Author's School

Graduate School of Arts & Sciences

Author's Department/Program

Biology and Biomedical Sciences: Evolution, Ecology and Population Biology

Language

English (en)

Date of Award

5-24-2010

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Chair and Committee

Justin Fay

Abstract

Traditionally, Saccharomyces cerevisiae has been associated with wine, beer and bread production, yet wild strains have also been isolated from natural habitats. While all strains of S. cerevisiae as well as other Saccharomyces species are capable of wine fermentation, a genetically distinct group of S. cerevisiae strains is primarily used to produce wine. These strains exhibit an apparent genetic bottleneck, which led to the hypothesis that wine stains have been domesticated from `wild' natural strains. However, it is unknown whether the genetic bottleneck was accompanied by selection for phenotypic differences. In this study we tested for phenotypes correlated with the genetic bottleneck observed for wine strains. First, growth and fitness parameters: e.g. growth rate) of yeast strains were evaluated on different media types that simulated winemaking and natural habitats. Results provided no evidence that `wine' or `wild' strains have greater fitness in their respective environments, and suggest that the putative domestication has not resulted in habitat specific growth adaptation. Second, we tested for phenotypes associated with human perception of wine aroma and flavor characteristics using discriminatory and descriptive sensory analysis. The results from this study established human perception as a selectable yeast phenotype, and demonstrated that divergence in wine aroma and flavor attributes is consistent with the domestication hypothesis. The isolates used to infer domestication are geographically broad, but ecologically undersampled. We tested the relevance of global population genetic patterns in S. cerevisiae by conducting a population genetic study of S. cerevisiae isolated from vineyard and non-vineyard locations in North America. We used genome-wide single nucleotide markers to determine if the domestication hypothesis is supported at a local scale. Results demonstrate that two distinct populations of S. cerevisiae exist in North America, corresponding to European `wine' and North American `wild' genotypes. We provide evidence for genetic exchange between populations, suggesting a lack of physical or temporal barriers to gene flow. While wine strains exhibit a population genetic pattern consistent with previous studies, we find that the wild population is dominated by a few clonal genotypes, identifying new questions regarding the domestication hypothesis and the genetic structure of other wild populations.

Comments

Permanent URL: http://dx.doi.org/10.7936/K7XD0ZP9

Share

COinS