Combining omics data and genome-scale models to understand metabolic adaptations in lactic acid bacteria to changing environments

Lactic acid bacteria are a group of bacteria that share the characteristic of lactate fermentation, and are in particular focus of microbiological research not only because of their involvement in human health, but also due to their role in the food industry. On the one hand, they can be used as probiotics, contributing to healthy micro flora in the human body. On the other hand, they can take part in the production of fermented foods and flavour development. En- terococcus faecalis and Streptococcus pyogenes are two lactic acid bacteria that cause several infections in the human body. Therefore, they have been in the focus of clinical studies for the past few decades. The rising trend of resistance to multiple antibiotics makes the treatment of the infections caused by theses two pathogens very hard. To overcome this progressive trend of resistance, it is important to find novel drug targets in these pathogens. In the present study, I investigated the metabolic characteristics of these two pathogens using an integrative method, comprising multi-omics data integrated with the respective genome-scale metabolic models un- der the conditions comparable to different tracts in the human body. First, I investigated the effect of glutamine auxotrophy on the metabolic adjustments of E. faecalis (in the case of a ∆glnA mutant) in response to a change in environmental pH, using an integrative approach combining metabolic and proteome data with genome-scale modelling. The result suggested that the higher energy demand in the ∆glnA mutant of E. faecalis is most likely due to the lack of control on glutamine transport system as a result of the absence of glnA in the mutant. In the next part, I developed a method for functional analysis of the solution space of the genome-scale metabolic models. This method employs random perturbation to discover the reliability of flux distribution in the network. Additionally, it allows to find out which type of experimental data is most effective in limiting the solution space when the data are used as constraints. Finally, I generated tract-specific genome-scale metabolic models for E. faecalis and S. pyogenes in or- der to find tract-specific drug targets in their metabolic networks. I used multi-omics profiles (metabolic, transcriptome and proteome data) obtained under the conditions comparable to nat- ural physiological condition in the human body, namely root canal, unrinary tract and plasma, and used the data to constraint the respective genome-scale metabolic models. The models were used to find potential drug targets using different levels of threshold for metabolic flux values and growth rate of the bacteria. The results suggested that there exist potential drug targets in different subsystems in the metabolic network, from central carbon metabolism to transport system. The presented profiles of drug targets have to be validated experimentally in order to be used for the development of new treatment approaches