Oxidative Status Determines the Cytotoxicity of Ascorbic Acid in Human Oral Normal and Cancer Cells

Oral squamous cell carcinoma (OSCC) can arise anywhere in the oral cavity. OSCC’s molecular pathogenesis is complex, resulting from a wide range of events that involve the interplay between genetic mutations and altered levels of transcripts, proteins, and metabolites. Platinum-based drugs are the first-line treatment for OSCC; however, severe side-effects and resistance are challenging issues. Thus, there is an urgent clinical need to develop novel and/or combinatory therapeutics. In this study, we investigated the cytotoxic effects of pharmacological concentrations of ascorbate on two human oral cell lines, the oral epidermoid carcinoma meng-1 (OECM-1) cell and the Smulow–Glickman (SG) human normal gingival epithelial cell. Our study examined the potential functional impact of pharmacological concentrations of ascorbates on the cell-cycle profiles, mitochondrial-membrane potential, oxidative response, the synergistic effect of cisplatin, and the differential responsiveness between OECM-1 and SG cells. Two forms of ascorbate, free and sodium forms, were applied to examine the cytotoxic effect and it was found that both forms had a similar higher sensitivity to OECM-1 cells than to SG cells. In addition, our study data suggest that the determinant factor of cell density is important for ascorbate-induced cytotoxicity in OECM-1 and SG cells. Our findings further revealed that the cytotoxic effect might be mediated through the induction of mitochondrial reactive oxygen species (ROS) generation and the reduction in cytosolic ROS generation. The combination index supported the agonistic effect between sodium ascorbate and cisplatin in OECM-1 cells, but not in SG cells. In summary, our current findings provide supporting evidence for ascorbate to serve as a sensitizer for platinum-based treatment of OSCC. Hence, our work provides not only repurposing of the drug, ascorbate, but also an opportunity to decrease the side-effects of, and risk of resistance to, platinum-based treatment for OSCC.


Introduction
Oral cancer is the sixth leading cause of global cancer-related deaths [1,2]. Oral squamous carcinogenesis is a multistep process involved in multiple genetic events including the mutation of oncogenes and tumor-suppressor genes. Oral squamous cell carcinoma (OSCC) can arise anywhere in the oral cavity, including the tongue, upper and lower gingiva, and buccal mucosa [3][4][5]. Excessive alcohol consumption, betel nut chewing, and smoking are the three major risk factors for OSCC [6]. OSCC's molecular pathogenesis involves the interplay between genetic mutations and altered levels of transcripts, proteins, and metabolites. The overall 5-year survival rate of OSCC has remained at 50%, despite intensive research [7]. Platinum-based drugs are the first-line treatment for OSCC; however,

The Effect of Ascorbate on the Metabolic Activity, Cell-Cycle Profile, and Cellular Proliferation in OECM-1 and SG Cells
To investigate the cytotoxicity of ascorbate in oral cells, two forms of ascorbates, Lfree acid and sodium base, were applied to monitor their cytotoxic effects on one oral epidermoid carcinoma meng-1 (OECM-1) cell, originally derived from the primary culture of an oral cancer patient by Professor Meng CL, and compared it with a human Smulow-Glickman (SG) normal gingival epithelial cell [35], using MTT analysis. We observed the differential cytotoxicity of both cell lines where L-ascorbic acid and sodium ascorbate were applied to cells over 2 mM (Figure 1). The responsiveness of the OECM-1 cells to both forms of ascorbates was more sensitive than that of the SG cells ( Figure 1A,B). Sodium ascorbate had a similar cytotoxic effect to that of L-ascorbic acid on OECM-1 and SG cells.
ascorbate were applied to cells over 2 mM (Figure 1). The responsiveness of the OECM-1 cells to both forms of ascorbates was more sensitive than that of the SG cells ( Figure 1A,B). Sodium ascorbate had a similar cytotoxic effect to that of L-ascorbic acid on OECM-1 and SG cells.  (6 × 10 4 ) and SG (6 × 10 4 ) cells were cultured in 24-well plates and treated with the indicated concentrations of L-ascorbic acid (A) for 5 h and sodium ascorbate (B) for 3 h. Cell viability was measured using the MTT method. Bars depict the mean ± SD of three independent experiments. Student's t-tests were analyzed and compared with vehicle. * p < 0.05, ** p < 0.01, and *** p < 0.001.
We further examined the effect of ascorbates on the cell-cycle profiles of OECM-1 and SG cells ( Figure 2). Sodium ascorbate decreased the populations of the G1 and G2/M phases and increased the populations of the subG1 and S phases in OECM-1 and SG cells (Figure 2A,B). These cell-cycle-related proteins were examined for their expression changes using RT-PCR and Western Blot analysis ( Figure 2C,D). Dose-dependent decreasing trends in p53, p21, and cyclin D1 mRNAs were more apparent in SG cells than in OECM-1 cells compared with consistent amount of internal control β-actin mRNA ( Figure  2C). In Western Blot analysis, dose-dependent decreasing trends in p21 and cyclin D1 proteins were found in OECM-1 cells, and similar trends were found in p53 and cyclin D1 proteins in SG cells ( Figure 2D). and SG (6 × 10 4 ) cells were cultured in 24-well plates and treated with the indicated concentrations of L-ascorbic acid (A) for 5 h and sodium ascorbate (B) for 3 h. Cell viability was measured using the MTT method. Bars depict the mean ± SD of three independent experiments. Student's t-tests were analyzed and compared with vehicle. * p < 0.05, ** p < 0.01, and *** p < 0.001.
We further examined the effect of ascorbates on the cell-cycle profiles of OECM-1 and SG cells ( Figure 2). Sodium ascorbate decreased the populations of the G1 and G2/M phases and increased the populations of the subG1 and S phases in OECM-1 and SG cells (Figure 2A,B). These cell-cycle-related proteins were examined for their expression changes using RT-PCR and Western Blot analysis ( Figure 2C,D). Dose-dependent decreasing trends in p53, p21, and cyclin D1 mRNAs were more apparent in SG cells than in OECM-1 cells compared with consistent amount of internal control β-actin mRNA ( Figure 2C). In Western Blot analysis, dose-dependent decreasing trends in p21 and cyclin D1 proteins were found in OECM-1 cells, and similar trends were found in p53 and cyclin D1 proteins in SG cells ( Figure 2D).
The dose-dependent increasing trend of the population of the S phase suggested that cellular proliferation might be involved in the cytotoxicity of oral cells. Hence, we applied BrdU proliferation analysis to examine the effect of sodium ascorbate in OECM-1 and SG cells ( Figure 3A-D). Our BrdU results showed that sodium ascorbate first increased the proliferation rate and then decreased the proliferation rate in both cells ( Figure 3B,D). However, the decreasing effect in the SG cells was more apparent than in the OECM-1 cells (Figure 3 compared B and D).

Modulation of Oxidative-Stress Proteins and Cell Density by Ascorbatein OECM-1 and SG Cells
We sought to elucidate the differential cytotoxic effect between OECM-1 and SG cells. We applied L-ascorbic acid and sodium ascorbate to examine the change in p-ERK/ERK (cell-survival marker), γH2A.x (DNA-damage marker), p21 (senescence marker), p62 (autophagy marker), HO-1 (oxidative-stress marker), and Nrf2 (oxidative-stress marker) using Western Blot analysis ( Figure 4A). We observed an increasing p-Erk/Erk ratio and γH2A.x, and decreasing p21 proteins, in OECM-1 and SG cells ( Figure 4B). The decreasing p62 and HO-1 proteins were only found in treatment with sodium ascorbate, not in treatment with L-ascorbic acid, in both cell types. A decrease in Nrf2 proteins was found in OECM-1 treated with sodium ascorbate and L-ascorbic acid ( Figure 4). The effects of ascorbate on Nrf2 whether shorter exposure or longer exposure showed that the decrease of in Nrf2 (molecular weight of approximately 60 kDa) and the increase in Nrf2 (molecular weight of approximately 70 kDa) in OECM-1 cells.

Figure 2.
Effects of L-ascorbic acid and sodium ascorbate on the cell-cycle profile and related gene and protein expressions in OECM-1 and SG cells. OECM-1 (3.5 × 10 5 ) and SG (3.5 × 10 5 ) cells were cultured in 6-well plates and treated with the indicated concentrations of L-ascorbic acid and sodium ascorbate for 4 h. They were then subjected to (A,B) cell-cycle profile analysis, (C) RT-PCR analysis, and (D) Western Blot analysis. (C) RT-PCR analysis was performed with 1 µ g total RNA and b-actin was a loading mRNA control. (D) The cell lysates (30 µ g total proteins) were subjected to Western Blot analysis using antibodies against the indicated proteins. ACTN was a loading-protein control. (A,B) Bars depict the mean ± SD of three independent experiments. Student's t-tests were analyzed and compared with vehicle. * p < 0.05, ** p < 0.01, and *** p < 0.001. The mRNA and protein bands (C,D) were quantified through pixel density scanning and evaluated using Image J, version 1.44a (http://imagej.nih.gov/ij/) (accessed on 1 February 2023). The ratios of mRNA/β-actin (C) and protein/ACTN (D) were listed in the OECM-1 and SG cells.
The dose-dependent increasing trend of the population of the S phase suggested that cellular proliferation might be involved in the cytotoxicity of oral cells. Hence, we applied BrdU proliferation analysis to examine the effect of sodium ascorbate in OECM-1 and SG Figure 2. Effects of L-ascorbic acid and sodium ascorbate on the cell-cycle profile and related gene and protein expressions in OECM-1 and SG cells. OECM-1 (3.5 × 10 5 ) and SG (3.5 × 10 5 ) cells were cultured in 6-well plates and treated with the indicated concentrations of L-ascorbic acid and sodium ascorbate for 4 h. They were then subjected to (A,B) cell-cycle profile analysis, (C) RT-PCR analysis, and (D) Western Blot analysis. (C) RT-PCR analysis was performed with 1 µg total RNA and b-actin was a loading mRNA control. (D) The cell lysates (30 µg total proteins) were subjected to Western Blot analysis using antibodies against the indicated proteins. ACTN was a loading-protein control. (A,B) Bars depict the mean ± SD of three independent experiments. Student's t-tests were analyzed and compared with vehicle. * p < 0.05, ** p < 0.01, and *** p < 0.001. The mRNA and protein bands (C,D) were quantified through pixel density scanning and evaluated using Image J, version 1.44a (http://imagej.nih.gov/ij/) (accessed on 1 February 2023). The ratios of mRNA/β-actin (C) and protein/ACTN (D) were listed in the OECM-1 and SG cells. Figure 3A-D). Our BrdU results showed that sodium ascorbate first increased the proliferation rate and then decreased the proliferation rate in both cells ( Figure 3B,D). However, the decreasing effect in the SG cells was more apparent than in the OECM-1 cells (Figure 3 compared B and D).

Modulation of Oxidative-Stress Proteins and Cell Density by Ascorbatein OECM-1 and SG Cells
We sought to elucidate the differential cytotoxic effect between OECM-1 and SG cells. We applied L-ascorbic acid and sodium ascorbate to examine the change in p-ERK/ERK (cell-survival marker), γH2A.x (DNA-damage marker), p21 (senescence marker), p62 (autophagy marker), HO-1 (oxidative-stress marker), and Nrf2 (oxidative-stress marker) using Western Blot analysis ( Figure 4A). We observed an increasing p-Erk/Erk ratio and γH2A.x, and decreasing p21 proteins, in OECM-1 and SG cells ( Figure 4B). The decreasing However, the changes in Nrf2, HO-1, and the p-Erk/Erk ratio were unstable in our analytic systems. Based on the fact that cell density and the responsiveness of ascorbate are related [36], we examined the differential responsiveness in various cell densities of the OECM-1 and SG cells. In sodium-ascorbate-treated OECM-1 cells, decreasing Nrf2 and HO-1 proteins were found in all testing-cell densities, but an increased p-Erk/Erk ratio was found in higher cell densities ( Figure 5A). In sodium-ascorbate-treated SG cells, decreasing Nrf2 and HO-1 proteins were found in lower cell densities, but an increased p-Erk/Erk ratio was found in higher (2.5 × 10 5 ) cell densities. In L-ascorbic acid-treated OECM-1 and SG cells, the increased p-Erk/Erk ratio was found in all testing-cell densities and the increased HO-1 proteins were found in higher (2.5 × 10 5 ) cell densities ( Figure 5B). The decreased Nrf2 proteins resulting from L-ascorbic acid were found in the OECM-1 cells, but not in the SG cells. Our current data showed that the effects of sodium ascorbate and L-ascorbic acid varied in different cell densities of OECM-1 and SG cells, suggesting that relative cell numbers might be a determinant for the action of ascorbate in cells.
p62 and HO-1 proteins were only found in treatment with sodium ascorbate, not in treatment with L-ascorbic acid, in both cell types. A decrease in Nrf2 proteins was found in OECM-1 treated with sodium ascorbate and L-ascorbic acid ( Figure 4). The effects of ascorbate on Nrf2 whether shorter exposure or longer exposure showed that the decrease of in Nrf2 (molecular weight of approximately 60 kDa) and the increase in Nrf2 (molecular weight of approximately 70 kDa) in OECM-1 cells. Figure 4. Effects of sodium ascorbate and L-ascorbic acid on the oxidative-stress proteins in OECM-1 and SG cells. OECM-1 (2.5 × 10 5 ) and SG cells (2.5 × 10 5 ) were cultured in 6-well plates and treated with 10 mM sodium ascorbate for 3.5 h or L-ascorbic acid for 5.5 h. (A) The cell lysates (30 µ g total proteins) were subjected to Western Blot analysis using antibodies against the indicated proteins. ACTN was a loading-protein control. The protein bands (B) were quantified through pixel density scanning and evaluated using Image J, version 1.44a (http://imagej.nih.gov/ij/) (accessed on 1 February 2023). The ratios of p-Erk/Erk and protein/ACTN were plotted in the OECM-1 and SG cells.
However, the changes in Nrf2, HO-1, and the p-Erk/Erk ratio were unstable in our analytic systems. Based on the fact that cell density and the responsiveness of ascorbate are related [36], we examined the differential responsiveness in various cell densities of the OECM-1 and SG cells. In sodium-ascorbate-treated OECM-1 cells, decreasing Nrf2 and HO-1 proteins were found in all testing-cell densities, but an increased p-Erk/Erk ratio was found in higher cell densities ( Figure 5A). In sodium-ascorbate-treated SG cells, decreasing Nrf2 and HO-1 proteins were found in lower cell densities, but an increased p-Erk/Erk ratio was found in higher (2.5 × 10 5 ) cell densities. In L-ascorbic acid-treated OECM-1 and SG cells, the increased p-Erk/Erk ratio was found in all testing-cell densities and the increased HO-1 proteins were found in higher (2.5 × 10 5 ) cell densities ( Figure 5B). The decreased Nrf2 proteins resulting from L-ascorbic acid were found in the OECM-1 cells, but not in the SG cells. Our current data showed that the effects of sodium ascorbate and L-ascorbic acid varied in different cell densities of OECM-1 and SG cells, suggesting that relative cell numbers might be a determinant for the action of ascorbate in cells.

The Effect of Ascorbate on Cytosolic and Mitochondrial ROS Generation in OECM-1 and SG Cells
Nrf2 is a master protein that responds to the status of reactive oxygen species (ROS in cells. It was interesting to address the status of ROS at a pharmacologic concentration of L-ascorbic acid and sodium ascorbate in OECM-1 and SG cells. We measured the cyto solic ROS levels of both types of cells using flow-cytometry analysis with DCFH-DA dye ( Figure 6A-D). Our data showed that sodium ascorbate decreased the ROS levels in both cell lines in a dose-dependent manner ( Figure 6A,C). However, L-ascorbic acid first de creased the ROS level and then returned to the increasing level in both cells ( Figure 6B,D) The cell lysates (30 µg total proteins) were subjected to Western Blot analysis using antibodies against the indicated proteins. ACTN was a loading-protein control. The protein bands (A) and (B) were quantified through pixel density scanning and evaluated using Image J, version 1.44a (http://imagej.nih.gov/ij/) (accessed on 1 February 2023). The ratios of protein/ACTN were listed in the OECM-1 and SG cells.

The Effect of Ascorbate on Cytosolic and Mitochondrial ROS Generation in OECM-1 and SG Cells
Nrf2 is a master protein that responds to the status of reactive oxygen species (ROS) in cells. It was interesting to address the status of ROS at a pharmacologic concentration of L-ascorbic acid and sodium ascorbate in OECM-1 and SG cells. We measured the cytosolic ROS levels of both types of cells using flow-cytometry analysis with DCFH-DA dye ( Figure 6A-D). Our data showed that sodium ascorbate decreased the ROS levels in both cell lines in a dose-dependent manner ( Figure 6A,C). However, L-ascorbic acid first decreased the ROS level and then returned to the increasing level in both cells ( Figure 6B,D).
We further measured the increasing trend in mitochondrial ROS, compared with the cytosolic ROS status following treatment with sodium ascorbate of the OECM-1 and SG cells (Figure 7). We observed the induction of mitochondrial ROS by sodium ascorbate in a dose-dependent manner in OECM-1 cells ( Figure 7A). The amount of mitochondrial ROS first decreased and then returned to its basal level following treatment with sodium ascorbate of the SG cells ( Figure 7B).
The increasing trend in subG1 by sodium ascorbate revealed that mitochondrial dysfunction might be involved. Hence, we measured mitochondrial-membrane potential using flow-cytometry analysis with JC-1 dye ( Figure 8A-D). Our data showed that the mitochondrial-membrane potential was significantly disrupted by sodium ascorbate in the OECM-1 and SG cells ( Figure 8A,B). However, the disruption of mitochondrial-membrane potential was observed at the level of 1 mM sodium ascorbate in the OECM-1 cells, but not in the SG cells ( Figure 8C,D). We further measured the increasing trend in mitochondrial ROS, compared with the cytosolic ROS status following treatment with sodium ascorbate of the OECM-1 and SG cells (Figure 7). We observed the induction of mitochondrial ROS by sodium ascorbate in a dose-dependent manner in OECM-1 cells ( Figure 7A). The amount of mitochondrial ROS first decreased and then returned to its basal level following treatment with sodium ascorbate of the SG cells ( Figure 7B). The increasing trend in subG1 by sodium ascorbate revealed that mitochondrial dysfunction might be involved. Hence, we measured mitochondrial-membrane potential using flow-cytometry analysis with JC-1 dye ( Figure 8A-D). Our data showed that the mitochondrial-membrane potential was significantly disrupted by sodium ascorbate in the OECM-1 and SG cells ( Figure 8A,B). However, the disruption of mitochondrial-membrane potential was observed at the level of 1 mM sodium ascorbate in the OECM-1 cells, but not in the SG cells ( Figure 8C,D).

The Combination Index between Cisplatin and Ascorbate in OECM-1 and SG Cells
Cisplatin is a standard OSCC chemotherapeutic agent and has the ability to induce oxidative stress via the production of ROS [32][33][34]. It was interesting to address the combination of sodium ascorbate with cisplatin in OECM-1 and SG cells. Isobologram analysis has been mathematically proven and widely used to evaluate drug interactions [37]. The dose of the same efficacy when two drugs are used alone, which is generally expressed as half-effective dose (ED50). Based on IC50 values ( Figure 1) and the classic experimental design to calculate the combination index (CI) [38], we designed combinations of concentrations and calculated the CI between cisplatin and sodium ascorbate in the OECM-1 and SG cells. The definition of CI < 1 as a synergistic effect was observed in the OECM-1 cells and the combination of cisplatin plus sodium ascorbate was CI > 1 in the SG cells ( Figure  9A,B). The ED50 of cisplatin decreased from 16.2 to 0.7 μM in the presence of 1.41 mM sodium ascorbate (ED50 = 3.5 mM) in the OECM-1 cells ( Figure 9A). Although the combination of cisplatin with sodium ascorbate was antagonistic, the ED50 of cisplatin decreased from 58.6 to 1.6 μM in the presence of 3.1 mM sodium ascorbate (ED50 = 2 mM) in the SG cells ( Figure 9B).

The Combination Index between Cisplatin and Ascorbate in OECM-1 and SG Cells
Cisplatin is a standard OSCC chemotherapeutic agent and has the ability to induce oxidative stress via the production of ROS [32][33][34]. It was interesting to address the combination of sodium ascorbate with cisplatin in OECM-1 and SG cells. Isobologram analysis has been mathematically proven and widely used to evaluate drug interactions [37]. The dose of the same efficacy when two drugs are used alone, which is generally expressed as half-effective dose (ED 50 ). Based on IC 50 values ( Figure 1) and the classic experimental design to calculate the combination index (CI) [38], we designed combinations of concentrations and calculated the CI between cisplatin and sodium ascorbate in the OECM-1 and SG cells. The definition of CI < 1 as a synergistic effect was observed in the OECM-1 cells and the combination of cisplatin plus sodium ascorbate was CI > 1 in the SG cells ( Figure 9A,B). The ED 50 of cisplatin decreased from 16.2 to 0.7 µM in the presence of 1.41 mM sodium ascorbate (ED 50 = 3.5 mM) in the OECM-1 cells ( Figure 9A). Although the combination of cisplatin with sodium ascorbate was antagonistic, the ED 50 of cisplatin decreased from 58.6 to 1.6 µM in the presence of 3.1 mM sodium ascorbate (ED 50 = 2 mM) in the SG cells ( Figure 9B).

Discussion
Ascorbate has strong antioxidant properties at physiological concentrations, b may have pro-oxidant effects at pharmacological concentrations. Hence, we investig the cytotoxic effects of pharmacological concentrations of ascorbate on OECM-1 an cells. Our study examined the potential functional impact of pharmacological conce tions of ascorbates on the cell-cycle profiles, mitochondrial-membrane potential, oxid response, the synergistic effect of cisplatin, and the differential responsiveness bet OECM-1 and SG cells. Our study data suggest that the determinant factor of cell de is important in ascorbate-induced cytotoxicity in OECM-1 and SG cells. Our finding ther revealed that the cytotoxic effect might be mediated through the induction of chondrial ROS generation and the reduction in cytosolic ROS generation. The com tion index supported the agonistic effect between sodium ascorbate and cisplat OECM-1 cells, but not in SG cells. In summary, our current findings provide suppo evidence for ascorbate to serve as a sensitizer for platinum-based treatment of O Hence, our work provides not only the repurposing of the drug, ascorbate, but al opportunity to decrease the side-effects of, and risk of resistance to, platinum-based ment for OSCC.
Endogenous ROS has been reported to be generated, primarily, in mitochondri tosol, and peroxisomes. ROS has been demonstrated to be produced mainly in the

Discussion
Ascorbate has strong antioxidant properties at physiological concentrations, but it may have pro-oxidant effects at pharmacological concentrations. Hence, we investigated the cytotoxic effects of pharmacological concentrations of ascorbate on OECM-1 and SG cells. Our study examined the potential functional impact of pharmacological concentrations of ascorbates on the cell-cycle profiles, mitochondrial-membrane potential, oxidative response, the synergistic effect of cisplatin, and the differential responsiveness between OECM-1 and SG cells. Our study data suggest that the determinant factor of cell density is important in ascorbate-induced cytotoxicity in OECM-1 and SG cells. Our findings further revealed that the cytotoxic effect might be mediated through the induction of mitochondrial ROS generation and the reduction in cytosolic ROS generation. The combination index supported the agonistic effect between sodium ascorbate and cisplatin in OECM-1 cells, but not in SG cells. In summary, our current findings provide supporting evidence for ascorbate to serve as a sensitizer for platinum-based treatment of OSCC. Hence, our work provides not only the repurposing of the drug, ascorbate, but also an opportunity to decrease the side-effects of, and risk of resistance to, platinum-based treatment for OSCC.
Endogenous ROS has been reported to be generated, primarily, in mitochondria, cytosol, and peroxisomes. ROS has been demonstrated to be produced mainly in the mitochondrial-electron transport chain in prostate cancer cells treated with cisplatin. Consistent with our current findings for OSCC, the selectivity of mitochondria ROS is not parallel to cytosol ROS, and this might result from the complexity of the functional role of ascorbate within cellular redox homeostasis. Another issue is the differential effect of L-ascorbic acid and sodium ascorbic on the trends in cytosolic ROS generation. However, the acidity of free-form ascorbate has been demonstrated as not being the reason for the cytotoxicity for human cervical cancer cells. The transformation of Fe 3+ into Fe 2+ can be induced by antioxidant ascorbate, and the ascorbate-Fe 2+ complex may catalyze ROS generation via Fenton's reaction. Furthermore, the oxidation of ascorbate results in the formation of the ascorbate radical and a high flux of H 2 O 2 . Therefore, we should pay attention to whether our current ROS status findings result directly from ascorbate or possibly from ROS generated during different compartmentation in oral cells.
There are three lines of defense for a living cell's antioxidant system, including (I) biosynthesis and activation of antioxidant enzymes; (II) free radical scavenging; and (III) the repair of oxidative damage. The functional role of ascorbate is as a molecule involved in all these stages. Another aspect of the action of ascorbate is to regulate the biosynthesis of antioxidant proteins as an antioxidant. Three important transcriptional factors involved in the cellular antioxidant response are Nrf2, redox effector factor 1, and activator protein 1. The free form of Nrf2 from the Keap1/Nrf2 complex transfers to the nucleus and binds to the antioxidant response element to start the biosynthesis of antioxidant protein. Ascorbate is known as an Nrf2 activator, and its deficiency leads to impaired Nrf2 action, resulting in inflammation and apoptosis. One of the challenges is the multiple bands of Nrf2 in Western Blot analysis. Based on its target gene, HO-1, expression, we observed consistent changes in Nrf2/HO-1 by sodium ascorbic, not by L-ascorbic acid, in the OECM-1 cells. It was difficult to determine the relationship between Nrf2 and HO-1 in the SG cells. Much attention is focused on research literature discussing numerous commercial Nrf2 antibodies [39,40]. However, we need to determine the status of Nrf2, including its potential isoform(s) and different modifications in human oral cells. More importantly, we need to understand Nrf2 functions resulting from its transcriptional activity, target gene expression, or others, that may play a dual role in different cancer cells.
Intravenous L-ascorbic acid is able to bypass the tight control of the intestine, leading to higher plasma levels [41,42]. Hence, repurposing ascorbate for cancer therapy is an accessible and valuable means of treatment [22][23][24]. It was hypothesized that cancer cells generally demonstrate higher steady-state levels of ROS stress than normal cells because of defects in oxidative metabolism and the accumulation of labile iron. Detailed mechanisms related to hydrogen peroxide of ascorbate killing some cancer cells but having little effect on normal cells are not well known [27]. L-ascorbic acid promotes oxidation via hydrogen peroxide generation through pH-dependent auto-oxidation in the presence of a catalytic metal [43]. In addition to oxidative homeostasis, ascorbate also functions as a cofactor for the Fe 2+ -2-oxoglutarate-dependent dioxygenase family, involved to the stability of HIF-1α and Tet2. In general, pharmacologic doses of ascorbate exhibit anticancer effects through the induction of both oxidative stress and DNA demethylation.

Cell Viability Analysis
OECM-1 (6 × 10 4 ) and SG (6 × 10 4 ) cells were plated in 24-well culture plates and cultured for the indicated drug treatment. Ascorbate was first removed from the ascorbatetreated cells and then incubated with MTT solution for 1 h at 37 • C. Dimethyl sulfoxide (DMSO; 200 µL) was then added, and the absorbances at 570 nm and 650 nm were measured using an ELISA plate reader (Multiskan EX, Thermo, Waltham, MA, USA). The control group containing cells cultured in medium only was defined as 100% cell survival. Cell viability was calculated based on the absorbance ratio between the cells cultured with the selected drugs and the untreated controls, which were assigned a value of 100. The combination index of cisplatin plus specific drug in OECM-1 (2 × 10 4 ) and SG (1.7 × 10 4 ) cells. The combination index (CI) was calculated utilizing CalcuSyn (Biosoft, Cambridge, UK) to generate Isobolograms. Typically, a CI value of <1 denotes a synergistic combination effect and a CI value of >1 denotes an antagonistic combination effect [38].

Fluorescence-Activated Cell Sorting (FACS) for Flow Cytometry Analyses of Cell-Cycle Profiles, Proliferation, and ROS
The cell-cycle profiles were measured according to their cellular DNA content using FACS. Briefly, OECM-1 (3.5 × 10 5 ) and SG (3.5 × 10 5 ) cells were seeded in 6-well culture plates and treated with the selected drugs for 24 h before being harvested. The cells were fixed, permeabilized, and stained with 7-Aminoactinomycin (7-AAD, BD Biosciences, San Jose, CA, USA). The cell-cycle distribution was then evaluated using FACS, based on cellular DNA content. Cell proliferation was assessed using immunofluorescent staining with incorporated bromodeoxyuridine (BrdU) (BD Pharmingen™ BrdU Flow Kit) (BD Biosciences) and flow cytometry, according to the manufacturer's instructions. Briefly, OECM-1 (3.5 × 10 5 ) and SG (3.5 × 10 5 ) cells were seeded in 6-well culture plates and treated with the selected drugs for 24 h. After incubation, the cells were stained with BrdU, harvested, washed with PBS, and then fixed and permeabilized before being stained with BrdU fluorescent antibodies. The cells were resuspended in staining buffer and an FITC-BrdU fluorescence analysis was performed using a FACSCalibur flow cytometer and Cell Quest Pro software, version 6.1 (BD Biosciences). The intracellular ROS levels were determined using the fluorescent marker, DCFH-DA. Briefly, the cells were treated with the selected drugs for 24 h, stained with DCFH-DA (20 µM) for 40 min at 37 • C, and then harvested. Afterwards, the cells were washed once with PBS and then the DCFH-DA fluorescence intensity was analyzed on the FL-1 channel of the FACSCalibur flow cytometer using the Cell Quest Pro software, version 6.1 (BD Biosciences). The median fluorescence intensity of the vehicle was used as the starting point for M1 gating. Procedural details were described previously [45,46].
Mitochondrial superoxide production is an important source of reactive oxygen species in cells that may cause disease. MitoSOX TM Red (Invitrogen, Thermo Fisher Scientific, M36008) mitochondrial superoxide indicator is a fluorogenic dye for the highly selective detection of superoxide in the mitochondria of live cells. Once in the mitochondria, MitoSOX TM Red reagent is oxidized by the superoxide and exhibits red fluorescence. Briefly, OECM-1 (3.5 × 10 5 ) and SG (3.5 × 10 5 ) cells were seeded in 6-well culture plates and treated with indicated sodium ascorbate dosages for 4 h. After incubation, the cells were harvested and stained with 5 mM MitoSOX TM Red 37 • C for 20 min, and washed once with PBS, and resuspended in PBS. The MitoSOX TM Red fluorescence was then analyzed via flow cytometry (FACSCalibur, BD Biosciences); fluorescence intensity was analyzed on the FL-2 channel of the FACSCalibur flow cytometer using the Cell Quest Pro software, version 6.1 (BD Biosciences). The fluorescence intensity of the vehicle was used as the starting point for M1 gating.

Mitochondrial-Membrane Potential (MMP) Analysis
Mitochondrial depolarization was measured as a function of the decrease in the red/green fluorescence intensity ratio. OECM-1 (3.5 × 10 5 ) and SG (3.5 × 10 5 ) cells were cultured in 6-well plates and treated with the indicated concentrations of sodium ascorbate for 4 h. All dead and viable cells were harvested, washed with PBS, and incubated with 1× binding buffer containing the MMP-sensitive fluorescent dye JC-1 for 30 min at 37 • C in the dark. After washing the cells once with PBS, JC-1 fluorescence was analyzed on channels FL-1 and FL-2 of the FACSCalibur flow cytometer using Cell Quest Pro software, version 6.1 (BD Biosciences) to detect monomer (green fluorescence) and aggregate (red fluorescence) forms of the dye, respectively. The cell-volume gating strategy involved forward scatter height (FSC-H) and side scatter height (SSC-H), and the median fluorescence intensity of the vehicle was used as the starting point for M2 gating.

Reverse Transcription-Polymerase Chain Reaction (RT-PCR)
OECM-1 (3.5 × 10 5 ) and SG (3.5 × 10 5 ) cells were cultured in 6-well plates and treated with the indicated concentrations of L-ascorbic acid and sodium ascorbate for 4 h. OECM-1 and SG cells were lysed in TRIzol reagent (Invitrogen) to isolate total RNAs. Reverse transcription for first strand cDNA synthesis was conducted using MMLV reverse transcriptase (Epicentre Biotechnologies, Madison, WI, USA) with 1 µg of total RNA for 60 min at 37 • C. PCR reactions were operated on a Veriti Thermal Cycler (Applied Biosystems, Waltham, MA, USA). The mRNA bands were quantified through pixel density scanning and evaluated using Image J, version 1.44a. The PCR primers are listed in Table 1.

Statistical Analysis
Values are expressed as the mean ± SD of at least three independent experiments. All comparisons between groups were made using Student's t-tests. Statistical significance was set at p < 0.05.

Conclusions
Our study first revealed that the level of cell density was an important determinant for the cytotoxic effect of ascorbate on OECM-1 and SG cells. The differential effects on cell-cycle profiles and mitochondrial-membrane potential were verified in these two cell lines. Our findings further revealed that the cytotoxic effect might be mediated through the induction of mitochondrial ROS generation and the reduction in cytosolic ROS generation. The combination index supported the agonistic effect between sodium ascorbate and cisplatin in OECM-1 cells, but not in SG cells. In summary, our current findings provide supporting evidence for ascorbate to serve as a sensitizer for platinum-based treatment of OSCC.