IRF-8 extinguishes neutrophil production and promotes dendritic cell lineage commitment in both myeloid and lymphoid mouse progenitors

Abstract

While most blood lineages are assumed to mature through a single cellular and developmental route downstream of HSCs, dendritic cells (DCs) can be derived from both myeloid and lymphoid progenitors in vivo. To determine how distinct progenitors can generate similar downstream lineages, we examined the transcriptional changes that accompany loss of in vivo myeloid potential as common myeloid progenitors differentiate into common DC progenitors (CDPs), and as lymphoid-primed multipotent progenitors (LMPPs) differentiate into all lymphoid progenitors (ALPs). Microarray studies revealed that IFN regulatory factor 8 (IRF-8) expression increased during each of these transitions. Competitive reconstitutions using Irf8(-/-) BM demonstrated cell-intrinsic defects in the formation of CDPs and all splenic DC subsets. Irf8(-/-) common myeloid progenitors and, unexpectedly, Irf8(-/-) ALPs produced more neutrophils in vivo than their wild-type counterparts at the expense of DCs. Retroviral expression of IRF-8 in multiple progenitors led to reduced neutrophil production and increased numbers of DCs, even in the granulocyte-macrophage progenitor (GMP), which does not normally possess conventional DC potential. These data suggest that IRF-8 represses a neutrophil module of development and promotes convergent DC development from multiple lymphoid and myeloid progenitors autonomously of cellular context.

Figure 1

IRF-8 expression increases during the common myeloid progenitor to CDP transition. Microarray analysis was carried out on double-sorted common myeloid progenitors (CMPs) and CDPs from BM. (A) Common myeloid progenitors were lineage (Lin)⁻c-kit^hiSca-1⁻CD16/32⁻CD34⁺ Flk2⁺CD115⁻ and CDP were were Lin⁻c-kit^intermediateSca-1⁻CD16/32⁻CD34⁺Flk2⁺CD115⁺. (B) In vitro differentiation of common myeloid progenitors and CDPs. Cells were cultured for 7 days with IL-3, GM-CSF, Flt3 ligand, and stem cell factor (all 10 ng/mL). DC (CD11c⁺Gr-1⁻) and neutrophil (CD11c⁻Gr-1⁺) potential was analyzed by flow cytometry. (C) Heat map of microarray data showing selected transcription factors that were differentially regulated at least 5-fold between the common myeloid progenitor and CDP subsets. (D) Quantitative RT-PCR analysis of IRF-8 in double-sorted LMPP, common myeloid progenitor, GMP, and CDP. Data represent 3 or 6 independent sorts for GMP and CDP or LMPP and common myeloid progenitor, respectively, analyzed in 3 separate assays; **P < .05.

Figure 2

Irf8^−/− mice have reduced numbers of CDPs and increased numbers of GMPs. Progenitor cells in the BM of Irf8^−/− and wild-type mice were analyzed by flow cytometry. Graphs show the frequency (left) and absolute number (right) of common myeloid progenitor (CMP), GMP, and CDP. Horizontal lines indicate the mean and error bars show SEM. Each data point represents one mouse, with data for 14 wild-type and 13 IRF8^−/− shown; **P < .05.

Figure 3

IRF-8 drives DC commitment in a cell-intrinsic manner. Competitive mixed BM chimeras were carried out by injecting 2 × 10⁶ B6.SJL (CD45.1⁺) wild-type whole BM cells with Irf8^−/− or wild-type (CD45.2⁺) BM that were matched for HSC numbers into 800 cGy-irradiated (C57BL/6 × B6.SJL) F1 recipients (CD45.1⁺CD45.2⁺). Seven weeks after transplantation, BM and spleens were harvested, and progenitor, DC, and neutrophil development was analyzed by flow cytometry. (A,C) Representative plots showing CD45.2 and CD45.1 chimerism in one representative mouse. Graphs show (B) chimerism of common myeloid progenitor (CMPs), CDPs, and GMPs or (D) splenic CD8α⁻ DC and BM neutrophils from wild-type (top) or Irf8^−/− (bottom) reconstituted mice. Each point represents 1 mouse, with lines connecting values in a single mouse. Data are pooled from 3 experiments of 1-4 mice per group. Data for all 9 mice in each group are shown; **P < .05 based on paired students 2-tailed t test.

Figure 4

Common myeloid progenitors from Irf8^−/− mice overproduce neutrophils at the expense of DCs in vitro and in vivo. Common myeloid progenitors (CMP) were double-sorted from the BM of Irf8^−/− or wild-type controls. (A-B) One thousand progenitors were cultured for 7 days in the presence of Flt3 ligand, IL-3, GM-CSF, and SCF (all 10 ng/mL). DC and neutrophil output was analyzed by flow cytometry. Data show averages from 5 wells and represent 3 independent assays. (C-D) Four thousand common myeloid progenitors from Irf8^+/+ or Irf8^−/− were injected intravenously into each 800 cGy-irradiated (C57BL/6 × B6.SJL) F1 recipient (CD45.1⁺CD45.2⁺). BM and spleens were harvested 10 days after transplantation and the development of DCs and neutrophils was analyzed by flow cytometry. Data shown are for 6 mice in each experimental line pooled from 2 independent experiments with 3 mice per group; **P < .05.

Figure 5

ALPs from Irf8^−/− mice but not wild-type mice produce neutrophils in vivo. (A) LMPPs and ALPs were double-sorted from the BM of Irf8^−/− or wild-type controls and quantitative real-time PCR analysis of IRF-8 expression was carried out. Data are representative of 6 independent experiments. (B-C) Four thousand ALPs purified from either Irf8^−/− or wild-type mice were injected intravenously into each 800 cGy-irradiated F1 recipient (CD45.1⁺CD45.2⁺). BM and spleen were harvested 10 days after transplantation and the development of DCs, neutrophils, and B cells was analyzed by flow cytometry. (B) Representative contour plots. (C) Pooled data from 3 independent experiments. Data represent 5 wild-type and 6 Irf8^−/− recipients. Error bars show calculated SEM; **P < .05.

Figure 6

Ectopic expression of IRF-8 extinguishes neutrophil potential and induces DC potential in LMPPs, common myeloid progenitors, and GMPs. LMPPs, common myeloid progenitors (CMPs), or GMPs were double-sorted from the BM of wild-type mice. One thousand cells per well were cultured in the presence of Flt3 ligand, GM-CSF, IL-3, and stem cell factor (all 10 ng/mL) for 7 days, and transduced by retrovirus as described in “In vitro culture and retroviral transductions.” Cells were harvested and analyzed for neutrophil (MHC class II⁻GR-1⁺) and DC (MHC class II⁺Gr-1⁻) production in virus-infected (GFP⁺) cells. Representative FACS plots showing the development of GFP⁺ MHC class II⁻ Gr-1⁺ neutrophils and GFP⁺ MHC class II⁺ Gr-1⁻ DCs in (A) LMPPs and common myeloid progenitors or (B) GMPs. Data represent 4 independent experiments with progenitors analyzed in triplicate. (C) GFP⁺ cells were sorted from GMP cultures and morphology of cells was analyzed by Giemsa stain and microscopy.

Figure 7

IRF-8 regulates distinct transcriptional programs in myeloid and lymphoid progenitors. ALPs or common myeloid progenitors (CMPs) were double-sorted from Irf8^+/+ or Irf8^−/− mice and gene expression profiling was carried out by microarray analysis. Venn diagram shows genes that are specifically modulated in the common myeloid progenitor or ALP pathways, as well as genes that are coordinately dysregulated in both pathways. The genes listed in the tables show known transcription factors. Numbers represent fold changes in Irf8^−/− compared with Irf8^+/+ cells.