One-third of the world's population is estimated to be latently infected with Mycobacterium tuberculosis. This reservoir of bacteria is largely resistant to antimicrobial treatment that often only targets actively replicating mycobacteria, with current treatment for latent infection revolving around inhibiting the resuscitation event rather than preventing or treating latent infection. As a result, antimicrobials that target latent infection often have little to no activity in vivo. Here we report a method of in vitro analysis of physiologically relevant non-replicating persistence (NRP) utilizing cholesterol as the sole carbon source, alongside hypoxia as a driver of Mycobacterium bovis BCG into the NRP state. Using the minimal cholesterol media NRP assay, we observed an increased state of in vitro resistance to front-line anti-tubercular compounds. However, following a phenotypic screen of an approved-drug library, we identified dapsone as a bactericidal active molecule against cholesterol-dependent NRP M. bovis BCG. Through an overexpression trial of probable antimicrobial target enzymes, we further identified FolP2, a non-functional dihydropteroate synthase homologue, as the likely target of dapsone under cholesterol-NRP due to a significant increase in bacterial resistance when overexpressed. These results highlight the possible reason for little in vivo activity seen for current front-line anti-NRP drugs, and we introduce a new methodology for future drug screening as well as a potential role for dapsone inclusion within the current treatment regime.
Keywords: cholesterol metabolism; dapsone; latent tuberculosis infection; non-replicating persistence; tuberculosis.