Ximenynic acid was obtained (99.5% purity) from the seed oil of Santalum spicatum by low temperature recrystallization, as described previously (17). Oleic acid (Sigma-Aldrich; Merck Millipore, Darmstadt, Germany) was selected as the non-alkynyl-fatty acid control for its similar structure to ximenynic acid. All other chemicals were purchased commercially as analytical grade reagents.

HepG2 human hepatoma cell line was obtained from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). Cells were grown in high glucose Dulbecco's modified Eagle's medium (DMEM; Hyclone; GE Healthcare Life Sciences, Logan, UT, USA) supplemented with 110 mg/l sodium pyruvate, 4 mM L-glutamine, 10% fetal bovine serum (Biowest, Nuaillé, France), 10 mM HEPES (Corning Incorporated, Corning, NY, USA), 100 U/ml penicillin and 0.1 mg/ml streptomycin at 37°C in a humidified incubator with 5% CO2. The medium was changed every 2 days. Cells were passaged when they reached 70–80% confluence, and the culture medium was changed 1 day before.

Cells were seeded overnight in standard medium and then change to serum-free medium supplemented with 1% insulin-transferrin-selenium (ITS; Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA), 0.1 mg/ml fatty acid-free bovine serum albumin (BSA; MP Biomedicals, LLC, Santa Ana, CA, USA) for synchronizing cell cycle division. After 24 h serum starvation, cells were incubated in experimental media for indicated times. The experimental media was DMEM supplemented with 1% ITS, 0.1 mg/ml BSA and various concentrations of ximenynic acid or oleic acid with a molar ratio of 4:1 to BSA. Cells treated with medium not containing fatty acids were termed the vehicle control group.

Cells were cultured in 96-well plates overnight with ~1500 cells/well. After serum-starving for 24 h, serum-free medium was removed and ximenynic acid or oleic acid at 25, 50, 100 and 150 µM were added to the cells. After 72 h incubation, the media were replaced with serum-free medium diluted methyl thiazolyl tetrazolium (MTT; Amresco, LLC, Solon, OH, USA) solution (5 mg/ml) and incubated at 37°C for 4 h. Subsequently, MTT medium was removed and the cells were dissolved with dimethyl sulfoxide followed by detection of absorption at 595 nm with the Bio-Rad iMark Microplate Absorbance Reader (Bio-Rad Laboratories, Inc., Hercules, CA, USA).

Cell cycle distribution was analyzed by propidium iodide (PI) staining. Cells were seeded in the 6-well plates overnight, and then incubated with serum-free media containing 50 or 150 µM ximenynic acid or oleic acid for 24 h and 36 h following rinsing with phosphate-buffered saline (PBS). The harvested cells were slowly fixed in pre-chilled 70% ethanol following washing with PBS. The fixed cells were incubated at −20°C overnight. Subsequently to rinsing with PBS, cells were re-suspended with PI solution from the Cell Cycle Staining kit [CCS012; Multi Sciences (Lianke) Biotech Co., Ltd., Hangzhou, China] and incubated in a dark at room temperature for 30 min. Cell cycle distribution was then analyzed using BD FACSCalibur (Becton Dickinson, USA).

Cell apoptosis was analyzed with Annexin V-FITC (fluorescein isothiocyanate) and PI kit (Multisciences, China). Cells were cultured in 6-well plates overnight, and then starved for 24 h with serum-free medium followed by culture with different concentrations of ximenynic acid or oleic acid for 72 h. The harvested cells were gently rinsed with PBS. Cells were stained with Annexin V-FITC and PI in dark for 15 min. The level of apoptosis was then analyzed by BD FACSCalibur (BD Biosciences, Franklin Lakes, NJ, USA).

Cells were seeded in 6-well plates overnight. After starving for 24 h with serum-free medium, cells were treated with 50 or 150 µM ximenynic acid or oleic acid for 72 h. Total RNA was extracted by HP Total RNA kit (Omega Bio-Tek, Inc., Norcross, GA, USA) according to the manufacturer's protocol, and the RNA concentration was determined using a NanoDrop 2000 (Thermo Fisher Scientific, Inc., Wilmington, DE, USA). RT was performed on total RNA to produce cDNA with using a Transcriptor First Strand cDNA Synthesis kit (Roche Diagnostics, Basel, Switzerland). The expression of genes was analyzed by qPCR. The sequences of primers are presented in Table I. The qPCR analysis was performed using the SsoFast EvaGreen Supermix (Bio-Rad Laboratories, Inc.), according to the manufacturer's protocol, with a CFX96™ Real Time PCR Detection System (Bio-Rad Laboratories, Inc.). The qPCR program commenced with initial denaturation at 95°C for 30 sec followed by 40 amplification cycles of 95°C for 5 sec (denaturation) and 60°C for 10 sec (annealing and extension). The melting curve program was from 65°C to 95°C for 5 sec with an increment rate of 0.5°C/sec (18). The RPLPO gene was used as the reference gene. Data were selected from a minimum of 3 experiments.

Cells were cultured in 100 mm dishes overnight and then starved with serum-free medium for 24 h. After treatment with 50 or 150 µM ximenynic acid or oleic acid for 72 h, cells were harvested and proteins were extracted using cell lysis buffer supplemented with phenylmethylsulfonyl fluoride (Beyotime Institute of Biotechnology, Haimen, China). Whole-cell lysates were centrifuged at 20,000 × g at 4°C, for 10 min to collect the supernatant. The protein concentration was determined by bicinchoninic acid protein assay (Beyotime Institute of Biotechnology). Protein (20 µg) were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis separation and transferred to polyvinylidene difluoride membrane (Bio-Rad Laboratories, Inc.) for probing with antibodies. The antibodies for caspase-3 (cat. no. 9662s), silent information regulator T1 (SIRT1; cat. no. 2496p), COX-2 (cat. no. 12282s), general control of amino acid synthesis yeast homolog like 2 (GCN5L2; cat. no. 3305p), and GAPDH (cat. no. 2118s) were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA), and COX-1 (cat. no. ab109025) from Abcam (Cambridge, UK). The secondary antibodies were anti-mouse IgG (cat. no. #7076) and anti-rabbit IgG (cat. no. #7074) from Cell Signaling Technology, Inc. These antibodies were diluted in 5% BSA (1:1,000; Beyotime Institute of Biotechnology). The immunoreactive bands were incubated with the primary antibodies overnight at 4°C, then were washed three times with PBS with Tween-20 (PBST). The secondary antibody was incubated for 1 h at room temperature, and also washed three time with PBST. The protein bands were visualized using the ECL reagent (Bio-Rad Laboratories, Inc.). The intensity of immunoreactive bands was analyzed by Quantity One software (Bio-Rad Laboratories, Inc.). Western blotting data were selected from a minimum of three independent experiments.

Data are expressed as the mean ± standard deviation of three independent experiments. The significance of the difference between groups was analyzed by SPSS software version 18 (SPSS, Inc., Chicago, IL, USA) using the Bonferroni or Dunnett's T3 method of single factor analysis of variance (19). P<0.05 was considered to indicate a statistically significant difference.

HepG2 cells were treated with different concentrations (0–150 µM) of ximenynic acid or oleic acid for 72 h, and the cell viability was measured by MTT assay. As shown in Fig. 1, ximenynic acid significantly inhibited the growth of HepG2 cells from 0 to 150 µM in a dose-dependent manner compared with the control group (P=0.12, 25 µM; P=0.04, 50 µM; P=0.03, 100 µM; P=0.12, 150 µM), while oleic acid could not inhibit HepG2 cell growth at every concentration compared with the control group (Fig. 1).

Considering the anti-proliferation activity of ximenynic acid demonstrated in Fig. 1, it was suspected that ximenynic acid may also lead to apoptosis of HepG2 cells. Thus, the HepG2 cells were treated with ximenynic acid or oleic acid (50 and 150 µM) for 72 h followed by PI staining and flow cytometry analysis (Fig. 2A). It was observed that the percentage of apoptotic cells was significantly increased by ximenynic acid at a concentration of 150 µM in a dose-dependent manner (P=0.018; Fig. 2B), and accordingly the percentage of live cells was significantly reduced at the same concentration (P=0.004; Fig. 2C) compared with the vehicle control group. In the oleic acid group, the percentage of apoptotic cells and living cells was not significantly different compared with the control group (Fig. 2B and C). The anti-cancer activity of 150 µM ximenynic acid was greater than that of oleic acid.

The degradation of pro-caspase-3 protein occurs during the process of cell apoptosis, which results in protein cleavage into several fragments (20) and indirectly induces a decrease in the pro-protein. The cleaved caspase-3 (17 KDa) was not detected in the ximenynic acid group, however western blot analysis demonstrated that the level of pro-caspase-3 (35 KDa) was significantly decreased by 150 µM ximenynic acid compared with the vehicle control (P=0.001, P<0.05; Fig. 2D). Together, flow cytometry and western blot analyses indicated that ximenynic acid promotes apoptosis of HepG2 cells in a dose-dependent manner, and was more effective than oleic acid.

The upstream signals that mediate the effect of ximenynic acid on apoptosis remain unclear. SIRT1 is a NAD-dependent deacetylase that inhibits apoptosis by downregulating transcriptional activity of p53 (21). In addition, SIRT1 silencing induces activation of caspase-3, indicating it is a downstream target of SIRT1 (22). The current study demonstrated that SIRT1 protein expression was significantly reduced by 150 µM ximenynic acid compared with the vehicle control group (P=0.013, 50 µM; P=0.019, 150 µM), whereas there was no significant difference in the oleic acid groups compared with vehicle (Fig. 3). These results indicate that ximenynic acid-induced apoptosis is associated with SIRT1 inhibition in HepG2 cells.

Cell cycle arrest in G1/S transition can induce cancer cell apoptosis (16). The current study investigated whether ximenynic acid affects the cell cycle distribution. After treatment with ximenynic acid for 24 or 36 h, the cell cycle distributions of HepG2 cells were observably changed (Fig. 4A-C). The percentage of cells in G0/G1 phase was significantly increased in the ximenynic acid groups (P=0.002, 24 h, 50 µM; P=0.001, 24 h, 150 µM; P=0.001, 36 h, 50 µM; P=0.001, 36 h, 150 µM; Fig. 4A), and significantly decreased in S phase (P=0.002, 24 h, 50 µM; P=0.001, 24 h, 150 µM; P=0.001, 36 h, 50 µM; P=0.001, 36 h, 150 µM; Fig. 4B) and G2/M phase (P=0.014, 24 h, 50 µM; P=0.001, 24 h, 150 µM; P=0.001, 36 h, 50 µM; P=0.005, 36 h, 150 µM; Fig. 4C) compared with the vehicle control group in a dose- and time-dependent manner. The structures of ximenynic acid and oleic acid are similar, however, oleic acid did not observably alter the cell cycle distributions of HepG2 cells at 50 or 150 µM (Fig. 4A-C). These results indicate that ximenynic acid change the cell cycle distribution of HepG2 cells depending on the dose and duration of ximenynic acid treatment.

To understand whether ximenynic acid affects the expression of cell cycle-associated genes in HepG2 cells, qPCR analysis was performed to determine the expression variation of cyclin genes, including cyclin E1 (CCNE1) and cyclin D3 (CCND3). After treatment for 36 h, the mRNA levels of CCNE1 and CCND3 were significantly decreased in the50 and 150 µM ximenynic acid group compared with the vehicle control group (CCNE1, P=0.001; CCND3, P=0.043; Fig. 4D and E), and its inhibitory effect on CCND3 and CCNE1 was stronger than oleic acid.

GCN5L2 (also termed lysine acetyltransferase 2A) is a typical histone acetyltransferase (HAT), which promotes cell cycle progression by acetylation and deacetylation of histones (23). The expression of GCN5L2 protein was significantly downregulated in the 150 µM ximenynic acid group compared with the vehicle control group (P=0.026; Fig. 4F).

COX-1 and COX-2 mRNA and protein were demonstrated to be expressed in the HepG2 cells (Fig. 5). After serum starvation for 24 h, HepG2 cells were incubated with fatty acids for 72 h. Although COX-2 protein expression was significantly increased in oleic acid group at 50 µM (P=0.01, P<0.05), it was not changed in the ximenynic acid group compared with the vehicle control group (Fig. 5A). These results were largely in keeping with the COX-2 gene expression of PTGS2 [prostaglandin-endoperoxide synthase 2 (PTGS2)] in HepG2 (Fig. 5C).

Unexpectedly, compared with COX-2, ximenynic acid exerted a greater effect on COX-1, with PTGS1 mRNA (P=0.009, 50 µM; P=0.001, 150 µM; Fig. 5D) and COX-1 protein (P=0.005, 50 µM; P=0.001, 150 µM; Fig. 5B) levels significantly decreased by 50 to 150 µM ximenynic acid compared with the vehicle control group, whereas PTGS1 levels were significantly increased by oleic acid treatment at the concentrations of 50 and 150 µM (P=0.001, 50 µM; P=0.001, 150 µM; Fig. 5D). These results indicate that ximenynic acid selectively inhibits COX-1 expression.

PGE2 is the product of arachidonic acid metabolism by COXs (24). Prostaglandin E receptor 2 (EP2) a prostaglandin E receptor 4 (EP4) are PGE2 receptors involved in various physiological and pathophysiological processes (25). EP2 and EP4 mRNAs were significantly downregulated in the 150 µM ximenynic acid group compared with the vehicle control (EP2, P=0.03; EP4, P=0.001), while they were unchanged by oleic acid (Fig. 5E and F). It was indicated that the inhibitory effect of ximenynic acid on EP2 and EP4 mRNA levels was more effective than that of oleic acid.

Vascular endothelial growth factor (VEGF)-B and VEGF-C are two members of the VEGF family that induce angiogenesis and inhibit apoptosis (26). The VEGF-B and VEGF-C mRNA levels were significantly decreased in the 50 and 150 µM ximenynic acid groups compared with the vehicle control group (P<0.05) and oleic acid (P<0.05) groups (VEGF-B: P=0.001, 50 µM; P=0.001, 150 µM; VEGF-C: P=0.049, 50 µM; P=0.044, 150 µM), whereas VEGF-B was significantly upregulated in the oleic acid (P<0.05) groups (Fig. 6A and B).

Numerous factors modulate angiogenesis in cancer. The current study investigated the expression of several angiogenesis-associated genes including chemokine (C-C motif) ligand 2 (CCL2), Hypoxia inducible factor 1, α submit (HIF-1α), high mobility group box 1 (HMGB1), and transforming growth factor-β1 (TGF-β1). CCL2 is recognized for its role in the inflammatory response and modulation of angiogenesis (27). HMGB1, a proinflammatory cytokine, is crucial for inflammation-associated diseases and involved in cell apoptosis and angiogenesis (28). HIF-1α, an oxygen balance regulator, regulates VEGF transcription and is involved in numerous cell functions, including angiogenesis (29). TGF-β1, a multifunction factor for cells, has also been previously demonstrated to induce angiogenesis progression (30). In the present study, these four mRNA levels were significantly inhibited by 150 µM ximenynic acid compared with the vehicle control group (CCL2, P=0.016; TGF-β1, P=0.004; HMGB1, P=0.007; HIF-1α, P=0.044; Fig. 6C-F). By contrast, oleic acid acid significantly promoted the expression of HMGB1 genes at 50 µM (P=0.002), however did not show significant effects on TGF-β1, CCL2 and HIF-1α genes compared with the control group (Fig. 6C-F).