The curcumin analog (PAC) suppressed cell survival and induced apoptosis and autophagy in oral cancer cells

PAC (3,5-Bis (4-hydroxy-3-methoxybenzylidene)-N-methyl-4-piperidone), a novel bioactive curcumin analog, has been reported to have anticancer properties against various tumors. However, the anti-cancer effects of PAC on oral cavity squamous cell carcinoma were not studied yet. Our aim is to investigate the anti-oral cancer properties of PAC in vitro, and determine the molecular mechanisms underlying these effects. Viability assays including MTT and LDH were conducted to measure cell proliferation. Flow cytometry-based cytotoxicity assay was performed to detect autophagic cell death and oxidative stress markers. Western blotting was used for measuring protein expression/activation in apoptotic, autophagic and pro-carcinogenic cellular signaling pathways. We demonstrated that PAC preferentially and, in a dose, -dependent way kills oral cancer cells, but was not toxic to normal human gingival cells. PAC destabilizes cell-cycle distributions, inhibits the expression of oncogenes (cyclin D1) and that of cyclin-dependent kinase inhibitor (p21WAF1) is upregulated, increases the expression of p53 gene, and inhibits epithelial-mesenchymal transition markers in oral cancer cells. The PAC effect involve various signaling pathways including NF-κB, MAPK, Wnt, caspase-3/9 and PARP1. Finally, PAC demonstrated ability to induce autophagy, decrease production of reactive oxygen species, increase intracellular glutathione (GSH) activity, and reduce mitochondrial membrane potential in oral cancer cells. In conclusion, PAC inhibits the proliferation and increases the apoptosis and autophagy and oxidative stress of oral cancer cells. These effects involve ERK1/2, p38/JNK, NF-κB and Wnt cellular signaling pathways. Overall, our study suggests the potential use of PAC to treat oral cancer.

Cell culture and PAC stimulation. Ca9-22 cells and GEC were sub-cultured respectively in RPMI-1640, F12 or DMEM media supplemented with 5% (v/v) FBS for Ca9-22 and 10% for GEC, 100 units/mL penicillin, and 100 µg/mL streptomycin in T75 flasks 26,27 . All cells were maintained in a humidified atmosphere of 5% CO 2 at 37 °C. All experiments conducted with GEC were at passages 3 to 4, and GF at passages 4 to 5. PAC was used at (0, 1, 2.5, 5 and 10 μM in DMSO) with all experiments, except for Western blotting where we used PAC at 5 μM, only.
Cell viability assay. Cells were seeded at 3 × 10 5 cells/well in 6-well cell culture plates, and culture for 24 h. Culture were then stimulated exposed to vehicle control (DMSO) or various concentrations of PAC (0, 1, 2.5, 5 and 10 μM) for 24 h. At the end of the stimulation period, the culture medium of each condition was supplemented with MTT solution at 0.5 mg/mL then incubated for 3 h at 37 °C in the dark. The cells were washed, then 1 ml of 0.04 N HCl in isopropanol was added to the culture wells, followed by an extended 15-min incubation. Finally, 200 µl of the reaction mixture (in triplicate) was transferred to the wells of a 96-well flat-bottom plate and the absorbance was measured at 550 nm using iMark reader (Bio-Rad). The cell viability (percentage) was determined by using the following formula: % of cell viability = [(OD 550 nm (treated cell)/(OD (control cell)] × 100. This experiment was repeated 7 times. To support the cell viability assay we performed lactate dehydrogenase (LDH) activity measurement using culture supernatants. Supernatants were collected from each condition after exposure or not of the cells to PAC for 24 h. The LDH activity was measured by means of an LDH kit from (Sigma-Aldrich), as outlined in our previous work 28 . The experiment was repeated 4 independent times.
Colony formation assay. Ca9-22 cells were seeded into 6-well plates at 10 3 cells per well for 24 h, before being treated with various concentrations of PAC. The culture medium was changed every 2 or 3 days. After 2 weeks, the cell clones that survived were fixed by 100% ethanol for 10 min. After washing twice by PBS, cells were then stained with 0.5% crystal violet for 30 min and they were photographed with a digital camera.
Annexin V-FITC apoptosis assay with propidium iodide. Ca9-22 cells were seeded into T25 flasks the exposed to 1 µM, 2.5 µM, 5 µM or 10 µM of PAC, for 24 h and 48 h, at 37 °C. After exposure, cells were detached from the flasks using 0.05% trypsin and 0.01% EDTA, centrifuged, washed by PBS and the pellet was suspended in 300 µl phosphate-buffered saline (PBS) and incubated with 0.5 µl annexin V-FITC and 0.5 µl of propidium iodide at room temperature for 30 min in the dark. Stained cells were subjected to flow cytometry analyses, using a "LSRII" or "CantoII" cytometer instrument from BD Biosciences. Data analysis was made by FACS Diva software v. 6.1.3. The experiments were repeated 5 independent times.
Cell cycle distribution. Cell percentage depending on the various phases of the cell cycle was measured by using the fluorescent DNA dye 7-aminoactinomycin D (7-AAD) (Sigma, USA), as previously described 29 . Briefly, oral cancer cells were treated with different PAC concentrations (1 µM, 2.5 µM, 5 µM or 10 µM) for 24 h. They were then incubated in the presence of 7-AAD at 1 μg/mL for 30 min at 37 °C. The fluorescence intensities of DNA were detected with "LSRII" or "CantoII" (BD Biosciences). Data were analyzed by FACSDiva software v. 6 www.nature.com/scientificreports/ in the dark. After washing with PBS twice, fluorescence intensity was determined by Flow Cytometry System (BD Biosciences), and the percentage of GSH-positive cells was calculated. Experiments were done in triplicates.
Assessment of mitochondrial superoxide. To evaluate the mitochondrial superoxide production following treatments with PAC, we used an indicator named MitoSOX Red (Molecular Probes, Invitrogen), according to the manufacturer's instructions, and afterwards performed a flow cytometry analysis. PAC-treated cells (1µ M, 2.5 µM, 5 µM or 10 µM) were incubated for 30 min with 5 μM MitoSOX at 37 °C in 5% CO 2 . They were then suspended in 1 mL PBS, and analyses by means of flow cytometry with "LSRII" or "CantoII" (BD Biosciences) (n = 3).
Western blotting. The effect of PAC at 5 µM on various proteins involved in MAPK pathways, apoptosis, autophagy, and cell cycle was evaluated by making use of Western blots. Briefly, 10 6 cells (stimulated and unstimulated) were harvested for extraction by RIPA lysis buffer. Protein concentrations were determined by the Bradford assay. For each sample, a mixture of 20 μg to 60 μg was added to an equal amount of loading buffer containing 10% β-mercaptoethanol and migrated through a 7-15% acrylamide gel. After electrophoretic separation, proteins were transferred from the gel to a nitrocellulose membrane to be blocked in a 5% milk solution for 1 h at 24 °C and put overnight in an incubator with appropriate concentrations of antibodies. Finally, a band detection was performed by an ECL Imaging System (EMD, Millipore, Billerica, MA, USA), according to instructions from its manufacturer.
Statistical analysis. All our experiments were repeated more than three times and the statistical significance obtained between controls (untreated cells) and tests (treated) with various concentrations of PAC was expressed as mean ± SD. Differences between values were determined if statistically significant using a one-way ANOVA.
A p-value < 0.05 was defined as significant.

PAC selectively inhibits the growth of oral cancer cells, and increases apoptosis.
To evaluate the effect of PAC on normal and gingival cells affected by cancer to proliferate, we used the MTT assay. As presented in Fig. 1, PAC at various concentrations decrease the growth of oral cancer cells. This decrease was more important with higher concentration of PAC. Indeed, at 10 µM oral cancer cell growth was almost drastically reduced. Interesting, the decrease of normal human gingival epithelial cells growth following the exposure to 10 uM of PAC was too low compared to cancer cells. With 5 µM of PAC, there was an 80% growth reduction of oral cancer cells, but only 10% to 20% growth reduction of primary human gingival epithelial cells. These results demonstrated that normal human cells exhibit great resistance to PAC, in comparison to Ca9-22 cells. The effect of PAC on cancer cells was observable even at low concentration with an LC 50 is around 3 µM. Its proliferation level decreased, from 86.21% ± 4.03% (p = 0.007), 78.41% ± 8.23% (p = 0.02), 24.37% ± 7.31% (p = 2.39 × 10 −5 ) to 9.85% ± 2.16% (p = 6.46 × 10 −9 ) respectively with PAC concentrations of 1, 2.5, 5 or 10 μM (Fig. 1A). With GEC cells, a slight decrease was observed only at the highest concentration of PAC (10 μM) after 48 h (p = 0.02) and 72 h (p = 0.03), but any significant inhibition for all treatments (1 μM to 10 μM) was noticeable after 24 h. (Fig. 1B). These outcomes were confirmed using a cytotoxicity assay (Fig. 1C,D), a treatment with PAC was inducing dose-and time-dependent cell death in Ca9-22 ( Fig. 1C) but not in GEC cells (Fig. 1D), suggesting a selective toxic effect of PAC on cancer rather than normal gingival epithelial cells. We also demonstrated a significant effect of PAC reducing the capacity of oral cancer cells to form colony but not in GEC (Fig. 1E). These outcomes clearly indicate as well that PAC is having preferentially anti-proliferative activity and cytotoxicity against Ca9-22 cells, but not on normal human gingival epithelial cells. We demonstrated that anti-proliferative effect of PAC on oral cancer cells goes through a cell cycle modification. Indeed, Ca9-22 cells treated with 5 µM showed a cell-cycle arrest, by increasing the expression of cyclin-dependent kinase inhibitors (CDKIs) such as p21, p27, p16 or p53 and Rb (retinoblastoma protein) which are tumor suppressor genes known to be inactive in most cancer types, while it inhibits oncogenes such as cyclin D1 or c-Myc. In addition, PAC can to reduce DNA damage on oral cancer cells by inhibition of H2A.X protein expression ( Fig. 2A). The effect of PAC on cancer cells is involving specific inhibition of cell proliferation signaling pathways. Indeed, we demonstrated a significant decrease of ERK1/2 phosphorylation, p38, STAT and AKT phosphorylation. Also, PAC at 5 µM was decreasing the expression of NF-kB and Wnt/β-catenin signaling pathways (Fig. 2B).

Effect of treatments with PAC on cell cycle distribution in oral cells. Cell cycle distribution, on
Ca9-22 PAC-treated cells, was evaluated by flow cytometry using the fluorescent DNA dye 7-aminoactinomycin D (7-AAD). Briefly, these were under treatment with different PAC concentrations for 24 h. Then, they were responding to PBS buffer before being incubated with 7-AAD Fig. 3  www.nature.com/scientificreports/ PAC. In contrary, the S population increases from 6.85% ± 3.74% in controls to 38.95% ± 17.18%, p < 0.005 using the same concentration it. The G2/M population is also doing it as the previous one, from 5.05% ± 1.34% to 48.7% ± 13.85% p < 0.005 with the same treatment (Fig. 3A,B).

PAC induces apoptosis in oral cancer cells by targeting the intrinsic mitochondrial pathway.
In determining the type of death induced by various concentrations of PAC on oral cancer cells, the fluorochrome-labeled annexin V with PI was used. Our results show the presence of 4 different sub-populations referring to live cells (annexin V-/PI-), early apoptotic cells (annexin V + /PI−), late apoptotic cells (annexin V + / PI +) and necrotic cells (annexin V−/PI +). After 24 h of exposure to PAC, the percentage for the first population decreases from 96.48% for untreated to 71.29% with 5 µM, and 18.32% with 10 µM. Conversely, the early apoptotic cells were increasing from 2.56% with the control to 18.62 with 5 µM. In late apoptotic population was much higher with PAC exposure ranging from, 1.09% with the controls to 55.46% with 10 µM of PAC. The necrotic population was also increasing from 0.87% with the untreated cells to 24.48% with 10 µM PAC treated oral cancer cells (Fig. 4A). To confirm apoptosis-inducing ability of PAC concentration on oral cancer cells, we  (Fig. 4B).
The effect of PAC of caspase signaling for oral cancer cell apoptosis was confirmed by Western blotting (Fig. 4C) showing that PAC at 5 µM strongly decreases Bcl-2 expression and increases that of pro-apoptotic Bax and cytochrome C as well as allows cleavage of PARP-1. The effect of PAC promoting cancer cell death involves caspase 3/9 (Fig. 4C).

PAC promotes autophagy in oral cancer cells by targeting LC3B and p62 protein expression.
To determine the different cellular mechanisms by which PAC exerts its anti-oral cancer function, we investigated the effect of PAC on cell autophagy. As shown in Fig. 5A,B, PAC at 5 μM strongly leads to autophagy in cancer cells. The percentage of positive cells detected by the Autophagy Assay (Red) was increasing from 0.1125% ± 0.15% in controls to 10.57% ± 3.35% (p = 0.00066) and 28.59 ± 13.78% (p = 0.0030) respectively for 5 µM or 10 µM of PAC. These observations were confirmed by Western Blot showing an overexpression (high protein levels) of two autophagy markers (LC3B and p62) in Ca9-22 cells (Fig. 5C). Fig. 6A, we demonstrated that PAC at 5 µM inhibits the EMT in oral cancer cells, as shown by the increase of E-cadherin, while vimentin expression was decreasing. Metastatic cancer is also closely associated with cell migration. In this sense, we investigated the effect of various PAC concentrations on Ca9-22 cells migrating under treatment by using the Wound Healing (Scratch) Assay. As shown Fig. 6B, PAC at various concentrations was able to decrease the migration of oral cancer cells. Indeed, after 6 h of treatments with PAC, 100% of the scratch area was covered due to cell migration, in the controls, while less than 47% and 33% of the scratched area being covered following exposure to 5 µM or 10 µM, respectively.

ROS production during treatments with PAC in oral cancer cells.
Our data shows that the intracellular ROS accumulation induced by the treatment for 24 h with PAC was evaluated using a flow cytometer (Fig. 7). The percentage of ROS ( +) is low with increased concentration of PAC as treatment. In addition, the percentage of positive ROS staining in untreated cells was 81.8% ± 10.6% decreases to 64.3% ± 4.13%, p = 0.02 and to 35.8% ± 8.8%, p = 0.002 respectively when the cells were treated by 5 and 10 µM of PAC. However, any difference was observec when the cells were treated at low concentration of PAC (1 and 2.5 µM). We, therefore, suspect that PAC acts as a powerful antioxidant by producing intracellular glutathione (GSH).  The effect of PAC of caspase signaling for oral cancer cell apoptosis was confirmed by Western blotting. PAC at 5 µM strongly decreases Bcl-2 expression and increases that of pro-apoptotic Bax and cytochrome C as well as allows cleavage of PARP-1. The effect of PAC promoting cancer cell death involves caspase-3/9 (n = 3). All Blots for each apoptosis protein derive from the same experiment and were exposed at same time of phosphor imager exposition. www.nature.com/scientificreports/ GSH generation during treatments with PAC in oral cancer cells. Figure 8 shows that PAC strongly increase the GSH expression with 2.5 µM (55.8% ± 4.6%, p = 0.01compared to controls (26.1% ± 5.6%). The percentage of GSH positive is going up, from 26.1% with non-treated cells to 70.3% ± 1.2%, p = 0.004 and 83.8% ± 5.4%, p = 0.004), respectively after cell exposure to 5 µM or 10 µM of PAC, for 24 h. PAC is then confirmed to play a crucial role in producing GSH, an antioxidant.
Mitochondrial membrane potential is reduced during treatments with PAC in oral cancer cells. The MMP expression that is reduced by PAC was measured by the MitoProbe DiOC 2 (3) Assay Kit for Flow Cytometry. As shown in Fig. 9, there is a concentration-dependent effect of PAC on MMP intensity in Ca9-22 cells. Indeed, the MMP intensity decreases from 71.5% ± 4.3% for controls to 4.3% ± 9.4%, p = 0.01 and 9.6% ± 8.5%, p = 0.003 for Ca9-22 treated cells with 5 µM or 10 µM respectively for 24 h. In addition, any significant difference was observed when the cells were treated by PAC at 1 and 2.5 µM (Fig. 9). in untreated cells to 17.57% ± 7.3%, p = 0.01 and to 42.75% ± 11.5%, p = 0.003 respectively when the cells were treated by 5 and 10 µM of PAC concentration (Fig. 10). The key role of PAC inducing mitochondrial superoxide production was clarified.

Discussion
Conventional oral cancer treatments rely on surgery, along with radiation therapy (external beam radiotherapy and/or brachytherapy), as well as coadjutant therapy (chemotherapy with agents like cisplatin), it still involves high economic costs and highly damaging options 30,31 . To reduce this burden, plant products as complementary and alternative therapies could become solutions. Some recent discoveries demonstrated the superiority of the plant world to prevent cancer 32,33 . Several research studies over the last century have revealed several important functions of curcuminoids. And it has been shown that curcumin has a capacity to influence multiple natural targets and seemed to have activity against different diseases 12,15 . Although it was evaluated with success for www.nature.com/scientificreports/ widespread biological activities, those of concern are its low bioavailability and rapid metabolization encouraged researchers to look for new synthetic analogs 23,34 . In a search for novel compounds active against cancer, PAC was found out. This curcumin analog demonstrated that it has a great potential as an anti-breast cancer agent 21,24 . Until today, there had been no available study investigating the properties of PAC to be a possible complementary treatment for oral cancer. Our goal was to examine its inhibitory effects, on cells that proliferate or are in apoptosis, as well as the mechanisms of action involved as an interesting avenue to develop treatments for oral cancer and study its potential to be an active supplement ingredient in some drugs and products for mouth cleaning. Our data shows that PAC is exhibiting preferentially ability to kill oral cancer cells. Our data are supportive to those reported by Al-Howail et al. (2016) and Al-Hujaily et al. (2011) showing that PAC has similar effect against breast cancer 21,24 . PAC decrease the cell proliferation through the deregulation of the cell cycle. Indeed, PAC treated cells showed high expression of CDKIs including p21, p27, p16 or p53 and Rb. It is well known that these genes suppress tumors, known for being inactive in most types of cancer and also for inhibiting genes with oncogenic functions like cyclin D1 or c-Myc. These molecules have a history of playing a major role in the cellcycle arrest at G2/M, and ultimately induce inhibition of oral cancer cell proliferation. This is mediated by inhibiting several signaling pathways involved in cancer such as MAPK, Wnt and NF-κB. The same results were reported by Al-Hujaily et al. (2011) and Al-Howail et al. (2016) reporting that PAC could be considered as a powerful nontoxic new chemotherapeutic agent against ER-negative tumors by triggering breast cancer cell apoptosis more than curcumin 21,24 , downregulating ERα, c-Myc, cyclin D1 24 and inducing cell-cycle arrest at G2/M of this cancer cells 21 . Also, our results demonstrate that PAC strongly induces oral cancer cell apoptosis mediated via the mitochondrial pathway by increasing the Bax/Bcl-2 protein expression ratios, activating caspase-3/9 and cleaving PARP-1. We conclude that the PAC anti-oral cancer properties were controlled by impairing pro-survival signaling pathways and activating apoptosis pathways such as ERK1/2 activation, but not MEK1/2, AKT activation, NF-κB 35,36,37 and β-catenin 38 , known to be implicated in tumor progression and as well for their resistance to chemo-and radiotherapy in several cancer types. Indeed, it is clearly reported that several oncoproteins such as cyclin D1 and p21 are strongly destabilized by inhibition of ERK1/2 activation. Wnt/βcatenin is the key oncogenic signaling pathway found to be highly active in various types of cancer, including oral cancer 38,39 . Similar observation was reported on breast cancer cells being treated with PAC 21,24 . Therefore, the strong PAC-dependent apoptosis could be of great value for inhibiting of oral cancer cell proliferation and inducing their death. In a recent study carried out by Al-Qasem et al. (2016), they have published the same effect of PAC on colon cancer cells. They concluded that PAC exhibits anti-colon cancer properties that have potential, by targeting cyclin D1 and suppressing JAK2/STAT3, AKT/mTOR and MEK/ERK signaling pathways as well as their common downstream effector which is cyclin D1 25 . We suggest that PAC suppresses the activation of MAPK and the induction of caspase cleavage, leading to the downregulation of several MAPK targets including cyclin D1, p53, and caspases, as Bax, Bcl-2 and cytochrome C. All these molecules are playing key roles in oral . All Blots derive from the same experiment and were exposed at same time of phosphor imager exposition.  [40][41][42] . In other hand, we demonstrated that PAC promoted oral cancer cell autophagy and reduced oxidative stress via ROS production. Apoptosis and autophagy are two important cellular processes play major roles in determining cellular fate 43 particulary in cancer development.. Cancer is defined as the balance between cell proliferation and apoptosis. Developing a new drug that activates apoptosis in cells could be of great value to target the screening of therapeutic agents against cancer, specifically when we are focusing to find out compounds able to affect mitochondria, which play an important role in controlling this cellular suicide. The reduction of the ROS intracellular by PAC treatment can be explained by the fact that, cancer cells generate high levels of reactive oxygen species (ROS), as an unavoidable consequence of metabolism. These are considered as potentially harmful and usually linked to cancer progression. In our study, we demonstrated that PAC decrease intracellular ROS production and activate GSH, an antioxidant. We suggested that hyper proliferation of Ca9-22 cells is accompanied by a high rate of ROS production, but when treated with 10 μM of PAC, they adapted under similar normal conditions; they achieve this by increasing their antioxidant status in order to optimize ROSdriven proliferation, as reported by Dodson et al. (2019) 44 . Most publications indicate that an increasing GSH synthesis leads to improving NADPH production and trans-repression of pro-oxidant genes like NOS2 and COX2 45 . These results are in disagreement with several current research studies focusing on the use of anticancer treatments having an effect while apoptosis in cancer cells is induced by increased production of reactive oxygen species (ROS). However, our findings are also in concordance with others recent studies reporting that increase in cellular O2-and H2O2 by NADPH oxidase 46 are leading the stimulation of cancer cell proliferation by ROSsensitive AKT/ERK signaling pathways and the reduction of cancer cell growth by blocking this enzyme 47 ; Dikalova et al. (2010) by using mitoTEMPO, they significantly reduced ROS in cells and inhibited NADPH oxidase activity 46 . As for Nazarewicz et al. (2013), they also reported that mitochondrial ROS scavenging by mitoTEMPO can block ROS-sensitive signaling pathways in cancer cells, cancer metabolism and induce cell death 48 . Our data suggests that diminishing ROS production, by PAC-treated oral cancer cells, is altering cell signaling mediated by ROS-sensitive AKT, ERK1/2, and p38 as well as inducing cell death. The PAC-attenuation effect from these signaling pathways was the principal factor leading to dramatic changes in oral cancer cell   www.nature.com/scientificreports/ metabolism. As well, its anticancer activity toward mitochondria-targeted antioxidants (GSH) is closely associated with inhibition of ROS-sensitive AKT/ERK survival signaling pathways and glycolysis in human oral cancer cells. GSH is our body's antioxidant powerhouse that is protecting our cells from free radical damage. It has been well reported that antioxidants have beneficial health effects at physiological concentrations 49 . However, in normal situations, the basal level of ROS is crucially good to cell viability. It may prevent oxidative stress by maintaining the balance in ROS homeostasis, especially in cancer cells known for having much faster metabolism and higher ROS levels compared to ourcontrols 49 . Inhibition of mitochondrial superoxide may represent a novel, specific anticancer treatment strategy reducing side effects that are cytotoxic. Indeed, our work as well shows that PAC reduced mitochondrial ROS and decreased MMP. Oxidative stress is also known to induce disruption of the latter 50 MMP is occurring to play a key role in mitochondrial homeostasis through selective elimination of those that are dysfunctional 51 . But MMP was considered as the central bioenergetic parameter for transporting ions and proteins that are indispensable for the control of normal mitochondrial function and are correlating with cell differentiation, tumorigenesis and being malignant 52 . Many other studies reported that carcinomaderived cells possess a higher mitochondrial membrane potential than normal ones 53,54 In addition, Schieke et al. 55 published that higher MMP is linked with a tumorigenic property 55 . After inhibition of MMP by rapamycin treatment on stem cells having a high protonmotive force, its tumorigenicity decreased drastically. Moreover, our findings indicate that PAC induces intracellular ROS, generates mitochondrial superoxide and diminishes MMP leading to the release of cytochrome C in mitochondria and to the enhancement of oxidative stress in oral cancer cells.

Conclusion
The present study examined the antioral cancer and preferential killing effects of PAC, new analog of curcumine. Our results demonstrate the PAC preferential killing against oral cancer cells without effects on normal oral cells involving the regulation of proliferation, apoptosis, autophagy and oxidative stress. Therefore, PAC inhibits intracellular ROS, increased MitoSOX, diminished MMP, and PAC plays a crucial role in producing GSH, an antioxidant. With its preferential killing ability, PAC may provide a potential for improving oral cancer therapy without a cytotoxic side effect to normal oral cells (Fig. 11). The future studies will be to combine cisplatin with this new curcumin analog (PAC), to normalize cisplatin side effects and potential drug resistance on oral cancer.