1

A DOWNSHIFT IN TEMPERATURE ACTIVATES THE HOG PATHWAY, WHICH DETERMINES FREEZE TOLERANCE IN Saccharomyces cerevisiae*

Joaquín Panadero, Claudia Pallotti, Sonia Rodríguez-Vargas,

Francisca Randez-Gil, and Jose A. Prieto#

From the Department of Biotechnology, Instituto de Agroquímica y Tecnología de los Alimentos, Consejo Superior de Investigaciones Científicas, P. O. Box 73, E-46100-Burjassot

Valencia, Spain. #To whom correspondence should be addressed: Department of Biotechnology, Instituto de Agroquímica y Tecnología de los Alimentos (CSIC), P.O. Box 73, 46100-Burjassot, Valencia, Spain. Tel. +34-963900022. Fax. +34-963636301. E-mail: prieto@iata.csic.es

Running title: Freeze-tolerance regulation by the MAPK Hog1

The molecular mechanisms that enable

yeast cells to detect and transmit cold signals and their physiological significance in the adaptive response to low temperatures are unknown. Here, we demonstrate that the MAPK Hog1p is specifically activated in response to cold. Phosphorylation of Hog1p was dependent on Pbs2p, the MAPKK of the HOG pathway, and Ssk1p, the response regulator of the two-component system Sln1p-Ypd1p. However, Sho1p was not required. Interestingly, phosphorylation of Hog1p was stimulated at 30ºC in cells exposed to the membrane rigidifier agent dimethyl sulfoxide. Moreover, Hog1p activation occurred specifically through the Sln1-branch. This suggests that Sln1p monitors changes in membrane fluidity caused by cold. Quite remarkably, activation of Hog1p at low temperatures affected the transcriptional response to cold-shock. Indeed, absence of Hog1p impaired the cold-instigated expression of genes for trehalose- and glycerol-synthesizing enzymes and small chaperones. Moreover, a downward transfer to 12 or 4ºC stimulated the overproduction of glycerol in a Hog1p-dependent manner. However, hog1∆ mutant cells showed no growth defects at 12ºC as compared to the wild-type. On the contrary, deletion of HOG1 or GPD1 decreased tolerance to freezing of wild-type cells pre-incubated at a low temperature, while no differences could be detected in cells shifted directly from 30 to -20ºC. Thus, exposure to low temperatures triggers a Hog1p-dependent accumulation of glycerol, which is essential for freeze protection.

Variations in the surrounding temperature are probably the most common stress for all living organisms. In particular, a downshift in temperature leads to a reduction in membrane fluidity, impaired protein biosynthesis and stabilization of secondary structures of DNA and RNA (1). Consequently, the adaptive response to cold-shock in most organisms, studied so far, includes a change in the lipid composition of membranes and the remodeling of the transcriptional and translational machinery. These changes are mainly triggered by a drastic variation in the gene expression program, which leads to both survival and adaptation to low temperatures.

Unlike other stress conditions, the biochemical mechanisms by which eukaryotic cells sense changes and respond to a downshift in temperature are poorly understood. Using the cyanobacteria Synechocystis as an experimental model, Murata and Los (2) suggested that a phase transition of the plasma membrane upon cold-shock would trigger a conformational change of a putative cold-sensor, this being the primary event in the transduction of the temperature signal. According to this hypothesis, a membrane-bound histidine kinase Hik33, has been identified as a cold-sensor in this organism, which is able to detect changes in the physical state of the membrane and regulate the expression of most cold-induced genes (3, 4). Other putative cold-sensors, like DesK in Bacillus subtilis (5) and TRP-channels in the mammalian nervous system (6) have also been characterized as integral membrane proteins. Recently, Hik33 has been shown to regulate the expression in Synechocystis of osmostress-inducible genes (7), suggesting that both osmotic

2

and cold-stress could be perceived in this organism by common mechanisms and sensing elements.

In the yeast Saccharomyces cerevisiae, an increase in environmental osmolarity is perceived by one of two osmosensors, Sln1p, the only known yeast histidine kinase sensor, and Sho1p, a transmembrane protein (reviewed in references 8-10). Recently, a third osmosensor Msb2p, displaying a redundant function with Sho1p, has also been characterized (11). Sln1p, together with Ypd1p and Ssk1p forms a so-called two-component system (12), a typical signal transduction system found in other fungi, plants and bacteria (13). Once the osmosensors are activated, they transmit the osmostress signal through different elements. This signal is integrated at the level of the MAPKK Pbs2p. Then, Pbs2p phosphorylates the Hog1p MAPK, which gives the name to the osmoregulatory MAPK cascade, the HOG pathway.

Although the HOG pathway has classically been considered as specific to osmotic stress, recent studies have revealed new functions for this MAPK route. Evidence suggests that the HOG pathway is essential for regulating adaptation to citric acid stress in yeast (14). Heat stress also activates the HOG pathway, through the membrane-bound osmosensor Sho1p (15). HOG basal activity is also involved in methylglyoxal resistance (16), distribution of proteins within the Golgi (17) and cell-wall maintenance (18).

Additional functions to that of osmostress signaling have been identified for HOG homologous pathways in other species. In the fission yeast S. pombe, the Wak1p/Win1p-Wis1p-Sty1p SAPK pathway drives the transcriptional response to different stimuli, like osmostress, heat-shock, oxidative and UV injury integrating the response to multiple stresses into a common cascade (for a review see reference 19). The central element of this cascade is the Sty1p MAPK (20), the functional homologue to mammalian p54 JNK and p38, two of the three mammalian stress-activated MAPKs, SAPKs (8) and yeast Hog1p. Recently, Sty1p has been reported to be phosphorylated in response to cold-shock, leading to the activation of cold-induced genes (21). Similarly, a downshift in temperature triggers the activation of p54 JNK (22). However, the physiological significance of these responses in cold-adaptation and the mechanisms that trigger activation upon cold-

shock of osmoregulatory MAPK pathways remain unclear.

Our objective in this study was to characterize the molecular mechanisms mediating cold response in S. cerevisiae. We hypothesized that the HOG MAPK pathway plays an important role in the phenotypic adaptation of yeast cells to this stress. Our results have shown that the HOG pathway mediates in transmitting the cold signal, regulates the expression of a sub-set of cold-induced genes and determines viability on freezing. These findings support the idea that the HOG pathway plays a more general role in stress response in yeast, thus representing an important advance to our limited knowledge of the strategies and targets involved in cell protection against freeze injury.

EXPERIMENTAL PROCEDURES

Yeast Strains and Culture Conditions - S. cerevisiae W303-1A and BY4741 wild-type strains and isogenic mutants used in this study are listed in TABLE I. Cells were grown in YNB [0.67% yeast nitrogen base without amino acids (DIFCO) plus 2% glucose] supplemented with the appropriate concentrations of essential nutrients (25) or YPD (1% yeast extract, 2% peptone, 2% glucose). Cells were grown routinely in Erlenmeyer flasks at 30 or 12ºC on an orbital shaker (250 rpm).

For cold-shock experiments, cells grown at 30ºC to mid-log phase (OD600 = 0.3-0.5) were collected by centrifugation, resuspended in 12ºC-pre-cooled culture medium and incubated at the same temperature for different periods. Phenotype experiments on solid media were made by adjusting 30ºC-exponential cultures to OD600 = 0.3, and spotting serial dilutions onto culture plates. Cells were incubated at 30 or 12ºC for 2 and 6 days, respectively. E. coli (DH10B strain) was grown in LB medium (1% peptone, 0.5% yeast extract, 0.5% NaCl). Plasmids - Plasmids for monitoring Hog1p nuclear accumulation were pVR65, containing HOG1 fusion with GFP in vector YCp111 (26), and pGP22, a derivative of vector pRS413, which contains the PTP2-coding sequence under the control of the pGAL1 (H. Saito). Yeast cells were transformed according to Ito et al. (27), and transformants were selected by auxotrophic complementation in YNB plates. E. coli was transformed by electroporation following the manufacturer’s instructions (Eppendorf).

3

RNA Purification and Northern Blot Analysis - Cells from 30 ml culture samples were harvested

by centrifugation (5,500 x g, 5 min, 4ºC), resuspended in 0.5 ml LETS buffer (200 mM LiCl, 20 mM EDTA, 20 mM Tris-HCl [pH 8.0], 0.4% SDS) and transferred to a screw-cap Eppendorf tube containing 0.5 ml phenol and 0.5 ml glass beads (acid-washed, 0.4-mm diameter). The suspension was mixed vigorously twice, for 45 s each time in a Mini Bead-Beater homogenizer (BioSpec, Bartlesville, OK) and total RNA was purified as previously described (25). Purified RNA was diluted in 30-50 µl of sterile DEPC-treated water.

Equal amounts of RNA (30 µg) were separated in 1% (w/v) agarose gels, containing formaldehyde (2.5% v/v), transferred to a Nylon membrane and hybridized with 32P-labelled probes. PCR-amplified DNA fragments containing sequences of TPS1 (-2 to +1,492), NSR1 (-48 to +536), OLE1 (+111 to +1,121), HSP12 (+1 to +329), GLO1 (+5 to +975), TIP1 (-139 to +1,507), GRE1 (+1 to +506), GPD1 (+23 to +848) and ACT1 (+10 to +1083) were used to probe mRNA levels. DNA sequences were obtained from the MIPS database (available at http://mips.gsf.de). Probes were radiolabeled with the random primer Ready-to-Go kit (Amersham Biosciences, Chalfont-St Giles, England) and [α32P]dCTP (Amersham Biosciences). Filters were exposed to a high-resolution BAS-MP 2040S imaging plate (Fuji, Kyoto, Japan) for 24-48 h and scanned in a Phosphorimager (Fuji, FLA-3000). Spot intensities were quantified with the Image Gauge software version 3.12 (Fuji). Values of spot intensity were corrected according to the ACT1 mRNA level and represented as the relative mRNA level. The highest relative mRNA for each gene and sample analyzed was set at 100. Preparation of Yeast Extracts and Western Blot - To prepare whole-cell extracts for Hog1p detection, cells were harvested by filtration, transferred to a tube containing 300 µl of SDS loading buffer (50 mM Tris–HCl, pH 6.8, 100 mM dithiothreitol, 2% SDS, 0.1% bromophenol blue, 20% glycerol) and boiled at 95°C for 10 min. Ten µl of each sample was separated by SDS–PAGE and blotted onto nitrocellulose membranes. Filters were blocked with 5% BSA in TBST (10 mM Tris-HCl, pH 8.0, 150 mM NaCl, 0.5% Tween-20).

Dual phosphorylated Hog1p was detected by an antibody specific to phosphorylated p38 MAPK (Cat# 9215, Cell signaling, Beverly,

MA). A rabbit polyclonal antibody raised against a recombinant protein corresponding to the carboxy terminus (221-435) of S. cerevisiae Hog1p was used as a loading control (Cat# sc-9079, Santa Cruz Biotechnology, Santa Cruz, CA). The antisera were applied at 1:1000 (phosphorylated Hog1p) and 1:6000 (total Hog1p) dilutions according to the manufacturer’s instructions. Blots were developed using the ECL Western blotting detection kit from Amersham Biosciences. Films were scanned in a HP scanjet 5370c (Hewlett-Packard, Palo Alto, CA) and quantified with the Image Gauge software version 3.12 (Fuji). Intensity values of phospho-Hog1p were corrected according to the total Hog1p level. The highest relative value for each protein and sample analyzed was set at 100. Fluorescence Microscopy - Cells were cultured as described, and DAPI (2,5 µg/ml final concentration) was added to the culture 1h before microscopy. One ml samples were taken at different times before and after DMSO addition. Samples were spun for 10 s, the supernatant decanted, and the cells were resuspended in the residual liquid. Aliquots of 4 µl were directly observed in a microscope Leica DM IRB, employing filters for GFP and DAPI fluorescence. False positives were detected by comparison to DAPI pictures, which revealed the position of the nuclei. At least 100 cells were visualized each time. Glycerol Assay - To determine intracellular glycerol content under cold-stress conditions, YPD-grown cells (OD600 = 1.0) were collected by centrifugation, resuspended in 12ºC-or 4ºC-pre-cooled YPD culture medium and incubated at the indicated temperature. At different times, cells (10 units of OD600) were collected by filtration and quickly transferred to a cold tube containing 1 ml of distilled water. Then, the yeast suspension was boiled for 10 min, cooled on ice, and centrifuged at 15,300 x g for 10 min (4ºC). Finally, the supernatant was collected and used for further analysis. Glycerol was determined colorimetrically with a commercial kit (Boehringer), following the manufacturer’s instructions. The Values obtained are expressed as µg of glycerol per mg of yeast cells, dry weight. To do this, we used a cell mass value of 0.3 mg cells dry weight/ml per unit of OD600 (28). Viability Experiments - Cell cultures were grown in YPD liquid medium at 30ºC (OD600 = 0.4-0.6) and samples were transferred directly to

4

-20 or pre-incubated at 12ºC or 4ºC for 6 h or 48 h, respectively, prior the shift to -20ºC. At different times, samples were thawed at 30ºC for 30 min and diluted, and cells were plated onto solid YPD. After 2 days, colonies were counted and the percent of viable cells was measured.

RESULTS

A Downshift in Temperature Triggers the Phosphorylation of Hog1p via the Sln1-Branch of the HOG Pathway - We examined the effect of a temperature downshift on the phosphorylation state of Hog1p in cells of the parental strain BY4741. Fig. 1 shows the results of a Western blot analysis of protein extracts analyzed with anti-phospho p38 antibody, which recognizes the dual-phosphorylated form (Thr174, Tyr176) of Hog1p, and polyclonal anti-Hog1p as a loading control. Hog1p was clearly activated in response to a thermal transfer from 30ºC to 12ºC. The kinetics of Hog1p phosphorylation was similar to that previously reported for yeast cells subjected to mild osmotic stress (29). Thus, the phosphorylation level reached a maximum within 5-10 min of exposure to low temperature and decreased markedly after 30 min. Similar results were found when cells were transferred to 15ºC or 10ºC (data not shown).

We further investigated whether the cold-instigated phosphorylation of Hog1p was dependent on upstream components of the HOG pathway. The mechanisms and elements implied in the Hog1p MAPK activation by osmostress have been widely characterized (8-10). Phosphorylation of Hog1p is directly dependent on Pbs2p MAPKK, which is, in turn, activated by any of the two feeding signals derived from Sho1p or the Sln1p-Ypd1p two-component system. As can be seen in Fig. 1, phosphorylation of Hog1p at 12ºC was completely absent in a pbs2∆ mutant. Then, we analyzed the signal transfer from the two putative osmosensing receptors, Sho1p and Sln1p. Disruption of SLN1 histidine kinase is lethal (12), and therefore, we visualized the phosphorylation of Hog1p in mutants of the SSK1 response regulator. As shown in Fig. 1, deletion of SHO1 had no major effects on the phospho-Hog1p accumulation after a temperature downshift. By contrast, deletion of SSK1 eliminated the cold-instigated phosphorylation of the MAPK. Therefore, a downward shift to low temperatures drives the

phosphorylation of the osmosensitive HOG MAPK pathway. However, only the Sln1-branch appears to be absolutely essential for this response. Exposure of Yeast Cells to DMSO Leads to the Phosphorylation and Nuclear Import of the MAPK Hog1p – The absence of cold-instigated phosphorylation of Hog1p in the sho1∆ mutant suggested that osmotic imbalance and cold stress are recognized differently by yeast cells. We tried to find further support for this idea by analyzing what effects a change in the physical state of the yeast membrane had on the phosphorylation state of Hog1p. Cold-stress is assumed to cause rapid but reversible rigidification of the plasma membrane, which can be mimicked by treating the membrane with the rigidifier DMSO. This hypothesis has been validated by different studies on plant cells (30, 31). Like exposure to low temperature, treatment with DMSO induces rigidification of the plant cell membrane leading to activation of several cold-induced MAPKs (31). In S. cerevisiae, there are reports of a connection between membrane physical state and heat-shock gene transcription (32). Consequently, we tested whether the HOG pathway could be activated by chemically modulated changes in membrane fluidity. As shown in Fig. 2A, Western blot revealed that the rapid and transient phosphorylation of Hog1p was activated by adding 250 mM DMSO to a yeast culture of the BY4741 wild-type strain. Similar results were observed by exposure to sub-lethal DMSO concentrations in the range 125-500 mM (data not shown). Quite remarkably, DMSO activated Hog1p in sho1∆ mutant cells, but not in the ssk1∆ mutant (Fig. 2A). Moreover, phosphorylation of Hog1p was completely dependent on Pbs2p. Thus, our results suggest that DMSO selectively stimulates the Sln1-branch of the HOG pathway.

Hog1 protein is reported to move into the nucleus upon phosphorylation (24). Therefore, nuclear localization of the MAPK by treating the cells with DMSO would be further evidence of its activation. In order to test this, we used a hog1∆ mutant strain (W303-1A background) transformed with plasmid pVR65, carrying a HOG1-GFP fusion (26), and DMSO-exposed cells were visualized by fluorescent microscopy. As shown, the rigidifying agent induced the nuclear accumulation of Hog1p-GFP just 10 min after the treatment (Fig. 2B). We also observed that nuclear localization of Hog1p did not

5

decrease at all in 30 min. This fact was in consonance with the phosphorylation kinetics displayed in this strain. Indeed, Hog1p phosphorylation by DMSO in cells of the S. cerevisiae W303-1A strain lasted longer (Fig. 2B) as compared to that seen in the BY4741 strain (Fig. 2A). Hence, if yeast cells are exposed to DMSO at 30ºC, the cold-induced phosphorylation of Hog1p is mimicked and the nuclear localization of the MAPK is determined. A functional HOG Pathway is Dispensable for Growth at Low Temperature - S. pombe responds to cold by inducing the SAPK pathway. However, such activation is not necessary for growth at low temperatures (21). Therefore, we were interested in testing the effects of HOG pathway activation in the cold response of S. cerevisiae. Exponentially growing cells of the parental BY4741 and HOG pathway mutant strains were plated on YPD medium and cultivated at 30 or 12ºC (Fig. 3). As can be seen, deletion of HOG1, PBS2, SHO1or SSK1 had no effects on the growth of the parental strain at 12ºC. Similar results were observed at 15ºC or 10ºC, or when growth at a low temperature was tested in either YPD or YNB liquid medium (data not shown). The HOG Pathway is Involved in the Genetic Response to Cold Stress - Under osmotic stress, activation of Hog1p alters the expression of hundred of genes (33). Therefore, we looked into how the HOG pathway is involved in the transcriptional response of S. cerevisiae to cold-shock. Three genes, OLE1, TIP1 and NSR1, previously reported as cold-induced and traditionally used as cold-shock response markers (34-36) were probed by Northern blot of mRNA samples from wild-type and hog1∆ mutant cells. OLE1 encodes the stearoyl-CoA 9-desaturase (37), TIP1 encodes a mannoprotein of the cell wall (36) and NSR1 is a nucleolar protein involved in pre-rRNA processing (35). As expected, transcription of OLE1, TIP1 and NSR1 was fully induced in wild-type cells within 2-4 h after a shift from 30ºC to 12ºC (Fig. 4A). However, no effects were found on deleting HOG1. Indeed, all three marker genes displayed similar kinetics of mRNA variation when samples from wild-type and hog1∆ strains were compared (Fig. 4A).

Next we examined the cold-inducible expression of a different set of genes. In particular, we analyzed the transcriptional profile of GLO1, GPD1, TPS1, HSP12 and GRE1. GLO1 encodes glyoxalase I (38). The

product of GPD1 catalyzes the first step in the glycerol pathway (39). TPS1 encodes the trehalose-6-phosphate synthase (40). HSP12 encodes a small heat-shock protein (41) and GRE1 encodes for a protein of unknown function. It has recently been reported that all of them are induced by 6-8 h of incubation at 10ºC (42), a regulation pattern that was missed in earlier studies of the cold-shock response (35, 36, 43). Findings also show that TPS1 expression is activated below 10ºC and even at 0ºC (45). As is shown in Fig. 4B, the expression level of all the marker genes tested increased markedly in cold-shocked cells of the wild-type strain. Nevertheless, they showed different kinetics of mRNA variation. The transcript accumulation of GLO1, HSP12 and TPS1 only appeared to increase significantly at the late stage, 4-8 h, of the incubation period at 12ºC. By contrast, the induction curve of GRE1 and especially of GPD1, showed two phases (Fig. 4B). mRNA levels increased rapidly within the first 60 min shift, then dropped and increased again by 2 h. By comparison, the hog1∆ mutation strongly affected the expression of the genes analyzed (Fig. 4B). Indeed, there was no apparent cold-induction of GPD1, GRE1, GLO1 and HSP12. Only the induction profile of TPS1 remained unaltered, although the maximal mRNA accumulation was clearly diminished (Fig. 4B). Yeast Cells Accumulate Glycerol in Response to a Downward Shift in Temperature – The observation that a number of genes are coordinately induced at low temperature via the HOG pathway led us to investigate the functionality of this response. In particular, we tested whether up-regulation of GPD1 at 12ºC might determine enhanced levels of glycerol accumulation at low temperature. As shown in Fig. 5, when logarithmic growing cells of the BY4741 wild-type strain were shifted to 12ºC, intracellular glycerol content started to increase and by 3 h reached levels three times higher than those at 30ºC. Later, glycerol accumulated at lower relative rates reaching near maximum values after 9 h (Fig. 5). This pronounced increase did not occur by deletion of HOG1. Thus, glycerol content hardly varied in hog1∆ mutant cells after 3 h at 12ºC, and by 15 h had increased only about 2-fold over the 30ºC levels (Fig. 5).

Then we went on to test whether glycerol content could also increase at lower temperatures where yeast growth stops. Recent

6

studies by Kandror et al. (44) have shown that yeast cells incubated at near-freezing temperatures can undergo a strong increase in their trehalose content. Therefore, we observed glycerol content in wild-type and hog1∆ mutant cells shifted to 4ºC. As shown in Fig. 5, glycerol clearly accumulated upon shifting yeast cells of the parental strain to this temperature. Quite remarkably, the increase in glycerol content was higher at 4ºC than that observed at 12ºC (Fig. 5). Thus, the intracellular level of glycerol showed a 7-fold increase after 24 h at 4ºC and by 48 h had enhanced about 9-fold. As expected, this large increase was not observed in mutant cells lacking Hog1p (Fig. 5). Moreover, the effects of HOG1 deletion on glycerol content appeared to be much more pronounced at 4ºC than at 12ºC. While a slight but significant increase in glycerol content was measured at 12ºC, on transference to 4ºC no variations in the 30ºC levels could be detected after 48 h (Fig. 5). Hog1p Plays a Functional Role in Freeze Tolerance – Overaccumulation of trehalose at near-freezing temperatures is critical for the viability of yeast cells exposed to freezing (44). Similarly, the Hog1p-mediated increase in glycerol content after a downward shift in temperature may provide freeze tolerance. To test this possibility, cultures of wild-type and hog1∆ mutant cells were transferred from 30ºC to 4ºC for 2 days, and then frozen at -20ºC. Samples were thawed after 2 h, 1 or 4 days, and analyzed for cell viability. Cells shifted directly from 30ºC to -20ºC were used as control. As shown in Fig. 6A, similar degrees of sensitivity to freezing were displayed by control cells of the hog1∆ or wild-type strain. For instance, 15% of wild-type and 10% of mutant cells survived after 4 days at -20ºC. The incubation of wild-type cells at 4ºC for 2 days enhanced freeze tolerance (Fig. 6B), which was to be expected, given such conditions induce maximal glycerol accumulation (Fig. 5). In a representative experiment, about 40% of 4ºC-treated cells survived for 4 days at -20ºC. The strain without Hog1p, in which no variation in glycerol content could be measured at 4ºC (Fig. 5), also increased its resistance to freezing upon incubation at this temperature for 48 h (Fig. 6B). Nevertheless, this effect was less pronounced than that observed in the wild-type strain, again suggesting that a high level of intracellular glycerol helps to protect yeast cells against freeze damage.

To further investigate this view, we analyzed the behavior of a gpd1∆ mutant, which lacks glycerol-3-phosphate dehydrogenase (39). As expected, gpd1∆ mutant cells shifted directly from 30ºC to -20ºC displayed a similar degree of sensitivity to that observed for wild-type and hog1∆ cells (Fig. 6A). Cell survival was also followed up after incubation for 2 days at 4ºC and frozen storage (Fig. 6B). Again, it was evident that incubation at 4ºC improved freeze tolerance. However, the increase in freeze tolerance was less marked than that in wild-type cells, being closer to that observed for the hog1∆ mutant strain, especially after 4 days at -20ºC.

DISCUSSION Traditionally, the HOG pathway, a

conserved MAPK cascade in S. cerevisiae, has been considered a specific signaling route responding to changes in external solute concentrations (8-10). Far from this classical concept, our results demonstrate that the HOG pathway functions as a transducer of cold stimuli. By using genetic and biochemical approaches, we showed that Hog1p is phosphorylated by Pbs2p, the MAPKK of the HOG pathway, after a downward shift in temperature. Phosphorylation of Hog1p was also found to be induced by the membrane-rigidifier agent DMSO. Several lines of research in plants and cyanobacteria have shown that this chemical indeed mimics the changes in membrane fluidity caused by a sharp decrease in temperature (2, 45-47), and triggers the activation of several cold-induced MAPKs (31). Furthermore, our results revealed that both cold-shock and DMSO activate Hog1p specifically via the Sln1-branch of the MAPK cascade. Thus, osmotic-stress and cold appear to be perceived by yeast cells as different stimuli and generate stress signals that are transmitted by specific elements of the HOG pathway.

Quite remarkably, this activation led to the transcriptional induction of a specific group of cold-responsive genes and the intracellular accumulation of glycerol, a key process leading to freeze resistance. This is consistent with the function of Hog1p orthologs, like Sty1p in the fission yeast S. pombe or p38 in mammals, which relay different environmental signals, including cold, to the transcriptional machinery (8, 19). These results also support recent reports showing that the MAPK Hog1p plays a role in response to different stimuli (14-16). Hence, the

7

HOG pathway has a functional role in the propagation of a variety of signals to changing environmental conditions, and determines tolerance to different stresses.

The finding that the two-component osmosensor Sln1p-Ypd1p, participates in cold perception was not wholly surprising. Sln1p, a typical histidine kinase, is the only two-component sensor molecule known in S. cerevisiae, and there is considerable evidence that low-temperature signals are regulated by two-component systems in both prokaryotic and eukaryotic cells (48, 49). Histidine kinases that act as cold sensors, such as Hik33 in Synechocystis (3) and DesK in B. subtilis (5) have also been identified. Moreover, Hik33 appears to be responsible for detecting both cold and osmotic stress (7). Like this cyanobacterial protein, Sln1p appears to function under either of these stressful conditions and it is likely, therefore, that the basic mechanism of its activation by low temperature and osmotic pressure could be similar, or even identical.

High osmotic stress causes a rapid efflux of water with associated cell volume shrinkage and reduction of turgor pressure. Recently, it has been shown that the HOG pathway is activated by reductions in turgor, through yeast-cell exposure to nystatin, without any application of osmotic stress (50). Nystatin is a well known yeast membrane-permeabilizing agent that induces cell volume shrinkage (51). Quite remarkably, nystatin gave rise to Hog1p phosphorylation specifically through the Sln1-branch of the MAPK cascade (50). Thus, this observation suggests that there is indeed a causal link between the stimulation of Hog1p through the Sln1p-branch upon cold-shock and osmotic stress.

The fluidity state of the cell membrane might be a key factor to integrate the sensing mechanism of cold and hyperosmolarity. Several studies in Synechocystis (52, 53) and bacteria (5) indicate that cold-sensors perceive a decrease in membrane fluidity as the primary signal of cold stress. There is also evidence that different osmosensors, like EnvZ in E. coli or OpuA, a transmembrane transporter of L. lactis with osmosensor and osmoregulator properties (54), are stimulated by changes in the fluidity of membrane lipids (55, 56). Moreover, studies in B. subtilis (57) and S. cerevisiae (58) indicate that osmotic stress reduces cell-membrane fluidity. Altogether, these observations suggest a model in which Sln1p monitors the changes in

membrane fluidity caused by different stressors. The fact that Sln1p responds to DMSO supports this view. However, more work is required to confirm this idea and to understand the exact molecular mechanism of signal perception.

As shown in our study, signaling through the HOG pathway was only required for the proper expression of a specific set of cold-induced genes. This suggests that additional cold-signal transduction pathways differing in targets and likely function are operative in S. cerevisiae. Hog1p-regulated genes might be important for growth at low temperature. However, experiments with HOG-pathway mutants failed to show a cold-sensitive phenotype, even though lack of Hog1p had dramatic effects on the transcriptional induction of these genes. Another possibility is that these genes could be necessary to protect yeast cells against cold-shock damage. TPS1, HSP12, GRE1, GLO1 and GPD1, encode proteins with functions in or related to stress protection. Notably, all these genes are also induced by osmotic stress in a Hog1p-dependent manner (33). However, a sharp decrease in temperature arrests or delays growth but has no major effects on cell viability. Although we can not discard a regulatory role of the HOG pathway in the cold-adaptation mechanism, this function remains unclear.

Recently, Kandror et al. (44) have shown that yeast cells display an adaptive response at low temperature, the termed “near-freezing response”, which increases tolerance to freezing. This response is dependent on the transcriptional regulators MSN2/MSN4, the two factors mediating the general stress response pathway in S. cerevisiae (59), and involves the production of high amounts of trehalose and synthesis of certain heat-shock proteins (44). On the other hand, previous studies have shown there is a functional link between Hog1p and Msn2p/Msn4p in the yeast osmotic response of GPD1 (60, 61). Glycerol is the only compatible osmolyte in S. cerevisiae and its production plays a fundamental role in osmoadaptation. Moreover, much of the damage to cells during freezing/thawing, in particular at low freezing rates, is a result of osmotic shrinkage (62). Such evidence and the findings reported here would strongly suggest that Hog1p could be an important factor in the adaptive response to freezing.

In support of this, we observed a high accumulation of glycerol in cells that were

REFERENCES

8

transferred to 12ºC or 4ºC. Notably, this accumulation was higher at the lowest temperature. These results could be explained by lower levels of activity in the glycerol export channel Fps1p at 4ºC as compared to 12ºC. The glycerol facilitator Fps1p in S. cerevisiae plays a central role in the accumulation of glycerol under osmotic stress (63, 64). Moreover, absence of Fps1p confers increased freeze tolerance to yeast cells (65). As expected, glycerol accumulation upon cold-shock at 12ºC or 4ºC was wholly dependent on Hog1p and enhanced cell viability upon freezing. Indeed, the hog1∆ mutant strain demonstrated lower freeze tolerance than the wild-type strain when cells were incubated at 4ºC for 48 h prior the transfer to -20ºC. Similar results were found at 12ºC (data not shown). More importantly, absence of Gpd1p led to approximately the same loss of viability in cells pre-adapted at 4ºC, but not in

cells transferred directly from 30ºC to -20ºC. Hence, glycerol accumulation at low temperature contributes to protecting yeast cells against freezing injury.

Although cold-induction of a number of genes was found to be dependent on Hog1p, the most important factor contributing to enhanced freeze tolerance was the increase in the intracellular glycerol content. This result and previous findings (44) suggest than glycerol acts together with trehalose and perhaps heat-shock proteins in the adaptation mechanism of S. cerevisiae, and probably other organisms, to temperatures below 0ºC. Freezing is a complex and multifaceted stress, in which different stressors and stress responses appear to play important roles (66). It is not surprising; therefore, that tolerance to freezing is likely to involve different mechanisms and protective molecules working in concert.

Acknowledgements We thank A. Blasco for technical assistance and M. Tamás and S. Hohmann for providing plasmids and yeast strains.

1. Thieringer, H. A., Jones, P. G., and Inouye, M. (1998) BioEssays 20, 49-57 2. Murata, N., and Los, D. A. (1997) Plant Physiol. 115, 875-879 3. Suzuki, I., Los, D. A., Kanesaki, Y., Mikami, K., and Murata, N. (2000) EMBO J. 19, 1327-1334 4. Suzuki, I., Kanesaki, Y., Mikami, K., Kanehisa, M., and Murata, N. (2001) Mol. Microbiol. 40, 235-244 5. Aguilar, P. S., Hernandez-Arriaga, A. M., Cybulski, L. E., Erazo, A. C., and de Mendoza, D. (2001) EMBO J. 20, 1681-1691 6. McKemy, D. D., Neuhausser, W. M., and Julius, D. (2002) Nature 416, 52-58 7. Mikami, K., Kanesaki, Y., Suzuki, I., and Murata, N. (2002) Mol. Microbiol. 46, 905-915 8. de Nadal, E., Alepuz, P. M., and Posas, F. (2002) EMBO Rep. 3, 735-740 9. Hohmann, S. (2002) Microbiol. Mol. Biol. Rev. 66, 300-372 10. Westfall, P.J., Ballon, D. R., and Thorner, J. (2004) Science 306, 1511-1512 11. O'Rourke, S. M., and Herskowitz, I. (2002) Mol. Cell. Biol. 22, 4739-4749 12. Maeda, T., Wurgler-Murphy, S. M., and Saito, H. (1994) Nature 19, 242-245 13. Gustin, M. C., Albertyn, J., Alexander, M., and Davenport, K. (1998) Microbiol. Mol. Biol. Rev. 62, 1264-1300 14. Lawrence, C. L., Botting, C. H., Antrobus, R., and Coote, P. J. (2004) Mol. Cell. Biol. 24, 3307-3323 15. Winkler, A., Arkind, C., Mattison, C. P., Burkholder, A., Knoche, K., and Ota, I. (2002) Eukaryot. Cell 1, 163-173 16. Aguilera, J., Rodríguez-Vargas, S., and Prieto, J. A. (2005) Mol. Microbiol. 56, 228-239 17. Reynolds, T. B., Hopkins, D. B., Lyons, M. R., and Graham, T. R. (1998) J. Cell. Biol. 143, 935-946 18. García-Rodríguez, L. J., Duran, A., and Roncero, C. (2000) J. Bacteriol. 182, 2428-2437 19. Gacto, M., Soto, T., Vicente-Soler, J., Villa, T. G., and Cansado, J. (2003) Int. Microbiol. 6, 211-219 20. Millar, J. B., Buck, V., and Wilkinson, M. G. (1995) Genes Dev. 9, 2117-2130

9

21. Soto, T., Beltran, F. F., Paredes, V., Madrid, M., Millar, J. B., Vicente-Soler, J., Cansado, J., and Gacto, M. (2002) Eur. J. Biochem. 269, 5056-5065 22. Ohsaka, Y., Ohgiya, S., Hoshino, T., and Ishizaki, K. (2002) Cell. Physiol. Biochem. 12, 111-118 23. Thomas, B.J., and Rothstein, R. (1989) Cell 56, 619-630 24. Cvrcková, F., de Virgilio, C., Manser, E., Pringle, J.R., and Nasmyth, K. (1995) Genes Dev. 9, 1817-1830 25. Sherman, F., Fink, G. R., and Hicks, J. B. (1986) Methods in Yeast Genetics, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY 26. Reiser, V., Ruis, H., and Ammerer, G. (1999) Mol. Biol. Cell 10, 1147-1161 27. Ito, H., Jukuda, K., Murata, K., and Kimura, A. (1983) J. Bacteriol. 153, 163-168 28. Ansell, R., Granath, K., Hohmann, S., Thevelein, J.M., and Adler, L. (1997) EMBO J. 16, 2179-2187 29. van Wuytswinkel, O., Reiser, V., Siderius, M., Kelders, M. C., Ammerer, G., Ruis, H., Mager, W. H. (2000) Mol. Microbiol. 37, 382-397 30. Örvar, B. L., Sangwan, V., Omann, F., and Dhindsa, R. S. (2000) Plant J. 23, 785-794 31. Sangwan, V., Örvar, B. L., Beyerly, J., Hirt, H., and Dhindsa, R. S. (2002) Plant J. 31, 629-638 32. Carratu, L., Franceschelli, S., Pardini, C. L., Kobayashi, G. S., Horvath, I., Vigh, L., and Maresca, B. (1996) Proc. Natl. Acad. Sci. USA 93, 3870-3875 33. Rep, M., Krantz, M., Thevelein, J. M., and Hohmann, S. (2000) J. Biol. Chem. 275, 8290-8300 34. Kondo, K., and Inouye, M. (1991) J. Biol. Chem. 266, 17537-17544 35. Kondo, K., and Inouye, M. (1992) J. Biol. Chem. 267, 16252-16258 36. Kowalski, L. R., Kondo, K., and Inouye, M. (1995) Mol. Microbiol. 15, 341-353 37. Stukey, J. E., McDonough, V. M., and Martin, C. E. (1990) J. Biol. Chem. 265, 20144-20149 38. Inoue, Y., and Kimura, A. (1996) J. Biol. Chem. 271, 25958-25965 39. Albertyn, J., Hohmann, S., Thevelein, J. M., and Prior, B. A. (1994) Mol. Cell. Biol. 14, 4135-4144 40. Bell, W., Klaassen, P., Ohnacker, M., Boller, T., Herweijer, M., Schoppink, P., Van der Zee, P. and Wiemken, A. (1992) Eur. J. Biochem. 209, 951-959 41. Praekelt, U. M., and Meacock, P. A. (1990) Mol. Gen. Genet. 223, 97-106 42. Sahara, T., Goda, T., and Ohgiya, S. (2002) J. Biol. Chem. 277, 50015-50021 43. Zhang, L., Ohta, A., Horiuchi, H., Takagi, M., and Imai, R. (2001) Biochem. Biophys. Res. Commun. 283:531-535 44. Kandror, O., Bretschneider, N., Kreydin, E., Cavalieri, D., and Goldberg, A. L. (2004) Mol. Cell. 13, 771-781 45. Vigh, L., Maresca, B., and Harwood, J. L. (1998) Trends Biochem. Sci. 23, 369-374 46. Horvath, I., Glatz, A., Varvasovszki, V., Torok, Z., Pali, T., Balogh, G., Kovacs, E., Nadasdi, L., Benko, S., Joo, F., and Vigh, L. (1998) Proc. Natl. Acad. Sci. USA 95, 3513-3518 47. Los, D. A., and Murata, N. (2004) Biochim. Biophys. Acta 1666, 142-157 48. Perraud, A. L., Weiss, V., and Gross, R. (1999) Trends Microbiol. 7, 115-120 49. Browse, J., and Xin, Z. (2001) Curr. Opin. Plant Biol. 4:241-246 50. Reiser, V., Raitt, D. C., and Saito, H. (2003) J. Cell. Biol. 161, 1035-1040 51. Bolard, J. (1986) Biochim. Biophys. Acta 864, 257-304 52. Szalontai, B., Nishiyama, Y., Gombos, Z., Murata, N. (2000) Biochim Biophys Acta 1509, 409-419 53. Inaba, M., Suzuki, I., Szalontai, B., Kanesaki, Y., Los, D. A., Hayashi, H., Murata, N. (2003) J Biol Chem. 278, 12191-12198 54. van der Heide, T., Stuart, M. C., and Poolman, B. (2001) EMBO J. 20, 7022-7032 55. Tokishita, S., and Mizuno, T. (1994) Mol. Microbiol. 13, 435-444 56. van der Heide, T., and Poolman, B. (2000) Proc. Natl. Acad. Sci. USA 97, 7102-7106 57. Lopez, C. S., Heras, H., Garda, H., Ruzal, S., Sanchez-Rivas, C., and Rivas, E. (2000) Int. J. Food Microbiol. 55, 137-142 58. Laroche, C., Beney, L., Marechal, P. A., and Gervais, P. (2001) Appl. Microbiol. Biotechnol. 56, 249-254 59. Estruch, F. (2000) Stress-controlled transcription factors, stress-induced genes and stress tolerance in budding yeast. FEMS Microbiol. Rev. 24, 469-486

10

60. Alepuz, P. M., Jovanovic, A., Reiser, V., and Ammerer, G. (2001) Mol. Cell. 7, 767-777 61. Alepuz, P. M., de Nadal, E., Zapater, M., Ammerer, G., and Posas, F. (2003) EMBO J. 22, 2433-2442 62. Wolfe, J., and Bryant, G. (1999) Criobiology 39, 103-129 63. Luyten, K., Albertyn, J., Skibbe, W. F., Prior, B. A., Ramos, J., Thevelein, J. M., and Hohmann, S. (1995) EMBO J. 14, 1360-1371 64. Tamas, M. J., Luyten, K., Sutherland, F. C., Hernandez, A., Albertyn, J., Valadi, H., Li, H., Prior, B. A., Kilian, S. G., Ramos, J., Gustafsson, L., Thevelein, J. M., and Hohmann, S. (1999) Mol. Microbiol. 31, 1087-1104 65. Izawa, S., Ikeda, K., Maeta, K., and Inoue, Y. (2004) Appl. Microbiol. Biotechnol. 66, 303-305 66. Randez-Gil, F., Aguilera, J., Codón, A., Rincón, A. M., Estruch, F., and Prieto, J. A. (2003) Functional Genetics of Industrial Yeasts, de Winde, J. H. (Ed.), Springer-Verlag, Heidelberg, pp. 57-97

FOOTNOTES

* This research was funded by the Comisión Interministerial de Ciencia y Tecnología project (AGL2001-1203) from the Ministry of Science and Technology of Spain. 1 The abbreviations used are: DMSO, dimethyl sulfoxide; MAPK, mitogen-activated protein kinase; MAPKK, MAPK kinase; HOG pathway, High Osmolarity Glycerol pathway; S. pombe, Schizosaccharomyces pombe; SAPK, stress-activated protein kinase; JNK, c-Jun N-terminal kinase; S. cerevisiae, Saccharomyces cerevisiae; YNB, yeast nitrogen base; YPD, yeast extract/peptone/dextrose; E. coli, Escherichia coli; LB, Luria Bertani; GFP, green fluorescent protein; DEPC, diethyl pyrocarbonate; PCR, polymerase chain reaction; MIPS, munich information center for protein sequences; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; BSA, bovine serum albumin; TBST, tris-buffered saline tween-20; ECL, enhanced chemiluminescence; DAPI, 4’,6-diamidin 2-phenyldiol; L. lactis, Lactococcus lactis; B. subtilis, Bacillus subtilis

FIGURE LEGENDS

FIG. 1. A downward shift in temperature triggers the phosphorylation of Hog1p. Cells of the BY4741 wild-type and corresponding mutant strains, pbs2∆, ssk1∆ and sho1∆ were grown in YPD at 30ºC until early-exponential phase (OD600 = 0.4-0.6) and then cold-shocked at 12ºC. At the indicated times, cells were harvested, processed and the crude protein extracts analyzed by Western blot. Cells grown at 30ºC (time 0) and 0.5 M NaCl-stressed cells (NaCl) were used as control. Total protein extracts were analyzed by SDS-PAGE and blotting with anti-phospho-p38 antibody, which cross-reacts with the dual phosphorylated form of Hog1p (P-Hog1p). The membrane was then re-blotted with anti-Hog1p antibody as loading control (Hog1p). The graphic represents the phospho-Hog1p relative levels in cold-shocked cells of the wild-type strain. Values of phospho-Hog1p were normalized with respect to the total Hog1p corresponding values. The highest relative value was set at 100. For more details see the Experimental Procedures section. A representative experiment is shown. Independent experiments revealed similar kinetics and phospho-Hog1p values. FIG. 2. Exposure to DMSO induces the phosphorylation and nuclear import of Hog1p. A, The level of phospho-Hog1p (P-Hog1p) was examined in cells of the BY4741 wild-type (wt), pbs2∆, sho1∆ and ssk1∆ strains exposed to 250 mM DMSO (final concentration). YPD-cultures were grown until early-exponential phase and then transferred to the same medium containing DMSO. At the indicated times aliquots were withdrawn for protein extraction and analysis by Western blot. Phospho-Hog1p (P-Hog1p) and total Hog1p (Hog1p) were visualized as described in Fig 1. B, Wild-type cells of the W303-1A strain were transformed with the plasmid pVR65 expressing a Hog1-GFP protein fusion and grown on galactose as carbon source to mid-exponential phase (OD600 = 0.3-0.4). DMSO was added to 250 mM (final concentration), and samples were taken at the indicated times for Western blot and fluorescence microscopy. DAPI (2.5 µg/ml) was added to the culture 1 hour before DMSO treatment. Cells were processed for fluorescence microscopy as described in the Experimental Procedures section. Cells were directly observed and photographed. DAPI controls are not shown.

11

Time 0 corresponds to untreated cells. Phospho-Hog1p (P-Hog1p) and total Hog1p (Hog1p) were visualized as described in Fig 1. In each case, a representative experiment is shown. FIG. 3. The activity of the HOG pathway has no effects on growth at low temperature. Cells of the wild-type BY4741 and corresponding mutant strains hog1∆, pbs2∆, sho1∆, and ssk1∆ were assayed for growth at low temperature. Cultures were grown in YPD at 30ºC until early-exponential phase and adjusted to OD600 = 0.3. Serial dilutions (1-10-4) of the adjusted cultures were spotted (3 µl) onto YPD plates and incubated at 30ºC and 12ºC for 2 or 6 days, respectively. FIG. 4. The activity of Hog1p modulates the transcriptional response to cold-shock. RNA samples from cells of the wild-type BY4741 (wt) and hog1∆ mutant strains were analyzed by Northern blot. YPD-cultures were grown at 30ºC (DO600 = 0.4-0.6) and then transferred to 12ºC for the indicated times. Filters were probed for NSR1, TIP1 and OLE1 (A) or GLO1, TPS1, HSP12, GRE1 and GPD1 (B) mRNA. Graphs represent quantification of the mRNA levels of each gene relative to those of ACT1 in wt (●) or hog1∆ (○) samples. Cells grown at 30ºC (time 0) were used as control. Levels of ACT1 mRNA are shown in panel B. A representative experiment is shown. FIG. 5. Glycerol accumulates in cold-shocked cells. Cell cultures of the wild-type BY4741 strain (●) and corresponding hog1∆ (○) mutant were grown in YPD at 30ºC (DO600 = 0.4-0.6) and transferred to 12ºC or 4ºC. At the indicated times, aliquots of the cultures were withdrawn, filtered and the cell pellet processed for glycerol assay. Values represent the means of at least two independent experiments. The error associated with the points was calculated by using the formula: (1.96 × SD) / √n, where n is the number of measurements. dw, dry weight. FIG. 6. Hog1p is essential for freezing tolerance in S. cerevisiae. Cell cultures of the wild-type BY4741 strain (●) and corresponding hog1∆ (○) and gpd1∆ (▲) mutant strains were grown at 30ºC (DO600 = 0.4-0.6) and transferred directly to -20ºC (A) or pre-incubated at 4ºC for 48 h, prior the shift to -20ºC (B). After 3 h, 1 or 4 days, cell samples were thawed at 30ºC for 30 min, diluted and plated onto solid YPD and the percent of survival was determined. Values represent the means of at least three independent experiments. The error associated with the points was calculated by using the formula: (1.96 × SD) / √n, where n is the number of measurements.

12

TABLE I

Saccharomyces cerevisiae strains used in this study

Strain

Genotype

Reference or source

W303-1A MATa ade2-1 his3-11,15 leu-2-3,112 trp1-1 ura3-1 can1-100 GAL mal SUC2 (23)

W303 hog1∆ MATa hog1∆::TRP1 (24)

BY4741 MATa his3∆1 leu2∆0 met15∆0 ura3∆0 EUROSCARF

BY4741 hog1∆ MATa hog1∆::kanMX4 EUROSCARF

BY4741 pbs2∆ MATa pbs2∆::kanMX4 EUROSCARF

BY4741 sho1∆ MATa sho1∆::kanMX4 EUROSCARF

BY4741 ssk1∆ MATa ssk1∆::kanMX4 EUROSCARF

ssk1∆sho1∆

0 5 10 30 0 5 10 30

pbs2∆

P-Hog1p

Hog1p

0 5 10 30Time (min)

Time (min)

Hog1p

0 5 10 30 60

P-Hog1p

5

NaCl

0 20 40Time (min)

020406080

100

P-H

og1p

rela

tive

leve

lwtBY4741

Cold

60

ssk1∆sho1∆

0 5 10 30 0 5 10 30

pbs2∆

P-Hog1p

Hog1p

0 5 10 30Time (min)

Time (min)

Hog1p

0 5 10 30 60

P-Hog1p

5

NaCl

0 20 40Time (min)

020406080

100

P-H

og1p

rela

tive

leve

lwtBY4741

Cold

60

Fig. 1 Panadero et al.

13

ssk1∆

0 1 3 6 12

sho1∆

P-Hog1p

Hog1p

Time (min)

Time (min)

Hog1p

P-Hog1p

DMSO

0 1 3 6 12 pbs2∆

0 1 3 6 12

B

0 10 30

A

0 5 10 30Time (min)

P-Hog1p

Hog1p

0 1 3 6 12

wtBY4741

wtW303-1A

Time (min)

DMSO

ssk1∆

0 1 3 6 12

sho1∆

P-Hog1p

Hog1p

Time (min)

Time (min)

Hog1p

P-Hog1p

DMSO

0 1 3 6 12 pbs2∆

0 1 3 6 12

B

0 10 30

A

0 5 10 30Time (min)

P-Hog1p

Hog1p

0 1 3 6 12

wtBY4741

wtW303-1A

Time (min)

DMSO

Fig. 2 Panadero et al.

14

wt

hog1∆

pbs2∆

sho1∆

ssk1∆

30ºC 12ºCYPD

wt

hog1∆

pbs2∆

sho1∆

ssk1∆

30ºC 12ºCYPD

Fig. 3 Panadero et al.

15

NSR1

wt

hog1∆

TIP1 OLE1

Time (hours)

Time (hours)

0 0.5 1 2 4 8 0 0.5 1 2 4 8

GLO1 TPS1

HSP12

wt

hog1∆

A

B

GRE1 GPD1

ACT1

wt

hog1∆

wt

hog1∆

GLO1 TPS1

GRE1 GPD1

HSP12

0 0.5 1 2 4 80

20406080

100

020406080

100

020406080

100

Relative mRNA level (%)

0 0.5 1 2 4 80 0.5 1 2 4 8

NSR1 TIP1 OLE1Relative mRNA level (%)

0 0.5 1 2 4 8 0 0.5 1 2 4 8 0 0.5 1 2 4 8

0 0.5 1 2 4 8 0 0.5 1 2 4 8 0 0.5 1 2 4 80

20406080

100

Time (hours)

NSR1

wt

hog1∆

TIP1 OLE1

Time (hours)

Time (hours)

0 0.5 1 2 4 8 0 0.5 1 2 4 8

GLO1 TPS1

HSP12

wt

hog1∆

A

B

GRE1 GPD1

ACT1

wt

hog1∆

wt

hog1∆

GLO1GLO1 TPS1TPS1

GRE1GRE1 GPD1GPD1

HSP12HSP12

0 0.5 1 2 4 80

20406080

100

020406080

100

020406080

100

Relative mRNA level (%)

0 0.5 1 2 4 80 0.5 1 2 4 8

NSR1NSR1 TIP1TIP1 OLE1OLE1Relative mRNA level (%)

0 0.5 1 2 4 8 0 0.5 1 2 4 8 0 0.5 1 2 4 8

0 0.5 1 2 4 8 0 0.5 1 2 4 8 0 0.5 1 2 4 80

20406080

100

Time (hours)

Fig. 4 Panadero et al.

16

Time (hours)

0

10

20

30

40

50

0 10 20 30 40 50

0

5

10

15

20

25

0 5 10 15 20 25

Gly

cero

l(µg

/mg

ofce

lls, d

w)

12ºC

4ºC

Time (hours)

0

10

20

30

40

50

0 10 20 30 40 50

0

5

10

15

20

25

0 5 10 15 20 25

Gly

cero

l(µg

/mg

ofce

lls, d

w)

12ºC

4ºC

Fig. 5 Panadero et al.

17

020406080

100

0 1 2 3 4

Time (days)

Rel

ativ

esu

rviv

al(%

)

0 1 2 30

2040

6080

100

30ºC -20ºC

4

30ºC 4ºC -20ºC

020406080

100

0 1 2 3 4

Time (days)

Rel

ativ

esu

rviv

al(%

)

0 1 2 30

2040

6080

100

40 1 2 30

2040

6080

100

30ºC -20ºC

30ºC 4ºC -20ºC

4

Fig. 6 Panadero et al.

18