High-Throughput and Rapid Screening of Novel ACE Inhibitory

Subscriber access provided by READING UNIV

Article

High-throughput and Rapid Screening of Novel ACE Inhibitory Peptides from Sericin Source and Inhibition Mechanism by Using in Silico and in Vitro Prescriptions Huaju Sun, Qing Chang, Long Liu, Kungang Chai, Guangyan Lin, Qingling Huo, Zhenxia Zhao, and Zhongxing Zhao J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.7b04043 • Publication Date (Web): 31 Oct 2017 Downloaded from http://pubs.acs.org on November 1, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Agricultural and Food Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 37

Journal of Agricultural and Food Chemistry

High-throughput and Rapid Screening of Novel ACE Inhibitory Peptides from Sericin Source and Inhibition Mechanism by Using in Silico and in Vitro Prescriptions Huaju Sun, Qing Chang, Long Liu, Kungang Chai, Guangyan Lin, Qingling Huo, *

Zhenxia Zhao, Zhongxing Zhao

Guangxi Colleges and Universities Key Laboratory of New Technology and Application in Resource Chemical Engineering, School of Chemistry and Chemical Engineering, Guangxi University, Nanning 530004, China

*

Corresponding Authors Phone: +86-771-3233718; Fax: [email protected] (Zhongxing Zhao)

+86-771-3233718;

1

ACS Paragon Plus Environment

E-mail:


1

ABSTRACT: Several novel peptides with high ACE-I inhibitory activity were

2

successfully screened from sericin hydrolysate (SH) by coupling in silico and in vitro

3

approaches for the first time. Most screening processes for ACE-I inhibitory peptides

4

were achieved through high-throughput in silico simulation followed by in vitro

5

verification. QSAR models based predicted results indicated that the ACE-I inhibitory

6

activity of these SH peptides, and six chosen peptides exhibited moderate high ACE-I

7

inhibitory activities (LogIC50 values: 1.63 to 2.34). Moreover, two tripeptides among

8

the chosen six peptides were selected for ACE-I inhibition mechanism analysis which

9

based on Lineweaver-Burk plots indicated to behave as competitive ACE-I inhibitor.

10

The C-terminal residues of short-chain peptides that contain more H-bond acceptor

11

groups could easily form hydrogen bonds with ACE-I and have higher ACE-I

12

inhibitory activity. Overall, sericin protein as a strong ACE-I inhibition source could

13

be deemed a promising agent for antihypertension applications.

14

KEYWORDS: sericin hydrolysis, in silico prediction, LC-MS, ACE-I inhibitory

15

peptides, molecular docking, inhibitory mechanism

16

2


Page 2 of 37

Page 3 of 37


17

■ INTRODUCTION

18

Hypertension is a major risk factor for developing cardiovascular diseases, and has

19

become a serious threat to human health and quality of life recently. According to

20

incomplete statistic from the World Health Organization (WHO), this chronic disease

21

will increase up to 29 % of the world’s adult population till 2025.1 It has been

22

reported that angiotensin-I-converting enzyme (ACE-I, EC 3.4.23.15) plays a key role

23

in blood pressure control in renin-angiotensin-aldosterone system, which alternatively

24

can monitor and attenuate hypertension.2 Recently, search for new ACE-I inhibitor

25

peptides with high bioactivity from natural sources has attracted greater attention for

26

hypertension treatment attributed to their minimal side-effects.3

27

Up till now, more than 700 types of ACE-I inhibitory peptides have been reported

28

in BIOPEP database (http://www.uwm.edu.pl/biochemia/index.php/en/biopep). Most

29

potent peptides were directly screened from enzymatic hydrolysates of traditionally

30

edible proteins, such as milk proteins,4,5 marine organisms,6-11 meat,12,13 and plant

31

proteins.14,15 However, purification process like gel chromatography, ion exchange

32

chromatography and reverse phase high-performance liquid chromatography are

33

hindered by complicated and time-consuming nature. Furthermore, the low

34

purification efficiency and high cost of these approaches have seriously restricted

35

their applications in industrial development. Moreover, screening for ACE-I

36

inhibitory peptides is usually estimated from in vitro ACE-I inhibitory activity of

37

hydrolysates, where the fraction with highest activity is chosen for further separation.

38

However, among other problem associated with conventional extraction strategies,

39

there exist a chance of missing some peptides with high activity.16 Thus, high

40

throughput and rapid screening of novel peptides with high ACE-I inhibitory activity 3



Page 4 of 37

41

from natural sources has become a center of rich research area among scientific

42

community dealing in separation and purification of peptides.

43

With the development of sequencing technology and biomolecular simulation, an

44

increasing number of researches are booming for predicting activity of peptides

45

binding with ACE-I by using computer simulation. Compared to traditional methods,

46

in silico analysis is considered as a high-throughput and rapid technique to

47

preliminarily select bioactive peptides from known protein sequences. Based on some

48

typical enzymes with known cleavage specificities, in silico approach has been used

49

to predict bioactive peptides from natural proteins sources, such as prolyl

50

endopeptidase inhibitory peptides,17 peptidase-IV inhibitory peptides and ACE-I

51

inhibitory peptides from edible sources including meat by-products and milk

52

proteins.18,19. However, not many reports are available about screening of novel and

53

highly-active peptides from natural source with the assistance of in silico method.

54

Sericin, as a major component of silkworm cocoon, has been reported to have great

55

potential

for

many

biomedical

and

biological

56

antibiotic-antibacterial activity,20 antioxidant behavior,21 anti-tyrosinase activity22 and

57

anti-carcinogenic effects.23 To the best of our knowledge, its potential activity for

58

ACE-I inhibitory property has not been reported so far. Thus, in this work, several

59

novel peptides were successfully screened from sericin hydrolysate (SH) for ACE-I

60

inhibition by coupling in silico and in vitro approaches. In silico was used to screen

61

hydrolysate enzyme (PeptideCutter) for sericin protein, predict ACE-I inhibitory

62

activity in terms of Quantitative structure activity relationship (QSAR), analyze

63

toxicity and anti-digestion activity (ToxinPred and PeptideCutter) of the obtained

64

peptides from virtual SH source using liquid chromatography – mass spectrometry

65

(LC-MS) technique. The selected peptides with theoretically high ACE-I inhibitory 4


applications,

such

as

Page 5 of 37


66

activity were synthesized and verified in vitro studies. Moreover, ACE-I inhibition

67

pattern and ACE-I inhibition mechanism of the selected peptides were also

68

systematically studied from classic Lineweaver-Burk model and molecular docking

69

simulation.

70 71

■ MATERIALS AND METHODS

72

Materials. The silkworm cocoon waste was provided by Maoyuan silk cocoon Co.,

73

Ltd. (Yizhou, China). Trypsin (EC 3.4.21.4, 250 U/mg) was purchased from Biodee

74

Biotechnology Co. Ltd. (Beijing, China), while proteinase K (EC 3.4.21.64, 40 U/mg)

75

and pepsin (pH>2) (EC 3.4.23.1, 2500 U/mg) were purchased from Solarbio Science

76

& Technology Co., Ltd. (Beijing, China). The hippuryl-L-histidyl-L-leucine (HHL)

77

and ACE-I (from rabbit lung) was supplied by Sigma-Aldrich Chemical Co. (St. Louis,

78

MO, USA). The identified peptides were synthesized (98% purity) by GL. Biochem.

79

Co., Ltd. Methanol for HPLC analysis was supplied by Thermo Fisher Scientific Co.,

80

Ltd. (USA). All other chemicals and reagents used were of analytical grade without

81

further treatment.

82

Selection of proteases for hydrolysis using PeptideCutter analysis. Sericin was 24,25

83

mainly composed of three proteins (Ser1, Ser2 and Ser3) as reported previously

84

Among which, Ser1 contributed 90% of the total weight of cocoon proteins in the last

85

larval instar. Therefore, we extracted the sequence of Ser1 from protein data

86

UniProtKB (http://www.uniprot.org/) and protein database of the National Center for

87

Biotechnology Information (NCBI) (http://www.ncbi.nlm.nih.gov/protein).15 Then,

88

Ser1 was consequently subjected to in silico hydrolysis using the program of

89

PeptideCutter (http://web.expasy.org/peptide_cutter/) to identify the most appropriate

5



90

proteases for sericin hydrolysis (SH). The top three proteases with the largest cleavage

91

sites were chosen for sericin hydrolysis in vitro.

92

Preparation of sericin hydrolysis in vitro. Sericin was extracted from silkworm

93

cocoon waste autoclaving at 121 °C and 0.2 MPa (mention autoclave environment)

94

for 2 h.26 Its ratio of solid to liquid was adjusted to 1:30 (W/V). After filtration, the

95

extraction solution was lyophilized and stored under 5.0 °C. The extraction ratio of

96

sericin was approximately 260-300 g/kg from silkworm cocoon. The hydrolyzed

97

using the determined three proteases of pepsin (P), trypsin (T), proteinase K (K) and

98

their bienzyme T+K system under hydrolysis conditions as shown in Table 1. In this

99

work, both of single protease and bienzyme were used to hydrolyze sericin, which

100

were named as one-stage and two-stage enzymatic hydrolysis, respectively. Finally,

101

four SHs were obtained by using different protease systems. The ACE-I inhibitory

102

activity of different SHs was tested, and the one with highest ACE-I inhibitory

103

activity was further studied.

104

Measurement of ACE-I inhibitory activity. The assay of ACE-I inhibitory activity

105

in vitro was measured by Cushman’s method 27 with slight modifications. It is mainly

106

based on the liberation of hippuric acid from hippuryl-L-histidyl-L-leucine

107

(Hip-His-Leu) catalyzed by ACE-I. The IC50 value is defined as the concentration of

108

peptides (µM) required to inhibit 50% of ACE-I activity.28 Each assay was performed

109

in triplicate of independent determinations. The IC50 value of each SH was used to

110

screen protease and resource of ACE-I inhibitory peptide (hydrolysate).

111

Construction of QSAR model for rapid screening peptides with high ACE-I

112

inhibitory activity. QSAR models of peptides with ACE-I inhibitory activity were

113

constructed according to the method of Wei Qi 29 with slight modifications. Firstly, a 6


Page 6 of 37

Page 7 of 37


114

total of 122 ACE-I inhibitory peptides (from di- to hexapeptide) with known ACE-I

115

inhibitory activity were retrieved from both the BIOPEP database and published

116

literature.30 These peptides were classified into three data sets (train sets and test sets)

117

according

118

tetra/penta/hexa/peptides. Then, the most active peptide of each data set was chosen

119

as the template, and the rest 119 peptides were aligned on template to form three

120

common structures using Sybyl X-2.1.1 (Tripos Inc., St. Louis, MO, USA).

to

peptide

length,

including

dipeptides,

tripeptides

and

121

Next, CoMFA and CoMSIA descriptor fields (steric, electrostatic, hydrophobic,

122

donor and acceptor) were used to analyze three training sets, and the best QSAR

123

model was identified by the partial least squares (PLS) methodology analysis with the

124

leave-one-out (LOO) cross-validation procedure. The cross-validated coefficient (Q2)

125

and standard deviation of error prediction (R2) were calculated according to Eq. (1)

126

and Eq. (2):31 ∑( − ) =1− ∑( − )

= 1 −

∑( − ) ∑( − )

Eq. (1)

Eq. (2)

127

where Ymean is average activity of the entire data set, while Yobs, Ypred and YCVpred

128

represent the observed, predicted and cross-validated activity values, respectively.

129

LC-MS analysis for Sericin hydrolysate. SH with highest ACE-I inhibitory activity

130

was used for peptide sequence identification by a nanoLC LTQ Orbitrap MS (Thermo

131

Scientific, USA), following the stepwise method as: (1) SH was separated with a

132

Hypersil C18 column (250×4.6 mm, particle size 5 µm); (2) the adsorbed components

133

were eluted with a two solvent system: (A) 0.1% formic acid and 2.0% acetonitrile in

134

water and (B) 0.1% formic acid and 2.0% water in acetonitrile with a gradient from 7



Page 8 of 37

135

2.0% solvent B to 15.0% solvent B over 5.0 min, to 35.0% over 20 min, to 98.0%

136

over 5 min, maintained for 10 min at a constant flow rate of 250 nL/min; (3)

137

ionization was performed by nanolockspray ionization source in positive ion mode

138

(capillary voltage 3.80 kV and a source temperature of 280 °C; (4) an MS full-scan

139

was performed for each sample with an acquisition range m/z of 50-800 Da.

140

ACE-I inhibitory activity prediction and validation. The constructed QSAR model

141

was used to predict ACE-I inhibitory activity of identified peptides from SH

142

hydrolysate by the combination of T and K. Nine peptides with various ACE-I

143

inhibitory activities from QSAR prediction were chosen, and their toxicity and

144

digestive

145

ToxinPred(http://www.imtech.res.in/raghava/toxinpred/)32

146

Finally, eight peptides with nontoxicity and anti-digestion activity were synthesized

147

by GL. Biochem. Co., Ltd. (Shanghai, China) using conventional solid-phase

148

chemistry and validated their activities in vitro.

stability

were

subsequently

simulated and

by

PeptideCutter.18

149

Kinetics of ACE-I Inhibition. Among the six selected peptides with the highest

150

ACE-I inhibitory activity, the highest and lowest ACE-I inhibitory active ones (named

151

A and B) were inspected for the inhibition kinetics by Lineweaver-Burk plot.33

152

Briefly, ACE-I inhibition tests were conducted in the presence of different

153

concentrations of HHL and peptides solution. The varying concentration of the

154

enzyme substrate HHL (0.62, 1.10, 1.64 and 1.97 mM) was reacted with ACE-I in the

155

absence and presence of two different concentrations of inhibitory peptide A (108.89

156

µM and 163.34 µM) and peptide B (93.77 µM and 140.65 µM).34,35

157

Molecular docking of the selected peptides with ACE-I. Molecular docking was

158

performed to investigate conformation between ACE-I active sites and inhibitors 8


Page 9 of 37


159

using the flexible docking tool of Sybyl X-2.1.1 program package, and the structure

160

was energy minimized using the Powell conjugate gradient optimization algorithm

161

with the Tripos force field.36 The three-dimensional crystal structure of

162

ACE-I-lisinopril complex was obtained from the Protein Data Bank (PDB: 1O86,

163

http://www.rcsb.org/pdb/home/home.do) by only removing the inhibitor lisinopril and

164

water molecule. Then the protein structure was pre-analyzed and prepared for the

165

docking runs using biopolymer structure preparation tool with default settings and the

166

protocol was created automatically. The Surflex-Dock program was used for docking

167

studies. The receptor-ligand interactions of the ligand were evaluated by the software

168

in terms of Total Score and expressed as LogKd, where Kd is binding constant, where

169

high value of Total Score represents good protein-ligand binding and vice versa. The

170

conformations of ranked No. 1 (the highest Total Score) was selected and the

171

interaction model between the ACE-I residues and peptides was shown by PyMOL

172

v1.8.6 (http://www.pymol.org/).

173

Statistical analysis. All assays of ACE-I inhibitory activity were conducted in

174

triplicates. Data were presented as mean ± standard deviation. The statistical analysis

175

was performed using SPSS 10.0 software (SPSS Inc., Chicago, IL, USA). Significant

176

difference in means between the samples was determined at a 5% confidence level (p

177

< 0.05).

178 179

■ RESULTS AND DISCUSSION

180

Amino acid composition of Ser1 and screening enzymes for sericin hydrolysis.

181

Figure S1(A) shows the sequence of Ser1, obtained from protein data UniProtKB and 9



182

protein database of NCBI. Its amino acid (AA) composition was counted using

183

ProtParam (http://web.expasy.org/protparam/), and the results are shown in Figure 1

184

(A). It can be seen that sericin possesses the most abundant AA in Ser1 protein,

185

holding a minimum 32 % of Ser1. Hydrophobic AAs (red columns) account for 30.0%

186

of Ser1 except serine, which could count for high activity of ACE-I inhibitors due to

187

the presence of more hydrophobic AAs in the protein.37

188

Ser1 protein was consequently subjected to in silico hydrolysis using PeptideCutter

189

to determine the most suitable proteases for in vitro sericin hydrolysis. In this work,

190

about 38 available enzymes and chemicals (Table S1) supplied from this program,

191

were all chosen for theoretic hydrolysis simulation. The top five proteases with largest

192

cleavage sites are shown in Table S2. The order of cleavage site number was:

193

proteinase K > pepsin (pH>2) > trypsin > clostripain > pepsin (pH 1.3). Next,

194

proteases of proteinase K (K), trypsin (T), and pepsin (pH>2) (P) were chosen for

195

sericin hydrolysis in experiment due to its possessing most cleavage sites which

196

possibly release most ACE-I inhibitory peptides from protein.38

197

After being hydrolyzed with different proteases in vitro, four hydrolysates were

198

prepared and their ACE-I inhibitory activities were tested. Figure 1B shows their

199

LogIC50 values corresponding to different hydrolysis conditions with single protease

200

and biprotease. In one-stage hydrolysis with single protease, hydrolysates show the

201

ACE-I inhibitory activity in the following order of preference: K > T > P. In the next

202

experiment, proteinase K and trypsin (K+T) were used to hydrolyze sericin in

203

two-stage hydrolysis. As expected, its hydrolysate shows the highest ACE-I inhibition

204

activity. About, 137 peptides in its hydrolysate were obtained and their sequence

205

distribution is shown in Figure S1(B). Consistent with that observed by simulation 10


Page 10 of 37

Page 11 of 37


206

(Table S2), more cleavage sites may have contributed towards the higher ACE-I

207

inhibition activity.

208

QSAR model for prediction of peptide ACE-I inhibitory activity. Amino acids at

209

C-terminal end usually contribute significantly to ACE-I inhibitory activity.39,40 The

210

C-terminal residues of 137 peptides from Ser1 hydrolysates (K+T) using in silico

211

were analyzed and shown in Table S3, where most were composed of hydrophobic

212

AA (50-60 % of the total), basic AA and acidic AA. Based on these results, some

213

peptides with known ACE-I inhibitory activity

214

useful peptides as the training and test sets for QSAR model construction were chosen.

215

These 122 peptides possessing the same C-terminal residues, which composed of 38

216

dipeptides, 39 tripeptides and 45 peptides length between 4 and 6, are listed in Table

217

S4. Peptides of 13th (LogIC50=0.21), 57th (LogIC50=0.36) and 96th (LogIC50=0.18)

218

have the highest activity of ACE-I inhibition for each training set. Thus, these three

219

peptides were selected as the template, and the rest of peptides in each training set

220

were aligned onto templates (peptide 13th, 57th and 96th) to form three

221

superimpositions. Based on three common structures, 122 peptides were analyzed by

222

CoMFA and CoMSIA, respectively.

30

were collected, among which 122

223

Subsequently, the stepwise development of CoMFA and CoMSIA models using

224

five descriptor fields i.e. steric, electrostatic, hydrophobic, acceptor and donor fields,

225

41,42

226

tetra/penta/hexapeptides, respectively, while the statistical parameters for the best

227

QSAR CoMFA/CoMSIA models of each data set are shown in Table 2. As seen in

228

QSAR

229

tetra/penta/hexa-peptides were corresponding to CoMSIA, CoMFA and CoMSIA

is

presented

analysis,

in

the

Table

best

S5-S7

for

predictions

dipeptides,

for

11


tripeptides,

dipeptides,

and

tripeptides,


230

models, respectively. The optimum model for each data set showed all high statistical

231

significances (Q2 >0.73 and R2 >0.99), indicating a proof of high predictive ability of

232

the constructed model.43

233

Moreover, the contributions of five descriptor fields for each data set are also

234

shown in Table 2. Clearly, the contribution of each field to peptide ACE-I inhibitory

235

activity was different for each data set. Dipeptide set was contributed by the steric,

236

electrostatic and acceptor fields prominently; tripeptide set was completely controlled

237

by steric field; while tetra/penta/hexapeptide set was mainly influenced by

238

electrostatic, hydrophobic and donor fields. The scatter plots of observed against

239

predicted LogIC50 values for the best QSAR model were displayed in Figure

240

2(A-C).44,45 This analysis showed that the obtained data points were uniformly

241

distributed along the regression line with small prediction errors ( 2), trypsin and chymotrypsin,

303

among which, only SSY could be cleaved during simulation of GI digestion. Thus, the

304

remaining eight stable peptides (SSR, SSK, GSR, SGR, NPR, SSSNSV, SSDA and 14


Page 14 of 37

Page 15 of 37


305

DE) were further synthesized and validated for their ACE-I inhibitory activities in

306

vitro (Table 4). It can be seen that all the synthesized peptides exhibited similar tested

307

LogIC50 value to predicted one with small prediction errors (

High-Throughput and Rapid Screening of Novel ACE Inhibitory

Recommend Documents