Mannosylerythritol Lipid Is a Potent Inducer of Apoptosis and Differentiation of Mouse Melanoma Cells in Culture¹

The microbial extracellular glycolipid known as MEL³ is a biosurfactant composed of both lipophilic and hydrophilic moieties. MEL is produced in large amounts by the yeast Candida antarctica T-34 when this microorganism is grown on myristic acid as a source of carbon. The hydrophilic moiety of MEL was identified as 4-O-(di-O-acetyl-di-O-alkanoyl-β-d-mannopyranosyl)-erythritol (1). Recently, we reported that MEL induces the differentiation in granulocytes of HL-60 promyelocytic leukemia cells (2).

Gangliosides and GSLs, which are ubiquitous constituents of the plasma membranes of mammalian cells, were recently found to be active in the modulation of cell proliferation in oncogenesis and differentiation (3, 4). Cellular differentiation and oncogenic transformation are accompanied by dramatic changes in absolute and relative levels of GSLs (5). Moreover, ceramide, the product of the hydrolysis of sphingomyelin in the sphingolipid cycle, was recently reported to induce apoptosis in human neuroepithelioma cells (6). The accumulation of ceramide also seems to be associated with several antiproliferative responses, which include the differentiation of cells, apoptosis, and cell cycle arrest (7, 8). Mammalian and microbial glycolipids differ in the details of specific residues, but their backbones are similar. Thus, we postulated that MEL might have biological effects and most likely has antiproliferative effects on tumor cells in addition to inducing cell differentiation.

Apoptosis has become a focus of attention in studies of the biology of cancer cells, and it has been proposed that the progression of a tumor might not only be a function of cell proliferation but might also be a product of the aberrant survival of cells that results from the inappropriate suppression of apoptosis (9). Various compounds, including glucocorticoids and cytostatic drugs, as well as the withdrawal of hormones, serum, or individual growth factors (10) can trigger apoptotic events. The product of the proto-oncogene bcl-2 protects many cell lines from apoptosis (11) induced by extracellular agents and certain responses to stress (12, 13, 14). It was reported that bcl-2 interrupts the ceramide-mediated apoptosis of pre-B leukemia cells (15). To our knowledge, there are no reports to date of the induction of an apoptotic response in mammalian cells by microbial extracellular glycolipids.

In the present study, we found that the treatment of malignant melanoma B16 cells with MEL resulted in a profound, dose-dependent inhibition of growth. MEL was also found to be a potent inducer of both apoptosis and differentiation in B16 cells, inducing significant tyrosinase activity and enhanced production of melanin. Overexpression of human bcl-2 conferred resistance to MEL-induced apoptosis but further enhanced the expression of the differentiation-associated markers. These findings might provide the groundwork for the use of microbial extracellular glycolipids as novel therapeutic reagents in the treatment of melanoma.

MEL was prepared and purified essentially as described by Kitamoto et al. (1).

Mouse melanoma B16 4A5 cells (referred to hereafter as B16 cells) were obtained from the RIKEN Cell Bank (Tsukuba, Ibaraki, Japan) and maintained in culture in DMEM supplemented with 10% fetal bovine serum or in serum-free DMEM-ITES (insulin, transferrin, ethanolamine, and selenite) medium as described elsewhere (2). A cell counting kit (WST-1; Dojin Laboratories, Kumamoto, Japan) was used to monitor the number of viable cells.

B16 cells cultured in the presence or absence of MEL were collected and washed with PBS without Mg²⁺ and Ca²⁺ ions. For the detection of chromatin condensation, cells were fixed in formalin, stained with 100 μm Hoechst 33342 (Sigma Chemical Co., St. Louis, MO), and examined with a fluorescence microscope (Olympus, Inc., Tokyo, Japan). Low molecular weight chromosomal DNA from apoptotic cells was purified and subjected to electrophoresis on a 2% agarose gel as described elsewhere (16).

B16 cells (1 × 10⁶ cells) cultured with the indicated concentration of MEL were washed with PBS, fixed with 70% ethanol, resuspended in 1 ml of PBS, and stained with PI (50 μg/ml) for 30 min at room temperature in darkness. Cells were then filtered through nylon mesh (35 μm) and used to study apoptosis. For the analysis of the cell cycle, BrdUrd (Sigma Chemical Co.) was added at a final concentration of 5 μg/ml at indicated times, and cells were incubated for 30 min (17), fixed in 70% ethanol, and stored at −20°C. Before the next step in the analysis, samples were treated with 0.1% (w/v) RNase A and 4 n HCl for 15 min at 37°C and 20 min at room temperature, respectively, and suspended in 1 ml of 0.1 m Na₂B₄0₇. After microcentrifugation, cells were resuspended in 1 ml of 0.5% (v/v) Tween 20 in PBS and washed repeatedly until the supernatant reached a final pH of 7.0. After the addition of 20 μl of a solution of FITC-conjugated BrdUrd-specific antibodies and incubation at room temperature for 20 min, cells were washed twice with PBS and stained with PI (5 μg/ml) at 4°C for 15 min. Cells were filtered through nylon mesh (35 μm), and flow cytometric analysis was performed with a FACSort flow cytometer (Becton Dickinson, San Jose, CA). Data from a minimum of 10,000 cells were collected and analyzed with LYSIS II software (Becton Dickinson).

B16 cells were cotransfected with the pCAGGS vector (18) that included the full-length human bcl-2 gene and pcDNA3, which conferred resistance to Geneticin (G418; Life Technologies, Inc., Rockville, MD). Control cells were cotransfected with pcDNA3 and pCAGGS without the bcl-2 gene. After selection with 500 μg/ml G418, the expression of bcl-2 was confirmed by reverse transcription-PCR (one-step reverse transcription-PCR kit; Toyobo Co., Osaka, Japan) using total RNA as template and primers 5′-ACGCTGGGAGAACGGGGTAC-3′ (sense primer) and 5′-GCGGCTGTATGGGGCGTGTG-3′ (antisense primer), which are specific for human bcl-2. Whole cell extracts were prepared, and Western blots were carried out to verify the expression of bcl-2 protein using the antibodies against human bcl-2 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA).

Tyrosinase activity was measured as described by Shoji et al. (19). One unit of tyrosinase activity was defined as the activity that caused an increase in absorbance at 280 nm of 0.001/min under the condition of the reaction. The melanin content of cells was determined as described by Johnston et al. (20).

We examined the effects of MEL on the proliferation and viability of mouse melanoma B16 cells. We incubated B16 cells with 2.5–10 μm MEL and monitored changes in the number of viable cells during a 3-day incubation in the absence of serum (Fig. 1). At 2.5 μm, MEL had no significant inhibitory effects on cell proliferation. At 5 μm, MEL suppressed growth significantly during the exposure of cells to MEL for 48–72 h, with a slight effect on cell viability as determined by the trypan blue exclusion test. At 10 μm, MEL completely blocked cell proliferation, with cytotoxicity becoming apparent after exposure to MEL for 24 h. In the presence of serum, the growth-inhibitory effect of MEL was seen only at a concentration of MEL higher than that at which the same suppression of growth was observed in the absence of serum (data not shown).

We next examined whether the accumulation of dead cells that occurred in response to 10 μm MEL was a result of apoptotic cell death. Because apoptosis can be identified by numerous phenomena, none of which by itself is a sufficient criterion, we used three different types of analysis to detect changes in the condensation of chromatin, DNA structure, and DNA content. Staining cells with Hoechst 33342 revealed nuclear condensation and the fragmentation of cells after treatment with 10 μm MEL for 24 h, whereas untreated cells appeared normal (Fig. 2,A). Electrophoresis on an agarose gel of genomic DNA revealed the fragmentation of DNA in MEL-treated cells that was typical of apoptotic nuclei (Fig. 2,B). Finally, flow cytometric analysis revealed a prominent sub-G₁ peak, which is a hallmark of cells that are undergoing apoptosis (21, 22), in the case of cells treated with 12.5 μm MEL, but not in the case of cells treated with 2.5 μm MEL (Fig. 2 C). Taken together, these results demonstrated that MEL acted as a potent trigger of apoptotic cell death.

We next examined whether the MEL-induced interruption of growth and apoptosis were accompanied by an accumulation of cells in one or more specific phases of the cell cycle. As shown by the result of our flow cytometric study (Fig. 3), when cells were exposed to 10 μm MEL for 24 h, there was a significant decrease in the size of cell populations in the S phase and G₂-M phases (from 30% to 7% and from 9% to 6%, respectively) and a marked increase in the relative number of cells in the G₀-G₁ phase (from 52% to 84%). Thus, the antiproliferative effect of MEL was partly attributable to the induction of cell cycle arrest at the G₀-G₁ phase of the cycle.

In an attempt to gain some insight into the molecular pathways that lead to cell death in response to MEL, we induced the overexpression of bcl-2 in B16 cells by stably transfecting them with full-length cDNA for human bcl-2 and then challenged the cells with MEL. Expression of bcl-2 protein in bcl-2-transfected B16 cells is shown in Fig. 4,A. B16 cells that overexpressed bcl-2 proliferated much more rapidly than B16 cells transfected with the empty control vector (vector only) when treated with 10 μm MEL for 24 and 48 h (Fig. 4, B and C). In addition, ectopic expression of bcl-2 effectively blocked MEL-induced apoptosis as assessed in terms of the number of cells in the sub-G₀-G₁ peak after flow cytometry (Fig. 5).

We recently demonstrated the differentiation-inducing potential of MEL in leukemia cells (2). Differentiation of melanoma cells toward melanocytes can be followed by monitoring the accumulation of melanin and increased tyrosinase activity (23). As shown in Fig. 6,A, the tyrosinase activity of B16 cells that had been treated with 10 μm MEL for 24 h was about three times that of untreated cells. Under the same conditions, we detected a slight but significant increase in melanin content (Fig. 6,B) and in the number of phenotypically melanin-positive cells (Fig. 6,C, top right panel). A more profound effect was observed when MEL-induced differentiation was examined in cells that expressed bcl-2 [Fig. 6, A, B, and C (bottom right panel)]. Our results suggesting that cells had escaped MEL-induced apoptosis as a result of the protective expression of bcl-2 and had undergone further differentiation indicated that MEL can trigger both cell death and differentiation.

In this study, we found that MEL had a dose-dependent antiproliferative effect on mouse B16 melanoma cells in serum-free medium (see Fig. 1). Such an inhibition of cell growth was also observed in medium supplemented with 10% fetal bovine serum (data not shown). The growth-suppressive effect of MEL on B16 cells was significantly greater at 5 μm MEL than at 2.5 μm MEL. The CMC of MEL in aqueous solution is 2.7 μm (24). Thus, the concentration at which MEL was effective in inhibiting the growth of B16 cells was greater than the CMC. Therefore, it seems likely that MEL formed multiple micelles in solution at the effective concentration. Biosurfactants have relatively low CMCs and high surface tension. Their interfacial tension-lowering actions can be attributed to their excellent ability to orient themselves at interfaces (25). MEL is a glycolipid-type biosurfactant; however, the mechanistic links between the formation of micelles of MEL on cell membranes and the observed biological effects have yet to be clarified. One possible mode of action of MEL might involve qualitative and/or quantitative changes of membrane-associated compounds, because we have shown previously that the MEL-induced differentiation of HL-60 cells is accompanied by changes in the composition of cell-surface GSLs (2).

To our knowledge, this report is the first to demonstrate the induction of apoptosis in cancer cells by a microbial extracellular glycolipid. The condensation of chromatin, DNA fragmentation, and a sub-G₁ peak during flow cytometric analysis demonstrated clearly that B16 cells underwent apoptosis on exposure to concentrations of ≥10 μm MEL. Analysis of the cell cycle also showed clearly that upon treatment with 10 μm MEL, B16 cells accumulated at the G₁ phase of the cell cycle within 24 h. This result suggests that MEL interfered with proliferation and induced apoptosis in B16 cells in close association with the G₁ arrest. MEL (10 μm) stimulated tyrosinase activity and the production of melanin (Fig. 6), which are markers of the differentiation of melanoma cells. We also found that MEL could trigger the differentiation of B16 cells, and that only a limited number of cells had apoptotic features when the concentration of MEL was 3–5 μm.⁴ Thus, MEL had various effects on B16 melanoma cells, inducing cell cycle arrest and apoptosis as well as differentiation. Such bioactive effects of MEL were dose dependent, and different effects were distinctly dominant at different concentrations of MEL.

By introducing a bcl-2 expression plasmid into B16 cells, we found that the extent of MEL-induced apoptosis in melanoma cells was diminished by bcl-2. It has been reported that bcl-2 protects against multiple signals that lead to cell death, suggesting that bcl-2 regulates a common cell death pathway and functions at a point where various signals converge (26). The results of the present study support the existence of a MEL-induced pathway of programmed cell death that acts upstream of the commitment point defined by the action of bcl-2. Similar pathways to apoptosis have been reported in response to a number of extracellular agents, such as dexamethasone, chemotherapeutic agents, and tumor necrosis factor α, as well as the activation of Fas and irradiation (12, 13, 14). The mechanism(s) by which MEL acts to induce apoptosis in B16 cells remains to be elucidated. Several candidates for direct targets of MEL can be proposed, including, for example, an isoform of PKC (PKCα) whose expression was found to be enhanced when B16 cells were exposed to 5 μm MEL for 2 days.⁴ There are contradictory reports about the role of PKC in apoptosis. The effect of the activation of PKC is often antiapoptotic, but PKC does initiate apoptosis in a few types of cells (27, 28). MEL-induced apoptosis of B16 cells might also be associated with mitogen-activated protein kinase, because we found that stimulation of PC12 cells with 5 μm MEL caused the temporary activation and enhanced phosphorylation of p42/p44 mitogen-activated protein kinase.⁵ Furthermore, the biosynthetic pathway to membrane GSLs might be an additional target for MEL. bcl-2-transfected B16 cells had a more differentiated phenotype than control cells after both had been exposed to an apoptosis-inducing concentration of MEL. However, the relationship between MEL-induced apoptosis and differentiation is still unclear. It is possible that B16 cells might activate different dominant pathways that are regulated by related signal cascades in response to specific concentrations of MEL.

In conclusion, the microbial glycolipid MEL induced apoptosis in melanoma cells. Enhanced expression of bcl-2 interfered with MEL-mediated apoptosis and stimulated the MEL-induced differentiation of B16 cells. Additional studies are needed to elucidate the precise mechanism(s) of action of MEL in apoptosis, G₁ arrest, and cell differentiation.

We thank Dr. Y. Tsujimoto for the gift of bcl-2 expression vector and for helpful discussions.