Induction of Tumor Immunity by Removing CD25⁺CD4⁺ T Cells: A Common Basis Between Tumor Immunity and Autoimmunity

Two types of tumor-killing activity in the nude mice having rejected tumors after transfer of CD25⁻ cells. A, Spleen cells from individual nude mice (n = 6) that had received CD25⁻ cells and rejected RL♂1 (as shown in Fig. 1) were cultured for 7 days with or without RL♂1 (filled or open circles, respectively) and assessed for killing activity against RL♂1 or B16. The group of mice enclosed with circles or rectangles (designated I or II, respectively) in the right and left figures are the same. Dotted lines connect cytotoxic activities of RL♂1-stimulated or nonstimulated cultures from the same mice. B, Activity to kill RL♂1 or B16 in each representative mouse from the group enclosed with circles (type I) or rectangles (type II) in A is shown. Spleen cells from each mouse were cultured for 7 days with (•) or without RL♂1 (○). C, CD4/CD8 phenotype of cytotoxic cells. Individual mice from group I or II in A were assessed for cytotoxic activity at E:T ratio of 100 after depleting CD4⁺ cells, CD8⁺ cells, or both (designated CD4⁻ cells, CD8⁻ cells, or CD4⁻8⁻ cells, respectively) from the spleen cells cultured for 7 days with RL♂1, as shown in B.

Induction of tumor immunity by administering anti-CD25 mAb to normal mice. Eight-week-old BALB/c or B6 mice were i.v. injected with 1 mg each of purified PC61 on 4 and 2 days (filled arrows) before s.c. inoculation of 1 × 10⁵ RL♂1 or B16 cells (open arrow), and tumor growth was monitored for individual mice (five mice per group).

In vitro induction of cytotoxic lymphocytes by eliminating CD25⁺4⁺ T cells. A, CD25⁻ or uneliminated spleen cells prepared from normal BALB/c mice were cultured with or without RL♂1 for 5 or 7 days and assessed for activity to kill RL♂1 at E:T ratio of 100. Each circle represents the activity of a single culture. Vertical bars are SDs of the mean. The inset shows a representative cytotoxic activity of CD25⁻ (square) or uneliminated (circle) spleen cells, or CD25⁻ cells mixed with purified CD25⁺CD4⁺ T cells (triangle), cultured for 7 days with (filled symbols) or without RL♂1 (open symbols). B, CD25⁻ spleen cells from various strains were cultured for 7 days and assessed at E:T ratio of 100 for activity to kill various syngeneic or allogeneic tumor cells. Average activity of four independent experiments and SDs is shown.

In vitro generation of TCR⁻ CD4⁻8⁻ NK cells. A, BALB/c CD25⁻ cells were cultured for 7 days without tumor stimulation and assessed for activity to kill RL♂1 (○) or B16 (•) after depleting CD4⁺ cells, CD8⁺ cells, or both, by treating cells with specific Ab and C: a, C treatment alone; b, anti-CD4 and C; c, anti-CD8 and C; d, anti-CD4, anti-CD8, and C. The results of five independent experiments are shown. B, Anti-CD4 (○), anti-CD8 (•), anti-α/β-TCR (□), or anti-γ/δ-TCR Ab (▵) was added at various concentrations to the cytotoxicity assay with BALB/c CD25⁻ cells cultured for 7 days and RL♂1 cells as the targets. As a positive control of blocking by anti-CD8 or anti-α/β-TCR Ab, Abs were added to the killing of RL♂1 cells by B6 spleen cells stimulated for 7 days with allogeneic RL♂1 cells (asterisks). A representative result of three independent experiments is shown. C, CD25⁻ spleen cells from B6 mice were depleted of B cells, cultured for 7 days with APCs, and stained with FITC anti-CD4, PE anti-CD8, and biotinylated anti-α/β-TCR, anti-γ/δ-TCR, or anti-NK1.1 Ab with Cy5-streptavidin as the secondary reagent. Expression of α/β-TCR, γ/δ-TCR, or NK1.1 on CD4⁻8⁻ cells in the left figure is shown as histograms (abscissa is staining intensity in logarithmic scale; ordinate is the number of cells as arbitrary units). A representative staining of three independent experiments is shown.

Absence of promiscuous killing activity of cultured CD25⁻ cells in beige mice. CD25⁻ or uneliminated spleen cells from B6-bg or B6 mice were cultured for 7 days, and assessed for killing activity against RL♂1 or B16 cells. Allospecific killing activity (asterisks) of uneliminated B6-bg or B6 spleen cells stimulated for 7 days with RL♂1 was also assessed against RL♂1 cells as the targets.

Requirement of CD25⁻4⁺ T cells or IL-2 for induction of CD4⁻8⁻ NK cells. A and B, CD25⁻ or uneliminated BALB/c spleen cells (filled or open circle, respectively) were cultured for various days and assessed for proliferative activity (A) and IL-2 activity in supernatants (B). These activities in the culture of CD25⁻ cells were also assessed after depleting CD4⁺ cells or CD8⁺ cells (filled or open triangle, respectively) on day 7 or 5 at the peak of proliferation or IL-2 secretion, respectively. C, Pan anti-class II MHC mAb (CA4) or control rat IgG was added at graded concentrations to 1-wk culture of CD25⁻ or uneliminated BALB/c spleen cells (filled or open columns, respectively) and assessed for the effect on proliferation. D, BALB/c spleen cells were depleted of a particular cell population(s) by treatment with specific Ab and C, then cultured for 7 days and assessed for activity to kill RL♂1 at E:T ratio of 100. Treatments of spleen cells before culture were as follows: a, C treatment alone; b, anti-CD25 and C; c, anti-CD25, anti-CD8, and C; d, anti-CD25, anti-CD4, and C; e, anti-CD4, anti-CD8, and C. IL-2 (1000 U/ml) was added to the culture of spleen cells treated with: f, anti-CD25, anti-CD4 Ab plus C; g, anti-CD4, anti-CD8 Ab plus C; or h, C alone. A representative result of three independent experiments is shown.

Inability of CD25⁻4⁺ T cells from IL-2-deficient mice to generate LAK/NK-like cells from normal CD4⁻8⁻ cells. CD25⁻4⁺ spleen cells from IL-2^−/−, IL-4^−/−, IFN-γ^−/−, or control BALB/c mice were mixed with an equal number of CD4⁻8⁻ spleen cells prepared from normal BALB/c mice, cultured for 7 days, and assessed for the activity to kill RL♂1 or B16 cells. A representative result of three independent experiments is shown.