Abstract
A significant increase in the incidence of liver cancer is projected over the next few decades with limited therapeutic options for patients. Like many other cancer types, liver cancer can be promoted by “Western-style” diets (WD) that are high in fat and processed sugars, such as fructose. Studies indicate that chronic WD consumption can lead to systemic dysregulation of insulin signalling and lipid metabolism that increase the risk for development of liver diseases such as non-alcoholic fatty liver disease. Liver tissue damage and the ensuing inflammation has been shown to promote mutations that can lead to oncogenic transformation in hepatocytes that lead to the formation of malignancies. A particular challenge in understanding the dietary impact on liver and tumour metabolism is how dietary nutrients fuel metabolic pathways and how this synergises with gene expression changes caused by cell-autonomous and systemic signals because of diet. In order to address this it is necessary survey metabolic fluxes on a global scale and genome scale metabolic models (GSMMs) used in systems biology provide rigorous mathematical frameworks for this purpose.To study how diet impacts liver and liver tumour metabolism, a new mouse GSMM, Mouse Metabolic Reaction Network (MMRN), was reconstructed using orthology between mouse and human. MMRN was constrained with gene expression data from the tissues of a carcinogen-induced mouse model of liver cancer. By using in silico constraint-based modelling approaches, the effect of gene expression and dietary composition, alone and in combination, on liver and tumour metabolism were investigated. The WD was shown to lead to distinct metabolic phenotypes irrespective of gene expression, but also, that gene expression in the tumour drive flux through specific metabolic pathways compared to liver tissue. These observations were validated experimentally. A novel computational approach, termed Systematic Diet Composition Swap (SyDiCoS), was developed to investigate the impact of WD by swapping out individual nutrients to its corresponding composition in a CD and investigate its impact on metabolic flux. This approach made it possible to deconvolute the effect of specific dietary nutrients on metabolic flux in a given genetic background. SyDiCoS showed that carbohydrates and lipids in WD increases glycerol and succinate production, respectively, and that both these nutrient classes are required to increase biomass production in the tumour.
Using dietary content as input for tissue GSMMs is the state of the art in metabolic modelling and has been shown to result in accurate model predictions. However, dietary nutrients are first metabolised by microorganisms in the gut, and the diet itself has a significant impact in the composition of the gut microbiome that, in turn, effects its functional metabolic capability.
To this end, metagenomics data were used to reconstruct gut microbiome community GSMMs to study how the gut microbiome composition impacts gut metabolism under different dietary regiments. Using this approach together with SyDiCoS and systematic removal of specific microorganisms from the community, the interplay between diet and gut microbiome composition on metabolism was investigated. Finally, community GSMMs and tissue GSMMs were integrated into a single multi-tissue GSMM to establish a diet-microbiome-liver metabolic axis. Several novel and previously identified metabolic interactions between the microbiome and the liver as well as tumour-host metabolic interactions were identified. The diet was shown to influence similar pathways as in single-tissue GSMMs, but its integration into a multi-tissue GSMM effected exchanges of metabolites involved in these pathways. This indicated that the integration of tissues into a multi-tissue framework impacts predictions made in single-tissue GSMM predictions.
| Date of Award | 1 Jan 2023 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Saeed Shoaie (Supervisor) & Dimitrios Anastasiou (Supervisor) |