Accepted Manuscript

Antioxidant and anti-freezing peptides from salmon collagen hydrolysate pre-

pared by bacterial extracellular protease

RiBang Wu, CuiLing Wu, Dan Liu, XingHao Yang, JiaFeng Huang, Jiang
Zhang, Binqiang Liao, HaiLun He

PII: S0308-8146(17)31984-2
DOI: https://doi.org/10.1016/j.foodchem.2017.12.035
Reference: FOCH 22132

To appear in: Food Chemistry

Received Date: 11 July 2016

Revised Date: 26 September 2017
Accepted Date: 10 December 2017

Please cite this article as: Wu, R., Wu, C., Liu, D., Yang, X., Huang, J., Zhang, J., Liao, B., He, H., Antioxidant and
anti-freezing peptides from salmon collagen hydrolysate prepared by bacterial extracellular protease, Food
Chemistry (2017), doi: https://doi.org/10.1016/j.foodchem.2017.12.035

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.
 Antioxidant and anti-freezing peptides from salmon collagen hydrolysate

prepared by bacterial extracellular protease

RiBang Wu1, CuiLing Wu1, Dan Liu 1, XingHao Yang1, JiaFeng Huang1, Jiang Zhang1,

Binqiang Liao1, HaiLun He1*

1
School of Life Science, State Key Laboratory of Medical Genetics, Central South

University, Changsha 410013 ，China

*Corresponding author:

Hailun He，Tel：+86-0731-82650230; Fax：+86-0731-82650230; E-mail address:

helenhe@csu.edu.cn

1
Abstract

Extracted salmon skin collagen was hydrolysed with the free or immobilized

extracellular protease of Vibrio sp. SQS2-3. The hydrolysate exhibited anti-freezing

activity (> 3 kDa) and antioxidant activity (< 3000 Da) after ultrafiltration. The

antioxidant peptide was further purified by size-exclusion chromatography and found

to scavenge DPPH (73.29 ± 1.03%), ·OH (72.73 ± 3.34%,), and intracellular ROS in

HUVECs; protect DNA against oxidation-induced damage; and have an ORAC of

2.78 ± 0.28 mmol TE/g. The antioxidant peptide fraction was identified using mass

spectrometry, and nineteen salmon collagen-sourced peptides were obtained. Of these,

the peptide Pro-Met-Arg-Gly-Gly-Gly-Gly-Tyr-His-Tyr is a novel sequence and was

the major component; this peptide was shown to have antioxidant activity via the

ORAC assay (2.51 ± 0.14 mmol TE/g). These results suggested that the protease from

Vibrio sp. SQS2-3 is suitable for preparation of anti-freezing peptides and antioxidant

peptides in a single step and represents a comprehensive use of fish skin collagen.

Keywords:

Salmon skin collagen; bacterial extracellular protease; enzymatic hydrolysis;

antioxidant peptide; anti-freezing

1. Introduction

Reactive oxygen species (ROS), including hydroxyl radicals (·OH), superoxide

anion radicals (O2· −), hydrogen peroxide (H2O2), peroxyl radicals (ROO· −), and

2
nitroxide radicals (NO·), are strongly related to aging and diseases, such as cancer,

Alzheimer's disease, cardiovascular disease, diabetes and weakening of the immune

system (Stohs, 1995; Diaz, 1997; Spolarics, 1998). Antioxidant substances

(glutathione, β-carotene, vitamin C and vitamin E) and antioxidant enzymes

(superoxide dismutase, glutathione S transferase and catalase) can remove excess free

radicals to prevent oxidative stress. In addition, free radical-mediated oxidation in

food also leads to potential toxicity, which is a concern of the food industry and

consumers (Frankel, 2005). Therefore, synthetic antioxidants are utilized in the food

industry, but their potential toxicity requires strict regulation; examples include

butylated-hydroxytoluene (BHT), butylated-hydroxyanisole (BHA),

tertbutylhydroquinone (TBHQ), and propyl gallate (PG). Peptides with antioxidant

activity represent novel antioxidants that can replace synthetics. Many studies have

reported that peptides released from parental protein by hydrolysis could be an

effective way to obtain antioxidant peptides.

Hydrogen bonds, electrostatic interactions, and disulphide bonds are the major

forces that stabilize the structure of proteins (Benjakul, 2000). These bonds and

interactions can be disrupted by a sharp temperature change during frozen storage

(Benjakul, 2003; Benjakul, 2009). Cryoprotectant can protect proteins against

freezing damage by inhibiting the generation of ice crystals (Alvarez, 2010), e.g.,

sorbitol and sucrose, which are common additives in the food storage industry (Jin,

2010). However, these cryoprotectants make food products sweet, which is often

unacceptable to consumers (Ruttanapornvareesakul, 2006). It has been reported that a

3
protein hydrolysate containing oligopeptides is able to inhibit protein denaturation,

including peptides from blacktip shark skin (Kittiphattanabawon, 2012), gelatine from

bovine hide (Wang, 2009) and sericin (Wu, 2015).

Recent studies have shown that collagen hydrolysates have various biological

activities, such as gelatin hydrolysis with anti-freezing activity from blacktip shark

skin (Phanat, 2012), antioxidant collagen peptides from croceine croaker (Wang,

2013), and angiotensin I-converting enzyme-inhibitory antioxidant peptides from

skate skin gelatin (Ngo, 2014). Skin from salmon contains by-products that are rich in

collagen, which can create value-added products by enzymatic hydrolysis. However,

most salmon skin is discarded and/or used as animal feed. Therefore, this study

focused on value-added utilization of collagen from salmon skin through enzymatic

hydrolysis. An antioxidant peptide from ultrafiltration fractions with a small molecule

weight was identified; ultrafiltration fractions containing larger peptides that were

identified as potential cryoprotectants.

2. Materials and methods

2.1 Materials

Fresh salmon by-products with skin were purchased from a seafood market in

Shanghai and stored at -20 °C prior to use. Soybean meal was purchased from a

supermarket in Changsha, Hunan Province, China, and stored at -20 °C. The

protease-producing strain Vibrio sp. SQS2-3 was isolated from the water of the

intertidal zone in Hainan Province, China (GenBank: KF220487.1).

4
 Tryptone and yeast extract were purchased from Oxoid (Basingstoke, UK).

1,1-diphenyl-2-picrylhydrazyl (DPPH), fluorescein, 2,20-azobis (2-amidinopropane)

dihydrochloride (AAPH) and 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic

acid (Trolox) were purchased from Sigma-Aldrich, Ltd. (Shanghai, China). Ninhydrin,

1,10-phenanthroline monohydrate was purchased from Aladdin® (Shanghai, China).

Sephadex LH-20 was purchased from GE Healthcare Life Sciences (Uppsala,

Sweden). RPMI 1640 medium and foetal bovine serum was purchased from Gibco®

ThermoFisher Scientific Company (Grand Island, USA).

2.2 Methods

2.2.1 Preparation of salmon skin collagen

Byproducts with skin were washed with flowing water for 5 min. The muscles

were removed thoroughly. The skin was sheared into 5-mm pieces and washed with

cold distilled water 5-7 times to remove fish oil. Salmon skin was mixed with distilled

water at a ratio of 1:10 (w/v, g/ml), heated for 30 min at 75 °C, and centrifuged at

10,000 × g for 10 min to extract soluble collagen. The supernatant was dialysed using

a dialysis bag (with 8 kDa to 14 k Da molecule weight cut-off) with 4 °C distilled

water and lyophilized into collagen solids.

2.2.2 Preparation and Purification of bacterial extracellular protease

Extracellular protease was obtained through bacterial fermentation according to

the methods of Dan Liu (Liu, 2014) with some modifications. The protease-producing

5
strain Vibrio sp. SQS2-3 was activated at 15 °C with shaking at 200 rpm for 24 h in

2216E medium. The activated strain was seeded into fermentation broth (0.5 % corn

powder, 0.5 % bean powder, 0.25 % wheat bran, 0.1 % CaCl2, 0.4 % Na2HPO4, and

0.03 % KH2PO4, prepared with sea water) at a ratio of 1:50 (v/v, ml/ml) and cultured

at 15 °C with shaking at 200 rpm. The supernatant was collected by centrifugation

(10,000 × g, 4 °C, 20 min) after 120 h and dialysed in Tris-HCl (20 mM, pH 7.8) at

4 °C. The crude enzyme from Vibrio sp. SQS2-3 was purified in a HiTrap DEAE FF

(5 ml, 16 × 25 mm) anion exchange column with an NGC chromatography system

(Bio-Rad, Hercules, CA, USA). The chromatography column was equilibrated with

20 mM Tris-HCl (pH 7.8) at a flow rate of 2 ml/min for 10 min. The crude enzyme

was loaded into the pre-equilibrated column at a flow rate of 2 ml/min and washed

with 20 mM Tris-HCl (pH 7.8) for 10 min. The column was eluted with a linear

gradient of 1 M NaCl (0-100%) at a flow rate of 2 ml/min for 12.5 min. Protein

fractions were monitored at 280 nm. The protease activity and total protein content

were measured by the Folin phenol assay (Anson, 1938) and BCA method,

respectively.

2.2.3 Immobilization of extracellular protease

Enzyme immobilization on chitosan was performed according to Singh et al.

Chitosan powder was dissolved in 1.5% (v/v) glacial acetic acid to prepare the

chitosan solution (1.5%, w/v). The solution was dropped into a 1 M KOH solution

containing 25% (v/v) ethanol and incubated for 1 h to prepare chitosan beads. The

6
beads were washed with 0.1 M sodium phosphate buffer (pH 7.8) and activated with 2%

(v/v) glutaraldehyde at room temperature for 1 h. Subsequently, the activated chitosan

beads were washed extensively with distilled water for complete removal of unreacted

glutaraldehyde. For immobilization, the activated beads were incubated with purified

protease at 4 °C for 12 h and washed with 0.1 M sodium phosphate buffer (pH 7.8) to

remove unbound protease. The protease activity and total protein content were

measured using the Folin phenol assay (Anson, 1938) and BCA method, respectively.

The efficiency of enzyme immobilization was calculated as follows:

speci ic activity of immobilized protease
immobilization ef icience % = × 100
speci ic activity of soluble protease

2.2.4 Enzymatic hydrolysis

Salmon skin collagen was digested with the protease from Vibrio sp. SQS2-3 at

an enzyme to substrate ratio of 1:10 (v/w, ml/mg) and temperature of 45 °C for 30, 60,

90, 120, and 150 min. The reaction was terminated at 95 °C for 10 min. The

hydrolysate was collected and centrifuged at 10,000 × g at 4 °C for 10 min. The

hydrolysis degree at various treatment times was analysed using the ninhydrin

coloration method. The standard curve was determined using leucine at

concentrations of 0, 0.4, 0.8, 1.2, 1.6, 2.0 mM.

2.2.5 Purification of bioactive fractions

Ultrafiltration tubes with 3,000 Da MWCO (Amicon Ultra 14 ml 3K, Millipore,

USA) were utilized for preliminary separation. The supernatant of the hydrolysate

7
was added to the upper casings of the ultrafiltration tubes, which were subsequently

centrifuged at 4500 × g and 4°C for 60 min. Fractions in the upper casings and lower

casings after ultrafiltration were collected and named UF-1 and UF-2. UF-1, with

larger molecule weight, was considered to be a potential anti-freezing agent. Distilled

water was added to UF-1 to replace the salt-containing solution using ultrafiltration

tubes with 3k Da MWCO at 4,500 × g and 4°C for 45 min.

UF-2, which had smaller molecule size, was further purified by size exclusion

liquid chromatography. The fraction was loaded (1 ml) onto a Sephadex LH-20

column (16 × 600 mm), equilibrated with distilled water and eluted with distilled

water at a flow rate of 0.75 ml/min. Each fraction was monitored at 220 nm, collected

at a volume of 5 ml and lyophilized. Each fraction peak was prepared at the same

concentration (150 µg/ml), and the antioxidant activity was further detected in

subsequent assays.

2.2.6 Measuring Ca2+-ATPase activity of myoglobulin after freeze-thawing cycles

Fresh porcine muscle was washed with cold distilled water and sliced into pieces

(5 mm × 5 mm). The muscle was homogenized in distilled water (2 °C) at a ratio of

1:10 (w/v, g/ml) by a homogenizer at a speed of 11,000 rpm for 1 min. The

homogenate was then centrifuged at 10,000 × g for 30 min at 2 °C, and the precipitate

was collected. To extract myoglobulin, the precipitate was further homogenized with

1.2 M KCl, pH 7.0 for 4 min at a ratio of 1:10 (w/v) in an ice bath. The extract was

centrifuged at 5000 × g for 30 min at 2 °C. Myoglobulin was precipitated with three

8
volumes of distilled water and collected by centrifuging at 5000 × g for 30 min at

2 °C. Then, myoglobulin was dissolved in an appropriate volume of KCl (1.2 M, pH

7.0) by stirring for 30 min at 2 °C. Myoglobulin was kept at 4 °C and was used within

12 h.

The upper fraction from ultrafiltration was added to the myoglobulin solution at

a ratio of 1:1 (v/v). The mixture was frozen at -20 °C for 20 min. The frozen sample

was thawed at 37 °C in a water bath for 10 min. Freeze-thawing was performed for 0,

3, 6 and 9 cycles. Thawed samples were kept on ice until analysis. The upper fraction

was replaced by distilled water in the blank.

The Ca2+-ATPase activity of myoglobulin was determined using the method of

Benjakul et al. (Benjakul, 1997) The freeze-thaw sample (25 µl) was mixed with 15 µl

of 0.5 M Tris-maleate (pH 7.0), 25 µl of 0.1 M CaCl2 and 172.5 µl of distilled water.

The mixture was mixed in a vortex mixer. ATP (20 mM, 12.5 µl) was added to initiate

the reaction, which was incubated at room temperature for 10 min. The reaction was

terminated by adding 125 µl of trichloroacetic acid (100 g/ml, 4 °C) and centrifuged

at 5000 × g for 5 min. The inorganic phosphate liberated in the supernatant was

measured by the method of Fiske and Subbarow (1925). The supernatant (50 µl) was

added to 80 µl of 2.5% ammonium molybdate, 20 µl of 1% stannous chloride

(prepared at a ratio of 1:10:25, SnCl2: HCl (conc.): 1 N H2SO4), and 900 µl of

distilled water, and the mixture was incubated at room temperature for 10 min. The

absorbance of the resulting mixture was measured at 720 nm by a multiplate reader. A

blank solution was prepared by adding TCA prior to the addition of ATP. The

9
Ca2+-ATPase activity of myoglobulin after freeze-thawing for 3, 6 and 9 cycles was

reported as residual Ca2+-ATPase activity (%), which was calculated in comparison to

the sample without freeze-thawing.

2.2.7 DPPH radical scavenging activity assay

The DPPH radical scavenging activity assay was conducted according to the

method of Shimada et al. (1992). DPPH solution (100 µl, 0.1 mM in 95% ethanol)

was mixed with 20 µl of the purified fraction solution (100 µg/ml) and incubated at

room temperature for 1 h. The absorbance of the resulting solution was measured at

517 nm using a microplate reader (Enspire 2300, Multimode Plate Reader, Perkin

Elmer, USA). For the blank, the purified fraction was replaced with distilled water.

Vitamin C at a concentration of 200 µg/ml was used as a positive control. The DPPH

radical scavenging activity was calculated using the following formula:

DPPH radical scavenging activity (%) = [1-Abs sample /Abs blank] × 100

2.2.8 Hydroxyl radicals scavenging activity assay

The hydroxyl radical scavenging activity was measured according to the method

developed by Wang et al. (2012) A FeSO4 solution (40 µl, 2 mM), 1,

10-phenanthroline (40 µl, 2 mM) and sample (80 µl) were mixed stepwise. The

reaction was initiated by adding 40 µl of H2O2 (0.03% v/v). After incubation at 37 °C

for 60 min, the absorbance of the resulting solution was measured at 536 nm using a

microplate reader (Enspire 2300, Multimode Plate Reader, Perkin Elmer, USA). The

10
group without any antioxidant was used as the negative control, while the mixture

without H2O2 was used as the blank. The hydroxyl radical scavenging activity (HRSA)

was calculated as follows:

HRSA (%) = [(As - An)/(Ab - An)] × 100

where As, An, and Ab are the absorbance values determined at 536 nm for the sample,

negative control and blank after reaction, respectively.

2.2.9 Protection effect on oxidation-induced DNA damage

The protection effect assay for oxidation-induced damage followed the method

described by Qian et al. (2008). Plasmid DNA has different structures according to the

degree of damage, which can be detected by agarose gel electrophoresis. The reaction

system included 8 µl of pET-22b DNA, 2 µl of 2 mM FeSO4, 8 µl of antioxidant and 2

µl of 0.1 mM H2O2. The mixture was incubated at 37 °C for 10 min and analysed by 1%

agarose gel electrophoresis.

2.2.10 Oxygen radical absorbance capacity (ORAC) assay

The ORAC assay was measured according to the method of Alberto et al. (2004).

The reaction was performed in 75 mM phosphate buffer (pH 7.4). The sample

solution (20 µl) and fluorescein (100 µl, 96 nM) was added to a 96-well microplate,

and the reaction was initiated by adding 30 µl of AAPH (120 mM). The fluorescence

intensity was measured every 30 s for 180 cycles with excitation and emission

wavelengths of 485 nm and 538 nm, respectively, at 37°C in a microplate reader

11
(Enspire 2300, Multimode Plate Reader, Perkin Elmer, USA). Trolox was used as the

positive control. The ORAC was defined as trolox equivalents (mmol TE/g peptide or

mmol TE/mmol peptide) according to the area under the curve (AUC) and calculated

as follows:

ORAC = (AUCsample-AUCcontrol) / (AUCTrolox-AUCcontrol) × (MTrolox / Msample)

where AUCsample, AUCcontrol and AUCTrolox were the integral areas under the

fluorescence decay curve of the peptide with 75 mM PBS (pH 7.4) and Trolox,

respectively. MTrolox and Msample were the concentrations of trolox and peptide,

respectively.

2.2.11 Cell culture and cytotoxicity determination

Human umbilical vein endothelial cells (HUVECs) were grown in Roswell Park

Memorial Institute (RPMI) 1640 medium containing 10% (v/v) foetal bovine serum

(FBS), 100 µg/ml penicillin-streptomycin and 5% CO2 at 37 °C. The cytotoxicity

levels of samples on cells were measured using the MTT method. HUVECs were

treated with 0.25% trypsin containing 0.02% EDTA and re-suspended with RPMI

1640 medium containing 10% (v/v) FBS. Cells (1 × 105 cells/ml) were seeded into

96-well plates and incubated for 24 h with peptides (20 and 100 µg/ml，respectively).

After various treatments, the medium was removed and cells were incubated in a

solution of 1 mg/ml MTT at 37 °C for 4 h. The supernatant was removed, and 150 µl

of DMSO was added to solubilize the formed formazan salt. The concentration of

formazan salt was determined by measuring the absorbance at 490 nm using a

12
microplate reader (Enspire 2300, Multimode Plate Reader, Perkin Elmer, USA). The

viability of cells was quantified as a percentage compared to the blank (peptides

replaced with 0.01 M phosphate buffer saline).

2.2.12 Determination of ROS production

Intracellular formation of ROS was determined by employing the oxidation

sensitive dye 2’,7’-dichloro-fluorescence diacetate (DCFH-DA) as a substrate.

HUVECs (1 × 105 cells/ml) were seeded into a 24-well plate and grown in RPMI

1640 medium containing 10% (v/v) FBS, 100 µg/ml penicillin-streptomycin and 5%

CO2 at 37 °C for 12 h. The medium was replaced by RPMI 1640 medium (without

FBS, but with added glucose), and the cells were incubated for 24 h. Subsequently, 10

µM DCFH-DA (prepared with RPMI 1640 medium) was added to each well and

incubated for 1 h at 37 °C. Excess DCFH-DA was removed and cells were washed

with RPMI 1640 medium three times. Treated cells were immersed in 400 µl of 0.01

M phosphate buffer saline, and images of stimulated HUVECs were collected using a

Nikon ECLIPSE TE2000-U with a digital CCD camera (DS-U2, Nikon, Japan) under

fluorescence.

2.2.13 Identification of antioxidant fraction and solid phase synthesis of peptides

The active peak (F5) was further analysed by liquid chromatography-mass

spectrometry to identify the amino acid sequences and parent proteins’ peptide

composition (analysed by Sangon Biotech Co., Ltd). Peptides were synthetized

13
selectively through solid phase synthesis (synthetized by China Peptides Co., Ltd.)

according to the amino acid sequences determined by liquid chromatography-mass

spectrometry.

3. Results and discussion

3.1 Purification and immobilization of extracellular protease

As some of the most important industrial enzymes, proteases have long been

used for industrial purposes, and the use of immobilized enzymes in industry is a

fascinating area of research. In this study, proteases were purified using anion

exchange chromatography. The specific activity of purified protease was 2586.97 ±

67.35 U/mg, while the specific activity of crude enzyme was 902.38 ± 30.32 U/mg.

Proteases were immobilized on chitosan beads via covalent binding. The specific

activity of immobilized proteases was up to 1700.42 ± 32.31 U/mg, with an

immobilization of 65.73 ± 1.38%. Thus, the protease from Vibrio sp. SQS2-3 in its

immobilized form could potentially be applied in industrial processes.

3.2 Preparation and biological activity of salmon skin collagen hydrolysate

The hydrolysate of salmon skin collagen was obtained after digestion with the

protease of Vibrio sp. SQS2-3. As depicted in Fig. 1a, collagen from salmon skin can

be digested by the protease. After 150 min of treatment, the hydrolysis degree did not

significantly increase, which indicated that the hydrolysis had reached its maximum.

To completely utilize collagen hydrolysate, it was subsequently fractionated into two

14
parts (designated as UF-1 and UF-2) by ultrafiltration.

Research has shown that collagen can be a source of anti-freezing peptides. For

example, Phanat et al. found peptides from the hydrolysate of blacktip shark skin that

exhibited a cytoprotective effect on surimi subjected to various freeze-thaw cycles

(Phanat, 2012). UF-1 was considered to be a potential anti-freezing fraction. As

shown in Fig. 2, the residual Ca2+-ATPase activity of myoglobin decreased obviously

after 3, 6, and 9 freeze-thaw cycles with distilled water and UF-2. When UF-1 was

added to myoglobin, the residual Ca2+-ATPase activity decreased significantly slower

than those with distilled water or UF-2 after freeze-thaw cycles.

During freeze-thaw cycles, myosin may denature, likely because ice crystals are

generated and the ionic strength increases, which destroys the original protein

structure (Benjakul & Sutthipan, 2009). Peptides released from collagen commonly

contain hydrophilic amino acid residues, which can bind water to prevent ice crystal

formation. In addition, small, short peptides are considered to have lower

anti-freezing activity because small peptides prefer to localize in the aqueous phase

(Phanat, 2012). Peptides with more amino acid residues can form α-helical structures

more easily. Jeong et al. reported that type I anti-freezing peptides with Ala and Thr

exist in an α-helical conformation to form antiparallel homo-dimers. Thus, type I

anti-freezing peptides could enhance their ice-binding and antifreeze activity by

increasing the amount of active sites. (Jeong, 2013). Damodaran also reported that the

-Gly-Pro-X- or -Gly-X-X- tripeptide repeat sequences in collagen peptides likely

15
make great contributions to the inhibition of ice crystal generation (Damodaran,

2007).

Fraction UF-1 (> 3 kDa) displayed persistent low DPPH scavenging activity

during hydrolysis, while the DPPH scavenging activity of the small fraction UF-2 (<

3 kDa) increased gradually with an improvement in the degree of hydrolysis (Fig. 1b).

After 150 min of treatment, the antioxidant activity of the product reached a

maximum (51.27 ± 1.97% in DPPH scavenging activity), close to that of the 180-240

min treatment, which indicated that the hydrolysis had reached a near-steady state.

3.3 Purification of antioxidant peptides from UF-2 and detection of antioxidant

activity

To obtain the antioxidant peptide, the smaller peptide fraction from ultrafiltration

(UF-2) was further purified using size exclusion liquid chromatography. As shown in

Fig. 3a, seven peptides fractions (F1-F7) were isolated from the smaller fraction of

ultrafiltration (UF-2) after UF-2 was purified by Sephadex LH-20. The antioxidant

activity assay showed that F5 exhibited much higher DPPH radical and hydroxyl

radical scavenging activities (73.29 ± 1.03% and 72.73 ± 3.34%, respectively) than

other isolated fractions (shown in Fig. 3b and 3c). The smallest fraction, F7, displayed

the second strongest antioxidant activity, which was obviously lower than that of F5.

From these results, we concluded that although F7 was the smallest peptide fraction

and displayed a certain degree of antioxidant activity, F5 possibly contained more

16
effective antioxidant peptides that could convert DPPH radicals and hydroxyl radicals

into stable products and terminate the radical chain reaction. Both F5 and F7 were

smaller than F1-F4, which proved that small peptides are more likely to exhibit

antioxidant activity.

To accurately evaluate the antioxidant activity of F5, two other assays were

performed, including the ORAC assay and an assay to evaluate the protective effect

against oxidation-induced DNA damage. The protective effect against

oxidation-induced DNA damage was analysed to elucidate the positive role of the

antioxidant peptide. In this study, the hydroxyl radical-generating system was based

on the Fenton reaction (Fe2+ + H2O2). The running rate of plasmid DNA in agarose gel

electrophoresis was in the order supercoiled DNA > linear DNA > open circular DNA

due to the differences in the spatial structures. In addition, the degree of DNA damage

was in the order linear DNA > open circular DNA > supercoil DNA. As shown in Fig.

4, most supercoiled DNA was converted into the open circular form due to hydroxyl

radical damage. The purified fraction F5 had a significant protective effect on

oxidation-induced DNA damage.

The ORAC assay was used to detect the antioxidant activity of the peptide to

quench peroxyl radicals (Elias, 2006). The protective effect of an antioxidant can be

measured by assessing the fluorescence decay curve of the sample compared to a

blank in which no antioxidant is present (Sheih, 2009). The time-dependent

fluorescence decay curves induced by AAPH are shown in Fig. 5. Fig. 5a shows that

the fluorescence intensity of the PBS group decreased rapidly, while F5 showed a

17
dose-dependent increase in the inhibition of fluorescence decay. The ORAC assay

was initiated by peroxyl radicals (LOO·) that also formed during lipid peroxidation.

Radical chain reactions caused by peroxyl radicals were blocked by the antioxidant

donating hydrogen. After that, the hydrogen donor formed a stable structure or

became a more stable radical.

3.4 Cytotoxicity test of purified peptides on human umbilical vein endothelial cell

The purified peptide fraction F5 was tested for its effect on the viability of

HUVECs using the MTT assay. Fraction F5 did not display significant cytotoxicity at

20 and 100 µg/ml (Fig. 6). In addition, it showed even higher cell viability in test

groups. This confirmed that the purified peptide fraction is safe towards HUVECs in

terms of cytotoxicity and can be used to evaluate the intracellular radical-scavenging

activity.

3.5 Intracellular radical-scavenging effects of peptides in human umbilical vein

endothelial cells

Hyperglycaemia is a factor that induces endothelial dysfunction in diabetes due to

oxidative stress (Hemang, 2013). To investigate intracellular radical-scavenging

effects, HUVECs were labelled with DCFH-DA. DCFH-DA can penetrate cells freely,

where it is hydrolysed by an intracellular esterase to DCFH, which is then trapped

inside cells. DCFH is non-fluorescent but is oxidized by intracellular ROS to form

fluorescing DCF. As shown in Fig. 7, cells treated with RPMI 1640 medium

18
containing 35 mM glucose displayed a stronger DCF-fluorescence intensity than cells

in normal RPMI 1640 medium, which indicated that glucose at a high concentration

could induce oxidative stress in HUVECs. Cells treated with different concentrations

of purified peptide fraction F5 displayed less fluorescence intensity than the

PBS-treated control group, suggesting that the purified peptide fractions could

scavenge intracellular ROS to protect HUVECs from radical-mediated damage in

hyperglycaemia.

3.6 Identification and verification of antioxidant peptide

Purified fraction F5 was considered the most important contributor to the

antioxidant activity of collagen hydrolysate. The peptide components of F5 were

analysed using liquid chromatography mass spectrometry. Nineteen peptides were

found in F5 (shown in Table. 1), and the peptide PMRGGGGYHY accounted for

more than 86% of the content of F5.

The sequence alignment showed that the peptide (PMRGGGGYHY) was part of

the alpha-3 chain of collagen type I. The peptide was synthetized according to the

sequence, and its antioxidant activity was detected using the ORAC assay. As shown

in Fig. 5b, the synthetized peptide displayed dose-dependent antioxidant activity.

Compared to the antioxidant activity of F5, as shown in Fig. 5c, the peroxyl radical

scavenging activity of PMRGGGGYHY (2.51 ± 0.14 mmol TE/g) was lower than the

antioxidant activity of F5 (2.78 ± 0.28 mmol TE/g), indicating that PMRGGGGYHY

may be one of the major peptides that contributes to the antioxidant activity of F5.

19
However, many other antioxidant peptides have not yet been identified.

Notably, this peptide contains several amino acid residues, including methionine,

tyrosine and histidine. Previous studies have reported that methionine residues

constitute an important antioxidant defence mechanism. Exposed methionine residues

work as a reactive site that scavenges oxidants through the formation of a sulfoxide

structure after oxidation to stop free-radical chain reactions (Rodney, 1996). The

methionine in PMRGGGGYHY was detected as oxidized, indicating that it may be an

active site of this antioxidant peptide. In addition, aromatic amino acids, such as

tyrosine, also play an important role in antioxidant activity as hydrogen donors (Wang,

2014). Tyrosine residues remove free radicals and change them into phenoxy radicals,

which are much more stable. This is because the reactive activity of unpaired

electrons is decreased by the formation of an electron-conjugated system, so the

propagation of the radical-mediated peroxidizing chain reaction could be inhibited

(Sheih, 2009). Histidine has been reported to be an important active site in antioxidant

peptides due to its imidazole characteristics; it can also function as a proton donor,

hydrogen donor, and lipid peroxyl radical trap (Niranjan, 2005; Je, 2007; Li, 2007). In

addition, Glycine is also considered to be a potential target site of free radicals by

supporting protons (Zhang, 2009). The backbone of peptides containing glycine are

more flexible, exposing functional groups to free radicals (Wu, 2015).

4. Conclusion

Many studies have shown that collagen is an ideal low-value protein resource that

20
can be used in bioactive peptide preparations. However, new bacterial proteases have

seldom been applied in previous studies. In addition, it is worth studying how to use

these protein resources comprehensively. This study aimed to explore the potential

application of bacterial extracellular protease in preparing a bioactive peptide. This

study investigated whether bacterial extracellular protease can be used to hydrolyse

collagen to prepare antioxidant and anti-freezing peptides in a single process. These

peptides were considered for use in reducing oxidative damage and as cryoprotectants

in protein storage. The characteristics and molecular mechanism will be further

studied in the future.

Acknowledgements

The work was supported by the National Natural Science Foundation of China

(31370104, 21205142), National Sparking Plan Project (2013GA770009), Opening

Foundation of the Chinese National Engineering Research Center for Control and

Treatment of Heavy Metal Pollution (No. 2015CNERC-CTHMP-07), Open-End Fund

for Valuable and Precision Instruments of Central South University (CSUZC201640),

and the Fundamental Research Funds for the Central Universities of Central South

University (2015zzts273, 2017zzts351, 2017zzts076).

21
References

Abhay N. S., Sushant S., Neeraj S., Vikash K. D. (2011). Glutaraldehyde-activated

chitosan matrix for immobilization of a novel cysteine protease, procerain B. Journal

of Agricultural Food Chemistry, 59, 6256-6262.

Alberto, D., Carmen, G.C., Begona, B. (2004). Extending applicability of the oxygen

radical absorbance capacity (ORAC-fluorescein) assay. Journal of Agricultural Food

Chemistry, 52, 48-54.

Alvarez, M. D., Fernández, C., & Canet, W. (2010). Oscillatory rheological properties

of fresh and frozen/thawed mashed potatoes as modified by different cryoprotectants.

Food and Bioprocess Technology, 3(1), 55-70.

Anson, M. (1938). The estimation of pepsin, trypsin, papain, and cathepsin with

hemoglobin. The Journal of General Physiology, 22, 79-89.

Benjakul, S., Seymour, T. A., Morrissey, M. T., An, H. (1997). Physicochemical

changes in Pacific whiting muscle proteins during iced storage. Journal of Food

Science, 62(4), 729-733.

Benjakul, S., Bauer, F. (2000). Physicochemical and enzymatic changes of cod

muscle proteins subjected to different freeze-thaw cycles. Journal of the Science of

Food and Agriculture, 80(8), 1143-1150.

Benjakul, S., Visessanguan, W., Thongkaew, C., Tanaka, M. (2003). Comparative

study on physicochemical changes of muscle proteins from some tropical fish during

22
frozen storage. Food Research International, 36(8), 787-795.

Benjakul, S., Sutthipan, N. (2009). Muscle changes in hard and soft shell crabs during

frozen storage. LWT-Food Science and Technology, 42(3), 723-729.

Damodaran, S. (2007). Inhibition of ice crystal growth in ice cream mix by gelatin

hydrolysate. Journal of Agricultural Food Chemistry, 55(26), 10918-10923.

Diaz, M. N., Frei, B., Vita, J. A., & Keaney, J. F. (1997). Antioxidant and

atherosclerotic heart disease. The New England Journal of Medicine, 337, 408–416.

Fiske, C. H., Subbarow, Y. (1925). The colorimetric determination of phosphorus.

Journal of Biological Chemistry, 66(2), 375-400.

Frankel, E. N. (2005). Lipid oxidation (2nd ed.). Bridgwater: The Oily Press.

Hemang P., Chen J., Kumuda C. D. and Mahendra K. (2013). Hyperglycemia induces

differential change in oxidative stress at gene expression and functional levels in

HUVEC and HMVEC. Cardiovascular Diabetology, 12, 142.

Je, J. Y., Qian, Z. J., Byun, H. G., Kim, S. K. (2007). Purification and characterization

of an antioxidant peptide obtained from tuna backbone protein by enzymatic

hydrolysis. Process Biochemistry, 42, 840-846.

Jeong, K. B., Jun, H. L., Ravichandran, N. M., Lee, S. G., Hackwon, D., Koh, H. Y.,

Shim, H. E., Kim, H. C., Kim, H. J. (2013). Antifreeze Peptides and Glycopeptides,

and Their Derivatives: Potential Uses in Biotechnology. Marine Drugs, 11, 2041;

23
doi:10.3390/md11062013.

Jin, S. K., Kim, I. S., Jung, H. J., Kim, D. H., Choi, Y. J., Hur, S. J. (2010). Effect of

cryoprotectants on chemical, mechanical and sensorial characteristics of spent laying

hen surimi. Food and Bioprocess Technology.

doi:10.1007/s11947-11010-10392-11947.

Li, B., Chen, F.,Wang, X., Ji, B., Wu, Y. (2007). Isolation and identification of

antioxidative peptides from porcine collagen hydrolysate by consecutive

chromatography and electrospray ionization-mass spectrometry. Food Chemistry, 102,

1135-1143.

Liu, D., Yang, X. H., Huang, J. F., Wu, R. B., Wu, C. L., He, H. L., Li, H. (2014). In

situ Demonstration and characteristic analysis of the protease components from

marine bacteria using substrate immersing zymography. Applied Biochemistry and

Biotechnology. doi: 10.1007/s12010-014-1287-2.

Ngo, D. H., Ryu, B. M., Kim, S. K. (2014). Active peptides from skate (Okamejei

kenojei) skin gelatin diminish angiotensin-I converting enzyme activity and

intracellular free radical-mediated oxidation. Food Chemistry, 143, 246-255.

Niranjan, R., Eresha, M., Jung, W. K., Je, J. Y., Kim, S. K. (2005). Purification of a

radical scavenging peptide from fermented mussel sauce and its antioxidant properties.

Food Research International, 38, 175-182.

Phanat, K., Soottawat, B., Wonnop, V., Fereidoon, S. (2012). Cryoprotective effect of

24
gelatin hydrolysate from blacktip shark skin on surimi subjected to different

freeze-thaw cycles. LWT - Food Science and Technology, 47, 437-442.

Qian, Z. J., Jung, W. K., Byun, H. G., Kim, S. K., (2008). Protective effect of an

antioxidative peptide purified from gastrointestinal digests of oyster, Crassostrea

gigas against free radical induced DNA damage. Bioresource Technology, 99,

3365-3371.

Ren, Y., Wu, H., Li, X. F., Lai, F. R., Zhao, G. L., Xiao, X. L. (2014). A two-step,

one-pot enzymatic method for preparation of duck egg white protein hydrolysates

with high antioxidant activity. Applied Biochemistry and Biotechnology, 172,

1227-1240.

Ruttanapornvareesakul, Y., Somjit, K., Otsuka, A., Hara, K., Osatomi, K., Osako, K.,

et al. (2006). Cryoprotective effects of shrimp head protein hydrolysate on gel

forming ability and protein denaturation of lizardfish surimi during frozen storage.

Fisheries Science, 72(2), 421-428.

Sheih, I. C., Wu, T. K., Fang, T. J. 2009. Antioxidant properties of a new

antioxidative peptide from algae protein waste hydrolysate in different oxidation

systems. Bioresource Technology, 100, 3419-3425.

Shimada, K., Fujikawa, K., Yahara, K., Nakamura, T. (1992). Antioxidative

properties of xanthan on the antioxidation of soybean oil in cyclodextrin emulsion.

Journal of Agricultural Food and Chemistry, 40, 945-948.

25
Spolarics, Z. (1998). Endotoxemia, pentose cycle, and the oxidant/antioxidant balance

in the hepatic sinusoid. Journal of Leukocyte Biology, 63, 534–541.

Stohs, S. J. (1995). The role of free radicals in toxicity and disease. Journal of Basic

and Clinical Physiology and Pharmacology, 6, 205–228.

Wang, B., Li, Z. R., Chi, C. F., Zhang, Q. H., Luo, H. Y. (2012). Preparation and

evaluation of antioxidant peptides from ethanol-soluble proteins hydrolysate of

Sphyrna lewini muscle. Peptides, 36, 240-250.

Wang, B., Wang, Y. M., Chi, C. F., Luo, H. Y., Deng, S. G., Ma, J. Y. (2013).

Isolation and Characterization of Collagen and Antioxidant Collagen Peptides from

Scales of Croceine Croaker (Pseudosciaena crocea). Marine Drugs, 11, 4641-4661;

doi:10.3390/md11114641.

Wang, B., Gong, Y. D., Li, Z. R., Yu, D., Chi, C. F., Ma, J. Y. (2014). Isolation and

characterisation of five novel antioxidant peptides from ethanol-soluble proteins

hydrolysate of spotless smoothhound (Mustelus griseus) muscle. Journal of

Functional Food, 6, 176-185.

Wu, R. B., Wu, C. L., Liu, D., Yang, X. H., Huang, J. F., Zhang, J., Liao, B. Q., He,

H. L., Li, H. (2015). Overview of antioxidant peptides derived from marine resources:

the sources, characteristic, purification, and evaluation methods. Applied Biochemistry

and Biotechnology, 176: 1815-1833.

Zhang, J., Zhang, H., Wang, L., Guo, X., Wang, X., Yao, H. (2009). Antioxidant

26
activities of the rice endosperm protein hydrolysate: identification of the active

peptide. European Food Research and Technology, 229, 709-719.

27
Figure legends

Figure 1. (a) Hydrolysis degree of collagen treated for 30, 60, 90, 120, 150, 180, 210
and 240 min. (b) DPPH radical scavenging activity of the ultrafiltration fraction of
collagen hydrolysates treated for 30, 60, 90, 120, 150, 180, 210 and 240 min. Values
are expressed as the mean ± SD (n=3).

Figure 2. Anti-freezing activity of UF-1-protecting Ca2+-ATPase activity of

myoglobin against freeze-thawing circles. Values are expressed as the mean ± SD
(n=3).

Figure 3. (a) Size-exclusion liquid chromatography of the UF-2 fraction from

ultrafiltration on a Sephadex LH-20 column. The F1 - F7 fractions represent purified
fractions; (b) DPPH radical scavenging activity of the purified fraction from size
exclusion liquid chromatography; (c) hydroxyl radical scavenging activity of the
purified fraction from size exclusion liquid chromatography. Values are expressed as
the mean ± SD (n=3).

Figure 4. Protective effect of purified peptide on hydroxyl radical-induced oxidation

of plasmid DNA. Lane 1, pET-22b plasmid DNA; Lane 2, Purified fraction F5; Lane
3, FeSO4 and H2O2 treatment (as DNA damage group); Lanes 4-5, FeSO4 and H2O2
treatment in the presence of purified fraction F5 at concentrations of 100 µg/ml and
150 µg/ml, respectively.

Figure 5. Peroxyl radical scavenging activity (ORAC assay) of (a) purified fraction F5
and (b) synthesized peptide PMRGGGGYHY. (c) Comparison of the peroxyl radical
scavenging activity of purified fraction F5 and synthesized peptide PMRGGGGYHY.

Figure 6. Cell viability determined by MTT assay. Cytotoxicity of purified peptides

from collagen hydrolysates on HUVECs was compared to non-treated group (blank).
Values are expressed as the mean ± SD (n=3).

Figure 7. Intracellular ROS in HUVECs indicated as green fluorescence by

28
DCFH-DA.

29
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Table 1

Table 1. Peptides identified through liquid chromatography mass spectrum

Amino acid sequence Source

PMRGGGGYHY (Oxidation @Met) alpha 3 type I collagen [Oncorhynchus mykiss]

PHTGIFNVPMEGHY EMILIN-2-like [Salmo salar]

PGTIVGSAAFN sodium/calcium exchanger 2-like [Salmo salar]

VGKVTNGGYH unnamed protein product [Oncorhynchus mykiss]

GITGPAGPRGPAGPHGPPGKDGRAGGH alpha 2 type I collagen [Oncorhynchus keta]

GESGPAGPAGPNGPA alpha 3 type I collagen [Oncorhynchus mykiss]

GIAGPAGPRGPSGPA alpha 1 type I collagen [Oncorhynchus mykiss]

SGPAGPAGPNGPA alpha 3 type I collagen [Oncorhynchus mykiss]

AGARSGADVDEGME unnamed protein product [Oncorhynchus mykiss]

GARGADGSTGPAGPAGPL type I collagen alpha 2 chain [Oncorhynchus keta]

LGFTQGSAAAGGG tumor necrosis factor receptor associated factor 2

[Oncorhynchus mykiss]

GPAGPAGPPGPKGPVGPPGPA unnamed protein product [Oncorhynchus mykiss]

GEGLLAVQITDPEGK unnamed protein product [Oncorhynchus mykiss]

GLNGPSGPRGPHGS alpha 3 type I collagen [Oncorhynchus mykiss]

DGRAGGHGAIGPVGH alpha 2 type I collagen [Oncorhynchus keta]

IAGPAGPRGPSGPA alpha 1 type I collagen [Oncorhynchus mykiss]

GPAGPRGPAGPHGPPG alpha 2 type I collagen [Oncorhynchus keta]

GSVGIAGGPGHQGPG alpha 2 type I collagen [Oncorhynchus keta]

AGGGYDQSGGYD alpha 2 type I collagen [Oncorhynchus keta]

Highlights

Salmon skin collagen was hydrolysed using immobilized/free marine bacterial

collagenase

The hydrolysates showed excellent anti-freezing (>3 kDa) and antioxidant
activity (< 3 kDa).

Activities of the purified peptides fraction included scavenging free radicals,
protecting DNA from oxidative damage, and reducing oxidative stress, based on
measures using ORAC and HUVEC.

19 novel peptides sequences were identified using mass spectrum.

30