Hepatic stroma-educated regulatory DCs suppress CD8+ T cell proliferation in mice

Abstract

Liver dendritic cells (DCs) display immunosuppressive activities and inhibit the CD4⁺ T cell response. The present study assessed whether and how liver DCs suppress CD8⁺ T cells. We found that bone marrow-derived mature DCs incubated with liver stromal cells were characterized by a longer life span, reduced CD11c, IA/IE, CD80, CD86, and CD40 expression, and increased CD11b expression. These unique liver stromal cell-educated mature DCs (LSed-DCs) stimulated CD8⁺ T cells to express CD25 and CD69, but inhibited their proliferation. CD8⁺ T cell suppression depended on soluble factors released by LSed-DCs, but not cell-cell contact. Compared with mature DCs, LSed-DCs produced more nitric oxide and IL-10. Addition of a nitric oxide synthase inhibitor, PBIT, but not an IL-10-blocking mAb, reversed LSed-DC inhibition of CD8⁺ T cell proliferation. We also found that LSed-DCs reduced CD8⁺ T cell-mediated liver damage in a mouse model of autoimmune hepatitis. These results demonstrate that the liver stroma induces mature DCs to differentiate into regulatory DCs that suppress CD8⁺ T cell proliferation, and thus contribute to liver tolerance.

Keywords: CD8+ T cells; Immune response; Immunity; Immunology and Microbiology Section; autoimmune hepatitis; liver; nitric oxide; regulatory dendritic cells.

Figure 1. LSed-DC morphology and phenotype

Purified BM-derived mDCs were seeded onto LSC monolayers at 2×10⁶ cells/well in 6-well plates. mDC morphology was monitored by phase-contrast microscopy (400×) A. After two weeks of incubation with LSCs, LSed-DCs were detected by flow cytometry B. BM-derived mDCs and imDCs were used as controls. Data are representative of at least three independent experiments.

Figure 2. CD8⁺ T cell activation by LSed-DCs

Purified OT-1CD8⁺ T cells (2×10⁵) were co-cultured with OVA_257-264-loaded mDCs (2×10⁴) and/or OVA_257-264-loaded LSed-DCs (2×10⁴) for 48 h. Cells were collected and CD3⁺/CD8⁺ T cells were gated for analysis of CD25 and CD69 expression by flow cytometry A. Co-culture supernatants were collected for analysis of IL-2 and IFN-γ via ELISA B. Data are presented as means±SD of triplicate wells, and represent three independent experiments. ***P < 0.001, ANOVA.

Figure 3. CD8⁺ T cell suppression by LSed-DCs in vitro and in vivo

CFSE-labeled OT-1 splenic CD8⁺ T cells (2×10⁵) were co-cultured with OVA_257-264-loaded mDCs (2×10⁴) in the presence or absence of LSed-DCs (2×10⁴) for 48 h in vitro. CFSE was analyzed in gated CD3⁺/CD8⁺ T cells, and histograms showed relative CD8⁺ T cell numbers as counted by flow cytometry A. OT-1 CD8⁺ T cells (1.5×10⁶) with OVA_257-264-loaded mDCs (1.5×10⁵) and/or LSed-DCs (1.5×10⁵) were transferred intravenously into naive C57BL/6 mice for three days. Mononuclear cells from blood were separated and stained for analysis of OT-1 CD8⁺ T cell (CD3⁺/CD8⁺/Vβ5.1/5.2 TCR⁺) frequency B. Histograms showed relative transferred OT-1 CD8⁺ T cell numbers as counted by flow cytometry. Data are presented as means±SD of triplicate wells, and are representative of at least two independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001, ANOVA.

Figure 4. The role of IL-10 in LSed-DC-mediated CD8⁺ T cell suppression

Purified LSed-DCs (2×10⁴) were cultured for 48 h. CD8⁺ T cell proliferation was assessed following stimulation with paraformaldehyde-fixed LSed-DCs or LSed-DC culture supernatants A. Supernatants from cultures containing mDCs or LSed-DCs were collected at 48 h for analysis of IL-10 via ELISA B. Anti-IL-10 mAb and matched isotype were added to co-cultures containing LSed-DCs for 48 h, and relative CD8⁺ T cell numbers and IL-2 production were examined by flow cytometry and ELISA C. Data are presented as means±SD of triplicate wells, and represent three independent experiments. **P < 0.01, ***P < 0.001 ANOVA.

Figure 5. The role of NO in LSed-DC-mediated CD8⁺ T cell suppression

Supernatants from cultures containing mDCs (2×10⁴), or LSed-DCs (2×10⁴) with or without LPS (0.5 μg/mL) were collected at 24 h for analysis of NO via Griess assay A. The NO donor, NOC-18 (10 μM), was added to mDC/CD8⁺ T cell co-cultures, and the NOS inhibitor, PBIT (10μM), was added to LSed-DC/mDC/CD8⁺ T cell co-cultures. 48 h later, relative CD8⁺ T cell numbers were examined using flow cytometry B. Data are presented as means±SD of triplicate wells, and represent three independent experiments. *P < 0.05, **P < 0.01, ANOVA.

Figure 6. Suppression of AIH by liver LSed-DCs

AIH was induced by transfer of HBV-specific CD8⁺ T cells (1×10⁸) into HBV transgenic mice. LSed-DCs (2×10⁷) were also transferred together with HBV-specific CD8⁺ T cells into HBV transgenic mice. HBV transgenic mice receiving CD8⁺ T cells from WT mice or no CD8⁺ T cells, and WT mice receiving HBV-specific CD8⁺ T cells were used as negative controls. Serum AST was detected at different time points. Data are presented as means±SD of triplicate wells, and represent at least two separate experiments. *P < 0.05, ANOVA.