Mapping tuberculosis granulomas in human tissue

Tuberculosis and related mycobacterial infections leave a distinctive mark in the body called a granuloma: a tightly organized cluster of immune cells that both contains bacteria and can help the disease persist. Understanding how those structures are arranged and how the different parts behave has been a major challenge because granulomas are complex and vary from patient to patient. To tackle that problem, a team led by Xun Jiang applied high-resolution spatial RNA sequencing to samples from 38 patients with mycobacterial disease. Instead of analyzing mixed tissue as a whole, spatial RNA sequencing keeps the location of gene activity in the tissue, letting researchers see which cells and signals are present in each part of a lesion. Using this approach, the team set out to describe the spatial and functional organization of granulomas in human disease, compare lesions caused by Mycobacterium tuberculosis (MTB) and non-tuberculous mycobacteria (NTM), examine how lymph node granulomas differ, and compare these mycobacterial lesions with those found in sarcoidosis. The study produced a detailed map of where immune activity, tissue remodeling, and antigen presentation occur inside granulomas.

The core methods and major findings come directly from the spatial RNA sequencing analysis of 38 patient samples. The study identified five distinct granuloma niches: a necrotic core, an immune-activated inner niche, an inflammatory and extracellular matrix (ECM)-remodeling middle niche, an outer structural niche, and a tertiary lymphoid structure niche that supports antigen presentation. Patterns of gene activity showed that immune activity peaks in the inner niche and then gives way to fibrosis toward the granuloma periphery. When granulomas were located in lymph nodes, they showed reduced involvement of fibroblasts but stronger activation of the JAK-STAT pathway. Granulomas from Mycobacterium tuberculosis (MTB) cases showed heightened JAK-STAT and IFN-γ signaling, while those from non-tuberculous mycobacteria (NTM) cases carried stronger hypoxia signatures. Compared with sarcoidosis lesions, mycobacterial granulomas featured a more structured adaptive immune response, including clustering of plasma cells. The data and interactive atlas are made available at https://lab-li.ciim-hannover.de/mb-granuloma/.

These findings provide a clearer picture of the spatial architecture and functional programs inside human granulomas and define specific disease signatures that could be used as biomarkers or treatment targets. By showing where immune activation, antigen presentation, fibrosis, and hypoxia occur within lesions, the study points researchers toward the most relevant cells and signaling pathways for future work. The differences between MTB and NTM granulomas—stronger JAK-STAT and IFN-γ signaling in MTB, and more hypoxia in NTM—suggest that diagnostics or therapies might need to be tailored to the type of mycobacterial infection. Likewise, the reduced fibroblast involvement but stronger JAK-STAT activation in lymph node granulomas highlights how location matters for disease biology. Because the atlas is publicly available, clinicians and scientists can explore the spatial signatures directly to guide biomarker discovery and to design therapeutic strategies that target the right niche within a granuloma.