Improving the antioxidant properties of quinoa flour through fermentation with selected autochthonous lactic acid bacteria

International Journal of Food Microbiology 241 (2017) 252–261

Contents lists available at ScienceDirect

International Journal of Food Microbiology journal homepage: www.elsevier.com/locate/ijfoodmicro

Improving the antioxidant properties of quinoa flour through fermentation with selected autochthonous lactic acid bacteria Carlo Giuseppe Rizzello a,⁎, Anna Lorusso a, Vito Russo a, Daniela Pinto b, Barbara Marzani b, Marco Gobbetti a a b

Dipartimento di Scienze del Suolo, della Pianta e degli Alimenti, University of Bari, 70126 Bari, Italy Giuliani S.p.A., 20191 Milano, Italy

a r t i c l e

i n f o

Article history: Received 5 March 2016 Received in revised form 14 August 2016 Accepted 27 October 2016 Available online 29 October 2016 Keywords: Bioactive peptides Quinoa Lactic acid bacteria Antioxidant Sourdough

a b s t r a c t Lactic acid bacteria strains, previously isolated from the same matrix, were used to ferment quinoa flour aiming at exploiting the antioxidant potential. As in vitro determined on DPPH and ABTS radicals, the scavenging activity of water/salt-soluble extracts (WSE) from fermented doughs was significantly (P b 0.05) higher than that of noninoculated doughs. The highest inhibition of linoleic acid autoxidation was found for the quinoa dough fermented with Lactobacillus plantarum T0A10. The corresponding WSE was subjected to Reverse Phase Fast Protein Liquid Chromatography, and 32 fractions were collected and subjected to in vitro assays. The most active fraction was resistant to further hydrolysis by digestive enzymes. Five peptides, having sizes from 5 to 9 amino acid residues, were identified by nano-Liquid Chromatography-Electrospray Ionisation-Mass Spectra/Mass Spectra. The sequences shared compositional features which are typical of antioxidant peptides. As shown by determining cell viability and radical scavenging activity (MTT and DCFH-DA assays, respectively), the purified fraction showed antioxidant activity on human keratinocytes NCTC 2544 artificially subjected to oxidative stress. This study demonstrated the capacity of autochthonous lactic acid bacteria to release peptides with antioxidant activity through proteolysis of native quinoa proteins. Fermentation of the quinoa flour with a selected starter might be considered suitable for novel applications as functional food ingredient, dietary supplement or pharmaceutical preparations. © 2016 Elsevier B.V. All rights reserved.

1. Introduction The interest for health-promoting functional foods, dietary supplements and pharmaceutical preparations, which contain peptides deriving from food proteins, is markedly increasing (Shadidi and Li, 2015). Bioactive peptides are defined as specific protein fragments that have a positive impact on the body function or condition, and may, ultimately, influence the human health (Kits and Weiler, 2003). Among the various bioactivities described, antioxidant activity, together with antihypertensive, immunomodulatory, antimicrobial, and antitumoral activities, gained a growing interest from the scientific community, Abbreviations: ABTS, 2,2′-azino-di-[3-ethylbenzthiazoline sulphonate]; ADI, acceptable daily intake; BHT, butylated hydroxytoluene; DCFH-DA, 2′,7′dichlorofluorescein diacetate; DPPH, 1,1-diphenyl-2-picrylhydrazyl; DY, dough yield; FBS, fetal bovine serum; GRAS, generally recognized as safe; MTT, (3-(4,5-dimethyl-2yl)-2,5-diphenyltetrazolium bromide); nano-LC-ESI-MS/MS, nano-Liquid Chromatography-Electrospray Ionisation-Mass Spectra/Mass Spectra; OPA, ophtaldialdehyde; PUFA, polyunsaturated fatty acids; QF, quinoa flour; ROS, reactive oxygen species; RP-FPLC, Reversed-Phase Fast Performance Liquid Chromatography; RSA, radical scavenging activity; TFA, trifluoroacetic acid; TFAA, total free amino acids; TTA, total titratable acidity; WSE, water/salt-soluble extracts. ⁎ Corresponding author at: Dipartimento di Scienze del Suolo, della Pianta e degli Alimenti, Via G. Amendola 165/a, 70126 Bari, Italy. E-mail address: [email protected] (C.G. Rizzello).

http://dx.doi.org/10.1016/j.ijfoodmicro.2016.10.035 0168-1605/© 2016 Elsevier B.V. All rights reserved.

food industry, and consumers (Korhonen and Pihlanto, 2007; Sarmadi and Ismail, 2010; Shadidi and Li, 2015). Overall, the interest for antioxidant peptides has increased thanks to the abundant evidences of the in vivo prevention of oxidative stresses, which are mainly associated to degenerative aging diseases (e.g., cancer and arteriosclerosis) (Adebiyi et al., 2009). Antioxidants have a large potential for food industries. Delay of food discoloration and deterioration, which occur because of the oxidation, undoubtedly enhances food preservation. Radical mediated oxidation of fats and oils is one of the major causes of spoilage for lipid containing foods during processing and storage (Rajapakse et al., 2005). Biologically active peptides with potential antioxidant activity were derived from many animal and plant protein sources (Korhonen and Pihlanto, 2007; Shadidi and Li, 2015). Production and/or isolation from peanut kernels, rice bran, sunflower protein, alfalfa leaf protein, corn gluten meal, frog skin, yam, egg-yolk protein, milk-kefir and soymilk kefir, mushroom, mackerel, curry leaves, cotton leaf worm, casein, algae protein waste, wheat gluten and buckwheat protein were already reported (Sarmadi and Ismail, 2010). Recently, antioxidant peptides were isolated from various cereal flours fermented with sourdoughs (Coda et al., 2012). In particular, a pool of lactic acid bacteria, selected based on proteolytic activities, had the capacity to release antioxidant peptides (8 to 57 amino acid residues) during sourdough

C.G. Rizzello et al. / International Journal of Food Microbiology 241 (2017) 252–261

fermentation of whole wheat, spelt, rye, and kamut doughs. The antioxidant activity of these peptides on mouse fibroblasts, which were artificially subjected to oxidative stress, was also described (Coda et al., 2012). Quinoa (Chenopodium quinoa Willd.) is a seed crop, which is traditionally cultivated in the Andean region since thousands of years. Commonly, quinoa grains and flour are used for human consumption and animal feeding (Rizzello et al., 2015a). Quinoa has the capacity of adapting to a range of agro-ecological conditions, showing tolerance to frost, salinity and drought, and having the potential to grow on marginal soils. These features, together with an undoubtedly high nutritional value, determine the worldwide interest for this crop (Stikic et al., 2012). During the last years, the production of quinoa markedly increased, thus emphasizing the opportunity to cultivate this crop in various regions (Stikic et al., 2012). FAO selected the quinoa as one of the crops that are destined to offer food security in the 21st century (Jacobsen et al., 2003). The high nutritional value of quinoa seeds is mainly due to the high concentrations of proteins, minerals, and vitamins (Fleming and Galwey, 1995). Quinoa proteins are rich in amino acids like lysine, threonine and methionine, which are deficient in cereals. Recently (Rizzello et al., 2015a), autochthonous lactic acid bacteria were isolated from quinoa flour and spontaneously fermented doughs (Rizzello et al., 2015a). Strains, selected for pro-technological features, were used as starters to get quinoa sourdough. Free amino acids, soluble fibers, total phenols, phytase and antioxidant activities, and the in vitro protein digestibility markedly increased during fermentation. These results encouraged the use of quinoa and selected starters for the manufacture of novel and healthy leavened baked goods (Rizzello et al., 2015a). This study aimed at investigating the antioxidant potential of quinoa flour, which was subjected to fermentation with autochthonous and selected lactic acid bacteria. Bioactive peptides were purified, identified and characterized for the antioxidant properties in vitro, also using human keratinocytes NCTC 2544.

253

seeds through the laboratory mill Ika-Werke M20 (GMBH, and Co. KG, Staufen, Germany). Protein (total nitrogen × 5.7), lipids, ash and moisture contents were determined according to the AACC approved methods 46-11A, 30-10.01, 08-01, and 44-15A, respectively (AACC, 2010). The determination of insoluble and soluble dietary fibers was carried out by AOAC approved methods 991.42 and 993.19, respectively (Horwitz and Latimer, 2006). Total carbohydrates (%) were calculated as the difference: 100 − (moisture + proteins + lipids + ash). The characteristics of the flour used in this study are reported in Table 1. The 26 autochthonous lactic acid bacteria strains were cultivated into their respective media at 30 °C for 24 h. Cells were harvested singly by centrifugation (10,000 × g, 10 min, 4 °C), washed twice in 50 mM sterile potassium phosphate buffer (pH 7.0) and re-suspended in tap water at the cell density of ca. 8.0 log cfu/mL. Quinoa flour (30 g) and 70 mL of tap water, containing the above cellular suspension of each lactic acid bacterium (cell density in the dough of ca. log 7.0 cfu/g), were used to prepare 100 g of dough (DY of 330). Mixing was done for 5 min. Doughs were fermented at 37 °C for 24 h, under stirring conditions (ca. 200 rpm), according to the conditions previously set-up by Coda et al. (2011b) to maximize the release of antioxidant peptides. Two not inoculated quinoa doughs (DY 330) were produced. Incubation was lasting 0 (Ct0) and 24 h (Ct24) at 37 °C and doughs were used as controls. The pH value of doughs was determined by a pH meter (Model 507, Crison, Milan, Italy) with a food penetration probe. Total titratable acidity (TTA) was determined after homogenization of 10 g of dough with 90 mL of distilled water, and expressed as the amount (mL) of 0.1 M NaOH needed to reach the value of pH of 8.3. Lactic acid bacteria were enumerated by plating serial dilutions of doughs into MRS, mMRS, or SDB agar media, supplemented with cycloheximide (0.1 g L). Plates were incubated at 30 °C for 48 h, under anaerobiosis (AnaeroGen and AnaeroJar, Oxoid).

2.3. Water/salt-soluble extracts 2. Materials and methods 2.1. Microorganisms Lactic acid bacteria strains, previously (Rizzello et al., 2015a) isolated from quinoa, were used in this study. Lactobacillus plantarum T0B3, T0A10, T0A6, T0A2, T0C2, T0C3 and T0C1, and Lactobacillus rossiae T0A16 were from quinoa flour (T0); Lactobacillus plantarum T1B6, T1B16, T1A14, T1A12, T1A11 and T1C17, and Pediococcus pentosaceus T1B11, T1A13 and T1C1 were from quinoa flour dough (DY, dough weight × 100 / flour weight, of 160) subjected to spontaneous fermentation at 30 °C for 16 h (T1); and L. plantarum T6B4, T6B14, T6B10, T6A14, T6A10, T6A4, T6C16, T6C20 and T6C5 were from quinoa type I sourdough (T6), which was made and propagated through the traditional protocol commonly used for wheat flour fermentation, without using starter cultures or baker's yeast (Pontonio et al., 2015). All the strains were previously identified genotypically through sequencing of the 16S rDNA gene (Pontonio et al., 2015). Lactic acid bacteria were cultivated on different culture broth, depending on the isolation medium (Rizzello et al., 2015a): MRS (Oxoid, Basingstoke, Hampshire, United Kingdom) (strains labeled with letter A); modified MRS (mMRS), containing 1% [wt/vol] maltose, and 5% [vol/vol] fresh yeast extract, pH 5.6, (Oxoid) (strains labeled with letter B), and SDB (sourdough bacteria broth) (strains labeled with letter C). 2.2. Fermentation Organic quinoa (Chenopodium quinoa) dehulled seeds, imported from Argentina (Fundacion Nuevagestion, San Ignacio de Loyola, Jujuy), were used in this study. Quinoa flour (QF) was obtained from

Water/salt-soluble extracts (WSE) were prepared from each dough, according to the method originally described by Osborne (1907) and further modified by Weiss et al. (1993), at the end of incubation. An aliquot of each dough (containing 3.75 g of flour) was diluted with 15 mL of 50 mM Tris-HCl (pH 8.8), held at 4 °C for 1 h, vortexing at 15-min intervals, and centrifuged at 20,000 ×g for 20 min. The supernatants, containing the water/salt-soluble nitrogen fraction, were stored at −20 °C before the in vitro assay for determining the antioxidant activity. The peptide concentration of WSE was determined by the ophtaldialdehyde (OPA) method (Church et al., 1983). A standard curve was prepared using tryptone (0.25 to 1.5 mg/mL) and used as the reference. The use of peptone gave a similar standard curve. The concentration of total free amino acids (TFAA) of WSE was analyzed by a Biochrom 30 series Amino Acid Analyzer (Biochrom Ltd., Cambridge Science Park, England) with a Na-cation-exchange column (20 by 0.46 cm internal diameter), as described by Rizzello et al. (2010).

Table 1 Characteristics of the quinoa flour used for lactic acid bacteria fermentation. Proximal composition Moisture (%) Proteins (%) Lipids (%) Carbohydrates (%) Soluble fibers (%) Insoluble fibers (%) Ash (%)

11.4 ± 0.6 12.5 ± 0.8 5.3 ± 0.4 69.3 ± 3.5 1.2 ± 0.4 7.9 ± 2.0 1.9 ± 0.2

Three samples were twice analyzed. Mean values ± standard deviations were reported.

254

C.G. Rizzello et al. / International Journal of Food Microbiology 241 (2017) 252–261

2.4. Radical scavenging activity

2.7. Purification of antioxidant peptides

The scavenging effect of WSE obtained from quinoa doughs was assayed using DPPH free radical and ABTS cation radical assays. In details, the scavenging activity on 1,1-diphenyl-2-picrylhydrazyl (DPPH) free radical was measured according to the method of Shimada et al. (1992) with some modifications. Freeze-dried WSE were dissolved in 0.1 M phosphate buffer at pH 7.0, at the final concentration of 1 mg/mL of peptides, and 2 mL of each solution were added to 2 mL of 0.1 mM DPPH dissolved in 95% ethanol. The mixture was shaken and left for 30 min at room temperature, and the absorbance of resulting solution was read at 517 nm. The absorbance measured after 10 min was used for the calculation of the DPPH scavenged by WSE (Rizzello et al., 2010). A lower absorbance represents a higher DPPH scavenging activity. The scavenging effect was expressed as shown in the following equation: DPPH scavenging activity (%) = [(blank absorbance − sample absorbance) / blank absorbance] × 100. Butylated hydroxytoluene (BHT) and α-tocopherol (1 mg/mL) were also assayed as antioxidant references. In order to describe correlations among descriptors, values of pH, peptides and FAA concentrations, and DPPH radical scavenging activity of all quinoa doughs were subjected to Pearson correlation function by using Microsoft Xlstat software (Version 2014.5.03). The WSE and the related partially purified fractions, were further characterized by the ABTS assay (Rizzello et al., 2010). The radical cation (2,2′-azino-di-[3-ethylbenzthiazoline sulphonate]) (ABTS+) scavenging capacity was measured using the Antioxidant Assay Kit CSO790 (Sigma Chemical Co.), following the manufacturer's instruction. Trolox (6-hydroxy 2,4,7,8-tetramethylchroman-2-carboxylic acid) was used as the antioxidant standard. The scavenging activity was expressed as Trolox equivalent.

The WSE obtained from the quinoa dough fermented with L. plantarum T0A10 was fractionated by ultra-filtration (Ultrafree-MC centrifugal filter units, Millipore) using three different membrane sizes: 50, 30 and 10 kDa cut-off. After ultra-filtration, the permeate collected at 50, 30 and 10 kDa cut-off (fractions A, B and C, respectively) was used for ABTS scavenging activity assay as described above. The 10 kDa partially purified fraction was further fractionated (32 fractions) by Reversed-Phase Fast Performance Liquid Chromatography (RP-FPLC), using a Resource RPC column and an ÄKTA FPLC equipment, with the UV detector operating at 214 nm (GE Healthcare Bio-Sciences AB, Uppsala, Sweden). Aliquots, containing 1 mg/mL of peptides, were added to 0.05% (v/v) trifluoroacetic acid (TFA) and centrifuged at 10,000 ×g for 10 min. The supernatant was filtered with a 0.22 μm pore size filter and loaded onto the column. Gradient elution was performed at a flow rate of 1 mL/min using a mobile phase composed of water and acetonitrile (CH3CN), containing 0.05% TFA. The concentration of CH3CN was increased linearly from 5 to 46% between 16 and 62 min, and from 46 to 100% between 62 and 72 min. Solvents were removed from collected fractions by freeze drying. Fractions were re-dissolved in sterile water to determine the peptide concentration through the OPA method, and subjected to in vitro assays for antioxidant activity (ABTS) (Fig. 5).

2.5. Radical scavenging activity on methanol extracts The radical scavenging activity was also determined on methanol extract (ME) of quinoa doughs. Five grams of sample were mixed with 50 mL of 80% methanol to get ME. The mixture was purged with nitrogen stream for 30 min, under stirring condition, and centrifuged at 4600 ×g for 20 min. ME were transferred into test tubes, purged with nitrogen stream and stored at ca. 4 °C before analysis. The free radical scavenging capacity was determined using DPPH, as reported by Yu et al. (2003). A blank reagent was used to verify the stability of DPPH• over the test time. 75 ppm butylated hydroxytoluene (BHT) as the antioxidant reference. The scavenging activity was calculated as described above.

2.8. Proteolysis and heat stability of purified fraction The purified fraction from WSE, which showed the highest antioxidant activity, was subjected to sequential protein hydrolysis by digestive enzymes according to the method described by Pasini et al. (2001). Briefly, a freeze-dried aliquot of the partially purified fraction, corresponding to 10 mg of peptides, was suspended in 400 μL of 0.2 N HCl (pH 2.0), containing 0.05 mg/mL of pepsin (EC 3.4.23.1) (Sigma Aldrich CO., St. Louis, MO), and homogenized in a Sterilmixer Lab (PBI International). After 30 min of incubation at 37 °C under stirring conditions (150 rpm), 115 μL of a solution of 1 M boric acid and 0.5 N NaOH, adjusted to pH 6.8 with 5 N HCl, containing 0.25 mg/mL of pancreatin (Sigma), were added. The resulting pH was 7.6. Pancreatic digestion was lasting 150 min. Digested sample was heated for 5 min at 100 °C and centrifuged at 12,000 ×g for 20 min, to recover the supernatant. After treatments, sample was transferred into dialysis tubes (0.5 kDa cut-off, Spectrum Lab, Rancho Dominguez, CA) and dialyzed against water for 24 h at 4 °C. Before in vitro assays for antioxidant activity, sample was freeze-dried and dissolved in 0.1 M phosphate buffer at pH 7.0, at the final concentration of 2.15 mg/mL, corresponding to the concentration of the partially purified fraction tested before the enzymatic digestion.

2.6. Inhibition of linoleic acid autoxidation 2.9. Identification of antioxidant peptides The antioxidant activity of WSE was also measured according to the method of Osawa and Namiki (1985), with some modifications. After freeze-drying, 1.0 mg of each sample was suspended in 1.0 mL of 0.1 M phosphate buffer (pH 7.0), and added to 1 mL of linoleic acid (50 mM), previously dissolved on ethanol (99.5%). Incubation in a glass test tube, tightly sealed with silicon rubber cap, was allowed at 60 °C in the dark for 8 days. The degree of oxidation was determined by measuring the values of ferric thiocyanate according to the method described by Mitsuta et al. (1996). One hundred microliters of the above sample were mixed with 4.7 mL of 75% (v/v) ethanol, 0.1 mL of 30% (w/v) ammonium thiocyanate, and 0.1 mL of 0.02 M ferrous chloride, dissolved in 1 M HCl. After 3 min, the degree of color development that represents the oxidation of linoleic acid was measured spectrophotometrically at 500 nm. Butylated hydroxytoluene (BHT) (1 mg/mL) was also assayed as antioxidant references. A reference sample (without the addition of antioxidants) was included in the assay as negative control.

The fraction of WSE with the highest radical-scavenging activity was subjected to a second step of purification through RP-HPLC, under the conditions described previously, and using an ÄKTA Purifier apparatus (GE Healthcare Bio-Sciences Corp., Piscataway, New Jersey, USA). The centers of the peaks were collected, freeze dried and used for mass spectrometry analysis. Identification of peptides was carried out by nanoLiquid Chromatography-Electrospray Ionisation-Mass Spectra/Mass Spectra (nano-LC-ESI-MS/MS), using a Finnigan LCQ Deca XP Max ion trap mass spectrometer (ThermoElectron) through the nano-ESI interface. According to manufacturer's instrument settings for nano-LC-ESIMSMS analyses, MS spectra were automatically taken by Xcalibur software (ThermoElectron), in positive ion mode. MS/MS spectra were processed using the software BioWorks 3.2 (ThermoElectron) generating peaklists suitable for database searches. Peptides were identified using MS/MS ion search of Mascot search engine (Matrix Science, London, England) and NCBInr protein database (National Centre for

C.G. Rizzello et al. / International Journal of Food Microbiology 241 (2017) 252–261

Biotechnology Information, Bethesda, USA). For identification of peptides the following parameters were considered: enzyme: “none”; instrument type: “ESI-trap”; peptide mass tolerance: ± 0.1% and fragment mass tolerance: ±0.5 Da. Results from peptide identification were subjected to a manual evaluation, as described by Chen et al. (2005), and the validated peptide sequences explained all the major peaks in the MS/MS spectrum. 2.10. Protective effect on oxidative-induced stress in keratinocytes NCTC 2544 The human keratinocyte cell line NCTC 2544 was obtained from the National Institute for Cancer Research of Genoa, Italy. Cells were cultivated in RPMI medium supplemented with 10% fetal bovine serum (FBS), 2 mM L-glutamine, 1% penicillin (10,000 U/mL)/ streptomycin (10,000 μg/mL) mixture and maintained in 25 cm2 culture flasks at 37 °C, 5% CO2. Every two days confluent cultures were splitted 1:3–1:6, after washing with phosphate buffered saline (PBS) (without Ca2+ and Mg2+), using trypsin/EDTA, and seeded into 96-well plates at 2–5 · 104 cells/well. MTT assay was used for the determination of the viability of H2O2stressed NCTC 2544 cells (Coda et al., 2012; Rizzello et al., 2013). Cells were incubated with a purified peptide fraction for 16 h. The concentrations of freeze-dried peptide fraction in the reaction mixture were 0.01, 0.1, and 1 mg/mL, while α-tocopherol (250, 500 and 1000 μg/mL) was used as the positive control. After treatment, medium was removed from each well and, after washing, cells were exposed to 1 mM hydrogen peroxide (100 μL/well) for 2 h. Two controls, one without addition of the peptide fraction, and another without hydrogen peroxide treatment, were used. After the incubation, MTT (3-(4,5-dimethyl-2-yl)-2,5-diphenyltetrazolium bromide) assay was performed (Mosmann, 1983). The capacity of succinate dehydrogenase to convert 3-(4,5-dimethylthiazol-2-yl)2,5-diphenyltetrazolium bromide into visible formazan crystals was assessed (Curiel et al., 2015). Data were expressed as the percentage of viable cells compared to the control culture, not subjected oxidative stress. Each experiment was carried out in triplicate. 2.11. Intracellular reactive oxygen species (ROS) generation Production of reactive oxygen species (ROS) was monitored spectrofluorometrically on human keratinocytes NCTC 2544 using the

255

2′,7′-dichlorofluorescein diacetate (DCFH-DA) assay, as described by Tobi et al. (2000) and Curiel et al. (2015). In details, cells were cultivated as described above and treated for 16 h with 0.01, 0.1, and 1 mg/mL of the purified peptide fraction as described above. H2O2-stressed cells were used as control. After the incubation with the peptide fraction, the medium was removed and washed twice with PBS. Cells were treated with DCFH-DA dissolved in DMSO (final concentration 100 μM) for 30 min at 37 °C, in the dark. Then, cells were treated with 100 μL of pre-warmed RPMI (2,5% FBS) containing 1 mM H2O2 for 2 h at 37 °C, in the dark. At the end of the treatment, cells were washed twice, lysed with Cell Lytic M lysis buffer (Sigma Aldrich), added with 1% protease inhibitor cocktail (Sigma Aldrich) and transferred into a black 96-well plate. Fluorescent 2′,7′dichlorofluorescein (DCF) was read fluorometrically using a Fluoroskan Ascent FL Microplate Fluorescence Reader (Thermo Scientific) at excitation and emission wavelengths of 485 and 538 nm, respectively. Each experiment was carried out in triplicate. 2.12. Statistical analysis Data were subjected to one-way ANOVA; pair-comparison of treatment means was achieved by Tukey's procedure at P b 0.05, using the statistical software Statistica (Statistica7.0 per Windows). 3. Results 3.1. Quinoa flour fermentation After 24 h of fermentation at 37 °C, the inoculated doughs harbored a cell density of lactic acid bacteria ranging from 3.2 ± 0.4 to 4.3 ± 0.5 × 109 cfu/g. No significant (P N 0.05) differences were found between fermented quinoa doughs. Before fermentation, the value of pH was 5.94 ± 0.04 (Ct0). After fermentation, the values of pH varied from 2.60 ± 0.03 to 3.16 ± 0.02, with a median value of 2.70 ± 0.12 (Fig. 1A). The lowest value of pH was found when Lactobacillus plantarum T0A10 was used as starter. The highest values were found with L. plantarum T6B14 (3.16 ± 0.01), followed by Pediococcus pentosaceus T1C1 (2.95 ± 0.02), representing the extreme and the outlier values in the box-plot analysis, respectively (Fig. 1A). Nevertheless, the values of pH found for fermented doughs were lower than those found for control doughs Ct0 and Ct24 (5.94 ± 0.02 and 4.22 ± 0.02, respectively). TTA of control doughs was 20.0 ± 0.2 and 26.8 ± 0.3 for Ct0 and Ct24, respectively. The use of starter lactic acid bacteria significantly

Fig. 1. pH (A); total titratable acidity, TTA (B); peptide concentration (mg/mL of water/salt soluble extract, C), and total free amino acid concentration (g/kg of dough, D) of quinoa doughs (DY 330) incubated at 37 °C for 24 h and inoculated with lactic acid bacteria strains isolated from raw quinoa flour, quinoa flour dough incubated at 30 °C for 16 h, and quinoa type I sourdough. Red and blue lines refer to the values of not inoculated quinoa doughs before (Ct0) and after (Ct24) incubation at 37 °C for 24 h, respectively. Median values are represented (■). The top and the bottom of the box represent the 75th and 25th percentile of the data, respectively. The top and the bottom of the bars represent the 5th and the 95th percentile of the data, respectively. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

256

C.G. Rizzello et al. / International Journal of Food Microbiology 241 (2017) 252–261

Fig. 2. DPPH radical scavenging activity of the water/salt-soluble extracts from quinoa doughs (DY 330) incubated at 37 °C for 24 h and inoculated with lactic acid bacteria strains isolated from raw quinoa flour, spontaneously fermented quinoa dough, and quinoa type I sourdough. Two not inoculated quinoa doughs were obtained at 0 (Ct0) and 24 h (Ct24) of incubation at 37 °C and used as controls. BHT (1 mg/mL) was used as the positive control. Data are the means from three independent experiments. Bars represent standard deviation. a–cColumns with different superscript letters differ significantly (P b 0.05).

(P b 0.05) increased TTA. In particular, the values ranged from 27.2 ± 0.2 (L. plantarum T6B10) to 54.0 ± 0.1 (L. plantarum T6C5), with a median value of 47.1 ± 5.98 (Fig. 1B). The peptide concentration of all fermented doughs significantly (P b 0.05) increased compared to Ct0 (15.08 ± 0.01 mg/mL) and Ct24 (20.92 ± 0.01) (Fig. 1C). The highest concentrations were found for L. plantarum T6A4, L. plantarum T6A14, L. plantarum T6B10, and L. plantarum T0A10. The median value was 31.55 ± 7.78 mg/mL. The concentration of total free amino acids (TFAA) of quinoa dough before incubation (Ct0) was 400 ± 23 mg/kg. A significant (P b 0.05) increase was found during incubation of Ct24, probably due to endogenous protease activity (Fig. 1D). The concentration of TFAA further increased with lactic acid bacteria. The range for fermented doughs was very wide, varying from 1565 ± 44 to 3959 ± 38 mg/kg of dough, with a median value was 2075 ± 585 mg/kg (Fig. 1D). The highest and lowest values were found, respectively, for doughs fermented with L. plantarum T0A6, T6B10 and T1A11, and L. plantarum TOC3. 3.2. In vitro antioxidant activity Despite the above differences and due to extraction with Tris-HCl buffer, WSE from control and fermented doughs had values of pH ranging from 6.6 ± 0.1 to 7.0 ± 0.3. During radical scavenging assay, the colored stable DPPH radical is reduced to non-radical DPPH-H, when in the presence of an antioxidant or a hydrogen donor. DPPH radical without antioxidants was stable over the time. Under the assay conditions, the 100% of activity corresponds to the complete scavenging of DPPH radical (0.1 mM final concentration) after 10 min of incubation by the antioxidant compounds. According to previous studies (Rizzello et al., 2008; Wanita and Lorenz, 1996), the color intensity of DPPH• showed a logarithmic decline when in the presence of butylated hydroxytoluene (BHT, 1 mg/mL). In particular, BHT showed a scavenging activity of 84.8 ± 0.8%. The WSE from Ct0 and Ct24 had the lowest values of antioxidant activity (9.5 ± 0.9% and 19.3 ± 1.1%, respectively). The scavenging activity of the fermented quinoa doughs varied between 32.7 ± 0.6 and 84.3 ± 1.0% (Fig. 2). The doughs inoculated with L. plantarum T6A10, T0C1, and T6C5 showed the lowest values: 32.8 ± 0.6, 34.6 ± 0.8 and 43.6 ± 0.6%, respectively. The highest scavenging activity was found when fermentation was carried out with L. plantarum T6B4 (84.3 ± 1.0%), T6B4 (81.9 ± 1.3%), T6C16 (81.9 ± 0.9%) and T0A10 (81.3 ± 0.7%). No Pearson correlations between the scavenging activity and pH, peptides and FAA concentrations were found (Table 1S). Compared to WSE, the ME from fermented quinoa doughs had a markedly lower radical scavenging activity (Fig. 1S), that ranged from 49 ± 2.0 to 55 ± 1.4%. In any case, the values were significantly (P b 0.05) lower than BHT (78 ± 1.0%), and no significant (P N 0.05) differences were found among the fermented samples. A very slight increase in the antioxidant activity was found in fermented doughs

compared to Ct0 and Ct24, thus hypothesizing a marginal role of the lactic acid bacteria fermentation on the extractable methanol compounds (e.g. polyphenols) Due to these reasons, the further characterization was performed only on WSE. ABTS assay is based on the formation of ferryl myoglobin radical from metmyoglobin and hydrogen peroxide, which cause the oxidation of ABTS to ABTS+, the chromogen radical cation. In the presence of an antioxidant agent such as Trolox (water-soluble vitamin E analog), the chromatogenic reaction is suppressed. BHT (1 mg/mL) was used as the antioxidant standard, and an ABTS scavenging activity of 0.278 ± 0.020 mmol/L Trolox was observed under the experimental conditions of this study. As expected, the activity of Ct0 was the lowest (0.015 ± 0.001 mmol/L Trolox) (Fig. 3). A slight increase was found for Ct24 (0.045 ± 0.003 mmol/L Trolox). The antioxidant activity of the WSE from doughs fermented with L. plantarum T1B6, T6B4, and T0A10 did not differ significantly (P b 0.05), and ranged from 0.272 ± 0.003 to 0.291 ± 0.002 mmol/L Trolox. All the other WSE showed values significantly (P N 0.05) lower (0.151 ± 0.002–0.180 ± 0.002 mmol/L Trolox). Lipid peroxidation is thought to proceed via radical mediated abstraction of hydrogen atoms from methylene carbons in polyunsaturated fatty acids (Qian et al., 2008). Compared to the reference (reaction mixture without antioxidants), the presence of WSE from quinoa doughs inhibited the linoleic acid autoxidation (Fig. 4). In agreement with the previous findings, the oxidation of linoleic acid was markedly inhibited by the addition of WSE from doughs fermented with L. plantarum T0A10 and T6B4. Both the WSE inhibited the oxidation more effectively than BHT. The lowest antioxidant activity

Fig. 3. Scavenging capacity (on ABTS+) of the water/salt-soluble extracts from selected quinoa doughs (DY 330) incubated at 37 °C for 24 h and inoculated with Lactobacillus plantarum T1B6, T6B4, T0A10, and T6C16. BHT (1 mg/mL) was used as the antioxidant standard. The scavenging activity was expressed as Trolox equivalent (mmol/L). Data are the means from three independent experiments. Bars represent standard deviation. a–c Columns with different superscript letters differ significantly (P b 0.05).

C.G. Rizzello et al. / International Journal of Food Microbiology 241 (2017) 252–261

257

3.3. Purification and characterization of antioxidant peptide fractions

Fig. 4. Lipid peroxidation inhibitory activity of the WSE from quinoa doughs (DY 330) incubated at 37 °C for 24 h and inoculated with Lactobacillus plantarum T1B6, T6B4, T0A10, and T6C16. The activity was measured under linoleic acid oxidation system for 8 days. BHT (1 mg/mL) was used as the positive control. The reaction mixture without the addition of the antioxidant was considered as a negative control (reference, rf). Data are the means from three independent experiments. Bars represent standard deviation. a–d Points with different superscript letters differ significantly (P b 0.05).

corresponded to Ct24, while L. plantarum T1B6 and T6C16 had an intermediate behavior. Aiming at identifying antioxidant peptides, the WSE from the quinoa dough fermented with L. plantarum T0A10, showing the highest in vitro activity in all the assays, including the inhibition of linoleic acid, was subjected to ultra-filtration (cut-off 50, 30 and 10 kDa) and further assayed for radical scavenging activity. All the fractions, corresponding to the permeates from ultra-filtration, showed the same (P N 0.05) activity towards ABTS, since the active compounds were not retained by any of the membranes used, suggesting that the molecular mass of the antioxidant compounds was lower than 10 kDa (data not shown).

Thirty-two fractions were collected through RP-FPLC separation of WSE from the quinoa dough fermented with L. plantarum T0A10. In particular, an aliquot of the WSE corresponding to 10 mg of peptides (as determined by the OPA method) was fractionated. Collected fractions were freeze-dried, dissolved in ca. 600 μL of distilled water, and assayed for ABTS scavenging activity. The activity of the fractions ranged from 0.004 ± 0.001 to 0.311 ± 0.035 mmol/L Trolox. Fraction n.2 (qf2), which showed the highest activity, was further characterized. Its peptide concentration was 2.15 ± 0.09 mg/mL. No statistical correlation was found between the concentration of peptides and the antioxidant activities (data not shown), thus hypothesizing that the chemical characteristics of the peptides collected in specific fraction play a determinant role on the antioxidant activity (Coda et al., 2012; Sarmadi and Ismail, 2010). The purified fraction was subjected to sequential hydrolysis by pepsin, trypsin and pancreatin, which mimicked the digestive process. As determined by the ABTS scavenging assay, fraction qf2 showed a decrease of the antioxidant activity lower than 8% compared to undigested fraction (0.290 ± 0.011 vs 0.311 ± 0.035 mmol/L Trolox). Moreover, the antioxidant activity of qf2 was not affected by heating for 5 min at 100 °C. 3.4. Identification of antioxidant peptides Five peptides, having sizes from 5 to 9 amino acid residues, and hydrophobic ratios from 11 to 80%, were identified by nano-LC-ESI-MS/MS analysis (Table 2). All the peptides were found in the NCBInr database as encrypted in different quinoa proteins (the accession numbers are reported in Table 2). IVLVQEG and TLFRPEN, which are encrypted in a quinoa globulin B (seed storage protein), showed molecular masses of 756.9 and 896.9 Da respectively. The total net charge was − 1 and 0. VGFGI and FTLIIN, both released from a salt overly sensitive protein, had molecular masses of 537.7 and 719.881 Da, respectively. A neutral

Fig. 5. Reverse Phase-Fast Protein Liquid Chromatography (RP-FPLC) chromatogram of the purified fraction (ultrafiltration through membrane size of 10 kDa cut-off) from the water/saltsoluble extract obtained from quinoa dough fermented at 37 °C for 24 h with Lactobacillus plantarum T0A10. The dashed line refers to the gradient of eluent B. The 32 fractions collected and the corresponding scavenging activity on ABTS+ were represented (♦). Data are the means from three independent experiments. Bars represent standard deviation. a–ePoints with different superscript letters differ significantly (P b 0.05).

258

C.G. Rizzello et al. / International Journal of Food Microbiology 241 (2017) 252–261

Table 2 Sequences of peptides contained in the purified fraction qf2 of the water/salt-soluble extract obtained from quinoa dough fermented at 37 °C for 24 h with Lactobacillus plantarum T0A10.

Peptide number

Sequencea

Score

Calculated mass

Expected mass

Delta

Source protein NCBI accession number

1 2 3 4 5

IVLVQEG TLFRPEN VGFGI FTLIIN LENSGDKKY

24 21 19 22 23

757.403 870.954 538.834 719.149 1053.653

756.900 869.982 537.682 719.881 1053.130

−0.503 −0.975 −1.152 0.732 −0.523

11S seed storage globulin B; ABI94736.1 11S seed storage globulin B; ABI94736.1 Salt overly sensitive; ABS72166.1 Salt overly sensitive; ABS72166.1 Maturase K CCI55135.1

a

The single-letter amino acid code is used.

total net charge was determined. LENSGDKKY (1053.1 Da), also having total neutral net charge, was encrypted in a maturase K protein. VGFGI, FTLIIN, and IVLVQEG were characterized by the highest hydrophobic ratio: 80, 66, and 57%, respectively. TLFRPEN and LENSGDKKY showed lower values: 28 and 11%, respectively. 3.5. Effect of purified peptides on viability of oxidation induced cells and intracellular reactive oxygen species (ROS) generation To further investigate the capacity of the purified mixture of peptides to act as radical scavenger, keratinocytes NCTC 2544 were grown in the presence of freeze-dried fraction qf2. Afterwards, cells were treated with hydrogen peroxide. Cell viability was assayed through the capacity of functional mitochondria to catalyze the reduction of MTT to formazan salt via mitochondrial dehydrogenases. Compared to cell viability after oxidative stress from reference sample (17.2 ± 2.6%, Fig. 6A), α-tocopherol and qf2 significantly (P b 0.05) increased cell survival (Fig. 6A). At all concentrations assayed, the protective effect of the purified fraction qf2 was significantly (P b 0.05) higher than α-tocopherol

250 μM. No significant (P N 0.05) differences were found compared to α-tocopherol at 500 μM (Fig. 6A). To determine the potential of the peptide mixture against oxidative stress-mediated injuries, keratinocytes NCTC 2544 were pre-treated with the purified freeze-dried fraction qf2 and α-tocopherol, further incubated with DCFH-DA, and stressed with H2O2. DCFH-DA was used as a probe to assess the formation of intracellular hydrogen peroxide by the fluorescence reader. The assay correlates to the emitted fluorescence with the ROS intracellular production and, consequently, to the antioxidant activity (Tobi et al., 2000). Data are reported as percentage of radical scavenging activity (RSA) compared to cells stressed with H2O2, without antioxidant. As expected, cells pre-treated with 500 and 1000 μM of α-tocopherol showed RSA higher than 70% (Fig. 6B), while a lower value was found when 250 μM were used. No significant (P N 0.05) differences were found between treatments with 0.01 or 1 mg/mL of the purified fraction. The highest activity (68.7 ± 5.5%) was found when pre-treatment was carried out with the intermediate concentration of 0.1 mg/mL. In this case, the activity was statistically similar (P b 0.05) to those found for treatments with 500 and 1000 μM of α-tocopherol. The appearance of the prooxidant activity at the highest concentration of the peptide fraction was already reported for other natural antioxidant compounds when tested in a biological system at high doses (Bouayed and Bohn, 2010). It was hypothesized that it can be due to the occurrence of different mechanisms related to cytotoxicity and apoptosis (Bouayed and Bohn, 2010). 4. Discussion

Fig. 6. In vitro assay on human keratinocytes NCTC2544. A) Effect of purified peptide fraction qf2 on cell viability. Oxidative stress was artificially induced with hydroxide peroxide treatment. The percentage of viable cells was measured with the 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. rf, reference culture incubated without antioxidants; αtp, treatments with α-tocopherol (250, 500 μM and 1 mM); qf2, treatments with the purified peptide fraction (0.01, 0.1 and 1 mg/mL). Each sample was tested in triplicate. Bars represent standard deviation. a–c Columns with different superscript letters differ significantly (P b 0.05). B) Effect of purified peptide fraction qf2 on the radical scavenging activity (RSA) of mouse fibroblasts after oxidative stress, as estimated by 2′,7′-dichlorofluorescein diacetate (DCFH-DA) assay. Oxidative stress was artificially induced with hydroxide peroxide treatment. The RSA of the H2O2-stressed cells incubated without antioxidant compounds (reference, rf) was also included. Data are reported as percentage of radical scavenging activity (RSA) compared to cells stressed with H2O2, without antioxidant. Each sample was tested in triplicate. Bars represent standard deviation. a–cColumns with different superscript letters differ significantly (P b 0.05).

The legal regulation for using antioxidants in foods varies depending on the countries (Mikovà, 2007). Although many synthetic compounds have an antioxidant potential, only a few of them have the GRAS (generally recognized as safe) status to be permitted in food products by international organizations such as the Joint FAO/WHO Expert Committee on Food Additives and the European Community's Scientific Committee for Food. Toxicological studies are crucial to determine the safety and the acceptable daily intake (ADI) level (Mikovà, 2007). ADI for the most common antioxidants (e.g., BHA, BHT and gallates) changed over the years because of the toxicological effects (Hettiarachchy et al., 1996; Mikovà, 2007; Park et al., 2001; Wanita and Lorenz, 1996), thus limiting the use (Mikovà, 2007). The negative interference with food flavor is another limit (Mikovà, 2007). Recently, natural antioxidants attracted the interest of many food industries because of the consumers request for healthy foods (Mikovà, 2007). Although synthetic antioxidants might be more effective, natural antioxidants have simpler structure, higher stability and non-hazardous immune-reaction (Sarmadi and Ismail, 2010). The most attractive natural antioxidants mainly belong to the chemical classes of phenols and peptides. Usually, bioactive peptides correspond to cryptic sequences from native proteins, which are released through hydrolysis by digestive enzymes, microbial and plant proteolytic enzymes and during food fermentation (Korhonen and Pihlanto, 2007). Overall, antioxidant peptides have low molecular mass (b6.0 kDa), low cost and high activity (Sarmadi and Ismail, 2010). They act as inhibitors of lipid peroxidation, direct scavengers of free radicals and/or as agents for chelating transition metal ions that catalyze the generation of radical species

C.G. Rizzello et al. / International Journal of Food Microbiology 241 (2017) 252–261

(Sarmadi and Ismail, 2010). Usually, the antioxidant activity is correlated with the amino acid composition, conformation and hydrophobicity (Chen et al., 2005). Despite the large literature on bioactive peptides from milk or from other animal proteins, the preparation of bioactive peptides from vegetable proteins through proteolysis by lactic acid bacteria is recommended (Korhonen and Pihlanto, 2007; Rizzello et al., 2011, 2015b). Despite quinoa is recognized as a rich source of proteins, to the best of our knowledge no studies have investigated the potential release of antioxidant peptides during fermentation. Recently, the antioxidant activity was only related to phenolic components (Swieca et al., 2014; Tang et al., 2015). Lactic acid bacteria fermentation has a well-known role in improving the nutritional properties of cereal flours and related baked goods (Katina et al., 2005). The breakdown of proteins (proteolysis) by lactic acid bacteria plays an important role in generating peptides and amino acids for bacterial growth. The proteolytic system of lactic acid bacteria comprises three major components: (i) cell-wall bound proteinases that initiate the degradation of extracellular proteins into oligopeptides, (ii) peptide transporters that take up the peptides into the cell, and (iii) various intracellular peptidases that degrade the peptides into shorter peptides and amino acids (Liu et al., 2010). The presence and the expression of proteinases and peptidases with different specificity largely vary among lactic acid bacteria strains (De Angelis et al., 2007a, 2007b; Liu et al., 2010; Savijoki et al., 2006), making necessary specific strain selection for different targets. First, this study reported the capacity of lactic acid bacteria to release peptides with antioxidant activity through proteolysis of native quinoa proteins. According to previous studies (Coda et al., 2012; Rizzello et al., 2008), fermentation with selected lactic acid bacteria was allowed for long time and under semi-liquid conditions to fully exploit microbial proteolysis (Coda et al., 2012). Proteinase activity and, especially, a large portfolio of peptidases are the pre-requisites to release bioactive peptides during sourdough fermentation (De Angelis et al., 2007a, 2007b; Gobbetti et al., 2010). Compared to allochthonous or commercial starters, which are mainly tailored for wheat flour fermentation, the selection of strains within the autochthonous microbiota is another important pre-requisite (Coda et al., 2010) for rapid adaptation, intense acidification and positive influence on the nutritional and technological properties (Rizzello et al., 2015a). The use of the selected strain as starter allows the obtainment of standardized and repeatable results (Coda et al., 2014). Based on these considerations, strains isolated from quinoa flour, spontaneously fermented quinoa dough and type I quinoa sourdough (Rizzello et al., 2015a) were used in this study. A preliminary evaluation of the main pro-technological characteristics (Coda et al., 2014) of the autochthonous strains was made, confirming for all a growth of ca. 2 logarithmic cycles, and a different impact on acidification and release of peptides and free amino acids, that are considered as crucial characteristics of the starters selected for fermented food applications, such as the baked goods production (Rizzello et al., 2015a). The antioxidant activity of the not inoculated dough (control) was very low. This suggested that quinoa endogenous enzymes and spontaneous fermentation did not allow a significant release of antioxidant compounds. Although the strains used in the study have been isolated from quinoa matrices, the final cell density obtained by a spontaneous fermentation cycle is lower of several logarithmic cycles, compared to the inoculated doughs. Moreover, in the absence of the inoculation, it is unlikely that the selected biotype, if present, certainly dominates the fermentation process (Coda et al., 2014). Water/salt-soluble extracts (WSE) from several fermented doughs showed elevated in vitro radical scavenging activity and inhibition of linoleic acid autoxidation. Doughs fermented with L. plantarum T0A10, T1B6 and T6B4 had the highest in vitro activities, which were comparable to or higher than 1 mg/mL BHT. The contribution of the polyphenols on the antioxidant activity was investigated on methanol extracts. Contrary to WSE, only a slight increase of the activity was found on methanol extract as the consequence of fermentation, without significant differences among the

259

strains inoculated as starters. Overall, lactic acid bacteria might affect the polyphenols profile through the release of phenolic compounds through the hydrolysis of more complex and glycosylated forms (Nionelli et al., 2014) or through the acidification, that can improve their solubilization (Nionelli et al., 2014). Nevertheless, due to the low contribution of the fermentation to the antioxidant activity of the methanol extracts, only the WSE, containing the proteolysis derivatives, were further characterized. Many factors such as the level and type of proteins (e.g., polymorphism, ratio between protein fractions, amino acid composition and sequence, and molecular masses of individual polypeptides) differentiate the technological, structural, nutritional and functional properties of the various food matrices (Coda et al., 2012). Therefore, differences on the functional features of peptides deriving from different flours (e.g., cereal, pseudocereals, and legumes) are expected (Coda et al., 2012; Rizzello et al., 2015b, 2015c). Aiming at purifying antioxidant peptides, fraction 2 (peptide concentration lower than 3 mg/mL) from quinoa dough fermented with L. plantarum T0A10 was selected through RP-FPLC separation. This fraction eluted in the early zone of the acetonitrile gradient. The in vitro antioxidant activity of this fraction was higher than that of the synthetic antioxidant used as the positive control. As previously reported (Minervini et al., 2003; Rizzello et al., 2008), purified fractions might show higher bioactivity than the respective raw extract, due to the highest concentration of active compounds. To be active, antioxidant peptides should have the capacity to overcome hydrolysis and modifications at the intestine level, and to reach their targets (Sarmadi and Ismail, 2010). The antioxidant activity of the purified fraction seemed to be not affected by sequential in vitro treatments with digestive enzymes. The antioxidant activity of the purified fractions was also determined towards human keratinocytes NCTC 2544, which were artificially subjected to oxidative stress. As shown by MTT assays, the purified fraction exhibited a marked protective effect at all the concentration tested. The effect on the survival of keratinocytes was comparable to that of 500 μM α-tocopherol. The protective effect was also investigated through the determination of the intracellular ROS production and detoxification by DCFH-DA assay. Also in this case and at low concentrations, the purified fraction showed antioxidant activity on human keratinocytes similar to that of α-tocopherol. Nevertheless, the capacity of peptides to cross the intestinal barrier should be evaluated by simulation of intestinal absorption experiments, and the potential towards human health validated through in vivo assays. As shown by nano-LC-ESI-MS/MS analysis, the purified fraction contained five peptides. None of these peptides was previously reported as antioxidant (Sarmadi and Ismail, 2010; Shadidi and Li, 2015). Overall, it was hypothesized that the strongest antioxidant activity had to be ascribed to synergic effect between peptides rather than to the activity of a single peptide (Coda et al., 2012). The identified peptides showed low molecular masses as typical features of well-known antioxidant peptides (Sarmadi and Ismail, 2010). The presence of residues like Tyr (Y), Trp (W), Met (M), Lys (K), Pro (P), Cys (C), His (H), Val (V), Leu (L) and Ala (A) confirmed the antioxidant features (SalmenkallioMarttila et al., 2001). In particular, the composition of IVLVQEG and LENSGDKKY mainly consisted of the above amino acid residues (43 and 44%, respectively). Hydrophobic amino acids enhance the solubility of peptides in lipids, thus facilitating the access to hydrophobic radical species and to hydrophobic PUFA (polyunsaturated fatty acids) (Sarmadi and Ismail, 2010). Indeed, three of the identified sequences had hydrophobic ratios higher than 50% (IVLVQEG, VGFGI, and FTLIIN). The highest level of hydrophobicity was found for VGFGI (80%). Nevertheless, residues such as Cys, which directly interacts with SH group of radicals (Pasini et al., 2001), and His, which confers antioxidant activity by the hydrogen-donating, lipid peroxyl radical trapping and/or the metal ionchelating ability of the imidazole group (Rajapakse et al., 2005), were found in the identified peptides. LENSGDKKY has Leu at the N-terminal, which was already shown as another typical feature for antioxidant peptides (Sarmadi and Ismail, 2010; Suetsuna and Chen, 2002). With the only

260

C.G. Rizzello et al. / International Journal of Food Microbiology 241 (2017) 252–261

exception of IVLVQEG, all the identified peptides contain aromatic amino acids, which may donate protons to electron deficient radicals. This feature markedly improves the radical scavenging activity (Sarmadi and Ismail, 2010). Recently, Orsini Delgado et al. (2016) identified a mixture of antioxidant peptides obtained after a simulated gastrointestinal digestion of amaranth (a pseudocereal belonging to Amaranthaceae, the same family of quinoa) proteins, that are characterized by similar number of amino acidic residues and same molecular mass range compared to those we found after quinoa fermentation. In particular, the major part of the peptides they identified derives from the hydrolysis of the amaranth 11S globulin, characterized by similarity N70% to the quinoa 11S globulin (in which IVLVQEG and TLFRPEN are encrypted). Nutraceutical industry and preventive medicine are currently showing a marked interest for natural antioxidant compounds because of their potential application in food, cosmetic and pharmaceutical products to replace synthetic somewhat carcinogenous antioxidants (Aleksic and Knezevic, 2014). The demand for dietary phytonutrients encourages the exploitation of plant potential through lactic acid fermentation (Curiel et al., 2015; Rizzello et al., 2013). In agreement with this perspective, this study shows that selected lactic acid bacteria have the capacity to generate antioxidant peptides during fermentation of quinoa flour. The fermentation conditions that were set up in this study are easily applicable at industrial level for making novel bakery products with high nutritional and functional values. The purified peptides exhibited bioactive features that are compatible with various antioxidant mechanisms, thus indicating a presumptive protection against free radicals. Besides the application in foods, these features could lead to the production of innovative formulations and to the design of new synthetic peptides for cosmetic and pharmaceutical applications. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ijfoodmicro.2016.10.035.

References AACC - American Association of Cereal Chemists, 2010. Approved methods of analysis. St. Paul. Approved Methods Committee. Available from: http://methods.aaccnet.org/. Adebiyi, A.P., Adebiyi, A.O., Yamashita, J., Ogawa, T., Muramoto, K., 2009. Purification and characterization of antioxidative peptides derived from rice bran protein hydrolysates. Eur. Food Res. Technol. 228, 553–563. Aleksic, V., Knezevic, P., 2014. Antimicrobial and antioxidative activity of extracts and essential oils of Myrtus communis L. Microbiol. Res. 169, 240–254. Bouayed, J., Bohn, T., 2010. Exogenous antioxidants—double-edged swords in cellular redox state. Health beneficial effects at physiologic doses versus deleterious effects at high doses. Oxidative Med. Cell. Longev. 3, 228–237. Chen, Y., Know, S.W., Kim, S.C., Zhao, Y., 2005. Integrated approach for manual evaluation of peptides identified by searching protein sequence databases with tandem mass spectra. J. Proteome Res. 4, 998–1005. Church, F.C., Swaisgood, H.E., Porter, D.H., Catignani, G.L., 1983. Spectrophotometric assay using o-phthaldialdehyde for determination of proteolysis in milk and isolated milk proteins. J. Dairy Sci. 66, 1219–1227. Coda, R., Nionelli, L., Rizzello, C.G., De Angelis, M., Tossut, P., Gobbetti, M., 2010. Spelt and emmer flours: characterization of the lactic acid bacteria microbiota and selection of mixed starters for bread making. J. Appl. Microbiol. 108, 925–935. Coda, R., Di Cagno, R., Rizzello, C.G., Nionelli, L., Edema, M.O., Gobbetti, M., 2011b. Utilization of African grains for sourdough bread making. J. Food Sci. 76, 329–335. Coda, R., Rizzello, C.G., Pinto, D., Gobbetti, M., 2012. Selected lactic acid bacteria synthesize antioxidant peptides during sourdough fermentation of cereal flours. Appl. Environ. Microbiol. 78, 1087–1096. Coda, R., Di Cagno, R., Gobbetti, M., Rizzello, C.G., 2014. Sourdough lactic acid bacteria: exploration of non-wheat cereal-based fermentation. Food Microbiol. 37, 51–58. Curiel, J.A., Pinto, D., Marzani, B., Filannino, P., Farris, G.A., Gobbetti, M., Rizzello, C.G., 2015. Lactic acid fermentation as a tool to enhance the antioxidant properties of Myrtus communis berries. Microb. Cell Factories 14, 67. De Angelis, M., Di Cagno, R., Gallo, G., Curci, M., Siragusa, S., Crecchio, C., Parente, E., Gobbetti, M., 2007a. Molecular and functional characterization of Lactobacillus sanfranciscensis strains isolated from sourdoughs. Int. J. Food Microbiol. 114, 69–82. De Angelis, M., Rizzello, C.G., Alfonsi, G., Arnault, P., Cappelle, S., Di Cagno, R., Gobbetti, M., 2007b. Use of sourdough lactobacilli and oat fibre to decrease the glycemic index of white wheat bread. Br. J. Nutr. 98, 1196–1205.

Fleming, J.E., Galwey, N.W., 1995. Quinoa (Chenopodium quinoa). In: Williams, J.T. (Ed.), Cereals and Pseudocereals. Champman & Hall, London, pp. 3–83. Gobbetti, M., Di Cagno, R., De Angelis, M., 2010. Functional microorganisms for functional food quality. Crit. Rev. Food Sci. Nutr. 50, 716–727. Hettiarachchy, N.S., Glenn, K.C., Gnanasambandam, R., Johnson, M.G., 1996. Natural antioxidant extract from fenugreek (Trigonella foenumgraecum) for ground beef patties. J. Food Sci. 61, 516–519. Horwitz, W., Latimer, G., 2006. Official Methods of Analysis of AOAC International. 18th ed. AOAC International, Gaithersburg, MD. Jacobsen, S.E., Mujica, A., Jensen, C.R., 2003. The resistance of quinoa (Chenopodium quinoa Willd.) to adverse abiotic factors. Food Rev. Int. 19, 99–109. Katina, K., Arendtb, E., Liukkonena, K.-H., Autioa, K., Flandera, L., Poutanen, K., 2005. Potential of sourdough for healthier cereal products. Trends Food Sci. Technol. 16, 104–112. Kits, D.D., Weiler, K., 2003. Bioactive proteins and peptides from food sources. Applications of bioprocesses used in isolation and recovery. Curr. Pharm. Des. 9, 1309–1323. Korhonen, H., Pihlanto, A., 2007. Bioactive peptides from food proteins. In: YH, H. (Ed.), Handbook of Food Products Manufacturing. Health, Meat, Milk, Poultry, Seafood and Vegetables. Wiley Interscience, Hoboken, New Jersey, pp. 5–39. Liu, M., Bayjanov, J.R., Renckens, B., Nauta, A., Siezen, R.J., 2010. The proteolytic system of lactic acid bacteria revisited: a genomic comparison. BMC Genomics 11. Mikovà, K., 2007. The regulation of antioxidants in foods. In: Rahman'S (Ed.), Handbook of Food Preservation, 2nd ed. Taylor & Francis Group, Boca Raton, Florida, pp. 267–283. Minervini, F., Algaron, F., Rizzello, C.G., Fox, P.F., Monnet, V., Gobbetti, M., 2003. Angiotensin I-converting-enzyme-inhibitory and antibacterial peptides from Lactobacillus helveticus PR4 proteinase-hydrolized caseins of milk from six species. Appl. Environ. Microbiol. 69, 5297–5305. Mitsuta, H., Yasumoto, K., Iwami, K., 1996. Antioxidative action of indole compounds during the autoxidation of linoleic acid. Eiyo to Shohkuryo 19, 210–214. Mosmann, T., 1983. Toxicity determined in vitro by morphological alterations and neutral red absorption. J. Immunol. Methods 65, 56–63. Nionelli, L., Curri, N., Curiel, J.A., Di Cagno, R., Pontonio, E., Cavoski, I., Gobbetti, M., Rizzello, C.G., 2014. Exploitation of Albanian wheat cultivars: characterization of the flours and lactic acid bacteria microbiota, and selection of starters for sourdough fermentation. Food Microbiol. 44, 96–107. Orsini Delgado, M.C., Nardo, A., Pavlovic, M., Rogniaux, H., Añón, M.C., Tironi, V.A., 2016. Identification and characterization of antioxidant peptides obtained by gastrointestinal digestion of amaranth proteins. Food Chem. 197, 1160–1167. Osawa, T., Namiki, M., 1985. Natural antioxidant isolated from Eucalyptus leaf waxes. J. Agric. Food Chem. 33, 777–780. Osborne, T.B., 1907. The Proteins of the Wheat Kernel. Carnegie Institute of Washington, publication 84. Judd and Detweiler, Washington, DC. Park, P.J., Jung, W.K., Nam, K.S., Shahidi, F., Kim, S.K., 2001. Purification and characterization of antioxidative peptides from protein hydrolysate of lecithin-free egg yolk. J. Am. Oil Chem. Soc. 78, 651–656. Pasini, G., Simonato, B., Giannattasio, M., Peruffo, A.D.B., Curioni, A., 2001. Modifications of wheat flour proteins during in vitro digestion of bread dough, crumb, and crust: an electrophoretic and immunological study. J. Agric. Food Chem. 49, 2254–2261. Pontonio, E., Nionelli, L., Curiel, J.A., Sadeghi, A., Di Cagno, R., Gobbetti, M., Rizzello, C.G., 2015. Iranian wheat flours from rural and industrial mills: exploitation of the chemical and technology features, and selection of autochthonous sourdough starters for making breads. Food Microbiol. 47, 99–110. Qian, Z.J., Jung, W.K., Kim, S.K., 2008. Free radical scavenging activity of a novel antioxidative peptide purified from hydrolysate of bullfrog skin, Rana catesbeiana Shaw. Bioresour. Technol. 6, 1690–1698. Rajapakse, N., Mendis, E., Jung, W.K., Je, J.Y., Kim, S.K., 2005. Purification of a radical scavenging peptide from fermented mussel sauce and its antioxidant properties. Food Res. Int. 38, 175–182. Rizzello, C.G., Cassone, A., Di Cagno, R., Gobbetti, M., 2008. Synthesis of angiotensin Iconverting enzyme (ACE)-inhibitory peptides and _-aminobutyric acid (GABA) during sourdough fermentation by selected lactic acid bacteria. J. Agric. Food Chem. 16, 6936–6943. Rizzello, C.G., Nionelli, L., Coda, R., De Angelis, M., Gobbetti, M., 2010. Effect of sourdough fermentation on stabilisation, chemical and nutritional characteristics of wheat germ. Food Chem. 119, 1079–1089. Rizzello, C.G., Cassone, A., Coda, R., Gobbetti, M., 2011. Antifungal activity of sourdough fermented wheat germ used as an ingredient for bread making. Food Chem. 127, 952–959. Rizzello, C.G., Coda, R., Sánchez-Macías, D., Pinto, D., Marzani, B., Filanino, P., Giuliani, G., Paradiso, V.M., Di Cagno, R., Gobbetti, M., 2013. Lactic acid fermentation as a tool to enhance the functional features of Echinacea spp. Microb. Cell Factories 12, 44. Rizzello, C.G., Lorusso, A., Montemurro, M., Gobbetti, M., 2015a. Use of sourdough made with quinoa (Chenopodium quinoa) flour and autochthonous selected lactic acid bacteria for enhancing the nutritional, textural and sensory features of white bread. Food Microbiol. 56, 1–13. Rizzello, C.G., Hernández-Ledesma, B., Fernández-Tomé, S., Curiel, J.A., Pinto, D., Marzani, B., Coda, R., Gobbetti, M., 2015b. Italian legumes: effect of sourdough fermentation on lunasin-like polypeptides. Microb. Cell Factories 14, 168. Rizzello, C.G., Lavecchia, A., Gramaglia, V., Gobbetti, M., 2015c. Long-term fungal inhibition by Pisum sativum flour hydrolysate during storage of wheat flour bread. Appl. Environ. Microbiol. 81, 4195–4206. Salmenkallio-Marttila, M., Katina, K., Autio, K., 2001. Effect of bran fermentation on quality and microstructure of high-fibre wheat bread. Cereal Chem. 78, 429–435. Sarmadi, B.H., Ismail, A., 2010. Antioxidative peptides from food proteins: a review. Peptides 31, 1949–1956.

C.G. Rizzello et al. / International Journal of Food Microbiology 241 (2017) 252–261 Savijoki, K., Ingmer, H., Varmanen, P., 2006. Proteolytic systems of lactic acid bacteria. Appl. Microbiol. Biotechnol. 71, 394–406. Shadidi, F., Li, Q., 2015. Biologically active peptides from foods. In: Ustunol, Z. (Ed.), Applied Food Protein Chemistry. John Wiley & Sons, Ltd, Chichester, UK. Shimada, K., Fujikawa, K., Yahara, K., Nakamura, T., 1992. Antioxidative properties of xanthan on the antioxidation of soybean oil in cyclodextrin emulsion. J. Agric. Food Chem. 40, 945–948. Stikic, R., Glamoclija, D., Demin, M., Vucelic-Radovic, B., Jovanovic, Z., Milojkovic-Opsenica, D., Jacobsen, S.E., Milovanovic, M., 2012. Agronomical and nutritional evaluation of quinoa seeds (Chenopodium quinoa Willd.) as an ingredient in bread formulations. J. Cereal Sci. 55, 132–138. Suetsuna, K., Chen, J.R., 2002. Isolation and characterization of peptides with antioxidant activity derived from wheat gluten. Food Sci. Technol. Res. 8, 227–230. Swieca, M., Seczyk, L., Gawlik-Dziki, U., Dziki, D., 2014. Bread enriched with quinoa leaves – the influence of protein – phenolics interactions on the nutritional and antioxidant quality. Food Chem. 162, 54–62.

261

Tang, Y., Li, X., Zhang, B., Chen, P.X., Liu, R., Tsao, R., 2015. Characterization of phenolics, betanins and antioxidant activities in seed of three Chenopodium quinoa Willd. genotypes. Food Chem. 166, 380–388. Tobi, S.E., Paul, N., McMillan, T.J., 2000. Glutathione modulates the level of free radicals produced in UVA-irradiated cells. J. Photochem. Photobiol. B 57, 102–112. Wanita, A., Lorenz, K., 1996. Antioxidant potential of 5-Npentadecylresorcinol. J. Food Process. Preserv. 20, 417–429. Weiss, W., Vogelmeier, C., Gorg, A., 1993. Electrophoretic characterization of wheat grain allergens from different cultivars involved in bakers' asthma. Electrophoresis 14, 805–816. Yu, L., Perret, J., Harris, M., Wilson, J., Haley, S., 2003. Antioxidant properties of bran extracts from Akron wheat grown at different locations. J. Agric. Food Chem. 51, 1566–1570.