ATG5 promotes eosinopoiesis but inhibits eosinophil effector functions

Abstract

Eosinophils are white blood cells contributing to the regulation of immunity and they are involved in the pathogenesis of numerous inflammatory diseases. In contrast to other cells of the immune system, no information is available about the role of autophagy in eosinophil differentiation and functions. In order to study the autophagic pathway in eosinophils, we generated conditional knockout mice in which Atg5 is deleted within the eosinophil lineage only (designated Atg5eoΔ mice). Eosinophilia was provoked by crossbreeding Atg5eoΔ mice with Il5 (IL-5) overexpressing transgenic mice (designated Atg5eoΔIl5tg mice). Deletion of Atg5 in eosinophils resulted in a dramatic reduction in the number of mature eosinophils in blood and in an increase of immature eosinophils in the bone marrow. Atg5-knockout eosinophil precursors exhibited reduced proliferation under both in vitro and in vivo conditions, but no increased cell death. Moreover, reduced differentiation of eosinophils in the absence of Atg5 was also seen in mouse and human models of chronic eosinophilic leukemia. Atg5-knockout blood eosinophils demonstrated augmented levels of degranulation and bacterial killing in vitro. Moreover, in an experimental in vivo model, we observed that Atg5eoΔ mice achieve better clearance of the local and systemic bacterial infection with Citrobacter (C.) rodentium. Evidence for increased degranulation of ATG5low-expressing human eosinophils was also obtained in both tissues and blood. Taken together, mouse and human eosinophil hematopoiesis and effector functions are regulated by ATG5 which controls the amplitude of overall antibacterial eosinophil immune responses.

Regulation of eosinophil functions by autophagy

“…Similar results were obtained in EoL-1 cells, a model of an established human CEL. Upon induced differentiation of EoL-1 cells, significantly lower levels of surface markers CD11b, Siglec-8 and CCR3 were observed in ATG5-knockout EoL-1 cells, indicating decreased maturation of eosinophil precursors [89] (Fig. 2).…”

“…Cre recombinase is expressed only after commitment to the eosinophil lineage and is absolutely specific for eosinophils (eoCre mice) [22]. Results obtained from Atg5 flox/flox eoCre mice showed elevated numbers of immature eosinophils in the bone marrow and a significant drop of mature eosinophils in the circulation [89] (Fig. 2).…”

“…In contrast to other cells of the immune system, the relevance of autophagy and ATG proteins for the biology of eosinophils has long remained elusive. However, insightful evidence was recently generated from experimental mouse models in which Atg5 was specifically depleted in the eosinophil lineage, resulting in eosinophil-specific autophagy deficiency [89]. As discussed below, these models allowed testing the function of eosinophils under in vitro conditions in the presence and absence of ATG5.…”

“…During in vitro eosinophil differentiation, a reduced phosphorylation of p38 and p44/42 mitogen-activated protein kinases (MAPKs) was detected in eosinophil precursors lacking Atg5, which might explain, at least partially, the observed phenotype. Eosinophil populations were further purified from the bone marrow of hypereosinophilic II5 (IL-5) transgenic mice (NJ.1638), showing downregulation of Gata-1, C/ebpε, Pu.1, and Trib1 transcription factors in the absence of Atg5, which might reflect the reduced and delayed eosinophil differentiation [89]. Moreover, the differentiation potential of Atg5-knockout eosinophil precursors was tested under pathologic conditions, employing an established mouse model of chronic eosinophilic leukemia (CEL).…”

“…The effect of autophagy on eosinophil degranulation has been studied in Atg5-knockout eosinophils, which intriguingly exhibited an enhanced capacity to degranulate in vitro as measured by increased CD63 surface expression following GM-CSF priming and C5a stimulation [89] (Fig. 2).…”