CX3CR1+ CD115+ CD135+ common macrophage/DC precursors and the role of CX3CR1 in their response to inflammation

Abstract

CX(3)CR1 expression is associated with the commitment of CSF-1R(+) myeloid precursors to the macrophage/dendritic cell (DC) lineage. However, the relationship of the CSF-1R(+) CX(3)CR1(+) macrophage/DC precursor (MDP) with other DC precursors and the role of CX(3)CR1 in macrophage and DC development remain unclear. We show that MDPs give rise to conventional DCs (cDCs), plasmacytoid DCs (PDCs), and monocytes, including Gr1(+) inflammatory monocytes that differentiate into TipDCs during infection. CX(3)CR1 deficiency selectively impairs the recruitment of blood Gr1(+) monocytes in the spleen after transfer and during acute Listeria monocytogenes infection but does not affect the development of monocytes, cDCs, and PDCs.

Figure 1.

MDP and CDP share a similar phenotype. BM cells from Cx₃cr1^gfp/+ reporter mice were labeled with antibodies against lineage markers (CD11c, CD11b, NK1.1, CD3, Ter119, CD19, and Gr1), ckit (CD117), FLT3 (CD135), and CSF-1-R (CD115) and analyzed by FACS. CDP and MDP are overlapping populations; lineage⁻ CD117^low CD135⁺ CD115⁺ cells (CDP) express CX₃CR1 (gfp), whereas CD117^int CX₃CR1 CD115⁺ cells (MDP) express CD135. Results are representative of >10 experiments.

Figure 2.

Differentiation potential of MDP in vivo. (a–c) MDPs from Cd45.1/2 or Cd45.2 Cx₃cr1^gfp/+ reporter mice were purified as described in Fig. S2 (available at http://www.jem.org/cgi/content/full/jem.20081385/DC1) and were adoptively transferred into irradiated (900 rad) C57BL/6 Cd45.2 congenic recipients. Spleens of recipient mice were analyzed at day 7 after transfer by flow cytometry, using lineage marker (NK1.1 CD3 CD19), CD11b, CD11c, and CD8-α antibodies. R1 corresponds to PDCs (Lin⁻ CD11c^int CD11b⁻ CX₃CR1⁺ cells), R2 and R3 correspond to cDCs (lin⁻ CD11c^high CD11b⁻ CX₃CR1⁻ cells and lin⁻ CD11c^high CD11b⁺ CX₃CR1⁺ cells), and R4 corresponds to monocytes (lin⁻ CD11b⁺ CD11c⁻ CX₃CR1⁺ cells). The experiment was repeated five times with two to three mice per group and with similar results. (b) Donor-derived Lin⁻ CD11c^int CD11b⁻ CX₃CR1⁺ cells express PDCA1. (c) Role of anti-CD115 Ab. The number of donor-derived cells per spleen are represented after adoptive transfer of MDP, purified with or without CD115 antibody (n = at least 3 mice per group from two experiments). Error bars show SD. (d) The flow diagram represent the Lin⁻ CD117^int CD115⁺ CD135⁺ CX₃CR1⁺ MDPs that give rise to monocytes, cDCs, and PDCs and their putative relationship with other myeloid precursors.

Figure 3.

CX₃CR1 is important for the development of CD11b⁺ CD11c⁻ monocytes. (a) MDP numbers in BM from CX₃CR1^+/− (black) and CX₃CR1^−/− (white) mice. Data are the mean ± SD of five mice per group. (b–d) Competitive adoptive transfer of MDP into irradiated host (900 rad). Equal numbers (10⁴) of MDPs from Cd45.1/2 mice (blue) and Cd45.2/2 mice (red) were mixed and adoptively transferred into a Cd45.1 congenic recipient. Spleens of recipient mice were analyzed at day 7 after transfer by flow cytometry, using lineage marker (NK1.1 CD3 CD19), CD11b, CD11c, and CD8-α antibodies. (b and c) When both Cd45.1/2 and Cd45.2/2 donor mice were of the Cx₃cr1^+/− genotype, both donors contributed equally to NK1.1⁻ CD3⁻ CD19⁻ CD11c^high CD11b⁻ (R1, cDC) and NK1.1⁻ CD3⁻ CD19⁻ CD11c^high CD11b⁺ (R2, cDC), NK1.1⁻ CD3⁻ CD19⁻ CD11c^int CD11b⁻ (R3, PDC), and NK1.1⁻ CD3⁻ CD19⁻ CD11b⁺ CD11c⁻ splenocytes (R4). (b and d) In contrast, when Cd45.1/2 donor mice were of the Cx₃cr1^+/− genotype and Cd45.2/2 donor mice were of the Cx₃cr1^−/− genotype (CX₃CR1 deficient), both donors contributed equally to CD11c⁺ cDCs in R1 and R2 but Cd45.1/2 Cx₃cr1^+/− MDPs were 10× more efficient than Cd45.2/2 Cx₃cr1-deficient MDPs in generating CD11b⁺ CD11c⁻ splenocytes in R3. Results in b are from one representative experiment out of three, with two to three mice per experimental group, and bar graphs in c and d represent the mean and SD from three independent experiments. The asterisk indicates a significative difference between groups (P < 0.05 using the Wilcoxon test).

Figure 4.

CX₃CR1 is important for the recruitment of CD11b⁺ CD11c⁻ monocytes to the spleen during infection. (a and b) Mouse fractalkine/CX₃CL1 transcripts are detected in spleen tissue sections from WT mice around B cells follicles in the marginal zone/T cell area. Bars: (a) 500 µm; (b) 200 µm. (c) Short-term adoptive transfer of BM monocytes into irradiated WT recipient mice. Equal numbers (2 × 10⁵) of BM monocytes from Cx₃cr1^+/− Cd45.1/2 mice (blue) and Cx₃cr1^−/− Cd45.2/2 mice (red) were mixed and adoptively transferred into a Cd45.1 congenic recipient. CX₃CR1-expressing CD45.1/2 Gr1⁺ monocytes are recruited in the spleen of irradiated host 10× more efficiently than CX₃CR1-deficient monocytes. The experiment was performed three times with two to three mice per group with similar results. (d–f) Monocyte recruitment in the spleen of infected mice. (d) BALB/c Cx₃cr1^+/− and Cx₃cr1^−/− mice (seven per group) were infected i.v. with live Lm (3 × 10⁵), and monocytes (CX₃CR1⁻ gfp⁺ CD11b⁺ Ly-6C⁺ Ly-6G⁻, CD19⁻, CD3⁻, and NK1.1⁻) were enumerated after 24 h in the blood and spleen of infected and control uninfected mice. (e) C57BL/6 Cx₃cr1^+/− and Cx₃cr1^−/− mice (seven per group) were injected with 7 × 10⁵ Lm, and monocytes were enumerated after 24 h in the blood and spleen of infected mice. (f) C57BL/6 Cx₃cr1^+/− and Cx₃cr1^−/− mice (seven per group) were injected with 7 × 10³ Lm, and monocytes were enumerated after 48 h in the spleen of infected mice. Lm growth in the spleen of infected mice is shown. The asterisks indicate a significative difference between groups (P < 0.05). (g) Absence of proliferation of monocytes in the spleen of Lm-infected mice. BALB/c mice (n = 6 per group) were infected with 3 × 10³ bacteria, spleen and leg bones were harvested 16 and 48 h later, and cells were processed as indicated in Fig. S5 (available at http://www.jem.org/cgi/content/full/jem.20081385/DC1). Data indicate the percentage of BM precursors and spleen monocytes in G0/G1 and in G2⁺S as analyzed by flow cytometry after DNA labeling with PI. The experiment was performed three times with similar results. (h) BALB/c Cx₃cr1^+/+ and Cx₃cr1^−/− mice were infected i.v. with live Lm (3 × 10³). Data show the number of bacteria (mean ± SE) in the spleen 24 h after infection. (i) C57BL/6 Cx₃cr1^+/+ and Cx₃cr1^−/− mice were injected with 10⁴ Lm. Data show the number of bacteria (mean ± SE) in the spleen 24 h after infection. Circles represent individual mice.

Figure 5.

Functions of spleen CD115⁺ Gr1⁺/TipDCs. (a) CD11b⁺ spleen cells from Cx₃cr1^gfp/+ reporter mice negative for lineage marker (NK1.1, CD3, B220, CD19, and CD11c) were analyzed for Ly6G, Ly6C, F4/80, CD115, and Cx₃CR1 (gfp) expression by multicolor flow cytometry. (b) TNF production by lineage⁻ CD11c^low CD11b⁺ cells 24 h after i.v. infection with Lm. Gated cells (circled) are displayed in red. Data show representative FACS profiles from three experiments. (c and d) Production of TNF-α, iNOS, and ROI by lineage⁻ CD11c^low CD11b⁺ Gr1⁺/Ly6C⁺ spleen cells from C57BL/6 mice 24 (c) and 48 h (d) after i.v. infection with 7 × 10⁵ bacteria and 7 × 10³ bacteria, respectively. The experiment was repeated five times with similar results. Error bars show SD. (e) iNOS⁺ cells in the spleen of BALB/c mice 24 h after infection with 10⁵ WT Lm (red) express CD11b and Ly6C and are found in the perifollicular area and T cell area. B cell follicles are labeled with anti-B220 antibodies (green) and T cells are labeled with anti-CD3 antibodies (blue). Bar, 100 µm. The experiment was repeated five times with similar results.