Two Novel Antioxidant Nonapeptides from Protein Hydrolysate of Skate (Raja porosa) Muscle

In the current study, the preparation conditions of neutrase hydrolysate (SMH) from skate (Raja porosa) muscle protein were optimized using orthogonal L9(3)4 tests, and R values indicated that pH was the most important factor affecting HO· scavenging activity of SMH. Under the optimum conditions of pH 7.0, enzymolysis temperature 60 °C, enzyme/substrate ratio (E/S) 2%, and enzymolysis time 5 h, EC50 of SMH on HO· was 2.14 ± 0.17 mg/mL. Using ultrafiltration, gel filtration chromatography, and RP-HPLC, two novel antioxidant nonapeptides (SP-A and SP-B) were isolated from SMH and their amino acid sequences were found to be APPTAYAQS (SP-A) and NWDMEKIWD (SP-B) with calculated molecular masses of 904.98 Da and 1236.38 Da, respectively. Both showed strong antioxidant activities. SP-A and SP-B exhibited good scavenging activities on HO· (EC50 0.390 and 0.176 mg/mL), DPPH· (EC50 0.614 and 0.289 mg/mL), and O2−· (EC50 0.215 and 0.132 mg/mL) in a dose-dependent manner. SP-B was also effective against lipid peroxidation in the model system. The aromatic (2Trp), acidic (2Asp and Glu), and basic (Lys) amino acid residues within the sequences of SP-B might account for its pronounced antioxidant activity. The results of this study suggested that protein hydrolysate and peptides from skate muscle might be effective as food additives for retarding lipid peroxidation occurring in foodstuffs.

familiar proteases were used for the hydrolysis of skate muscle. The enzymolysis conditions, degree of hydrolysis (DH) and HO· scavenging activities of hydrolysates are shown in Table 1. Among the hydrolysates of skate muscle prepared with different enzymes, the neutrase hydrolysate showed the highest DH (14.07% ± 0.86%) and HO· scavenging activities with EC50 of 2.57 ± 0.23 mg/mL. These results were consistent with previous reports that the antioxidant capacities of protein hydrolysates were significantly influenced by the enzymes used and positively correlated with the DHs of hydrolysates [16,23]. For these reasons, neutrase was chosen for the preparation of the hydrolysate and its enzymolytic conditions were optimized using an orthogonal L9(3) 4 test.  The R range analysis showed that the influence of four selected factors on HO· scavenging activity (EC50) was: A (pH) > D (Temp.) > B (E/S) > C (Time). Hydrolysates (SMH) with the most pronounced antioxidant activity were produced when the hydrolysis conditions for neutrase were A2B3C3D2. Under these optimal conditions, the EC50 of SMH on HO· scavenging activity was confirmed to be 2.14 ± 0.17 mg/mL, which is to say, the optimum condition of obtaining the highest antioxidative protein hydrolysate of skate muscle was enzymolysis time 5 h, total enzyme dose 2%, enzymolysis temperature 60 °C and pH 7.0.

Fractionation of SMH by Ultrafiltration
Ultrafiltration is a technique used to separate substances dissolved in solution based on molecular size. It is widely used to purify and concentrate target fraction from protein hydrolysates on the molecular-weight-cutoff (MWCO) membrane. Therefore, the neutrase hydrolysate was fractionated into three different MW fractions through 3 k and 10 kDa ultrafiltration (UF) membranes and named as SMH-I (MW > 10 kDa), SMH-II (MW 3-10 kDa), and SMH-III (MW < 3 kDa), respectively. SMH, SMH-I, SMH-II, and SMH-III scavenged HO· in a concentration-dependent manner with EC50 values of 2.14 ± 0.17, 5.93 ± 0.26, 2.47 ± 0.13, and 1.84 ± 0.11 mg/mL, respectively. HO· scavenging assay showed the highest antioxidant capacity for SMH-III with peptides of MW below 3 kDa, followed by SMH-II containing peptides of MW ranged from 3 to 10 kDa. This result was consistent with some earlier reports showing that the antioxidant activity of hydrolysates relied on their MW distributions, and hydrolysates and peptides with lower MW could cross the intestinal barrier more easily [7,16], and interact more effectively with free radicals interfering in oxidative process [24].

Anion-Exchange Chromatography of SMH-III
To produce sub-fractions with more anti-oxidant activity, SMH-III was fractionated using DEAE-52 cellulose anion-exchange chromatography and separated into five fractions ( Figure 1). Fr. 1 and Fr. 2 were washed out by deionized water, Fr. 3 and Fr. 4 were washed out by 0.1 M NaCl, and Fr. 5 was washed out by 0.5 M NaCl. Figure 1 indicates that Fr. 5 possessed the highest HO· scavenging rate with EC50 of 1.218 ± 0.10 mg/mL among all the samples.

Figure 1.
Elution profile of SMH-III as indicated by DEAE-52 cellulose anion-exchange chromatography, and upper panel represented HO· scavenging activity (EC50 (mg/mL)). All the values were mean ± SD; SD: standard deviation. (a-f) Values with same letters indicated no significant difference of different sample at same concentrations (p < 0.05).

Amino Acid Sequence Analysis and Molecular Mass Determination
The N-terminal amino acid sequences and molecular mass of SP-A and SP-B were determined using a Protein/Peptide Sequencer and Electrospray ionization mass spectrometry (ESI-MS) (Figure 4)

Antioxidant Activity of SP-A and SP-B
To evaluate the antioxidant activities of SP-A and SP-B, assays of DPPH· , HO· and O2 − · scavenging activities and lipid peroxidation inhibition activity were performed and the fractions were compared to ascorbic acid (AA) and BHT as positive control ( Figure 5).

Hydroxyl Radical Scavenging Activity
HO· is the most reactive of the oxygen radicals, which can react with almost all the substances in the cells and induce severe damage to biomacromolecules in cells, such as enzymes, DNA, and lipids. As shown in Figure 5A, SP-A and SP-B scavenged HO· in a concentration-dependent manner with EC50 values of 0.390 and 0.176 mg/mL, respectively. SP-B showed the most pronounced activity of any of the hydrolysates, fractions, or peptides. Many antioxidant peptides have been isolated from other aquatic substances and their by-products. The EC50 of SP-B was lower than those of GFRGTIGLVG (0.293 mg/mL) and GPAGPAG (0.240 mg/mL) from protein hydrolysate of croceine croaker scales [2], PSYV (2.64 mg/mL) from protein hydrolysate of loach (Misgurnus anguillicaudatus) [25], YPPAK (0.228 mg/mL) from protein hydrolysate of blue mussel (Mytilus edulis) [10], PYFNK (0.24 mg/mL) from protein hydrolysate of Sphyrna lewini [11], and NGLEGLK (0.313 mg/mL) and NADFGLNGLEGLA (0.612 mg/mL) from protein hydrolysate of giant squid [26,27], but it was higher than those of WDR (0.15 mg/mL) from protein hydrolysate of S. lewini [11], GFPSG (0.107 mg/mL) from protein hydrolysate of croceine croaker scales [2], MQIFVKTLTG (0.005 mg/mL) and DLSDGEQGVL (0.007 mg/mL) from protein hydrolysate of venison [28], and NGPLQAGQPGER (0.123 mg/mL) and FDSGPAGVL (0.078 mg/mL) from protein hydrolysate of giant squid [26,27]. SP-B showed good activity against HO· , which indicated that it could be used as the scavenging agent for preventing HO· -induced damage to living cells.

1,1-Diphenyl-2-picrylhydrazyl (DPPH) Radical Scavenging Activity
Because of its convenience and efficiency, DPPH· scavenging assay is widely used to evaluate the antioxidant ability of protein hydrolysates, fractions, and purified peptides and their suitability for use as free radical scavengers and hydrogen donors. As shown in Figure 5B, SP-A and SP-B were able to reduce stable DPPH· to yellow-colored diphenylpicrylhydrazine. They both showed concentration-dependent anti-DPPH· activity. The EC50 of SP-A and SP-B were 0.614 and 0.289 mg/mL, respectively, and SP-B showed most pronounced DPPH· scavenging activity of any of the hydrolysates, fractions, or peptides except for the positive control, AA. The EC50 of SP-B was lower than those of GFRGTIGLVG (1.271 mg/mL) and GPAGPAG (0.675 mg/mL) from protein hydrolysate of croceine croaker scales [2], PSYV (17.0 mg/mL) from protein hydrolysate of loach [25], YPPAK (2.62 mg/mL) from protein hydrolysate of blue mussel [10], and WDR (3.63 mg/mL) and PYFNK (4.11 mg/mL) from protein hydrolysate of S. lewini [11], but higher than those of GFPSG (0.283 mg/mL) from protein hydrolysate of croceine croaker scales, and VCSV (0.045 mg/mL) and CAAP 0.009 mg/mL) from protein hydrolysate of flounder fish muscle [22]. The data suggested that SP-A and SP-B could convert DPPH· to less harmful or non-harmful products and break the radical chain reaction.

Superoxide Anion Radical Scavenging Activity
Superoxide radicals (O2 − · ) can release protein-bound metals and form perhydroxyl radicals, which will initiate lipid oxidation in organism and food systems. The O2 − · scavenging effects of SP-A and SP-B were evaluated at concentrations ranging from 0.065 to 1.000 mg/mL, and the antioxidative activity increased with increasing concentration of the peptides ( Figure 5C). SP-A and SP-B could effectively scavenge O2 − · with EC50 of 0.215 and 0.132 mg/mL, respectively. The EC50 of SP-B was lower than GFRGTIGLVG (0.463 mg/mL) and GFPSG (0.151 mg/mL) from protein hydrolysate of croceine croaker scales [2], and NADFGLNGLEGLA (EC50 0.864 mg/mL) and NGLEGLK (EC50 0.419 mg/mL) from protein hydrolysate of giant squid muscle [27], but it was higher than those of GPAGPAG (0.099 mg/mL) from protein hydrolysate of croceine croaker scales [2], YPPAK (0.072 mg/mL) from protein hydrolysate of blue mussel [10], LKQELEDLLEKQE (EC50 0.128 mg/mL) from protein hydrolysate of oyster [29], and WDR (EC50 0.09 mg/mL) and PYFNK (EC50 0.11 mg/mL) from protein hydrolysate of S. lewini [11]. In living cells, the neutralization of O2 − · to hydrogen peroxide can be catalyzed by superoxide dismutase (SOD), which is one of the defense mechanisms of cytoprotection against reactive oxygen. SP-B showed strong O2 − · scavenging activity and might have SOD-like activity, thereby scavenging O2 − · in biological organisms.

Lipid Peroxidation Inhibition Assay
Each radical scavenging assay represented only one kind of antioxidant mechanism, which could not reflect the comprehensive capacity of samples that acted as antioxidants to retard or inhibit lipid oxidation in biological organisms or food systems. Lipid peroxidation is a complicated process involving the formation and propagation of lipid radicals and hydroperoxides in the presence of oxygen. The activities of SP-A and SP-B against the peroxidation of linoleic acid were investigated and compared to that of BHT in a subsequent experiment.
As shown in Figure 5D, the control (not exposed to antioxidant) had the highest absorbance value in the test after a seven-day incubation, indicating the highest degree of oxidation among the samples, but SP-A and SP-B inhibited lipid peroxidation in linoleic acid emulsion system effectively for up to seven days. The activity of SP-B was similar to that of positive control (BHT). Results indicated that SP-B had the same comprehensive capacity as BHT in the linoleic acid peroxidation system.

Relationship between Antioxidant Activities and Amino Acid Compositions of Peptides
The antioxidant activities of peptides may rely on molecular size, amino acid sequence, and the presence of amino acids capable of involvement in oxidative reactions [30,31]. In context, the amino acid compositions of SP-A (APPTAYAQS) and SP-B (NWDMEKIWD), such as hydrophobic/aromatic, acidic, and basic amino acids, were different. This could be the main cause of their antioxidant properties.
Antioxidant peptides possess some metal chelation and hydrogen-and electron-donating activity, which allows them to interact with free radicals and either of them terminates radical chain reactions or prevents them from taking place [25,32]. Sarmadi and Ismail reported that aromatic amino acids such as Trp, Tyr, and Phe, could donate protons to electron-deficient radicals and therefore enhanced the radical scavenging activities of peptides [33]. Dá valos et al. reported that Met, Trp, and Tyr showed the most pronounced antioxidant activity of any amino acids [34]. Ren et al. reported that one or more residues of Trp, Met, Tyr, His, Pro, Cys, and Phe could enhance the activity of antioxidant peptides [32]. This finding was further supported by the research reported by Zhang, Mu, and Sun, who found that antioxidant amino acids, such as Trp, Met, Tyr, and Phe, were responsible for their activities of the identified peptides (YYIVS, TYQTF, SGQYFL, YYDPL and YMVSAIWG) [35]. One Tyr and two

Pro in sequences of SP-A (APPTAYAQS) and two Trp and one Met in sequence of SP-B (NWDMEKIWD) should be responsible for their high antioxidant activities.
In many previous studies, hydrophobic amino acids were considered the main factors affecting the antioxidant activities of peptides [25,36,37]. Although, there were more hydrophobic amino acids in the sequence of SP-A (2Pro, 3Ala, and Tyr) than those in the SP-B (2Trp, Ile, and Met), SP-A showed less free radical scavenging and less lipid peroxidation inhibition activity than SP-B in the current study. There might be other factors affecting the activity of SP-B. Carefully analyzing the amino acid compositions of SP-A and SP-B, the significant difference between the two peptides was the presence of basic (Lys) and acidic (2Asp and Glu) amino acid residues in the sequence of SP-B. Byun et al. and Gimenez et al. have reported that acidic and basic amino acid residues played a role in metal ion chelating activity, which is related to antioxidant activity [38,39]. Ren et al. have reported that basic peptides had greater capacity to scavenge hydroxyl radicals than acidic or neutral peptides [32]. In addition, Klompong et al. reported that the ratio of hydrophilic and hydrophobic forces of peptides had another crucial influence on the peptide solubility in aqueous solution and lipids [23]. It might be speculated that the acidic and basic amino acid residues in SP-B made it easier for the structure to interact with free radicals and exert antioxidant activity at the cost of their own antioxidant activity. SP-B was found to have more long side chains than SP-A did. Carboxyl and amino groups in the side chains can act as hydrogen donors and metal ion chelators [40]. These findings were expected because amino acids with long side chains can interact with free radicals very effectively and inhibit the propagation cycles of lipid peroxidation.

Preparation of Protein Hydrolysates Using Five Proteases
The skate muscle was separated manually, pounded to homogenate, and defatted as described in a previous work [10]. The homogenate was mixed with isopropanol at a ratio of 1:4 (w/v), homogenized, and allowed to stand at 35 °C for 1 h. The mixture was filtered and the resulted residue was defatted at 35 °C for 1.5 h using isopropanol at a ratio of 1:4. Finally, the mixture was centrifuged at 4500× g for 15 min and the precipitate (defatted protein) was freeze-dried and stored at −20 °C. Defatted protein was dissolved in different buffer solution at a ratio of 1:4 (w/v) and hydrolyzed for 3 h separately using five proteases with an enzyme/substrate (E/S) ratio of 1.5% (w/w) ( Table 1). Hydrolysate samples were taken out, inactivated in boiling water bath for 10 min, and centrifuged at 10,000× g for 15 min. The resulting supernatants were prepared for measurement of the degree of hydrolysis (DH) and HO· scavenging activity.

Optimum Conditions for the Active Enzymatic Hydrolysate
To prepare defatted protein hydrolysates of skate muscle; enzymatic hydrolysis was performed using neutrase. An orthogonal L9(3) 4 test design was used to optimize the conditions of enzymatic hydrolysis [11,12]. Four controllable variables, including pH, enzyme/substrate ratio (E/S; %), time (h), and temperature (°C), were selected. The selected variables and their levels were listed in Table 2. Defatted protein (5 g) was put into 20 mL distilled water and hydrolyzed with neutrase according to the experimental runs ( Table 2). The subsequent processing was the same as the method described at the section of preparation of protein hydrolysates using five proteases. The hydrolysate prepared under the optimum conditions was referred to as SMH.

Fractionation of SMH by Ultrafiltration
SMH solution was filtered using two ultrafiltration membranes (Diaflo membranes, 76 mm diameter, MilliporeGlostrup, Denmark) with 3 and 10 kDa MWCO to produce three fractions corresponding to molecular weights (MW) above 10 kDa (SMH-I), between 3 and 10 kDa (SMH-II), and below 3 kDa (SMH-III). Three fractions were dialyzed, concentrated, and lyophilized. SMH-III was chosen for further study on HO· scavenging activity.

Gel Filtration Chromatography of SMH-III
Gel filtration chromatography was performed according to the procedure of Luo et al. [12]. Fr. 5 (100.0 mg) was dissolved in 5 mL deionized water and purified by Sephadex G-15 gel filtration column (2.6 × 80 cm) equilibrated previously with deionized water. The column was eluted with deionized water at a flow rate of 1 mL/min, and the eluate was collected every 3 mL and measured at 280 nm. Three subfractions (Fr. 5-1, Fr. 5-2, and Fr. 5-3) were collected, lyophilized, and their HO· scavenging activities were determined.

Degree of Hydrolysis (DH)
DH was determined according to the method described by Chi et al. [21]. Hydrolysate (50 μL) was mixed with 0.5 mL of phosphate buffer (0.2 M, pH 8.2) and 0.5 mL of TNBS reagent (0.05%) freshly prepared by diluting with deionized water. After being incubated at 50 °C for 1 h in a water bath, the reaction in mixture was stopped by adding 1 mL of HCl (0.1 M) and incubated at 25 °C for 30 min. The absorbance was measured at 420 nm. L-Leucine was used as a standard. To determine the total amino acid content, defatted skate muscle was completely hydrolyzed with 6 M HCl with a sample/acid ratio of 1:100 at 120 °C for 24 h. DH (%) was calculated using the following equation: Where, At is the number of α-amino acids released at time t, A0 is the number of α-amino acids in the supernatant at 0 h, and Amax is the total number of α-amino acids obtained after acid hydrolysis at 120 °C for 24 h.

Amino Acid Sequence Analysis and Molecular Mass Determination
The antioxidant peptides of SP-A and SP-B were subjected to N-terminal amino acid sequencing on an Applied Biosystems 494 protein sequencer (Perkin Elmer Wallace Inc., Akron, OH, USA) according to the previous method [2]. Edman degradation was performed according to the standard program supplied by Applied Biosystems.
Accurate molecular masses of SP-A and SP-B were determined using a Q-TOF mass spectrometer (Micromass, Waters, Los Angeles, CA, USA) coupled with an electrospray ionization (ESI) source on the previous method developed by Luo et al. [11]. Ionization was carried out in positive mode with a capillary voltage of 3500 V. Nitrogen was maintained at 40 psi for nebulization and 9 L/min at 350 °C for evaporation temperature. Data were collected in centroid mode from m/z 100 to 2000.

Antioxidative Activity
The DPPH· , HO· and O2 − · scavenging activities were measured according to the method described by Wang et al. [11], and the results were expressed as a half elimination ratio (EC50) defined as the concentration where a sample caused a 50% decrease of the initial concentration of DPPH· , HO· and O2 − · .
3.6.1. HO· Scavenging Activity HO· scavenging activity was measured using the method developed by Wang et al. [10]. In this system, HO· is generated by the Fenton reaction. HO· can oxidize Fe 2+ to Fe 3+ , and only Fe 2+ can combine with 1,10-phenanthroline to form a red compound (1,10-phenanthroline-Fe 2+ ). It shows maximum absorbance at 536 nm. The concentration of HO· is reflected by the degree of decolorization of the reaction solution. Briefly, 1,10-phenanthroline solution (1.0 mL, 1.865 mM) and the sample (2.0 mL) were added into a screw-capped tube and mixed. The FeSO4· 7H2O solution (1.0 mL, 1.865 mM) was then pipetted into the mixture. The reaction was initiated by adding 1.0 mL H2O2 (0.03% v/v). After incubation at 37 °C for 60 min in a water bath, the absorbance of the reaction mixture was measured at 536 nm against a reagent blank. Reaction mixture without any antioxidant served as the negative control, and mixture without H2O2 served as the blank. HO· scavenging activity (HRSA) was calculated by the following formula: HRSA(%) = [(As − Ac)/(Ab − Ac)] × 100% (2) where As is the sample absorbance; Ac is the control group absorbance; and Ab is the blank absorbance.

DPPH· Scavenging Activity
DPPH· scavenging activity was tested as previously described [10]. Two milliliters of deionized water containing different concentrations of sample (from 0.125 to 2.0 mg/mL) were placed in the cuvettes, and then 500 μL of ethanol solution of DPPH (0.02%) and 1.0 mL of ethanol were added. The control containing DPPH solution without sample was also prepared. In blank, DPPH solution was replaced with ethanol. The antioxidant activity of the sample was evaluated using the relative amount of inhibition of DPPH· with the following equation: DPPH· scavenging activity (%) = (Ac + Ab − As)/Ac × 100% (3) where As is the sample absorbance; Ac is the control group absorbance; and Ab is the blank absorbance.

O2 − · Scavenging Activity
The O2 − · scavenging activity of the sample was investigated by the previous method [21]. In the experiment, O2 − · was generated in 1 mL of nitrotetrazolium blue chloride (2.52 mM), 1 mL of NADH (624 mM), and 1 mL of different concentrations of samples (from 0.065 to 1.0 mg/mL). The reaction was initiated by adding 1 mL of phenazine methosulfate solution (120 μg/mL) to the reaction mixture. The absorbance was measured at 560 nm against the corresponding blank after 5 min incubation at 25 °C. The capacity of scavenging O2 − · was calculated using the following equation: O2 − · scavenging activity (%) = [(Ac − As)/Ac] × 100% (4) where As is the absorbance with sample; and Ac is the absorbance without sample.