Introduction

Gentiana lutea is a herbaceous perennial plant that grows in mountains of Central and Southern Europe and Western Asia. The roots of G. lutea, owning bitter properties, are frequently utilized in alcoholic beverages and food. They are also used in traditional medicine for stomachic preparations. According to phytochemical and pharmacological studies, G. lutea is rich in compounds that show several biological activities, such as hepatoprotective, antimicrobial, anti-inflammatory, antioxidant, radioprotective, hypoglycemic, anti-atherosclerotic and immunomodulatory activities (Pan et al. 2016). These multiple effects are mainly due to secondary metabolites. The most represented groups of secondary metabolites in G. lutea roots include secoiridoid-glycosides (gentiopicroside, gentiopicrin, swertiamarin, sweroside and amarogentin), iridoids (loganic acid), xanthones (gentisin and isogentisin), xanthone-glycosides (gentioside) (Mustafa et al. 2015a; Aberham et al. 2007).

A protective role of G. lutea extract against ketoconazole-induced testicular damage in rats, and the cytotoxic effect of x-ray irradiation on normal human immunocompetent cells, have been observed (Amin 2008; Menkovic et al. 2010).

Some studies suggested that Gentiana species may have beneficial cardio-vascular effects and anti-atherosclerotic activities (Schmieder et al. 2007; Rojas et al. 2000; Kesavan et al. 2013).

G. lutea extracts showed inhibitory activity on the myeloperoxidase and aldose reductase enzymes (Nastasijević et al. 2012; Akileshwari et al. 2012). A protective effect of gentiopicroside, the most abundant compound in G. lutea root extracts (Mustafa et al. 2015a), in experimental colitis has been associated to its anti-inflammatory activity through the down-regulation of TNF-α, IL-1β, IL-6, COX-2 and iNOS expression (Niu et al. 2016).

Up to now few studies have investigated the neuroprotective activity of extracts of the genus Gentiana. However, beneficial effects on central nervous system and antidepressant activity by inhibition of MAO-A and MAO-B have been described (Haraguchi et al. 2004).

Gentisides from G. rigescens and root methanol extract of G. lutea have shown to be potent inducers of neurite outgrowth on rat pheochromocytoma PC12 cells (Gao et al. 2010; Mustafa et al. 2015b). Bellidifolin, a xanthone compound derived from plants of Gentiana species, could exert protective effect on nerve injury produced by hypoxia (Zhao et al. 2017).

Neuroprotection can be achieved by an anti-apoptotic activity or through antioxidant properties. The purpose of this work was to evaluate the effect of G. lutea root extract on anti-proliferative activity of pro-apoptotic agents in the human SH-SY5Y neuroblastoma cell line. This kind of tumor cells is believed quite similar to normal neurons with respect to their morphological/neurochemical properties.

Inflammation is associated with many neurodegenerative disorders, including Parkinson’s disease, Alzheimer’s disease, multiple sclerosis and amyotrophic lateral sclerosis (Glass et al. 2010). Neurons can be exposed to inflammatory mediators that cause cell death. Glial cells are the main responsible of neuroinflammation. Neverthless, in neurodegenerative diseases, such as Alzheimer’s disease, CNS-infiltrating peripheral macrophages may also contribute to neurodegeneration (Steiner et al. 2016). Finally, we also investigated the effect of G. lutea extract treatment on the TNF-α release from activated RAW264.7 macrophage-like cells.

Materials and methods

Materials

Dulbecco’s modified Eagle’s medium (DMEM) and fetal bovine serum were from Euroclone. Acridine orange, aluminium chloride, ascorbic acid, dithiotreitol (DTT), dimethyl sulfoxide (DMSO), DPPH (2,2-diphenyl-1-picrylhydrazyl), ethidium bromide, Folin–Ciocalteau’s reagent, glutathione reductase, MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide], NADPH, Triton X-100, tannic acid, quercetin, sulfosalicilic acid, reduced glutathione (GSH), 5,5′-dithiobis(2-nitrobenzoic acid), lipopolysaccharide (LPS) and tert-butyl hydroperoxide (t-BHP), HPLC-grade acetonitrile (≥ 99.9) and formic acid were purchased from Sigma-Aldrich. 2′,7′-dichlorofluorescein diacetate (DCFH2-DA) was purchased from Cayman Chemical. Fluorogenic caspase-3 substrate, acetyl-Asp-Glu-Val-aspaminomethylcoumarin (Ac-DEVD-AMC), doxorubicin and vinblastine were from Alexis Biochemicals. Polyvinylidene difluoride (PVDF) was from Bio-Rad. Anti-Bcl-2 (cat. no. sc-509), anti-Bax (cat. no. sc-493), anti-Sirt-1 (cat. no. sc-15404) and anti-actin (cat. no. sc-1616) antibodies were from Santa Cruz Biotechnology, Inc.

The analytical standards of loganic acid, swertiamarin, sweroside, gentiopicroside, amarogentin and isogentisin were purchased from PhytoLab (Germany). The purity of all standard compounds was ≥ 97% (determined by HPLC). All other chemicals were reagent grade. Deionized water (> 18-MΩ cm resistivity) was obtained from a Milli-Q SP Reagent Water System (Millipore, Bedford, MA, USA). Stock solutions of vinblastine and doxorubicin were prepared in DMSO and stored in the dark at − 20 °C. All solvents and solutions were filtered before HPLC analysis through 0.45-μm PTFE filters purchased from Phenomenex (Bologna, Italy).

Preparation of G. lutea extracts

Commercial dried roots of G. lutea, collected in French Alpes, were purchased from an herbal shop. A voucher specimen was retained in the Laboratory of Cellular Biochemistry of Department of Biotechnological and Applied Clinical Sciences, University of L’Aquila (FBGL-0004).

Roots were cut into small pieces and finely powered using a mechanical blender.

Aliquots of 10 g of obtained powder were extracted with methanol (3 × 50 ml, 24 h each, at room temperature) or with H2O (3 × 50 ml, 24 h each, at room temperature or 1 × 50 ml, for 10 min, at 95 °C). After filtration through Whatman No. 1 filter paper, the alcoholic extracts were collected and concentrated under vacuum by using a rotary evaporator, yielding brown residues. The aqueous extracts were collected, frozen and lyophilized. The extracts were stored in darkness at 4 °C until used.

The methanolic and aqueous extracts were reconstituted, at a concentration of 50 mg/ml, in DMSO and distilled water, respectively.

Determination of total phenolic content

Total phenolic content was determined using the Folin–Ciocalteau method with some modifications (Stratil et al. 2006). Briefly, the reaction mixture with 1 mg of extract, 100 µl of Folin–Ciocalteau’s reagent and 800 µl of 2.5% sodium carbonate was incubated at 25 °C for 1 h and then the absorbance was measured at 765 nm in a Perkin–Elmer Lambda 19 spectrophotometer. Tannic acid was used as a standard. The results were compared with the tannic acid calibration curve and expressed as milligrams of tannic acid equivalents (TAE)/g of sample.

Determination of total flavonoid content

The total flavonoid content of extracts was determined by the aluminium chloride (AlCl3) method (Ordoñez et al. 2006). The reaction mixture with aliquots of 500 µl of extract solution, containing 2 mg of extract, and an equal volume of 2% AlCl3 in ethanol was incubated for 1 h at 25 °C. The absorbance was recorded at 420 nm in a Perkin–Elmer Lambda 19 spectrophotometer.

Known concentrations of quercetin were used to generate a standard calibration plot. The concentrations of flavonoids in the samples were calculated from the calibration plot and expressed as milligrams of quercetin equivalent (QE)/g of sample.

Determination of total flavonol content

Flavonols were determined according to Kumaran and Karunakaran (2007), with some modifications. Aliquots of 200 µl of extract solution, containing 5 mg of extract, were incubated with an equal volume of 2% AlCl3 and 600 µl of 5% sodium acetate. The reaction mixture was incubated at 25 °C for 150 min and the absorbance was recorded at 440 nm in a Perkin–Elmer Lambda 19 spectrophotometer. Known concentrations of quercetin were used to generate a standard calibration plot. The concentrations of flavonol in the test samples were calculated from the calibration plot and expressed as milligrams of quercetin equivalent (QE)/g of sample.

DPPH free radical scavenging assay

DPPH quenching activity of extracts was evaluated by spectrophotometric assays. Increasing concentrations of samples in methanol were mixed with 100 µL of 1 mM DPPH solution in methanol. After 30 min of incubation at room temperature, absorbance was recorded at 517 nm on a Perkin-Elmer spectrophotometer Lambda 19. Ascorbic acid was used as a reference standard. Radical-scavenging ability was calculated according to the following formula:

$$\% \;of\;Inhibition = \, \left[ {\left( {A_{control} - A_{sample} } \right)/A_{control} } \right] \, \times \, 100$$

where Acontrol is the absorbance of the DPPH solution without the samples, and Asample is the absorbance of the tested samples. The antioxidant activity was expressed as IC50 value, calculated as the minimum concentration of samples required to inhibit 50% of the DPPH radical.

HPLC–MS triple quadrupole

HPLC–MS studies were performed using an Agilent 1290 Infinity series and a Triple Quadrupole 6420 from Agilent Technology (Santa Clara, CA) equipped with an ESI source operating in negative ionisation mode. The separation was achieved on a Synergi Polar-RP C18 (4.6 mm × 150 mm, 4 µm) analytical column from Phenomenex (Chesire, UK). The mobile phases for HPLC–ESI–MS (triple quadrupole) analyses were water with 0.1% formic acid (80) and acetonitrile (20) at a flow rate of 0.6 mL/min in isocratic conditions. The solvent composition varied as follows: 0–7 min, 20% B; 7–15 min, 20–90% B; 15–18 min, 90% B; 18–25 min, 90–20% B. The column temperature was set at 30 °C and the injection volume was 10 μl. In HPLC-ESI–MS, ion source was operating in negative ionization (NI) mode. Optimization of the HPLC/ESI–MS conditions was carried out by flow injection analysis (FIA) of the analytes (1 μl of a 50 μg/ml individual standard solutions). The optimum ESI ion source conditions were as follows: gas temperature, 350 °C; nebulizer gas (nitrogen) pressure, 50 psi; drying gas (nitrogen) flow rate, 11 ml/min; capillary voltage, 4000 V. Quantifications were performed by using selected ion monitoring (SIM) mode, fragmentor 70 V, dwell time 150 and all the analytes monitored in two time segment, loganic acid, gentiopicroside swertiamarin and sweroside from 0 to 9 min, and amarogentin, gentiopicroside and isogentisin from 9 until the end of the run. The different ions were: the deprotonated molecular ions [M–H]− of loganic acid, amarogentin and isogentisin at m/z 375, 585 and 257, respectively and the [M + HCOOH–H]− ions of swertiamarin, sweroside and gentiopicroside at m/z 419, 403 and 401, respectively.

Cell culture

The human SH-SY5Y neuroblastoma cell line and the murine RAW 264.7 macrophage-like cell line, obtained from the American Type Culture Collection (ATCC) and the European Collection of Cell Cultures (ECACC), respectively, were grown in DMEM medium. The media were supplemented with 10% heat-inactivated fetal bovine serum, 100 U/ml penicillin, 100 μg/ml streptomycin and 2 mM glutamine. Cells were maintained at 37 °C in a humidified 5% CO2 atmosphere. Cell viability was determined by trypan blue exclusion test.

Cytotoxicity assay

The effects of G. lutea extract on cell viability were evaluated in vitro using the MTT colorimetric method (Mosmann 1983). Exponentially growing cells were seeded in 96-well plates and, after 24 h of growth, were treated with increasing concentrations of commercial G. lutea methanolic extract, from 50 to 800 µg/ml. The same amount of DMSO was added to negative controls. After treatment, MTT reagent, at a concentration of 0.5 mg/ml, was added to each well and the cells were incubated for additional 3 h at 37 °C. The absorbance at 570 nm was determined in a microplate reader (BioRad, Model 550) after solubilizing the formazan crystals by addition of 100 μl of acidified isopropanol (0.04 M HCl in isopropanol). Viability of untreated cells was considered 100%. Cell survival was determined comparing the absorbance of treated and untreated cells.

For evaluation of combined effects of the extract and some pro-apoptotic agents, SHSY-5Y cells were treated with increasing concentrations, from 50 to 800 µg/ml, of methanolic extract of G. lutea root, alone or in combination with concentrations from 0.05 to 0.8 µM of doxorubicin and vinblastine for 48 h. At the end of incubation, cell survival, compared with untreated controls, was evaluated using the MTT assay.

Apoptosis evaluation

Nuclear morphology was assessed by acridine orange/ethidium bromide double staining assay. After washing with PBS, cells were stained with a fluorescent solution containing 100 µg/ml ethidium bromide and 100 µg/ml acridine orange in PBS and immediately observed with a fluorescence microscope. Cells showing condensed and fragmented chromatin were considered apoptotic. A minimum of 400 cells were counted for each determination.

Caspase-3 activity

Cells were washed with PBS and then resuspended in a lysis buffer, containing 50 mM Tris–HCl, pH 7.4, 10 mM EGTA, 1 mM EDTA, 10 mM DTT, 1% (v/v) Triton X-100, for 30 min at 4 °C. After centrifugation at 15,000 g for 15 min at 4 °C, supernatants were collected and used for detection of caspase activity.

Cell lysate (60 µg of proteins) was mixed with 20 μM fluorogenic caspase-3 peptide substrate, Ac-DEVD-AMC, in the reaction buffer (50 mM Tris–HCl, pH 7.4, 10 mM EGTA, 1 mM EDTA, 10 mM DTT). The reaction mixture was incubated for 30 min, at 37 °C (Köhlex et al. 2002).

Fluorescence was measured on a Perkin-Elmer LS-50B spectrofluorometer, setting excitation at 380 nm and emission at 460 nm.

Detection of intracellular reactive oxygen species (ROS)

DCF fluorescence was used to detect the generation of cellular ROS (LeBel et al. 1992). Cells were exposed to 200 µg/ml of G. lutea root extract alone or in combination with 0.1 µM vinblastine for 12, 24 and 36 h at 37 °C and to 200 µM t-BHP for 2 h at 37 °C, as positive control. After treatments, cells were incubated with DCFH2-DA to a final concentration of 20 µM at 37 °C for 30 min. After washing with PBS, fluorescence intensity of cells was analyzed with a Perkin–Elmer LS-50B spectrofluorometer, setting excitation at 485 nm and emission at 530 nm.

Glutathione determination

Cells were washed with PBS and resuspended in 5 mM EDTA and 5% (w/v) sulfosalicilic acid. Total glutathione (GSH) concentrations was measured as previously reported (Rahman et al. 2006) with some modifications.

An aliquot (10 µl) of cell extract was added to 930 µl of 0.1 M potassium phosphate buffer (pH 7.5), containing 5 mM EDTA, 58 µM NADPH and 40 µg/ml 5,5′-dithiobis(2-nitrobenzoic acid). The assay was initiated by addition of 60 µl of glutathione reductase (3 U/ml). The rate of formation of 5-thio-2-nitrobenzoic acid, proportional to total glutathione concentration, was followed at 412 nm for 2 min at 25 °C in a Perkin–Elmer Lambda 19 spectrophotometer. GSH concentration values were calculated from a standard curve and expressed as nmol/106 cells.

Western blot analysis

Cells were washed with PBS and lysed in a buffer containing 10 mM Hepes, pH 7.2, 5 mM MgCl2, 142 mM KCl, 0.2% (v/v) Nonidet P-40, 1 mM EDTA and a suitable cocktail of protease inhibitors, at 4 °C for 30 min. Proteins were separated by SDS-polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride (PVDF) membranes. The blots were incubated with monoclonal anti-Bcl-2, anti-Bax and anti-Sirt-1 antibodies, and polyclonal anti-actin antibody, for 1 h at room temperature. The membranes were then probed with the appropriate peroxidase-conjugated secondary IgG antibodies. The blots were visualized using an enhanced chemiluminescent detection system (Thermo Scientific, Rockford, USA) and the intensity of each band was quantified by ImageJ software and normalized with actin levels.

TNF-α determination

RAW 264.7 cells were seeded at a density of 1.5 × 105 cells/well in 24-well plates for 24 h. The following treatments were performed: (1) cell stimulation with 200 ng/ml lipopolysaccharide (LPS) from Pseudomonas aeruginosa for 1 h, after pre-treatment with 200 µg/ml of G. lutea root methanolic extract for 7 h; (2) simultaneous exposure of cells to LPS and methanolic extract for 1 h.

Cell-free supernatants of treated and untreated cells were collected and TNF-α levels were quantified using an enzyme-linked immunoadsorbent assay (ELISA) kit (PeproTech, Rocky Hill, NJ) according to the manufacturer’s instructions.

Statistical analysis

Data are reported as mean ± SD. Statistical differences were calculated using the Student’s t test. Results were considered statistically significant at p value < 0.05.

Results

Total phenolic, flavonoid and flavonol content and free radical-scavenging capacity

Polyphenols represent a group of at least 10,000 different compounds. These secondary metabolites are associated to many beneficial effects of plant extracts in the prevention of cancer, cardiovascular and neurodegenerative diseases.

Besides the polyphenol concentration in G. lutea extracts, the content of flavonoids, the most commonly occurring dietary polyphenols, and flavonols (a class of flavonoids) was also determined.

Assays used in the present study demonstrated that phenolic contents were different between methanolic and aqueous extracts. As shown in Table 1, methanolic extract showed the highest amounts of polyphenols (92.46 ± 6.69 mg tannic acid equivalent/g extract), flavonoids (8.74 ± 0.49 mg quercetin equivalent/g extract) and flavonols (0.98 ± 0.27 mg quercetin equivalent/g extract).

Table 1 Content of polyphenols, flavonoids and flavonols and DPPH scavenging activity in G. lutea root extracts
Full size table

The antioxidant activity of G. lutea extracts was assessed by 2,2-diphenyl-1-picrylhydrazyl (DPPH) assay. Although the antioxidant potential of the extract fractions was found to be lower than that of ascorbic acid, used as positive control, all extracts showed some antioxidant activities. Methanolic extract, with an IC50 value of 59.0 ± 13.2 μg/ml, exhibited the major antioxidant activity (Table 1).

Since methanolic extract of G. lutea roots exhibited the highest amount of polyphenols and the highest free radical scavenging activity, the biological activity of only this extract was further investigated.

Method validation

Calibration curves of the analyzed compounds were constructed injecting 10 µl of standard solutions at eight different concentrations, from 0.005 or 0.01 mg/l, according to the different analytes, in HPLC/MS technique (Table 2). Five replicates for each concentration were performed and the coefficients of variation (CVs) ranged from 1.03 to 4.39% for run-to-run precision, and from 4.4 to 8.7% for day-to day precision. All the calibration curves of the analyzed compounds showed a correlation coefficient greater than 0.9924. LOD and LOQ were estimated on the basis of 3:1 and 10:1 S/Ns, obtained with standards containing the compounds of interest at low concentration levels. The LODs of analytes were in the range 0.00003–0.0006 mg/l, while the LOQs were in the range 0.0001–0.002 mg/l.

Table 2 Method validation data of HPLC–MS analytical method for the six analytes: concentration range of calibration curve, regression equation, correlation coefficient (R2), limit of detections (LODs) and limit of quantifications (LOQs)
Full size table

Quantification of bitter compounds in methanolic extract of G. lutea

HPLC/ESI–MS triple quadrupole method was used to quantify six bitter compounds in G. lutea methanolic extract (Table 3).

Table 3 Quantitative determination of the analyzed marker compounds in Gentiana lutea L. extract (mg/kg and percent %, are reported) by LC/ESI–MS; (results are given together with standard deviation, n = 3)
Full size table

The analyzed sample showed a high content of gentiopicroside, the major bitter principle of gentian plant, of 60,253.9 mg/kg (6.03%). The found level of the bitterest natural product to date, amarogentin, is of 8337.6 mg/kg (0.83%). The content of the other two secoiridoids, swertiamarin and sweroside, was 6222.6 and 16,520.4 mg/kg, respectively. Sweroside was the second most abundant compound after gentiopicroside, with a percentage of 1.65%. Loganic acid showed a level of 4235.6 mg/kg (0.42%), similar to isogentisin, i.e. 4929.5 mg/kg (0.49%).

Effect of methanolic extract of G. lutea roots on SH-SY5Y cell survival

To verify the effect of extract on cell growth and viability, SH-SY5Y cells were exposed to increasing concentrations of extract (ranging from 50 to 800 µg/ml) for 48 h. Cell survival, evaluated using the MTT assay, was compared with untreated controls.

Cell viability was unaffected by the presence of methanolic extract (Fig. 1). Only a weak anti-proliferative activity was observed at the highest concentration of 800 µg/ml.

Fig. 1
figure 1

Effect of Gentiana lutea methanolic extract treatment for 48 h on SH-SY5Y cell viability. Cell survival was determined by the MTT assay

Full size image

This result indicates that our G. lutea extract did not exhibit cytotoxic effect on SH-SY5Y cell; rather, at low concentration, it seemed to increase cellular proliferation.

Influence of G. lutea extract on the cytotoxic effect of pro-apoptotic agents in SH-SY5Y cells

Experiments were performed to evaluate methanol extract effects on cytotoxicity induced by doxorubicin and vinblastine that are important chemotherapeutic and pro-apoptotic agents used to treat a wide spectrum of human cancers. The doxorubicin and vinblastine concentrations were chosen on the basis of the IC50 values obtained from preliminary experiments (data not shown).

The activities of 200 and 400 µg/ml of methanol extract on SH-SY-5Y cells treated with 0.2 µM doxorubicin and 0.1 µM vinblastine for 48 h were investigated. Cell survival, compared with untreated controls, was evaluated using the MTT assay.

G. lutea methanol extract exhibited protective effect from both vinblastine and doxorubicin treatments, but significant results were obtained only on the anti-proliferative activity of vinblastine (Fig. 2).

Fig. 2
figure 2

Combined treatment of SH-SY5Y cells with 200 and 400 µg/ml of G. lutea root methanol extract and 0.2 µM doxorubicin or 0.1 µM vinblastine for 48 h. Cell survival was determined by the MTT assay. The results represent the mean ± SD of three independent experiments. *Data are statistically different from vinblastine alone-treated cells (p < 0.05). #Data are statistically different from untreated cells (p < 0.05)

Full size image

Extract enhanced viability of vinblastine-treated cells from 41.5 to 60.6%, with the concentration of 200 µg/ml, and to 61.8% with the concentration of 400 µg/ml, respectively.

Since extract, at both concentrations, exhibited significant protective effect against vinblastine treatment, the activity of the lower concentration (200 µg/ml) of extract, in combination with vinblastine, was further investigated.

G. lutea extract reduced vinblastine-induced apoptosis in SH-SY5Y cells

The effect of 200 µg/ml G. lutea extract on viability of vinblastine-treated SH-SY5Y cells was also evaluated, by trypan blue exclusion test, after 12, 24 and 36 h of treatment. G. lutea extract did not show any cytotoxic effect on SH-SY5Y cells. By contrast, in the combined treatment it showed a life-promoting activity if compared with the vinblastine alone treatment, increasing cell survival. After 36 h, the combination of the two drugs caused a significant increase in the number of viable cells (71.3%) as compared to vinblastine alone (45.6%) (Fig. 3A). To determine the mechanism by which methanolic G. lutea extract improved cell growth, SH-SY5Y cells treated with vinblastine alone or in combination with G. lutea extract were analyzed for the occurrence of apoptotic process. Apoptotic markers such as condensation and fragmentation of nuclear chromatin and activation of caspase-3 were evaluated. The analysis of nuclear morphology showed that the percentage of apoptotic cells increased from 12 to 36 h of treatment with vinblastine alone, reaching 40.5%, after 36 h. The addiction of 200 µg/ml extract in combination with vinblastine to cell samples resulted in a significant decrease of apoptosis as compared to vinblastine alone, reaching 15.9% of apoptotic cells after 36 h of treatment (Fig. 3B, D).

Fig. 3
figure 3

Effect of 200 µg/ml G. lutea root extract and 0.1 µM vinblastine alone or in combination on SH-SY5Y cell viability, apoptosis and caspase-3 activity. A Cells were counted and the percentage of trypan blue-negative cells were determined at the indicated times. B Analysis of characteristic nuclear morphological features of apoptosis. The percentage of condensed and fragmented nuclei was estimated by fluorescence microscope analysis of acridine orange and ethidium bromide double-stained cells observed at the indicate times. C Time course of caspase-3 activity measured by using DEVD-aminomethylcoumarin as substrate. Results represent the mean ± SD of three independent experiments. *Data are significantly different from vinblastine alone-treatment (p < 0.05). #Data are significantly different from untreated cells (p < 0.05). D Fluorescence microscope pictures of SH-SY5Y cells untreated (control) and treated with extract and vinblastine alone or in combination for 36 h

Full size image

To confirm the protection from apoptosis, the effect of G. lutea root extract on caspase-3 activity in SH-SY5Y cells was also investigated. After incubation with the extract, a significant progressive reduction of vinblastine-induced caspase-3 activity was observed. In fact, compared to treatment with vinblastine alone, the combined treatment showed, at 24 and 36 h, a significant decrease of caspase-3 activity (almost 1.8- and 2-fold, respectively) (Fig. 3C).

Effect of G. lutea root extract on Bcl-2, Bax and Sirt-1 expression

To further clarify the mechanism of G. lutea root extract protection from vinblastine-induced apoptosis, the expression of Bcl-2 and Bax proteins was evaluated by western blot analysis after treatment of SH-SY5Y with extract and vinblastine either alone or in combination for 24 h. Both extract and vinblastine, alone or in combination, did not influence the expression of Bax protein (Fig. 4). On the contrary, either the extract alone and in combination induced a fairly but significant increase of Bcl-2 expression as compared to untreated cells (Fig. 4A, C). In vinblastine alone treated-cell samples, immunoblot analysis showed the presence of a second, slower migrating Bcl-2 band, that can probably be attributed to protein phophorylation, in line with the well-known post-translational modification of this protein in cells treated with agents that prevent microtubule assembly (Terrano et al. 2010). By contrast, no detectable Bcl-2 band with higher molecular weight was observed after exposure of SH-SY5Y to G. lutea extract in combination with vinblastine. Thus, Gentiana extract could prevent Bcl-2 phosphorylation that antagonizes its anti-apoptotic activity.

Fig. 4
figure 4

Western blotting analysis of Bcl-2, Bax and Sirt-1 expression in SH-SY5Y cells treated for 24 h with methanolic extract of G. lutea roots and vinblastine alone or in combination. A Representative immunoblots of three independent experiments with similar results of Bcl-2 and Bax expression. B Representative immunoblot of three independent experiments with similar results of Sirt-1 expression. C Densitometric analysis of immunoreactive bands normalized with actin protein levels. Data have been obtained arbitrarily assigning the value 1 to the control. Results represent the mean ± SD of three independent experiments. #Data are significantly different from untreated cells (p < 0.05); *Data are significantly different from vinblastine-treated cells

Full size image

Sirt-1 is implicated in the regulation of many cellular processes including apoptosis, cellular survival and senescence, endocrine signaling, aging and longevity (Chang and Guarente 2014). Therefore, the expression of Sirt-1 protein in SH-SY5Y cells was evaluated by western blot analysis after treatment for 24 h with G. lutea root extract and vinblastine either alone or in combination. The exposure to 0.1 µM vinblastine alone induced a significantly decrease in protein expression as compared to untreated cells (Fig. 4b, c). This result suggests that cytotoxic effect of vinblastine may be associated to a Sirt-1 decreased expression.

The exposure to 200 µg/ml G. lutea root extract alone did not influence the expression of Sirt-1, but in the combined treatment, the extract reverted the reduction of Sirt-1 expression. The presence of the extract, in fact, stabilized Sirt-1 protein to levels similar to those of basal untreated cells.

Effect of G. lutea extract on intracellular GSH and ROS levels

Since increase of ROS production is involved in the cytotoxic activity and in the initiation and execution phases of apoptosis of several anticancer agents (Fang et al. 2007), we explored the possible modulation of ROS levels associated to the protective effect of G. lutea methanol extract in SH-SY5Y cells.

Cells were exposed to 200 µg/ml extract and 0.1 µM vinblastine, alone or in combination, for 12, 24 and 36 h, loaded with DCFH2-DA and then analyzed for DCF fluorescence. No significant alteration of the intracellular ROS level has been observed, whereas a tenfold increase of intracellular ROS level was detected in cells treated with t-BHP, used as positive control.

To verify if the major resistance of SH-SY5Y cells to apoptosis was due to a higher content of reducing power, intracellular glutathione concentration was determined. After 24 h exposure of SH-SY5Y cells to 200 µg/ml extract and 0.1 µM vinblastine, either alone and in combination, a significant increase of GSH levels in all treated cells was observed as compared to untreated control. Even if vinblastine, administered alone, induced a mild stimulation (1.5 fold with respect to untreated cells) of GSH production, G. lutea extract treated cells exhibited higher increase (twofold with respect to untreated cells) of GSH concentration. This increase of GSH concentration (2.3 fold with respect to control) was also induced by the combined treatment (Fig. 5).

Fig. 5
figure 5

Levels of total glutathione in SH-SY5Y cells untreated (control) or treated with 200 µg/ml G. lutea root methanol extract and 0.1 µM vinblastine alone or in combination for 24 h. Results represent the mean ± SD of triplicate experiments. #Data are significantly different from untreated control (p < 0.05); *Data are significantly different from vinblastine alone treatment (p < 0.05)

Full size image

Anti-inflammatory activity of G. lutea root extract

Several plant compounds have been shown to exhibit anti-inflammatory activity. Roots of many species of Gentiana have traditionally been used in Europe and Western Asia as herbal medicines because of their anti-inflammatory effects. To examine the anti-inflammatory activity of G. lutea root extract, the ability to inhibit LPS-induced expression of TNF-α in RAW264.7 macrophage cells has been investigated.

Macrophages, a standard model for studying anti-inflammatory drugs, were treated with 200 µg/ml extract and 200 ng/ml LPS from Pseudomonas aeruginosa, then the concentrations of secreted TNF-α in the supernatants were determined.

The following treatments were performed: (1) cell stimulation with 200 ng/ml LPS for 1 h, after pre-treatment with 200 µg/ml methanolic extract of commercial G. lutea roots for 7 h; (2) simultaneous exposure to LPS and methanolic extract for 1 h. In both experiments (with pre-treatment and simultaneous treatment), LPS produced a considerable increase of TNF-α in culture media, 903.3 ± 110.6 and 910 ± 141.4 pg/ml, respectively, as compared to media of untreated cells (136.0 ± 33.3 and 130 ± 14.1, respectively).

The exposure of RAW264.7 to extract alone did not affect the levels of TNF-α, but pre-treatment of cells with G. lutea extract significantly decreased the LPS stimulation (about 60%). This inhibitory effect was much less evident and not significant with the simultaneous treatment, showing a reduction of only 23% (Table 4).

Table 4 Effect of methanolic extract of G. lutea roots on TNF-α levels in LPS stimulated murine macrophage-like cell line RAW 264.7
Full size table

Discussion

Biological properties of crude extracts, that contain several compounds, could be due not only to the nature and abundance of particular metabolites but also to their synergism. Our methanol extract showed a significant polyphenol content and antiradical activity. A high content of gentiopicroside, the major bitter compound of gentian plant, was found in the methanolic extract in good agreement with previous studies, while the content of amarogentin and sweroside was higher than that previously reported (Mustafa et al. 2015a; Aberham et al. 2007).

In neuroblastoma cell line SH-SY5Y, G. lutea extract not only did not exhibit any cytotoxic effect, but instead it produced a significant protection of cells from apoptotic agents, such as vinblastine. In our cellular model, vinblastine cytotoxicity was not associated to an increase of intracellular ROS, thus indicating that the anti-apoptotic activity of G. lutea extract could not be due to its antioxidant capacity.

G. lutea extract, instead, seemed to regulate the balance between pro-/anti-apoptotic markers, that has been involved in the maintenance of the mitochondrial integrity (Youle and Strasser 2008). It has been reported that Bcl-2, besides its function on the mitochondrial membrane (Ouyang and Giffard 2004), activates NF-κB, AKT and IKK in cancer (Mortenson et al. 2007). Bcl-2 over-expression leads to the maintenance of cyclin D1a expression in human lymphoma cell lines (Tucker et al. 2008). The increase of Bcl-2 expression induced by G. lutea extract in our cell model is particularly interesting since SH-SY5Y cells derive from a human neuroblastoma and they have been extensively used to evaluate neuronal injury or death in neurodegenerative diseases. Then, it was observed that the anti-apoptotic protein Bcl-2 improves neuron survival upon various stressing injuries (Ouyang and Giffard 2004, 2014).

Moreover, G. lutea extract treatment antagonized the phosphorylation of Bcl-2 protein, typically induced by the antimitotic agent vinblastine, that could disable its anti-apoptotic activity.

Sirt-1 has been reported to be a pivotal protein for cell survival and for the improvement of defense against the damage induced by oxidative stress or radiation (Hisahara et al. 2005).

Proteins involved in the control of cell proliferation and survival, such as the tumor suppressor p53, the transcription factor NF-kB and the FOXO family of transcription factors, are some non-histone cellular substrates of Sirt-1.

Several studies demonstrated that Sirt-1 plays a neuroprotective role and shields neurons from apoptosis induced by different insults, such as mechanical trauma, ischemia or neurotoxins (Zhao et al. 2012; Hernandez-Jimenez et al. 2013; Wang et al. 2018; Zou et al. 2016). Our results showed that vinblastine-induced apoptosis was associated to reduced Sirt-1 expression while, the simultaneous addition of G. lutea extract markedly reversed this down-regulation. Therefore, Sirt-1 seems to mediate the protection of G. lutea extract against vinblastine-induced toxicity in SH-SY5Y cells.

An increase of GSH, induced by G. lutea extract, could also contribute to the protective activity of the extract. GSH participates not only to the cellular antioxidant defense pathway, but high levels of this metabolite could be involved in detoxification of chemotherapeutic agents, decreasing their effectiveness (Traverso et al. 2013).

Vinca alkaloids, such as vinblastine, vinorelbine, vincristine, are widely used in the treatment of different types of cancers. Vinca alkaloids trigger neuropathy, as unwished side effect, in about 20% of cancer patients and there is not, up to now, an efficient strategy to prevent this detrimental effect (Banach et al. 2016). So far, a few studies about neuroprotective activity against vinca alkaloids neurotoxicity by natural metabolites have been published (Majithia et al. 2016; Schröder et al. 2013).

Since G. lutea extract attenuates apoptosis in neuroblastoma SH-SY5Y cells, it might be considered as a promising potential tool in chemotherapy-induced neuropathies.

Several plant metabolites also exhibit anti-inflammatory activity. This activity, observed in several species of Gentiana (Wang et al. 2013; Yamada et al. 2014; He et al. 2015; Jia et al. 2016) was often associated with the presence of iridoid and secoiridoid glycosides. Kesavan et al. (2013) observed that PDGF-BB-induced production of intracellular nitric oxide was significantly decreased by treatment with G. lutea extract in aortic smooth muscle cells.

In our study, we showed that even roots of G. lutea have anti-inflammatory activity. In fact, treatment with G. lutea extract decreased the levels of TNF-α released from macrophages cell line RAW264.7 stimulated with LPS from Pseudomonas aeruginosa.

The extract did not seem to affect the interaction of LPS with membranes since short and simultaneous treatment of these cells showed only a weak anti-inflammatory activity, whereas prolonged exposure of RAW264.7 to G. lutea extract triggered a higher anti-inflammatory activity decreasing TNF-α production.

The isolation and characterization of bioactive substances of G. lutea extract should be objectives of further studies to verify the potential beneficial activity of these compounds as well as their possible synergism.