(C) PLOS One
This story was originally published by PLOS One and is unaltered.
 . . . . . . . . . .



Wild Patagonian yeast improve the evolutionary potential of novel interspecific hybrid strains for lager brewing [1]

['Jennifer Molinet', 'Anid-Millennium Science Initiative-Millennium Institute For Integrative Biology', 'Ibio', 'Santiago', 'Departamento De Biología', 'Facultad De Química Y Biología', 'Universidad De Santiago De Chile', 'Department Of Zoology', 'Stockholm University', 'Stockholm']

Date: 2024-07 

Lager yeasts are limited to a few strains worldwide, imposing restrictions on flavour and aroma diversity and hindering our understanding of the complex evolutionary mechanisms during yeast domestication. The recent finding of diverse S. eubayanus lineages from Patagonia offers potential for generating new lager yeasts with different flavour profiles. Here, we leverage the natural genetic diversity of S. eubayanus and expand the lager yeast repertoire by including three distinct Patagonian S. eubayanus lineages. We used experimental evolution and selection on desirable traits to enhance the fermentation profiles of novel S. cerevisiae x S. eubayanus hybrids. Our analyses reveal an intricate interplay of pre-existing diversity, selection on species-specific mitochondria, de-novo mutations, and gene copy variations in sugar metabolism genes, resulting in high ethanol production and unique aroma profiles. Hybrids with S. eubayanus mitochondria exhibited greater evolutionary potential and superior fitness post-evolution, analogous to commercial lager hybrids. Using genome-wide screens of the parental subgenomes, we identified genetic changes in IRA2, IMA1, and MALX genes that influence maltose metabolism, and increase glycolytic flux and sugar consumption in the evolved hybrids. Functional validation and transcriptome analyses confirmed increased maltose-related gene expression, influencing greater maltotriose consumption in evolved hybrids. This study demonstrates the potential for generating industrially viable lager yeast hybrids from wild Patagonian strains. Our hybridization, evolution, and mitochondrial selection approach produced hybrids with high fermentation capacity and expands lager beer brewing options.

Lager beer dominates the global market, accounting for over 90% of commercial beer varieties. The main player in lager fermentation is the yeast Saccharomyces pastorianus, an interspecific hybrid between S. cerevisiae and S. eubayanus. Despite its popularity, the range of flavours and aromas found in lager beers is restricted by the low genetic diversity of available lager strains. Here, we explored if lager yeast profiles can be diversified by leveraging natural isolates of S. eubayanus from Chilean Patagonia. We generated de novo hybrids between S. cerevisiae and three distinct S. eubayanus Patagonian lineages. Through experimental evolution and selection on fermentation traits, we improved the fermentation profiles of the hybrids. We found that mutations in IRA2, IMA1, and MALX genes enhanced their maltose and maltotriose metabolism, resulting in higher ethanol production and unique aroma profiles. Our results also confirm that S. eubayanus mitochondria confer a greater evolutionary potential than S. cerevisiae mitochondria. The current study encourages the use of wild yeast strains to develop new brewing applications to expand the repertoire of de novo lager yeasts.

Funding: This research was funded by Agencia Nacional de Investigación y Desarrollo (ANID) FONDECYT program and ANID-Programa Iniciativa Científica Milenio – ICN17_022 and NCN2021_050. FAC is supported by FONDECYT grant N° 1220026, JM by FONDECYT POSTDOCTORADO grant N° 3200545 and PV by ANID FONDECYT POSTDOCTORADO grant N° 3200575. CAV is supported by FONDECYT INICIACIÓN grant N° 11230724. RFN is supported by FONDECYT grant N° 1221073. RS and JM’s work is supported by the Swedish Research Council (2022-03427) and the Knut and Alice Wallenberg Foundation (2017.0163). FIS, JM, and PV received a salary from FONDECYT grant N° 1220026, FONDECYT POSTDOCTORADO grants N° 3200545 and N° 3200575, respectively. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Here, we hybridized three different S. cerevisiae and S. eubayanus strains to generate genetically and phenotypically diverse novel lager hybrids via spore-to-spore mating. The initial de novo hybrids had fermentation capacities comparable to those of their parental strains and did not show positive heterosis. However, when we subjected hybrids to a ‘fast motion’ improvement process using experimental evolution under different fermentation conditions for 250 generations, they exceeded the fitness of the ancestral hybrids, particularly those retaining the S. eubayanus mitochondria. Superior hybrid fitness was explained by faster fermentation performance and greater maltose and maltotriose consumption We found that copy number variation in MAL genes in the S. cerevisiae subgenome, together with SNPs in genes related to glycolytic flux, induced significantly greater expression levels of MAL and IMA1 genes, and led to improved fitness under fermentative conditions in these novel S. cerevisiae x S. eubayanus yeast hybrids. Furthermore, evolved hybrids had significantly distinct aroma profiles, varying significantly from the established profiles found in lager beer.

Lager yeast hybrids experienced an intense domestication process through selection and re-pitching during beer fermentation since the 17 th century [ 9 , 10 , 12 , 19 , 20 ], a process similar to experimental evolution [ 21 , 22 ]. Experimental evolution with microbes is a powerful tool to study adaptive responses to selection under environmental constraints [ 23 – 26 ]. Recent studies on novel S. cerevisiae x S. eubayanus hybrids suggest that hybrid fermentative vigour at low temperature results from a variety of genetic changes, including loss of heterozygosity (LOH), ectopic recombination, transcriptional rewiring, selection of superior parental alleles [ 27 ], heterozygote advantage due to the complementation of loss-of-function mutations in the F1 hybrid genome [ 28 ], and novel structural and single nucleotide variants in the hybrid genome [ 29 ]. A recent transcriptome analysis of a laboratory-made lager hybrid strain under fermentation conditions highlighted that the regulatory ‘cross-talk’ between the parental subgenomes caused a novel sugar consumption phenotype in the hybrid (maltotriose utilization, essential for lager fermentation), which was absent in both parental strains [ 28 ]. Although these studies have greatly contributed to our understanding of the genetic basis of different lager phenotypes, most studies only considered a single S. eubayanus genetic background (type strain CBS 12357), which alone is not representative of the species-rich genetic diversity.

The discovery of S. eubayanus in Patagonia in 2011 [ 14 ], opened new possibilities for creating novel hybrid strains by using the full range of natural genetic diversity found in this species. Phylogenetic analyses have revealed six distinct lineages of S. eubayanus, including China, Patagonia A (‘PA’), Holarctic, and Patagonia B, ‘PB-1’, ‘PB-2’ and ‘PB-3’, and some admixed strains derived from ancient crosses [ 15 , 16 ]. Of these, S. eubayanus from Patagonia displays the broadest phenotypic diversity for a wide range of traits, including high maltose consumption, aroma profiles, and fermentation capacity [ 15 , 17 , 18 ]. The distinctive traits of wild Patagonian S. eubayanus strains indicate their potential for crafting new lager beer styles. These strains could yield novel taste and aroma profiles, approaching similar complexity and diversity in flavour, appearance, and mouthfeel as Ale beers.

Humans have paved the way for microbes, such as yeast, to evolve desirable features for making bread, wine, beer, and many other fermented beverages for millennia [ 1 ]. The fermentation environment, characterized by limited oxygen, high ethanol concentrations, and microbial competition for nutrients (typically yeasts, molds, and bacteria) can be considered stressful [ 2 ]. One evolutionary mechanism to overcome harsh conditions is hybridization, because it rapidly combines beneficial phenotypic features of distantly related species and generates large amounts of genetic variation available for natural selection to act on [ 3 – 5 ]. Hybrids can also express unique phenotypic traits not seen in the parental populations through the recombination of parental genetic material, enabling them to thrive in different ecological niches [ 4 , 6 – 8 ]. An iconic example is the domesticated hybrid yeast Saccharomyces pastorianus to produce modern lager (pilsner) beers. S. pastorianus results from the successful interspecies hybridization between S. cerevisiae and S. eubayanus [ 9 , 10 ]. Hybrids have been shown to benefit from the cold tolerance of S. eubayanus and the superior fermentation kinetics of S. cerevisiae [ 11 ]. We now know that domestication over the last 500 years has generated lager yeast strains with the unique ability to rapidly ferment at lower temperatures resulting in a crisp flavour profile and efficient sedimentation, improving the clarity of the final product. However, the genetic diversity of commercial lager yeast strains is extremely limited, mainly due to the standardization of industrial lager production during the nineteenth century in Germany [ 9 , 12 ]. This gave rise to only two genetically distinct S. pastorianus subgroups, Group 1 strains (‘Saaz’) and Group 2 strains (‘Frohberg’). The poor genetic diversity of lager strains used in commercial brewing today (85 lager strains commercially available versus 358 ale strains [ 13 ]) puts tight constraints on the variety of flavours and aromas found in lager beer. At the same time, it limits our understanding of the evolutionary mechanism operating during the yeast domestication process.

Data visualization and statistical analyses were performed with R software version 4.03. Maximum specific growth rates and total CO 2 loss were compared using an analysis of variance (ANOVA) and differences between the mean values of three replicates were tested using Student’s t-test and corrected for multiple comparisons using the Benjamini-Hochberg method. A p-value less than 0.05 (p<0.05) was considered statistically significant. Heatmaps were generated using the ComplexHeatmap package version 2.6.2. A principal component analysis (PCA) was performed on phenotypic data using the FactoMineR package version 2.4 and the factoextra package version 1.07 for extracting, visualizing and interpreting the results.

DNA content was assessed through the propidium iodide (PI) staining assay, as previously described [ 67 ]. Initially, cells were recovered from glycerol stocks on YPD solid media and allowed to grow overnight at 25°C. Subsequently, a portion of each patch was transferred into liquid YPD media and incubated overnight at 25°C. Then, 1 ml of each culture was harvested and suspended in 2.3 ml of cold 70% ethanol for fixation during 48h h at 4°C. Following fixation, cells were washed with sodium citrate (50 mM, pH 7) and 100 μl of cells resuspended in the same solution, and 1 μL of RNAse A (100 mg/mL) were incubated for 2 h at 37°C. Then, cells were stained with a solution containing PI (final concentration of 50 μg/mL) and sodium citrate (50 mM, pH 7), and incubated for 40 min at room temperature in darkness. Analysis was conducted on a BD FACSCanto II flow cytometer with excitation at 488 nm and fluorescence collection using an FL2-A filter, analyzing ten thousand cells per sample. Three strains with known ploidy (two S. cerevisiae -n and 2n- and one S. pastorianus -4n-) were employed as controls.

The S. cerevisiae IRA2 polymorphism was validated by Sanger sequencing. PCR products were purified and sequenced by KIGene, Karolinska Institutet (Sweden). The presence of the SNP in the evolved hybrid strains was checked by visual inspection of the electropherograms. Null mutants for the IRA2 gene in the S. cerevisiae subgenome were generated using CRISPR-Cas9 [ 64 ] as previously described [ 36 ]. Briefly, the gRNAs were designed using the Benchling online tool ( https://www.benchling.com/ ) and cloned into the pAEF5 plasmid [ 65 ], using standard “Golden Gate Assembly” [ 66 ]. Ancestral and evolved hybrids were co-transformed with the pAEF5 plasmid carrying the gRNA and the Cas9 gene, together with a double-stranded DNA fragment (donor DNA). The donor DNA contained nourseothricin (NAT) resistance cassette, obtained from the pAG25 plasmid (Addgene plasmid #35121), flanked with sequences of the target allele, corresponding to 50-pb upstream of start codon and 50-pb downstream of the stop codon. Correct gene deletion was confirmed by standard colony PCR. All primers, gRNAs, and donor DNA are listed in Tab B in S1 Table .

Reads quality was evaluated using the fastqc tool ( https://www.bioinformatics.babraham.ac.uk/projects/fastqc/ ) and processed using fastp (-3 l 40) [ 57 ]. Reads were mapped to a concatenated fasta file of the DBVPG6765 and CL216.1 genome sequences. To account for mapping bias due to the different genetic distances of the parental strains to their reference strains, the L3 and CL710.1 parental strains were re-sequenced using WGS, after which genomic reads were mapped with BWA [ 47 ] to the DBVPG6765 and CL216 references and SNPs were called using freebayes [ 58 ]. These SNPs were used to correct the hybrid genome sequence using the GATK FastaAlternateReferenceMaker tool. RNAseq reads were mapped to this hybrid reference using STAR (-outSAMmultNmax 1, -outMultimapperOrder random) [ 59 ]. Counts were obtained with featureCounts using a concatenated annotation file [ 60 ]. Counts were further analyzed in R using de DESeq package [ 61 ]. A PCA analysis to evaluate the reproducibility of replicates was performed, after which two outlier replicates (H3-A replicate 3 and H3-E replicate 2) were removed. To analyze differences in allele expression, a list of 1-to-1 orthologous genes between both parental strains was identified using OMA [ 62 ]. Orthologous genes that differ more than 5% on their gene lengths were excluded. The differential allelic expression of these orthologous genes was determined using design = ~parental, with parental being “L3” or “CL710”. Furthermore, orthologous genes that showed differential allele expression depending on the ancestral or evolved strain background were assessed using an interaction term (~ parental:condition), with condition being “ancestral” or “evolved”. Finally, to evaluate differences between ancestral and evolved hybrid strains, all 11,047 hybrid genes (5,508 S. eubayanus and 5,539 S. cerevisiae) were individually tested for differential expression using DESeq2. Overall gene expression differences were evaluated using the design ~condition. For all analyzes an FDR < 0.05 was used to consider statistical differences. GO term enrichment analyzes on differentially expressed genes were calculated using the package TOPGO [ 63 ].

Gene expression analysis was performed on ancestral and evolved hybrid strains H3-A and H3-E. RNA was obtained and processed after 24 h under beer wort fermentation in triplicates, using the E.Z.N.A Total RNA kit I (OMEGA) as previously described [ 37 , 56 ]. Total RNA was recovered using the RNA Clean and Concentrator Kit (Zymo Research). RNA integrity was confirmed using a Fragment Analyzer (Agilent). Illumina sequencing was performed in NextSeq500 platform.

Since S. cerevisiae and S. eubayanus show divergence greater than 20% we could not detect loss of heterozygosity (LOH) by mapping to a single parental reference. As both hybrids appear to have one copy of each parental genome with no aneuploidies, LOH should appear as a loss of coverage segments that can be detected as copy number losses. LOH regions were detected by mapping reads to the concatenated genome and we used CNVkit [ 55 ] to detect genomic segments (1000 bp windows) showing CNVs with log2FC less than -2. Comparisons were made between parental strains vs ancestral hybrids and between ancestral hybrids with evolved hybrids.

We used sppIDer [ 54 ] to assess the proportional genomic contribution of each species to the nuclear and mitochondrial genomes in each sequenced hybrid. In addition, we used the tool to identify potential aneuploidies within these genomes. CNVs were called using CNVkit (—method wgs,—-target-avg-size 1000) [ 55 ]. As the analysis was performed on a haploid reference (both parental genomes were present), a CNV of log2 = 1 corresponds to a duplication.

Genomic DNA was obtained for whole-genome sequencing using the YeaStar Genomic DNA Kit (Zymo Research, USA) and sequenced in an Illumina NextSeq500 following the manufacturer’s instructions. Variant calling and filtering were done with GATK version 4.3.0.0 [ 45 ]. Briefly, cleaned reads were mapped to a concatenated reference genome consisting of S. cerevisiae strain DBVPG6765 [ 46 ] and S. eubayanus strain CL216.1 [ 17 ] using BWA mem 0.7.17 [ 47 ], after which output bam files were sorted and indexed using Samtools 1.13 [ 48 ]. Variants were called per sample using HaplotypeCaller (default settings) generating g.vcf files. Variant databases were built using GenomicsDBImport and genotypes were called using GenotypeGVCFs (-G StandardAnnotation). SNPs and INDELs were extracted and filtered out separately using SelectVariants. We then applied recommended filters with the following options: QD < 2.0, FS > 60.0, MQ < 40.0, SOR > 4.0, MQRankSum < -12.5, ReadPosRankSum < -8.0. This vcf file was further filtered by removing missing data using the option–max-missing 1, filtering out sites with a coverage below 5 th or above the 95 th coverage sample percentile using the options–min-meanDP and–max-meanDP, and minimum site quality of 30 (—minQ 30) in vcftools 0.1.16 [ 49 ]. Sites with a mappability less than 1 calculated by GenMap 1.3.0 [ 50 ] were filtered using bedtools 2.18 [ 51 ]. As an additional filtering step, the ancestral and evolved vcf files were intersected using BCFtools 1.3.1 [ 52 ] and variants with shared positions were extracted from the vcf files of the evolved hybrids. Annotation and effect prediction of the variants were performed with SnpEff [ 53 ].

Experimental evolution was carried out at 20°C under two different media conditions (M and T): 1) YNB + 2% maltose supplemented with 9% ethanol (M) and 2) YNB + 1% maltose + 1% maltotriose supplemented with 9% ethanol (T). Experimental evolution assays in maltose were performed in a final volume of 1 mL in 2 mL tubes, while those in maltose and maltotriose were performed in a 96-well plate under a final volume of 200 μL. Each hybrid strain was first grown in 0.67% YNB medium with 2% maltose at 25°C for 24 h with constant agitation at 150 rpm. Each pre-inoculum was then used to inoculate each evolution line at an initial OD 600nm of 0.1, with three replicate lines per strain in medium M and four replicate lines in medium T. Lines in medium M were incubated at 20°C for 72 h. Lines in medium T were incubated for 144 h at 20°C. After this, cultures were then serially transferred into fresh medium for an initial OD 600nm of 0.1. Serial transfers were repeated for 250 generations in total (approximately seven months). The number of generations was determined using the formula log(final cells–initial cells)/log 2 [ 44 ]. Population samples were stored at -80°C in 20% glycerol stocks after 50, 100, 150, 200 and 250 generations. After 250 generations, three colonies were isolated for each replicate line on YPM solid medium (1% yeast extract, 2% peptone, 2% maltose and 2% agar) supplemented with 6% ethanol. The fastest growing colonies were stored at -80°C in 20% glycerol stocks. The fitness increase of each the 28 evolved line was determined as the ratio between the phenotypic value of a given line and the equivalent of its respective ancestral hybrid.

Hybrids and parental strains were phenotypically characterized under microculture conditions as previously described [ 36 ]. Briefly, we estimated mitotic growth in 96-well plates containing Yeast Nitrogen Base (YNB) supplemented with 2% glucose, 2% maltose, 2% maltotriose, 2% glucose and 9% ethanol, 2% glucose and 10% sorbitol, and under carbon source switching (diauxic shift) from glucose to maltose as previously described [ 37 ]. All conditions were evaluated at 25°C. Lag phase, growth efficiency, and the maximum specific growth rate (μ max ) were determined as previously described [ 38 , 39 ]. For the diauxic shift between glucose and maltose, lag time and μ max were determined during growth in maltose. The parameters were calculated following curve fitting (OD values were transformed to ln) using the Gompertz function [ 40 ] in R (version 4.03).

Sugar (glucose, fructose, maltose and maltotriose) consumption and ethanol production were determined by High-Performance Liquid Chromatography (HPLC) after 14 days of fermentation. Filtered samples (20 μL) were injected in a Shimadzu Prominence HPLC (Shimadzu, USA) with a BioRad HPX-87H column using 5 mM sulfuric acid and 4 mL acetonitrile per liter of sulfuric acid as the mobile phase at a 0.5 mL/min flow rate. Volatile compound production was determined by using HeadSpace Solid-Phase MicroExtraction followed by Gas Chromatography-Mass Spectrometry (HS-SPME-GC/MS) after 14 days of fermentation as previously described [ 18 ].

Fermentations were carried out in three biological replicates using previously oxygenated (15 mg/L) 12°P wort, supplemented with 0.3 ppm ZnCl 2 as previously described [ 17 ]. Briefly, pre-cultures were grown in 5 mL 6°P wort for 24 h at 20°C with constant agitation at 150 rpm. Cells were then transferred to 50 mL 12°P wort and incubated for 24 h at 20°C with constant agitation at 150 rpm. Cells were collected by centrifugation and used to calculate the final cell concentration to inoculate the subsequent fermentation according to the formula described by White and Zainasheff [ 35 ]. Cells were inoculated into 50 mL 12°P wort in 250 mL bottles covered by airlocks containing 30% glycerol. The fermentations were incubated at 12 or 20°C, with no agitation for 15 days and monitored by weighing the bottles daily to determine weight loss over time.

Parental strains were sporulated on 2% potassium acetate agar plates (2% agar) for at least seven days at 20°C. Interspecific F1 hybrids were generated through spore-spore mating between S. eubayanus strains and S. cerevisiae strains ( S1 Fig ). For this, tetrads were treated with 10 μL Zymolyase 100 T (50 mg/mL) and spores of opposite species were dissected and placed next to each other on a YPD agar plates using a SporePlay micromanipulator (Singer Instruments, UK). Plates were incubated at two different temperatures, 12 and 20°C, for 2–5 days to preserve the cold- and heat-tolerant mitochondria, respectively, as previously described [ 31 , 32 ], resulting in nine different F1 hybrids (ranging from H1 until H9, Tab A in S1 Table ). This procedure was repeated on 25 tetrads of each species, for each type of cross (H1 to H9) and temperature (12 and 20°C), resulting in 18 different cross x temperature combinations. Finally, colonies were isolated, re-streaked on fresh YPD agar plates, and continued to be incubated at 12 and 20°C. The hybrid status of isolated colonies was confirmed by amplification of rDNA-PCR (ITS1, 5.8S, and ITS2) using universal fungal primers ITS1 and ITS4 [ 33 ], followed by digestion of the amplicon using the HaeIII restriction enzyme (Promega, USA) as previously described [ 34 ] on one colony for each cross attempt ( S1 Fig ). Confirmed F1 hybrids were designated as H1 to H9 based on parental strains, followed by the hybridization temperature (12 or 20) and the colony number (i.e. H1.20–1 depicts cross 1 at 20°C (Tab A in S1 Table )). We identified the mitochondrial genotype by Sanger sequencing the mitochondrial COX3 gene as previously described [ 32 ].

Three S. cerevisiae strains were selected for hybridization from a collection of 15 strains isolated from different wine-producing areas in Central Chile and previously described by Martinez et al. [ 30 ]. Similarly, three S. eubayanus parental strains were selected from a collection of strains isolated from different locations in Chilean Patagonia, exhibiting high fermentative capacity and representative of the different Patagonia-B lineages (PB-1, PB-2, and PB-3) [ 15 ]. The S. pastorianus Saflager W34/70 (Fermentis, France) strain was used as a commercial lager fermentation control. All strains were maintained in YPD agar (1% yeast extract, 2% peptone, 2% glucose and 2% agar) and stored at -80°C in 20% glycerol stocks. Strains are listed in Tab A in S1 Table .

Results

Evolved lines carrying the S. eubayanus mitochondria exhibit a greater fitness under fermentation All results so far indicated that the de novo interspecific hybrids did not show any hybrid vigour, in none of the phenotypes assessed. We thus decided to subject hybrids to experimental evolution to enhance their fermentative capacity. We specifically selected four hybrids (H3.12–3, H4.12–4, H6.20–2, and H8.20–5) because they demonstrated the highest phenotypic values across kinetic parameters. From here on we will refer to these strains as H3-A, H4-A, H6-A, and H8-A (A for ‘ancestral’ or unevolved hybrid). These four hybrids completely consumed the sugars present in the beer wort, except for maltotriose, which may explain the lower fermentative capacity of the hybrids compared to the commercial strain W34/70 (Tab E in S4 Table). Furthermore, these four hybrids represent crosses made at 12°C and 20°C and they encompass all six parental genetic backgrounds. To enhance the fermentative capacity of these selected hybrids, they were subjected to adaptive evolution at 20°C for 250 generations under two distinct conditions: i) YNB supplemented with 2% maltose and 9% ethanol (referred to as "M" medium), and ii) YNB supplemented with 1% maltose, 1% maltotriose, and 9% ethanol (referred to as “T” medium). We evolved three lines independently per cross in medium M, and four independent lines per cross in medium T. These conditions were chosen because maltose is the main sugar in beer wort (approximately 60%) [68]. Considering that yeast typically consume carbon sources in a specific order (glucose, fructose, maltose, and maltotriose), we employed a combination of maltose and maltotriose to facilitate the utilization of the latter carbon source. After 250 generations, the evolved lines showed different levels of fitness improvements, depending on the environmental conditions and their genetic background (Figs 2A and S4), with distinct fitness trajectories over time (S5 Fig). All evolved lines significantly increased in fitness in at least one of the evolution media and/or kinetic parameters assessed compared to their respective ancestral hybrids (Fig 2A and Tabs A and B in S5 Table; p-value < 0.05, one-way ANOVA). Interestingly, evolved lines from hybrids made at 12°C mating temperature (H3-A and H4-A) showed a more pronounced fitness increase in the T medium compared to those generated at 20°C (p-value = 3.327e-08, one-way ANOVA, Figs 2B and S4B), suggesting that hybrids with S. eubayanus mitochondria have greater potential for improvement than hybrids with S. cerevisiae mitochondria. We verified that the two ancestral H3-A and H4-A hybrids carried only S. eubayanus mitochondria by sequencing the COX3 gene, while H6-A and H8-A inherited the mitochondria from S. cerevisiae (Tabs C and D in S5 Table). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 2. Fitness of evolved lines under microcultures and fermentation conditions. (A) Mean relative fitness (maximum OD 600nm ) of evolved lines after 250 generations to their respective ancestral hybrids under microculture conditions. Evolved lines were evaluated in the same medium where they were evolved (M or T medium). (B) Comparison of mean relative fitness (maximum OD 600nm ) shown in (A) between evolved lines from hybrids with S. eubayanus (Se) and S. cerevisiae (Sc) mitochondria. (C) Mean relative fitness (maximum CO 2 loss) of evolved lines after 250 generations to their respective ancestral hybrids under fermentation conditions at 12°C. (D) Comparison of mean relative fitness (maximum CO 2 loss) shown in (C) between evolved lines from hybrids with S. eubayanus and S. cerevisiae mitochondria. (E) Maltotriose uptake of evolved hybrid lines in maltose (M) and maltose/maltotriose (T), relative to the commercial lager strain W34/70. Ancestral hybrids are shown in grey, and hybrid lines with S. eubayanus and S. cerevisiae mitochondria are shown in blue and red, respectively. (F) The fermentative capacity of evolved individuals relative to the commercial lager strain W34/70 grouped according to the environmental condition used during experimental evolution and inherited mitochondria. Plotted values correspond to the mean of three independent biological replicates of each evolved line or strain. Asterisk indicates significant statistical differences between evolved lines and their respective ancestral hybrids in (A) and (C), between evolved lines with different inherited mitochondria in (B) and (D), and between evolved lines and the commercial lager strain in (E) and (F). Purple depicts Parental strains, brown the ancestral hybrid, and red and blue the Sc and Se evolved lines carrying mitochondria, respectively. Asterisk represents different levels of significance (Students t-test, * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001, ns not significant). https://doi.org/10.1371/journal.pgen.1011154.g002 Next, we assessed the fermentative capacity of the evolved lines under conditions resembling beer wort fermentation (12°Brix and 12°C) (Figs 2C and S6 and Tabs A and B in S6 Table). We did not observe a significant increase in CO 2 production levels in the evolved lines of the H6-A and H8-A hybrids in either M or T media (Figs 2C and S6 and Tab B in S6 Table, p-value < 0.05, one-way ANOVA). However, we found a significant greater CO 2 production in the evolved lines of H4-A, evident in both evolution media, indicative of higher fermentation activity. The evolved lines of H3-A under T media also demonstrated a slightly higher CO 2 production (Fig 2C and Tab B in S6 Table, p-value < 0.05, one-way ANOVA, for H4 evolved lines and p-values of 0.0708 and 0.05149 for H3 evolved lines in M and T, respectively). Thus, both evolved hybrid lines generated at cold-temperature, carrying S. eubayanus mitochondria, showed a greater increase in CO 2 production than hybrids carrying the S. cerevisiae mitochondria (Fig 2C). Specifically, hybrids with S. eubayanus mitochondria increased their maximum CO 2 loss by 10.6% when evolving in M medium (p-value = 0.003698, one-way ANOVA) and by 13% in T medium (p-value = 1.328e-08, one-way ANOVA) (Fig 2D). This was predominantly due to an elevated maltotriose uptake (Fig 2E and Tab C in S6 Table). Notably, the fermentative capacity of these hybrids reached that of the commercial strain (Tab D in S6 Table, p-value > 0.05, one-way ANOVA). These findings strongly suggest that lines derived from hybrids generated at colder temperatures carrying S. eubayanus mitochondria and evolved in a complex maltose/maltotriose medium (T), significantly enhanced their lager fermentative capacity due to an increased maltotriose uptake during beer wort fermentation.

[END]
---
[1] Url: https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1011154

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

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