In vitro antithrombotic activities of peanut protein hydrolysates

Accepted Manuscript In vitro antithrombotic activities of peanut protein hydrolysates Shao Bing Zhang PII: DOI: Reference:

S0308-8146(16)30106-6 http://dx.doi.org/10.1016/j.foodchem.2016.01.108 FOCH 18682

To appear in:

Food Chemistry

Received Date: Revised Date: Accepted Date:

6 September 2015 14 December 2015 26 January 2016

Please cite this article as: Zhang, S.B., In vitro antithrombotic activities of peanut protein hydrolysates, Food Chemistry (2016), doi: http://dx.doi.org/10.1016/j.foodchem.2016.01.108

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

In vitro antithrombotic activities of peanut protein hydrolysates

Shao Bing Zhang School of Food Science and Technology, Henan University of Technology, Zhengzhou-450001 Henan Province, People's Republic of China

School of Food Science and Technology, Henan University of Technology, Zhengzhou-450001, Henan Province, People's Republic of China. Tel/Fax +86-371-67758022. E-mail: [email protected]

ABSTRACT The antithrombotic activities of peanut protein hydrolysates were investigated using a microplates assay. When peanut proteins were hydrolyzed to a limited extent by various enzymes, their thrombin inhibitory abilities were significantly enhanced. However, the resultant hydrolysates showed significantly different activities even at the same degrees of hydrolysis. The hydrolysates generated by Alcalase 2.4L displayed the best antithrombotic activities and the hydrolysis process was further optimized by response surface methodology. The antithrombotic activities were increased to 86% based on a protein concentration of 50 mg/ml under the optimal conditions: pH 8.5, enzyme concentration of 5000 IU/g of peanut proteins, and 2 h hydrolysis time at 50°C. The Alcalase 2.4L crude hydrolysates were then fractionated

successively

by

preparative

and

semi-preparative

reverse-phase

high-performance liquid chromatography (RP-HPLC). The peptide fraction collected inhibited thrombin-catalyzed coagulation of fibrinogen completely at a concentration of 0.4 mg/ml, with an antithrombotic activity close to that of heparin at quite a low concentration (0.2 mg/ml). This peptide fraction was further analyzed by online reverse-phase ultra-performance liquid chromatography (RP-UPLC) coupled to matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS), and three new peptides were identified as Ser-Trp-Ala-Gln-Leu, Gly-Asn-His-Glu-Ala-Gly-Glu and Cys-Phe-Asn-Glu-Tyr-Glu, respectively. This research provided an effective way to produce antithrombotic peptides from peanut proteins, and also helped to elucidate the structure-function relationships of peanut peptides.

2

Keywords: Antithrombotic activities; Peanut; Protein hydrolysates; Enzymatic hydrolysis 1. Introduction Peanuts are one of the most important oilseeds in the world. In China, the annual production of peanuts exceeds 14 million tons. About 50% of this supply is processed as edible oil by pressing or solvent extraction. Although defatted peanut flour consists of 47-55% protein, it is underutilized compared to soybean as a source of protein (Yu, Aahmedna, & Goktepe, 2007). It is probably because peanut protein products prepared from defatted peanut flour generally show poorer functionalities than those of soybean protein products. Therefore, further modification of peanut proteins to exploit their bioactivities could enhance the utilization of peanut proteins. Enzymatic hydrolysis of peanut proteins to obtain bioactive peptides has been widely reported. These studies mainly focussed on producing antioxidant (Hwang, Shyu, Wang, & Hsu, 2010; Jamdar et al., 2010; Ji, Sun, Zhao, Xiong, & Sun, 2014) and antihypertensive (Huang, Lu, Zhang, Sun, & Song, 2013; Jamdar et al., 2010; White, Sanders, & Davis, 2014) peptides. It is recognized that blood clots may cause heart problems if they lodge in the coronary artery. The blood clotting process relies heavily on the formation of a fibrin clot which results from the interaction between thrombin and fibrinogen (Scheraga, 2004). If thrombin activity is inhibited the fibrin clot will not form. Researchers have found antithrombotic activity from peptides originating from animals, such as casein (Bouhallab & Touzé, 1995; Ronquillo et al., 2012), pork meat (Shimizu et al., 2009), goby muscle (Nasri et al., 2012), Scolopendra subspinipes mutilans (Kong, Huang, Shao, Li, & Wei, 2013), and scorpion proteins (Ren et al., 2014). The amino acid sequences of some active peptides were identified

3

with proposed structure-function relationships. Compared to peptides from plants, limited information on animal peptides is available. Lee and Kim (2005) purified and reported two new peptides from soy protein hydrolysate that could inhibit platelet aggregation. We previously reported that crude rapeseed peptides obtained by aqueous enzymatic extraction showed significant thrombin inhibitory abilities (Zhang, Wang, & Xu, 2008). The rapeseed peptides, however, have not been purified for further research. To our knowledge,with respect to peanuts, little research has focussed on the enzymatic preparation of antithrombotic peptides and their structure identification. In the present study, peanut protein isolates (PPI) were produced from the defatted peanut meal and then the conditions for their enzymatic hydrolysis were optimized by response surface methodology (RSM). In addition, the highly antithrombotic peptides were purified using consecutive chromatography and their primary structure was identified by reverse-phase matrix-assisted

ultra-performance laser

liquid

chromatography

desorption/ionization

(RP-UPLC)

time-of-flight

mass

coupled

to

spectrometry

(MALDI-TOF-MS). 2. Materials and methods 2.1. Materials Deshelled peanut seeds were purchased from a local market. The peanuts contained 48.6% oil and 27.3% protein (N×5.46), on a dry basis. Enzymes used in this research included: Protex 6L (an alkaline serine endopeptidase from Bacillus licheniformis, recommended pH 8.0, temperature 50°C, with enzyme activity of 265,882 IU/g) was provided by Genencor Division of Danisco (Wuxi, China). Alcalase 2.4L (an alkaline serine endopeptidase from

4

Bacillus licheniformis, recommended pH 8.0, temperature 50°C, with enzyme activity of 280,800 IU/g) was purchased from Novo-Nodisk A/S (Tianjin, China). Mifong®2709 (an alkaline endopeptidase from Bacillus licheniformis, recommended pH 9.0, temperature 50°C, with enzyme activity of 112,000 IU/g) was purchased from Donghua-Qiangsheng Biotechnology Co., Ltd (Beijing, China). The peanut protein isolates (PPI) containing 85.4% protein (N×5.46) was obtained according to the method of Zhao, Liu, Zhao, Ren, and Yang (2011). Thrombin and fibrinogen were purchased from Sigma Chemical Co. (St. Louis, MO, USA). All other chemicals used in the experiments were of analytical grade.

2.2. Experimental design for optimizing the enzyme hydrolysis of PPI The enzyme treatment conditions were optimized according to the Box-Benken design (Myers & Montgomery, 1995). The effects of these three independent parameters (pH, hydrolysis time, and enzyme concentration) at three variations were evaluated. The parameters and their levels were chosen by preliminary experiments. The ability of inhibiting the coagulation of fibrinogen was determined. Design-Expert version 6.0.10 (Stat-Ease Inc., Minneapolis, MN, USA) software was used for the estimation of the coefficients and regression analysis of the data. The actual and coded levels of three variables are identified in Table 1.

2.3. Enzyme treatment PPI (5 g) were placed into 250 ml plastic centrifuge tubes and diluted with 100 ml distilled water. The centrifuge tubes were placed in a water bath, which was equipped with a stirring

5

rod for constant agitation. The enzymes were then added to commence the reaction at their recommended pH and temperature. After attaining the desirable degrees of hydrolysis (DH), the solution was immediately heated to 85°C and the temperature kept constant for 20 min, followed by centrifugation at 1819×g (3000 rpm) for 10 min. The supernatant was then sampled for its ability to inhibit the thrombin-catalyzed coagulation. The control was treated under the same conditions as the PPI solution at 50°C for 30 min, but without enzyme addition. The hydrolysis reaction was carried out in accordance with the RSM experimental design once the optimum enzyme was identified. The pH, hydrolysis time, and enzyme concentration were varied from 8.0 to 9.0, from 5000 to 7000 IU/g of PPI, and from 1.0 to 2.0 h, respectively.

2.4. Degree of hydrolysis (DH) determination for proteins The DH of proteins was determined by assay according to the pH-stat method (Adler-Nissen, 1986). The DH was analyzed based on the volume of base (B) added to maintain constant pH and using Eq. 1. DH = BNB

1 1 1 ×100% α MP ht

(1)

where B is the consumption (ml) base, NB is the normality of the base, MP is the protein mass (g) in the sample,

α is the average dissociation degree of the terminal amino peptides

(at pH 9 and 50°C, estimated to be 0.99), and ht is the total number of peptide bonds per gram of protein (taken as 7.8 meq/g).

6

2.5. Evaluation of thrombin inhibition activity In order to evaluate the ability of samples to inhibit the coagulation of fibrinogen, each was assayed according to the method of Yang, Wang, and Xu (2007) with some modifications. A microplate reader was set at 37°C. The peptide sample, thrombin, and fibrinogen were independently dissolved in 0.05 M Tris-HCl buffer (pH 7.2) containing 0.12 mM NaCl. A 0.1% fibrinogen solution (140 µL) and 40 µL sample solution were first injected into the plate wells and mixed, then the absorbance at 405 nm was recorded (blank reading). Next, 10µL thrombin solution (12 IU/ml) was added into the wells to commence the reaction of thrombin-catalyzed fibrinogen coagulation. After 10 min of incubation, the absorbance of the sample was recorded again. Instead of adding the sample solution, the control treatment contained 40 µL of Tris-HCl buffer (pH 7.2, 0.05 M) for the control blank and control absorbance value. Triplicate determinations were carried out at each concentration of the hydrolysates. The inhibitory abilities were calculated according to Eq. 2. inhibitory effect, % =

[C - CB]-[S - SB] ×100% [C - CB]

(2)

where S, SB, C, and CB represent the absorbance of the sample, the sample blank, the control, and the control blank, respectively.

2.6. Purification of peanut peptides by reverse-phase high performance liquid chromatography (RP-HPLC)

The crude peanut hydrolysates obtained under optimal Alcalase 2.4L hydrolysis conditions were first purified using preparative RP-HPLC on a Hedern ODS C18 column (100×250 mm). Based on preliminary tests, optimal separation conditions were: 1 ml (20 mg/ml) of peptide 7

solution loaded onto the column, a mobile phase of 35% (v/v) acetonitrile, a flow rate of 10 ml/min, and detection at 220 nm. The main fractionation peaks were collected and lyophilized for evaluation of inhibitory effect. The fraction showing the highest antithrombotic activity obtained by preparative RP-HPLC was further purified using semi-preparative RP-HPLC on a Yinluo ODS C18 column (10×250 mm). A peptide solution (8 mg/ml) of 100 µL was loaded onto the column. The elution after optimization was performed using a mobile phase (A) consisting of 5% (v/v) acetonitrile containing 0.05% (v/v) trifluoroacetic acid (TFA), and another mobile phase (B) consisting of 80% (v/v) acetonitrile containing 0.05% (v/v) TFA. Gradient elution was carried out as follows: 0-5 min, 100-80% A; 5-20 min, 80-50% A; 20-24 min, 50-0% A; 24-26 min, 0% A; 26-30 min, 0-100% A; 30-40 min, 100% A with a flow speed of 1.5 ml/min detected at 220 nm. Again, the major peaks were collected, lyophilized, and analyzed for inhibitory effect. The most active fraction was further analysed to determine amino acid sequence.

2.7. Determination of the amino acid sequence of the purified peptides by RP-UPLC coupled to MALDI-TOF-MS

Analysis of the purified peptides by online RP-UPLC coupled to MALDI-TOF-MS was carried out according to Yang et al. (2012) with some modifications. The peptide fraction (0.5 µL) was subjected to UPLC for peptide isolation on an Aquity UPLC BEH C18 column (2.1×150 mm) (Waters Corporation, Milford, MA). Gradient elution was carried out from 0.1% formic acid to 40% acetonitrile containing 0.1% formic acid (20 min) with a flow speed

8

of 0.3 ml/min. The sequence of selected peptides was identified by a Waters Synapt mass quadrupole time-of-flight mass spectrometer (Waters Corporation, Milford, MA). The spectra were recorded over a range of 50 to 1000 mass/charge (m/z). The eluent of UPLC was introduced to the mass spectrometer by an electrospray ion source. Peptide fragmentation was obtained by collision-induced dissociation (CID) with air as the collision gas. The collision energies were set at 6 eV and 20 eV, respectively. Data collection and analysis were made using Mass Lynx software version 4.1 (Micromass UK Ltd., Wythenshawe, Manchester, UK).

2.8. Statistical analysis

Experimental results were reported as the means and standard deviations for at least two replicates. A second-order polynomial was tested using an F-test at 95% confidence level in the Box-Benken design. The data was compared by ANOVA using Duncan’s multiple range tests. Significant differences were defined when p < 0.05.

3. Results and Discussion Three reversible phases were included in the conversion of fibrinogen to fibrin catalyzed by thrombin. Firstly, fibrinopeptides A (FpA) and B (FpB) were released from fibrinogen by thrombin hydrolysis to produce fibrin monomer. Secondly, intermediate polymers were formed by fibrin monomers through noncovalent interactions. Thirdly, the fibrin clot was formed due to the aggregation of intermediate polymers (Scheraga, 2004). The thrombin inhibitory activity of peanut protein and protein hydrolysates was analyzed by assaying the

9

decrease or total absence of turbidity (Sabbione, Scilingo, & Añón, 2015). As shown in Fig. 1, when PPI was hydrolyzed to a limited extent by alkaline proteases, the resultant hydrolysates generally exhibited higher thrombin inhibitory activities than PPI (only 3.86±0.15%). This suggested that some active peptide sequences originally buried in the protein molecules were released upon hydrolysis. However, the antithrombotic activities of hydrolysates prepared by various proteases differed significantly even at the same DH. Since the same DH generally indicates a similar molecular weight distribution of hydrolysates, these results demonstrated that the antithrombotic activities might not solely rely on the reduction of protein molecular weight. In fact, enzymes act on different sites of a protein due to selectivity, resulting in crude peptides with different structures and activities. Alcalase 2.4L hydrolysates generally had stronger inhibitory effects compared to other hydrolysates, probably due to the special selectivity of Alcalase 2.4L during hydrolysis of PPI, which is in agreement with the results of a study into scorpion protein by Ren et al. (2014). Fig. 1 shows the highest inhibitory activity of Alcalase 2.4L hydrolysates was 68% when the DH was 6%, and it was then significantly reduced with a further increase in DH value. Similar change trends were also found in the other protein hydrolysates. Excessive hydrolysis might reduce the content of bioactive groups for inhibiting thrombin. However, these results were obtained only under limited hydrolysis conditions (the DH scope was not big enough) for screening the enzymes. PPI should be hydrolyzed by the optimal enzyme under more extensive conditions to further investigate their thrombin inhibitory abilities. Alcalase 2.4L was selected in the following optimization process due to its good performance in limited hydrolysis. Table 1 presents the experimental conditions and the

10

corresponding results from the RSM optimization. The inhibitory values (y) ranged from 42.90 to 86.72% based on a protein concentration of 50 mg/ml. The maximum inhibitory effect was achieved at pH 8.5, 5000 IU/g of PPI enzyme concentration, and 2 h hydrolysis time. A second-order polynomial was tested using an F-test at 95% confidence level. The second-order polynomial below satisfactorily explained the inhibitory effect. Y is the predicted value for inhibitory effect (%) and X1, X2, and X3 are the coded variables as showed in Table 1. This predictive equation had a good fit with an R2 = 0.9088. Y (%) = 56.70+8.39 X1+8.22 X2-9.73 X3-5.35 X12+1.05 X22+3.74 X32+5.71 X1X2-8.73 X1X3 -4.26 X2X3

(3)

ANOVA (data not shown) indicated that the regression model was significant with a level of p < 0.01 and it had insignificant (p > 0.05) lack of fit. The inhibitory effects were significantly influenced by pH, enzyme concentration, and hydrolysis time (their p values were 0.0057, 0.0063 and 0.0026, respectively). Fig. 2A indicates that when enzyme concentration was used at 6000 IU/g of PPI, hydrolysis time did not markedly affect thrombin inhibitory abilities at low pH values, whereas at higher pH the inhibitory abilities were significantly increased with prolonged reaction time. The results depicted in Fig. 2B show that when hydrolysis time was fixed at 1.5 h, enzymatic concentration had insignificant effects on the inhibitory activities at low pH values. However, as the pH values were increased, the inhibitory activities were significantly lowered at higher enzyme concentrations. Generally, to hydrolyze the proteins to a large degree, we may choose high enzyme concentration combined with short hydrolysis time or low enzyme concentration combined with long hydrolysis time. In this work the latter conditions were desirable to

11

obtain protein hydrolysates with high antithrombotic activities. It was assumed that low enzyme concentration combined with long hydrolysis time would produce a greater amount of small peptides in the products, because the protein hydrolysis has been recommended as one-by-one mode (Adler-Nissen, 1986). Enough hydrolysis time would allow the extensive hydrolysis of protein molecules. The higher antithrombotic activities may be attributed to the existence of a larger amount of small peptides. The thrombin inhibitory effects of PPI hydrolysates were significantly enhanced from 68 to 86% based on the same protein concentration by RSM optimization, which was close to that of the crude rapeseed peptides but significantly weaker than that of heparin, a general antithrombotic drug (Zhang, Wang, & Xu, 2008). To achieve more potent antithrombotic peptide fractions, PPI hydrolysates were further purified by consecutive chromatography. During the first purification, PPI hydrolysates were fractionated by preparative RP-HPLC with three distinct fractions (P1, P2, and P3) obtained after optimization of the separation conditions (Fig. 3). As shown in Fig. 3B, fraction P2 showed higher antithrombotic activities than P1 and P3 at all tested concentrations. The highest inhibitory value was about 95% at a concentration of 40 mg/ml. The P2 potent fraction was then loaded onto a semi-preparative C18 HPLC column. It was further separated into 5 different peaks (Fig. 4A). The other four fractions showed excellent antithrombotic activities at a low concentration of 0.4 mg/ml, except for fraction R1 (Fig. 4B). Fraction R3 inhibited thrombin-catalyzed coagulation of fibrinogen completely. As a plant-derived peptide fraction, it displayed better antithrombotic activity than some hydrolysates from animal origins, such as egg white protein hydrolysates (Yang, Wang, & Xu, 2007) and goby protein hydrolysates (Nasri et al., 2012). However, its

12

inhibitory ability is lower than that of a purified fraction from scorpion protein hydrolysates with an IC50 value of 0.012 mg/ml (Ren et al., 2014). Heparin exhibited a 100% thrombin inhibitory effect at the concentration of 0.2 mg/ml under the same assay conditions (Zhang, Wang, & Xu, 2008). The R3 isolated peanut peptide fraction showed only slightly lower antithrombotic activity (65%) than heparin at the same concentration (0.2 mg/ml). To identify the peptide structures, fraction R3 was analyzed by online RP-UPLC coupled to MALDI-TOF-MS in positive mode. The RP-UPLC spectrum of R3 is presented in Fig. 5A. The R3 fraction did not show a single peptide fraction, but contained three main peptides. The respective MS spectra of U1, U2 and U3 are shown in Fig. 5B, C and D. The amino acid sequences of these fractions were proposed by the Biolynx peptide sequencer according to their MS spectrums. The primary structures of the U1, U2 and U3 fractions were Ser-Trp-Ala-Gln-Leu, Gly-Asn-His-Glu-Ala-Gly-Glu and Cys-Phe-Asn-Glu-Tyr-Glu, respectively. Their molecular mass was 603, 712 and 803 Da, respectively. In general, the thrombin inhibitory action of peptides may be attributed to two different mechanisms: 1) their interaction with the active sites of thrombin via certain important amino acid residues, whereby the proteolysis of the fibrinogen was prevented; 2) their binding with the already formed fibrin monomers to prevent the polymerization of the latter. Laudano and Doolittle (1978) synthesized some short peptides starting with the sequence Gly-L-Pro-L-Arg, which corresponded to the N-terminal segment of the fibrin monomers. They found these peptides could effectively prevent the polymerization of fibrin monomers. It was further reported by Mao et al. (1987) that polypeptide hirudin had negatively charged C-terminal and hydrophobic N-terminal by analysis of its complete amino

13

acid sequence. These unique structures may incline to bind with thrombin, which generally carries a positive charge, and therefore play an active role in anticoagulant activity. More recently, Lee and Kim (2005) obtained two new peptides (Ser-Ser-Gly-Glu and Asp-Glu-Glu) that could inhibit platelet aggregation from soy protein hydrolysate. Ren et al. (2014) reported a novel anticoagulant peptide (Val-Glu-Pro-Val-Thr-Val-Asn-Pro-His-Glu) from scorpion protein hydrolysates. They believed that the high anticoagulant activity of this peptide was because it contained five hydrophobic amino acids (three Val and two Pro) and two negatively charged amino acids (two Glu). In the present study, fractions U2 and U3 contained two Glu and fraction U1 contained one Gln. Gln is probably hydrolyzed and transformed into Glu. Therefore, we assume that plentiful Glu residues in the purified peanut peptides may contribute much to their excellent antithrombotic activities. However, the antithrombotic activities of these peptides should depend on not only the amino acid compositions, but also their unique sequences. The underlying antithrombotic mechanisms of peptides still require further studies.

4. Conclusions The hydrolysates from peanut proteins, utilizing Alcalase 2.4L, exhibited high antithrombotic activities. After separation and purification, the peptide fractions showed comparable activities to heparin. Investigating the potent peptide fraction revealed three novel antithrombotic peptides that were identified to have the following sequences of Ser-Trp-Ala-Gln-Leu,

Gly-Asn-His-Glu-Ala-Gly-Glu

and

Cys-Phe-Asn-Glu-Tyr-Glu.

Negatively charged amino acids in these peptides may influence their antithrombotic potency.

14

To our knowledge, this is the first report that shows the potent antithrombotic activity of peptides purified from peanut proteins. These results demonstrate that peanut peptides may potentially be utilized to develop functional foods for thrombosis prevention.

Acknowledgments This research has been financially supported by the Program for Science & Technology Innovation Talents in Universities of Henan Province (13HASTIT005) and National Natural Science Foundation of China (31171652).

References Adler-Nissen, J. (1986). Enzymatic hydrolysis of food proteins. Elsevier Applied Science Publishers, New York.

Bouhallab, S., & Touzé, C. (1995). Continuous hydrolysis of caseinomacropeptide in a membrane reactor: kinetic study and gram-scale production of antithrombotic peptides. Lait, 75:251-258.

Huang, J. N., Lu, X., Zhang, L. X., Sun, Q., & Song, G. H. (2013). Preparation of antihypertensive peptide from hydrolyzing peanut protein by trypsin covalently immobilized on chemically modified chitosan-coated Fe3O4 particles. Advance Journal of Food Science and Technology, 3:361-369.

Hwang, J. Y., Shyu, Y. S., Wang, Y. T., & Hsu, C. K. (2010). Antioxidative properties of protein hydrolysate from defatted peanut kernels treated with esperase. LWT- Food Science and Technology, 43:285-290.

Jamdar, S. N., Rajalakshmi, V., Pednekar, M. D., Juan, F., Yardi, V., & Sharma, A. (2010). 15

Influence of degree of hydrolysis on functional properties, antioxidant activity and ACE inhibitory activity of peanut protein hydrolysate. Food Chemistry, 121:178-184. Ji, N., Sun, C., Zhao, Y., Xiong, L., & Sun, Q. (2014). Purification and identification of antioxidant peptides from peanut protein isolate hydrolysates using UHR-Q-TOF mass spectrometer. Food Chemistry, 161:148-154. Kong, Y., Huang, S. L., Shao, Y., Li, S., & Wei J. F. (2013). Purification and characterization of a novel antithrombotic peptide from Scolopendra subspinipes mutilans. Journal of Ethnopharmacology, 145:182-186.

Laudano, A. P., & Doolittle, R. F. (1978). Synthetic peptides derivatives that bind to fibrinogen and prevent the polymerization of fibrin monomers. Proceedings of the National Academy of Sciences USA, 75 :3085-3089.

Lee, K. A., & Kim, S. H. (2005). SSGE and DEE, new peptides isolated from a soy protein hydrolysate that inhibit platelet aggregation. Food Chemistry, 90 :389–393. Mao, S. J., Yates, M. T., Blankenship, D. T., Cardin, A.D., Krstenansky, J. L., Lovenberg, W., et al. (1987). Rapid purification and revised amino-terminal sequence of hirudin: A specific thrombin inhibitor of the bloodsucking leech. Analytical Biochemistry, 161:514–518.

Myers, R. H., & Montgomery, D. C. (1995). Response surface methodology: process and product optimization using designed experiments. John Wiley & Sons, Inc., Chichester. Nasri, R., Amor, I. B., Bougatef, A., Arroume, N. N., Dhulster, P., Gargouri, J., et al. (2012). Anticoagulant activities of goby muscle protein hydrolysates. Food Chemistry, 133:835-841.

16

Rao, S., Sun, J., Liu, Y., Zeng, H., Su, Y., & Yang, Y. (2012). ACE inhibitory peptides and antioxidant peptides derived from in vitro digestion hydrolysate of hen egg white lysozyme. Food Chemistry, 135, 1245-1252. Ren, Y., Wu, H., Lai, F., Yang, M., Li, X., & Tang, Y. (2014). Isolation and identification of a novel anticoagulant peptide from enzymatic hydrolysates of scorpion (Buthus martensii Karsch) protein. Food Research International, 64:931-938.

Ronquillo, R. R., Guerrero, A. C., Nájera, A. F., Serrano, G. R., Ruiz, L. G, Grajeda, J. P. R., et al. (2012). Antithrombotic and angiotension-converting enzyme inhibitory properties of peptides released from bovine casein by Lactobacillus casei Shirota. International Dairy Journal, 26:147-154.

Sabbione, A. C., Scilingo, A., & Añón, M. C. (2015). Potential antithrombotic activity detected in amaranth proteins and its hydrolysates. LWT- Food Science and Technology, 60 :171-177.

Scheraga, H. A. (2004). The thrombin-fibrinogen interaction. Biophysical Chemistry, 112:117-130.

Shimizu, M., Sawashita, N., Morimatsu, F., Ichikawa, J., Taguchi, Y., Ijiri, Y., et al. (2009). Antithrombotic papain-hydrolyzed peptides isolated from pork meat. Thrombosis Research, 123 :753-757.

White, B. L., Sanders, T. H., & Davis, J. P. (2014) Potential ACE-inhibitory activity and nanoLC-MS/MS sequencing of peptides derived from aflatoxin contaminated peanut meal. LWT- Food Science and Technology, 56:537-542. Yang, W. G., Wang, Z., & Xu, S. Y. (2007). A new method for determination of

17

antithrombotic activity of egg white protein hydrolysate by microplate reader. Chinese Chem Letters, 18, 449-451.

Yu, J., Aahmedna, M., & Goktepe, I. (2007). Peanut protein concentrate: Production and functional properties as affected by processing. Food Chemistry, 103 :121–129. Zhang, S. B., Wang, Z., & Xu, S. Y. (2008). Antioxidant and antithrombotic activities of rapeseed peptides. Journal of the American Oil Chemists' Society, 85 : 521–527. Zhao, G., Liu, Y., Zhao, M., Ren, J., & Yang, B. (2011). Enzymatic hydrolysis and their effects on conformation and functional properties of peanut protein isolate. Food Chemistry, 127, 1438-1443.

18

Figure captions

Fig. 1 Antithrombotic activities of peanut protein hydrolysates at different degree of

hydrolysis (DH) obtained by various alkaline proteases. The control was the antithrombotic activity of peanut protein isolates. Results are expressed as means ± SD. Data marked with different letters indicate significant differences (p < 0.05).

Fig. 2 (A) Response surface for the inhibition effect on the thrombin-catalyzed coagulation

of fibrinogen as a function of pH and hydrolysis time with enzyme concentration of 6000 IU/g of PPI; (B) as a function of pH and enzyme concentration with hydrolysis time fixed at 1.5 h. The used protease was Alcalase 2.4L.

Fig. 3 (A) Elution profile of crude peanut protein hydrolysates separated by preparative

RP-HPLC (the arrow points to the most potent fraction); (B) antithrombotic activities of separated fractions (P1, P2, and P3). Results are expressed as means ± SD. Data marked with different letters indicate significant differences (p < 0.05).

Fig. 4 (A) Elution profile of peptide fraction P2 separated by semi-preparative RP-HPLC

(the arrow points to the most potent fraction); (B) antithrombotic activities of separated fractions (R1, R2, R3, R4 and R5) based on a concentration of 0.4 mg/mL. Results are expressed as means ± SD. Data marked with different letters indicate significant differences (p < 0.05).

Fig. 5 Elution profile of peptide fraction R3 separated by RP-UPLC (A) and mass spectrums

of the resultant peptide fractions U1 (B), U2 (C) and U3 (D).

19

Fig. 1

20

Fig. 2

(A) 74.7263

Y: inhibitory effect

66.4181 58.1100

49.8019 41.4938

2.00 9.00

1.75

8.75

1.50

X2: hydrolysis

8.50

1.25

time

1.00

8.25 8.00

X1: pH

(B) 81.9363

Y: inhibitory effect

72.1741 62.4119 52.6497 42.8875

7000.00 9.00

6500.00

8.75

6000.00

X3: enzyme concentration

8.50

5500.00

8.25

5000.00 8.00

21

X1: pH

Fig. 3

(A) P1

P3 P2

(B)

22

Fig. 4

(A) R3 R5

R2

R4 R1

(B) d

c b

b

a

23

Fig. 5

(A)

3: Diode Array 280 Range: 2.944e-2

U2 U3 2.5e-2 U1

2.0e-2

AU

1.5e-2 1.0e-2 5.0e-3

0.0 0.00

Time 2.00

6.00

4.00

8.00

11.5min 20150320-4 MaxEnt 3 84 [Ev-136887,It50,En1] (0.050,200.00,0.200,1400.00,2,Cmp) SW

1: TOFMSES+

AQ

L

bMax yMax 604.19(M+H) +

LQAWS

(B)

100

333.17 274.14 b2 %

542.15

105.05

215.12 257.12

0 50

317.19 a3

159.05 142.08 W

100

150

200

250

560.21 473.22 498.23 b4 440.46 499.26

334.19 336.10

300

350

400

24

450

500

550

660.24 586.32 600

650

759.26 685.31 700

754.34 750

760.33 796.29 M/z

11. 5mi n 20150320- 4

MaxEnt 3 GN

100

90 [ Ev- 170181, I t 50, En1] ( 0. 050, 200. 00, 0. 200, 1400. 00, 1, Cmp) H EGAE

E

A H

G

2: TOF MS ES+ E

bMax G yMax 713. 19( M+H) +

N

(C)

562. 17

110. 06 H

%

676. 20 604. 19 438. 11 b4

675. 22 560. 22

291. 08

164. 10 211. 09

267. 08 108. 03

309. 09 b3

375. 16

674. 24

487. 20

714. 25 640. 28 718. 26 M/ z

0 50

100

150

200

250

300

350

400

450

500

550

600

650

700

11.5min 20150320-4 MaxEnt 3 100 [Ev-235324,It50,En1] (0.050,200.00,0.200,1400.00,2,Cmp) CF 100

N

(D)

E EYENF 466.17 a4

YE C

1: TOFMSES+ bMax yMax

804.23(M+H) +

%

430.18 251.07 343.17 133.04 178.06 b2

607.14

701.32 662.23 y5

550.13

704.19

805.29 875.21 933.60

0 100

200

300

400

500

600

25

700

800

900

1000

M/z 1100

Table 1 Experimental design and results obtained from the hydrolysis of peanut protein isolates (PPI) by using Alcalase 2.4L coded variable

actual variable

Run

X1

X2

X3

x1

x2

x3

y (%)

1

0

0

0

8.50

1.50

6000

63.15±4.96

2

-1

0

1

8.00

1.50

7000

43.45±4.36

3

0

-1

-1

8.50

1.00

5000

52.88±0.72

4

1

1

0

9.00

2.00

6000

69.47±1.61

5

0

-1

1

8.50

1.00

7000

44.77±6.43

6

0

0

0

8.50

1.50

6000

53.54±4.55

7

1

-1

0

9.00

1.00

6000

50.48±3.09

8

0

0

0

8.50

1.50

6000

56.55±1.24

9

0

0

0

8.50

1.50

6000

58.25±2.47

10

0

0

0

8.50

1.50

6000

51.98±4.70

11

-1

0

-1

8.00

1.50

5000

48.30±1.44

12

0

1

-1

8.50

2.00

5000

86.72±4.91

13

1

0

-1

9.00

1.50

5000

84.17±1.49

14

0

1

1

8.50

2.00

7000

61.57±0.93

15

-1

-1

0

8.00

1.00

6000

46.75±0.63

16

-1

1

0

8.00

2.00

6000

42.90±3.56

17

1

0

1

9.00

1.50

7000

44.42±1.61

Values represent the means of two experiments with standard deviations. X1, X2 ,and X3 represent the coded variables for pH, hydrolysis time, and enzyme concentration; x1, x2,and x3 represent the actual variables for pH, hydrolysis time (h), and enzyme concentration (IU/g of PPI); y represents the inhibition effect on the thrombin-catalyzed coagulation of fibrinogen. y was calculated using Equation 2.

26

Highlights
A method for preparation of peanut antithrombotic peptides is developed
The purified peptides show slightly lower antithrombotic activities than heparin
Three new peanut antithrombotic peptide sequences are identified

27