Down-Regulation of CD62L Shedding in T Cells by CD39+ Regulatory T Cells Leads to Defective Sensitization in Contact Hypersensitivity Reactions

Abstract

Injection of regulatory T cells (Tregs) followed by sensitization with 2,4,6-trinitrochlorobenzene induced a transient increase in size and cellularity of skin-draining lymph nodes (LNs) in mice. This led us to hypothesize that Tregs may affect the trafficking of T cells from and to peripheral LNs. Two to three hours after sensitization, we found fewer CD8⁺ T cells expressing CD62L in LNs compared with untreated controls. Injection of wild-type Tregs prevented this down-regulation of CD62L. In contrast, Tregs devoid of the adenosine triphosphate (ATP)-degrading ecto-enzyme CD39 were unable to do so. As for the mechanism of CD62L regulation, we found that ATP, which is released in skin upon hapten-exposure, is inducing the protease ADAM17 in LN-residing T cells via engagement of P₂X₇ ATP receptors. ADAM17 cleaves CD62L from the surface of CD8⁺ T cells, which in turn provide a signal for T cells to leave the LNs. This regulation of CD62L is disturbed by the presence of Tregs, because Tregs remove extracellular ATP from the tissue by activity of CD39 and, therefore, abrogate the shedding of CD62L. Thus, these data indicate that the regulation of ATP turnover by Tregs in skin and LNs is an important modulator for immune responses.

Figure 1. Tregs degrade extracellular ATP and induce accumulation of CD8⁺ T cells in draining LNs

(a) LNs were isolated from mice sensitized at the abdomen (15 μl; 1% TNCB) and killed after indicated time points. Single-cell suspensions were prepared by disrupting the LNs with tweezers, and ATP was measured immediately *P < 0.05 (t test). (b) Mice were intravenously injected with 3 × 10⁶ Treg and sensitized with TNCB at the abdomen. After 2 hours, LNs were treated as in a. *P < 0.05 (t test). (c) Mice were treated with Tregs and sensitized at the ears (10 μl; 1% TNCB). After 3 hours, ear-draining LNs were removed, and cells were counted. *P < 0.05 (analysis of variance). (d) Cells prepared as in c were analyzed by FACS. *P < 0.05 (analysis of variance). All experiments were carried out five times. ATP, adenosine triphosphate; h, hours; LN, lymph node; M, mol/L; TNCB, 2,4,6-trinitrochlorobenzene; Treg, regulatory T cell; wt, wild type.

Figure 2. CD39^−/− Tregs fail to block down-regulation of CD62L in CD8⁺ T cells and suppress egress of CD8⁺ T cells

(a) LN cells from mice treated with 3 × 10⁶ Tregs as indicated and sensitized at the ears (10 μl; 1% TNCB) were analyzed for CD8 and CD62L expression by flow cytometry. Numbers in quadrants depict percentages of CD62L⁺ cells among CD8⁺ cells. Bar graph depicts the mean values of five independent experiments. *P < 0.05 (t test). (b) A total of 3 × 10⁶ CD8⁺PKH-PE⁺ T cells were intravenously injected into recipients, followed by intravenous injection of Tregs and sensitization at the ears. 2 hours later, CD8⁺PKH-PE⁺ T cells were analyzed in the LNs. Numbers reflects percentages of PKH-PE⁺ cells among all CD8⁺ T cells. Bar graph shows mean values of four independent experiments. *P < 0.05 (t test). h, hours; TNCB, 2,4,6-trinitrochlorobenzene; Treg, regulatory T cell; WT, wild type.

Figure 3. Degradation of extracellular ATP by Tregs down-regulates IFN-γ secretion by CD8⁺ T cells

(a) Animals (n = 4) were injected with Tregs and sensitized with TNCB. After 3 hours, CD8⁺ LN cells were prepared, and 1 × 10⁶ T cells were co-cultured with 5 × 10⁴ hapten-pulsed DCs in 1 ml complete medium. Tissue culture supernatants were tested for IL-10 and IFN-γ by commercial ELISA after 48 hours. Shown are means ± standard deviations of four independent experiments. *P < 0.05 (t test). (b) 1 × 10⁶ CD8⁺ T cells from sensitized animals (n = 4) were isolated and restimulated with 5 × 10⁵ hapten-pulsed DC_hap or control DCs in 1 ml of complete medium. To some cultures anti-CD62L or isotype antibodies (10μg/ml) were added as indicated. Tissue culture supernatant was tested for IFN-γ after 48 hours. Shown are means ± standard deviations of four independent experiments. *P < 0.05 (t test). ATP, adenosine triphosphate; DC, dendritic cell; h, hours; LN, lymph node; TNCB, 2,4,6-trinitrochlorobenzene; Treg, regulatory T cell; wt, wild type.

Figure 4. ATP induces shedding of CD62L that is blocked by Tregs

(a) CD8⁺ cells were cultivated with ATP (3 hours) and CD62L expression was determined. Shown are means ± standard deviations of four experiments. (b) LN cells were cultured with 500 μmol/L ATP, Tregs, KN62 (10 μmol/L), or ATPase (2U/ml) (3 hours), and CD8/CD62L expression was analyzed. (c) Ratios of CD8⁺ T cells and Tregs (left) and LN cells and Tregs (4:1; right) were cultivated with 500 μmol/L of ATP (3 hours), and CD62L⁺ expression was analyzed. Means ± standard deviations of four experiments are shown. (d) LN cells were incubated with TNBS and Tregs as indicated, and ATP was determined. Dot plots show CD8 and CD62L expression, and numbers indicate percentage of CD62L⁺CD8⁺ cells. (e) Tissue culture supernatants of 5 × 10⁶ control and/or ATP-treated LN cells were immunoprecipitated with anti-CD62L antibodies followed by SDS-PAGE. ATP, adenosine triphosphate; LN, lymph node; M, mol/L; ns, not significant; TNCB, 2,4,6-trinitrochlorobenzene; Treg, regulatory T cell; WT, wild type.

Figure 5. Sensitization induces phosphorylation of ERK and increased activation of ADAM17

(a) CD8⁺ T cells isolated from LNs were treated with graded doses of ATP for 3 hours as indicated. Thereafter, cell lysate was analyzed for ADAM17, actin, ERK, and pERKexpression by Western blot (n =3; shown is a typical result). (b) CD8⁺ T cells isolated from LNs were treated with ATP and the ADAM17 antagonist TAPI for 3 hours as indicated. Thereafter, expression of CD62L by CD8⁺ T cells was analyzed (n = 3). (c) Mice were intravenously injected with 3 × 10⁶ Tregs or phosphate buffered saline and sensitized (10 μl; 1% TNCB) at the ear. 3 hours later, CD8⁺ T cells were isolated from LNs, and ADAM17 and actin expressions were determined by Western blot (n = 3; shown is a typical experiment). ATP, adenosine triphosphate; ERK, extracellular signal-regulated kinase; LN, lymph node; M, mol/L; pERK, phosphorylated extracellular signal-regulated kinase; TAPI, N-(R)-[2-(hydroxyaminocarbonyl) methyl]-4-methylpentanoyl-L-naphthylalanyl-L-alanine, 2-aminoethyl amide; TNCB, 2,4,6-trinitrochlorobenzene; WT, wild type.