Preparation of bioactive peptides with antidiabetic, antihypertensive, and antioxidant activities and identification of α‐glucosidase inhibitory peptides from soy protein

Abstract In this study, the peptides of soy protein obtained by enzymatic digestion with proteases were analyzed for their antidiabetic, antihypertensive, and antioxidant activities. Peptides prepared with alkaline proteinase (AP) exhibited the highest α‐glucosidase inhibitory activity compared with those from papain and trypsin digestion. AP hydrolysates also exhibited dipeptidyl peptidase IV (DPP‐IV) inhibitory, angiotensin‐converting enzyme (ACE) inhibitory, and antioxidant activities. Gastrointestinal digestion of peptides enhanced α‐glucosidase, DPP‐IV, and ACE inhibitory activities compared with AP hydrolysates. AP peptides showing highest α‐glucosidase inhibitory activity were purified by anion‐exchange and size‐exclusion chromatography， and identified using tandem MS. We found three novel α‐glucosidase inhibitory peptides with sequences LLPLPVLK, SWLRL, and WLRL with IC50 of 237.43 ± 0.52, 182.05 ± 0.74, and 162.29 ± 0.74 μmol/L, respectively. Therefore, peptides hydrolyzed from soy protein are promising natural ingredients for nutraceutical applications assisting in the management of diabetes.


| INTRODUC TI ON
Diabetes mellitus (DM) is a common chronic metabolic disorder characterized by dysregulation of blood glucose and has been considered as a major global health issue now and in the future. DM presents in two main forms: Type 1 insulin-dependent DM (T1DM) constitute 10% of all DM cases, and type 2 non-insulin-dependent DM (T2DM) constitute 90% of all DM cases. It is predicted that the number of patients with diabetes will increase from 350 million today to 592 million by 2035 (Lee et al., 2016). Management of T2DM normally involves lifestyle modification and pharmacologic therapy. Several drugs, which are approved to treat hyperglycemia in T2DM, are costly and have serious side effects. For example, glucagon-like peptide-1 (GLP-1) receptor agonists have been shown to increase the risk of kidney injury (Qaseem et al., 2017). The serious side effect and the toxicity associated with some therapeutic drugs indicate the demand for diet-derived antidiabetic agents or approaches that are considered natural and safe. Enzymes associated with the regulation of postprandial hyperglycemia, such as α-glucosidase, α-amylase, and dipeptidyl peptidase IV (DPP-IV), have been recognized as therapeutic target. α-glucosidase is a membrane-bound enzyme located in the epithelium of the small intestine, which catalyses the cleavage of glucose from disaccharides (Matsui, Yoshimoto, Osajima, Oki, & Osajima, 1996). Inhibition of this enzyme has been recognized as an effective approach for lowering serum glucose level (Johnson, Lucius, Meyer, & Gonzalez De Mejia, 2011). α-glucosidase, which catalyses the reaction from dextrin to glucose in the small intestine, plays a key role in the digestion and absorption of carbohydrates.
Thereby, α-glucosidase inhibitors such as acarbose, voglibose, and miglitol effectively diminish postprandial hyperglycemia by impeding the digestion of carbohydrates. However, these types of inhibitors have unpleasant and potentially serious gastrointestinal side effects which limit use. Other therapeutic targets include dipeptidyl peptidase IV (DPP-IV), which is known to cause the inactivation of incretin hormones GLP-1 and gastric inhibitory peptide (GIP) (Drucker, 2007). By inhibiting DPP-IV, it is possible to increase the half-life of GLP-1 and GIP resulting in enhanced insulin secretion (Nongonierma & FitzGerald, 2013). There is also growing evidence indicating that the development of diabetes is related to oxidative stress (Asmat, Abad, & Ismail, 2016). Therefore, it is significant for identifying diet-related strategies that could address these different pathologies.
Bioactive peptides have a wide range of functional properties, including antimicrobial, antihypertension, hypoglycemic activity, immunomodulation, and antioxidative functions. Due to their lower relative molecular weight, higher absorption, and bioavailability (Koopman et al., 2009), food-derived bioactive peptides have shown potential to serve as natural alternatives or complements to synthetic drugs. Many studies have demonstrated hypoglycemic effects of bioactive peptides from hemp seed protein, goby fish, and egg-yolk protein (Nasri et al., 2015;Ren et al., 2016;Zambrowicz et al., 2015).
The objective of this research was to prepare peptides from soy protein hydrolysates with enzymolysis and to screen bioactive peptides with high α-glucosidase inhibitory activities. Successive separation and purification were carried out, and the α-glucosidase inhibitory peptides were identified. We also measured the in vitro antioxidative activity and the angiotensin-converting enzyme (ACE) inhibitory activity of the soy protein hydrolysates. The potential of soy bioactive peptides as ingredients for nutraceutical or functional food formulation is discussed.

| Preparation of soy protein peptides
The protein content of the raw material was determined by using the Kjeldahl method. Preparation of the protein hydrolysates was performed in a temperature-and pH-controlled 500-mL reaction vessel equipped with a stirrer. Three grams of soy protein powder was dispersed in distilled water to obtain 2% protein slurry (w/v).
Subsequently, the protein slurry was heated to 100°C for 10 min for denaturation. After cooling down, 0.5 mol/L sodium hydroxide (NaOH) and 0.5 mol/L hydrochloric acid (HCl) aqueous solutions were used to adjust the pH of the slurry to the optimum conditions recommended by the manufacturers for each protease: AP (pH 9, 50°C), papain (pH 6.5, 60°C), and trypsin (pH 7, 37°C). The protease (6,000 U/g, on the basis of the protein content of slurry) was added to initiate the protein enzymatic hydrolysis. Temperature and pH values were kept constant until the degree of hydrolysis no longer changed. Then, the hydrolysates were heated in a boiling water bath for 10 min to inactive the protease. After removal of the precipitate by centrifugation (4,436 g, 10 min), it was collected supernatant and stored at −20°C until use. The degrees of hydrolysis (DH) values were measured by "pH-stat" method described by Adler-Nissen (1986).

| Molecular weight distribution
The molecular weight distributions of hydrolysates were evaluated by size-exclusion HPLC (Waters) on a TSKgel 2000 SWxl column (7.8 mm × 300 mm). One hundred milligram of hydrolysate powder was dissolved in 10 ml mobile phase (acetonitrile/water/ trifluoroacetic acid, 40/60/0.1), sonicated for 5 min, centrifuged at 8,000 rpm for 10 min, and then filtered through the microporous membrane. The injection quantity was 10 μl with the flow rate of 0.5 ml/min. The effluent was monitored by UV detector at 220 nm.
The α-glucosidase inhibitory rate of the hydrolysate was estimated in a 96-well microplate using a MR-96A microplate reader (Mindray) by monitoring the amount of 4-nitrophenol (PNP) released from 4-nitrophenol-α-D-glucopyranoside (PNPG). A dose (50 μl) of sodium phosphate buffer (pH 6.8, 0.2 M), 50 μL of PNPG (1 mg/ml) dissolved in sodium phosphate buffer, and 50 μl of the samples dissolved in the same buffer were mixed adequately in a 96-well microplate. Then, it was incubated for 5 min at 37°C, and the reaction started immediately followed by the addition of 10 μl of α-glucosidase solution (0.2 U/ml of the buffer). The plate was incubated for another 30 min at 37°C, and then, the reaction was terminated by the addition of 80 μl of 1 M Na 2 CO 3 solution. The absorbance of released product PNP was measured immediately at 405 nm. Sodium phosphate buffer and acarbose (10 mg/ml of the buffer) were used as the negative and positive control, respectively. The percentage of α-glucosidase inhibition was calculated as follows: where OD A is the absorbance of the supernatant in which the sample is replaced by the same amount of buffer; OD S is the absorbance of the supernatant with the tested sample; and OD B is the absorbance of the supernatant in which the tested sample and α-glucosidase solution are replaced by the same amount of buffer. IC 50 value was obtained and used to evaluate the α-glucosidase inhibition rate of the sample, which was calculated as the concentration of the tested sample required for α-glucosidase inhibition rate of 50%.

| Dipeptidyl peptidase IV (DPP-IV) inhibition assay
DPP-IV inhibition assay was according to the method from Harnedy et al with minor modifications (Harnedy, O'Keeffe, & Fitzgerald, 2015). The assay was performed in a 96-well microplate; 25 μl of test sample was preincubated with 25 μl of substrate gly-pro-p-nitroanilide (12 mM) at 37°C for 10 min, after added 50 μl of DPP-IV (0.02 U/ml of Tris-HCl buffer, pH 8.0); the mixture was incubated at 37°C for 30 min; and then, 100 μl acetic acid-sodium acetate (1 M) was added to terminate the reaction. The DPP-IV inhibition rate (%) was calculated as following: where OD A is the absorbance of the supernatant in which the tested sample was replaced by the same amount of buffer; OD S is the absorbance of the supernatant with the tested sample; OD B is the absorbance of the supernatant in which the tested sample and DPP-IV solution are replaced by the same amount of buffer; and OD N is the absorbance of the supernatant in which the DPP-IV solution is replaced by the same amount of buffer.

| ACE inhibition assay
The determination of ACE inhibitory activity was performed by using in vitro method described by Wu and Ding with slight modifications (Wu & Ding, 2002). This assay was based on the release of HA from HHL with ACE enzymolysis. For the assay, 200 μl ACE (0.1 U/ml 0.1 M borate buffer containing 0.3 M NaCl at pH 8.3) was added to the sample (100 μl), and the mixture was incubated at 37°C for 10 min. Then, 200 μl of HHL (5 mM in 0.1 M borate buffer containing 0.3 M NaCl at pH 8.3) was added into each well as substrate. The enzymatic reaction subsequently carried out at 37°C for 60 min. The reaction was terminated by adding 300 μl of 1 M HCl.
Borate buffer was used as the control. The final HA was extracted with 1.5 ml of ethyl acetate; after 90°C of incubation, the organic layer would be dry and dissolved in 2 ml of distilled H 2 O. The amount of HA produced was determined by an HPLC system. An aliquot of 50 μl from the reaction mixture was analyzed by an HPLC system equipped with Waters C18 column (ODS, 150 × 4.6 mm, 5 μm particle size) using an elution of formic acid (0.4%): methanol (3:7 v/v) at a constant flow rate of 0.5 ml/min, and HA was detected by UV detector at 228 nm. The ACE inhibition rate (%) was calculated as follows: where A is the chromatography peak area of the supernatant in which the tested sample is replaced by the same amount of buffer; C is the chromatography peak area of the supernatant with tested sample; and B is the chromatography peak area of the supernatant with no tested sample and ACE solution.

| Hydroxyl radical scavenging ability
The antioxidant activity of the soy protein peptides was determined by a hydroxyl radical scavenging assay involving in the Fenton reaction, according to Hanasaki (Hanasaki, Ogava, & Fukui, 1994). The samples were prepared into 0.1, 0.5, 2, 5, and 10 mg/ml. Then, 1 ml of 9 µmol/L FeSO 4 solution and 1 ml of salicylic acid-ethanol solution (9 µmol/L，50% ethanol solution) were added respectively and mixed well, with the addition of 1 ml of 8.8 µmol/L H 2 O 2 (0.03%) reaction initiated. The mixture was incubated in a 37℃ water bath for 30 min and then centrifuged at 2,896 g for 5 min to remove the precipitation.
The absorbance of the supernatant was measured at 510 nm. The hydroxyl radical clearance rate was calculated as follows: where A 1 is the absorbance of the sample supernatant, A 2 is the absorbance of the supernatant in which the salicylic acid-ethanol is replaced by the same amount of ethanol, and A 0 is the absorbance of the supernatant in which the sample is replaced by distilled water.
The ascorbic acid was used as a positive control.

| Ferric reducing antioxidant power (FRAP) activity
The FRAP assay was used to determine the antioxidative of the samples using the method described by Tsai et al with some modifications Tsai, Huang, & Mau, 2006) and to configure the sample solution with the concentration of 1, 2, 5, 10, and 20 mg/ml. Using 1 ml of sample mixed with 2.5 ml of 0.2 mol/L sodium phosphate buffer (pH 6.6) and 2.5 ml of 1% potassium ferricyanide solution incubated at 50℃ for 30 min, add 2.5 ml of 10% trichloroacetic acid solution, centrifuged at 1,042 g for 10 min, and then collected 2.5 ml of supernatant mixed with 2.5 ml of distilled water and 0.5 ml of 0.1% ferric chloride solution. The absorbance at 700 nm was measured after 10 min by a 754 PC spectrophotometer (Shanghai, China). FeSO 4 aqueous solution was used as positive control.

| In vitro digestion simulation
According to the in vitro harmonized protocol described by Minekus (Minekus et al. (2014), the soy peptides were diluted to 10 mg/ml with deionized water and adjusted to pH 2.0 with 1 M HCL. Then, hydrolysis was started by adding the pepsin (2.5 U/mg substrate) from porcine gastric mucosa, the mixture was incubated for 2 hr at 37°C in a water bath with stirring, and pH value was adjusted to 7.2 with 1 M NaOH solution. To simulate the digestion of small intestine, pancreatin (E: S = 4:100) from porcine pancreas was added in the mixture to start the hydrolysis sequentially. The mixture was incubated at 37°C for 2 hr, and the pH value was kept to 7.2 with 1 M NaOH. After the incubation, the digestion process was terminated after enzyme denaturation through increasing the temperature to 95°C for 20 min. The soy peptides were separated from the reaction mixtures after digestion by centrifugation at 8,000 rpm for 10 min.
The inhibitory activities of ACE, α-glucosidase, and DPP-IV of the supernatants were determined by the methods mentioned above.

| Gel filtration chromatography (G-15)
Three microliters of fraction from ion-exchange chromatography (10 mg/ml) was loaded into a chromatography column (6 mm × 60 cm) packed with the pretreated gel Sephadex G-15 (Beijing Solarbio Science & Technology Co., Ltd.). Then, the sample eluted using distilled water with the flow rate of 0.5 ml/min. The peptide elution was monitored by measurement of the absorbance at 220 nm, and fractions were collected. Separation process was repeated multiple times, and the α-glucosidase inhibitory activities of collected fractions were analyzed. The fraction with the αglucosidase inhibitory activity marked by H 1-1 was concentrated by rotary evaporation and subjected to the subsequent analysis and identification procedure.

| Identification of peptide sequence (LC-MS/ MS)
The

| Peptide synthesis
Synthetic peptides were provided by Qingdao China Peptides Co., Ltd. using the conventional Fmoc solid-phase synthesis method.
Synthetic peptides were purified and determined by HPLC-MS analysis with the purification rate over than 98%.

| Statistical analysis
The data were expressed as the mean of three replicates and standard deviation (SD). Statistical calculation was performed by ANOVA with IBM SPSS statistics, version 25 (IBM Inc.). Differences were considered to be significant at p < 0.05.

| Preparation of peptides with α-glucosidase inhibitory activity
To prepare the antidiabetic bioactive peptides, soy protein isolate (protein content 86.62%) was hydrolyzed with AP, papain, and trypsin, respectively, and α-glucosidase inhibitory activity of hydrolysates was analyzed.
While trypsin cleaves solely C-terminal to Arg and Lys, papain prefers to cleave the peptides bonds between the carboxylic acid group of Lys or Arg and adjacent amino acid residue. The highest α-glucosidase inhibition rate (53.79 ± 3.24%) of the hydrolysate released by AP was derived at the DH value of 36.84 ± 1.20%. Figure 1 also shows that the DH value positively affected the inhibition rates of the hydrolysates. Table 1 shows the results of the molecular weight distribution.
There were 95.39% of the hydrolysates with molecular weight lower than 2,000 Da and 69.76% with molecular weight lower than 500 Da.

| Inhibitory activities of soy protein peptides against α-glucosidase and DPP-IV
The soy protein peptides obtained by AP hydrolysis were freezedried after the removal of insoluble precipitation. Inhibitory activities against α-glucosidase and DPP-IV of soy protein peptides were measured. α-Glucosidase is a membrane-bound enzyme located in the epithelium of the small intestine that catalyses the cleavage of glucose from disaccharides (Matsui et al., 1996). Inhibition of this enzyme has been recognized as an effective approach to reduce the serum glucose level (Johnson et al., 2011). Since hydrolysates are crude mixtures of proteins, the α-glucosidase inhibitory activity of acarbose (IC 50 = 0.52 ± 0.05 mg/ml) was higher than the hydrolysates that the IC 50 of α-glucosidase inhibitory activity was 4.94 ± 0.07 mg/ml.

DPP-IV is known for its inactivation of incretin hormones GLP-1
and GIP (Drucker, 2007). By inhibiting DPP-IV, it is possible to increase the half-life of GLP-1 and GIP resulting in enhanced insulin secretion (Nongonierma & FitzGerald, 2013). The same preparation had high DPP-IV inhibition activity with the IC 50 value of 2.73 ± 0.08 mg/ml.
The soy peptides we screened contained small peptides with a molecular weight between 180 and 500 Da, that is, about 2-5 amino acids.

| Antioxidation and ACE inhibition activities of soy protein hydrolysates
Since oxidative stress also affects the development of diabetes, the antioxidative activities of hydrolysate were evaluated. The soy hydrolysates obtained by AP exhibited hydroxyl radical scavenging ability (IC 50 = 5.30 ± 0.15 mg/ml), while it was weaker than that of vitamin C (IC 50 = 0.54 ± 0.04 mg/ml). As shown in Figure Wu and Ding (2002). Therefore, soy protein hydrolysates showed the potential to be a multifunction nutraceutical agent. The structure of the peptides may be closely related to ACE inhibitory effects. It is reported that the position of a hydrophobic amino acid at the C-terminal end is associated with ACE inhibitory activity (Escudero, Toldra, Sentandreu, Nishimura, & Arihara, 2012).
Soy protein belongs to plant protein and contains more hydrophobic amino acids (Zhang et al., 2017).

| Simulated gastrointestinal digestion
Before and after in vitro simulated digestion, the inhibitory activities against α-glucosidase, DPP-IV, and ACE of soy protein peptides were measured, and the results are shown in Figure 3. Before digestion, soy peptides with a concentration of 10.0 mg/ml indicated that the inhibition rate was 29.04 ± 1.49%, 40.85 ± 0.82%, and 20.31 ± 2.37% against α-glucosidase, DPP-IV, and ACE, respectively.
After gastric phase digestion, α-glucosidase inhibitory activity of the peptides from whey protein after the digestion (Tavares et al., 2011).
Although we do not know the reasons leading to different results, the amino acid sequence of peptides and inhibitory mechanism of peptides against ACE should be considered.
In general, this result demonstrated that the peptides could endure the whole digestive environment and still maintain or enhance their bioactivity.

| Separation and purification
In order to purify peptides with α-glucosidase inhibitory activity, successive chromatographic separations were performed with anion-exchange resin DEAE-52 and Sephadex-15 gel. As shown in  Figure 4. After the H 1 was washed off by distilled water, it was found that H 1 was only the effective inhibitor on the α-glucosidase with inhibition rate 87.10 ± 2.70%, and the inhibitory activity was higher than the original hydrolysates. All other fractions were not able to inhibit α-glucosidase.
We collected and concentrated H 1 fraction by rotary evaporation, then purified further by size-exclusion chromatography. As shown in Figure 5 inset, fractionation by size of the H 1 fraction revealed a relatively clear molecular weight distribution profiles with only two main peaks (H 1-1 , H 1-2 ). Figure 5 illustrates these two fractions both had α-glucosidase inhibition activity, and H 1-2 showed high α-glucosidase inhibition activity up to 95.35 ± 2.70% inhibitory rate, which was much higher than H 1-1 with 26.08 ± 2.12% inhibitory rate.

| Identification of purified peptides
We determined molecular weight of the purified fraction H We obtained the complete amino acid sequences by amino acid composition analysis and manual calculation of the MS/MS spectrum. As shown in Figure 6, the first octapeptide was identified as LLPLPVLK, while the second peptide consisted of four amino acids with the sequence of WLRL. Besides, peptides with five amino acid SWLRL and hexapeptide MLPVMR were also identified.
Compared with the previously reported hypoglycemic peptides, such as N159-1 (PFP separated from Aspergillus oryzae) (Kang, Yi, & Lee, 2013), albumin (KLPGF) (Yu, Yin, Zhao, Liu, & Chen, 2012), and egg white protein (RVPSLM) (Yu et al., 2011), there appear to be no primary structural homology to the four oligopeptides we found. It is not difficult to point out that hydrophobic amino acids were common in these hypoglycemic peptides. In particular, Leu and Pro showed outstanding contributions to these oligopeptides separated from soy protein (Sadri, Larki, & Kolahian, 2017).
It was reported that Pro and Leu in peptides were considered as vital amino acids playing inhibition effect against α-glucosidase separately or synergistically, which should be taken into account in the synthesis of novel α-glucosidase inhibitory peptides for antidiabetic treatment.

| Antidiabetic properties of synthetic peptides
The synthesized peptides have been confirmed the great inhibitory effect on α-glucosidase activity. The hexapeptides, MLPVMR, did not show inhibitory activity. However, other three peptides, LLPLPVLK, SWLRL, and WLRL, had great inhibitory influence on α-glucosidase with the IC 50 value of 237.43 ± 0.52, 182.05 ± 0.74, and 165.29 ± 0.74 μmol/L. It has been suggested that the inhibitory mechanism could be through the hydrophobic interactions of the polypeptide with the active site of the enzyme, as it has been reported for other inhibitors (Bharatham, Bharatham, Park, & Lee, 2008). The peptides we found are rich in hydrophobic amino acids, especially the Leu, which might play an important role to inhibit α-glucosidase.

| CON CLUS IONS
In summary, the peptides possessing high α-glucosidase inhibitory effect were successfully prepared from soy protein treated via an AP process at the DH value of 36.84 ± 1.20% in the present study. The soy protein peptides also exhibited antioxidative activity and inhibition activities against DPP-IV and ACE. The peptides with α-glucosidase inhibition activity were separated with anionexchange resin, Sephadex G-15 gel, which was determined using LC-MS/MS sequentially. The IC 50 α-glucosidase inhibitory activity values of LLPLPVLK， SWLRL, and WLRL were 237.43 ± 0.52, 182.05 ± 0.74, and 165.29 ± 0.74 μmol/L, respectively. Compared with other α-glucosidase inhibitory peptides, these three peptides we found showed different structural homology from them, as well as lower IC 50 value than acarbose. Therefore, this research can be referenced for the development of novel antidiabetic peptide nutraceuticals. Also, in order to validate the findings of the present study, further research in cell culture and in vivo should be needed.

ACK N OWLED G EM ENTS
This study is supported by The National Key Technologies R&D Program of China (No. 2017YFC160090-02).

CO N FLI C T O F I NTE R E S T
The authors declare that they do not have any conflict of interests.

E TH I C A L S TATEM ENT
Human testing and animal testing were unnecessary in this study.