Genomic and Phenotypic Studies of Traits Affecting Cold Tolerance in Natural Yeast Species and Hybrids

Abstract

Over the past decades, scientific communities have put a significant amount of interest in genetic variation within and between Saccharomyces species and how it can be used for industries. In this thesis, the effect of inter and intraspecific variation on the phenotypes of natural yeast species and hybrids is explored. To study intraspecific variation, fitness profiles of S. cerevisiae strain Σ1278b were constructed and compared with BY4743 from previous studies under nutrient starvation and temperature stress. It was found that genes involved in genome integrity maintenance and general metabolism are important for the growth of Σ1278b in complete and nutrient limited media, respectively. Moreover, the results also show that there is little fitness variation between the two yeast strains. Given that, less conserved genes are likely to produce background-specific phenotypes. To study interspecific variation, experiments were carried out on independent lineages of a cross between S. cerevisiae (Sc) and S. uvarum (Su). These hybrids were constructed to resemble the natural hybrid S. pastorianus at the genome level. To examine the effect of chimeric protein complexes, hybrids that carry mitochondrial DNA (mtDNA) from S. cerevisiae were selected. It was found that combining GCN1 from S. cerevisiae and GCN20 from S. uvarum appeared to make the hybrids more robust under amino acid starvation and cold stress condition. In addition, it was found that variation in phenotypic traits could be caused by structural differences of the protein GCN1 and differential transcriptional regulation of the gene GCN20. Interestingly, this may explain biased retention of this protein complex in the natural hybrid S. pastorianus. To study another aspect of interspecific variation, spot assays and RNA sequencing were carried out to determine the effect of mitotypes on the expression of nuclear genes in yeast hybrids. The results show that fitness is affected heavily by mitotypes as hybrids with mtDNA from S. cerevisiae is more favourable at warm and vice versa for mtDNA from S. uvarum. In addition it was found that mitotypes appear to have an influence on allelic expression bias in several biological pathways. This is true especially in respiration as the expression of orthologs of this process from Sc and Su subgenomes was suppressed in complete medium at warm for Sc mtDNA and at cold for Su mtDNA, respectively. And finally, as a side project, an open-source application, called gcatR, is developed to analyse growth curve data. This application is created in R language and contains several features including interactive data table, growth curve parameter estimation, time series clustering and classification, and fitting based on generalised additive models. What’s more, it creates a set of diagrams (such as line graph, bar chart and heatmap) to represent the obtained results. More importantly, this application is designed to be user friendly and does not require users to have prior knowledge in any programming language. gcatR is provided free of charge and can be accessed at http://www.gcatr.manchester.ac.uk/ or downloaded to install on a local Windows device.

Date of Award	31 Dec 2018
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Daniela Delneri (Supervisor) & Jean-Marc Schwartz (Supervisor)