Abstract Dekkera bruxellensis is important for lambic beer fermentation but is considered a spoilage yeast in wine fermentation. We compared two D. bruxellensis strains isolated from wine and found that they differ in some basic properties, including osmotolerance. The genomes of both strains contain two highly similar copies of genes encoding putative glycerol–proton symporters from the STL family that are important for yeast osmotolerance. Cloning of the two DbSTL genes and their expression in suitable osmosensitive Saccharomyces cerevisiae mutants revealed that both identified genes encode functional glycerol uptake systems, but only DbStl2 has the capacity to improve the osmotolerance of S. cerevisiae cells. glycerol uptake, osmotolerance, non-conventional yeasts INTRODUCTION Dekkera bruxellensis is a non-conventional yeast often found in beverages and industrial fermentations. For a long time it has been regarded as a spoilage yeast, mainly in wine production or the bioethanol industry. This yeast produces not only ethanol, similarly to Saccharomyces cerevisiae, but also volatile phenols (usually unpleasant in wine) or the acetic acid that contributes to the pleasant flavour of Belgian lambic beers or fermented Kombucha tea (Schifferdecker et al.2014; Blomqvist and Passoth 2015). Dekkera bruxellensis has been shown to be resistant to hostile environments, which makes it suitable for various industrial applications and emphasizes its biotechnological potential. The ancestors of D. bruxellensis and S. cerevisiae separated before the whole-genome duplication event, but they both developed a similar fermentative lifestyle with the production of ethanol. Dekkera bruxellensis is a Crabtree-positive yeast (Blomqvist and Passoth 2015) and is able to use nitrate as a nitrogen source (de Barros Pita et al.2011). Several D. bruxellensis strains have been isolated from various niches and their fermentation capacity and specific features, mainly related to the ability to adapt to harsh and limiting environmental conditions, studied in detail (Schifferdecker et al.2014; Blomqvist and Passoth 2015). One of the studied features was the relatively high osmotolerance of D. bruxellensis. It has been shown that D. bruxellensis uses a similar approach to S. cerevisiae and many other yeast species, i.e. the induction of glycerol synthesis and intracellular accumulation upon osmotic stress. A detailed study showed the conserved role of the high osmolarity glycerol (HOG) mitogen-activated protein kinase (MAPK) pathway (Hohmann 2002), which upon activation leads to an increased expression of genes encoding the enzymes necessary for glycerol production in D. bruxellensis cells (Galafassi et al.2013). Upon osmotic stress, yeast cells not only induce the synthesis of glycerol, but they also try to accumulate glycerol from their environment using specific transporters present in their plasma membranes. Stl1 transporters mediate the active uptake of glycerol with protons and have already been characterized in detail in three different yeast species: S. cerevisiae (Ferreira et al.2005), in pathogenic Candida albicans (Kayingo et al.2009) and in osmotolerant Zygosaccharomyces rouxii (Duskova et al.2015b). In all three yeast species, Stl proteins contribute to the accumulation of glycerol and thus osmotolerance, and moreover, in Z. rouxii, which is not able to use glycerol as a source of carbon, they are involved in the regulation of intracellular pH even under non-stressed conditions (Duskova et al.2015b). In this work we compared the osmotolerance of two D. bruxellensis strains isolated from wine and chose the more osmotolerant one for the characterization of its glycerol uptake systems. The analysis revealed that the genome of D. bruxellensis contains two copies of STL genes. Their heterologous expression in S. cerevisiae (lacking its own ability to accumulate glycerol) showed that they both encode a functional glycerol uptake system. Nevertheless, only DbStl2 was able to improve the osmotolerance of S. cerevisiae cells. MATERIALS AND METHODS Strains and plasmids The Dekkera bruxellensis strains were Y879 (CBS 2499) and Y881 (CBS 2796), kindly provided by Prof. Jure Piskur and Prof. Concetta Compagno. For the heterologous expression of DbSTL genes, Saccharomyces cerevisiae strains BY4741 hog1Δ stl1Δ (Duskova et al.2015a), YSH 818 hog1Δ (derivative of W303-1A; Eberhardt and Hohmann 1995) and its derivative YSH hog1Δ stl1Δ (constructed by homologous recombination with the use of primers listed in Table S1 in the online supplementary material) were used. Escherichia coli XL1-Blue (Stratagene, La Jolla, CF) was routinely used as the host for plasmid amplification. Plasmids for the expression of DbSTL genes were generated by homologous recombination in S. cerevisiae. The primers used for DNA-fragment amplification and diagnostic PCR are listed in Supplementary Table S1. Successful cloning was verified by diagnostic PCR and sequencing. The amplified open reading frames (ORFs) were cloned into YEp352 and pGRU1 (to tag the coding sequences with the sequence of the green fluorescent protein (GFP)) vectors behind the NHA1 promoter (replacing the NHA1 ORF in pNHA1-985 and pNHA1-985GFP, respectively; Kinclova et al.2001) resulting in the pDbSTL1, pDbSTL1-GFP, pDbSTL2 and pDbSTL2-GFP plasmids. As a negative control, the empty backbone YEp352 was used for transformation. Media and growth conditions Yeast cell cultures were grown in standard liquid or solid (supplemented with 20 g L−1 agar) media: YPD (10 g L−1 yeast extract, 20 g L−1 bacto peptone, 20 g L−1 glucose); YNB (6.7 g L−1 yeast nitrogen base without amino acids, 20 g L−1 glucose) supplemented as indicated. For testing the uptake of glycerol, 20 g L−1 glucose was replaced with 10 mM glycerol. All growth experiments were performed at 30°C. For growth-curve measurements, cells were inoculated to optical density at 600 nm (OD600) = 0.02 in 100 μL aliquots of media in a 96-well microplate and cell growth was monitored in an ELx808 Absorbance Microplate Reader (BioTek Instruments, Winooski, VT) as described previously (Maresova and Sychrova 2007). For drop tests on solid media, cells were suspended in sterile water to OD600 = 1. Serial 10-fold dilutions of cell suspensions were prepared and spotted on YPD or YNB plates supplemented as indicated. All experiments were repeated at least twice and representative results are shown. For estimating the hyper- and subsequent hypo-osmotic stress survival, cells were grown in liquid YPD media to OD600 = 0.5 (step A), a 10-mL aliquot was diluted with 10 mL of YPD + 2 M KCl and incubated for 2 h (shaking at 30°C, step B). Then an aliquot of 1 mL was withdrawn, transferred to 10 mL of water and incubated for 10 min (shaking at 30°C, step C). Three 10 μL aliquots of cultures withdrawn at the end of the three steps were appropriately diluted and plated on YPD plates to estimate the colony-forming units. Average values and standard deviations were calculated from three independent biological replicates and are presented as means ± standard deviations. Fluorescence microscopy Yeast cells producing Stl proteins tagged with GFP at their C-terminus were harvested in the exponential phase of growth in YNB medium. The fluorescence signal was observed under an Olympus AX 70 microscope using a U-MWB cube with a 450–480 nm excitation filter and 515 nm barrier filter. Bioinformatic analysis The DNA sequences were retrieved from the NCBI GenBank (Benson et al.2013). Their accession numbers are DbSTL1 JX965362 and DbSTL2 AHMD01000100.1 complement (33076..34911) for the Y879 strain, and MDGX01000493.1 (50..1843) and MDGX01000058.1 (6213..8048) for the DbSTL1 and DbSTL2 of the Y881 strain, respectively. The multiple alignments were calculated by MUSCLE algorithm (Edgar 2004). The protein topology was predicted by the TMHMM2.0 algorithm (Krogh et al.2001) and the Protter tool was used for the protein structure visualization (Omasits et al.2014). RESULTS AND DISCUSSION Two D. bruxellensis strains differ in osmotolerance To assess the general osmotolerance of D. bruxellensis and its ability to survive a hyperosmotic shock followed by a hypo-osmotic shock, two strains were used, both originating from wine production: Y879 (CBS 2499) in France and Y881 (CBS 2796) in Germany. To distinguish the general osmotic stress from the salt stress and from sodium toxicity, three different compounds were used to compare the effects of long-term exposure to high osmotic conditions (KCl, NaCl and sorbitol). Though both strains grew similarly under non-stressed conditions (controls, Fig. 1A left panel), the Y879 strain was much more tolerant to all three compounds than Y881 (Fig. 1A), and the difference was proportional to the strength of osmotic stress. With a mild stress (e.g. 0.5 M KCl or 1 M sorbitol, Fig. 1B), the effect of stress on the strains’ growth rates was almost identical, but with higher salt or sorbitol concentrations (e.g. 1 M KCl or 1.5 M sorbitol, Fig. 1A), the growth of Y881 was more affected than the growth of Y879. When we estimated the number of cells surviving two subsequent shocks (hyper- and hypo-osmotic), we saw again a higher sensitivity of Y881 to hyperosmotic shock compared to Y879 (20 vs. 60% survival after 2 h in 1 M KCl; Fig. 1C). Surprisingly, the subsequent hypo-osmotic shock (20 min in water) did not kill a statistically significant number of Y881 or Y879 cells. This resistance to a hypo-osmotic stress (after the hyperosmotic growth conditions) is unique for D. bruxellensis. For other species such as Z. rouxii, S. cerevisiae and various non-cerevisiae Saccharomyces, the transfer from hyper- to hypo-osmotic conditions affects the cell survival significantly (Duskova et al.2015a,b; Zemancikova et al.2018). Altogether, comparison of the two Dekkera strains isolated from wine showed significant differences in their osmotolerance. Figure 1. View largeDownload slide Tolerance to osmotic stress. (A) Tolerance to osmotic stress on YPD plates. (B) Growth in liquid media estimated as OD after 41 h of cultivation. YPD, white columns; YPD + 0.5 M KCl, grey columns; YPD + 1 M sorbitol, black columns. The average results of three independent experiments are presented as mean ± standard deviation. (C) Survival of subsequent hyperosmotic (YPD + 1 M KCl for 2 h) and hypo-osmotic (H2O for 10 min) shocks. White columns, control without shocks; grey columns, after hyperosmotic shock; black columns, after hypo-osmotic shock. The average results of three independent experiments are presented as mean ± standard deviation. Figure 1. View largeDownload slide Tolerance to osmotic stress. (A) Tolerance to osmotic stress on YPD plates. (B) Growth in liquid media estimated as OD after 41 h of cultivation. YPD, white columns; YPD + 0.5 M KCl, grey columns; YPD + 1 M sorbitol, black columns. The average results of three independent experiments are presented as mean ± standard deviation. (C) Survival of subsequent hyperosmotic (YPD + 1 M KCl for 2 h) and hypo-osmotic (H2O for 10 min) shocks. White columns, control without shocks; grey columns, after hyperosmotic shock; black columns, after hypo-osmotic shock. The average results of three independent experiments are presented as mean ± standard deviation. Genomes of both strains contain two STL genes One of the reasons for the higher osmotolerance of the Y879 strain could be a better intracellular accumulation of glycerol as the main osmoprotectant of yeast cells. Though the existence of only one putative STL1 gene in the genome of CBS 2499 (Y879 in our work) was reported (Piskur et al.2012; Galafassi et al.2013), we found two homologous copies of the ScSTL1 in the available genome sequence of the more osmotolerant D. bruxellensis Y879 strain, which could be the reason for the relatively high osmotolerance of this strain. Nevertheless, a detailed search in the genome databases revealed that the Y881 strain also contains two highly similar STL genes. We named the identified sequences DbSTL1 and DbSTL2. The STL1 genes of the two D. bruxellensis strains are 1794 nt long and they differ in 10 nucleotides resulting in only one change in protein sequence (amino acid residue 305 in the 6th transmembrane domain—Ala in Y879 and Val in Y881). The STL2 genes (1836 nt long) of both strains are also almost identical, with only five different nucleotides and one amino acid residue changed (amino acid 29 in the N-terminus—Phe in Y879 and Ile in Y881). DbSTL1 encodes a protein 598 amino acids long (Stl1), and the other one a slightly longer Stl2 (611 amino acids). The sequence analysis showed that both DbStl proteins have a typical structure with 12 putative transmembrane domains and N- and C-termini oriented to the cytosol (Fig. 2A), and DbStl2 having a slightly longer hydrophilic N-terminus. Comparison of DbStl protein sequences with the sequences of four previously characterized glycerol–proton symporters from other yeast species revealed that they share highly homologous transmembrane domains and connecting loops and the biggest differences can be found in the length and composition of their N- and C-termini (Fig. 2A). Accordingly, D. bruxellensis Stl proteins also contain the five conserved structural motifs (Fig. 2A) characteristic for sugar transporters and yeast glycerol–proton symporters (Leandro, Fonseca and Goncalves 2009). Figure 2B summarizes the level of identity between the Stl proteins and shows that it is higher between the two D. bruxellensis Stl proteins than between the two Z. rouxii Stl proteins (70.73 vs. 63.85%, respectively), and that D. bruxellensis Stl proteins share the highest level of identitity with C. albicans Stl1. Figure 2. View largeDownload slide Structure analysis of Stl proteins. (A) Representation of topology model and sequence conservation of Stl proteins. Consensus sequence of multiple alignment was calculated as the percentage of the modal residue per column for Stl sequences of D. bruxellensis, C. albicans, S. cerevisiae and Z. rouxii. Protein topology was predicted by the TMHMM2.0; transmembrane domains are numbered; yeast sugar-transporter conserved motifs are highlighted in orange. (B) Level of identity (%) between the primary structures of yeast Stl proteins. Figure 2. View largeDownload slide Structure analysis of Stl proteins. (A) Representation of topology model and sequence conservation of Stl proteins. Consensus sequence of multiple alignment was calculated as the percentage of the modal residue per column for Stl sequences of D. bruxellensis, C. albicans, S. cerevisiae and Z. rouxii. Protein topology was predicted by the TMHMM2.0; transmembrane domains are numbered; yeast sugar-transporter conserved motifs are highlighted in orange. (B) Level of identity (%) between the primary structures of yeast Stl proteins. The fact that both studied D. bruxellensis strains contain almost identical STL1 and STL2 genes suggests that the lower osmotolerance of the Y881 is not due to the absence or malformation of the STL gene(s) in its genome. Also the first analysis of corresponding promoter regions did not reveal significant differences. The promoter regions differ in few nucleotides and surprisingly, they apparently do not contain motifs known to be involved in upregulation upon osmotic stress in various yeast species, e.g. STRE elements. As no data related to the expression levels of the two DbSTL genes in the two studied strains are available in public databases, we plan to study and compare the regulation of expression of DbSTL1 and DbSTL2 in Y879, Y881 and also other D. bruxellensis strains (originating from different niches) under various stress conditions in the near future. Both DbStl proteins are functional in S. cerevisiae cells, but only DbStl2 improves the osmotolerance of the host cells To verify the function of both identified DbSTL genes, the genomic DNA of the more osmotolerant strain (Y879) was used for the amplification of both ORFs, which were further cloned into vectors enabling the expression and visualization of gene products in S. cerevisiae. As the host, two S. cerevisiae osmosensitive strains lacking the HOG1 and STL1 genes were used. These two strains differ in their genetic background, which is BY4741 and W303, respectively. The expression of constructed plasmids in these strains revealed that both DbStl1 and DbStl2 are correctly expressed and targeted to the plasma membrane (Fig. 3A), and a comparison of the transformants’ growth showed that both DbStl proteins improve (compared to cells transformed with the empty YEp352) the growth of S. cerevisiae hog1Δstl1Δ cells on minimal medium supplemented with standard (20 mM) KCl concentration and with 10 mM glycerol as a carbon source (Fig. 3B, left panels). If the growth medium was supplemented with 400 mM KCl, then only the presence of DbStl2 enabled the growth of cells (Fig. 3B, right panels), suggesting that DbStl1 has only a low capacity to provide cells with external glycerol as a carbon source. DbStl1 improved the osmotolerance of S. cerevisiae cells only slightly at relatively low concentrations of sorbitol (600 mM), and this improvement was surprisingly strain dependent. As shown in Fig. 3C, the BY4741-derived cells benefited from the presence of DbStl1 (right panel), but not the W303-derived cells (left panel). This difference could be caused by a different lipid composition of the plasma membrane of the two S. cerevisiae strains, which might be reflected in the folding and consequently the activity of the DbStl1. On the other hand, the presence of DbStl2 improved the growth of both S. cerevisiae strains under all the tested conditions of higher osmotic pressure. Figure 3. View largeDownload slide Expression of DbSTL1 and DbSTL2 in S. cerevisiae strains lacking their own HOG1 and STL1 genes. (A) Microscopic images of BY4741 hog1Δ stl1Δ expressing DbSTL genes tagged with GFP sequence. (B) Growth of BY4741 hog1Δ stl1Δ transformants on YNB plates without glucose supplemented with KCl and glycerol. YEp352, empty plasmid, negative control. (C) Growth of YSH hog1Δ stl1Δ and BY4741 hog1Δ stl1Δ cells on YNB plates without glucose supplemented with 600 mM sorbitol and 10 mM glycerol. Figure 3. View largeDownload slide Expression of DbSTL1 and DbSTL2 in S. cerevisiae strains lacking their own HOG1 and STL1 genes. (A) Microscopic images of BY4741 hog1Δ stl1Δ expressing DbSTL genes tagged with GFP sequence. (B) Growth of BY4741 hog1Δ stl1Δ transformants on YNB plates without glucose supplemented with KCl and glycerol. YEp352, empty plasmid, negative control. (C) Growth of YSH hog1Δ stl1Δ and BY4741 hog1Δ stl1Δ cells on YNB plates without glucose supplemented with 600 mM sorbitol and 10 mM glycerol. Pathogenic Candida albicans and osmotolerant Zygosaccharomyces rouxii also possess two copies of STL genes. Nevertheless, the predicted ability of CaStl2 to transport glycerol was not experimentally proven (Kayingo et al.2009) and although both studied ZrStl proteins mediated glycerol uptake in Z. rouxii, only one of them (ZrStl1) was functional upon expression in S. cerevisiae (Duskova et al.2015b). Our results confirm that both identified D. bruxellensis STL genes encode glycerol uptake systems that are functional upon expression in S. cerevisiae cells. SUPPLEMENTARY DATA Supplementary data are available at FEMSLE online. Acknowledgements The authors wish to thank Anna Schifferdecker, Jure Piskur and Concetta Compagno for the D. bruxellensis strains and many very fruitful discussions. FUNDING This work was supported by the Czech National Science Foundation (grant number GA CR 15-03708S) and the European Commission (grant number FP7-ITN-264717 Cornucopia). Conflict of interest. None declared. REFERENCES Benson DA, Cavanaugh M, Clark K et al. GenBank. Nucleic Acids Res 2013; 41: D36– 42. Google Scholar CrossRef Search ADS PubMed Blomqvist J, Passoth V. Dekkera bruxellensis—spoilage yeast with biotechnological potential, and a model for yeast evolution, physiology and competitiveness. FEMS Yeast Res 2015; 15: fov021. Google Scholar CrossRef Search ADS PubMed de Barros Pita W, Leite FC, de Souza Liberal AT et al. The ability to use nitrate confers advantage to Dekkera bruxellensis over S. cerevisiae and can explain its adaptation to industrial fermentation processes. Antonie Van Leeuwenhoek 2011; 100: 99– 107. Google Scholar CrossRef Search ADS PubMed Duskova M, Borovikova D, Herynkova P et al. The role of glycerol transporters in yeast cells in various physiological and stress conditions. FEMS Microbiol Lett 2015; 362: 1– 8. Google Scholar CrossRef Search ADS PubMed Duskova M, Ferreira C, Lucas C et al. Two glycerol uptake systems contribute to the high osmotolerance of Zygosaccharomyces rouxii. Mol Microbiol 2015; 97: 541– 59. Google Scholar CrossRef Search ADS PubMed Eberhardt I, Hohmann S. Strategy for deletion of complete open reading frames in Saccharomyces cerevisiae. Curr Genet 1995; 27: 306– 8. Google Scholar CrossRef Search ADS PubMed Edgar RC. MUSCLE: a multiple sequence alignment method with reduced time and space complexity. BMC Bioinformatics 2004; 5: 113. Google Scholar CrossRef Search ADS PubMed Ferreira C, van Voorst F, Martins A et al. A member of the sugar transporter family, Stl1p is the glycerol/H+ symporter in Saccharomyces cerevisiae. Mol Biol Cell 2005; 16: 2068– 76. Google Scholar CrossRef Search ADS PubMed Galafassi S, Toscano M, Vigentini I et al. Osmotic stress response in the wine yeast Dekkera bruxellensis. Food Microbiol 2013; 36: 316– 9. Google Scholar CrossRef Search ADS PubMed Hohmann S. Osmotic stress signaling and osmoadaptation in yeasts. Microbiol Mol Biol Rev 2002; 66: 300– 72. Google Scholar CrossRef Search ADS PubMed Kayingo G, Martins A, Andrie R et al. A permease encoded by STL1 is required for active glycerol uptake by Candida albicans. Microbiology 2009; 155: 1547– 57. Google Scholar CrossRef Search ADS PubMed Kinclova O, Ramos J, Potier S et al. Functional study of the Saccharomyces cerevisiae Nha1p C-terminus. Mol Microbiol 2001; 40: 656– 68. Google Scholar CrossRef Search ADS PubMed Krogh A, Larsson B, von Heijne G et al. Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes. J Mol Biol 2001; 305: 567– 80. Google Scholar CrossRef Search ADS PubMed Leandro MJ, Fonseca C, Goncalves P. Hexose and pentose transport in ascomycetous yeasts: an overview. FEMS Yeast Res 2009; 9: 511– 25. Google Scholar CrossRef Search ADS PubMed Maresova L, Sychrova H. Applications of a microplate reader in yeast physiology research. Biotechniques 2007; 43: 667– 72. Google Scholar CrossRef Search ADS PubMed Omasits U, Ahrens CH, Muller S et al. Protter: interactive protein feature visualization and integration with experimental proteomic data. Bioinformatics 2014; 30: 884– 6. Google Scholar CrossRef Search ADS PubMed Piskur J, Ling ZH, Marcet-Houben M et al. The genome of wine yeast Dekkera bruxellensis provides a tool to explore its food-related properties. Int J Food Microbiol 2012; 157: 202– 9. Google Scholar CrossRef Search ADS PubMed Schifferdecker AJ, Dashko S, Ishchuk OP et al. The wine and beer yeast Dekkera bruxellensis. Yeast 2014; 31: 323– 32. Google Scholar CrossRef Search ADS PubMed Zemancikova J, Kodedova M, Papouskova K et al. Four Saccharomyces species differ in their tolerance to various stresses though they have similar basic physiological parameters. Folia Microbiol (Praha) 2018; 63: 217– 227. Google Scholar CrossRef Search ADS PubMed © FEMS 2018. All rights reserved. For permissions, please e-mail: email@example.com
FEMS Microbiology Letters – Oxford University Press
Published: Mar 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera