LWT - Food Science and Technology 158 (2022) 113162 Available online 28 January 2022 0023-6438/© 2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Development of antioxidant peptides from brewers’ spent grain proteins Ranithri Abeynayake a, Sitian Zhang a, Wenzhu Yang b,**, Lingyun Chen a,* a Department of Agricultural, Food & Nutritional Science, University of Alberta, Edmonton, Alberta, T6G 2P5, Canada b Lethbridge Research and Development Centre, Lethbridge, AB T1J 4B1, Canada A R T I C L E I N F O Keywords: Brewers’ spent grain (BSG) Protein hydrolysates Protease hydrolysis Antioxidant activities A B S T R A C T Brewers’ spent grain (BSG), the most abundant brewing by-product contains up to 24% (w/w) of protein on a dry basis but is used as low-value animal feed. This study was conducted to develop antioxidant peptides from BSG proteins. Protease hydrolysis significantly increased BSG protein solubility to 94.4% at neutral pH. Peptides prepared by Alcalase, and its combination with Neutrase, Flavourzyme, or Everlase showed the highest DPPH radical scavenging activities ranging between 72.6 and 74.9%. The highest superoxide radical scavenging ac- tivity of 19.3% was observed in the hydrolysate resulted from Alcalase and Flavourzyme combination. Everlase and FoodPro PHT combined treatment was the most effective in producing ferrous ion chelating peptides. Molecular structures analysis suggests that histidine significantly contributed to DPPH radical scavenging ac- tivity of BSG peptides due to the high proton donation ability of its imidazole ring. Highly hydrolyzed BSG protein could have more positive charges to stabilize negatively charged superoxide radicals. Ferrous ion chelating ability was negatively correlated to degree of hydrolysis, suggesting that longer peptides are more likely to form compact structures to trap ferrous ions. This research has demonstrated the potential to use BSG as a cost-effective raw material to generate natural antioxidants for food applications. 1. Introduction The cereal grain barley (Hordeum vulgare L.) was one of the first agriculturally domesticated grains as early as 10,000 years ago. Apart from its main use as animal feed, major food applications include brewing. Barley, the primary grain source used in brewing, contains 10–20 g/100 g protein on a dry basis (Houde, Khodaei, Benkerroum, & Karboune, 2018). Hordeins (A, B, C, D) also known as barley prolamins are the main barley protein, and the rest are glutelins, globulins, and albumins. In brewing, malted barley is produced by controlled germi- nation of barley for a short period. This is followed by mashing to remove soluble constituents, mainly sugars that resulted from starch hydrolysis. Although barley proteins are partially hydrolyzed to amino acids and peptides by endogenous proteases during malting, most of the proteins remain insoluble. On the other hand, due to removal of car- bohydrates by endosperm solubilization, the protein content is increased in mashing. Brewers’ spent grain (BSG) removed after mashing is the most abundant brewing by-product, and its global production is esti- mated to be more than 180 million tons per year (Patrignani, Brantsen, Awika, & Conforti, 2021). It is reported to contain 15–24 g/100 g protein on a dry basis (Mussatto & Roberto, 2005). BSG consists of al- bumin, globulin, hordein, and gluteline according to Osborne classifi- cation. All these fractions have good solubility in alkaline medium (Osborne, 1924). In this work, 53.7% of protein remained in insoluble components, likely due to the heat treatment during brewing that can trigger protein unfolding and aggregation and promote protein-cellulose interactions in BSG that restricted protein extraction. Also, previous work observed failure to extract B and C hordeins under non-reducing conditions, suggesting C hordeins are entrapped in disulfide linked ag- gregates formed by B hordeins during mashing (Celus, Brijs, & Delcour, 2006). Although, numerous attempts have been made to explore the commercial applications of BSG, still the main use of this protein rich by-product is low value animal feed, with the rest deposited in the landfill. Numerous benefits can be expected to gain by identifying higher value-added applications of BSG. With increasing research pieces of evidence to associate oxidative stress to occurrence and progression of chronic diseases (Carocho, Mo- rales, & Ferreira, 2018), the consumer interest in antioxidant containing foods as well as supplements is increasing. Oxidative stress is created by continuous exposure to reactive species or decreased functioning of * Corresponding author. ** Corresponding author. E-mail addresses: wenzhu.yang@canada.ca (W. Yang), lingyun.chen@ualberta.ca (L. Chen). Contents lists available at ScienceDirect LWT journal homepage: www.elsevier.com/locate/lwt https://doi.org/10.1016/j.lwt.2022.113162 Received 4 August 2021; Received in revised form 17 January 2022; Accepted 26 January 2022 mailto:wenzhu.yang@canada.ca mailto:lingyun.chen@ualberta.ca www.sciencedirect.com/science/journal/00236438 https://www.elsevier.com/locate/lwt https://doi.org/10.1016/j.lwt.2022.113162 https://doi.org/10.1016/j.lwt.2022.113162 https://doi.org/10.1016/j.lwt.2022.113162 http://crossmark.crossref.org/dialog/?doi=10.1016/j.lwt.2022.113162&domain=pdf http://creativecommons.org/licenses/by/4.0/ LWT 158 (2022) 113162 2 antioxidant defense mechanisms (Wong, Xiao, Wang, Ee, & Chai, 2020). A higher degree of oxidative stress can destruct macromolecules in biological systems leading to chronic diseases (Carocho et al., 2018). On the other hand, autoxidation of polyunsaturated lipids in food systems involves a free radical chain reaction leading to undesirable sensory properties, and deterioration of nutrition quality of foods (Zhong & Shahidi, 2015, pp. 287–333). Over the past several decades, natural antioxidants such as polyphenols, vitamin C, carotenoids, proteins, and peptides have attracted attention as they are generally considered safe (Yang et al., 2019). Antioxidant peptides have been reported from different sources such as soy protein (Moure, Domínguez, & Parajó, 2006), fish protein (Mendis, Rajapakse, Byun, & Kim, 2005; Sun, Zhang, & Zhuang, 2013), and barley hordein (Bamdad, Wu, & Chen, 2011). Compared to other natural antioxidant ingredients, peptides can exert antioxidant effects through multiple mechanisms, including chelating metal ions, sup- pressing reactive oxygen species, scavenging free radicals, and partici- pating in redox reactions. These activities are closely related to amino acid composition, hydrolyzation degree and molecular weight of pep- tides. Bioactive peptides containing a higher percentage of hydrophobic amino acids may have significantly higher antioxidant activity (Bam- dad, Shin, Suh, Nimalaratne, & Sunwoo, 2017). Thus, it is expected that BSG with high content of hydrophobic amino acids might be a good protein source to prepare hydrolysates with antioxidant effects. The overall aim of the project was to develop value-added applica- tions of protein from BSG, and high efficacy feed from the residue after protein removal. The objective of this study was to study the effect of protease hydrolysis in improving antioxidant activity of BSG peptides and to develop high efficacy feed from the residue after protein removal. Different proteases were used to prepare bioactive peptides including Alcalase (A), Flavourzyme (F), Neutrase (N), Everlase (E), FoodPro PHT (P), and their combinations (A+F, A+N, A+E, F+N, and E+P). Combi- nations were decided based on the optimum temperature and pH ranges of the enzymes reported in literature. Enzyme combinations were used to obtain a wide range of hydrolysis degrees and peptides of different sequences within the same hydrolysis duration, thus increasing the chance of obtaining peptides with desirable bioactivities. The peptide antioxidant activities were studied by reducing power, DPPH radical scavenging activity, ferrous chelating activity, and superoxide radical scavenging assay. The effects of the amino acid composition, solubility, hydrolysis degree, and molecular weight on the peptide antioxidant properties were discussed. Furthermore, the commercial antioxidants; butylated hydroxytoluene (BHT), and L-ascorbic acid were used as positive controls to understand the potential of using BSG antioxidant protein fractions as an antioxidant ingredient for food applications. 2. Materials and methods 2.1. Materials Brewers’ spent grain was kindly provided by a local brewing com- pany (Edmonton, AB, Canada). FoodPro PHT was obtained from DuPont Industrial Biosciences (Denmark). Flavourzyme (500 U/g), Alcalase (≥2.4 U/g), Everlase (≥16 U/g), Neutrase (≥0.8 U/g), sodium dodecyl sulfate (SDS), 2,4,6-trinitrobenzene sulfonic acid (TNBS), L-leucine, fluorescent dye 8-anilino-1-naphthalene sulphonic acid (ANS), Folin- Ciocalteau reagent, gallic acid, 1,1- diphenyl-2-picryl hydrazyl (DPPH), butylated hydroxytoluene (BHT), 3-(2-pyridyl)-5,6-bis (4-phenyl-sul- phonic acid)-1,2,4- triazine (ferrozine), ethylenediaminetetraacetic acid (EDTA), trichloroacetic acid (TCA), and standard molecular markers for HPLC analysis were obtained from Sigma Aldrich (St. Louis, MO, USA). L-ascorbic acid, potassium ferricyanide, and pyrogallol were obtained from Fischer Scientific (Edmonton, AB, Canada). All the other chemicals were of analytical grade. 2.2. Preparation of BSG bioactive peptides Brewers spent grain protein was extracted using alkali method fol- lowed by acid precipitation. In brief, BSG grounded to 0.5 mm was dispersed in 0.1 mol/L NaOH to make 20% (w/v) solution. Protein was extracted at 50 ◦C for 2 h. Followed by centrifugal separation of su- pernatant at 8,000×g for 15 min at 20 ◦C (Beckman Coulter Avanti J-E Centrifuge System, CA, USA). After adjusting supernatant pH to 4.0, protein was precipitated by centrifugation at 8,000×g for 15 min at 20 ◦C. Precipitated protein was freeze-dried followed by storage at 4 ◦C until further used. An aqueous suspension of 2 g/100 mL BSG protein was hydrolyzed with different proteases or protease combinations at their optimum pH and temperatures for 4 h (Table 1). The substrate was mixed with 4 g/ 100 g enzyme (based on protein dry weight). For treatment with two combined proteases, 2 g/100 g each enzyme was added to BSG sus- pension. The pH and temperature were monitored throughout the hy- drolysis. At the end of hydrolysis, enzymes were inactivated by heating at 80 ◦C for 20 min. Centrifugation at 8,000×g for 15 min at 20 ◦C was done to separate solubilized protein from insoluble substances. Collected supernatant was adjusted to pH 7.0 and freeze-dried (Fig. 1). Protein content of the extract and the hydrolysates was determined by nitrogen analyzer (FP-428, Leco Corporation, St. Joseph, MI, USA) using nitrogen conversion factor 6.25. 2.3. Degree of hydrolysis (DH) The DH was determined according to TNBS method (Adler-Nissen, 1979) with modifications. Samples (50 μL) were taken at different time intervals during hydrolysis and heated at 80 ◦C for 20 min to inactivate enzymes, then mixed with 950 μL of 1 g/100 mL sodium dodecyl sulfate (SDS) at 75 ◦C for 20 min to make a better sample suspension in SDS. Sample of 0.25 mL was mixed with 2.0 mL sodium phosphate buffer (0.2 mol/L, pH 8.2), and 2.0 ml of 100 mg/100 mL TNBS reagent, followed by incubation at 50 ◦C for 60 min. At the end of incubation, 4.0 mL of Table 1 Protease treatment conditions used to prepare BGS protein hydrolysates, and the relevant degree of hydrolysis (DH) values. Enzyme/ Enzyme combination Origin Specificity Conditions DH% pH Temp (oC) Flavourzyme Aspergillus oryzae Acidic endo and exopeptidase 6 50 24.1a Alcalase Bacillus licheniformis Alkaline endopeptidase specific 8 50 17.7b Everlase Bacillus sp. Alkaline endopeptidase 9.5 50 8.1f Neutrase Bacillus amyloliquefaciens Neutral endopeptidase 7 50 6.7g FoodPro PHT Geobacillus sp. Thermostable endopeptidase 8 65 5.9g Alcalase and Neutrase (A+N) 7 50 11.4e Alcalase and Flavourzyme (A+F) 7 50 23.2a Alcalase and Everlase (A+E) 8 50 13.0d Flavourzyme and Neutrase (F+N) 7 50 15.3c Everlase and FoodPro PHT (E+P) 8 65 6.5g Means with different superscript letters differ significantly at (p < 0.05). R. Abeynayake et al. LWT 158 (2022) 113162 3 Fig. 1. Preparation of BSG antioxidant peptides. R. Abeynayake et al. LWT 158 (2022) 113162 4 0.1 mol/L HCl was added to terminate the reaction. Solutions were cooled to room temperature for 30 min, and the absorbance was measured at 340 nm. A standard curve was constructed using L-leucine ranging from 0 to 4 mmol/L. The DH values were calculated using the following equation: %DH = h h total × 100 where h is the number of peptide bonds broken during hydrolysis (mmol/L/g of protein) and h total is the total amount of peptide bonds in BSG protein (7.75 mmol/L/g of protein) (Celus et al., 2006). 2.4. Solubility Solubility was determined by the method described by Barbin et al. (Barbin, Natsch, & MÜLler, 2011). Aqueous samples of 1 g/100 mL were stirred for 1 h after adjusting to pH 7.0. Then, the resulting solutions were centrifuged for 30 min at 4000×g and 30 ◦C. Precipitates obtained after removing supernatants were freeze-dried until reaching to a con- stant weight. Protein content was determined by nitrogen analyzer using nitrogen conversion factor 6.25. Solubility was calculated using the following equation: %Solubility= ( 1 − Weight of precipitate × Protein content Weight of protein in initial sample ) × 100 2.5. Size exclusion high-performance liquid chromatography (SE-HPLC) Samples were analyzed for average molecular weight (Mw) by SE- HPLC using an Agilent 1100 series HPLC system coupled with a Bio- suiteTM 125/5 mm HR-SEC column (7.8 300 mm, Waters Corp., Mass., USA). The mobile phase, phosphate buffer (0.1 mol/L) at pH 7.0 con- taining 0.1 mol/L NaCl flowed at a rate of 0.5 mL/min at 25 ◦C. Twenty microliters of samples were injected into the HPLC system and absor- bance was monitored at a UV wavelength of 220 nm. Protein Standard Mix 15–600 kDa (#69385, Sigma-Aldrich) with the following range of molecular weights: thyroglobulin (670 kDa), γ-globulins (150 kDa), al- bumin (44.3 kDa), ribonuclease A (13.7 kDa), and p-aminobenzoic acid (137 Da), which were also analyzed using the same procedure. Log Mw of the markers was plotted against their respective elution times. 2.6. Amino acid composition For amino acid analysis, samples were dissolved in deionized water to a concentration of 10 mg/mL. Thirty microliters of each sample were dried and hydrolyzed under vacuum in 6 mol/L HCl for 24 h at 110 ◦C. Hydrolyzed samples were again dried and then dissolved in 20 mmol/L HCl. Amino acid analysis was done using Waters AccQ-Tag method. High-performance liquid chromatography system (Agilent series 1200, Palo Alto, CA, USA) equipped with AccQ-Tag 3.9 × 150 mm C18 column was used in separating the derivatives. Samples were pumped at a flow rate of 1.5 mL/min using gradient solvent system (AccQ-Tag eluent, acetonitrile, and water), and finally detected at 254 nm wavelength. Data processing was controlled by ChemStation software. 2.7. Antioxidant activity 2.7.1. DPPH free radical scavenging activity Free radical scavenging activity of samples was evaluated by DPPH assay according to the method described by Blois (Blois, 1958). One milliliter of samples (1.0 mg/mL) was mixed with 1 mL of 0.1 mmol/L DPPH• in anhydrous ethanol. Mixture was incubated at room temper- ature for 30 min. BHT and ascorbic acid at concentrations of 0.1 and 0.01 mg/mL were used as positive controls. Reduction of DPPH free radicals was determined by measuring the absorbance at 517 nm. The ability of the samples to scavenge DPPH free radicals was calculated according to the following equation: %DPPH radical scavenging activity= A0 − A A0 × 100 Where A0 and A represent the absorbance of the control and hydrolysate sample, respectively. 2.7.2. Superoxide radical scavenging activity Superoxide radical scavenging activity was measured by monitoring the inhibition of pyrogallol autoxidation and polymerization (Marklund & Marklund, 1974). Briefly, 80 μL of 2.0 mg/mL sample was mixed with 80 μL of 0.05 mol/L Tris–HCl buffer (pH 8.3) containing 1 mM EDTA, followed by addition of 40 μL of 1.5 mmol/L pyrogallol in 1.0 mmol/L HCl. Absorbance was measured at 320 nm for 5 min at room tempera- ture. Ascorbic acid at 0.01 and 0.1 mg/mL was used as the positive control. Superoxide radical scavenging activity was calculated using the following equation: %Superoxide radical scavenging activity= (ΔA0/min) − (ΔAs/min) ΔA0/min × 100 Where A0 and A represent absorbance of the blank and the sample, respectively. 2.7.3. Ferrous ion chelating activity Ferrous ion chelating activity was measured according to the method described by (Kong & Xiong, 2006). Briefly, 1 mL of 20 μmol/L FeCl2 was mixed with 0.5 mL of 1.0 mg/mL sample, and then 1 mL of 0.5 mmol/L ferrozine was added to initiate the reaction. Mixture was incubated at room temperature for 15 min before measuring absorbance at 562 nm. The EDTA at concentrations of 0.1 and 0.01 mg/mL were used as positive controls. Ferrous ion chelating ability was calculated by the following equation: %Ferrous ion chelating ability= B0 − B B0 × 100 Where B0 and B represent the absorbance of the sample and the control, respectively. 2.7.4. Reducing power Reducing power was measured according to the method of Oyaizu (Oyaizu, 1986). In short, 1 mL of 1 g/100 mL sample was mixed with 1 mL of 1% (w/v) potassium ferricyanide and 1 mL of 0.2 mol/L phos- phate buffer (pH 6.6). The mixture was incubated at 50 ◦C for 20 min. Reaction was stopped by adding 1 mL of 10 g/100 mL TCA. After centrifugation at 1000×g, 10 min at 20 ◦C, 1 mL of supernatant was collected. It was mixed with 1 mL of distilled water and 0.2 mL of 100 mg/100 mL FeCl3. After 10 min reaction, absorbance of the resulting solution was measured at 700 nm. BHT and ascorbic acid at concen- trations of 0.1 and 0.01 mg/mL were used as positive controls. Results were expressed as absorbance at 700 nm. An increased absorbance of the reaction mixture indicated increased reducing power. 2.8. Statistical analysis All experiments were performed at least in three independent batches. Data were presented as the mean ± standard deviation (SD). Comparison of sample means was made through one-way analysis of variance (ANOVA) followed by Tukey’s pos-hoc test at a significance level of 0.05. Statistical evaluation was conducted by Student’s t-test and analysis of variance (ANOVA). The correlations between peptide structures (molecular weight, hydrolysis degree, amino acid composi- tion, solubility) and antioxidant activities (reducing power, ferrous ion chelating activity, superoxide and DPPH radical scavenging activities) were studied by Pearson analysis, and the correlation coefficient r values were presented. Statistical analyses were performed using SAS 9.0 for R. Abeynayake et al. LWT 158 (2022) 113162 5 Windows (SAS Institute Inc., Cary, NC). 3. Results and discussion 3.1. Extraction and hydrolysis of BSG protein The BSG had higher protein content compared to that of barley grain. The increase of protein content from 10 to 20 to 24 g/100 g on a dry basis is due to removal of sugars yielded from hydrolysis of carbohy- drates, mainly starch stored in endosperm (Cermeno et al., 2021; Houde et al., 2018; Kanauchi, Mitsuyama, & Araki, 2001; Mussatto & Roberto, 2005). The protein extract of BSG contained 62.6 g/100 g of protein and the hydrolysates contain 65.2–69.6 g/100 g protein on a dry basis. The protein recovery from BSG was 46.3%, which was lower than that of soy and pea protein (more than 80%). It is always a challenge to purify proteins from agricultural processing by-products due to structural and chemical modifications in processing. Some portion of insoluble proteins in barley undergo modifications during brewing including aggregation and binding to other matrix components. This made it harder to purify them without intense chemical treatments. However, such treatments may result in uncontrolled hydrolysis of proteins, leading to losing their bioactivity. Thus, in this study mild alkaline treatment was applied in extracting proteins from BSG. This could possibly explain the low pro- tein extraction efficacy. Antioxidant activities of peptides are known to have significant variations upon proteolytic hydrolysis depending on the enzyme speci- ficity and hydrolysis conditions applied. Therefore, BSG bioactive pep- tides were prepared by hydrolyzing with selected proteases or protease combinations having different specificities. Table 1 shows the hydrolysis conditions for the selected enzymes. Combined enzyme treatments were developed by combining two proteases with similar optimum pH and temperature ranges. According to our preliminary studies, low concen- trations of some proteases were not very effective in hydrolyzing pro- teins remained in BSG. This could be due to the resistant nature of proteins remaining after brewing or formation of complex protein structures during brewing. Thus, a relatively higher amount of enzymes were applied in this study. As shown in Table 1, DH ranging from 5.9% to 24.1% was observed at the end of 4 h hydrolysis. Moreover, hydro- lysis beyond 4 h did not increase DH% significantly. Flavourzyme, the fungal protease was the most effective with a DH value of 24.07%. This is explained by endo and exopeptidase activity of Flavorzyme, while all the other proteases used were endo peptidases. The Alcalase and Fla- vourzyme (A+F) combined treatment resulted in 23.2% of DH. The DH values of other combined protease treatments were fallen in between the values of individual enzymes. In general, hydrolysates with higher DH can be expected to have lower molecular weight. Not only the peptides of different molecular weights but also the specificity of proteases might have resulted in bioactive peptides containing different amino acid sequences. The antioxidant activity of BSG peptides could be closely related to their structural and chemical properties. Therefore, the antioxidant peptides resulting from protease hydrolysis were analyzed for solubility, molecular weight distribution, and amino acid composition. 3.2. Solubility In general, poor water solubility at neutral pH is one of the factors that limit the applications of most plant proteins. Fig. 2a compares the solubility of BSG protein and bioactive peptides in water at neutral pH. It is noticed that enzyme treatments applied in this work greatly increased BSG protein solubility up to 94.4%. Such high solubility will allow BSG bioactive peptides to be used in various food formulations for value- added applications. Water solubility of barley proteins extracted with NaOH was found to be around 20% at pH 7.0 (Bilgi & Çelik, 2004). This low solubility could result from the high hydrophobic amino acid con- tent of barley hordein and glutelin such as phenylalanine, valine, leucine, isoleucine, and methionine. Elevated solubility of BSG protein (36.2%) compared to that of barley protein (20%) (Bilgi & Çelik, 2004) can be attributed to hydrolysis by endogenous proteases during brewing. Especially, hordein and glutelin are partially decomposed to peptides by endoproteases, (Osman et al., 2002). Enzyme hydrolysis has been widely applied to increase the solubility and functional properties of proteins. 3.3. Molecular weight distribution The SE-HPLC chromatograms of BSG protein and bioactive peptides are shown in Fig. 2b. The limited solubility of BSG protein at pH 7.0 made it difficult to obtain molecular weight distribution of all constit- uent protein fractions. Chromatogram of soluble BSG protein consisted of a broad peak with an average molecular weight of 23.9 kDa. Inter- estingly, peaks associated with protein aggregates were observed at 1061–1792 kDa. Although the mechanism is not known, unfolded pro- teins can self-assemble to form soluble or insoluble protein aggregates during brewing. All enzyme treatments were capable of producing peptides of 23.9 kDa, and smaller peptides with an average molecular weight of 8.0 kDa. Aggregates were also observed in protease hydroly- sates. Peptides resulting from hydrolysis might have combined with existing aggregates to form larger and more complex structures. It seems that peptides prepared under acidic or neutral pH (Flavourzyme, Neu- trase, and F+N) contained a higher amount of protein aggregates compared to those treated under alkaline conditions. Previous study also observed the increased soluble protein aggregate formation between pH 6.6–6.8 for whey proteins (Hollar, Parris, Hsieh, & Cockley, 1995). 3.4. Amino acid composition Amino acid profiles of proteins and bioactive peptides were identi- fied to play a significant role in several antioxidant mechanisms. Amino acid composition expressed as mM/100g of protein is shown in Table 2. Hydrophobic amino acids were observed to be potent antioxidants, probably because of their ability to stabilize free radicals by donating protons (Chi, Hu, Wang, Li, & Luo, 2015; Nourmohammadi, Sadeghi- Mahoonak, Alami, & Ghorbani, 2017; Pownall, Udenigwe, & Aluko, 2010). Also, due to its non-polar nature, hydrophobic amino acids can be expected to have a great affinity towards fatty acid radicals formed in lipid oxidation. As shown in Fig. 2c, hydrophobic amino acid content increased from 50.4 to 164.5 mmol/100g upon protease hydrolysis. Peptides from A+E treatment contained the highest hydrophobic amino acid content. Be- sides the hydrophobic nature, Pownall et al. (Pownall et al., 2010) observed a higher amount of aromatic amino acids: tyrosine and phenylalanine in protein fractions with high radical scavenging and metal chelating properties. Donation of protons to electron-deficient species while maintaining stability via resonance structures gives aro- matic amino acids the ability to scavenge free radicals. Peptides ob- tained from A+E treatment had the highest aromatic amino acid content (51.5 mmol/100g). Histidine was observed to involve in several anti- oxidant mechanisms. The comparatively high proton donor activity of histidine is believed to be due to presence of imidazole ring in the side chain (Mendis et al., 2005). All the treatments showed significantly (p < 0.05) higher histidine content compared to that of BSG protein (2.4 mmol/100g). The highest histidine content (11.4 mmol/100g) was observed in F+N treatment followed by A+E combination (11.1 mmol/100g). Although, barley contains all the essential amino acids (Biel & Jacyno, 2013), BSG protein was found to lack in cysteine. Apart from that, Flavourzyme hydrolysate did not contain methionine. 3.5. Antioxidant activity Different assays were performed since antioxidants were found to take different pathways in reacting with or neutralizing reactive species. Commercial antioxidants were tested as positive controls to see the R. Abeynayake et al. LWT 158 (2022) 113162 6 Fig. 2. Solubility (a), molecular weight distribution (b), and amino acid contents (c) of BSG protein and hydrolysates. Means with different letters differ significantly at (p < 0.05). R. Abeynayake et al. LWT 158 (2022) 113162 7 potentiality of BSG peptides to be used in commercial applications. The EDTA which is a strong metal chelator was used as a positive control in ferrous ion chelating ability assay. Ascorbic acid, a natural antioxidant known to play a pivotal role to suppress reactive species developed in biological systems including superoxide radical, was compared with the superoxide radical scavenging activity of test samples. BHT and ascorbic acid were used representing a synthetic and natural antioxidant respectively, as the positive controls in DPPH radical scavenging and reducing power assays. 3.5.1. DPPH radical scavenging activity The DPPH assay is commonly used to analyze the ability of a com- pound to act as an antioxidant by donating protons in a non-aqueous medium. Thus, it is an indicator of antioxidant activity in non-polar environments including fatty tissues and lipid-containing foods. As shown in Fig. 3a, scavenging activity was 66.2% for BSG protein at the concentration of 1 mg/mL. Protease hydrolysis significantly (p < 0.05) increased DPPH radical scavenging activity of BSG protein due to release of bioactive peptides responsible for antioxidant effect. The peptides that reported the highest DPPH scavenging activity were produced by Alcalase or its combination with other enzymes (A+N, A+F, A+E). It could be due to the specific activity of Alcalase to give rise to amino acid sequence with greater DPPH scavenging potential. Previous studies also reported relatively higher DPPH radical scavenging activity of Alcalase peptides (Chi, Hu, et al., 2015; Nourmohammadi et al., 2017). Possibly release of hydrophobic amino acids by Alcalase treatment could exert good proton donation capacity in a hydrophobic environment. The released aromatic amino acids could donate protons to electron-deficient species like DPPH, maintaining stability via resonance structures. The significantly increased histidine could also contribute to Table 2 Amino acid composition of BSG protein and hydrolysates given as mmol/100 g. Residue BSG Flaourzyme Everlase Neutrase Alcalase FoodPro PHT A+N A+F F+N E+P E+A Asx 15.2 29.9 21.2 18.3 39.4 26.8 26.9 28.6 39.7 41.2 43.8 Ser 11.2 23.1 16.3 14.2 30.5 20.8 20.6 22.9 31.4 32.0 34.8 Glx 41.8 86.4 61.4 51.9 116.8 78.4 78.3 84.1 117.4 120.4 134.6 Gly 12.0 27.9 18.0 17.7 35.6 23.1 24.2 23.6 35.8 36.3 40.9 His 2.4 7.5 4.2 6.2 9.2 5.4 6.9 6.4 11.4 9.1 11.1 Arg 6.3 13.5 5.7 6.1 11.8 6.6 6.9 8.3 12.4 11.9 13.5 Thr 8.8 16.8 12.3 11.1 21.4 13.3 14.3 15.8 22.3 22.6 25.3 Ala 14.4 28.2 19.3 17.2 37.0 23.5 24.2 26.5 37.3 38.5 42.2 Pro 25.3 52.9 36.3 31.9 70.9 46.6 47.2 50.5 71.0 75.0 82.5 Cys 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Tyr 5.9 12.5 8.9 7.8 16.3 10.7 10.6 12.5 18.2 17.0 18.9 Val 12.5 26.8 17.9 15.4 34.6 23.5 22.6 25.8 36.1 36.2 39.2 Met 3.1 0.0 4.9 4.1 9.3 5.8 6.1 7.2 9.6 9.9 10.7 Lys 6.4 12.8 8.4 7.5 15.8 10.6 10.8 11.7 15.9 16.0 18.2 Ile 8.3 19.3 12.7 10.9 23.3 15.9 14.4 18.5 25.0 24.3 27.4 Leu 17.0 35.2 24.9 21.7 45.2 30.7 29.1 32.5 46.6 47.7 54.9 Phe 9.5 20.4 14.3 12.3 27.6 17.3 18.2 19.6 28.1 28.0 32.3 Asx represents Asn and Asp; Glx represents Gln and Glu. Fig. 3. Antioxidant activities of BSG protein and hydrolysates. (a) DPPH radical scavenging activity, (b) superoxide radical scavenging activity, (c) ferrous ion chelating activity, and (d) reducing power of BSG protein and hydrolysates. Means with different letters differ significantly at (p < 0.05). R. Abeynayake et al. LWT 158 (2022) 113162 8 the DPPH radical scavenging activity. As suggested by Mendis et al., (Mendis et al., 2005), the greater DPPH radical scavenging activity of histidine could be due to the high proton donation ability of the imid- azole ring present in the side chain. 3.5.2. Superoxide radical scavenging activity The superoxide anion radical produced in aerobic respiration is a signaling molecule, and also essential in regulating apoptosis, and aging. However, over-production or malfunctioning of antioxidant defense mechanisms may cause oxidative stress induced by superoxide radicals. As shown in Fig. 3b, protease hydrolysis increased superoxide radical scavenging activity of BSG protein from 7.8% to 19.3%. The highest activity was reported by bioactive peptides prepared with A+F treat- ment. Pownall et al. (Pownall et al., 2010) related higher concentrations of proline and hydrophobic amino acids in pea peptide fractions to the highest superoxide radical scavenging activity. Another study conducted in our lab proposed that higher superoxide radical scavenging activity of hordein peptides could be due to high histidine, proline, tyrosine, and tryptophan content in hordein (Bamdad et al., 2011). However, the present study found no relationship between amino acid composition and superoxide radical scavenging activity of BSG protein. Moreover, there was no correlation observed between molecular weights and antioxidant activities. This could be due to aggregate formation of BSG hydrolysates as observed by SE-HPLC chromatograms. Instead, bioac- tive peptides samples with high DH showed higher superoxide radical scavenging activity in general. The protein hydrolysates of skipjack tuna prepared with five proteases also showed a correlation between DH, and superoxide radical scavenging capacity (Chi, Hu, et al., 2015). It can be suggested that with the increasing DH, there were more positive charges available on the amine groups of peptides which could form electrostatic interactions with negatively charged superoxide radicals for their sta- bilization. The lower activity of Flavourzyme, F+N, and Neutrase pep- tides could be due to formation of a larger amount of protein aggregates, making positive charges less available to form interactions with super- oxide radicals. 3.5.3. Ferrous ion chelating activity Iron is essential in regulating normal body functions however, in excess it shows pro-oxidant activity to catalyze various oxidation re- actions in the body, and also takes part in Fenton reaction. Hydrogen peroxide produced as a by-product in aerobic respiration can be con- verted to hydroxyl radicals via the Fenton reaction when the ferrous ion is available. Removal of excess ferrous ions by metal chelators has the potential to cease reactive species generation. In this study, ferrous chelating activity was measured by the ability of samples to inhibit formation of ferrous-ferrozine complex. As shown in Fig. 3c, protease hydrolysis was effective in increasing ferrous ion chelating activity of BSG protein. Upon protease hydrolysis, ferrous ion chelating activity was approximately increased by two folds. The highest activity, 35.1% was reported for E+P treatment. Other authors also observed the favorable effects of protease hydrolysis on metal chelation (Eckert, Bamdad, & Chen, 2014). Although several studies (Nourmohammadi et al., 2017; Pownall et al., 2010) reported that high hydrophobic and aromatic amino acids in peptides showed greater ferrous ion chelating activity, there was no such relationship observed in this study. Instead, a strong and negative correlation (R = - 0.84) existed between the DH and ferrous ion chelating activity of bioactive peptides. Thus, peptides with long chains showed greater ferrous ion chelating activity. Metal chelating activity of proteins and peptides could be via structures that can trap metal ions (Zhang et al., 2009). Long peptide chains might have a comparatively higher ability to form such structures than shorter peptides and amino acids. Although BSG protein must have contained the longest peptide chains compared to bioactive peptides, its poor ferrous ion chelating activity can be explained by its low solubility. 3.5.4. Reducing power Among several pathways, donation of protons to neutralize reactive species is one of the most potent antioxidant mechanisms. Reducing power assay quantifies the ability of a compound to reduce ferric ion to ferrous ion by donating protons. As given in Fig. 3d, bioactive peptides showed significantly (p < 0.05) higher reducing power compared to that of BSG protein (0.3). The reducing power of the peptides ranged be- tween 0.34 and 0.44. FoodPro PHT was the most effective treatment in producing peptides with reducing power, followed by Neutrase and E+P treatments. Previously, several attempts have been made to identify the factors that change the reducing power upon protease hydrolysis. The increase of reducing power can be attributed to exposure of electron- dense amino acid side chains moieties as a result of hydrolysis (Bam- dad et al., 2011). Free amino acids also can be expected to elevate reducing power as they are a source of protons and electrons to maintain high redox potential (Zhu, Chen, Tang, & Xiong, 2008). Another study suggested that the strong reducing power observed in loach protein peptides could be due to amino acids: tyrosine, methionine, histidine, lysine, and tryptophan (You, Zhao, Cui, Zhao, & Yang, 2009). However, in this study, there was no strong correlation observed between reducing power and amino acid composition or structural properties of the sam- ples. Possibly multiple factors contributed to the increased reducing power of BSG after hydrolysis. 3.6. Comparison with commercial antioxidants Incorporation of natural as well as synthetic antioxidants into lipid- containing foods and nutraceuticals is widely practiced to retard oxidation of biological molecules and associated negative changes. However, safety concerns, availability, and low cost take natural anti- oxidants such as proteins and peptides to the platform. To assess the applicability on an industrial scale, BSG bioactive peptides with the highest activity in each antioxidant assay were compared with selected positive controls, which are known commercial antioxidants. The highest DPPH radical scavenging activity reported by A+N treatment (74.9%) at 1.0 mg/mL was comparable to that of BHT (42.4%) and ascorbic acid (76.2%) at 0.1 mg/mL. The A+F bioactive peptides at 2.0 mg/mL showed the best capacity to scavenge superoxide radicals (19.3%). It was comparable to ascorbic acid (11.4%) at 0.01 mg/mL, a widely used natural antioxidant in food systems, and also a key contributor to antioxidant defense mechanisms in biological systems. The EDTA, a strong metal chelator was used as the positive control in ferrous ion chelating ability assay. None of the peptides were effective in chelating ferrous ions compared to EDTA. In this study, all samples at 1.0 mg/mL showed significantly (p < 0.05) higher reducing power compared to that of ascorbic acid at 0.01 mg/mL (0.16). Even, the reducing power of FoodPro PHT peptides, 0.44 was closer to the reducing power of 0.1 mg/mL BHT. Although the antioxidant activities of peptides were comparable to lower concentrations of positive con- trols, they have the potential to be used in food, feed, and nutraceutical applications as they can be incorporated at high concentrations. On the other hand, the same bioactive peptides can be used to give functional properties and enhance the nutrition profile of food systems. 4. Conclusions The high fiber content in BSG leads to reduced protein digestibility and functional properties. Also, BSG is regarded as a less attractive food ingredient considering sensory and functional attributes. This manu- script demonstrates the potentiality of BSG as a low-cost protein source to develop antioxidant peptides for food applications. In addition, the remaining fiber-rich residue was tested for feed applications in another lab. The results show substantially increased neutral detergent fiber content and improved in vitro neutral detergent fiber digestibility due to protease hydrolysis (Ran, Jin, Abeynayake, Saleem, Zhang et al., 2021). This study shows a recovery of 46.3% proteins from BSG using a R. Abeynayake et al. LWT 158 (2022) 113162 9 non-toxic, simple, and less expensive alkaline treatment. The extracted protein has an economic potential to be used as a functional ingredient in food preparations and also as a protein supplement in food and feed formulas. Moreover, protease hydrolysis is an effective way to generate peptides with antioxidant activities from BSG protein. Hydrolysates prepared by Alcalase, and its combination with Neutrase and Fla- vourzyme were more effective in scavenging DPPH and superoxide radical. FoodPro PHT and its combination with Everlase were the most effective in producing hydrolysates with high ferrous ion chelating ca- pacity and reducing power. Although the antioxidant activities of peptides are comparable to lower concentrations of positive controls and the use of proteases adds extra cost to the process, they have a good potential to be used in food, feed, and nutraceutical applications as they can be incorporated at high concentrations. Also, the consumer preference for natural food additives over synthetic ones will further strengthen the applicability of BSG protein hydrolysates as antioxidants. Further in vivo studies are required to understand the actual efficacy of BSG peptides as antioxidants in biological systems. CRediT authorship contribution statement Galhenage Thanusha Ranithri Abeynayake: Conceptualization, Methodology, Validation, Formal analysis, Writing – original draft. Sitian Zhang: Formal analysis, Writing – review & editing. Wenzhu Yang: Resources, Writing – review & editing, Supervision, Project administration. Lingyun Chen: Conceptualization, Resources, Writing – review & editing, Supervision, Project administration, Funding acquisition. Declaration of competing interest The authors have declared no conflicts of interest for this article. Acknowledgments The authors are grateful to the Natural Sciences and Engineering Research Council of Canada (NSERC) and the Agriculture and Agri-Food Canada (AAFC) Growing forward program (GF2#1542) for financial support. Dr. Lingyun Chen would like to thank the Natural Sciences and Engineering Research Council of Canada (NSERC)-Canada Research Chairs Program for its financial support. References Adler-Nissen, J. (1979). Determination of the degree of hydrolysis of food protein hydrolysates by trinitrobenzenesulfonic acid. Journal of Agricultural and Food Chemistry, 27(6), 1256–1262. Bamdad, F., Shin, S. H., Suh, J.-W., Nimalaratne, C., & Sunwoo, H. (2017). Anti- inflammatory and antioxidant properties of casein hydrolysate produced using high hydrostatic pressure combined with proteolytic enzymes. Molecules, 22(4), 609. Bamdad, F., Wu, J., & Chen, L. (2011). Effects of enzymatic hydrolysis on molecular structure and antioxidant activity of barley hordein. Journal of Cereal Science, 54(1), 20–28. Barbin, D., Natsch, A., & MÜLler, K. (2011). Improvement of functional properties of rapeseed protein concentrates produced via alcoholic processes by thermal and mechanical treatments. Journal of Food Processing and Preservation, 35(3), 369–375. Biel, W., & Jacyno, E. (2013). Chemical composition and nutritive value of spring hulled barley varieties. Bulgarian Journal of Agricultural Science, 19(4), 721–727. Bilgi, B., & Çelik, S. (2004). Solubility and emulsifying properties of barley protein concentrate. European Food Research and Technology, 218(5), 437–441. Blois, M. S. (1958). Antioxidant determinations by the use of a stable free radical. Nature, 181(4617), 1199–1200. Carocho, M., Morales, P., & Ferreira, I. C. (2018). Antioxidants: Reviewing the chemistry, food applications, legislation and role as preservatives. Trends in Food Science & Technology, 71, 107–120. Celus, I., Brijs, K., & Delcour, J. A. (2006). The effects of malting and mashing on barley protein extractability. Journal of Cereal Science, 44(2), 203–211. Cermeno, P. M., Dermiki, M., Kleekayai, T., Cope, L., McManus, R., Ryan, C., et al. (2021). Effect of enzymatically hydrolysed brewers’ spent grain supplementation on the rheological, textural and sensory properties of muffins. Future Foods, 4, 100085. Chi, C.-F., Hu, F.-Y., Wang, B., Li, Z.-R., & Luo, H.-Y. (2015). Influence of amino acid compositions and peptide profiles on antioxidant capacities of two protein hydrolysates from skipjack tuna (Katsuwonus pelamis) dark muscle. Marine Drugs, 13 (5), 2580–2601. Eckert, E., Bamdad, F., & Chen, L. (2014). Metal solubility enhancing peptides derived from barley protein. Food Chemistry, 150, 498–506. Hollar, C., Parris, N., Hsieh, A., & Cockley, K. (1995). Factors affecting the denaturation and aggregation of whey proteins in heated whey protein concentrate mixtures. Journal of Dairy Science, 78(2), 260–267. Houde, M., Khodaei, N., Benkerroum, N., & Karboune, S. (2018). Barley protein concentrates: Extraction, structural and functional properties. Food Chemistry, 254, 367–376. Kanauchi, O., Mitsuyama, K., & Araki, Y. (2001). Development of a functional germinated barley foodstuff from brewer’s spent grain for the treatment of ulcerative colitis. Journal of the American Society of Brewing Chemists, 59(2), 59–62. Kong, B., & Xiong, Y. L. (2006). Antioxidant activity of zein hydrolysates in a liposome system and the possible mode of action. Journal of Agricultural and Food Chemistry, 54 (16), 6059–6068. Marklund, S., & Marklund, G. (1974). Involvement of the superoxide anion radical in the autoxidation of pyrogallol and a convenient assay for superoxide dismutase. European Journal of Biochemistry, 47(3), 469–474. Mendis, E., Rajapakse, N., Byun, H.-G., & Kim, S.-K. (2005). Investigation of jumbo squid (Dosidicus gigas) skin gelatin peptides for their in vitro antioxidant effects. Life Sciences, 77(17), 2166–2178. Moure, A., Domínguez, H., & Parajó, J. C. (2006). Antioxidant properties of ultrafiltration-recovered soy protein fractions from industrial effluents and their hydrolysates. Process Biochemistry, 41(2), 447–456. Mussatto, S. I., & Roberto, I. C. (2005). Acid hydrolysis and fermentation of brewer’s spent grain to produce xylitol. Journal of the Science of Food and Agriculture, 85(14), 2453–2460. Nourmohammadi, E., SadeghiMahoonak, A., Alami, M., & Ghorbani, M. (2017). Amino acid composition and antioxidative properties of hydrolysed pumpkin (Cucurbita pepo L.) oil cake protein. International Journal of Food Properties, 20(12), 3244–3255. Osborne, T. B. (1924). The vegetable proteins: Longmans. Green and Company. Osman, A., Coverdale, S., Cole, N., Hamilton, S., De Jersey, J., & Inkerman, P. (2002). Characterisation and assessment of the role of barley malt endoproteases during malting and mashing 1. Journal of the Institute of Brewing, 108(1), 62–67. Oyaizu, M. (1986). Studies on products of browning reaction antioxidative activities of products of browning reaction prepared from glucosamine. The Japanese Journal of Nutrition and Dietetics, 44(6), 307–315. Patrignani, M., Brantsen, J. F., Awika, J. M., & Conforti, P. A. (2021). Application of a novel microwave energy treatment on brewers’ spent grain (BSG): Effect on its functionality and chemical characteristics. Food Chemistry, 346, 128935. Pownall, T. L., Udenigwe, C. C., & Aluko, R. E. (2010). Amino acid composition and antioxidant properties of pea seed (Pisum sativum L.) enzymatic protein hydrolysate fractions. Journal of Agricultural and Food Chemistry, 58(8), 4712–4718. Ran, T., Jin, L., Abeynayake, R., Saleem, A. M., Zhang, X., Niu, D., et al. (2021). Effects of brewers’ spent grain protein hydrolysates on gas production, ruminal fermentation characteristics, microbial protein synthesis and microbial community in an artificial rumen fed a high grain diet. Journal of Animal Science and Biotechnology, 12(1). https://doi.org/10.1186/s40104-020-00531-5 Sun, L., Zhang, Y., & Zhuang, Y. (2013). Antiphotoaging effect and purification of an antioxidant peptide from tilapia (Oreochromis niloticus) gelatin peptides. Journal of Functional Foods, 5(1), 154–162. Wong, F.-C., Xiao, J., Wang, S., Ee, K.-Y., & Chai, T.-T. (2020). Advances on the antioxidant peptides from edible plant sources. Trends in Food Science & Technology, 99, 44–57. Yang, X.-R., Zhang, L., Ding, D.-G., Chi, C.-F., Wang, B., & Huo, J.-C. (2019). Preparation, identification, and activity evaluation of eight antioxidant peptides from protein hydrolysate of hairtail (Trichiurus japonicas) muscle. Marine Drugs, 17(1), 23. You, L., Zhao, M., Cui, C., Zhao, H., & Yang, B. (2009). Effect of degree of hydrolysis on the antioxidant activity of loach (Misgurnus anguillicaudatus) protein hydrolysates. Innovative Food Science & Emerging Technologies, 10(2), 235–240. Zhang, J., Zhang, H., Wang, L., Guo, X., Wang, X., & Yao, H. (2009). Antioxidant activities of the rice endosperm protein hydrolysate: Identification of the active peptide. European Food Research and Technology, 229(4), 709–719. Zhong, Y., & Shahidi, F. (2015). Methods for the assessment of antioxidant activity in foods Handbook of antioxidants for food preservation. Elsevier. Zhu, L., Chen, J., Tang, X., & Xiong, Y. L. (2008). Reducing, radical scavenging, and chelation properties of in vitro digests of Alcalase-treated zein hydrolysate. Journal of Agricultural and Food Chemistry, 56(8), 2714–2721. R. Abeynayake et al. http://refhub.elsevier.com/S0023-6438(22)00097-4/sref1 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref1 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref1 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref4 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref4 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref4 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref5 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref5 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref5 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref6 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref6 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref6 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref7 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref7 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref8 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref8 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref9 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref9 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref10 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref10 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref10 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref11 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref11 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref12 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref12 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref12 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref13 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref13 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref13 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref13 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref15 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref15 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref18 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref18 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref18 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref19 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref19 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref19 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref21 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref21 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref21 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref22 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref22 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref22 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref24 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref24 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref24 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref26 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref26 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref26 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref27 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref27 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref27 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref28 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref28 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref28 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref30 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref30 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref30 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref32 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref33 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref33 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref33 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref34 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref34 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref34 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref36 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref36 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref36 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref38 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref38 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref38 https://doi.org/10.1186/s40104-020-00531-5 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref42 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref42 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref42 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref45 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref45 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref45 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref47 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref47 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref47 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref48 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref48 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref48 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref49 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref49 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref49 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref50 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref50 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref51 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref51 http://refhub.elsevier.com/S0023-6438(22)00097-4/sref51 Development of antioxidant peptides from brewers’ spent grain proteins 1 Introduction 2 Materials and methods 2.1 Materials 2.2 Preparation of BSG bioactive peptides 2.3 Degree of hydrolysis (DH) 2.4 Solubility 2.5 Size exclusion high-performance liquid chromatography (SE-HPLC) 2.6 Amino acid composition 2.7 Antioxidant activity 2.7.1 DPPH free radical scavenging activity 2.7.2 Superoxide radical scavenging activity 2.7.3 Ferrous ion chelating activity 2.7.4 Reducing power 2.8 Statistical analysis 3 Results and discussion 3.1 Extraction and hydrolysis of BSG protein 3.2 Solubility 3.3 Molecular weight distribution 3.4 Amino acid composition 3.5 Antioxidant activity 3.5.1 DPPH radical scavenging activity 3.5.2 Superoxide radical scavenging activity 3.5.3 Ferrous ion chelating activity 3.5.4 Reducing power 3.6 Comparison with commercial antioxidants 4 Conclusions CRediT authorship contribution statement Declaration of competing interest Acknowledgments References