Antibodies against: cleaved Asp175 caspase3 (no. 9661), cleaved caspase 8 (Asp391, 18C8, no. 9496), cleaved caspase 9 (Asp330, no. 9501), phospho-Stat3 (Tyr705)(D3A7, no. 9145), Stat3 (no. 9132) and phospho-Stat5 (Tyr694)(C11C5, 9359) were from Cell Signaling (Danvers, MA, USA). PARP-1 (F-2, sc-8007), Cdc25A (F-6, sc-7389), cyclin D1 (M-20, sc-718), p21 (C-19, sc-397), α-tubulin (DM1A, sc-32293), β-tubulin (H-235, sc-9104), Stat5 (C-17, sc-835), c-Jun (H-79, sc-1694) and Jun B (210, sc-73) were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA). c-Myc Ab-2 (9E10.3, no. MS-139-P1) was from Thermo Fisher Scientific (Fremont, CA, USA), phospho- Ser177 Cdc25A (no. AP3046a) from Abgent (San Diego, CA, USA), phospho-Ser139 H2AX (DR 1017) from Calbiochem (San Diego, CA, USA) and β-actin (AC-15, A5441) as well as anti-acetylated tubulin (clone6-11B-1, T6793) were from Sigma (St. Louis, MO, USA). The secondary antibodies peroxidase-conjugated anti-rabbit IgG and anti-mouse IgG were purchased from Dako (Glostrup, Denmark).

HL-60 (human promyelocytic leukaemia) cells and human MCF-7 and MDA-MB231 breast cancer cell lines were purchased from American Type Culture Collection (ATCC, Rockville, MD, USA) and grown in RPMI-1640 medium (HL-60). The other cell lines were grown in MEM medium, which was supplemented with 10% heat-inactivated foetal calf serum (FCS), 1% Glutamax and 1% penicillin/streptomycin (Life Technologies, Carlsbad, CA, USA), 1% NEAA (Invitrogen, Karlsruhe, Germany). Cells were kept in a humidified atmosphere at 37°C containing 5% CO₂.

Parts of the rhizomes of Smilax spinosa Miller (vernacular name: ‘Kokolmeka roja’) were collected in Guatemala, Departamento Petén, at the north-western shore of Lago Petén Itzá, San José, ∼1 km north of the road from San José to La Nueva San José (16 59′30″ N, 89 54′00″ W). Voucher specimens (leg. G. Krupitza and R. O. Frisch, Nr. 4-2009, 19. 04. 2009, Herbarium W, det. B. Wallnöfer (W) 26.1.2010) were archived at the Museum of Natural History, Vienna, Austria.

Rhizomes of S. spinosa were cut, dried by lyophilisation and then pulverized. Twenty grams of the obtained powder were mixed with 200 ml solvent (1:10; Table I) extracted in an ultra sonic bath for 10 min and afterwards under reflux for 1 h in the water bath. The solution was filtered and the retained plant material (residue) was dried at room temperature before being re-extracted with the next more polar solvent. The liquid extract was evaporated under reduced pressure to give a crude fraction (0.67 mg corresponding to 1 g rhizome; Table I) (15,16). The extract weights obtained from serial extraction of the dried rhizomes of S. spinosa with five solvents of increasing polarity are presented in Table I, which illustrates that the weight of the methanol extract corresponds to ∼20.6% of the dried rhizome.

The methanol extract was the most active of the S. spinosa extracts, therefore it was further fractionated by dissolving 4.1 g in 60 ml of a water-methanol mixture (9:1). After threefold extraction with 60 ml petroleum ether each for the removal of chlorophyll, waxes and fats, the remaining fraction was diluted with 60 ml of water. Subsequently, this aqueous solution was extracted three times with 120 ml chloroform each. The collected chloroform layers were washed three times with 360 ml sodium chloride solution (1%). After drying with sodium sulphate, the solution was filtered and the chloroform was evaporated under reduced pressure. The weights of the obtained sub-fractions are listed in Table II. Approximately 10% of the starting extract was lost during the fractionation process.

Thin layer chromatography with chloroform: methanol: water (70:22:3.5) revealed that the petroleum ether fraction (F1) exhibited an almost identical banding pattern as the petroleum ether extract (detection under visible light and UV366 with ASR; data not shown), but with some additional bands, which seemed to be responsible for the higher activity of the F1 fraction (data not shown) compared to the petroleum ether extract.

HL-60 cells were seeded in 24-well plates at a concentration of 1×10⁵ cells/ml allowing logarithmic growth within the next 48 h. Cells were then incubated with increasing concentrations of plant extracts (5 μg/ml, 15 μg/ml, 30 μg/ml, 60 μg/ml) for 48 h. After 24 and 48 h, the cell number was counted using a KX-21 N microcell counter (Sysmex Corporation, Kobe, Japan) and the percent of cell divisions compared to the untreated control were calculated as follows: [(C48h + drug - C24h + drug)/(C48h- drug - C24h - drug)] x 100=% cell division, whereby C48h + drug or C48h - drug were the cell numbers after 48 h with or without extract treatment, respectively. C24h + drug or C24h - drug were the respective cell numbers after 24 h (17,18).

Hoechst 33258 (HO) and propidium iodide (PI) double staining (Sigma, St. Louis, MO) determine the type of death the cell is undergoing, i.e. apoptosis (early or late) or necrosis (19,20). HL-60 cells were seeded in a 24-well plate at a concentration of 1×10⁵ cells/ml and treated with increasing concentrations of fractions F1, F2 and F3. After 24 h, 48 h and 72 h of incubation, 100 μl cell suspension of each well was transferred into separate wells of a 96-well plate and HO and PI were added at final concentrations of 5 μg/ml and 2 μg/ml, respectively. After 1 h of incubation at 37°C, stained cells were examined and photographed on a fluorescence microscope (Axiovert, Zeiss, Jena, Germany) equipped with a DAPI filter. Cell death was evaluated and counted by visual examination of the photographs according to the morphological characteristics revealed by HOPI staining. Experiments were performed in triplicate.

HL-60 were seeded in T-75 tissue culture flasks at a concentration of 1.8×10⁵ cells/ml and treated with the indicated concentration of methanol extract or fraction F2. Cells were harvested after 0.5, 2, 4, 8 and 24 h. Then, cells were washed twice with cold PBS and centrifuged at 1,000 rpm for 5 min at 4°C. The cell pellet was lysed in a buffer containing 150 mM NaCl, 50 mM Tris pH 8.0, 1% Triton-X-100, 1 mM phenylmethylsulfonyl fluoride (PMSF) and 1 mM Protease Inhibitor Cocktail (PIC), (Sigma, Schnelldorf, Germany). The lysate was centrifuged at 12,000 rpm for 20 min at 4°C. Supernatant was transferred into a 1.5 ml tube and stored at −20°C until further analysis. Equal amounts of protein lysate were mixed with SDS (sodium dodecyl sulphate) sample buffer and loaded onto a 10% polyacrylamide gel. Proteins were separated by polyacrylamide gel electrophoresis (PAGE) at 120 Volt and electro-transferred onto a PVDF (polyvinylidene difluoride) membrane (Hybond, Amersham, Buckinghamshire, UK) at 95 Volt for 80 min. Membranes were allowed to dry for at least 30 min up to 2 h to provide fixing of the proteins to the membrane. Methanol was used to remoisten the membranes. Equal sample loading was checked by staining the membrane with Ponceau S (Sigma, Schnelldorf, Germany). After removing Ponceau S with PBS or TBS (Tris buffered saline, pH 7.6), membranes were blocked in PBS- or TBS-milk (5% non-fat dry milk in PBS containing 0.5% Tween-20 or TBS containing 0.1% Tween-20) for 1 h. Then, membranes were washed with PBS/T (PBS containing 0.5% Tween-20) or TBS/T (TBS containing 0.1% Tween-20), changing the washing solution 4–5 times, for at least 20 min. Next, membranes were incubated with the primary antibody in blocking solution (according to the data sheet TBS-, PBS-milk or TBS-, PBS-BSA) diluted 1:500 - 1:1000, gently shaking at 4°C, overnight. Thereafter, membranes were washed again with PBS/T or TBS/T and incubated with the secondary antibody (peroxidase conjugated anti-rabbit IgG or anti-mouse IgG) diluted 1:2000 in PBS- or TBS-milk at room temperature for 1 h. Chemiluminescence was developed by the ECL detection kit (Amersham, Buckinghamshire, UK) and membranes were exposed to Amersham Hyperfilm.

MDA-MB-231 and MCF-7 breast cancer cells were grown in phenol red-free DMEM/F12 culture medium (Invitrogen, Karlsruhe, Germany) supplemented with 10% FCS and 1% PS (Invitrogen, Karlsruhe, Germany) under standard conditions at 37°C in a humidified atmosphere containing 5% CO₂ and 95% air. Twenty-four hours prior to treatment, the cells were transferred to DMEM/F12 culture medium supplemented with 2.5% charcoal-stripped FCS (PAN Biotech, Aldenbach, Germany) and 1% PS. The MeOH extract and F2 were dissolved in DMSO and diluted with medium (final DMSO concentration <0.1%) to 30 and 60 μg/ml. Experiments under each set of conditions were carried out in triplicate. Blanks contained DMSO in the medium of the test compounds. After 18 h of incubation, ethoxyresorufin (final concentration 5.0 μM, Sigma-Aldrich, Munich, Germany) was added and 0.4 ml aliquots of the medium were sampled after 200 min. Subsequently, the formation of resorufin was analyzed by spectrofluorometry (PerkinElmer LS50B, Waltham, MA, USA) with an excitation wavelength of 530 nm and an emission wavelength of 585 nm.

For statistical analyses Excel 2003 software and Prism 5 software package (GraphPad, San Diego, CA, USA) were used. The values are expressed as the mean ± SEM and the Student’s t-test was applied to compare differences between control samples and treatment groups. Statistical significance level was set at p<0.05.

To determine the anti-proliferative effects in HL-60 cells, extracts were applied at increasing concentrations (5, 10, 30, 60 μg/ml and partly 90 and 120 μg/ml) for 24, 48, and 72 h (Fig. 1a–e). The methanol extract was the most potent and inhibited cell proliferation dose-dependently with an IC₅₀ (the concentration inhibiting 50% proliferation) of ∼60 μg/ml. The extract exhibited the highest activity within the first 48 h, which decreased thereafter (data not shown). Although the extract concentration at which proliferation was significantly inhibited was rather high, the corresponding weight of the rhizome was low, as the methanol extract was almost one fifth of the whole lyophilized rhizome substance. This corresponds to 300–500 mg of dried rhizome, or 370–630 mg of fresh rhizome per kg body weight; a person with an average body weight has to consume the alcoholic extract derived from only 20 g of rhizome within two days. Inhibition of cell proliferation was accompanied by a rapid upregulation of Cdc25A expression within 2 h of treatment (Fig. 1f). It seems that the high Cdc25A levels were the result of an increased protein synthesis rather than an inhibited protein degradation, as Ser177 phosphorylation, which tags Cdc25A for recruitment of the proteasome and subsequent proteolysis (21), was sustained throughout 2 and 8 h of treatment, whereas the degradation of the Cdc25A proto-oncogene below control levels was observed after 24 h. The increase in Cdc25A was followed by cyclin D1 upregulation within 4 h. The protooncogene cyclin D1 is necessary for the transit from early G1 to beginning of S phase and induction of cyclin D1 expression is indicative for cell cycle activation. p21 inhibits the Cdk2/Cyclin E kinase complex that normally cooperates with cyclin D1 at a later stage to facilitate G1-S transition and then S-phase progression (22). p21 was induced within 2 h of extract treatment and this accurately counteracted the induction of Cdc25A phosphatase, which causes the activation of Cdk2/Cyclin E. (23,24) and therefore, cell cycle progression was blocked. Additionally, p21 binds to the Cdk4/Cyclin D complex. This results in a hypo-phosphorylation and activation of pRb and thereby the suppression of the E2F pathway and cessation of the cell cycle (25). However, when cyclin D1 became induced, p21 levels were already back to control levels. An additional band below 21 kD became visible, which was most likely a degradation product of p21. Expression of p21 transcription is widely regulated by p53. Since HL-60 cells are p53 deficient (26) p21 must have been controlled in a p53-independent manner (27) and it was shown that the proto-oncogene c-Myc negatively regulates p21 (28). The chaotic expression of prominent cell cycle protagonists and proto-oncogenes together with p21 induction undoubtedly affected DNA replication and cell duplication thereby triggering growth arrest.

Growth arrest due to extra-cellular stressors often elicits apoptosis. Therefore, HL-60 cells were treated with increasing concentrations (60, 90, 120 μg/ml) of the methanol extract to analyse cell viability (Fig. 2a). The time- and dose-dependent increase in the number of dead cells showed a morphology which is typical for apoptosis, whereas a small number of necrotic cells were only observed at the highest dose (120 μg/ml) after 72 h (Fig. 2b). In an attempt to increase the pro-apoptotic activity the methanol extract was subjected to a fractionation procedure (as described in Materials and methods) obtaining three fractions. Fraction 2 (F2) exhibited the strongest pro-apoptotic effect whereby 40% apoptotic cells were observed upon treatment with 90 μg/ml F2 for 24 h (Fig. 3a) and this was a ∼2.5 fold increase compared to the primary methanol extract (<15% apoptotic cells upon treatment with 90 μg/ml methanol extract for 24 h; Fig. 2a). F2 inhibited cell proliferation less efficiently than the original methanol extract (data not shown). Therefore, a cell cycle inhibitory property was separated from a pro-apoptotic property which was less dependent on the cell cycle. This is of particular significance as tumour cells, which are not cycling, could be targeted by this fraction. To get further insight into the mechanisms of F2, the expression of pro-apoptotic proteins as well as of markers indicating genotoxicity and microfilament stress was investigated. The increased cleavage of caspase 9 was observed after 2 h of incubation, whereas caspase 8 was cleaved after 4 h. Finally the executor, caspase 3, was activated after 24 h. The caspase cascade i.e. the activation of caspase 9 and 8, which both cause the cleavage and activation of caspase 3 and the subsequent induction of apoptosis (2) suggested that F2 induced the intrinsic pathway. The activity of caspase 3 was reflected by the cleavage of its target PARP (120 kD) into a smaller 85 kD fragment (29) (Fig. 3b). The phosphorylation of H2AX (γH2AX) is a sensitive and commonly used marker for the presence of DNA-double-strand breaks (30). During this experiment, the phosphorylation of H2AX was induced after 8 and 24 h, whereas caspase 9 was activated before phosphorylation of H2AX implicating that the increase of γH2AX levels was the consequence of caspase 3 activation and the subsequent induction of nucleases causing DNA degradation. Therefore, F2 itself did not induce DNA double strand breaks but this does not exclude the possibility of a genotoxic property of F2 triggering DNA single strand breaks or the generation of DNA adducts.

Several plant compounds were shown to affect the equilibrium of microtubule polymerization. Tilting this fine-tuned equilibrium of polymerized-depolymerized microfilaments is incompatible with normal cell division and causes cell cycle arrest and apoptosis. The acetylation of α-tubulin reflects the polymerization status of the microtubule meshwork (31). However, F2 did not alter tubulin acetylation and therefore did not target the spindle apparatus.

Stat family proteins are transcription factors involved in normal and pathological cellular processes. The Stat3 proto-oncogene product is the most activated Stat protein in human cancers (32) accelerating cell proliferation, preventing apoptosis (33) and playing a role in angiogenesis (34). Stat3, which is phosphorylated at Tyr 705, shifts into the nucleus activating target genes (35) and overexpression of Stat3 was found in leukaemia, breast, pancreatic and prostate cancer, as well as in melanoma (36). In F2-treated HL-60 cells the constitutive Tyr 705 phosphorylation of Stat3 was downregulated after 2 h (Fig. 4) which suggests that the anti-apoptotic activity was also decreased temporally correlating with the activation of caspase 9. Also, the Stat5 proto-oncogene induces anti-apoptotic genes such as Bcl-X (37), thereby maintaining cell survival despite drug-induced stresses (38). However, Stat5 was neither constitutively phosphorylated (activated) nor was the protein level decreased by F2. By contrast, the expression of c-Myc was strongly suppressed after 2 h of F2 treatment and was re-expressed (still below control levels) after 24 h. c-Myc causes an abnormal proliferation rate and is overexpressed in many tumour types and influences cell differentiation and apoptosis (39).

c-Jun and JunB belong to the family of Jun transcription factors, which are components of the activating protein-1 (AP1) transcription factor complexes. AP1 heterodimers are important for cell proliferation, differentiation, and activated c-Jun promotes cell cycle progression and neoplastic transformation (40). Markedly, c-Jun was strongly increased between 2 and 8 h of F2 treatment, whereas JunB expression remained unchanged. The strictly inverse expression of c-Jun and c-Myc upon F2 treatment was most likely independent of each other, as it has not been reported that c-Jun and the AP1 complex suppress c-Myc, nor has it been shown that c-Myc negatively regulates c-Jun. However, when in complex with ATF2 and c-Myc, c-Jun binds to the ATF/CRE site of ATF3 and c-Jun/ATF2/c-Myc induce cell proliferation (41). The dramatic disproportional expressions of c-Myc and c-Jun upon F2 treatment excluded the possibility of a transcriptional active c-Myc/ATF2/c-Jun complex and hence, in case of such a scenario, proliferation was most likely compromised.

Smilacacea species are reported to contain steroidal compounds with estrogenic and anti-estrogenic effects (42,43). A subset of cytochrome P450 (CYP) enzymes are important regulators of estrogen and phyto-estrogen metabolism and CYP1A1 plays a role in estrogen receptor (ER) pathway-dependent synergic carcinogenic action of xeno-estrogens (44). Caucasian individuals with polymorphic CYP1A1 (homozygous for A2455G) stand an increased risk for breast cancer (45,46). Therefore, we analysed the activity of CYP1A1 on S. spinosa extract treatment in ER positive (MCF-7) and ER negative (MDA-MB231) breast cancer cell lines (Fig. 5). In MCF-7 cells the crude MeOH extract weakly induced CYP1A1, whereas CYP1A1 was severely induced by F2. This effect was not observed in MDA-MB231 cells. CYP1A1 inhibition augments LPS-triggered fever, whereas induction of CYP1A1 controls fever and inhibits inflammation (47,48). This seems to depend on the expression of ER and suggests that this remedy is particularly effective in women and could explain its use against internal haemorrhaging during menstruation or after childbirth (14). CYPs are phase 1 enzymes and detoxify xenobiotics and contribute to drug clearance but they also activate pro-carcinogens. Therefore, it has to be considered that the intake of this remedy can cause unwanted interactions with other drugs or environmental (nutritional) compounds.

Among different extracts of increasing polarity the methanol extract of the rhizome of S. spinosa inhibited cell proliferation most significantly, which was associated with the induction of p21. The induction of the proto-oncogenes Cdc25A and cyclin D1 may have counteracted cell cycle arrest, yet they did not prevent it. Further fractionation of the methanol extract increased the apoptotic property (∼2.5 fold at 60 μg/ml), which correlated with the transient inactivation of the proto-oncogene Stat3 and the activation of caspase 9, followed by the induction of caspase 8 and 3. Overexpression of c-Jun did not abrogate, but most likely attenuated apoptosis. Reportedly, the mere overexpression of proto-oncogenes can trigger apoptosis when other co-operating side parameters are limited. The traditional use of ‘Kokolmeka roja’ for many generations proves that the intake of this remedy is safe and the healing properties prevail over potential adverse effects. Although it is generally used against inflammatory ailments, in the present study we have shown that the rhizome of S. spinosa exhibits significant potential as anti-neoplastic concept and should therefore be tested in vivo.