PROFILE OF PHYTOCHEMISTRY AND ANTIOXIDANT ACTIVITY OF SORREL IN FUNCTION OF ORGANIC FERTILIZATION AND DENSITY OF PLANTIO PERFIL DE FITOQUÍMICA E ATIVIDADE ANTIOXIDANTE DE AZEDINHA EM FUNÇÃO DA ADUBAÇÃO ORGÂNICA E DENSIDADE DE PLANTIO

The revival of non-conventional edible plants enables important gains in nutritional, social, cultural, economic, and environmental points of view. Among them, sorrel — Rumex acetosa L. — has been deemed pleasant by consumers due to its acid taste. As we do not have a lot of phytotechnical and nutritional information about this plant, the objective was to assess the influence of different spacing sets and cattle manure dosages on their physical-chemical characteristics and bioactive compounds content with antioxidant action. Randomized blocks with four repetitions was chosen as experimental design, using a scheme of subdivided plots, with five dosages of weather-beaten cattle manure in the plots and two sets of spacing in the subplots. Data was submitted to a variance analysis with two factors (two-way ANOVA). It was not verified any significant effect (P>0.05) in the interaction Dosage x Spacing. Fertilization dosages and spacing sets did not have any effect on color and total soluble solids average contents, humidity, titratable acidity, and phenolic compounds. Increasing dosages of organic fertilizer, up to 75 kg ha, reduced the plant’s pH; above that amount, pH rose. Presence of anthocyanins and vitamin C was not detected. The total carotenoids average content was not influenced by the manure factor, and the plant presented greater total carotenoids content in the smallest spacing. Antioxidant activity [1.0 mg mL], although inferior to the standards that were used (BHT, vitamin C and quercetin), was quite representative. A conclusion was reached that fertilizing with weather-beaten cattle manure had an effect only on pH, and cropping spacing had an influence only on carotenoids content. Other phytochemicals and physical-chemical characteristics were not influenced in sorrel plants with different levels of fertilization and sets of spacing.


INTRODUCTION
Preservation of local species and varieties, as well as knowledge about neglected edible plants, in addition to contributing to maintain resilience of small-scale agricultural systems, they also help to stabilize production, minimize risks, as well as profit from available micro-environments with varieties of more suitable crops (PILLA; AMOROSO, 2009). There are several vegetable species that are not used or are underused with high nutritional qualities that often are greater than those of plants usually cropped (LADIO, 2005). Non-conventional vegetables such as purslane (Portulaca oleracea), Malabar speech (Basella rubra), green amaranth (Amaranthus viridis), lamb's ear (Stachis lanata), and sorrel (Rumex acetosa) present considerable contents of minerals and natural antioxidants such as carotenoids and phenolic compounds in their constitution, and its regular consumption is associated with benefits for human health (VIANA et. al, 2015).
Sorrel (Rumex acetosa L.), having characteristic rusticity and flavor, has been found pleasant by consumers, presents good marketing potential and can contribute to enrich human diet, 776 Profile of phytochemistry TORRES, M. P. et al. because it presents relevant amounts of macro and micronutrients in its constitution , besides natural antioxidant compounds such as phenolic and carotenoid compounds (VIANA et al., 2015). Sorrel belongs to the Poligonaceae family and is herbaceous, perennial, about 20 cm (8 in) high, forming clumps with several propagules. It is usually found in the wilds in regions with mild climate in Europe and Asia; in Brazil, it is cropped from the State of Rio Grande do Sul to the State of Minas Gerais. However, it rarely blossoms in Brazilian climate conditions (PEDROSA et al. 2012;SILVEIRA et al., 2013;MADEIRA et al., 2013).
In recent years, a greater conscience about the influence of food in health and its direct reflexes in quality of life has been generating an increase on the demand for healthier food. It is also possible to see that farmers are adopting a more conscious attitude about their products' nutrient content (RODRIGUES-AMAYA et al., 2008). However, according to Miliauskas et al. (2004), there is few scientific information about the nutritional composition and the bioactive substances with health-promoting proprieties content, especially regarding plants less used in medicine and culinary.
Food's healing and preventive power have been emphatically associated with the presence of antioxidants, and sometimes correlated to bioactive substances concentrations, even if the antioxidant effect in a piece of food is clearly differentiated from its effect in health (RODRIGUEZ-AMAYA et al., 2008). According to Bernardes et al. (2011), the ingestion of food containing bioactive compounds such as carotenoids, phenolic compounds and vitamins constitutes the main form for the organism to obtain antioxidants. Therefore, the determination of bioactive compounds in vegetables is essential to assess the source food of those compounds, add scientific knowledge about food nutritional composition and the benefits in preventing diseases and reinforcing the importance of daily consuming these products (FALLER; FIALHO, 2009).
However, it is worth highlighting that environmental conditions, fertilization, genetic factors, plant variety, among others, widely influence phytochemical content in vegetables (RAMOS et al. 2011). In addition, studies that correlate cropping practices with phytochemicals production are rare (ARBOS et al., 2010).
The development of non-conventional vegetables organic cropping technologies can increase the diversity of quality food products, contribute to population's food and nutritional safety, provide technical assistance to the producer, favor environmental conservation, encourage production and consumption of those plants.
Thus, this present study aimed to assess physical-chemical characteristics and bioactive compounds content in sorrel cropped in different sets of spacing and with different dosages of fertilization with weather-beaten cattle manure.
This study assessed sorrel (Rumex acetosa L.) type I, according to Silva et al. (2013), whose exsiccated sample is deposited in EPAMIG's herbarium with the following registration: PAMG 57029. Matrixes that were cropped in the nonconventional vegetables center in EPAMIG's Fazenda Experimental Santa Rita in the City of Prudente de Morais were used to produce seedlings. Plastic bags measuring approximately 20 cm x 10 cm (8 in x 4 in) were used, with a substrate composed by the mixture of ravine soil and weatherbeaten cattle manure in a 1:1 proportion and placed in a vegetation housing.
Randomized blocks with four repetitions was chosen as experimental design, with a scheme of subdivided plots. The plots consisted of five levels of weather-beaten cattle manure fertilization factor translated into dosages of weather-beaten cattle manure (0, 25, 50, 75, and 100 t ha -1 ), and subplots were two levels of spacing factor (2 cm x 25 cm and 30 cm x 30 cm/0.8 in x 9.4 in and 11.8 in x 11.8 in).
In the plots, which consisted of seedbeds of 1.20 m x 4.00 m (3.9 ft x 13.1 ft) and 0.20 m (0.6 ft) high, weather-beaten cattle manure dosages were distributed, being 2/3 of the dosage applied five days before transplantation and 1/3 of it applied in coverage, 20 days after planting the seedlings. In the subplots, with an area of 2.4 m 2 (25.8 ft²), 24 seedlings were planted, distributed in four lines of six plants, in the respective sets of spacing. The eight central plants were considered for the composition of the useful plot.
Periodical hoeing was used as a farming practice, and it was not necessary to perform plague and disease control. Sprinkling irrigation was performed using a SANTENO I -09 / 1 Kgf cm -2 hose. Plants were cut at 3 cm (1.1 in) from soil level, on the 57th day after transplanting, when they reached a point of harvesting.
Leaves were harvested by chance and transported under refrigeration to Federal University of São João del-Rei's (UFSJ) Sete Lagoas Campus, where analysis was carried out.
After verifying the assumptions of normality and homogeneity through Shapiro-Wilk and Levene tests, dada collected was submitted to a variance analysis with two factors (two-way ANOVA), according to the statistical model for randomized blocks with a scheme of subdivided plots.
Facing the statistical significance of the fertilization factor, a regression study was performed on tendency analysis between dosages of organic fertilizer. For the spacing factor, the amount of statistic F was considered; and to compare treatment averages against control groups, Tukey test was used. In every analysis, a nominal level of 5% of significance was adopted. The statistical analysis was carried out using R Software (R CORE TEAM, 2014).
These were assessed during the experiment: physical-chemical characteristics, bioactive compounds and antioxidant activity, which will be described as it follows.

Physical-chemical assessment
The instrumental color was measured in three different points of full leaves' abaxial surfaces, with the aid of a Konica Minolta colorimeter, model CR 410with D65/10 illumination pattern, to read parameters L*, a*, and b*.
Total soluble solids content was determined through a direct reading from the filtered portion, obtained from grounded leaves, placed on digital Reichert model R2mini refractometer prism, with automatic temperature compensation (AOAC, 2012). Results were expressed in °Brix.
Humidity was obtained from the difference between initial and final mass of the sample dried in a sterilization greenhouse and drying process at 105º C (221º F) until constant weight (AOAC, 2012). pH was determined by potentiometry (AOAC, 2012), with the aid of a digital pH meter, through direct immersion of the electrode in the homogenized in natura samples that were added 5 mL (0.18 oz) of deionized water.
Titratable acidity was determined by titulometry using NaOH 0.01 N as solution and phenolphthalein as indicator, according to the methodology described in AOAC (2012). Results were expressed in percentage of citric acid.

Assessment of bioactive compounds
Total carotenoids were extracted with acetone p.a. and quantified by spectophotometry at 450 nm (RODRIGUEZ-AMAYA, 1999). Results were expressed in µg of carotenoids by 100 grams (0.22 lb) of fresh mass.
Total phenolic compounds content in methanolic extracts of the plants was determined by Folin-Denis method, according to AOAC (2012). Results were expressed in mg of gallic acid equivalents (GAE) by 100 grams (0.22 lb) of dry mass.
Anthocyanins were assessed through differential pH method, as described by Giusti & Wrolstad, (2001). Absorbance was calculated from the equation , where Aλ510 = absorbance in wavelength of 510 nm, Aλ700 = absorbance in wavelength of 700 nm, (Aλ510 -Aλ700) pH 1 = difference of absorbance amounts read in the solution pH 1.0 and (Aλ510 -Aλ700) pH 4.5 = difference of absorbance amounts read in the solution pH 4.5.
Concentration of monomeric pigments in the extract was represented in cyanidin-3-glycoside and calculated using the equation MA= Ascorbic acid content was assessed by highperformance liquid chromatography (HPLC), in a SHIMADZU chromatographer equipped with UV-VIS detector, involving the following conditions: C 18 column (ID 4.6 mm x 150 mm/0.02 ft x 0.49 ft), loop of 20 microliters, wavelength of 254 nm, flow of 1mL min -1 . The mobile phase was prepared from cetyltrimethylammonium bromide (5mmol L -1 ) and monobasic potassium phosphate (50 mmol L -1 ) 1:1, according to the technique described by Benlloch et al. (1993).

Assessment of antioxidant activity
Methanolic extract's antioxidant activity was determined by the DPPH method, using as positive control BHT, vitamin C and quercetin. This technique consists of adding 1 mL (0.04 oz) of the extract in concentrations which vary from 0.1-1,000 µg mL -1 to 1 mL (0.04 oz) of a DPPH methanolic solution (0.1 mM). This reaction was processed within 1 hour at room temperature. DPPH absorption was immediately verifiedat 515 nm in a FEMTO® 700S spectophotometer. All of the experiments were carried out in triplicate. The capacity to kidnap free radicals was expressed as radical oxidation percentage of inhibition and calculated by the equation %Inhibition=100x [(ADPPH -Aextr)/ADPPH], where ADPPH is the absorbance of DPPH solution and AExtr is the absorbance of the sample in solution (ROESLER, et al., 2007). The results were expressed in % and compared to reference patterns.

RESULTS AND DISCUSSION
The variance analysis did not reveal any significant effect (p>0.05) in the interaction Fertilization x Spacing, which indicates that these factors act independently on every assessed characteristic. Thus, the effects of the main (Fertilization) and secondary (Spacing) factors were separately studied.
The dosages of fertilization and sets of spacing that were used did not have any effect on sorrel's coloration, whose averages of the amounts were equivalent to: L* = 48.50, a* = -16.91 and b* = 28.43 (Table 1). It was verified that the amounts obtained were very close to those found by other authors for sorrel type I, such as those observed by Silva et al. (2013): L* = 47.12; a*= -17.67 and b*= 30.95, even though plants were cropped in different time, place and system. Viana et al. (2014) obtained the amounts L* = 43.13, a*= -16.45, b*= 25.95 immediately after harvest in a refrigerate storage study. Table 1. Average content of L* (luminosity), a* and b* (chromaticity), total soluble solids TSS (°Brix), Humidity (H, %); pH; Titratable acidity (TA, g citric acid/100 g fresh mass), Total Carotenoids (TC, µg total carotenoids/100 g fresh mass), Total Phenols (TP, mg total phenols/100 g dry mass), Antioxidant Activity from the extract in a 1.0 mg.mL -1 concentration (AA, % The average content of total soluble solids, humidity and titratable acidity did not present significant differences (p>0.05), according to what was presented in Table 1, which indicates that fertilization and spacing factors, in the assessed conditions, did not interfere on those characteristics of the plant. Sorrel leaves' acidity, which is an important characteristic of this non-conventional vegetable, in this study presented an average of 1.22 g (0.04 ounce) of citric acid/100 g of fresh matter (Table 1), an amount greatly lot superior to that reported by Silva et al. (2013) for sorrel from EPAMIG-Prudente de Morais-MG's nonconventional vegetables center (0.11 g of citric acid/100 g of fresh matter) and by Viana et al. (2014) for sorrel from Sete Lagoas (0.81 g of citric acid/100 g of fresh matter). Acidity in vegetables is attributed especially to organic acids that are found dissolved in cells vacuoles and it is a quality attribute that is related to aroma and taste of food (CHITARRA; CHITARRA, 2005).
It was verified a significant effect (P = 0.05) in the fertilization factor for the variable pH, indicating that this sorrel characteristic responds to the application of different fertilization dosages. An average amount of 3.11 for pH was observed, but a decrease on the pH of the plant was verified with the input of weather-beaten cattle manure up to the The increasing application of weatherbeaten cattle manure probably had an influence on the quantity and quality of the nutrients absorbed by the plant, which interfered in its pH. The tendency of the pH of the plant to increase in dosages above 75 t ha -1 may be related to soil reactions, considering that, in the soil, the increase or reduction of pH, due to application of organic matter, depends on the predominance of the processes that consume or release H + (SILVA; MENDONÇA, 2007). That interferes in the availability of the nutrients to plants, such as those of Cu, Fe, Mn and Zn cations, which are converted into insoluble forms with the elevation of the pH of the soil (RESENDE, 2009). PH amounts observed in this study are in accordance with those detected in sorrel leaves in other studies -2.98 and 3.22 VIANA et al., 2014), respectively.
The factors of fertilization and spacing did not present any significant effect (p>0.05) on phenolic compounds. The average content of total phenols found was equivalent to 840.03 or 870.03 mg. GAE 100 g -1 dry mass (Table 1), an amount inferior to that related by Vieira et al. (2010) to Barbados gooseberry (1,693.33 mg GAE 100g -1 dry mass), a species that also integrates the list of nonconventional vegetables, and to contents quantified by Asolini et al. (2006) in yerba mate (14,500 mg GAE g -1 dry leaf), rosemary (8,000 mg GAE 100 g -1 dry leaf) and broadleaf plantain (6,500 mg GAE 100 g -1 dry leaf), species considered sources of phenolic compounds. It is worth highlighting that those authors used Folin-Ciocalteau spectophotometric method. The average content of total phenols found in sorrel becomes a lot more expressive when compared to the amounts that Arbos et al. (2010), who also used Folin-Ciocalteau reagent, obtained in vegetables, such as organic arugula (126.84 mg GAE.100 g -1 ), organic lettuce (108.72 mg GAE.100 g -1 ) and organic common chicory (92.15 mg GAE.100 g -1 ).
The presence of anthocyanins and vitamin C was not detected in sorrel leaves assessed in this present study. Viana et al. (2015) also reported the absence of vitamin C in sorrel type I cropped in the Non-Conventional Vegetables Center (EPAMIG-Prudente de Morais-MG). However, they reported the presence of the same in leaves of sorrel type II at the concentration of 72.45 mg of vitamin C/g of leaf in dry basis. Ascorbic acid or vitamin C is present in several fruits and vegetables, such as broccoli, cauliflower, spinach, pepper, bell pepper, cabbage, among others; and many pre-and post-harvest factors -cultivar used, climate conditions, planting practices, etc.-have an influence on its concentration (CARVALHO et al., 2006).
A significant difference was revealed about total carotenoids content (P = 0.0057) according to the spacing factor, presenting greater average content (5,294.23 µg of carotenoids. 100 g -1 fresh mass) in the smallest spacing (25 cm/9.8 in).
Regarding the capacity to kidnap DPPH free radical from sorrel methanolic extracts, it was verified that antioxidant action was detected only when assessed on the greatest concentration (1.0 mg.mL -1 ); this was not observed on the smallest concentrations (0.1 mg.mL -1 and 0.01 mg.mL -1 ), and there was a significant difference (P<0.05) in relation to fertilization dosages (P<0.0001).
The average amount of antioxidant activity in the dosages of 25 and 50 t ha -1 of weather-beaten cattle manure was equal to 73.0% (Table 2), and it was the one closest to the average amounts presented by BHT (butylhydroxytoluene) patterns, vitamin C and quercetin equal to 83.85, 91.72 and 86.61%, respectively. There was a significant variation between the amounts of antioxidant activity in relation to the fertilization dosages and reference patterns that were assessed. There was not a significant difference between the dosages of 25 and 50, and they showed an antioxidant capacity superior to that of the control group (without application of weatherbeaten cattle manure) and to those of the plants that received dosages of 100 and 75 (t ha -1 ) of weatherbeaten cattle manure. Similar results were described by Arbos and assistants (2010), who detected an antioxidant capacity of 72.3, 77.2 and 77.3% at a 1.0 mg mL -1 concentration for organic-produced lettuce, common chicory and arugula, respectively. These amounts were superior to those obtained in conventional cropping -62.7, 66.6 and 70.1% for lettuce, common chicory and arugula, respectively. Therefore, results of organic lettuce produced with cattle-manure-based bio-fertilizers were the closest ones to sorrel samples when cropped with 25 and 50 t ha -1 of weather-beaten cattle manure.
Antioxidants have several modes of action, and the different methods used to determine these activities are performed under different conditions (RODRIGUES-AMAYA et al., 2008). There was no correlation between carotenoids (correlation coefficient: -0.0737; P=0.604) nor between phenolic compounds (correlation coefficient: 0.1661; P=0.3056) and sorrel's antioxidant capacity. The correlation between total phenolic compounds content and antioxidant capacity is still very controversial, given that it is possible to stablish a good correlation between those two variables in several fruits and vegetables, but in some cases that correlation is low (VIZZOTTO et al., 2012). These authors attribute it to the presence of several other phytochemicals, such as carotenoids or vitamins (C and E) that may influence on the antioxidant activity. Thus, the detected antioxidant activity can be attributed to other compounds that were not assessed.

CONCLUSIONS
The fertilization with weather-beaten cattle manure and cropping spacing have an influence on carotenoids content, antioxidant activity and pH.
The other sorrel phytochemicals and physical-chemical characteristics were not altered by fertilization variations with weather-beaten cattle manure and spacing, which allows us to consider that growing sorrel is viable when smaller spacing is used, with the application of small dosages of manure.