In vitro and in vivo assessment of genotoxic activity of Petiveria alliacea

Petiveria alliacea L. (Phytolaccaceae) is an Amazonian shrub used in traditional medicine for many purposes. This study investigated the genotoxicity of the hydroalcoholic extract of P. alliacea (EHPa). In addition, we conducted pharmacognostic characterization and phytochemical investigation. Phytochemical screening and thin layer chromatography (TLC) were used to determine the chemical composition of the extract. Genotoxicity was evaluated using an in vitro comet assay in human lymphocytes (50, 100, and 250 mg/ml) and an in vivo micronucleus assay in mice orally treated with the extract (50, 100, and 250 mg/kg b.w., p.o.). The phytochemical screening identified the metabolites reducing sugars, alkaloids and saponins. TLC analyses identified organosulfur compounds in EHPa. Comet assay data showed that EHPa induced exacerbated DNA damage at 100 mg/ml treatments (EHPa 100 mg/ml: TL 219.13 ± 48.38 mm, % DNA-t 59.33 ± 18.98%, TM 134.99 ± 61.34, OM 87.48 ± 32.98) as compared with negative control (Roswell Park Memorial Institute (RPMI): TL 37.79 ± 24.79 mm, % DNA-t 9.73 ± 6.56%, TM 4.86 ± 5.79, OM 6.90 ± 5.80; p<0.01). These effects were similar to positive control (hydrogen peroxide). The micronucleus assay, however, showed that EHPa induce slight genotoxicity in vivo. However, EHPa induced an exacerbated DNA damage in vitro, but this effect was slight in the animal model. These in vitro effects may be related to the chemical composition of the extract.


INTRODUCTION
Petiveria alliacea L. (Phytolaccaceae) is an Amazonian indigenous perennial shrub, slender and erect, which reaches about 1 m in height (Rocha et al., 2006).This variety of names is directly related to its large geographical distribution that includes countries in tropical America, Africa, Europe, and the Caribbean, West Indies, and India (Alonso, 2007).
The DNA of all living organisms is constantly exposed to injury.Genomic and cellular integrity is maintained by effective maintenance and repair mechanisms.Despite the existence of such pathways, some substances found in medicinal plants and herbal drugs can damage the genetic material and provide genotoxic stress triggering cell aging and carcinogenic processes (Abdelmigid, 2013).International organizations, such as the International Committee on Harmonization (ICH), have recommended the investigation of the potential DNA damaging effects of medicinal plants, through in vitro and in vivo experimental models (ICH, 2008).In this regard, these in vitro and in vivo models in the field of genetic toxicology can identify and measure the DNA damage, including frequency of DNA adducts, DNA strand break (single or double), DNA cross-linking, mutations or chromosomal aberrations (Abdelmigid, 2013).
Although P. alliacea posses great pharmacological potential, some studies have pointed out its genotoxic hazards.In this sense, Hoyos et al. (1992) have demonstrated that ethanol extract of P. alliacea induces DNA damage evaluated by sister-chromatid exchange assay in in vitro and in vivo models, and also Soares et al. (2014) reported that different aqueous extracts of the plant caused single-and double-strand breaks in plasmid DNA from an Escherichia coli strain.However, both investigations evaluated the genotoxicity of low concentrations of P. alliacea extracts in nonhuman models.
In this spite, this study was conducted to predict the potential genotoxic hazards of high concentrations of the hydroalcoholic extract of P. alliacea (EHPa) through in vitro and in vivo experimental models.In addition, the pharmacognostic parameters of the dry powder of the aerial parts of P. alliacea and the phytochemical composition of EHPa was examined.

Plant
The aerial parts of P. alliacea (stem, leaves, and flowers) were collected in April, 2011 at Acará, Pará, Brazil (S 01°29′09.0/W48°17′94.8).A voucher specimen was identified by the Botanist consultant Mário Augusto G. Jardim and was deposited in the herbarium of the Museu Paraense Emílio Goeldi (MPEG) under registry MG 94354.After collection, the plant material was first dried at room temperature (25°C/24 h) and subsequently, was dried in a forced-air oven to 40°C for 15 h.P. alliacea dry aerial parts were then powdered in a knife mill (Willey, Marconi MA 680 model).

Pharmacognostic characterization
The pharmacognostic characterization of P. alliacea was conducted from its aerial parts and encompassed the following determinations: granulometric analysis and mean particle diameter, loss on drying (oven/infrared moisture balance), total ash content, foaming index, and content of extractable matter (Allen et al., 2000;Brasil, 2010;World Health Organization 2011).

Extract preparation
The hydroalcoholic extract was choosen because previous studies of our group have demonstrated its potential effects on central nervous system when comparing with traditionally preparations (Andrade et al., 2012;Silva et al., 2015).Crude EHPa was prepared through the maceration method.Thus, plant drug (930 g) was extracted with 70% ethanol (v/v) for 7 days.The hydroalcoholic extract was filtered and evaporated under vacuum (40°C, 167 mbar, and 120 rpm) to provide 92.07 g of EHPa (1 g of extract equivalent to 10.1 g of dried aerial parts).

Preliminary phytochemical screening
The phytochemical screening of EHPa was conducted according to Table 1.Reactions, reagents, and positive controls used in preliminary phytochemical screening of EHPa according to Barbosa et al. (2001).

Metabolites
Reagent

Thin-layer chromatography of EHPa
Thin-layer chromatography (TLC) was performed on a precoated 20 × 20 cm silica gel 60 F254 plate (Merck, Germany) for the identification of sulfur compounds in EHPa as described by Wagner and Bladt (2001).Briefly, aliquots of EHPa dissolved in dichloromethane (10 mg/ml) were spotted on the plate.TLC analysis was developed using toluene-ethyl acetate (100:30, v/v) as mobile phase.As positive controls, the organic phase of aqueous extracts of Allium sativum and Allium cepa fresh bulbs extracted with dichloromethane were used.After chromatography, the plate was sprayed with 10 ml of vanillin-glacial acid reagent (VGA) and heated at 110°C for 3 min.Then the spots were observed under visible light.The retention factor values (Rf) of each spot were calculated using the equation: Rf = distance the spot moved above the origin/distance the solvent front moved above the origin.

Cell culture and sample preparation for comet assay
Peripheral human blood samples were obtained from healthy volunteers (non-smoking and non-drinking males and females between 20 and 45 years old) after approval by the Research Ethics Committee of UFPA (CAEE 0154.0.073.000-11,Registry 165/11 CEP-ICS/UFPA).The blood samples were obtained as previously described by Albuquerque et al. (2015) for lymphocyte culture.In brief, 0.3 ml of venous blood were added to 4 ml of RPMI-1640 medium containing 20% of Fetal bovine solution (FBS) and 50 μg/ml phytohemagglutinin (PHA) to stimulate lymphocyte proliferation.The mixture was then incubated under humidified environment at 37°C and 5% CO2 for 24 h.The lymphocytes cultures (4.5 ml) were then treated with different concentrations of EHPa (50, 100, and 250 mg/ml; 0.4 ml) for 3 h at 37°C.As a negative and positive control, the lymphocytes were treated with RPMI-1640 medium and H2O2 at 35%, respectively, for 30 min at 37°C.Afterwards, cell viability was assessed with trypan blue dye exclusion test.Only cell samples whose viability was over 90% were measured by the comet assay (single cell gel electrophoresis).

DNA damage using comet assay
The comet assay was performed as described by Singh et al. (1988) and modified by Anderson et al. (1994).Treated lymphocytes, as already described were mixed with 0.5% low melting-point agarose, and the mixture (100 µl) was pipetted onto pre-coated slides with 1.5% normal melting point agarose.The drop containing the cells was placed under a glass cover slip (24 mm × 24 mm) and left at 4°C for 5 min.The cover slips were gently removed and the slides were treated with a lysis solution (2.5 M NaCl, 100 mM EDTA, 100 mM TRIS, 1% Triton X-100 and 10% DMSO, pH 10.2) for 24 h at 4°C.The slides were immersed in freshly prepared electrophoresis alkaline buffer (2.5 M NaOH, 100 mM EDTA, pH>13) and electrophoresis was performed using an electric field of 23 V/cm for 20 min.After electrophoresis, the slides were neutralized with distilled water for 5 min, fixed with ethanol for 3 min and air-dried.The microscope slides were then stained with ethidium bromide (20 µg/ml), and the cells were visualized and photographed under a fluorescence microscope (ZEISS AxioCam HRc) using a 510 to 560 nm emission/barrier filter at 400× magnification.A total number of 100 cell images (50 cells/slide) were collected randomly for each treatment and analyzed using Comet Score 1.6 (Tritek) software.The parameters of DNA damage analyzed included tail length (TL), percentage of DNA in the tail (% tail DNA), tail moment (TM), and Olive moment (OM).To avoid interference of additional DNA damage this experiment was conducted in a dark room and performed in duplicate.

Micronucleus assay
The micronucleus assay protocol was modified from Schmid (1975).Briefly, 30 mice were separated into five groups of six mice each.Thus, three groups were administered by oral gavage with EHPa at 50, 100, and 250 mg/kg b.w., and two groups received physiological saline (0.9% NaCl) and cyclophosphamide (50 mg/kg b.w., i.p.).After 24 h the animals were euthanized by cervical dislocation, and their femurs were dissected immediately for bone marrow extraction.The bone marrow samples were homogenized twice with 1 ml of fetal bovine serum (FBS) and centrifuged for 10 min at 1000 rpm.The supernatant was discarded, and the pellet was re-suspended in 0.5 ml of FBS.The cells were dropped on microscope slides and air-dried.Afterward, the slides were fixed with methanol and stained in two different concentrations of Leishman stain to allow visual differentiation between immature polychromatic erythrocytes (PCE) and mature normochromatic erythrocytes (NCE): (a) Leishman at 100% for 3 min., and (b) Leishman-distilled water (1:6, v/v) for 15 min (Krishna and Hayashi, 2000).The microscope slides were air-dried and then analyzed under light microscopy (Olympus, CX41) at a magnification of 1000×.To evaluate the genotoxicity of EHPa, the ratio of micronucleated polychromatic erythrocytes (MNPCE) per 2000 PCE was considered for each treatment and animal.To evaluate the cytotoxic effects of EHPa, 200 erythrocytes (PCE and NCE) were counted for the PCE/(PCE + NCE) ratio following Hayashi et al. (2000).

Data analysis
All results are expressed as mean ± SD values.For the in vitro comet assay, statistical analyze was performed using one-way analysis of variance (ANOVA) followed by Tukey post hoc test for multiple comparisons.The data was analyzed by Graph-Pad Prism software version 5.0 (San Diego, USA).For the in vivo micronucleus assay, normality test followed by Kruskal-Wallis test for nonparametric data were performed using Assistat software version 7.6 beta (UFPB, Brazil).In all experiments, differences were considered statistically significant if p < 0.01.

Pharmacognostic characterization
Table 2 summarizes the results obtained from pharmacognostic characterization of the dry powder of the aerial parts of P. alliacea.As shown, the powder possessed a mean particle diameter (D 50 ) of 0.275 mm.The powder was considered as moderately coarse according to the classification of pharmacopeia.All values for moisture content and total ash content, using the Brazilian Pharmacopoeia (Brazil, 2010) techniques, were within the limits established for plant drugs.The foaming index and the content of extractable matter corresponded to the values for qualitative assays of the plant drug.

Phytochemical screening of EHPa
Preliminary phytochemical screening of the EHPa was regarded as positive for the presence of metabolites, reducing sugars, alkaloids, and saponins.

Thin-layer chromatography of EHPa
TLC of the organic phase of A. sativum and A. cepa extracts showed the development of organosulfur compounds spots as previously reported by Wagner and Bladt (2001).TLC chromatogram of EHPa revealed the presence of three blue or blue-gray colored majority spots having R f values of 0.55, 0.87, and 0.95, after spraying with VGA reagent.These spots were similar to the positive controls used and indicate the presence of sulfur-compounds in EHPa (for example, thiosulfinates).

In vitro comet assay
The comet assay was performed for in vitro evaluation of some potential DNA damage of EHPa on human lymphocytes.According to the parameters for DNA damage analyzed (Figure 1), EHPa exhibited a dual effect on DNA.Treatment with EHPa at the concentrations of 50 and 100 mg/ml induced significant DNA damage when compared to the negative control (EHPa 50 mg/ml: TL 231.

In vivo micronucleus assay
The micronucleus assay was carried out in rodents for assessment of genotoxic and cytotoxic activities of EHPa.Table 3 summarizes the micronucleated polychromatic erythrocytes (MNPCE) frequencies and PCE/(PCE + NCE) ratio -ratio between immature polychromatic erythrocytes (PCE) and immature polychromatic erythrocytes (PCE) plus mature normochromatic erythrocytes (NCE); ratio in bone marrow of mice treated with different doses of EHPa.As for the genotoxic parameter, our results demonstrated that EHPa at doses of 50 and 250 mg/ml induced changes in MNPCE frequency compared to negative control (p < 0.01); however, these values were at least threefold lower than the positive control (p < 0.01).With regards to the cytotoxicity parameter, EHPa did not decrease significantly the PCE/(PCE + NCE) ratio when comparing mice treated with the negative control for all tested doses (p > 0.01).As observed with cyclophosphamide treatment, the PCE/(PCE + NCE) ratio was statistically lower than negative control and EHPa 50 mg/ml (p < 0.01), but not significant, compared to the EHPa 100 mg/ml and EHPa 250 mg/ml treatment groups.

DISCUSSION
Petiveria alliacea is an important Amazonian herb with several pharmacological properties already reported.Recent investigations of our group have demonstrated that ethanolic preparations of P. alliacea possess anxiolytic, antidepressant and mnemonic effects upon central nervous system (Andrade et al., 2012;Silva et al., 2015).Although previous studies have demonstrated that P. alliacea induces genotoxic hazards in low concentrations, there is a lack of information about the effects of high concentrations of ethanolic preparations in both in vitro and in vivo models.
As alluded to earlier, we performed some pharmacognostic tests aiming to determine the quality control parameters of the powder obtained from P. alliacea.With regards to pharmacognostic characterization, powdered dry aerial parts of P. alliacea showed quality control parameters which were within the limits set forth in Brazilian Pharmacopoeia.These results are also similar to those observed by Audi et al. (2001) for loss on drying and content of extractable matter tests.
Phytochemical screening encompasses a set of qualitative tests for the identification of metabolites present in medicinal plants preparations.In this study EHPa showed positive results for the presence of reducing sugars, alkaloids, and saponins.Carillo et al. (1997) also identified the presence of alkaloids and saponins in ethanol extract of P. alliacea, differently however, we did not identify tannins, flavonoids, and sesquiterpene lactones.On the other hand, Villar et al. (1997) did not identify saponins and alkaloids in P. alliacea extracts.
Thin layer chromatography was also performed in order to identify the presence of organosulfur compounds in EHPa.TLC chromatograms demonstrated the presence of organosulfur compounds similar to Kubec and Musah (2001) with a spot of R f = 0.55.Other sulfur compound spots identified in the chromatogram may belong to the class of thiosulfinates and other polysulfides, when compared to positive controls A. cepa and A. sativum spots (Wagner and Bladt, 2001).In this regard, several studies have reported the identification of organosulfur compounds in EHPa, including polysulfides (for example, dibenzyl trisulfide) and thiosulfinates (Benevides et al., 2001;Hernández et al., 2014;Kubec et al., 2002Kubec et al., , 2010;;Rosado-Aguilar et al., 2010;Urueña et al., 2008;Cifuentes et al., 2009).
In addition to pharmacognostic and phytochemical investigations, the evaluation of toxicity of medicinal plants and their derivatives is an important step in predicting potential hazards to human health (Abdelmigid, 2013;Netto et al., 2006).In the acute toxicity assay (data not shown), we did not observe any death in animals treated with EHPa even at the maximum dose administered (1000 mg/kg b.w.).Hernández et al. (2014) evaluating the acute toxicity of a P. alliacea fraction (doses ranged between 5 to 2000 mg/kg b.w., i.p.), calculated an LD 50 of 1545 mg/kg b.w.
The possible genotoxic effects of P. alliacea was evaluated through the in vitro comet assay in the present study; through the in vitro comet assay it was observed that EHPa significantly induced DNA damage of human lymphocytes.However, the EHPa induced high DNA damage in low concentrations (50 to 100 mg/ml), while the greatest concentration (250 mg/ml) induced a small DNA damage.This different effect could be due to the synergistic and antagonistic relationship between some substances in the extract, like thiosulfinates (Xiao and Parkin, 2002), that can lead to an oxidative stress imbalance and can also activate DNA repair pathways, reducing the genotoxic effects (Cooke et al., 2003).In this context, Hageman et al. (1997) reported that the water extract of raw garlic and two organosulfur compounds, diallyl sulfide and S-allylcysteine, were able to inhibit the DNA adduct formation induced by benzo[a]pyrene by antioxidant mechanisms, such as increase of the glutathione S-transferase activity and reduction of ROS.In addition, other studies reported that another antioxidant compound, such as phenolic compounds, mainly caffeoylquinic acid derivatives, from root extracts of Rhaponticum carthamoides may stimulate in CHO cells to repair oxidative DNA damage and protect DNA by increased antioxidant gene and protein expression levels (SOD2 and CAT) regulating intracellular antioxidant capacity (Skała et al., 2016).
In this regards, reactive oxygen species (ROS) and free radicals generation may play a pivotal role in DNA damage (Cadet et al., 1999;Del-Rio et al., 2005;Subhashinee et al., 2005).Previous studies have demonstrated that P. alliacea exerted both pro-oxidant and antioxidant effects on in vitro and in vivo models (Desmarchelier et al., 1997;Andrade et al., 2012).Desmarchelier et al. (1997) observed that preparations obtained from aerial parts of P. alliacea (1 mg/ml) presented slight anti-oxidant activity with lowest TRAP values (aqueous and dichloromethane extracts: <100 µM Trolox; methanol extract: 275 µM Trolox).On the other hand, our group also reported that animals orally treated with 900 mg/kg of P. alliacea L. hydroalcoholic extract exhibited pro-oxidant status (Andrade et al., 2012).In this study, it was shown that the extract was able to increase Trolox equivalent antioxidant capacity (TEAC), nitric oxide (NO) and malondialdehyde (MDA) levels, as well as increase methemoglobin formation in the human plasma.These data suggest that P. alliacea extract has pro-oxidant action and can lead to cell death (cytotoxicity) in vivo, because the increase of MDA is used as a marker of lipid peroxidation and cell damage caused by the action of reactive species in the body, as well as oxidative processes in the membrane (Janero, 1990;Del-Rio et al., 2005).In addition, in vitro and in vivo studies have reported that the MDA is able to interact with nucleic acid bases, as deoxyguanosine and deoxyadenosine, to form several different adducts, which may be mutagenic and these were quantified in several human tissues (Marnett, 1999a;Marnett, 1999b;Lykkesfeldt, 2007).Thus, high levels of reactive species and MDA formation induced by P. alliacea may be one of the mechanisms that lead to the high EHPa-induced DNA damage in vitro observed in this study.
In order to confirm the results from in vitro comet assay the in vivo micronucleus assay was performed using mouse bone marrow.In this study, only the treatment with EHPa at the concentrations of 50 and 250 mg/kg b.w.induced small changes in the frequency of MNPCE compared to negative control.In this regard, the micronucleus assay has been employed for genotoxicity and mutagenicity detection of substances that can cause cytogenetic damage, thus the increased MNPCE frequency can be related to cancer, though the MN can be a target of carcinogenesis (Bonassi et al., 2006).On the other hand, the comet assay was used to detect genotoxic effect of compound by measuring its DNA damage at single cell (Hartmann et al., 2003).Therefore, the present data showed that P. alliacea induces a slight mutagenic effect in vivo, but a high genotoxic effect that was detected by Comet assay.Thus, the data suggest that P. alliacea is associated with high genotoxicity, but a low risk for carcinogenesis, as described by Lee et al., (2015).These different effects detected by in vitro (comet assay) and in vivo (micronucleus assay) assays, also was reported by other studies, as Schauss et al. (2015) that showed that the Dahurian larch tree derived lavitol led to DNA damage in marrow cells, liver and blood, but did not increase the number of binucleate cells containing micronuclei.These effects can be explained by the concentrations/doses used in both assays.As well established, herbal derivatives, including plant extracts, possess poor lipid solubility or improper molecular size, which results in poor absorption and consequently poor bioavailability (Kesarwani and Gupta, 2013).In this regard, the activity of some enzymes on P. alliacea compounds could decrease their bioavailability on tissues, as well as the time of administration of the plant extract, and as mentioned before, the possible synergistic and antagonistic interactions between the compounds presents in the EHPa and their effects on oxidative balance.These findings may suggest that the oral administration route can allow some substances present in the plant extract to suffer from drug biotransforming reactions involving hepatic enzymes, which can facilitate their excretion and consequently the decrease of plasmatic concentrations.In spite of this, our results suggest that at least a concentration of 50 mg in plasma level could trigger DNA damage, as observed through comet assay.
Furthermore, other factors that justify the differences obtained in the in vivo versus in vitro experiments cannot be excluded.These factors include the DNA repair mechanisms present in biological environment, namely base excision repair, nucleotide excision repair, mismatch repair, direct repair, and recombination repair, which could repair some DNA damage induced by EHPa in animal model (Ramos et al., 2011).Furthermore, it was hypothesized that the DNA damage induced by EHPa can also lead to cell death because it is already wellreported that P. alliacea extracts exhibited cytotoxic activity against normal fibroblasts and peripheral blood mononuclear cells (Urueña et al., 2008), and several tumor cell lines, such as melanoma, leukemia, and lymphoma (Pérez-Leal et al., 2006;Rossi, 1990).Regarding cellular toxicity, our data showed that none of the doses led to cytotoxic effects.However, there was a tendency to increase the cytotoxicity with increasing concentration of extract, as seen in Table 3. Urueña et al. (2008) observed that an aerial parts fraction of P. alliacea exhibited low toxicity to normal fibroblasts and peripheral blood mononuclear cells in the presence or absence of phytohemagglutinin stimulation (IC 50 of 440, 151, and 121 µg/ml, respectively).Differing from our results, previous studies using the crustaceans Artemia salina and Thamnocephalus platyurus have demonstrated cytotoxic activity in P. alliacea extracts (Desmarchelier et al., 1996;Berger et al., 1998;Mayorga et al., 2010).These differences could be due to the different variables involved in the preparation of P. alliacea extracts (that is, plant part, period of collection, nature of solvent, method of extraction, and time of extraction).
In this context, Hernández et al. (2014) also reported other mechanisms by which the P. alliacea induced 4T1 cells apoptosis, such as caspase-3 activation and DNA fragmentation without mitochondrial membrane depolarization, this was also shown by Urueña et al. (2008) using a leukemia cell line.Accordingly, Hartmann et al. (2003) and Azqueta and Collins (2013) also reported that the cytotoxicity is associated with increased levels of DNA damage, which leads to the generation of false positive results.In addition, a previous study showed that a P. alliacea ethanol extract was able to induce sisterchromatid exchanges in human lymphocytes at the final concentration of 100 µg/ml, as compared with the positive (N-ethyl-N-nitrosourea) and saline control groups (Hoyos et al., 1992).
Therefore, the variation in these factors may yield extracts with different chemical composition which could explain the different pharmacological profiles found in the cytotoxic evaluation.The susceptibility of each cell model employed in these studies may also explain these differences.

Conclusion
Our data suggest that P. alliacea induces DNA damage in vitro, but this effect was slight in an in vivo model.The different effect of the EHPa in in vitro model may be related to the chemical composition of the extract, which includes thiosulfinates compounds.Further investigations on the mechanism of the DNA damage induced by P. alliacea are required to understand if it could trigger deleterious effects, such as an induction of mutations and carcinogenic processes.

Figure 1 .
Figure 1.In vitro genotoxic effects of P. alliacea.The parameters tail length (A), percentage of DNA in the tail (B), tail moment (C) and Olive moment (D) were used as markers for DNA damage in human lymphocytes through comet assay.All values are represented as mean ± SD. # p < 0.01 compared to the negative control (RPMI); *p<0.01 compared to the positive control group (H2O2); and **p < 0.01 compared to greater EHPa concentration.

Table 2 .
Pharmacognostic characterization of the dry powder of aerial parts of P. alliacea.
In vivo micronucleus assaysExperimental animalsAll experimental protocols were conducted in accordance with the National Institute of Health (Institute of Laboratory Animal Resources) for animal research and were approved by the Committee for Animal Care and Use of the Universidade Federal do Espírito Santo (UFES, Approval number: 074/2011).Eight-week-old adult male Swiss mice (Mus musculus) weighing between 25 and 35 g were obtained from the animal colony at UFES.The animals were housed in polypropylene cages, maintained under ad libitum food and water diet, and controlled laboratory conditions (22 to 25°C) with an artificial 12 h light/dark cycle.