A combined systems biology and genomics approach to the study of metabolism in Kluyveromyces marxianus

English: The yeast Kluyveromyces marxianus has become an important micro-organism for industrial applications, as have other non-conventional yeasts. It has the advantages over Saccharomyces cerevisiae (baker’s yeast) in that it is more thermotolerant, has a much higher growth rate and can utilise a wider range of sugars, including the pentose D-xylose, which is found abundantly in lignocellulosic biomass. Although considerable advances have been made in engineering S. cerevisiae strains to ferment pentose sugars, their performance in this respect still does not approach that of glucose fermentation. S. cerevisiae is the model Crabtree positive yeast, meaning that it naturally ferments glucose even if oxygen is present at a high level. Crabtree negative yeasts, such as K. marxianus, have to be forced into a fermentative metabolism by imposing oxygen-limited conditions, which is impractical on industrial scale. Thus, a tremendous amount of knowledge needs to be gained regarding the regulation of metabolism in this non-conventional yeast before success could be expected in the re-programming of K. marxianus strains into xylose fermenting, Crabtree positive strains. The challenge of bringing a non-model species such as K. marxianus to the point of identifying key regulators affecting central metabolic pathways seems formidable. The aims of this work was to firstly harness the new technology of next-generation sequencing (NGS) to create a first draft genome for K. marxianus strain UFS-Y2791 and to generate high-quality RNA-seq differential transcriptome datasets, simultaneously capturing a tremendous amount of information. Efficient analytical methods and software implementations were also developed to explore these large datasets in an efficient manner, revealing new insights into the response of this species to glucose and xylose as carbon sources. RNA-seq data revealed a striking resemblance with the pattern of glucose derepression in the xylose medium, with up-regulation of genes for alternative carbon source utilisation, especially in the peroxisomes. Subsequently, two independent approaches were taken to identify differentially active transcription factors regulating the response. The first was the enumerative method of heptamer frequency comparisons, revealing the most likely regulators of differentially expressed genes. Secondly, a likelihood statistical approach was designed that employs multiple sources of evidence, which resulted in the construction of the first genome-wide gene regulatory network for K. marxianus. The method bridges the gap between the new NGS-based methods, which can rapidly generate data on any non-model species, and the wealth of experimental data that exist for a model species such as S. cerevisiae. Gene set enrichment statistics of the transcription factor target sets showed a general pattern that the activities of differentially active transcription factors were regulated primarily by post translational modifications instead of gene regulation. The use of RNA-seq was further expanded to the elucidation of the kinases that regulate transcription factors. The chromosomal context of differential gene expression was also investigated. Clusters of genes were identified, similar to the sub-telomeric regions previously identified in S. cerevisiae, but not close to telomeres. These regions contain industrially important enzymes and the potential binding sites for differentially active transcription factors. Finally, the possible roles of cofactor balances were investigated. Flux balance analysis was demonstrated here in rationalising the genetic response observed in RNA-seq transcriptomics and to understand the complex interplay between ATP, NADPH and NADH, the cofactor specificity of the oxidative pentose phosphate pathway, as well as the role of fructose-1,6-bisphosphatase. New roles are proposed for the latter enzyme, which differs from the currently accepted norm. A strategy for the metabolic engineering of a future xylose fermenting K. marxianus strain is also presented. The integrated analysis presented here expands our knowledge base of this yeast species, which is set to become increasingly important in a future bio-economy.