Okra seeds (Abelmoschus esculentus variety) were purchased from a local market in Baghdad, Iraq. The enzymes pepsin (4500 U/g), trypsin (9500 U/g) and papain (8500 U/g) were purchased from Sigma Aldrich, USA. All chemicals used were of analytical grade.

Okra protein isolate (OPI) and concentrate (OPC) were prepared according to Kareem and Shaker [11] from defatted okra powder. Briefly, defatted okra powder was prepared by removal of the seed peel then grinding the seed meal. The ground meal was defatted with ethanol to produce OPC which contained 72.1% protein, 0.86% fat, 9.94% ash and 12.37% carbohydrate. OPI was prepared from OPC by precipitation with HCl at pH 4 resulting in a composition of 91.2% protein, 0.32% fat, 2.67% ash and 3.2% carbohydrate.

Enzymatic hydrolysis of OPI and OPC was conducted as described by Fan et al. [3] and named OPIH and OPCH. Five grams of protein sample was dissolved in distilled water then brought to 100 mL with Glycine-HCl buffer (0.5 M, pH 2). Pepsin was added resulting in an enzyme:protein ratio of 1:50 (g/g). The sample was placed in a shaking incubator at 37˚C for 360 min. Aliquots, 10 mL, were removed at times 0, 15, 30, 60, 90, 120, 180, 240, 300, and 360 min. Aliquots were heated to 90˚C for 5 min to stop the enzymatic reaction. For trypsin hydrolysis, 5 g of protein sample was dissolved in distilled water and brought to 100 mL with sodium phosphate buffer (0.05 M, pH 7.5) and trypsin was added resulting in an enzyme:protein ratio of 1:100 (g/g) and incubated in a shaking incubator at 45˚C for 360 min. Aliquots were removed at the 10 time points and the reaction stopped as described above for pepsin. For papain hydrolysis, 5 g of protein sample was dissolved in distilled water and brought to 100 mL with sodium phosphate buffer (0.05 M pH 6.0) and papain was added for an enzyme:protein ratio of 1:100 (g/g). Aliquots were removed at the 10 time points and the reaction stopped as described above for pepsin. Each hydrolysis reaction was replicated and the aliquots at each time point were pooled.

The degree of hydrolysis (DH) of each protein sample at each time point with each enzyme was determined according to Benjakul and Morrissey [13] . Briefly, OPCH and OPIH, 0.125 mL (or 10 mg sample) was added to 2 mL sodium phosphate buffer (0.2125 M, pH 8.2) and 1 mL of 2,4,6-trinitrobenzenesulfonic acid (TNBS) (0.1% solution in water) were added. Samples were placed in a water bath at 50˚C for 30 min in the dark. Adding 2.0 mL of 0.1 M sodium sulfite terminated the reactions. The mixtures were cooled to room temperature for 15 min. The absorbance was measured at 420 nm and alpha-amino acid was expressed in terms of L-leucine. The DH was determined using the following equation:

DH% = [(L_t − L₀)/(L_max − L₀)] × 100

L_t corresponds to the amount of amino acid liberated at time t. L₀ was the amount of amino acid in the original samples. L_max was the total amount of amino acid in the original OPC and OPI after acid hydrolysis. Lmax was determined by mixing 50 mg protein from OPC or OPI with 4.5 mL of 6 N HCl. Tubes were flushed with nitrogen gas and sealed in a screw-cap tube. The hydrolysis was run at 100˚C for 24 hrs., then filtered with Whatman paper No. 1 and neutralized with 6 N NaOH before amino acid determination as described above [13] .

The radical scavenging activity of OPC and OPI hydrolysates was determined using the DPPH assay as described by Thiansilakul et al. [14] . A sample of 1.5 mL of hydrolysate (50 mg sample) was mixed with 1.5 mL DPPH (0.15 mM in 95% ethanol) and held for 30 min at room temperature in the dark. A control was prepared with 1.5 mL of distilled water and 1.5 mL DPPH. The absorbance at 517 nm was recorded and compared to unhydrolyzed OPC or OPI dissolved in distilled water. The scavenging activity was calculated using the following equation where A₅₁₇ is the absorbance at 517 nm of the sample and B₅₁₇ is the absorbance at 517 nm of the control.

Radical scavenging activity (%) = [(B₅₁₇ − A₅₁₇)/B₅₁₇] × 100

The reducing power of OPIH and OPCH was conducted according to Wu et al. [15] . A sample (1 mL or 0.05 mg) was added to 1 mL phosphate buffer (0.2 M, pH 6.6) and 1 mL 1% potassium ferrocyanide. This was incubated at 50˚C for 20 min then 1 mL of a 10% solution of tricloroacetic acid was added along with 1 mL of distilled water and 0.2 mL of 1% FeCl₃. A control sample was prepared using distilled water in place of the protein hydrolysates. After 10 min at room temperature, the absorbance at 700 nm was recorded.

The metal chelating activity of OPCH and OPIH was determined according to Boyer and McCleary [16] . Aliquots, 4.7 mL (240 mg sample) was combined with 0.1 mL of FeCl₂ (2 mM) and 0.2 mL of 5 mM ferrozine solution. The control samples were 4.7 mL of distilled water in place of the protein hydrolysates. The samples were kept at room temperature for 20 min then the absorbance at 562 nm was recorded. The following equation was used to determine chelating activity with B₅₆₂ as the absorbance at 562 nm of the control and A₅₆₂ as the absorbance of the hydrolysate samples at 562 nm.

Chelating activity (%) = [(B₅₆₂ − A₅₆₂)/B₅₆₂] × 100

All antioxidant activity assays were conducted in triplicate and averaged and the data was normalized by dividing the activity by the amount of protein in the OPCH and OPIH samples used.

Okra seed protein was purified into a concentrate which contained 72.1% protein and an isolate which contained 91.2% protein. The protein samples were hydrolyzed with three different proteases, pepsin, trypsin, and papain under optimal conditions at times from 0 to 360 min (15 hrs.). Figure 1 shows the results of the DH for each hydrolysate. For pepsin hydrolysis, (Figure 1(a)) the hydrolysis rate increased rapidly in a linearly fashion until 120 min (DH of 29.28%), then there was a slow increase in DH reaching a maximum of 35.20% at 360 min for OPI. For OPC, a similar trend was observed with a DH of 19.89% at 120 min and a maximum of 26.28% at 360 min. For OPI, 83% of the total hydrolysis was completed at 120 min and for the OPC, 76.5% of the total hydrolysis was completed by 120 min. For hydrolysis with trypsin (Figure 1(c)), at 120 min the DH of OPI was 7.68% (73% of the total hydrolysis) and the DH for OPC was 4.85% (75% of the total hydrolysis). The maximum DH values for OPI and OPC were 10.53% and 6.47% respectively. For hydrolysis with papain (Figure 1(c)), at 180 min the DH of OPI was 32.84% (93% of the total hydrolysis) and the DH for OPC was 25.86% (90.5% of the total hydrolysis). The maximum DH values for OPI and OPC were 35.21% and 28.59% respectively. Hydrolysis with pepsin and papain were similar with DH values for OPI at 35%. The least effective protease

was trypsin. We are unclear why the DH for the OPC was consistently lower than that of the isolate but we believe it may be due to inhibitory compounds present in the OPC which were removed during the preparation of the OPI.

Ktari et al. [4] investigated the hydrolysis of cuttlefish protein with a variety of proteases including trypsin, chymotrypsin and proteases from various Bacillus species. For each protease, there was rapid hydrolysis up to 200 min, but slowed to negligible increases in DH after that. One of the least effective protease was trypsin which reached only a 4% DH. Yet in a study by Fan et al. [3] who hydrolyzed tilapia protein with various enzymes including pepsin, trypsin and papain, the trypsin hydrolysates showed the highest DH values. The DH values in this study [3] for pepsin, trypsin and papain were 5.3%, 15.1% and 9.5% respectively. Pena-Ramos and Xiong [5] showed that the DH of soy protein isolate with papain was almost double that of the pepsin hydrolysate. These studies indicating that the DH value is related to the specificity of the protease and will vary with each protein substrate used.

The radical scavenging activity (RSA) of okra protein concentrate hydrolysate (OPCH) and okra protein isolate hydrolysate (OPIH) using the DPPH assay over hydrolysis time for each protease is shown in Figure 2. On average the RSA

of hydrolysates prepared using papain and pepsin were the highest in both the OPCH and OPIH. This may be due to the DH of the hydrolysates, which were similar in both papain and pepsin and the lowest with trypsin. Although on average, the DH of the trypsin hydrolysates were about 3.5 times lower than those of papain and pepsin, the RSA values for trypsin are approximately half of the values for papain and trypsin initially and 1.5 times lower with longer hydrolysis times. Therefore, the peptides released during the trypsin hydrolysis have higher RSA activities than those released with pepsin and papain. In general, there is a slight increase in RSA activity with an increase in hydrolysis time, which is most noticeable in the trypsin hydrolysates. On average, considering the data was normalized for the protein concentration, there are little differences between the RSA values of the isolate and concentrate hydrolysates even though the concentrates had lower DH values. The only exception is the initial RSA values for trypsin hydrolysates which are lower in the concentrate samples compared to the isolates. Hydrolysate samples after times 120 - 180 min show RSA activities that are similar. For each protease, at times greater than 120 - 180 min, there were negligible increases in DH, therefore as mentioned above, the RSA activity of the hydrolysates is related to the DH values.

The reducing power (RP) of OPCH and OPIH over hydrolysis time for each protease is shown in Figure 3. Overall, the RP of the pepsin hydrolysates were higher than for the papain and trypsin hydrolysates. Trypsin RP values were similar to the values for papain, which is not expected since the DH for trypsin was approximately 3.5 times lower than for papain. Again, as stated above for RSA activity, the peptides released by trypsin have higher RP activities than those released by papain. In general, there is an increase in RP activity with an increase in time until 60 min for OPIH and 120 min for OPCH which is expected since at times after these, there were negligible increases in DH in the hydrolysates. As stated above for RSA, the RP activity of hydrolysates increases with DH. The RP values of the OPIH and OPCH are similar at the higher DH values. The only differences in RP values are at times less than 120 min, where the RP values for the concentrates are lower, presumably due to the lower DH values of these samples compared to the isolates.

The metal chelating activity (MCA) of the OPCH and OPIH samples is shown in Figure 4. In Figure 4(b), for OPCH, the MCA values for trypsin are consistently lower than those for papain and pepsin. This may be due to the lower DH in the trypsin hydrolysates. Additionally, the MCA values for trypsin are similar over the time period while the MCA values for papain and pepsin show a slight increase up to 30 min, which is related to the DH. The MCA values in Figure 4(a) for OPIH show a different trend, the pepsin samples have higher MCA values compared to the papain and trypsin values, which are similar to each other. This is not expected considering the results for OPCH and we do not have an explanation. In general the MCA values for pepsin increase up to 120 min, then remain steady which is related to the DH. There are slight increases in the MCA

values for papain and trypsin, but not as dramatic as seen for pepsin. In general the MCA values for OPIH are much higher than those for OPCH. This may be due to the difference in the ash content between the two. The ash in the OPC is 9.4% and 2.6% in the OPI. With a higher amount of minerals in the OPC sample, the lower MCA values for the OPCH samples are due to the excess minerals. In general, the carboxyl and amino groups in proteins and the acidic and basic amino acids groups in amino acids play important roles in metal chelation and an increase in DH generally results in an increase in MCA activity due to the increase number of carboxyl and amino groups [7] .

Jamdar et al. [6] reported that peanut protein hydrolysate (using alcalase) showed that with an increase in DH, there was an increase in RSC, MCA, and ACE inhibitory activity while RP decreased. Ajibola et al. [7] , showed the effect of peptide size on the antioxidant properties of yam bean seed treated with alcalase. There was an increase in RSA and RP with samples containing peptides less

than 3 kDa which is in agreement with results reported by Girgih et al. [8] . This might explain the trypsin samples that result in more active antioxidant activity than expected based on the low DH. Trypsin hydrolysis may have resulted in more peptides less than 3 kDa and free amino acids which have more RSA activity. But, they [7] did find that MCA was not influenced by the size of the hydrolysates in this yam protein system. They showed that small peptide size and high level of hydrophobicity are important for RP and DPPH activity.

In general, the DH is related to antioxidant properties with higher DH values resulting in higher antioxidant values. Fan et al. [3] hydrolyzed tilapia protein with pepsin, trypsin and papain and the trypsin hydrolysate showed the highest DH (15.1%) and antioxidant activity as measured with RSA, superoxide anion scavenging, hydrogen peroxide scavenging, and the hydroxyl radical scavenging assays. Hydrolysates obtained with papain (DH of 9.5%) showed higher antioxidant activity than those obtained from pepsin hydrolysis (DH of 5.3%) but both were lower than the trypsin hydrolysates. While the study by Fan et al. [3] showed that a high degree of hydrolysis is related to a high DH, this is not always true. Ktari et al. [4] hydrolyzed cuttlefish protein with a variety of enzymes. Chymosin hydrolysates had higher DH values than those produced by trypsin (about 2 times higher DH with chymosin) yet the RSA values of trypsin hydrolysates were significantly higher than chymosin produced hydrolysates, while the RP activity of both were similar. Ktari et al. [4] stated that peptides generated during hydrolysis by different proteases have different antioxidant capacities.

Therefore, the antioxidant activity of protein hydrolysates is related to the protease used which results in a variety of peptides of different molecular weights and hydrophobicities. Low molecular weight fractions and free amino acids in protein hydrolysates generally contain more antioxidant activities depending on the assay used to test the the activity. Generally low molecular weight peptides result in high RCA activity, but not always metal chelating activity [3] [7] . In this study, pepsin hydrolysates with a final DH of 35.2% showed higher RP and MCA than trypsin and papain hydrolysates.