Chemical characterisation and toxicity assessment in vitro and in vivo of the hydroethanolic extract of Terminalia argentea Mart. leaves

Abstract

Ethnopharmacological relevance: Terminalia argentea Mart. (Combretaceae), known mainly as “capitão” is a native tree, not endemic, that occurs in the Amazon, Caatinga, Cerrado and Atlantic Forest in Brazil. Leaf infusion is popularly mentioned by riverine communities that inhabit the microregion of Northern Araguaia (Mato Grosso, Brazil) for the treatment of gastric ulcer, bronchitis and haemorrhage. Considering the wide medicinal use, lack of studies that evaluate the safety of use and the scarcity of phytochemical studies of T. argentea leaves, this work was carried out with the objective of evaluating the toxicity of the hydroethanolic extract of the leaves of T. argentea Mart. (HETa) in experimental models in vivo and in vitro, as well as to advance the phytochemical analysis of HETa. Materials and methods: HETa was prepared by macerating the leaf powder in hydroethanolic solution. Phytochemical characterisation was carried out by thin-layer chromatography (TLC), high-performance liquid chromatography (HPLC) and mass spectrometry through direct flow infusion coupled with electrospray ionization and ion-trap analyzer (DFI-ESI-IT-MS analyses) The contents of phenols, flavonoids and phytosterols were analysed by colorimetric methods. Cytotoxicity was assessed by the Alamar blue assay on Chinese hamster ovary epithelial cells (CHO-K1) and human gastric adenocarcinoma cells (AGS). In vitro genotoxicity of HETa (10, 30 or 100 μg/mL) was assessed by micronucleus (MN) and comet tests using CHO-K1 cells. The acute toxicity assessment was performed by oral administration of HETa in single dose Swiss mice (males and females) up to 2000 mg/kg and sub-chronic toxicity by daily oral administration of HETa (50, 200 and 800 mg/kg) in Wistar rats for 30 days. The parameters related to the clinical and toxicological observations were determined every 6 days and at the end of the treatment the blood was collected for biochemical and haematological analysis, and some organs were removed for macroscopic and histopathological analysis. Results: Preliminary phytochemistry and TLC analysis of HETa revealed the presence of phenolic compounds (18.8%), flavonoids (10.8%), saponins, tannins and phytosterols (19%). The HPLC data revealed the presence of gallic acid, rutin, ellagic acid, catechin, quercetin and kaempferol. In the analysis by DFI-ESI-IT-MS, the presence of gallic acid, rutin, ellagic acid and quercetin was confirmed and identified caffeic acid, quinic acid, galloylmucic acid, quercetin xyloside, quercetin rhamnoside, quercetin glucoside, caffeoyl ellagic acid, quercetin galloyl xyloside, terminalin, quercetin galloyl glucose, corilagin, quercetin digalloyl xyloside, quercetin digalloyl glucoside, punicalin and punicalagin. HETa showed no cytotoxic effect on CHO-K1 and AGS cells. In the MN assay, HETa increased the number of MNs and nuclear buds (NBUDs) in binucleate cells at the three concentrations tested and the nucleoplasmic bridges (NPBs) number at 30 μg/mL. In the comet test, HETa (10 and 100 μg/mL) alone showed a genotoxic effect on CHO-K1 cells. In pre-treatment, HETa at all concentrations tested prevented DNA damage induced by H2O2. In co-treatment with H2O2, HETa showed genotoxic effects at the three concentrations, and post-treatment DNA damage in exposed CHO-K1 cells to H2O2 was repaired in 22.5% with 10 μg/mL HETa. In the acute toxicity test, the HETa did not cause death in the mice, being verified only by piloerection and reversible in 2 h in males and in 4 days in females. No macroscopic changes were observed in the analysed organs. In the sub-chronic toxicity test, the HETa did not cause death in the rats after 30 days and the few changes were: absolute (103/mm3) and relative (%) values of basophils increased by 477.8% and 423% (p < 0.001), respectively, with 50 mg/kg; reduction in feed intake (23.6%, p < 0.01) only on day 18; total cholesterol concentration (13.1%, p < 0.05) and relative heart weight (13.2% %, p < 0.05) at a dose of 800 mg/kg. These effects were not dose-dependent nor followed by clinical signs and symptoms of intoxication, nor of macroscopic and histopathological changes in the organs of animals treated with HETa. Conclusions: The results demonstrated that HETa had no cytotoxic in vitro effects for CHO-K1 and AGS cells. In in vitro genotoxicity assays, the HETa induced different responses, according to concentration and experimental condition. In the MN test the HETa presented genotoxic potential by increasing the number of MNs, NBUDs and NPBs. In the comet assay, HETa was genotoxic by itself and in the co-treatment protocol with H2O2. In pre-treatment or post-treatment protocols with H2O2, HETa presented an antigenotoxic effect by preventing or repairing, respectively, the genotoxicity induced by H2O2. In the in vivo models, HETa was shown to be relatively safe after acute administration in mice [no-observed-adverse effect level (NOAEL) of 2000 mg/kg] and sub-chronic in rats (NOAEL of 800 mg/kg), confirming the riverine information that it is non-toxic in the dosage used. Phytochemical analysis of HETa revealed the presence of phenolic compounds, flavonoids, saponins, tannins and phytosterols. Among the flavonoids and tannins, we highlight gallic acid, rutin, ellagic acid, quercetin, caffeic acid, quinic acid, corilagin, punicalin and punicalagin. Thus, it can be stated that HETa has a good safety margin for therapeutic use.