Cytotoxic effect of Vernonanthura polyanthes leaves aqueous extracts

Vernonanthura polyanthes, popularly known as assa-peixe, is a medicinal plant that has been widely used by Brazilian Cerrado population for treatment of diseases without a detailed evaluation of their effectiveness, toxicity, and proper dosage. Thus, more studies investigating the safety of V. polyanthes aqueous extract before the use are needed. The purpose of this study was to evaluate the toxicity, cytotoxicity and genotoxicity of V. polyanthes leaves aqueous extract using the Artemia salina and Allium cepa assays. For the A. salina assay, three groups of 10 larvae were exposed to V. polyanthes leaves aqueous extract at the concentrations of 5, 10, 20, 40, and 80 mg/ml. For the A. cepa assay, 5 onion bulbs were exposed to V. polyanthes leaves aqueous extract at 10, 20, and 40 mg/ml, and then submitted to macroscopic and microscopic analysis. As result it was identified a toxicity and cytotoxicity of V. polyanthes dependent on the extract concentration. The A. salina assay suggests that the concentration of 24 mg/ml of the V. polyanthes extract is able to kill 50% of naupllis; while the A. cepa assay suggests that V. polyanthes leaves aqueous extract is toxic at concentrations higher than 20 mg/ml; however the cytotoxic effect in A. cepa root cells was observed at 40 mg/ml of the extract. It is important to say that the V. polyanthes leaves aqueous extract concentration commonly used in popular medicine is 20 mg/ml. Thus, the popular concentration used is very close to toxicity limit in A. salina model (24 mg/ml) and is the concentration which showed toxic effect in A. cepa root cells (20 mg/ml). No genotoxic activity of V. polyantes leaves aqueous extract was observed in the conditions used in this study. Because of the antiproliferative action and no genotoxic activity, V. polyanthes leaves aqueous extract may present compounds with potential use for human medicine. However more detailed studies need to be performed to confirm this potential.


Introduction
Medicinal plants have played an important role in the prevention, treatment and also cure of different diseases. The plants used in medicine, food supplements, cosmetics, and other health related products has increased along the time (Jones et al., 2004). Approximately 50% of all drugs currently in clinical trials are derived from plants (Shakya, 2016) and there is speculation that more than two billion people worldwide resort to medicinal plants to treat diseases (Lambert et al., 1997).
The bioactive phytocompounds present in medicinal plants include an array of compounds, such as tannins, lignans, coumarins, quinones, stilbenes, xanthones, phenolic acids, flavones, flavonols, catechins, anthocyanins, and proanthocyanins. However, it is important to say that those same compounds also can cause adverse issues for human health, if bad administered. For example, rosemary (Cardoso et al., 2014), and guaco (Dalla Nora et al., 2010), commonly considered medicinal plants, reveals toxic effects dependent on the dose.
The present study is designed to appraise the toxicity and cytogenotoxicity potential of a folk medicinal plant, Vernonanthura polyanthes (Spreng.) A.J. Vega & Dematt. known commonly as assa-peixe. The genus Vernonia went through a taxonomical reclassification and the South America species were segregated in 22 new genera, including Vernonathura (Robinson, 1999). The taxonomic classification of Vernonia and Vernonathura genus are yet complex and unclear (Martucci et al., 2104). But the compounds found in species of booth genus have been usually used in popular medicine. Vernonanthura polyanthes presents oval leaves, rough and hairy spear-shaped and occurs primarily in Brazilian Cerrado biome (Vega and Dematteis, 2010). The white or pink inflorescences are arranged at the apices of the branches in small capitula (Alves and Neves, 2003).
In relation to the toxicity, previous study identified that the V. polyanthes aqueous extract showed no toxic, genotoxic, and antigenotoxic activity in the experimental conditions tested using the wing somatic mutation and recombination test SMART/wing (Guerra-Santos et al., 2016). On the other hand, other study demonstrated that the V. polyanthes leaves ethanolic extract was toxic to mice with LD 50 of 2.78 g/kg (Temponi et al., 2012). That contradictory information needs to be best analyzed to verify the V. polyanthes toxicological potential. Actually, several different methodologies should be used to ensure accurate information about the toxicity of medicinal plants.
Here we selected two different methodologies to evaluate the V. polyanthes toxicological potential: Allium cepa and Artemia salina assays. Those methods are very useful for toxicity and cytogenotoxicity screening and they are simple, inexpensive and minimum laboratory facilities are required for its performance (Ribeiro et al., 2016;Neves et al., 2014;Leme and Marin-Morales, 2009). Moreover, the results obtained using those in vivo models show a high degree of conformity with the results obtained from mammalian assays (Freires et al., 2017;Fedel-Miyasato et al., 2014;Leme and Marin-Morales, 2009).

Plant collection and preparation of aqueous extract
Samples of Vernonanthura polyanthes (Spreng.) A.J. Vega & Dematt leaves were collected from specimes located at Campus Henrique Santillo, Universidade Estadual de Goiás, Anápolis, Goiás, Brasil (16° 23' 0,16"S 48° 56' 37,8"W, 1073 m) in November 2015 and plants no present flowers at the moment of collect. The collected leaves were identified and a voucher specimen was deposited in UEG Herbarium under identification number 10512. After the identification, the leaves were dried under air circulation and powdered using a knife mill. The aqueous extract solution was prepared by infusion with boiling water. After a rest of 20 min, the solution was filtered using the Millipore membrane filter (0.45 μm pore size) and briefly used. The V. polyanthes leaves aqueous extract concentration commonly used in popular medicine is 20 mg/ml (ANVISA, 2011, Guerra-Santos et al., 2016. Then, it was tested five different concentrations in A. salina assay (5, 10, 20, 40, and 80 mg/ml) and three different concentrations in A. cepa assay (10, 20, and 40 mg/ml).

Toxicity evaluation using Artemia salina
Eggs (25 mg) from A. salina were acquired from local pet shops and hatched at 25 -30°C in saline water (pH 8.0). After 24 h, the newly hatched larvae were collected and used in the lethality assay according to procedures described by Meyer et al. (1982). Groups of 10 larvae were exposed to V. polyanthes leaves aqueous extract (5, 10, 20, 40, and 80 mg/ml) diluted in natural seawater and, after 24 h, the survival rates (%) were recorded. The negative control (NC) was the saline water used and the positive control (PC) was a potassium dichromate (K 2 Cr 2 O 7 ) solution (5 mg/ml). Three independent experiments were performed in experimental triplicate. The averages for the number of dead individuals of A. salina were compared by ANOVA followed by Tukey's test, using the statistical program SISVAR (Ferreira, 2011). In addition, Statistica software was used to calculate the minimum lethal concentration (LC 50 ).

Mitochondrial activity evaluation using Allium cepa
The mitochondrial respiratory chain activity using A. cepa was adapted from the method reported by Prajitha and Thoppil (2017). Briefly, A. cepa bulbs were growth in water for 72 h. The germinated bulbs were exposed to V. polyanthes aqueous extract at 10, 20, and 40 mg/ml for 3 h. Sterile mineral water and sodium azide (1 mg/mL) were used as negative (NC) and positive (PC) controls, respectively. To evaluate the mitochondrial activity, the roots with the same size were stained with 2,3,5-triphenyl tetrazolium chloride (TTC, 0.5%) for 15 min at 35 °C in the dark. After the staining, the roots images were captured in stereoscope and the quantitative analyses were performed measuring complex triphenyl formazan released after washing with 95% ethanol for 30 min in dark. The detection of the dye was made at 490 nm in spectrophotometer, and the experiment was performed in triplicate. The treated and control groups were compared using one way analysis of variance (ANOVA) of ranks, followed by Tukey's test. P values of less than 0.05 (P<0.05) were considered statistically significant.

Cytotoxic and genotoxic evaluation using Allium cepa assay
Allium cepa bulbs were germinated in water for 48 h, and after exposed to three different concentrations of V. polyanthes leaves aqueous extract (10, 20, and 40 mg/mL). The controls were: sterile mineral water (NC) and sodium azide (PC, 1mg/ml). After treatment, the roots were collected for the microscopic analyses. Then, the treated roots were collected and placed in Carnoy's fixative solution (3:1 ethanol:glacial acetic acid v/v) at -2 °C. The roots were subsequently hydrolyzed in HCl 5N for 1 min and washed with sterile mineral water. After this, the roots were placed on a microscope slide and one drop of acetic acid 45% was added. Then, the roots were macerated with rusty needles. A coverslip were placed on the material and a pressure was made to help separate the cells. After, the cover slips were removed after freezing in liquid nitrogen and stained with Giemsa (5%) for 5-10 min. The microscope slides were observed under an optical microscope. The following parameters were observed: mitotic index (MI); frequency of chromosomal aberrations (CA); and nuclear abnormalities (NA). To obtain those information 1000 cells were evaluated from each bulb, totalizing 5 bulbs per treatment (5,000 cells). The data were compared using t test and p < 0.05 was considered statistically significant.

Results
The phytochemical screening of V. polyanthes leaves aqueous extract revealed the presence of phenolic compounds, flavonoids, terpenoids, tannins and coumarins. Alkaloids and saponins were not identified in this samples. The total phenolic content, flavonoids and tannins were determined, indicating 27.73 ± 0.11%, 2.35 ±0.00% and 4.81 ± 0.02% in V. polyanthes leaves aqueous extract, respectively. After first screening for secondary metabolites, the V. polyanthes leaves aqueous extract was submitted to Artemia salina and Allium cepa assay to assess its toxicological potential.
The lethality of V. polyanthes leaves extract on brine shrimps nauplii was directly proportional to the concentrations of the extract used in this work (Figure 1). All A. salina nauplii were alive after 24 h of experiment in the negative control, and all nauplii exposed to the positive control died. The LC 50 obtained for V. polyanthes leaves aqueous extracts was 24 mg/ml. This means that the medicinal popular dose (20 mg/ml) is very close to toxicity limit identified for Artemia salina at the present study (LC 50 = 24 mg/ml). However, the toxicity of V. polyanthes leaves aqueous extract was smaller than the toxicity of positive control (LC 50 = 0.19 mg/ml) which Averages were compared using one-way ANOVA followed by Tukey´s test. Averages followed by the same letters do not differ statistically among themselves.
is a drug commonly used in this assay able to cause nauplii death at low concentration (potassium dichromate, 5 mg/ml). In addition, statistical analysis showed no significant differences in survival rate between negative control and 5 mg/ml of V. polyanthes, showing that in this concentration this extract is not toxic to A. salina nauplii. On the other hand, no statistical differences were observed between positive control and 40, and 80 mg/ml of V. polyanthes, which means that in this concentration the V. polyanthes extract is highly toxic.
The effect of V. polyanthes aqueous extract on the viability of A. cepa cells was also investigated using a mitochondrial respiratory chain activity assay. For this, the conversion of TTC to insoluble red colored triphenyl formazan by the meristematic cells was measured in the root. We observed a concentration-dependent decrease in mitochondrial activity in the onion roots treated with V. polyanthes aqueous extract (Figure 2A). The results obtained are in agreement with the A. salina assay and showed a significant decrease in viability of A. cepa cells at 20 and 40 mg/ml. In this assay, the V. polyanthes aqueous extract was not toxic at 10 mg/ml.
With the objective of investigating the possible mechanism involved in toxicity of the V. polyanthes aqueous extract we performed a microscopic analysis of the A. cepa meristematic cells. The results showed no statistical difference in MI between 10 and 20 mg/ml and the negative control (Table 1), which suggested that on those concentrations the V. polyanthes extracts is not cytotoxic. However, it was observed a decrease in the MI rate in cells subjected to the concentration of 40 mg/ml (Table 1), which suggest that the highest concentration of the V. polyanthes aqueous extract used in this work was able to inhibit the cell division and so is cytotoxic.
The genotoxicity was evaluated by the frequency of chromosomal aberrations and nuclear abnormalities. V. polyanthes leaves aqueous extracts at all concentrations showed very few chromosomal aberrations and nuclear abnormality frequency, not differing statistically from the negative control (Table 1). Three main types of chromosome aberrations were recorded: stickiness, bridges and lagging chromosomes (Figure 3). This result indicates that V. polyanthes leaves aqueous extracts is no genotoxic at all the concentrations tested in this work. Averages were compared using one-way ANOVA followed by Tukey´s test. Averages followed by the same letters do not differ statistically among themselves; (B) Representative pictures of Allium cepa root tips after TTC staining.

Discussion
In this study, the in vivo A. salina and A. cepa assays were used to assess the toxic, cytotoxic, and genotoxic potential of the V. polyanthes leaves aqueous extract. This species has been empirically used by Brazilian Cerrado population for treatment of several diseases without a detailed evaluation of their effectiveness, toxicity, and proper dosage. Then, more studies investigating the safety of V. polyanthes aqueous extract are needed to safeguard the wellbeing of its users. Trying to assess the effects of V. polyanthes in the form it is actually administered we prepared plant extract in the same way as the therapeutic teas used by the general population In addition, we used different concentrations from that of the popular use to evaluate the effect of concentration in tea toxicity.
The A. salina assay suggests that concentrations over 24 mg/ml are toxic for shrimp and kill 50% of nauplii; while the viability of A. cepa assay suggests that concentrations higher than 20 mg/ml cause significant cell death. Since the concentration commonly used in popular medicine is 20 mg/ml, it is necessary a precaution in the administration of this infusion and more studies using human cells need to be performed to detect the proper dosage for humans.
A previous study using Drosophila as a model to evaluate the toxic potential of V. polyanthes leaves aqueous extract showed no toxic, genotoxic, and antigenotoxic activity in the experimental conditions tested using the wing somatic mutation and recombination test SMART/wing (Guerra-Santos et al., 2016). Different from that, at the present work we identified a toxic and cytotoxic activity of V. polyanthes. This disagreement can happen, once there is no single test to detect the full spectrum of toxicity of the substance (Dearfield et al., 2002).
The cytotoxic and antiproliferative activity of the V. polyanthes leaves aqueous extract observed in the Same letters represent no significant difference using one-way ANOVA followed by Tukey´s test (p < 0.05). NC: negative control; PC: positive control; and 10, 20, and 40 mg/ml of V. polyanthes leaves aqueous extract. present work can be due to the phytochemicals present in this species. We identified the presence of phenolic compounds, flavonoids, terpenoids, tannins and coumarins. The total phenolic content was 27.73 ± 0.11%. Phenolic compounds consist of minimum of one aromatic ring and can be classified into several groups such as phenolic acids, flavonoids, coumarins, tannins, lignins and stilbenes (Shahidi and Yeo, 2016). The phenolic compounds have attracted attention of scientific community due to their effective antioxidant capacity (Shahidi and Yeo, 2016). We identified that V. polyanthes leaves aqueous extracts has 2.35% of flavonoids and 4.81% of tannins.
Up to now, it was identified more than 8000 flavonoids, which have been presented different biological activities . The wide functional and pharmacological activities of flavonoids are mainly due to their structural variations and chemical modifications, such as acylation, methylation, glycosylation and hydroxylation (Malla et al., 2012). Those wide conformational variations allow the flavonoids to exhibit different activities, such as: antioxidant, antitumor, antiallergic, anti-inflammatory, antibacterial, antiarteriosclerotic and antiestrogenic (Bhouri et al., 2011;Zhu et al., 2015). Then, flavonoids can interfere in biological cell activity depending on their nature and concentration.
Regard to terpenes, in plants, in high concentration they can be toxic, and are important to defense against herbivores and pathogens (Paduch et al., 2007). For human helthcare, the terpenoids exhibit important pharmacological activities, such as antimicrobial, antiviral, anti-fungal, anti-inflammatory, antihyperglycemic and antiparasitic activities, as well as are effective in chemoprevention and chemotherapy (Paduch et al., 2007). There are some terpenoids such as paclitaxel and its derivate which are commonly used as anticancer drugs (Perveen, 2018). However, complementary studies are necessary to determine the correlation of the flavonoids and terpenoids with the antiproliferative cell activity.
In addition, our results showed that V. polyanthes aqueous extract present no genotoxic effect on A. cepa root. As discussed before, the presence of flavonoids and terpenoids could account for the lack of mutagenic effects caused by infusions of these extract (Teixeira et al., 2003;Perveen, 2018). Similarly to what was observed in this work, Costus spiralis , Sambucus australis (Tedesco et al., 2017), Luchea divaricata (Frescura et al., 2012) and Solidago microglossa (Bagatini et al., 2009) medicinal extracts also presented cytotoxic but no genotoxic or mutagenic effect using A. cepa test.
The scientific literature has been shown that many of the agents used in cancer therapy are derived from natural compounds extract from plants, such as: vinca alkaloid family isolated from Catharanthus roseus (Noble 1990;Stanton et al., 2011), and the taxanes paclitaxel originally identified from plants of the genus Taxus (Baird et al., 2010). Vernonia genus which is a sister genus of Vernonanthura also has different species with antitumor properties, such as V. zeylanica (Abeysinghe et al., 2019), V. extensa (Thongnest et al., 2019), V. cinerea (Pouyfung et al., 2019), V. amygdalina (Yedjou et al., 2018), V. anthelmintica (Ito et al., 2016), V. condensate (Thomas et al., 2016), V. divaricata (Lowe et al., 2014) and V. scorpioides (Pagno et al., 2006). In the similar way to those species, the identification of cytotoxic and antiproliferative properties of V. polyanthes leaves aqueous extract may suggest that this species also could be able to inhibit the tumor cells proliferation, however this capacity must be yet deeply investigated with other assays using cancer cells.