Genome-Scale Metabolic Modeling of Lipid Metabolism in Mycobacterium tuberculosis Clinical Strains from Nigeria for Host-Directed Metabolic Intervention Target Discovery

Mycobacterium tuberculosis exploits host lipid metabolism, particularly cholesterol and fatty acid catabolism, as primary carbon and energy sources during chronic infection of macrophages. Genome-scale metabolic models (GEMs) represent comprehensive computational representations of an organism's metabolic network, enabling flux balance analysis and targeted prediction of essential reactions amenable to metabolic intervention. No GEM has been constructed or applied to M. tuberculosis clinical strains from Nigeria, where locally circulating lineages may harbor metabolic network variations influencing drug susceptibility. This dissertation develops and applies genome-scale metabolic models of lipid metabolism to M. tuberculosis clinical strains isolated from newly diagnosed drug-susceptible and drug-resistant TB patients across Northeast Nigeria, covering Borno, Adamawa, and Yobe States. Whole-genome sequences of 50 clinical M. tuberculosis isolates were generated on Illumina MiSeq. Strain-specific GEMs were constructed by integrating annotated genome content with the reference iEK1011 M. tuberculosis GEM framework, using ModelSEED for gap-filling. Constraint-based flux balance analysis was performed using COBRApy in Python. Comparative lipid pathway flux profiling between drug-susceptible and drug-resistant isolates was conducted. Flux variability analysis identified 47 reactions with significantly differential flux ranges between drug-susceptible and multi-drug-resistant (MDR) strain GEMs, concentrated in the methylmalonyl-CoA pathway and propionate catabolism. In silico gene deletion analysis identified 14 genes essential specifically in MDR strains and absent from drug-susceptible essential gene sets, representing synthetic lethality candidates. Three of these candidates (fadA5, prpC, sdhA) encode enzymes with existing structural data and druggable active sites, prioritized for experimental validation. These GEM-derived discoveries represent original metabolic drug target predictions specific to Northeast Nigerian MDR M. tuberculosis strains. Keywords: genome-scale metabolic model, Mycobacterium tuberculosis, lipid metabolism, drug resistance, Northeast Nigeria.