(C) PLOS One [1]. This unaltered content originally appeared in journals.plosone.org.
Licensed under Creative Commons Attribution (CC BY) license.
url:https://journals.plos.org/plosone/s/licenses-and-copyright

------------



Convergent adaptation of Saccharomyces uvarum to sulfite, an antimicrobial preservative widely used in human-driven fermentations

['Laura G. Macías', 'Instituto De Agroquímica Y Tecnología De Los Alimentos', 'Iata-Csic', 'Paterna', 'Departament De Genètica', 'Universitat De València', 'Valencia', 'Melisa González Flores', 'Instituto De Investigación Y Desarrollo En Ingeniería De Procesos', 'Biotecnología Y Energías Alternativas']

Date: 2022-01 

Abstract Different species can find convergent solutions to adapt their genome to the same evolutionary constraints, although functional convergence promoted by chromosomal rearrangements in different species has not previously been found. In this work, we discovered that two domesticated yeast species, Saccharomyces cerevisiae, and Saccharomyces uvarum, acquired chromosomal rearrangements to convergently adapt to the presence of sulfite in fermentation environments. We found two new heterologous chromosomal translocations in fermentative strains of S. uvarum at the SSU1 locus, involved in sulfite resistance, an antimicrobial additive widely used in food production. These are convergent events that share similarities with other SSU1 locus chromosomal translocations previously described in domesticated S. cerevisiae strains. In S. uvarum, the newly described VIIXVI and XIXVI chromosomal translocations generate an overexpression of the SSU1 gene and confer increased sulfite resistance. This study highlights the relevance of chromosomal rearrangements to promote the adaptation of yeast to anthropic environments.

Author summary It is known that genetically distant species can arrive to similar evolutionary solutions during the adaptation to a specific environment, a phenomena known as convergent adaptation, and this frequently occurs after point mutations, gene duplications, or species hybridizations. In this work, we discovered a new example of convergent evolution in the adaptation of two wine fermentation yeast species to the presence of sulfite, an antimicrobial additive widely used in food production. We observed that two species, Saccharomyces cerevisiae and Saccharomyces uvarum, acquired chromosomal rearrangements to convergently adapt to the presence of sulfite in fermentative environments. We describe new chromosomal translocations in S. uvarum strains that generate an overexpression of the SSU1 gene and confer increased sulfite resistance, a similar event that was already described in S. cerevisiae. Overall, this study describes a new case of convergent evolution in which the chromosomal rearrangements have a significant role in the adaptation of yeast to an environment created by humans to produce food.

Citation: Macías LG, Flores MG, Adam AC, Rodríguez ME, Querol A, Barrio E, et al. (2021) Convergent adaptation of Saccharomyces uvarum to sulfite, an antimicrobial preservative widely used in human-driven fermentations. PLoS Genet 17(11): e1009872. https://doi.org/10.1371/journal.pgen.1009872 Editor: Justin C. Fay, University of Rochester, UNITED STATES Received: February 5, 2021; Accepted: October 11, 2021; Published: November 11, 2021 Copyright: © 2021 Macías et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: Raw sequencing data of S. uvarum strains sequenced in this study were submitted under BioProject PRJNA471597. Funding: This work was supported by grants from the Ministerio de Ciencia, Innovación y Universidades to A.Q (RTI2018-093744-B-C31), EB (RTI2018-093744-B-C32), and by Conselleria d'Educació, Investigació, Cultura i Esport grant PROMETEO/2020/014 to A.Q and EB, as well as grants PICT 2015-1198 from the Fondo para la investigación Científica y Tecnológica, PIP 2015-555 from Comisión de Investigaciónes Científicas and PI04-A128 from Universidad Nacional de Comahue to CL. MGF also thanks CONICET for a postdoctoral fellowship. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist.

Introduction Organisms belonging to different lineages can evolve independently to overcome similar environmental pressures through different molecular mechanisms. This convergent evolution has been seen as evidence of the action of natural selection [1,2]. In recent years, comparative genomics studies have suggested that convergent adaptations occur more frequently than previously expected [3,4]. For example, species of insects spanning multiple orders have independently evolved higher tolerance to toxic compounds produced by plants after different amino acid substitutions that might lower sensitivity to cardenolides [5] demonstrating that convergent adaptation can occur in nature between organisms belonging to different taxonomic levels. In the case of yeasts, convergent evolution by point mutations has been described both in evolving yeast species in nature [6] and in short-term evolutionary studies in the species Saccharomyces cerevisiae [7], for example in populations evolved under glucose limitation that increased fitness after alternative mutations in the genes MTH1 and HXT6/HXT7 [8]. Convergent evolution can occur through different mechanisms, including point mutations, gene duplications, and species hybridizations. Examples of convergent evolution via chromosomal rearrangements are rare, a single study has suggested that an intrachromosomal translocation is responsible of a convergent evolution in independent lineages in the case of the major histocompatibility complex [9]. A second study has suggested that amylase evolution in fish may have converged though a putative chromosomal translocation, although this has not yet been confirmed [10]. The genus Saccharomyces is composed of eight species including the model organism S. cerevisiae [11]. There is a substantial nucleotide divergence displayed for example between S. cerevisiae and the species S. uvarum and S. eubayanus, comparable to the divergence found between humans and birds [12]. S. cerevisiae has traditionally been associated with food and beverage fermentations which have been traced back to 5,000–10,000 years ago [13,14]. This domestication of S. cerevisiae by humans has left footprints that characterize their genome [15,16,17]. Along with S. cerevisiae, the species S. uvarum is the only natural species of the Saccharomyces genus that shows ecological success in human-driven fermentative environments [18]. S. uvarum coexists and even replaces S. cerevisiae in wine and cider fermentations performed at low temperatures, in particular at regions with oceanic or continental climate [19–21]. Genomic footprints of domestication, like introgressions, have also been reported in S. uvarum genomes [22]. During fermentation processes, yeast cells face adverse conditions such as osmotic stress due to high sugar concentrations, low temperatures, low pH, and the presence of certain toxic compounds used as preservatives. One of the most common preservatives used in wine and cider fermentations is sulfite [23]. The most common molecular mechanism to deal with the presence of sulfite in the media in yeasts involves the sulfite efflux with a plasma membrane pump encoded by the gene SSU1 [24,25]. The strains lacking this gene showed a higher sensitivity to sulfite due to the intracellular accumulation of this compound [26]. The transcription factor encoded by the FZF1 gene has been reported to interact with the upstream promoter region of the gene SSU1 to increase its transcription [26]. Mutations causing large-scale chromosomal rearrangements often occur in yeast populations rather than less frequent small-scale changes [27]⁠. Even though most large-scale changes are deleterious and, therefore, quickly removed from the population, these mutations contribute to the genetic variation within the population facilitating the rapid adaptation to novel environments [28,29]. It has been reported that specific chromosomal rearrangements in S. cerevisiae wine strains generate an overexpression of the SSU1 gene that increases the tolerance to sulfite [30], although it has been suggested that other unrelated sulfite tolerance adaptations could be present in the genome of the wine strains [31]. A reciprocal translocation between chromosomes VIII and XVI replaced the promoter of the SSU1 gene, encoding a sulfite transporter [30]. This modification causes an increased expression of SSU1 and, as a consequence, a greater resistance to sulfite [30]. After this first evidence, several groups have confirmed both the presence of this rearrangement in different strains belonging to the S. cerevisiae wine yeast subpopulation and the advantage that sulfite resistance confers to yeasts during their competition in wine fermentation [32–34]. Translocation VIIIXVI has been proposed not only to contribute to the ecological differentiation of wine yeasts but also to the partial reproductive isolation between wine and wild subpopulations of S. cerevisiae [35,36]. Years later, another translocation event, between chromosomes XV and XVI, was described and associated with an increase in the expression of the SSU1 gene in S. cerevisiae [37]. Another molecular mechanism causing the overexpression of this gene found in S. cerevisiae is an inversion in chromosome XVI [38]. A recent study with hundreds of strains confirmed the dominant presence of these SSU1 locus rearrangement in the wine strains population, specially in commercial starters [39]. The promoter region of the SSU1 gene has been demonstrated to be a hotspot of evolution in S. cerevisiae leading to different chromosomal rearrangements with a common phenotypic outcome: an increased sulfite tolerance. This work aims to test the evidence of convergent evolution at a higher taxonomic level by using another Saccharomyces species isolated from human-driven environments, S. uvarum. In this study, several strains of S. uvarum isolated from a wide range of environments and geographic locations have been used to identify high sulfite tolerant strains and the underlying molecular mechanisms associated with this trait.

Discussion In this work, we present a case of a convergent adaptation of S. uvarum strains, isolated from fermentation environments, to grow in sulfite containing media, a preservative usually added in industrial processes such as wine or cider fermentation. This is the first example reported in which different chromosomal rearrangements originated by two different chromosomal translocation events resulted in the over-expression of the SSU1 gene and, therefore, an increase of the sulfite tolerance in the strains carrying the translocations. In S. cerevisiae, different cases of structural variations have been described in the promoter of the SSU1 gene. These variations include chromosomal [30,34,37], which involve different chromosomes than those reported for S. uvarum, and a chromosomal inversion [38]. These SSU1 promoter variants described for S. cerevisiae have been reported to cause the overexpression of this gene being those strains much more tolerant to the presence of sulfites in the culture media. This is the first time that a chromosomal translocation event in the SSU1 promoter, providing an adaptive value, is described for another Saccharomyces species, different from S. cerevisiae. As far as we know, our work describes the first example of a phenotypic convergence produced by independent chromosomal rearrangements in two of the most divergent Saccharomyces species, S. cerevisiae, and S. uvarum (20% of nucleotide divergence). In fact, the last common ancestor existed 20 million years ago [12]. Strains of both species exhibit rearrangements at different locations in the promoter of the SSU1 gene that allows adaptation to tolerate high sulfite concentrations. It is well known the enormous adaptive role that exerts the overexpression of the SSU1 gene in industrial strains [35,36]. This effect would explain why it has been favored the appearance of molecular mechanisms, as the chromosomal translocation at the SSU1 locus, resulting in a phenotypic convergence. Interestingly, the four chromosomal translocation events described so far are independent, produced at different locations of the SSU1 promoter, and involving reciprocal translocations between chromosome XVI and different partners. Our results, including several complementary approaches, confirm the strong selection pressure that the antimicrobial effect of sulfite imposes on yeasts in human-driven fermentations, as well as remarks on the role of chromosomal rearrangements as a source of variation to promote yeast adaptations in fast-evolving environments. The molecular mechanisms that produced the overexpression of the SSU1 gene remains unclear. The regulation mechanism of the SSU1 gene known until now is mediated by the five-zinc-finger transcription factor codified by the FZF1 gene. This gene acts as a positive regulator of the SSU1 by binding directly to its upstream promoter [26]. The Fzf1p binding sequence has been described as 5’-CTATCA-3’. This sequence is present at many sites throughout the genome but SSU1 is the only demonstrated target. We have identified the binding sequence in the ancestral promoter SSU1 version of strains without chromosomal rearrangements. Interestingly, both rearrangements described in this work, occurred before the FZF1 binding site, like in S. cerevisiae, hence, the SSU1 promoter region lost the Fzf1p binding site due to the chromosomal rearrangements. Our main hypothesis is that FZF1 is not regulating the expression of the SSU1 gene in these S. uvarum strains. Instead of that, this gene could be possibly constitutively active or being regulated by another of several transcription factors that have not been identified yet. We can also conclude from our experiments that the overexpression effect of the SSU1 gene is not dependent on the presence of sulfite in the media as this gene is highly expressed from the early stages of fermentation with and without sulfite. The XIXVI translocation was found in a unique European strain isolated from a cider fermentation while the VIIXVI translocation event is shared among European and South American strains. Previous population analyses performed on the S. uvarum species classify them into four differentiated populations: Australasian, South America B, South America A, and Holarctic [22]. In a recent study [41], the existence of South America A population, genetically differentiated from the Holarctic population has been questioned and the authors suggest that these strains are the result of the genetic admixture of Holarctic and South America B strains. This fact, together with the high incongruence observed in our phylogenic reconstruction, leads us to think that they should not be properly considered as two different populations because they are, indeed, a mixed population. This idea is supported by the shared chromosomal rearrangement described in this study between strains isolated in Europe and Argentina. We hypothesize that these strains probably coexisted at the same location. This rearrangement was spread by sexual reproduction among different strains and it became fixed later in those strains grown in human-related environments where sulfite is used as a microbial preservative. Our data suggest that the VIIXVI recombination had a unique and recent origin in a European strain, and then, it was inherited by these South American strains due to hybridizations between European and South American strains. This premise is supported by the conserved region observed in the SSU1 surrounding gene sequences of NPCC1417 with respect to the translocated regions of the European strains. The conservation of this large segment could be due to a reduction of the recombination rate between the translocated and the standard chromosome alleles in the regions flanking the translocation point or to genetic hitchhiking in the surroundings of the translocated SSU1 gene as the target of selection. However, the fact that the conserved region surrounding the reciprocal translocation site is significantly smaller does not support a lower recombination rate in the regions flanking the translocation points and, hence, is compatible with the presence of a large, linked region swept along with the selectively favored recombinant SSU1 allele. Finally, our discovery highlights the role of the SSU1 gene promoter as a hotspot of evolution at different taxonomic levels. S. cerevisiae is the predominant species in sulfite-containing environments as wine, cider, and other fermented beverages. However, S. uvarum can be also dominant in certain types of fermentation, especially those performed at lower temperatures [19,20,42]. This abundance can explain the detection of the SSU1 locus chromosomal translocation events exactly in those species, as an adaptation to sulfite. Other species such as Hanseniospora uvarum, Metschnikowia pulcherrima, Bretanomyces sp. among others can be found in relatively high numbers in those environments at the beginning and even at more advanced stages of fermentations [43,44]. Future studies should examine chromosomal rearrangements involving the gene responsible for sulfite detoxification in these species.

Materials and methods Yeast strains, media, and fermentations Information about the yeast strains used in this study is summarized in S2 Table. Strains were maintained and propagated in GPYD media (5 g/L yeast extract, 5 g/L peptone, 20 g/L glucose). Wine fermentations were carried out in 100 mL bottles filled with 90 ml of synthetic must (100 g/L glucose, 100 g/L fructose, 6 g/L citric acid, 6 g/L malic acid, mineral salts, vitamins, anaerobic growth factors, 300 mg/L assimilable nitrogen) that simulates standard grape juice [45]. Fermentations were inoculated at 5.0 × 106 cells/ml density from overnight precultures determined by measuring OD 600 . Bottles were closed with Muller valve caps and incubated at 25°C with gentle agitation. Fermentation progress was followed by daily measuring bottle weight loss. In the fermentations with MBS, after preliminary tests, a sub-lethal concentration (15 mg/l) of MBS that allow the four strains used (BMV58, CECT12600, NPCC1290, and NPCC1314) to grow was selected. All wine fermentations were performed at least in independent triplicates. Edited strains construction To modify SSU1 promoters in the CBS7001 strain we used the CRISPR-Cas9 technique as described by Generoso et al. [46]. Primers used are listed in S3 Table. The plasmid pRCCN (Addgene) was used to target the SSU1 promoter to integrate the recombinant fragments, amplified from BMV58 or BR6-2 strains. The protospacer sequences were chosen according to Doench et al. [47] using CBS7001 genome sequence as reference to avoid selecting unspecific gRNA. Then we amplified by PCR the plasmid pRRC-N, which carries the natMX resistance marker, with primers carrying the protospacer sequence at their 5’ ends [46]. The PCR was carried out with Phusion High-Fidelity Polymerase following the provider instructions using the primers listed in S3 Table. Before addition to the transformation mix, we treated 30 μL of the PCR product with 10 U of DpnI restriction enzyme (Thermo Scientific) for 3 h to guarantee the degradation of pRRC-N template. To ensure the reparation by homologous recombination we used PCR amplified fragments of the SSU1 promoter from BMV58 or BR6-2 strains whose 40 nucleotides of each side are homologous to both upstream and downstream sequences of the target sequence [48]. 1 mmol of the PCR fragment was added to the transformation mix, performed following Gietz and Schiestl method [49]. Transformants were selected in ClonNat (Sigma) GPY agar plates and verified by PCR using diagnostic primers (S3 Table) and sanger sequencing. Finally, the positive strains were cured of the pRCCN vector.

Genome sequencing, assembly, and annotation Strains were sequenced by Illumina HiSeq 2000 with paired-end reads of 100 bp long at the Genomics section from the Central Service of Experimental Research Support (SCSIE), University of Valencia. SPAdes [50], with default parameters, was used for de novo assembly. BR6-2 strain and NPCC1314 were sequenced using PacBio sequencing Single Molecule, Real-Time (SMRT) DNA sequencing technology (platform: PacBio RS II; chemistry: P4-C2 for the pilot phase and P6-C4 for the main phase). The raw reads were processed using the standard SMRT analysis pipeline (v2.3.0). The de novo assembly was done using Flye (version 2.7) with 3 polishing iterations and default parameters [51]. MUMmer [52] was used to get the homology between the strains sequenced in this study and the reference S. uvarum strain CBS7001 [53]. This information was used to get scaffolds into chromosome structure (note that, in Scannel et al. [53] annotation, chromosome X was mislabeled as chromosome XII and vice-versa). Annotation was performed as described in [54]. We used a combination of two approaches including transferring the annotation from the S. cerevisiae S288c based on synteny conservation. The annotated assemblies were used to identify the ultrascaffolds containing the SSU1 gene and the surrounding annotated genes. We identified the position of the SSU1 gene and then we selected for further investigation those assemblies whose SSU1 gene position and surrounding genes does not match with the reference strain position (chromosome XVI). Phylogenetic analyses Annotated genomes sequenced in this study as well as collected data from previous studies [22,53] were used for phylogeny reconstruction. A list of the genomes used in this analysis can be found in supplementary S1 Table. Introgressed genes from other Saccharomyces species were removed from the analysis. A total number of 1265 orthologous genes were found among the 21 S. uvarum strains. Nucleotide sequences were translated into amino-acids and aligned with Mafft [55]. Aligned protein sequences were back-translated into codons. Maximum-Likelihood (ML) phylogeny reconstruction was performed for each gene using RAxML [56] with the GTRCAT model and 100 bootstrap replicates. ML-trees were concatenated to infer a coalescence-based phylogeny using ASTRAL-III, version 5.6.3 [57]. Tree was visualized using iTOL [58]. Analyses of the origin of the shared chromosomal rearrangement among BMV58, CECT12600, and NPCC1417 strains Gene sequences upstream and downstream of the SSU1 gene were extracted to calculate genetic distances among the strains BMV58, CECT12600, and NPCC1417. Distances were calculated using the “dist.dna” function from the ape R package [59] under the “K81” model [60]. This method was repeated to calculate pairwise genetic distances using the BMV58 as a reference against NPCC1309 and NPCC1314 strains. An in-house python script was used to select 1,000 random windows of 20 genes within BMV58 and NPCC1417 genomes to calculated pairwise genetic distances. Southern blot analysis We performed Southern blot analyses with karyotyping gels. Pulsed-field gel electrophoresis was performed under these conditions: 60 seconds during 12 h and 120 seconds during 14 h with an angle of 150° and a velocity of 6V/cm. The strains included were BMV58, CECT12600, NPCC1290, and NPCC1314. DNA was transferred to a nylon membrane Amersham Hybond -N+ (GE Healthcare Europe GmbH, Barcelona, Spain) according to manufactures protocol. We construct the probes using the primers listed in S3 Table and the PCR DIG Probe Synthesis Kit (Roche Applied Science, Mannheim, Germany). Each Southern blot analysis was done with high stringency conditions to be sure of the specificity of the probe. Hybridization was prepared with DIG Easy Hyb Granules (Roche Applied Science), following recommendations of the manufacturer for prehybridization, hybridization, and post hybridization washes. For washing, blocking, and detection of DIG-labeled probes DIG Wash and Block Buffer Set (Roche Applied Science) was used. For the detection of DIG-labeled molecules an Anti-Digoxigenin-AP, Fab fragment (1,10.000) (Roche Applied Science), was used. Finally, CDP-Star Set (Roche Applied Science), a chemiluminescent substrate for alkaline phosphatase was used at 1:100 dilution, and images were stored after 30 min of exposition. Gene expression determination For each culture, a 10–20-ml sample was taken each day of wine fermentation. The cells were quickly collected by centrifugation, washed, and frozen with liquid N 2 . Then, frozen cells were homogenized with a FastPrep-24 (MP Biomedicals, Santa Ana, USA) device with acid-washed glass beads (0.4 mm diameter; Sigma-Aldrich, Madrid, Spain) in LETS buffer (10 mm Tris pH 7.4, 10 mM lithium-EDTA, 100 mM lithium chloride, 1% lithium lauryl sulfate) for 30 s alternating with ice incubation six times. The phenol:chloroform method with minor modifications [61] was used to extract and purify total RNA. Then, cDNA was synthesized from the RNA and the expression of SSU1 genes was quantified by qRT-PCR (quantitative real-time PCR). cDNA was synthesized in 13 μl using 2 μg of RNA mixed with 0.8 mM dNTP’s and 80 pmol Oligo (dT). The mixture was incubated at 65°C for 5 min and in ice for 1 min. Then, 5 mM dithiothreitol (DTT), 50 U of RNase inhibitor (Invitrogen, Waltham, USA), 1 × First-Strand Buffer (Invitrogen), and 200 U Superscript III (Invitrogen) were added in 20 μl mixture and this was incubated at 50°C for 60 min and 15 min at 70°C. qRT-PCR gene-specific primers (200 nM), designed (S3 Table) from consensus sequences between the different strains, were used in 10 μl reactions, using the Light Cycler FastStart DNA MasterPLUS SYBR green (Roche Applied Science) in a LightCycler 2.0 System (Roche Applied Science). All samples were processed for DNA concentration determination, amplification efficiency, and melting curve analysis. To obtain a standard curve, serial dilutions (10−1 to 10−5) of a mixture of all samples was used. The average of ACT1 and RDN18-1 constitutive genes was used to normalize the amount of mRNA and to safeguard repeatability, correct interpretation, and accuracy [62]. Sulfite tolerance assay Sulfite tolerance was tested in YEPD +TA (tartaric acid) agar plates as described by Park et. al. [63]. YEPD (2% dextrose, 2% peptone and 1% yeast extract) was supplemented with L- tartaric acid at 75 mM buffered at pH 3.5 and potassium metabisulfite (K 2 S 2 O 5 , MBS) was added to each plate to a final concentration of 0, 0.05, 0.10, 0.15, 0.20, 0.25, 0.30, 0.35 or 0.40, g/L. Yeast precultures were grown overnight in GPY medium. Cell cultures were diluted to OD 600 = 1. Then, serial 1:5 dilutions of cells were inoculated in MBS YEPD plates and incubated at 25°C for a week.

Acknowledgments Genome sequences were obtained at the Genomics section from the Central Service of Experimental Research Support (SCSIE), University of Valencia. We also thank Chris Todd Hittinger, Philippe Marullo and Diego Libkind for kindly providing strains.

[END]

[1] Url: https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1009872

(C) Plos One. "Accelerating the publication of peer-reviewed science."
Licensed under Creative Commons Attribution (CC BY 4.0)
URL: https://creativecommons.org/licenses/by/4.0/


via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/