The human gingival epithelial carcinoma cell line Ca9-22 was purchased from RIKEN BioResource Research Center (Tsukuba, Japan). The culture of the Ca9-22 was in RPMI-1640 supplemented with 5% fetal bovine serum (FBS). We also included in this study human polyclonal oral epithelial cell line (GMSM-K) provided by Dr. Grenier (Université Laval). The GMSM-K cell line was constructed by Gilchrist et al. (2000) who transfected oral epithelial cells with the shuttle vector plasmid, pZ189, containing the T Antigen Coding Region and Replication Origin from the simian virus 40 (SV40). This cell line was grown in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 5% FBS and 1% penicillin-streptomycin solution (Sigma Aldrich, St. Louis, MO, USA) both cell lines were maintained in culture at 37°C in a 5% CO₂ atmosphere.

Rapamycin was purchased from MedChemExpress LLC (NJ 08540, USA). MTT, protease and phosphatase inhibitors were from Sigma-Aldrich (Oakville, Ontario, Canada). LDH-Cytotoxicity Colorimetric Assay Kit II was purchased from BioVision (Milpitas, California, USA), while Annexin V-FITC/PI Kit was from BD Bioscience. (Mississauga, ON, Canada). Autophagy Assay, Red (Cat. #9156), Intracellular Total ROS Activity Assay (Cat. #9144) and Intracellular GSH Assay (Cat. #9137) were purchased from ImmunoChemistry Technologies (Davis, CA, USA). Autophagy Inhibitor, 3-MA (Cat. #189490) and N-acetylcysteine (NAC) were from Sigma. ECL system was acquired from EMD Millipore (Billerica, MA, USA). The primary antibodies as procaspase 3 (sc-56046), procaspase 9 (sc-17784), NF-κB (sc-8008) and β-catenin (sc-59737) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA), E-cadherin (8834), pERK1/2 (4370), ERK1/2 (4695), pp38 (4631), p38 (9212), cleaved caspase-3 (9664S), cleaved caspase-9 (20750S) were from Cell Signaling Technology (Danvers, MA, USA), LC3B (2775) and p62 (39,749) were all from Cell Signaling Technology (Danvers, MA, USA)and β-actin (A5441) was from Sigma-Aldrich (Oakville, ON, Canada). The secondary goat anti-mouse (554002) and anti-rabbit (554021) were from BD Pharmingen (Mississauga, ON, Canada). VersaDoc™ MP 5000 system was from Bio-Rad (Mississauga, ON, Canada).

Cell proliferation was evaluated using MTT assay as well as nucleus staining. Briefly, Ca9-22 and GMSM-K cells were seeded into 12-well plates at the density of 3 x 10⁵ cells/well, cultured overnight and then exposed to different concentrations of rapamycin (from 0.1 to 100 µM) for 24 h. After incubation, the culture medium was replaced with a new one containing MTT solution of 5 mg/ml in PBS for 3 h at 37°C in the dark, as described by Semlali and al. previously (13, 14). The cells were incubated for 15 min in 1 ml HCl 0.05 N-isopropanol solution to lyse the cells and release the formed formazan. The solution was transferred to 96-well microplate with 200 µl per well, and the absorbance was measured at 550 nm by an iMark reader (Bio-Rad). Percentage of viable proliferating cells was determined by using the following formula: % of cell viability = [(OD_550 nm (treated cell) − OD (blank)/(OD (control cell) − OD (blank))] × 100. The IC50 of rapamycin was obtained by plotting the percentage inhibition of cell proliferation against rapamycin concentration. The experiment was repeated eight times.

For nucleus staining, oral cancer cells (Ca9-22) at 10⁵ were seeded into sterile glass slides immersed in RPMI-1640 medium supplemented or not with different concentrations of rapamycin (0, 0.1, 1, 10, or 20 and 100 μM). The cells were cultured for 24 h at 37°C in a 5% CO₂ incubator. After 24h the cells were fixed with 4% paraformaldehyde for 60 min at room temperature before Hoechst staining (10 μg/ml) for 15 min. Finally, the slides were washed three times with the phosphate buffered saline (PBS), observed under an epifluorescence microscope (Nikon Optiphot) and photographed with a digital camera (Nikon COOLPIX 995). The experiment was repeated three times.

Cellular toxicity was determined by the LDH-Cytotoxicity Detection Kit from BioVision, which allows to directly quantify cell death in culture, based on the measurement of lactate dehydrogenase (LDH) released into growth media (15). Briefly, 3.10⁵ cells per well were to be seeded into six-well plates and incubated for 24h, before being exposed to different concentrations of rapamycin for an extra 24 h. Afterward, 50 μl of each supernatant was transferred in triplicates to a 96-well plate and supplemented with 50 μl reconstituted substrate mix. Then, the plates were incubated for 30 min at room temperature in the dark until the yellow color developed, before reading at 490 nm with a xMark microplate absorbance spectrophotometer (Bio-Rad, Mississauga, ON, Canada). Triton X-100 (1%) was used as a positive control for LDH and the negative one was obtained with untreated cells. LDH release was calculated using the following formula: % of LDH activity = [rapamycin (absorbance) − negative control (absorbance)] × 100)/[positive control (absorbance) − negative control (absorbance)]. The experiment was repeated four times.

Ca9-22 cells were seeded into 6-well plates at the density of 2000 cells per well for 24 h. They were then stimulated with different concentrations of rapamycin, ranging from 0 to 100 μM. The cells were incubated for two weeks at 37°C in 5% CO_2. The culture medium was changed every 2 or 3 days. The colonies were fixed with 100% ethanol and then stained with 0.5% crystal violet solution, as described by our previous works (13, 14). The colonies were subsequently washed twice with deionized water, dried at room temperature, observed under an optical microscope, and finally photographed. The experiment was repeated three times.

The cells were cultured and stimulated with different concentrations of rapamycin for 24 h at 37°C and 5% CO₂. They were then detached with a solution of 0.05% trypsin and 0.01% EDTA, incubated with Annexin V-FITC and propidium iodide at room temperature for 30 min in the dark. Finally, cells were resuspended in 300 µl of the phosphate buffered saline (PBS), to perform a flow cytometry, using either BD LSR II or BD FACSCanto II cytometer (BD Bioscience) equipped with FACSDiva Software v. 6.1.3. The experiment was repeated four times.

Cell migration assay was performed as described previously by Semlali and al. (13, 14). Ca9-22 cells were seeded into 6-well plates and cultured until they reached 100% confluence. Cell monolayers were subjected to a scratch in the shape of a cross with a sterile pipette tip. The cells were then stimulated with different concentrations of rapamycin, ranging from 0 to 100 μM and incubated at 37°C in a humid atmosphere containing 5% CO₂. Photographs were taken of each well with an inverted microscope after 0, 6, 12 and 24 h after the scratch was made. The cell migration was analyzed by image processing software that was able to measure the distance between opposite edges of the scratch at each time point. Each well was then compared, based on their percentage of closure. The experiment was repeated three times.

To evaluate the effect of rapamycin on autophagy in Ca9-22 cells, we used flow cytometry analyses as described previously (13, 14). Briefly, Ca9-22 cells were seeded into 60 mm Petri dishes for adhesion overnight. Afterward, cells were treated with controls of vehicle alone (0.2% of DMSO) or with 10 and 20 μM of rapamycin for 24 h. Following the rapamycin treatment, cells were resuspended in 500 μl of culture medium containing 1/5 Red staining solution. Cells were incubated for 60 minutes at 37°C in the dark and then collected by low centrifugation. The cell pellet was washed with 500 μl of the 1X Assay Buffer three times and suspended in 500 μl fresh 1X Assay Buffer before analyzing it with the green (FL1) channel of a flow cytometer using BD LSR II or BD FACSCanto II system (BD Bioscience) equipped with FACSDiva Software v. 6.1.3. The experiment was repeated four times.

Oxidative stress was assessed by flow cytometry using ROS marker protocol from ImmunoChemistry Technologies. After stimulation with rapamycin at (0, 10 and 20 μM) for 24 h, cells were detached with trypsin, washed with PBS, and were then resuspended in 490 μl of culture medium supplemented with 10 μl Green ROS stain solution and incubated in the dark for 1 h at 37°C. Fluorescent intensity of labeled cells was analyzed by flow cytometry at 488 nm using BD LSR II or BD FACSCanto II cytometer (BD Bioscience). The percentage of positive results was calculated in living cells with FACSDiva Software v. 6.1.3. This experiment was repeated four times.

Generation of mitochondria-mediated ROS was evaluated by using the MitoSOX-Red Mitochondrial Superoxide Indicator (Invitrogen). Firstly, Ca9-22 cells were treated with different concentrations of rapamycin (10 and 20 µM) for 24h. Subsequently, cells were harvested, washed twice in PBS and incubated with 5 mmol/l of mitochondrial dye (MitoSOX Red; Molecular Probes, Invitrogen) for 30 min at 37°C in the dark, followed by analysis on a flow cytometer to calculate the percentage of MitoSox-positive cells. This experiment was repeated four times.

As described by our previous studies (13, 14, 16), to evaluate the effect of rapamycin on damage to oral cancer cells, a H2A.X flow cytometry was performed. In addition, after treating Ca9-22 cells with the studied concentrations of rapamycin, they were trypsinized and then fixed with 75% ethanol for 15 min. The centrifugation of the samples was carried out to eliminate the fixative solution. Afterward, a permeabilization solution containing 1% BSA/0.2% Triton/1X PBS was added to the cells and they were then incubated in the dark at 4°C overnight with the first phospho-histone H2A.X (Ser139) monoclonal antibody from Santa Cruz Biotechnology at a dilution 1/100, washed twice with PBS and incubated with the secondary antibody conjugated to Alexa Fluor 488 from Santa Cruz Biotechnology in a 1:100 ratio for 1h before analyzing them with the BD flow cytometry system (BD FACS Canto II) and the percentage of positive cells was calculated. This experiment was repeated three times.

Total RNA was extracted from treated and untreated cells by using the RNeasy Mini Kit from Qiagen (Toronto, Ontario, Canada). An amount of 1 μg total RNA was reverse-transcribed into a cDNA copy with the High-Capacity cDNA Reverse Transcription Kit from Applied Biosystems (Thermo Fisher Scientific, USA) according to the manufacturer’s instructions, as described by our previous works (17–19). The RNA concentration and purity were determined by using a Nanodrop 8000 spectrophotometer (Thermo Fisher Scientific, USA). The qPCR protocol was performed with a 7500 Real-Time PCR System (Applied Biosystems). The reaction volume for each sample consisted of 12.5 µl SYBR Green Master Mix 2X, 0.5 µl primer (Forward and Reverse) (See Table 1), 7 µl distilled water, and 5 µl cDNA. The conditions for the PCR hold were 95°C for 5 min, followed by 40 cycles at 95°C for 15 sec, 60°C for 30 minutes and 30 sec at 72°C. The results were then analyzed using the Livak method for relative expression. The experiment was repeated three times.

Ca9-22 cells at the density of 10⁶ were harvested for extraction by a lysis buffer. A Bradford protein assay was then conducted to determine the protein concentrations of each sample. An amount varying between 20 μg and 60 μg of proteins were separated by SDS-PAGE with 8–15% of acrylamide, electro transferred onto nitrocellulose membrane. The membrane was blocked in a 5% milk solution at room temperature for 1 h and then incubated overnight with the specific primary antibodies. The membrane was then left in the secondary antibody solution for 1 h before being rinsed four times with washing solution. Detection of proteins was carried out with an enhanced chemiluminescence (ECL) Western Blotting Kit according to the manufacturer’s instructions and revelation by Versa Doc™ MP 5000 system (Bio-Rad, Mississauga, ON, Canada).

The significant difference between experimental (treated) groups and controls (untreated) was evaluated by Student’s t-test in GraphPad Prism 7 Software. Error bar represented mean ± SEM. ^* P- value < 0.05 was statistically significant.

We first investigate the effect of rapamycin on the proliferation of human gingival epithelial carcinoma cells (Ca9-22) and human polyclonal oral epithelial cells (GMSM-K). As shown in Figure 1A, rapamycin inhibited the proliferation of Ca9-22 cells in a dose-dependent manner. The half-maximal inhibitory concentration (IC₅₀) value was around 15 μM of rapamycin (Figure 1A). In addition, only the high concentration of rapamycin (100 µM) affects the proliferation of normal gingival epithelial cells (GMSM-K). These results were confirmed with the nucleus staining assay. Figure 1B shows that the number of nuclei in cancer cells dramatically decreases in the presence of rapamycin, while no effect was seen with GMSM-K cells (Figure 1B). The anti-proliferation effect of rapamycin evaluated by MTT was confirmed by LDH assay. As shown in Figure 1C, rapamycin increased the LDH activity of Ca9-22 cells in a dose-dependent manner. The median inhibitory concentration was also calculated and was around 15 µM. For GMSM-K, rapamycin induced cell toxicity only for 100 µM; the same effect was observed by MTT assay for this cell type (Figure 1C). These results show that a high dose of rapamycin non-selectively inhibits the viability of human oral cancer and normal cells. However, when the cells were treated with lower concentrations of rapamycin (non-cytotoxic to normal oral cells), we witnessed a significant inhibitory effect on oral cancer cell proliferation. We next investigated whether rapamycin affected the cell morphology.

Differences in cell morphology were observed between rapamycin-treated oral cancer cells and controls (treated with DMSO) under light microscopy (Data not shown), the most dramatic morphology changes were seen with rapamycin-treated cells at low concentrations of rapamycin (10, 20 µM) and high concentration (100 µM) in Ca9-22 cells, but only at 100 µM for normal oral cells. These morphology changes are manifested by cell shrinkage, cells became round and lost their integrity to promote cancer progression as well as their number was significantly reduced, and there was an extensive detachment from the cell culture substratum compared with the vehicle. The morphology changes observed in Ca9-22 cells are probably characteristics of oral cancer cell death and apoptosis.

The mechanism by which rapamycin inhibits oral cancer cell proliferation is probably through the cell cycle. This was confirmed by studying the effect of rapamycin on cell-cycle inhibitors. As shown in Figure 1D, rapamycin at 20 µM (a concentration close to the IC₅₀ value) was inhibiting cell proliferation by induction of cell-cycle inhibitors such as p21, p15 and p27 and the repression of cyclin D1 expression. Therefore, rapamycin inhibits proliferation by blocking the cell cycle.

As shown in Figure 2, rapamycin inhibits colony formation in a dose-dependent manner compared with controls, especially in oral cancer cells. No effect was seen in normal cells for the low concentrations (≤ 20 µM), only at the high concentration of rapamycin (100 μM) (Figures 2A, B). These results obtained with the clonogenic assay were consistent with MTT and LDH assays. For the rest of the study, we used 10 and 20 μM as an average of the IC₅₀. With these two concentrations, we next investigated whether the inhibition of oral cancer cell growth/survival was accompanied by induction of their apoptosis through rapamycin treatment.

As shown in Figure 3A, the stimulation of cancer cells with 10 and 20 μM of rapamycin was causing an increase in cell death and a decrease in living cells. The percentage of the apoptosis cells in untreated cells was 13%, representing necrotic and early and late apoptosis cells. When Ca9-22 cells were exposed to 20 μM of rapamycin, the level of apoptotic cells increased to 41.8% (Figure 3). To further confirm the effect of rapamycin on cell apoptosis in more detail, we investigated the effects of 20 µM rapamycin on inactive and cleaved caspase-3 and -9. Figure 10 shows a decrease in the expression of procaspase 3 and 9 with an increasing concentration of rapamycin. Also, the cleaved forms of caspase-3 and -9 increased following Ca9-22 treatment with rapamycin. Overall, rapamycin promotes cancer cell death through caspases-3 and -9 signaling pathways.

It was clearly reported that, mTOR is one of the key autophagy inhibitors and rapamycin is one of the best characterized autophagy inducers, we investigated whether rapamycin-induced cell death occurred via induction of autophagy. As shown in Figure 4A, rapamycin increased the percentage of Ca9-22 cells undergoing autophagy. However, that at basal level was 0.9%. Rapamycin treatment for 24 h significantly increased the percentage of Ca9-22 cells autophagic death from 0.9% with the control to 3.1% with 10 μM, and 82.2% with 20 μM. These results were confirmed by the study of LC3 and p62 gene expression study Figure 4B. As shown in Figure 4C, rapamycin at 20 μM caused an increase in the expression of both genes LC3B-II and p62. Thus, rapamycin is a promotor of cancer cell autophagy. The autophagosome accumulation induced by rapamycin measured by flux cytometry is probably linked to an increase in autophagy due to increased autophagosome formation. In addition, Figure 7B shows that rapamycin at 20µM increases autophagy to 32.9% in ca9-22 compared to 5.9% in untreated cells. However, a pre-treatment with 50µM of 3-MA decreases the percentage of cell autophagic to 13.7%. This autophagy appears to be associated with oxidative stress. In addition, a pre-treatment with 10 mM of NAC also decreases the percentage of cell autophagic from 32.9% in cells treated with 20 µM of rapamycin to 7.6% when the cells were treated by NAC and rapamycin (Figure 7B).

As shown in Figure 5A there was a significant induction of ROS in Ca9-22 cells being treated with rapamycin compared to cells untreated. The percentage of ROS increased to 57.9% and 66.1% respectively when the cells were treated with 10 and 20 μM of rapamycin compared to 17.6% in controls (with the vehicle). The generation of mitochondrial ROS in rapamycin-treated Ca9-22 cells for 24h was also assessed by flow cytometry using MitoSOX staining. The percentage of MitoSOX-positive cells also increased with rapamycin concentration. Figure 5B shows an increase from 5.5% in untreated cells to 10.3% with 10 μM, and 25.1% with 20 μM of rapamycin. Overall, these results suggest that rapamycin-mediated oral cancer cell apoptosis by mitochondrial-derived ROS production, which occurred upstream of the mitochondrial apoptosis.

ROS generation has been recognized to induce stress-mediated cell death in a variety of cancer types. Our hypothesis was that rapamycin triggers cell death via ROS production. As shown in Figures 6 and 7, the toxicity rate significantly increased with rapamycin concentration. The inhibition of oxidative stress by 10 mM of N-acetylcysteine (NAC) dramatically reversed the effect of rapamycin. In addition, the percentage of toxicity cutoff in cells decreased from 73.08% ± 12.57% with 20 µM of rapamycin to 41.18% ± 1.18% with the same concentration but having a pretreatment with 10 mM of NAC for 60 min. Similar results were observed for autophagy inhibition, a pretreatment of Ca9-22 cells was carried out with 50 µM of 3-methyladenine (3-MA); this inhibitor decreased the above percentage from 73.08% ± 12.57% with 20 µM of rapamycin to 24.96% ± 4.51% when cells were pretreated with 3-methyladenine for 60 min and then stimulated by 20 µM of rapamycin (Figure 6). Inhibition of oxidative stress or autophagy also reversed rapamycin-induced apoptosis. As shown in Figure 7, the percentage of apoptotic cells increased with rapamycin concentrations, particularly with 20 µM (28.2%). This induction of apoptosis was inhibited up to 9.3% and 10% when cells were pretreated respectively with NAC and 3-MA.

Figure 8 shows the effect of rapamycin on the expression of g-H2AX by Ca9-22. The percentage of g-H2AX-positive cells increased significantly following exposure of oral cancer cells to rapamycin. Indeed, the percentage of g-H2AX-positive ranged from 11.6% in control cells to 75% with 10 μM, to 96% with 20 μM of rapamycin (Figure 8).

The effect of rapamycin on cell migration was determined by the scratch method. Non-stimulated cells were compared to Ca9-22 stimulated with concentrations of 1, 10, 20, and 100 μM of rapamycin. As shown in Figure 9. the wound-healing assay demonstrated that rapamycin inhibited the Ca9-22 cell migration and invasion. Non-stimulated cells migrated over the scratch made. However, migration of rapamycin-treated cells was inhibited in a dose-dependent manner.

To investigate what signaling pathways related to cancer progression were targeted by rapamycin in oral cancer cells, we analyzed β-catenin, NF-κB, MAP kinase (ERK1/2 and p38) and that of caspases. As shown in Figure 10, there is a decrease in the expression of β-catenin, a pathway involved in cell adhesion as well as for NF-κB a pathway implicated in inflammation. Also, rapamycin decreased the activation of ERK1/2 and p38 pathways but was not affecting the total ERK1/2 and p38. Unlike proliferation pathways, rapamycin is thus activating two key pathways involved in apoptosis, particularly cleaved caspase-3 and cleaved caspase-9 pathways, confirming the results presented in (Figure 10).