The contents of total polyphenols, flavonoids, condensed tannins and alkaloids are shown in Table 1. The AcOEt fraction presented the highest content of total polyphenols (546.57 ± 2.35 mg/g gallic acid equivalents (GAE)). The chloroform fraction, despite presenting the lowest content of total polyphenols, demonstrated the highest content of flavonoids, tannins and alkaloids. Similar findings were related by Janovik et al. [17], analyzing different fractions of Cariniana domestica. Alkaloids were only found in the CE and CHCl₃ fractions.

The DPPH assay is based on the measurement of the ability of an antioxidant substance to scavenge the radical, reducing it to hydrazine, with a simultaneous change in color from violet to yellow, which is measured spectrophotometrically. The AcOEt fraction showed excellent antioxidant capacity, superior even to the standard ascorbic acid, with an IC₅₀ of 9.11 ± 0.75 µg/mL while the standard had an IC₅₀ of 15.48 ± 1.28 µg/mL (Figure 1 and Table 2). The CE and n-BuOH fractions showed very similar IC₅₀ values, and finally, with higher IC₅₀ but still a relatively low value, came the CHCl₃ fraction (44.46 ± 1.27 µg/mL). These results are in agreement with the content of polyphenols found in the fractions (AcOEt > CE ≈ n-BuOH > CHCl₃), so is possible to attribute the antioxidant capacity seen in the DPPH assay to the presence of these compounds. This relationship is already well established and has been described by other authors, using similar assay systems [18,19].

Dichlorofluorescein (DCFH) is widely used to determine oxidative stress in cells. When its diacetate form (DCFH-DA) is added to the cells, it diffuses through the cell membrane and is hydrolyzed by intracellular esterase to liberate DCFH, which reacts with oxidizing species (particularly H₂O₂ and hydroxyl radical) forming the fluorescent compound 2′,7′-dichlorofluorescein (DCF) [20,21]. Thus, the principle of this assay is to evaluate the ability of antioxidant molecules in the extract to scavenge ROS produced by normal metabolism by cells, and then inhibit the oxidation of DCFH to DCF, observed by the decrease in fluorescence intensity.

The CHCl₃ fraction at a concentration of 250 µg/mL was able to significantly reduce the oxidation of DCFH and consequently reduce the oxidative stress observed in supernatant of rat brain homogenate, compared to the basal group (p < 0.05), demonstrating pronounced antioxidant activity. Ascorbic acid reduced in a significant manner the oxidation of DCFH-DA, at 125 and 250 μg/mL (p < 0.05). The CE and other fractions did not demonstrate this ability; no statistically significant difference compared to the basal group at the same concentrations tested was observed (Figure 2).

The antioxidant potential of the CE and fractions were evaluated through the TBARS assay based on the formation of malondialdehyde (MDA), a subproduct of lipid peroxidation. The lipid peroxidation was stimulated with FeSO₄ addition to brain supernatant, and the extracts were able to significantly decrease of MDA formation at all concentrations tested when compared with the induced control, as can be observed in Figure 3. Similarly, all the extracts reduced MDA formation at the basal levels (non-induced group), especially the CHCl₃ fraction, which showed a very low IC₅₀ (Table 2). It is suggested that this effect is due to the large amounts of flavonoids, tannins and alkaloids found in this fraction, contributing to the antioxidant activity, and also due to the demonstrated ability to scavenge ROS. In our study, ascorbic acid showed an IC₅₀ value of 117.81 ± 1.23 µg/mL, these results were in accordance with Liang et al., [22] who found moderate inhibition of lipid peroxidation for ascorbic acid using a similar method. The vehicle (ethanol) in which all the extracts were dissolved had no effect when administered alone. Sini and Devi [23] observing the antioxidant activity of the chloroform extract of Solanum trilobatum, verified the capacity to inhibit lipid peroxidation in rat liver homogenate, and also the scavenging effect on superoxide and hydroxyl radicals whose results were higher than the standards butylated hydroxytoluene (BHT) and catechin.

Protein carbonyl content is the most commonly used marker of protein oxidation, and its accumulation has been observed in several human diseases, including Alzheimer’s disease, diabetes, arthritis and others [24]. All the fractions were able to reduce protein carbonyl content in serum compared to the induced group (H₂O₂), in a dose-dependent way, but only the CHCl₃ fraction, at the concentration of 250 µg/mL completely inhibited the oxidative damage caused (Figure 4). This fraction showed the second lowest IC₅₀ for this assay (67.65 ± 0.82 µg/mL), slightly larger than the standard ascorbic acid (61.80 ± 0.54 µg/mL), however the difference between these two values was not significant (p < 0.05) (Table 2, Figure 4). Similarly to what was observed in the lipid peroxidation assay, this protective effect may be attributed to the phytochemical composition of the fraction and also the ability of scavenging ROS observed by DCFH-DA method, reducing the oxidative stress.

In our study, the CHCl₃ fraction showed the best results by all antioxidant activity methods, except in the DPPH assay. The steric accessibility of DPPH radical is a major determinant of the reaction, since small molecules that have better access to the radical site have relatively higher antioxidant capacity. Moreover, many antioxidant compounds that react quickly with peroxyl radicals may react slowly or may even be inert in this assay. The inexistence of DPPH or similar radicals in biological systems is also a shortcoming. Despite these limitations, the DPPH radical is a simple, stable and useful spectrophotometric method with regard to screening measuring the antioxidant capacity of both pure compounds and complex samples [25,26].

To verify the presence of phenolic compounds in the CE and fractions of S. guaraniticum, the samples were subjected to HPLC/DAD analysis. The phenolic acids such as chlorogenic, caffeic and rosmarinic were identified by comparison of their retention’s time and UV spectrum with those of the standards. The chromatograms and quantification of the compounds are shown in Figure 5 and Table 3, respectively.

Phenolic compounds may have different antioxidant capacities, depending on their structural conformation, number of hydroxyl groups and their distribution in the structure. In general, phenolic acids are considered efficient hydrogen donors due to their characteristic carboxylic acid group, which is easily ionized [27]. In this study, chlorogenic acid was identified in all extracts. However, the AcOEt fraction presented the highest amount of this phenolic acid (21.55 ± 0.73 mg/g), followed by CE (11.15 ± 0.14 mg/g), n-BuOH (11.60 ± 1.95 mg/g) and CHCl₃ (3.23 ± 1.04 mg/g). These results follow the same trend of the antioxidant capacity by the DPPH assay, suggesting that this polyphenol was a significantly contributor to the low IC₅₀ value presented by the AcOEt fraction in this method.

Rosmarinic acid was present in high amounts in the CE and n-BuOH fractions. This phenolic acid is known to have many biological activities, including hepatoprotective activities in liver diseases [28] which could be related with the popular use of this plant. In fact, Sabir and Rocha [15] demonstrated the hepatoprotective activity of jurubeba in an experimental model using mice. Our findings corroborate the results obtained by those authors. Similarly, in a study of Lin et al. [29], flavonoids and phenolic acids in the ethanol extract from lemon balm leaves (Melissa officinalis L.) were evaluated, and rosmarinic acid, present in the amount of 83.33 ± 3.46 mg/g of freeze dried sample extract and 78.40 ± 4.13 mg/g of extract of hot air dried sample, was considered as the major component. In addition, caffeic acid has also been found in the CE and n-BuOH fractions, in small amounts, and may also have contributed to the activities shown.

Similar behaviors were found comparing the CE and n-BuOH fractions in relation to the content of polyphenols and flavonoids, antioxidant capacity by DPPH assay and lipid peroxidation. This fact can be explained, in part, by similarity of their chromatographic profiles and also by very similar amounts of the three phenolic acids as quantified by HPLC/DAD.

All chemicals were of analytical grade. The solvents and reference compounds used for the extractions and analytical procedures such as chloroform, ethyl acetate, ethanol, n-butanol, gallic, chlorogenic and rosmarinic acids, and spectrophotometric grade methanol were purchased from Merck (Darmstadt, Germany). Folin-Ciocalteau reagent 2N, DPPH radical, bismuth nitrate, rutin, catechin, thiobarbituric acid and DCFH-DA were acquired from Sigma Chemical Co. (St. Louis, MO, USA). All others chemicals and reagents were purchased locally.

The leaves of Solanum guaraniticum were collected in December 2011, in the city of Guaporé (Rio Grande do Sul, Brazil). A voucher specimen is deposited in the herbarium of the Biology Department of Federal University of Santa Maria, and cataloged under the registration number SMDB 13158. Plant material (645.58 g) was dried in a stove at controlled temperature (38 °C), triturated and subjected to the maceration with 70% ethanol (4.3 L) for seven days, with daily agitation. The material was filtered and concentrated under reduced pressure, in order to obtain an aqueous extract. Part of this was taken to complete dryness, yielding the crude extract (CE), and another part was successively partitioned with chloroform, ethyl acetate and n-butanol, which also were dried to give each corresponding fraction. The fractionation method is illustrated below (Figure 6).

Male Wistar rats (3.0–3.5 months of age and weighing 270–320 g) were maintained groups of 3–4 rats per cage. They had continuous access to food and water in a room with controlled temperature (22 ± 3 °C) and on a 12 h light/dark cycle. The animals were maintained and used in accordance to the guidelines of the Brazilian Association for Laboratory Animal Science (COBEA). The rats were killed by decapitation and the brain tissue was rapidly dissected, weighed and immediately homogenized in Tris-HCl 10 mM, pH 7.5 (1/10, w/v). The homogenate was centrifuged for 10 min at 4,000 rpm and the supernatant was used for tests of scavenging of ROS by DCFH-DA method and inhibition of lipid peroxidation.

For inhibition of protein oxidation (carbonyl content) assay, human blood samples were collected by venipuncture and separated by centrifugation to obtain the serum. The experiments were carried out according to the research ethics committee of the Federal University of Pampa (Rio Grande do Sul, Brazil), and approved under number 23,081.

For phytochemical composition, antioxidant capacity by DPPH assay and HPLC/DAD analysis, was used a calibration curve, and the experimental values were statistically analyzed by analysis of variance. Data were expressed as mean ± S.D. (n = 3). For the data analysis of inhibition of lipid peroxidation, protein carbonyl content and scavenging of ROS by DCFH-DA method, was used analysis of variance (one-way ANOVA) followed by Tukey test, and p < 0.05 were considered significant. Data were expressed as mean ± S.E.M.