Tumor Rejection by in Vivo Administration of Anti-CD25 (Interleukin-2 Receptor α) Monoclonal Antibody¹

Immune regulation has been shown to be involved in the progressive growth of some murine tumors. North and Bursuker (1) demonstrated that CD4⁺ T cells from BALB/c mice carrying syngeneic methylcholanthrene-induced Meth A sarcoma down-regulated the activation of immune effector cells against the tumor. Similar findings were obtained with other murine tumors (2, 3, 4). We previously demonstrated a dominant antigen peptide, pRL1a, that was recognized by cytotoxic T lymphocytes on BALB/c radiation-induced leukemia RL♂1 (5). The pRL1a peptide was derived from the 5′-untranslated region of c-akt that became translated by insertion of the long terminal repeat (6). Overexpression of the altered Akt molecules seemed to induce CD4⁺ immunoregulatory cells, which resulted in progressive RL♂1 growth in BALB/c mice. In vivo depletion of CD4⁺ T cells from BALB/c mice caused RL♂1 regression (7).

Recently, CD4⁺CD25⁺ cells have been shown to represent a unique population of immunoregulatory cells (8, 9, 10, 11, 12, 13). Transfer of BALB/c spleen cells depleted of CD25⁺ cells into BALB/c nu/nu mice induced various autoimmune diseases (11). In addition, in vivo administration of anti-CD25 mAb³ induced autoimmune diseases in (B6 × A/J)F₁ mice (14).

In this study, we investigated the effect of in vivo administration of anti-CD25 mAb on the growth of eight tumors—RL♂1 and four other leukemias, a myeloma, and two fibrosarcomas—that grew progressively in syngeneic mice. We found that a single injection of less than 0.125 mg of anti-CD25 mAb (PC61) caused regression in six of the eight tumors, including RL♂1. After antibody treatment, a reduction in the number of CD4⁺CD25⁺ cells was observed by flow cytometry, which suggested that effective tumor rejection responses resulted from a depletion of CD4⁺CD25⁺ immunoregulatory cells.

A/St strain of mice was derived from a colony at the Texas University College of Medicine (Houston, TX). BALB/c and B6 mice were purchased from Japan SLC (Shizuoka, Japan). Breeding pairs of AKR/J mice were provided by Dr. T. Shiroishi (National Institute of Genetics, Mishima, Japan). Breeding pairs of CB-17 SCID mice were provided by Dr. K. Kuribayashi (Mie University School of Medicine, Mie, Japan). BALB/c athymic (nu/nu) mice were produced by breeding male athymic (nu/nu) mice with female nu/+ mice at our animal center.

The tumor cell lines used and their derivation are listed in Table 1.

Anti-L3T4 (CD4) mAb GK1.5 (15) was provided by Dr. F. Fitch (University of Chicago, Chicago, IL). Anti-Lyt-2.2 (CD8) mAb 19/178 (16) was provided by Dr. U. Hämmerling (Memorial Sloan-Kettering Cancer Center, New York, NY). Anti-CD25 IL-2Rα mAb produced by hybridoma PC61 (17) was a rat IgG1 antibody. Another anti-CD25 mAb produced by hybridoma, 7D4 (18), was a rat IgM antibody. For in vivo administration, anti-CD25 mAb (PC61) was used after purification. The hybridoma ascites produced in CB-17 SCID mice was purified to homogeneity by ammonium sulfate precipitation, followed by chromatography on a DEAE Toyopearl 650S column (Tosoh, Tokyo, Japan). The concentration of IgG was determined from its absorbance at 280 nm as an absorption coefficient value of 1.5.

Anti-L3T4 (CD4) mAb and anti-Lyt-2.2 (CD8) mAb were used in the form of ascites from hybridoma-bearing mice as described previously (19). Depletion of CD4 and/or CD8 T cells by in vivo administration of its respective mAb was confirmed as described previously (19). Normal rat IgG was obtained from Caltag (Burlingame, CA).

Cells (1 × 10⁶) were washed and incubated with mAb for 30 min at 4°C in 2% FCS-containing PBS. The following mAbs were used: (a) PE-conjugated anti-L3T4 (CD4) mAb (GK1.5; Becton Dickinson Co., Mountain View, CA); (b) PE-conjugated anti-Lyt-2.2 (CD8) mAb (KT15; Serotec Ltd., Kidlington, Oxford, England); (c) PE-conjugated anti-CD3ε mAb (145-2C11); and (d) FITC-conjugated anti-CD25 (IL-2Rα) mAb (7D4; PharMingen Co., San Diego, CA). After treatment, the cells were washed, suspended in PBS, and analyzed on a FACScan (Becton Dickinson).

Tumor cells (in 0.2 ml) were injected intradermally into the backs of mice with a 27-gauge needle. Before inoculation of tumor cells, the hair was cut with clippers. The diameter of the tumors was measured with Vernier calipers twice at right angles to calculate the mean diameter.

The mice were anesthetized with ether, and a volume of 0.2 ml of mAb diluted in PBS was injected through the retrobulbar venous plexus.

We first investigated the effect of the in vivo administration of anti-CD25 mAb (PC61) on a CD25⁺ population in lymphoid tissues. PC61 antibody blocks IL-2 binding to the receptor (17), and anti-CD25 mAb 7D4 does not block IL-2 binding (18). For detection of CD25⁺ cells in lymph node cells from PC61-treated mice, we used 7D4 with flow cytometry. As shown in Fig. 1 and Fig. 2, CD25⁺ cells consisted of ∼10% CD4⁺ cells and less than 1% CD8⁺ cells among the lymph node cells from untreated mice, which was consistent with previous results (11, 12, 13). CD4⁺CD25⁺ cells reduced maximally on days 3–4 and fully recovered by day 9 after a single in vivo administration of 0.25 mg anti-CD25 mAb (PC61). The reduction was observed in the range of 70–80% at doses between 0.125 and 0.75 mg. For subsequent analyses, we used a single injection of 0.25 mg PC61 on day −4 unless otherwise stated.

The tumors used were listed in Table 1. We used two spontaneously occurring leukemias, two radiation-induced leukemias, a dimethylbenzanthracene-induced leukemia, a mineral oil-induced myeloma and two methylcholanthrene-induced sarcomas. As shown in Fig. 3 and 4, all of the tumors grew progressively in syngeneic mice and eventually killed them. A single administration of 0.25 mg anti-CD25 mAb (PC61) on day −4 caused regression in six of the eight tumors. Administration of normal (control) rat IgG had no effect. No recurrence of tumors was observed thereafter. Among the tumors used, only ASL1 and RL♂1 expressed CD25 on the cell surface. Administration of anti-CD25 mAb (PC61) had no effect on the growth of tumors with either CD25⁺ or CD25⁻ phenotype in BALB/c nu/nu mice.

Next, the timing of the administration of anti-CD25 mAb (PC61) on tumor growth was examined. As shown in Fig. 5, administration on days −4, −2, 0, and 1, but not later than day 2, even with reduction of CD4⁺CD25⁺ T cells after MOPC-70A inoculation, caused regression.

The dose effect of anti-CD25 mAb (PC61) on tumor growth was then examined. As shown in Table 2, a single administration of 0.125, 0.25, 0.5, or 0.75 mg on day −4 caused MOPC-70A regression. At doses of 0.03 and 0.06 mg, regression was observed in 0 of 3 and 2 of 3 mice inoculated with the tumor, respectively.

To investigate the involvement of CD4 and CD8 T cells in tumor regression caused by the administration of anti-CD25 mAb (PC61), the effect of the coadministration of anti-CD25 mAb (PC61) and anti-CD4 mAb and/or anti-CD8 mAb was investigated. We showed previously (20, 21) that in high responder (BALB/c × B6)F₁ mice, CD8 T cells were required for rejection in all of the seven tumors from the BALB/c, B6, and (BALB/c × B6)F₁ mice investigated, but the requirement of CD4 T cells differed depending on the tumor. Meth A and MOPC-70A were representative tumors that did and did not require CD4 T cells, respectively. As shown in Fig. 6, in the case of MOPC-70A, coadministration with anti-Lyt-2.2 (CD8) mAb, but not anti-L3T4 (CD4) mAb, inhibited the regression by anti-CD25 mAb (PC61). On the other hand, in the case of Meth A, coadministration with either anti-Lyt-2.2 (CD8) mAb or anti-L3T4 (CD4) mAb inhibited the regression.

As shown in Fig. 7, no MOPC-70A or RL♂1 growth was observed even at higher doses in BALB/c mice that rejected MOPC-70A or RL♂1, respectively, by anti-CD25 mAb (PC61).

In this study, we demonstrated that in vivo administration of anti-CD25 mAb (PC61) caused regression of tumors that grew progressively in syngeneic mice and killed them eventually. The tumors used were five leukemias, a myeloma, and two sarcomas derived from four different inbred mouse strains. The effect of anti-CD25 mAb (PC61) was observed on six of the eight tumors. Administration of anti-CD25 mAb (PC61) resulted in a reduction in CD4⁺CD25⁺ cells in the peripheral lymphoid tissues. These findings suggested that CD4⁺CD25⁺ immunoregulatory cells were involved in the growth of those tumors. Tumor regression was observed even 2–3 months after PC61-treatment. Kinetic analysis showed that the administration of anti-CD25 mAb (PC61) later than day 2 after tumor inoculation caused no tumor regression. This could be due to an insufficient activation of effector cells for proliferating tumor cells by late depletion of CD4⁺CD25⁺ T cells. Alternatively, it could be due to a failure in the activation of effector cells by depletion of CD4⁺CD25⁺ immunoregulatory cells after tumor antigen stimulation.

We previously identified a dominant rejection antigen peptide recognized by CTL on RL♂1 leukemia cells (5). Irrespective of the presence of the rejection antigen, RL♂1 continued to grow in syngeneic BALB/c mice and killed them eventually. Depletion of CD4⁺ T cells from the mice resulted in tumor regression (7), consistent with the present findings.

Thymectomy at day 3 after birth caused various autoimmune diseases (9, 22, 23, 24). CD4⁺CD25⁻ T cells were shown to be responsible for causing the diseases (8, 9). Transfer of CD4⁺CD25⁺ cells to those mice inhibited the occurrence of the autoimmune diseases (9). The CD4⁺CD25⁺ cells that appeared to represent a distinct lineage (11, 12, 13) down-regulated the induction and/or activation of those autoreactive CD4⁺ T cells from the CD4⁺CD25⁻ cell pool. Thymectomy at day 3 resulted in the disappearance of CD4⁺CD25⁺ cells, which constituted ∼10% of the CD4⁺ T cells in the peripheral lymphoid tissues, which suggests that those cells migrated from the thymus to those tissues on about day 3 after birth (9).

Taguchi and Takahashi (14) demonstrated the depletion of CD25⁺ cells and the occurrence of autoimmune diseases in (B6 × A/J)F₁ mice by in vivo administration of anti-CD25 mAb (PC61) 11 consecutive times every other day at a dose of 2 mg. In our study, a single injection at a dose of 0.125 mg was sufficient to cause regression of the tumors, and no histological indication of autoimmune disease and no autoantibody formation were observed in the mice 3 months after the antibody treatment (data not shown). These findings suggested that the effect of the PC61-treatment seemed to differ between the multitargeted autoimmune responses and the responses against the tumor.

Although the exact mechanisms of suppression by CD4⁺CD25⁺ cells in vivo are presently unknown, the in vitro studies by Thornton and Shevach (12) and Takahashi et al. (13) demonstrated that CD4⁺CD25⁺ cells suppressed the proliferation of CD4⁺CD25⁻ cells by specifically inhibiting the production of IL-2. Moreover, the inhibition required the activation of CD4⁺CD25⁺ suppressor cells via T-cell receptor for antigen, and mediation by cell contact but not by cytokines.

Coadministration with anti-CD8 mAb inhibited tumor regression by anti-CD25 mAb (PC61) alone, which suggests that CD8 T cells were responsible for those tumor regressions. Coadministration with anti-CD4 mAb had no effect on the regression of MOPC-70A but inhibited the regression of Meth A by PC61 alone. This suggested that the relative involvement of CD4⁺ T cells depended on the tumor, probably as helper T cells for the generation of CD8 effector cells, and was consistent with our previous results (20, 21). The lack of regression of AKSL2, a spontaneous leukemia derived from an AKR mouse and a RL♀8, a radiation-induced leukemia derived from a BALB/c mouse by PC61-treatment, together with the normal expression of H-2 class I antigens on those tumors (data not shown) suggested low or no antigenicity of those tumors for eliciting effective rejection responses in syngeneic mice.

We thank Drs. O. Taguchi and T. Takahashi (Aichi Cancer Institute, Nagoya, Japan) for providing antibodies. We also thank M. Isobe for excellent technical assistance and J. Mizuuchi for preparation of the manuscript.