Microwave-assisted extraction in dry fruit of andean species Vaccinium meridionale : Experimental conditions on the recovery of total polyphenols

Due to their nutraceutical properties, the fruits of the species Vaccinium meridionale, which inhabit the Andean region, are of scientific interest. Microwave-assisted extraction has been applied to different vegetal matrices for to extract efficiently polyphenolic compounds. In this work we study in microwave assisted extraction processes, the effect on the extraction of total polyphenols in dried fruits of Vaccinium meridionale, using response surface methodology. The main objective of this research was to analyze the relationships between power, temperature, solid-liquid ratio, time and ethanol concentration in aqueous media on the effectiveness of total polyphenols recovery in dry fruits of Vaccinium meridionale. A central composite design face-centered with three levels for each variable was used: P = 300 – 900 W; T = 70 –110 °C; L:Srat. = 30:1 – 70:1 w/w; t = 5 – 15 min; [EtOH] = 0 – 80%. In this sense, the effect of the input variables on the recovery of total polyphenols, the optimization process for maximum extraction, and the comparison to other solid-liquid extractions in terms of quantity of total polyphenols, are reported. It was found that the significant variables in the extraction process were ethanol concentration, temperature and time; the best yields were obtained in the range [EtOH] = 44 – 49%, T = 110 °C, and t = 10 – 15 min. In terms of quantity, time and consumption of energy, microwave-assisted extraction technique is more efficient than other solid-liquid extraction processes for the extraction of total polyphenols.


INTRODUCTION
Over the last decades, a greater interest has been placed on the importance of consuming fruits of Vaccinium genus due to their therapeutic effects.Compounds from the family of flavonoids and benzoic acids contained in those foods have shown antioxidant activity (Shahidi;Ambigaipalan 2015;Zielinska;Michalska 2016;Wang et al., 2017).These molecules are associated with oxidative stress phenomena in the organism, due to the delay and inhibition of the action of free radicals (FR), which, in turn, has been associated with the development of multiple diseases (Sen et al., 2010;Nardi et al., 2016;Wu et al., 2016).On the other hand, some flavonoid contained in dry fruits of this genera have both antimutagenic and antitumoral activity.Preservation and normalization effects have been observed in the cellular cycle (Vattem;Ghaedian;Shetty 2005;Gambini et al., 2015); DNA molecule repair in cellular lines of colon cancer, induction of apoptosis, inhibition of tumor invasion and angiogenesis, as well as decrease in swelling processes (Kumar;Dhatwalia;Dhawan, 2016;Kazan et al., 2016;Kazan et al., 2017).Inhibition of atherogenesis, quercetin-mediated plaque aggregation, increase in neuronal signalization and increase in insulin secretion, have also been found (Martineau et al., 2006;Perez;Duarte 2010;Krikorian et al., 2010).Some anthocyanins present in fruits of Vaccinium genus, aid in the treatment of microcirculation diseases in diabetic patients and provide stabilization effects of collagen in varicose veins and hemorrhoids.Furthermore, they prevent gastrointestinal disorders and presence of proanthocyanidins microbial activity (Seeram, 2006;Kaufman et al., 2006;Khoo;Falk, 2014).
The fruit of Vaccinium meridionale, a native species of the Andes, has caught attention due to its high content of anthocyanins, and in general, total polyphenols (TP).These fresh fruits are consumed on a regular basis in preparations, such as juice and jelly (Garzón et al., 2010).The fruit of Vaccinium meridionale has shown activity in the inhibition of lipid peroxidation of corn oil (Gaviria et al., 2009a), growth inhibition in colon cancer cell lines (Maldonado;Arango;Rojano, 2014), protective effects in HT1080 cells lines against reduction of viability induced by the rotenone (Sequeda et al., 2016), a significant reduction in cell viability from SW480 cancer cell line (Zapata et al., 2016;Agudelo et al., 2017); and cytotoxic activity of extracts in transformed leukemic cell lines (González et al., 2017).
Regarding the process of TP obtainment in plants, microwave-assisted extraction (MAE) is highlighted because it possesses several advantages as compared to other conventional techniques.For instance, MAE requires short time periods, low solvent and energy consumption.Irradiation with microwaves generates matter and thermal energy transport from the inside of the cell towards the outside, extracting the metabolites of interest in a homogeneous and efficient manner (Wang;Weller, 2006;Mandal;Mohan;Hemalatha, 2007;Dean, 2012;Leonelli;Veronesi;Cravotto, 2013).
In this paper the effect of power (P), temperature (T), the liquid-solid ratio (L:S rat.), time (t), and ethanol concentration ([EtOH]) in the solvent system on the amount of TP extracted in dry fruits of Vaccinium meridionale with MAE technique, are discussed.Response surface methodology (RSM) was employed in order to analyze, with a relatively small number of experimental points (Song et al., 2011), the relationships between the independent variables or factors and the response variable.In this work, a central composite design face-centered with three levels for each variable was used: P = 300-900 W; T = 70 -110 °C; L:S rat.= 30:1-70:1 w/w; t = 5-15 min; [EtOH] = 0-80%.Quantification of TP was performed by applying the colorimetric system of Folin-Ciocalteu (F-C) reagent, and gallic acid (GA) as a reference molecule.The analytical method was previously validated by Espinosa, Garzón and Medina (2016), in terms of selectivity, linearity, repeatability and accuracy.Experimental design, treatment of experimental data and optimization process were carried out using Design-Expert ® software, trial version 9. The recovery efficiency is compared with other solid-liquid extraction processes.

Collection of Vaccinium meridionale fresh fruits
Fresh fruits of Vaccinium meridionale were collected from a native habitat located in the municipality of Chiquinquirá (Boyacá, Colombia), latitude 5600, longitude 73800, and altitude between 2500 and 2600 masl; average precipitation 990 mm a year, average temperature 12.9 °C, maximum 19.3 °C and minimum 7.4 °C.Once collected, they were placed in polypropylene plastic bags and immediately carried out to the laboratory.Leaves and stems were removed manually and consequently washing with type I ultrapure degasified water.

Vaccinium meridionale dry fruit processing
The fresh fruits of Vaccinium meridionale were dried in a Memmert UM400 oven at 40 °C, until the mass of the plant material remained constant for 12 hours.Consequently, the dried fruit was milled in an IKAA 11 analytical mill, and the sample was sieved through a sieve of 0.5 mm pore size.Particles of less than or equal to 0.5 mm were stored in a desiccator and frozen at -4 °C.

Microwave-assisted Vaccinium meridionale fruit extracts
Dry fruit (Dw) were weighed in a range of 281.7-645.2± 0.1 mg and mixed in 19354.8-19718.3± 0.1 mg of solvent system (0% EtOH, 40% EtOH and 80% EtOH).All weights were measured in an Ohaus explorer EX 224 analytical balance, with a capacity of 220 g, and uncertainty ± 10 -4 g.The mixture obtained was homogenized at room temperature (12-20 °C) in an incubator shaker IKA KS 4000 ic control for 5 minutes.Subsequently, it was led to the Sineo MDS-8G extraction/digestion microwave system, according to the required extraction conditions.Finally, the mixture was centrifuged at 4000 rpm for 20 minutes in a Hettich rotina 46 S centrifuge.The obtained supernatant was tagged as MAE fruit extract.If it was necessary, aqueous extracts were stored in a refrigerator at 4 °C and ethanolic extracts in a freezer at -4 °C.In any case, the time passed between obtaining and quantifying the extract did not exceed 24 hours.Figure 1 shows the general scheme.

Experimental design
Central composite design face-centered was performed in order to assess the effect of five independent variables: power (P), temperature (T), liquid-solid ratio (L:S rat.), time (t) and ethanol concentration ([EtOH]) in the ethanol-water solvent system.Response or dependent variable was TP recovered in mg of gallic acid per g of dry fruit (mg GA/g Dw).In total, 50 experimental points were carried out (N = 2 k + 2k + n 0 = 2 5 + 2•5 + 8 = 50).Levels selected in each of the input factors were supported on the analysis of previous studies of TP extraction with MAE (Song et al., 2011;Ballard et al., 2010;Nayak et al., 2015;Gallo et al., 2010): P = (300, 600, 900) W; T = (70, 90, 110) °C; L:S rat.= (30:1, 50:1; 70:1) w/w; t = (5, 10, 15) min; [EtOH] = (0, 40, 80) %.This study was limited to a maximum temperature of 110 °C, because the fruit of Vaccinium meridionale contains thermolabile compounds of a great therapeutic importance, such as resveratrol, catechin, and epicatechin (Garzón et al., 2010;Liazid et al., 2007).Ethanol was the solvent selected for its low toxicity, which makes it an ideal solvent to be used in pharmaceutical, cosmetic and food products (Mandal;Mohan;Hemalatha, 2007).The order of the experimental points, the modeling of experimental data and the optimization of process variables were performed using the Design-Expert ® software, trial version 9. Statistical significance of the model and model parameters was determined at the 5% probability level.The adequacy of the model was determined by calculating the lack of fit, coefficient of determination (R 2 ) and the Fisher test value (F-value) obtained from the analysis of variance (ANOVA) generated by the software.Table 1 shows the experimental design used in this study.Regarding the model generated by the software, the degree of agreement between the experimental results and the predicted quantities was evaluated by calculating the relative error.Six new extractions were also performed in the maximum recovery region predicted by the software and their differences were evaluated.

Quantification of total polyphenols in extracts
TP quantification was carried out through the colorimetric F-C method, according to analytical method validation proposed by Espinosa, Garzón and Medina (2016).Selectivity, linearity, repeatability and accuracy were the parameters validated.The major improvement of this quantification process was the employ of a gravimetric aliquot of the dry fruit extract instead of volumetric aliquot, which is commonly described in the literature (Turkmen;Sari;Velioglu, 2006;Rafiee et al., 2011;Blainski et al., 2013;Li et al., 2013;Hatami et al., 2014;Sim et al., 2016).Quantification process was performed as follow: fruit extracts were weighed directly into a 5-mL volumetric flask (25 -140 ± 0.1 mg of extract).125 μL of de F-C 2N reagent were added to the flask and homogenized for 1 minute.Afterward, 400 μL of sodium carbonate aqueous solution at 10% were added and diluted to the mark.It was homogenized for 30 seconds and left in the dark for 1 hour.Absorbance of the solution was measured at 760 nm wavelength in a Hach DR 5000 spectrophotometer.The quantities of TP contained in the extract, with gallic acid as a reference molecule, was calculated as shown in Equation 1: a great influence of the experimental conditions over TP quantity recovered.By applying multiple regression analysis on the experimental data with Design-Expert ® software trial version 9, the response variable and the independent factors were related by a second-order polynomial equation.Table 3 shows the analysis of variance (ANOVA) for response surface quadratic model.The model F-value of 70.18 implies the model is significant.There is only a 0.01% chance that a "model F-value" this large could occur due to noise.The "lack of fit F-value" of 1.33 implies the lack of fit is not significant relative to the pure error.There is a 36.94%chance that a "lack of fit F-value" this large could occur due to noise.The significance of each coefficient was evaluated using the F-test and P-values.It can be seen that all the linear terms, three quadratic and four interaction terms are significant (P-values greater than 0.1000 indicate the model terms are not significant).
The reduced regression equation in terms of actual factors is presented as follows (R 2 = 0.9746) (Equation 2): Where mg GA/g Dw is the TP quantity recovered in mg of gallic acid per g of dry fruit, mg solv is the mass of the solvent used to produce the extract (mg), A is absorbance of the diluted solution, b the intercept of the calibration curve (b = 0.241 ± 0.007), m the slope of the calibration curve (m = 0.869 ± 0.001), mg ext the extract aliquot (mg), and g Dw corresponds to the mass of the dry fruit used to produce the extract (g).

Microwave-assisted extraction in dry fruits of Vaccinium meridionale
Table 2 shows the TP obtained in each of the 50 experiments realized.The response variable obtained is found in the range of (18.8-40.2) mg GA/g Dw, showing Power is expressed in W, temperature in K, L:S rat. in g/g, time in minutes, and [EtOH] in %.In agreement with experimental results, a relation directly proportional to power and TP quantity along the interval studied, was found.Using quadratic polynomial by Equation 2, the pass from 300 to 900 W increases on average, extraction capacity in 1.4 ± 0.7 mg GA/g Dw.Temperature also presents a directly proportional relation.According to Equation 2, the pass from 70 °C to 110 °C generates, on average, an increase of 5.2 ± 0.9 mg GA/g Dw in extraction capacity.Conversely, there were no meaningful effects in the L:S rat. .The ratio from 1:30 to 1:70 generates, on average, an increase in the extraction capacity of 0.8 ± 1.4 mg GA/g Dw.Extractive capacity enhancements by the increase in the L:S rat. in systems with the power of 300 W; if power is 600 W there is an inverse relation if [EtOH] is 80%; and if power is 900 W, the relationship is inverse when [EtOH] is 40 and 80%.Time factor displays a direct relation up to 11.8 min, after which it will be in the inverse way until the end of the interval studied.The span from 5 to 11.8 minutes generates, on average, an increase in the extraction capacity of 2.3 ± 0.7 mg GA/g Dw.Finally, [EtOH]  a u(TP) / TP = ± 1.1%; u(P) = ± 10 W; u(T) = ± 1 °C; u(L:S rat.): = ± 2•10 -4 g/g;  on the efficiency of the process.Several authors have previously reported the importance of [EtOH] on the recovery of TP with MAE.For example, the extraction of TP from leaves of Ipomoea batatas (Song et al., 2011), in green tea leaves (Camellia sinensis) (Quan et al., 2006), and blackberry fruits Morus alba L. (Teng;Lee, 2013).Also, Xiao et al. (2008), highlighted the effect of [EtOH] on the extraction of flavonoids in Radix astragali.
Figure 3 shows an example of the behavior of temperature relative to TP extraction.Variables P and [EtOH] are included; variables t and L:S rat.remain constant at their central points (t = 10 min; L:S rat.= 50:1).The first figure shows the response surface obtained when the temperature is at the lowest limit of the experimental design (T = 70 °C), and the last figure represents the surface obtained at the highest limit (T = 110 °C).In this example, we note the increase in TP obtainment by temperature increase under all experimental conditions, in which P and [EtOH] are considered.Recovery is minimum in the lowest limit and maximum in the highest limit.Destruction of vegetable cells exposed to microwaves is promoted when the temperature of the system increases, due to an enhancement in solute mass transfer rate into the matrix, increasing its solubility and decreasing solvent viscosity and superficial tension (Ballard et al., 2010;Leonelli;Veronesi;Cravotto, 2013;Dahmoune et al., 2015).It is possible to expect in dry fruits of Vaccinium meridionale, an increase in the TP quantity extracted when the temperature is higher than 110 °C such as has been reported in previous work (Liazid et al., 2007;Inglet et al., 2010).However, an increase in system temperature would affect the chemical stability of bioactive compounds such as resveratrol, catechin and epicatechin (Liazid et al., 2007;Garzón et al., 2010).
With respect to the time factor, the maximum recoveries have been obtained in conditions nearing at 11.8 minutes.Time of exposition is a key factor because, in an inefficient extraction process, overheating could extend thermal degradation of phenolic compounds (Liazid et al. 2007).

Optimization of MAE conditions and verification of the model
In order to establish the optimal conditions of TP microwave-assisted extraction in dry fruits of Vaccinium meridionale, an optimal range was determined using Design-Expert ® software, trial version 9. Efficient extraction of TP requires a temperature of 110 °C (the highest one used in this research), [EtOH] ranging between 44 -49 %, and time between 10 -15 minutes.Both remaining factors power and liquid:solid ratio are not determinant variables, since high recoveries of TP are possible along all the intervals considered.In order to evaluate the suitability of the quadratic equation for described and optimized the amount of TP extracted, six different extractions were carried out in the range of maximum recovery.Table 4 shows the experimental quantity of TP extracted at each determination, the quantity of TP predicted by Equation 2, and the percentage difference between them.As it can be seen, the model predicts successfully the behavior obtained.
On average, the differences between the experimental results and the values predicted by Equation 2 are similar to those found in the experimental design.We also found in three of these six assays a major recovery of TP compared to the quantities obtained in the experimental design (the maximum quantity of TP recovered was 40.2 mgGA/gDw at 600 W, 110 °C, 50:1 w/w liquid:solid ratio, 10 minutes, and 40 % EtOH (run 2)).In this sense, response surface methodology was successfully applied for modelling and optimization of TP extraction in dry fruits of Vaccinium meridionale with the MAE technique.
Table 5 shows a comparison between the efficiency of extraction with MAE and solid-liquid extraction respect to TP quantity recovered in fruits of Vaccinium meridionale.Maximum recovery in each case was reported.For MAE results, assay 3 of Table 4 was considered as the maximum.In terms of efficiency, MAE is found to be between 8 to 35% better, as compared to solid-liquid extraction processes.It is possible that efficiency in the extraction of TP with MAE improves if EtOH is replaced by MeOH in the solvent system, as in the studies reported in literature.As previously indicated, EtOH was selected due to its low toxicity.It is also important to highlight the difference of time in the extraction processes.TP obtainment in dry fruits of Vaccinium meridionale with MAE is significantly higher than solid-liquid extraction, as proven in previous studies (Pan;Niu;Liu;2002;Afoakwah et al., 2012;Chan et al., 2011;Li et al., 2011;Routray;Orsat, 2012).In energetic terms, MAE is also more efficient, since it requires lower energy consumption.a Average value from experimental results for recovery of TP in mg GA per 100 g of fresh fruits, with an average humidity content of 80% (Gaviria et al., 2009b).

CONCLUSIONS
In this study, we found that [EtOH] is the most important variable for TP extraction in dry fruit of Vaccinium meridionale with MAE, followed by temperature, time, power and L:S ratio.Experimental results were described by a quadratic polynomial equation.A range of optimal conditions for TP obtainment was found and experimentally validated: 110 °C of temperature, ethanol concentration between 44-49% and time between 10-15 minutes.MAE proved to be an effective technique for TP extraction in dry fruits of Vaccinium meridionale.TP extraction is more efficient between 8.4% and 34.8% than solid-liquid extraction processes wherever studied.Additionally, it is important to note the short periods for extraction and subsequent decrease in energy consumptions.

Figure 1 :
Figure 1: Scheme of the microwave-assisted extraction of total polyphenols in Vaccinium meridionale dry fruit.

Figure 2
Figure 2 shows, an example of the behavior of [EtOH] on recovery capacity.The variables considered are T and t, whereas P and L:S rat.remain constant at their central points (P = 600 W; L:S rat.= 50:1).The upper figure shows the response surface obtained when [EtOH] = 0%; the figure placed on the middle shows maximum recovery surface when [EtOH] = 44%; and the lower figure shows response surface when [EtOH] = 80%.It is to focal point the great influence of [EtOH]on the efficiency of the process.Several authors have previously reported the importance of[EtOH]  on the recovery of TP with MAE.For example, the extraction of TP from leaves of Ipomoea batatas(Song et al., 2011), in green tea leaves (Camellia sinensis)(Quan et al., 2006), and blackberry fruits Morus alba L.(Teng;Lee, 2013).Also,Xiao et al. (2008), highlighted the effect of[EtOH]  on the extraction of flavonoids in Radix astragali.Figure3shows an example of the behavior of temperature relative to TP extraction.Variables P and [EtOH] are included; variables t and L:S rat.remain constant at their central points (t = 10 min; L:S rat.= 50:1).The first figure shows the response surface obtained when the temperature is at the lowest limit of the experimental design (T = 70 °C), and the last figure represents the surface obtained at the highest limit (T = 110 °C).In this example, we note the increase in TP obtainment by temperature increase under all experimental conditions, in which P and [EtOH] are considered.Recovery is minimum in the lowest limit and maximum in the highest limit.Destruction of

Table 1 :
Experimental design used in the extraction of TP with MAE.

Table 2 :
is directly proportional to the of extraction, until it reaches a concentration of 45%; afterward, the relation is inverted until the end of the interval considered.The change in concentration from 0% to 45% gives, on average, an increase in the extraction capacity of 12.4 ± 0.7 mg GA/g Dw.According to results described previously, we found that [EtOH] is the most important variable in the extraction process, followed by T, t, P, and L:S rat..TP (mg GA/g Dw) obtained in extraction processes with MAE, according to experimental design described in Table1. capacity

Table 3 :
ANOVA for response surface quadratic model for the experimental results of TP contents from dry fruits Vaccinium meridionale extracts with MAE. Continue...

Table 4 :
TP (mg GA/g Dw) obtained in the range of experimental conditions with maximum recovery described by Equation2.

Table 5 :
Comparison of TP quantity recovered from dry fruits of Vaccinium meridionale with MAE and solid-liquid extraction processes.