OSU-2S (Figure 1A) was synthesized in Dr. Chen’s lab at The Ohio State University as previously described (Omar et al., 2011). The identity and purity of OSU-2S were verified by mass spectrometry analysis and HPLC, respectively. Sorafenib (BAY 43-9006) (Figure 1A) was purchased from BioVision^® (Milpitas, CA, USA). OSU-2S and sorafenib were dissolved in DMSO and diluted in culture medium. Fetal bovine serum and MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide] were purchased from (Sigma-Aldrich, St. Louis, MO, USA). The enhanced chemiluminescence system, Matrigel and 24-well modified Boyden chambers (8 μm pore size) were obtained from GE Healthcare Bioscience (Piscataway, NJ, USA), BD Biosciences (Bedford, MA, USA) and Corning Costar (Cambridge, MA, USA), respectively. Antibodies against various biomarkers were obtained from the following sources: PKCδ, ERKs, pERKs, from cell Signaling Technologies (Beverly, MA, USA); Poly(ADP-ribose) polymerase from Pharmingen (San Diego, CA, USA); β-actin from Sigma-Aldrich (St. Louis, MO, USA); Caspase 3 and p53 from Novus Biologicals (Littleton, CO, USA). Mammalian PKCδ shRNA expression plasmid (pKD-PKCδ-v2) and random shRNA (pKD-NegCon-v1) were purchased from Upstate (Temecula, CA, USA). Mammalian p53 shRNA expression plasmid, shp53 pLKO.1 puro was a gift from Bob Weinberg (Addgene plasmid # 19119), pLKO.1 – TRC control non-silencing plasmid was a gift from David Root (Addgene plasmid # 10879), GFP-p53 was a gift from Tyler Jacks (Addgene plasmid # 12091) and the empty vector, pEGFP-N1-FLAG was a gift from Patrick Calsou (Addgene plasmid # 60360). Other chemicals and reagents were obtained from Sigma-Aldrich unless otherwise mentioned.

The HCC cells were cultured in Dulbecco’s Modified Eagle’s medium (DMEM, Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% fetal bovine serum, 1.5 g/L sodium bicarbonate and 1% penicillin/streptomycin. PLC5 and HepG2 cells were obtained from the American Type Culture Collection (Manassas, VA, USA) while Hep3B and Huh7 cells were purchased from Sigma-Aldrich (St. Louis, MO, USA). Cell lines were maintained at 37°C in a humidified incubator containing 5% CO₂.

The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bro-mide (MTT) assay was used for cell viability analysis as described before (Hung et al., 2015). In summary, HCC cells were seeded in 10% FBS supplemented DMEM at 1 × 10⁴ cells per well density in 96-well flat-bottomed plates. The cells were treated with different concentrations of OSU-2S, sorafenib or their combination after 24 h of cell seeding. An equivalent volume of the used vehicle (DMSO) was used for the control treatment. After 48 h of treatment, the media were removed by aspiration and replaced by 200 μL fresh media containing 0.5 mg/mL of MTT then incubated in the CO₂ incubator at 37°C for 2 h. At the end of the experiment, the supernatants were removed and the formed formazn crystals were dissolved in 200 μL/well DMSO. The intensity of the formed violet color was measured at 570 nm using a plate reader. Following plate reading, the data were analyzed by CalcuSyn software package version 2.1 (Biosoft, Cambridge, UK), which is based on the median effect equation to calculate the combination index (CI) of different treatments. The cell viability was expressed as percent cell vialbilty relative to the vehicle-treated control group.

Lysates of OSU-2S-treated HCC cells at the indicated concentrations for 48 h were prepared for Western blotting of PARP, PKCδ, ERK1/2, pERK1/2, caspase 3, p53, and β-actin. Western blot analysis was performed as formerly reported (Arafa el et al., 2014).

Caspase-3/7 activities in HCC cells treated with OSU-2S, sorafenib, or their combination were measured using Caspase-Glo 3/7 luminescence assay kit according to the manufacturer’s directions (Promega, Madison, WI, USA). The vehicle (DMSO) was used as negative control. In a brief, cells were seeded at 1 × 10⁴ (100 μl/well) into clear bottom, opaque wall 96-well tissue culture plates and incubated for 24 h. Cells were treated for 24 h and caspase-3/7 activities were assayed a plate luminometer.

The assay with performed essentially as detailed before (Omar et al., 2013) with minor modifications. Hep3B cells were trypsinized and suspended in 0.5 ml of serum-free medium containing different concentrations of OSU-2S, sorafenib or their combination. The cell suspensions were seeded onto the membranes of the upper chambers of modified Boyden chambers (8 μm; Corning Costar, Cambridge, MA, USA) which were pre-coated with Matrigel. The lower chambers contained the same concentrations of the used agents in 10% FBS-containing medium. The cells were then incubated at 37°C for 24 h. After incubation, the cells remaining on the upper surface of the membranes were removed gently with cotton swabs. Cells which invaded into the lower surface of the membrane were fixed in 90% methanol and stained with 0.1% crystal violet. Stained cells were counted in at least ten 200x fields.

For the measurement of the ability of test compounds to affect cancer cell migration, the Modified Boyden chambers were used as mentioned before (Omar et al., 2009). Briefly, Hep3B in 0.5 ml of serum-free DMEM containing different concentrations of the used agents were seeded into the upper chamber membranes. The cells were incubated at 37°C for 60 min, then transferred to new wells containing the same concentrations of the used agents in 10% FBS-supplemented DMEM, and then incubated for 8 h. Non-migrated cells on the upper surface of each membrane were swabbed gently, while migrated cells into the lower side of the membrane were fixed, stained and counted as mentioned above.

Hepatocellular carcinoma cells were transfected with shRNA plasmids for the knockdown of PKCδ, p53 or control vector using Lipofectamine 2000 (Life Technologies) according to the manufacturer protocol. Transfected cells with pKD-PKCδ-v2 were further subjected to stable clone isolation by 500 μg/ml geneticin (Invitrogen, Carlsbad, CA, USA) and antibiotic-resistant colonies were isolated after 2–3 weeks. The knockdown of the corresponding protein was confirmed by immunoblotting.

Hep3B cells were transfected with 1 μg/ml of plasmids DNA encoding GFP-p53 or the empty vector, pEGFP-N1-FLAG using Lipofectamine 2000 according to the manufacturer’s instructions and as mentioned before (Cai and Liu, 2008). The expression of p53 was confirmed by both Western blotting and fluorescence microscopy.

The analysis of statistical significance between different treatments was performed using one-way ANOVA followed by the Neuman–Keuls test for multiple comparisons. Differences were considered significant at P < 0.05. Statistical analysis was performed using SPSS for Windows (SPSS, Inc., Chicago, IL, USA).

The ability of OSU-2S and sorafenib as a single agent to inhibit the cell viability of HCC cell lines has been reported before (Ng and Chen, 2006; Omar et al., 2011). In order to select a suitable range of drug concentrations for combination experiments, the effect of both sorafenib and OSU-2S on the cell viability of four different HCC cell lines was initially investigated using MTT assay. The dose response curve of sorafenib or OSU-2S was assessed relative to vehicle control treatment (Figures 1B,C). The half maximum inhibitory concentration (IC₅₀) for OSU-2S was in the range of 1.8–3.9 μM with the HepG2 cells being the most sensitive and the PLC-5 cells being the most resistant. For sorafenib, the IC₅₀ was in the range of 6.2–10.4 μM with HepG2 cells being the most sensitive and Hep3B cells being the most resistant (Figure 1D). Sorafenib-mediated anti-proliferative effect was significantly enhanced upon the combination with OSU-2S. Combination indexes (CI) were calculated using Calcusyn software for each dose combination. Values of CI <1 indicate synergy, =1 indicate additive effect and >1 indicate antagonism. Synergistic effects were observed in the four used HCC cell lines with different degrees. For example, in HepG2, Hep3B, and PLC-5 cells, almost all the selected dose levels showed synergistic effect. While in Huh7 cells, 2 out of 4 of the combination concentrations showed additive effects (Figure 2).

The ability of sorafenib/OSU-2S combination to elicit apoptotic cell death compared to single drug treatment was initially tested using caspases 3/7 activity assay. For this experiment, Hep3B cells were selected since it was the most resistant to sorafenib as a single agent. Results showed that increasing doses of OSU-2S dramatically increased sorafenib-induced activation of caspases 3/7 (1.5- to 3-fold increase) especially at 2.5 μM dose level of OSU-2S (Figure 3A). These results were confirmed by Western blotting of two hallmarks of apoptosis, caspase 3 and its downstream target protein, PARP. The results showed a significant increase in caspase 3 activation through cleavage with a parallel increase in PARP cleavage (Figure 3B).

In addition, the possible modulatory effects of OSU-2S on the reported anticancer mechanism of sorafenib were investigated by Western blotting of ERK1/2 phosphorylation, which is considered as a major target of sorafenib (Adnane et al., 2006; Manov et al., 2011). Results showed that OSU-2S increased the inhibitory effect of sorafenib on ERK1/2 phosphorylation (Figure 3B). At the same time, sorafenib/OSU-2S combination displayed a significant increase in PKCδ activation as a reported mechanism for OSU-2S and another biomarker for apoptosis (Omar et al., 2011) (Figure 3B).

The ability of OSU-2S to synergize the effect of sorafenib on endothelial cell migration/invasion was analyzed by modified Boyden’s chamber assay. Only cancer cells with high migratory ability can pass through the Boyden’s chamber membrane of 8 μm pore. Sorafenib effectively inhibited the ability of HCC cells to invade Matrigel-coated membranes in a dose-dependent manner (Figure 3C). In addition, sorafenib inhibited the migratory ability of HCC cells through the porous inserts (Figure 3D). The sorafenib/OSU-2S combination showed a significant synergy in sorafenib-mediated inhibition of both migration and invasion.

PKCδ, a pro-apoptotic kinase, is involved in caspase-3-dependent apoptotic pathway in HCC (Reyland, 2007; Hung et al., 2008). The role of PKCδ activation as a possible mechanism for the synergy between sorafenib and OSU-2S was studied through the knockdown of PKCδ of from Hep3B and Huh7 cell lines. The knockdown of PKCδ protein was initially confirmed by Western blotting and the stable clones with the lowest expression levels of PKCδ was used for the following MTT analysis (Figure 4A). PKCδ knockdown caused a significant increase in the resistance of both Hep3B and Huh7 to sorafenib and OSU-2S with about 2- to 3.5-fold increase in the IC₅₀ (Figures 4B,C). In addition, PKCδ knockdown completely eliminated the synergy between sorafenib and OSU-2S in their combination as indicated by all CI values over 1 (Figure 4D). These results suggested the activation of PKCδ as a putative mechanism for synergistic sorafenib/OSU-2S combination.

The used HCC cell lines showed differential sensitivity to both sorafenib and OSU-2S. HepG2 was the most sensitive to both drugs. Among the four used HCC cell lines in this study, only HepG2 cells have wild type functional p53 while the others lack functional p53 due to deletion or mutation (Lee et al., 2002). Based on this observation and based on the central role of p53 in apoptotic cell death, the lack of functional p53 in Hep3B postulated as a mechanism of resistance. To elucidate the possible role of the presence of functional p53 in HCC cells in the sensitivity or the resistance of sorafenib and OSU-2S, p53 knockdown was performed in HepG2 cells followed by MTT assay. Results showed that the p53 knockdown caused a significant increase in the resistance of HepG2 cells toward both sorafenib and OSU-2S with IC₅₀ values of 9.9 and 6.1 μM, respectively, which is almost doubling of the IC₅₀ values (Figures 5A–C). In addition, p53 over-expression was performed in Hep3B cells, which were the most resistant to sorafenib followed the MTT assay. The results showed that restoring the activity of p53 caused the Hep3B cells to be much more sensitive to both sorafenib and OSU-2S with IC₅₀ values of 3.2 and 1.9 μM, respectively (Figures 6A–C). Furthermore, the anti-proliferative activity of sorafenib/OSU-2S combination was significantly increased upon p53 restoration by the overexpression as indicated by the progressive morphological changes from flat to round in Hep3B-p53-GFP which are characteristic of apoptosis (Figure 6D). These results suggested a significant role of p53 in the sensitivity of HCC cells to both sorafenib and OSU-2S as single agents or in combination.