Abstract Faithful mimics of tuberculosis (TB) infection are needed to provide mechanistic insights into the complex host-pathogen interactions and accelerate drug discovery. Current in vitro models only allow short investigation durations, present divergent transcriptional signatures to human infections, and are unreliable drug discovery platforms. We developed a 3D collagen culture system mimicking the lung microenvironment (collagen fibres, pore size and stiffness), where we incorporated Mycobacterium tuberculosis (Mtb) infected human THP-1 or primary monocytes. Dual RNA-sequencing revealed high mammalian gene expression similarity with patient samples compared to 2D macrophage infections. Similarly, gene expression of bacteria was much more representative to in vivo gene expression compared to bacteria in 2D cultures (114 genes in 3D vs 21 genes in 2D). Key phenotypes observed in humans, such as foamy macrophages and mycobacterial cords (never seen in any other in vitro culture system), were reproduced in our model. Our system overcomes many challenges associated with the traditional platforms, including showing remarkable efficacy with clinically relevant concentrations of first-line anti-TB drug pyrazinamide, not seen in any other in vitro model, making it reliable, readily adoptable for tuberculosis studies and drug screening. Significance statement Mtb is a slow-growing pathogen which modulates host response over time. The current in vitro platforms offer a very short study duration to study, are unreliable as drug discovery platforms, and the phenotypic and genotypic traits of the host and pathogen differ. The collagen-I hydrogel culture system developed in this study addresses these challenges by successfully recapitulating several key phenotypes observed in human infections. Dual RNA sequence also showed excellent gene expression similarities for both the host and the bacteria. Furthermore, remarkable efficacy with the antibiotic Pyrazinamide was demonstrated, a first for in vitro cultures despite over 50 years of clinical use of the drug. We expect our platform to be exploited widely for drug discovery and understanding host-pathogen interactions.
