Obtenção, caracterização química e potencial antioxidante de extrato hidroalcoólico de folha e inflorescência de Amaranthus deflexus

Abstract

There are many species of weeds present throughout the national territory from crops, vacant lots and sidewalks in urban areas. The genus Amaranto has species cultivated across the globe, however, on a small scale of production. The species known as Amaranthus deflexus commonly develops in the South and Southeast regions of the country and is mainly considered as a harmful plant. Studies indicate potential for food and antioxidant for species of this plant. The project, divided into 3 parts, proposed to analyze the extract of the leaves and inflorescence of this species, collected in the city of Francisco Beltrão-PR. In part 1, the proximate composition of the leaves and inflorescence was determined in terms of the presence and amount of moisture, proteins, lipids, ash and carbohydrates. In addition to pH and titratable acidity analyses. In part 2 we sought to apply a Fractional Factorial Design (FFD) to optimize the proposed extraction conditions for quantification of phenolic compounds (CFT) and antioxidant activity (AA). Finally, part 3 sought to generate predictive models using images of extracts and leaf and inflorescence tests together with a chemometric approach applying the Partial Least Squares (PLS) method. The results showed an ash content above that of some vegetables consumed in the diet. Furthermore, through the optimization it was possible to affirm that concentrations of 50% for the leaf and between 0 to 50% for the inflorescence are better in the extraction of bioactive compounds from Amaranthus deflexus. In the last step, the predictive models were considered excellent by analyzing some factors, indicating a good prediction of phenolic compounds content and antioxidant activity for both leaf and inflorescence images. Furthermore, they also indicated that by using only the extract images, without the need to use the test images, it is already possible to predict the CFT and AA content due to the excellent prediction obtained in this method.