Seventy-three immortalized human cancer cell lines from 15 solid tumor models were analysed (Table I). The U87MG was purchased from American Type Culture Collection (ATCC HTB-14; Manassas, VA, USA). Authentication of all cell lines was carried out by the Center for Molecular Diagnostics of Barretos Cancer Hospital (São Paulo, Brazil) as previously reported (20). Shorthly, short tandem repeat (STR) DNA typing was performed according to the International Reference Standard for Authentication of Human Cell Lines previously described (21). The identity of all cell lines was confirmed by genotyping, with the exception of U373, which was shown to be a sub-clone of U251 cell line. All the cell lines were maintained in Dulbecco's modified Eagle's medium (DMEM 1X, high glucose; Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) or RPMI-1640 (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin/streptomycin solution (P/S; Gibco; Thermo Fisher Scientific, Inc.), at 37°C and 5% CO₂.

The sap from E. tirucalli L. (accepted name record 82539-Sp. Pl. 452 1753.) was initially extracted with hexane and the resulting precipitate was extracted with n-butanol. The most lipophilic compounds present in the butanol fraction were purified by means of high performance liquid chromatography analysis (HPLC). Further purification of the compounds was carried out using a Sephadex G75 column in a mixture of hexane-ethyl acetate. Recrystallization of the acetate fraction from butanol gave 3.5 g crystals, comprising euphol acetate and filtrate (1.5 g). The chemical structure of the euphol, representeted in Fig. 1, was determined by elemental analyses of 1H NMR and 13C NMR spectral data, and by comparison with their respective authentic compounds using Chemdraw software version 7.0 (22,23) (PubChem CID: 441678). The 1H NMR (500 MHz) and 13C NMR (126 MHz) spectra were recorded on a Bruker 500 MHz instrument. The MS spectra were recorded on a Perkin Elmer instrument, model API 150 and run in ES-MS positive mode: MH+ 427 m/e, MH+ _H2O 409 m/e. The NMR parameters were 13C NMR (CDCl3): 15.7 (C29), 15.8 (C18), 17.9 (C26), 19.1 (C21), 19.2 (C6), 20.4 (C19), 21.7 (C11), 24.9 (C30), 25.0 (C23), 25.9 (C27), 28.0 (C2, C7), 28.2 (C28), 28.3 (C16), 30.0 (C15), 31.1 (C12), 35.4 (C1), 35.7 (C22), 36.1 (C20), 37.5 (C10), 39.2 (C4), 44.3 (C13), 49.9 (C17), 50.2 (C14), 51.2 (C5), 79.2 (C3), 125.4 (C24), 131.1 (C25), 133.8 (C9), 134.3 (C8) and 1H NMR (CDCl3): 0.75 (3H, s, H-18), 0.80 (3H, s, H-29), 0.85 (3H, d, H-21), 0.87 (3H, s, H-30), 0.95 (3H, s, H-19), 1.00 (3H, s, H-28), 1.50 (3H, s, H-26), 1.68 (3H, s, H-27), 3.23 (1H, dd, H-3), 5.09 (1H, bt, H-24). The tetracyclic triterpene euphol used in this study showed >95% purity.

The extract fraction was initially dissolved in dimethyl sulfoxide (DMSO) at a concentration of 50 mg/ml and stored at −20°C. The intermediate dilutions of the experimental compound were prepared to obtain a concentration of 1% DMSO.

The cytotoxicity effect of euphol was assessed by Cell Titer 96 Aqueous cell proliferation assay (MTS assay; Promega Corporation, Madison, WI, USA), following the manufacturer's instructions as previous described (20,24). For experiments, the cells (third and fifth passage) were plated into 96-well plates (until a maximum 5×10³ cells/well). The plate was incubated overnight for optimal attachment of adherent lines, and then placed under low serum-starved conditions for 24 h (DMEM supplemented with 0.5% of FBS). Subsequently, the cells were treated with increasing concentrations of the test compound diluted in DMEM (0.5% FBS) and incubated for 72 h. The control groups received the same amount of vehicle (1% DMSO, final concentration). Growth of tumoral cells was quantified by ability of living cells to reduce the yellow dye MTS to a blue formazan product. Two hours before the end of incubation, 10 µl of MTS were added to each well, and the plate further incubated for 2 h at 37°C. Absorbance was measured in automatic microplate reader Varioskan (Thermo Fisher Scientific, Inc.) at 490 nm. The response to euphol treatment was assessed by standardizing treated groups to the untreated control, and were expressed as a percentage relatively to control cells, in DMSO alone (considered as 100% viability) ± SD.

The results of absorbance values of treated cells were converted to percentage of cell viability in cells in with presence of the vehicle (DMSO), which were used as control, corresponding to 100% survival. The analysis of the non-linear regression curve using GraphPad PRISM version 7 (GraphPad Software, Inc., La Jolla, CA, USA) the was carried out on results of viability, yielding an equation used to calculate the concentration of substance required to produce 50% reduction in cell viability (IC₅₀) as previous determined (25,26).

Combination studies were done with fixed concentrations (determinated IC₅₀ value) of standard chemotherapeutic paclitaxel (Sigma-T7402; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) and gemcitabine hydrochloride (Sigma-G6423; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) (data not shown), exposed simultaneously to increasing concentrations of euphol. The drug interactions were evaluated by the combination index (CI) that was calculated by the Chou-Talalay equation, which takes into account both the potency (Dm or IC₅₀) and the shape of the dose-effect curve (27,28), using CalcuSyn software version 2.0 (Biosoft, Ferguson, MO, USA). In CI analysis, synergy was defined as CI values significantly lower than 1.0; antagonism as CI values significantly higher than 1.0; and additivity as CI values equal to 1.0 at drug IC₅₀ value for each cell line.

The proliferation effect of euphol was assessed by BrdU assay kit (Roche Applied Science, Mannheim, Germany) following the manufacturer's instructions. In brief, 5×10³ cells/well were seeded in a 96-well, flat-bottom plate. The plate was incubated overnight for optimal attachment of adherent lines, and then placed under low serum-starved conditions for 24 h (DMEM-0.5% FBS). Subsequently, the cells were treated with increasing concentrations of the test compound diluted in DMEM (0.5% FBS) and incubated for 72 h. The control groups received the same amount of vehicle (1% DMSO, final concentration). Therefore, the proliferation effect was assessed by BrdU assay kit following the manufacturer's instructions. Experiments were done in triplicate in three independent experiments for each cell line.

The criterion of GI to ascertain the cell line sensibility to euphol was previously described (29). Mean GI values was calculated at fixed dose of 15 µM of euphol (concentration closer to the mean IC₅₀ value for all cell lines) and established as 100%-percentage of viable cells at this dose. Samples exhibiting more than 60% GI in the presence of 15 µM euphol were classified as highly sensitive (HS); moderately sensitive (MS) when located between 40–60%; and resistant when showing less than 40%. All the assays were done in triplicate and repeated at least three times for each cell line.

Cell migration properties were evaluated by wound-healing assay, as previsously described by our group (26,30). The pancreatic cancer cell lines, Mia-Pa-Ca-2 and Panc-1, were plated in 6-well plates and grown to confluence. A sterile tip was used to create a scratch in monolayer cells. Cells were then incubated with euphol at 8.46 and 21.47 µM. The ‘wounded’ areas were photographed by phase contrast microscopy (Model IX71; Olympus Corporation, Tokyo, Japan) to evaluate wound closure (0, 24, 48 and 72 h). The migration rate of individual cells was determined by measuring the distances covered from the initial time to the selected time-points (bar of distance tool, DP2-BSW Olympus version 2.2). The percentage of the relative migration distance was calculated as wound area at a given time compared to the initial wound surface. Pictures shown are representative of three independent experiments performed in triplicates.

Inhibition of anchorage-independent was assesed by soft-type-agar assay as reported (30). We placed 1 ml of acellular solution of 0.6% agar (combining equal volumes of 1.2% Noble agar with 2× concentrated DMEM with 20% FBS) into a six-well plate and incubated at 37°C for 10 min; 2×10⁴ cells of Mia-Pa-Ca-2 and Panc-1 were homogenized in solution containing DMEM supplemented with 0.35% agar (upper layer agar; equal volumes of 0.7% Noble agar and 2X concentrated DMEM with 20% FBS) and seeded onto acellular coating. After solidification 0.5 ml of DMEM medium + 10% FBS was added. The medium was changed every two days, and DMEM medium + 0.5% euphol at 3 and 10 µM was added. The cells were incubated at 37°C in a humidified atmosphere of 5% CO₂ for 20 days; colonies formed were stained with 0.05% crystal violet for 15 min. Photo-documented colonies were analyzed using the Image J Software. The assay was performed in two biological replicates and the experiments were done in duplicate.

The results of in vitro experiments are expressed as mean ± standard deviation (SD) of three independent experiments. IC₅₀ values were obtained by nonlinear regression. We applied Student's t-test for comparing two different conditions whereas two-way analysis of variance with Tukey's post hoc test was used for assessing differences between more groups. P<0.05 was considered to indicate a statistically significant difference. All statistical analyzes were performed using GraphPad PRISM version 7 (GraphPad Software, Inc.).

The antitumor effect of euphol in vitro was assessed using MTS assay on 73 human cancer lines from 15 solid tumor models (breast, colon, bladder, prostate, lung, pancreas, esophageal, head and neck, cervical, epidermoid carcinoma, meduloblastoma, placental choriocarcinoma, ovarian carcinoma, glioblastoma, and melanoma) (Table I). We generated complete dose-response curves and IC₅₀ values for this euphol treatment. Among each tumor type, the distinct cell lines exhibited a heterogeneous profile of response to euphol (Fig. 2A). The mean of IC₅₀ values was 15.14 (6.47 µg/ml), but varied significantly between individual cell lines with up to a more than 27-fold difference in the IC₅₀ values [IC₅₀ range: 1.41–38.89 µM (0.60–16.62 µg/ml)]. Esophageal squamous cell carcinoma and pancreatic carcinomas showed the most sensitive profiles (IC₅₀ mean 11.08 and 6.84 µM, respectively, Fig. 2A and Table I), followed by prostate, melanoma and colon cancer cell lines.

To allow a better classification of the cell lines response to euphol, we determined their GI. We found that 50.68% (37/77) were classified as HS, 21.92% (16/73) were MS, and 27.4% (20/77) were resistant (Fig. 2B and Table I). Esophageal (100%), prostate (100%) and pancreatic (80%) cancer models showed a higher percentage of HS cell lines. At variance, glioma (54.5%) followed by breast tumor type (42%) has the most cancer cell lines scored as resistant.

We further investigated the biological effect of euphol on pancreatic cancer cell lines, the most sensitive tumoral type in our study. To determine whether the effect of euphol on cancer cells is cytotoxic or cytostatic, its effect was also evaluated on the proliferation of pancreatic cell lines by BrdU incorporation. As shown in Fig. 3A, euphol exhibited dose-dependent effects on proliferation of pancreatic cancer cell lines (Panc-1 and Mia-Pa-Ca-2) but varied significantly between individual cell lines. In both cancer cell lines, low doses of euphol slightly decreased proliferation and the dose of 17.51 µM was able to inhibit almost 50% of the proliferation.

Euphol also exhibited dose-dependent effects on cell viability on pancreatic cancer cell lines (Panc-1 and Mia-Pa-Ca-2). Although euphol decreased the proliferation, the strongest inhibition of proliferation was seen at 35.1 µM concentration of euphol after 72 h, with almost 39.4% for Mia-Pa-Ca-2 cells and 51% for Panc-1 cells (Fig. 3A), whereas the same concentration suppressed cell viability of Mia-Pa-Ca-2 cells by 10.7% and Panc-1 cells by 22.4% (Fig. 3B). Thus, comparing the effects on cell viability in concentrations of the same magnitude, euphol seemed to inhibit growth through a more cytotoxic than cytostatic fashion.

Next, we investigated its biological effect on pancreatic cancer cell lines. In untreated conditions for both cell lines (Mia-Pa-Ca-2 and Panc-1, complete close up of the wound took longer than 72 h, suggesting that these cell lines have an intrinsic low migratory capability. Those two cells lines were treated with euphol at 8.46 and 21.47 µM. In spite of the low migratory feature of the cells, euphol treatment significantly inhibited cell migration in Mia-Pa-Ca-2 at 72 h when compared to control cells (Fig. 3C). We could not evaluate the effect of euphol in Panc-1 cell motility, since this cell present an inherent low motility phenotype hampered an adequate assessment of any inhibition. The findings suggest that euphol has a cytotoxic effect but were not able to inhibit the migration of this cell line (Fig. 3D).

In addition to cell proliferation (anchorage-dependent growth) and cell migration, colony formation (anchorage-independent growth) is one of the typical characteristics of the metastatic potential of cancer cells in vitro and strongly correlates with tumorigenesis in vivo (31). Therefore, we evaluated the effect of euphol in anchorage-independent growth of Panc-1 cancer cells. Panc-1 cells formed colonies on agar after 20 days of incubation, and the presence of euphol at IC₅₀ value resulted in a significant suppression of number and size of colonies (P<0.05; Fig. 3E). The ability to form colony was reduced by 90% compared to the untreated control. Similarly, we proceeded the same way with Mia-Pa-Ca-2 cell line, however we can not observe the colonies formation even in the control condition suggesting a low tumorigenic potential of this cell line (data not shown).

We also evaluated the potential combinatorial value of euphol in the context of standard esophageal and pancreatic tumor therapy. We found that euphol and gemcitabine combination treatment showed a synergistic effect (CI<1) in 50% of pancreatic cells lines investigated (mean CI values, range: 0.76–0.8; Table II), being the combination more effective than single agents. Likewise, euphol was able to synergistically sensitize most esophageal cells lines to paclitaxel treatment (mean CI values, range: 0.37–0.55; Table III).