Abstract:Inflammatory responses must be tightly coordinated with the activation of emergency myelopoiesis to produce potent myeloid cells that fight infection without causing excessive host damage. Here, we show that granulocyte-macrophage colony-stimulating factor (GM-CSF) programs myeloid-committed progenitors to produce trained macrophages (increased cytokine response), but programs the upstream noncommitted LKS+ progenitors (defined as Lin- c-Kit+ Sca-1+ cells) to produce tolerized macrophages (decreased cytokine response). In myeloid progenitors, GM-CSF strongly activates signal transducer and activator of transcription 5 (STAT5), Ras-Raf-extracellular signal regulated kinase (ERK), and Akt-mTOR signaling pathways, which are essential to establish a training program, whereas in LKS+ progenitors, GM-CSF induces NF-κB translocation to the nucleus to establish a tolerization program. These differences arise from higher GM-CSF receptor expression in myeloid progenitors compared with LKS+ cells. We demonstrate that β-catenin regulation of NF-κB nuclear translocation is central in this process. In myeloid progenitors, glycogen synthase kinase 3 (GSK3) inactivation by strong ERK and phosphatidylinositol 3 kinase (PI3K)-Akt signaling increases cytoplasmic β-catenin levels to block NF-κB nuclear translocation. In contrast, when ERK and PI3K-Akt signaling are weak, active GSK3 causes a decrease in β-catenin, allowing NF-κB nuclear translocation in LKS+ progenitors. Finally, GM-CSF-induced LKS+ tolerization takes place in several murine models of trained immunity and in human CD34+ CD38- progenitors. Our study reveals that in addition to activating myelopoiesis, GM-CSF also programs early and immediate myeloid progenitors to produce opposing immune memory phenotypes. We propose that the inflammatory response from immediate myeloid progenitors may be balanced by the tolerized phenotype of early progenitors, thus providing a mechanism for appropriate resolution of inflammation and protection against a prolonged cytokine storm.

Transcription Factor RUNX3 Mediates Plasticity of ThGM Cells Toward Th1 Phenotype

A distinct GM-CSF<SUP>+</SUP> T helper cell subset requires T-bet to adopt a T<sub>H</sub>1 phenotype and promote neuroinflammation

ThG cells: a distinct T helper cell subset expressing GM-CSF, TNF, and IL-2, but not Th lineage-specific cytokines

STAT5 programs a distinct subset of GM-CSF-producing T helper cells that is essential for autoimmune neuroinflammation.

GATA3 induces the pathogenicity of Th17 cells via regulating GM-CSF expression

Gene Regulatory Network of Human GM-CSF-Secreting T Helper Cells

Gene regulatory network of human GM-CSF secreting T helper cells

GATA3, HDAC6, and BCL6 Regulate FOXP3+ Treg Plasticity and Determine Treg Conversion into Either Novel Antigen-Presenting Cell-Like Treg or Th1-Treg.

Lncrna-Gm Targets Foxo1 to Promote T Cell–mediated Autoimmunity

Generation of ROR gamma t(+) Antigen Specific T Regulatory 17 Cells from Foxp3(+) Precursors in Autoimmunity

The Role of STAT3 in Antigen-Igg Inducing Regulatory CD4+Foxp3+T Cells

GM-CSF receptor expression determines opposing innate memory phenotypes at different stages of myelopoiesis

Transcription factor TCF1 binds to RORγt and orchestrates a regulatory network that determines homeostatic Th17 cell state

An autoimmune transcriptional circuit drives FOXP3 + regulatory T cell dysfunction

GM-CSF and CXCR4 define a T helper cell signature in multiple sclerosis

Histone Methylation Mediates Plasticity of Human Foxp3+ Regulatory T Cells by Modulating Signature Gene Expressions

Interactions among the Transcription Factors Runx1, RORγt and Foxp3 Regulate the Differentiation of Interleukin 17–producing T Cells

Restriction of innate Tγδ17 cell plasticity by an AP-1 regulatory axis

Effector T cell differentiation: are master regulators of effector T cells still the masters?

Plasticity and functionality of Th17 subsets in EAE

Transcription factor EGR2 controls homing and pathogenicity of TH17 cells in the central nervous system