Inducing lymphopenia before adoptive cell transfer can improve the antitumor effect of donor immune cells. It was recently reported that lymphopenic conditions can initiate the differentiation of naive T cells into effector cells. Although T cells require a specific “strong” signal via TCR as well as costimulatory signals during Ag-driven differentiation, there has been little evidence to suggest any requirement for costimulatory signaling for the differentiation of naive T cells in a lymphopenic host. In this study, we demonstrate that naive CD8+ T cells are indispensable for induction of antitumor effect, and, in addition to Ag-driven differentiation, CD28 signaling is essential for the differentiation of naive CD8+ T cells into functional effector CTLs during homeostatic proliferation (HP). The systemic administration of IL-2 did not restore the antitumor effect induced by HP in the absence of CD28 signaling. These results suggest that homeostatic cytokines enable CD8+ T cells to expand and survive, and that TCR and the CD28 signal initiate the differentiation of effector functions. A deeper understanding of the mechanisms underlying enhanced induction of the antitumor immune response with accompanying HP may allow us to more precisely induce enhanced immunity with costimulation signaling and the administration of common γ-chain cytokines.

The antitumor immune response is usually insufficient to bring about the regression of growing tumors in cancer-bearing patients, even when it appears that tumor-specific CTLs are being generated. This is because tumor antigenicity may frequently be weak or the tumor may have escape mechanisms that shield it from immune surveillance (1). Most cancer immunotherapies have focused on the induction and enhancement of cellular immune responses (2), since these responses play a key role in the immunologic rejection of allogeneic grafts and syngeneic tumors. To eradicate established tumors through an antitumor immune response, three conditions must be satisfied. First, sufficient numbers of immune cells with high affinity/avidity to tumor Ags must be generated in vivo. Second, tumor Ag-specific immune T lymphocytes must infiltrate the site of tumor growth. Finally, immune T cells must be activated at the tumor site where they can exert effects such as direct lysis or cytokine secretion, causing the ultimate distraction of the tumor. To meet these requirements, an alternate approach is being developed that involves the adoptive transfer of immune T lymphocytes, vaccination with tumor Ag peptides, and the administration of immunological cytokines. Ongoing clinical trials continue in some facilities (3, 4, 5).

The majority of attempts focus on increasing the number of immune T lymphocytes to allow for a sufficient quality to interact with tumor Ags. The adoptive transfer of immune T lymphocytes, which are generated with antigenic peptides and Th1 cytokines in vitro, augments the repertoire of tumor-specific CTLs in vivo (3, 5). However, infused T cells generally do not survive and expand efficiently in tumor-bearing patients (3, 4, 5, 6). In one study that used a mouse immunotherapy model, the systemic administration of cytokines, TNF-α, IL-12, and high dose IL-2 was conducted. However, because of the nonspecific activation of lymphocytes, patients undergoing immunotherapy with cytokines experience multiple side effects (i.e., fever, chills, diarrhea, confusion, skin rashes, and hepatic and renal dysfunction). Thus, systemic administration of IL-2 combined with peptide vaccination or cell transfer may provide a better approach to tumor immunotherapy (3, 7, 8). During the last decade, one strategy that has emerged focuses on manipulations that reduce the number of regulatory T cells before initiating immunotherapy to augment the immune responses of T lymphocytes (8). Based on this type of strategy, several studies have shown that nonmyeloablative chemotherapeutic treatments as well as irradiation can modulate the immune response and enhance the antitumor activity of adoptively transferred lymphocytes. These treatments induce severe lymphopenic conditions in the host and the subsequent spontaneous expansion of T cells in the periphery, a process known as homeostatic proliferation (9). Under the induced lymphopenia, a beneficial antitumor immune response is triggered by the concomitant induction of homeostatic T cell proliferation and the presentation of tumor-associated Ag (TAA)2 in the lymph nodes (3, 5, 10, 11).

Recent studies have revealed that this homeostatic proliferation of T cells requires a “weak” signal via TCR/CD3 bound to self-peptide/MHC complexes that control their positive selection in the thymus. Homeostatic proliferation also requires cytokine stimulation with the homeostatic cytokines IL-7 and/or IL-15 (9, 12, 13, 14, 15). As naive T cells undergo homeostatic proliferation in lymphopenic mice in the absence of overt antigenic stimulation, they progressively acquire the phenotypic and functional characteristics of Ag-specific memory T cells without developing of “effector” functions (16, 17). Thus, upon successive division, Ag-independent homing T cell proliferation does not result in up-regulation of activation markers, such as CD69 and CD25, but rather up-regulation of several memory markers, such as CD44 and CD122 (16, 17, 18). After acquisition of a memory-like phenotype, T cells exhibit a rapid anamnestic response upon exposure to cognate Ag (16, 17, 19).

Interestingly, unlike Ag-driven proliferation and differentiation, there has been little evidence to suggest any requirement for costimulatory signals and IL-2 signaling for the expansion and phenotypical changes of naive CD8+ and CD4+ T cells in various lymphopenic models (20, 21). Using CD28 knockout mice, one recent study has noted that costimulation is not required for homeostatic proliferation of polyclonal populations of peripheral CD8+ T lymphocytes. During homeostatic proliferation, the differentiation of naive CD8+ T cells into functional memory phenotype CD8+ T cells is dependent on T cell proliferation and, initially, on the presence of appropriate MHC/self-peptide complexes, without requiring IL-2 or costimulation via CD28 (16).

In contrast to homeostatic proliferation, T cells require a specific “strong” signal, delivered via the TCR as well as costimulatory signals. CD28 is a particularly important costimulatory molecule because it interacts with B7-1 (CD80) and B7-2 (CD86) on APCs to provide a key signal for the generation of T cell immunity (22, 23, 24). TCR/CD28 ligation up-regulates the transcription of several important genes (e.g., Bcl-xL, IL-2Rα, and IL-2), all of which are necessary for the proliferation and functional differentiation of naive T cells (22, 25). TCR/CD28 ligation also leads to IL-2 production in T cells, which is important since IL-2 is a key cytokine that induces precursor cells to differentiate into CD8+ CTLs.

Although homeostatic T cell proliferation in lymphopenic hosts may not require CD28 and IL-2, our results strongly argue that CD28 signaling on donor CD8+ T cells is indispensable for functional differentiation of tumor Ag-specific CTLs from precursor cells. Homeostatic T cell proliferation also requires the availability of homeostatic cytokines by lymphodepletion to induce the antitumor immunity observed in this study. Furthermore, using other lines of hard-to-eradicate tumor cells, the administration of IL-2 was found to enhance the antitumor immune response induced by homeostatic proliferation, resulting in suppression of tumor growth. Although IL-2 did not enhance the homeostatic proliferation and differentiation of CTLs in the absence of CD28 signaling, systemic administration of IL-2 significantly augmented lymphopenia-induced antitumor effect in the presence of CD28 signaling. Hence, we demonstrate that the antitumor immune response is dramatically initiated by physiological mechanisms of homeostatic proliferation, unlike Ag-driven expansion, and that terminal differentiation into functional tumor-specific CTLs is directed by CD28 signaling.

C57BL/6 (Ly5.2+) and B6.SJL (Ly5.1+) mice were purchased from Japan SLC. B6 Ly5.1+/Ly5.2+ heterozygous mice were generated by the backcross of B6.SJL to C57BL/6. CD28−/− (B6.CD28−/−) mice were generated as previously described (26) and were kindly provided by Drs. K. Lee and C. June (Naval Medical Research Institute, Bethesda, MD). These mice were backcrossed more than 10 times to the C57BL/6 mice. These animals were maintained in our mouse facility under specific pathogen-free conditions. All experiments were performed in accordance with protocols approved by the Animal Care and Use Committee of the Tokyo University of Science.

MCA102 fibrosarcoma lines (27) and MCA102gp33 cells, which were derived from parental tumor cells by gene transfection using lymphocytic choriomeningitis virus gp33 minigenes, were kindly provided by S. Rosenberg (National Institutes of Health, Bethesda, MD) (28). B16 melanoma cells were obtained from American Type Culture Collection (ATCC).

20G2 (rat IgG2a anti-(4-hydroxy-3-nitrophenyl acethyl) mAb was generated in our laboratory and used as a rat IgG2a control (control Ig). Rat anti-mouse B7-1/CD80 (RM80) and B7-2/CD86 (GL-1) mAbs were generously provided by Dr. K. Okumura (Juntendo University School of Medicine, Tokyo, Japan) and Dr. R. J. Hodes (National Cancer Institute and National Institutes of Health, Bethesda, MD), respectively (29, 30). The B cell hybridomas, anti-CD8 (3–155) rat IgM, anti-CD4 (RL-172) rat IgM, anti-CD44, anti-CD62L (Mel-14) rat IgG, and anti-CD25 (7D4) rat IgM were purchased from ATCC. PE, FITC, or APC-conjugated anti-CD4 (GK1.5), anti-CD8α (53-6.7.2), anti-CD44, and anti-CD62L (Mel-14) were purchased from BD Pharmingen. The B cell hybridomas, anti-I-Ab (Y-3P), anti-H-2Db/Kb (28-8-6), and anti-FcγR (2.4G2), were obtained from ATCC. The mAbs were fluorescein-labeled according to standard techniques. The intracellular fluorescent dye, CFSE, was purchased from Invitrogen: Molecular Probes. An rmIL-2-producing cell line, P3U1 BCMGS-mIL2, was provided by Dr. H. Karasuyama (Tokyo Medical and Dental University, Tokyo, Japan). rmIL-2 was purified from the culture supernatant of P3U1 BCMGS-mIL2 and resuspended in 0.5% C57BL/6 serum/PBS.

Whole splenocytes harvested from wild-type or CD28−/− mice were adoptively transferred by i.v. injection (2 × 107) into syngeneic recipients. Lymphopenia was induced by sublethal irradiation (6.5 Gy) of B6 mice on the same day that the donor cells were injected. For purification of CD8+ T cells, whole splenocytes were added to plates (Iwaki) coated with rabbit Abs specific for mouse IgM (Cappel) and incubated for 40 min at 37°C, followed by negative panning on BD Falcon anti-CD4 mAb-coated plates (BD Biosciences). After negative panning, CD8+ T cells were purified by positive panning on BD Falcon anti-CD8 (3–155) coating plates. For purification of CD44low and CD44high cells, cells were enriched by negative selection with anti-CD44 or anti-CD62L (Mel-14) conjugated with BioMag goat anti-rat IgG (Qiagen). For analysis of cell division, donor cells were labeled with CFSE before adoptive transfer, as described previously (31). Donor cell proliferation was measured by a stepwise reduction of the CFSE intensity using FACS analysis on day 14 or on other days, and distinct peaks represented the number of cell divisions in the lymph node and spleen. Using B6 Ly5.1+/Ly5.2+ mice as a host, CFSE levels and phenotypical changes of donor cells were measured on gated CD4+ or CD8+ donor cells (Ly5.1/Ly5.2+).

Lymphocytes or spleen cell suspensions from donor-cell-injected mice were prepared in FACS medium (PBS containing 0.5% calf serum and 0.1% sodium azide). Cells were incubated first with unlabeled anti-FcγR (2.4G2) to block nonspecific binding and then stained with each Ab. We used a FACSCalibur with CellQuest software (BD Biosciences) for four-color flow cytometric analysis.

Cells were stimulated in vitro with 1 μg/ml gp33 peptide in the presence of 2 μM monensin at 37°C for 6 h (Boehringer Mannheim). After staining for surface receptors, the cells were fixed with PBS containing 4% paraformaldehyde at room temperature for 10 min, and were then incubated in permeabilization buffer (containing 50 mM NaCl, 5 mM EDTA, 0.02% NaN3, and 0.5% Triton X (adjusted to pH 7.5)) at 4°C for 10 min. After incubation in PBS containing 3% BSA for blocking, cells were incubated with rat anti-mouse IFN-γ Ab (XMG1.2) conjugated with Cy5 dye at 4°C for 45 min. Cells were washed with PBS three times and analyzed using FACSCalibur with CellQuest software.

Each experiment was performed in triplicate or greater to confirm the reproducibility of the results, and representative data were collected. The repeated-measures ANOVA–Fisher’s protected least significant difference test was employed. Student’s t test was used to examine the significance of the data when applicable. A difference was considered to be statistically significant for p < 0.05.

Various other groups have reported that T cell expansion in lymphopenic hosts can induce an antitumor immune response (10, 11, 32). In this study, mice with irradiation-induced lymphopenia were used for the induction of homeostatic T cell proliferation to confirm whether T cell homeostasis can enhance suppression of tumor growth. Nonirradiated or sublethally (6.5 Gy) irradiated C57BL/6 (B6) mice under went adoptive transfer of syngeneic splenocytes and were then challenged with intradermal (i.d.) injection of 1 × 106 MCA102gp33 fibrosarcoma cells (Fig. 1,A). As shown in Fig. 1, B and C, control nonirradiated mice developed large tumors with a mean size of 312.4 ± 77.9 mm2 at 36 days after tumor inoculation. Mice treated only with sublethal irradiation developed tumors that were similar in size to those of control mice (327.6 ± 23.2 mm2). Nonirradiated mice administered splenocytes were also not capable of suppressing tumor growth (248.6 ± 105.8 mm2). In agreement with the findings of other groups, irradiated mice that had undergone adoptive transfer of splenocytes not only had no tumor growth but also completely rejected the tumor cells.

FIGURE 1.

Inhibition of MCA102gp33 growth by homeostatic proliferation. A, Experimental designs for induction of homeostatic proliferation and the antitumor effect. Host C57BL/6 mice were sublethally irradiated at 6.5 Gy, and syngeneic wild-type or CD28-deficient splenocytes (1.5–2 × 107) were adoptively transferred to irradiated mice. Mice were challenged with an i.d. injection of 106 MCA102gp33 fibrosarcoma cells in a PBS suspension. Some of these mice were i.p. administered anti-B7-1 and anti-B7-2 mAbs (200 μg each) on days 0, 2, 4, and 6 after treatment. B, The antitumor effect is significantly induced by homeostatic proliferation. Nonirradiated (⋄) or sublethally irradiated (♦) B6 mice were challenged i.d. with MCA102gp33 fibrosarcoma cells as described above. In the treated group, nonirradiated (□) or irradiated (▪) mice were adoptively transferred syngeneic splenocytes (1.5–2 × 107), followed by tumor inoculation. Data represent one of three similar experiments in which n = 4 or 2 mice per experimental group. C, Sublethal irradiation and syngeneic splenocyte infusion inhibit tumor growth significantly. Photographs show representative images of the tumor focus (B).

FIGURE 1.

Inhibition of MCA102gp33 growth by homeostatic proliferation. A, Experimental designs for induction of homeostatic proliferation and the antitumor effect. Host C57BL/6 mice were sublethally irradiated at 6.5 Gy, and syngeneic wild-type or CD28-deficient splenocytes (1.5–2 × 107) were adoptively transferred to irradiated mice. Mice were challenged with an i.d. injection of 106 MCA102gp33 fibrosarcoma cells in a PBS suspension. Some of these mice were i.p. administered anti-B7-1 and anti-B7-2 mAbs (200 μg each) on days 0, 2, 4, and 6 after treatment. B, The antitumor effect is significantly induced by homeostatic proliferation. Nonirradiated (⋄) or sublethally irradiated (♦) B6 mice were challenged i.d. with MCA102gp33 fibrosarcoma cells as described above. In the treated group, nonirradiated (□) or irradiated (▪) mice were adoptively transferred syngeneic splenocytes (1.5–2 × 107), followed by tumor inoculation. Data represent one of three similar experiments in which n = 4 or 2 mice per experimental group. C, Sublethal irradiation and syngeneic splenocyte infusion inhibit tumor growth significantly. Photographs show representative images of the tumor focus (B).

Close modal

It has been reported that effective antitumor immune responses against melanomas achieved by various vaccination protocols require CD8+, CD4+, or both T cell subsets. To evaluate which T cell subset is responsible for inhibition of tumor growth, we examined tumor growth in irradiated recipients transfused with 2 × 106 purified CD8+ T cells (>93% purity), 2 × 106 purified CD4+ T cells (>98% purity), or 4 × 106 mixed CD4+ and CD8+ T cells, followed by i.d. inoculation with 1 × 106 MCA102gp33 cells. Compared with control mice, tumor growth was equally suppressed by the transfer of whole splenocytes, mixed CD4+ and CD8+ T cells, or even CD8+ T cells alone (132 ± 46 vs 33 ± 39 mm2, vs 40 ± 34 mm2, vs 50 ± 28 mm2; p < 0.05,). No antitumor effect was evident with the transfer of CD4+ T cells alone (Fig. 2 A).

FIGURE 2.

Donor CD44low naive CD8+ T cells are indispensable for induction of the antitumor effect induced by homeostatic proliferation. Data represent one of two similar experiments. A, Transfer of CD8+ T cells alone induces potent antitumor effect. Sublethally irradiated mice were adoptively transferred whole splenocytes (1.5–2 × 107) (⋄), purified CD4+ (•) T cells, or CD8+ (▵) T cells (1.5–2 × 106). Some of the irradiated mice were administered mixed CD4+ and CD8+ T cells (each 1.5–2 × 106) (▴). Following treatment, recipient mice were injected with MCA102gp33 fibrosarcoma cells i.d. on day 0. As a control, sublethally irradiated mice were transplanted with tumor cells (♦). B, The growth of MCA102gp33 is significantly suppressed after adoptive transfer of CD44low naive phenotype T cells. Sublethally irradiated host mice were adoptively transferred whole splenocytes (1.5–2 × 107) (⋄), purified CD44low T cells (3–3.5 × 106) (▪), or CD44high T cells (5–7 × 105) (○). Some irradiated mice were adoptively transferred mixed CD44low (3–3.5 × 106) and CD44high T cells (5–7 × 105) (□). After treatment, mice were transplanted with MCA102gp33 cells. C, CD44low naive CD8+ T cells are indispensable for induction of the antitumor effect induced by homeostatic proliferation. Irradiated mice were transferred whole splenocytes (1.5–2 × 107) (⋄) or total (▪) or CD44low naive (□) CD8+ T cells (1.5–2 × 106 each), followed by i.d. inoculation with MCA102gp33. D, Tumor growth is suppressed by adoptive transfer with Ag-exposed “true” memory CD8+ T cells, but not with CD44high memory phenotype CD8+ T cells. Irradiated mice were adoptively transferred total (▵) or CD44high (•) memory phenotype CD8+ T cells, or true memory CD8+ T cells (○) (1.5–2 × 106 each), followed by challenged with MCA102gp33. Ag-exposed CD44high true memory CD8+ T cells were purified from mice subjected to induced homeostatic proliferation and that completely rejected MCA102gp33.

FIGURE 2.

Donor CD44low naive CD8+ T cells are indispensable for induction of the antitumor effect induced by homeostatic proliferation. Data represent one of two similar experiments. A, Transfer of CD8+ T cells alone induces potent antitumor effect. Sublethally irradiated mice were adoptively transferred whole splenocytes (1.5–2 × 107) (⋄), purified CD4+ (•) T cells, or CD8+ (▵) T cells (1.5–2 × 106). Some of the irradiated mice were administered mixed CD4+ and CD8+ T cells (each 1.5–2 × 106) (▴). Following treatment, recipient mice were injected with MCA102gp33 fibrosarcoma cells i.d. on day 0. As a control, sublethally irradiated mice were transplanted with tumor cells (♦). B, The growth of MCA102gp33 is significantly suppressed after adoptive transfer of CD44low naive phenotype T cells. Sublethally irradiated host mice were adoptively transferred whole splenocytes (1.5–2 × 107) (⋄), purified CD44low T cells (3–3.5 × 106) (▪), or CD44high T cells (5–7 × 105) (○). Some irradiated mice were adoptively transferred mixed CD44low (3–3.5 × 106) and CD44high T cells (5–7 × 105) (□). After treatment, mice were transplanted with MCA102gp33 cells. C, CD44low naive CD8+ T cells are indispensable for induction of the antitumor effect induced by homeostatic proliferation. Irradiated mice were transferred whole splenocytes (1.5–2 × 107) (⋄) or total (▪) or CD44low naive (□) CD8+ T cells (1.5–2 × 106 each), followed by i.d. inoculation with MCA102gp33. D, Tumor growth is suppressed by adoptive transfer with Ag-exposed “true” memory CD8+ T cells, but not with CD44high memory phenotype CD8+ T cells. Irradiated mice were adoptively transferred total (▵) or CD44high (•) memory phenotype CD8+ T cells, or true memory CD8+ T cells (○) (1.5–2 × 106 each), followed by challenged with MCA102gp33. Ag-exposed CD44high true memory CD8+ T cells were purified from mice subjected to induced homeostatic proliferation and that completely rejected MCA102gp33.

Close modal

To examine which population of donor T cells (naive or memory) in the phenotype could induce the antitumor immune response in the lymphopenic host, we purified CD44low naive phenotype T cells or CD44high memory phenotype T cells and transferred each purified population into irradiated mice, followed by i.d. transplantation of MCA102gp33 cells (Fig. 2,B). Surprisingly, transfusion of CD44low naive T cells suppressed tumor growth significantly compared with the control irradiated mice, whereas CD44high memory phenotype T cells failed to induce the antitumor effect (199 ± 32 vs 72 ± 62 mm2 (p < 0.05), vs 248 ± 23 mm2; Fig. 2,B). To confirm the data shown in Fig. 2, A and B, we purified CD44low naive CD8+ T cells, then transfused them into a lymphopenic host, and found that these cells could significantly suppress tumor growth (271.3 ± 118.8 vs 98.7 ± 24.6 mm2; Fig. 2,C). As shown in Fig. 2 D, large numbers of CD44high memory phenotype CD8+ T cells failed to induce antitumor effect. Thus, antitumor immune response induced by homeostatic proliferation is largely aided by the transferred CD44low naive CD8+ T cells.

For CD8+ T cells to differentiate into CTLs and carry out their cytotoxic functions, they require the costimulatory signal via CD28 and IL-2. Homeostatic proliferation of T cells, however, does not require CD28 signaling and the presence of IL-2. Furthermore, it is not accompanied by differentiation to effector T cells. Thus, to resolve discrepancy in the role of CD28, we used CD28-deficient donor T cells or B7 blockade through the combined administration of anti-B7-1 and anti-B7-2 mAb. When CD28−/− cells were transferred to irradiated C57BL/6 hosts challenged with MCA102gp33 cells, CD28−/− splenocytes did not induce suppression of tumor growth (218.0 ± 38.4 vs 194.7 ± 17.4 mm2; Fig. 3,A), whereas transfusion of wild-type splenocytes inhibited tumor growth significantly (218.0 ± 38.4 vs 16.6 ± 28.7 mm2 (p < 0.05); Fig. 3,A). To confirm the essential role of CD28 signaling, we transferred wild-type splenocytes into lymphopenic hosts and/or blocked B7-mediated costimulation with the combined administration of anti-B7-1 and anti-B7-2 mAb. B7 blockage completely abrogated the antitumor immune response induced by homeostatic proliferation (218.0 ± 38.4 vs 248.6 ± 12.3 mm2; Fig. 3 B). Both approaches show that CD28 signals are essential for the homeostatic proliferation-induced antitumor immune response.

FIGURE 3.

B7-CD28 interaction is essential for induction of the antitumor immune response under lymphopenic conditions. Data represent one of three similar experiments in which n = 3 mice per experimental group. A, Adoptive transfer of CD28-deficient splenocytes failed to suppress tumor growth. Wild-type (WT) (▪) or CD28-deficient (CD28−/−) (□) splenocytes were transferred (2 × 107 each) into irradiated hosts and challenged i.d. with MCA102gp33. As a control, sublethally irradiated mice were transplanted with tumor cells (♦). B, The homeostatic proliferation-induced antitumor effect disappears by costimulation blockage with anti-B7-1 and anti-B7-2 mAbs. Wild-type splenocytes (2 × 107) were transferred into irradiated hosts, followed by i.d. inoculation of MCA102gp33. Some animals were treated with 200 μg control Ig (▪) or anti-B7-1 and anti-B7-2 mAbs (□) each on days 0, 2, 4, and 6. As a control, sublethally irradiated mice were transplanted with tumor cells (♦).

FIGURE 3.

B7-CD28 interaction is essential for induction of the antitumor immune response under lymphopenic conditions. Data represent one of three similar experiments in which n = 3 mice per experimental group. A, Adoptive transfer of CD28-deficient splenocytes failed to suppress tumor growth. Wild-type (WT) (▪) or CD28-deficient (CD28−/−) (□) splenocytes were transferred (2 × 107 each) into irradiated hosts and challenged i.d. with MCA102gp33. As a control, sublethally irradiated mice were transplanted with tumor cells (♦). B, The homeostatic proliferation-induced antitumor effect disappears by costimulation blockage with anti-B7-1 and anti-B7-2 mAbs. Wild-type splenocytes (2 × 107) were transferred into irradiated hosts, followed by i.d. inoculation of MCA102gp33. Some animals were treated with 200 μg control Ig (▪) or anti-B7-1 and anti-B7-2 mAbs (□) each on days 0, 2, 4, and 6. As a control, sublethally irradiated mice were transplanted with tumor cells (♦).

Close modal

Our results strongly indicate the necessity for CD28 signaling for the lymphopenia-induced antitumor immune response; however, CD28 signaling is not required for homeostatic proliferation of naive T cells (19, 20). Therefore, we confirmed the expansion and phenotypical changes of donor T cells during homeostatic proliferation in the presence or absence of CD28 signaling. Purified T cells recovered from wild-type (Ly5.1) or CD28−/− (Ly5.1) mice were labeled with CFSE and adoptively transferred into irradiated B6 heterozygous mice (Ly5.1+/Ly5.2+). Some mice transplanted with wild-type T cells were administered a combination of anti-B7-1 and anti-B7-2 mAb on days 0, 2, 4, and 6. Fourteen days after transfer, spleen and lymph node cells were analyzed for division and for phenotypical changes in their population of donor CD8+ T cells (Ly5.1/Ly5.2+ CD8+ cells). In tumor-bearing hosts, donor cells could divide in the absence of CD28 signaling; however, the highly proliferating CFSE population was significantly diminished as compared with normal CD28+ T cells (Fig. 4,A). The reduction in this highly dividing population due to defective CD28 signaling was also observed in tumor-free wild-type and RAG−/− mice. Thus, the CFSE population is dependent on CD28 signals. To analyze the phenotypical changes in donor CD8+ T cells during homeostatic proliferation, CFSE-labeled donor cells were recovered and stained with mAb specific for CD25, CD44, CD62L, and CD122. In agreement with other investigators, proliferating donor CD8+ T cells showed an up-regulation of CD44 and CD122, but not CD25, during homeostatic proliferation in the presence or absence of CD28 signaling (Fig. 4 B). The expression pattern of these cell-surface molecules is independent of CD28 signaling.

FIGURE 4.

Role of CD28 signaling in the expansion and differentiation of donor CD8+ T cells during homeostatic proliferation. Normal or CD28-deficient donor splenocytes (Ly5.1/Ly5.2+) (2 × 107) were labeled with CFSE and adoptively transferred to irradiated Ly5.1+/Ly5.2+ congenic mice or C57BL/6 RAG knockout mice. Some mice were challenged with MCA102gp33. A costimulation group was treated with anti-B7-1/anti-B7-2 mAbs at 200 μg each on days 0, 2, 4, and 6 after treatment. Fourteen days after treatment, the expansion and phenotypical changes of donor CD8+ T cells (gated Ly5.1CD8+ cells) were analyzed. Data represent one of three similar experiments in which n = 1 or 2 mice per experimental group. A, The highly proliferating CFSE population is significantly diminished in the absence of CD28 signaling. B, Phenotypical changes in donor CD8+ T cells during homeostatic proliferation. C, Tumor-associated, Ag-specific IFN-γ-producing CD8+ T cells disappear in the absence of CD28 signal. Total splenocytes recovered from each mouse were restimulated with gp33 peptides in vitro, and intracellular staining of IFN-γ was performed.

FIGURE 4.

Role of CD28 signaling in the expansion and differentiation of donor CD8+ T cells during homeostatic proliferation. Normal or CD28-deficient donor splenocytes (Ly5.1/Ly5.2+) (2 × 107) were labeled with CFSE and adoptively transferred to irradiated Ly5.1+/Ly5.2+ congenic mice or C57BL/6 RAG knockout mice. Some mice were challenged with MCA102gp33. A costimulation group was treated with anti-B7-1/anti-B7-2 mAbs at 200 μg each on days 0, 2, 4, and 6 after treatment. Fourteen days after treatment, the expansion and phenotypical changes of donor CD8+ T cells (gated Ly5.1CD8+ cells) were analyzed. Data represent one of three similar experiments in which n = 1 or 2 mice per experimental group. A, The highly proliferating CFSE population is significantly diminished in the absence of CD28 signaling. B, Phenotypical changes in donor CD8+ T cells during homeostatic proliferation. C, Tumor-associated, Ag-specific IFN-γ-producing CD8+ T cells disappear in the absence of CD28 signal. Total splenocytes recovered from each mouse were restimulated with gp33 peptides in vitro, and intracellular staining of IFN-γ was performed.

Close modal

Note that although the largest population of naive donor CD8+ T cells undergoing homeostatic proliferation acquired CD44highCD62Lhigh central memory phenotype (TCMP), a small fraction of CFSE cells showed down-regulated CD62L and acquired CD44highCD62Llow effector memory phenotype (TEMP). Interestingly, these CD44highCD62Llow CD8+ donor T cells were found to belong to the highly dividing population, which lacked the defect in CD28 signaling (Fig. 4,B, bold squares). To determine whether tumor Ag-specific CTLs included CD44highCD62Llow CD8+ TEMP population, we analyzed the IFN-γ production of donor T cells stimulated with gp33 peptide using intracellular cytokine staining. Fourteen days after cell transfer, splenocytes were stimulated with gp33 peptide for 6 h in the presence of monensin, and stained for intracellular IFN-γ, as described in Materials and Methods. As expected, the IFN-γ-producing cells had accumulated in the highly dividing population and were not detected in the absence of CD28 signaling (Fig. 4 C). These data indicate that CD28 signaling is essential for the differentiation of tumor Ag-specific CTLs during homeostatic proliferation.

CD28 signaling on T cells plays a key role in the up-regulation of IL-2 gene transcription, causing the differentiation of precursor cells into functional CTLs in response to foreign Ag. Therefore, we examined whether systemic administration of IL-2 could compensate for the accumulation of CD44highCD62Llow CD8+ TEMP in the absence of CD28 signaling. Purified T cells from wild-type or CD28-deficient B6 mice were transferred to irradiated mice, and IL-2 was administered systemically for 5 days at a concentration of 10,000 U/day. Interestingly, the antitumor effect induced by homeostatic proliferation could not be achieved with IL-2 administration in the absence of CD28 signaling (Fig. 5,A). Sequential administration of IL-2 could neither overcome the lack of CD44highCD62Llow CD8+ TEMP nor restore the population of IFN-γ-producing cells that respond to gp33 peptide in donor cells (Fig. 5 B). These data show that a loss of IL-2 production is not critical for the failure of the antitumor immune response induced by homeostatic proliferation and generation of CD8+ TEMP cells.

FIGURE 5.

IL-2 administration fails to overcome the defect in effector memory phenotype CD8+ T cells in the absence of CD28 signaling. A, Administration of IL-2 fails to induce antitumor effect with adoptive transfer of CD28-deficient splenocytes. Irradiated hosts were administered CD28-deficient splenocytes (2 × 107) and challenged i.d. with MCA102gp33. Some groups of mice treated with CD28−/− splenocytes were i.p. administered buffer (PBS containing 0.5% B6 serum) (□) or 10,000 U/day rmIL-2 at 0–5 (○) or 10–15 (•) days. As a control, irradiated mice were transplanted with MCA102gp33 (♦). Also shown is a group in which wild-type whole splenocytes (2 × 107) were adoptively transferred (▪). B, IL-2 administration fails to restore the highly proliferating and IFN-γ-producing population of donor CD8+ T cells in the absence of CD28 signaling. The experimental procedure is similar to that in Fig. 4. At 0–5 days after treatment, some mice were administered buffer or 10,000 U/day rmIL-2, and 14 days after treatment, the expansion and IFN-γ production of donor (Ly5.1) CD8+ T cells were analyzed by FACS.

FIGURE 5.

IL-2 administration fails to overcome the defect in effector memory phenotype CD8+ T cells in the absence of CD28 signaling. A, Administration of IL-2 fails to induce antitumor effect with adoptive transfer of CD28-deficient splenocytes. Irradiated hosts were administered CD28-deficient splenocytes (2 × 107) and challenged i.d. with MCA102gp33. Some groups of mice treated with CD28−/− splenocytes were i.p. administered buffer (PBS containing 0.5% B6 serum) (□) or 10,000 U/day rmIL-2 at 0–5 (○) or 10–15 (•) days. As a control, irradiated mice were transplanted with MCA102gp33 (♦). Also shown is a group in which wild-type whole splenocytes (2 × 107) were adoptively transferred (▪). B, IL-2 administration fails to restore the highly proliferating and IFN-γ-producing population of donor CD8+ T cells in the absence of CD28 signaling. The experimental procedure is similar to that in Fig. 4. At 0–5 days after treatment, some mice were administered buffer or 10,000 U/day rmIL-2, and 14 days after treatment, the expansion and IFN-γ production of donor (Ly5.1) CD8+ T cells were analyzed by FACS.

Close modal

One recent clinical trial demonstrated that, after lymphodepletion chemotherapy, adoptive transfer of autologous activated CTLs, which were generated from tumor-infiltrating T lymphocytes (33) by stimulation with IL-2 and tumor Ag peptides in vitro, caused the regression of large, vascularized tumors in patients with refractory metastatic melanoma. Furthermore, to achieve a sufficient antitumor response, the adoptive transfer of activated CTLs and IL-2 administration were combined in this trial (3). Based on this report, we examined whether administration of a low dose of IL-2 could enhance the survival and activation of CTLs in the tumor focus, which was triggered by the concomitant stimulation of homeostatic expansion and CD28-B7 interaction in vivo. Sublethally irradiated B6 mice underwent adoptive transfer of 2 × 107 syngeneic splenocytes and, on the same day, were challenged with i.d. injection of 1 × 106 B16 melanoma cells. Ten days after tumor inoculation, mice were injected with varying doses of IL-2 approximately every 12 h for 6 or 9 days. The dose of IL-2 needed to induce lymphokine-activated killer activity in vivo was reported previously (34). Using B16 melanoma, our results indicated that the minimum dose of IL-2 for enhancing the homeostatic proliferation-induced antitumor effect was 5000 U/day (311.1 ± 116.3 vs 77.9 ± 8.4 mm2 (p < 0.05); Fig. 6,A), and that the administration of 10,000 U/day of IL-2 could induce complete rejection (394.3 ± 100.8 vs 21.4 ± 26.4 mm2 (p < 0.05); Fig. 6 B).

FIGURE 6.

Systemic IL-2 administration enhances the antitumor effect induced by homeostatic proliferation. The growth of B16 melanoma cells is significantly suppressed under conditions of homeostatic proliferation and IL-2 administration. Mice were sublethally irradiated at 6.5 Gy, and syngeneic splenocytes (2 × 107) were adoptively transferred to irradiated mice. Mice were challenged with i.d. injection of 106 B16 melanoma cells. Some of these mice were i.p. administered buffer (▴) (A) 5000 U/day at day 10–15 or (B) 10,000 U/day rmIL-2 (▵) at day 10–18 after treatment. As a control, irradiated mice were transplanted with B16 (♦). Data represent one of two similar experiments in which n = 3 or 4 mice per experimental group.

FIGURE 6.

Systemic IL-2 administration enhances the antitumor effect induced by homeostatic proliferation. The growth of B16 melanoma cells is significantly suppressed under conditions of homeostatic proliferation and IL-2 administration. Mice were sublethally irradiated at 6.5 Gy, and syngeneic splenocytes (2 × 107) were adoptively transferred to irradiated mice. Mice were challenged with i.d. injection of 106 B16 melanoma cells. Some of these mice were i.p. administered buffer (▴) (A) 5000 U/day at day 10–15 or (B) 10,000 U/day rmIL-2 (▵) at day 10–18 after treatment. As a control, irradiated mice were transplanted with B16 (♦). Data represent one of two similar experiments in which n = 3 or 4 mice per experimental group.

Close modal

Recently, it was demonstrated that adoptively transferred antitumor immune T cells mediate the rejection of large vascularized tumors in mice under appropriate conditions of host immune suppression and Ag stimulation (32, 35). In addition to augmentation of the expansion and survival of activated immune cells in vivo, our observations and those of other groups indicate that the multiple availability of homeostatic cytokines leads to the differentiation of naive T cells into functional effector T cells in the lymphopenic host, and that this lymphopenia-induced immune response can regress tumor growth significantly (11) (Fig. 1). These findings suggest that the elimination of “cellular sinks” for homeostatic cytokines may enable naive T cells to be activated and to differentiate into functional CTLs accompanied by the enhanced reactivity of naive T cells to weak Ags, such as TAAs (36, 37). Because naive T cells have the potential to differentiate into functional CTLs in the lymphopenic host, we investigated the mechanisms that could initiate this differentiation during homeostatic proliferation.

It is well known that suboptimal TCR signaling and costimulation by B7-CD28 interaction are essential for the induction of functional effector CTLs during Ag stimulation (23, 38, 39). Prevention of costimulatory signaling, especially by CD28-B7 interaction, results in the induction of anergy or activation-induced cell death in both CD4+ and CD8+ immune responses (22, 25, 40, 41). In contrast to Ag-driven expansion and differentiation, there has been little evidence suggesting any requirement for costimulatory signaling for the expansion and phenotypical changes of naive T cells in various models of lymphopenic host (20, 21). Therefore, we first examined whether CD28-B7 interaction is required for the induction of antitumor immune response, as induced by homeostatic proliferation. Our results strongly indicated that CD28 signaling is indispensable for the induction of antitumor immune response in a lymphopenic host (Fig. 3, A and B).

On confirming which population of donor cells was involved, our results demonstrated that adoptive transfer of only donor naive CD8+ T cells could induce a sufficiently strong antitumor effect in a lymphodepleted host (Fig. 2,A). Because gp33 peptides are only antigenic to CD8+ T cells bound to H2-Db, one might argue that CD4+ T cells may not have been able to participate in the immune response in these experiments. Considering that TAAs are recognized by CD4+ T cells, we performed the same experiment using B16 melanoma cell lines. Similar to the previous results, the infusion of only CD8+ T cells into the lymphopenic host clearly induced antitumor effect; however, “help” from host CD4+ T cells could not be excluded (data not shown). To confirm the differentiation of naive T cells into functional effector T cells during homeostatic proliferation, purified naive CD44lowCD8+ T cells were adoptively transferred to lymphopenic mice, and MCA102gp33 was inoculated into the host mice. As expected, only adoptive transfer of naive CD8+ T cells could induce a potent antitumor effect (Fig. 2,C). Although the transfer of memory phenotype CD44high T cells, which were purified from tumor-free mice, did not induce antitumor immune response, Ag-driven CD44high “true” memory T cells recovered from tumor-rejected mice caused significant regression of tumor growth under lymphopenic conditions (Fig. 2 D). Therefore, we conclude that inherent mechanisms of homeostatic proliferation can initiate the differentiation of naive CD8+ T cells into functional effector CTLs and also enhance the expansion and secondary response of Ag-generated true memory CD8+ T cells. Using Friend leukemia cells, another group has shown that homeostatic proliferation of donor CD4+ T cells was essential for tumor rejection (42). This discrepancy in the requirement for donor T cells may depend on tumor immunogenicity. A recent study has shown that homeostatic proliferation-induced memory phenotype CD8+ T cells controlled bacterial infection as effectively as did true memory CD8+ T cells; however, their protective capacity required the presence of CD4+ T cells during homeostatic proliferation (43). It was also reported that adoptive transfer of naive 2C TCR Tg CD8+ T cells to a RAG−/− host led to a strong antitumor immune response, followed by regression of the tumor growth (35). In vitro experiments indicated that the Ag-specific signal mediated by the corresponding TCR and an additional signal provided by CD28-B7 interaction was sufficient for CD8+ T cell activation in the presence of exogenous IL-2, and that CD4+ cells were not required for the generation of the CTL response (44). These observations suggest that CD4+ T cells may not be involved in the CTL response during homeostatic proliferation, whereas they would be important for the maintenance of non-Ag-exposed memory CD8+ T cells generated under lymphopenic conditions (43).

In our analysis of donor naive CD8+ T cells during homeostatic proliferation, we found that the rapidly proliferating population, which had CD44highCD62LCD25 effector memory phenotype (CD8+ TEMP), disappeared in the absence of CD28 signaling (Fig. 4, A and B). This highly dividing population of donor CD8+ T cells was capable of producing IFN-γ in response to gp33 peptide stimulation in vitro, and the defect in CD28 signaling completely prevented the appearance of Ag-specific IFN-γ-producing cells (Fig. 4 C). Thus, we conclude that CD28 signaling is essential for the differentiation of naive CD8+ T cells into functional CTLs with accompanying homeostatic proliferation. Other groups have suggested the possibility that the only effect of costimulation is merely an acceleration of homeostatic proliferation of CD4+ T cells (20, 21). CD4+ T cells within the highly diving fraction can become independent on costimulation in their ability to produce IL-2 and IFN-γ following Ag or PMA/ionomycin stimulation (21). To determine whether the CD28 signal is important for IFN-γ production and homeostatic proliferation of donor CD8+ T cells in a tumor-free host, we adoptively transferred wild-type polyclonal CD8+ T cells to a RAG−/− host. Similar to findings using a tumor-bearing mouse as a host, our preliminary data indicated that the lack of CD28 signaling resulted in the reduction of CD44highCD62LCD8+ TEMP cells, and in the accumulation of CD44highCD62L+ central memory phenotype CD8+ T cells (CD8+ TCMP). The subset of the most highly dividing fraction was considerably smaller than that of normal cells. IFN-γ production in dividing CD8+ T cells stimulated with PMA/ionomycin was also decreased in the absence of CD28 signaling (data not shown). CD8+ TCM and TEM cells have similar potential for IFN-γ production, however, TEM cells have been shown to develop effector functions more rapidly than TCM cells (45). Therefore, it is suggested that CD28 signaling is essential for differentiation into fully functional memory T cells in the presence and absence of Ag stimulation during homeostatic proliferation. Min et al. also reported that two distinct types of proliferation occurred upon transfer of naive T cells into lymphopenic host; one is the slow “homeostatic” proliferation that is fully dependent on IL-7, whereas the rapid “spontaneous” proliferation dose not require IL-7 (46). One would argue that the rapid proliferation of CD8+ TEMP population regulated by CD28 signaling may be considered as spontaneous proliferation (46). However, using IL-7 and IL-15 knockout mice as hosts, Gattinoni et al. demonstrated that both IL-15 and IL-7 are critical for sustaining the proliferation and function of transferred CD8+ T cells to induce effective antitumor immune response (32). Therefore, we conclude that rapid proliferation accompanied by differentiation of naive CD8+ T cells into TEMP cells upon cognate recognition of tumor-associated Ags is distinct from spontaneous proliferation of naive T cells in lymphopenic condition. Our results suggested that the physiological mechanisms involved in homeostatic proliferation, which is independent of CD28 signaling, induces the differentiation of naive CD8+ T cells into TCMP cells. During homeostatic proliferation, CD28 signaling may regulate the differentiation of CD8+ TCMP cells into functional CD8+ TEMP cells. However, the ambient concentration of homeostatic cytokines (i.e., IL-7 and IL-15) is known to be elevated in lymphopenic hosts and might be sufficient enough to induce tumor-associated Ag-specific expansion and differentiation of naive CD8+ T cells into CTL directly (37, 47, 48). Our experiments do not rule out this possibility, although CD62L down-regulation of naive CD8+ T cells occurs within a day and stays low upon Ag-specific normal responses (data not shown). Furthermore, our preliminary data indicated that Ag-driven CD8+ T cells proliferate much more rapidly than does preferential proliferation of CD8+ T cells that was observed in lymphopenic hosts.

In the course of homeostatic proliferation, CD44 expression was up-regulated and CD62L expression remained at a high level (Fig. 4,B). Only the rapidly dividing cell population that disappeared without CD28 signal down-regulated CD62L and was capable of producing IFN-γ (Fig. 4,C). This is supported by the unexpected evidence that regression of the antitumor effect caused by the lack of CD28 signaling was not restored by administration of a low dose of IL-2 (Fig. 5,A); that is, for the first few divisions of naive CD8+ T cells, IL-2 treatment did not induce any phenotypical or functional changes in donor CD8+ T cells. The rapidly proliferating population with a potent capacity for IFN-γ production on restimulation with Ag or PMA/ionomycin in vitro was not restored with IL-2 administration (Fig. 5,B). These findings are consistent with observations that proliferation and functional differentiation of naive T cells in the lymphopenic host occur in an IL-2-independent manner. Previous studies demonstrate that when immune T cells, which had been activated by antigenic peptides and IL-2 in vitro, were transferred to mice who were then administrated IL-2, a much stronger antitumor effect occurred (5, 7). Therefore, we examined whether systemic IL-2 administration could enhance the antitumor immune response induced by homeostatic proliferation and CD28 signaling, even though it is not important for the initiation of proliferation and differentiation of naive T cells in the lymphopenic host. Systemic IL-2 administration between 10 and 15 or 18 days after adoptive transfer resulted in the augmentation of the antitumor immune response induced by homeostatic proliferation in the presence of CD28 signaling (Fig. 6, A and B). Therefore, it is likely that, during this process, naive CD8+ T cells differentiate into CD8+ TEM cells either directly or indirectly in secondary lymph nodes presenting TAA, followed by their infiltration into the tumor focus. On reaching tumor focus, TEM CD8+ T cells encounter TAA peptide presented by MHC molecules on the tumor cells, and rapidly differentiate into effector T cells, in which CD25 expression is up-regulated. The survival and expansion of effector CTL in tumor focus may be enhanced by systemic IL-2, and finally the antitumor effect is significantly augmented.

The findings of the present study further indicate that naive CD8+ T cells have a strong ability to induce effective antitumor immune response to TAA under lymphopenic conditions, and that CD28 signaling is essential for the conclusive differentiation of naive CD8+ T cells into effector CTLs during homeostatic proliferation. We also showed that systemic IL-2 administration at appropriate times can greatly augment the antitumor response induced by homeostatic proliferation. It is well known that anticancer chemotherapy with nonmyeloablative drugs induces lymphopenia, as does total body irradiation. Taken together, these facts provide a strong rationale for establishing a novel treatment combining immunotherapy and chemotherapy to induce an effective antitumor immune response in cancer patients. With this approach, naive PBLs used for adoptive transfer are collected and maintained before the initiation of anticancer chemotherapy. A deeper understanding of the mechanisms underlying the enhanced induction of antitumor immune response induced by homeostatic proliferation may allow us to more precisely achieve our goals of enhanced immunity with the combined use of costimulation signaling and administration of common γ-chain cytokines.

We thank Dr. H. Karasuyama for providing the BCMGS-mIL2, Dr. Richard Hodes for the GL-1 Ab, and Dr. Ko Okumura for the RM80 Ab. This work was aided by Sakiko Kobayashi for preparation of Abs and cytokines, and a member of Science Service, Inc. for care of experimental animals.

The authors have no financial conflicts of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

2

Abbreviations used in this paper: TAA, tumor-associated Ag; HP, homeostatic proliferation; i.d., intradermal(ly); TCM, central memory T; TCMP, central memory phenotype T; TEM, effector memory T; TEMP, effector memory phenotype T.

1
Khong, H. T., N. P. Restifo.
2002
. Natural selection of tumor variants in the generation of “tumor escape” phenotypes.
Nat. Immunol.
3
:
999
-1005.
2
Overwijk, W. W., M. R. Theoret, S. E. Finkelstein, D. R. Surman, L. A. de Jong, F. A. Vyth-Dreese, T. A. Dellemijn, P. A. Antony, P. J. Spiess, D. C. Palmer, et al
2003
. Tumor regression and autoimmunity after reversal of a functionally tolerant state of self-reactive CD8+ T cells.
J. Exp. Med.
198
:
569
-580.
3
Rosenberg, S. A., M. E. Dudley.
2004
. Cancer regression in patients with metastatic melanoma after the transfer of autologous antitumor lymphocytes.
Proc. Natl. Acad. Sci. USA
101
: (Suppl. 2):
14639
-14645.
4
Rosenberg, S. A., J. C. Yang, N. P. Restifo.
2004
. Cancer immunotherapy: moving beyond current vaccines.
Nat. Med.
10
:
909
-915.
5
Dudley, M. E., J. R. Wunderlich, P. F. Robbins, J. C. Yang, P. Hwu, D. J. Schwartzentruber, S. L. Topalian, R. Sherry, N. P. Restifo, A. M. Hubicki, et al
2002
. Cancer regression and autoimmunity in patients after clonal repopulation with antitumor lymphocytes.
Science
298
:
850
-854.
6
Toes, R. E., R. Offringa, R. J. Blom, C. J. Melief, W. M. Kast.
1996
. Peptide vaccination can lead to enhanced tumor growth through specific T-cell tolerance induction.
Proc. Natl. Acad. Sci. USA
93
:
7855
-7860.
7
Rosenberg, S. A..
1988
. Cancer therapy with interleukin-2: immunologic manipulations can mediate the regression of cancer in humans.
J. Clin. Oncol.
6
:
403
-406.
8
Dudley, M. E., J. Wunderlich, M. I. Nishimura, D. Yu, J. C. Yang, S. L. Topalian, D. J. Schwartzentruber, P. Hwu, F. M. Marincola, R. Sherry, et al
2001
. Adoptive transfer of cloned melanoma-reactive T lymphocytes for the treatment of patients with metastatic melanoma.
J. Immunother.
24
:
363
-373.
9
Surh, C. D., J. Sprent.
2005
. Regulation of mature T cell homeostasis.
Semin. Immunol.
17
:
183
-191.
10
Dummer, W., A. G. Niethammer, R. Baccala, B. R. Lawson, N. Wagner, R. A. Reisfeld, A. N. Theofilopoulos.
2002
. T cell homeostatic proliferation elicits effective antitumor autoimmunity.
J. Clin. Invest.
110
:
185
-192.
11
Wang, L. X., R. Li, G. Yang, M. Lim, A. O’Hara, Y. Chu, B. A. Fox, N. P. Restifo, W. J. Urba, H. M. Hu.
2005
. Interleukin-7-dependent expansion and persistence of melanoma-specific T cells in lymphodepleted mice lead to tumor regression and editing.
Cancer Res.
65
:
10569
-10577.
12
Tan, J. T., B. Ernst, W. C. Kieper, E. LeRoy, J. Sprent, C. D. Surh.
2002
. Interleukin (IL)-15 and IL-7 jointly regulate homeostatic proliferation of memory phenotype CD8+ cells but are not required for memory phenotype CD4+ cells.
J. Exp. Med.
195
:
1523
-1532.
13
Ku, C. C., M. Murakami, A. Sakamoto, J. Kappler, P. Marrack.
2000
. Control of homeostasis of CD8+ memory T cells by opposing cytokines.
Science
288
:
675
-678.
14
Tan, J. T., E. Dudl, E. LeRoy, R. Murray, J. Sprent, K. I. Weinberg, C. D. Surh.
2001
. IL-7 is critical for homeostatic proliferation and survival of naive T cells.
Proc. Natl. Acad. Sci. USA
98
:
8732
-8737.
15
Seddon, B., P. Tomlinson, R. Zamoyska.
2003
. Interleukin 7 and T cell receptor signals regulate homeostasis of CD4 memory cells.
Nat. Immunol.
4
:
680
-686.
16
Cho, B. K., V. P. Rao, Q. Ge, H. N. Eisen, J. Chen.
2000
. Homeostasis-stimulated proliferation drives naive T cells to differentiate directly into memory T cells.
J. Exp. Med.
192
:
549
-565.
17
Goldrath, A. W., L. Y. Bogatzki, M. J. Bevan.
2000
. Naive T cells transiently acquire a memory-like phenotype during homeostasis-driven proliferation.
J. Exp. Med.
192
:
557
-564.
18
Murali-Krishna, K., R. Ahmed.
2000
. Cutting edge: Naive T cells masquerading as memory cells.
J. Immunol.
165
:
1733
-1737.
19
Gudmundsdottir, H., L. A. Turka.
2001
. A closer look at homeostatic proliferation of CD4+ T cells: costimulatory requirements and role in memory formation.
J. Immunol.
167
:
3699
-3707.
20
Prlic, M., B. R. Blazar, A. Khoruts, T. Zell, S. C. Jameson.
2001
. Homeostatic expansion occurs independently of costimulatory signals.
J. Immunol.
167
:
5664
-5668.
21
Hagen, K. A., C. T. Moses, E. F. Drasler, K. M. Podetz-Pedersen, S. C. Jameson, A. Khoruts.
2004
. A role for CD28 in lymphopenia-induced proliferation of CD4 T cells.
J. Immunol.
173
:
3909
-3915.
22
Hellstrom, K. E., I. Hellstrom, L. Chen.
1995
. Can co-stimulated tumor immunity be therapeutically efficacious?.
Immunol. Rev.
145
:
123
-145.
23
Chen, L., P. S. Linsley, K. E. Hellstrom.
1993
. Costimulation of T cells for tumor immunity.
Immunol. Today
14
:
483
-486.
24
Watts, T. H., M. A. DeBenedette.
1999
. T cell co-stimulatory molecules other than CD28.
Curr. Opin. Immunol.
11
:
286
-293.
25
Lenschow, D. J., T. L. Walunas, J. A. Bluestone.
1996
. CD28/B7 system of T cell costimulation.
Annu. Rev. Immunol.
14
:
233
-258.
26
Shahinian, A., K. Pfeffer, K. P. Lee, T. M. Kundig, K. Kishihara, A. Wakeham, K. Kawai, P. S. Ohashi, C. B. Thompson, T. W. Mak.
1993
. Differential T cell costimulatory requirements in CD28-deficient mice.
Science
261
:
609
-612.
27
Shu, S. Y., T. Chou, K. Sakai.
1989
. Lymphocytes generated by in vivo priming and in vitro sensitization demonstrate therapeutic efficacy against a murine tumor that lacks apparent immunogenicity.
J. Immunol.
143
:
740
-748.
28
Prevost-Blondel, A., C. Zimmermann, C. Stemmer, P. Kulmburg, F. M. Rosenthal, H. Pircher.
1998
. Tumor-infiltrating lymphocytes exhibiting high ex vivo cytolytic activity fail to prevent murine melanoma tumor growth in vivo.
J. Immunol.
161
:
2187
-2194.
29
Nakajima, A., M. Azuma, S. Kodera, S. Nuriya, A. Terashi, M. Abe, S. Hirose, T. Shirai, H. Yagita, K. Okumura.
1995
. Preferential dependence of autoantibody production in murine lupus on CD86 costimulatory molecule.
Eur. J. Immunol.
25
:
3060
-3069.
30
Hathcock, K. S., G. Laszlo, H. B. Dickler, J. Bradshaw, P. Linsley, R. J. Hodes.
1993
. Identification of an alternative CTLA-4 ligand costimulatory for T cell activation.
Science
262
:
905
-907.
31
Lyons, A. B., C. R. Parish.
1994
. Determination of lymphocyte division by flow cytometry.
J. Immunol. Methods
171
:
131
-137.
32
Gattinoni, L., S. E. Finkelstein, C. A. Klebanoff, P. A. Antony, D. C. Palmer, P. J. Spiess, L. N. Hwang, Z. Yu, C. Wrzesinski, D. M. Heimann, et al
2005
. Removal of homeostatic cytokine sinks by lymphodepletion enhances the efficacy of adoptively transferred tumor-specific CD8+ T cells.
J. Exp. Med.
202
:
907
-912.
33
Bracci, L., F. Moschella, P. Sestili, V. La Sorsa, M. Valentini, I. Canini, S. Baccarini, S. Maccari, C. Ramoni, F. Belardelli, E. Proietti.
2007
. Cyclophosphamide enhances the antitumor efficacy of adoptively transferred immune cells through the induction of cytokine expression, B-cell and T-cell homeostatic proliferation, and specific tumor infiltration.
Clin. Cancer Res.
13
:
644
-653.
34
Rosenberg, S. A., J. J. Mule, P. J. Spiess, C. M. Reichert, S. L. Schwarz.
1985
. Regression of established pulmonary metastases and subcutaneous tumor mediated by the systemic administration of high-dose recombinant interleukin 2.
J. Exp. Med.
161
:
1169
-1188.
35
Brown, I. E., C. Blank, J. Kline, A. K. Kacha, T. F. Gajewski.
2006
. Homeostatic proliferation as an isolated variable reverses CD8+ T cell anergy and promotes tumor rejection.
J. Immunol.
177
:
4521
-4529.
36
Bradley, L. M., L. Haynes, S. L. Swain.
2005
. IL-7: maintaining T-cell memory and achieving homeostasis.
Trends Immunol.
26
:
172
-176.
37
Li, J., G. Huston, S. L. Swain.
2003
. IL-7 promotes the transition of CD4 effectors to persistent memory cells.
J. Exp. Med.
198
:
1807
-1815.
38
Yu, X., R. Abe, R. J. Hodes.
1998
. The role of B7-CD28 co-stimulation in tumor rejection.
Int. Immunol.
10
:
791
-797.
39
Chen, L., P. McGowan, S. Ashe, J. Johnston, Y. Li, I. Hellstrom, K. E. Hellstrom.
1994
. Tumor immunogenicity determines the effect of B7 costimulation on T cell-mediated tumor immunity.
J. Exp. Med.
179
:
523
-532.
40
Gimmi, C. D., G. J. Freeman, J. G. Gribben, G. Gray, L. M. Nadler.
1993
. Human T-cell clonal anergy is induced by antigen presentation in the absence of B7 costimulation.
Proc. Natl. Acad. Sci. USA
90
:
6586
-6590.
41
Linsley, P. S., J. A. Ledbetter.
1993
. The role of the CD28 receptor during T cell responses to antigen.
Annu. Rev. Immunol.
11
:
191
-212.
42
North, R. J..
1982
. Cyclophosphamide-facilitated adoptive immunotherapy of an established tumor depends on elimination of tumor-induced suppressor T cells.
J. Exp. Med.
155
:
1063
-1074.
43
Hamilton, S. E., M. C. Wolkers, S. P. Schoenberger, S. C. Jameson.
2006
. The generation of protective memory-like CD8+ T cells during homeostatic proliferation requires CD4+ T cells.
Nat. Immunol.
7
:
475
-481.
44
Harding, F. A., J. P. Allison.
1993
. CD28–B7 interactions allow the induction of CD8+ cytotoxic T lymphocytes in the absence of exogenous help.
J. Exp. Med.
177
:
1791
-1796.
45
Sallusto, F., J. Geginat, A. Lanzavecchia.
2004
. Central memory and effector memory T cell subsets: function, generation, and maintenance.
Annu. Rev. Immunol.
22
:
745
-763.
46
Min, B., H. Yamane, J. Hu-Li, W. E. Paul.
2005
. Spontaneous and homeostatic proliferation of CD4 T cells are regulated by different mechanisms.
J. Immunol.
174
:
6039
-6044.
47
Klebanoff, C. A., S. E. Finkelstein, D. R. Surman, M. K. Lichtman, L. Gattinoni, M. R. Theoret, N. Grewal, P. J. Spiess, P. A. Antony, D. C. Palmer, et al
2004
. IL-15 enhances the in vivo antitumor activity of tumor-reactive CD8+ T cells.
Proc. Natl. Acad. Sci. USA
101
:
1969
-1974.
48
Alpdogan, O., M. R. van den Brink.
2005
. IL-7 and IL-15: therapeutic cytokines for immunodeficiency.
Trends Immunol.
26
:
56
-64.