Radiation resistance of human melanoma analysed by retroviral insertional mutagenesis reveals a possible role for dopachrome tautomerase

Article metrics

Abstract

While melanomas are resistant to the cytotoxic effects of radiotherapy, little is known about the molecular mechanisms underlying this intrinsic resistance. Here, we describe the utilization of retroviral insertional mutagenesis to facilitate the analysis of genetic changes that are associated with radioresistance in human melanoma. A radial growth phase human melanoma cell line, WM35, was infected with a replication-defective amphotropic murine retrovirus and subsequently selected for X-ray radiation-resistant variants. Several radiation-resistant clones were independently isolated and characterized. Interestingly, these clones also displayed resistance to ultraviolet radiation and to the chemotherapeutic drug cis-diamminedichloroplatinum(II) (CDDP). By Northern and Western analyses, we showed that the expression of DOPAchrome tautomerase (DCT), also known as tyrosinase-related protein 2 (TYRP2), an enzyme that functions in eumelanin synthesis, was significantly elevated in the radiation-resistant clones relative to the parental WM35 cells. Moreover, the levels of DCT in a variety of human melanoma cell lines correlated with their relative levels of radioresistance and the enforced expression of DCT conferred increased resistance to UV(B) treatment. An analysis of stress signaling induced by radiation as well other cytotoxic stressors showed that resistance associated with DCT overexpression applied specifically to treatments that activate the ERK/MAPK pathway. Indeed, DCT overexpression in a melanoma cell line resulted in increased ERK activity. Moreover, ectopic expression of dominant-active MEK in this melanoma cell line conferred UV(B) resistance suggesting that the ERK/MAPK pathway downstream of DCT may play a critical role in radiation and drug resistance. Overall, given that each gamma- and UV(B)-resistant cell line also exhibited resistance to CDDP and that CDDP-resistant clones showed increased resistance to UV(B) irradiation, these results suggest a common mechanism underlying radio- and chemoresistance, which is mediated by DCT expression.

Introduction

A hallmark of malignant melanoma is its high degree of intrinsic resistance to treatment with virtually all conventionally used chemotherapeutic drugs as well as radiation. This represents the principal obstacle in the clinical management of this disease, the molecular basis of which is largely unknown.

With the goal of identifying candidate genes associated with intrinsic drug resistance in melanomas, we previously reported the utilization of retroviral insertional mutagenesis to generate mutant melanoma cell lines displaying resistance to the chemotherapeutic agent cis-diamminedichloroplatinum(II) (CDDP) (Lu et al., 1995; Chu et al., 2000). The rationale of this method is based on the random integration of proviruses into the host cell genome, which results in the activation or inactivation of cellular genes, depending on the site(s) of proviral integration. Subsequently, mutant cells expressing a dominantly selectable phenotype, such as increased drug resistance, can be selected and analysed for common sites of proviral integration. This approach has previously been successfully utilized to identify several genes associated with malignant transformation and tumor progression of murine cells including Fli-1, Spi-1, p53 and NF-E2/p45 (Mowat et al., 1985; Ben-David et al., 1988; Moreau-Gachelin et al., 1988; Ben-David and Bernstein, 1991; Ben-David et al., 1991; Lu et al., 1994). A comparison of global gene expression between retrovirus-derived CDDP-resistant melanoma cells and their drug-sensitive parental cells, WM35, revealed that the transcript encoding DOPAchrome tautomerase (DCT), also known as tyrosinase-related protein 2 (TYRP2), was upregulated in each of the CDDP-resistant mutants analysed (Chu et al., 2000). These results were confirmed by Northern and Western blot analyses.

DCT is a well-characterized melanocyte-specific enzyme that, in conjunction with tyrosinase and tyrosinase-related protein 1 (TYRP1), functions in the conversion of L-tyrosine to the pigment melanin in melanocytes (Jara et al., 1990; Hearing and Tsukamoto, 1991; Cooksey et al., 1997; Olivares et al., 2001). DCT specifically catalyses the tautomerization of the melanin precursors L-DOPAchrome to 5,6-dihydroindole-2-carboxylic acid (DHICA) (Tsukamoto et al., 1992), which is subsequently oxidized by TYRP1 to form the black/brown pigment, eumelanin.

In addition to demonstrating that DCT was upregulated in the virally derived CDDP-resistant clones, we showed that constitutive levels of DCT expression in a variety of independently derived human melanoma cell lines closely correlated with their relative levels of CDDP resistance (Chu et al., 2000; Pak et al., 2000). This correlation applied only to DCT, and was independent of the expressions of tyrosinase, TYRP1 and of total cellular melanin content (Pak et al., 2000). Resistance in DCT overexpressing cells was not restricted to CDDP, but also included several other chemotherapeutic drugs such as carboplatin and methotrexate, both of which induce cytotoxicity by inducing DNA damage. Interestingly, however, no correlation was observed between DCT expression and resistance to the microtubule stabilizer, taxol (Chu et al., 2000). Based on these results, we postulated that DCT may represent a novel mediator of drug resistance in melanoma cells, with specificity for DNA-damaging agents. This postulate was supported by studies in which the enforced expression of DCT in WM35 cells by transfection rendered these cells more resistant to CDDP treatment (Chu et al., 2000). Since DCT has frequently been reported to be highly expressed in melanomas, this melanocyte-specific enzyme may play an important role contributing to intrinsic resistance phenotype of melanomas to various anticancer DNA-damaging drugs.

In this study, we utilized retroviral insertional mutagenesis to generate radiation-resistant variants of the WM35 melanoma cell line and subsequently demonstrated that DCT overexpression is associated with radiation as well as drug resistance. We also showed that the radiation-resistant variants display aberrant ERK activation in response to UV(B) irradiation compared to WM35 cells. The activation of ERK was shown to be mediated through DCT expression in melanoma cells. These results suggest a possible function for DCT in both radiation and drug resistance, which may be associated, at least in part, by differential activation of the ERK/MAPK stress response pathway.

Materials and methods

Cell lines and retroviral insertional mutagenesis

All human melanoma cell lines were cultured in RPMI 1640 medium (Life Technologies) supplemented with 5% fetal bovine serum. Cells were routinely maintained at 37°C in a humidified chamber containing 5% CO2.

For retroviral insertional mutagenesis, a replication-defective neo-containing murine stem cell virus (MSCV) with an amphotrophic host range was produced by infection of GP+envAM12 helper-free packaging cells (Markowitz et al., 1988) with an ecotropic virus produced by GP+E-86 cells transfected with an env-version of MSCV (Figure 1a), as described previously (Lu et al., 1995).

Figure 1
figure1

(a) Schematic representation of the MSCV v2.1 retroviral vector. This amphotrophic retroviral vector is devoid of env sequences. It includes an extended packaging region (P+) for high viral titer and a modified 5′-untranslated region with a primer-binding site for tRNAGln instead of the usual tRNAPro. Arrows indicate long terminal repeats (LTR) and pgk promoters utilized for the expression of viral and neo RNA transcripts, respectively. (b) Patterns of proviral integration in the radiation-resistant clones. Genomic DNA (10 μg) was isolated from four radiation-resistant variant cell lines (XR5, XR6, XR9, XR10), a CDDP-resistant variant (clone E) and the parental cell line WM35, and digested with EcoRI. Southern blot analysis using pgk-neo as a probe shows hybridization to integrated MSCV proviruses

For retroviral insertional mutagenesis, 2 × 105 WM35 cells were infected for 4–8 h with the recombinant virus having a titer of 0.5–1.0 × 106 G418-resistant colony-forming units/ml. After 24 h, the medium was removed and replaced with a fresh medium containing 800 μg/ml G418 (Life Technologies) to select for infected cells. Subsequent to G418 selection, WM35 cells and three independent retroviral-infected WM35-neo cells were plated at a density of 105 cells/dish and were treated with 8 Gy X-ray radiation, a dose that is lethal to wild-type WM35 cells. Following radiation treatment, cells were cultured for several weeks and the resulting colonies were isolated and expanded in culture. Resistance of these cell lines to X-rays was compared to parental WM35 cells in colony-forming efficiency assays. Resistant cell lines were chosen for further analysis.

Southern blot analysis

High molecular weight genomic DNA was isolated by the proteinase K-phenol : chloroform method. DNA (20 μg) were cut with restriction endonuclease and fractionated on 1% agarose gels. DNA were then transferred onto nylon filters and hybridized with a 32P-labeled pgk-neo probe that maps within the MSCV retrovirus (Figure 1a). Following hybridization, filters were washed and exposed onto X-ray film at −70°C with an intensifying screen.

UV(B) irradiation and cell viability assay

For UV(B) irradiation, cells were plated at a density of 5.0 × 105 cells/60-mm dish. At 24 h after plating, the culture medium was removed and replaced with 2 ml PBS. Cells were then briefly treated (approximately 5–10 s) with 5 mJ/cm2 UV(B) radiation using a chamber containing three General Electric UV(B) lights that emit between 280 and 320 nm with peak emission at 310 nm. The treatment dose was measured using an International Light radiometer/photometer (Model IL1400A). Following treatment, the PBS was aspirated and replaced with RPMI 1640 media containing 5% FBS. Control dishes were treated in the same manner but were not exposed to UV(B). At 24 h after treatment, viable cells were counted by trypan blue exclusion.

Northern blot analysis

Total RNA was extracted from cultured cells using Trizol (Life Technologies). RNA extracts were quantified by spectrophotometric analysis and 10 μg of each extract was fractionated on 1% agarose/formaldehyde gels. Following electrophoresis, RNA was transferred to nylon filters (ZetaProbe, BioRad Laboratories) by capillary elution in 20 × SSC and then baked for 1 h at 80°C in a vacuum oven. Prehybridization and hybridization with 32P-labeled cDNA probes were performed at 42°C in 50% formamide, 5 × SSPE, 1 × Denhardt's solution, 5% dextran sulfate and 1% SDS. Blots were washed with high stringency at 65°C and exposed onto an X-ray film at −70°C with an intensifying screen.

cDNA probes and melanoma cells transfection

The cDNA probes for TYRP1 and DCT were provided by Dr Rick Sturm (University of Queensland, Australia) and have been described previously (Pak et al., 2000). The tyrosinase cDNA probe was purchased from American Type Culture Collection. These cDNAs were radiolabeled by random priming using the Oligolabeling Kit (Pharmacia) and purified on Sephadex G-50 spun columns (Pharmacia).

Transfection with MEKEE and DCT

The melanoma cell line WM35 was transfected with pcDNA DCT or pcDNA vector alone as previously described (Chu et al., 2000). For MEKEE overexpression, WM35 cells were transfected with the expression vector pIND-MEKEE (Yan and Templeton, 1994), using lipofectamine (Gibco/BRL), and G418-resistant clones were isolated and expanded.

SDS–PAGE and Western blot analysis

Cells were lysed in RIPA buffer (50 mM Tris, pH 7.4, 50 mM NaCl, 2 mM MgCl2, 1 mM EDTA, 0.5% IGEPAL CA-630) containing the following protease inhibitors: 0.1 mM PMSF, 10 μ M sodium orthovanadate and 20 μg/ml leupeptin. Protein concentrations were determined by the Bradford assay (BioRad Laboratories) and 50 μg of each extract was resolved by SDS–polyacrylamide gel electrophoresis (SDS–PAGE) and transferred onto Immobilon-P filters (Millipore Corp.) using a semidry transfer cell (BioRad Laboratories). Blots were blocked in 5% milk powder in Tris-buffered saline containing 0.01% Tween-20 (TTBS), and then incubated in the same solution containing αPEP8 anti-DCT antibody at a dilution of 1 : 4000 (kindly provided by Dr Vincent Hearing), Phospho-Erk and Erk-2 (Transduction laboratories) and β-action (Sigma) antibodies. Blots were subsequently incubated with a horseradish peroxidase-conjugated goat anti-rabbit secondary antibody (1 : 5000 dilution; StressGen) and immunoreactive sites were detected by enhanced chemiluminescence and exposed onto an X-ray film.

CDDP cytotoxicity assay

Cells were plated at a density of 1 × 105 cells/well in flat-bottom 96-well plates and allowed to attach for 24 h. The cells were then treated with 15 μ M CDDP for 24 h. Following CDDP treatment, the quantity of viable cells was determined using the CellTiter96 Aqueous Non-Radioactive MTS Proliferation Assay (Promega) according to the manufacturer's instructions.

Results

Derivation of gamma radiation-resistant melanoma variants by retroviral insertional mutagenesis

The strategy of retroviral insertional mutagenesis has been shown previously to be an effective means of inducing mutations in cells that lead to a dominantly selectable phenotype. We have successfully utilized this strategy using the replication-defective MSCV retrovirus harboring the neomycin phosphotransferase (neo) gene to establish tumorigenic (Bani et al., 1996) and CDDP-resistant (Lu et al., 1995) variants of the normally nontumorigenic, CDDP-sensitive WM35 human melanoma cell line.

Cells from three independent infections were subjected to selection for X-radiation-resistant variants. Treatment with 8 Gy radiation yielded several variant cell lines that survived this normally lethal dose of radiation, presumably due to mutation(s) acquired by proviral integration. These variants were isolated and expanded in culture for further investigation.

Identification of proviral integration sites in resistant clones

The clonality of the radiation-resistant variants was determined by Southern blot analysis. Blots containing EcoRI digested genomic DNA from these variants were hybridized with a pgk-neo probe that maps within the MSCV retrovirus (Figure 1a). As shown in Figure 1b, each of the four radiation resistant variants displayed a distinct banding pattern that suggests that they originated independently. In fact, two distinct DNA fragments obtained from the variant XR9 hybridized with the pgk-neo probe, suggesting that this clone had acquired the integration of two proviruses. Interestingly, none of the banding patterns for the radiation-resistant variants corresponded to that of the CDDP-resistant variant, Clone E, which we have previously shown to possess a single provirus at a common site of integration for CDDP resistance, which we designated, CRL-1 (Chu et al., 2000). DNA from the parental cell line, WM35, was included as a negative control and did not hybridize with the pgk-neo probe.

Cross-resistance of gamma radiation-resistant clones to UV(B) irradiation

Cytotoxicity induced by gamma radiation results primarily from the generation of DNA strand breaks. However, other forms of radiation are known to induce cytotoxicity by inflicting different forms of DNA damage. For example, cytotoxicity induced by UV(B) radiation results from the formation of pyrimidine dimers. Therefore, we sought to determine if the virally derived X-ray-resistant variants were cross-resistant to DNA damage caused by UV(B) radiation. Following treatment with 5 and 10 mJ/cm2 UV(B), the X-ray-resistant clones XR9 and XR10 displayed increased resistance relative to the parental WM35 cells (Figure 2). Clone D1, a tumorigenic variant of WM35 generated by retroviral insertional mutagenesis using the MSCV retrovirus (Bani et al., 1996) was included as a negative control. The relative resistance ratio was approximately twofold at 5 mJ/cm2 and lower at 10 mJ/cm2, which is relatively low for most in vitro models of resistance. However, similar degrees of resistance were obtained in CDDP-resistant mutants generated by retroviral insertional mutagenesis (Lu et al., 1995), which was subsequently shown to have significant effects on tumor sensitivity to CDDP in vivo (unpublished data). Moreover, this level of resistance is comparable to that observed with respect to clinical levels of acquired resistance (Wolf et al., 1987; Mazzoni et al., 1990; Kuboda et al., 1991).

Figure 2
figure2

Increased resistance of gamma radiation-resistant clones to UV(B) treatment. Two virally derived gamma radiation-resistant variants (XR9, XR10) were irradiated with 5 mJ/cm2 UV(B) and the number of viable cells were determined 24 h following treatment. Each of the gamma radiation-resistant variants showed cross-resistance to UV(B) treatment compared to the parental WM35 cells and tumorigenic variant of this cell line clone D1. Data are presented as mean±s.e.m

Upregulation of DCT in gamma radiation-resistant clones

We have previously reported that CDDP resistance in melanoma cells correlated with the overexpression DCT, a melanocyte-specific enzyme that functions in the biosynthesis of melanin (Chu et al., 2000). We further showed that resistance associated with DCT overexpression was not restricted to CDDP, but applied to other DNA-damaging agents. In contrast, resistance to microtubule reactive drugs such as taxol, which exert their cytotoxic effects independent of DNA damage, was not related to DCT expression. Based on these findings, we speculated that DCT overexpression may mediate resistance specifically to agents that induce DNA damage. To determine if DCT expression is also associated with resistance to radiation-induced DNA damage, the levels of DCT were determined in the X-ray-resistant variants. Northern and Western blot analyses revealed that DCT was overexpressed in each radiation-resistant clone examined at the RNA and protein levels, respectively (Figure 3). Moreover, radiation resistance was shown to be independent of the expressions of tyrosinase and TYRP1 (Figure 4). These results suggest that radiation resistance is specifically associated with DCT expression and is not attributable to an overall activation of melanin synthesis. This is consistent with our previous findings, which showed that CDDP resistance correlated only with DCT expression and not with tyrosinase or TYRP1 (Pak et al., 2000). Taken together, these results suggest a possible association between DCT and radioresistance.

Figure 3
figure3

Upregulation of DCT in radiation-resistant clones. (a) Northern and (b) Western blot analyses showed that DCT expression was upregulated at both the mRNA and protein levels in each of the radiation-resistant clones (XR5, XR6, XR9, XR10) relative to the parental WM35 cells. The upregulation of DCT in the CDDP-resistant variant, clone E, was previously shown and is included as a positive control. Levels of GAPDH and Erk2 show uniformity of RNA loading per lane for the Northern and Western blots, respectively

Figure 4
figure4

Expression of tyrosinase and TYRP1 in CDDP- and UV(B)-resistant clones. Total RNA (10 μg) obtained from the parental WM35 human melanoma cell line, clone E (CDDP-resistant variant) and XR5, XR6, XR9, XR10 (UV(B)-resistant variants) were fractionated on agarose gels, transferred to nylon filters and hybridized with cDNA probes for human tyrosinase and TYRP1. Hybridization with the GAPDH probe demonstrates the uniformity of RNA loading. The pattern of expression of the mRNA encoding these melanogenic enzymes displayed no relationship with radiation resistance

Radioresistance correlates with DCT in a panel of melanoma cell lines

We have previously shown that the constitutive levels of DCT expression in a variety of different human melanoma cell lines, obtained from patients at various stages of disease progression, correlated well with drug resistance (Chu et al., 2000; Pak et al., 2000). These results demonstrated that the relationship between DCT and drug resistance was not a phenomenon restricted to the WM35 cell line and its derivatives generated by retroviral insertional mutagenesis, but possibly a generally applicable characteristic of melanoma cells. In order to determine the scope of the association between DCT expression and radioresistance, several human melanoma cell lines were subjected to UV(B) treatment and their levels of resistance were compared to DCT expression. As shown in Figure 5a, the levels of DCT expression paralleled radioresistance in each melanoma cell line examined. For example, the metastatic cell line WM9, which expresses very low levels of DCT was relatively sensitive to treatment with 5 mJ/cm2 UV(B) (survival=16.7±1.5%), while the DCT overexpressing cell line, MeWo, was relatively resistant to UV(B) irradiation (survival=63.5±4.0%). The levels of DCT in each of these cell lines were previously shown to correlate with their relative levels of CDDP resistance (Pak et al., 2000). In both these cell lines, the levels of tyrosinase and TYRP1 expression did not correlate with their degree of drug resistance (Pak et al., 2000).

Figure 5
figure5

(a) Constitutive levels of DCT expression correlated with UV(B) resistance in several human melanoma cell lines. The percentage of viable cells following treatment with 5 mJ/cm2 (solid bars) were determined for several human melanoma cell lines and compared to their respective levels of DCT expression (striped bars). Cells expressing high relative levels of DCT (WM1341B, MeWo) showed greater resistance to treatment with 5 mJ/cm2 UV(B) compared to cells that express lower levels of DCT (WM35, WM9). (b) The relative sensitivity of WM35 cells that were transfected with DCT was determined following UV(B) treatment. The WM35 DCT transfectant cell line (C8) showed higher levels of resistance to UV(B) irradiation compared to wild-type WM35 cells and the vector-alone transfectants (solid bars). The striped bars indicate the relative levels of DCT expression in these cell lines. Data are presented as mean±s.e.m

In order to demonstrate that resistance is directly attributable to DCT overexpression, the resistance of WM35 cells transfected with DCT were determined following UV(B) irradiation. We have characterized these cells previously and have shown that they overexpress DCT and display increased resistance to CDDP treatment (Chu et al., 2000). As summarized in Figure 5b, DCT-transfected cells exhibited greater levels of resistance to UV(B) irradiation compared to the parental WM35 cells and to vector-alone transfectants. These results are in agreement with those reported by Nishioka et al. (1999), who have also shown that DCT transfection in a melanoma cell line confers resistance to UV(B) irradiation. Therefore, taken together, our finding suggests that DCT overexpression may contribute to resistance to DNA damage induced by various forms of radiation.

Association between DCT-mediated resistance and ERK activation

Cytotoxicity induced by treatment with either CDDP or UV(B) has been shown to be mediated, at least in part, by the activation of various cellular stress response pathways (Nehme et al., 1997; Sanchez-Perez et al., 1998; Pillaire et al., 2000; Hayakawa et al., 2003). As such, the manner in which cancer cells respond to stress may have major implications for relative levels of resistance to anticancer treatment. We have previously noted that DCT expression in human melanoma cells also correlated with resistance to heat stress, in addition to drug and radiation treatment (unpublished data), suggesting that stress pathways that are activated by each of these stressors may be altered in DCT overexpressing cells.

Since the primary function of DCT is to convert DOPAchrome to DHICA, a recent study has shown that these diffusible melanin-related metabolites do not serve merely as pigment precursors, but may also act as modulators of the response of the pigmentary cell melanocytes to external stimuli, such as inflammation and stress (Fang et al., 2001). Therefore, we examined the effect of L-cysteine, a reducing agent that has been shown to increase the conversion of L-DOPAchrome to DHICA by DCT, on the activation of the ERK pathway in a melanoma cell line. Western blot analysis using antibodies against phosphorylated ERK showed that ERK is dramatically activated in melanoma cells following treatment with L-cysteine (Figure 6a). To investigate whether ERK activation is mediated by DCT, we examine the phosphorylation status of ERK in DCT overexpressing cells. Figure 6b shows that three clones of WM35 overexpressing DCT display elevated levels of ERK activation compared to the vector-alone mock-transfected WM35 cells (Figure 6b).

Figure 6
figure6

(a) Effect of DHICA reducing agents L-cysteine on ERK/MAPK expression. WM35 cells were grown as controls, or treated with L-cysteine for 3 and 24 h and lysates prepared from these cells were subjected to Western blot analysis using DCT or phospho-Erk antibodies (New England Biolab). Later, the blot was hybridized with an Erk-2 antibody (Transduction laboratory). (b) Overexpression of DCT in WM35 cells increases ERK/MAPK activity leading to UV(B) resistance. Clones isolated fromWM35 cells transfected with DCT expression vector or pooled PcDNA vector transfected WM35 cells were subjected to Western blot analysis using the DCT, phospho-Erk and Erk-2 antibodies, as described above

Differential activation of ERK in radiation-resistant variants in response to UV(B)

To determine whether ERK is differentially activated in the DCT overexpressing, radiation-resistant variants compared to WM35, cells were treated with 20 mJ/cm2 UV(B) and the levels of ERK phosphorylation were determined over a 2-h time course. Constitutively, XR9 and XR10 cells showed a significantly higher level of ERK activation relative to WM35 (Figure 7). Following UV(B) irradiation, ERK was activated in both the WM35 and radiation-resistant cells, but the fold increase in activation was markedly higher in WM35 cells compared to XR9 and XR10, respectively (Figure 7). We also observed a similar pattern of ERK activity in WM35 cells following CDDP treatment (data not shown). It is therefore plausible that differences in ERK activation between the parental WM35 cells and the radiation- and CDDP-resistant variants may modulate resistance to radiation treatment. To prove further that ERK activation is responsible for UV(B) resistance, the WM35 cell line were transfected with the MEKEE expression vector, a dominant-active mutant of ERK (Yan and Templeton, 1994; Donovan et al., 2001). Western blot analysis showed that the levels of ERK activation in the MEKEE transfectants were significantly higher compared to the parental cells (Figure 8a). MEKEE transfectants also showed increased resistance to UV(B) treatment (Figure 8b). Previously it has been shown that the MEK1 inhibitor PD98059 sensitizes C8161 melanoma cells to CDDP-induced apoptosis (Mandic et al., 2001). We have also shown that a similar dose of PD98059 sensitized WM35 cells to CDDP-induced apoptosis (data not shown). These results support the importance of the ERK pathway in UV(B) and drug resistance.

Figure 7
figure7

Increased ERK/MAPK activity in melanoma cells treated with UV(B). WM35 cells and the radiation-resistant variant XR9 and XR10 cell lines were treated with 5 mJ/cm2 UV(B); after the indicated time, the cells were lysed and subjected to Western blot analysis using phospho-Erk and Erk-2, as described above. To show equal loading, the blot was hybridized with β-actin antibody

Figure 8
figure8

(a) WM35 cells overexpressing MEKEE are more resistant to UV(B). WM35 cells were transfected with MEKEE, a dominant mutant form of MEK, and the isolated clones were examined by Western blot analysis for ERK/MAPK activity using a phospho-Erk antibody. Two clones (C1 and C7) showing higher levels of ERK/MAPK were chosen for further studies. Equal loading was determined by staining the blot with Coomassie blue (data not shown) (b). Survival of WM35 cells overexpressing MEKEE (clones C1 and C7) were examined in response to UV(B) using MTS assay. Data are presented as mean±s.e.m

Discussion

Retroviral insertional mutagenesis has proven to be an effective means of generating drug-resistant mutant cell lines, which can subsequently be analysed to identify possible resistance-associated genes. We have previously utilized this approach to generate mutants of the WM35 human melanoma cell line that are resistant to the chemotherapeutic drug CDDP. Subsequent studies showed that each of these CDDP-resistant clones overexpressed DCT (Chu et al., 2000). We further showed that the constitutive levels of DCT expression in a panel of human melanoma cell lines correlated with its respective levels of drug resistance and that the enforced expression of DCT in a CDDP-sensitive cell line rendered it more CDDP resistant. Here, a similar approach was utilized to investigate the mechanisms of radiation resistance in melanomas. The gamma radiation-resistant mutants generated in this manner were also shown to be more resistant to treatment with UV(B), and were each shown to overexpress DCT at the mRNA and protein levels. We further found that DCT expression correlated with radiation resistance in several human melanoma cell lines and that DCT overexpression by transfection increased radiation resistance.

Taken together, these data implicate a role for DCT in radiation resistance in melanoma cells. This is in agreement with a recent report by Nishioka et al. (1999), which showed that DCT expression was markedly reduced in UV(B)-irradiated melanoma cells and that transfection with the DCT expression vector was able to rescue these cells from UV(B)-induced apoptosis. We also demonstrated that each of the DCT overexpressing radiation-resistant cell lines examined displayed increased resistance to CDDP treatment, while DCT overexpressing CDDP-resistant cells previously established by retroviral insertional mutagenesis (Lu et al., 1995) exhibited increased radiation resistance. These results, along with our previous report implicating DCT in drug resistance (Chu et al., 2000), suggest that DCT may be a broad mediator of radiation and drug resistance in melanoma cells and that its upregulation in melanomas may contribute to the intrinsic resistance phenotype of this type of malignancy in a tissue-specific manner.

Currently, the mechanisms by which DCT overexpression leads to increased radiation and drug resistance remain unclear. Based on our previous finding that DCT expression confers resistance specifically to DNA-damaging drugs such as CDDP, carboplatin and methotrexate, but not to tubulin targeting agents, for example, taxol (Chu et al., 2000), and given that radiation can also induce DNA lesions, it is possible that DCT overexpression is associated with resistance pathways associated with DNA damage. Several cellular DNA repair pathways exist that serve to protect cells from various forms of DNA damage. DNA repair capacity is frequently enhanced in cancer cells that display resistance to DNA lesions induced by specific chemotherapeutic drugs and radiation treatment. For example, Ferry et al. (2000) reported increased nucleotide excision repair (NER) activity in CDDP-resistant ovarian cancer cells. Similarly, loss of DNA mismatch repair capacity was shown to be associated with the drug-resistant variants in human colon cancer and endometrial adenocarcinoma cells (Lin and Howell, 1999). Thus, the possibility that these DNA repair mechanisms may play an important role in DCT-mediated drug and radiation resistance in melanomas warrants further investigation.

Although the function of DCT in the nondecarboxylative tautomerization of L-DOPAchrome to the highly stable product DHICA in melanin synthesis has been well characterized (Tsukamoto et al., 1992; Lin and Howell, 1999), the results presented here and previously clearly demonstrate an association between the expression of this melanogenic enzyme and increased resistance to DNA damage induced by chemo- and radiotherapy. In contrast, we have demonstrated that the expressions of the other melanogenic enzymes, tyrosinase and TYRP1 bore no relationship to radioresistance. Furthermore, we have previously shown that CDDP resistance correlated specifically with DCT expression, but was independent of tyrosinase and TYRP1 expression, and of cellular melanin content (Pak et al., 2000). Thus, increased resistance mediated by DCT overexpression appears to be distinct from the other melanogenic parameters. Whether DCT expression directly or indirectly induces resistance remains to be determined.

Due to the restricted subcellular localization of DCT to the melanin-producing compartment (melanosomes), we speculate that DCT may induce resistance pathways indirectly through intermediate molecule(s) that can enter other cellular compartments. A possible candidate is the melanin precursor DHICA, which is the product of the DCT-mediated tautomerization of L-DOPAchrome. DHICA is a stable, highly diffusible, antioxidant molecule that has previously been shown to exhibit marked reactivity to nitrogen oxides produced by autoxidation of nitric oxide (NO) (Jara et al., 1990), stimulate NO production by lipopolysaccharide (LPS)-induced-NO synthase (iNOS) (D'Acquisto et al., 1995) and function as a potent inhibitor of lipid peroxidation (Memoli et al., 1997). Interestingly, we have shown that the reducing compound L-cysteine, which has been shown to enhance the DCT-catalysed conversion of L-dopachrome to DHICA (Rosengren and Rorsman, 1998), is capable of increasing the ERK/MAPK activity in melanoma cells. Since the ERK/MAPK pathways have been reported to be activated by CDDP, UV(B) and heat stress in other systems (Chen et al., 1995; Merienne et al., 2000; Wang et al., 2000), it is therefore plausible that increased levels of DHICA by DCT may mediate resistance. This possibility was shown to be correct as DCT induction due to UV(B) and drug treatment as well as overexpression of DCT in melanoma cells activates the ERK/MAPK pathway. Moreover, melanoma cells transfected with MEKEE, a constitutively active mutant of MEK (Yan and Templeton, 1994; Donovan et al., 2001), increases Phospho-Erk activity resulting in increased UV(B) resistance in melanoma cells. This result is consistent with a recent report, which shows that the inhibitors of the ERK/MAPK pathway can sensitize melanoma cells to the toxic effect of chemotherapeutic drugs (Mandic et al., 2001).

We have previously identified the common site of proviral integration, CRL-1, which was rearranged in five out of the nine CDDP-resistant clones generated by retroviral insertional mutagenesis (Chu et al., 2000). While all nine clones were shown to overexpress DCT, fluorescence in situ hybridization (FISH) analysis showed that CRL-1 and DCT map to different chromosomes (unpublished data). Although it is possible that the gene(s) mutated by CRL-1 rearrangement may regulate DCT in these five clones, the fact that the remaining four clones each showed different proviral integration patterns suggests that either viral integration occurs in a large genomic region within the CRL-1 loci or provirus inserted in several loci that may alter the expression of gene(s) resulting in DCT overexpression and subsequently drug resistance. The data presented in this study, which showed that each of the DCT overexpressing, radiation-resistant variants derived by retroviral insertional mutagenesis had distinct sites of proviral integration, support this notion. The characterization of these integration sites should provide insights into the regulation of DCT.

In summary, we have shown that radiation-resistant variants generated by retroviral insertional mutagenesis overexpressed DCT and that the constitutive levels of DCT expression in a panel of human melanoma cell lines correlated with radiation resistance. We further showed that the enforced expression of DCT in radiation-sensitive melanoma cells by transfection increased their resistance to radiation treatment. Most interestingly, we showed that DCT overexpressing melanoma cells exhibit cross-resistance to both radiation and drug treatment. Upregulation of DCT through transfection or by UV(B) treatment was shown to increase the ERK/MAPK pathways leading to increase in both drug and UV(B) resistance. Together these results suggest an important role for DCT in both intrinsic radioresistance and resistance to DNA-damaging chemotherapeutic drugs. Given that melanomas are characteristically resistant to these types of treatment, the specific targeting of DCT by such drugs as specific antisense oligonucleotides may represent a possible sensitization strategy to render melanomas more responsive to conventional therapies.

References

  1. Bani MR, Rak J, Adachi D, Wiltshire R, Trent JM, Kerbel RS and Ben-David Y . (1996). Cancer Res., 56, 3075–3086.

  2. Ben-David Y and Bernstein A . (1991). Cell, 66, 831–834.

  3. Ben-David Y, Giddens EG, Letwin K and Bernstein A . (1991). Genes Dev., 5, 908–918.

  4. Ben-David Y, Prideaux VR, Chow V, Benchimol S and Bernstein A . (1988). Oncogene, 3, 179–185.

  5. Chen F, Torres M and Duncan RF . (1995). Biochem J., 312 (Part 2), 341–349.

  6. Chu W, Pak BJ, Bani MR, Kapoor M, Lu SJ, Tamir A, Kerbel RS and Ben-David Y . (2000). Oncogene, 19, 395–402.

  7. Cooksey CJ, Garratt PJ, Land EJ, Pavel S, Ramsden CA, Riley PA and Smit NP . (1997). J. Biol. Chem., 272, 26226–26235.

  8. D'Acquisto F, Carnuccio R, d'Ischia M and Misuraca G . (1995). Life Sci., 57, L401–L406.

  9. Donovan JC, Milic A and Slingerland JM . (2001). J. Cell. Biol., 276, 40888–40895.

  10. Fang D, Kute T and Setaluri V . (2001). Pigment Cell Res., 14, 132–139.

  11. Ferry KV, Hamilton TC and Johnson SW . (2000). Biochem. Pharmacol., 60, 1305–1313.

  12. Hayakawa J, Depatie C, Ohmichi M and Mercola D . (2003). J. Biol. Chem., 278, 20582–20592.

  13. Hearing VJ and Tsukamoto K . (1991). FASEB J., 5, 2902–2909.

  14. Jara JR, Solano F, Garcia-Borron JC, Aroca P and Lozano JA . (1990). Biochim. Biophys. Acta, 1035, 276–285.

  15. Kuboda H, Sugimoto T, Ueda K, Tsuchida S, Horri Y, Inazawa J, Sato K and Sawada T . (1991). Int. J. Cancer, 47, 732–737.

  16. Lin X and Howell SB . (1999). Mol. Pharmacol., 56, 390–395.

  17. Lu S, Man S, Bani MR, Adachi D, Hawley RG, Kerbel RS and Ben-David Y . (1995). Cancer Res., 55, 1139–1145.

  18. Lu SJ, Rowan S, Bani MR and Ben-David Y . (1994). Proc. Natl. Acad. Sci. USA, 91, 8398–8402.

  19. Mandic A, Viktorsson K, Heiden T, Hansson J and Shoshan MC . (2001). Melanoma Res., 11, 11–19.

  20. Markowitz D, Goff S and Bank A . (1988). Virology, 167, 400–406.

  21. Mazzoni A, Trave F, Russo P, Nicolin A and Rustum YM . (1990). Oncology, 47, 488–494.

  22. Memoli S, Napolitano A, d'Ischia M, Misuraca G, Palumbo A and Prota G . (1997). Biochim. Biophys. Acta, 1346, 61–68.

  23. Merienne K, Jacquot S, Zeniou M, Pannetier S, Sassone-Corsi P and Hanauer A . (2000). Oncogene, 19, 4221–4229.

  24. Moreau-Gachelin F, Tavitian A and Tambourin P . (1988). Nature, 331, 277–280.

  25. Mowat M, Cheng A, Kimura N, Bernstein A and Benchimol S . (1985). Nature, 314, 633–636.

  26. Nehme A, Baskaran R, Aebi S, Fink D, Nebel S, Cenni B, Wang JY, Howell SB and Christen RD . (1997). Cancer Res., 57, 3253–3257.

  27. Nishioka E, Funasaka Y, Kondoh H, Chakraborty AK, Mishima Y and Ichihashi M . (1999). Melanoma Res., 9, 433–443.

  28. Olivares C, Jimenez-Cervantes C, Lozano JA, Solano F and Garcia-Borron JC . (2001). Biochem. J., 354, 131–139.

  29. Pak BJ, Li Q, Kerbel RS and Ben-David Y . (2000). Melanoma Res., 10, 499–505.

  30. Pillaire MJ, Nebreda AR and Darbon JM . (2000). Biochem. Biophys. Res. Commun., 278, 724–728.

  31. Rosengren E and Rorsman H . (1998). Melanoma Res., 8, 469–470.

  32. Sanchez-Perez I, Murguia JR and Perona R . (1998). Oncogene, 16, 533–540.

  33. Tsukamoto K, Jackson IJ, Urabe K, Montague PM and Hearing VJ . (1992). EMBO J., 11, 519–526.

  34. Wang X, Martindale JL and Holbrook NJ . (2000). J. Biol. Chem., 275, 39435–39443.

  35. Wolf CR, Hayward IP, Lawrie SS, Buckton K, McIntyre MA, Adams DJ, Lewis AD, Scott ARR and Smyth JF . (1987). Int. J. Cancer, 39, 695–702.

  36. Yan M and Templeton DJ . (1994). J. Biol. Chem., 269, 19067–19073.

Download references

Acknowledgements

We thank Dr Dan Sauder (Sunnybrook and Women's College Health Sciences Centre) for use of the UV(B) lamp and radiometer/photometer. The DCT antibody was kindly provided by Dr Vincent Hearing and cDNA for DCT was provided by Dr Rick Sturm and expression vector pIND-MEKEE by Dr Dennis Templeton. This work was supported by grants from the National Cancer Institute of Canada (NCIC) and Canadian Institute of Health Research (CIHR) to YBD and the National Institutes of Health, USA (CA-41233) to RSK. BJP is a recipient of the Sunnybrook Trust Fellowship for Medical Research.

Author information

Correspondence to Yaacov Ben-David.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Pak, B., Lee, J., Thai, B. et al. Radiation resistance of human melanoma analysed by retroviral insertional mutagenesis reveals a possible role for dopachrome tautomerase. Oncogene 23, 30–38 (2004) doi:10.1038/sj.onc.1207007

Download citation

Keywords

  • radiation resistance
  • malignant melanoma
  • DOPAchrome tautomerase
  • mitogen-activated protein kinase

Further reading