Apoptosis, or programmed cell death, is critical for epidermal homeostasis. Keratinocyte proliferation is balanced by terminal differentiation in the granular layer that leads to nuclear fragmentation (Polakowska et al, 1994) and caspase activation (Weil et al, 1999), hallmarks of apoptosis. Apoptotic keratinocyte cell death also results in the formation of the stratum corneum (Ishida-Yamamoto et al, 1999). In addition to these homeostatic functions in normal skin, keratinocyte apoptosis also represents an important response to environmental injury such as ultraviolet (UV) radiation. Apoptosis of keratinocytes that have sustained UV-induced DNA damage requires p53 and represents a key protective mechanism against squamous cell carcinoma by removing premalignant cells that have acquired mutations (Ziegler et al, 1994).
By contrast, melanocytes represent a static component of the epidermis and the role of apoptosis in basal melanocyte function and melanocytic tumor formation has not been well characterized. Nevertheless, it is generally thought that acquisition of resistance to apoptosis is important in the transition from normal melanocyte to melanoma (Gilchrest et al, 1999). This notion is supported by studies demonstrating apoptosis resistance of melanocytic nevus cells compared with melanocytes in collagen gel (Alanko et al, 1999) and resistance of melanoma cell lines to drug-induced apoptosis (Li et al, 1998). Moreover, melanoma cells in vivo demonstrate low levels of spontaneous apoptosis compared with other tumor cell types (Mooney et al, 1995;Staunton and Gaffney, 1995) and resistance to apoptosis is associated with increased metastasis in animal models of melanoma (Glinsky et al, 1997). The poor prognosis for patients with advanced melanoma generally results from lack of therapeutic response to conventional chemotherapy (Koh, 1991). Most chemotherapeutic drugs function by inducing apoptosis in malignant cells (Fisher, 1994), and dysfunctional apoptosis is the most likely explanation for drug resistance in melanoma (Grossman and Altieri, 2001;Satyamoorthy et al, 2001).
The molecular basis for apoptosis resistance in melanoma has been examined in previous studies, demonstrating in some cases increased expression of apoptotic inhibitors or loss of pro-apoptotic factors that may result in a dysfunctional apoptotic program (reviewed inGrossman and Altieri, 2001). With respect to the "extrinsic" pathway, initiated by binding of tumor necrosis factor or related cytokines (i.e., tumor necrosis factor-related apoptosis-inducing ligand) to cell surface death receptors leading to activation of caspase-8 (Ashkenazi and Dixit, 1998), several groups have reported defective death receptor signaling or downregulation in melanoma cells (Griffith et al, 1998;Thomas and Hersey, 1998;Ugurel et al, 1999). In addition, the caspase-8 inhibitor FLIP is highly expressed in melanoma (Irmler et al, 1997). There has been considerable interest in the intracellular factors regulating the "intrinsic" pathway, initiated by mitochondrial cytochrome c release and leading to activation of caspase-9 (Green and Reed, 1998). The anti-apoptotic Bcl-2, which blocks cytochrome c release (Adams and Cory, 1998), was found increased in melanoma compared with melanocytes in some but not all studies (Cerroni et al, 1995;Collins and White, 1995;Selzer et al, 1998;Tang et al, 1998). Similarly for Bcl-2-related proteins, the anti-apoptotic Bcl-XL and Mcl-1 and the pro-apoptotic Bax have been reported elevated in some but not all melanomas examined (Selzer et al, 1998;Tang et al, 1998). Deletion or inactivation by methylation of Apaf-1, a cofactor for caspase-9 (Zou et al, 1997), has also been described in melanoma (Soengas et al, 2001). It has previously been reported (Grossman et al, 1999a) that Survivin, a newly recognized inhibitor of apoptosis (IAP) protein (LaCasse et al, 1998) that blocks caspase-9 (O'Connor et al, 2000), was found expressed in melanoma but not normal melanocytes. Finally, another IAP protein, Livin/ML-IAP, has also been detected in melanoma cells but not melanocytes (Vucic et al, 2000). With the exception of these particular IAP proteins and Bcl-2 (Morales-Ducret et al, 1995;Kim et al, 2000;van den Wijngaard et al, 2000), however, there is little information on the expression of apoptotic regulators in melanocytes.
In this study we examine the expression of apoptotic regulatory molecules and apoptotic responses in melanocytes, melanoma cells, and keratinocytes. We found that melanocytic and keratinocytic cells exhibit different expression patterns of apoptosis regulatory proteins and differential susceptibilities to various apoptotic stimuli.
Materials and methods
Drugs and chemicals
Cisplatin (Sigma, St Louis, MO) was dissolved in dimethyl formamide and stored at 4°C. Etoposide (Sigma) was dissolved in dimethyl sulfoxide and stored at -20°C. Staurosporine (Sigma) was dissolved in dimethyl sulfoxide and stored at 4°C. 4-tert-butylphenol (4-TBP; Aldrich, Milwaukee, WI) was prepared monthly in 70% ethanol and stored at -20°C.
Antibodies
Rabbit polyclonal antibodies to the following apoptotic regulators were purchased from Santa Cruz Biotechnology (Santa Cruz, CA): Bcl-2 (n-19, sc-492), Mcl-1 (S-19, sc-819), Bcl-XL (H-62, sc-7195), FLIPS/L (H-202, sc-8347), c-IAP-1 (h-83, sc-7943), c-IAP-2 (H-85, sc-7944), and Bax (N-20). Mouse monoclonal anti-hILP/XIAP (clone 48) was obtained from Transduction Laboratories (San Diego, CA). Mouse monoclonal anti-p53 (Ab-6) was obtained from Calbiochem (San Diego, CA). Rabbit anti-serum generated against Apaf-1 (clone 846) and Livin were kindly provided by Dr Yuri Lazebnik (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY) and Dr Gary Kasof (AstraZeneca, Worcester, MA), respectively. Purified rabbit polyclonal antibody recognizing Survivin has been described previously (Grossman et al, 1999a).
Melanocytic cells
Human melanoma lines YUSAC2, YUSIT1, YUGEN8, and LOX (Grossman et al, 1999a) were maintained as described previously. Normal human melanocytes were obtained from discarded foreskins, as approved by the Institutional Review Board at the University of Utah (no. 8476). Foreskins were collected in either keratinocyte growth medium (KGM)-2 (Clonetics Corp., San Diego, CA) or melanocyte medium consisting of Ham's F10 (Gibco BRL, Grand Island, NY) containing 7.5% heat-inactivated fetal calf serum (Gibco), 0.1 mM 3-isobutyl-1-methylxanthine (Sigma), 1 mM dibutyryl cyclic adenosine monophosphate (Sigma), 1 
M sodium orthovanadate (Sigma), 2.5 nM cholera toxin (Sigma), 50 ng phorbol 12-myristate 13-acetate per ml (Sigma), and 1% penicillin-streptomycin-glutamine (Gibco-BRL). After scraping away excess fat and subcutaneous tissue, foreskins were floated dermis side down in 25 mg Dispase II per ml (Roche Pharmaceuticals, Nutley, NJ) in phosphate-buffered saline (PBS) at 4°C overnight. The epidermis was lifted off, placed in trypsin–ethylenediamine tetraacetic acid solution (0.05% trypsin, 0.53 mM ethylenediamine tetraacetic acid; Gibco-BRL) at 37°C, and occasionally agitated for 10 min. An equal volume of melanocyte medium was then added, epidermal fragments were carefully removed, and the cell suspension was pelleted, resuspended in melanocyte medium, and then cultured in a humidified incubator with 5% CO2 at 37°C. Over the first 48 h the cell suspension was triturated twice daily to break up cell clumps and improve cell attachment. The medium was changed every 2–3 d. After 1 wk, 0.4 mg G418 sulfate per ml (Gibco-BRL) and an equal volume of 1 M sodium hydroxide were added to the medium for 48 h to eliminate potentially contaminating fibroblasts. Melanocytes were pigmented and reacted with antibody to the melanocyte markers S-100 and Melan-A (Dako, Carpinteria, CA) by immunohistochemistry (not shown). The dibutyryl cyclic adenosine monophosphate was found dispensable for melanocyte growth, and was omitted from the medium in later experiments. Early passage (1–3) melanocytes were seeded at 3
105 per 35 mm dish (Falcon 353001) 1 d prior to exposure to apoptosis-inducing agents.
Keratinocytic cells
The human keratinocyte line HaCat was maintained as described previously (Grossman et al, 1999b). For the preparation of normal human keratinocytes, foreskins were collected in KGM-2 medium and an epidermal suspension was obtained by trypsinization as above. Cells from two foreskins were pooled to increase yield and plating density. After diluting 1: 2 in soybean trypsin inhibitor (10 mg per ml in PBS, Gibco-BRL), keratinocytes were pelleted, washed once in KGM-2 medium, and transferred in KGM-2 to flasks (for lysate preparation) or 35 mm dishes for apoptosis experiments. Cells were incubated as above, and medium was changed every 2–3 d until cells were used at 30–70% confluency.
UVB irradiation
Immediately prior to irradiation, cells were washed with Dulbecco's PBS (Gibco-BRL), 1 ml PBS was added, and dishes were uncovered. The irradiation chamber contained four UVB lamps (FS20T12-UVB, National Biological Corporation, Twinsburg, OH) that emit wavelengths between 250 and 420 nm (72.6% UVB, 27.4% UVA, 0.01% UVC), with peak emission at 313 nm, according to the manufacturer. A filter (Kodacel K6808, Eastman Kodak, Rochester, NY) was placed over the dishes to block residual UVC. Dosimetry was determined using a newly calibrated UVB-500C meter (National Biological Corp.). The irradiation rate was approximately 0.2 mJ per m2 per s. After UVB exposure, the medium was replaced and the recovered dishes were returned to the incubator.
Apoptosis detection
Cells exposed to drugs or UVB were recovered by trypsinization (including nonadherents), washed in PBS, resuspended in cold 70% ethanol, and stored on ice. Apoptosis was assessed by DNA content analysis as described previously (Grossman et al, 1999a). Briefly, cells were ethanol-fixed and resuspended in staining solution consisting of 25 g propidium iodide per ml (Sigma), 0.05% Triton X-100, and 100 g RNAse per ml (Gibco-BRL) in PBS. Flow cytometry was performed on a FACScan (Becton Dickinson, San Jose, CA) using Cell Quest software (Becton Dickinson).
Western blotting
Cell lysates were prepared and protein content was measured as described previously (Grossman et al, 1999a). Samples (25 g) were electrophoresed under reducing conditions at 200 V through a 5% stacking and 15% resolving acrylamide gel, and transferred on to Immobilon-P membranes (Millipore, Bedford, MA). Membrane blots were placed in PBS containing 5% nonfat milk and 0.1% Tween-20, and incubated at 4°C overnight. Blots were then incubated 2 h at room temperature in 5% nonfat milk and 0.1% Tween-20 containing primary antibodies to Bcl-2 (0.2 g per ml), Mcl-1 (0.2 g per ml), Bcl-XL (0.5 g per ml), C-IAP-1 (0.5 g per ml), C-IAP-2 (0.2 g per ml), XIAP (0.25 g per ml), Survivin (0.5 g per ml), Livin (1: 3000 dilution), FLIP (0.5 g per ml), Apaf-1 (1: 3000 dilution), p53 (0.1 g per ml), or Bax (0.3 g per ml). After washing in PBS/Tween, blots were reacted with 1: 10,000 dilution horseradish peroxidase-conjugated goat anti-rabbit or goat anti-mouse immunoglobulin (New England Nuclear, Boston, MA) in blocking solution for 30 min at room temperature. After washing in PBS/Tween, bands were visualized by enhanced chemiluminescence (Perkin-Elmer Life Sciences Inc., Boston, MA) and autoradiography. To confirm equivalent protein loading per lane, blots were subsequently washed and reacted with a 1: 10,000 dilution of (if previously exposed to rabbit antibody) mouse anti-
-actin (clone AC-74, Sigma) or (if previously exposed to mouse antibody) rabbit anti--actin (Sigma), and bands visualized as above.
Results
Expression of apoptosis regulators
We assessed the expression of a panel of apoptotic regulators in human melanocytes, melanoma cells, keratinocytes, and the keratinocyte line HaCat by western blotting. For melanocytes and keratinocytes, lysates were independently prepared from three different donors given the possibility of individual-specific effects. We first examined apoptotic inhibitors of the Bcl-2 family, negative regulators of the mitochondrial apoptotic pathway. As shown in Figure 1(a), Bcl-2 displayed a striking melanocytic-specific pattern with expression in all three melanocyte and four melanoma lines; it was barely detected in two of three keratinocyte lines and absent in HaCat cells. The Bcl-2 homolog Mcl-1 was similarly present in melanocytes and melanoma cells, and absent from keratinocytic cells (Figure 1a). Bcl-XL, by contrast, was seen in all cell types with strongest expression in the melanoma lines (Figure 1a). We then looked at members of the IAP family of caspase inhibitors. Both c-IAP1 and c-IAP2 were detected in all cell types, with strongest expression seen in melanocytes and melanoma cells and relatively weak expression in two of three keratinocyte lines (Figure 1a). Expression of XIAP was greatest in the melanoma lines and absent in HaCat cells, whereas melanocytes and keratinocytes expressed low but comparable levels (Figure 1a). Survivin was generally seen in only the transformed cells (three melanoma lines and HaCat cells) and not in either melanocytes or keratinocytes (Figure 1a), although low levels could be detected in the fourth melanoma line (YUSIT1) and three of seven melanocyte lines on overexposed blots (not shown). Livin, on the other hand was found in all cell types, with relatively weaker expression in two of three keratinocyte lines (Figure 1a). The caspase-8 inhibitor FLIP was highly expressed in keratinocytes, weakly expressed in HaCat cells, and notably absent from both melanocytes and melanoma cells (Figure 1a).
Figure 1.
Expression of apoptotic regulators. Cell lysates (25 g) prepared from subconfluent cultures of melanocytes (MC, lanes 1–3), melanoma lines (YUSIT1, lane 4; YUSAC2, lane 5; YUGEN8, lane 6; LOX, lane 7), keratinocytes (KC, lanes 8–10), and HaCat cells (lane 11) were subjected to western blotting using a panel of antibodies against (A) apoptosis inhibitors and (B) pro-apoptotic molecules as indicated. Approximate molecular weights of reactive species are indicated. (C) Each blot was restained with antibody to actin to confirm equivalent protein loading in each lane. A representative blot is shown.
Next, we examined expression of several pro-apoptotic regulatory molecules. The tumor suppressor p53 was highly expressed in two melanoma lines (YUSAC2, YUGEN8) and relatively weakly expressed in all other cells (Figure 1b). The p53-dependent caspase-9 cofactor Apaf-1 was strongly expressed in melanocytes and melanoma cells, with keratinocytes demonstrating relatively weak expression (Figure 1b). The Bcl-2 family member Bax was notably expressed in three melanocyte lines and YUSIT1 melanoma cells, and absent from keratinocytes (Figure 1b). Equivalent protein loading was confirmed for each blot by staining for actin (Figure 1c). Thus melanocytic cells expressed high levels of anti-apoptotic Bcl-2 and IAP family proteins, and the pro-apoptotic protein Apaf-1, whereas keratinocytic cells expressed high levels of the anti-apoptotic protein FLIP and were relatively deficient in Bcl-2 family proteins. These expression studies are summarized in Table I.
Cell cycle analysis
Given recent insights into the relationship between apoptosis and cell cycle control (Evan and Vousden, 2001), we determined if there were any baseline differences in cell cycle progression or proliferation status among these cell types prior to examining the differential sensitivities of these cell types to various apoptotic stimuli. Analysis of independent subcultures of the cells used in these experiments, however, did not reveal significant differences (ANOVA, p>0.05) in mitotic (G2M) fractions among the melanocytic and keratinocytic cell types (Figure 2). There was more variability in G1 fractions among the cell types, with that of HaCat being relatively lower than the rest (ANOVA, p=0.001), suggesting a higher percentage of HaCat cells may be in S phase.
Figure 2.
Cell cycle analysis. Subconfluent untreated cultures of melanocytes (MC), keratinocytes (KC), HaCat cells, and melanoma lines (YUSIT1, YUSAC2, YUGEN8, LOX) were collected, fixed, stained with propidium iodide, and analyzed by flow cytometry. Data are expressed as percent interphase (G1, open bars) and dividing (G2M, filled bars) cells as indicated. Error bars reflect SEM from six independent subcultures.
Full figure and legend (16K)Apoptotic responses to UVB
UVB radiation is known to be a potent physiologic inducer of apoptosis in keratinocytes, and we confirmed that normal keratinocytes exhibited a obse-dependent (30–120 mJ per cm2) apoptotic response, as determined by nuclear propidium iodide staining and flow cytometry, which was maximal 24 h following UV exposure (Figure 3a). HaCat cells were similarly sensitive to UV-induced apoptosis (Figure 3a). Normal melanocytes, on the other hand, were markedly resistant to UV-induced apoptosis, requiring 120 mJ per cm2 for a maximal response that was not evident until 48 h following exposure (Figure 3a). We also examined four melanoma cell lines under these same conditions. As shown in Figure 3(b), peak apoptotic responses for all melanoma lines were induced at doses (60–120 mJ per cm2) comparable with that required by keratinocytes and HaCat cells, but like melanocytes, were not maximal until 48 h following UV exposure.
Figure 3.
Melanocytes are resistant to UVB-induced apoptosis. (A) Subconfluent cultures of keratinocytes (KC), HaCat cells, and melanocytes (MC) were exposed to increasing doses of UVB as indicated. Cells treated at each dose were collected and fixed either 24 (open circles) or 48 h (filled circles) after UV exposure. Data are expressed as percent apoptotic cells, derived from the sub-G1 fraction obtained by flow cytometry of propidium iodide-stained cells. Error bars reflect SEM from three independent experiments, and cells (KC, MC) were derived from different donor foreskins for each experiment. (B) Subconfluent cultures of melanoma lines YUSIT1, YUSAC2, YUGEN8, and LOX were treated and analyzed as in A above. Error bars reflect SEM from three independent experiments.
Full figure and legend (34K)Apoptotic responses to 4-TBP
Next, we examined apoptotic responses to the tyrosine analog 4-TBP, previously shown to induce apoptosis in melanocytes (Yang et al, 2000). Similarly, we demonstrated that 4-TBP (0.6–0.9 mM) readily induced apoptosis in melanocytes with responses peaking 48 h after initial exposure (Figure 4a). Keratinocytes, on the other hand, were unresponsive (Figure 4a). There was some variability in both melanocyte and keratinocyte responses to 4-TBP, particularly at the highest concentration tested, due to an aberrant response of cells from one of five individuals. HaCat cells were highly sensitive, with maximal responses peaking at 24 h after initial exposure (Figure 4a). As shown in Figure 4(b), the four melanoma lines were acutely sensitive, responding to the lowest concentration (0.3 mM) tested, and apoptotic responses at each dose were generally maximal 24 h following initial exposure.
Figure 4.
Keratinocytes are unresponsive to 4-TBP-induced apoptosis. (A) Subconfluent cultures of melanocytes (MC), keratinocytes (KC), and HaCat cells were incubated with increasing doses of 4-TBP as indicated. Cells treated at each dose were collected and fixed either 24 (open circles) or 48 h (filled circles) later. Percent apoptotic cells were determined as in Figure 3. Error bars reflect SEM from five independent experiments, and cells (KC, MC) were derived from different donor foreskins for each experiment. (B) Subconfluent cultures of melanoma lines YUSIT1, YUSAC2, YUGEN8, and LOX were treated and analyzed as in A above. Error bars reflect SEM from three independent experiments.
Full figure and legend (33K)Drug-induced apoptosis
We then determined the susceptibility of these cell types to several apoptosis-inducing drugs, each thought to initiate apoptosis by distinct mechanisms. Cells were separately incubated with increasing concentrations of drug, and apoptotic responses measured at 24 and 48 h. First, cells were exposed to etoposide (5–50 M), which binds topoisomerase II and blocks DNA religation. Melanocytes, keratinocytes, and two of the melanoma lines (YUSAC2 and YUGEN8) were generally unresponsive, whereas another (LOX) was minimally responsive (Figure 5a). By contrast, the lowest concentration (5 M) readily induced apoptosis in HaCat cells and the melanoma line YUSIT1 (Figure 5a). Second, cells were exposed to the DNA-damaging agent cisplatin (10–90 M). Keratinocytes and one melanoma line (YUSAC2) responded poorly, whereas other cell types exhibited moderate responses peaking 48 h after initial exposure (Figure 5b). Finally, apoptotic responses to staurosporine (0.2–2 M), an inhibitor of phospholipid/calcium-dependent protein kinase, were examined. Melanocytes, keratinocytes, and YUSAC2 melanoma cells responded minimally, whereas all other cell types exhibited robust responses generally peaking 24 h after initial exposure (Figure 5c). These responses are summarized, along with the other apoptotic responses observed in this study, in Table II.
Figure 5.
Drug-induced apoptosis. Subconfluent cultures of (A) melanocytes (MC), (B) keratinocytes (KC), (C) HaCat cells, and (D–G) melanoma lines YUSIT1, YUSAC2, YUGEN8, and LOX were exposed to increasing concentrations of etoposide, cisplatin, or staurosporine as indicated. Cells treated at each dose were collected and fixed either 24 (open circles) or 48 h (filled circles) later. Percent apoptotic cells were determined as in Figure 3. Error bars reflect SEM from two or three independent experiments, and cells (KC, MC) were derived from different donor foreskins for each experiment.
Full figure and legend (40K)Apoptotic regulators during apoptotic stimulation
Finally, we investigated potential relationships in these cells between apoptosis induction and expression of apoptotic regulators. For these experiments, we sought to examine each cell type under conditions in which an apoptotic response would be initiated but not maximal or complete. Based on the responses to apoptotic stimuli described above, we chose to prepare lysates from melanocytes, YUSAC2 melanoma cells, and keratinocytes 24 h after exposure to 95 mJ per cm2 UVB. Lysates were simultaneously prepared from sham-treated cells to serve as controls for baseline levels of apoptotic regulators. Two experiments were performed, using melanocytes and keratinocytes derived from different donors. In melanocytes, UVB-induced apoptotic stimulation was associated with reduced expression of Mcl-1, XIAP, and Livin, and induction of p53 (Figure 6). Keratinocytes similarly exhibited reduced expression of XIAP and Livin, and increased expression of p53. In YUSAC2 cells, UVB led to slightly decreased levels of Mcl-1, XIAP, Livin, and p53 that were associated with the appearance of putative breakdown products of 25–30 kDa (Figure 6). Levels of Bcl-2, Bcl-XL, c-IAP1, c-IAP2, Survivin, FLIP, Apaf-1, and Bax were not appreciably changed upon UVB exposure in any of the cell types (not shown).
Figure 6.
Expression of apoptotic regulators during apoptotic induction. Cell lysates (25 g), prepared from subconfluent cultures of melanocytes (MC), YUSAC2 melanoma cells (YUS), and keratinocytes (KC) 24 h following sham treatment (-) or exposure to 95 mJ per cm2 UVB (+), were subjected to western blotting for apoptosis inhibitors and pro-apoptotic molecules as indicated. Adequate visualization of the Livin blot for keratinocytes required relatively longer exposure time than that for melanocytes and YUSAC2, thus optimal exposures are shown for each. Approximate molecular weights of reactive species are noted. Blots were restained with antibody to actin to confirm equivalent protein loading in each lane. A representative blot is shown. Results are representative of two independent experiments performed.
Discussion
Apoptosis is known to be a key factor in keratinocyte homeostasis and squamous cell carcinoma development, yet the role of cell death in the maintenance and transformation of melanocytes is poorly understood. To define the basic apoptotic programs in these cell types, we have characterized their expression of apoptosis regulatory proteins and their responses to a variety of apoptotic stimuli. Consistent with their different functional roles, we found that melanocytic and keratinocytic cells exhibit differential expression patterns of apoptotic regulators that correlate with different susceptibilities to apoptotic signals.
Several groups (Morales-Ducret et al, 1995;Kim et al, 2000;van den Wijngaard et al, 2000) had previously demonstrated Bcl-2 expression in melanocytes. These observations were confirmed, and we also found that melanocytes expressed relatively high levels of additional Bcl-2 family proteins Bcl-XL and Mcl-1, as well as the pro-apoptotic Bax, whereas keratinocytic cells were relatively deficient in these proteins. Keratinocytic cells, in contrast to melanocytic cells, expressed high levels of the anti-apoptotic protein FLIP. We did not detect significant FLIP expression in either melanocytes or melanoma cells, althoughIrmler et al (1997) reported FLIP expression in both melanoma tumors and cell lines. In the three of four melanoma lines examined, levels of Bcl-2 were comparable with that in melanocytes, whereas both Mcl-1 and Bcl-XL were generally elevated. Bax was highly expressed in only one of four melanoma lines. These findings are consistent with previous expression studies in melanoma lesions on Bcl-2 (Selzer et al, 1998;Tang et al, 1998), Bcl-XL (Selzer et al, 1998;Leiter et al, 2000), Mcl-1 (Tang et al, 1998), and Bax (Selzer et al, 1998;Leiter et al, 2000).Raisova et al (2001) have shown that susceptibility to death receptor signaling in melanoma lines correlates with a high Bax/Bcl-2 ratio. We did not observe such a correlation among the lines studied here with sensitivity to various apoptotic stimuli, but it is intriguing that the LOX melanoma line was responsive to lower doses of UVB and cytotoxic drugs and was the only line that did not express Bcl-2 or Mcl-1. Of all the apoptotic regulators examined, Bax was the only protein consistently expressed in melanocytes and not melanoma cells.
Melanocytic cells also consistently expressed multiple IAP proteins, including C-IAP-1, C-IAP-2, XIAP, and Livin. The c-IAP proteins were expressed at roughly equivalent levels in melanocytes and melanoma cells, whereas XIAP and Livin were found at higher levels in melanoma cells compared with melanocytes. Initial studies (Vucic et al, 2000;Kasof and Gomes, 2001) on Livin (or ML-IAP) noted expression in melanoma cells but not melanocytes (Vucic et al, 2000). Our ability to detect Livin expression in melanocytes may be due to our use of a polyclonal reagent that would be expected to react with both isoforms that have been reported (Ashhab et al, 2001). Consistent with previous work (Grossman et al, 1999a), the only apoptotic regulator that was generally expressed in melanoma cells and not melanocytes was the IAP protein Survivin, suggesting that it may be important at a relatively later stage in melanocyte transformation.
The pro-apoptotic protein p53 was highly expressed in HaCat cells and two of the melanoma lines, but not in either melanocytes or keratinocytes. High levels of p53 generally reflect mutant protein, and it is known that both alleles of the p53 gene are mutated in HaCat cells (Lehman et al, 1993). It is likely that p53 mutation accounts for the high levels detected in YUSAC2 and YUGEN8 melanoma lines. Finally, althoughSoengas et al (2001) reported loss of Apaf-1 in melanoma lesions and 10 of 19 melanoma lines, we found Apaf-1 consistently expressed in both melanocytes and all four melanoma cell lines.
UVB radiation is the primary mutagen involved in nonmelanoma skin cancer (Grossman and Leffell, 1997) and a physiologic apoptotic stimulus. Most premalignant keratinocytes with UVB-induced mutations are eliminated rapidly by apoptosis, or lost slowly through normal epidermal turnover. Consistent with this scenario and previous observations (Jost et al, 2001), UVB produced a rapid apoptotic response in both normal keratinocytes and the keratinocyte line HaCat. By comparison, we found that normal melanocytes were relatively resistant to UVB-induced apoptosis. Notable distinctions between keratinocytes and melanocytes are that the former are present in large numbers and chronically renewed in the epidermis, whereas the latter sparsely populate the epidermis and are thought to be long-lived (Gilchrest et al, 1999). Melanocytes could be viewed as less dispensable than keratinocytes, and thus it makes teleologic sense that they would have a much higher threshold than keratinocytes for UVB-induced apoptosis.
The blunted apoptotic response of melanocytes to UVB was not indicative of resistance to all apoptotic stimuli, as these cells were quite sensitive to 4-TBP-induced apoptosis. We found that normal keratinocytes, on the other hand, were refractory to 4-TBP, consistent with a previous report (Yang et al, 2000).Yang et al (2000) also first reported that melanocyte cytotoxicity induced by 4-TBP represented apoptotic cell death, demonstrating DNA fragmentation and phosphatidylserine relocalization in melanocytes exposed to 4-TBP. Depigmenting agents, such as 4-TBP, were originally proposed to act as substrates for tyrosinase (McGuire and Hendee, 1971). The mechanism by which 4-TBP induces apoptosis in melanocytes, however, is unclear and apparently not mediated by tyrosinase (Yang et al, 2000). The A2b adenosine receptor may be involved, asLe Poole et al (1999) showed it is upregulated in 4-TBP-treated melanocytes and not keratinocytes.
A curious feature of both UVB and 4-TBP apoptotic responses is that melanoma and HaCat cells were more sensitive, respectively, than normal melanocytes and keratinocytes. This was somewhat unexpected, as we would have predicted that the malignant cells would be more resistant to these apoptotic stimuli than the untransformed cells from which they were derived. One possibility is that long-term culture of cell lines selects for clones with a proliferative advantage in vitro that may be attained at the expense of apoptotic resistance. The mechanistic basis for increased susceptibility to UVB/4-TBP-induced apoptosis in melanoma and HaCat cells is unclear, however, as we did not consistently find increased expression of pro-apoptotic molecules or decreased expression of apoptotic inhibitors compared with normal melanocytes and keratinocytes. Similarly, HaCat cells were consistently more sensitive to drug-induced apoptosis than normal keratinocytes. When comparing drug responses of the melanoma cell lines with normal melanocytes, on the other hand, no consistent pattern emerged. For example, two of four melanoma lines responded to etoposide, whereas melanocytes were unresponsive; melanocytes responded better to cisplatin than two of the four melanoma lines tested; and finally, three of the four melanoma lines responded better to staurosporine than did melanocytes. The susceptibility of these cell types to drug-induced apoptosis was not related to the number of proliferating cells in culture, as they displayed comparable G2M fractions despite very disparate responses.
The expression and activity of apoptotic regulatory proteins maintains a critical balance in cells, with anti-apoptotic and pro-apoptotic regulators potentially exerting influence at multiple control points in apoptotic pathways. Although melanocytes and keratinocytes displayed almost reciprocal expression patterns for many of these proteins, we found that apoptotic stimulation had similar effects in both. Specifically, UVB exposure in both cell types was associated with downregulation of two apoptotic inhibitors (XIAP and Livin) and upregulation of a pro-apoptotic molecule (p53). Thus these changes are consistent with the known functional activities of these particular regulatory proteins. It is intriguing that only selected members of the IAP family, namely XIAP and Livin, were altered, whereas others such as the C-IAPs and Survivin were not. Similarly, it is unclear why apoptotic stimulation of melanocytes, for example, would result in selective reduction in Mcl-1 without affecting other Bcl-2 family proteins. It is interesting that in YUSAC2 cells, UVB exposure was associated with what appears to be limited proteolysis or cleavage of these regulators, as was recently described for XIAP in tumor necrosis factor-related apoptosis-inducing ligand-stimulated melanoma cells (Zhang et al, 2001). It is also possible that the lower band in these blots may represent induction of smaller cross-reactive proteins.
In conclusion, the differences reported here between melanocytes and keratinocytes, with respect to expression of apoptotic regulators and apoptotic responses, suggest that these cell types are endowed with different apoptotic programs to meet different functional needs in the skin. The relative deficiency of apoptotic inhibitors in keratinocytes may maintain a low apoptotic threshold that is required to sustain rapid turnover and efficiently remove damaged cells. The broad expression of apoptotic inhibitors in melanocytes, on the other hand, may serve to maintain their longevity, but this could come at the expense of retaining damaged cells with the potential of developing into melanoma. Our data do not support the notion of reduced susceptibility to apoptosis in the progression from melanocyte to melanoma, but rather suggest that the apoptosis resistance characteristic of melanoma clinically may merely reflect an anti-apoptotic state that is well established in precursor melanocytes. The ability to precipitate apoptosis, however, in malignant melanocytes and keratinocytes by expressing p53 (Yamashita et al, 2001) or Apaf-1 (Soengas et al, 2001), or by blocking individual apoptosis inhibitors of either the Bcl-2 (Olie et al, 2002) or IAP (Grossman et al, 1999a;Grossman et al, 1999b;Kasof and Gomes, 2001) families, suggests that the apoptotic balance in these cells is easily perturbed and thus could potentially be exploited for therapeutic gain.
References
| 1. | Adams JM & Cory S. The Bcl-2 protein family: arbiters of cell survival. Science (1998) 281: 1322–1326. | Article | PubMed | ISI | ChemPort | |
| 2. | Alanko T, Rosenberg M & Saksela O. FGF expression allows nevus cells to survive in three-dimensional collagen gel under conditions that induce apoptosis in normal human melanocytes. J Invest Dermatol (1999) 113: 111–116. | Article | PubMed | ISI | ChemPort | |
| 3. | Ashhab Y, Alian A, Polliack A, Panet A & Yehuda DB. Two splicing variants of a new inhibitor of apoptosis gene with different biological properties and tissue distribution pattern. FEBS Lett (2001) 495: 56–60. | Article | PubMed | ISI | ChemPort | |
| 4. | Ashkenazi A & Dixit VM. Death receptors: signaling and modulation. Science (1998) 281: 1305–1308. | Article | PubMed | ISI | ChemPort | |
| 5. | Cerroni L, Soyer HP & Kerl H. bcl-2 protein expression in cutaneous malignant melanoma and benign melanocytic nevi. Am J Dermatopathol (1995) 17: 7–11. | PubMed | ISI | ChemPort | |
| 6. | Collins KA & White WL. Intercellular adhesion molecule 1 (ICAM-1) and bcl-2 are differentially expressed in early evolving malignant melanoma. Am J Dermatopathol (1995) 17: 429–438. | PubMed | ISI | ChemPort | |
| 7. | Evan GI & Vousden KH. Proliferation, cell cycle and apoptosis in cancer. Nature (2001) 411: 342–348. | Article | PubMed | ISI | ChemPort | |
| 8. | Fisher DE. Apoptosis in cancer therapy: crossing the threshold. Cell (1994) 78: 539–542. | Article | PubMed | ISI | ChemPort | |
| 9. | Gilchrest BA, Eller MS, Geller AC & Yaar M. The pathogenesis of melanoma induced by ultraviolet radiation. N Engl J Med (1999) 340: 1341–1348. | Article | PubMed | ISI | ChemPort | |
| 10. | Glinsky GV, Glinsky VV, Ivanova AB & Hueser CJ. Apoptosis and metastasis: increased apoptosis resistance of metastatic cancer cells is associated with the profound deficiency of apoptosis execution mechanisms. Cancer Lett (1997) 115: 185–193. | Article | PubMed | ISI | ChemPort | |
| 11. | Green DR & Reed JC. Mitochondria and apoptosis. Science (1998) 281: 1309–1312. | PubMed | ISI | ChemPort | |
| 12. | Griffith TS, Chin WA, Jackson GC, Lynch DH & Kubin MZ. Intracellular regulation of TRAIL-induced apoptosis in human melanoma cells. J Immunol (1998) 161: 2833–2840. | PubMed | ISI | ChemPort | |
| 13. | Grossman D & Altieri DC. Drug resistance in melanoma: mechanisms, apoptosis, and new potential therapeutic targets. Cancer Metastasis Rev (2001) 20: 3–11. | Article | PubMed | ISI | ChemPort | |
| 14. | Grossman D & Leffell DJ. The molecular basis of nonmelanoma skin cancer. Arch Dermatol (1997) 133: 1263–1270. | Article | PubMed | ISI | ChemPort | |
| 15. | Grossman D, McNiff JM, Li F & Altieri DC. Expression and targeting of the apoptosis inhibitor, survivin, in human melanoma. J Invest Dermatol (1999a) 113: 1076–1081. | Article | PubMed | ISI | ChemPort | |
| 16. | Grossman D, McNiff JM, Li F & Altieri DC. Expression of the apoptosis inhibitor, survivin, in nonmelanoma skin cancer and gene targeting in a keratinocyte cell line. Lab Invest (1999b) 79: 1121–1126. | PubMed | ISI | ChemPort | |
| 17. | Irmler M, Thome M & Hahne M et al. Inhibition of death receptor signals by cellular FLIP. Nature (1997) 388: 190–195. | Article | PubMed | ISI | ChemPort | |
| 18. | Ishida-Yamamoto A, Tanaka H, Nakane H, Takahashi H, Hashimoto Y & Iizuka H. Programmed cell death in normal epidermis and loricrin keratoderma. Multiple functions of profilaggrin in keratinization. J Invest Dermatol Symp Proc (1999) 4: 145–149. | ISI | ChemPort | |
| 19. | Jost M, Gasparro FP, Jensen PJ & Rodeck U. Keratinocyte apoptosis induced by ultraviolet B radiation and CD95 ligation—differential protection through epidermal growth factor receptor activation and Bcl-x(L) expression. J Invest Dermatol (2001) 116: 860–866. | Article | PubMed | ISI | ChemPort | |
| 20. | Kasof GM & Gomes BC. Livin, a novel inhibitor of apoptosis protein family member. J Biol Chem (2001) 276: 3238–3246. | Article | PubMed | ISI | ChemPort | |
| 21. | Kim YG, Kim HJ & Kim DS et al. Up-regulation and redistribution of Bax in ultraviolet B-irradiated melanocytes. Pigment Cell Res (2000) 13: 352–357. | Article | PubMed | ISI | ChemPort | |
| 22. | Koh HK. Cutaneous melanoma. N Engl J Med (1991) 325: 171–182. | PubMed | ISI | ChemPort | |
| 23. | LaCasse EC, Baird S, Korneluk RG & MacKenzie AE. The inhibitors of apoptosis (IAPs) and their emerging role in cancer. Oncogene (1998) 17: 3247–3259. | Article | PubMed | ISI | |
| 24. | Le Poole IC, Yang F, Brown TL, Cornelius J, Babcock GF, Das PK & Boissy RE. Altered gene expression in melanocytes exposed to 4-tertiary butyl phenol (4-TBP). upregulation of the A2b adenosine receptor 1. J Invest Dermatol (1999) 113: 725–731. | Article | PubMed | ISI | ChemPort | |
| 25. | Lehman TA, Modali R & Boukamp P et al. p53 mutations in human immortalized epithelial cell lines. Carcinogenesis (1993) 14: 833–839. | PubMed | ISI | ChemPort | |
| 26. | Leiter U, Schmid RM, Kaskel P, Peter RU & Krahn G. Anti-apoptotic bcl-2 and bcl-xL in advanced malignant melanoma. Arch Dermatol Res (2000) 292: 225–232. | Article | PubMed | ISI | ChemPort | |
| 27. | Li G, Tang L, Zhou X, Tron V & Ho V. Chemotherapy-induced apoptosis in melanoma cells is p53 dependent. Melanoma Res (1998) 8: 17–23. | PubMed | ISI | ChemPort | |
| 28. | McGuire J & Hendee J. Biochemical basis for depigmentation of skin by phenolic germicides. J Invest Dermatol (1971) 57: 256–261. | Article | PubMed | ISI | ChemPort | |
| 29. | Mooney EE, Ruis Peris JM, O'Neill A & Sweeney EC. Apoptotic and mitotic indices in malignant melanoma and basal cell carcinoma. J Clin Pathol (1995) 48: 242–244. | PubMed | ISI | ChemPort | |
| 30. | Morales-Ducret CR, van de Rijn M & Smoller BR. bcl-2 expression in melanocytic nevi. Insights into the biology of dermal maturation. Arch Dermatol (1995) 131: 915–918. | Article | PubMed | ChemPort | |
| 31. | O'Connor DS, Grossman D & Plescia J et al. Regulation of apoptosis at cell division by p34cdc2 phosphorylation of survivin. Proc Natl Acad Sci USA (2000) 97: 13103–13107. | Article | PubMed | ChemPort | |
| 32. | Olie RA, Hafner C & Kuttel R et al. Bcl-2 and bcl-xL antisense oligonucleotides induce apoptosis in melanoma cells of different clinical stages. J Invest Dermatol (2002) 118: 505–512. | Article | PubMed | ISI | ChemPort | |
| 33. | Polakowska RR, Piacentini M, Bartlett R, Goldsmith LA & Haake AR. Apoptosis in human skin development: morphogenesis, periderm, and stem cells. Dev Dyn (1994) 199: 176–188. | PubMed | ISI | ChemPort | |
| 34. | Raisova M, Hossini AM & Eberle J et al. The Bax/Bcl-2 ratio determines the susceptibility of human melanoma cells to CD95/Fas-mediated apoptosis. J Invest Dermatol (2001) 117: 333–340. | Article | PubMed | ISI | ChemPort | |
| 35. | Satyamoorthy K, Bogenrieder T & Herlyn M. No longer a molecular black box—new clues to apoptosis and drug resistance in melanoma. Trends Mol Med (2001) 7: 191–194. | Article | PubMed | ISI | ChemPort | |
| 36. | Selzer E, Schlagbauer-Wadl H, Okamoto I, Pehamberger H, Potter R & Jansen B. Expression of Bcl-2 family members in human melanocytes, in melanoma metastases and in melanoma cell lines. Melanoma Res (1998) 8: 197–203. | PubMed | ISI | ChemPort | |
| 37. | Soengas MS, Capodieci P & Polsky D et al. Inactivation of the apoptosis effector Apaf-1 in malignant melanoma. Nature (2001) 409: 207–211. | Article | PubMed | ISI | ChemPort | |
| 38. | Staunton MJ & Gaffney EF. Tumor type is a determinant of susceptibility to apoptosis. Am J Clin Pathol (1995) 103: 300–307. | PubMed | ISI | ChemPort | |
| 39. | Tang L, Tron VA & Reed JC et al. Expression of apoptosis regulators in cutaneous malignant melanoma. Clin Cancer Res (1998) 4: 1865–1871. | PubMed | ISI | ChemPort | |
| 40. | Thomas WD & Hersey P. TNF-related apoptosis-inducing ligand (TRAIL) induces apoptosis in Fas ligand-resistant melanoma cells and mediates CD4 T cell killing of target cells. J Immunol (1998) 161: 2195–2200. | PubMed | ISI | ChemPort | |
| 41. | Ugurel S, Seiter S, Rappl G, Stark A, Tilgen W & Reinhold U. Heterogeneous susceptibility to CD95-induced apoptosis in melanoma cells correlates with bcl-2 and bcl-X expression and is sensitive to modulation by interferon-gamma. Int J Cancer (1999) 82: 727–736. | Article | PubMed | ISI | ChemPort | |
| 42. | Vucic D, Stennicke HR, Pisabarro MT, Salvesen GS & Dixit VM. ML-IAP, a novel inhibitor of apoptosis that is preferentially expressed in human melanomas. Curr Biol (2000) 10: 1359–1366. | Article | PubMed | ISI | ChemPort | |
| 43. | Weil M, Raff MC & Braga VM. Caspase activation in the terminal differentiation of human epidermal keratinocytes. Curr Biol (1999) 9: 361–364. | Article | PubMed | ISI | ChemPort | |
| 44. | van den Wijngaard RM, Aten J, Scheepmaker A, Le Poole IC, Tigges AJ, Westerhof W & Das PK. Expression and modulation of apoptosis regulatory molecules in human melanocytes: significance in vitiligo. Br J Dermatol (2000) 143: 573–581. | Article | PubMed | ISI | ChemPort | |
| 45. | Yamashita T, Tokino T, Tonoki H, Moriuchi T, Jin HY, Omori F & Jimbow K. Induction of apoptosis in melanoma cell lines by p53 and its related proteins. J Invest Dermatol (2001) 117: 914–919. | Article | PubMed | ISI | ChemPort | |
| 46. | Yang F, Sarangarajan R, Le Poole IC, Medrano EE & Boissy RE. The cytotoxicity and apoptosis induced by 4-tertiary butylphenol in human melanocytes are independent of tyrosinase activity. J Invest Dermatol (2000) 114: 157–164. | Article | PubMed | ISI | ChemPort | |
| 47. | Zhang XD, Zhang XY, Gray CP, Nguyen T & Hersey P. Tumor necrosis factor-related apoptosis-inducing ligand-induced apoptosis of human melanoma is regulated by Smac/DIABLO release from mitochondria. Cancer Res (2001) 61: 7339–7348. | PubMed | ISI | ChemPort | |
| 48. | Ziegler A, Jonason AS & Leffell DJ et al. Sunburn and p53 in the onset of skin cancer. Nature (1994) 372: 773–776. | Article | PubMed | ISI | ChemPort | |
| 49. | Zou H, Henzel WJ, Liu X, Lutschg A & Wang X. Apaf-1, a human protein homologous to C. elegans CED-4, participates in cytochrome c-dependent activation of caspase-3. Cell (1997) 90: 405–413. | Article | PubMed | ISI | ChemPort | |
Acknowledgments
We thank Drs Yuri Lazebnik for Apaf-1 antibody and Gary Kasof for Livin antibody. We also thank Dr Sancy Leachman for critically reading the manuscript. Supported by NIH grant KO8AR48618, the Huntsman Cancer Foundation, and a Fellowship-to-Faculty Transition Award from the University of Utah funded in part by the Howard Hughes Medical Institute.
