Antioxidant properties and inhibition of angiotensin-converting enzyme by highly active peptides from wheat gluten

This study aimed to focus on the high-value utilization of raw wheat gluten by determining the potent antioxidant peptides and angiotensin I-converting enzyme (ACE) inhibitory peptides from wheat gluten oligopeptides (WOP). WOP were analyzed for in vitro antioxidant activity and inhibition of ACE, and the identification of active peptides was performed by reversed-phase high-performance liquid chromatography and mass spectrometry. Quantitative analysis was performed for highly active peptides. Five potent antioxidant peptides, Leu-Tyr, Pro-Tyr, Tyr-Gln, Ala-Pro-Ser-Tyr and Arg-Gly-Gly-Tyr (6.07 ± 0.38, 7.28 ± 0.29, 11.18 ± 1.02, 5.93 ± 0.20 and 9.04 ± 0.47 mmol 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox) equivalent/g sample, respectively), and five potent ACE inhibitory peptides, Leu-Tyr, Leu-Val-Ser, Tyr-Gln, Ala-Pro-Ser-Tyr and Arg-Gly-Gly-Tyr (half maximal inhibitory concentration (IC50) values = 0.31 ± 0.02, 0.60 ± 0.03, 2.00 ± 0.13, 1.47 ± 0.08 and 1.48 ± 0.11 mmol/L, respectively), were observed. The contents of Leu-Tyr, Pro-Tyr, Tyr-Gln, Ala-Pro-Ser-Tyr, Arg-Gly-Gly-Tyr, and Leu-Val-Ser were 155.04 ± 8.36, 2.08 ± 0.12, 1.95 ± 0.06, 22.70 ± 1.35, 0.25 ± 0.01, and 53.01 ± 2.73 μg/g, respectively, in the WOP. Pro-Tyr, Tyr-Gln, Ala-Pro-Ser-Tyr, Arg-Gly-Gly-Tyr, and Leu-Val-Ser are novel antioxidative/ACE inhibitory peptides that have not been previously reported. The results suggest that WOP could potentially be applied in the food industry as a functional additive.


Results and discussion
Chemical composition and amino acid composition of the WOP. The WOP sample contained the following components (dry basis): proteins (peptides), 94.52 ± 0.56%; lipids, 0.05 ± 0.01%; ash, 2.32 ± 0.25%; and moisture, 2.35 ± 0.23%. As expected, peptides were the main component of WOP. The dissolution of wheat gluten samples during hydrolysis and the elimination of insoluble, undigested nonprotein substances after hydrolysis led to a high protein content 25 . Nanofiltration removed most of the mineral salts, and Na + and Cl − caused the remaining ash.
As summarized in Table 1, WOP were rich in glutamic acid (35.23 ± 0.12%), proline (10.62 ± 0.08%), leucine (5.03 ± 0.05%), serine (4.12 ± 0.01%) and phenylalanine (3.81 ± 0.05%). Furthermore, the content ratio of essential (lysine, tryptophan, phenylalanine, methionine, threonine, isoleucine, leucine and valine) and semiessential (arginine and histidine) amino acids was 21.95% and 4.48%, respectively. A rich amino acid composition exerts a significant impact on the regulation of human physiological functions. For example, glutamate is the main excitatory neurotransmitter used at synapses in the central nervous system and exerts an important effect on maintaining nerve cell signaling 25 . As the first limiting amino acid, lysine can affect the absorption and utilization of other amino acids as well as metabolism and immune function 26 . Leucine, isoleucine, and proline are branched-chain amino acids and are important sources of metabolic energy 27 . The functional activity of a peptide is critically influenced by the type and amount of amino acids constituting its makeup. Prior work has suggested Table 1. Amino acid composition (dry basis) of WOP (%) a . a Data correspond to the average and standard deviation of three independent experiments. b The content of amino acids in WOP. c Aspartic acid + asparagine. d Glutamine acid + glutamine. www.nature.com/scientificreports/ that many amino acids and their derivatives, such as glutamic acid, leucine, histidine, arginine, tyrosine, lysine, valine, alanine, cysteine, and proline, have antioxidant effects. Kong et al. 28 reported that corn oligopeptides have a considerable antioxidant capacity, with amino acids such as leucine having strong chelation effects. In WOP, the relative content of antioxidant amino acids is as high as 65.18%, such as leucine, arginine, tyrosine, cysteine, valine, alanine and other amino acids, potentially leading to chelation of metal ions, so WOP may be a natural substance with antioxidant activity. In addition, according to previous studies, many ACE inhibitory peptides contain valine, alanine, leucine, tyrosine, and phenylalanine 29 , which suggests that WOP may contain ACE inhibiting peptides and have potential ACE inhibiting activity.
Relative molecular weight distribution of WOP. Size exclusion chromatography with a high-performance liquid chromatography (HPLC) system was applied to analyze the molecular weight distribution of WOP. According to the results (Fig. 1), peptides with molecular weights below 1000 (94.32%) dominated WOP. Most of the peptides in WOP (61.23%) were in the molecular weight range of 500-140 and were mainly di-and tripeptides. According to reports, di-and tripeptides are actively transported in intestinal epithelial cells by specific peptide transporters (e.g., PepT1, PepT2), and amino acid residues of di-and tripeptides are systemically absorbed faster than free amino acids 30,31 .
These findings indicate that a two-step enzymatic hydrolysis followed by a multistep separation may efficiently remove large peptides or undigested proteins and produce oligopeptides. This method of enzymatic hydrolysis has also been carried out in past studies with good success 32,33 . The molecular structure and weight highly exerted an influence on the functional characteristics of peptides, which are greatly affected by processing conditions. It is widely known that most ACE inhibitory peptides from food proteins have relatively low molecular weights, most of which are short peptides less than 1000 Da 34 . Moreover, previous studies have demonstrated that compared with relatively large peptides, small peptides have higher antioxidant activities, and several di-and tripeptides have shown greater antioxidant activities than those their constituent amino acids 35,36 . According to recent advances in oxidant-antioxidant test systems, the effects of small peptides were greater than those of larger peptides and proteins 37 . Antioxidant properties of WOP. As an enzymatic hydrolysate, WOP showed strong scavenging ability against 1,1-diphenyl-2-picrylhydrazyl (DPPH), hydroxyl and 2,2′-azinobis (3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS) free radicals. The concentration for 50% of maximal effect (EC 50 ) of DPPH and hydroxyl free radical scavenging activity were 1.52 ± 0.09 and 6.23 ± 0.18 mg/mL, respectively, and ABTS free radical scavenging activity was 1.23 ± 0.03 mmol Trolox equivalent (TE)/g sample. Zhang et al. hydrolyzed wheat germ protein by Alcalase 2.4 L to obtain wheat germ peptides (molecular weight < 1000 Da), and the EC 50 value for DPPH were found to be 2.77 mg/mL 38 . Karami et al. evaluated the DPPH free radical scavenging activity of three hydrolysates of wheat germ protein prepared with Alcalase, pepsin or proteinase K, and the EC 50 values were 1.56, 1.78 and 1.83 mg/mL, respectively 24 . Compared with the above studies, the DPPH free radical scavenging activity of WOP in this study was slightly higher. Additionally, The EC 50 value of WOP for hydroxyl free radicals was better than those of peptides obtained from defatted wheat germ (6.04 mg/mL) 39 . Besides, WOP exhibited a high oxygen radical absorption capacity (ORAC) value (1216.87 ± 11.51 mmol TE/g sample) ( Table 2). The high antioxidant peptides obtained by hydrolyzing chicken egg whites with different proteases had an ORAC value between 900-1300 mmol Trolox equivalent (TE)/g sample, which is close to WOP 40 . Moreover, these antioxidant properties of WOP showed a strong dose-dependent relationship. The results revealed that two-step enzymatic hydrolysis was an effective approach to obtain an antioxidant hydrolysate from wheat gluten. Given its apparent antioxidant properties, amino acids or peptides contained in WOP act as electron donors and react with free radicals to generate stable products and end radical chain reactions. Among the four types of antioxidant assay system results, the ABTS free radical scavenging activity of WOP was remarkably strong. Moreover, the ABTS experiment was seemingly very effective in assessing antioxidant capacity. Therefore, the ABTS free radical antioxidant assay was used in further experiments.   33 showed that the IC 50 value of oligopeptides derived from black-bone silky fowl for ACE was 2.86 mg/mL. Moreover, after measuring the ACE inhibitory activity of 48 marine protein hydrolysates, He et al. 42 found IC 50 values ranging from 0.17 to 501.7 mg/ mL. Compared with these data, WOP had a relatively high ACE inhibitory activity. As there were many kinds of peptides in WOP, we speculated that some of the peptides in WOP may be competitive inhibitors with greater affinity for the ACE active region than WOP. According to studies 32,34 , the ACE inhibitory activity of peptides may be connected with the length of their peptide chains. The molecular weights of ACE inhibitory peptides are relatively low, generally less than 1000. Furthermore, hydrophobic amino acids in the C-terminus of antihypertensive peptides are significant for the activity of antihypertensive peptides 33 . WOP may contain a relatively large amount of peptides that can inhibit ACE. Therefore, there is strong reason to continue to identify and purify ACE inhibitory peptides from WOP.

ABTS free radical scavenging and ACE inhibitory characteristics of reversed-phase high performance liquid chromatography (RP-HPLC) fractions.
An RP-HPLC column was applied to fractionate WOP to determine the possible impacts of peptide composition on antioxidant activity and ACE inhibitory activity. Figure 3A shows the elution profiles of WOP. According to peak shape, height, area, and resolution, six component peaks were collected, fractions 1-6, where each component peak did not necessarily represent a single peptide. Repeated chromatography was used to separately collect six fractions, and their ABTS free radical scavenging and ACE inhibitory activities were determined. According to Fig. 3B, fractions 2-6 expressed equal or even higher ABTS radical scavenging activity (0.91-1.62 mmol TE/g sample) than that of WOP (1.23 ± 0.03 mmol TE/g sample), but fraction 1 had a lower activity. The ACE inhibition results were similar to those above, where fractions 2-6 exhibited higher ACE inhibitory activities (61.21-89.36%) than that of WOP, and fractions 1 showed a lower ACE inhibitory activity (34.18 ± 2.06%) at the concentration of 0.7 mg/mL (Fig. 3C). The inhibitory activities of fractions 5 and 6 were 1.39 and 1.74 times that of WOP, respectively.
According to these results, some peptides in certain fractions may have relatively higher ABTS free radical scavenging activity and ACE inhibition (or are present at a higher concentration) than those of peptides in other fractions. Therefore, for further identification and screening of active peptides, fractions 1-6 were all chosen.

Purification and recognition of antioxidative peptides and ACE inhibitory peptides in WOP.
For fractions 1-6, an electrospray ionization-quadrupole time-of-flight mass spectrometer (ESI-Q-TOF2) was used to recognize active peptides, resulting in the identification of 40 active peptides from WOP whose amino acid sequences and relative molecular masses are listed in Table 3. These peptides were composed Table 2. Antioxidant activities of WOP using four types of assay systems. a EC 50 value (mg/mL), defined as the concentration of samples needed to scavenge 50% of the free radicals. b mmol Trolox equivalent (TE)/g sample. c μmol TE/g sample. www.nature.com/scientificreports/ of 2-6 amino acids, which was consistent with the relative molecular mass analysis described above. It was reported that antioxidant and ACE inhibitory activities of peptides were highly influenced by peptide molecular weight. Peng et al. 43 reported that the 100-2800 Da fraction of peptides purified from the hydrolysate of whey protein had the strongest antioxidant activity. Saiga et al. 44 studied the peptides in hydrolyzed chicken breast muscle extract, and the result indicated that the small peptides (< 1000 Da) showed higher activity than those with molecular weight of > 1000 Da. The peptides identified in our study were all small peptides with molecular weight of < 1000 Da. Therefore, it was expected that the peptides identified in this study would exhibit antioxidant and ACE inhibitory activities. Selections of the peptides for ABTS free radical scavenging activity and ACE inhibitory activity assays were based on previous experience and related references. Therefore, we conducted a  www.nature.com/scientificreports/ preliminary analysis of the structural characteristics and amino acid composition of antioxidant peptides and ACE inhibitory peptides. Specifically, according to previous studies, some di-and tripeptides with aromatic amino acid residues (Tyr or Trp) and peptides containing Arg, Pro, Gln, or Asn are highly likely to have strong antioxidant activity 1,10,12 . In addition, peptides carrying a Tyr, Phe, Trp, Pro, or a hydrophobic amino acid residue at the C-terminus have strong ACE inhibitory activity 45 . On the basis of these reports, we selected and synthesized 15 peptides as potential candidates (Table 4) with these preferred characteristics from fractions 2-6 for further testing to determine their ABTS free radical scavenging activity and ACE inhibitory activity. Total antioxidant capacities can be assessed conveniently by virtue of the ABTS experiment. Figure 4A shows the ABTS free radical scavenging capacity, in which five peptides (Leu-Tyr, Pro-Tyr, Tyr-Gln, Ala-Pro-Ser-Tyr, Arg-Gly-Gly-Tyr) had extremely strong performance. Their ABTS free radical scavenging activities were 6.07 ± 0.38, 7.28 ± 0.29, 11.18 ± 1.02, 5.93 ± 0.20 and 9.04 ± 0.47 mmol Trolox equivalent (TE)/g sample, respectively. The free radical scavenging activity was 4.93, 5.92, 9.08, 4.82, and 7.35 times the WOP activity for these Table 3. Identification of peptides by quadrupole time-of-flight mass spectrometry (Q-TOF2) of WOP fractions collected from reversed-phase high-performance liquid chromatography (RP-HPLC). a Fractions are indicated in Fig. 3 b Three-letter representations of amino acids, and single-letter representations in parentheses. c Average mass. www.nature.com/scientificreports/  Fig. 3 b Three-letter representations of amino acids and single-letter representations in parentheses.  www.nature.com/scientificreports/ peptides, respectively, and the other peptides had weaker ABTS free radical scavenging abilities. Although some peptides do not exhibit strong ABTS free radical scavenging activity, we are unsure if they have synergistic effects with other peptides to enhance the overall antioxidant activity of WOP. On the other hand, based on these results, we were confident that at least five peptides (Leu-Tyr, Pro-Tyr, Tyr-Gln, Ala-Pro-Ser-Tyr and Arg-Gly-Gly-Tyr) in WOP are major contributors to its ABTS free radical scavenging activities. The active sites and groups they exposed may be able to act as electron donors, stop the chain reaction of free radicals, and ultimately achieve antioxidant effects 33 . Beermann et al. 46 isolated and purified soy peptides and found that the dipeptide Leu-Tyr has excellent antioxidant effects, which was in good agreement with our experiments herein. To the best of our knowledge, the other four peptides (Pro-Tyr, Tyr-Gln, Ala-Pro-Ser-Tyr, Arg-Gly-Gly-Tyr) have not been previously reported as antioxidant active peptides. Chen et al. 47 previously speculated that short peptides with aromatic amino acid residues (Trp or Tyr) may have potent antioxidant activity. The C-termini of four of the five antioxidant peptides identified in the present study were Tyr. Saito et al. 48 reported that the tripeptide Tyr-His-Tyr, separated from soy protein hydrolysate, has strong antioxidant capacity and great synergistic effects with phenolic antioxidants (e.g., BHA and δ-tocopherol). Interpreted by the special abilities of phenolic and indolic groups, the antioxidant activities of Trp and Tyr may act as hydrogen donors. With longer lifetimes than those of simple peroxy radicals, much more stable phenoxyl and indoyl radicals inhibit any reverse reaction or the propagation of the radical-mediated peroxidizing chain reaction 48 . In addition, oxidation of specific sites may be due to the binding of metal ions or other starting materials at specific sites on the polypeptides. The selective modification of a peptide to His, Pro, Met, Cys, Arg, Lys, and Trp residues may also be related to its antioxidant activity 48 . Figure 4B shows that 15 peptides exhibited different ACE inhibitory activities at 0.7 mg/mL (ranging from 1.94 to 83.13%). Among them, five peptides, Leu-Tyr, Leu-Val-Ser, Tyr-Gln, Ala-Pro-Ser-Tyr, and Arg-Gly-Gly-Tyr, attracted our attention due to their increased inhibitory activity against ACE compared to that of WOP. At this concentration, their activities were 1.59, 1.64, 1.18, 1.04, and 1.01 times that of WOP, respectively. Furthermore, we determined the IC 50 values of the five active peptides: Leu-Tyr (0.31 ± 0.02 mmol/L), Leu-Val-Ser (0.60 ± 0.03 mmol/L), Tyr-Gln (2.00 ± 0.13 mmol/L), Ala-Pro-Ser-Tyr (1.47 ± 0.08 mmol/L) and Arg-Gly-Gly-Tyr (1.48 ± 0.11 mmol/L). The IC 50 values of the five active peptides were calculated to be 0.09, 0.19, 0.62, 0.64 and 0.67 mg/mL, respectively, in the units of mg/mL, which were 7.56, 3.58, 1.10, 1.06, and 1.01 times that of WOP (0.68 ± 0.03 mg/mL), respectively. To the best of our knowledge, there is no report that the three peptides Leu-Val-Ser, Ala-Pro-Ser-Tyr and Arg-Gly-Gly-Tyr have ACE inhibitory activity. However, there are reports that the dipeptide Leu-Tyr had ACE inhibitory activity (IC 50 = 38.5 μM) and could lower blood pressure 49,50 . In addition, Leu-Tyr and Leu-Val-Arg, previously isolated from sardine muscle hydrolysate, have good ACE inhibitory effects 41 . The dipeptide Tyr-Gln was also already known to be isolated from buckwheat as an ACE inhibitory peptide (IC 50 = 628 μM) 51 . In the study of Balti et al. 52 , Ala-His-Ser-Tyr isolated from cuttlefish was found to have ACE inhibitory activity (IC 50 = 11.6 μM). Interestingly, the structure of Ala-Pro-Ser-Tyr found in the present study was similar to that of Ala-His-Ser-Tyr. Therefore, it was inferred that these two peptide sequences may result in good binding to ACE. For Arg-Gly-Gly-Tyr, another similar tetrapeptide, Tyr-Gly-Gly-Tyr, was found to be an effective ACE inhibitor in vitro by Saito et al. 53 . In addition, the digested product of Tyr-Gly-Gly-Tyr, Gly-Gly-Tyr , a tripeptide, also exerted good inhibitory activity (IC 50 = 1.3 μM), as shown by Shamloo et al. 54 . To a lesser extent, the released amino acid tyrosine can also exhibit ACE inhibitory activity on its own 55 . In the present study, Arg-Gly-Gly-Tyr, which has the Gly-Gly-Tyr sequence, displayed potent ACE inhibitory activity. Its activity may be due to the preferred amino acid sequence (Gly-Gly-Tyr) or amino acid (tyrosine) and their synergistic effects with the amino acid arginine at the N-terminus. Regarding the association between the structure and activity of ACE inhibitory peptides, Ryan et al. 56 suggested that hydrophobic, branched-chain or aromatic amino acids were important components of ACE inhibitory peptides, while hydrophilic amino acids had weak ACE inhibitory activity, as they were incompatible with the ACE active site. Cheung et al. 57 noted that active peptides tended to have Pro, Phe, or Tyr at the C-terminus and Val or Ile at the N-terminus. Some ACE inhibitory peptides with similar structures at the C/N termini, such as the lactotripeptides Val-Pro-Pro and Ile-Pro-Pro in yogurt, Ile-Ala-Pro in gliadin hydrolysate, Ala-Pro and Val-Arg in salmon hydrolysate, Tyr-Ala-Pro and Val-Ile-Ile in cuttlefish hydrolysate, and Ile-Gln-Pro in algae, have been isolated and identified 22,[58][59][60][61] . Among these peptides, it has been shown that lactotripeptides Val-Pro-Pro and Ile-Pro-Pro have good antihypertensive effects in clinical trials. However, the nature of the C/N terminal amino acids and whether they contain hydrophobic amino acids may not connect with the ACE inhibitory activity of short peptides. Despite the small number of amino acids in oligopeptides, simple secondary structures are formed due to the interactions between them and affect competition for the ACE active region. The antihypertensive drug captopril is synthesized according to the proline configuration 62 . The ACE inhibitory peptide is a competitive inhibitor with strong affinity to the active region of ACE. Its affinity with ACE is stronger than that of angiotensin I or bradykinin, and it is not easy to release from the ACE binding region, thus hindering the two biochemical reaction processes of catalytic hydrolysis of angiotensin I into angiotensin II and catalytic hydrolysis of bradykinin into inactivated fragment, thus lowering blood pressure 16,17 . Quantitative analysis of active peptides. Five antioxidant peptides and five ACE inhibitory peptides were identified from WOP. In particular, four peptides, Leu-Tyr, Tyr-Gln, Ala-Pro-Ser-Tyr and Arg-Gly-Gly-Tyr, had antioxidant and ACE inhibitory activities. High-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) was applied to estimate the concentration of highly active peptides in WOP. The results showed that the contents of Leu-Tyr, Pro-Tyr, Tyr-Gln, Ala-Pro-Ser-Tyr, Arg-Gly-Gly-Tyr, and Leu-Val-Ser were 155.04 ± 8.36, 2.08 ± 0.12, 1.95 ± 0.06, 22.70 ± 1.35, 0.25 ± 0.01, and 53.01 ± 2.73 μg/g of WOP, respectively (Table 5). Interestingly, it was found that four peptides have both good antioxidant activity and ACE inhibitory activity. These peptides can potentially be incorporated into functional foods for antioxidation or to www.nature.com/scientificreports/ partially address high blood pressure. In this study, our findings were limited in vitro. It was not clear whether the identified active peptides function in vivo. Therefore, further studies are needed to evaluate the in vivo effects of these active peptides. Previous studies have reported that the bioactivity of low-molecular-weight peptides can be retained during gastrointestinal digestion 32,33 . The bioavailability and metabolism of active peptides will also be further studied to better understand the multifunctionality of peptides.

Conclusion
WOP has good antioxidant activity and ACE inhibitory activity. We identified six highly active peptides (Leu-Tyr, Pro-Tyr, Tyr-Gln, Ala-Pro-Ser-Tyr, Arg-Gly-Gly-Tyr, and Leu-Val-Ser) from WOP, four of which have dual activities, most of which were novel active peptides. This study expanded the potential types of peptides exerting antioxidant and antihypertensive activities. WOP and these active peptides may be beneficial for the future development of physiologically functional foods. The in vivo effects of each peptide and the mechanism by which these active peptides carry out their effect will be researched by further studies. Enzymolysis and preparation of WOP. Wheat gluten was suspended with distilled water at a ratio of 1:10 (w/w) by stirring in a homogenizer (Donghua Homogenizer Factory, Shanghai, China) at 25,000 × g for 10 min. The slurry was digested with α-amylase using an enzyme-to-substrate protein ratio of 1:100 (w/w) at pH 5.0 and a temperature of 75 °C for 1 h to hydrolyze the starch in wheat gluten. The hydrolysate was then centrifuged at 3000 × g (LG10-2.4A, Beijing LAB Centrifuge Co., Ltd., China) for 10 min to remove nonprotein substances such as carbohydrates. The insoluble protein precipitate was recovered and resuspended with distilled water in the homogenizer with the same procedures as above. Two-step enzymolysis was adopted for the resulting slurry of highly pure proteins. The first hydrolysis was performed with Alcalase 2.4 L (enzyme/substrate ratio of 1:100 [w/w]) at pH 8.5 at 55 °C for 2 h, and then the second hydrolysis was performed with Protex 7L for another 2 h at pH 6.5 at 60 °C at an enzyme/substrate ratio of 1:100 (w/w). The pH value was kept unchanged during the process of hydrolysis by adding 1.0 mol/L HCl or NaOH. Afterwards, hydrolysis was ceased by heating at 100 °C for 10 min. The hydrolysate was centrifuged at 3,000 × g using the same centrifuge as above for 15 min. Molecular weight cut-off (MWCO) ceramic membranes of 10,000 and 1000 Da were applied to filter the supernatant successively to obtain a filtrate containing oligopeptides. Next, the filtrate with oligopeptides was subjected to nanofiltration to remove mineral salts and free amino acids. An R-151 rotary evaporator (BUCHI Co., Ltd., Switzerland) was used to concentrate the resulting filtrate, and activated carbon was applied to decolor it. Then, an L-217 Lab spray dryer (Beijing Laiheng Lab-Equipments Co., Ltd., Beijing, China) was employed to spray dry the concentrated oligopeptide solution to obtain oligopeptide powder. Spray drying was carried out at 25 mL/min feed temperature, 14 kPa feed rate, 135 °C inlet air temperature and 20 °C inlet air pressure.  63 . An 835-50 automatic amino acid analyzer (Hitachi, Ltd., Tokyo, Japan) was applied to determine the amino acid composition in accordance with the approaches of Yang et al. 64 . DPPH free radical scavenging activity assay. The approach of Kedare et al. 65 was applied to measure the DPPH free radical scavenging ability of the samples. Briefly, 100 μL of 0.1 mmol/L DPPH in 95% ethanol was added to 100 μL of each sample at various solid concentrations. After shaking, the mixture stood for 30 min in the dark at room temperature, and then a Spectra MR multimode reader (DYNEX Technologies, Inc., Chantilly, VA, USA) was applied to assess the absorbance at 517 nm (A S ). Samples in the control experiment (A C ) were replaced by ethanol, and a blank excluding DPPH (A B ) was prepared as above. The calculation of scavenging activity was subject to [ ABTS free radical scavenging activity assay. The ABTS assay kit instructions were applied to assess the ABTS free radical scavenging activity. Briefly, the 7 mmol/L ABTS stock solution in 2.45 mmol/L potassium persulfate solution constituted the stock solution of ABTS free radicals, and the resulting mixture was placed in the dark at room temperature for 16 h. The ABTS free radical solution was diluted with distilled water to an absorbance of 0.70 ± 0.05 at 734 nm. The diluted ABTS free radical solution (200 μL) was mixed with the sample (10 μL), and the absorbance was recorded at 734 nm after 10 min (distilled water was used as the control). The final scavenging activity was expressed as mmol TE/g sample 33 .

Determination of chemical composition and amino acid composition.
ORAC-based assay. For the ORAC-based assay, the approach of Schauss et al. 67  respectively, and the fluorescence was measured every 2 min. The total read time was 150 min, and PBS was used as a blank control. In addition, a control without AAPH was used. We calculated the relative fluorescence value and the integral area (AUC) under the fluorescence decay curve with and without antioxidant. Then, we subtracted the control AUC without the antioxidant to calculate the protective area of the antioxidant (net AUC). The Trolox antioxidant standard curve was applied to obtain the ORAC antioxidant index of the sample.
Determination of ACE inhibitory activity. The modified approach of Cushman and Cheung 68 was used to analyze ACE inhibitory activity. Briefly, the preincubation of 20 µL of sample and 30 µL of ACE solution (60 mU/mL) was performed at 37 °C for 5 min. After adding 50 µL of the substrate (7.6 mmol/L HHL in 50 mmol/L sodium borate buffer containing 300 mmol/L NaCl at pH 8.3), the 30-min incubation of the mixture was performed at ambient temperature. The reaction was ceased with 100 µL of 1.0 mol/L HCl. RP-HPLC was employed to determine the amount of HA liberated by ACE on an Inertsil ODS-SP C18 column (4.6 mm × 150 mm, Shimadzu, Kyoto, Japan). The peak area was subject to the calculation of ACE inhibitory activity.
Isolation of WOP with RP-HPLC. An RP-HPLC system (Shmadzu, Kyoto, Japan) with an XBridge BEH130 C18 column (4.6 × 250 mm, Waters, USA) was used to separate WOP. Eluent A (Milli-Q water with 0.1% (v/v) TFA) and eluent B (80% (v/v) acetonitrile with 0.1% (v/v) TFA) were used for gradient elution with the following program: 0-15% B for 0-50 min, 15-40% B for 50-120 min, and 40-80% B for 120-140 min. The flow rate was 0.6 mL/min. The effluent was monitored at 220 nm. Fifty microliters of the 10 mg/mL sample was injected. The separation procedure was repeated successively for six times using automatic collector (Fraction Collecter10A, Shmadzu, Kyoto, Japan) by setting up the automatic collection time program. For further analysis, the fractions from the RP-HPLC system were freeze-dried. www.nature.com/scientificreports/ 5 µL/min. The ionization potential was 3,000 V in positive mode. The atomization gas and collision gas were N 2 and Ar, respectively. Firstly, the ESI-MS spectrogram was obtained by first-order mass spectrometry. The ions to be measured were selected from the ESI-MS spectrogram, and then analyzed to obtain the ESI-MS/MS spectrogram. The mass spectrum was analyzed using MassLynX software (Waters, USA).
Peptide syntheses. Peptides were synthesized using the conventional solid-phase synthesis (fluorenyl methoxycarbonyl (Fmoc) strategy) with a peptide synthesizer (model 90, AAPPTec, USA) by GL Biochem (Shanghai) Ltd. The synthesized peptides were purified and detected by HPLC using a C18 column (ProStar, Varian Co., USA). The molecular masses of the peptides were determined by a mass spectrometer (Voyager DE-Pro, ABI, USA). The synthesized peptides with the purity of higher than 98% were stored at − 80 °C until use 2 .
Quantitative analysis of active peptides. HPLC-MS/MS was employed to quantitatively analyze the peptides in WOP. A Dionex Ultimate 3000 HPLC system (Thermo Scientific, Waltham, MA, USA) was applied to separate peptides on an MSLAB HP-C18 column (150 mm × 4.6 mm) by a 10 min gradient elution at a flow rate of 1.0 mL/min. Eluent A (water with 2 mmol/L ammonium formate) and eluent B (acetonitrile with 2 mmol/L ammonium formate) were used to elute peptides. Ten microliters of the sample was injected. An API 3200 Q TRAP HPLC-MS/MS (Applied Biosystems, Foster City, CA, USA) with an ESI source was used to perform tandem mass spectrometry. The collision energy was 30 eV. The ionspray voltage was 4500 V in positive mode. The atomization gas and auxiliary gas were 45 and 50 psi, respectively. The atomization temperature was 550 ℃. The mass spectrum was analyzed using Analyst Software 1.5 (Applied Biosystems, USA). WOP and the synthesized peptides were dissolved in distilled water, and subjected to HPLC-MS/MS. The standard curve equation of each peptide was obtained. Then, the quantification of each peptide in WOP was performed and measured. www.nature.com/scientificreports/