The osmoadaptive response of the wine yeast Saccharomyces cerevisiae K1-V1116 during icewine fermentation

The adapted metabolic response of commercial wine yeast under prolonged exposure to concentrated solutes present in Icewine juice is not fully understood. Presently, there is no information regarding the transcriptomic changes in gene expression associated with the adaptive stress response ofwine yeast during Icewine fermentation compared to table wine fermentation. To understand how and why wine yeast respond differently at the genomic level and ultimately at the metabolic level during Icewine fermentation, the focus ofthis project was to identify and compare these differences in the wine yeast Saccharomyces cerevisiae KI-Vll16 using cDNA microarray technology during the first five days of fermentation. Significant differences in yeast gene expression patterns between fermentation conditions were correlated to differences in nutrient utilization and metabolite production. Sugar consumption, nitrogen usage and metabolite levels were measured using enzyme assays and HPLC. Also, a small subset of differentially expressed genes was verified using Northern analysis. The high osmotic stress experienced by wine yeast throughout Icewine fermentation elicited changes in cell growth and metabolism correlating to several fermentation difficulties, including reduced biomass accumulation and fermentation rate. Genes associated with carbohydrate and nitrogen transport and metabolism were expressed at lower levels in Icewine juice fermenting cells compared to dilute juice fermenting cells. Osmotic stress, not nutrient availability during Icewine fermentation appears to impede sugar and nitrogen utilization. Previous studies have established that glycerol and acetic acid production are increased in yeast during Icewine fermentation. A gene encoding for a glycerollW symporter (STL1) was found to be highly expressed up to 25-fold in the i Icewine juice condition using microarray and Northern analysis. Active glycerol transport by yeast under hyperosmotic conditions to increase cytosolic glycerol concentration may contribute to reduced cell growth observed in the Icewine juice condition. Additionally, genes encoding for two acetyl CoA synthetase isoforms (ACSl and ACS2) were found to be highly expressed, 19- and II-fold respectively, in dilute juice fermenting cells relative to the Icewine juice condition. Therefore, decreased conversion of acetate to acetyl-CoA may contribute to increased acetic acid production during Icewine fermentation. These results further help to explain the response of wine yeast as they adapt to Icewine juice fermentation. ii

The role of Stl1p in glycerol accumulation in osmotically stressed icewine yeast Saccharomyces cerevisiae K1 V1116

The high sugar concentration in Icewine juice exerts hyperosmotic stress in the wine yeast causing water loss and cell shrinkage. To counteract the dehydration, yeast synthesize and accumulate glycerol as an internal osmolyte. In a laboratory strain of S. cerevisiae, STLl encodes for Stllp, an H+ /glycerol symporter that is glucose inactivated, but induced upon hyperosmotic stress. STLl, was found to be a highly upregulated gene in Icewine fermenting cells and its expression was 25-fold greater than in yeast cells fermenting diluted Icewine juice, making it one of the most differentially expressed genes between the two fermentation conditions. In addition, Icewine fermenting cells showed a two-fold higher glycerol production in the wine compared to yeast fermenting diluted Icewine juice. We proposed that Stllp is (1) active during Icewine fermentation and is not glucose inactivated and (2) its activity contributes to the limited cell growth observed during Icewine fermentation as a result of the dissipation of the plasma membrane proton gradient. To measure the contribution ofStl1p in active glycerol transport (energy dependent) during Icewine fermentation, we first developed an Stllp-dependent (14C]glycerol uptake assay using a laboratory strain of S. cerevisiae (BY 4742 and LiSTLl) that was dependent on the plasma membrane proton gradient and therefore energy-dependent. Wine yeast K1-Vll16 was also shown to have this energy dependent glycerol uptake induced under salt stress. The expression of STLl and Stllp activity were compared between yeast cells harvested from Icewine and diluted Icewine fermentations. Northern blot analysis revealed that STLl was expressed in cells fermenting Icewine juice but not expressed under the diluted juice conditions. Glycerol uptake by cells fermenting Icewine juice was not significantly different than cells fermenting diluted Icewine juice on day 4 and day 7 of Vidal and Riesling fermentations respectively, despite encountering greater hyperosmotic stress. Furthermore, energy- dependent glycerol uptake was not detected under either fermentation conditions. Because our findings show that active glycerol uptake was not detected in yeast cells harvested from Icewine fermentation, it is likely that Stllp was glucose inactivated despite the hyperosmotic stress induced by the Icewine juice and therefore did not play a role in active glycerol uptake during Icewine fermentation.

Real-Time Automatic Object Classification and Tracking using Genetic Programming and NVIDIA R CUDA TM

Genetic Programming (GP) is a widely used methodology for solving various computational problems. GP's problem solving ability is usually hindered by its long execution times. In this thesis, GP is applied toward real-time computer vision. In particular, object classification and tracking using a parallel GP system is discussed. First, a study of suitable GP languages for object classification is presented. Two main GP approaches for visual pattern classification, namely the block-classifiers and the pixel-classifiers, were studied. Results showed that the pixel-classifiers generally performed better. Using these results, a suitable language was selected for the real-time implementation. Synthetic video data was used in the experiments. The goal of the experiments was to evolve a unique classifier for each texture pattern that existed in the video. The experiments revealed that the system was capable of correctly tracking the textures in the video. The performance of the system was on-par with real-time requirements.