Fruits of F. vulgare subsp. piperitum were collected on the southern slopes of the limestone massif of Rocca Busambra (Corleone, Palermo, Italy). Typical specimens (PAL 109709), identified by Prof. Vincenzo Ilardi, have been deposited in Herbarium Mediterraneum Panormitanum of the ‘Orto Botanico’ (Palermo, Italy). A total of 136 g F. vulgare subsp. piperitum fruits were hydro-distillated for 3 h using Clevenger's apparatus. The oil (yield 1.36%) was dried with Na₂SO₄, filtered and stored in the freezer at −20°C, until the time of analysis. The chemical composition of FVPEO was performed as previously reported (12).

MDA-MB231 TNBC and estrogen-positive MCF7 cells were obtained from ‘Istituto Scientifico Tumori’ (Genoa, Italy) and cultured as monolayers in DMEM medium (cat. no. ECM0749L, Euroclone SpA) supplemented with 10% (v/v) heat-inactivated FCS (cat. no. ECS0180L; Euroclone SpA), 1% non-essential amino acids (cat. no. ECB3054D; Euroclone SpA), 2 mM glutamine (cat. no. ECB3000D, Euroclone SpA) and 1% penicillin/streptomycin solution (cat. no. ECB3001D; Euroclone SpA). Cells were plated on 96-well microplates or on 6-well cell culture plates and allowed to adhere overnight in culture medium at 37°C in a humidified atmosphere containing 5% CO₂, followed by treatment with FVPEO or vehicle only. Media and cell culture reagents were purchased from Euroclone SpA. All other chemicals and reagents were provided by Millipore Sigma.

In order to assess the viability of FVPEO-treated breast cancer cells, an MTT assay was performed, as previously described (19). Briefly, 8×10³ cells/well were plated in 200 µl DMEM in a 96-well plate and treated. At the end, 4 µl MTT solution (5 mg/ml in PBS) were added to the cell medium and the incubation was protracted for 2 h at 37°C in the dark. Mitochondria dehydrogenase activity of viable cells converts MTT to formazan, which is soluble in tissue culture medium. Cells were then lysed in lysis buffer and absorbance was read at 570 and 690 nm using an automatic ELISA plate reader (OPSYS MR; Dynex Technologies).

In order to determine either changes in nuclear morphology or plasma membrane damage, the cells were stained with Hoechst 33342 (cat. no. H3570; Invitrogen; Thermo Fisher Scientific, Inc.), a cell permeant fluorochrome emitting blue fluorescence when bound to dsDNA and excited by ultraviolet light. For these assays, cells (8×10³) were incubated with Hoechst 33342 (2.5 µg/ml medium) for 30 min, washed with PBS and suspended in culture medium prior to FVPEO treatment. Morphological changes in apoptotic cells, manifesting as chromatin condensation and fragmentation, were detected by fluorescence microscopy using an excitation wavelength of 372 nm and emission wavelength of 456 nm. All images were captured by Leica Q Fluoro Software (Leica Microsystems, Inc.). For this analysis, at least 10 fields were considered for each sample and apoptotic cells were counted in random fields at a magnification of ×100.

The apoptotic cell morphology was also studied by acridine orange and ethidium bromide double staining, as reported by Liu et al (20).

The detection of intracellular reactive oxygen species production was carried out using 2′,7′-dichlorodihydrofluorescein diacetate (H₂-DCFDA) staining, as previously described (21). H₂-DCFDA (cat. no. D399; Molecular Probe; Thermo Fisher Scientific, Inc.) is a non-polar dye that can easily cross the cell membrane and can be oxidized to DCFDA in the presence of reactive oxygen species (ROS) emitting green fluorescence.

After incubating the cells with FVPEO, the medium was removed and cells were incubated with 10 µM H₂-DCFDA for 30 min at 37°C. Next, positive cells were analyzed under a Leica fluorescence microscope (Leica Microsystems, Inc.) with excitation at 485 nm and emission at 530 nm, as previously described (22).

For western blot analysis, cells were lysed in RIPA lysis buffer containing 50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.1% sodium dodecyl sulfate, 0.5% sodium deoxycholate, and 1 mM EDTA supplemented with phosphatase inhibitor mix (Sigma-Aldrich; Merck KGaA). Extracts were sonicated thrice and protein content was determined using the Bradford assay (Bio-Rad Laboratories, Inc.) using a BSA (Sigma-Aldrich; Merck KGaA) standard curve. Next, 30 µg/lane of protein sample were subjected to SDS polyacrylamide gel electrophoresis and transferred onto a nitrocellulose membrane. Manganese superoxide dismutase (MnSOD; cat. no. sc-133254; 1:200), c-Jun (cat. no. sc-1694; 1:200), phospho-c-Jun N-terminal kinases (pJNK; cat. no. sc-6254; 1:200), γ-H2A histone family member X (H2AX; cat. no. sc-517348; 1:200), p53 (cat. no. sc-126; 1:200), pro-caspase-3 (cat. no. sc-65497; 1:200) and PARP-1 (cat. no. sc-53643; 1:200) were detected using specific antibodies produced by Santa Cruz Biotechnology, Inc. The antibody for Heme oxygenase (HO-1; cat. no. orb5455; 1:1,000) was provided by Biorbyt Ltd., and that for NQO1 (cat. no. 3187S; 1:1,000) by Cell Signaling Technology, while the nuclear factor E2-related factor-2 (Nrf-2) antibody (cat. no. NBP1-32822, 1:1,000) was purchased from Novus Biologicals. Next, the nitrocellulose filters were incubated with anti-rabbit IgG (H+L) HRP conjugate (cat. no. W4011; 1:5,000; Promega) or Anti-Mouse IgG (H+L) HRP conjugate (cat. no. W4021; 1:5,000; Promega) secondary antibody for 1 h. Protein bands were detected using ECL^™ Prime Western Blotting System (cat. no. GERPN2232; Cytiva), and quantified using Quantity One software 4.6.6 (Bio-Rad Laboratories, Inc.). The correct protein loading was examined using immunoblotting for γ-tubulin (cat. no. T3559; 1:2,000, Sigma-Aldrich; Merck KGaA). All blots shown are representative of at least three separate experiments.

Statistical analysis of data was performed using GraphPad Prism 5.0. software (GraphPad Software Inc.). A Student's t-test was applied to evaluate significant differences between untreated and treated samples. For the analysis of multiple groups a one-way ANOVA test was used. Data are expressed as the mean ± SD. The statistical significance threshold was set at P<0.05.

Hydro distillation of the fruits of F. vulgare subsp. piperitum gave a pale-yellow oil. The EO composition was previously reported (12). As shown in the pie chart of Fig. 1, FVFEO was particularly enriched in monoterpene hydrocarbons (71.01%), with terpinolene (20.10%), limonene (17.84%), α-phellandrene (10.53%) and γ-terpinene (10.43%) as the main components of EO. The second most abundant class was phenylpropanoids (14.36%), typical metabolites of F. vulgare subsp. vulgare (12), with estragole (10.96%) and myristicin (3.09%) as the main products of this class of compounds. Oxygenated monoterpenes were present at a lower amount (10.78%) with fenchone (8.83%) as the principal metabolite of this class. Based on these observations, the anti-tumor potential of FVFEO was explored.

To demonstrate a possible anti-proliferative effect of the EO of the fruits of F. vulgare subsp. piperitum (FVPEO), the present study focused on MDA-MB231, a very aggressive and poorly differentiated breast cancer cell line, which does not express estrogen, progesterone and human epidermal growth factor receptor 2 (HER2)/receptors (23). MDA-MB231 cells were treated with increasing concentrations of FVPEO (125–2,000 µg/ml) for various periods of time, and the viability was assessed using MTT assay, as reported in the Materials and methods. As shown in Fig. 2A, the cell survival rate displayed a marked dose- and time-dependent decrease following FVPEO treatment, as compared to the untreated control. Following 24 h of treatment the viability of MDA-MB231 cells was reduced by 20% of the control with 125 µg/ml FVPEO. Increasing the treatment dose, the viability diminished progressively and a consistent cytotoxic effect was reached at the highest concentration examined (only ~5% of viable cells with 2,000 µg/ml). The cytotoxic effect of FVPEO was further increased as treatment time increased to up to 48 h, while the viability was decreased to 15% with 500 µg/ml FVPEO.

Light microscopy findings showed that, following exposure to FVPEO, MDA-MB231 cells underwent morphological changes. As shown in Fig. 2B, cells treated with lower doses (125–250 µg/ml) of FVPEO appeared elongated, as compared with untreated cells. As the dose of FVPEO increased, typical morphological changes of apoptotic cells (i.e. cell shrinkage and roundness) appeared and a marked reduction in cell number was observed.

A growth inhibition effect was also observed when FVPEO was tested on MCF7 cells, an estrogen- and progesterone-positive breast cancer cell line (Fig. 2). Following incubation with FVPEO, the cell viability declined in a dose- and time-dependent manner, and cells showed clear signs of death, supporting the anti-cancer potential of FVPEO.

Next, it was explored whether the cytotoxic effect of FVPEO was dependent on oxidative stress. To this end, MDA-MB231 cells were pre-incubated for 2 h with NAC, a ROS scavenger; different doses of FVPEO were then added for another 24 h. Our data demonstrated that the addition of NAC counteracted the cytotoxic effect of FVPEO. In particular, as shown in Fig. 3A, 10 mM NAC prevented the cytotoxic effect induced by low concentrations (125–250 µg/ml) of FVPEO and notably reduced that exerted by high concentrations (500–1,000 µg/ml). To better explore these effects, the generation of ROS by H2-DCFDA, a fluorochrome that binds ROS and emits green fluorescence in its oxidized form, was also evaluated. Using this experimental approach, a clear rise in green fluorescence, indicative of ROS production in FVPEO-treated cells, was observed (Fig. 3B). The increase, which had already appeared at 30 min of incubation with 250 and 500 µg/ml FVPEO, peaked at 60 min after application. Western blot analysis was also performed to evaluate whether FVPEO treatment modified the level of MnSOD, one of the main cellular antioxidant enzymes (24). As shown in Fig. 3C, an increased level of MnSOD was observed following FVPEO incubation.

In light of the observed data demonstrating ROS production in MDA-MB231-treated cells, it was explored whether the cytotoxic effect induced by FVPEO can be accompanied by the activation of stress-related proteins (Fig. 4).

First, the level of c-Jun and pJNK was analysed. c-Jun is a member of the activating protein transcription factor that can be activated in response to different extracellular stimuli, such as pro-inflammatory cytokines, ultraviolet radiation and several different forms of cellular stress (25). Its activation has been correlated to the signalling of JNKs, a family of stress-mediated kinases capable of integrating many different cellular stimuli, including mitogenic signals, environmental stresses and different apoptotic insults (26).

The present data provided evidence that FVPEO treatment caused a modest increase in pJNK, but a consistent increase in the c-Jun level that was already visible at 24 h at a dose of 250 µg/ml and further increased with longer periods of incubation.

The involvement of stress in FVPEO-treated cells was also confirmed by the upregulation of NF-E2 p45-related factor 2 (Nrf2), a transcription factor that has been considered one of the main regulatory factors of redox homeostasis that controls a battery of detoxification and cytoprotective genes (27). In our experimental conditions, an increase in Nrf-2 content that was associated to an upregulation of its target genes HO-1 and NQO1 was observed in FVPEO-treated cells.

To investigate whether the loss of viability of MDA-MB231 cells under FVPEO treatment was due to the induction of apoptosis, cells were stained with Hoechst 33342, a fluorescent dye that binds to DNA and can identify nuclear apoptotic changes. According to fluorescence microscopy images, MDA-MB231 cells treated with FVPEO exhibited a clear nuclear fragmentation and condensation, as compared with untreated control cells (Fig. 5A). The proportion of cells with condensed and fragmented nuclei increased as the FVPEO dose increased (Fig. 5B). Such an effect was also confirmed by acridine orange/ethidium bromide dual staining showing the presence of typical morphological features of apoptosis in MDA-MB231-treated cells (Fig. 5C). Indeed, some FVPEO-treated cells were positive for a yellow-green acridine orange nuclear staining (early apoptotic cells), with a considerable proportion of them (~45%) showing a concentrated orange nuclear ethidium bromide staining (late apoptotic cells).

To further explore the underlying mechanism of FVPEO-induced apoptosis, it was examined whether the observed effects and association with DNA injury can be linked to the recruitment of DNA damage markers. When a double-strand break occurs in DNA, alteration in chromatin structure promotes the phosphorylation of the histone variant H2AX at the Ser-139 residue, a form known as γH2AX (28). This event is induced by the kinases ataxia-telangiectasia mutated, ataxia-telangiectasia mutated and Rad3-related protein and DNA-dependent protein kinase, allowing the formation of H2AX phosphorylated on serine 139, which is considered a marker of DNA damage (29). The present data showed that the treatment of MDA-MB231 cells with FVPEO induced a strong phosphorylation of H2AX at Ser139 in a dose-dependent manner (Fig. 6).

Next, possible changes in the level of p53 protein, a key factor involved in the induction of apoptosis in response to DNA damage, was examined (30). As shown in Fig. 6, the level of p53 markedly increased in MDA-MB231-treated cells compared with the untreated control.

In the present study, the effect of FVPEO on caspase-3, a key mediator of apoptosis of mammalian cells, whose activation by the cleavage of procaspase-3 is responsible for chromatin condensation, was also analysed (31). These results indicated that FVPEO treatment induced a dose-dependent decrease in the inactive procaspase-3, indicating the activation of caspase-3 (Fig. 6). This suggestion was confirmed by the cleavage of PARP-1, a well-known target of caspase-3 (32), which was observed following FVPEO treatment. In combination, these data indicated that FVPEO induced caspase-dependent apoptosis triggered by DNA damage.