Optimization of Ultrasound-assisted Extraction of Phenolic Compounds from Myrcia amazonica DC. (Myrtaceae) Leaves

INTRODUCTION

The Myrtaceae is a family constituted by 130 genera and 5700 species, it is very common in Brazil, and several of its species can be found in the Amazon region. The Myrcia genus, with 374 species occurs mainly in the Central America to Northern Argentina.[1] The species of the Myrtaceae family are popularly known for having woody aspect, however, the Brazilian generally do not produce valuable timber, restricting themselves to providing firewood for use in small parts or objects and other forms of local use.[2] In Brazil, the largest use of plants in this family is related to their fruits, they are usually marketed and known to possess very characteristic aroma and flavor (e.g., guava, Psidium guajava L., Jabuticabeira, Myrciaria cauliflora (Mart.) O. Berg, Surinam cherry, Eugenia uniflora L.).[34] This family has a diversity of species with medicinal properties already confirmed such as the works diuretic action, anti-hemorrhagic, astringent, hypoglycemic, and antihypertensive among others.[5]

Secondary metabolites are directly related to the development, protection, and dissemination of the plant, as a result, these compounds play a series of therapeutic activities to the human body. Among the secondary metabolites, phenolic compounds are substances that present an important role in protection against pathogens and predators, these compounds are very diverse and can be divided into several classes (including the flavonoids [Fv], and condensed and hydrolysable tannins).[6] Several Fv has antioxidant activity and has a high redox potential, thus they can reduce the free radicals that can be harmful and also has a great potential for chelating metals.[7] The main class of tannins are hydrolysable and condensed being that the former are derived from a gallic acid,[8] they precipitate with proteins, anti-oxidants are also its antimicrobial power has been much studies in recent times and agents.[9] The Myrtaceae family is characterized by the presence of secondary metabolites such as phenolic compounds (mainly Fv and tannins) and essential oils.[410]

The evaluation of the influence of several factors simultaneously must be necessary to find optimum conditions in several processes. Thus, currently several statistical models have been employed to solve this problem, among them the Box–Behnken design and response surface methodology (RSM), allow for the studies on factors and its multiple interactions.[1112]

Myrcia amazonica DC. is a species predominantly found in northern Brazil, and there are few studies in the literature. Thus, this study aimed to perform an optimization of extraction of phenolic compounds, Fv, and tannins in leaves of Myrcia amazoncia DC. employing ultrasound-assisted extraction.

MATERIALS AND METHODS

Experimental design

Box–Behnken experimental design (3³) with three factors and three levels was used to optimize and evaluate main effects, interaction effects, and quadratic effects of the process variables.[17] The factors studied were: (i) Extraction times (ETs) of 5, 10, and 15 min, (ii) solid: Liquid ratio of 100, 150, and 200 mg/mL, and (iii) ethanolic strength (ES) 65, 80, and 95% (v/v), show in Table 1. Experiments were performed using a 40 kHz ultrasonic UNIQUE^® (USC 4800). The experimental runs were randomized to satisfy the statistical requirement of independence of observations, show in Table 2. A second-order polynomial regression model was used to express the yield as a function of the independent variables as follows:

Table 1

Code and level of factors chosen for the experiments

Table 2

Box-Behnken design and observed responses of ultrasound-assisted extraction of Myrcia amazonica leaves

Where y represents the response variables, β₀ is a constant, β_i, β_ii, and β_ij are the linear, quadratic and interactive coefficients, respectively. x_i and x_j are the levels of the independent variables. All statistical analysis was conducted using the software Design Expert^® version 7.0.[18] The factors with a significance higher than 5% (P < 0.05) was considered.

RESULTS AND DISCUSSION

Table 2 shows the results with the corresponding responses data for the M. amazonica leaves, regarding TPc, TTc, and TFc. The highest value obtained of TPc under the experimental design was 13.66% (Run 2). Regarding total tannins (TT) content, experiment 3 showed the highest value, which was 39.12%, and in this extraction was employed the same alcohol degree that the experiment described above.

The highest content of TF found was 2.60% (Run 4), and this value was also obtained with the 80% alcohol degree, the same of the runs early. The similarity of these experiments regarding the highest values of the responses suggests the influence of alcohol in the extraction process, and this fact will be discussed later.

The second-order polynomial model was fitted to each of the response variables (Equation 1). The ANOVA provides the summary of main effects and their significance in Table 3, which the level of significance are represented as percentages. This analysis showed that the ES had a significant effect on the total phenols, TT, and TF content at the 5% level.

Table 3

Summary of factor effects on Myrcia amazonica powder properties

The ANOVA revealed that only ET² exerted an influence on the TPc at a significance level of 5%. None of the others effects and the interactive terms were significant. Thus, in the levels investigated, this squared term had a negative influence over the total phenols content. The coefficient of this independent variable is shown in the following fitted equation:

Figure 1a and b show the surfaces plots of TTc as a function of ET, solid to liquid ratio (SLR), and ES. These graphs showed that ET and SLR exerted a nonlinear effect on TTc. In general, higher and lower levels of SLR provided better yields of the tannins by ultrasound-assisted extraction, thus its possible make the extraction in these two regions of the surface. The fact of the obtainment of great amounts of tannins with lower SLR could present an economy of this raw material. Sousa et al. found similar results with the stem barks of the species Stryphnodendron adstringens (Mar.) Coville, which lower levels of SLR results in great yields of tannins.[19] Other studies investigated the influence of various parameters on the tannins obtained by ultrasound-assisted extraction, such the article of Martins et al., where was studied the parameters ET, ES, and particle size of raw material on the tannins amount of Dipteryx alata Vogel fruits.[20]

Figure 1

(a) Surface plot of total tannin content as a function of solid to liquid ratio and extraction time (b) surface plot of total tannin content as a function of ethanolic strength and extraction time

The application of experimental design to increase the yield of extraction markers has been very employed in early years. The ultrasound-assisted extraction of the phenolic compound 4-nerolidylcatechol was investigated in roots of species Pothomorphe umbellata, and the main factors that exerted influence in this extraction process were ethanol to water ratio and ET.[21]

According to Figure 1b, it can observed that lower amounts of ethanol in extraction solvent had a positive influence in yields of TT, which could be explained by the higher affinity to these polar compounds with polar solvents (with more water).[1922]

The fitted equation of TT is given by:

Figure 2 presents the surface responses of TFc as a function of the terms ET, ES, and SLR. The graphs show that lowest amounts of SLR result in better yields of Fv extraction. Moreover, intermediate values of ES (80%, v/v) associated superior ET (45 min) provided improve yields of Fv extraction. Figure 2 shows the interaction of ES and SLR, and from this response surface, it can be observed that the better condition to obtainment higher amounts of Fv. The second-order behavior in the optimization of Fv extraction by ultrasound was also found in a study conducted by Zhang et al., which SLR and ES were the quadratic terms.[23] The coefficients of independent variables to this model are shown in the following equation:

Figure 2

Surface plot of total flavonoid content as a function of ethanolic strength and extraction time

The dried extract obtained presented a yield of 68.40% (w/w). The levels of total polyphenols, TT, and total FV were, respectively 18.875%, 18.33%, and 7.95% (w/w). These contents of phenolic compounds found were relevant because the values were greater than other species of the same genus.[24]

In this work, the ultrasound-assisted extraction of total phenols, TT, and TF from M. amazonica leaves was investigated with a three-variable experiment Box–Behnken design based on an RSM in enhancing the yields of these marker classes. The best extraction parameters in the levels studied were obtained. The knowledge gained in this study has a great importance for further investigations and application of this vegetal species.

ABOUT AUTHOR

Mariana Cristina de Morais Rodrigues

Mariana Cristina de Morais Rodrigues, is pharmacist graduated in the State University of Goiás, Brazil, and currently is a Master degree student in the Federal University of Goiás, under guidance of Prof. Dr. Edemilson Cardoso da Conceição. Her research it's concentrated mainly with pharmaceutical technology applied to natural products and development of intermediate and final products from medicinal plants.