A Quasi-Domesticate Relic Hybrid Population of Saccharomyces cerevisiae × S. paradoxus Adapted to Olive Brine

Abstract

The adaptation of the yeast Saccharomyces cerevisiae to man-made environments for the fermentation of foodstuffs and beverages illustrates the scientific, social, and economic relevance of microbe domestication. Here we address a yet unexplored aspect of S. cerevisiae domestication, that of the emergence of lineages harboring some domestication signatures but that do not fit completely in the archetype of a domesticated yeast, by studying S. cerevisiae strains associated with processed olives, namely table olives, olive brine, olive oil, and alpechin. We confirmed earlier observations that reported that the Olives population results from a hybridization between S. cerevisiae and S. paradoxus. We concluded that the olive hybrids form a monophyletic lineage and that the S. cerevisiae progenitor belonged to the wine population of this species. We propose that homoploid hybridization gave rise to a diploid hybrid genome, which subsequently underwent the loss of most of the S. paradoxus sub-genome. Such a massive loss of heterozygosity was probably driven by adaptation to the new niche. We observed that olive strains are more fit to grow and survive in olive brine than control S. cerevisiae wine strains and that they appear to be adapted to cope with the presence of NaCl in olive brine through expansion of copy number of ENA genes. We also investigated whether the S. paradoxus HXT alleles retained by the Olives population were likely to contribute to the observed superior ability of these strains to consume sugars in brine. Our experiments indicate that sugar consumption profiles in the presence of NaCl are different between members of the Olives and Wine populations and only when cells are cultivated in nutritional conditions that support adaptation of their proteome to the high salt environment, which suggests that the observed differences are due to a better overall fitness of olives strains in the presence of high NaCl concentrations. Although relic olive hybrids exhibit several characteristics of a domesticated lineage, tangible benefits to humans cannot be associated with their phenotypes. These strains can be seen as a case of adaptation without positive or negative consequences to humans, that we define as a quasi-domestication.

Keywords: Saccharomyces cerevisiae; hybridization; microbe population genomics; microbiology of olive brine; yeast.

FIGURE 1

Hybrid olive strains form a monophyletic group and have a similar genomic organization. (A) Phylogenetic placement of hybrid olive strains among the known lineages of Saccharomyces cerevisiae (MO, Mediterranean oaks; NA-JP, North America – Japan). Whole-genome phylogenetic tree constructed after discarding S. paradoxus regions in all genomes. The phylogeny was inferred from of 93 sequences and 913590 SNPs using the Maximum Likelihood method as implemented in IQ-TREE with the TVM+F+G4 model of sequence evolution and was rooted with S. paradoxus. Branch lengths correspond to the expected number of substitutions per site and black dots in tree nodes depict bootstrap support values above 85% (1000 replicates). Strains isolated from the olive niche are distinguished based on the specific isolation source (see color codes). (B) Similar divergence plots of the genomes of selected hybrid strains (highlighted in the phylogeny) to the reference genome of S. paradoxus CBS 432. The dotted lines depict the 10% divergence threshold that represents the average divergence between S. cerevisiae and S. paradoxus. The substantially distinct divergence plot of a Brazilian S. cerevisiae × S. paradoxus hybrid strain (UFMG-CM-Y651) previously reported by us (Barbosa et al., 2016) is included for comparison.

FIGURE 2

Gene ontology of S. paradoxus genes found in hybrid strains. (A) Gene ontology terms with p-value < 0.01 and organized under “Component,” Process,” and “Function” for the 103 S. paradoxus genes shared between the 23 hybrid genomes analyzed. The size of the circles is proportional to the number of genomes that contribute for that term. (B) Number of S. paradoxus genes and number of strains by GO (gene ontology) term.

FIGURE 3

Whiskers plots of the relative fitness (growth and survival) in olive brine of representatives of the Olives (AP 5.4, AP 7.2, YO 654, ZIM 2580, PYCC 4891, and PYCC 6732) and Wine (AWRI 1631, Lalvin W15, PR, PYCC 4072, TUM V1, and Uvaferm VRB) populations of S. cerevisiae. The results are based on counts of colony forming units/ml of 6 strains from each group inoculated individually in two duplicate and independent experiments.

FIGURE 4

Comparison of glucose (initial concentration 0.6% w/v) and fructose (initial concentration 0.1% w/v) consumption by representatives of the Olives and Wine population of S. cerevisiae in different conditions. (A) Olive brine. (B) Phosphate buffer supplemented with 8% w/v NaCl. (C) Phosphate buffer supplemented with 8% w/v NaCl and 0.1% w/v yeast extract.

FIGURE 5

Copy number variation (CNV) of ENA genes and growth rates in the presence of NaCl of strains from the Olives and Wine populations. (A) Violin plots describing the number of ENA genes (ENA1, ENA2 and ENA5, average value for each strain) among olive brine, intestinal tract, olive oil-alpechin, and wine strains (black circles indicate the median within each group). (B) Violin plots of relative growth rates in the presence of 6 and 8% (w/v) NaCl (reference: medium without NaCl) among olive brine, intestinal tract, olive oil-alpechin, and wine strains. (C) Numbers of ENA copies shown in tabular format for each strain. Darker green color shades correspond to increased numbers of gene copies. CNV of actin (ACT1) is indicated as reference. Statistical significant differences of CNV between groups of strains are highlighted.

FIGURE 6

Model for the origin of relict S. cerevisiae hybrids as a consequence of homoploid hybridization between a S. cerevisiae wine strain and a S. paradoxus member of the European population, followed by adaptive LOH corresponding to a massive loss of the S. paradoxus sub-genome.