Up-Regulation by Ammonium Trichloro(Dioxoethylene-0,0′) Tellurate (AS101) of Fas/Apo-1 Expression on B16 Melanoma Cells: Implications for the Antitumor Effects of AS101¹

The expansion of a neoplastic clone is determined by the equilibrium between proliferation, on the one hand, and induction or blockade of apoptosis, on the other. The Apo-1/Fas Ag is a member of the TNF/nerve growth factor supergene family (1). Binding of the cross-linking Apo-1/Fas or Fas mAbs (2) or the physiologic ligand (3) induces programmed cell death in many cell lines positive for the Apo-1/Fas Ag (2, 4). Apo-1/Fas-induced cell death has been recognized as an effector mechanism involved in T cell-mediated target cell killing (5), with T cells being the main reservoir of the Apo-1/Fas ligand (FasL)³ (6). The sensitivity of tumor cells to cross-linking of the Apo-1/Fas Ag might therefore determine at least partly the effects of the natural immune surveillance on a tumor.

As might be anticipated from their normal tissue counterparts, cells from NK lymphoma and large granular leukemia of T and NK origin were shown to express functional FasL (7). Furthermore, soluble FasL was detected in the sera of these patients (7), which may be derived from membrane FasL cleaved from the tumor cells. Considering nonlymphoid tumors, functional FasL expression has to date been reported in several distinct lineages of tumors: colon carcinoma (8), melanoma (9), hepatocellular carcinoma (10), and astrocytoma (11). Evidence is now accumulating that other tumors may also express FasL (ovarian carcinoma and head and neck squamous cell carcinoma) (12). The most enticing interpretation of FasL expression by tumor cells is that it offers a further novel mechanism of immune evasion. In vitro functional studies lend indirect evidence in support of such a hypothesis, as tumor cell lines (8, 9, 10, 11) and/or tumor cells tested ex vivo (10, 11) can deliver a death signal to Fas-expressing target cells in vitro, suggesting that potential effector cells in antitumor immune responses may be inactivated. Clearly, these FasL-expressing tumors (like many other cancers) possess other potential mechanisms of immune escape.

The in vivo consequences of FasL expression by unmodified tumor cells have only been directly investigated in the murine melanoma B16-F10, which produced tumors later in lpr mice (expressing minimal or no Fas) compared with normal or gld mice (9).

The AS101 compound, first synthesized at Bar-Ilan University (Ramat-Gan, Israel) has previously been shown to have immunomodulating properties and minimal toxicity. The compound has been shown to stimulate both mouse and human cells to proliferate and to secrete a variety of cytokines both in vitro and in vivo (13, 14, 15, 16, 17, 18, 19). AS101 administered systematically to mice transplanted with fibrosarcoma or Lewis lung carcinoma (3LL) (13) mediated significant antitumor effects that could be ascribed to its immunomodulatory properties. AS101 has been found to have radioprotective (14, 17, 20) and chemoprotective (21, 22, 23, 24) properties when injected into mice before sublethal and lethal doses of irradiation or various chemotherapeutic drugs.

AS101 was found to improve the survival of Madison lung carcinoma-bearing mice when given in combination with chemotherapy (21, 25). Given these preclinical findings, phase I clinical trials on advanced cancer patients treated with AS101 were conducted to determine the maximum tolerated dose and the optimal dose required for immune modulation with minimal toxicity. The results demonstrated that AS101 had no serious side effects. At 1 to 3 mg/m², it was shown to induce a clear predominance in type 1 responses, with a concomitant decrease in type 2 responses (19). This was reflected by a significant enhancement in IL-2 and IFN-γ levels paralleled by a substantial decrease in IL-4 and IL-10 (19). On the basis of these studies, phase II clinical trials on patients with non-small cell lung cancer receiving chemotherapy alone or chemotherapy plus AS101 were initiated and showed statistically significant prevention of bone marrow toxicity induced by chemotherapy. The response rate of patients was higher, although not significantly, in patients treated with chemotherapy and AS101 than in patients treated with chemotherapy alone (24).

Recently, AS101 was shown to significantly inhibit the development of B16-F10 melanoma lung metastases when given at daily injections to syngeneic mice and to directly inhibit the development of B16 colony development in semisolid cultures (26).

The present study was designed to evaluate the ability of AS101 to break the immunologic unresponsiveness to melanoma by altering at least in part their Fas expression, thus affecting the interactions between tumor cells and the host immune response.

Eight-week-old male C57BL/6 (wt), C57BL/6-lpr, and C57BL/6-gld mice were obtained from The Jackson Laboratory (Bar Harbor, ME). Experiments were performed in accordance with approved institutional protocols and the guidelines of the institutional animal care and use committee.

AS101 was supplied in a solution of PBS, pH 7.4, and was maintained at 4°C. Before use, AS101 was diluted in PBS, and the appropriate concentrations in a 0.2-ml volume were administered to mice by i.p. injections.

The B16-F10 (B16) melanoma cell line was used for all experiments. The cells possessed heavy pigment and produced typical melanoma tumors after s.c. inoculation of 5 × 10³ cells. B16 cells were grown in RPMI 1640 medium (Life Technologies, Grand Island, NY), supplemented with 10% heat-inactivated FCS (Biologic Industries, Kibbutz Beit Haemek, Israel), 2 mM glutamine, 1 mM nonessential amino acids, 1 mM sodium pyruvate, 100 U/ml penicillin, and 100 μg/ml streptomycin (all from Biologic Industries).

Animals were unilaterally and s.c. injected with a low inoculum (5 × 10³ cells) of B16-F10 melanoma cells of C57BL/6 origin on day 0. AS101, at different doses, or PBS, in a 0.2-ml volume, was injected i.p. starting on either day −1 or day 3 following tumor implantation and then every other day until the end of the experiment. The diameters of s.c. tumors were measured daily with calipers, and the volume was calculated by the formula: tumor volume = (longest diameter) × (shortest diameter)².

B16 cells were incubated at 37°C for 24 h in the presence of various concentrations of AS101. Cell extracts were prepared by suspension in ice-cold lysis buffer containing 10 mM Tris (pH 7.4), 150 mM NaCl, 1% Triton X, 0.1% SDS, 5 mM EDTA, 0.5% deoxycorticosterone, 1 mM PMSF, 2 mM NaF, 100 μM sodium vanadate, 5 μg/ml aprotinin, and 5 μg/ml leupeptin. Samples were immunoprecipitated using either mouse IgG (PAb 246) wild-type P53 Abs (Oncogene, Cambridge, MA) or goat IgG anti-Fas Abs (Santa Cruz Biotechnology, Santa Cruz, CA). Immune complexes were precipitated with protein A-Sepharose (Pharmacia, Piscataway, NJ) and washed four times with the lysis buffer. The pellets were boiled for 4 min, electrophoresed on 12.5 or 7.5% SDS-PAGE, and immunoblotted with the specific Abs. Blots were developed using horseradish peroxidase-conjugated secondary Abs and the ECL detection system (Amersham, Arlington Heights, IL).

B16 melanoma cells were cultured for 24 h with different concentrations of AS101. The anti-mouse Fas (clone Jo2, hamster IgG) mAb or the isotype-matched control anti-mouse IgG was added to the various cell samples (10⁶/sample) for 30 min and counterstained with FITC-labeled anti-hamster IgG. For FasL staining, the rabbit IgG anti-mouse FasL was used (Santa Cruz Biotechnology). For each sample 10⁴ cells were analyzed on a FACScan (Becton Dickinson, Mountain View, CA). Dead cells were excluded from analysis by staining with propidium iodide.

B16 cells were maintained as stock cultures in RPMI 1640 supplemented with 10% FCS and antibiotics. A number of 100-mm petri dishes were plated with 5 × 10⁵ cells in medium or in medium containing AS101 and incubated for 24 h. The cells were rinsed, trypsinized, counted, plated at 500 cells/plate, and incubated for macroscopic colony formation. Following 8 or 9 days of incubation, colonies were fixed with methanol and stained with Giemsa. Colonies with >50 cells were counted. All survival points were performed in duplicate, and experiments were conducted three times.

Following incubation with AS101, anti-Fas IgG was added to the cultures at 5 μg/ml for 4 h. In some experiments (Fig. 7) apoptosis was detected without ligation with anti-Fas Abs. The percentage of cells undergoing apoptosis was quantitatively determined using an apoptosis detection kit (R & D, Minneapolis, MN) by virtue of their ability to bind annexin V and exclude propidium iodide.

Survival curves were tested by comparing the cumulative percentage of survival using the Gehan-Wilcoxon test. For comparisons of means of the various groups the pairwise t test was used. Median values of tumor weight were analyzed according to the nonparametric Wilcoxon rank sum test.

In these experiments AS101 at different concentrations was injected into mice transplanted with a low inoculum of B16-F10 melanoma cells. The dose of 10 μg/mouse/injection was found to be optimal and therefore was applied in two treatment protocols. This dose has previously been shown to be optimal in exerting immunomodulating and antitumor activities in various mouse models. The results shown in Figure 1 clearly demonstrate the antitumor effects of AS101. Injection of AS101 at 10 μg/mouse starting 1 day before tumor cell inoculation significantly decreased tumor volume. At the same dose of AS101 injected 3 days following transplantation of the tumor, the decrease in tumor volume, although significant, was less apparent.

The in vivo effects of AS101 led us to consider direct effects of the compound on tumor cell clonogenicity. As shown in Figure 2, a dose-dependent inhibitory effect of AS101 was observed when B16 cells were cocultured with AS101. The decrease in clonogenicity was significant at doses of 0.5 and 1 μg/ml (p < 0.01). It should be noted that both concentrations of AS101 were previously shown to elicit optimal immunomodulatory effects when introduced to normal splenocytes, bone marrow, or bone marrow stromal cells (13, 14, 16, 22). These results suggest that the cytotoxic activity of AS101 can be partly explained by a direct effect on B16-F10 cells.

Stimulation of B16 cells with AS101 results in the accumulation of B16 cells in G₁ (data not shown). This prompted us to evaluate p53 expression in B16 cells. Moreover, AS101 has previously been shown to up-regulate p53 protein expression in M109 lung adenocarcinoma cells (25). Figure 3 shows that stimulation of B16 cells with AS101 for 24 h increased wild-type p53 protein expression in a dose-dependent manner, with 0.5 μg/ml being the optimal dose of AS101. Unstimulated B16 cells expressed low, but detectable, levels of p53.

There is mounting evidence that wild-type p53 is required to induce Fas expression in malignant cells (27, 28). Accumulating data reveal that certain anticancer drugs that affect the expression of wild-type p53 or microinjection of p53 into tumor cells result in up-regulation of Fas expression in these cells (27, 28). The increase in p53 expression in B16 cells treated with AS101 led us to evaluate the induction of Fas expression by the compound.

Flow cytometric analysis showed that B16 cells do not express Fas. Exposure of the cells to AS101 revealed a dose-dependent increase in Fas expression (Fig. 4,a). Expression of Fas protein following incubation with AS101 was validated by Western blot analysis (Fig. 5). In contrast, untreated B16 cells expressed abundant amounts of FasL. Treatment with AS101 did not further increase FasL expression (Fig. 4 b).

It has previously been shown that although the presence of Fas/Apo-1 is clearly a prerequisite for biologic responses triggered by specific Ab binding, signaling may not uniformly result in programmed cell death (4). We therefore investigated whether Fas receptors induced by AS101 were functional.

Figure 6 shows that up-regulation of CD95 receptor expression after AS101 treatment resulted in increased responsiveness to CD95 stimulation. Treatment with anti Apo-1 alone led to early and late apoptosis in up to 13.4% of B16 cells. In B16 cells treated with AS101 for 24 h, receptor stimulation with anti Apo-1 resulted in induction of early and late apoptosis in up to 44.4% of the cells, depending on the AS101 concentration. Control cells (RPMI) to which anti-Fas was not added contained only 10.1% apoptotic cells, while AS101 treated cells under these conditions included only 11%. Treatment of cells with isotype-matched control Ab did not significantly affect these values (data not shown). Longer incubation with AS101 (72 h) resulted in significantly increased rates of apoptosis without ligation with anti-Apo-1. Incubation of 72-h AS101-stimulated cells with anti-FasL Abs significantly decreased the rate of apoptosis, suggesting that apoptosis was induced by AS101 in B16 cells by triggering the Fas-FasL system (Fig. 7).

It has recently been shown that FasL, but not Fas, is abundantly expressed on human metastatic melanoma lesions (9). In contrast, the majority of immune cells infiltrating the tumor mass are Fas positive. Thus, B16 cells expressing little or no Fas can escape immune surveillance. We therefore hypothesized that the antitumor effects of AS101 could be partly ascribed to breaking the immunologic unresponsiveness to melanoma by altering the expression of Fas on target cells, thus affecting the interactions between tumor cells and the host immune response.

We tested this hypothesis by s.c. injecting B16-F10 mouse melanoma cells into syngeneic C57BL/6 mice carrying the lpr mutation in the Fas gene and into gld mutant mice lacking functional FasL. Mice were treated with either AS101 or PBS. As shown in Figure 8, tumor development among control groups was lowest in the lpr mice as recently reported (9), while no difference was observed between gld and wild-type mice. This points to a crucial involvement of the Fas system in tumor rejection. Among the AS101-treated groups, the most pronounced effect appeared in the lpr mice, while the lowest effect of AS101 was seen in the gld mutant mice.

To differentiate between the effects of AS101 on target and effector cells, B16-F10 melanoma cells were cultured in vitro for 48 h with AS101 and then injected into C57BL/6 syngeneic mice. As shown in Figure 9, a significant decrease in tumor volume occurred in mice transplanted with AS101-treated vs PBS-treated cells. This decrease was significant (p < 0.05; Fig. 9).

Research data suggest that most cells of the immune system express Fas and/or FasL either constitutively or after activation (29, 30, 31, 32, 33). Apart from the many regulatory possibilities between immune cells that this suggests, the immune system also interacts with nonlymphoid tissue via Fas and FasL. A number of tumors have recently been reported to express FasL (7, 8, 9, 10, 11), enabling them potential mechanisms of immune escape. For this reason, modifiers of the sensitivity of neoplastic cells to the immune response have attracted growing attention.

In the present study we present evidence that the immunomodulator AS101 exerts antitumor effects both directly and indirectly. As shown in the clonogenic assays and in the ability of AS101 to induce autocrine apoptosis (Fig. 7), AS101 elicited a cytotoxic response from B16 melanoma cells in a dose-dependent manner. Moreover, injection of AS101-treated B16 cells resulted in decreased tumor development (Fig. 9). These data imply a direct anti-tumor effect of AS101 with no mediation of immune effector cells. AS101 is shown in this study to up-regulate Fas/Apo-1 receptor and protein expression in B16-F10 melanoma cells. Moreover, these receptors were functional, since ligation with anti-Fas Abs resulted in enhanced apoptosis. Furthermore, prolonged incubation with AS101 resulted in enhanced spontaneous apoptosis triggered by the autocrine Fas/FasL system, since inhibition of the latter by anti-FasL Abs significantly decreased the rate of apoptosis.

Although Fas/Apo-1 can be induced in response to cellular activation, expression of Fas/Apo-1 is not predictive of an apoptotic response initiated by specific Ab engagement (4). The dissociation of Fas/Apo-1 function and expression was observed in normal and malignant lymphocytes (34) as well as in nonhemopoietic tumors (2, 4). Therefore, the ability of AS101 to up-regulate functional Fas receptor expression in B16-F10 melanoma cells is crucial both for maintaining efficient immune surveillance on B16 tumors and for inducing autocrine apoptosis by the Fas/FasL system.

The in vivo consequences of AS101-induced up-regulation of Fas expression by B16 melanoma cells were directly investigated in this study using lpr mice (expressing minimal or no Fas) compared with normal or gld mice (lacking functional FasL). The results of those experiments imply both a direct and an indirect antitumor effect of AS101. The strongest antitumor effect of AS101 was observed in lpr mice and was probably the result of the insensitivity of infiltrating lymphocytes to FasL-dependent apoptosis by tumor cells, on the one hand, and increasing the sensitivity of Fas-expressing tumor cells to FasL-dependent apoptosis by infiltrating T cells, on the other.

The weakest effect of AS101 was observed in the gld mutant mice and was probably due to the inability of the effector T cells, lacking FasL, to induce the apoptotic signal to Fas-expressing tumor cells. In addition to the increase in tumor sensitivity to the immune response, this study shows that AS101 could directly induce in vitro an autocrine apoptotic cell death involving the Fas/FasL system. We may therefore postulate that the in vivo antitumor effects of AS101 in wild-type or mutant C57BL/6 mice are partly due to autocrine apoptotic cell death induced by Fas (up-regulated by AS101) and FasL (naturally expressed), which are both present on B16 melanoma cells.

We show that stimulation of B16 cells with AS101 results in up-regulation of p53 protein expression. The mechanisms by which p53 orchestrates diverse cellular programs, ultimately culminating in apoptotic cell death, is the subject of intense investigation. The p53 may repress cellular survival genes such as bcl-2 (35) but may also transcriptionally activate target genes that can mediate apoptosis. One of these target genes is Fas (27). It has recently been demonstrated that direct microinjection of wild-type p53 cDNA into the nucleus of p53 null cells or wild-type p53-containing tumor cells results in up-regulation of Fas/Apo-1 expression (27). Microinjection experiments also showed increased susceptibility toward CD95-mediated apoptosis. In addition, chemotherapeutic drugs have been shown to up-regulate Fas expression in tumor cells via accumulation of wild-type p53 (28). We may therefore postulate that up-regulation by AS101 of p53 expression in B16 melanoma cells could be a major determinant in the induction of Fas/Apo expression and function in these cells. As shown in this study and previously (9), B16 melanoma cells express FasL in abundance. We are currently studying the ability of AS101 to up-regulate FasL expression in tumor cells that do not constitutively express this protein.

p53 is frequently mutated or deleted in human cancers. Other mechanisms of action have previously been proposed to explain the antitumoral effects of AS101.

Our preliminary data show that AS101 may induce the expression of WAF1 mRNA in primary fibroblasts from p53 knockout mice. The WAF1/C1P1 gene that encodes P21 WAF1/C1P1 protein, an inhibitor of cyclin-dependent kinases, is a downstream mediator of p53 function. WAF1/C1P1 was recently identified as one of the wild-type p53 targets that appear to mediate the tumor-suppressing effects of p53 (36). It was recently reported that WAF1/C1P1 gene expression can be regulated through multiple mechanisms that are p53 independent (37, 38). We therefore do not rule out the possibility that AS101’s direct activation of the cell death program is regulated by both p53-dependent or p53-independent pathways.

Aside from the direct effect of AS101 on tumor cells and its ability to increase tumor sensitivity to immune effector actions via the Fas/FasL system, other mediators of AS101’s antitumoral effects have been described. AS101 treatment of cancer patients or tumor-bearing mice led to a clear predominance of Th1 responses, with a concomitant and significant decrease in Th2 responses (19). This effect was shown to be associated with AS101’s antitumoral effects. Cancer patients thus treated showed increased IL-2 and IFN-γ levels and a substantial decrease in IL-4 and IL-10 (19). The concentration of IL-12, a key component in the cytokine network and a potent antitumor agent, was significantly increased in AS101-treated patients who also expressed enhanced levels of LAK and NK cell-mediated cytotoxicity. AS101 was previously also shown to increase the tumoricidal activity of macrophages via nitric oxide production (25). Taken together, it seems that AS101 can selectively activate type 1 responses, whether directly or indirectly, by decreasing Th2 responses, which may thus render it effective in activating NK cells, CTL, and tumorcidal macrophages, which together increase its ability to kill tumor cells.

The differential ability of Th1 and Th2 T cells to express FasL has recently been reported (39). Cloned Th1 cells express high levels of functional FasL, whereas cloned Th2 cells express only low levels (39). Thus, the Fas/FasL pathway appears to be differentially regulated in cells committed to either Th1 or Th2 differentiation. Since, as noted earlier, AS101 treatment causes a predominance in Th1 responses (19), AS101 may successfully up-regulate FasL expression on Th1 effector cells and thus improve immune surveillance by the Fas/FasL system.

The results of our in vivo and in vitro studies suggest that by up-regulating Fas expression on B16 melanoma cells, AS101 can both induce autocrine apoptotic cell death and augment the ability of effector cells to eliminate target tumor cells via the Fas/FasL system. Our study suggests that pharmacologic agents such as AS101 may render melanoma tumor cells more sensitive to Fas/FasL-induced apoptosis, breaking the immunologic unresponsiveness to melanoma, and may provide a complementary approach to treatment of malignant melanoma.

We thank Mr. Uri Karo from the Department of Life Sciences for his assistance with the FACS analysis.