The effect of L-dopa on the potentiation of radiation damage to human melanoma cells.

Since L-dopa (L-3,4-dihydroxyphenylalanine) has been shown to possess a selective toxicity for melanoma cells both in vitro and in vivo, we have examined the combined effect of L-dopa and radiation on human melanoma cells. It was found that the combined use of L-dopa potentiated the radiation cytotoxicity to HMV-I human melanoma cells, compared with the response seen in non-melanoma HeLa S3 cells. In HMV-I cells during their exponential phase, L-dopa decreased the shoulder width of the radiation survival curve significantly. In addition, L-dopa significantly inhibited the repair of potentially lethal damage (PLD) in HMV-I cells during their plateau phase. When the distributions of the G1, S, and G2-M cells were measured 24 h after combined L-dopa and radiation treatment, there was significant increase in the accumulation of cells in the G2-M phase of the cell cycle, compared to cells that received either L-dopa or radiation treatment only.

The effect of L-dopa on the potentiation of radiation damage to human melanoma cells I. Yamada',S. Seki2,S. Ito3,S. Suzuki',0. Matsubara2 & T. Kasuga2 'Department of Radiology and 2Second Department of Pathology, School of Medicine, Tokyo Medical and Dental University, Tokyo 113,and 3School of Hygiene, Fujita-Gakuen Health University, Toyoake, Aichi 470-11, Japan. Summary Since L-dopa (L-3,4-dihydroxyphenylalanine) has been shown to possess a selective toxicity for melanoma cells both in vitro and in vivo, we have examined the combined effect of L-dopa and radiation on human melanoma cells. It was found that the combined use of L-dopa potentiated the radiation cytotoxicity to HMV-I human melanoma cells, compared with the response seen in non-melanoma HeLa S3 cells. In HMV-I cells during their exponential phase, L-dopa decreased the shoulder width of the radiation survival curve significantly. In addition, L-dopa significantly inhibited the repair of potentially lethal damage (PLD) in HMV-I cells during their plateau phase. When the distributions of the G,, S, and G2-M cells were measured 24 h after combined L-dopa and radiation treatment, there was significant increase in the accumulation of cells in the G2-M phase of the cell cycle, compared to cells that received either L-dopa or radiation treatment only.
L-Dopa (L-3,4-dihydroxyphenlalanine) has been shown to be selectively toxic to melanoma cells in vitro (Wick et al., 1977). Further, it has been reported that L-dopa and its chemical analogues inhibit the growth of murine melanoma in vivo and prolong the survival span of melanoma-bearing mice (Wick, 1978). The mechanism of such action has been postulated to involve a tyrosinase-mediated oxidation of L-dopa to a quinone with subsequent sulphydryl group scavenging and inhibition of enzymes central to DNA synthesis (Graham et al., 1978;Wick, 1980). Unfortunately, while the clinical investigation of the effect of L-dopa on advanced human malignant melanomas has already started (Wick, 1983), no study has yet appeared dealing with the effect of L-dopa when combined with other therapeutic modalities in the management of melanomas. Thus, in this report, we have examined the effect of L-dopa on the potentiation of radiation damage to human melanoma cells. We also have evaluated the redistribution of cells in different phases of the cell cycle as a possible mechanism for the interaction between L-dopa and radiation.

Cells
We used the HMV-I human melanoma cell line that was established from a black-brown malignant melanoma in the vaginal wall of a woman (Yamada et al., 1987). HeLa S3 cells were used as the non-melanoma control cells. The two cell lines were maintained in Ham's F-10 medium, supplemented with 10% calf serum (Flow Laboratories), penicillin (100 U ml -'), and streptomycin (100 fig ml -'), and incubated in a humidified atmosphere of 95% air/5% CO2 at 37°C. Chemicals L-Dopa was purchased from Sigma Chemical Co. (St Louis, MO, USA). The drug solution was freshly prepared in Ham's F-10 medium just before use at the beginning of each experiment.

Irradiation
Cells that were grown in plastic Petri dishes were irradiated at room temperature, using a '"Co '-ray unit at a dose rate of 1.44 Gy min-'. For radiation survival studies, the cells were irradiated with doses of 1, 2, 4, 6, 8 and 10 Gy.
Effects of L-dopa on radiosensitivity Two x 105 cells were inoculated into 60-mm Petri dishes and incubated for 48 hours. One hundred Ag of L-dopa per millilitre was added to the culture medium immediately before irradiation. The cells were then irradiated with graded doses and incubated for 4 h at 37°C.
Effects of L-dopa on potentially lethal damage (PLD) repair For PLD repair studies, 2 x 105 cells were inoculated into 60-mm Petri dishes and grown to confluence. During this period, the medium was changed on alternate days. Cells in the confluent state were irradiated with graded doses and incubated for 6 h at 37'C. To examine the effect of L-dopa on the PLD repair, 100 fig of L-dopa per millilitre was added to the culture medium immediately before irradiation and similarly incubated for 6 h.

Colony formation
After irradiation and L-dopa exposure, the cells of each treated group were washed and trypsinised, and an appropriate number of cells were plated in duplicate 60 mm Petri dishes containing 5 ml of the complete medium. The dishes were incubated at 37°C in an atmosphere of 95% air/5% CO2 for 14 days. The resulting colonies that contained more than 50 cells were counted and the survival fraction of each group was calculated in reference to the untreated control group. At least three replicate experiments were conducted for each treatment. The respective survival curves then were constructed by plotting the surviving fraction as a function of radiation dose. The slope of the linear portion of the survival curves was fitted by a least-squares linear regression analysis and the Do, Dq and n values were calculated. Further, a linear quadratic analysis was carried out for these survival curves, and a and values were calculated.
Cell cycle analysis Cell cycle distributions of melanoma cells treated with Ldopa and radiation were determined from DNA histograms measured by flow cytometry. Exponentially growing cells were exposed to 10 Gy radiation and then incubated with 100 pg ml-' L-dopa for four hours. The L-dopa was then removed by changing the medium. Twenty-four hours later, the cells were trypsinised from the dish. The cell suspension was washed twice with a phosphate buffered saline (PBS, pH 7.2), after which the cells (1 x 106) in 1 ml of PBS were mixed with 3 ml of cold 95% ethanol, and incubated at -20C for 60 min for fixation. The cells then were washed twice with PBS and incubated in a solution of 1 mg ml' of ribonuclease (RNase A, 4396 U mg-', Worthington Biochemical Corp., Freehold, NJ, USA) in PBS for 30 min at room temperature. After enzyme treatment, the sample was mixed with 1 ml of 50 g ml-' of propidium iodide (Calbiochem, San Diego, CA, USA) in PBS and kept at room temperature for 60 min for a DNA assay (Dean et al., 1982). The DNA content per cell was assayed by flow cytometry, using a FACScan (Becton-Dickinson, Sunnyvale, CA, USA), with collection of fluorescence emissions having wavelengths longer than 590 nm. Some I05 cells were analysed and the distribution histograms of the fluorescence intensity in linear scale were obtained. Cell cycle analysis by DNA distribution was performed by using the 'CCANA 1' program reported by Dean (1980), and the proportions in the G,, S and G2-M phases were calculated. Each data point represents the mean of three experiments. The same experiment and analysis were carried out for the untreated controls, the L-dopa only, and the radiation only groups.

Results
Killing effects of L-dopa In order to examine the killing effects of L-dopa alone, exponentially growing cells were exposed to 100 Lg ml-' of L-dopa for 0-6 hours and survival fractions were determined using a colony-forming assay. The plating efficiency of the untreated HMV-I cells was 78 ± 8%, and that of the HeLa S3 cells, 60 ± 7%. There was no significant reduction in the survival fraction in cells treated with L-dopa.
Next, HMV-I cells in the confluent state were exposed to I00 Lg ml' of L-dopa for 0-6 hours. The plating efficiency of the untreated HMV-I cells in the confluent state was 76 ± 3%, and there was no significant reduction in the survival fraction in cells treated with L-dopa.

Radiation sensitivities
The radiation dose-response curve of the HMV-I and HeLa S3 cells are shown in Figure 1. Using a least-squares regression analysis to fit the survival curve, HMV-I cells had a Do of 1.49 ± 0.21 Gy, a Dq of 2.92 ± 0.23 Gy, and an n of 7.0 ± 1.33. Thus, the survival curve of the HMV-I cells had a broader shoulder region than that of the HeLa S3 cells which showed a Do of 1.41 ± 0.13 Gy, a Dq of 1.89 ± 0.18 Gy, and an n of 3.8 ± 1.20.
When the HMV-I cells were irradiated and then exposed to 100 lig ml-' of L-dopa for 4 h, there was significant reduction in the cell survival in each of the radiation doses examined, compared with the untreated cells (P <0.05) (Figure 1). Further, L-dopa decreased the shoulder width of the survival curve considerably, as indicated by the values of a Do of 1.42±0.11 Gy, a Dq of 0.61 ±0.24Gy, and an n of 1.5 ± 1.12 (Table I). However, the same treatment with Ldopa did not significantly affect the radiation sensitivity of the HeLa S3 cells, i.e. a Do of 1.41 Gy ± 0.18, a Dq of 1.90 ± 0.20 Gy, and an n of 4.0 ± 1.63.  Table III. Discussion PLD repair Figure 2 shows the survival curves of the HMV-I cells in the confluent state. When the HMV-I cells in the confluent state were incubated for 6 h after irradiation before replating, the repair of PLD was prominent; the ratio of the Do values before and after incubation being 1.7 (Table II). However, when L-dopa was added to the cell cultures immediately before irradiation and remained for 6 h before replating, PLD repair was significantly inhibited (P <0.01).

Inhibition of cell cycle progression
The DNA histograms shown in Figure 3 indicate that L-dopa alone did not produce significant changes in the cell cycle distribution compared to the untreated controls. An Melanoma cells possess a unique metabolic pathway for the conversion of L-dopa to melanin that is said to be mediated by tyrosinase (Pawelek, 1976). Further, it has been shown that L-dopa is selectively incorporated by melanoma cells and that it exhibits a selective cytotoxicity (Wick et al., 1977). Our present study has indicated that L-dopa potentiated radiation cytotoxicity towards human melanoma cells. The survival curve of exponentially growing HMV-I melanoma cells was modified by L-dopa treatment, and a decrease in the shoulder portion was especially prominent. Many investigators have demonstrated that the large shoulder in the survival curve of melanoma cells may be related to the poor radiation response that has been clinically observed in human melanomas (Barranco et al., 1971;Fertil & Malaise, 1981). Furthermore, Sasaki (1987) has demonstrated that tumour   E Z l cells. However, Weichselbaum and 'Little (1982a,b) have de-6 8 10 12 monstrated that, the larger the fraction dose, the more promse (Gy) inent the repair of the PLD in the melanoma cells, and they have suggested that the greater repair of PLD may be cells and its inhibition by L-another important factor determining the poor radiation related into 60 mm Petri disheS sponse in melanomas. In the management of melanomas, re irradiated with graded doses therefore, L-dopa may also be an effective agent for r immediately after irradiation inhibiting the PLD repair.
'LD repair (0), the cells were In addition, the enhancement of cell killing was found to 60-mdsefoanasyo'* 60-mm dishes for an assay of be associated with increased blockage of the HMV-I cells by 100fgfmlof L-dopa for 6Ph L-dopa in the G2-M phase. Our data indicated that L-dopa *esent a mean of three to five exerted a similar degree of inhibition in cell cycle progression at different phases of the cell cycle, and that the HMV-I cells were most sensitive to radiation damage during the G2-M phase. Thus, the increased blockage caused by L-dopa during treated with L-dopa the G2-M phase may indicate increased cell damage and an inability to proliferate or, alternatively, may merely reflect oa (Gy -') (Gy-2) the accumulation of dead cells during the G2-M phase, and 0. 175 ± 0.010 0.044 ± 0.010 hence the potentiation of radiation cell killing by the L-dopa.
The biochemical mechanism of radiosensitisation by L-0.143 ± 0.007 0.026 ± 0.004 dopa remains to be studied. Wick et al. (1977) and Wick (1978) have postulated that L-dopa acts on melanoma cells 0.209 ± 0.012 0.035 ± 0.005 through an initial conversion to quinones mediated by tryosinase and a subsequent scavenging action on the sulphydryl groups, by which DNA polymerase a may be inactivated. This hypothesis is supported by the fact that 1,2-benzoquinones have a marked affinity for DNA iave remarkably high survival polymerase a (Graham et al., 1978) and that L-dopa inhibits 'ing a course of fractionated the activity of DNA polymerase a only in the presence of n, 30 fractions), when com-tyrosinase (Wick, 1980). Recently, Lonn and Lonn (1985) ig a smaller shoulder. Thus, have demonstrated that DNA polymerase a is involved in the kay be an effective means of repair process of DNA lesions induced by X-ray irradiation n the radiation survival curve in human melanoma cells. Hence, it might be possible that seen in melanomas. Our data indicated that non-melanoma control cells were unaffected by the same L-dopa treatment. This suggests that L-dopa may potentiate the radiation toxicity selectively in melanoma cells, and probably does not affect normal tissue that has no tyrosinase. Thus, the therapeutic ratio in the management of melanomas may be enhanced by the use of L-dopa.
Our present data have also demonstrated that the PLD repair of the HMV-I melanoma cells was significantly inhibited by post-radiation incubation of the irradiated cells with L-dopa. Recently, radiotherapy using large dose per fraction has been proposed for the therapy of human melanoma (Habermalz & Fischer, 1976;Overgaard, 1980), aiming at overcoming the large shoulder in the melanoma Radiation (10 Gy) 47 ± 5 25 ± 8 28 ± 2 Radiation + L-dopa 44 ± 7 18 ± 4 38 ± 3 I a L-dopa inhibits the repair of radiation-induced DNA lesions through inactivating DNA polymerase a, and thus potentiates the radiation cytotoxicity in melanoma cells.