The GM-CSF receptor utilizes β-catenin and Tcf4 to specify macrophage lineage differentiation

Abstract

Granulocyte-macrophage colony stimulating factor (GM-CSF) promotes the growth, survival, differentiation and activation of normal myeloid cells and is essential for fully functional macrophage differentiation in vivo. To better understand the mechanisms by which growth factors control the balance between proliferation and self-renewal versus growth-suppression and differentiation we have used the bi-potent FDB1 myeloid cell line, which proliferates in IL-3 and differentiates to granulocytes and macrophages in response to GM-CSF. This provides a manipulable model in which to dissect the switch between growth and differentiation. We show that, in the context of signaling from an activating mutant of the GM-CSF receptor β subunit, a single intracellular tyrosine residue (Y577) mediates the granulocyte fate decision. Loss of granulocyte differentiation in a Y577F second-site mutant is accompanied by enhanced macrophage differentiation and accumulation of β-catenin together with activation of Tcf4 and other Wnt target genes. These include the known macrophage lineage inducer, Egr1. We show that forced expression of Tcf4 or a stabilised β-catenin mutant is sufficient to promote macrophage differentiation in response to GM-CSF and that GM-CSF can regulate β-catenin stability, most likely via GSK3β. Consistent with this pathway being active in primary cells we show that inhibition of GSK3β activity promotes the formation of macrophage colonies at the expense of granulocyte colonies in response to GM-CSF. This study therefore identifies a novel pathway through which growth factor receptor signaling can interact with transcriptional regulators to influence lineage choice during myeloid differentiation.

Fig. 1. Y577 regulated genes

A. Heatmap showing the expression of the top 25 up and down regulated genes associated FIΔ induced factor independent granulocyte differentiation over 72 hours (see materials and methods for details of gene-set derivations). B. IPA defined canonical pathways significantly associated with the neutrophil gene set Lysine= lysine degradation, LES= leukocyte extravasation signaling, CES= caveolar-mediated endocytosis signaling. C. Heatmap showing the expression of the top 25 up and down regulated genes associated with FIΔ-Y577F induced factor independent macrophage differentiation over 72 hours, * marks Tcf4. D. IPA defined canonical pathways associated with the macrophage gene set, AP= acute phase response signaling, RA= rheumatoid arthritis. The M value (log2 fold change) for both FIΔ or FIΔY577F expressing cells was converted to a colour scale with red indicating that gene expression increases between 0 and 72 hours and green indicating that the gene expression decreases with the colour intensity indicating the magnitude of regulation (range log2, −3 to +3). Black boxes indicate individual genes involved in the indicated pathway.

Fig. 2. Tcf4 is regulated by GM-CSF and promotes macrophage differentiation

Quantitative real-time PCR was used to determine the expression of the indicated genes. Expression was normalised to β-actin and expression is shown relative to 0 hours. Results shown are the average of two experiments A. FDB1 cells expressing FIΔ and FIΔY577F were cultured without factor and RNA extracted at 0, 24 and 72 hours. B. Parental FDB1 cells were cultured in GM-CSF and RNA extracted at 0, 24, 72 and 120 hours. C. Full length TCF4 and the engineered form lacking the N-terminal β-catenin interaction domain are depicted. The β-catenin interaction domain is indicated by a black box and the DNA binding domain is indicated by a grey box. D. FDB1 parental cells were transduced with a MSCV-IRES-GFP (MIG) retrovirus encoding TCF4, ΔN-TCF4 or a control retrovirus and selected for GFP expression by FACS. Cells were washed and placed in the indicated growth conditions. Expansion of GFP positive cells was determined by assessing viable cell number using trypan blue exclusion and flow cytometry to determine the proportion of GFP positive cells at the indicated time points. E. At day 5, cells were cytocentrifuged, Wright-Giemsa stained and 200 cells for each condition were scored microscopically for morphology. F. At day 5 cells were stained with PE-conjugated anti-c-fms and the percentage of c-fms positive cells in the GFP positive population was determined for each cell line. A representative flow histogram is shown. Error bars represent SEM (n=4) and *=p<0.05, **=<0.01, ***=p<0.001 (unpaired, two sided t-test).

Fig. 3. β-catenin stabilization is associated with increased macrophage differentiation

A. FDB1 cells expressing FIΔ and FIΔY577F were cultured without factor and parental cells in GM-CSF and cell lysates were made at 0, 24 and 72 hours. Lysates were western blotted with the indicated antibodies. Representative photo-micrographs were taken at the 72 hour timepoint for FDB1 FIΔ and FIΔY577F cells and at 120 hours for parental cells in GM-CSF. B. FDB1 cells growing in response to IL-3 or undergoing GM differentiation in response to GM-CSF were incubated with 2 µM of the GSK3β inhibitor BIO ((2’Z,3’E)-6-Bromoindirubin-3’-oxime) or N-methylated control analogue (Me-BIO). Activity of BIO at 16 hours was assessed by detection of β-catenin protein by western analysis. C. Cell growth was determined at days 2 and 5 of GSK3β inhibition by trypan blue exclusion. D. At day 5 cells were cytocentrifuged, Wright-Giemsa stained and 200 cells for each condition were scored microscopically for morphology. E. At day 5 cells were stained with PE-conjugated anti-c-fms and the percentage of c-fms positive cells was determined for each cell population. Error bars represent SEM (n=3) and *=p<0.05, **=<0.01, ***=p<0.001 (unpaired, two sided t-test).

Fig. 4. β-catenin directs macrophage differentiation in response to GM-CSF receptor signaling

FDB1 cells were transduced with MIG encoding β-catenin with mutations S33A, S37A, T41A, S45A (βcatS33A) or vector alone. A. GFP positive cells were placed in IL-3 or GM-CSF and after 5 days were assessed for expression of stabilised β-catenin by western analysis. B. Growth was assessed over 4 days using AQueous one solution proliferation assay. C. Photomicrographs of cytocentrifuged, Wright-Giemsa stained cells at day 5. D. At day 5 200 cells for each condition were scored by morphology. Error bars represent SEM (n=2) and *=p<0.05 (unpaired, two sided t-test).

Fig. 5. GSK3β inhibition increases macrophage differentiation of primary cells

A. Bone marrow cells were plated in methylcellulose medium containing GM-CSF and the indicated concentrations of BIO or the control analogue Me-BIO and scored for colony formation 8 days later. CFU=colony forming unit, M=macrophage, G=granulocyte, GM= granulocyte-macrophage. Error bars represent SEM (n=2) and *=p<0.05, **=<0.01, (unpaired, two sided t-test). B. Overlapping TCF4/β-catenin targets and macrophage genes with an FDR p<0.05 (113) were subjected to Ingenuity Pathway Analysis (IPA). The genes were subject to IPA mapping of network interactions and filtered for direct interactions and networks containing genes identified as involved in Wnt/β-catenin signaling. The colour indicates the direction of differential expression in our macrophage regulated gene list with red indicating that gene expression increases between 0 and 72 hours and green indicating that the gene expression decreases with the colour intensity indicating the magnitude of regulation. CTNNB1= β-catenin.

Fig. 6. Novel molecular interactions in monopoiesis regulated by GM-CSF signaling

Signaling emanating from Tyr577 of the GMR leads to stabilisation of the β-catenin protein (possibly though GSK3β inactivation) which, in combination with Tcf4, leads to differentiation of the macrophage lineage at the expense of neutrophil lineage.