Megadoses of Ascorbate as a New Chemotherapeutic Approach in Uveal Melanoma: A Preliminary In Vitro Investigation

Background: Despite the more recent advances, there is still no effective systemic therapy for Uveal Melanoma (UM). However, a better understanding of the role of oxidative stress in cancer, has more recently led to a completely new approach to the systemic therapy of cancer, and modulators of the oxidative balance, such as sodium ascorbate (ASC) or arsenic trioxide (ATO), have already entered advanced phases of preclinical and clinical development.

Introduction and the resulting cell suspension, centrifuged for 5 minutes at 500 × g. After a second PBS 1× washing and centrifugation step, the pellet was re-suspended in fresh culture medium, at the concentration of about 3-4 × 10 5 /ml, before exposure to ASC or ATO.

Cell count and viability
Cell counting, before and after exposure to increasing doses of either ASC or ATO, was performed with the automated ("Muse"™) method according to the instructions supplied by the manufacturer, which encompass an in house method of nuclear staining for the assessment of cell viability.

Experimental protocol
During the phase of exponential growth, the cells, detached and collected as described, where counted with Trypan Blue, diluted to a concentration of about 3-4 × 10 5 /ml, and plated, in flat bottom twelvewell plates, at a volume of 1 ml (3-4 × 10 5 cells) per well.
Four dilutions of 2, 4, 6, and 8 μg/ml of ATO were used, starting from a stock solution of ATO commercially available at a concentration of 1 mg/ml (Trisenox™), by adding 2, 4, 6, and 8 μl of Trisenox™, respectively, to the four different wells, each containing 1 ml (3-4 × 10 5 OCM1 or C918 cells) of cell suspension. The concentrations of ATO were chosen according to the data reported by Yediou and coll. in acute promyelocytic leukemia cell lines (HL60) [28]. ATO can be stored at room temperature for up to 36 months.
A 1 M solution of ASC was prepared, fresh each time, as described elsewhere [29], and aliquots of 1, 3, 5, and 7 μl of this stock solution were added to the four wells, each containing 1 ml (3-4 × 10 5 OCM1 or C918 cells) of cell suspension, to a final concentration of 1, 3, 5, and 7 mM, respectively.
Both OCM1 and C918 cells were exposed for 18-24 hours. One control (no treatment) sample was also included in each experiment. At the end of the incubation period the cells were collected in vials and mixed with the Muse™ Count & Viability Reagent, according to the procedure supplied by the manufacturer, for automated counting and viability analysis. Namely, 10 μl of each sample were added to 190 μl of the Muse™ Count & Viability Reagent, which differentially stains viable and non-viable cells based on their permeability to the two DNA binding dyes present in the reagent. A specific software module then performs calculations automatically and displays data in two different types of dot plots, one of which is reported in Figure 2.
Each count and viability test was repeated at least twice, and a total of 96 tests were carried out, with the automated cell counter ("Muse"™), with and without (control sample) scalar doses of ATO and ASC, as reported.
A second group of tests was also performed, but this time by exposing both OCM1 and C918 cell lines to both ASC and ATO, for only two hours. Moreover, in order to explore potential synergisms between both ASC and ATO, part of these tests was carried out by mixing the lowest concentrations of both ATO (2 and 4 μg/ml) and ASC (1, 3, and 5 mM/ml).
After exposure, the cells were washed twice with PBS 1×, resuspended in fresh medium, and further incubated for 18-24 hours. Even in this case, each count and viability test was repeated at least twice and a total of 72 tests were performed with the automated cell counter ("Muse"™). involvement of alterations of the redox (reduction/oxidation) balance in tumor progression, particularly for oncological indications with little treatment options, including metastatic melanoma [13][14][15][16][17].
In this perspective, arsenic trioxide (ATO) (chemical formula As 2 O 3 ), a pro-oxidant drug, introduced a few years ago in the treatment of acute promyelocytic leukemia [18], has undergone clinical testing in the treatment of metastatic melanoma [19,20] with some results, particularly in in vitro studies [21,22].
More importantly, sodium ascorbate (ASC) delivered at very high doses that raise plasma concentrations to millimolar (mM) levels, not achievable by oral administration, has shown significant anticancer effects in vitro, due to its pro-oxidant activity, with increased production of intracellular H 2 O 2 , and consequent oxidative damage [23][24][25].
Given all the above, we have undertaken the present in vitro investigation to test the effectiveness of both ATO and high doses of ASC in killing UM cells.

Cells, reagents, and equipment
OCM1 and C918 UM cell lines used in this study ( Figure 1) originate from primary uveal melanomas and have undergone extensive authentication including genetic analysis as described elsewhere [26].
All reagents, including culture media, sodium ascorbate (ASC), calcium chloride, sodium citrate, sodium phosphate, sodium chloride, potassium chloride, and Trypan Blue, were purchased from Sigma-Aldrich (St. Louis MO). ATO (Trisenox™) was kindly supplied by the dept. of Hematology of the hospital of Pescara (Italy). Automated cell count and viability was performed by using the "Muse"™ (Merck-Millipore) automated cell analyzer.
For count and viability, the cells were detached/disaggregated by preparing a 10× solution containing 1.35 M potassium chloride, and 0.15 M sodium citrate dehydrate (citrate saline). This stock solution was diluted to 1× just before use [27]. To detach UM cells growing in monolayer from the bottom of flasks, they were incubated for 5 minutes with citric saline solution at 37°C in 5% CO 2 /95% air, and then mechanically detached by gently tapping the flasks, as for the trypsinization procedure. An equal volume of PBS 2× was then added,

Statistical analysis
The statistical analysis was performed, for each independent sample (Table 1), by comparing the proportion of viable cells at any given dosage (of either ASC or ATO), with the effect of the chosen dose on cell viability (response). The Chi Square (ChiSq) statistics with three degrees of freedom (DF) was used to summarize the information for each experiment at a suitable scale level of cell count.

Results
After exposure to increasing concentrations of ASC (1, 3, 5, and 7 mM), both OCM1 and C918 UM cell lines showed a consistent decrease in cell viability, with an inverse relationship between the concentration of ASC (in mM/ml) and the number of live cells in culture. The same did not apply to ATO in which, after an initial decrease of cell viability, at the concentration of 2 μg/ml, the percentage of live cells in culture   Figure  2, which shows the distribution of live/dead cells in a representative sample analyzed by flow cytometry, and in Table 1, reporting the percentages of live cells after exposure to any given concentration of either ASC or ATO.
As shown in Table 2, when the number of viable cells in each independent sample, at any given dosage of either ASC or ATO, were compared with the effect of that dose on cell viability, a high statistically significant (p<<0.01) dose-response relationship could be found for ASC in all (6/6) experiments, while this was true only in two out of six experiments performed with ATO. This was a clear indication that at the chosen dosages, ASC killed a significantly higher number of UM cells than ATO, in vitro, and that this (cytotoxic) effect was strictly proportional to the dose of ASC (but not of ATO) employed.
To further confirm the statistical analysis, the data reported in Table  1 clearly show that while ATO never reaches the LC 50 , ASC reaches its LC 50 at around 3 mM/ml, i.e. less than a half of the highest concentration (7 mM) used in this experiment.
When the responses of both OCM1 and C918 to ASC and ATO were compared (Figure 3), a rather similar pattern was observed, with both cell lines; i.e.: ATO showed some effects on cell viability at the dosage of 2 and 4 μg/ml, but no effect at all at higher doses, while ASC produced a constant and progressive decrease in cell survival at any of the chosen concentrations.
By reducing the exposure of both UM cell lines to two hours, an evident decrease in the effectiveness of both ATO and ASC could be appreciated, even if ASC still killed a considerably higher number of UM cells than ATO. This led us to conclude that ASC is rapidly internalized, after exposure, and remains inside the cells even after washing it out from the culture medium.
More importantly, when the lowest doses of ATO and ASC are mixed together, they kill significantly more UM cells than if taken alone, thus indicating a clear cytotoxic synergism of both ATO and ASC, when administered together to either OCM1 or C918 UM cells (Figure 4).

Discussion
The levels of ROS, and particularly of H 2 O 2 are tightly controlled under physiological conditions, to maintain redox homeostasis in the melanosome and regulate the melanogenesis process [30].

Run 1 Run1
Control  The human melanocyte is continually exposed to ROS, and the oxidative balance is maintained by intracellular antioxidants, but this control is usually lost during melanoma development [16] and the disruption of this balance seems to be an early event in the etiology of melanoma, leading to increased oxidative stress and mutation [31].
Epidemiological investigations implicate UVB in non-melanoma and UVA in melanoma skin cancer, and while UVB are known to mediate their action through photoproducts, such as pyrimidine dimers, it is well known that UVA acts mainly through the induction of ROS [28] and unbalance of the cellular redox steady state.
On the other hand, redox deregulation due to metabolic alterations, and the role of ROS in signal transduction, represent a specific vulnerability of malignant cells, which can be selectively targeted by redox chemotherapeutics [18,32] aimed at either increasing the levels of intracellular ROS [32] or decreasing the amount of intracellular antioxidants in cancer cells [18,33].
Among the candidate molecules in redox directed cancer chemotherapeutics, ATO (As 2 O 3 ) and high doses of ASC have more recently received the attention of the scientific community [18].
ATO is studied as an anticancer agent whose use originated in traditional Chinese Medicine. It is an established cell growth inhibitor and apoptosis inducer [34,35] and, in patients with Acute Promyelocytic Leukemia (APL) it has been variously defined as the "most effective single agent" [36], "the most biologically active single drug" [37], and more recently, "a drug from a poison" to be used not only in the treatment of this leukemia, but also in other types of cancer [38], for its efficacy, cost effectiveness, and low toxicity [39,40].
As reported by different authors [18,[41][42][43] oxidative stress is critical to the ATO-induced apoptotic process, with generation of ROS, including, among others, H 2 O 2 , which is considered a metabolite of the drug [44].
Interestingly, ATO shares this pro-oxidant activity with a number of other compounds [18], including ASC when administered in high doses by intravenous injection; at these "pharmacological" concentrations, ASC is a powerful pro-oxidant and is considered a pro-drug of H 2 O 2 [23,25,[45][46][47].
In the experiments reported herein, ASC is significantly more efficient than ATO in killing both OCM1 and C918 UM cells in vitro, when used at doses commonly employed in clinical settings.
Although ASC has shown its selective cytotoxicity for different cancer cell types [23][24][25] and ATO has been already used to treat UM cell lines in vitro [21,22], this is, to our knowledge, the first report ever, concerning the effects of ASC alone or in combination with ATO on UM cells lines.
In this study, ASC has been shown significantly more efficient than ATO in killing UM cells, with an LC 50 of less than 3 mM/ml, a plasma concentration which can be easily reached by intravenous administration of this vitamin.
Interestingly, by reducing the exposure of UM cell lines to two hours, we have been able to see the same effects obtained after 18-24 of continuous exposure of UM cells to ASC, thus indicating that ASC is captured by UM cells, as also reported in the literature on this subject [18].
Regarding the comparison between the doses of ASC (in mM/ml) and ATO (in μg/ml), it may be argued that is not appropriate, given the difference in the dosage used, but this is not the case since the doses of both compounds used herein, are those commonly employed for routine therapeutic purposes [28]. With specific reference to ASC, it should be emphasized that in a recent clinical trial on metastatic pancreatic cancer, ASC has been administered by intravenous injection in doses ranging from 50 to 100 grams per infusion, three days a week for eight weeks, without major side effects, and plasma concentrations of up to 30 mM/ml [48], i.e. more than fourfold the maximal dose (7 mM) of ASC used in the present investigation.
More importantly, when used in combination, both ASC and ATO show a clear synergism of action as illustrated in the diagrams of Figure 4, showing the cytotoxic effects of ATO and ASC alone and in comparison with the combination of both ("mix").
When considering that: 1. ASC in high intravenous doses is a powerful pro-oxidant with cytotoxic effects on a number of different human tumor cell lines [18,[23][24][25] (including retinoblastoma [28], and promyelocytic leukemia [49]) and no harmful effects on normal cells [