Effects of bergamot essential oil and its extractive fractions on SH-SY5Y human neuroblastoma cell growth

Abstract

Objectives

The goals were to investigate the mechanisms underlying the antiproliferative effects of bergamot essential oil (BEO) and to identify the compounds mainly responsible for its SH-SY5Y cells growth rate inhibition.

Methods

Five BEO extractive fractions (BEOs) differing in their chemical composition were used. Cell proliferation was determined by 3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and cell count assays. Trypan blue exclusion test and Annexin V/PI staining were performed to assess their cytotoxic activity. Genotoxicity was detected by comet assay. The cell cycle was checked cytofluorimetrically. Reactive oxygen species (ROS) and Δψm were measured fluorimetrically. Western blotting analyses for some apoptosis-related proteins were carried out.

Key findings

Treatment of SH-SY5Y cells with some types of BEOs decreased cell growth rate by a mechanism correlated to both apoptotic and necrotic cell death. Coloured BEOs act by increasing ROS generation, responsible for the drop in Δψm, and modulate p38 and extracellular signal-regulated kinases (ERK ½) mitogen-activated protein kinases, p53, Bcl-2 and Bax signalling pathways. Finally, we identify bergamottin and 5-geranyloxy-7-methoxycoumarin as the bioactive molecules that could play a pivotal role in the antiproliferative effects exerted by coloured BEOs.

Conclusions

Our study provides novel insights into the field of the antiproliferative effects of BEO, which could be exploited in the context of a multitarget pharmacological strategy.

antiproliferative activity, bergamot essential oil, Citrus bergamia, complementary and alternative medicine, SH-SY5Y cells

Introduction

Bergamot essential oil (BEO) is obtained by rasping the fruit peel of Citrus bergamia Risso & Poiteau (bergamot), a small tree belonging to the Rutaceae family. Bergamot is cultivated almost exclusively along the southern coast of the Calabria region (Italy). Bergamot essential oil is very appreciated for its organoleptic properties and is be widely used in the manufacture of perfumes. It also exerts biological activities upon the cardiovascular system,^[1] possesses antimicrobial activity against a number of common pathogens,^[2–4] shows neuropharmacological effects^[5,6] and in aromatherapy is used to minimize symptoms of stress-induced anxiety and mild mood disorders and cancer pain.^[7,8] Bergamot juice (BJ), obtained by squeezing the fruits, is seldom employed by the food industry. Recently, we showed that BJ is able to inhibit important molecular pathways related to cancer-associated aggressive phenotype, thus reducing cell growth, adhesion and migration in different in-vitro^[9] and in-vivo models.^[10] In addition, we demonstrated that flavonoid fraction of BJ (BJe) inhibits proliferation and induces apoptosis in human colon cancer cells.^[11] Also, low concentrations of BJe reduces bacterial lipopolysaccharide (LPS)-induced inflammatory response in THP-1 monocytes, through Sirtuin 1 (SIRT 1)-mediated Nuclear Factor-kappa B (NF-κB) inhibition,^[12] and exerted anti-inflammatory and antioxidant effects also in an animal model of inflammatory bowel disease.^[13]

Several reports have demonstrated that natural compounds possess a broad spectrum of biological activities widely used in traditional, folk and alternative medicine. Thus, many drugs developed from plant sources have been extensively studied, clinically tested, and some of these received clinical approval or are currently enrolled in clinical trials.^[14,15] In particular, in the field of cancer, many of the drugs are natural or semisynthetic products.^[16] These include vinblastine, vincristine, the camptothecin derivatives, topotecan and irinotecan, etoposide, derived from epipodophyllotoxin and paclitaxel, while some agents which failed in clinical studies are stimulating renewed interest.^[17] In this contest, essential oils have drawn the attention of researchers taking advantage of the fact that molecules that constitute the phytocomplex may have antitumour property.^[18,19] Essential oils are phytocomplexes composed mainly by terpenes and cumarins, two classes of non-nutritive dietary components exhibiting important biological properties including anticancer activity.^[20–23]

Monoterpenes are plants' secondary metabolites contributing to their flavour and aroma. Chemically, they are a class of aromatic hydrocarbons (terpenes) derived from the condensation of two isoprene units. Experimental studies performed both in vitro and in vivo have demonstrated that some monoterpenes possess anticarcinogenic properties, acting as chemopreventive as well as chemotherapeutic agents. It is well known that monoterpenes such as citral, d-limonene, β-myrcene or perillyl alcohol induce the glutathione-S-transferase and specific cytochrome P450 isozymes.^[24–27] Moreover, monoterpenes may stop cancer at the progression stage through several cellular and molecular activities such as the block of cell cycle leading apoptosis.^[20]

Coumarins are a group of plant-derived polyphenolic compounds belonging to the benzopyrones family. They occur in fruits, vegetables, olive oil, wine and beverages like tea and coffee. The furanocoumarins, or furocoumarins, consist of a furan ring fused with coumarin. Both coumarins and furocoumarins are generally distributed throughout the Citrus species, exspecially in the peel oils. Psoralen, such as 5-methoxypsoralen (5-MOP; bergapten), 8-methoxypsoralen (8-MOP) and bergamottin, are linear furocoumarins very abundant in Citrus sp, while citropten and 5-geranyloxy-7-methoxycoumarin are the most representative. Coumarins and coumarin-related compounds are reported to have a broad spectrum of biological activities, including antimicrobial, antiplatelet aggregation, antimutagenic and anticancer activities.^[28]

Recently, we^[29,30] and other authors^[31] have shown that BEO reduced survival and proliferation of SH-SY5Y neuroblastoma cells by the activation of multiple pathways leading to both necrotic and apoptotic cell death. Although Russo et al.^[32] have attempted to identify the compounds responsible for BEO antiproliferative effects, the issue has not yet been completely understood. Therefore, our study was designed to identify the compounds mostly responsible for the growth rate inhibition induced by BEO in SH-SY5Y cultures. To this objective, we used five BEO extractive fractions differing in their chemical composition that have allowed us to identify the compounds that could be mainly responsible for the antiproliferative effect of BEO.

Materials and Methods

Drug

Bergamot essential oil and its extractive fractions were kindly provided by ‘Simone Gatto’ (Messina, Italy), a company that processes genuine bergamot oils obtained from the ‘Consorzio del Bergamotto’ of Reggio Calabria (Italy). In this study, five different types of BEOs were used: cold-pressed (BEO); terpeneless coloured (BEO-TF); terpeneless colourless (BEO-TFi); furocoumarin-free colourless (BEO-FFi); furocoumarin free (BEO-FF). Their chemical composition has described by Costa R. et al.^[33] and succinctly reported in Table 1.

Table 1

Bergamot essential oils used in this study and their summarized chemical composition (from Costa et al., Flav. Fragr. J., 2010)

	BEO	BEO-FF	BEO-TF	BEO-FFi	BEO-TFi
	Content of coumarins and psoralens (g/l)
citropten	1.92	traces	6.13	traces	–
bergapten	2.07	traces	4.21	traces	–
bergamottin	21.68	18.19	39.20	traces	–
5- Geranyloxy-7-methoxycoumarin	1.42	1.29	2.82	traces	–
	Volatile constituents (%)
β-pinene	5.08	4.62	traces	4.85	traces
limonene	38.89	37.40	traces	36.76	traces
γ-terpinene	5.62	5.25	traces	4.78	traces
linalool	6.55	11.86	20.3	11.91	16.79
linalyl acetate	37	34.02	72	34.57	77

BEO, bergamot essential oil; BEO-FF, bergamot essential oil- furocoumarin free; BEO-FFi, bergamot essential oil-furocoumarin-free colourless; BEO-TF, bergamot essential oil-terpeneless coloured; BEO-TFi, bergamot essential oil-terpeneless colourless.

Cell culture and treatments

Experiments were performed on the SH-SY5Y human neuroblastoma cell line, obtained originally from ATCC (Rockville, MD, USA). The cultures were grown as reported.^[5] SH-SY5Y cells were seeded onto 96-well plates at a density of 5 × 10³ cells/well for the 3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyltetrazolium bromide (MTT) test or at 1.5 × 10⁴ cells/well for the reactive oxygen species (ROS) and mitochondrial membrane potential (Δψ_m) determinations. For the cell count assay, the cytofluorimetric analyses and the comet assay, the cells were seeded onto six-well plates (1 × 10⁵ cells/well). The next day, the media were replaced with fresh medium (untreated cultures) or with medium containing increased concentrations of BEOs ranging from 0.01–0.03% (0.1–0.3 μl/ml), and the cells were incubated for 24, 48 and 72 h (proliferation assays) or for 24 h (for the other assays). Bergamot essential oil and its extractive fractions were diluted 1 : 1 in a 1 : 9 water/dimethylsulfoxide (DMSO) solution and then further diluted in culture media to obtain the final concentrations reported above. The same DMSO concentrations used to dissolve BEOs to the final concentration of 0.01, 0.02 and 0.03% served as vehicle controls (0.009, 0.018, 0.027% DMSO in culture media, respectively). In comparison with untreated cultures, DMSO 0.027% (1 : 3703 dilution) did not exert significant influence on any parameters analysed in this study (data not shown). Therefore, the experimental data showed in this study refer to those obtained in untreated cultures assumed as control.

Proliferation assays

Growth rate of SH-SY5Y cells was evaluated by both MTT and cell count tests. The MTT assay was performed as reported^[34] with modifications. Cells were seeded and treated as described above. After 24, 48 and 72 h of incubation with the BEOs, the plates were centrifuged, and the supernatants were replaced with 100 μl of fresh medium without phenol red containing 0.5 mg/ml of MTT. After 4 h, crystals of formazan were solubilized by HCl/isopropanol 0.1 N buffer and quantified spectrophotometrically at a wavelength of 570 nm (reference at 690 nm).

Cell growth was also detected by the cell counted assay^[35] by plating and treating the cells as already described.

Cytotoxicity and genotoxicity assays

Cytotoxic effect induced by BEOs was evaluated in term of cell death assessed after 24 h of treatment by the cell exclusion of trypan blue assay (0.4% w/v).^[36]

Furthermore, the genotoxic effect of BEOs was assessed by detecting deoxyribonucleic acid (DNA) damage by the comet assay.^[37] The considered parameters were tail length, percentage of DNA in the head (HDNA%), percentage of DNA in the tail (TDNA%), tail moment and olive tail moment. Both cytotoxic and genotoxic assays were performed on the SH-SY5Y cells plated and treated for 24 h as previously described.

Fluorescence-activated cell sorting analysis of apoptosis and cell cycles

SH-SY5Y cells were seeded and exposed to BEOs for 24 h as previously described. Annexin V fluorescein isothiocyanate/propidium iodide (PI) staining was performed to detect both apoptotic and necrotic cell death as previously described.^[9] Cytofluorimetric studies were also used to assess the DNA content for the evaluation of cell cycle, by using the CycleTEST PLUS DNA reagent (BD Biosciences, Milan, Italy).

Spectrofluorimetric determination of oxidative stress biomarkers

Intracellular production of ROS and mitochondrial membrane potential (Δψm) were measured fluorimetrically. Accumulation of ROS was estimated using 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA, Sigma-Aldrich, Milan, Italy), as previously described.^[5] The cells were seeded onto 96-well plates and the next day treated with BEOs. Fluorescence was monitored every 15 min for 90 min at 485 nm excitation and 535 nm emission.

Change in Δψm, as a result of mitochondrial perturbation, was evaluated by measuring the fluorescence of incorporated rhodamine 123 (R123; Invitrogen, Life Technologies). The fluorochrome (R123 10 μm) was added to cells treated or not with BEOs for 24 h and left for an additional 20 min at 37°C. Fluorescence was captured at 535 nm excitation and 595 nm emission.

Western blot analysis

The SH-SY5Y cells (1 × 10⁶) were seeded in a 100 mm plate with complete media overnight and then were treated with 0.02% of BEO, BEO-TF or BEO-FF for 24 h. Control cells received only complete fresh media. Total proteins were extracted from the cells as previously described.^[38] Proteins (30 μg/lane) were separate on 10% Sodium Dodecyl Sulphate-PolyAcrylamide Gel Electrophoresis (SDS-PAGE), electro-transferred on polyvinylidene difluoride membranes. The latter were incubated 1 h with 5% non-fat milk to block non-specific binding sites and then incubated overnight at 4°C with the following antibodies: mouse monoclonal antiphospho p44/42 mitogen-activated protein kinase (MAPK) (Thr 202/Try 204) and anti-p44/42 MAPK (Cell Signaling Technology, Beverly, MA, USA); rabbit monoclonal anti-phospho p38 and rabbit polyclonal anti-p38 (AbCam, Cambridge, UK); mouse monoclonal anti-Bax (Thermo Fisher Scientific, Waltham, MA, USA); mouse monoclonal anti-Bcl-2 (Thermo Fisher Scientific); mouse monoclonal anti-p53 and rabbit polyclonat anti-phospho-p53 (AbCam); rabbit polyclonal anti-β-actin (Sigma-Aldrich). Membranes were incubated with peroxidase-conjugated goat anti-rabbit or anti-mouse immunoglobulin G (IgG) secondary antibody (Jackson ImmunoResearch, West Grove, PA, USA) for 2 h at room temperature. Protein bands were visualized using a chemiluminescent detection system (SuperSignal West Pico Chemiluminescent Substrate; Thermo Fisher Scientific).

Statistical analysis

Data were expressed as mean ± SEM and statistically evaluated for differences using one-way analysis of variance, followed by Turkey–Kramer multiple comparison test (GrafPAD Software for Science, San Diego, CA, USA). P values less than or equal to 0.05 were considered significant.

Results

Bergamot essential oils reduce SH-SY5Y cell proliferation

Both MTT test and the cell count assay were performed to determine cell proliferation. As shown in Figure 1, all the coloured BEO extractive fractions (BEO, BEO‐TF and BEO‐FF) exerted concentration and time‐dependent growth inhibitory activity. In particular, as assessed by the MTT assay, BEO‐TF produced the greater antiproliferative effect, as it reaches statistical significance already after 24 h of incubation with the 0.02% concentration (P < 0.05 vs the untreated cultures; Figure 1a). Moreover, BEO‐TF inhibits the SH‐SY5Y growth rate in a statistically significant manner also after exposure for 48 and 72 h at the 0.01% concentration (P < 0.05 and P < 0.01 vs the untreated cultures, respectively; Figure 1a). Although with less efficacy, also BEO and BEO‐FF decreased cell proliferation in a statistically significant way (Figure 1a). On the contrary, colourless extracts (BEO‐TFi and BEO‐FFi) did not exert significant effects on cell proliferation (Figure 1a).

Figure 1

Antiproliferative effects of bergamot essential oils (BEOs) on SH-SY5Y cell line. SH-SY5Y were exposed to BEOs (0.01–0.03%) for the indicated times. Proliferation rate was assessed by 3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay (a) and cell count (b). MTT results are expressed as percentages ± SEM of optical density in treated cells. Each concentration was eightfold tested, and three independent experiments were carried out. Data from cell count are expressed as mean ± SEM of three independent experiments performed in triplicate. *P < 0.05, **P < 0.01 and ***P < 0.001 vs control, respectively (analysis of variance followed by Tukey–Kramer multiple comparisons test).

3‐(4,5‐Dimethylthiazole‐2‐yl)‐2,5‐diphenyltetrazolium bromide data were established by counting the cells in a Neubauer hemocytometer chamber (Figure 1b).

Bergamot essential oils induce cytotoxic effects in SH-SY5Y cells

The trypan blue dye exclusion assays was the first test performed to evaluate the cytotoxic effects induced by BEOs. Figure 2a–c shows that 24 h treatment of SH-SY5Y cells with increasing concentration of the coloured BEOs (0.01–0.03%) induced significant rise in cell death starting from the 0.02% concentration (P < 0.01 vs the untreated cultures). A dramatic increase in cell death was observed when the cultures were treated with the 0.03% of BEO, BEO-FF or BEO-TF (about 70%; P < 0.001). On the contrary, exposure of SH-SY5Y cells to BEO-TFi and BEO-FFi did not cause cell death (data not shown).

Figure 2

Cytotoxic effects of bergamot essential oils (BEOs). Treatment of SH-SY5Y cells with increasing concentration of BEOs (0.01–0.03%) for 24 h induces cytotoxic effects. (a) Data from trypan blue test are expressed as the percentage of stained cells (death cells) vs total cell counted, and are the mean ± SEM of three experiments performed in triplicate. **P < 0.01 and ***P < 0.001 vs control (analysis of variance followed by Turkey–Kramer multiple comparisons test). (b) In the panel are reported the comet assay images captured by fluorescence microscopy after 24 h with treatment of 0.03% of BEOs. A representative experiment that was replicated three times with similar results is shown. The images display rounds and intact nucleus in untreated cells and in both BEO-furocoumarin-free colourless (FFi) and BEO-terpeneless colourless (TFi)-treated cells. Otherwise cells exposed to BEO, BEO – furocoumarin free (FF) and BEO-terpeneless coloured (TF) showing DNA fragments in the tail. TDNA: % tail DNA ± SEM.

Moreover, results of comet assay suggested that BEO, BEO‐FF or BEO‐TF at the higher concentration used in this study (0.03%) were able to induce DNA injury as demonstrated by the presence of a high percentage of DNA fragments in the tail (Figure 2d). Instead, data collected from the cells treated with the 0.02% coloured BEOs showed a slight DNA damage, while the 0.01% concentration did not exert any genotoxic effect (data not shown). On the contrary, both BEO‐FFi and BEO‐TFi lacked genotoxicity, as indicated by the rounds and intact nucleus displayed.

Bergamot essential oils alter SH-SY5Y distribution within cell cycle and cause apoptosis

In order to detect the mechanisms underlying the reduction of cell growth exerted by the coloured BEOs, we examined the progression through the cell cycle by flow cytometry. Data analysis showed that BEO, BEO‐TF and BEO‐FF arrest the SH‐SY5Y cell cycles in G₀/G₁ phase, causing a decrease of cell population in G₂/M phase (Table 2). In parallel, we observed the increase in the hypodiploid events, indicative of DNA fragmentation, corresponding to the cells in the sub‐G₀ phase (Table 2). Conversely, the colourless BEO‐TFi and BEO‐FFi did not cause important perturbation of the cell cycles (data not shown).

Table 2

Cell cycle analysis of SH-SY5Y cells treated with colored BEOs. Fluorescence-activated cell sorting analysis (FACS) analysis show that BEO, BEO-TF and BEO-FF block the cells in G₀/G₁ phase and increase cellular population in sub-G₀ phase. Results are represented as percentages of the total cell population and are the mean ± SEM of three independent experiments

Cycle phase	Control	0.01	0.02	0.03
			BEOs (%)
BEO
Sub G0	–	–	14 ± 0.3%	53 ± 5.9%
G0/G1	65 ± 5.8%	70 ± 5.1%	75 ± 6.6%	41 ± 2.4%
S	20 ± 2.1%	18 ± 1.1%	7 ± 1.6%	4 ± 0.3%
G2/M	15 ± 1.3%	12 ± 1.9%	4 ± 1.1%	2 ± 1.3%
BEO-TF
Sub G0	–	–	16 ± 0.2%	57 ± 3.3%
G0/G1	63 ± 6.4%	68 ± 7.1%	71 ± 7.6%	32 ± 7.0%
S	23 ± 3.2%	18 ± 2.2%	8 ± 2.4%	8 ± 2.1%
G2/M	14 ± 0.7%	14 ± 1.4%	5 ± 0.6%	3 ± 1.1%
BEO-FF
Sub G0	–	–	11 ± 0.4%	48 ± 4.3%
G0/G1	65 ± 5.6%	69 ± 4.1%	73 ± 7.1%	40 ± 3.1%
S	23 ± 2.1%	20 ± 2.1%	12 ± 1.2%	8 ± 1.1%
G2/M	12 ± 0.9%	11 ± 0.6%	4 ± 0.7%	4 ± 0.3%

BEO, bergamot essential oil; BEO-FF, bergamot essential oil-furocoumarin free; BEO-TF, bergamot essential oil-terpeneless coloured; Gap 0 (G₀), Gap 1 (G₁), DNA synthesis phase (S), Gap 2 (G₂) and mitosis phase (M).

The cytotoxic activity of coloured BEOs was also assessed by cytofluorimetric analysis using Annexin V/PI staining, an assay that allowed us to discriminate dead cells to apoptosis or necrosis. As shown in Figure 3, exposure of SH‐SY5Y cells to coloured BEOs caused the reduction of viable cells and the increase in both necrotic and apoptotic cells compared with control cultures. At 0.02% concentration of BEO, BEO‐TF and BEO‐FF, in comparison to control cultures, the viable cells (Annexin V‐/PI‐) were greatly reduced, and both apoptotic and necrotic cells increased considerably. At 0.03% concentration of colored BEOs the rate of viable cells has been further reduced, while dead cells rose dramatically. In particular, after 24 h of incubation with BEO, the population in early apoptosis (Annexin V+/PI‐) increased to 12% of total cells. The events in late apoptosis (Annexin V+/PI+) were 30%, and the necrotic cells (Annexin V‐/PI+) were 30%. Similar results were observed using BEO‐TF (early apoptosis: 10%; late apoptosis: 35%; necrosis: 36%) and BEO‐FF (early apoptosis: 8%; late apoptosis: 20%; necrosis: 30%). At 0.01% concentration there was not evident effect on cell viability (Figure 3), and the exposure to BEO‐TFi and BEO‐FFi did not cause any change in viable, necrotic and apoptotic cells at any concentration (data not shown).

Figure 3

Cytofluorimetric evaluation of apoptosis. The SH-SY5Y cells were exposed to bergamot essential oils (BEOs) for 24 h in a range of 0.01–0.03%. Data are presented as percentages of events Annexin-V and PI negative (viable cells), Annexin V positive and PI negative (early apoptotic cells), Annexin V and PI positive (late apoptotic cells) and Annexin V negative and PI positive (necrotic cells). The results are representative of three independent experiments. A great increase in the proportion of both apoptotic and necrotic events was observed in the cells treated with BEO, BEO-furocoumarin free (FF) and BEO terpeneless coloured (TF).

Apoptosis-related proteins modulated by bergamot essential oils

In the aim to investigate the intracellular pathways involved in BEOs‐induced activation of programmed cell death, we evaluated the expression of proteins involved in the apoptotic machinery by western blot. As presented in Figure 4, exposure of SH‐SY5Y to 0.02% BEO, BEO‐TF or BEO BF for 24 h produced activation of p53 by phosphorylation, increased levels of the pro‐apoptotic Bax and reduced the anti‐apoptotic protein Bcl‐2. In the same experimental condition, we also observed a reduced phosphorylation of p38 and extracellular signal‐regulated kinases (ERK 1/2) MAPK as compared with the untreated cultures (Figure 4). The results obtained with the highest concentration of coloured BEOs used in this study (0.03%) are similar to those obtained with the 0.02%, while 0.01% has had no effect on the apoptosis‐related proteins (data not shown).

Figure 4

Effect of colored bergamot essential oils (BEOs) on mitogen-activated protein kinases (MAPKs) and apoptosis-related protein in SH-SY5Y cells. The cells were treated with 0.01–0.03% of BEO, BEO-furocoumarin free (FF) and BEO terpeneless coloured (TF) for 24 h and then subjected to immunoblotting as described in methods. Three independent experiments were carried out. A representative immunoblot image of homogenate from cells exposed to 0.02% BEOs is shown.

Bergamot essential oils increase reactive oxygen species production and induce alteration of mitochondrial transmembrane potential

To further elucidate the mechanisms underlying the BEOs antiproliferative activity, we measured the ROS production in a time‐course experiment. Exposure of SH‐SY5Y to 0.02% coloured BEOs for 15–90 min increased ROS generation over time in a statistically significant way. Results shown in Figure 5a reported data collected after 90 min of exposure (+44% for the BEO and BEO‐FF or +37% for the BEO‐TF; P < 0.01 vs untreated cultures). The intracellular ROS production following incubation with 0.03% BEO, BEO‐TF or BEO‐FF increased even more up to about 80% of rise compared with the control without significant differences among them (Figure 5a). The exposure to coloured BEOs at a concentration of 0.01% (Figure 5a), as well as the BEO‐TFi and BEO‐FFi at any concentration (data not shown) did not cause significant ROS generation.

Figure 5

Oxidative stress and mitochondrial impairment induced in SH-SY5Y cells by colored BEOs. (a) Cells were treated with BEO, BEO-furocoumarin free (FF) and BEO terpeneless coloured (TF) and reactive oxygen species (ROS) production was measured by the probe 2′,7′-dichlorodihydrofluorescein diacetate throughout the 15–90 min followed the addition of BEOs. The graphic show data collected after 90 min of exposure. (b) Changes in Δψm was assessed by means of the cationic fluorochrome R123. Data of both ROS and Δψm determinations are reported as percentage of the levels detected in control cultures and reported as the mean ± SEM of eight wells per experimental group (n = 3). *P < 0.05, **P < 0.01 and ***P < 0.001 vs untreated cultures.

Rise of ROS production by BEO, BEO‐TF or BEO‐FF was accompanied by a fall in Δψm (Figure 5b). Indeed, concentration of 0.03% of coloured BEOs reduced Δψm of SH‐SY5Y cells by about 40% after 24 h of exposure (P < 0.01 vs untreated cultures; Figure 5b). A milder but significant Δψm reduction (about 20%; P < 0.05) was also observed following incubation with the 0.02%. No evident alteration in mitochondrial integrity was observed when the cells were exposed to colourless BEOs.

Discussion

This study shows the effects of five BEO extractive fractions made with different chemical compositions on SH-SY5Y neuroblastoma cells. First, we demonstrated that BEO, BEO-TF and BEO-FF reduced cell proliferation by a cytotoxic action mediated by both apoptotic and necrotic cell death. This issue was partially known since Berliocchi et al.^[31] already documented the toxic profile of bergamot essential oil on survival and proliferation of SH-SY5Y cells. Our data are substantially in agreement with Berliocchi's study, but the authors did not find increased ROS production.^[31] On the contrary, our experiment showed a significant increase in ROS generation after treatment with BEO at 0.02% and 0.03% concentration. Interestingly, our data are in agreement with a recent study that reported the ability of BEO to stimulate the production of ROS, albeit in another cell type.^[39] Moreover, the peculiarity of our study is that we used five extractive fractions of BEO with different chemical compositions that allowed us to identify the compounds mainly responsible for the antiproliferative effect of BEO. Experiments performed with BEO, BEO-FF and BEO-TF showed that all the coloured BEOs reduced SH-SY5Y growth rate with the same strength, while BEO-TFi and BEO-FFi were lacking any cytotoxicity and antiproliferative effect. Also, colourless BEOs were unable to produce genotoxicity, block cell cycle and induce apoptosis, whereas 0.02% and 0.03% concentration of coloured ones determined DNA damage and loss of cell viability due to a dramatic increase of both apoptotic and necrotic cells. However, up to 0.01% concentrations of any BEOs exerted no effects on the cell parameters tested in this study (data not shown).

Apoptosis aims to prevent proliferation of malignant transformed cells thus playing an important role in cancer. Along this line, almost all tumour cells lacking the programmed cell death as well as carcinogenesis is enhanced when apoptosis is missing.^[40] Therefore, apoptosis is considered an important goal of many anticancer chemotherapeutic drugs. The apoptotic process is closely regulated by several factors, including tumour suppressor and inducer genes such as the Bcl-2 family proteins which could both promote survival of malignant cells, conferring resistance to chemotherapy (Bcl-2 and Bcl-XL) or induce apoptosis (Bax and Bad).^[41] The imbalance of pro- and anti-apoptotic Bcl-2 family proteins leads to the loss of Δψm which in turn determines mitochondrial outer membrane permeabilization with consequent release of pro-apoptotic factors into the cytosol. For that reason, dissipation of Δψm may be considered an objective of chemotherapeutic agents,^[42] and inhibition of pro-survival Bcl-2 proteins is considered a promising avenue for cancer therapy.^[43] On the other hand, there is growing evidences that high levels of ROS can cause apoptosis by triggering mitochondrial permeability transition pore opening and release of pro-apoptotic factors.^[44] Many reports have revealed that diverse cancer chemotherapeutic agents targeting ROS metabolism can kill cancer cells by raising the oxidant stress beyond the toxic threshold.^[45,46] Since cancer cells show higher levels of endogenous ROS compared with normal cells, the toxic limit can be easily achieved in cancer cells.^[45,47] On the other hand, ROS accumulation is also the result of a dysfunction in the mitochondrial respiratory chain produced by the mitochondria and can be involved in cell death.^[48] Reactive oxygen species is the mediator of intracellular signalling cascades and some apoptotic key regulators, such as proteins of the Bcl-2 family.^[49]

In our study, we demonstrated that in SH-SY5Y cells high concentrations of BEO, BEO-TF or BEO-FF increase intracellular ROS generation, which contributes to determining the drop in Δψm. Of note, 0.01% concentration of coloured BEOs do not alter ROS production and slightly increase the Δψm. These observations, along with the results of cytotoxic assays, suggest that the cytotoxic activity of BEO, BEO-TF or BEO-FF occurs only at the higher concentration. Moreover, western blot analysis showed that coloured BEOs may modulate important signalling pathways playing a roles in apoptosis. In particular, they (1) reduced the phosphorylation of both p38 and Erk 1/2 MAPKs, (2) increased both the phosphorylation of p53 and the levels of the pro-apoptotic Bax and (3) decreased the anti-apoptotic protein Bcl-2. Therefore, we hypothesized that BEO, BEO-TF or BEO-FF caused apoptosis of SH-SY5Y cells by both interfering with the Bcl-2 family proteins and increasing ROS production. These two events in turn caused a drop in Δψm that contributes to supporting the apoptotic process. However, extensive mitochondrial damage caused by the highest concentrations of coloured BEOs may reduce energy production, thereby hindering the apoptotic execution and switching cell death towards a necrotic phenotype. The pivotal role exerted by MAPKs in the initiate and progression of cancer is well known. Depending on the cell type, stimuli and the latency of the activation of MAPKs, they may either protect or enhance sensitivity to apoptosis.^[50] Here, we demonstrated that coloured BEOs reduce phosphorylation of Erk1/2-MAPK, suggesting their involvement in the SH-SY5Y apoptosis triggered by the phytocomplexes. Interestingly, evidence that coloured BEOs reduce p38 phosphorylation could support the hypothesis of their anti-angiogenic potential.^[51]

Mitochondria are both major cellular sources and targets of free radicals, as well as effectors of the intrinsic apoptotic pathway, thus playing a pivotal role in several pathophysiological conditions, such as neurodegeneration or cancer, where hormetic stimulation of the vitagene pathway is strongly warranted. Vitagenes are genes involved in the cellular defensive system in response to stressful conditions, encoding for heat shock proteins (Hsp) Hsp32, Hsp70, the thioredoxin/thioredoxin reductase (TrxR) and the sirtuin protein systems.^[52] In recent years, it has been shown that some phytochemicals have the ability to modulate the vitagene pathway, generally by activating it, thus conferring protection against oxidative stress.^[53] Interestingly, it has been proposed that the cancer chemopreventive activity of Curcumin, at least in part, may be due to its ability to irreversibly inhibit TrxR activity, which in turn strongly increases reduced nicotinamide adenine dinucleotide phosphate (NADPH) oxidase activity leading to a high ROS production^[54] to which cancer cells appear more sensitive.^[45,46]

We believe that the complex phytochemical composition of BEOs allows them to interact with different cellular targets that contribute in various ways to the biological activity. At the same time, this makes it difficult to identify one or more bioactive molecules mainly responsible for the biological effect.

The second goal of the study was to identify the compounds mainly responsible for the antiproliferative effect of BEOs. It comprises more than 345 compounds, divided in a volatile fraction (93–96%) containing monoterpene and sesquiterpene hydrocarbons, such as limonene, α- and β-pinene, β-myrcene, γ-terpinene, linalool and linalyl acetate and a non-volatile fraction (4–7%) consisting of polymethoxylated flavones, coumarins and psoralens, such as bergamottin and bergaptene.^[33] As shown in Table 1, the five BEO extractive fractions used in our experiments differ in the presence or absence of some phytochemicals. Since only BEO, BEO-TF and BEO-FF reduced the growth rate of SH-SY5Y cells, the bioactive molecules involved in this action must be present in all the three active BEOs and absent or poorly present in BEO-TFi and BEO-FFi which have not shown antiproliferative activity. The most abundant compounds found in BEO are d-limonene and linalyl acetate, which together account for about 75% of the whole oil. Of interest, both molecules exhibited growth inhibition of cancer cells.^[55,56] Recently, Russo and co-workers^[32] attempted to identify the components of BEO involved in SH-SY5Y cell death, individually testing d-limonene, linalyl acetate, linalool, γ-terpinene, β-pinene and bergapten. None of these compounds caused cell death when used alone, but significant cytotoxicity was found after a co-treatment with limonene and linalyl acetate. Moreover, the addition of linalool did not have further effects on cell viability, thus suggesting involvement of other components that might participate in the cellular demise induced by BEO.^[32] In our study, we demonstrated that BEO, BEO-FF and BEO-TF reduced the SH-SY5Y growth rate with the same strength and by the same molecular mechanisms. On the contrary, BEO-TFi and BEO-FFi were lacking any cytotoxicity and antiproliferative effect. Chemical composition of the five BEO extractive fractions illustrated in Table 1 show that both limonene and linalyl acetate are present in BEO-FFi that do not exert any cytotoxic effect on SH-SY5Y cells. Therefore, based on our results, we suggest that molecules present in the BEO-FFi may antagonize the cytotoxic effect of limonene and linalyl acetate reported by Russo et al.^[32] Moreover, limonene is poorly present in BEO-TF although the latter inhibited SH-SY5Y growth rate. Interestingly, two compounds present in active BEOs and absent in the other fractions without anticancer activity are bergamottin and 5-geranyloxy-7-methoxycoumarin. Therefore, we believe that these two coumarins may play a pivotal role in the antiproliferative and cytotoxic effects exerted by coloured BEOs on SH-SY5Y cells, amplifying the biological activities of the other compounds. Of note, the anticancer properties of these two compounds are already reported. Bergamottin has been shown to possess antitumour activity both in in vitro and in vivo models^[57,58] by influencing the metabolic activation of chemical carcinogens. Recently, Hwang et al.,^[59] demonstrated the inhibitory effects of bergamottin on tumour invasion and migration of human fibrosarcoma HT-1080 cells. Also 5-geranyloxy-7-methoxycoumarin has reported as potential anticancer agent, since it suppresses human colon cancer (SW-480) cell proliferation.^[60]

Conclusion

Our study adds new findings on the pharmaco-toxicological profile of BEO suggesting a potential role against cancer, demonstrating for the first time that bergamottin and 5-geranyloxy-7-methoxycoumarin may greatly contribute to the BEO-induced antiproliferative effect. Also, we strengthen the concept that identifying one or more components of a phytochemical complex such as the plant essential oils and dissecting their molecular mechanisms of action is very arduous. Moreover, often, the complex mixtures of phytochemicals could be more effective than their individual constituents in preventing cancer through both additive and synergistic effects.^[61,62] Hence, it is important to study the potential anticancer activity of whole extracts containing all phytochemicals in a context of a multitarget pharmacological strategy.

Declarations

Conflict of interest

The Author(s) declare(s) that they have no conflicts of interest to disclose.

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.