Figure 2.

Short-term homing of adoptively transferred L-selectin⁺CCR2⁺ monocytes to inflamed PLNs is mediated by MCP-1. (A) Flow cytometry of PBMCs from CX3CR1^+/GFP knockin mice stained for the monocyte-macrophage markers F4/80 or CD11b (Mac-1) reveals two populations of GFP⁺ leukocytes based upon differential expression of GFP, L-selectin (CD62L), and CCR2. The F4/80⁺GFP⁻ cells were identified as eosinophils by their high side scatter (data not shown). (B) Flow cytometric analysis of leukocyte populations in homing assays. CX3CR1^+/GFP PBMCs were injected intravenously into wild-type recipients 5–7 d after induction of skin inflammation by CFA/KLH injection. 4 h later, GFP⁺ donor populations in recipient blood and inflamed PLNs were compared with input PBMCs. To avoid counting of contaminating GFP⁺ NK cells, all samples were stained with anti-NK1.1. Numbers in dot plots indicate the percentage of GFP⁺ cells in the GFP^MED and GFP^HIGH gates in one representative experiment (out of 6). (C) GFP^MED cells (white bars), but not GFP^HIGH cells (black bars) were preferentially recruited to inflamed PLNs. Data presented as mean ± SEM, n = 6 mice per group. **P < 0.01 vs. frequency of input GFP^MED CX3CR1^+/GFP PBMCs in the input. ^## P < 0.01 vs. input frequency of GFP^HIGH CX3CR1^+/GFP PBMCs. (D) Homing of all GFP⁺ and GFP^MED CX3CR1^+/GFP NK1.1⁻ leukocytes to inflamed subiliac PLNs (black bars) and to contralateral noninflamed PLNs (white bars). The inflammation-induced increased homing of the GFP^MED subset was completely blocked by anti–MCP-1. Anti–MCP-1 treatment did not affect homing of GFP^HIGH cells and did not alter CX3CR1^+/GFP NK1.1⁻ cell numbers in recipient blood or spleen (data not shown). Data presented as mean ± SEM, n = 5 mice per group. *P < 0.05 control mAb inflamed versus contralateral PLNs. ^# P < 0.05 vs. control mAb-treated inflamed PLNs.