Host-strain compatibility shapes host transcriptional responses in macrophage in vitro infection models of Mycobacterium tuberculosis.

Differences in host distribution are observed across Mycobacterium tuberculosis complex (MTBC) lineages, but investigating the molecular basis of host compatibility in vivo is challenging. Here, we establish an epidemiology-guided in vitro framework to operationalize 'host compatibility' as a theoretical parameter that can be approximated experimentally using macrophage infection readouts. We infected two widely used but biologically distinct models, human THP-1-derived macrophages (THP-1m) and bovine macrophages (BoMac), with four MTBC human and animal strains. We quantified infected-cell frequency, intracellular bacterial burden dynamics, cell death and host transcriptome by RNA-seq. Across both models, infection phenotypes were concordant with the a priori epidemiology-informed reference pairings: the most compatible strain showed a higher late intracellular bacterial burden in its respective host, and the corresponding transcriptomes were enriched for replication, repair and cell-cycle signatures, whereas the opposite pairings showed lower bacterial burdens and stronger enrichment for endosomal, antigen-presentation and immune-signalling pathways in each cell type. Given the role of phospholipase C and virulence, we assessed plc expression and performed pharmacological inhibition experiments. These analyses revealed lineage- and macrophage-context-dependent expression patterns and reduced intracellular burden and cytotoxicity in THP-1m but not in BoMac. This framework links MTBC strain diversity to macrophage infection phenotypes and prioritizes candidate pathways potentially involved in bacterial-host interactions.