Systemic immune dysregulation in severe tuberculosis patients revealed by a single-cell transcriptome atlas

J Infect. 2023 May;86(5):421-438. doi: 10.1016/j.jinf.2023.03.020. Epub 2023 Mar 30.


Tuberculosis (TB), caused by Mycobacterium tuberculosis (Mtb) infection, is currently the deadliest infectious disease in human that can evolve to severe forms. A comprehensive immune landscape for Mtb infection is critical for achieving TB cure, especially for severe TB patients. We performed single-cell RNA transcriptome and T-cell/B-cell receptor (TCR/BCR) sequencing of 213,358 cells from 27 samples, including 6 healthy donors and 21 active TB patients with varying severity (6 mild, 6 moderate and 9 severe cases). Two published profiles of latent TB infection were integrated for the analysis. We observed an obviously elevated proportion of inflammatory immune cells (e.g., monocytes), as well as a markedly decreased abundance of various lymphocytes (e.g., NK and γδT cells) in severe patients, revealing that lymphopenia might be a prominent feature of severe disease. Further analyses indicated that significant activation of cell apoptosis pathways, including perforin/granzyme-, TNF-, FAS- and XAF1-induced apoptosis, as well as cell migration pathways might confer this reduction. The immune landscape in severe patients was characterized by widespread immune exhaustion in Th1, CD8+T and NK cells as well as high cytotoxic state in CD8+T and NK cells. We also discovered that myeloid cells in severe TB patients may involve in the immune paralysis. Systemic upregulation of S100A12 and TNFSF13B, mainly by monocytes in the peripheral blood, may contribute to the inflammatory cytokine storms in severe patients. Our data offered a rich resource for understanding of TB immunopathogenesis and designing effective therapeutic strategies for TB, especially for severe patients.

Keywords: Cytokine storm; Immunological responses; Mycobacterium tuberculosis; ScRNA-seq; Severe patients; Tuberculosis.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Humans
  • Killer Cells, Natural
  • Latent Tuberculosis*
  • Mycobacterium tuberculosis*
  • Transcriptome
  • Tuberculosis*