Type 1 and type 2 CD8+ effector T cell subpopulations promote long-term tumor immunity and protection to progressively growing tumor

Tumor-specific Tc1, but not Tc2, cells deliver protective antitumor immunity

We investigated whether secretion of multiple cytokines by CD8+ T cells is associated with improved protection against tumor challenge. We show that antitumor immunity induced by immunization with dendritic cells and a MHC class I-binding tumor peptide are dependent on secretion of IFN-gamma but not IL-4 or IL-5 by host cells. To further address the role of IL-4 and IL-5 in antitumor immunity, tumor-specific TCR-transgenic CD8+ T cells were activated in vitro to generate cytotoxic T (Tc) 1 cells that secrete high IFN-gamma and no IL-4 or IL-5 or Tc2 cells that secrete IL-4, IL-5, and some IFN-gamma. Both cell types killed target cells in vitro. Tc1 and Tc2 cells were adoptively transferred into syngeneic hosts, and their ability to protect against tumor challenge was compared. Tc1 cells were able to significantly delay tumor growth, whereas Tc2 cells or Tc2 cells from IFN-gamma(-/-) donors had no effect. This was due to neither the inability of Tc2 cells to survive in vivo or to migrate to the tumor site nor their inability to secrete IL-4 and/or IL-5 in the presence of limiting amounts of anti-CD3. However, IFN-gamma secretion by Tc2 cells was triggered inefficiently by restimulation with Ag compared with anti-CD3. We conclude that the ability to secrete "type 2" cytokines, and cytotoxic ability, have a limited role in antitumor immune responses mediated by CD8+ T cells, whereas the capacity to secrete high amounts of IFN-gamma remains the most critical antitumor effector mechanism in vivo.

Therapy of lung metastases through combined vaccination with carcinoma cells engineered to release IL-13 and IFN-gamma

TS/A spontaneous mouse mammary adenocarcinoma cells were engineered to release interferon-gamma (IFN-gamma), a Th1 cytokine (TS/A-IFNgamma) and interleukin-13 (IL-13), a Th2 cytokine (TS/A-IL13). Mice bearing lung micrometastases induced by parental TS/A cells received repeated subcutaneous vaccinations with TS/A-IFN-gamma admixed with TS/A-IL13 engineered cells. This combined treatment cured up to 75% of mice, whereas vaccinations with either TS/A-IFNgamma or TS/A-IL13 alone cured only 20-40% of mice. Combined TS/A-IL13 and TS/A-IFNgamma therapeutic vaccinations elicited a reactive infiltrate of CD4+ and CD8+ lymphocytes in lung metastases and an increased production of IFN-gamma in the spleen and lung, suggesting a shift of the immune response toward the Th1 type. The type of infiltrating cells along with the lack of efficacy in T cell-deficient mice point to a major role of T cells. In conclusion, no antagonism but a synergistic and effective definitive cure stems from the combined vaccination with tumor cells engineered to release a Th1 and a Th2 cytokine.

Gene microarrays reveal extensive differential gene expression in both CD4(+) and CD8(+) type 1 and type 2 T cells

An important subdivision of effector T cells can be made based on patterns of cytokine production and functional programs. Type 1 T cells produce IFN-gamma and protect against viral pathogens, whereas type 2 cells produce cytokines such as IL-4 and IL-5 and protect against large extracellular parasites. Both CD4(+) and CD8(+) T cells can be polarized into type 1 or type 2 cytokine-secreting cells, suggesting that both populations play a regulatory role in immune responses. In this study, we used high-density oligonucleotide arrays to produce a comprehensive picture of gene expression in murine CD4(+) Th1 and Th2 cells, as well as CD8(+) type 1 and type 2 T cells. Polarized type 1 and 2 cells transcribed mRNA for an unexpectedly large number of genes, most of which were expressed in a similar fashion between type 1 and type 2 cells. However, >100 differentially expressed genes were identified for both the CD4(+) and CD8(+) type 1 and 2 subsets, many of which have not been associated with T cell polarization. These genes included cytokines, transcription factors, molecules involved in cell migration, as well as genes with unknown function. The program for type 1 or type 2 polarization was similar for CD4(+) and CD8(+) cells, since gene expression patterns were roughly the same. The expression of select genes was confirmed using real-time PCR. The identification of genes associated with T cell polarization may give important insights into functional and phenotypic differences between effector T cell subsets and their role in normal responses and inflammatory disease.

Role of effector cell-derived IL-4, IL-5, and perforin in early and late stages of type 2 CD8 effector cell-mediated tumor rejection

Type 2 CD8 T cells (Tc2) secrete IL-4 and IL-5 and display perforin-dependent cytolysis in vitro. Using an OVA-transfected B16-melanoma model, we show that tumor-reactive Tc2 effector cells accumulated at the tumor site and induced tumor regression that enhanced survival in mice with pulmonary tumors. Transfer of perforin-deficient Tc2 cells generated from perforin gene knockout mice showed no differences in therapeutic efficiency when compared with wild-type Tc2 cells. In contrast, Tc2 cells derived from select cytokine gene-deficient mice showed that therapeutic effects were dependent on effector cell-derived IL-4 and IL-5 that led to a local elevation in lung-derived chemoattractants and accumulation of activated host-derived CD8/CD44(high), CD4/CD44(high), and OVA-specific tetramer-positive CD8 cells in vivo. Host-derived T and non-T immune cells increased in the lung over time and correlated with an elevated production of type 1-related chemokines. Conversely, donor Tc2 cell numbers markedly diminished at later times, suggesting that prolonged therapeutic responses were due to host-derived mechanisms. Moreover, type 1 host responses were detectable with increased levels of IFN-gamma production by lung-derived CD4 and CD8 T cells from surviving Tc2-treated mice. Transfer of Tc2 cells into IFN-gamma-deficient tumor-bearing mice was markedly less effective then into wild-type mice, suggesting that host-derived IFN-gamma-dependent mechanisms play a role in Tc2-mediated antitumor responses.

Immunotherapy of melanoma: a dichotomy in the requirement for IFN-gamma in vaccine-induced antitumor immunity versus adoptive immunotherapy

The mechanism by which tumors are rejected following the adoptive transfer of tumor-specific T cells is not well characterized. Recent work has challenged the requirement for cytotoxicity mediated by either the perforin/granzyme or Fas/Fas ligand pathway in T cell-mediated tumor regression. Many reports, including ours, suggest that tumor-specific production of IFN-gamma is critical for T cell-mediated tumor regression. However, in most of these studies the evidence to support the role for IFN-gamma is only indirect. We have directly examined the requirement for IFN-gamma using IFN-gamma knockout (GKO) mice. The results show an interesting dichotomy in the requirement for IFN-gamma: Antitumor immunity induced by active-specific immunotherapy (vaccination) required IFN-gamma, whereas adoptive immunotherapy did not. In GKO mice vaccination with the GM-CSF gene-modified B16BL6-D5 tumor (D5-G6) failed to induce protective immunity against parental D5 tumor. However, adoptive transfer of effector T cells from GKO mice cured 100% of GKO mice with established pulmonary metastases and induced long term antitumor immunity and depigmentation of skin. Furthermore, in vivo neutralization of IFN-gamma by mAb treatment or adoptive transfer into IFN-gamma receptor knockout mice failed to block the therapeutic efficacy of effector T cells generated from wild-type or perforin knockout mice. Analysis of regressing metastases revealed similar infiltrates of macrophages and granulocytes in both wild-type and GKO mice. These results indicate that in this adoptive immunotherapy model, neither a direct effect on the tumor nor an indirect effect of IFN-gamma through activation of myeloid or lymphoid cells is critical for therapeutic efficacy.

The Role of Adoptively Transferred CD8 T Cells and Host Cells in the Control of the Growth of the EG7 Thymoma: Factors That Determine the Relative Effectiveness and Homing Properties of Tc1 and Tc2 Effectors

We had previously examined the factors that regulate the response of OVA-specific TCR-transgenic CD8 T cells to the B16 OVA melanoma, growing as lung metastases. We examine here whether the same parameters operate for EG7, growing intradermally. Tc1 or Tc2 CD8 effector cells from OT-1 mice were injected either mixed with the tumor or i.v. at day 0 or 7. Tc2 were one-fifth to one-tenth as effective as Tc1 when injected with the tumor, in controlling tumor growth, but were only 1/20 to 1/100 injected i.v. Tc1 injected i.v. entered the draining lymph nodes faster than Tc2 and caused a faster accumulation of host cells. Both caused an abrupt termination of host cell entry into lymph nodes and spleen after tumor elimination, but this occurred earlier for Tc1 than for Tc2. Host responses were ineffective in the absence of adoptive transfer but were essential after transfer. Perforin expression in the donor cells plays no role in adoptively transferred Tc1 or Tc2 control of the tumor, and neither IL-4 nor IL5 is needed for Tc1 or Tc2 function. Tc1 cells from mice lacking IFN-gamma, however, control tumor growth less well, whereas Tc2 effectors lacking IFN-gamma are unaffected. Tc1 from IFN-gamma-deficient mice attract fewer host cells to the draining lymph node, whereas Tc1 cells from perforin-deficient donors are unimpaired. We conclude that host cell recruitment is a crucial element in adoptive immunotherapy. The differences between the EG7 and the previous B16 melanoma model are discussed.

Immunopotentiating role of IFN-gamma in early and late stages of type 1 CD8 effector cell-mediated tumor rejection

Type 1 cytolytic CD8 effector T cells (Tc1) characteristically secrete IFN-gamma. Using an OVA-transfected B16 melanoma lung tumor model, we show that OVA Ag-specific Tc1 cells mediate a reduction in tumor growth that significantly prolongs survival in tumor-bearing mice. Transfer of Tc1 cells from OT-I mice crossed to IFN-gamma-KO mice showed that IFN-gamma-deficient Tc1 effector cells were less therapeutically effective than corresponding cells from wildtype mice. Therapeutic effects were dependent, in part, on effector cell-derived IFN-gamma, which not only induced elevated levels of lung-derived IP-10 and RANTES chemokine message in vivo, but also increased the local accumulation of activated host-derived CD4(+)/CD44(High), CD8(+)/CD44(High), and non-T-immune cell populations at the tumor site. Over time, the numbers of host-derived immune cells increased in the lung, which correlated with an elevated production of IP-10 and RANTES and a continued reduction in tumor burden. Conversely, donor Tc1 cell numbers markedly diminished at corresponding times, suggesting that prolonged therapeutic responses were due to the presence of host-derived antitumor mechanisms. Moreover, adoptive transfer of IFN-gamma-deficient Tc1 cells into tumor-bearing IFN-gamma-KO recipients showed that both recipient and donor-derived IFN-gamma play a significant role in Tc1-mediated responses and that Tc1 effector cell immunotherapy is predominantly mediated by IFN-gamma-dependent mechanisms.

IFN-gamma can promote tumor evasion of the immune system in vivo by down-regulating cellular levels of an endogenous tumor antigen

Although IFN-gamma has been generally thought to enhance antitumor immune responses, we found that IFN-gamma can promote tumor escape in the CT26 colon carcinoma by down-regulating the protein expression of an endogenous tumor Ag. gp70, the env product of the endogenous ecotropic murine leukemia virus, has been reported to be the immunodominant Ag of CT26. We show that IFN-gamma down-regulates intracellular and surface levels of gp70 protein resulting in a reduced lysis by CTL, which is restored by pulsing IFN-gamma-treated CT26 with the L(d)-restricted immunodominant AH1 epitope derived from gp70. To investigate the role of CT26 sensitivity to IFN-gamma in vivo, we constructed two variants of CT26, CT26.mugR and CT26.IFN, that are unresponsive to IFN-gamma or express IFN-gamma, respectively. We demonstrate using these variants that tumor responsiveness to IFN-gamma promotes a reduction in tumor immunogenicity in vivo that is correlated with an increased tumor incidence in immune mice. Analysis of the tumors from mice challenged with CT26 or CT26.mugR revealed infiltration of CD8 T cells secreting IFN-gamma. We conclude that IFN-gamma secreted by tumor-infiltrating T cells promotes tumor escape through the down-regulation of the endogenous tumor Ag gp70. These findings have impact on the design of effective antitumor vaccine strategies.