Chemical Composition and Biological Activity of A . cepa L . and A . × cornutum ( Clementi ex Visiani 1842 ) Methanolic Extracts

Here we report a comparative study of the pyhtochemical profile and a biological activity of the two onion extracts, namely A. cepa L. and A. × cornutum (Clementi ex Visiani 1842). Identification of flavonoids and anthocyanins and their individual quantities were determined by high performance liquid chromatography (HPLC). The potency of both extracts to scavenge free radicals was determined by DPPH (2,2'-diphenyl-1-picrylhydrazyl) radical-scavenging activity and oxygen radical absorbance capacity (ORAC) methods. DNA protective role was further tested by single-cell gel electrophoresis (COMET) assay and by Fenton’s reagent causing double strand brakes on pUC19 plasmid. In the presence of both extracts, a significant decrease in the DNA damage was observed, which indicates a protective role of A. cepa and A. × cornutum on the DNA strand brakes. Additionally, cytotoxicity was tested on glioblastoma and breast cancer cell lines. The results showed that both extracts had antiproliferative effect, but the most prominent decrease in cellular growth was observed with glioblastoma cells.


Introduction
Oxidative stress is the disturbance between the production of reactive oxygen species (ROS) and antioxidant defenses.Overproduction of ROS can cause damage of important biomolecules such as DNA, proteins, lipids, and carbohydrates resulting in several diseases [1].For this reason, living cells have developed antioxidant systems to control free radicals, lipid peroxidation and maintain oxidative-antioxidative balance [2].Dietary antioxidants also play an important role in the suppression of oxidative stress that may cause initiation and progression of various diseases, including cancer.Therefore, the consumption of dietary antioxidants is considered to be an effective tool in preventing diseases that may be caused by oxidative stress [3].The genus Allium covers more than 750 species distributed all over the northern hemisphere [4].The members of this genus are known not only as flavored vegetables and spice, but also as medical plants that have been used in traditional medicine [5,6].Among other Allium species, Allium cepa L. (common onion) is one of the oldest plants cultivated worldwide [7].Many epidemiological studies confirmed that dietary consumption of onions is associated with reduced risk for developing many forms of cancer, cardiovascular and neurodegenerative diseases [8][9][10].Their beneficial effect on health is attributed to a high content of biologically active phytomolecules such as phenolic compounds, especially flavonoids and several organosulfur compounds [11].The most abundant flavonoids found in onions are quercetins, namely quercetin-4'-monoglucoside and quercetin-3,4'-diglucoside which account for more than 85% of total flavonoid content [12,13].Besides flavonoids, onions, especially the red varieties, are a rich source of anthocyanins.The most frequently reported anthocyanins in red onions are cyanindin derivatives, although minor amounts of peonidin, petunidin and delphidin derivatives have also been identified [14].A. × cornutum (Clementi ex Visiani 1842) is a triploid hybrid onion (2n=3x=24) originating from three parental species (A.cepa, A. pskemense B. Fedt.and A. roylei Stearn), traditionally cultivated in costal Croatia under the name 'Ljutika' (shallot) [15].Due to its tasty bulbs and leaves it is widely used as a spice and similarly cultivated in other parts of the world such as South-East Asia and Europe.To our knowledge, the chemical composition and biological activity of A. × cornutum have not been studied so far.This motivated us to identify and quantify major phenolic compounds present in bulbs of A. × cornutum and A. cepa and to assess the biological activity (free radical scavenging potential, antimutagenic activity, antiproliferative activity on cancer cells, and antigenotoxic activity) of these widely used plants.

HPLC analysis of flavonols and anthocyanin and total phenolic content of the two onion species
Two major and three minor fine peaks were successfully resolved by HPLC analysis of flavonols in A. × cornutum and A. cepa methanolic extracts (Figure 1).The resulting chromatograms were qualitatively similar, which is in good agreement with other reported studies [16][17][18][19][20]. Here, two major peaks were identified as quercetin 3,4'-diglucoside (1) (tR for A. cepa 33.26 min; tR for A. × cornutum 33.30 min) and quercetin 4'-monoglucoside (2) (tR for A. cepa 41.41 min; tR for A. × cornutum 41.45 min).Three minor peaks were identified as myricetin (3) (tR for A. cepa 43.34 min; tR for A. × cornutum 43.37 min), quercetin aglycone (4) (tR for A. cepa 48.13 min; tR for A. × cornutum 48.17 min) and isorhamnetin (5) (tR for A. cepa 52.94 min; tR for A. x cornutum 52.97 min).Two main quercetin conjugates, (1) and ( 2), together account for about 80% of total flavonol content in both onions (Table 1).Similar concentration of the two quercetin conjugates was previously observed in other studies [16][17][18][19], confirming that the two are indeed predominant flavonols in onions.However, in this work somewhat higher concentration of (1) and ( 2) was observed in A. × cornutum which is in agreement with higher total phenolic content (A. × cornutum TPC 6.63 ± 0.31 mg CA/g DW; A. cepa TPC 6.24 ± 0.23 CA/g DW) accounted for this onion species.Our obtained TPC values show good correlation to those reported by Santas, Carbo, Gordon, and Almajano (2008) [21].In their study, TPC values of the two methanolic extracts of Spanish onion varieties ranged from 5.15 to 6.33 mg GAE/g DW for white onion, and 2.48 to 2.58 mg GAE/g DW for Calçot de Valls onion.However, it should be noted that TPC values could vary depending on the location of the cultivar and on the solvent used during the extraction process [1,[21][22][23][24].
Furthermore, quantification of individual anthocyanins was carried out on HPLC (Table 1) whereat three anthocyanins in A. cepa and four in A. × cornutum extract were identified.Three anthocyanins were the same for both onions and were further identified as peonidin glucoside (6), petunidin glucoside (7) and malvidin glucoside (8).Interestingly, the fourth anthocyanin, delphidin glucoside (9), was found only in the A. × cornutum extract.The only study reporting the presence of the delphidin glucoside in Tropea red onion is the one reported by Gennaro et al. (2002) [14].Beside delphidin glucosides, they also identified a petunidin derivatives and together they accounted for ~30% of total anthocyanins in onion bulbs.

Antioxidant activity
In order to determine the antioxidant potential of our extracts, two tests were performed, DPPH and ORAC assay.Both onion showed mutually similar DPPH (% DPPH inhibition for A. cepa 64.82 ± 5.31 and A. × cornutum 60.50 ± 3.84) and ORAC values (µmol TE/mL for A. cepa 17.62 ± 0.57 and A. × cornutum 19.38 ± 2.21).However, slightly higher antioxidant activity was observed for A. × cornutum in ORAC assay, which can be attributed to a bit higher total phenolic content.The high antioxidant scavenging activity of Allium species was reported in other studies and it was shown that it depends on the existence of both phenolic and organosulfur compounds [24][25][26][27][28][29].

DNA nicking assay
As it was shown that both onions have relatively high amount of phenolic compounds, we further wanted to investigate their potential in protecting DNA from ROS.In that sense, pUC19 plasmid exposed to Fenton's reagent served as a biological probe.
Hydroxyl radicals generated by the Fenton reaction cause oxidatively induced DNA strand breaks to yield its open circular or relaxed form [1]. Figure 2. show the DNA protecting effect of A. cepa and A. × cornutum where it can be seen that lower extract concentrations show poor DNA protection and an increase in open circular and linear forms of DNA.On the contrary, higher concentrations show significant free radical scavenging activity and the ability to maintain a supercoiled form of DNA.Our results clearly show that preservation of intact circular plasmid is concentration dependent (5-100 µg/ml).Since both extracts have a relative high amount of phenolic compounds (quercetin and its derivatives) which act as antioxidants, we presume they can be responsible for such a good scavenging activity.Several other previous studies have in turn also reported that high phenolic content prevents DNA damage caused by hydroxyl radicals [1,30].
The antioxidant activity of phenolic compounds is associated with the total number and position of hydroxyl functional groups that are able to reduce highly oxidizing free radicals such as superoxide, peroxyl, aloxyl and hydroxyl radicals [31].Other than ROS reduction by hydrogen donation, antioxidant action can include suppression of the ROS formation by chelating metal ions, scavenging ROS and inhibition of oxidases [6].Since phenolic compounds act as antioxidants, it is justified to conclude that higher phenolic content leads to stronger antioxidant capacity.This conclusion agrees with our findings that TPC values are in correlation with DPPH-scavenging and ORAC values for both onions.

Comet assay
The comet assay was performed to determine a possible protective effect of both onions on DNA damage in human leukocytes.Incubation of leukocytes with DNA damage inducing reagent (H2O2) resulted in a significant increase of DNA damage comparing to the untreated control (p < 0.001) (Figure 3.).Cells treated with A. × cornutum or A. cepa extracts showed only moderate level of the DNA damage indicating that some of the phytochemicals present in onions can also provoke oxidative DNA damage.This is considered to be a structural property of phenolic compounds that can act as, both, antioxidants and prooxidants at the same time.In the presence of oxygen, some transition metals such as Cu and Fe can cause oxidation of flavonoids leading to the formation of ROS, which can damage DNA molecule [32].When leukocytes were simultaneously incubated with onion extracts and H2O2, a significant decrease in DNA damage was observed (p < 0.0001 for A. × cornutum and A. cepa).
Hydrogen peroxide generates hydroxyl radicals (OH • ) that cause DNA strand breaks and fragmentation.Thanks to the high levels of phenolic compounds, especially quercetin and its derivatives, A. × cornutum and A. cepa are able to catch and inactivate those radicals before they can induce DNA damage.Quercetin and its glycosides may work in a way that donate a hydrogen atom from its phenolic hydroxyl group in B-ring to remove hydroxyl radicals generated from hydrogen peroxide.Similar protective effect of quercetin on hydrogen peroxide induced DNA damage was reported before [33][34][35].Our result demonstrate that both extracts are able to prevent DNA damage caused by oxidative DNA damaging agents such as H2O2 in human leukocytes.

Cell proliferation assay
To evaluate whether A. × cornutum and A. cepa methanolic extract have cytotoxic effect on breast and glioblastoma cancer cell lines, cells were treated with onion extracts for 4, 24, 48 and 72 hours and the amount of metabolically active cells was determined by MTT assay.
As shown in Figure 4. both extracts significantly reduced (p < 0.001) the number of viable cells, albeit A. × cornutum which showed slightly stronger inhibitory effect on all cancer cell lines.The most striking observation was the low viability of glioblastoma, which among other tested cancer cells, showed the highest susceptibility in the presence of both onion extracts.Our results indicate that both plants, A. × cornutum and A. cepa, are effective inhibitors of tumor cell proliferation which is in correlation with study of Boivin et al., (2009) [26], who have also shown strong antiproliferative effect of Allium vegetables on different cancer cell lines.It's worth noting that Allium extracts used in their study, also displayed strong inhibitory effect against glioblastoma cancer cells.Most importantly, the authors showed that onion extract had no impact on the growth of normal fibroblast, suggesting their selective antiproliferative activity exclusively on tumor cells.Similar results were obtained by Yang, Meyers, van der Heide, and Liu (2004) [36] as they confirmed antiproliferative effect on human epithelial colorectal adenocarcinoma cells (Caco-2) and liver hepatocellular carcinoma cells (HepG2).The authors also observed that different range of antiproliferative activities depends on onion cultivar and the type of cancer cells used in the experimental process.Therefore, the authors hypothesized that the reason for these differences could be the attributed to a different type and concentration of bioactive phytochemicals, which can target different molecules or different signaling pathways in distinct cancer cell lines.Millet, Lamy, Jonas, Stintzing, Mersch-Sundermann, and, Merfort (2012) [37] tested antiproliferative activity of three onion extracts: fermented aqueous extract (FAE), aqueous (AE) and methanolic (ME).Only FAE extract showed significant toxicity on HepG2 cancer cell growth, while the other two did not have any impact on cell proliferation.It should also be emphasized that the inhibition of cancer cell proliferation by the two tested extracts may not be exclusively due to their polyphenolic content, but might be also attributed to their other bioactive compounds, for example organosulfur compounds, typical for the majority of Allium species.These findings are in agreement with known anticancer properties of Allium species noticed in many epidemiological studies [9].

Plant material
Allium × cornutum was obtained from local gardens of Croatian coasts and islands and the Allium cepa plants were purchased at a local market.

Preparation of onion extracts
The phenolic compounds were extracted from the homogenized dry plant material (10 g) using 70% v/v methanol/water (100 ml).After 30 min of extraction with magnetic stirring at room temperature (ca 20°C), the extract was centrifuged at 3000 rpm for 15 min.All three supernatants were pooled and dried under a vacuum using rotary evaporator at 50°C.The dry residues were dissolved with 10% DMSO.For HPLC analysis, the solution was filtered through a 0.45 µm nylon filter disc prior to analysis.

HPLC analysis of flavonols
High performance liquid chromatography (HPLC) measurement of flavonols were carried out on Perkin Elmer HPLC system (Waltham, Massachusetts, USA) consisting of a binary pump Series 200, an autosampler, Peltier column oven Series 200, UV/VIS detector Series 200 and UltraAqueous C18 column (250 x 4.6 mm, Resek).TotalChrom Workstation software was used to process the chromatographic data.After filtration through 0.45 µm syringe filter, extract was injected directly through a 20 µl fixed loop into a guard of C18 column.Each sample was injected three times in order to check reproducibility.A gradient consisting of a solvent A (0.2% H3PO4) and solvent B (MeOH/acetonitrile, 1:1 v/v) was applied at the flow rate of 0.8 ml/min as follows: 0-0.5 min 96% A and 4% B; 0.5-40 min 50% A and 50% B; 40-45 min 40% A and 60% B; 45-60 min 0% A and 100% B; 60-68 min 0% A and 100% B; 68-70 min 96% A and 4% B; 70-80 min 96% A and 4% B. Detection of the elution peaks was at 360 nm.Flavonoid compounds were identified on the basis of their retention times and quantified using external standard calibration curves.Standards for identification purposes were: quercetin, quercetin 4'-monoglucoside, quercetin 3,4'-diglucoside prepared in methanol.The resultant concentrations are expressed as mg/100 g of dry weight.

HPLC analysis of anthocyanins
HPLC analysis of anthocyanins was performed using a Varian HPLC system (Varian, Inc., CA, USA), consisting of a Star 9010 pump, a Rheodyne 7125 syringe loading sample injector, a 500-LC module for a column oven, a ProStar 330 Photodiode Array Detector, and a Star Chromatography workstation, version 5. The separation was carried out using a Kinetex C18 core-shell column (150 x 4.6 mm), filled with 5 µm particle size, and furnished with the SecurityGuard ULTRA Cartridge UHPLC C18 for 4,6 mm ID column (Phenomenex, USA), both termostated at 35ºC.Two eluents were used: A was 0.3% HClO4 and B was MeOH.The linear gradient was as follows: from 28% B to 51% B in 42 min, than to 69% in 3 min and to 80% B in 1 min 80% B for 3 min.The time of equilibration for the column to the initial gradient was 6 min, and the injection volume was 10 µL.The flow rate was 0.6 mL/min.Samples and the standards were filtered before analysis through a 0.45 µm pore size membrane syringe filters.Anthocyanins were identified according to retention times and UV-DAD spectra of each peak at 520 nm.Quantifications were performed using a standard curve of malvidin-3-O-glucoside chloride.The resultant concentrations are expressed as mg/100 g of dry weight.

Determination of total phenolic content
The total phenolic content (TPC) of A. × cornutum and A. cepa methanolic extracts was determined using Folin-Ciocalteu method described by Singleton and Rossi (1965) [38].Briefly, determination of TPC in tested samples was carried out using 10 mL of previously diluted Folin-Ciocalteu reagent (1:20 v/v) and 50 µL of onion extract.After 3 min of incubation, 1 mL of saturated sodium carbonate was added after which the reaction mixture was incubated for another 60 min in the dark and the absorbance was measured at 725 nm using the Perkin Elmer UV/VIS Lambda Bio 40 spectrophotometer.Caffeic acid (CA) served as the standard and results were expressed as mg of caffeic acid equivalents (CA) per g of dry weight.All measurements were carried out in triplicate and the results are expressed as mean values ± SD.

Antioxidant activity
3.5.1.Measurement of the DPPH (2,2'-diphenyl-1-picrylhydrazyl) radical-scavenging activity Antioxidant capacity of extracts was evaluated using the DPPH method previously described by Kulišić, Dragović-Uzelac, and Miloš, (2006) [39].Briefly, the method is based on the reduction of alcoholic DPPH • ˉ solution in the presence of a hydrogen donating antioxidant.An aliquot (50 µL) of the onion extract was mixed with methanolic solution of DPPH (1 mL, 0.1 mM) and the initial absorbance at 517 nm was measured immediately using EtOH as a blank.After an hour of incubation, the absorbance was measured again and the percentage of DPPH inhibition was calculated according to the formula by Yen and Duh (1994) [40]: where AC(0) is the absorbance of the control at t=0 min, and AA(t) is the absorbance of the antioxidant at t=1 h.All measurements were done in triplicate.

Oxygen Radical Absorbance Capacity (ORAC) assay
The assay was performed in Perkin-Elmer LS55 spectrofluorimeter, using 96-well white polystyrene microtiter plates (Porvair Sciences).Each reaction contained 190 µL of fluorescein (160 µM), 60 µL AAPH (150 mM) and 30 µL of plant extracts or reference standard Trolox (6.25-50 µM).All experimental solutions and samples were prepared in phosphate buffer (0.075 mM, pH 7.0).The measurements were performed in triplicates at 37°C using the excitation wavelength of 485 nm and the fluorescence decay was monitored at 530 nm during the period of 60 min.The obtained fluorescence decay curves were analyzed by FL WinLab software and the resultant area under the curve (AUC) of each standard or sample was acquired after the blank AUC subtraction (Supplementary Fig. 1.).The standard curve was generated by plotting AUC of standards with corresponding nmol of trolox.The ORAC values of onion extracts were expressed as µmol of trolox equivalents (TE) per mL of tested sample.The results were obtained from three independent experiments.Blood samples were obtained from a healthy female donor (age 28, non-smoker, first author of this paper and protocol was approved by the Ethics Committee of School of Medicine, University of Zagreb).Venous blood was collected into heparinized vacutainer tubes (Becton Dickeson, Plymouth, UK) under sterile conditions.Leukocytes were isolated by gradient centrifugation with Histopaque-1077 (Sigma, St. Luis, USA), 400 x g for 30 min and washed twice with PBS.Cell viability was determined with trypan blue exclusion assay.Leukocytes (2 x 10 6 cells/mL) were then incubated in RPMI-1640 medium (Gibco, Invitrogen, USA) supplemented with A. × cornutum and A. cepa methanolic extract (100 µg/mL) for 30 min at 37°C in a humidified atmosphere with 5% CO2.Cells growing in a medium supplemented with 1% DMSO were used as the negative control and those incubated with 200 µM of H2O2 (for 5 min, 4°C) served as the positive control.After incubation, cells were centrifuged at 300 x g for 5 min at 4°C and washed twice with PBS.Supernatant was discarded and pellet was placed on ice and resuspended in LMP agarose.All treatments were performed in duplicates.

Comet assay (single-cell gel electrophoresis)
The alkaline comet assay was carried out according to the procedure of Singh, McCoy, Tice, and Schneider (1988) [41].Conventional microscope slides were precoated by dipping in a solution of 1% normal melting point (NMP) agarose (Sigma-Aldrich, MO, USA) and dried overnight.After treatment, cells were centrifuged (300 x g, 5 min, 4ºC) and the supernatant was removed.The pellet was resuspended in RPMI medium and the cells were counted.Approximately 1 x 10 6 cells/mL were mixed with 100 µL 0.5% low melting point agarose (LMP) and placed on a precoated slide.After solidification of the agarose, the slides were immersed in fresh ice-cold lysis solution (2.5 M NaCl, 0.1 Na2EDTA, 10 mM Tris-Cl, 10% DMSO, 1% sodium lauryl sarcosine, 1% Triton X-100, pH 10) overnight at 4ºC.Alkaline denaturation was performed in an electrophoresis buffer solution (1 mM Na2EDTA and 300 mM NaOH, pH ≥ 13) for 20 min.Electrophoresis was carried out in a chilled electrophoresis buffer for 20 min (25 V, 300 mA, 4ºC).The slides were then washed three times for 5 min in 0.4 M Tris-Cl, pH 7, treated with ethanol for another 5 min and dried.For comet analysis, slides were stained with DAPI (5 µg/mL).A total of 100 comets (50 cells from each of the two replicated slides) were scored visually in five classes according to tail size and intensity (from undamaged 0; to maximally damaged 4) as depicted in Figure 5   DNA nicking assay was performed using supercoiled pUC19 plasmid DNA by the method of Prakash et al. (2007).A reaction mixture containing different concentrations of plant extracts (10-100 µg/mL) and pUC19 plasmid DNA (0.5 µg) was incubated for 10 min at room temperature, followed by the addition of Fenton's reagent (30 mM H2O2, 50 µM ascorbic acid and 80 µM of FeCl3).The reaction mixture was then incubated for 30 min at 37 ºC and the DNA was analyzed on a 1% agarose gel.

Cell proliferation assay
Cells were resuspended in a diluted solution of trypan blue and counted by a binocular inverted microscope, MOTIC AE30, using Neubauer chambers.The cell number was calculated according to the formula: number of counted cells × 10 4 /mL.The cells were then plated in 96-well plates at a density of 11,000 cells/well and incubated overnight.The cells were treated with A. × cornutum and A. cepa methanolic extract at a concentration of 100 µg/ml in a complete medium (in triplicate) for 4, 24, 48 and 72 h.Then, the MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide] assay was performed in such a manner that after the treatment with onion extracts, the cells were incubated with 0.5 g MTT/L at 37°C for 2 hours.After that, the medium was removed and dimethylsulphoxide (10% DMSO) was added and incubated for another 10 min at 37°C while shaking.The degree of formazan formation, an indicator of living and metabolically active cells, was measured photometrically at 570 nm.The data was calculated in relation to the untreated control (100%) from three independent measurements.

Statistical analysis
Results are expressed as mean values with depicted standard deviation.Microsoft Excel Student t-test was used to analyze data and to discriminate statistically significant results.

Conclusion
This was the first known thoroughly comprehensive study of A. × cornutum phytochemical composition and biological activity.HPLC analysis revealed two major quercetin conjugates (quercetin 3,4'-diglucoside and quercetin 4'-monoglucoside) as the most abundant flavonols in A. × cornutum and A. cepa extracts.Additionally, we successfully identified three anthocyanins in both onions, while the forth anthocyanin, delphydin glucoside, was observed only in A. × cornutum.Overall, A. × cornutum showed slightly higher concentrations of all identified phenolic compounds that might be a cause of a higher bioactivity profile accounted for this onion.
Our results clearly show that both onions have strong protective effects on DNA molecule as we have proven with several in vitro experiments.Both onions have also shown strong antiproliferative activity on tested human cancer cell lines.Susceptibility of glioblastoma cells to both extracts was higher than that of a breast cancer cell lines.Antiproliferative activity of the tested methanol extracts could be mediated by induction of apoptosis, alterations of the cell cycle or some other mechanism, and further research is needed to clarify the exact mechanism(s) of their anti-proliferative activity in vitro.Therefore, it can be concluded that A. × cornutum and A. cepa methanolic extracts have comparable composition and concentrations of phenolic compounds, antioxidant, antigenotoxic and antiproliferative effects.This research demonstrated that A. × cornutum and common onion might be considered as an important source of natural antioxidants that might have a beneficial protective effect on human health.

Figure 4 .
Figure 4. Cytotoxic effect of A. × cornutum and A. cepa methanolic extracts on two human cancer cell lines: beast and glioblastoma cancer cells.The antiproliferative effect of A. × cornutum (A) and A.cepa (B) after 4, 24, 48 and 72 h of exposure.The percentage of metabolically active cells is expressed in comparison to the untreated control.The data are presented as a mean value ± SD of the three independent experiments.The statistically significant difference of glioblastoma cell line is represented as *p < 0.001 for A. × cornutum and A. cepa.
. DNA damage index was determined according to the equation: DI = 1n1 + 2n2 + 3n3 + 4n4, where DI is damage index in arbitrary unit (AU), n1-n4 are the number of comets with damage levels 1, 2, 3 and 4. DI values can indicate various situations, from all undamaged cells (class 0; 0 AU) to highly damaged cells (class 4; 400 AU).

Table 1 .
HPLC quantification of flavonols and anthocyanins in A. cepa and A. × cornutum extracts.