PHYTOCHEMICAL SCREENING, ANTIOXIDANT, ANTI-CYTOTOXIC AND ANTICANCER EFFECTS OF GALINSOGA PARVIFLORA AND VERNONIA POLYANTHES (ASTERACEAE) EXTRACTS

The study was designed to investigate the chemical composition and the biological effects of G. parviflora and V. polyanthes ethanolic extracts in vitro. Total content of phenols, flavonoids and tannins was quantified by spectrophotometry; chemical characterization was permed by mass spectrometry (ESI (-) FT-ICR MS and APCI (+) FT-ICR MS analysis). Antioxidant activities were determined by FRAP and Fe2+ chelating methods. Extracts cytotoxicity was evaluated in human lymphocytes, sarcoma-180 (S-180) and human gastric adenocarcinoma (AGS) cells, by MTT assay. V. polyanthes presented higher total content of tannins and G. parviflora presented higher amount of phenols and flavonoids. Chemical characterization showed the presence of flavonoids, phenolic acids and sesquiterpene lactones in V. polyanthes extract, and steroids, phenolic acids and fatty acids (Poly Unsaturated Fatty Acids PUFA) in G. parviflora extract. V. polyanthes extract stood out in the Fe2+ chelation test. G. parviflora extract did not present outstanding antioxidant results in the tested protocols. Both species showed a tendency to promote cytotoxicity in human lymphocyte cells. Regarding the antiproliferative effect, both species were able to reduce SPhytochemical Screening, Antioxidant, Anti-Cytotoxic and Anticancer Effects of Galinsoga Parviflora and Vernonia Polyanthes (Asteraceae) Extracts International Journal of Research -GRANTHAALAYAH 85 180 cell viability and G. parviflora extract showed high antiproliferative potential in the assay with AGS cells. These findings reinforce the medicinal use of these plants, as well as suggest their potential use for the development of new drugs and for the treatment of cancers.


Chemical Analysis of Extracts Total Phytocompounds Content
Total content of phenolic compounds was tested according to the Zhang et al. [36], by Folin -Ciocalteu method. The total flavonoid and tannin content were assessed by the protocols of Zhishen et al. [37] and Pansera et al. [38], respectively. The readings were taken on an ELISA microplate spectrophotometer, at the wavelengths recommended for each protocol. All analyses were performed in triplicate and the standards used were the recommended in the protocols.

ESI (-) FT-ICR MS analysis
The ethanolic extracts were analysed by Mass Spectrometry with Fourier Transform Ion Cycle (FT-ICR MS) to determine the chemical profile. The samples were solubilized (1.0 mg.mL -1 ) in a methanol solution, which was infused at a rate of 2.0 µL/min in the negative mode electrospray (ESI). The Solarix model 9.4 T mass spectrometer, Bruker Daltonics, Bremen, Germany, was programmed to operate in a range of m/z 150 -1500. The conditions of the ESI source used in the analysis were: nebulizer gas pressure of 1.0 bar, capillary voltage 3.8 kV and capillary transfer temperature of 200 °C. The ion accumulation time was 0.010 s, and each spectrum was acquired by accumulating 32 scans with a 4M time domain (mega-point). Mass spectra were processed using Data Analysis software (Bruker Daltonics, Bremen, Germany).

APCI (+) FT-ICR MS analysis
For sample analysis, 1.0 mg of the extracts of V. polyanthes and G. parviflora were individually solubilized in 1.0 mL of methanol (99.5%, Vetec® Química Fina Ltda, Brazil). The samples were injected directly into the APCI (+) source at a flow rate of 20 µL.min -1 . The dynamic range of ion acquisition in the ICR cell was m/z 200 -1200.
Other APCI source parameters were: voltage in the capillary (cone): 2,100.0 V; end plate offset = -500 V; drying gas temperature and flow: 180 ºC and 4 L min -1 ; nebulizer gas pressure and temperature: 320 ºC and 2.0 bar; skimmer = 25 V; collision voltage = -2 V and corona discharge: 3000 nA. In ion transmission, the ion accumulation time in the hexapole (ion accumulation time) and TOF were 0.020 s and a range of 0.850 -0.900 ms, respectively. Each spectrum was acquired from the accumulation of 32 scans with a time domain of 4M (mega-point). Mass spectra were acquired and processed using Data Analysis software (Bruker Daltonics, Bremen, Germany).

Antioxidant Activity
Evaluation of the antioxidant activity of the crude extracts was performed by FRAP and Fe 2+ chelating activity. FRAP (Ferric Reducing Antioxidant Power) is also known as Iron Reducing Antioxidant Power test. Following the protocol of Rufino et al. [39], with modifications, FRAP reagent was obtained from the combination of 0.3 mM acetate buffer solution, 10 mM TPTZ (2,4,6-Tri (2-pyridyl) 1,3,5-triazine) solution and aqueous chloride solution ferric 20 mM. In 2.0 mL microtubes, it was added 30 µL of the samples, 90 µL of distilled water and 900 µL of the FRAP reagent. Microtubes were vortexed and incubated in an oven at 37 ºC for 30 minutes. Thus, 250 µL of this solution was added to a 96-well microplate, being performed the same with the blank, reaction control and standards gallic acid, ferrous sulfate and Trolox. The reading was performed on a spectrophotometer for ELISA microplate at 595 nm. The calculation of Antioxidant Activity (AA%) was performed using the equation below, with values expressed in EC50 (μg.mL -1 ).
Fe 2+ ion chelating activity test was performed as established by Tang et al. [40], with modifications. In a 1.5 mL microtube, it was added 1000 µL of the sample, 50 µL of FeCl2 and 200 µL of ferrozine. The microtube was vortexed, allowed to react for 10 minutes, the solution was placed in 96-well microplates and read on an Epoch ELISA spectrophotometer at 595 nm. Ascorbic acid, gallic acid and EDTA were used as standards. All tests were performed in triplicate and the calculations for assessing chelating activity (CA%) were based on the equation below, with values expressed in EC50 (µg.mL -1 ).
In vitro cytotoxicity Cancer cells Anticancer in vitro experiments were performed with sarcoma 180 (S-180) and human gastric adenocarcinoma cells (AGS; ATCC CRL-1739). Sarcoma-180 cells were acquired from the Banco de Células do Rio de Janeiro, Brazil, and the AGS cells were supplied by the Laboratório de Triagem Biológica de Produtos Naturais from UFES. S-180 cells were previous cultured with culture medium RPMI 1640 and AGS cells with DMEM medium, both cell lines supplied with 10% fetal bovine serum. Cells were seeded in 96-well microplate, S-180 at 2.10 5 cells.mL -1 and AGS at 6.10 4 cells.mL -1 in each well, and previous maintained at 37 ºC and atmosphere of 5% CO2.
After 24 hours, cells were treated with seven different concentrations of the extracts for 48h, starting from the initial concentration of 200 µg.mL -1 for V. polyanthes and 400 µg.mL -1 for G. parviflora. Assay was performed in triplicate and cell viability was assessed using the MTT reduction method. All protocols were in accordance to the Ethics Committee of Humans and Animals Use.

Human lymphocytes
To assess the cytotoxicity in healthy cells, it was used the protocol of Marullo et al. [41], with modifications by Dutra et. al. [42]. Peripheral blood was collected from a healthy donor, aged between 18 and 30 years, with free and informed consent. Possible donors with history of recent disease, exposure to radiation or drug use and alcohol ingestion thirty days before blood donating were excluded from donating. All protocols were approved by the Research Ethics Committee of the Universidade Federal do Espírito Santo Santo.
In order to compare the effects of extracts in cancer and healthy cells, human lymphocytes were cultured under the same growth conditions of S-180 and AGS cells. To evaluate the anti-cytotoxicity, human lymphocytes were treated with extract concentrations at 5.00, 25.00 or 50.00 µg.mL -1 and cisplatin at 50.0 µg.mL -1 . Following the protocol of pre-treatment, human lymphocytes received extract concentrations and after 24 hours from the incubation received cisplatin; simultaneous treatment, in which the extract concentrations and cisplatin were placed at the same time; and post-treatment, where the lymphocytes were initially treated with cisplatin and, after 24 h, received extract concentrations. Untreated cells were used as a negative control (NC) and cells treated with cisplatin were used as a positive control (PC). To evaluate the perceptual of cytotoxic damage reduction it was used the formula from Serpeloni et al., adpted for Dutra et al. [42], [43]: Where "A" is the cell group treated with cisplatin; "B" is the cell group treated with plant extracts more cisplatin; and "C" is the negative control group of cells.

MTT assay
The method is based on the reduction of MTT ((3-(4,5-dimethylthiazol-2yl) -2,5-diphenyl tetrazoline bromide) in a violet-colored product (formazan) by the mitochondrial enzyme succinate-dehydrogenase, a reaction that can only occur in viable cells. Thus, after 24h or 48h of the last treatment, human lymphocytes, S-180 and AGS cells were subjected to the cell viability test using the MTT assay.
Microplates were centrifuged at 860 rcf for 10 minutes and the supernatant was discarded. 20 µL of MTT at 5 mg.mL -1 were added to each well and incubated for 3 hours at 37 ºC and an atmosphere of 5% CO2. After the period, the plates were centrifuged at 860 rcf for 5 minutes, the supernatant was discarded and 100 µL of DMSO was added. The reading was performed on an Epoch ELISA spectrophotometer at 595 nm. The experiments were carried out in triplicate and the evaluation of cell cytotoxicity was calculated by the equation below and expressed as a percentage of viable cells (% VC) and IC50 (µg.mL -1 ):

Statistical analysis
Results were presented as mean ± standard error of the mean (SE). After verifying the normality of the data, the comparison of means was performed by one-way analysis of variance (ANOVA), followed by Tukey's test (p<0.05). In order to establish relationships between total phenols, flavonoids and tannins content, antioxidant activities, anticytotoxicity and anticancer effects of the extracts of G. parviflora and V. polyanthes, principal component analysis (PCA) and Pearson correlation were performed. For PCA analysis and Pearson correlation, the results of S-180 anticancer effects at the concentration of 50.0 μg.mL -1 were used, since this concentration corresponds to the best results for anticancer activity.

Chemical analysis of V. polyanthes and G. parviflora extracts
In a comparison, the evaluation of the total content indicates that G. parviflora presents higher values of phenolic compounds and total flavonoids and that V. polyanthes stands out for the total tannin content ( Table 1). Compounds or classes of metabolites presented in extracts were proposed based on the ions generated, number of unsaturations and rings (DBE) and data from the literature. Fig. 1A and Table 2 show the chemical composition of the extract of the species V. polyanthes by the negative ESI ionization source, and Fig. 2 A and Table 3 indicate the results obtained from the APCI (+) FT-ICR MS. In addition, Fig. 1B and Table 4 summarizes the chemical composition of G. parviflora by the negative ESI ionization source, and Fig. 2B and Table 5 summarizes the results obtained by the positive APCI ionization source.      Chanaj-Kaczmareck et al. [9] quantified phenolic compounds and total flavonoids in G. parviflora hydromethanolic extract and, compared to our study, obtained low values of phenolic compounds (29.55 mg GAE.g -1 ) and flavonoids (2.48 mg QE. g -1 ). It may occur due to the use of different solvents or because the use of ethanol seems to favor the extraction of the phenolic and flavonoid compounds of G. parviflora. In addition, differences in climatic and edaphic variables and possible differences in genotypes may have contributed to the variation in the content of secondary metabolites [44], [45], [46].
In previous investigations, chemical analyzes of ethanol extracts from V. polyanthes leaves showed lower levels of flavonoids compared to our study [ [23], [47]. In contrast, in the study by Rodrigues [48] with ethanolic extract of the leaves of V. polyanthes, it was described levels of phenolic and flavonoids compounds higher than those presented in our investigation. This finds reinforces the possible influence of different study conditions on the total contents of phenolic and flavonoid compounds.
V. polyanthes is rich in flavonoids, phenolic acids, chlorogenic acids and sesquiterpene lactones [49], [50], [51]. Our results is corroborated by the study of Martucci [21] and Martucci and Gobbo-Neto [52], which indicate the presence of several compounds, such as dicaffeoylquinic acid, monocafeoylquinic acid and luteolin-7-O-glucuronide in V. polyanthes extract ( Table 2). Other studies also report substances similar to those identified in our study, such as the dicaffeoylquinic acid and the flavonoid luteolin [48], [50], and phenolic acids derived from cinnamic acid, such as ferulic acid and caffeine [53].
Other studies indicate that G. parviflora exhibits a diversity of flavonoids derived from caffeic acid, steroids, among others [8], [55], [56] which is in accordance to the study of Bazylko et al. [57], with ethanolic extract, that identified phenolic acids, such as caffeic, caffeoylquinic and dicafeoylquinic acid, using HPTLC; and also in accordance to the results of Dudek et al. [55], with hydrophilic extract of aerial parts, that identified substances derived from caffeic acid, such as 5-O-caffeoylquinic acid (chlorogenic acid), 1, 3-O-dicaffeoylquinic acid, 3, 5-Odicafeoilquin, using spectrometric techniques. Meanwhile, the findings presented in our study (Table 5) corroborate those reported by Mostafa et al. [56] and Anwar et al. [58] that indicate the presence of stigmasterol, β-sitosterol and β-sitosterol in G. parviflora extracts. In addition, caffeic acid derivatives found in G. parviflora has been identified as an important protective factor for dermal fibroblasts against oxidative stress induced by ultraviolet radiation (UVA), by activating the cellular antioxidant system in a study by Parzonko and Kiss [59].
Polyunsaturated fatty acids (linolenic and linoleic acid) and some monounsaturated (palmitoleic, palmitic and ricinoleic acid) were also identified in G. parviflora (Table 4), which has not yet been reported for this species. Public health authorities consider nutraceuticals as powerful instruments in maintaining health against nutritional problems and chronic diseases, with improvement in the individual's quality of life [60]. Thus, the presence of fatty acids in G. parviflora indicates the promising uses of this herb as a nutraceutical.
Despite the controversies, omega-3 fatty acid supplementation has been recommended and studies have reported satisfactory results regarding its regular dietary intake, with favorable effects on triglyceride levels, coagulation and blood pressure, heart rate, cancer prevention, reduction in the incidence of arteriosclerosis and in the prognosis of symptomatic heart failure or myocardial infarction [64], [65], [66].

Antioxidant activity of V. polyanthes and G. parviflora extracts
Antioxidant activities of V. polyanthes and G. parviflora extracts, by FRAP and Fe 2+ chelating activity is shown in Table 6. G. parviflora demonstrates lower antioxidant activity, when compared to V. polyanthes, in both tests. Following statistical analysis, V. polyanthes extract presented antioxidant activity comparable to the standards in the Fe 2+ chelating test; while the same was not observed for G. parviflora extract in tested conditions. It was observed in the study of Studzinska-Sroka et al. [16], with hydroalcoholic extract of G. parviflora, considerable antioxidant activity in the FRAP assay, with EC50 = 498.20 µg.mL -1 . It was also verified in the G. parviflora extract, by liquid chromatography (UPLC-PDA), the presence of phenolic acids, such as chlorogenic, caffeic and isovanyl acids, and 4-hydroxybenzoic. In conclusion, the authors stated that the application of G. parviflora extract of in cutaneous lesions allowed the healing of wounds and exhibited antioxidant, anti-inflammatory and hyaluronidase inhibitory activities. In addition, studies indicate significant antioxidant activity of V. polyanthes extracts, such as in FRAP assay, different to the reported in our study, and correlate antioxidant effects to the levels of total phenolics and flavonoids, such as rutin and quercetin [23], [47], [58]. V. polyanthes extract stood out in the Fe 2+ chelating test (Table 6), a condition similar to the observed in two fractions of V. amygdalina (ethanolic polyphenol and acetone eluate) that exhibited high chelating power [67].
In a comparison, V. polyanthes and G. parviflora extracts showed similar anticancer activity against S-180 cells; however, the extract of V. polyanthes was more cytotoxic for AGS cells than the extract of G. parviflora. In addition, the cytotoxicity induced by extracts in human lymphocytes did not differ.
Following the protocols of the anti-cytotoxicity, the results showed that in the pre and simultaneous treatment, both species were not able to avoid the cytotoxic damage induced by cisplatin. In the post-treatment, it was observed, for the highest tested concentrations of V. polyanthes extract, a tendency to reverse cisplatin induced damage. For G. parviflora extract, only the concentration of 25µg.mL -1 showed a promising action against cisplatin damage (Table  7). Cisplatin is a highly reactive molecule used in the treatment of cancer due to the ability to bind to proteins and phospholipid membranes. In addition, cisplatin can interact with RNA and DNA, forming adducts that may inhibits the replication, transcription or interrupt the cell cycle and activation of apoptosis, generating genotoxic and cytotoxic effects [68], [69]. Our results suggest that V. polyanthes and G. parviflora act on human lymphocytes repair mechanism, positively interfering with the maintenance of cellular homeostasis, even after interaction with cisplatin.
To our knowledge, there are no studies involving simultaneously V. polyanthes extract, human lymphocytes, S-180 and AGS cells. However, studies with other species of the genus Vernonia have demonstrated the promising uses of this group of herbs in trials with several cancer cell lines. An example is the investigation of Amuthan et al. [70], comparing the cytoprotective activity of V. cinerea crude aqueous extract and its fractions in normal HEK293 kidney cells and HELA cell lines, against cisplatin-induced cytotoxicity. Also investigating anticancer activity, Siew al. [71], showed that the methanolic extract of V. amygdalina leaves showed antiproliferative activity against several cancer cell lines. Traditionally, V. amygdalina leaves are already used in popular Indian culture for the treatment of cancer, and based on the excellent results that this species, the authors claim that further studies are needed to understand its potential in the treatment of cancer.

Explorative analyses: correlations between phytochemicals, antioxidant activity and anticancer effects
Results obtained in different assays were correlated by PCA and Pearson correlation coefficient. The first principal component (F1) accounted 76.05% and second principal component (F2) accounted 15.67%, with a total variance of 91.72% (Fig. 3). Total phenol, flavonoid and tannins content, FRAP, Fe 2+ chelating, anticancer activity against AGS cells and pre-treatment assay were the variables that dominated PC1. S-180 anticancer, simultaneous and post-treatment assays were the variables that dominated PC2. These finds suggest that anticancer activity against AGS cells was correlated with total phenol and flavonoid content and FRAP and Fe 2+ chelating antioxidant activity, while anti-cytotoxic activity in the pre-treatment protocol was correlated to the total tannins content. Following PCA analysis, it was not possible establish correlations among chemical content of extracts and human lymphocytes cytotoxicity, simultaneous and post-treatment assays. In summary, V. polyanthes seems to be correlated with cytotoxicity in AGS cells and anti-cytotoxicity in the post-treatment protocol, while G. parviflora seems to be correlated with cytotoxicity in sarcoma 180 cells, and pre and simultaneous treatment. In Pearson correlation analysis (Table 8), values followed by positive sign suggest a directly proportional relation between the factors. Thus, the positive correlation between phenol and flavonoid total content and FRAP and Fe 2+ chelating assays suggest the increasing of antioxidant activity, as well as, the positive correlation between phenol and flavonoid and AGS cytotoxicity suggest the increasing of the anticancer effect. In addition, FRAP and Fe 2+ chelating antioxidants are strongly correlated to the anticancer effects against AGS cells. Anti-cytotoxicity protocols were used to evaluate the ability of extracts inhibit or prevent cytotoxic damage induced by cisplatin. Similarly, considering positive correlations between the total tannin content and pre and simultaneous treatment tests suggest that an increase in tannins availability interferes positively with cell viability of human lymphocytes, as well as, tannins content interfere with human lymphocytes cell viability.  Marzouk and Abd Elhalim [72], using NMR technique, found in the extract of V. leopoldii aerial parts compounds previously not reported in the literature, characterized as sesquiterpene lactones, triterpene tetracyclic and apigenin-7-O-glucide and luteolin-7-O-glucide. According to the authors, chemical and pharmacological investigations have shown that the sesquiterpene lactones found in plants of the genus Vernonia present antitumor activities, and some sesquiterpene lactones may exhibit strong cytotoxicity against tumor cell lines.
Sesquiterpene lactones identified in V. polyanthes by the APCI (+) FT-ICR MS technique (8β-2 methylacryloyloxy-isohirsutinolide, piptocarpine A and 10α-acetoxy-8α-methylacryloyloxy-1 α-13-O-acetate or 1βmethoxyhirsut -O-acetate) may have contributed to the antiproliferative effects on S-180 and gastric adenocarcinoma cells. Considering that sesquiterpene lactones are chemical markers of the Asteraceae family and of the genus Vernonia, it was expected to find compounds of this chemical class in the extracts of the herbs investigated in our study. In the review of Ghantous et al. [73], sesquiterpene lactones are led clinical trials of cancer, as well as, the authors relate the good performance of these substances in cancer clinical trials to the structureactivity of sesquiterpene lactones (lipophilicity and geometry).
Similar to studies with V. polyanthes, to our knowledge, there are not studies in the literature that relate G. parviflora extracts to the cytotoxicity in human lymphocyte cells, S-180 and AGS. The study of Bazylko et al. [57] evaluated the potential cytotoxicity of aqueous and ethanolic extracts of G. parviflora and its protective effect against damage caused by UV (ultraviolet) irradiation in human skin fibroblast cells. The authors concluded that the ethanolic extract was cytotoxic, presenting intense effect on the generation of reactive species in fibroblasts after UVB irradiation exposure. On the contrary, the aqueous extract exhibited protective activity in fibroblasts, preventing the decrease in proliferative activity and the increase in apoptosis caused by UVA and UVB irradiation, by the inhibition of EROS generation.
Parzonko and Kiss [59] investigated the photo-protective effects of two derivatives of caffeic acid isolated from aerial parts of G. parviflora (2,3,5-or 2,4,5-tricafeoilaltrarico acid (TCA) and 2,4-or 3,5 -dicafoilglicaric acid (DCG)), in human dermal fibroblasts. In conclusion, the study clearly demonstrated that the derivatives of caffeic acid found in G. parviflora, in particular TCA, protect cells against damage caused by UVA radiation.

CONCLUSIONS AND RECOMMENDATIONS
Despite belonging to the same family, our analyses suggest that the studied plants present different content of phenols, flavonoids and tannins. It was identified Poly Unsaturated Fatty Acids (PUFA) in G. parviflora extract, being the first record of such compound for this plant species. Regarding anticancer effects, both species were able to reduce S-180 cell viability, and V. polyanthes extract showed high antiproliferative potential against AGS cells. In the anti-cytotoxic assay, V. polyanthes was more efficient in repairing cytotoxic cisplatin-induced damage. Chemical content of G. parviflora and V. polyanthes seems interfere with antioxidant and cytotoxic effects exhibited for the extracts.
There are several drugs for the treatment of cancer. In this scenario, our results reinforce the use of G. parviflora and V. polyanthes for medicinal and nutraceutical purposes, as well as suggest their potential use for the development of new drugs for the treatment of cancer. However, further investigations are needed to verify the biological activities of these plants.

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