CTL Adoptive Immunotherapy Concurrently Mediates Tumor Regression and Tumor Escape

Kebin Liu, Sheila A. Caldwell, Kristy M. Greeneltch, Dafeng Yang and Scott I. Abrams
J Immunol March 15, 2006, 176 (6) 3374-3382; DOI: https://doi.org/10.4049/jimmunol.176.6.3374

Abstract

Tumor escape and recurrence are major impediments for successful immunotherapy. It is well-documented that the emergence of Ag-loss variants, as well as regulatory mechanisms suppressing T cell function, have been linked to inadequate antitumor activity. However, little is known regarding the role of Fas-mediated cytotoxicity by tumor-specific CD8+ CTL in causing immune evasion of Fas resistant variants during adoptive immunotherapy. In this study, we made use of an adoptive transfer model of experimental lung metastasis using tumor-specific CTL as a relevant immune-based selective pressure, and wherein the Fas ligand pathway was involved in the antitumor response. Surviving tumor cells were recovered and examined for alterations in antigenic, functional, and biologic properties. We showed that diminished susceptibility to Fas-mediated cytotoxicity in vivo was an important determinant of tumor escape following CTL-based immunotherapy. Tumor escape variants (TEV) recovered from the lungs of CTL-treated mice exhibited more aggressive behavior in vivo. However, these TEV retained relevant MHC class I and tumor Ag expression and sensitivity to CTL via the perforin pathway but reduced susceptibility to Fas-mediated lysis. Moreover, TEV were significantly less responsive to eradication by CTL adoptive immunotherapy paradigms as a consequence of increased Fas resistance. Overall, we identified that Faslow-TEV emerged as a direct consequence of CTL-tumor interactions in vivo, and that such an altered neoplastic Fas phenotype compromised immunotherapy efficacy. Together, these findings may have important implications for both tumor progression and the design of immunotherapeutic interventions to confront these selective pressures or escape mechanisms.

Tumor escape and recurrence are major challenges confronting effective immunotherapy of cancer (1, 2, 3, 4, 5, 6). Therefore, understanding the mechanisms of tumor escape and recurrence may offer important insights into the development or improvement of anti-neoplastic T cell-mediated therapies. Although a variety of molecular and biologic alterations have been observed in tumors as they become more aggressive, it remains to be fully understood how such changes are thought to occur initially. It is conceivable that immunologically driven events, in addition to genetic and epigenetic determinants, contribute to the generation of tumor escape variants (TEV)3 expressing those phenotypes. This type of selection pressure is consistent with the recent concept of “cancer immunoediting” (7). The basis of immunoediting stems from the observations that chemically induced tumors derived from immune-deficient mice were much less tumorigenic, compared with tumors similarly derived from immune-competent mice, when subsequently transplanted into naive wild-type (wt) mice (8). These findings suggested that tumors produced endogenously in wt mice underwent an immune-based selective pressure.

Thus, immunoselection has been proposed as a consequence of immune cell-tumor cell interactions, resulting in the emergence of TEV. This outcome may occur following intrinsic immune responses as just pointed out, as well as to experimental interventions involving immunotherapy. In such studies, TEV have been reported to lose expression of MHC molecules, Ag determinants, and/or other capabilities required for meaningful Ag recognition (9, 10, 11, 12, 13, 14, 15, 16, 17). However, what remains to be further understood is the underlying basis of TEV that arise in response to immunotherapy that exhibit no significant loss of MHC class I (MHC I) and Ag expression (2, 4, 18, 19, 20, 21, 22). Suppression of CTL function in the tumor microenvironment has been proposed as a mechanism for the outgrowth of such TEV (19, 20, 21, 22).

A number of studies have now indicated that down-regulation of Fas expression or loss of Fas function has been correlated with tumor progression in both mouse (23, 24, 25, 26, 27) and human (28, 29, 30, 31, 32, 33) systems. Furthermore, studies in mouse models unveiled that the Fas ligand (FasL) pathway was important for optimal tumor regression via adoptive CTL transfer (34). Consequently, the loss of Fas expression or function, which ordinarily serves to induce apoptotic cell death (35, 36), may provide a selective survival advantage for certain neoplastic subpopulations within the parental or primary tumor population resulting from immune attack. The unresolved nature or basis of MHC/Ag-competent TEV, along with the observation that down-regulation of Fas occurs in neoplastic progression, led us to directly determine whether adoptive CTL transfer can mediate the emergence of TEV in vivo with altered sensitivity to Fas-mediated death and whether these TEV, in turn, are more refractory to immunotherapy as a consequence of increased Fas resistance.

Materials and Methods

Mice

Female BALB/c (H-2d) mice were obtained from the National Cancer Institute/Frederick Cancer Research Animal Facility (Frederick, MD). Female FasL-deficient CPt.C3-Tnfsf6gld mice on a BALB/c background (henceforth termed gld) were obtained from The Jackson Laboratory. Female, perforin-deficient (pfp) mice on a BALB/c background were kindly provided by M. Smyth (Peter MacCallum Cancer Institute, East Melbourne, Australia) via R. Wiltrout (Laboratory of Experimental Immunology, Center for Cancer Research, National Cancer Institute). All animal studies were approved by the National Institutes of Health Animal Care and Use Committee.

Tumor cells

The CMS4 sarcoma cell line was provided by A. DeLeo (University of Pittsburgh, Pittsburgh, PA). A CMS4 subline, with aggressive ability to grow in the lungs, was produced from the parental population by one in vivo passage in the lungs of normal BALB/c mice, as described (37). Briefly, CMS4 cells (1.5–2.5 × 105 cells per mouse) were injected i.v. into the lateral tail vein. Mice were sacrificed 14 days later, and lungs were removed and digested for 4–6 h at room temperature with an enzyme mixture containing hyaluronidase (0.1 mg/ml), collagenase (1 mg/ml), and DNase I (30 U/ml), all obtained from Sigma-Aldrich. Tumor cells that outgrew from these lung digests, termed CMS4-met, were then maintained in culture. The P815 mastocytoma cell line was obtained from the American Type Culture Collection.

Production of tumor-reactive CD8+ CTL populations

CD8+ CTL lines reactive against the parental CMS4 population were established from normal BALB/c mice, BALB/c-gld mice, or BALB/c-pfp mice using an immunization strategy consisting of a viable tumor challenge (5 × 105 cells given s.c. on one flank) coadministered with anti-CTLA-4 mAb (affinity-purified hamster anti-mouse clone UC10-4F10-11, hybridoma line provided by J. Bluestone (University of California, San Francisco, CA) at 100 μg/inoculation/mouse given i.p. on days 0, 3, and 6 posttumor transplant in a manner similar to that described in Ref.37 . Mice that were exposed to this regimen and failed to display evidence of primary tumor growth were rechallenged on the contralateral flank (given in the absence of anti-CTLA-4 mAb). Splenic-derived CD8+ CTL lines were derived from rechallenged mice, which showed little to no additional tumor growth. CTL cultures (1–2 × 105 per well) were propagated in vitro in 24-well plates (Costar) by weekly stimulation with irradiated (20 Gy) syngeneic BALB/c splenocytes (5 × 106 per well) as APC, irradiated (200 Gy) CMS4 tumor cells (1 × 105/well) as a source of cognate Ag and IL-2 (60 IU/ml, Tecin; Hoffman-LaRoche).

Cell surface marker analysis

Tumor cells were stained by direct or indirect immunofluorescence with anti-Fas mAb (clone Jo2) or H-2Ld-reactive mAb (both from BD Pharmingen) or the appropriate isotype-matched Ab and analyzed by flow cytometry using a FACS Calibur (BD Biosciences). Evaluation of TCR binding of MHC/Ag by wt-CTL and pfp-CTL was performed using a PE-labeled fluorescent Pro5 MHC I (H-2Ld) pentamer complexed with the gp70 epitope peptide, SPSYVYHQF (H-2Ld/gp70423–431; ProImmune). Previously, CMS4-reactive CTL were shown to recognize this particular MHC/Ag complex (37). Briefly, CTL were recovered over a Ficoll-Hypaque gradient (MP Biomedicals) 4–5 days after Ag stimulation, washed, and incubated with 0.25 μg of pentamer (per 1 × 106 cells) for 2 h at 4°C, as described by the manufacturer. An influenza virus-specific CTL line (37) was used as a negative control. Cells were fixed with 2% paraformaldehyde and analyzed by flow cytometry using a FACSCalibur.

Cytotoxicity assays

CTL activity was assessed by 51Cr release assays. Target cells were labeled with 250 μCi Na251CrO4 (Amersham Biosciences). CTL were recovered from culture by centrifugation over a Ficoll-Hypaque gradient and coincubated with radiolabeled target cells in 96-well, U-bottom plates (Costar) at various E:T ratios. After incubation for 4 or 18 h as indicated in Results, supernatants were collected using a supernatant collection system (Molecular Devices). Radioactivity was quantitated using a gamma counter. The percentage of specific 51Cr release was calculated according to the following formula: percent specific lysis = ((experimental cpm-spontaneous cpm)/(total cpm-spontaneous cpm)) × 100%. Total 51Cr release was obtained by adding 0.2% Triton X-100 (final concentration) to the wells. Data are reported as the mean ± SD of triplicate wells and are representative of four independent experiments.

Measurement of Fas-mediated cell death

Cell death was measured by propidium iodide staining, as described (26, 27). Briefly, CMS4 sublines were either untreated or pretreated with recombinant mouse IFN-γ and TNF-α (each cytokine used at 100 U/ml; R&D Systems) overnight (24 h), followed by culture with different concentrations of recombinant human soluble FasL (sFasL) (PeproTech). Cells were collected after 20–24 h incubation and then stained with propidium iodide according to the manufacturer’s instructions (R&D Systems). After staining, the cells were washed and immediately analyzed by flow cytometry.

RT-PCR and quantitative real-time PCR analysis

Total RNA was isolated from tumor cells or the lungs of tumor-bearing mice using RNA STAT-60 reagent (Tel-Test) according to manufacturer’s instructions, and used for first-strand cDNA synthesis using the ThermoScript RT-PCR system (Invitrogen Life Technologies). The cDNA was then used as template for PCR amplification of mouse Fas, mouse β-actin, or gp70 mRNA. The following parameters were used: 30 s at 94°C, 30 s at 60°C, and 1 min at 72°C for various cycles: Fas, 30 cycles; β-actin, 26 cycles; and gp70, 24 cycles. The PCR primers for mouse Fas were as follows: forward, 5′-ATGCTGTGGATCTGGGCT-3′; and reverse, 5′-TCACTCCAGACATTGTCC-3′. The PCR primers for mouse β-actin were as follows: forward, 5′-ATTGTTACCAACTGGGACGACATG-3′; and reverse, 5′-CTTCATGAGGTAGTCTGTCAGGTC-3′. The PCR primers for gp70 were as follows: forward, 5′-ACCTTGTCCGAAGTGACCG-3′; and reverse, 5′-GTACCAATCCTGTGTGGTCG-3′. To quantify PCR band intensities, gel images were first captured with an Epi ChemiII digital image system (Ultraviolet Products). The individual PCR-amplified DNA fragment intensities were then analyzed with Image software (National Institutes of Health, Bethesda, MD), as described (27).

Quantitative real-time PCR was performed using an Opticon 2 system (MJ Research/Bio-Rad) and SYBR Green (Stratagene) as the fluorophore. Standards were prepared using serial dilutions of cDNA in the range of 1 million to 1 copy per reaction, and gene expression was quantified by comparison with the standard curve. All reactions were performed using parameters established by the manufacturer, and all gene expression levels for gp70 were normalized to mouse β-actin as the internal control.

Adoptive transfer experiments

Treatment of CMS4-tumor-bearing mice by adoptive immunotherapy (38) was performed in an experimental lung metastasis model (37). CMS4 sublines were resuspended in HBSS, and injected i.v. into the lateral tail vein (2.5 × 105 cells in 100 μl of total volume) of naive, immune competent female BALB/c mice. Three or 10 days later, CTL (4–5 days following in vitro stimulation) were prepared by centrifugation over a Ficoll-Hypaque gradient, washed, and resuspended in HBSS. CTL or HBSS were also injected i.v. into the tail vein (varying numbers of cells in 100 μl). Mice receiving CTL were euthanized 14 days after the adoptive transfer. In the minimal and extensive disease settings, those time points corresponded to days 17 and 24 posttumor transplant, respectively. Control mice receiving HBSS, in both experimental designs, were euthanized usually 17 days posttumor transplant due to disease burden. For enumeration of lung metastases, lungs were injected with a 15% solution of India ink, resected, and fixed in Fekete’s solution as described (34, 37). The number of lung nodules was enumerated in a single-blinded fashion under a dissecting microscope (Stemi SV6; Zeiss). Tumor foci considered too numerous to accurately count were recorded as ≥250.

The CMS4-met.cntl and CMS4-met.sel sublines were derived from CMS4-met. CMS4-met (2.5 × 105 cells per mouse) were injected into the lateral tail vein, then 10 days later, wt-CTL (3 × 106 cells of clone WT-3–4 from Ref.34 or HBSS was injected i.v. Control (HBSS) or CTL-treated mice were euthanized as described above; i.e., on days 17 and 24 posttumor transplant, respectively. Tumor sublines recovered from control or CTL-treated mice, designated as CMS4-met.cntl and CMS4-met.sel, respectively, were then isolated from the lung digests of independent mice also as described above.

Statistical analysis

Statistical analysis in the lung metastasis and quantitative real-time PCR experiments was determined using an unpaired, two-sided t test, whereas statistical analysis for comparison of the different CMS4-met sublines was determined by ANOVA. Values of p < 0.05 were considered statistically significant.

Results

Derivation of TEV and their biologic behavior in vivo and in vitro

To determine whether TEV emerge from interactions between CTL and their Ag-bearing targets in vivo, we made use of a CTL adoptive immunotherapy paradigm. In this model, mice received CMS4-met, and then 10 days later, when tumor burden was well established, these same mice received treatment with CTL (3 × 106) via adoptive transfer (34, 37). Under such conditions of a pre-existing extensive disease burden, CTL adoptive transfer led to significant but incomplete tumor regression. On the average, 10–35 tumor nodules were grossly visible and quantifiable (34, 37). This model thus allowed us to recover surviving tumor cells, which were then examined for potential alterations in antigenic and biologic properties. In fact, we established several CMS4-met sublines from independent mice after interaction with CTL in vivo. As controls, CMS4-met sublines were produced from mice receiving saline instead of CTL. The following studies were then conducted using representative sublines from CTL-treated and control mice, henceforth termed CMS4-met.sel and CMS4-met.cntl, respectively.

To assess tumorigenic behavior in vivo, both sublines were injected into separate groups of naive, immune-competent syngeneic BALB/c mice. Tumorigenicity was determined by survival following i.v. injection, which resulted in lung metastases (Fig. 1⇓A). Survival was dramatically shortened in mice receiving CMS4-met.sel i.v., even with 40% fewer cells, compared with mice receiving CMS4-met.cntl (Fig. 1⇓A). This difference in survival first became apparent by day 13 and was most dramatic by day 16, at which time all 10 mice that received CMS4-met.sel had already succumbed to tumor burden, whereas all 10 mice that received CMS4-met.cntl were still alive. The experiment was then arbitrarily terminated at day 18. Gross examination of the lungs of surviving mice from both groups at day 14 postimplantation revealed extensive tumor foci that were difficult to quantify accurately. Interestingly, in contrast to what was observed in vivo, both sublines displayed comparable growth in vitro (Fig. 1⇓B), suggesting that the disparities in malignant behavior in vivo (Fig. 1⇓A) were not simply due to intrinsic differences in tumor cell proliferation.

FIGURE 1.
FIGURE 1.

Generation of an aggressive tumor escape variant after interaction with CTL in vivo. Stable CMS4-met.cntl and CMS4-met.sel sublines were established from individual mice bearing CMS4-met. CMS4-met was injected into BALB/c mice, then 10 days later, wt-CTL or saline was injected i.v. Control (saline) or CTL-treated mice were euthanized, and the indicated tumor sublines were isolated from the lungs of mice, as described in Materials and Methods. A, CMS4-met.cntl (2.5 × 105 per mouse) or CMS4-met.sel (1.5 × 105 per mouse) was then injected i.v. into separate groups of BALB/c mice (10 mice per group) and survival recorded over an 18-day observation period. B, Both tumor sublines were examined for proliferation in vitro, as determined by [3H]thymidine incorporation assays. Cells were plated at different densities, and the radioisotope was added during the final 18 h of a 24-h incubation. Data are expressed as the mean ± SD of triplicate wells and are representative of two independent experiments.

Sensitivity of TEV to CTL in vitro and in vivo

The observation that CMS4-met.sel was more aggressive than CMS4-met.cntl in vivo (Fig. 1⇑) suggested that the former subline also may be lytically less sensitive to immune destruction. To explore that possibility, we examined the lytic susceptibility of these two sublines toward the same Ag-specific CTL used for the adoptive transfer studies (Fig. 2⇓A). However, CTL assays performed over a range of E:T ratios revealed that both sublines were equally sensitive to lysis (Fig. 2⇓A), suggesting that CMS4-met.sel did not express an obvious functional loss of Ag recognition capability or sensitivity to CTL effector mechanisms. P815 cells, which express the relevant MHC I molecule (H-2Ld), but not the relevant tumor Ag, were included as a negative control target to illustrate lytic specificity, as reported previously (34, 37). In addition to P815 cells, no specific CTL-mediated lysis was detected against syngeneic Con A-induced lymphoblasts, BALB-3T3 fibroblasts (37), or the A20 tumor cell line, a syngeneic B cell lymphoma (data not shown).

FIGURE 2.
FIGURE 2.

Susceptibility of CMS4-met.sel and CMS4-met.cntl sublines to CTL activity in vitro and in vivo. A, CTL tested for lytic activity against the indicated targets at varying E:T ratios in an 18-h 51Cr release assay. Data are shown as the mean ± SD of triplicate wells and are representative of four independent experiments. B, The susceptibility of the two sublines to CTL in vivo was tested in an adoptive transfer model (see Materials and Methods). Each subline (2.5 × 105 per mouse) was injected i.v. Three days later, varying concentrations of CTL also were injected i.v. Two weeks following CTL adoptive transfer, lungs were processed for enumeration of nodules. Mice receiving saline were typically euthanized 17 days posttumor transplant because of progressive disease burden. Each data point represents total lung nodule counts from a single mouse. Data at the 5 × 105 density and a corresponding saline-treated control group represent a compilation of two independent experiments. C, RT-PCR (left panel) and quantitative real-time PCR (right panel) analyses for the expression of the gp70 transcript, relative to β-actin expression levels, in the lungs of the indicated tumor-bearing mice. Each mouse was injected with a given subline, and the lungs were harvested 3 days later. Three mice were examined per tumor-bearing group, as indicated at the top of the gel (left panel). The data for quantitative real-time PCR analysis are presented as mean ± SD of the relative gp70 levels from three separate mice (right panel; p = 0.123).

To determine whether these two CMS4-met sublines differed in their response to CTL in vivo, we analyzed them in CTL adoptive transfer experiments at varying CTL concentrations (Fig. 2⇑B). The prediction was that, if both CMS4-met sublines were biologically similar, then similar CTL dose-response patterns should emerge. The converse would be expected if both sublines were dissimilar. Although similar antitumor effects were observed between both sublines at both the higher and lower CTL concentrations, a differential antitumor response was observed at the intermediate CTL concentration (5 × 105). In fact, the adoptive transfer of 5 × 105 CTL strongly inhibited the growth of CMS4-met.cntl but largely failed to alter the growth of CMS4-met.sel (Fig. 2⇑B). Noteworthy, this differential antitumor CTL response was unmasked by limiting the input CTL concentration.

The rationale that a minimal disease setting was appropriate for this in vivo comparison was supported by our observations that both CMS4-met sublines migrated to or colonized within the lung at similar efficiencies (Fig. 2⇑C). Within 3 days of tumor injection, which corresponded to the time of CTL transfer, tumor load was comparable between CMS4-met.cntl and CMS4-met.sel. This observation was based on quantitative real-time PCR analysis of mRNA levels of tumor-associated MuLV gp70, an indicator of tumor cell number in the lung (Fig. 2⇑C; p = 0.123) (27). Overall, because more tumor nodules were observed with CMS4-met.sel than with CMS4-met.cntl after CTL transfer of 5 × 105 cells per mouse (Fig. 2⇑B), these data indicate that, under certain conditions, adoptive immunotherapy can be less effective against such TEV.

Antigenic and phenotypic characteristics of TEV

To further delineate the basis for the biologic differences between these two sublines, we compared and examined them for expression of: 1) the relevant MHC I restriction element, H-2Ld, for CTL recognition (37); 2) a dominant tumor-associated Ag, MuLV-gp70, recognized by these CTL (37); and 3) Fas, a cell surface receptor recently demonstrated to be an important element for CTL killing in this model (34) (Figs. 3⇓ and 4⇓). We showed that both sublines expressed comparably high levels of cell surface H-2Ld. In fact, CMS4-met.sel reproducibly expressed higher levels of H-2Ld (Fig. 3⇓A). The intensity of expression of gp70 mRNA was marginally, but not significantly, reduced in CMS4-met.sel, compared with CMS4-met.cntl (Fig. 3⇓B). Nonetheless, because CTL lytic activity remained unaltered (Fig. 2⇑A), these data implied that the slight reduction in gp70 expression by CMS4-met.sel likely had little, if any, functional impact on Ag or epitope presentation and recognition.

FIGURE 3.
FIGURE 3.

MHC I and tumor Ag by the different sublines. A, H-2Ld expression by CMS4-met.sel (solid thick line) and CMS4-met.cntl (solid thin line) as determined by flow cytometry. Staining with the isotype control Ab (for CMS4-met.cntl as a reference) is shown as the shaded area. B, Expression of the gp70 transcript by RT-PCR analysis (upper panel). The numbers shown at the top of the gel reflect detection of gp70 message in a cDNA-titratable fashion. β-actin was used as normalization standard. gp70 levels shown in the panel above were quantified by analysis of the PCR band intensities (bottom panel). The relative expression level of gp70 at each cDNA dilution was obtained by dividing the band intensity of that dilution by the corresponding β-actin band intensity. The ratios were then averaged from the four cDNA dilutions. Each bar represents the mean ± SD of the ratios of two independent experiments.

FIGURE 4.
FIGURE 4.

Fas expression by the different sublines. A, RT-PCR analysis of Fas and β-actin mRNA levels (left panel). Cell surface Fas expression by CMS4-met.cntl and CMS4-met.sel (right panels), with (solid line) and without (dashed line) IFN-γ and TNF-α treatment overnight (100 U/ml each cytokine). Isotype-matched Ab is shown as the shaded area. The bar graph below the histogram depicts Fas MFI values ± SD of three separate experiments. B, CMS4-met.cntl (upper panel) and CMS4-met.sel (lower panel) were incubated overnight with both IFN-γ and TNF-α (as in A) to maximize sensitivity to Fas-mediated death. Cytokine-exposed cells were then treated with either nothing (shaded area) or recombinant sFasL (solid thick line; 100 ng/ml). Data are representative of three separate experiments. Data are expressed as the mean ± SD of the three separate experiments for cells treated with two different concentrations of recombinant sFasL protein (right panel). C, Additional CMS4-met sublines were isolated following CTL adoptive transfer from independent mice, as in Fig. 1⇑. Each subline was examined for sensitivity to Fas-mediated death as in B (sFasL, 100 ng/ml). CMS4-met.sel is designated as subline 1, whereas four additional similarly derived sublines are labeled sequentially from 2 to 5. D, Six control sublines, designated as a–f, were assayed for sensitivity to Fas-mediated death as in B and C. CMS4-met.cntl is designated as subline a. Data are expressed as the mean ± SEM of nine separate experiments for CMS4-met.cntl and three separate experiments for the other control sublines. Statistical significance of all CTL-derived sublines in C vs all control sublines in D was assessed by ANOVA (F1,9 = 5.12; p = 0.0009).

In contrast to aspects related to MHC/Ag expression, differences were observed in Fas expression and function. RT-PCR analysis showed that CMS4-met.sel expressed substantially lower levels of Fas mRNA, compared with CMS4-met.cntl (Fig. 4⇑A). Flow cytometric analysis revealed that both CMS4-met sublines were similarly weakly positive for cell surface Fas expression, based on their low mean fluorescence intensity (MFI) values (Fig. 4⇑A, upper and lower right panels). Because the proinflammatory cytokines, IFN-γ and TNF-α, have been shown to enhance Fas expression and function on neoplastic cells (26, 27, 39, 40), we also examined these sublines after such cytokine treatment. Despite the low MFI values, CMS4.met.cntl was more responsive to Fas up-regulation, compared with CMS4-met.sel (p = 0.029; Fig. 4⇑A, upper and lower right panels).

More importantly, functional studies demonstrated that both sublines following cytokine treatment showed differential sensitivities to Fas-mediated death (Fig. 4⇑B). Although both sublines did not undergo substantial cell death at the lower dose of recombinant sFasL protein (20 ng/ml), CMS4-met.sel reproducibly was less sensitive than the control subline (p = 0.03). When tested at the higher dose of recombinant sFasL (100 ng/ml), the magnitude of cell death increased in both sublines; however, CMS4-met.sel was still significantly (p = 0.0004) less sensitive to Fas-mediated death, compared with CMS4-met.cntl. Thus, a functional Fas response, which was observed more so with CMS4-met.cntl than with CMS4-met.sel, was unmasked or enhanced in vitro under proinflammatory cytokine-inducible conditions. In fact, previous studies revealed that CTL produced these proinflammatory cytokines in response to tumor-specific stimulation in vitro, which, in turn, enhanced Fas expression and function of CMS4-met (34) (i.e., the parental line from which both CMS4-met.cntl and CM4-met.sel sublines were derived). Furthermore, Fas expression was enhanced in vivo by CMS4-met-containing lung preparations following CTL adoptive transfer (34), consistent with the notion that CTL released such cytokines during effector-target interactions in vivo consequently boosting the frequency of tumor cells expressing Fas. Thus, Fas expression/function of CMS4-met.cntl and CMS4-met.sel correlated with their response to CTL adoptive immunotherapy (Figs. 2⇑B and 4⇑, A and B).

As noted earlier, we isolated several additional control and CTL-derived CMS4-met.sel sublines after CTL adoptive transfer from independent mice. In this study, we examined them for their sensitivity to Fas-mediated death following cytokine treatment (as in Fig. 4⇑B). Our findings indicated that four of five sublines recovered after CTL adoptive transfer (i.e., sublines 1–3 and 5; CMS4-met.sel denoted as subline 1) displayed a similarly diminished sensitivity to Fas-mediated death (Fig. 4⇑C; tested at 100 ng/ml). Moreover, comparison of all five CTL-derived sublines with six independently established control sublines (including CMS4-met.cntl indicated as subline a; compare Fig. 4⇑, C and D) revealed overall highly significant differences between these two groups in their sensitivity to Fas-mediated death (F1,9 = 5.12; p = 0.0009). The observation that the extent of Fas-induced death of a high frequency of such CTL-derived sublines was similarly reduced suggested that altered Fas sensitivity was a prevailing “tumor escape” phenotype or characteristic of this model. Collectively, the differential response between the control and various CMS4-met.sel sublines to Fas-mediated death appeared to be at least in part quantitative.

Reduced susceptibility of TEV to immunotherapy was consistent with a loss of Fas sensitivity

To assess a potential causal relationship between Fas expression/function in neoplastic cells and their response to adoptive immunotherapy, we made use of tumor-specific perforin-deficient (pfp)-CTL (see Materials and Methods) and FasL-deficient (gld)-CTL (34). Previous studies revealed that both perforin- and Fas-based effector mechanisms accounted for virtually all CTL activity against CMS4-met targets in vitro (34). Thus, we reasoned that if the reduction in therapeutic efficacy against CMS4-met.sel involved increased Fas resistance, then pfp-CTL, which could not kill through perforin but could do so via the FasL pathway, would now be unable to mediate efficient antitumor activity. In contrast, gld-CTL, which could not kill through FasL, but could do so via the perforin pathway, would still be able to mediate efficient antitumor activity.

CTL lytic assays in vitro, at least in part, supported that notion (Fig. 5⇓A). Our data indicated that gld-CTL mediated nearly comparable lytic activity against both CMS4-met.sel and CMS4-met.cntl, suggesting that both sublines were equally sensitive to perforin-based killing. In fact, earlier studies had revealed that gld-CTL lytic activity against CMS4-met targets was completely abrogated by concanamycin A (34), a potent pharmacologic inhibitor of the perforin pathway (41). In contrast with what was seen with gld-CTL, pfp-CTL mediated less efficient lysis against CMS4-met.sel than the control subline, which was discernable by diluting the E:T ratio. However, the observation that pfp-CTL killing of both sublines was comparable by increasing the E:T ratio was consistent with the notion that CMS4-met.sel was not completely Fas resistant (Fig. 4⇑, B and C).

FIGURE 5.
FIGURE 5.

Reduced susceptibility of CMS4-met.sel to CTL adoptive immunotherapy was linked to diminished Fas sensitivity. A, Lytic activity by gld-CTL or pfp-CTL against the indicated targets, as in Fig. 2⇑A. Data are expressed as the mean ± SD of triplicate wells and are representative of two independent experiments. B, The susceptibility of the two sublines to the indicated CTL populations in vivo was investigated, as in Fig. 2⇑B, at the CTL concentrations shown. Each data point represents total lung nodule counts from a single mouse. C, Pentamer (H-2Ld/gp70423–431) staining of wt-CTL (a, solid line) or pfp-CTL (b, solid line). An influenza-specific CTL line was included as a negative control (dashed line in each histogram). Shaded area represents unstained wt-CTL or pfp-CTL. c, Data are illustrated as the average ± SD of the MFI values of three separate experiments.

Interestingly, similar functional patterns were observed in vivo following CTL adoptive transfer (Fig. 5⇑B). In this study, we first used a CTL concentration (1 × 106 cells per mouse) that was shown earlier to mediate nearly complete tumor rejection by wt-CTL (Fig. 2⇑B). We reasoned that if pfp-CTL preferentially or exclusively mediated tumor rejection via the FasL pathway, then the extent of antitumor activity against CMS4-met.sel would be lower, compared with the control subline. At this CTL concentration, the adoptive transfer of all three CTL populations led to comparable and potent antitumor effects against the control subline, consistent with the notion that tumor rejection of the control subline under conditions of minimal disease could actually occur via the perforin or Fas pathway. However, at this same CTL concentration, the adoptive transfer of pfp-CTL mediated significantly (p = 0.001) less antitumor activity against CMS4-met.sel, compared with the control subline. In contrast, wt-CTL or gld-CTL still expressed strong antitumor effects against CMS4-met.sel. This experiment was repeated at a 3-fold higher CTL concentration (3 × 106 cells per mouse), and similar patterns were observed (Fig. 5⇑B, right panels). Although a further reduction in the number of tumor nodules was noted by increasing the pfp-CTL concentration injected, the differential antitumor pfp-CTL response was still observed between both sublines (p = 0.001). This suggested that limited amounts of transferred pfp-CTL were unlikely exclusively responsible for this differential antitumor response. As with the in vitro CTL assays (Fig. 5⇑A), the observation that pfp-CTL still expressed antitumor effects against CMS4-met.sel in vivo (Fig. 5⇑B) was consistent with the notion that CMS4-met.sel was not completely Fas resistant (Fig. 4⇑, B and C).

The observation that pfp-CTL efficiently mediated the rejection of the control subline indicated that these CTL migrated efficiently to the site of tumor growth in the lung. Lastly, pentamer staining of both wt-CTL and pfp-CTL lines demonstrated comparable levels of reactivity (Fig. 5⇑C), which was maintained in a pentamer-titrable fashion (data not shown). These data suggested that the T cell receptors of both of these CTL lines expressed similar levels of MHC/Ag recognition capability. Therefore, the underlying nature of this differential antitumor response likely reflected how these pfp-CTL responded to intrinsic differences in tumor-associated biologic properties.

Discussion

Many studies in the area of tumor escape have endorsed a role for Ag-loss variants as an important determinant of tumor progression (9, 10, 11, 12, 13, 14, 15, 16, 17). In this study, we showed for the first time that 1) adoptive transfer of CTL caused the emergence of naturally occurring Fas-refractory TEV with unaltered MHC I/tumor Ag expression levels; 2) such MHC/Ag-competent, but Faslow-TEV were more aggressive in vivo and significantly less responsive to adoptive immunotherapy as a result of reduced Fas sensitivity; and 3) such Faslow-TEV retained lytic susceptibility to the perforin pathway. Overall, we not only identified immunoselection of Faslow-TEV as a direct consequence of CTL-tumor interactions in vivo, but also demonstrated that such an altered neoplastic Fas phenotype compromised the immunotherapy outcome which was uncovered using pfp-CTL (for proof of principle) or wt-CTL under cell-limiting conditions. In contrast with this work, other studies have engineered mouse cell lines to express the cFLIP gene to abrogate Fas sensitivity (24, 25), but whether this molecular alteration occurs naturally or physiologically in murine malignancies remains unclear. Moreover, it seems that the loss of sensitivity to Fas-mediated death in combination with alterations in other tumor-associated genetic events are required for a more malignantly proficient phenotype (26, 27).

Although alterations in the genetic or epigenetic program of the neoplastic population are undoubtedly crucial for tumor progression (42), we further hypothesized that CTL (immune)-mediated selective pressures may help drive the emergence of such aggressive neoplastic subpopulations possessing those tumorigenic characteristics. To address that notion, we made use of a CTL adoptive immunotherapy model of experimental lung metastasis under the circumstances of a pre-existing extensive tumor burden. Thus, under conditions whereby the CTL-mediated tumor regression response was incomplete, we were able to recover and study the biology of potential TEV. We found that TEV were more malignantly proficient, as determined by reduced survival (Fig. 1⇑A). These data support the notion that an external CTL-selective pressure can influence the outgrowth of TEV possessing heightened malignant behavior. Interestingly, these TEV did not display an alteration in CTL lytic sensitivity (Fig. 2⇑A) or an obvious reduction in the expression of the relevant MHC I molecule (H-2Ld) or a tumor-associated Ag (gp70) recognized by these CTL (Fig. 3⇑). Collectively, these data suggested that these TEV retained Ag recognition capability as defined by these in vitro criteria.

To examine whether these TEV retained or lost susceptibility to eradication by CTL adoptive immunotherapy, we compared the extent of tumor rejection of CMS4-met.sel vs CMS4-met.cntl at three different CTL concentrations (Fig. 2⇑B). Thus, if both sublines were biologically similar, then comparable CTL dose-response patterns should emerge. The converse would be expected if both sublines were dissimilar. To reduce the likelihood that any observable differential antitumor response reflected disparities in malignant behavior in vivo, experiments were conducted under the conditions of minimal disease. Within 3 days of tumor injection (i.e., corresponding to the time of CTL transfer in a minimal disease setting), tumor load was comparable between the CMS4-met.cntl and CMS4-met.sel sublines (Fig. 2⇑C). At the highest concentration tested, the adoptive transfer of these CTL strongly rejected both tumor sublines with nearly comparable efficacy (Fig. 2⇑B). Similarly, both tumor sublines failed to be rejected at the lowest CTL concentration tested. On the contrary, at the intermediate CTL concentration, a distinct pattern was unveiled in that TEV were substantially less susceptible to eradication, compared with the control subline. These data indicated that both tumor sublines were not biologically similar, and that CMS4-met.sel was less responsive to eradication under conditions whereby delicate fluctuations in the concentrations of transferred CTL affected the functional outcome. These results were not readily predictable from the in vitro experiments (Fig. 2⇑A), suggesting that in vitro assays do not necessarily recapitulate the complexity of immune cell-tumor cell interactions in vivo. It also is important to point out that, despite a 2-fold dilution of transferred CTL (from 1 × 106 cells to 5 × 105 cells), this still represented a 5 × 105 cell reduction, which in this model was ostensibly sufficient to unmask the differential antitumor response against the two sublines.

To elucidate further possible mechanisms underlying this differential antitumor response in vivo, we next examined these two sublines for differences in lytic susceptibility via the perforin and/or Fas pathway. Previous studies revealed that both pathways accounted for virtually all CTL lytic activity against CMS4-met targets (34). Additionally, CMS4-met was found to be resistant to the cytotoxic effects of TNF-α (26). In these studies, the observations that both tumor sublines were lysed or eradicated equally well by wt-CTL (Fig. 2⇑A) and gld-CTL in vitro (Fig. 5⇑A, left panel) and in vivo (Fig. 5⇑B) suggested that these TEV retained sensitivity to the perforin pathway. In contrast, CMS4-met.sel, as well as several other sublines recovered after CTL adoptive transfer, were overall significantly less sensitive to Fas-mediated death, compared with several independently isolated CMS4-met.cntl sublines (compare Fig. 4⇑, C and D). The finding that such CTL-derived sublines overall possessed a significantly altered sensitivity to Fas-induced death, compared with the control sublines, suggested that this outcome may be a dominant mechanism of escape in this model.

Titration experiments with wt-CTL revealed that if the input effector cell concentration became limiting, immunotherapy efficacy against such Faslow-TEV was diminished when compared appropriately with the control subline at 5 × 105 cells per mouse (Fig. 2⇑B). Studies with pfp-CTL demonstrated that reduced immunotherapeutic efficacy was linked to an altered Fas sensitivity, and that increasing the pfp-CTL concentration was unable to completely overcome this neoplastic alteration (Fig. 5⇑B). In contrast, increasing the wt-CTL concentration did overcome this neoplastic deficiency (Fig. 2⇑B), which was likely due to the availability of greater numbers of CTL to mediate tumor rejection via a perforin-dependent mechanism. This was supported by the use of gld-CTL, which displayed comparable antitumor effects (Fig. 5⇑B). Thus, these data are consistent with the hypothesis that under conditions whereby putative CTL/target ratios in vivo would fall below a critical therapeutic threshold, immunotherapy efficacy against such Faslow-TEV may be compromised. This clearly reinforces the notion that an important parameter of effective immunotherapy involves the delivery and/or accumulation of adequate numbers of antitumor CTL at sites of neoplastic growth to lessen the likelihood for tumor escape and the generation of potential TEV.

In that context, it is noteworthy that the perforin pathway might circumvent or compensate for the reduction or loss of Fas function in neoplastic cells under conditions of higher antitumor CTL concentrations in vivo, such as that achieved by adoptive immunotherapy of TIL in advanced melanoma patients (21, 43). Although we showed that such Faslow-TEV were responsive to the perforin pathway in this model, it is conceivable that in circumstances whereby the Fas/FasL pathway may actually be the dominant mechanism of T cell-mediated tumor rejection (44, 45) or the perforin pathway may not be triggered efficiently (46), such a Fas-based escape mechanism may have broader adverse implications for the immunotherapy outcome.

Furthermore, it is important to point out that CMS4-met.sel sublines were produced from mice only after a 2-week CTL-tumor interaction in vivo. Consequently, it remains to be fully understood whether additional tumor escape mechanisms, besides those described in this study, may be operative, which have not yet been observed or may occur with a longer, more persistent adaptive immune (CTL-induced) selective pressure. It also is important to point out that, because these studies were conducted with an already existing tumor cell population, our observations defined changes to a malignancy that had previously likely developed effective escape mechanisms. Thus, in the face of additional intrinsic or extrinsic selective pressures, as shown in this study, pre-existing tumors may be further edited or reshaped, resulting in variants with progressively heightened malignant or immunotherapy-refractory properties. Moreover, although we focused on the role of CTL in this process, it is possible that other host-derived FasL-expressing lymphoid cells and/or nonlymphoid tissues (i.e., lung; see Ref.23) may contribute to the generation of such Faslow-TEV. In this setting, CTL may initiate this sequence of events, possibly by up-regulating FasL in the lung microenvironment via cytokine secretion and, consequently, engage other FasL-dependent interactions. Whether these complex interactions are involved, however, require further investigation, perhaps through comparison of wt to gld mice.

Overall, our results presented here implicate the Fas/FasL system as a potentially important element of the selective process. This represents the first physiologic illustration of a CTL-mediated selective pressure as a potentially important mechanism underlying the outgrowth of naturally occurring Faslow-TEV expressing a more aggressive or immunotherapy-refractory phenotype. The observation that TEV expressing a more Fas-refractory phenotype emerged in response to Ag-specific CTL in vivo also may provide a mechanistic basis, at least in part, for the reported inverse relationship between Fas expression or function and malignant/metastatic phenotype in mouse models (23, 24, 25, 26, 27), as well as human systems reflecting colorectal, breast, lung, and liver carcinoma (28, 29, 30, 31, 32, 33, 47, 48). Further molecular characterization of such TEV will improve not only our understanding of the neoplastic process, but perhaps also may aid in the design of effective combination therapies.

Acknowledgments

We thank Elwood McDuffie for technical assistance and Debra Weingarten for editorial assistance in the preparation of this manuscript. This research was supported by the Intramural Research Program of the National Institutes of Health, National Cancer Institute, Center for Cancer Research.

Disclosures

The authors have no financial conflict of interest.

Footnotes

  • 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 K.L. and S.A.C. contributed equally to this study.

  • ↵2 Address correspondence and reprint requests to Dr. Scott I. Abrams, Laboratory of Tumor Immunology and Biology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Building 10, Room 5B46, 10 Center Drive, Bethesda, MD 20892-1402. E-mail address: sa47z{at}nih.gov

  • ↵3 Abbreviations used in this paper used in this paper: TEV, tumor escape variant; FasL, Fas ligand; gld-CTL, FasL-deficient CTL; MFI, mean fluorescence intensity; pfp-CTL, perforin-deficient CTL; sFasL, soluble FasL; wt-CTL, wild-type CTL; MHC class I, MHC I.

  • Received July 18, 2005.
  • Accepted January 6, 2006.

References

  1. ↵
    Igney, F. H., P. H. Krammer. 2002. Immune escape of tumors: apoptosis resistance and tumor counterattack. J. Leukocyte Biol. 71: 907-920.
    OpenUrlAbstract/FREE Full Text
  2. ↵
    Yee, C., J. A. Thompson, D. Byrd, S. R. Riddell, P. Roche, E. Celis, P. D. Greenberg. 2002. Adoptive T cell therapy using antigen-specific CD8+ T cell clones for the treatment of patients with metastatic melanoma: in vivo persistence, migration, and antitumor effect of transferred T cells. Proc. Natl. Acad. Sci. USA 99: 16168-16173.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    Marincola, F. M., E. Wang, M. Herlyn, B. Seliger, S. Ferrone. 2003. Tumors as elusive targets of T-cell-based active immunotherapy. Trends Immunol. 24: 335-342.
    OpenUrlPubMed
  4. ↵
    Schmollinger, J. C., R. H. Vonderheide, K. M. Hoar, B. Maecker, J. L. Schultze, F. S. Hodi, R. J. Soiffer, K. Jung, M. J. Kuroda, N. L. Letvin, et al 2003. Melanoma inhibitor of apoptosis protein (ML-IAP) is a target for immune-mediated tumor destruction. Proc. Natl. Acad. Sci. USA 100: 3398-3403.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    Waldmann, T. A.. 2003. Immunotherapy: past, present and future. Nat. Med. 9: 269-277.
    OpenUrlCrossRefPubMed
  6. ↵
    Pawelec, G.. 2004. Immunotherapy and immunoselection: tumor escape as the final hurdle. FEBS Lett. 567: 63-66.
    OpenUrlCrossRefPubMed
  7. ↵
    Dunn, G. P., L. J. Old, R. D. Schreiber. 2004. The three Es of cancer immunoediting. Annu. Rev. Immunol. 22: 329-360.
    OpenUrlCrossRefPubMed
  8. ↵
    Shankaran, V., H. Ikeda, A. T. Bruce, J. M. White, P. E. Swanson, L. J. Old, R. D. Schreiber. 2001. IFN-γ and lymphocytes prevent primary tumour development and shape tumour immunogenicity. Nature 410: 1107-1111.
    OpenUrlCrossRefPubMed
  9. ↵
    Uyttenhove, C., J. Maryanski, T. Boon. 1983. Escape of mouse mastocytoma P815 after nearly complete rejection is due to antigen-loss variants rather than immunosuppression. J. Exp. Med. 157: 1040-1052.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    Vasmel, W. L., E. J. Sijts, C. J. Leupers, E. A. Matthews, C. J. Melief. 1989. Primary virus-induced lymphomas evade T cell immunity by failure to express viral antigens. J. Exp. Med. 169: 1233-1254.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    Ward, P. L., H. K. Koeppen, T. Hurteau, D. A. Rowley, H. Schreiber. 1990. Major histocompatibility complex class I and unique antigen expression by murine tumors that escaped from CD8+ T-cell-dependent surveillance. Cancer Res. 50: 3851-3858.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    Restifo, N. P., F. M. Marincola, Y. Kawakami, J. Taubenberger, J. R. Yannelli, S. A. Rosenberg. 1996. Loss of functional β2-microglobulin in metastatic melanomas from five patients receiving immunotherapy. J. Natl. Cancer Inst. 88: 100-108.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    Jager, E., M. Ringhoffer, M. Altmannsberger, M. Arand, J. Karbach, D. Jager, F. Oesch, A. Knuth. 1997. Immunoselection in vivo: independent loss of MHC class I and melanocyte differentiation antigen expression in metastatic melanoma. Int. J. Cancer. 71: 142-147.
    OpenUrlCrossRefPubMed
  14. ↵
    Hicklin, D. J., Z. Wang, F. Arienti, L. Rivoltini, G. Parmiani, S. Ferrone. 1998. β2-Microglobulin mutations, HLA class I antigen loss, and tumor progression in melanoma. J. Clin. Invest. 101: 2720-2729.
    OpenUrlCrossRefPubMed
  15. ↵
    Khong, H. T., N. P. Restifo. 2002. Natural selection of tumor variants in the generation of “tumor escape” phenotypes. Nat. Immunol. 3: 999-1005.
    OpenUrlCrossRefPubMed
  16. ↵
    Chang, C. C., M. Campoli, N. P. Restifo, X. Wang, S. Ferrone. 2005. Immune selection of hot-spot β2-microglobulin gene mutations, HLA-A2 allospecificity loss, and antigen-processing machinery component down-regulation in melanoma cells derived from recurrent metastases following immunotherapy. J. Immunol. 174: 1462-1471.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    Sanchez-Perez, L., T. Kottke, R. M. Diaz, A. Ahmed, J. Thompson, H. Chong, A. Melcher, S. Holmen, G. Daniels, R. G. Vile. 2005. Potent selection of antigen loss variants of B16 melanoma following inflammatory killing of melanocytes in vivo. Cancer Res. 65: 2009-2017.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    Boon, T., P. van der Bruggen. 1996. Human tumor antigens recognized by T lymphocytes. J. Exp. Med. 183: 725-729.
    OpenUrlFREE Full Text
  19. ↵
    Hanson, H. L., D. L. Donermeyer, H. Ikeda, J. M. White, V. Shankaran, L. J. Old, H. Shiku, R. D. Schreiber, P. M. Allen. 2000. Eradication of established tumors by CD8+ T cell adoptive immunotherapy. Immunity 13: 265-276.
    OpenUrlCrossRefPubMed
  20. ↵
    Ochsenbein, A. F., S. Sierro, B. Odermatt, M. Pericin, U. Karrer, J. Hermans, S. Hemmi, H. Hengartner, R. M. Zinkernagel. 2001. Roles of tumour localization, second signals and cross priming in cytotoxic T-cell induction. Nature 411: 1058-1064.
    OpenUrlCrossRefPubMed
  21. ↵
    Mortarini, R., A. Piris, A. Maurichi, A. Molla, I. Bersani, A. Bono, C. Bartoli, M. Santinami, C. Lombardo, F. Ravagnani, et al 2003. Lack of terminally differentiated tumor-specific CD8+ T cells at tumor site in spite of antitumor immunity to self-antigens in human metastatic melanoma. Cancer Res. 63: 2535-2545.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    Yu, P., Y. Lee, W. Liu, T. Krausz, A. Chong, H. Schreiber, Y. X. Fu. 2005. Intratumor depletion of CD4+ cells unmasks tumor immunogenicity leading to the rejection of late-stage tumors. J. Exp. Med. 201: 779-791.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    Owen-Schaub, L. B., K. L. van Golen, L. L. Hill, J. E. Price. 1998. Fas and Fas ligand interactions suppress melanoma lung metastasis. J. Exp. Med. 188: 1717-1723.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    Djerbi, M., V. Screpanti, A. I. Catrina, B. Bogen, P. Biberfeld, A. Grandien. 1999. The inhibitor of death receptor signaling, FLICE-inhibitory protein defines a new class of tumor progression factors. J. Exp. Med. 190: 1025-1032.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    Medema, J. P., J. de Jong, T. van Hall, C. J. Melief, R. Offringa. 1999. Immune escape of tumors in vivo by expression of cellular FLICE-inhibitory protein. J. Exp. Med. 190: 1033-1038.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    Liu, K., S. I. Abrams. 2003. Alterations in Fas expression are characteristic of, but not solely responsible for, enhanced metastatic competence. J. Immunol. 170: 5973-5980.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    Liu, K., S. A. Caldwell, S. I. Abrams. 2005. Cooperative disengagement of Fas and intercellular adhesion molecule-1 function in neoplastic cells confers enhanced colonization efficiency. Cancer Res. 65: 1045-1054.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    Keane, M. M., S. A. Ettenberg, G. A. Lowrey, E. K. Russell, S. Lipkowitz. 1996. Fas expression and function in normal and malignant breast cell lines. Cancer Res. 56: 4791-4798.
    OpenUrlAbstract/FREE Full Text
  29. ↵
    Houghton, J. A., F. G. Harwood, A. A. Gibson, D. M. Tillman. 1997. The Fas signaling pathway is functional in colon carcinoma cells and induces apoptosis. Clin. Cancer Res. 3: 2205-2209.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    Krammer, P. H., P. R. Galle, P. Moller, K. M. Debatin. 1998. CD95(APO-1/Fas)-mediated apoptosis in normal and malignant liver, colon, and hematopoietic cells. Adv. Cancer Res. 75: 251-273.
    OpenUrlCrossRefPubMed
  31. ↵
    von Reyher, U., J. Strater, W. Kittstein, M. Gschwendt, P. H. Krammer, P. Moller. 1998. Colon carcinoma cells use different mechanisms to escape CD95-mediated apoptosis. Cancer Res. 58: 526-534.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    Liu, K., S. I. Abrams. 2003. Coordinate regulation of IFN consensus sequence-binding protein and caspase-1 in the sensitization of human colon carcinoma cells to Fas-mediated apoptosis by IFN-γ. J. Immunol. 170: 6329-6337.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    Liu, K., E. McDuffie, S. I. Abrams. 2003. Exposure of human primary colon carcinoma cells to anti-Fas interactions influences the emergence of pre-existing Fas-resistant metastatic subpopulations. J. Immunol. 171: 4164-4174.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    Caldwell, S. A., M. H. Ryan, E. McDuffie, S. I. Abrams. 2003. The Fas/Fas ligand pathway is important for optimal tumor regression in a mouse model of CTL adoptive immunotherapy of experimental CMS4 lung metastases. J. Immunol. 171: 2402-2412.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    Nagata, S., P. Golstein. 1995. The Fas death factor. Science 267: 1449-1456.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    Schulze-Osthoff, K., D. Ferrari, M. Los, S. Wesselborg, M. E. Peter. 1998. Apoptosis signaling by death receptors. Eur. J. Biochem. 254: 439-459.
    OpenUrlPubMed
  37. ↵
    Ryan, M. H., J. A. Bristol, E. McDuffie, S. I. Abrams. 2001. Regression of extensive pulmonary metastases in mice by adoptive transfer of antigen-specific CD8+ CTL reactive against tumor cells expressing a naturally occurring rejection epitope. J. Immunol. 167: 4286-4292.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    Abrams, S. I., J. W. Hodge, J. P. McLaughlin, S. M. Steinberg, J. A. Kantor, J. Schlom. 1997. Adoptive immunotherapy as an in vivo model to explore antitumor mechanisms induced by a recombinant anticancer vaccine. J. Immunother. 20: 48-59.
    OpenUrlCrossRefPubMed
  39. ↵
    Owen-Schaub, L. B., R. Radinsky, E. Kruzel, K. Berry, S. Yonehara. 1994. Anti-Fas on nonhematopoietic tumors: levels of Fas/APO-1 and bcl-2 are not predictive of biological responsiveness. Cancer Res. 54: 1580-1586.
    OpenUrlAbstract/FREE Full Text
  40. ↵
    Lee, J. K., T. J. Sayers, A. D. Brooks, T. C. Back, H. A. Young, K. L. Komschlies, J. M. Wigginton, R. H. Wiltrout. 2000. IFN-γ-dependent delay of in vivo tumor progression by Fas overexpression on murine renal cancer cells. J. Immunol. 164: 231-239.
    OpenUrlAbstract/FREE Full Text
  41. ↵
    Kataoka, T., N. Shinohara, H. Takayama, K. Takaku, S. Kondo, S. Yonehara, K. Nagai. 1996. Concanamycin A, a powerful tool for characterization and estimation of contribution of perforin- and Fas-based lytic pathways in cell-mediated cytotoxicity. J. Immunol. 156: 3678-3686.
    OpenUrlAbstract
  42. ↵
    Hanahan, D., R. A. Weinberg. 2000. The hallmarks of cancer. Cell. 100: 57-70.
    OpenUrlCrossRefPubMed
  43. ↵
    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.
    OpenUrlAbstract/FREE Full Text
  44. ↵
    Chakraborty, M., S. I. Abrams, C. N. Coleman, K. Camphausen, J. Schlom, J. W. Hodge. 2004. External beam radiation of tumors alters phenotype of tumor cells to render them susceptible to vaccine-mediated T-cell killing. Cancer Res. 64: 4328-4337.
    OpenUrlAbstract/FREE Full Text
  45. ↵
    Dobrzanski, M. J., J. B. Reome, J. A. Hollenbaugh, J. C. Hylind, R. W. Dutton. 2004. Effector cell-derived lymphotoxin α and Fas ligand, but not perforin, promote Tc1 and Tc2 effector cell-mediated tumor therapy in established pulmonary metastases. Cancer Res. 64: 406-414.
    OpenUrlAbstract/FREE Full Text
  46. ↵
    Esser, M. T., B. Krishnamurthy, V. L. Braciale. 1996. Distinct T cell receptor signaling requirements for perforin- or FasL- mediated cytotoxicity. J. Exp. Med. 183: 1697-1706.
    OpenUrlAbstract/FREE Full Text
  47. ↵
    Friesen, C., S. Fulda, K. M. Debatin. 1999. Cytotoxic drugs and the CD95 pathway. Leukemia. 13: 1854-1858.
    OpenUrlCrossRefPubMed
  48. ↵
    Owen-Schaub, L., H. Chan, J. C. Cusack, J. Roth, L. L. Hill. 2000. Fas and Fas ligand interactions in malignant disease. Int. J. Oncol. 17: 5-12.
    OpenUrlPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section