Abstract

Eosinophil infiltration into the lungs is frequently related to type II reactions during allergic reactions and fungal and parasite diseases. However, in humans, macaques, and mice, type I inflammatory responses to Mycobacterium tuberculosis (Mtb) result in eosinophil accumulation in lung lesions, which supports host resistance. Here, we demonstrate that eosinophils enter the lungs of mice and macaques as soon as one week following Mtb exposure [1]. In mice, this influx is not dependent on CCR3, but rather needs the highly expressed oxysterol receptor GPR183 to be expressed intracellularly expressed on eosinophils from humans and macaques. Eosinophil recruitment is compromised in mice lacking the oxysterol-synthesizing enzyme Ch25h because of the direct interaction between murine eosinophils and bacilli-filled alveolar macrophages, which upregulate Ch25h [2]. Our research demonstrates that eosinophils are among the first circulation-derived cells to detect and react to Mtb infection of alveolar macrophages, and it implicates GPR183 in eosinophil migration into lung tissue.

As eosinophil migration into the lung parenchyma peaked around d14, we proposed that recruited eosinophils may be able to detect Mtb or tissue-dwelling Mtb-infected cells or perhaps interact with them. We observed dynamic interactions between eosinophils and Mtb-containing cells using live imaging of ex vivo lung explants from eosinophil EPX-reporter mice that were Mtb-cyan fluorescent protein (CFP)-infected. In the alveoli, we saw eosinophils directly engaging with Mtb or Mtb-infected cells by approaching, sluggishly approaching, and extending pseudopods toward bacilli .The frequency of Mtb-positive eosinophils was subsequently evaluated by single-cell analysis of fluorescent Mtb-mCherry and found to be significantly higher than background in a non-phagocytic SSClow, MHCIIneg,CD68neg, CD11bneg, Ly6Gneg control population (average 0.15%).