The correlation between immune profiles and pathological changes in pulmonary tuberculosis granulomas revealed by bioinformatic analysis and experimental validation.

Most of Mycobacterium tuberculosis(Mtb) infection result in the formation of granulomas, which are often rich in immune cells, with subsequent clinical symptoms. However, the role of the immune system in the formation of tuberculosis granuloma structures has not been fully revealed. Here we first analyzed single-cell transcriptome and microenvironment spatial characteristics to reveal the contribution of immune cells to granuloma expansion with validation by immunofluorescence. We then integrated published peripheral blood transcriptome data for Mtb-infected patients and healthy controls. Immune cell profiles were deconvoluted and results were validated on a local cohort using flow cytometry. At the same time, an in-depth evaluation of the changes in the population and function of multiple peripheral blood immune cells during tuberculosis infection were conducted to define correlation with granuloma area. Finally, we screened 6 cytokines (IL6, IL8, IL10, IFNγ, TNFα, TGFβ) through machine learning bioinformatics and analyzed their correlation with the size of tuberculosis granuloma. Based on these findings, we confirmed that the dynamic variation in proportion of immune cells in peripheral blood and the levels of cytokine profiles are closely related to the occurrence and development of tuberculosis granuloma. This study provides a theoretical basis for the molecular mechanism of tuberculosis granuloma.