Clonal deletion and the fate of autoreactive thymocytes that survive negative selection

Abstract

Clonal deletion of autoreactive thymocytes is important for self-tolerance, but the intrathymic signals that induce clonal deletion have not been clearly identified. We now report that clonal deletion during negative selection required CD28-mediated costimulation of autoreactive thymocytes at the CD4(+)CD8(lo) intermediate stage of differentiation. Autoreactive thymocytes were prevented from undergoing clonal deletion by either a lack of CD28 costimulation or transgenic overexpression of the antiapoptotic factors Bcl-2 or Mcl-1, with surviving thymocytes differentiating into anergic CD4(-)CD8(-) double-negative thymocytes positive for the T cell antigen receptor αβ subtype (TCRαβ) that 'preferentially' migrated to the intestine, where they re-expressed CD8α and were sequestered as CD8αα(+) intraepithelial lymphocytes (IELs). Our study identifies costimulation by CD28 as the intrathymic signal required for clonal deletion and identifies CD8αα(+) IELs as the developmental fate of autoreactive thymocytes that survive negative selection.

Figure 1

TCRαβ⁺DN thymocytes are increased in CD28 costimulation-deficient mice. (a) CD4 versus CD8 staining of thymocytes from wildtype and costimulation-deficient (Cd28^−/− and B7-deficient) mice on C57BL/6 (left) and BALB/c (right) backgrounds (top panels). Numbers of total thymocytes and frequencies of DN cells are indicated. Surface TCRβ expression on gated DN thymocytes is also shown (lower panels). (b) Quantitation (frequency and number) of TCRαβ⁺DN thymocytes in wildtype and costimulation-deficient mice. Displayed are the mean + S.E. of each group which was compared to that from WT mice. p-values were obtained by Student’s two-tailed t-test. *, p<0.01; **, p<0.001; ***, p<0.0001. Data represent at least five independent experiments. (c) Profile of CD1d-PBS57 tetramer staining of TCRαβ⁺DN thymocytes from wildtype (shaded curve) and B7-deficient (bold line) mice.

Figure 2

Effect of Bcl-2 and Mcl-1 pro-survival transgenes on the appearance of TCRαβ⁺DN thymocytes. (a) CD4 versus CD8 staining of thymocytes from wildtype and transgenic mice (top panels). Numbers of total thymocytes and frequencies of DN cells are indicated. Surface TCRβ and CD5 expression on gated DN thymocytes is also shown (middle and lower panels). (b) Quantitation of TCRαβ⁺DN thymocytes in wildtype and transgenic mice. Displayed are the mean + S.E. of each group which was compared to that from WT mice. *, p<0.05; **, p<0.001; Data represent four independent experiments.

Figure 3

Expression of Mtv-reactive TCR-Vβs in pre- and post-selection thymocyte subsets. (a) Expression of Mtv-reactive TCR-Vβs in different thymocyte subsets from wildtype (WT) and costimulation-deficient mice on the BALB/c background. Displayed are the mean + S.E. of each group which was compared within each group to that from WT mice. The horizontal dashed line indicates the frequency of pre-selection DP thymocytes in WT mice expressing the particular TCR-Vβ specificity. (b) Quantitation (frequency and number) of specific TCR-Vβs expressed by TCRαβ⁺DN thymocytes from BALB/c crossed with C57BL/6 (CB6) mice that were either WT or transgenic for Mcl-1 or Bcl-2. Displayed are the mean + S.E. of each group which was compared within each group to that from WT mice. Data represent at least three independent experiments with at least five mice per group. *, p<0.01; **, p<0.001; ***, p<0.0001.

Figure 4

TCRαβ⁺DN thymocytes are the progeny of DP thymocytes. (a) The methylation status of the Cd8b promoter in Bcl-2 transgenic thymocytes was determined by bisulfite conversion and sequencing of genomic DNA from sorted thymocyte subpopulations (n=3). Data are presented as the frequency of Cd8b promoter sequences isolated from each cell population that were methylated or unmethylated as described and displayed in Supplementary Figure S2. (b) CD4 versus CD8 staining of thymocytes from the indicated mouse strains (top panels). Numbers of total thymocytes and frequencies of DN cells are indicated. Surface TCRβ expression on gated DN thymocytes is also shown (lower panels). Data are from two independent experiments.

Figure 5

Impact of thymic selection on expression of Mtv-reactive TCR-Vβs. (a) The frequency of cells expressing Mtv-reactive TCR-Vβs among post-selection (CD69⁺CD4⁺CD8^lo intermediate (INT) and SP) thymocytes was expressed relative to that among pre-selection DP thymocytes which was set = 100%. Displayed are the mean + S.E. of each group. Data are summaries of five independent experiments. (b) Surface CD28 expression on INT and TCRαβ⁺DN thymocyte subsets from each mouse strain was quantified as mean fluorescence intensity and was expressed relative to that of preselection DP thymocytes which was set = 100%. Data are representative of two independent experiments. (c) TCR induced calcium mobilization was assessed in CD8- depleted thymocytes. Biotinylated anti-TCR mAb (5 μg/ml) was crosslinked by avidin as indicated by arrows. Data are representative of three independent experiments. (d) Purified TCRαβ⁺DN and SP8 thymocytes (2×10⁴/well) were stimulated with plate-bound anti-TCR+CD28 (5μg +10μg) in the presence or absence of exogenous IL-2 (200 U/ml). Data represent mean c.p.m. + S.E. of triplicate cultures.

Figure 6

Developmentally diverted TCRαβ⁺DN thymocytes migrate to the intestine where they become CD8αα⁺ IELs. (a) Homing and fate of developmentally diverted TCRαβ⁺DN thymocytes. 0.8×10⁶ sorted CD5⁺DN thymocytes or CD5⁺CD8⁺ LN T cells from B7-deficient mice (left panels) were adoptively transferred into Rag2^−/− recipients. Five weeks later lymph nodes and small intestine of recipient animals were evaluated for the presence of TCRαβ⁺ T cells (right panels). The frequency of cells in each gate is indicated. (b) Effect of in vitro TCR+IL-15 stimulation of developmentally diverted TCRαβ⁺DN thymocytes from B7-deficient and Bcl-2 transgenic mice. 2×10⁴ sorted CD5⁺DN thymocytes/well (left panels) were cultured for 4d in either medium alone or with platebound anti-TCRβ mAb and soluble IL-15 (100ng/ml) and then analyzed. (c) CD8α versus CD8β expression is shown on TCRβ⁺ cells from the small intestine of the indicated mouse strains. The frequency of cells in each gate is shown. Data are representative of two independent experiments.

Figure 7

Developmentally diverted TCRαβ⁺DN thymocytes become CD8αα IELs. (a) Expression of Mtv-reactive TCR-Vβ5,11, and 12 in different thymocyte subsets and IELs from B7-deficient (BALB/c) mice. Displayed are the mean + S.E. of each group. Data represent three independent experiments. (b) Total number of TCRαβ⁺CD8αβ and CD8αα IELs in BALB/c and B7-deficient (BALB/c) mice. Displayed are the mean + S.E. of each group from three independent experiments. **, p<0.002. (c) Expression of Mtv-reactive TCR-Vβ5,11, and 12 in different T cell subsets in BALB/c mice. Displayed are the mean + S.E. of each group from three independent experiments. *, p<0.05; **, p<0.01. (d) Sorted TCRαβ⁺CD8αα⁺ IELs and SP4 LNT cells (2.5×10⁴/well) were stimulated with soluble anti-CD3 mAb (5 μg) and irradiated syngenic APC in the presence or absence of recombinant IL-2 (200 U/ml) or IL-15 (100ng/ml). Proliferation was assessed by [³H]-thymidine incorporation and data represent mean c.p.m. + S.E. of triplicate wells.

Figure 8

Runx3 is required for differentiation of TCRαβ⁺DN thymocytes into developmentally diverted CD8αα⁺ IELs. (a) Runx3d reporter expression in thymocyte subsets and IELs. Runx3-YFP reporter expression was assessed in the indicated T cell populations from Bcl2 transgenic mice (left panel). Displayed are the mean + S.E. of Runx3-YFP expression in each group (right panel). Data are representative of three independent experiments. (b) Effect of in vitro TCR+IL-15 stimulation of developmentally diverted TCRαβ⁺DN thymocytes from Runx3-sufficient (Runx3^+/YFP) and Runx3-deficient (Runx3^YFP/YFP) Bcl-2 transgenic mice. 2×10⁴ sorted CD5⁺DN thymocytes/well (left panels) were cultured for 4d in either medium alone or with plate- bound anti-TCRβ mAb and soluble IL-15 (100 ng/ml) and then analyzed. Runx3d reporter expression was quantified as mean fluorescence intensity of YFP and is indicated. (c) Total number of TCRαβ⁺DN thymocytes and TCRαβ⁺CD8αα⁺ IELs in Runx3-sufficient and Runx3-deficient Bcl2 transgenic mice. Displayed are the mean + S.E. of each group. Data are representative of three independent experiments. *, p<0.01.