Following, extracts were filtered (Whatman No. 1) and concentrated under vacuum at 70 °C using a rotoevaporator (Buchi R-210 Rotavapor, Buchi Co., New Castel, DE). The concentrate was resuspended in 5 ml methanol and analysed by HPLC. For quantification of coumarin, 1 g of freeze-dried samples was suspended in 10 ml of extraction solution (ethanol:water at 1:1 ratio), macerated until completely homogeneous, and allowed to rest for 24 h at room temperature. The material was filtered (Whatman No. 1) and the extract obtained was analyzed by HPLC. Analysis were conducted in HPLC

Shimadzu LC-20A system equipped with LC-20DA pump, manual injector with selleck inhibitor a fixed volume of 100 μL, CTO-20A column oven set at 40 °C, running LC Solution software with UV–Vis detector model SPD-20A. The column was a Nova Pack C18 (CLC-ODS, 3 μm; 4.6 × 250 mm). For resveratrol quantification, the method described by Sautter et al. (2005), with modifications, was LBH589 chemical structure used. Briefly, the mobile phase consisted of water acidified to pH 2.9 using phosphoric acid (H3PO4) (Solution A) and acetonitrile (solution B) in a ratio of 75A:25B, isocratic with a flow rate of 1.2 ml/min, with an injection volume of 100 μL, UV detection at 306 nm, and a total run time of 15 min per sample at 40 °C. For coumarin quantification, the mobile phase consisted of water (Solution

A) and methanol (Solution B) in a ratio of 60A:40B, isocratic with a flow rate of 1.0 ml/min,

injection volume of 100 μl, UV detection at 274 nm, NADPH-cytochrome-c2 reductase and a total run time of 15 min per sample at 40 °C. The parameters obtained in the validation of the methods are shown on Table 1. Standard curves for resveratrol and coumarin were prepared under the same conditions. Resveratrol (0.46, 0.092, 0.046, 0.0184 and 0.0092 mg/ml) and coumarin (43.6, 21.8, 10.9, 5.45 and 2.73 mg/ml) standards were diluted in methanol. Initially, sample injections were made with resveratrol and coumarin standards, using the internal standard method, in order to identify these compounds in the sample runs. The co-injection consisted of sample and standard compound in a ratio of 1:1 with standard concentrations of 0.4 mg/ml and 1 mg/ml in methanol for resveratrol and coumarin, respectively. This experiment was based on a completely randomized design with equal replications. For all analyses, determinations were made in triplicate as independent experiments. Data analysis was performed using JMP v. 9 software (SAS Institute, Cary, NC) for anthocyanins, yellow flavonoids, β-carotene, lycopene, total phenolics, resveratrol, and coumarin. Differences between variables were tested for significance by one-way analysis of variance (ANOVA). Significantly different means (p < 0.05) were separated by the Tukey’s test. Data are presented as mean ± SD (standard deviation).