALD medium, upon over-expression KRH2 but not KRH1 suppresses filamentation. Three representative colonies are shown for each experimental condition

Full size image
Fig. 5
figure 5

KRH1/2 expression under different nutrient conditions KRH1 is inactivated in low ammonium concentration. His-tagged KRH1 as well as KRH2 were purified using Ni–NTA agarose column and probed using anti-His antibodies. Equal protein was loaded onto the Ni–NTA column as represented by the coommassie blue stained, corresponding input fraction which represents 0.1% of the total protein used for binding

Full size image
Fig. 6
figure 6

Effect of glucose replenishment Over-expression of KRH2 in the WT, krh1Δ, krh2Δ and krh1Δkrh2Δ strains respectively in SLAD (a, b, c, d) and SLALD (e, f, g, h) medium as indicated in the figure. Images represent colonies (in triplicates) after exposure to glucose addition

Full size image

Based on the results presented above, we surmised that KRH2 may function to prevent filamentation response when glucose is abundant while KRH1 may not be required for this function. To determine whether the above role of KRH2 is in any way modulated by KRH1, the effect of glucose replenishment on KRH2 over expression was monitored separately in the mutants defective for KRH1 and KRH2. Interestingly, glucose suppressed KRH2 more strongly in the krh2Δ strain resulting in increased pseudohyphae formation in a krh2Δ mutant compared to a krh1Δ strain (Fig. 6, compare f and g of right panel). Further, pseudohyphae formed in the krh2Δ mutant upon glucose replenishment when KRH2 was over-expressed were abolished in a krh1Δkrh2Δ double mutant (Fig. 6, compare g and h of right panel) indicating that filamentation occurring as a result of glucose mediated suppression of KRH2 is dependent on KRH1.

Discussion

There is a large body of evidence to show that transcriptional factors responsive to glucose deprivation regulate metabolism in S. cerevisiae (Soontorngun 2016). The fundamental question, however, is what would be the metabolic basis of pseudohyphal differentiation in response to decreasing concentration of glucose when ammonium concentration is low to begin with. While the biochemical interactions occurring during this process have been studied in great detail (Broach 2012; Cullen and Sprague 2012), the mechanism of how such interactions give rise to a specific phenotype depending upon the availability of glucose as well as ammonium is not clear. Here, to investigate the role of KRH1 and KRH2 in the context of depleting glucose as well as ammonium concentrations, we undertook a detailed study of different mutant strains under varying growth conditions.

Based on our previous data and the results presented in this study, we consider the following possibility (refer to Fig. 7) when a wild-type cell is allowed to grow in SLAD medium (i.e. when glucose is abundant to start with). GPR1 senses abundant glucose to activate GPA2 on one hand and inhibit KRH2 on the other. Simultaneous activation of GPA2 and inhibition of KRH2 triggers the cAMP pathway favouring the vegetative mode of growth. As the glucose concentration diminishes, glucose mediated inhibition of KRH2 is alleviated thereby allowing inhibition of GPA2 and PKA by KRH2. When ammonium concentration decreases, repression by MEP2 prevents KRH2 from inhibiting GPA2 and PKA. It appears that MEP2 mediated inhibition of KRH2 occurs only when glucose dependent inhibition of KRH2 is alleviated in response to decreasing glucose concentration. Under these conditions, KRH2 can no longer inhibit GPA2 and PKA, thereby triggering pseudohyphae formation. We had previously reported that gpr1∆ mutant exhibits profuse pseudohyphae in SLALD but not in SLAD medium (Iyer et al. 2008). Further, deletion of GPR1 suppressed the inability of the MEP2 mutant to put forth pseudohyphae in SLALD but not in SLAD medium. In the absence of MEP2 as well as GPR1, although KRH2 is free to suppress PKA signaling, it is possible that pseudohyphae are facilitated by alleviation of glucose mediated repression of FLO11 in SLALD medium. Thus, the above mechanism seems to operate by co-ordinating the alleviation of glucose repression of FLO11, with that of MEP2 dependent activation of PKA, through GPR1-GPA2 axis. Keeping FLO11 under tight glucose repression, probably allows the cells to grow rapidly even under low ammonium until glucose repression is alleviated. Once the glucose levels decrease pseudohyphal differentiation may be triggered through MEP2. Thus, we suggest that the cAMP-PKA pathway favours vegetative growth only as long as glucose is in abundance. However, as nutrients get depleted, signals from MEP2 as well as from low glucose impinge on the cAMP-PKA pathway so that it favours filamentation response.

Fig. 7
figure 7

Model depicting the shift in signaling as glucose and ammonium concentrations decrease Under condition of high glucose concentration, filamentation is suppressed by KRH1 mediated downregulation of GPR1-GPA2 signaling as well as by glucose mediated activation of KRH2 through GPR1. In response to depleting ammonium KRH1 is down-regulated thereby alleviating KRH1 mediated suppression of PKA. Simultaneously, suppression of KRH2 mediated by MEP2 occurs. As glucose is utilized by the cells, the concentration decreases and pseudohyphae formation is triggered by alleviation of glucose mediated repression of FLO11

Full size image

Our observations indicate that KRH2 is inhibited by MEP2 in addition to being modulated by glucose signaling. Overall, the results of this study suggest that KRH2 receives and integrates signals from both glucose as well as ammonium. On the other hand, our observation that KRH1 is inactivated by low ammonium suggests that KRH1 may be involved only in cAMP-PKA signaling in response to glucose and may not play a role in pseudohyphal differentiation. In contrast to the current understanding that KRH1 and KRH2 are functionally equivalent or that they are redundant (reviewed in Peeters et al. 2007), we propose that KRH1 and KRH2 are non-redundant with respect to their function in filamentation response. Our earlier results suggested a paradigm shift in that low glucose-low ammonium is a physiologically relevant signal for dimorphic transition in S. cerevisiae (Iyer et al. 2008). It is only in this changed paradigm, that we are able to dissect out the differential role of KRH1 and KRH2. Thus, genetic and biochemical evidence presented in this study suggests that cells constantly monitor the level of glucose to decide when they should quit from normal proliferation and switch over to pseudohyphal mode of growth. Our data imply that given an opportunity, the cell would prefer to grow vegetatively rather than put-forth pseudohyphae as long as sufficient glucose is available.