OUP user menu

ETP1/YHL010c is a novel gene needed for the adaptation of Saccharomyces cerevisiae to ethanol

Christopher Snowdon, Ryan Schierholtz, Peter Poliszczuk, Stephanie Hughes, George van der Merwe
DOI: http://dx.doi.org/10.1111/j.1567-1364.2009.00497.x 372-380 First published online: 1 May 2009

Abstract

Saccharomyces cerevisiae has the ability to use a variety of different carbon sources to support its growth. Abundant fermentable sugars such as glucose and fructose are metabolized to ethanol that accumulates in the environment. Upon glucose depletion, nonfermentable carbon sources, such as ethanol and glycerol, are sufficient to support growth. However, high ethanol concentrations inhibit yeast growth and can become toxic to the cell. Here we show that YHL010c, a previously uncharacterized gene of S. cerevisiae, is needed by the yeast to adapt to ethanol, either as a sole carbon source or as a stressor. We named the gene ETP1 (Ethanol Tolerance Protein 1) and show that the etp1Δ strain has a growth defect in the presence of ethanol, ETP1 is needed for the ethanol-induced transcriptional activation of the ENA1 promoter and heat shock protein genes (HSP12 and HSP26), and plays a role in ethanol-induced turnover of the low-affinity hexose transporter Hxt3p. In addition, the hypersensitivity of etp1Δ to ethanol stress is partly due to the inability of the mutant to control the level of the cation/H+ antiporter Nha1p in the cell.

Keywords
  • ETP1/YHL010c
  • Hxt3p
  • ethanol stress
  • heat shock genes
  • Nha1p

Introduction

The ability of cells to sense and respond to its environment is critical for cellular survival and growth. Environmental stimuli are internalized via signaling mechanisms that elicit a variety of cellular responses, ranging from transcriptional adaptation to regulation of protein function and stability, to adapt appropriately. When considering its carbon metabolism, the yeast Saccharomyces cerevisiae ferments simple sugars such as glucose or fructose to ethanol when these sugars are present in abundance. However, when glucose is limited the metabolism shifts to respiration to increase energy yield. Ultimately, in the absence of glucose, the yeast uses alternative carbon sources, including ethanol and glycerol, to support its growth (Schuller, 2003). During these different stages of growth the organism adjusts its gene expression patterns and protein complement to downregulate genes and proteins specific to glucose metabolism and increase the expression and/or activity of those needed for respiration.

Although S. cerevisiae is known to tolerate high concentrations of ethanol (15% v/v), the presence of this compound in the concentration range of 4–6% (v/v) in the environment reduce the yeast's growth rate by 50% (Piper, 1995; Kubota, 2004). The intracellular ion homeostasis, cytosolic pH and the fluidity of membranes are severely affected in response to high concentrations of ethanol, resulting in inhibition of many membrane-related functions, the denaturing of proteins and ultimately decreased growth and viability (van Uden & da Cruz Duarte, 1981; Lloyd, 1993; Rose, 1993; Rosa & Sa-Correia, 1996; Sharma, 1996; Birch & Walker, 2000). Interestingly, the response of yeast to heat shock and ethanol stress is very similar. Heat shock genes, including those encoding the heat shock proteins, are induced by increasing concentrations of ethanol to cope with the intracellular stresses experienced by the yeast (Mager & Ferreira, 1993; Piper, 1995; Alexandre, 2001).

Several membrane transporters are used to maintain intracellular ion homeostasis in a variety of different conditions, including ethanol stress. The plasma membrane H+-ATPase Pma1p is the major protein needed to continually adjust the intracellular proton concentrations to maintain membrane potential and cytosolic pH (Rosa & Sa-Correia, 1991; Alexandre, 1994). In addition, cation/H+ antiporters, such as Nha1p, contributes to maintaining cytosolic pH by exchanging monovalent cations, including Na+, K+ and Li+, for H+ in the environment (Sychrova, 1999). Also, it is known that the transcription of ENA1, which encodes the major Na+-ATPase, is induced by nonfermentable carbon sources (Alepuz, 1997). Several nutrient-sensing pathways, including the PKA, TOR and Snf1 pathways, are used to control the transcriptional activation of ENA1 (Alepuz, 1997; Crespo, 2001; Ruiz & Arino, 2007).

The multicomponent Vid30 complex (Vid30c), also known as the glucose-induced protein complex (GPC), is needed for the degradation of FBPase and Mdh2p in response to glucose replenishment following starvation, and Hxt7p turnover in response to nitrogen starvation (Regelmann, 2003; Snowdon, 2008). The Vid30c therefore plays a role in the adaptation of yeast to changing nutrient conditions. The complex has been identified in several protein–protein interaction studies (Ho, 2002; Krogan, 2006; Pitre, 2006). One if its components, Moh1p, was shown to be a part of a smaller protein complex comprising itself, Yhl010c, Ykr017c and Nha1p (Ho, 2002). Moh1p, unlike other components of the GPC/Vid30c, does not seem to play a role in the degradation of FBPase following a carbon shift (Regelmann, 2003), while YHL010c and YKR017c are uncharacterized genes. Interestingly, analysis of genome-wide gene expression in respiratory growth conditions revealed that the transcription of YHL010c, MOH1 and NHA1 was activated when ethanol was the sole carbon source (Roberts & Hudson, 2006). We hypothesized that these genes were needed by the yeast to adapt to nonfermentable carbon conditions. Here we report that Yhl010c is a novel protein required by the yeast to grow with ethanol as a sole carbon source and to combat high concentrations of ethanol. We therefore renamed YHL010c as ETP1 (Ethanol Tolerance Protein 1).

Materials and methods

Strains and media

All the strains used in this work are listed in Table 1. BY4742 was considered the wild-type strain. All strains were cultured at 30 °C. Strains were constructed by the integration of linear fragments as previously described (Longtine, 1998). PCR constructs were generated using pFA6a-natMX6 (Van Driessche, 2005) as a template for generating the disruption cassettes for NHA1 and NRG1, and pFA6a-GFP(S65T)-His3MX6 (Longtine, 1998) and pFA6a-GFP(S65T)-natMX6 (Van Driessche, 2005) for generating the green fluorescent protein (GFP) cassettes for the 3′ fusion to HXT3 and NHA1, respectively. Primer sequences are provided as Supporting Information (Table S1). PCR fragments were transformed by electroporation as previously described (Ausubel, 2002). Transformants were selected on YPD [1% (w/v) yeast extract, 2% (w/v) peptone and 2% (w/v) dextrose] media containing 100 μg mL−1 nourseothricin (Werner Bioreagents) or synthetic complete (SC) media lacking histidine. Correct integration events were confirmed by PCR.

View this table:
1

Yeast strains used in this study

Yeast strainGenotype
BY4742MATαhis3Δ1 leu2Δ0 lys2Δ0 ura3Δ0
BYetp1BY4742 etp1::kanMX4
BYmoh1BY4742 moh1::kanMX4
BYykr017cBY4742 ykr017c::kanMX4
BYnha1BY4742 nha1::natMX
BYetp1 nha1BYetp1 nha1::natMX
BYnrg1BY4742 nrg1::kanMX4
BYetp1 nrg1BYetp1 nrg1::natMX
BYHXT3-GFPBY4742 HXT3-GFP::HisMX6
BYetp1 HXT3-GFPBYetp1 HXT3-GFP::HisMX6
BYNHA1-GFPBY4742 NHA1-GFP::natMX4
BYetp1 NHA1-GFPBYetp1 NHA1-GFP::natMX4
  • * Open biosystems.

  • This study.

Growth assays

For growth analyses wild-type and mutant strains were precultured in YPD to exponential growth phase, washed with sterile water and cell densities adjusted to an OD600 nm of 1.0. Tenfold dilutions were transferred to solid YPD media containing either 300 mM LiCl, 1.2 M NaCl, or increasing concentrations of ethanol [5%, 7.5% or 10% (v/v)], and YP media containing either 2.5% ethanol, 2.5% glycerol or 2.5% glycerol and 2.5% ethanol. Plates were incubated at 30 °C and images were collected to record growth. Importantly, media containing 5% ethanol were incubated 2 days longer, and media containing salt, 7.5% ethanol, or 10% ethanol were incubated 4–6 days longer than the control (YPD) media to record growth.

β-Galactosidase assays

Transformants of the wild-type and etp1Δ strains with pKC201 (Cunningham & Fink, 1996) were precultured in synthetic media [0.17% YNB without amino acids and ammonium sulfate (Difco), 0.5% ammonium sulfate, 2% dextrose and amino acids added to complement auxotrophies] to early exponential phase. Cells were harvested, washed with sterile water and transferred to YPD, YPD containing 200 mM LiCl and YP containing 2.5% ethanol. All media used in these analyses were adjusted to pH 5.5. Cultures were incubated for a further 3 h when samples were collected for β-galactosidase assays. Assays were performed as described previously (van der Merwe, 2001b). The units of activity were normalized to OD600 nm of the culture and calculated as described previously (Miller, 1972). Averages and SDs were determined from replicates of at least three independent experiments. The β-galactosidase activity supported by the wild type exposed to YP media containing 2.5% ethanol was used as 100% relative activation to which all expression values were related.

Fluorescence microscopy

Overnight cultures of BY4742 HXT3-GFP and BYetp1 HXT3-GFP were harvested, washed and transferred to SC media containing 2.5% glucose for 4 h to induce HXT3-GFP expression. Samples were collected and analyzed by fluorescence microscopy to represent time zero. The remainder of the cultures were harvested and transferred to SC media containing 2% ethanol as the sole carbon source. Samples were collected at the indicated time points and analyzed by fluorescence microscopy to follow the internalization of Hxt3-GFP.

RNA extraction and Northern analysis

The BYNHA1-GFP and BYetp1 NHA1-GFP strains were precultured in YPD to early exponential phase. Cell samples were harvested, washed and equal volumes transferred to YPD and YPD containing 7.5% (v/v) ethanol media. Following 3 h of incubation, cells were harvested and RNA extracted using the phenol–glass beads method described previously (van der Merwe, 2001a). Equal amounts of total RNA were separated by electrophoresis, transferred to a nylon membrane and fixed. The procedures for digoxigenin-labeled probe preparation, blotting, washing and detection were performed as previously described (Snowdon, 2008). Membranes were probed for GFP (NHA1-GFP), HSP12, HSP26 and ACT1 using probes prepared with the PCR Digoxigenin Probe Synthesis kit (Roche) according to the manufacturer's recommendations. Primer sequences for the probes are supplied as Table S1.

Protein extraction and Western analysis

BYNHA1-GFP and BYetp1 NHA1-GFP strains were grown in YPD to exponential phase and samples were collected for protein extraction. The remaining cells were harvested, washed and transferred to YPD containing 7.5% (v/v) ethanol and further incubated for 3 h after which cells were collected for protein extractions. The collected biomass was used to enrich membrane fractions. Cell breakage was performed as described previously (Snowdon, 2008) and the lysate subjected to differential centrifugation to remove intact cells. Following centrifugation at 15 800 g for 20 min, the pellet was used as the source of protein. Protein concentrations were determined using the methods of Bradford (Biorad). Equal amounts of protein were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to nitrocellulose membranes for Western blotting. The primary antibody used was rabbit anti-GFP (Ab290; AbCam) to detect Nha1-GFP. Secondary rabbit immunoglobulin G conjugated to horseradish peroxidase were used for detection using an ECL kit (GE) according to the manufacturer's recommendations.

Results

ETP1 is needed for the yeast's response to ethanol either as a carbon source or as a stressor

As Etp1p was previously shown to interact with Nha1p, a cation/H+ antiporter (Prior, 1996; Sychrova, 1999), Moh1p and Ykr017c, we tested whether MOH1, ETP1 or YKR017c were needed by the yeast to cope with salt stress. Actively growing wild type, moh1Δ, etp1Δ and ykr017cΔ strains were transferred to solid YPD media containing high concentrations of LiCl (300 mM) or NaCl (1.2 M). These salt stresses were analyzed at both pH 5.5 and pH 6.85 as Nha1p functions at an acidic pH (Banuelos, 1998). All the strains showed similar growth patterns in the presence of salt stress (Fig. 1a) indicating that MOH1, ETP1 or YKR017c were not needed by the yeast to tolerate the presence of excess LiCl or NaCl in the growth media.

1

ETP1 is needed for growth with and tolerance to ethanol. BY4742 (WT) and the indicated mutant strains were grown in YPD to exponential phase, cell suspensions were prepared (see Materials and methods), and spotted on to: (a) YPD and YPD containing either 300 mM LiCl (+0.3 M LiCl) or 1.2 M NaCl (+1.2 M NaCl); (b) YPD (w/o EtOH), and YPD containing either 5% ethanol (+5% EtOH), 7.5% ethanol (+7.5% EtOH) or 10% ethanol (+10% EtOH); (c) YP-based media containing 2% glucose (2% Gluc), 2.5% ethanol and 2.5% glycerol (2.5% Glyc+2.5% EtOH), 2.5% ethanol (2.5% EtOH), or 2.5% glycerol (2.5% Glyc). Plates were incubated at 30°C and photographed to record growth.

High concentrations of ethanol in the growth medium are known to compromise the plasma membrane function, negatively impact intracellular ion homeostasis and decrease cytosolic pH (van Uden & da Cruz Duarte, 1981; Lloyd, 1993; Rose, 1993; Rosa & Sa-Correia, 1996; Sharma, 1996; Birch & Walker, 2000). The growths of the wild type, moh1Δ, etp1Δ and ykr017cΔ strains were tested in YPD containing increasing concentrations of ethanol [5%, 7.5% or 10% (v/v)] as a stressor. While all the strains had similar growth patterns with glucose as the carbon source, the etp1Δ mutant had a decreased ability to grow when increasing concentrations of ethanol were added to YPD (Fig. 1b).

The transcription of ETP1, MOH1 and NHA1 was activated in response to either ethanol or glycerol as a sole carbon source (Roberts & Hudson, 2006). The growth patterns of the wild type, moh1Δ, etp1Δ and ykr017cΔ mutant strains were consequently tested with YP-based media containing ethanol and glycerol, either alone or in combination, as nonfermentable carbon sources. While the wild type, moh1Δ and ykr017cΔ strains had similar growth patterns, the etp1Δ mutant was severely inhibited when glycerol and ethanol were supplied in combination as the carbon sources (Fig. 1c). Also, ethanol as a sole carbon source had a more severe inhibition of etp1Δ growth than glycerol. In combination, these observations suggest that ETP1 is needed by the yeast for optimal growth when ethanol was present either as the sole carbon source or as a stressor in the environment.

Etp1p participates in ethanol and salt-induced transcriptional activation

The transcription of ENA1 is activated by many stimuli including increasing concentrations of Na+ or Li+ in the growth media, growth with nonfermentable carbon sources or by alkaline environments (Ruiz & Arino, 2007). We used the transcriptional activation of ENA1-lacZ (pKC201) as a reporter to study the role of Etp1p in the cell's response to LiCl or ethanol. Transformants of the wild type and etp1Δ strains with pKC201 were precultured in selective media and exposed to LiCl stress or ethanol as the sole carbon source for 3 h. While both strains had similar levels of β-galactosidase activities in the control conditions, the etp1Δ supported 27% and 48% less β-galactosidase activity than the wild type in response to LiCl treatment and 2.5% ethanol as the sole carbon source, respectively (Fig. 2a).

2

The transcriptional activation of ENA1 in response to either salt or ethanol is at least partially dependent on ETP1. (a) Transformants of the BY4742 (open bars) and etp1Δ (filled bars) strains with pKC201 were precultured in selective media and shifted to YPD (YPD), YPD containing 200 mM LiCl (YPD+200 mM LiCl), and YP media containing 2.5% ethanol (YP+2.5% EtOH). (b) Transformants of BY4742 (WT), BYetp1 (etp1Δ), BYnrg1 (nrg1Δ) and BYetp1 nrg1 (etp1Δnrg1Δ) with pKC201 were precultured in selective media and shifted to YPD (open bars) and YP media containing 2.5% ethanol (solid bars). All β-galactosidase assays were performed with three independent transformants in duplicate and activities are expressed as a relative % (see Materials and methods). Asterisks indicate statistically significant differences between control and treatment conditions as determined by Student's t-tests with P<0.05.

It is known that the Snf1 signaling pathway impacts the transcription of ENA1 in response to changing carbon conditions (Ruiz & Arino, 2007). Several transcriptional repressors, including Nrg1p, are controlled by the Snf1 pathway (Berkey, 2004; Vyas, 2005) and the transcription of ENA1 is derepressed in the nrg1Δ mutant (Vyas, 2005; Platara, 2006). We reasoned that decreased transcriptional activation of ENA1 in an nrg1Δetp1Δ strain would indicate that ETP1 affects a transcriptional activator needed to activate the ENA1 promoter. Consistent with the reported function of Nrg1p as a transcriptional repressor, ENA1 transcription was derepressed in the nrg1Δ mutant when either glucose or ethanol was the sole carbon source (Fig. 2b) (Berkey, 2004; Vyas, 2005). However, the etp1Δnrg1Δ double mutant supported only 78% and 49% of the transcriptional activation supported by the nrg1Δ mutant in YPD and YPE, respectively (Fig. 2b). Etp1p is therefore needed for the transcriptional activation of the ENA1 promoter, specifically in the presence of ethanol as the sole carbon source.

Etp1p is needed for the degradation of Hxt3p when cells are shifted from glucose to ethanol as a carbon source

Glucose in abundance activates the expression of the low-affinity hexose transporter Hxt3p (Perez, 2005). When glucose becomes limiting Hxt3p is no longer needed as the cell shifts to using the high-affinity hexose transporters. We monitored the potential involvement of Etp1p in the internalization and degradation of Hxt3p in response to a shift from glucose to ethanol as the sole carbon source. The wild-type HXT3-GFP and etp1ΔHXT3-GFP strains were grown in SC media containing 2.5% (w/v) glucose for 4 h to induce HXT3-GFP expression. Cells were shifted to SC media containing 2% (v/v) ethanol as the sole carbon source and cells were analyzed by fluorescence microscopy before (0 h) and at the indicated times after the shift to ethanol. Hxt3-GFP is clearly present on the plasma membranes of both strains at time zero and 1.5 h after the shift to ethanol (Fig. 3a). However, Hxt3-GFP was internalized in the wild-type strain after 3 h exposure to ethanol, while it was still abundantly present in the plasma membrane of the etp1Δ at the same time point. After 6 h exposure to ethanol Hxt3-GFP was not visible on the plasma membrane of the wild type, but was still clearly present in the plasma membrane of the etp1Δ (Fig. 3a). We next tested the presence of Hxt3-GFP by performing Western analysis with proteins extracted from the wild type HXT3-GFP and etp1ΔHXT3-GFP strains before (0 h) and 3 and 6 h after a shift from glucose to ethanol as the sole carbon source. It was clear that the Hxt3-GFP levels in the wild-type strain decreased significantly after 3 h and was absent after 6 h in ethanol. In contrast, this degradation was significantly delayed in the etp1Δ strain where Hxt3-GFP was still abundantly present after 3 h and clearly present after 6 h with ethanol as the sole carbon source (Fig. 3b). These observations suggest that Etp1p is needed for the appropriate internalization and subsequent degradation of Hxt3p in response to a shift from glucose to ethanol as a sole carbon source.

3

ETP1 is needed for the turnover of Hxt3p when ethanol is the carbon source. The BYHXT3-GFP (WT) and BYetp1 HXT3-GFP (etp1Δ) strains were precultured in SC media containing 2.5% glucose for 4 h. Cells were harvested, washed and shifted to SC media containing 2% ethanol as the sole carbon source. Samples were collected at the indicated times and (a) analyzed by fluorescence microscopy, and (b) proteins extracted for Western analyses using rabbit anti-GFP (Ab290, AbCam) and rabbit anti-ADH (200–4144, Rockland; loading control) primary antibodies.

Etp1p is needed for the transcriptional activation of heat shock genes in response to ethanol stress

The growth deficiency of the etp1Δ strain in ethanol stress prompted us to investigate the role of Etp1p in the transcriptional activation of genes known to be induced by ethanol stress. Included in these are the heat shock protein genes HSP12 and HSP26 (Roberts & Hudson, 2006). Exponential phase cells of the wild type and etp1Δ strains were grown in rich media with or without ethanol stress (7.5% EtOH) for 3 h. Total RNA was extracted and probed for HSP12 and HSP26. The levels of HSP26 and HSP12 mRNA in the etp1Δ strain were significantly lower than the wild-type strain after 3 h of ethanol stress (Fig. 4a). In contrast, the transcription of these genes was not affected by ETP1 deletion in the absence of ethanol stress (YPD) (Fig. 4b). These observations indicate that Etp1p is needed for the transcriptional activation of ethanol-induced genes in response to ethanol stress.

4

The transcriptional activation of heat shock genes is dependent on ETP1 in response to ethanol stress. (a) BYNHA1-GFP (WT) and BYetp1 NHA1-GFP (etp1Δ) were grown in YPD to exponential phase. Cells were harvested and shifted to fresh YPD (YPD) and YPD containing 7.5% ethanol (YPD+7.5% EtOH) for 3 h before RNA extraction. RNAs were probed for HSP26, HSP12 and ACT1 as a loading control. (b) Overexposure of (a). The bar graphs indicate the intensity of the HSP26 and HSP12 standardized to ACT1. The expression of HSP26 and HSP12 in the WT was considered as 100% expression in each condition.

The ethanol stress sensitivity of etp1Δ is relieved by deletion of NHA1

The protein–protein interaction reported between Etp1p and Nha1p (Ho, 2002) suggested that these proteins were functionally linked. To further investigate this we analyzed the growth patterns of the etp1Δ, nha1Δ and etp1Δnha1Δ strains in comparison with the wild-type strain in the presence of ethanol. Strains were precultured in YPD and spotted to ethanol-containing media that either contained or lacked glucose. Our analysis revealed that the nha1Δ, unlike the etp1Δ, did not have a growth defect with either of the carbon source combinations as it had a similar growth profile as the wild-type strain in all conditions tested (Fig. 5). Interestingly, the deletion of NHA1 in the etp1Δ strain (etp1Δnha1Δ) restored almost a wild-type level of growth in glucose media containing either 5% or 7.5% ethanol (Fig. 5). In contrast, in the absence of glucose the etp1Δnha1Δ strain had a growth pattern similar to the etp1Δ single mutant (Fig. 5). The growth defect of the etp1Δ strain in the presence of a high concentration of ethanol in glucose-containing media was therefore at least partially dependent on NHA1, suggesting that the sensitivity of the etp1Δ to ethanol stress is linked to the abundance of Nha1p in the cell.

5

NHA1 contributes to the hypersensitivity of the etp1Δ to ethanol stress. The BY4742 (WT), BYetp1 (etp1Δ) and BYetp1 nha1 (etp1Δnha1Δ) strains were grown in YPD to exponential phase, cell suspensions prepared (see Materials and methods), and spotted on to: YPD (w/o EtOH), YPD containing either 5% ethanol (+5% EtOH), 7.5% ethanol (+7.5% EtOH), YP-based media containing 2.5% glycerol (+2.5% Glyc), 2.5% glycerol and 2.5% ethanol (+Glyc/EtOH), or 2.5% ethanol (+2.5% EtOH). Plates were incubated at 30°C and photographed to record growth.

Etp1p is needed to control the levels of Nha1p in response to ethanol stress

Cells use a variety of molecular mechanisms, including the turnover of proteins, to adapt to stress conditions. We investigated the potential role of Etp1p in regulating Nha1p levels by determining the levels of Nha1-GFP in the wild-type and etp1Δ strains after these strains were exposed to ethanol stress. Fluorescence microscopy was used to analyze the intensity of Nha1-GFP at each time point. These results were inconclusive due to the low expression of NHA1 (data not shown). We consequently used Western blotting to compare the levels of Nha1-GFP in the membrane fractions of the wild-type and the etp1Δ strains. While similar levels of Nha1-GFP were detected between these strains grown in YPD, more Nha1-GFP was present in etp1Δ than the wild-type strain after exposure to ethanol stress (Fig. 6a).

6

Etp1p participates in the turnover of Nha1p in response to ethanol stress. BYNHA1-GFP (WT) and BYetp1 NHA1-GFP (etp1Δ) strains were grown in YPD to exponential phase. Cells were harvested and shifted to fresh YPD (YPD) and YPD containing 7.5% ethanol (YPD+7.5% EtOH) for 3 h before samples were collected for protein and RNA extractions. (a) Whole cell lysates were enriched for membrane fractions that were used as a source of protein. Equal amounts of protein were probed with rabbit anti-GFP antibodies (Ab290, AbCam). The blot was analyzed by densitometry to determine the relative levels of Nha1-GFP detected. Levels of Nha1p detected are provided in the graph with the wild type (open bars) in YPD taken as 100% Nha1p detected. The levels of Nha1-GFP in the etp1Δ strain are shown as solid bars. (b) RNAs were probed for GFP (NHA1-GFP) and ACT1 as a loading control.

To further understand the higher levels of Nha1-GFP in the etp1Δ strain, we analyzed the transcription of NHA1 when the wild-type and etp1Δ strains were grown in conditions of ethanol stress. Cells were precultured in YPD to exponential phase and transferred to fresh YPD lacking or containing 7.5% ethanol, and incubated for 3 h. Total RNAs were extracted and subsequent Northern analysis revealed the etp1Δ and wild-type strains supported similar levels of NHA1 transcription in all conditions tested (Fig. 6b), indicating that ETP1 is not involved in the transcriptional regulation of NHA1. In combination, these findings suggest that Etp1p function is linked to controlling the levels of Nha1p in response to ethanol stress.

Discussion

The genome of S. cerevisiae contains many genes (c. 16%) that have not yet been characterized. ETP1/YHL010c is one of these genes. We provide evidence that identify ETP1 as a novel gene needed by yeast to adapt efficiently to ethanol, either as a sole carbon source or as a cell stressor. Yeast growth in the presence of ethanol was impaired in the absence of ETP1 (Fig. 1) and the etp1Δ strain showed a decreased ability to activate the transcription of some genes in response to ethanol (Figs 2 and 4). Also, upon a shift from glucose to ethanol the turnover of some proteins, specifically Hxt3p, was delayed in the absence of ETP1 (Fig. 3). Lastly, the role of Etp1p in the yeast's ability to cope with ethanol stress is linked to the control of Nha1p levels in the cell (Figs 5 and 6).

ETP1 is needed for the ethanol-dependent transcriptional activation of several genes (Figs 2a and 4). Also, the transcriptional derepression of ENA1 in nrg1Δ in ethanol is decreased significantly in the nrg1Δetp1Δ double mutant (Fig. 2b). These observations suggest that Etp1p enables transcriptional activation directly, rather than controlling the function of a negative regulator. Etp1p could therefore either enter the nucleus to bind to the promoters of its target genes to enable activation, or the transcriptional regulators of these genes are not optimally controlled in the absence of Etp1p. We could not find any evidence of Etp1-GFP entering the nucleus with ethanol as either a sole carbon source or a stressor and searches of the Etp1p protein sequence did not reveal a DNA-binding motif (data not shown). Etp1p could therefore function in the cytoplasm to control the functional activation or nuclear entry of a specific transcription factor.

The function of Etp1p extends beyond the regulation of growth with ethanol as a sole carbon source. Cells lacking ETP1 cannot tolerate ethanol stress. The previously reported interaction of Etp1p with Nha1p (Ho, 2002) suggested that these two proteins might be functionally linked. K+ ions are known to play a role in cell cycle progression and Nha1p regulates intracellular K+ concentrations. Nha1p is known to function in regulating cell cycle progression, but, surprisingly, in a mechanism independent of controlling intracellular K+ levels (Simon, 2001, 2003). Our findings showed that Etp1p is not needed by the cell to maintain the intracellular levels of K+ during ethanol stress (data not shown). Nonetheless, Nha1p levels decreased by 68% in the wild type upon a shift to ethanol stress (Fig. 6a). However, this turnover is delayed in the absence of ETP1 as the levels of Nha1p in the etp1Δ mutant decreased by only 25% in response to ethanol stress (Fig. 6a). Also, the deletion of NHA1 in the etp1Δ mutant restored some tolerance to ethanol stress (Fig. 5). These observations suggest that the level of Nha1p in the cell needs to be tightly controlled during ethanol stress and that this control is at least partially mediated by Etp1p. In combination, the increased levels of Nha1p and the impaired induction of the heat shock genes (Fig. 4) could explain the growth deficiency of the etp1Δ in ethanol stress.

The exact molecular mechanism of action of Etp1p is currently not known. Etp1p is well conserved among eukaryotic organisms with homologs ranging from yeast to humans. Data base searches revealed several interesting structural domains in Etp1p that could provide insight into a potential mechanism of function for this protein. Etp1p contains similarity to human BRAP2, a protein known to bind to the nuclear localization signal of BRCA1 (Li, 1998; Asada, 2004), a protein linked to breast cancer (Welcsh, 2000, 2002). It contains significant homology with the C3HC4-type ring finger domain of BRAP2 and the UBP-type zinc finger found in ubiquitin hydrolases ranging from yeast to humans (Cherry, 1998). These domains collectively indicate a protein–protein interaction function for Etp1p, potentially associated with the processing of ubiquitinated proteins.

Supporting Information

Additional Supporting Information may be found in the online version of this article:

Table S1. Primers used in this study.

Acknowledgements

The authors thank Dr Kyle Cunningham for the generous gift of pKC201. This work was funded by an American Wine Society Education Fund award to C.S., a University of Guelph FRA awarded to R.S. and research funds from NSERC and OMAFRA awarded to G.M.

Authors' contribution

C.S. and R.S. contributed equally to this study.

Footnotes

  • Editor: Jens Nielsen

References

View Abstract