Genome-wide transcriptional landscape of Mycobacterium tuberculosis during acute lung infection

Abstract

Tuberculosis (TB) remains a major global health burden, yet the mechanisms by which Mycobacterium tuberculosis (Mtb) adapts to host environments to drive disease pathology are incompletely defined. A key limitation has been reliance on axenic culture systems that fail to recapitulate the complex, host-imposed stresses encountered by Mtb in vivo. Here, we report the first microarray-based genome-wide transcriptomic profiling of Mtb in rabbit lungs with active TB, which closely mirrors human disease features, including granuloma heterogeneity, necrosis, and cavitation. Using Mtb RNA isolated from infected lung homogenates or broth-culture, we capture bacterial transcriptional states shaped by the host microenvironments. The transcriptional data analyses reveal extensive, context-dependent reprogramming of Mtb metabolic, respiratory, and stress-response networks that diverges markedly from in vitro expression profiles, including activation of stress adaptation, lipid catabolism, nucleic acid metabolism, and transcriptional regulation pathways. These data uncover pathways and networks that are selectively engaged in vivo and likely critical for Mtb survival within granulomatous lesions. Our findings demonstrate that transcriptional states most relevant to TB pathogenesis are underrepresented in standard lab-grown Mtb models and highlight the importance of in vivo bacterial profiling. By characterizing Mtb gene expression within diseased lungs, this study provides a systems-level framework for understanding TB pathogenesis and reveals in vivo-essential pathways, offering potential targets for translational drug discovery and the development of more effective anti-TB therapies.