Ethanolic extract of gabirobeira (Campomanesia adamantium) leaves reduces proliferation of osteosarcoma cells in vitro

Osteosarcoma is the most commonly diagnosed malignant bone tumor in humans, with a higher incidence in children and young people. It is highly aggressive and has a high metastatic potential. Its treatment is based on both chemotherapy and surgical intervention. However, currently used chemotherapeutic agents, such as doxorubicin, have several adverse effects on the patient. Therefore, there is a growing demand for new chemotherapeutic agents that stimulate new researches, such as those involving compounds extracted from plants, such as the gabirobeira. In this study, we aimed to evaluate the cytotoxic effects of ethanolic extract, both crude and ethyl acetate, of gabirobeira leaves on osteosarcoma cells in vitro. Cytotoxicity was evaluated using the Trypan blue exclusion method and the IC50 values were calculated using the tetrazolium reduction method. The ethanolic extract of gabirobeira leaves showed a cytotoxic effect on osteosarcoma cells in vitro. The group treated with the crude extract at 1. 0μL mL concentration for 48 hours showed higher cytotoxicity and the lowest IC50 value for this extract was found in the 24 to 48 hours interval. The ethanolic extract of gabirobeira leaves is cytotoxic for osteosarcoma cells. Keyword: plant extracts; tumor; Campomanesia adamantium. Received on March 25, 2020. Accepted on March 3, 2021.

Diagnosis of OS is based on medical history, laboratory, and complementary physical examinations, such as radiography and histopathology (Ritter & Bielack, 2010). Treatment is usually multimodal, including systemic chemotherapy and surgery (Mercatelli, Bortolotti, Bazzocchi, Bolognesi, & Polito 2018). The most commonly used chemotherapeutic agents are cisplatin (CDP), doxorubicin (DOX), methotrexate (MTX), and ifosfamide (IFO) that can be used as neoadjuvant (preoperative) and adjuvant (postoperative) therapy (Mercatelli et al., 2018;Harrison & Schwartz, 2017). Although important for the treatment of patients, these drugs may cause short-and long-term adverse effects, such as nausea, vomiting, alopecia, mucositis, nephrotoxicity, among others (Harrison & Schwartz, 2017). Moreover, its prognosis is reserved for patients with metastases. Survival rates at five years after OS diagnosis for patients with metastases are only 10% to 30%, and for patients without metastases 60% to 70% (Mercatelli et al., 2018).
The incidence of adverse reactions justifies the search for new antineoplastic agents. Recently, the number of researches towards the discovery of new chemotherapeutic agents has grown, often using plants and other natural sources (Zhang et al., 2013). In this regard, the Brazilian Cerrado is considered a source of biological resources (Silva, Boas, Barros, Rodrigues, & Siqueira, 2009). Its flora is rich in fruit species that have great therapeutic potentials, such as Campomanesia adamantium (Cambess. O. Berg), popularly known as the guabiroba, guavira, guariroba, guabiroba-do-campo, guabiroba-do-cerrado and guabiroba-lisa (Campos et al., 2017;Pascoal et al., 2014).
Campomanesia adamantium is a shrub, with thin and multiple branches, measuring up to 2m in height (Vieira et al., 2011;Amaral, Reis, Ressel, & Pinto, 2016). Its leaves are simple, opposite, oval, or elliptical. It blooms from September to November and has flowers that are white and solitary (Amaral et al., 2016). The fruits ripen from November to December, are round, dark to light green and yellow colors (Vieira et al., 2011;Amaral et al. 2016;Vallilo et al. 2006). They are edible by both animals and the local population and are used in food manufacturing of products such as juices, jams, liqueurs, and jellies (Pascoal et al. 2014;Vallilo, Lamardo, Gaberlotti, Oliveira, & Moreno 2006).
In traditional medicine, this plant is used as antidiarrheal, antimicrobial, antioxidant, antirheumatic, antitumor, as well as in urinary tract infections, reducing cholesterol levels and treating diabetes. Studies have demonstrated its antimicrobial, antiproliferative, anti-inflammatory and antinociceptive activity (Campos et al., 2017;Pascoal et al., 2014;Viscardi et al., 2017;Ferreira et al. 2013;Fernandes et al. 2015), besides the cytotoxic potential of its leaf extract. Chalcone cardamonin, isolated from the ethanolic extract of leaves, has been reported to have antiproliferative activity in prostate PC-3 cancer cells (Campos et al., 2017;Pascoal et al., 2014).
Therefore, in this study, we aimed to assess the in vitro cytotoxic effects of gabirobeira leaf extract on OS cells culture to identify compounds with antineoplastic potential and, possibly, demonstrate a viable alternative to the currently chemotherapy treatments.

Material and methods
Plant material and extraction -The crude ethanolic extract of gabirobeira leaves (EEG) was obtained from dry and fragmented leaves of C. adamantium using the maceration technique. The botanical material, in this case gabirobeira leaves, was obtained in the Universidade Federal de Goiás (UFG), Campus Jataí, Jataí, GO, Brazil (16° 17' 26.6" S, 048° 57' 21.1" W) and identified by the Prof. Dr. Luiza Francisca de Souza.
The leaves were washed and oven-dried at 35°C for seven days to eliminate moisture and stabilize the enzymatic content. Subsequently, they were ground in an electric knife mill. Extraction occurred by macerating the plant material in absolute alcohol at a rate of 2,200 mL of alcohol 480 g -1 of powder for 72 hours at room temperature. The extract was then filtered and concentrated in a rotary evaporator (Model Ika-Werke) at a constant temperature of 45°C.
To obtain the ethyl acetate extract (EAE), the crude extract was submitted to the solid-liquid extraction process. The extract was pelleted after being placed on the silica gel. Subsequently, it was placed on a chromatographic column with acetate added to the saturation point to obtain the ethyl acetate fraction.
Ethics -This study was approved by the Research Ethics Committee of the Universidade Federal de Goiás (CEP/UFG).
Cells and culture -OS cells (MG-63, BCRJ 0173, Lot 000888, Passage 104), from the ATCC (American Type Culture Collection -Manassas, VA, USA), were obtained at the Rio de Janeiro Cell Bank (UFRJ -Rio de Janeiro, Brazil). They were placed in a humidified CO 2 incubator at 37ºC and 5% CO 2 atmosphere. Cultivation in bottles was performed in Dulbecco's Modified Eagle Medium (DMEM) plus 20% fetal bovine serum, amphotericin B and L-glutamine solution (all from Cultilab, Campinas, Brazil) according to the method of Yu et al. (2014).
Experimental design -After the cell culture, the cells were quantified in the Neubauer chamber and seeded in 96-well plates containing 100μL DMEM medium at concentrations of 5x10 4 e 1x10 4 , respectively, for the evaluation of cell viability (CV) using the Trypan blue exclusion method and for the CV assay using the tetrazolium reduction method [MTT (3-(4,5-dimethyl-2-thiazolyl) -2,5-diphnyl-2H-tetrazolium]. The plate was placed in a humidified CO 2 incubator at 37ºC and 5% CO 2 atmosphere for 24 hours.
The medium was then discarded and the wells treated and not treated with the gabirobeira leaves extract (EEG and EAE) for 24 (G 24 ), 48 (G 48 ) and 72 (G 72 ) hours, at concentrations of 0.1 μL mL -1 , 1 μL mL -1 , 10 μL mL -1 and 100 μL mL -1 . Moreover, the negative control group (CG) was treated only with dimethyl sulfoxide (DMSO, Cultilab, Campinas, Brazil), aiming to verify whether cell death in the groups treated with the extracts is solely due to their effect since the extracts are diluted in DMSO.
After treatment with the extract, the medium was discarded and cells submitted to the following analyses: assessment of CV using the Trypan blue exclusion method and CV assay using the tetrazolium reduction method. Both assays were performed in three independent experiments and triplicates.
Cell viability using the Trypan blue exclusion method -For the assessment of CV using the Trypan blue exclusion method, the technique used was adapted from Mosman (1983) and Peres et al. (2015). After treatment, the wells were washed twice with 100 μL phosphate-buffered saline (PBS, Cultilab, Campinas, Brazil). Then, the cells were resuspended using 100 μL of trypsin (Cultilab, Campinas, Brazil) in each well. The plates were then kept in the humidified CO 2 incubator for five minutes, homogenized and thus, remained for a further five minutes in the incubator. Soon after, trypsin was inactivated with 100μL DMEM.
Subsequently, the material was centrifuged at 1400 rpm for 10 minutes and after that, the medium was discarded and cells resuspended in 1mL DMEM. From this new solution, 40 μL were transferred to microtubes microcentrifuge containing 40 μL Trypan blue (Trypan blue -Sigma-Aldrich, St. Louis, USA). Then, to assess the CV, 10 μL of this new solution was instilled in a slide to be read in the Luna Automated Cell Counter. Cytotoxicity (CT) was estimated by the equation: %CT = 100 -[(abs of treatment/abs of negative control) x 100] Where CT is cytotoxicity and abs is absorbance. We analyzed the values of each group using the Scott-Knott test. From the results, we calculated the CV, CT and cell size (CS) means for each group of exposure time (G 24 , G 48 , and G 72 ) and both extracts (EEG and EAE).
Thereafter, the optical density of the wells was quantified through a spectrophotometer (Awareness Technology Ine/Stat Fax 2100, 425-540 nm, Palm City, FL, USA) and the concentration value that inhibits 50% of cell viability (Inhibitory concentration -IC 50 in μL mL -1 ) was determined using the GraphPad Prism statistical program (GraphPad Software, San Diego, CA, USA).

Results and discussion
The Trypan blue exclusion method is used to evaluate the cytotoxicity by measuring cell viability. This analysis allowed the identification of non-viable cells. Viable cells did not allow the Trypan blue dye to penetrate the intracellular medium due to their membrane integrity. However, the non-viable cells, due to their damaged membranes, allowed the passage of dye and were stained blue.
From this assay, we calculated the CV, CT and CS (cell size) mean values at the concentrations used (0.1, 1, 10 and 100 μL mL -1 ) and at the three exposure times (G 24 , G 48 e G 72 ) of both extracts (EEG and EAE). Regarding the EAG, we found no statistical difference for the three parameters between the concentrations used for groups G 24 and G 72 , as well as the CS in G 48 . However, CV and CS for G 48 showed a statistical difference between the 0 μL mL -1 and 0.1 μL mL -1 concentrations in relation to the 1, 10 and 100 μL mL -1 concentrations, as shown in Table 1. For EEG, we found no statistical difference between the concentrations used (0.1, 1 , 10 and 100 μL mL -1 ) at the three exposure times (G 24 , G 48 and G 72 ) in any of the three analyzed parameters (CS, CV and CT), according to Table 2. Regarding CV and CT, groups G24 and G48 submitted to treatment with EEG showed higher CV and lower CT compared to EEG. However, the CV of G72 was lower and CT higher when submitted to the EAE. Moreover, regarding CV, G24 and G48 groups showed a statistical difference between the extracts, while G72 showed no difference. Regarding CT, only G48 showed a statistical difference between the treatments used. Regarding CS, groups G24, G48 and G72 treated with EEG had higher CS compared to groups treated with EAE. However, groups G24 and G72 showed no statistical difference between the extracts used, unlike G48, as shown in Figure 1.
Given these results, it appears that EEG has higher antiproliferative potential than EAE, because it showed a lower CV and higher CT. Moreover, when analyzing the cell groups treated with EEG, we verified that G48 showed higher potential than G24 and G72. Finally, among the concentrations used in G48, the concentrations 1 , 10 and 100 μL mL -1 showed significant results. However, we highlight the result at 1 μL mL -1 since it is the smallest concentration and therefore, in theory, has fewer adverse effects.
Moreover, the increase in CT, consequently the reduction in CV, progressively occurred for EAE. However, for EEG, a progressive increase of CT occurred only between G24 and G48, while between G48 and G72, the CT decreased. This fact, together with the absence of statistical difference between extracts in G72, may suggest that the cytotoxic effect of crude extract decreases with certain exposure time.
The concentration required to inhibit 50% of cells in vitro was determined by calculating the IC 50 values from TRM cell viability assay data using the GraphPad Prism statistical program. The TRM assay was one of the first methods to be accepted for cell viability and cytotoxicity. It is a colorimetric test, where the absorbance measurements are carried out at the end of the test. TRM uses tetrazolium, a yellow compound, that is cleaved by mitochondrial dehydrogenases and reducing agents present in formazan crystals of viable cells, a blue compound, insoluble in aqueous solution. Crystals accumulate in hydrophobic cytoplasmic domains, such as lipid droplets. Thus, tetrazolium acts as an indicator of intracellular reducing potential (Pascua-Maestro et al., 2018). IC 50 values are shown in Table 3. The IC 50 for EEG progressively increased with time exposure to extract. However, its G 48 value was within the proposed range of best cytotoxicity.
The high IC 50 value in G 72 corroborates the results found in the cytotoxicity evaluation, in which there was less cell death compared to G 48 . It reinforces the theory that the action of this extract decreases with the exposure time. Thus, it is necessary to investigate the antiproliferative activity of this extract by consecutive treatments throughout the exposure time.
A pertinent consideration for C. adamantium extract is the relationship between the plant material used and its cytotoxic potential. In the antiproliferative activity analysis of the leaves and roots aqueous extract in leukemia, we found IC 50 value of 40 and 80 μg mL -1 , respectively, in Jurkat cells (neoplastic T lymphocytes) (Silva et al., 2009). Thus, although the cytotoxic potential of gabirobeira leaves has already been demonstrated (Silva et al., 2009;Campos et al., 2017), other types of botanical materials are need to be studied.
Another crucial factor is the fraction of the extract used. EAE showed lower cytotoxic potential than EEG, suggesting that the compounds responsible for this effect are lower in the acetate fraction. The ethanolic extract of Mesona Palustris BL also showed higher cytotoxicity, IC 50 value of 146.0 μg mL -1 , compared to ethyl acetate, IC 50 value of 182.96 μg mL -1 (Widyaningsih, 2012). Figure 1. Graphical representation of cell size (CS), cell viability (CV) and cell cytotoxicity (CT) mean values obtained by the Trypan blue exclusion assay. Osteosarcoma cells were treated with ethanolic extract of gabirobeira leaves (EEG) and acetate ethanolic extract of Campomanesia adamantium leaves (EAE) at concentrations of 0.1 , 1, 10 and 100 μL mL -1 for 24, 48 and 72 hours. For the assay, three independent experiments in triplicates were performed for each extract. *Statistical difference between crude and acetate extracts. Table 3. Concentration values necessary to inhibit 50% (IC50 in μL mL -1 ), calculated through the cell viability using the tetrazolium reduction method (TRM). MG-63 osteosarcoma cells were subjected to treatment with ethanolic extract of gabirobeira leaves (EEG) and acetate ethanolic extract of C. adamantium leaves (EAE) at concentrations of 0.1, 1, 10 and 100 μL mL -1 for 24, 48 and 72 hours. For the assay, three independent experiments in triplicates were performed for each extract. The cell line employed in the analyses is another factor to consider. In leukemia, the aqueous extract of C. adamantium leaves showed no significant effect on peripheral blood mononuclear cells, while in Jurkat cells, it showed cytotoxic activity. The same study demonstrated that vinblastine promoted cell death in both peripheral mononuclear blood and Jurkat cells, but presented distinct IC 50 values, respectively 25.9 and 9.5 μg mL -1 (Silva et al., 2009).
On that basis, we emphasize the need to evaluate the cytotoxic potential of gabirobeira leaves through other fractions of the extract, as well as in other OS lines and other neoplasms. Besides, it is important to evaluate other types of botanical materials of this plant such as the roots, since the results suggest that this plant may constitute a new alternative to the currently used chemotherapeutic agents.

Conclusion
From the results obtained, we conclude that the ethanolic extract of gabirobeira leaves (Campomanesia adamantium (Cambess. O. Berg)) has a cytotoxic effect on MG-63 osteosarcoma cells, more pronounced after 48 hours of treatment.