Quality by Design in optimizing the extraction of (poly)phenolic compounds from Vaccinium myrtillus berries.

Quality by Design was adopted for developing an effective extraction procedure of (poly)phenolic compounds from bilberry (Vaccinium myrtillus L.) fruits, using a pooled sample of berries from different regions of Ukraine. Mechanical solvent extraction, microwave-assisted extraction (MAE) and ultrasonic-assisted extraction (UAE) were investigated by screening matrices. Extraction time (Time, from 5 to 15 min), organic solvent type (OS type, methanol, ethanol and acetone), organic solvent percentage (OS%, from 50% to 90%), sample/extractant ratio (S/E ratio, from 0.025 to 0.1 g mL), and, only for MAE, extraction temperature (T, from 30 to 60°C), were selected as critical method parameters (CMPs). The spectrophotometric assays total soluble polyphenols (TSP), total monomeric anthocyanins (TMA), and radical scavenging activity (evaluated by the 2,2-diphenyl-1-picrylhydrazyl radical (DPPH), the 2,2'-azino-di-(3-ethylbenzthiazoline sulfonic acid), and the ferric reducing antioxidant power methods) were chosen as critical method attributes (CMAs). The screening procedure allowed for selecting UAE and methanol, while the other CMPs underwent further optimization through Response Surface Methodology. Target values for TSP, TMA and DPPH were selected and the method operable design region (MODR) was defined by means of Monte-Carlo simulations. The optimized conditions, with the corresponding MODR intervals in bracket, were the following: (i) Time, 17 min (15-23 min); OS%, 56% (44-59%); S/E ratio, 0.030 (0.022-0.034) g mL. Under these experimental conditions, CMAs values of the pooled sample were the following (n = 3): TSP=4433±176 mg (+)-catechin eq/100 g dry weight (d.w.); TMA=3575±194 mg cyanidin-3-glucoside eq/100 g d.w.; DPPH=273±5 μg DPPH inhib./mg d.w. The optimized extraction method was tested for matrix effect (ME%) in the UHPLC-MS/MS analysis of 15 anthocyanins and 20 non-anthocyanins individual (poly)phenols commonly found in bilberries, as well as for luteolin, sinapic acid, and pelargonidin-3-glucoside, absent in this fruit and therefore added to the extracts as surrogate standards for evaluating apparent recovery (AR%). |ME%| was in any case ≤ 23% and AR% of the surrogate standards in the range 91-95%, confirming the very good performances of the optimized extraction method.