Fibronectin leucine-rich transmembrane protein 2 drives monocyte differentiation into macrophages via the UNC5B-Akt/mTOR axis

Abstract

Upon migrating into the tissues, hematopoietic stem cell (HSC)-derived monocytes differentiate into macrophages, playing a crucial role in determining innate immune responses towards external pathogens and internal stimuli. However, the regulatory mechanisms underlying monocyte-to-macrophage differentiation remain largely unexplored. Here we divulge a previously uncharacterized but essential role for an axon guidance molecule, fibronectin leucine-rich transmembrane protein 2 (FLRT2), in monocyte-to-macrophage maturation. FLRT2 is almost undetectable in human monocytic cell lines, human peripheral blood mononuclear cells (PBMCs), and mouse primary monocytes but significantly increases in fully differentiated macrophages. Myeloid-specific deletion of FLRT2 (Flrt2^ΔMyel ) contributes to decreased peritoneal monocyte-to-macrophage generation in mice in vivo, accompanied by impaired macrophage functions. Gain- and loss-of-function studies support the promoting effect of FLRT2 on THP-1 cell and human PBMC differentiation into macrophages. Mechanistically, FLRT2 directly interacts with Unc-5 netrin receptor B (UNC5B) via its extracellular domain (ECD) and activates Akt/mTOR signaling. In vivo administration of mTOR agonist MYH1485 reverses the impaired phenotypes observed in Flrt2^ΔMyel mice. Together, these results identify FLRT2 as a novel pivotal endogenous regulator of monocyte differentiation into macrophages. Targeting the FLRT2/UNC5B-Akt/mTOR axis may provide potential therapeutic strategies directly relevant to human diseases associated with aberrant monocyte/macrophage differentiation.

Keywords: FLRT2; MTOR signaling; UNC5B; differentiation; macrophage; monocyte.

Figure 1

FLRT2 is significantly induced during monocyte-to-macrophage differentiation in vitro and in vivo. (A, B) THP-1 or HL-60 cells were treated with DMSO or PMA (100 ng/ml) for 24 h. FLRT2 mRNA and protein levels were detected by qPCR (A, n = 4) and immunoblot (B, n = 5) analyses, respectively. (C, D) qPCR analysis of Flrt2 mRNA (C, n = 3) and immunoblot analysis of FLRT2 protein (D, n = 3) in human PBMCs induced by M-CSF (40 ng/ml) for 0 or 7 days. (E, F) qPCR analysis of Flrt2 mRNA (E, n = 3) and immunoblot analysis of FLRT2 protein (F, n = 4) in bone marrow (BM) cells induced by M-CSF (25 ng/ml) for 0 or 5 days. (G, H) qPCR analysis of Flrt2 mRNA (G, n = 5 mice per group) and immunoblot analysis of FLRT2 protein (H, n = 4 mice per group) in BM cells and peritoneal macrophages (PMs) isolated from C57BL/6J mice after intraperitoneal (i.p.) thioglycollate injection. mRNA and protein levels were normalized to both β-actin and CD11b. (I) Representative flow cytometric profiles and data plots showing the frequencies of FLRT2⁺CD45⁺CD11b⁺F4/80⁻ monocytes and FLRT2⁺CD45⁺CD11b⁺F4/80⁺ macrophages in the BM, peripheral blood, and peritoneal cavity of C57BL/6J mice injected i.p. with or without thioglycollate for 3 days (n = 4 or 7 mice per group). Data are means ± SD. P values were determined using unpaired, two-tailed Student’s t-tests. NS, not significant. ^* P < 0.05, ^** P < 0.01, ^*** P < 0.001, ^**** P < 0.0001.

Figure 2

Myeloid cell-specific FLRT2 knockout (Flrt2^ΔMyel ) leads to decreased peritoneal monocyte-to-macrophage differentiation and impaired peritoneal macrophage functions in vivo. (A) Representative flow cytometric profiles and data plots showing the absolute numbers of CD45⁺CD11b⁺F4/80⁻ monocytes and CD45⁺CD11b⁺F4/80⁺ macrophages in the peritoneal cavity and BM, and the frequencies of CD45⁺CD11b⁺F4/80⁻ monocytes and CD45⁺CD11b⁺F4/80⁺ macrophages in peripheral blood of Flrt2^fl/fl and Flrt2^ΔMyel mice treated with or without i.p. injection of thioglycollate for 3 days (n = 6 or 8 mice per group). (B) Phase-contrast microscopy images showing morphological alterations of PMs isolated from Flrt2^fl/fl and Flrt2^ΔMyel mice (n = 3 mice per group). Scale bar, 50 μm. (C) qPCR analysis of Itgam, Cd14, Csf1r, Cd36, Msr1, and Olr1 mRNA in PMs isolated from Flrt2^fl/fl and Flrt2^ΔMyel mice (n = 3 mice per group). (D) CD36 and SR-A protein levels in PMs isolated from Flrt2^fl/fl and Flrt2^ΔMyel mice were determined by immunoblot analysis (n = 3 mice per group). (E) PMs were collected from Flrt2^fl/fl and Flrt2^ΔMyel mice and then co-cultured with HUVECs expressing RFP for 6 h, followed by fluorescent microscopy (n = 3 mice per group). Scale bar, 5 mm. (F) PMs were isolated from Flrt2^fl/fl and Flrt2^ΔMyel mice and seeded in a Boyden chamber for 48 h, followed by crystal violet staining (n = 3 mice per group). Scale bar, 100 μm. (G) Zymosan was injected i.p. into Flrt2^fl/fl and Flrt2^ΔMyel mice for 2 h, and PMs were collected and placed in a 6-well plate for 6 h. The phagocytic particles were observed under a laser confocal scanning microscope (n = 3 mice per group). Scale bar, 50 μm. Data are means ± SD. P values were determined using unpaired, two-tailed Student’s t-tests. NS, not significant. ^* P < 0.05, ^** P < 0.01, ^*** P < 0.001, ^**** P < 0.0001.

Figure 3

FLRT2 induces PBMC differentiation into macrophages and enhances the adhesion, migration, and phagocytosis of M-CSF-primed PBMC-derived macrophages. (A–C, E, F) Human PBMCs were transfected with control or hFLRT2 vector for 48 h. (A) Representative phase-contrast light microscopy images showing morphological alterations of PBMCs transfected as indicated (n = 3). Scale bar, 50 μm. (B) qPCR analysis of Itgam, Cd14, Csf1r, Cd36, Msr1, Olr1, and Flrt2 mRNA in PBMCs cells transfected as indicated (n = 3). (C) Immunoblot analysis of CD36, SR-A, and Flag-FLRT2 proteins in PBMCs transfected as indicated (n = 3). (D) HUVECs expressing RFP were co-cultured with PBMCs expressing vector-GFP or hFLRT2-GFP. After 6 h co-culture, pictures were taken using a fluorescence microscope (n = 3). Scale bar, 5 mm. (E) Transwell cell migration assays were performed, and numbers of migrated cells were counted (n = 3). Scale bar, 100 μm. (F) Phagocytosis assays were performed by culturing the cells in Texas red-conjugated zymosan particles for 2 h at 37°C (n = 3). Cells were viewed for internalization of the particles by fluorescence microscopy. Scale bar, 50 μm. Data are means ± SD. P values were determined using unpaired, two-tailed Student’s t-tests. ^* P < 0.05, ^** P < 0.01, ^*** P < 0.001, ^**** P < 0.0001.

Figure 4

UNC5B mediates FLRT2-promoted THP-1 cell maturation into macrophages. FLRT2 was overexpressed by transfecting the hFLRT2 vector, and UNC5B was silenced by transfecting UNC5B-specific shRNA into THP-1 cells. (A) Representative phase-contrast light microscopy images showing morphological alterations of THP-1 cells transfected as indicated (n = 3). Scale bar, 50 μm. (B) qPCR analysis of Itgam, Csf1r, Cd36, Msr1, and Olr1 mRNA in THP-1 cells transfected as indicated (n = 3). (C) Immunoblot analysis of CD36, SR-A, UNC5B, and GFP-FLRT2 proteins in THP-1 cells transfected as indicated (n = 3). (D) HUVECs expressing RFP were co-cultured with THP-1 cells transfected as indicated. After 6 h co-culture, pictures were taken using a fluorescence microscope (n = 3). Scale bar, 5 mm. (E) Transwell cell migration assay was performed in THP-1 cells transfected as indicated, and the numbers of the migrated cells were counted (n = 3). Scale bar, 100 μm. (F) Phagocytosis assays were performed by culturing the cells transfected as indicated in Texas red-conjugated zymosan particles for 2 h at 37°C (n = 3). Cells were observed for internalization of the particles by fluorescence microscopy. Scale bar, 50 μm. Data are means ± SD. P values were determined using unpaired, two-tailed Student’s t-tests. ^* P < 0.05, ^** P < 0.01, ^*** P < 0.001, ^**** P < 0.0001.

Figure 5

The direct interaction of the FLRT2 ECD with UNC5B is required for FLRT2-promoted THP-1 cell differentiation into macrophages. (A) Myc-tagged hUNC5B and/or Flag-tagged hFLRT2 were transfected into HEK293T cells. Immunoprecipitation samples were collected using anti-Flag antibody, followed by immunoblotting for Myc and Flag (n = 3). (B) Immunolabeling of FLRT2 and UNC5B in PMA-treated THP-1 cells showing the colocalization of endogenous UNC5B and FLRT2. Scale bar, 10 μm. (C) UNC5B was immunoprecipitated from PMA-incubated THP-1 cells, followed by immunoblotting for FLRT2 and UNC5B (n = 3). (D) Schematic illustration of FLRT2 and UNC5B constructs. LRRNT, leucine-rich repeat N-terminal domain. LRR, leucine-rich repeat. LRRCT, leucine-rich repeat C-terminal domain. FNIII, fibronectin type III domain. (E) UNC5B interacts with hFLRT2-ECD but not hFLRT2-ICD (n = 3). (F, G) THP-1 cells were transfected with control, human FLRT2 extracellular domain (hFLRT2-ECD) or human FLRT2 intracellular domain (hFLRT2-ICD) vector for 48 h, followed by qPCR (F) and immunoblot (G) analyses. Data are means ± SD. P values were determined using unpaired, two-tailed Student’s t-tests. NS, not significant. ^** P < 0.01, ^**** P < 0.0001.

Figure 6

FLRT2 activates Akt/mTOR signaling via the binding of UNC5B with Rac1. (A) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis of the upregulated differentially expressed genes (DEGs) in PMs of Flrt2^fl/fl and Flrt2^ΔMyel mice. (B) Immunoblot analysis of p-Akt (Ser473), p-Akt (Thr308), Akt, p-mTOR (Ser2481), mTOR, p-4E-BP-1 (Thr37/46), 4E-BP-1, p-S6K (Thr389), S6K, p-S6 (Ser240/244), and S6 proteins in PMs isolated from Flrt2^fl/fl and Flrt2^ΔMyel mice i.p. injected with thioglycollate for 3 days (n = 3 mice per group). (C) Immunoblot analysis of p-Akt (Ser473), p-Akt (Thr308), Akt, p-mTOR (Ser2481), mTOR, p-4E-BP-1 (Thr37/46), 4E-BP-1, p-S6K (Thr389), S6K, p-S6 (Ser240/244), S6, Flag-FLRT2, and FLRT2 protein levels in THP-1 cells transfected with control or hFLRT2 vector for 48 h (n = 3). (D) Immunoblot analysis of the indicated protein levels in THP-1 cells transfected with shNC or shFLRT2 #3 vector for 48 h (n = 3). (E) Immunoprecipitation samples were collected using anti-Flag antibody from control vector and Flag-UNC5B-overexpressing HEK293T cells and subjected to mass spectrometry (MS) to identify potential binding partners of UNC5B protein. A unique peptide of Rac1 protein was identified in the UNC5B-overexpressing immunoprecipitation samples by analyzing the mass-to-charge ratio of the samples. (F) The endogenous interaction between Rac1 and UNC5B was confirmed in THP-1 cells by co-immunoprecipitation (co-IP) analysis. Data are means ± SD. P values were determined using unpaired, two-tailed Student’s t-tests. ^* P < 0.05, ^** P < 0.01, ^*** P < 0.001, ^**** P < 0.0001.

Figure 7

MHY1485, an mTOR agonist, abrogates abnormal peritoneal monocyte-macrophage differentiation and macrophage functions of Flrt2^ΔMyel mice in vivo. (A) Protocol for in vivo rescue experiments. Flrt2^fl/fl and Flrt2^ΔMyel mice were infused with vehicle or mTOR activator MYH1485 (0.25 mg/kg) for 7 successive days by i.p. injection and were injected i.p. with thioglycollate at day 4. At the end of day 7, samples were collected for flow cytometric and functional analyses. (B) Immunoblot analysis of p-mTOR (Ser2481), mTOR, p-4E-BP-1 (Thr37/46), 4E-BP-1, p-S6K (Thr389), S6K, p-S6 (Ser240/244), and S6 levels in PMs isolated from Flrt2^fl/fl and Flrt2^ΔMyel mice (n = 3 mice per group). (C) Representative flow cytometric profiles and data plots showing the absolute numbers of CD45⁺CD11b⁺F4/80⁻ monocytes and CD45⁺CD11b⁺F4/80⁺ macrophages in the peritoneal cavity and BM, and the frequencies of CD45⁺CD11b⁺F4/80⁻ monocytes and CD45⁺CD11b⁺F4/80⁺ macrophages in peripheral blood of Flrt2^fl/fl and Flrt2^ΔMyel mice (n = 5, 6, or 9 mice per group). (D) At the end of day 7, PMs were collected from the above treated Flrt2^fl/fl and Flrt2^ΔMyel mice and then co-cultured with HUVECs expressing RFP for 6 h, followed by fluorescent microscopy (n = 3 mice per group). Scale bar, 5 mm. (E) At the end of day 7, PMs were isolated from the above treated Flrt2^fl/fl and Flrt2^ΔMyel mice and seeded in Boyden chambers for 48 h, followed by crystal violet staining (n = 3 mice per group). Scale bar, 100 μm. (F) At the end of day 7, zymosan was injected i.p. into the above treated Flrt2^fl/fl and Flrt2^ΔMyel mice for 2 h. PMs were collected and placed in a 6-well plate for 6 h. The phagocytic particles were observed under a laser confocal scanning microscope (n = 3 mice per group). Scale bar, 50 μm. Data are means ± SD. P values were determined using unpaired, two-tailed Student’s t-tests. NS, not significant. ^* P < 0.05, ^** p < 0.01, ^*** P < 0.001, ^**** P < 0.0001.

Figure 8

Schematic diagram of our major findings. Pro-differentiation factors, such as PMA, M-CSF, and thioglycollate, upregulate Flrt2 expression in monocytes. FLRT2, in turn, accelerates monocyte-to-macrophage differentiation by binding to UNC5B via its ECD and subsequently activating the Akt/mTOR signaling pathway.