Phylogenomic Analysis of Oenococcus oeni Reveals Specific Domestication of Strains to Cider and Wines.
Bottom Line: A study on the orthologs and single nucleotide polymorphism contents of the genetic groups revealed that the domestication of some strains to products such as cider, wine, or champagne, is reflected at the genetic level.While group A strains proved to be predominant in wine and to form subgroups adapted to specific types of wine such as champagne, group B strains were found in wine and cider.The results suggest that ancestral O. oeni strains were adapted to low-ethanol containing environments such as overripe fruits, and that they were domesticated to cider and wine, with group A strains being naturally selected in a process of further domestication to specific wines such as champagne.
Affiliation: Univ. Bordeaux, ISVV, EA 4577 Œnologie, Villenave d'Ornon, France Research and Innovation Centre, Fondazione Edmund Mach, San Michele all'Adige, Italy.Show MeSH
Related in: MedlinePlus
Mentions: The SNP content of the genomes was analyzed to further investigate the population structure of O. oeni. Mapping all the genomes against the complete genome of strain PSU-1 revealed 47,621 SNP positions and a total of 48,230 alleles. A concatenated sequence of 47,621 bp was produced for each strain by extracting the alleles of all SNPs positions and the 50 sequences were used to reconstruct an unrooted tree by the neighbor joining method (fig. 2C). This tree has a slightly different topology from that of the MLST. Although they both agree in their two major branches A and B, the tree generated from SNPs clearly excludes strain IOEB_C52 from all rest, suggesting that this strain might actually be part of a third group C. Bootstrap values show a far more consistent tree than the one previously made by MLST. The fore mentioned trees are consistent with the results of previous studies (Bilhère et al. 2009; Borneman et al. 2012a), except for the newly sequenced strain IOEB_C52 that might be part of a genetic group that has not yet been described. SNP data was further processed by Structure software to infer the number of populations detected among the 50 strains. Structure is suited for inferring population structure since it works by probabilistically assigning individuals to populations by characterizing their allele frequencies at each locus. This method can be more reliable than distance-based methods such as neighbor-joining trees which do not let incorporate additional information, so they are more suited for exploratory analysis than for statistical inference (Pritchard et al. 2000). The result confirmed the presence of two populations corresponding to strains from groups A and B plus a third population represented by strain IOEB_C52 alone (fig. 3). For both A and B populations there is at least 70% of genetic contribution from their own group, and 0% to almost 25% contribution from group C. Strain IOEB_C52, the only individual of C group, has more than 80% of group C contribution and most of the contribution of the rest comes from B (fig. 3).Fig. 3.—
Affiliation: Univ. Bordeaux, ISVV, EA 4577 Œnologie, Villenave d'Ornon, France Research and Innovation Centre, Fondazione Edmund Mach, San Michele all'Adige, Italy.