Introduction

The transcription of a gene is initially controlled by its respective core promoter. Repressible promoters have been used to study the function of essential genes and to modulate the expression of target genes in biotechnological processes (Delic et al. 2013). In Saccharomyces cerevisiae, the promoters of MET3, CUP1, and PHO5 have been used to downregulate various target genes. The MET3 promoter is repressed by methionine (Mao et al. 2002), while the CUP1 promoter is regulated by Cu2+ (Labbe and Thiele 1999), and the PHO5 promoter is responsive to inorganic phosphate (Rudolph and Hinnen 1987). The MET3 gene encodes ATP sulfurylase, the first enzyme involved in the sulfate assimilation and methionine biosynthetic pathway. It catalyzes the production of adenosine 5′-phosphosulfate (APS) from inorganic sulfate and ATP (Marzluf 1997; Thomas and Surdin-Kerjan 1997). APS is a high-energy molecule required for sulfate activation and reduction in the cell, and thus ATP sulfurylase plays a crucial role in the sulfur amino acid biosynthetic pathway when inorganic sulfate is used as a sole sulfur source (Ullrich et al. 2001). In S. cerevisiae, MET3 expression is repressed by exogenous methionine and S-adenosyl methionine (Cherest et al. 1985; Thomas et al. 1989). Due to its strictly controlled property, the S. cerevisiae MET3 promoter has been widely used to generate conditional lethal strains for functional studies as well as to downregulate the expression of target genes for biotechnological processes (Mao et al. 2002; Asadollahi et al. 2008). The MET3 promoters of other yeast species, such as Ashbya gossypii, Candida albicans, and Pichia pastoris, have been also evaluated as a useful molecular tool for regulated gene expression (Asadollahi et al. 2008; Care et al. 1999; Delic et al. 2013; Dunkler and Wendland 2007; Mao et al. 2002).

The thermotolerant, methylotrophic yeast Hansenula polymorpha is characterized by its high tolerance to various stressors, such as heavy metals, oxidants, xenobiotics, and environmental pollutants. It has attracted attention as an industrial yeast strain for various biotechnological applications, including heterologous protein production, bioconversion, environmental remediation, and glutathione production (Bartelsen et al. 2002). In addition, it has been extensively used as a model organism for investigating peroxisome biology and methanol metabolism (van Zutphen et al. 2010). The methanol-inducible promoters of the MOX and FMD genes, which encode alcohol oxidase and formate dehydrogenase, respectively, have been employed in the high-level expression of heterologous genes in H. polymorpha (Gellissen and Hollenberg 1997, 1999). The heat-inducible TPS1 promoter (Amuel et al. 2000), the nitrogen-regulated promoters of nitrate and nitrite utilization genes (Brito et al. 1999), and the maltose- and sucrose-inducible, MAL1 promoter that can also be repressed with glucose (Alamae et al. 2003), have all been used to express various genes in H. polymorpha. In the current study, we investigated the characteristics of the HpMET3 promoter, and showed that it could be another tool that is suitable for modulating gene expression in H. polymorpha.

Materials and methods

Yeast strains, plasmids, primers, and culture media

The yeast strains used in this study are listed in Table 1. All plasmids and primers used in this study are listed in Supplementary Tables 1 and Table 2, respectively. Yeast cells were cultivated in YPD medium [1 % (w/v) yeast extract, 2 % (w/v) Bacto-peptone, and 2 % (w/v) glucose], synthetic defined (SD) medium [0.67 % (w/v) yeast nitrogen base without amino acid and 2 % (w/v) glucose], or sulfur-free B-medium (synthetic medium with 2 % glucose without any sulfur source) (Cherest and Surdin-Kerjan 1992). Inorganic (sulfate; 40 mM final concentration) and organic (methionine, cysteine; 0.1, 0.5, or 1 mM) sulfur sources were added to SD or B-medium. Uracil (20 mg/ml), leucine (100 mg/ml), lysine (30 mg/ml), and histidine (20 mg/ml) were added to growth media according to the auxotrophic requirements of individual strains.

Table 1 Yeast strains used in this study
Full size table

Construction of MET3 promoter-based plasmids

To construct pDLM3P600-lacZ, the H. polymorpha expression vector for lacZ under the control of HpMET3 promoter, a 0.6 kb DNA fragment corresponding to the promoter region of HpMET3 was obtained by PCR amplification with primers MET3p600_3F and MET3p2B_RNC from the genomic DNA of the DL1-L strain and digested with ClaI and SalI. The HpMET3 promoter fragment was inserted into the ClaI/SalI-digested pDLMOX-yEGFPm expression vector, which contains the selectable marker HpLEU2, to generate pDLM3P600. The EcoRI/NsiI fragment of lacZ, encoding β-galactosidase, generated from pZAM522, was introduced into pDLM3P600, yielding pDLM3P600-lacZ. To construct the pDLMP600-PMT2 plasmid, the H. polymorpha expression vector for PMT2 under the control of HpMET3 promoter, the 0.7 kb HpPMT2 fragment was amplified by PCR using primers HpPMT2_11F_RI and HpPMT2_12B_StuI, and cloned between the EcoRI and StuI sites of pDLM3P600.

To generate PScMET3-ScERG7(L), a 0.6 kb DNA fragment corresponding to the promoter of ScMET3 and a 0.8 kb fragment of the ScERG7 gene were amplified from the genomic DNA of the BY4742 strain. We used the primers SspI BglII pScMET3 1F and pScMET3 ATG EcoRI 2B to amplify the 0.6 kb ScMET3 promoter, and primers EcoRI ScERG7 1F and ScERG7 SalI 2B to amplify the 0.8 kb ScERG7 fragment. The amplified ScMET3 promoter fragment was digested with BglII and EcoRI, while the ScERG7 fragment was digested EcoRI and SalI. The ScMET3 promoter and ScERG7 fragments were then inserted between the BamHI and SalI sites of pBluescript SK(+), to generate pBluSKP-PScMET3-ScERG7. The SmaI/HpaI fragment containing the ScLEU2 gene was generated from YEp351GAPII and ligated into NaeI-digested pBluSKP-PScMET3-ScERG7, yielding PScMET3-ScERG7(L). Using the same strategy, plasmid PHpMET3-ScERG7(L) was constructed; however, the 0.6 kb HpMET3 promoter obtained from pDLM3P600 via digestion with BamHI and EcoRI, was used instead of the ScMET3 promoter. To construct YCpU-SM3P600-lacZ, a S. cerevisiae expression vector for lacZ under the control of the ScMET3 promoter, the BglII/EcoRI-digested 0.6 kb ScMET3 promoter obtained from PScMET3-ScERG7(L) and the EcoRI/NsiI-digested lacZ fragment from pDLM3P600-lacZ were ligated with the BamHI/NsiI-digested YCpU, a S. cerevisiae centromeric plasmid containing the URA3 marker, yielding YCpU-SM3P600-lacZ.

Generation of MET3 promoter-based conditional mutants

To generate HpMET3(p)-Hppmt2 strains in which the native PMT2 promoter was replaced with the HpMET3 promoter, the H. polymorpha wild-type (WT) DL1-L strain was transformed with the ClaI-linearized pDLMP600-PMT2. Transformants were selected on synthetic complete medium lacking leucine (SC-LEU). Replacement of the HpMET3 promoter with the native HpPMT2 promoter was confirmed by PCR using primers MET3p600_3F and HpPMT2_2B_AscI_50. For the HpMET3(p)-ERG7 and ScMET3(p)-ERG7 strains, the S. cerevisiae BY4742 strain was transformed with an NsiI-linearized PScMET3-ScERG7(L) and PHpMET3-ScERG7(L) to replace the native ERG7 promoter with the promoters of ScMET3 or HpMET3, respectively. The correct replacement of the ScMET3 promoter and the HpMET3 promoter was confirmed by PCR using specific primer pairs (pSM3-SE7integ5F and pSM3-SE7integ4B, and MET3p600_3F and pSM3-SE7integ2B, respectively).

β-Galactosidase activity assay

Yeast cells were harvested and cell lysates were assayed for β-galactosidase activity. The assay was carried out as previously described (Rose and Botstein 1983) and the units of β-galactosidase activity were calculated according to Miller units (Miller 1972): 1 unit 1000 × A420/time (min) × volume (ml) × A600. The control is the transformants bearing the backbone vector without lacZ gene.

Quantitative PCR (qPCR) analysis

S. cerevisiae was grown at 30 °C and H. polymorpha at 37 °C in SD medium to the early growth phase (OD600 = 0.3). The medium was then changed to SD medium without any supplementation or SD medium supplemented with methionine or cysteine. After incubation for 2 or 3 h, cells were harvested and total RNA was isolated using the hot phenol extraction method (Chen et al. 2003). Total RNA was treated with DNase I (Fermentas), and cDNA synthesized from 1 μg total RNA using RnaUsScript reverse transcriptase (LeGene Biosciences) and oligo(dT) primers. The resulting cDNA samples were used as templates in qPCR assay. Each amplification reaction contained 10 μl SYBR Premix Ex Taq (Takara) and 10 pmol of each forward and reverse oligonucleotide primer (Supplementary Table 2).

Results and discussion

Tight repression of the HpMET3 promoter by cysteine

We previously observed that the expression of a set of H. polymorpha genes involved in sulfur metabolic pathway, including the HpMET3 gene, was significantly repressed by exogenous cysteine, but only moderately repressed by methionine (Sohn et al. 2014). In silico analysis of the HpMET3 promoter revealed that the structure of the HpMET3 promoter is distinguishable from that of the ScMET3 promoter, with respect to the number and the location of binding sites for transcription factors. Some of these transcription factors included Cbf1p, Met31p/Met32p, Gcn4p, Msn2p/Msn4p, and Yap1p (Fig. 1a). The ScMET3 promoter contains three biding sites for Cbf1p, one for Met31p/Met32p, four for Gcn4p, one for Msn2p/Msn4p, and three for Yap1p. In contrast, the HpMET3 promoter contains a single Cbf1 binding site and four Gcn4p binding sites. The organization of these two promoters differs with respect to the Cbf1p and Met31p/Met32p binding sites, which interact with the sole transcriptional activator of sulfur metabolism, Met4p, and recruit the Met4p transcriptional complex to target promoters (Thomas and Surdin-Kerjan 1997).

Fig. 1
figure 1

HpMET3 and ScMET3 promoter analysis. a In silico analysis of the HpMET3 and ScMET3 promoters. Transcription factor binding sites were identified using RSAT (http://fungi.rsat.eu/) and Yeastract (http://www.yeastract.com/). b Activity analysis of the H. polymorpha MET3 promoter fused to lacZ in the presence of different sulfur sources. Two H. polymorpha transformants (#2 and #3) containing the HpMET3(p)-lacZ expression vector were individually grown in sulfur-free B medium (B), with 0.1 mM cysteine (Cys), 0.1 mM methionine (Met), cysteine and methionine (MetCys) or 40 mM ammonium sulfate (sulfate). (C) Activity analysis of the S. cerevisiae MET3 promoter fused to lacZ in S. cerevisiae in the presence of different sulfur sources. Two S. cerevisiae transformants (#2 and #4) containing the ScMET3(p)-lacZ expression vector were individually grown in sulfur-free B medium (B), with 0.1 mM cysteine (Cys), 0.1 mM methionine (Met), cysteine and methionine (MetCys) or 40 mM ammonium sulfate (sulfate), respectively. The β-galactosidase assay was carried out separately with each of transformants and the β-galactosidase activity presented is the average value from triplicate assays. The error bars represent the standard deviation from triplicate measurements

Full size image

Serial deletion analysis of the HpMET3 promoter fused to lacZ revealed that the 0.6 kb fragment of the HpMET3 promoter, containing the Cbf1 and Gcn4p binding sites, retains its regulation pattern to be strongly repressed by cysteine (Supplementary Fig. 1A). The repression of the 0.6 kb HpMET3 promoter by cysteine was shown to be concentration-dependent (Supplementary Fig. 1B). The HpMET3 promoter exhibited a gradual repression by the increase of cysteine concentration. In particular, the promoter activity appeared to be completely repressed by 0.1 mM cysteine, indicating that 0.1 mM might be the minimal dose of cysteine required to repress efficiently the HpMET3 promoter.

Transformants of H. polymorpha expressing β-galactosidase under the control of the 0.6 kb HpMET3 promoter were cultivated in sulfur-free B medium without any supplementation and B medium supplemented with methionine, cysteine, or sulfate. The β-galactosidase activity of transformants cultured in the absence of a sulfur source was approx. 13-fold reduced by addition of 0.1 mM cysteine, but only threefold deceased by addition of 0.1 mM methionine (Fig. 1b). When we supplemented the culture medium with both methionine and cysteine, β-galactosidase activity decreased to levels similar to those observed when only cysteine was added to the medium. The addition of sulfate, an inorganic sulfur source, had little influence on the HpMET3 promoter activity.

Our results show that the expression of β-galactosidase, which is driven by the HpMET3 promoter, was almost completely repressed by low levels of cysteine, indicating that cysteine is central to downregulating the activity of the HpMET3 promoter in H. polymorpha. This is a unique feature of H. polymorpha MET3 promoter, considering that the S. cerevisiae MET3 promoter is reported to be more efficiently repressed by methionine rather than by cysteine (Mao et al. 2002). We also analyzed S. cerevisiae transformants harboring ScMET3(p)-lacZ expression vector and confirmed the more tight repression of S. cerevisiae MET3 promoter by methionine (Fig. 1c).

A conditional lethal mutant of HpPMT2 with the HpMET3 promoter

Given that the HpMET3 promoter could be repressed with cysteine, we placed an essential gene under the control of this promoter to analyze its feasibility as a tool for generating a conditional lethal mutant. We selected the HpPMT2 gene because it is thought to be essential for cell viability in H. polymorpha (Kim et al. 2013). We replaced the native promoter of HpPMT2 with the HpMET3 promoter and observed the growth of two independent conditional Hppmt2 mutant strains under various conditions (Fig. 2a, b). The growth of the conditional Hppmt2 mutant strains was severely retarded in the presence of 1 mM cysteine, indicating that HpPMT2 is essential for cell growth. We determined the transcript levels of HpPMT2 in the conditional mutant strains that were cultured in the presence of cysteine, methionine, or cysteine and methionine. In the absence of sulfur compounds, expression levels of HpPMT2 mRNAs driven by the HpMET3 promoter were fivefold higher than those driven by the native HpPMT2 promoter (Fig. 2c). In the presence of 0.5 mM cysteine, expression levels of HpPMT2 mRNAs driven by the HpMET3 promoter were decreased to around half of those driven by the native HpPMT2 promoter. The addition of methionine had little effect on mRNA expression levels. Considering that only a 50 % decrease in HpPMT2 expression was achieved by repressing the HpMET3 promoter, compared with HpPMT2 expression directed by its native promoter, the severe growth defects observed for the HpMET3(p)-Hppmt2 mutant strains are strong indicators of the vital function of HpPMT2 for cell viability.

Fig. 2
figure 2

Growth phenotype of the conditional HpMET3(p)-HpPMT2 mutant strains. a Scheme of the conditional HpPMT2 mutant strain under the control of the HpMET3 promoter. b Ten-fold serial dilutions of wild-type (WT) H. polymorpha and conditional HpPMT2 mutant strains were spotted onto SD medium with and without cysteine (1 mM Cys). c We used quantitative PCR assays to analyze the transcriptional regulation of HpPMT2 driven by the HpMET3 promoter. WT and conditional HpPMT2 mutant cells were grown in SD medium with 0.5 mM cysteine (Cys), 1 mM methionine (Met), 0.5 mM cysteine and 1 mM methionine. Cells were harvested, and total RNA extracted and then subjected to qPCR analysis. Expression levels of HpPMT2 transcripts were normalized to β-actin mRNA levels. Induced expression levels are presented as relative ratios to the expression levels of the WT cultivated in SD. The error bars represent the standard deviation from triplicate measurements

Full size image

We examined the cellular morphology of HpMET3(p)-Hppmt2 mutants, to determine the physiological status of H. polymorpha when functional HpPmt2p was depleted. Confocal microscopy revealed that wild-type (WT) cells exhibited normal cell morphology with circular shape and smooth surface when cultured in the presence of 1 mM cysteine. In contrast, a high proportion of the HpMET3(p)-Hppmt2 mutants showed rough and crumpled cell surface, with some cells having lysed (Fig. 3, right panel, indicated by arrows); thereby, reflecting defects in the maintenance of cell integrity. To examine the distribution of chitin in the cell wall, cells were stained with Calcofluor white. The cell wall along the rim was clearly seen for WT cells. However, HpMET3(p)-Hppmt2 mutant cells were stained at the cell wall along the rim but also within enlarged vacuoles (Fig. 3, left panel).

Fig. 3
figure 3

Microscopy analysis of conditional HpMET3(p)-HpPMT2 mutant strains Representative confocal microscopy images of the wild-type DL1-L strain of H. polymorpha and two conditional HpMET3(p)-HpPMT2 mutants (#2 and #3). Cultures were grown in synthetic defined (SD) medium with 1 mM cysteine. Cells were stained with Calcofluor white (blue), adjusted to 2 × 108/ml and fixed with 3.7 % (v/v) formaldehyde in a rotator for 15 min at room temperature. After washing with phosphate-buffered saline (PBS) buffer (pH 7.4) three times, cells were stained with fluorescent brightener 28 (Calcofluor white, CFW; Sigma Aldrich) at 2 mg/ml in a rotator for 5 min and washed with PBS buffer (pH 7.4). Inset pictures on the right panel show yeast cells at a higher magnification. Phase differential interference contrast (DIC) and fluorescence microscopy pictures (CFW) were taken using a Zeiss LSM700 confocal equipped with Axio Observer. Images were processed by ZEN2011 software (Zeiss). Magnification = approx. x 1000

Full size image

These results indicate that loss of HpPmt2p function, which is involved in the O-mannosylation of proteins, resulted in major defects in cell wall integrity and the possible occurrence of adverse events during cell division. Taken together, these results strongly support the potential of the HpMET3 promoter as a useful tool in generating conditional lethal mutant strains to obtain information regarding the physiological functions of an essential gene.

Incompatibility of the HpMET3 promoter in S. cerevisiae

To determine whether the HpMET3 promoter could be used as a repressible promoter in the heterologous host S. cerevisiae, we generated S. cerevisiae strains in which the ERG7 gene encoding lanosterol synthase, an essential enzyme involved in ergosterol biosynthesis (Karst and Lacroute 1977; Shi et al. 1994), was placed under the control of the HpMET3 promoter or ScMET3 promoter. Transformants were cultured in SD medium in the presence or absence of cysteine or methionine. Transformants were also cultured in rich YPD medium to determine the basal transcript levels of ERG7 driven by the two MET3 promoters under sulfur-rich condition. In the absence of any sulfur-containing amino acids, ERG7 expression levels driven by the ScMET3 promoter were sixfold higher than those driven by the native promoter. Expression levels of S. cerevisiae ERG7 under the control of the ScMET3 promoter were downregulated to about 30 % of endogenous expression levels following the addition of methionine; ERG7 expression levels were barely reduced when cysteine was added, as indicated in the expression of lacZ under the control of the ScMET3 promoter (Fig. 1c). In contrast, the HpMET3 promoter failed to drive induced expression of ERG7. The HpMET3 promoter did not appear to downregulate expression of ERG7 upon the addition of methionine or cysteine (Fig. 4).

Fig. 4
figure 4

ScERG7 expression under the control of the ScMET3 or HpMET3 promoters in S. cerevisiae Wild-type S. cerevisiae BY4742, ScMET3(p)- and HpMET3(p)-ScERG7 integrated strains were grown in YPD or synthetic defined (SD) medium supplemented with 1 mM cysteine (Cys) or 1 mM methionine (Met). Expression levels of ScERG7 transcripts were determined by qPCR and normalized to β-actin mRNA levels. Induced expression levels are presented as relative ratios to the expression levels of ScERG7 mRNAs driven by the native promoter on YPD medium. The error bars represent the standard deviation from triplicate measurements

Full size image

The incompatibility of HpMET3 promoter in S. cerevisiae could be attributed to the notable differences in promoter features between S. cerevisiae and H. polymorpha (Fig. 1a). Quite interestingly, we also observed no induced activity of the ScMET3 promoter in H. polymorpha under sulfur-limited condition, supporting incompatibility of these two yeast MET3 promoters (Supplementary Fig. 2). The H. polymorpha MET3 promoter is unique, in that it is more efficiently regulated by cysteine than methionine. Most studies regarding MET3 promoter-based regulation in S. cerevisiae, C. albicans and A. gossypii have demonstrated the preferential use of methionine as an effective repressor (Thomas and Surdin-Kerjan 1997).

Conclusion

The cysteine-repressible HpMET3 promoter can be used to downregulate expression of target genes in H. polymorpha. The HpMET3 promoter can be efficiently repressed by low levels (0.1 mM) of cysteine, but is barely modulated by methionine in H. polymorpha. These findings are in contrast to those for the methionine-repressible MET3 promoters in other yeast species. The HpMET3 promoter can be exploited to generate conditional lethal mutants of an essential gene, allowing for the study of its physiological function. In addition, this promoter could also be used to investigate downregulation of target genes in metabolic pathways, thereby manipulating the metabolic flux in H. polymorpha.