Fig. 1.

Apoptotic cells induce CCL22 expression in splenic CD169⁺ metallophillic macrophages. (A) B6 mice were injected with 10⁷ apoptotic thymocytes i.v. and at indicated time points, whole-spleen lysates were tested for CCL22. (B) At 4 h after apoptotic cell injection as in A, phagocytes were sorted based on indicated markers by FACS, and CCL22 message was measured by sqPCR. (C) B6.CD169^DTR mice were depleted of MMΦs as described (13) and injected with apoptotic thymocytes as described in A. At 24 h later, CCL22 protein was measured in whole-spleen lysates by ELISA. Immunofluorescence shows representative spleen sections from B6.CD169^DTR and littermate control groups 48 h after last injection with diphtheria toxin stained for B-cell markers (αB220, red) and αCD169 (green). (D) Splenic CD169⁺ cells were analyzed by FACS for uptake of Pkh26-labeled apoptotic cells 30 min after injection with 10⁷ cells i.v. (E) CD169⁺ and CD11c⁺ cells were sorted from B6 mice and cultured in complete RPMI with apoptotic thymocytes at a 10:1 apoptotic cell/phagocyte ratio for 4 h. CCL22 message was then measured by sqPCR as described. (F) At 4 h after apoptotic cell injection as in A, CD169⁺ MΦs were sorted by FACS and chemokine message for the species indicated was measured by sqPCR. (G) FACS-purified splenic CD169⁺ macrophages were incubated with apoptotic cells at a 10:1 ratio. At 24 h, supernatant from the cultures were analyzed for chemotactic activity against MACS-purified CD4⁺CD25⁺ Tregs. In some wells, neutralizing αCCL22 antibody or CCR4 antagonist was added to confirm the dependence of migration on CCL22-dependent mechanisms as described in Materials and Methods. Bars represent mean value for five mice ± SD (A, C, and D), pooled samples from three mice (B and F), or triplicate wells (E and G). *P^val ≤ 0.05; **P^val < 0.01 as determined by Student t test. Experiments were repeated at least three times with similar results. ND, not detected.