All reagents were purchased from Sigma, Israel unless otherwise stated. Cell culture media and additives were obtained from Beit-Haemek, Israel. Annexin V and propidium iodide were purchased from Biotium.

The following materials were tested:

The ratio of each compound is described in Table 1.

Cells were seeded in 96-well plates (1x10⁴ cells/well) in complete medium. On the following day, different concentrations of the above-described agents were added to the cells in triplicates. At 72 h later, the medium was replaced by fresh media (100 μl per well) containing 1 mg/ml 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and incubated for 2–4 h. MTT-formazan crystals were dissolved by the addition of 100 μl extraction buffer. Absorbance at 570nm and a reference wavelength of 690nm were recorded on an automated microplate reader. The relative number of viable cells was compared to untreated cells.

Transfections were performed using jetPEI^™ (Polyplus-transfection Inc, NY, USA) according to the manufacturer's instructions. 5 x 10⁵HCT116 cells were seeded in 6-well plates for Luc assays. The next day, when the cells were about 50% confluent, co-transfection with 3 μg of PY4-SV40-LUC vector plus 0.3 ng of pRL-CMV (Promega) was performed. 24h after the transfection medium was replaced and the cells were exposed to cetuximab, different concentration of terpinen-4-ol or left untreated. Luc assay was performed 48 h post after the treatment. Briefly, cells were washed once with PBS and then lyzed in 250 ~μl of lysis buffer for 5 min at room temperature. The lysates were centrifuged at 14,000 rpm for 5 min, and 20 μl of each lysate were used to measure Luc reporter gene expression. The Luc activity was normalized to Renilla Luc activity from a parallel co-transfection of pRL-CMVDual Luc system, Promega).

Cells were seeded in 12-well plates (1x10⁵ cells/well) in complete medium and treated with terpinen-4-ol at several concentrations for 72 h. Annexin V was detected according to the manufacturer’s protocol. The cells were washed with PBS and then incubated in a solution of the membrane-impermeable nuclear dye propidium iodide. The cells were then immediately analyzed by flow cytometry [FACSCalibur (Becton Dickinson, CA)], and the results were analyzed with the CELLQuest program (Becton Dickinson).

Male athymic nude mice, 6–8 weeks old, (Harlan Laboratories) were housed in sterile cages and handled with aseptic precautions. They were fed ad libitum. For testing the therapeutic potential of terpinen-4-ol, exponentially growing cancer cells were harvested and resuspended at a final concentration of 5x10⁶ cells per 0.1 ml PBS per injection. The cells were injected subcutaneously into the flank of the mice. When tumors were palpable (~0.3 cm³), the mice were randomly divided into groups and the treatment was started (intraperitoneal and/or intratumoral injections). The animals were treated twice a week for 3 weeks. They were weighed and tumor volume was measured with a caliper and plotted every 3 days starting from the onset the treatment. Tumor volume was calculated as 4/3π∙a∙b². At the end of the experiment, the mice were anesthetized and sacrificed by cervical dislocation and the tumors were excised.

Data from the in vitro studies are presented as mean±SD of sets of data as determined in triplicates. Statistical significance between treatments was determined by Student’s t-test, and P values < .05 were considered significant. In the in vivo studies, the tumor-bearing mice were randomized into 5 treatment groups and the tumor volumes were periodically monitored and calculated as 4/3π∙a∙b². Significant differences between groups and at different time points were determined by Student’s t-test.

The study was approved by the institutional committee for animal welfare at Tel-Aviv Sourasky Medical Center.

Two mixtures with different monoterpens composition were tested (Table 1). Mixture 1 was significantly more effective and toxic than mixture 2 (P<0.005). The main difference between the two was terpinen-4-ol. The results (Fig 1A) indicated for significant differences in cell survival between the mixtures, allowing us to identify terpinen-4-ol and its contribution to the cytotoxic effect.

Terpinen-4-ol inhibited the growth of colorectal (Fig 2A), pancreatic (Fig 2B), gastric (Fig 2C) and prostate (Fig 2D) cancers in a dose-dependent fashion [10% (in 0.005%)-90 (in 0.1%) growth inhibition)], as measured qualitatively by microscopic observations (data not shown) and quantitatively by the enzymatic MTT assay (Fig 2).

Low concentrations of terpinen-4-ol (0.005–0.01%) inhibited HCT116 cell proliferation in a modest way (20–30%) as measured by the MTT viability assay. High concentrations (0.05–0.1%), induced massive cell death (75–90%). As can be seen in Fig 3, apoptosis is the cell death mechanism responsible for the cytotoxic effect induced by Terpinen-4-ol. Early apoptotic death was induced by low dose of terpinen-4-ol, whereas the percentages of late apoptosis increased at higher concentrations. No necrotic cells were observed.

Terpinen-4-ol was combined with different types of conventional chemotherapy, depending on the type of cancer being treated. For CRC therapy, terpinen-4-ol was combined with oxaliplatin (Fig 4A) and 5-FU (Fig 4B). An impressive synergistic growth inhibition effect was achieved (83% and 91%, for oxaliplatin and 5-FU, respectively) in the combined regimen as compared to each agent alone [12% (oxaliplatin), 25% (terpinen-4-ol), and 20% (5-FU)]. These results have achieved statistical significance, P<0.005. Similar results were obtained in pancreatic cancer cells (Fig 4C–4F). Terpinen-4-ol impressively increased the efficacy of gemcitabine and that of Tarceva (60–85% cell death).

Combining terpinen-4-ol with either humanized anti-CD24 matured antibody (Arber's lab, Tel Aviv, Israel) or the chimeric anti EGFR antibody (Cetuximab, Merck Serono) resulted in a remarkable synergistic growth inhibition effect (85–90%, P<0.005) on human CRC (Fig 5A and 5B) and prostate cancer cells (Fig 5C and 5D).

The DLD1 CRC cells carry a mutation in the KRAS oncogene. Therefore, they are resistant to anti-epidermal growth factor (EGFR) therapy. Combining terpinen-4-ol (0.01%) with cetuximab (1 μM) resulted in a rather impressive efficacy of a 85–90% growth inhibition. These results were confirmed in another KRAS mutated CRC cell line (HCT116) (Fig 5E), with an 80% growth inhibition (P<0.005) for the combined therapy.

For that purpose we used the (Ets/Ap1)₄ RAS-responsive element (Py4) construct, which we had previously constructed [24–26]. The activity of the KRAS pathway was evaluated in mutated CRC cells (HCT116) (Fig 5F). Transfection with Py4-SV40-Luc activity in the presence of terpinen-4-ol (0.005 and 0.01%) was 1.3 and 1.5 fold lower as compared to the activity in its absence (P<0.05)_. No effect was observed after exposure to cetuximab (1 μM).

Next, we tested the potential anti-tumor activity of terpinen-4-ol in vivo. Intratumoral injections (5 injections) of the compound (0.1% and 1%) were given twice weekly to nude mice (n = 6) bearing xenografts of CRC DLD1 cells. The treatment was started when tumors were 0.3–0.5 cm³. Significant inhibition of tumor development was observed; 40% and 70% reduction in tumor volume and about 25% and 50% reduction of tumor weight (Fig 6A). These results were confirmed in another experiment (Fig 6B). No significant adverse effects were observed (data not shown).

When one of tumor's diameter reached the size of 0.5 mm intratumoral injections of terpinen-4-ol (0.1%) along with systemic (IP) administration of cetuximab (10 mg/kg) resulted in a significant decrease in tumor volume (62% ±2.5%, P<0.05) and weight (62.5% ±3%, P<0.05) (Fig 6C)

Two different formulations of terpinen-4-ol were evaluated in vivo, a nano and suspension formulations. The latter emerged as being more effective and safer (data not shown). Systemic administration of the suspension (1 and 4 mg) to mice bearing xenografts of CRC (HCT116) cells resulted in an impressive decrease in tumor volume (32%±1.1%, P<0.05 and 72%± 0.7%, P<0.005, respectively) and weight (26% ±1.19 and 70% ± 0.71%, P<0.005, respectively) in a dose-dependent manner (Fig 7).