IFN-γ knockout mice (GKO) rejected C26 colon carcinoma cells transduced to secrete interleukin(IL)-12 but do not reject similarly transduced TSA mammary adenocarcinoma (C26/12 and TSA/12 cells, respectively). To determine whether such difference could be because of a different tumor response to IFN-γ, we injected BALB/c mice with TSA, C26, and their IL-12-transduced counterparts rendered unresponsive to IFN-γ by stable transduction of a dominant negative (DN), truncated IFN-γ receptor α chain. TSA/DN and C26/DN showed the same in vivo growth kinetics as parental cells, whereas coexpression of IL-12 induced rejection independent of tumor-cell responsiveness to IFN-γ. This suggests that the role of IFN-γ is primarily in activating the host immune response, which appears to depend on the intrinsic immunogenicity of the target tumor. C26 and TSA share a common tumor-associated antigen, yet C26 cells are more immunogenic than TSA. C26/12 expressed 10-fold higher levels of class I MHC molecules and induced higher CTL activity compared with TSA/12 cells in GKO mice. Moreover, whereas in GKO mice the TSA/12 tumor was associated with a greater number of infiltrating T cells, only those infiltrating C26/12 tumor expressed the activation marker OX40. The search for cytokine(s) that might contribute in determining the different T-cell response to these IL-12-transduced tumors in GKO mice revealed a role of IL-15. In situ hybridization showed IL-15 expression in C26/12 but not in TSA/12 tumors. In addition, injection of GKO mice with soluble IL-15 receptor-α to block IL-15 expression prevented rejection of C26/12 cells. Together, the results suggest that in the absence of IFN-γ, IL-12 can exert antitumor activity through alternative mechanisms, depending on the tumor cell type and the availability of cytokines that can replace IFN-γ in sustaining T-cell functions.

IL-123 is a proinflammatory cytokine that exerts antitumor activity on a variety of cancers through activation of innate and adaptive immunity, as well as through inhibition of angiogenesis (1, 2, 3, 4, 5). The antitumor effect of IL-12 rests largely in its ability to induce high levels of IFN-γ production by T and NK cells (6). IL-12-induced IFN-γ has a direct antiproliferative effect on several tumor cell types and up-regulates the expression of class I MHC, thus enhancing immune recognition of TAAs (7). IFN-γ also stimulates the production of nitric oxide by tumor and/or host cells, which inhibits tumor growth (8, 9). Furthermore, IFN-γ is involved in the production of chemokines such as IFN-γ-inducible protein-10 and a monokine induced by IFN-γ by a variety of cell types, including mononuclear cells, fibroblasts, keratinocytes, endothelial cells, T cells, and some tumor cells. These chemokines act as chemoattractants for lymphocytes, and inhibit the differentiation and proliferation of endothelial cells (10, 11, 12, 13, 14, 15, 16). IFN-γ mediates its effects through a receptor that is widely expressed on normal and malignant cells (17). The IFN-γR consists of α and β chains, which dimerize on ligand binding. Dimerization triggers receptor transphosphorylation and activation of the associated JAK1 and JAK2 tyrosine kinases (18, 19). Activated JAK kinases then phosphorylate downstream STAT1α, causing its dimerization and nuclear translocation (20). Nuclear STAT1α binds GAS in IFN-γ-responsive genes, thereby modulating their transcription (21).

Although IFN-γ is the main effector of IL-12 antitumor activities, systemic injection of exogenous IFN-γ cannot completely substitute for IL-12 for activity (7). This finding might reflect differences in the half-life of the two proteins, but more likely, the differing distribution of their receptors; IFN-γR is ubiquitously expressed, whereas the IL-12R β2-chain is expressed only on NK cells and activated lymphocytes (22). We demonstrated previously that GKO mice still partially reject growth of an IL-12-transduced colon carcinoma cell line (C26/12) through a mechanism that requires the presence of CD4+ T cells (23), which produce GM-CSF associated with the reduction of tumor blood vessels through the activity of polymorphonuclear leukocytes and CD8+ T cells (24). These findings suggest that the antitumor effect of IL-12 can be mediated via an IFN-γ-independent pathway at least in the C26 model.

Here, we show that, unlike C26/12 tumor cells, mammary adenocarcinoma TSA cells transduced with IL-12 genes (TSA/12) are fully tumorigenic in GKO mice. Recently, the ability of the immune system to control tumor development in carcinogen-treated or tumor-prone mice, as well as the rejection of transplantable tumor, have been correlated with the direct effect of IFN-γ on tumor cells (25, 26, 27). Using IL-12-transduced or nontransduced TSA and C26 tumor cells rendered unresponsive to IFN-γ by stable transduction and overexpression of a IFN-γRDN, we tested whether the differential rejection of C26/12 and TSA/12 cells observed in GKO mice might rest in different tumor sensitivity to IFN-γ. The results indicate that tumor cell type and intrinsic immunogenicity are the determinants of the need for IFN-γ in tumor rejection. In our experimental setting, IFN-γ acts on cells of the immune response rather than directly on tumor cells.

Tumors and Mice.

The murine colon adenocarcinoma cell line C26 was derived from BALB/c mice treated with N-nitroso-N-methylurethane (28). TSA is an aggressive and poorly immunogenic cell line established from the first in vivo transplant of a moderately differentiated mammary adenocarcinoma that arose spontaneously in a BALB/c mouse (29). Both TSA and C26 cells are negative for class II MHC molecule expression. Cells were cultured at 37°C in a 5% CO2 atmosphere in DMEM (Life Technologies, Inc., Paisley, United Kingdom) supplemented with 10% FCS (Life Technologies, Inc.). BALB/cnAnCr (Charles River, Calco, Italy) and GKO mice (30) on a BALB/c background (BALB/c-Ifg<tm 1 129>; The Jackson Laboratory, Bar Harbor, ME) were maintained at the Istituto Nazionale Tumori under standard conditions according to institutional guidelines. Tumorigenic activity of control and transduced TSA or C26 cells was assayed in mice injected s.c. in the left flank with 105 cells in 0.2 ml. Tumor growth and size were recorded twice each week, and tumor growth was expressed as percentage of tumor-free mice over total injected mice at the indicated time points.

Treatment with sIL-15Rα.

A soluble fragment of the murine IL-15Rα (31) and the functionally mutated form of sIL-15Rα (M4), containing an amino acid substitution at the Sushi domain necessary for binding to IL-15 were used to block endogenous IL-15 bioactivity on GKO mice (32). sIL-15Rα (40 μg/mouse) was injected i.p. daily for 20 days starting 2 days before C26/12 tumor inoculation. Untreated- or M4-treated mice were used as controls.

Vector Construction and Retroviral Infection.

An 860-bp fragment containing the extracellular and transmembrane domains of the IFN-γ receptor, together with the FLAG sequence, derived from pEF2-FlmugR-DNM (a kind gift of Sidney Pestka, Johnson Medical School, Piscatway, NJ), was excised with XbaI and KpnI, blunt-ended, and cloned into the HpaI site of the LXSH retroviral vector (33) to obtain the vector IFN-γRDNLSH. This retroviral vector was transfected into the gp+E86 packaging cell line (34) by standard calcium phosphate coprecipitation, and the 48-h culture supernatant was used to infect the amphotropic PA317 packaging cell line (35). Infected PA317 cells were selected with hygromycin and used to generate helper-free, virus-containing supernatants. TSA, TSA/12 (36), C26, and C26/12 (37) target cells were infected by four cycles of exposure to undiluted supernatant for 2 h in the presence of Polybrene (8 mg/ml). At 48 h after infection, cells were diluted and selected in hygromycin (500 μg/ml). Bulk cultures and single resistant colonies were expanded and screened by FACS analysis for IFN-γR and FLAG expression, and by ELISA for IL-12 production (106 cells/ml; 48 h at 37°C). Limiting dilution cloning was used to obtain optimal levels of IFN-γRDN expression.

Flow Cytometry.

Expression of IFN-γRDN in transduced cell lines was assayed by flow cytometry after conventional staining with biotin-conjugated anti-IFN-γR (CD119, clone GR20) followed by streptavidin-phycoerytrin (PharMingen, San Diego, CA) or with mouse anti-FLAG M2 (Kodak, New Haven, CT) followed by fluorescence-conjugated antimouse immunoglobulin (Biosource, Camarillo, CA). For assay of class I MHC expression, biotin-conjugated anti-H-2Ld/H-2Db clone 28–14-8 (PharMingen) was used. Hybridoma 35/299, which expresses a rat IgG2a mAb specific for gp70 was a gift from Drew Pardoll (Johns Hopkins University, Baltimore, MD). Isotype-matched mAbs of unrelated specificity were used as controls. Analysis was performed on a FACScan (Becton Dickinson, Franklin Lakes, NJ). Data were collected on 5,000–10,000 viable cells and analyzed using winMDI 2.8 software.

EMSA.

Nuclear extracts were isolated from control, and transduced C26 or TSA cells (2 × 106/ml) either unstimulated or stimulated with the indicated amount of mrIFN-γ (PharMingen) according to Schreiber et al.(38). End-labeled DNA probes (50,000 cpm/sample) were mixed with 4 μg of crude nuclear extract and incubated at room temperature for 20–30 min in the presence of 1 μg of poly(dI·dC; Boehringer Mannheim, Roche Diagnostics, Monza, Italy) in 10 μl of buffer C [20 mm HEPES (pH 7.9), 10% glycerol, 0.4 M NaCl, 0.1 mm EDTA, 0.1 m EGTA, 1 mm DTT, 1 mm phenylmethylsulfonyl fluoride, 10 μg/ml leupeptin, 10 μg/ml aprotinin, and 10 μg/ml pepstatin A]. Samples were resolved on a 6% polyacrylamide gel in running buffer (0.5 × Tris-borate EDTA) for 1 h at 200 V. The gel was dried and exposed in a PhosphorImager storage screen and visualized using PhosphorImager 445S1 (Molecular Dynamics, Sunnyvale, CA). Probe competition experiments were carried out using an unlabeled GAS probe (5′ AGC-TTG-TAT-TTC-CCA-GAA-AAG-GGA-TC; Ref. 39). Supershift experiments were performed by preincubating nuclear extracts with 1–2 μg of Abs for 30 min at 4°C before the probe was added. All of the polyclonal Abs recognizing STAT family protein and used in supershift experiments were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA).

NK Cell Assay.

To measure NK cell activity, fresh splenocytes from SCID mice (44% NK-positive cells) were incubated for 2 days in RPMI 1640 (Life Technologies, Inc.) supplemented with 10% FCS and 500 units/ml of human IL-2 (Chiron-Italia, Milan, Italy) in the presence or absence of murine rIL-12 (2.5 ng/ml; kindly provided by M. Gately, Roche, Nutley, NJ) and tested for cytotoxic activity against 51Cr-labeled YAC, TSA, or C26 target cells.

Cell-mediated Cytotoxicity Assay.

Naïve BALB/c or GKO mice were inoculated into the footpad with γ-irradiated (15,000 rad) C26/12 or TSA/12 cells at a dose of 5 × 106 irradiated cells/mouse. After 5 days, popliteal lymph nodes were removed, and lymphoid cells were suspended to 5 × 105 cells/ml in RPMI 1640 supplemented with 10% FCS and restimulated in vitro for 5 days at 37°C in 5% CO2 in the presence of the common tumor antigenic peptide AH1 (SPSYVYHQF) at a final concentration of 1 μg/ml (synthesized by Primm S.r.l., Milan, Italy). Effector cells were used in cell-mediated cytotoxicity assays, with AH1-pulsed BALB/c blasts. In selected experiments restimulation in vitro of effector cells was performed for 5 days in the presence of 5 × 104/ml irradiated (15,000 rad) C26 or TSA cells, and in these experiments C26 and TSA cells were used as specific targets.

Immunocytochemistry and in Situ Hybridization.

Tumor fragments, tumor-draining lymph nodes, and spleens were collected 1, 3, and 7 days after tumor inoculation, embedded in OCT compound (Miles Laboratories, Inc., Elkhart, IN), snap-frozen in liquid nitrogen, and stored at −80°C. Immunocytochemical analysis using the biotinylated antirat IgG/streptavidin-horseradish peroxidase method was performed as follows: 5-μm cryostat sections were fixed in acetone and immunostained with rat antimouse mAb against TCR Vβ6 (clone RR4-7; PharMingen), Vβ8.3 (clone 1B3.3; PharMingen), CD8 (53.6.72 hybridoma, Lyt2; American Type Culture Collection), CD4 (GK1.5 hybridoma, T200; American Type Culture Collection), OX40 (OX86 hybridoma; European Collection of Animal Cell Cultures, Salisbury, United Kingdom), and NK (pan-NK, clone DX5; PharMingen). Sections were sequentially blocked using a tissue biotin blocking kit (DAKO) and incubated with optimal dilutions of primary Abs, biotynilated-antirat IgG (DAKO), and streptavidin-horseradish peroxidase (DAKO). Each incubation step was for 30 min and was followed by a 10-min wash in PBS. Sections were then incubated with 0.03% H2O2 and 0.06% 3,3′-diaminobenzidine (BDH Chemicals, Poole, United Kingdom) for 2–5 min, washed in tap water, and counterstained with hematoxylin. The number of immunostained cells was determined by light microscopy (magnification ×400) in 5 fields on a 1-mm2 grid and given as cells/mm2 (mean ± SD).

The presence of IL-15 mRNA was investigated by in situ hybridization using IL-15 cDNA obtained by cutting the pVKLIL-15IRES1neo plasmid (kindly provided by Silvano Ferrini, National Cancer Institute, Genova, Italy) at EcoRI and BamHI sites. After digestion, the 675-bp fragment was run on 1% agarose gel, purified using the Qiaex II gel extraction kit (Qiagen, Hiden, Germany) and biotin-labeled using a random primer biotin labeling kit (NEL804; ENZO Diagnostics, Farmingdale, NY). Cryostat sections were harvested on RNA-grade slides, air-dried, and fixed in 4% buffered paraformaldehyde for 10 min, dehydrated in ethanol, sequentially washed in PBS/50 mm MgCl2 and 200 mm Tris-HCl-glycine, acetylated in 2× SSC, 0.1 m triethanolamine, and 0.5% acetic anhydride (pH 8.0), washed in 2× SSC, and finally dehydrated in ethanol. Slides were prehybridized for 10 min at 42°C in 2× SSC, 50% formamide, and 500 μg/ml salmon sperm DNA, and hybridized overnight at 42°C with biotin-labeled-specific IL-15 cDNA probe (20 ng/ml), 2× SSC, 500 μg/ml salmon sperm DNA, 5× Denhardt’s solution, 10 mm dithiothreitiol, and 10% dextran sulfate. Unbound and nonspecifically bound probes were removed by sequential washes in 2× SSC, 50% formamide, and 1× SSC, 50% formamide at 50°C, and in 0.1× SSC at room temperature. Slides were air-dried and the biotnylated signal was detected using a TSA biotin system kit (NEN700; ENZO Diagnostics). Slides were counterstained with hematoxylin. RNA-specific binding was controlled by previous digestion with 100 μg/ml RNase A and 10 units/ml RNase T (Sigma Chemical Co., Poole, United Kingdom). PstI-digested pUC9 plasmid fragments were used as negative controls.

Statistical analysis was performed using the Student t test (Microsoft Office).

Growth of IL-12-transduced TSA and C26 Tumor Cells in BALB/c and GKO Mice.

TSA/12 cells formed tumors in all of the GKO mice but only in 10% of BALB/c mice (Fig. 1,A), indicating that the IL-12 effect is strictly IFN-γ-dependent. In contrast, C26/12 tumor growth was rejected in 70% of both BALB/c and GKO mice (Fig. 1 B). These contradictory results cannot be explained by a disparity in IL-12 production, because both transduced cell lines were selected to release similar amounts of IL-12 (6 ± 0.5 ng/106 cells/48 h). No differences between TSA and C26 cells in the inhibition of cell proliferation in response to rIFN-γ even at high doses of rIFN-γ (2000 units/ml) was observed in vitro (data not shown). Thus, a direct antiproliferative effect of IFN-γ on tumor cells is unlikely.

In Vitro Characterization of IFN-γRDN-transduced TSA and C26 Cells.

To address whether in vivo IL-12-induced IFN-γ acts on host immune cells or directly on tumor cells, C26, C26/12, TSA, and TSA/12 tumor cells were rendered IFN-γ-unresponsive by stable transduction with an IFN-γRDN α chain construct (27). Transduced clones were selected for IFN-γR α chain overexpression and for FLAG reactivity to distinguish the transgene encoded protein (data not shown). Gel EMSA for inhibition of nuclear STAT1α was used to confirm the block in IFN-γ signaling in these clones. Murine rIFN-γ-treated TSA, TSA/12, C26, and C26/12 cells revealed a nuclear complex (Fig. 2, A and B) that was inhibited by anti-STAT1α Ab and by cold GAS probe but not by an anti-STAT4 Ab (Fig. 2 C), whereas no STAT1α induction was observed in rIFN-γ-treated nuclear extracts from IFN-γRDN-transfected cells. Involvement of a recently described novel STAT1-independent IFN-γ signaling pathway through c-myc and c-jun induction in the absence of STAT1 phosphorylation (40) seems unlikely because this signal requires an intact IFN-γR α chain (41) missing in the DN constructs (27).

In addition, IFN-γRDN-transduced TSA and C26 cells no longer up-regulated class I MHC in response to rIFN-γ, and this loss of responsiveness was not because of a general defect in the class I MHC induction pathway, because both cell lines responded normally to murine IFN-α (Fig. 3; data not shown).

Tumorigenicity of IFN-γRDN-transduced TSA and C26 Cells.

All of the normal BALB/c mice injected with 105 parental or IFN-γRDN-transduced TSA and C26 cells formed rapidly progressing tumors (Fig. 4, A and B). Identical results were obtained using other IFN-γRDN-transduced TSA and C26 cell clones (data not shown). Reverse transcription-PCR and immunostaining of C26/DN and TSA/DN tumors collected 15 days after injection demonstrated intact expression of the DN receptors (data not shown).

Tumorigenicity of IFN-γ-unresponsive TSA/12 and C26/12 in the Presence or Absence of Endogenous IFN-γ.

To determine whether IFN-γ responsiveness of TSA and C26 cells is important for local IL-12-induced rejection of these tumors, BALB/c mice were injected with TSA/12/DN and C26/12/DN cells and tumor growth was evaluated (Fig. 4, C and D).

TSA/12, TSA/12/DN, C26/12, and C26/12/DN tumors showed equally impaired tumor take, with ∼80% of mice remaining tumor-free at 70 days after inoculation. These results strongly suggests that the antitumor effect of IL-12-induced IFN-γ is most likely not directed to tumor cells but instead to host cells. Mice that rejected C26/12/DN cells did not reject a challenge with BALB/c fibrosarcoma cells either transduced with the IFN-γRDN vector or nontransduced (data not shown), ruling out the possibility that construct- (e.g., the FLAG epitope present in the extracellular portion of the DN chain) or vector-derived antigens increase the immunogenicity of DN-transduced cells.

Additional proof that the status of IFN-γ-responsiveness of TSA and C26 cells does not determine the take or rejection of the tumors came from analysis of GKO mice injected with TSA/12/DN or C26/12/DN cells (Fig. 5, A and B). As expected, all of these mice developed TSA/12/DN tumors, whereas after initial take, 80% of mice rejected C26/12/DN tumors. Thus, IL-12-induced rejection of TSA tumors requires the presence of endogenous IFN-γ, but rejection of C26 cells does not.

Susceptibility of TSA/12 and C26/12 to Innate and Adaptive Immunity.

On the basis of the results in GKO mice, we examined the effect of IL-12 on the ability of host T and NK cells to recognize and kill C26 and TSA cells in the presence or absence of endogenous IFN-γ. C26, C26/12, TSA, and TSA/12 were analyzed for expression of the shared TAA gp70, the envelope protein of an endogenous murine leukemia virus. The immunodominant epitope recognized within gp70 is AH1, a nonamer peptide presented in the context of class I MHC Ld(42). Immunofluorescence and FACS analysis revealed higher gp70 (Fig. 6,A) but lower class I MHC Ld (Fig. 6,B) expression in both TSA and TSA/12 as compared with C26 and C26/12 cells. In 51Cr release assays, splenocytes derived from SCID mice enriched in NK cells and activated in vitro with IL-2 or IL-2 + IL-12 killed TSA target cells more efficiently than C26 targets (Fig. 6,C), indicating that innate immunity is more effective against the tumor cells expressing fewer class I MHC molecules on their surface. We then compared C26/12 and TSA/12 cells for the ability to induce CTLs in both BALB/c and GKO mice. Irradiated tumor cells were injected into the footpad, and lymphocytes from draining popliteal nodes were restimulated in vitro with the AH1 peptide (Fig. 6,D), or irradiated TSA or C26 tumor cells accordingly to the tumor used for priming (Fig. 6, E and F). Whereas C26/12 cells primed CTL induction in both strains, TSA/12 cells induced weak or undetectable CTL activity against AH1-pulsed blast or TSA cells (Fig. 6, D–F).

In Vivo Characterization of TSA/12 and C26/12 Tumor-infiltrating Lymphocytes in BALB/c and GKO Mice.

Immunohistology has been used to characterize the kinetic and types of leukocytes infiltrating C26/12 and TSA/12 in BALB/c and GKO mice.

The site of injection was excised 1, 3, and 7 days after tumor cell inoculation. As shown in Table 1, TSA/12 tumors had more infiltrating T than C26/12. A more in depth analysis of the two major TCR types, Vβ6 and Vβ8.3, involved in recognition of the AH1 immunodominant peptide was performed (42, 43). Vβ6 was mainly present in C26/12 tumors, and in GKO mice these Vβ6 T lymphocytes accounted for almost all of the CD8+ tumor-associated lymphocytes. In contrast, Vβ8.3 was present in TSA/12 tumors regardless of whether the tumors were collected from BALB/c or GKO mice. This apparently controversial result was resolved by evaluating a functional marker of T cell activation, OX40 (44). Indeed, almost all of the T cells derived from the C26/12 tumor, either from BALB/c or GKO mice, stained positively for OX40 (Table 1; Fig. 7, B and D). In contrast, TSA/12 tumors from BALB/c mice, although richer in infiltrating T cells, had fewer OX40-positive cells (Table 1; Fig. 7,A). An almost complete depletion of OX40 expression was observed in TSA/12 tumors from GKO mice (Table 1; Fig. 7 C). In accordance with the analysis of this functional marker, TSA/12 is rejected in BALB/c but not in GKO mice.

IL-15 Is the Candidate Substitute for IFN-γ in Mediating IL-12 Antitumor Activity.

On the basis of: (a) our recent findings that TSA/12 and TSA transduced with IL-15 are not rejected in GKO mice, whereas TSA cells transduced with both IL-12 and IL-15 genes are promptly rejected;4 (b) the antitumor activity of IL-15 and its synergistic effects with IL-12 (45, 46, 47); and (c) the essential role of IL-15 in controlling CD8+ T-cell trafficking, and CD8+ memory T-cell survival and functions in vivo(48, 49), we analyzed IL-15 expression in C26/12- and TSA/12-injected mice by in situ hybridization. Indeed, IL-15 expression was induced in BALB/c (Fig. 7,F; Table 1) and, to a slightly lesser extents, in GKO mice injected with C26/12 cells (Fig. 7,H), whereas rare IL-15-positive cells with macrophage morphology were present in TSA/12-injected BALB/c mice (Fig. 7,E), and no IL-15 was detected in TSA/12-injected GKO mice (Fig. 7,G). To prove that IL-15 produced during the rejection of C26/12 in GKO mice is functionally needed, mice were treated with a sIL-15Rα (31). None of the treated GKO mice were able to reject C26/12, whereas mice treated with M4, a functionally mutated sIL-15Rα unable to bind IL-15, showed 50% of rejection as did untreated mice (Fig. 8). This observation provides new data implicating IL-15 in IL-12-mediated antitumor activity at least in the absence of endogenous IFN-γ.

C26 colon carcinoma and TSA mammary tumor cells respond differentially to systemic rIL-12 and, when transduced to express IL-12, differentially require the presence of IFN-γ for rejection by the host. In fact, rIL-12 treatment of s.c.-injected C26 cells was shown to be ineffective, whereas treatment of C26 transduced with IL-2, which changes the pattern and number of infiltrating cells without affecting tumor take, resulted in rejection by 80% of the mice (50). In a different approach in which the effect of rIL-12 was tested against C26 cells injected intrasplenically, systemic rIL-12 completely inhibited liver metastases because of the abundant presence of NK T cells (51). Unlike C26, TSA cells injected s.c. are highly susceptible to rIL-12 treatment, leading to T lymphocyte and macrophages infiltration associated with necrosis (36).

In our studies, both IL-12-transduced-C26 and -TSA cells selected to release the same amount of cytokine were rejected in BALB/c mice, but only the C26/12 tumor was rejected in the absence of IFN-γ (Fig. 1; Ref. 23). The role of IFN-γ and the nature of IFN-γ target cells in IL-12-induced tumor rejection have remained unclear because of contrasting results on the possible direct effect of IFN-γ on tumor cells. This issue has been addressed by overexpressing a mutant form of the IFN-γR α chain in transplantable tumors (26, 27, 52, 53) or by using mice knocked out for this molecule (53, 54). We have stably transduced the DN truncated IFN-γR α chain into TSA and C26 cells alone or together with IL-12 genes, and injected them into BALB/c and GKO mice. In BALB/c mice, tumor take and growth kinetics of TSA/DN and C26/DN were similar to that of their parental counterparts, and both tumors were rejected in the presence of locally released IL-12 (TSA/12/DN and C26/12/DN; Fig. 4). These results contrast with those of Coughlin et al.(52), who reported that IFN-γ-unresponsive tumor cells were more tumorigenic and less susceptible to rIL-12 therapy in vivo. This discrepancy likely rests in the use of different tumor models, and rejection of the K1735 melanoma and the SCK mammary carcinoma used in that study was through IFN-γ-inducible protein-10-mediated inhibition of angiogenesis rather than by leukocyte killing. In contrast, rejection of both C26/12 and TSA/12 tumors requires leukocyte infiltration and killing. Moreover, we used tumor cells that express IFN-γRDN and IL-12 simultaneously, thus colocalizing IL-12 induction of IFN-γ and the block in its effect on tumor cells at the injection site.

Our results are consistent with those of Mumberg et al.(53), who also reported that tumor rejection was independent of tumor sensitivity to IFN-γ. In the same context, Qin et al.(54), using a carcinogen-induced tumor from IFN-γR knockout mice, observed that tumor IFN-γR expression on nonhematopoietic host cells but not on tumor cells was required for CD4+ T-cell mediated tumor rejection. Thus, cell type characteristics of the target tumor represent a key factor in determining tumor responsiveness to IFN-γ. Indeed, IFN-γ was shown recently to promote tumor escape of CT26 colon carcinoma cells, which are closely related to the C26 tumor, by down-regulating surface expression of the endogenous TAA gp70 (55). Our C26 and TSA tumor cells express the gp70 antigen (Fig. 6), but down-modulation of this molecule by IFN-γ was slight in C26 cells and undetectable in TSA cells (data not shown).

Expression of gp70 on TSA and C26 cells did not correlate with their immunogenicity. The gp70 immunodominant epitope AH1 is presented in the context of class I MHC Ld(42), which was expressed at higher levels on C26 and C26/12 cells than on TSA and TSA/12 cells (Fig. 6,B). Although such MHC differences were between 8- and 10-fold, they exceeded the 5-fold difference that dictates the selection of effector cells mediating immunocytokine therapy of the CT26 colon carcinoma (56). This difference in class I MHC expression might explain why NK cells killed TSA cells more efficiently than C26 target cells, and why C26/12 but not TSA/12 cells were able to prime CTLs in the absence of IFN-γ (Fig. 6, C–F). An attempt to correlate these in vitro data with the in vivo tumor infiltration of T lymphocytes bearing TCR Vβ used for AH1 recognition (42, 43) gave apparently contradictory results. Vβ6 were mainly found in C26/12 tumors and accounted for almost all CD8+ T cells in GKO mice. The alternative TCR type, Vβ8.3 was dominant in TSA/12 tumors either from BALB/c or GKO mice (Table 1). Although the number of Vβ6+ T lymphocytes in C26/12 appears to correlate with tumor rejection, the number of Vβ8.3+ T lymphocytes does not, as TSA/12 is rejected in BALB/c only. A possible explanation of these results may be the activation status of infiltrating T cells. Analysis of OX40, an activation marker for CD4+ and CD8+ T cells (44), indeed revealed that all of the tumors, except TSA/12 from GKO mice, were positive for OX40, correlating with tumor rejection.

Our preliminary efforts to identify a cytokine(s) that might plays a role in determining the differential TCR in these tumors has thus far eliminated GM-CSF, which is involved in C26/12 tumor rejection in GKO mice but is similarly induced in TSA/12-injected GKO mice (data not shown). By contrast, TSA cells transduced to express IL-15, a cytokine that plays a role in activation of NK and T-cell effector function (reviewed in Ref. 57), are rejected in BALB/c mice but not in GKO mice (4). In situ hybridization analysis of C26/12- and TSA/12-injected GKO mice revealed IL-15 expression at the C26/12 but not the TSA/12 tumor site (Fig. 8; Table 1). Moreover, block of endogenous IL-15 resulted in complete loss of protection from C26/12 tumor development in GKO mice (Fig. 8). This data complemented well the demonstration that TSA double transduced with IL-12 and IL-15 is rejected in GKO mice through CD8 T cells (4).

Together, the results suggest that IL-12 can exert antitumor activity throughout alternative mechanisms independent from IFN-γ but related to the intrinsic tumor immunogenicity and the availability of cytokines, such as IL-15 and GM-CSF, that can replace IFN-γ in sustaining T-cell functions.

Fig. 1.

Tumor take and onset of TSA/12 (A) and C26/12 (B) cells in BALB/c (▪) or GKO (•) mice. Groups of 18 mice were injected with 105 tumor cells. TSA/12 and C26/12 cells produced comparable amounts of IL-12 (6 ± 0.5 ng/106 cells/ml) after 48 h of in vitro culture.

Fig. 1.

Tumor take and onset of TSA/12 (A) and C26/12 (B) cells in BALB/c (▪) or GKO (•) mice. Groups of 18 mice were injected with 105 tumor cells. TSA/12 and C26/12 cells produced comparable amounts of IL-12 (6 ± 0.5 ng/106 cells/ml) after 48 h of in vitro culture.

Close modal
Fig. 2.

Inhibition of STAT1α activation in IFN-γRDN-transduced TSA and C26 cells. EMSA was performed with a GAS probe, and nuclear extracts were isolated from (A) TSA, TSA/DN clone 12, TSA/12, and TSA/12/DN clone 4 or (B) C26, C26/DN clone 17, C26/12, and C26/12/DN clone 5. Cells were either unstimulated (−) or stimulated (+) with increasing amounts of rIFN-γ (200 units/ml, 1000 units/ml, and 2000 units/ml) for 30 min. C, nuclear extracts were incubated in the absence (no Ab) or presence of affinity-purified Abs recognizing STAT1α or STAT4 before addition of the GAS probe. In the competition lane (Comp 100X), binding was carried out in the presence of a 100-fold molar excess of unlabeled GAS probe.

Fig. 2.

Inhibition of STAT1α activation in IFN-γRDN-transduced TSA and C26 cells. EMSA was performed with a GAS probe, and nuclear extracts were isolated from (A) TSA, TSA/DN clone 12, TSA/12, and TSA/12/DN clone 4 or (B) C26, C26/DN clone 17, C26/12, and C26/12/DN clone 5. Cells were either unstimulated (−) or stimulated (+) with increasing amounts of rIFN-γ (200 units/ml, 1000 units/ml, and 2000 units/ml) for 30 min. C, nuclear extracts were incubated in the absence (no Ab) or presence of affinity-purified Abs recognizing STAT1α or STAT4 before addition of the GAS probe. In the competition lane (Comp 100X), binding was carried out in the presence of a 100-fold molar excess of unlabeled GAS probe.

Close modal
Fig. 3.

Class I MHC expression in IFN-γRDN-transduced TSA and C26 cells. Cells were examined for class I MHC expression by flow cytometry using an anti-H-2Ld mAb after treatment with 200 units/ml rIFN-γ (A and B) or murine IFN-α (C). Open and filled histograms represent the level of classes I MHC in cells incubated with medium alone or after IFN stimulation, respectively.

Fig. 3.

Class I MHC expression in IFN-γRDN-transduced TSA and C26 cells. Cells were examined for class I MHC expression by flow cytometry using an anti-H-2Ld mAb after treatment with 200 units/ml rIFN-γ (A and B) or murine IFN-α (C). Open and filled histograms represent the level of classes I MHC in cells incubated with medium alone or after IFN stimulation, respectively.

Close modal
Fig. 4.

Tumorigenicity of IFN-γ-responsive or -unresponsive TSA and C26 cells secreting IL-12 or nonsecreting. BALB/c mice were injected s.c. into the left flank with 105 cells of: (A) TSA or TSA/DN clone 12, (B) C26 or C26/DN clone 17 (C) TSA/12 or TSA/12/DN clone 4, (D) C26/12 or C26/12/DN clone 5 (19 mice/group).

Fig. 4.

Tumorigenicity of IFN-γ-responsive or -unresponsive TSA and C26 cells secreting IL-12 or nonsecreting. BALB/c mice were injected s.c. into the left flank with 105 cells of: (A) TSA or TSA/DN clone 12, (B) C26 or C26/DN clone 17 (C) TSA/12 or TSA/12/DN clone 4, (D) C26/12 or C26/12/DN clone 5 (19 mice/group).

Close modal
Fig. 5.

Tumorigenicity of IFN-γ-responsive and -unresponsive TSA/12 and C26/12 cells in the absence of endogenous IFN-γ. GKO mice were injected s.c. into the left flank with 105 cells of: (A) TSA/12 or TSA/12/DN clone 4, or (B) C26/12 or C26/12/DN clone 5 (10 mice/group).

Fig. 5.

Tumorigenicity of IFN-γ-responsive and -unresponsive TSA/12 and C26/12 cells in the absence of endogenous IFN-γ. GKO mice were injected s.c. into the left flank with 105 cells of: (A) TSA/12 or TSA/12/DN clone 4, or (B) C26/12 or C26/12/DN clone 5 (10 mice/group).

Close modal
Fig. 6.

Immunogenicity of TSA/12 and C26/12 cells. TSA, TSA/12, C26, and C26/12 were analyzed for membrane expression of (A) TAA gp70 and (B) class I MHC H2-Ld. C, NK-mediated cytotoxicity of SCID mice splenocytes (44% NK cells) stimulated with IL-2 alone or together with IL-12 against YAC (•), TSA (▪), and C26 (▴) target cells. D–F, CTL activity after in vivo priming of BALB/c (filled symbols) and GKO mice (open symbols) with irradiated TSA/12 (▪, □) or C26/12 cells (▴, ▵). D, target cells were AH1-pulsed BALB/c blasts. E and F, target cells were C26 (▴, ▵) or TSA (▪, □). E:T is indicated.

Fig. 6.

Immunogenicity of TSA/12 and C26/12 cells. TSA, TSA/12, C26, and C26/12 were analyzed for membrane expression of (A) TAA gp70 and (B) class I MHC H2-Ld. C, NK-mediated cytotoxicity of SCID mice splenocytes (44% NK cells) stimulated with IL-2 alone or together with IL-12 against YAC (•), TSA (▪), and C26 (▴) target cells. D–F, CTL activity after in vivo priming of BALB/c (filled symbols) and GKO mice (open symbols) with irradiated TSA/12 (▪, □) or C26/12 cells (▴, ▵). D, target cells were AH1-pulsed BALB/c blasts. E and F, target cells were C26 (▴, ▵) or TSA (▪, □). E:T is indicated.

Close modal
Fig. 7.

OX40 and IL-15 expression by 7-day infiltrating lymphocytes from TSA/12 or C26/12 tumors. Top panel, staining of tumor sections from BALB/c (A and B) and GKO (C and D) mice inoculated with C26/12 (B and D) or TSA/12 (A and C) revealed OX40-positive cells in both mouse strains injected with C26/12 but not in GKO mice bearing TSA/12 tumor. Bottom panel, in situ hybridization of 7-day tumor section from TSA/12- (E and G) or C26/12- (F and H) injected BALB/c (E and F) or GKO (G and H) mice, using biotinylated IL-15 probe. Several macrophage-like cells stained positive in C26/12 tumors from BALB/c (F) or GKO (H) mice, whereas such cells were sparse in TSA/12-induced-tumors from BALB/c mice (E) and completely absent in TSA/12 tumors from GKO mice (G). Frozen sections were counterstained with hematoxylin. Original magnification: A–H, ×400; inset in H, ×600.

Fig. 7.

OX40 and IL-15 expression by 7-day infiltrating lymphocytes from TSA/12 or C26/12 tumors. Top panel, staining of tumor sections from BALB/c (A and B) and GKO (C and D) mice inoculated with C26/12 (B and D) or TSA/12 (A and C) revealed OX40-positive cells in both mouse strains injected with C26/12 but not in GKO mice bearing TSA/12 tumor. Bottom panel, in situ hybridization of 7-day tumor section from TSA/12- (E and G) or C26/12- (F and H) injected BALB/c (E and F) or GKO (G and H) mice, using biotinylated IL-15 probe. Several macrophage-like cells stained positive in C26/12 tumors from BALB/c (F) or GKO (H) mice, whereas such cells were sparse in TSA/12-induced-tumors from BALB/c mice (E) and completely absent in TSA/12 tumors from GKO mice (G). Frozen sections were counterstained with hematoxylin. Original magnification: A–H, ×400; inset in H, ×600.

Close modal
Fig. 8.

sIL-15Rα inhibited the C26/12 tumor rejection in GKO mice. GKO mice were injected s. c. with 5 × 105 C26/12 cells and were left untreated (n = 10) or given 22 daily i. p. injections of 40 μg of sIL-15Rα or 40 μg of M4, a functionally mutated sIL-15Rα unable to bind IL-15, starting 2 days before tumor inoculation (n = 7).

Fig. 8.

sIL-15Rα inhibited the C26/12 tumor rejection in GKO mice. GKO mice were injected s. c. with 5 × 105 C26/12 cells and were left untreated (n = 10) or given 22 daily i. p. injections of 40 μg of sIL-15Rα or 40 μg of M4, a functionally mutated sIL-15Rα unable to bind IL-15, starting 2 days before tumor inoculation (n = 7).

Close modal

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.

1

Supported by Associazione Italiana Ricerca sul Cancro, ISS-ITA-USA n. T00.A15, Consiglio Nazionale delle Ricerche Finalized Project on Biotechnology (PF49). G. G. was supported by an Italian Foundation for Cancer Research fellowship.

3

The abbreviations used are: IL, interleukin; NK, natural killer; TAA, tumor-associated antigen; IFN-γR, functional IFN-γ receptor; FACS, fluorescence-activated cell sorter; IFN-γRDN, dominant-negative truncated IFN-γ receptor; Ab, antibody; sIL-15Rα, soluble fragment of the murine IL-15Rα; GAS, γ-activated sites; r, recombinant; GKO, IFN-γ knockout mice; GM-CSF, granulocyte-macrophage colony-stimulating factor; EMSA, electrophoretic mobility shift assay; SCID, severe combined immunodeficiency; JAK, Janus-activated kinase; STAT, signal transducers and activators of transcription; mAb, monoclonal antibody; DN, dominant-negative; TCR, T-cell receptor.

4

A. Comes, E. Di Carlo, P. Musiani, O. Rosso, R. Meazza, C. Chiodoni, M. P. Colombo, and S. Ferrini. IFN-γ independent synergistic effect of IL-12 and IL-15 induces antitumor responses in syngeneic mice, submitted for publication, 2001.

Table 1

Immunohistochemical analysis of area at different time points after TSA/12 and C26/12 cells challenge of BALB/c or GKO mice

TumorDayCD4aCD8Vβ6Vβ8.3OX40NK cellsIL-15
C26/12 in BALB/c +1 NDb ND ND 
 +3 84 ± 15 73 ± 8 9 ± 3 17 ± 5 52 ± 11 31 ± 3 14 ± 2 
 +7 161 ± 20 173 ± 11 70 ± 5 13 ± 2 133 ± 27 19 ± 3 22 ± 5 
C26/12 in GKO +1 2 ± 3 <1 ND ND <1 
 +3 56 ± 8 14 ± 3 30 ± 5 15 ± 6 59 ± 13 3 ± 1 7 ± 1 
 +7 197 ± 19 56 ± 9 52 ± 8 14 ± 8 159 ± 21 2 ± 1 18 ± 3 
TSA/12 in BALB/c +1 ND ND ND <1 
 +3 85 ± 3 28 ± 5 <1 55 ± 8c 10 ± 3 10 ± 3 7 ± 2 
 +7 385 ± 13 141 ± 16 28 ± 6 267 ± 21c 95 ± 15 14 ± 2 17 ± 7 
TSA/12 in GKO +1 ND ND ND 
 +3 100 ± 16 22 ± 7 0d 44 ± 5d 4 ± 1d 6 ± 3 <1d 
 +7 370 ± 15 85 ± 18 1 ± 1d 257 ± 24d 17 ± 4d <1 <1d 
TumorDayCD4aCD8Vβ6Vβ8.3OX40NK cellsIL-15
C26/12 in BALB/c +1 NDb ND ND 
 +3 84 ± 15 73 ± 8 9 ± 3 17 ± 5 52 ± 11 31 ± 3 14 ± 2 
 +7 161 ± 20 173 ± 11 70 ± 5 13 ± 2 133 ± 27 19 ± 3 22 ± 5 
C26/12 in GKO +1 2 ± 3 <1 ND ND <1 
 +3 56 ± 8 14 ± 3 30 ± 5 15 ± 6 59 ± 13 3 ± 1 7 ± 1 
 +7 197 ± 19 56 ± 9 52 ± 8 14 ± 8 159 ± 21 2 ± 1 18 ± 3 
TSA/12 in BALB/c +1 ND ND ND <1 
 +3 85 ± 3 28 ± 5 <1 55 ± 8c 10 ± 3 10 ± 3 7 ± 2 
 +7 385 ± 13 141 ± 16 28 ± 6 267 ± 21c 95 ± 15 14 ± 2 17 ± 7 
TSA/12 in GKO +1 ND ND ND 
 +3 100 ± 16 22 ± 7 0d 44 ± 5d 4 ± 1d 6 ± 3 <1d 
 +7 370 ± 15 85 ± 18 1 ± 1d 257 ± 24d 17 ± 4d <1 <1d 
a

Cryostat sections were stained with mAb against CD4, CD8, NK, TCR Vβ6, TCR Vβ8.3, and OX40 using streptoavidin-biotin horseradish peroxidase immunohistochemistry or by in situ hybridization with IL-15 biotinilated probe. Cell counts were performed at ×400 enlargement with the help of a 1-mm2 square greed. Results are reported as the mean ± SD of positive cells/mm2.

b

ND, not determined.

c

Values significantly different (P < 0.02) from corresponding values in C26/12 tumor growing in BALB/c mice.

d

Values significantly different (P < 0.02) from corresponding values in C26/12 tumor growing in GKO mice.

We thank Mariella Parenza and Ivano Arioli for technical support.

1
Brunda M. J., Luistro L., Warrier R. R., Wright R. B., Hubbard B. R., Murphy M., Wolf S. F., Gately M. K. Antitumor and antimetastatic activity of interleukin 12 against murine tumors.
J. Exp. Med.
,
178
:
1223
-1230,  
1993
.
2
Nastala C. L., Edington H. D., McKinney T. G., Tahara H., Nalesnik M. A., Brunda M. J., Gately M. K., Wolf S. F., Schreiber R. D., Storkus W. J., et al Recombinant IL-12 administration induces tumor regression in association with IFN-γ production.
J. Immunol.
,
153
:
1697
-706,  
1994
.
3
Verbik D. J., Stinson W. W., Brunda M. J., Kessinger A., Joshi S. S. In vivo therapeutic effects of interleukin-12 against highly metastatic residual lymphoma.
Clin. Exp. Metastasis
,
14
:
219
-229,  
1996
.
4
Zou J. P., Yamamoto N., Fujii T., Takenaka H., Kobayashi M., Herrmann S. H., Wolf S. F., Fujiwara H., Hamaoka T. Systemic administration of rIL-12 induces complete tumor regression and protective immunity: response is correlated with a striking reversal of suppressed IFN-γ production by anti-tumor T cells.
Int. Immunol.
,
7
:
1135
-1145,  
1995
.
5
Strasly M., Cavallo F., Geuna M., Mitola S., Colombo M. P., Forni G., Bussolino F. IL-12 inhibition of endothelial cell functions and angiogenesis depends on lymphocyte-endothelial cell cross-talk.
J. Immunol.
,
166
:
3890
-3899,  
2001
.
6
Trinchieri G. Interleukin-12: a proinflammatory cytokine with immunoregulatory functions that bridge innate resistance and antigen-specific adaptive immunity.
Annu. Rev. Immunol.
,
13
:
251
-276,  
1995
.
7
Brunda M. J., Luistro L., Hendrzak J. A., Fountoulakis M., Garotta G., Gately M. K. Role of interferon-γ in mediating the antitumor efficacy of interleukin-12.
J. Immunother. Emphas. Tumor Immunol.
,
17
:
71
-77,  
1995
.
8
Wigginton J. M., Kuhns D. B., Back T. C., Brunda M. J., Wiltrout R. H., Cox G. W. Interleukin 12 primes macrophages for nitric oxide production in vivo and restores depressed nitric oxide production by macrophages from tumor-bearing mice: implications for the antitumor activity of interleukin 12 and/or interleukin 2.
Cancer Res.
,
56
:
1131
-1136,  
1996
.
9
Yu W. G., Yamamoto N., Takenaka H., Mu J., Tai X. G., Zou J. P., Ogawa M., Tsutsui T., Wijesuriya R., Yoshida R., Herrmann S., Fujiwara H., Hamaoka T. Molecular mechanisms underlying IFN-γ-mediated tumor growth inhibition induced during tumor immunotherapy with rIL-12.
Int. Immunol.
,
8
:
855
-865,  
1996
.
10
Angiolillo A. L., Sgadari C., Taub D. D., Liao F., Farber J. M., Maheshwari S., Kleinman H. K., Reaman G. H., Tosato G. Human interferon-inducible protein 10 is a potent inhibitor of angiogenesis in vivo.
J. Exp. Med.
,
182
:
155
-162,  
1995
.
11
Luster A. D., Unkeless J. C., Ravetch J. V. γ-interferon transcriptionally regulates an early-response gene containing homology to platelet proteins.
Nature (Lond.)
,
315
:
672
-676,  
1985
.
12
Amichay D., Gazzinelli R. T., Karupiah G., Moench T. R., Sher A., Farber J. M. Genes for chemokines MuMig and Crg-2 are induced in protozoan and viral infections in response to IFN-γ with patterns of tissue expression that suggest nonredundant roles in vivo.
J. Immunol.
,
157
:
4511
-4520,  
1996
.
13
Tannenbaum C. S., Wicker N., Armstrong D., Tubbs R., Finke J., Bukowski R. M., Hamilton T. A. Cytokine and chemokine expression in tumors of mice receiving systemic therapy with IL-12.
J. Immunol.
,
156
:
693
-699,  
1996
.
14
Sgadari C., Farber J. M., Angiolillo A. L., Liao F., Teruya-Feldstein J., Burd P. R., Yao L., Gupta G., Kanegane C., Tosato G. Mig, the monokine induced by interferon-γ, promotes tumor necrosis in vivo.
Blood
,
89
:
2635
-2643,  
1997
.
15
Sgadari C., Angiolillo A. L., Tosato G. Inhibition of angiogenesis by interleukin-12 is mediated by the interferon-inducible protein 10.
Blood
,
87
:
3877
-3882,  
1996
.
16
Kanegane C., Sgadari C., Kanegane H., Teruya-Feldstein J., Yao L., Gupta G., Farber J. M., Liao F., Liu L., Tosato G. Contribution of the CXC chemokines IP-10 and Mig to the antitumor effects of IL-12.
J. Leukoc. Biol.
,
64
:
384
-392,  
1998
.
17
Boehm U., Klamp T., Groot M., Howard J. C. Cellular responses to interferon-γ.
Annu. Rev. Immunol.
,
15
:
749
-795,  
1997
.
18
Kotenko S. V., Izotova L. S., Pollack B. P., Mariano T. M., Donnelly R. J., Muthukumaran G., Cook J. R., Garotta G., Silvennoinen O., Ihle J. N., et al Interaction between the components of the interferon γ receptor complex.
J. Biol. Chem.
,
270
:
20915
-20921,  
1995
.
19
Kotenko S. V., Izotova L. S., Pollack B. P., Muthukumaran G., Paukku K., Silvennoinen O., Ihle J. N., Pestka S. Other kinases can substitute for Jak2 in signal transduction by interferon-γ.
J. Biol. Chem.
,
271
:
17174
-17182,  
1996
.
20
Pearse R. N., Feinman R., Shuai K., Darnell J. E., Jr, Ravetch J. V. Interferon γ-induced transcription of the high-affinity Fc receptor for IgG requires assembly of a complex that includes the 91-kDa subunit of transcription factor ISGF3.
Proc. Natl. Acad. Sci. USA
,
90
:
4314
-4318,  
1993
.
21
Darnell J. E., Jr., Kerr I. M., Stark G. R. Jak-STAT pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins.
Science (Wash. DC)
,
264
:
1415
-1421,  
1994
.
22
Presky D. H., Gubler U., Chizzonite R. A., Gately M. K. IL12 receptors and receptor antagonists.
Res. Immunol.
,
146
:
439
-445,  
1995
.
23
Zilocchi C., Stoppacciaro A., Chiodoni C., Parenza M., Terrazzini N., Colombo M. P. Interferon γ-independent rejection of interleukin 12-transduced carcinoma cells requires CD4+ T cells and granulocyte/macrophage colony-stimulating factor.
J. Exp. Med.
,
188
:
133
-143,  
1998
.
24
Westlin W. F., Gimbrone M. A., Jr. Neutrophil-mediated damage to human vascular endothelium. Role of cytokine activation.
Am. J. Pathol.
,
142
:
117
-128,  
1993
.
25
Shankaran V., Ikeda H., Bruce A. T., White J. M., Swanson P. E., Old L. J., Schreiber R. D. IFNγ and lymphocytes prevent primary tumour development and shape tumour immunogenicity.
Nature (Lond.)
,
410
:
1107
-1111,  
2001
.
26
Kaplan D. H., Shankaran V., Dighe A. S., Stockert E., Aguet M., Old L. J., Schreiber R. D. Demonstration of an interferon γ-dependent tumor surveillance system in immunocompetent mice.
Proc. Natl. Acad. Sci. USA
,
95
:
7556
-7561,  
1998
.
27
Dighe A. S., Richards E., Old L. J., Schreiber R. D. Enhanced in vivo growth and resistance to rejection of tumor cells expressing dominant negative IFN γ receptors.
Immunity
,
1
:
447
-456,  
1994
.
28
Corbett T. H., Griswold D. P., Jr, Roberts B. J., Peckham J. C., Schabel F. M., Jr. Tumor induction relationships in development of transplantable cancers of the colon in mice for chemotherapy assays, with a note on carcinogen structure.
Cancer Res.
,
35
:
2434
-2439,  
1975
.
29
Nanni P., de Giovanni C., Lollini P. L., Nicoletti G., Prodi G. TS/A: a new metastasizing cell line from a BALB/c spontaneous mammary adenocarcinoma.
Clin. Exp. Metastasis
,
1
:
373
-380,  
1983
.
30
Dalton D. K., Pitts-Meek S., Keshav S., Figari I. S., Bradley A., Stewart T. A. Multiple defects of immune cell function in mice with disrupted interferon-γ genes.
Science (Wash. DC)
,
259
:
1739
-1742,  
1993
.
31
Ruchatz H., Leung B. P., Wei X. Q., McInnes I. B., Liew F. Y. Soluble IL-15 receptor α-chain administration prevents murine collagen-induced arthritis: a role for IL-15 in development of antigen-induced immunopathology.
J. Immunol.
,
160
:
5654
-5660,  
1998
.
32
Wei X., Orchardson M., Gracie J. A., Leung B. P., Gao B., Guan H., Niedbala W., Paterson G. K., McInnes I. B., Liew F. Y. The Sushi domain of soluble IL-15 receptor α is essential for binding IL-15 and inhibiting inflammatory and allogenic responses in vitro and in vivo.
J. Immunol.
,
167
:
277
-282,  
2001
.
33
Miller A. D., Rosman G. J. Improved retroviral vectors for gene transfer and expression.
Biotechniques
,
7
:
980
-990,  
1989
.
34
Markowitz D., Goff S., Bank A. Construction and use of a safe and efficient amphotropic packaging cell line.
Virology
,
167
:
400
-406,  
1988
.
35
Colombo M. P., Ferrari G., Stoppacciaro A., Parenza M., Rodolfo M., Mavilio F., Parmiani G. Granulocyte colony-stimulating factor gene transfer suppresses tumorigenicity of a murine adenocarcinoma in vivo.
J. Exp. Med.
,
173
:
889
-897,  
1991
.
36
Cavallo F., Signorelli P., Giovarelli M., Musiani P., Modesti A., Brunda M. J., Colombo M. P., Forni G. Antitumor efficacy of adenocarcinoma cells engineered to produce interleukin 12 (IL-12) or other cytokines compared with exogenous IL-12.
J. Natl. Cancer Inst.
,
89
:
1049
-1058,  
1997
.
37
Colombo M. P., Vagliani M., Spreafico F., Parenza M., Chiodoni C., Melani C., Stoppacciaro A. Amount of interleukin 12 available at the tumor site is critical for tumor regression.
Cancer Res.
,
56
:
2531
-2534,  
1996
.
38
Schreiber E., Matthias P., Muller M. M., Schaffner W. Rapid detection of octamer binding proteins with “mini-extracts,” prepared from a small number of cells.
Nucleic Acids Res.
,
17
:
6419
1989
.
39
Decker T., Kovarik P., Meinke A. GAS elements: a few nucleotides with a major impact on cytokine-induced gene expression.
J. Interferon Cytokine Res.
,
17
:
121
-134,  
1997
.
40
Ramana C. V., Gil M. P., Han Y., Ransohoff R. M., Schreiber R. D., Stark G. R. Stat1-independent regulation of gene expression in response to IFN-γ.
Proc. Natl. Acad. Sci. USA
,
98
:
6674
-6679,  
2001
.
41
Gil M. P., Bohn E., O’Guin A. K., Ramana C. V., Levine B., Stark G. R., Virgin H. W., Schreiber R. D. Biologic consequences of Stat1-independent IFN signaling.
Proc. Natl. Acad. Sci. USA
,
98
:
6680
-6685,  
2001
.
42
Huang A. Y., Gulden P. H., Woods A. S., Thomas M. C., Tong C. D., Wang W., Engelhard V. H., Pasternack G., Cotter R., Hunt D., Pardoll D. M., Jaffee E. M. The immunodominant major histocompatibility complex class I-restricted antigen of a murine colon tumor derives from an endogenous retroviral gene product.
Proc. Natl. Acad. Sci. USA
,
93
:
9730
-9735,  
1996
.
43
Rodolfo M., Castelli C., Bassi C., Accornero P., Sensi M., Parmiani G. Cytotoxic T lymphocytes recognize tumor antigens of a murine colonic carcinoma by using different T-cell receptors.
Int. J. Cancer
,
57
:
440
-447,  
1994
.
44
al-Shamkhani A., Birkeland M. L., Puklavec M., Brown M. H., James W., Barclay A. N. OX40 is differentially expressed on activated rat and mouse T cells and is the sole receptor for the OX40 ligand.
Eur. J. Immunol.
,
26
:
1695
-1699,  
1996
.
45
Lasek W., Golab J., Maslinski W., Switaj T., Balkowiec E. Z., Stoklosa T., Giermasz A., Malejczyk M., Jakobisiak M. Subtherapeutic doses of interleukin-15 augment the antitumor effect of interleukin-12 in a B16F10 melanoma model in mice.
Eur. Cytokine Netw.
,
10
:
345
-356,  
1999
.
46
Kimura K., Nishimura H., Matsuzaki T., Yokokura T., Nimura Y., Yoshikai Y. Synergistic effect of interleukin-15 and interleukin-12 on antitumor activity in a murine malignant pleurisy model.
Cancer Immunol. Immunother.
,
49
:
71
-77,  
2000
.
47
Di Carlo E., Comes A., Basso S., De Ambrosis A., Meazza R., Musiani P., Moelling K., Albini A., Ferrini S. The combined action of IL-15 and IL-12 gene transfer can induce tumor cell rejection without T and NK cell involvement.
J. Immunol.
,
165
:
3111
-3118,  
2000
.
48
Lodolce J. P., Boone D. L., Chai S., Swain R. E., Dassopoulos T., Trettin S., Ma A. IL-15 receptor maintains lymphoid homeostasis by supporting lymphocyte homing and proliferation.
Immunity
,
9
:
669
-676,  
1998
.
49
Kennedy M. K., Glaccum M., Brown S. N., Butz E. A., Viney J. L., Embers M., Matsuki N., Charrier K., Sedger L., Willis C. R., Brasel K., Morrissey P. J., Stocking K., Schuh J. C., Joyce S., Peschon J. J. Reversible defects in natural killer and memory CD8 T cell lineages in interleukin 15-deficient mice.
J. Exp. Med.
,
191
:
771
-780,  
2000
.
50
Vagliani M., Rodolfo M., Cavallo F., Parenza M., Melani C., Parmiani G., Forni G., Colombo M. P. Interleukin 12 potentiates the curative effect of a vaccine based on interleukin 2-transduced tumor cells.
Cancer Res.
,
56
:
467
-470,  
1996
.
51
Chiodoni C., Stoppacciaro A., Sangaletti S., Gri G., Cappetti B., Koezuka Y., Colombo M. P. Different requirements for α-galactosylceramide and recombinant IL-12 antitumor activity in the treatment of C-26 colon carcinoma hepatic metastases.
Eur. J. Immunol.
,
31
:
3101
-3110,  
2001
.
52
Coughlin C. M., Salhany K. E., Gee M. S., LaTemple D. C., Kotenko S., Ma X., Gri G., Wysocka M., Kim J. E., Liu L., Liao F., Farber J. M., Pestka S., Trinchieri G., Lee W. M. Tumor cell responses to IFNγ affect tumorigenicity and response to IL-12 therapy and antiangiogenesis.
Immunity
,
9
:
25
-34,  
1998
.
53
Mumberg D., Monach P. A., Wanderling S., Philip M., Toledano A. Y., Schreiber R. D., Schreiber H. CD4(+) T cells eliminate MHC class II-negative cancer cells in vivo by indirect effects of IFN-γ.
Proc. Natl. Acad. Sci. USA
,
96
:
8633
-8638,  
1999
.
54
Qin Z., Blankenstein T. CD4+ T cell-mediated tumor rejection involves inhibition of angiogenesis that is dependent on IFN γ receptor expression by nonhematopoietic cells.
Immunity
,
12
:
677
-686,  
2000
.
55
Beatty G. L., Paterson Y. IFN can promote tumor evasion of the immune system in vivo by down-regulating cellular levels of an endogenous tumor antigen.
J. Immunol.
,
165
:
5502
-5508,  
2000
.
56
Imboden M., Murphy K. R., Rakhmilevich A. L., Neal Z. C., Xiang R., Reisfeld R. A., Gillies S. D., Sondel P. M. The level of MHC class I expression on murine adenocarcinoma can change the antitumor effector mechanism of immunocytokine therapy.
Cancer Res.
,
61
:
1500
-1507,  
2001
.
57
Fehniger T. A., Caligiuri M. A. Interleukin 15: biology and relevance to human disease.
Blood
,
97
:
14
-32,  
2001
.