Metabolic modeling with reverse mapping links host perturbations to Mycobacterium tuberculosis

Understanding host-pathogen interactions during Mycobacterium tuberculosis (Mtb) infection has largely relied on identifying bacterial genes essential for intracellular survival. However, conventional reverse genetics approaches often overlook host pathways that regulate Mtb growth. Here, we developed a backtracing strategy to link host metabolic modulators with specific Mtb proteins. A genome-scale host metabolic model integrated with lung transcriptomic data from H37Rv-infected mice predicted 18 host proteins essential for Mtb survival. We validated this through pharmacological inhibition, showing that the identified host metabolic regulators reduce the intracellular survival of wild-type H37Rv. Next, we constructed a host-pathogen protein interaction network that linked them to 9 Mtb proteins. Among them, mce3E , fadA2 , and ptpA were identified through knockdown (KD) experiments as host-response modulators that favor pathogen growth. Proteomic and metabolomic analyses revealed their underlying molecular mechanisms. This was evaluated experimentally by modulating the corresponding host metabolic regulators, thereby restoring intracellular growth defects in Mtb KD strains and uncovering previously unrecognized host-pathogen survival mechanisms. Exogenous PGE2 treatment further suppressed intracellular bacterial survival.