Article

Nature Cell Biology 9, 1263 - 1272 (2007)
Published online: 21 October 2007 | doi:10.1038/ncb1646

TOR signalling regulates mitotic commitment through the stress MAP kinase pathway and the Polo and Cdc2 kinases

Janni Petersen1,2 & Paul Nurse1

The coupling of growth to cell cycle progression allows eukaryotic cells to divide at particular sizes depending on nutrient availability. In fission yeast, this coupling involves the Spc1/Sty1 mitogen-activated protein kinase (MAPK) pathway working through Polo kinase recruitment to the spindle pole bodies (SPBs). Here we report that changes in nutrients influence TOR signalling, which modulates Spc1/Sty1 activity. Rapamycin-induced inhibition of TOR signalling advanced mitotic onset, mimicking the reduction in cell size at division seen after shifts to poor nitrogen sources. Gcn2, an effector of TOR signalling and modulator of translation, regulates the Pyp2 phosphatase that in turn modulates Spc1/Sty1 activity. Rapamycin- or nutrient-induced stimulation of Spc1/Sty1 activity promotes Polo kinase SPB recruitment and Cdc2 activation to advance mitotic onset. This advanced mitotic onset is abolished in cells depleted of Gcn2, Pyp2, or Spc1/Sty1 or on blockage of Spc1/Sty1-dependent Polo SPB recruitment. Therefore, TOR signalling modulates mitotic onset through the stress MAPK pathway via the Pyp2 phosphatase.

Many eukaryotic cells need to reach a critical size before they undergo mitosis and division1, 2, 3, 4, and this process is often modulated by changes in the availability of nutrients1, 3, 5. The size control can be easily investigated in the fission yeast Schizosaccharomyces pombe by measuring cell length at septation, because the rod-shaped cell grows by elongation and undergoes mitosis and division at a defined length in steady-state growth conditions1. When wild-type cells are shifted from a good to a poor nitrogen source, they advance into mitosis5, and proliferation continues at a reduced cell size.

TOR signalling is thought to be involved in nutrient sensing6. There are two TOR kinases, Tor1 and Tor2, in fission yeast, of which only Tor2 is essential for growth7. Both kinases have been implicated in nutrient sensing after withdrawal of nitrogen, which induces cell cycle exit and sexual differentiation. Loss of Tor2 activity has a similar effect to that of nitrogen withdrawal and promotes sexual differentiation, whereas cells lacking Tor1 are incapable of G1 arrest and are therefore sterile7, 8, 9, 10, 11, 12. Inhibition of TOR signalling by rapamycin has no effect on cell growth in fission yeast13, and it has been suggested that rapamycin-insensitive Tor2 can compensate for Tor1. Cell size at division is regulated by the S. pombe MAPK stress response pathway (SRP), and in unperturbed cell cycles this pathway promotes commitment to mitosis to allow cells to match their rate of division to nutrient availability14. Loss of SRP signalling delays mitosis, whereas boosting SRP signalling promotes mitosis14, 15, 16. A significant role of the SRP in control of mitotic onset is carried out through the serine 402 (S402)-dependent recruitment of the Polo kinase Plo1 to the non-separated SPBs at the G2/M transition17, 18, 19. Plo1 recruitment to the SPB modulates the timing with which the Cdc2 kinase20 regulator of the G2/M transition is activated18, 19. In this paper we investigate how TOR signalling and the MAPK stress response pathway regulate cell size at mitosis following alterations in nutrient availability.

Top

Results

Poor nutrients or addition of rapamycin promote mitotic onset

To examine the effect of a nutrient shift on cell cycle progression, a culture of wild-type synchronous cells was shifted from a good nitrogen source (glutamate) to a poor nitrogen source (proline) (Fig. 1a). As previously reported5, cells shifted into proline are advanced into mitosis. This can also be observed in asynchronous cultures5, both as a peak in the proportion of dividing cells 40–60 min after nutritional shift (Fig. 1b, a, blue solid squares) and by division at a reduced cell length (Fig. 1b, a, red solid circles, and Fig. 1c). By contrast, a control glutamate-to-glutamate shift shows no advancement of mitosis as shown by the constant proportion of dividing cells in the population and cell length at division (Fig. 1a–C). The length at division of wild-type cells was reduced from 13.8 plusminus 1 mum to 9.5 plusminus 1.1 mum when shifted from glutamate into proline after 120 min (Fig. 1b, a and Table 1), although the steady-state length at division in proline readjusts to 11.3 plusminus 0.9 mum during subsequent divisions5.

Figure 1: Addition of rapamycin to reduce TOR signalling advances mitotic onset, leading to reduced cell size at division.

Figure 1 : Addition of rapamycin to reduce TOR signalling advances mitotic onset, leading to reduced cell size at division.

(A) a, Glutamate-grown cells were synchronized by elutriation and allowed to go through one round of division before rapamycin was added or before filtration and shifting into proline or into glutamate to which solvent was added. b, Calcofluor staining showing cells from the 260- and 300-min time points. (B) The left-hand y axes show cell length at division in mum (red); right-hand y axes show frequency of dividing cells (blue). a, Wild-type cells grown in glutamate were collected by filtration and resuspended into glutamate or proline (filled symbols) at time 0. b, Wild-type cells grown in glutamate, and rapamycin (open symbols) or solvent (filled symbols) was added at time 0. c, Wild-type cells grown in glutamate were collected by filtration and resuspended into proline, and either rapamycin or solvent was added at time 0 (closed). d, Wild-type cells grown in proline were collected by filtration and resuspended into glutamate, and rapamycin or solvent (filled symbols) added at time 0. (C) Wild-type cells stained with calcofluor showing cell length at division before and after shift of medium or addition of rapamycin.

Full size image (79 KB)


To examine the role of TOR signalling in nutrient-induced mitotic onset, rapamycin was added to cells grown with glutamate as a nitrogen source. The cells were advanced into mitosis and underwent cell division with similar kinetics to that of cells shifted from a glutamate to a proline medium (Fig. 1a–c). Rapamycin had no effect on cell growth13, shown by an unchanged rate of biomass increase (attenuance (D595) measurements) and cell-tip extension after drug addition (see Supplementary Information, Fig. S1). In addition, adding rapamycin to cells shifted from glutamate to proline had no further additive effects on mitosis and cell division advancement (Fig. 1b, c). To examine further the effect of inhibiting TOR signalling on mitotic onset, a reciprocal shift from proline to glutamate was carried out in the presence or absence of rapamycin. In the absence of rapamycin, onset of mitosis and cell division was delayed, reducing the numbers of dividing cells and increasing cell length at division (Fig. 1b, d). By contrast, when rapamycin was present, there was only a modest delay in mitosis and a minor increase in cell length at division (Fig. 1b, d). The addition of rapamycin to cells grown with proline as the nitrogen source reduced their length at division from 11.3 plusminus 0.9 mum back to 9.2 plusminus 0.4 mum (see Supplementary Information, Fig. S1D). These data suggest that TOR signalling through either Tor1 or Tor2 regulates mitotic entry.

Tor2 is inhibited by rapamycin

We noticed that glutamate-grown wild-type homothallic (h90) strains promoted sexual differentiation in late exponential cultures. Cultures of h90 strains contain a mix of mating partners and so are capable of generating pheromone signalling. It has previously been shown that pheromone signalling accelerates mitosis and G1 arrest21, suggesting that the basal level of pheromone signalling is relatively high and that Tor2 activity is easily repressed in glutamate media. To test this, we added rapamycin to early exponential h90 cultures; 10 h after treatment, 31% of the cells had undergone sexual differentiation, and sporulating zygotes were observed (Fig. 2a). By contrast, only 3% of the cells had formed zygotes in the control culture. This indicates that Tor2 can be inhibited by rapamycin in glutamate media, and that rapamycin effects mimic those of nitrogen starvation sufficiently to induce pheromone signalling in h90 strains and, as a consequence, accelerate mitosis and G1 arrest to allow mating21.

Figure 2: Inhibition of TOR signalling after nutrient shifts or addition of rapamycin.

Figure 2 : Inhibition of TOR signalling after nutrient shifts or addition of rapamycin.

(a) Rapamycin or solvent was added to early exponential glutamate-grown h90 cells for 10 h. (b) Left-hand y axes show cell length at division (mum; red); right-hand y axes show frequency of dividing cells (blue). tor1Delta cells (upper) and tor1Delta cells with exogenous pRep81-Tor repressed (lower) were grown in glutamate. Either rapamycin was added or cells were filtered into proline at time 0. The tor1Delta strain with full exogenous tor1+ expression responded to both a shift of medium and rapamycin treatment with kinetics similar to that of the wild-type cells (see Supplementary Information, Fig. S3A). (c) Tor1-dependent kinase activity (see Supplementary Information, Fig. S3C) is reduced after a glutamate-to-proline shift. Cells were grown in glutamate and shifted into proline at time 0. Haemagglutinin (HA)–Tor1 was immunoprecipitated from cells at the indicated time points and used in the kinase assay. Tor1 is a homologue of phosphatidylinositol-3 kinases, and is inhibited by wortmannin48. The Tor1-dependent kinase activity was inhibited when wortmannin was added to lane 1. Uncropped images of blots are shown in Supplementary Information, Fig. S6A.

Full size image (61 KB)

Reduced Tor1 signalling promotes mitotic onset

Because rapamycin was previously shown to inhibit Tor1-mediated activity22, we assessed the effect of direct manipulation of Tor1 signalling. Tor1 is not essential for growth; however, glutamate-grown tor1Delta cells are elongated12, 23 with a generation time around 50% greater than that of isogenic wild-type cells (see Supplementary Information, Fig. S2A). This suggests that a basal level of Tor1 in the cells is needed to maintain a normal growth rate, presumably because of roles of the TOR signalling pathway that have previously been described6. The deficiency in TOR signalling in tor1Delta (ref. 7) can be rescued by exogenous tor1+ expression from the weak nmt81 promoter, leaving generation time unaffected (see Supplementary Information, Fig. S2A). The consequence of reducing tor1 levels in this strain was not an increase in the steady-state length of tor1Delta, but instead it mimicked the effect of a change in medium or addition of rapamycin because a reduction in cell length at mitotic onset was observed to a level below that at which wild-type cells divided (14.7 plusminus 1.1 mum to 11.9 plusminus 0.8 mum; see Supplementary Information, Fig. S2B; see Table 1 for wild type). To further analyse the relative role of Tor1 and Tor2 in the control of mitotic entry, the tor1Delta strain was shifted from glutamate to proline, or grown in glutamate to which rapamycin was added (Fig. 2b, upper). Although the cells were unable to accelerate mitosis after the shift in medium to proline, addition of rapamycin reduced cell length at division, which is indicative of Tor2 inhibition. Furthermore, a tor1Delta strain with exogenous rep81-tor1+ repressed, shifted from glutamate to proline or grown in glutamate added rapamycin (Fig. 2b, lower), increased cell length at division, suggesting that both treatments further reduced Tor1 activity. However, in the rapamycin-treated cells, a reduction in cell length at division back to the initial cell length was observed, which is indicative of Tor2 inhibition. As mentioned above, tor1Delta cells are unable to arrest in G1 after nitrogen withdrawal7; however, when tor1Delta cells were shifted into nitrogen-free medium to promote sexual differentiation, the cells accelerated mitotic commitment (Table 2). Thus the failure of tor1Delta cells to accelerate mitosis in a glutamate-to-proline shift is not because tor1Delta cells generally are unable to advance mitosis. After a medium shift from glutamate to proline, we observed a decline in Tor1-dependent kinase activity (Fig. 2c); the Tor1 protein level remain unchanged during the medium shift or after the addition of rapamycin (see Supplementary Information, Fig. S3B). Together, these data suggest that a reduction in Tor1 signalling rather than a complete block of its activity promotes mitotic entry in response to changes in nutrients that allow continued proliferation at a reduced cell size.


Reduced TOR signalling activates Cdc2

To understand how TOR signalling affects mitotic commitment, we examined the regulator of G2/M transition, the Cdc2 kinase20. In wild-type cells shifted from glutamate to proline, or in cells to which rapamycin has been added, we observed a loss of the inhibitory phosphorylation on tyrosine 15 (ref. 24), which is a modification present in G2 to inhibit the kinase (Fig. 3a). The timing at which this loss takes place, which leads to the activation of Cdc2, is regulated by the activating phosphatase Cdc25 and the inhibitory kinase Wee1 (ref. 25). After adding rapamycin or shifting cells from glutamate to proline, the mutants cdc2.1w, cdc2.3w (which are largely insensitive to Wee1 and Cdc25, respectively)26, wee1.50 (ref. 27) and cdc25.22 cells showed no increase in the proportion of dividing cells and no reduction in cell length at division (Table 1; see Supplementary Information, Figs S1C and S4B). These results indicate that both Cdc25 and Wee1 are required for the advancement of mitotic onset induced by rapamycin or medium shifts from glutamate to proline.

Figure 3: Changes in nutrients or addition of rapamycin promote mitotic commitment through stress-pathway-dependent phosphorylation of Plo1 serine 402 and Cdc2 activation.

Figure 3 : Changes in nutrients or addition of rapamycin promote mitotic commitment through stress-pathway-dependent phosphorylation of Plo1 serine 402 and Cdc2 activation.

(a) Western blots of Suc1 precipitated Cdc2 with anti-Y15P and anti-cdc2, shifted from glutamate to proline or after rapamycin addition. (b) Anti-Plo1 indirect immunofluorescence staining of cells 35 min after filtering into glutamate into proline or to which rapamycin was added. The total number of Plo1-positive staining cells and the ratio of cells with Plo1 on non-separated SPBs to separated SPBs are shown. (c) Western blots of total extracts from wild-type cells shifted from glutamate to proline or to which rapamycin was added using anti-S402P followed by anti-Plo1. In a total extract, anti-S402P antibodies recognize only phosphorylated Plo1 (ref. 18). (d, e) Wild-type, plo1.s402A, cdc2.59 and cdc2.59 plo1.s402A grown in glutamate at 34 °C shifted to 25 °C simultaneously with filtering and resuspension into the indicated medium containing glutamate or proline at time 0. The cold-sensitive mutant cdc2.59 was used because plo1.s402A cells stop growing when shifted to 37 °C (ref. 18). A control shift of wild-type cells from 34 °C to 25 °C into glutamate showed an increase in dividing cells owing to the temperature-induced activation of Spc1/Sty1. The combined medium shift and temperature change increased advancement in mitosis, and it was Plo1 S402 dependent. (e) The indicated time points show cells stained with calcofluor, showing the cell length and the presence of a division plane. Uncropped images of blots are shown in Supplementary Information, Fig. S6B, C.

Full size image (60 KB)

Plo1 recruitment to SPBs modulates Cdc25/Wee1-controlled Cdc2 activity17, 18, 19. After 35 min of either rapamycin treatment or a media shift of an asynchronous culture, the ratio of cells with Plo1 on one non-separated SPB dot (arrows) compared with two separated mitotic SPB dots had increased (Fig. 3b). A minor fraction of mitotic Plo1 is phosphorylated on S402 in wild-type synchronous cultures to increase the affinity of Plo1 for SPBs18. Shifting wild-type cells from glutamate to proline or the addition of rapamycin resulted in a small increase in Plo1 phosphorylation on S402 (Fig. 3c), consistent with the observed approx20% increase in the fraction of mitotic cells in the asynchronous population. To examine the role of Plo1 kinase further, we analysed strains in which S402 of Plo1 was changed to alanine (S402A) to reduce SPB recruitment, or changed to glutamic acid (S402E) to obviate the need for SRP signalling and enhance Plo1 recruitment to the SPBs18. When grown in glutamate medium, plo1.S402E mutant cells were advanced into mitosis and division compared with wild-type cells, whereas the plo1.S402A mutation delayed mitotic entry, increasing cell size at division (Table 1). The advancement into mitosis in wild-type cells on the shift from glutamate to proline or on the addition of rapamycin was largely absent in both Plo1 mutants (Table 1; see Supplementary Information, Fig. S4A), indicating that dynamic control of S402 phosphorylation status was important for the response. The elongated, glutamate-grown plo1.S402A cells did show a small reduction in cell length at division after both treatments, but 2 h after the treatments the cell length at division was similar to that of glutamate-grown wild-type cells before the shift (Table 1). Another indication that the type of nitrogen source modulates mitotic entry was the suppression of the cdc2.59 mutant strain28 by a medium shift from glutamate to proline at its restrictive temperature (Fig. 3d). Significantly, this rescue was due to Plo1 S402-dependent control of mitotic entry18, because, unlike cdc2.59, the cdc2.59 plo1.S402A double mutant failed to divide after the shift into proline (Fig. 3d, e). Together these data indicate that Plo1.S402-dependent control of mitotic entry18 modulates cell size at mitosis and division in response to nutrient availability.

Spc1/Sty1 is essential for nutrient- and rapmycin-induced mitotic onset

Because SRP signalling is linked to environmental control of cell size at division, we asked whether nutrient-modulated TOR signalling altered mitotic size control by regulating SRP signalling. Cells that lack the SRP MAP kinase Spc1/Sty1 or its activating MAPK kinase Wis1 (ref. 29) were shifted from glutamate to proline or grown in glutamate to which rapamycin was added. Both sty1Delta and wis1Delta strains failed to advance into mitosis or reduce their size at division in response to nutrient shift or the addition of rapamycin (Table 1; see Supplementary Information, Fig. S4A). This is not a general failure of SRP mutants to advance mitosis, because sty1Delta cells still advanced mitotic commitment after the complete withdrawal of nitrogen (Table 2). Phosphorylation of the activating residues of Spc1/Sty1 that promote its kinase activity increased after nutrient shift or addition of rapamycin (Fig. 4a). On stress, Spc1/Sty1 activates the Atf1 transcription factor29. On shift from glutamate to proline or after the addition of rapamycin, atf1Delta cells still showed advanced mitotic commitment and divided at a reduced cell length (Table 1, see Supplementary Information, Fig. S4A). We therefore conclude that SRP signalling itself, but not the transcription-factor target Atf1, is essential for nutrient-induced advancement into mitosis and division.

Figure 4: Changes in Pyp2 activity regulate Spc1/Sty1 activity and cell size at commitment to mitosis.

Figure 4 : Changes in Pyp2 activity regulate Spc1/Sty1 activity and cell size at commitment to mitosis.

(A) Sty1 activation during nutritional shift and addition of rapamycin. a, Immunoprecipitates of Myc-tagged Sty1 collected by filtration from wild-type cells or pyp2::kanMX6 cells filtered into proline, or from wild-type cells filtered into glutamate or to which rapamycin was added. Blots were probed with PTPY antibodies against phospho-p38 MAPK, an activating phosphorylation. b, Sty1- dependent kinase assay. Cells from Spc1/Sty1–Myc or control wild-type cells were collected by filtration from glutamate-grown cultures. Immunoprecipitates using anti-Myc antibodies were used for the kinase assays. Immunoprecipitates of Spc1/Sty1–Myc collected by filtration from glutamate-grown cells filtered into glutamate or proline, or to which rapamycin was added or Pyp2 expression were repressed. SB203580, a Spc1/Sty1-specific inhibitor, was added to glutamate-grown cells (lane 1). (B) a, Cells were grown in glutamate. Left-hand y axes show changes in cell length at division (red); right-hand y axes show frequency of dividing cells (blue). Thiamine was added to repress nmt81pyp2–myc (OFF) or not (ON). b, Western blots of Pyp2–Myc and tubulin as controls from 0 and 30 min after addition of thiamine. (C) Western blots of Pyp2–Myc and tubulin as controls from wild-type cells grown in glutamate to which cycloheximide was added at time 0 or cycloheximide was added at time 0 after filtering cells into proline or adding rapamycin. (D) Western blots of Pyp2–Myc and tubulin as controls from glutamate-grown wild-type and mts3.1 cells. In B–D slower-migrating Pyp2 bands are hyperphosphorylated (data not shown). Uncropped images of blots are shown in Supplementary Information, Fig. S6D, E.

Full size image (52 KB)

The Pyp2 phosphatase regulates mitotic onset

Spc1/Sty1 is negatively regulated by two phosphatases Pyp1 or Pyp2 (ref. 29), and so the increase in Spc1/Sty1 activity on nutritional shift or rapamycin addition (Fig. 4ab) could arise from a reduction in activity of these two phosphatases. A pyp1Delta strain was advanced into mitosis after changes in nutrients or addition of rapamycin (Table 1; see Supplementary Information, Fig. S4A). By contrast, pyp2Delta cells were unable to advance mitosis (Table 1; see Supplementary Information, Fig. S4A) on nutritional shift or rapamycin addition. The level of Pyp1 in proliferating cells is severalfold higher than the level of Pyp2 (ref. 30). Interestingly, glutamate-grown cells that lack pyp2Delta are advanced into mitosis (Table 1) even though they still have functional Pyp1. This suggests that cells are sensitive to either Pyp2 protein levels themselves or to the formation of specific Pyp2–MAP kinase complexes. When a pyp2Delta strain was shifted from glutamate to proline, there was no increase in the amount of phosphorylation of the activating residues of Spc1/Sty1, indicating a lack of activation (Fig. 4a). By contrast, when exogenous Pyp2 expression was reduced, Spc1/Sty1 activity increased (Fig. 4ab). We therefore propose that the increase in Spc1/Sty1 activity induced by nutritional shift or the addition of rapamycin is mediated through the regulation of Pyp2 activity.

To test whether regulation of Pyp2 activity modulates mitotic onset, we reduced Pyp2 levels by placing pyp2+ under the control of the weakest repressible nmt81 promoter integrated in the leu1 locus of a pyp2Delta strain. After addition of thiamine to repress the promoter (OFF) (thiamine reduces transcription by around 50% within 1 h)31, the level of exogenously expressed Pyp2 decreased and cells advanced into mitosis and division at a reduced cell size (Fig. 4b). The timing with which cells reduced their size after thiamine addition suggested that regulation of Pyp2 activity in the cells is rapid and thus that Pyp2 is unstable over the time course of the medium shift. To test this, cycloheximide was added to glutamate-grown asynchronous cultures: Pyp2 levels decreased by 42% within 60 min(Fig. 4c). Furthermore, Pyp2 levels were considerably higher than in wild type when the activity of the proteasome was compromised with the mts3.1 mutant32 (Fig. 4d). No further decrease in Pyp2 levels was observed when cycloheximide was added simultaneously with either a medium shift into proline or with the addition of rapamycin (Fig. 4c), indicating that a change in this proteasomal control of Pyp2 levels is unlikely to be responsible for nutrient- or rapamycin-induced acceleration of mitosis. We conclude that Pyp2 is regulated through additional mechanisms.

A shift from glutamate to proline accelerated mitosis, and a shift of proline-grown cells to glutamate blocked mitosis for 60 min (Fig. 1b). We observed decreased Pyp2 levels when cells were shifted from glutamate to proline or rapamycin was added. By contrast, the reciprocal shift from proline to glutamate led to an increase in the Pyp2 level (Fig. 5a). This link between a change in Pyp2 levels and the timing of mitotic onset is consistent with the acceleration of mitosis and activation of Spc1/Sty1 observed after a reduction in exogenous expression of pyp2+ (Fig. 4a, b).

Figure 5: Pyp2 protein is regulated by nutrient shifts and altered TOR signalling.

Figure 5 : Pyp2 protein is regulated by nutrient shifts and altered TOR signalling.

(a) Western blots of Pyp2–Myc and tubulin as control from wild-type cells grown in glutamate filtered into proline or after rapamycin was added, harvested after 30 min, and a western blot of proline-grown cells filtered into glutamate, harvested after 60 min. (b) Western blots of Pyp2–Myc and tubulin as controls from glutamate-grown wild-type and gcn2::ura4+ cells. (c) Cells were grown in glutamate. Left-hand y axes show changes in cell length at division (red); right-hand y axes show frequency of dividing cells (blue). Rapamycin was added to strains expressing pyp2–myc with or without native UTRs. (d) Western blots of Pyp2–Myc and tubulin as control from steady state growing cultures and from 0 and 30 min after addition of rapamycin. (e) Western blots of Eif2alpha and Eif2alpha phosphorylated at serine 52. Lane 1 is a total extract from the eif2alpha.S52A strain. Lanes 2–5 show total extracts from wild-type strains. H2O2 (10 mm) was added to the cells shown in lane 3. Cells were filtered into proline or added rapamycin for 20 min (lanes 4 and 5). As a control, addition of H2O2 to YES-grown cells or exposure to UV (254 nm) (Supplementary Information, Fig. S5C) did induce Eif2alpha phosphorylation. (f) Left-hand y axes show changes in cell length at division (red); right-hand y axes show frequency of dividing cells (blue). eif2alpha.S52A and wild-type strains were grown in glutamate to which rapamycin was added. Uncropped images of blots are shown in Supplementary Information, Fig. S6E, F.

Full size image (54 KB)

Gcn2 kinase controls Pyp2 levels

To further understand how a nutrient shift or addition of rapamycin regulated Pyp2 levels, we looked at the S. pombe homologue of a known effector of TOR signalling in budding yeast, the GCN2 kinase. GCN2 kinase is a negative regulator of translation, and TOR signalling reduces GCN2 activity33. We found that S. pombe Gcn2 has a role in mitotic onset shown by gcn2Delta cells dividing at an increased length (Table 1), and therefore we asked whether this delay in mitotic commitment was because Gcn2 controls Pyp2 levels. Pyp2 was found to be elevated in a gcn2Delta strain (Fig. 5b). Furthermore, gcn2Delta pyp2Delta double mutants divided at a reduced size similar to that of the single pyp2Delta mutant (Table 1). This indicates that the delay in mitotic onset in gcn2Delta cells arises mainly from an increase in Pyp2 levels that then reduced Spc1/Sty1 activity. Consistent with the notion of Pyp2 levels being under Gcn2 control, Gcn2 was required for the nutrient- or rapamycin-induced acceleration of mitosis. The advancement into mitosis and 30% reduction of cell length at division of wild-type cells after both treatments were not observed in gcn2Delta cells (Table 1; see Supplementary Information, Fig. S4A). We next tested whether pyp2+ messenger RNA was subject to post-transcriptional regulation. To do this we used two pyp2Delta strains in which pyp2+ was under control of the nmt81 promoter. In one of these the pyp2 5' and 3' untranslated regions (UTRs) were intact; in the other they were absent. The cell length at division and Pyp2 protein levels at steady-state growth were similar in both strains (Fig. 5c, d). Addition of rapamycin to these strains (Fig. 5c) demonstrated that cells could fully respond and instigate a drug-induced advancement into mitosis only when pyp2 RNA was transcribed with the correct pyp2 5' and 3' UTRs. Both the 5' and 3' UTRs were required for full drug-induced advancement into mitosis and division (see Supplementary Information, Fig. S5B). Together, these data support the view that Pyp2 protein levels are regulated through regulatory elements within the pyp2 UTRs. An extra copy of pyp2 under control of its own UTRs integrated into the genome increased cell length at division (see Supplementary Information, Fig. S5A).

Acceleration of mitotic onset by rapamycin is Eif2alpha.S51 independent

In Saccharomyces cerevisiae, rapamycin promotes Gcn2-dependent Eif2alpha phosphorylation on serine 51, which blocks translation and arrests the cell cycle34, 35. In fission yeast, besides Gcn2, two additional kinases Hri1 and Hri2 phosphorylate Eif2alpha at serine 51 (serine 52 in S. pombe). However, Gcn2 specifically responded to changes in nutrients36, and hri1Delta or hri2Delta strains still accelerated mitosis in response to rapamycin treatment (data not shown). When rapamycin was added to glutamate-grown cells there was no marked increase in Eif2alpha.S52 phosphorylation above the high basal level (Fig. 5e), which may explain why rapamycin does not arrest the cell cycle in fission yeast. We therefore tested whether an eif2alpha.S52A strain37 could accelerate mitosis when treated with rapamycin. Compared with wild-type control cells, the mutant response was only slightly reduced (Fig. 5f), suggesting that Gcn2 also regulates translation in an Eif2alpha.S52-independent way as suggested by Tvegard and collegues37 and that Gcn2 substrate recognition through complex formation and/or subcellular localization is regulated in response to different modes of activation.

Top

Discussion

From the data presented here, we conclude that nutrient-modulated control of mitotic entry is mediated through TOR signalling and the stress response MAP kinase pathway, leading to changes in the activity of the mitotic kinase Cdc2. We find that both Tor1 and Tor2 are inhibited by rapamycin, and therefore it is likely that separate distinct complexes similar to those described in S. cerevisiae38 are also present in fission yeast. Relevant to this is the fact that co-immunoprecipitation experiments have shown that both Tor kinases associate with Pop3/Wat1 and Mip1, proteins of the rapamycin-sensitive TORC1 complex, with the Mip1 association being much stronger for Tor2 (Refs 8, 10). Our data indicate that nutrient-induced activation of mitotic onset is brought about by changes in Tor1 signalling through Gcn2 to modulate Pyp2 levels, which then leads to an increase in Spc1/Sty1 activity, resulting in subsequent Plo1 serine 402 phosphorylation and Cdc2 activation (Fig. 6). It is also possible that there is additional nutrient-induced Pyp2 regulation. This pathway is only essential for modulating the cell-size control at mitotic onset under conditions that allow continuous proliferation, because mutants of all the components involved were still able to accelerate mitosis during a complete withdrawal of nitrogen, which promotes cell cycle exit and sexual differentiation (Table 2). In proliferating mammalian cells, mTOR-dependent signalling has also been shown to be linked to cell-size control, because inhibition of mTOR by rapamycin reduces cell size39, and so the mechanisms we describe here may also be conserved in mammalian cells.

Figure 6: Proposed signalling pathways.

Figure 6 : Proposed signalling pathways.

A diagram showing the proposed signalling pathways that control nutritional modulation of the cell-size control at mitotic onset and cell division. Arrows and lines indicate positive and negative signals, respectively, rather than direct interactions. X represents potential further nutrient-induced Pyp2 regulation; Y represents unknown molecules inhibited by Tor2 to block sexual differentiation.

Full size image (39 KB)

Top

Methods

Cell cultures and strains.

Strains are listed in Supplementary Information, Table 1. Cells were cultured at 28 °C in yeast extract medium (YES)40 or in EMM minimal medium with either glutamate or proline as a nitrogen source5 or in MSL and MSL-N41 (minimal sporulating medium (MSL) without nitrogen (MSL-N)) supplemented if necessary with 150 mg l-1 amino acids. Thiamine (4 muM) was added to minimal medium to repress the nmt1 promoters. Early exponential cultures of 1.2 times 106 cells ml-1 for wild-type cells (attenuance at 595 nm (D595) of 0.1) were used in all nutritional-shift or rapamycin experiments. To ensure that the same biomass was used in all shifts, all cultures were shifted once they had reached a D595 of 0.1. Cells were filtered and washed three times. Rapamycin (300 ng ml-1) was added from a 2 mg ml-1 stock solution dissolved in 1:1 methanol plus DMSO; as a control, an equal amount of 1:1 methanol:DMSO was added. Cycloheximide (100 mug ml-1) was added from a 1 mg ml-1 stock. cdc2.59 mutants and control strains in this experiment were grown at 34 °C and were simultaneously filtered into new medium and shifted to 25 °C.

Molecular genetics.

Deletion of wis1, pyp1 and pyp2 open reading frames (ORFs) was done by PCR-based gene targeting using the kanMX6 construct42 and primers specific for wis1, pyp1 or pyp2 to remove the ORF (Sanger Centre, Cambridge, UK; S. pombe gene genome database (GB)). The deletions were confirmed by PCR. The Pyp2 ORF (ttttgtgctcatatgctccatcttctgtc + cgcggatccaacataaacaagattcattaataac) and UTRs from an S. pombe cDNA library43 (gccgctcgaggtgtacgagatttagggatcgcatgg + cgcggatccaacataaacaagattcattaataac) (restriction sites in bold) was cloned by PCR, and sequenced and subcloned into pRep81 (ref. 44) or pRep81x. An Nde1 site was introduced by PCR at the Pyp2 stop codon and a 13Myc Nde1 fragment cloned by PCR from pFa6a-13Myc-kanMX45 was inserted. The Pst1–Sac1 fragment from the pRep81/81X containing Pyp2 was cloned into IntA46 to allow integration of the pyp2-containing Not1 fragment into the leu1 locus of JP265. The Xho1–Sac1 Tor1 fragment was subcloned from pSLF273-Tor1 (ref. 7) in pRep81x and transformed into a tor1::ura4 (ref. 7) leu1.32 strain, creating JP295.

Microscopy.

Calcofluor staining of septa was performed as described previously40, and at least 200 cells were counted for each time point. Images and cell length measurements were obtained by using a Quantix camera (Photometrics, Marlow, Bucks., UK) with Metamorph software (Universal Imaging, Molecular Devices, Downingtown, PA). More than 100 dividing cells per strain were measured for cell length data.

Biochemistry.

Total extracts were prepared by precipitation with trichloroacetic acid18. Detection of Plo1 S402 phosphorylation was carried out in an identical manner to that previously described for Plo1 (Refs 18, 19). Immunoprecipitation of Spc1/Sty1–Myc (Spc1-12Myc) was performed in accordance with the protocol of Nguyen and Shiozaki47 using magnetic protein A beads (Dynal, Invitrogen Ltd, Paisley, Renfrewshire, UK) to isolate the immune complexes. Phosphorylated Spc1/Sty1 was recognized with PTPY antibodies against phospho-p38 MAPK (catalogue no. 9211S; Cell Signalling, Denvers, MA). Cdc2 was purified on Suc1 agarose beads (catalogue no. 14-122; Upstate, Dundee, UK) from a S. pombe extract using Spc1/Sty1 immunoprecipitation and wash buffers47. Haemagglutinin–Tor-dependent kinase assays were performed as described48. Spc1/Sty1–Myc-dependent kinase assays were performed in accordance with the protocol of Nguyen and Shiozaki47, using MBP as substrate and Protein A (Dynal) beads for immunoprecipitations. Total Eif2alpha and Eif2alpha.S51 phosphorylation was detected from total extracts with antibodies from BioSource (Camarillo, CA).

Note: Supplementary Information is available on the Nature Cell Biology website.



Top

Acknowledgements

We thank Iain Hagan, Keith Gull, Hiro Ohkura, Paul Russell, Jonathan Millar, Ronit Weissmann, Mohan Balasubramanian, Erik Boye and Nic Jones for reagents; the Rockefeller University, the Breast Cancer Research Foundation, the University of Manchester and Nic Jones/Cancer Research UK for support; Iain Hagan for technical support; and Iain Hagan, Rafael Daga, Philip Woodman and Eileithyia Swanton for stimulating discussions and valuable comments on the manuscript.

Received 4 May 2007; Accepted 10 September 2007; Published online 21 October 2007.

Top

References

  1. Mitchison, J. M. Growth during the cell cycle. Int. Rev. Cytol. 226, 165–258 (2003). | Article | PubMed | ISI | ChemPort |
  2. Kim, D. H. et al. mTOR interacts with raptor to form a nutrient-sensitive complex that signals to the cell growth machinery. Cell 110, 163–175 (2002). | Article | PubMed | ISI | ChemPort |
  3. Dolznig, H., Grebien, F., Sauer, T., Beug, H. & Mullner, E. W. Evidence for a size-sensing mechanism in animal cells. Nature Cell Biol. 6, 899–905 (2004). | Article |
  4. Jorgensen, P. & Tyers, M. How cells coordinate growth and division. Curr. Biol. 14, R1014–R1027 (2004). | Article | PubMed | ISI | ChemPort |
  5. Fantes, P. & Nurse, P. Control of cell size at division in fission yeast by growth-modulated size control over nuclear division. Exp. Cell Res. 107, 377–386 (1977). | Article | PubMed | ISI | ChemPort |
  6. Fingar, D. C. & Blenis, J. Target of rapamycin (TOR): an integrator of nutrient and growth factor signals and coordinator of cell growth and cell cycle progression. Oncogene 23, 3151–3171 (2004). | Article | PubMed | ISI | ChemPort |
  7. Weisman, R. & Choder, M. The fission yeast TOR homolog, tor1+, is required for the response to starvation and other stresses via a conserved serine. J. Biol. Chem. 276, 7027–7032 (2001). | Article | PubMed | ISI | ChemPort |
  8. Matsuo, T., Otsubo, Y., Urano, J., Tamanoi, F. & Yamamoto, M. Loss of the TOR kinase Tor2 mimics nitrogen starvation and activates the sexual development pathway in fission yeast. Mol. Cell. Biol. 8, 3154–3159 (2007). | Article |
  9. Weisman, R., Roitburg, I., Schonburn, M., Harari, R. & Kupiec, M. Opposite effects of Tor1 and Tor2 on nitrogen starvation responses in fission yeast. Genetics 175, 1153–1162 (2006). | Article | PubMed | ISI | ChemPort |
  10. Alvarez, B. & Moreno, S. Fission yeast Tor2 promotes cell growth and represses cell differentiation. J. Cell Sci. 119, 4475–4485 (2006). | Article | PubMed | ISI | ChemPort |
  11. Uritani, M. et al. Fission yeast Tor2 links nitrogen signals to cell proliferation and acts downstream of the Rheb GTPase. Genes Cells 11, 1367–1379 (2006). | Article | PubMed | ISI | ChemPort |
  12. Kawai, M. et al. Fission yeast tor1 functions in response to various stresses including nitrogen starvation, high osmolarity, and high temperature. Curr. Genet. 39, 166–174 (2001). | Article | PubMed | ISI | ChemPort |
  13. Weisman, R., Choder, M. & Koltin, Y. Rapamycin specifically interferes with the developmental response of fission yeast to starvation. J. Bacteriol. 179, 6325–6334 (1997). | PubMed | ISI | ChemPort |
  14. Shiozaki, K. & Russell, P. Cell-cycle control linked to extracellular environment by MAP kinase pathway in fission yeast. Nature 378, 739–743 (1995). | Article | PubMed | ISI | ChemPort |
  15. Millar, J. B., Buck, V. & Wilkinson, M. G. Pyp1 and Pyp2 PTPases dephosphorylate an osmosensing MAP kinase controlling cell size at division in fission yeast. Genes Dev. 9, 2117–2130 (1995). | Article | PubMed | ISI | ChemPort |
  16. Shiozaki, K., Shiozaki, M. & Russell, P. Heat stress activates fission yeast Spc1/StyI MAPK by a MEKK-independent mechanism. Mol. Biol. Cell 9, 1339–1349 (1998). | PubMed | ISI | ChemPort |
  17. Mulvihill, D. P., Petersen, J., Ohkura, H., Glover, D. M. & Hagan, I. M. Plo1 kinase recruitment to the spindle pole body and its role in cell division in Schizosaccharomyces pombe. Mol. Biol. Cell 10, 2771–2785 (1999). | PubMed | ISI | ChemPort |
  18. Petersen, J. & Hagan, I. M. Polo kinase links the stress pathway to cell cycle control and tip growth in fission yeast. Nature 435, 507–512 (2005). | Article | PubMed | ISI | ChemPort |
  19. MacIver, F. H., Tanaka, K., Robertson, A. M. & Hagan, I.M. Physical and functional interactions between polo kinase and the spindle pole component Cut12 regulate mitotic commitment in S. pombe. Genes Dev. 17, 1507–1523 (2003). | Article | PubMed | ISI | ChemPort |
  20. Lee, M. G. & Nurse, P. Complementation used to clone a human homolog of the fission yeast-cell cycle control gene cdc2. Nature 327, 31–35 (1987). | Article | PubMed | ISI | ChemPort |
  21. Davey, J. & Nielsen, O. Mutations in cyr1 and pat1 reveal pheromone-induced G1 arrest in the fission yeast Schizosaccharomyces pombe. Curr. Genet. 26, 105–112 (1994). | Article | PubMed | ISI | ChemPort |
  22. Weisman, R., Roitburg, I., Nahari, T. & Kupiec, M. Regulation of leucine uptake by tor1+ in Schizosaccharomyces pombe is sensitive to rapamycin. Genetics 169, 539–550 (2005). | Article | PubMed | ISI | ChemPort |
  23. Matsuo, T., Kubo, Y., Watanabe, Y. & Yamamoto, M. Schizosaccharomyces pombe AGC family kinase Gad8p forms a conserved signaling module with TOR and PDK1-like kinases. EMBO J. 22, 3073–3083 (2003). | Article | PubMed | ISI | ChemPort |
  24. Gould, K. L. & Nurse, P. Tyrosine phosphorylation of the fission yeast cdc2+ protein kinase regulates entry into mitosis. Nature 342, 39–45 (1989). | Article | PubMed | ISI | ChemPort |
  25. Nurse, P. Universal control mechanism regulating onset of M-phase. Nature 344, 503–508 (1990). | Article | PubMed | ISI | ChemPort |
  26. Russell, P. & Nurse, P. Negative regulation of mitosis by wee1+, a gene encoding a protein-kinase homolog. Cell 49, 559–567 (1987). | Article | PubMed | ISI | ChemPort |
  27. Fantes, P. A. & Nurse, P. Control of the timing of cell division in fission yeast. Cell size mutants reveal a second control pathway. Exp. Cell Res. 115, 317–329 (1978). | Article | PubMed | ISI | ChemPort |
  28. Booher, R. & Beach, D. Interaction between cdc13+ and cdc2+ in the control of mitosis in fission yeast; dissociation of the G1 and G2 roles of the cdc2+ protein kinase. EMBO J. 6, 3441–3447 (1987). | PubMed | ISI | ChemPort |
  29. Toone, W. M. & Jones, N. in The Molecular Biology of Schizosaccharomyces pombe (ed. Egel, R.) 57-72 (Springer, Berlin, 2004).
  30. Gaits, F. & Russell, P. Active nucleocytoplasmic shuttling required for function and regulation of stress-activated kinase Spc1/StyI in fission yeast. Mol. Biol. Cell 10, 1395–1407 (1999). | PubMed | ISI | ChemPort |
  31. Maundrell, K. nmt1 of fission yeast. A highly transcribed gene completely repressed by thiamine. J. Biol. Chem. 265, 10857–10864 (1990). | PubMed | ISI | ChemPort |
  32. Gordon, C., McGurk, G., Wallace, M. & Hastie, N. D. A conditional-lethal mutant in the fission yeast 26-S protease subunit Mts3+ is defective in metaphase to anaphase transition. J. Biol. Chem. 271, 5704–5711 (1996). | Article | PubMed | ISI | ChemPort |
  33. Wek, R. C., Jiang, H. Y. & Anthony, T. G. Coping with stress: eIF2 kinases and translational control. Biochem. Soc. Trans. 34, 7–11 (2006). | PubMed | ISI | ChemPort |
  34. Heitman, J., Movva, N. R. & Hall, M. N. Targets for cell cycle arrest by the immunosuppressant rapamycin in yeast. Science 253, 905–909 (1991). | Article | PubMed | ISI | ChemPort |
  35. Cherkasova, V. A. & Hinnebusch, A. G. Translational control by TOR and TAP42 through dephosphorylation of eIF2alpha kinase GCN2. Genes Dev. 17, 859–872 (2003). | Article | PubMed | ISI | ChemPort |
  36. Zhan, K., Narasimhan, J. & Wek, R. C. Differential activation of eIF2 kinases in response to cellular stresses in Schizosaccharomyces pombe. Genetics 168, 1867–1875 (2004). | Article | PubMed | ISI | ChemPort |
  37. Tvegard, T. et al. A novel checkpoint mechanism regulating the G1/S transition. Genes Dev. 21, 649–654 (2007). | Article | PubMed | ISI | ChemPort |
  38. Xie, M. W. et al. Insights into TOR function and rapamycin response: chemical genomic profiling by using a high-density cell array method. Proc. Natl Acad. Sci. USA 102, 7215–7220 (2005). | Article | PubMed | ChemPort |
  39. Fingar, D. C., Salama, S., Tsou, C., Harlow, E. & Blenis, J. Mammalian cell size is controlled by mTOR and its downstream targets S6K1 and 4EBP1/eIF4E. Genes Dev. 16, 1472–1487 (2002). | Article | PubMed | ISI | ChemPort |
  40. Moreno, S., Klar, A. & Nurse, P. Molecular genetic analysis of fission yeast Schizosaccharomyces pombe. Methods Enzymol. 194, 795–823 (1991). | Article | PubMed | ISI | ChemPort |
  41. Egel, R., Willer, M., Kjaerulff, S., Davey, J. & Nielsen, O. Assessment of pheromone production and response in fission yeast by a halo test of induced sporulation. Yeast 10, 1347–1354 (1994). | Article | PubMed | ISI | ChemPort |
  42. Bahler, J. et al. Heterologous modules for efficient and versatile PCR-based gene targeting in Schizosaccharomyces pombe. Yeast 14, 943–951 (1998). | Article | PubMed |