Back

Single-Cell and Spatial Transcriptomic Analyses Deciphering the Three-Layer Architecture of Human Tuberculosis Granulomas

Yu, X.; Wang, J.; Wang, P.; Liu, X.; Li, C.; Ju, Y.; Liu, S.; Dong, Y.; Wang, J.; Tuohetaerbaike, B.; Wen, H.; Zhang, W.; Niu, H.; Xu, S.; Jiang, C.; Jiang, X.; Wu, J.; Huang, H.; Chen, F.

2024-07-15 microbiology
10.1101/2024.07.15.603490 bioRxiv
Show abstract

BackgroundGranulomas (defining tuberculosis histopathological feature) are central to the hosts defense against Mycobacterium tuberculosis, critically influencing patient outcomes. However, knowledge of human granulomas structure and function are incomplete. This study employs single-cell and spatial transcriptomics to dissect human granulomas cellular composition, structure, communication and function from 19 pulmonary, lymphatic and skeletal samples. ResultsOur study identified nine key immune-activated/signaling-active cell clusters. Notably, we delineated a three-layered granuloma structure: a core with macrophages (Macro-c09, Macro-c10) and occasional fibroblasts (Fib-c03); a fibroblast-rich (Fib-c01) periphery; and an immune-infiltrated intermediate layer comprising diverse immune-cells recruited by strong signaling-molecules (SPP1/MIF) from core/periphery cells. This study also shows granuloma heterogeneity across individuals and tissues. ConclusionsBy merging scRNA-seq with ST-seq, we offer an intricate single-cell perspective of granulomas spatial-structure and formation mechanisms, identify signaling-molecules and significantly changed genes as potential targets for host-directed tuberculosis immunotherapy, highlight fibroblasts crucial role in granuloma formation, and provide an important reference/improved understanding of TB.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.