Cellular differentiation is controlled largely by specific gene expression programs. In the fission yeast Schizosaccharomyces pombe, the high mobility group (HMG) protein Ste11 is the main transcription factor responsible for the switch from cellular proliferation to sexual differentiation (Sugimoto et al., 1991; Mata and Bähler, 2006). Ste11 activates many genes that are required for the initiation of sexual differentiation, including the mating-type genes and mei2, the latter of which encodes a master regulator of meiosis. Expression of ste11 is under the control of several signaling pathways, which respond to external factors, such as nutrition, stresses and mating pheromones. They include the cyclic adenosine monophosphate (cAMP) pathway, the target of rapamycin (TOR) complex 1 (TORC1) pathway and the two mitogen-activated protein kinase (MAPK) pathways (Yamamoto et al., 1997; Otsubo and Yamamoto, 2010). Recently, it has become apparent that phosphorylation of the C-terminal domain (CTD) of RNA polymerase II (Pol II) by CTD kinase I (Ctdk1) specifically affects the expression of ste11 (Coudreuse et al., 2010; Sukegawa et al., 2011). It has also been shown that the meiotic regulator Mei2, the expression of which is governed by Ste11, enhances expression of ste11 through a positive-feedback loop (Sukegawa et al., 2011). These new discoveries, together with previous findings, illustrate that a number of signal transduction pathways orchestrate the regulation of ste11 expression in response to intricate environmental changes. In this Cell Science at a Glance article, we summarize the current knowledge of these signaling pathways, and in the accompanying poster we present a comprehensive scheme showing the regulation of ste11 expression.
Ste11 – a key transcription factor for sexual differentiation
Fission yeast is a unicellular microorganism, which proliferates typically in the haploid state, and the haploid cell assumes one of the two mating types, h+ (P) or h− (M). Under nutrient-rich conditions, cells proliferate through the mitotic cell cycle. When starved of nutrients, in particular nitrogen, the cells begin sexual differentiation; cells arrest in G1 phase, and conjugation proceeds between h+ and h− cells to give rise to h+/h− zygotes. These resulting diploid zygotes initiate meiosis and form asci, each of which contains four haploid spores (see Poster).
Many genes have to be regulated in an organized manner to enable sexual differentiation. Ste11 activates a large number of genes that are required for mating and meiosis. The target genes identified to date, such as mei2, and the mating-type genes mat1-P and mat1-M, have nearly identical ten-base motifs (5′-TTCTTTGTTY-3′, the so-called TR box) in their 5′-upstream region (Sugimoto et al., 1991; Mata and Bähler, 2006). There are five TR boxes in mei2, and two TR boxes in both mat1-P and mat1-M. The ste11 gene itself also has a TR box in its 5′-upstream region, and Ste11 appears to activate its own transcription through this TR box (Kunitomo et al., 2000). Disruption of ste11 causes complete sterility of the cell (Sugimoto et al., 1991), indicating that Ste11 is a key transcription factor for the sexual differentiation of fission yeast.
Ste11 function is suppressed in multiple ways during mitotic growth. Transcription of ste11 is reduced under nutrient-rich conditions (Sugimoto et al., 1991). Ste11 is active only in the G1 phase; in the remaining phases of the cell cycle, Ste11 is phosphorylated by Cdk, which prevents it from binding to DNA (Kjaerulff et al., 2007) (see Poster). In addition, Pat1 kinase (also known as Ran1) phosphorylates Ste11 in vegetative cells (Li and McLeod, 1996). The 14-3-3 protein Rad24 binds to the phosphorylated form of Ste11 and inhibits its accumulation in the nucleus (Kitamura et al., 2001). As Ste11 stimulates the expression of ste11 itself, this nuclear exclusion contributes to reducing further the level of ste11 expression (Qin et al., 2003) (see Poster). Ste11 is also subject to destruction through the ubiquitin–proteasome pathway during vegetative growth (Kitamura et al., 2001; Kjaerulff et al., 2007). Furthermore, negative control of Ste11 by the RNA-binding protein Nrd1 (also known as Msa2) at the translation level has been reported (Tsukahara et al., 1998; Oowatari et al., 2011).
Upon nitrogen starvation, expression of ste11 is increased dramatically (Sugimoto et al., 1991). This is brought about by the cooperation of four major signaling pathways, the cAMP pathway, the TORC1 pathway, the mating pheromone-responsive MAPK pathway and the stress-responsive MAPK pathway (see Poster). We will focus on these signaling pathways one by one in the sections below.
The cAMP pathway
Many cell systems use cAMP as an important second messenger in the regulation of adaptation to external conditions. In fission yeast, depletion of the carbon source results in a rapid reduction in the level of intracellular cAMP. Depletion of the nitrogen source also reduces the cAMP level, but in this case, it takes hours to reduce the cAMP level by half (Mochizuki and Yamamoto, 1992; Isshiki et al., 1992). The decrease in cAMP leads to inactivation of cAMP-dependent protein kinase (Pka) (Maeda et al., 1990), which induces gluconeogenesis and sexual differentiation in fission yeast (Hoffman and Winston, 1991; Yamamoto et al., 1997).
When there is a sufficient supply of nutrients, the α-subunit of a heterotrimeric G-protein (Gpa2), is activated by the seven-transmembrane protein Git3, a putative glucose receptor (Isshiki et al., 1992; Welton and Hoffman, 2000; Hoffman, 2005). Activated Gpa2 stimulates adenylate cyclase Cyr1 (also known as Git2), which catalyzes the production of cAMP from ATP (Kawamukai et al., 1991). At high concentrations, cAMP associates with Cgs1, the regulatory subunit of Pka (DeVoti et al., 1991). This allows the catalytic Pka subunit Pka1 to be released from the inhibitory Cgs1 subunit and to exert its kinase activity (Maeda et al., 1994). Pka1 represses the expression of ste11 through phosphorylation and inactivation of the zinc-finger transcription factor Rst2 (Kunitomo et al., 2000; Higuchi et al., 2002; Matsuo et al., 2008; Gupta et al., 2011) (see Poster).
Rst2 activates transcription of ste11 by binding to the STREP motif (5′-CCCCTC-3′) in its 5′-upstream region. Fission yeast cells lacking Rst2 grow mitotically, but cannot express ste11 in sufficient amounts to initiate sexual differentiation (Kunitomo et al., 2000). The sterility of rst2-defective cells can be counteracted with an artificial overexpression of ste11 (Kunitomo et al., 2000). Thus, Rst2 is a transcription factor that regulates the expression of ste11 specifically through mediating cAMP signaling.
In addition to regulating Ste11 at the transcriptional level, it has been observed that the Pka pathway also inhibits the nuclear accumulation of Ste11 (Valbuena and Moreno, 2010).
The TORC1 pathway
The TORC1 pathway, like the cAMP pathway, promotes vegetative growth and suppresses sexual differentiation when it is active (Otsubo and Yamamoto, 2010). It downregulates expression of ste11 in the presence of nutrients, in particular in the presence of nitrogen (see Poster). TORC1 contains Tor2, one of the two TOR kinases identified in fission yeast, which is essential for vegetative growth (Weisman and Choder, 2001). Analysis of temperature-sensitive tor2 mutants has shown that they arrest at G1 phase of the cell cycle in nutrient medium if they are shifted to the restrictive temperature (Alvarez and Moreno, 2006; Uritani et al., 2006; Weisman et al., 2007; Matsuo et al., 2007). Furthermore, ste11 is expressed at the restrictive temperature in these cells, irrespective of the nutritional conditions, and results in the initiation of ectopic sexual differentiation in these cells (Alvarez and Moreno, 2006; Matsuo et al., 2007; Valbuena and Moreno, 2010).
It has been shown that Ste11 target genes, including ste11 itself, are upregulated by loss of Tor2 and downregulated by overexpression of Tor2 (Matsuo et al., 2007; Valbuena and Moreno, 2010). Tor2 can physically interact with Ste11 (Alvarez and Moreno, 2006), and Ste11 accumulates in the nucleus if Tor2 is inactivated (Valbuena and Moreno, 2010). Thus, TORC1 might also inhibit sexual differentiation by controlling the localization of Ste11, such as Pka and Pat1 kinase, in addition to regulating the level of ste11 transcription.
The mating pheromone-responsive MAPK pathway
The sexual differentiation of fission yeast is also under the control of mating pheromones (Yamamoto et al., 1997), which are short peptides necessary to elicit conjugation.
In fission yeast, the mat1 locus on chromosome II determines the mating type of the cell. If this locus contains the P sequence (mat1-P), the cell is h+ (P); if the locus contains the M sequence (mat1-M), the cell is h− (M). Both mat1-P and mat1-M consist of two transcription units, mat1-Pc and mat1-Pi (also known as mat1-Pm), and mat1-Mc and mat1-Mi (also known as mat1-Mm) (Kelly et al., 1988). Transcription of mat1 is upregulated by Ste11 through binding to its 5′-upstream TR-boxes (Sugimoto et al., 1991).
Upon nitrogen starvation, Ste11 activates the transcription of mat1-Pc in h+ cells and of mat1-Mc in h− cells. The products of these genes then elicit the expression of the respective mating-type-specific genes. The P-specific genes include map2, which encodes the mating pheromone P-factor, and map3, which encodes the M-factor receptor. M-specific genes include mfm1, mfm2 and mfm3, which all encode the mating pheromone M-factor, and mam2, which encodes the P-factor receptor (Tanaka et al., 1993; Kitamura and Shimoda, 1991; Davey, 1992; Imai and Yamamoto, 1994; Kjaerulff et al., 1994). Subsequently, P-factor, a simple peptide of 23 amino acids is secreted from h+ cells, and M-factor, a peptide of nine amino acids that is farnesylated and carboxymethylated at the C-terminus, is secreted from h− cells (Imai and Yamamoto, 1994; Davey, 1992).
P-factor binds to the P-factor receptor Mam2 on the surface of h− cells, whereas M-factor binds to the M-factor receptor Map3 on the surface of h+ cells (Tanaka et al., 1993; Kitamura and Shimoda, 1991; Toda et al., 1991). Upon pheromone binding, the two receptors dock to the same Gα protein, Gpa1, and activate it. Activated Gpa1 then stimulates a MAPK cascade, comprising Byr2 (MAPKKK), Byr1 (MAPKK) and Spk1 (MAPK) (Nadin-Davis and Nasim, 1988; Wang et al., 1991b; Gotoh et al., 1993; Neiman et al., 1993). The Byr2–Byr1–Spk1 MAPK cascade enhances expression of ste11 and of additional genes required for mating and meiosis in response to the mating pheromones (Aono et al., 1994; Kjaerulff et al., 2005; Ozoe et al., 2002; Xue-Franzén et al., 2006). The MAPKKK Byr2 is also stimulated by Ras1, the fission yeast homologue of the oncoprotein RAS (Fukui et al., 1986; Nadin-Davis and Nasim, 1988; Nadin-Davis et al., 1986; Wang et al., 1991b). The activity of Ras1 is positively controlled by the GDP-GTP exchange factor (GEF) Ste6, and is negatively regulated by the GTPase-activating protein Gap1 (Hughes et al., 1990; Hughes et al., 1994; Imai et al., 1991; Wang et al., 1991a). Ste11 activates transcription of ste6 (Hughes et al., 1994). Thus, the pheromone signaling appears to be reinforced and firmly established by a positive-feedback mechanism that involves expression of Ste11, the pheromone-sensing machinery, and activation of Ras1 (see Poster). It has been suggested that Ste11 is a direct target of Spk1 MAPK (Kjaerulff et al., 2005).
The stress-responsive MAPK pathway
Fission yeast has another MAPK pathway that controls expression of ste11, the stress-responsive Sty1 (also known as Spc1) MAPK pathway (Yamamoto et al., 1997). Sty1 is the fission yeast homologue of mammalian MAPK p38. It responds to diverse stimuli, including heat shock, high osmolarity, oxidative stress and nutritional starvation (Shiozaki and Russell, 1995; Millar et al., 1995; Shiozaki and Russell, 1996; Degols et al., 1996). When cells are exposed to stress, the response regulator protein Mcs4 associates with the MAPKKK Wis4 (also known as Wak1) and probably also with the MAPKKK Win1, and upregulates MAPKKK activity. Activated Wis4 and Win1 phosphorylate the MAPKK Wis1, which in turn phosphorylates Sty1 (Shiozaki and Russell, 1995; Millar et al., 1995; Samejima et al., 1998; Shieh et al., 1998b). It has been indicated that Sty1 phosphorylation is also regulated by the Pyp1 tyrosine phosphatase and a MAPKKK-independent pathway that might involve the Cdc37 chaperone (Samejima et al., 1997; Shiozaki et al., 1998; Shieh et al., 1998a; Tatebe and Shiozaki, 2003). Once activated by this phosphorylation, Sty1 accumulates in the nucleus and phosphorylates several substrates, including a bZIP transcription factor Atf1 (also known as Gad7) (Shiozaki and Russell, 1996; Wilkinson et al., 1996; Gaits et al., 1998). Atf1 associates with another bZIP protein, Pcr1, and binds to DNA as a heterodimer (Kanoh et al., 1996) (see Poster). It has been shown that activated Sty1 is recruited to the promoters of stress-induced genes to which Atf1 binds (Reiter et al., 2008; Sansó et al., 2011). Wis1, Sty1 and Atf1 are essential for appropriate G1 arrest and the initiation of sexual differentiation under nitrogen starvation (Shiozaki and Russell, 1996; Kanoh et al., 1996). The expression of ste11 in response to nitrogen starvation is impaired in cells that are defective in wis1, sty1, atf1 or pcr1 (Shiozaki and Russell, 1996; Kanoh et al., 1996; Watanabe and Yamamoto, 1996), suggesting a role for the stress-responsive MAPK pathway in the induction of ste11 expression.
Considerable progress has been made recently in understanding how the Sty1 MAPK pathway regulates ste11 expression. It has become clear that the phosphorylation of Ser-2 residues on the heptad repeats of Pol II CTD by CTDK-I is essential for the upregulation of ste11 transcription (Coudreuse et al., 2010; Sukegawa et al., 2011). Interestingly, CTDK-I is dispensable for vegetative growth, and the main biological role of the Ser-2 phosphorylation appears to be to activate transcription of ste11. Furthermore, the stress-responsive MAPK cascade is crucial for the activation of CTDK-I under nitrogen starvation (Sukegawa et al., 2011). It has also been revealed that activated Mei2 protein, which switches the cell cycle from a mitotic to a meiotic one, can stimulate this pathway through the MAPKKKs Wis4 and Win1 and, hence, contribute to increased transcription of ste11. Although the way in which Mei2 activates MAPKKKs remains unclear, the feedback loop that involves the activation of Wis4 and Win1 by Mei2, which in turn activates the Wis1–Sty1–CTDK-I cascade leading to CTD phosphorylation and Ste11 expression, appears to be important in reinforcing the decision of the cell to commit to meiosis (Sukegawa et al., 2011) (see Poster).
As discussed above, the regulation of ste11 expression is fairly complex and includes the orchestration of a number of signaling pathways. Because expression of ste11 is a hallmark for commitment of a cell to sexual differentiation, it is crucial for a cell to check various internal and external conditions and assess their balance before making a final decision to express ste11, and to initiate sexual differentiation. Another remarkable feature of the regulation of ste11 transcription is that it includes several feedback loops to further enhance its transcription once it is initiated. This feedback mechanism is physiologically important, as it would be fatal for a cell to abandon the developmental process after it has been initiated.
Although the accompanying poster outlines the current knowledge of regulatory systems that are relevant for ste11 expression, we acknowledge that many unresolved questions remain. The molecular mechanisms are not yet sufficiently clear for any of the signaling pathways described here. In addition, the exact way in which the signaling pathways cooperate to monitor the nutritional conditions remains elusive. Although the cAMP–PKA pathway rapidly responds to the availability of the carbon source, it also de-escalates gradually after depletion of the nitrogen source (Isshiki et al., 1992; Mochizuki and Yamamoto, 1992) and this decrease appears to be important for the initiation of sexual differentiation. However, how this gradual decrease, which is mediated by nitrogen depletion, affects the initiation of sexual differentiation is currently unknown.
From the observations made to date, it appears that the TORC1 pathway more or less directly detects and responds to the lack of a nitrogen source. As described above, this signal is also conveyed through the stress-responsive MAPK pathway, presumably as a ‘stress of starving’. However, the Sty1 MAPK pathway does not appear to upregulate expression of ste11 when nutrients other than nitrogen are limited (Shiozaki and Russell, 1996). Interestingly, specific transcription of a Sty1-target gene cta3 under elevated salt concentrations is compromised in the tup11-tup12 double-deletion mutant. In the absence of the transcriptional co-repressors Tup11 and Tup12, other stresses induce the expression of cta3. Thus, it is thought that Tup repressors function as specificity factors by preventing the induction of cat3 transcription in response to inappropriate stresses (Greenall et al., 2002; Hirota et al., 2003). A similar mechanism might operate in the control of ste11 expression in response to nitrogen starvation. Furthermore, it is noteworthy that the Sty1 pathway regulates transcription of some genes that are relevant for the cAMP–PKA pathway (Stiefel et al., 2004; Davidson et al., 2004), implicating that there is a complex crosstalk between these pathways.
Finally, we would like to mention that yet another signaling pathway that controls expression of ste11 might exist. TOR complex 2 (TORC2), which contains Tor1, the second TOR kinase in fission yeast, is also required for sexual differentiation (Kawai et al., 2001; Weisman and Choder, 2001). In contrast to the tor2 temperature-sensitive mutant, tor1 deletion mutants grow vegetatively, but cannot arrest at G1 phase and enter sexual differentiation under nitrogen starvation (Kawai et al., 2001; Weisman and Choder, 2001). Although little is known with regard to the activation of the TORC2 pathway, it can be presumed that it positively regulates expression of ste11.
Further analysis of the regulation of ste11 expression will certainly lead us to a deeper understanding of the mechanisms underlying the switching from proliferation to differentiation.
We thank Akira Yamashita for critical reading of the manuscript.
The work of our laboratory was supported by a Grant-in-Aid for Scientific Research [grant number (S) 21227007 to M.Y.] from the Japan Society for the Promotion of Science and also in part by the Ministry of Education, Culture, Sports, Science and Technology Global Centers of Excellence Program (Integrative Life Science Based on the Study of Biosignaling Mechanisms) [grant number A03; Program leader, Yasushi Miyashita, University of Tokyo].